Volumen 27 (2)Diciembre de 2008

MANUELA. S. GRAÇA. Coimbra

ARTURO ELOSEGI. Bilbao

Comité Editorial

J. Alba Tercedor. Granada, España L. Naselli-Flores. Palermo, Italia

M. J. Boavida. Lisboa, Portugal A. Palau. Lleida, España

J. Catalan. Barcelona, España C. Pedros-Alió. Barcelona, España

E. Chauvet. Toulouse, Francia. D. Planas. Montreal, Canadá

A. C. Encalada. Quito, Ecuador J. Pozo. Bilbao, España

T. Ferreira. Lisboa, Portugal N. Prat. Barcelona, España

G. George. Lancaster, UK A. Quesada. Madrid, España

P. Giller. Cork, Irlanda. A. Rodríguez Capítulo. La Plata, Argentina

H. L. Golterman. Arlés, Francia F. Rueda. Granada, España

M. A. S. Graça. Coimbra, Portugal K. Simek.

C. Granado. Sevilla, España J. G. Tundisi. Carlos, Brasil

D. Jakobsen. Hillerød, Dinamarca E. Vicente. Valencia, España

B. Malmqvist. , Suecia W. A. Wurtsbaugh. Logan, UT, USA

M. Molles. Albuquerque, NM, USA H. Zagarese. Chascomús, Argentina

Secretaría de Redacción

LIMNETICA está indexada en las siguientes bases de datos:

Editora adjunta

Umeå

Ceske Budejovice, Rep. Checa

Sao

Toda la correspondencia relativa a la ASOCIACION IBERICA DE LIMNOLOGIA incluida la peticion de altas y bajas desocios debe dirigirse a la Secretarıa de la Asociacion Iberica de Limnologıa. Arturo Elosegi. Departamento de Ecologıa.Facultad de Ciencia y Tecnologıa. Universidad del Paıs Vasco. Apartado de Correos 644, 48080-BILBAO. Pagina web dela Asociacion: http://www.limnologia.eu

Volumen 27. Numero 2. 2008

LIMNETICARevista de la

Asociacion Iberica de Limnologıa

c© Asociacion Iberica de Limnologıa

Deposito legal: V-2404-1986

ISSN: 0213-8409

Impresion: Graficas Rey, S.L.

Impreso en Espana/Printed in Spain

LIMNETICA Vol. 27 (2), 2008

INDICE

195 ARAUZO, M.; MARTINEZ-BASTIDA, J. J. Y M. VALLADOLID. Contaminacion por nitrogeno en el sistema “rıo-acuıferoaluvial” de la cuenca del Jarama (Comunidad de Madrid, Espana) ¿Origen agrıcola o urbano?

211 FRAILE, HENAR; LEONARDO, JOSE MANUEL; G. DE BIKUNA, BEGONA E ITZIAR LARUMBE. Seguimiento de la calidadde un embalse de abastecimiento de agua potable segun las directrices de la Directiva Marco (embalse del Anarbe.Cuenca Norte)

227 RUIZ GARCIA, ANTONIO AND MANUEL FERRERAS-ROMERO. Distribution patterns of Hydropsychids and Rhyacophi-lids species (Trichoptera) in a not regulated Mediterranean river (SW Spain)

239 SAHUQUILLO, MARIA; MIRACLE, MARIA ROSA; RIERADEVALL, MARIA AND RIYZARD KORNIJOW. Macroinverte-brates assemblages on reed beds, with special attention to Chironomidae (Diptera), in Mediterranean shallow lakes

251 TEIXEIRA-DE MELLO, FRANCO AND GABRIELA EGUREN. Prevalence and intensity of black-spot disease in fishcommunity from a subtropical stream (Santa Lucıa river basin, Uruguay)

259 GARCIA, LILIANA; DELGADO, CRISTINA AND ISABEL PARDO. Seasonal changes of benthic communities in a temporarystream of Ibiza (Balearic Islands)

273 GUISANDE, CASTOR; GRANADO-LORENCIO, CARLOS; TOJA, JULIA AND DAVID LEON. Identification of the mainfactors in structuring rotifer community assemblages in ponds of Donana National Park using the amino acid com-position of the species

285 FERNANDEZ-DIAZ, MARTA; BENETTI, CESAR JOAO AND JOSEFINA GARRIDO. Influence of iron and nitrate concentra-tion in water on aquatic Coleoptera community structure: Application to the Avia River (Ourense, NW. Spain)

299 GUTIERREZ-CANOVAS, CAYETANO; VELASCO GARCIA, JOSEFA AND ANDRES MILLAN SANCHEZ. SALINDEX: Amacroinvertebrate index for assessing the ecological status of saline “ramblas” from SE of the Iberian Peninsula

317 GONCALVES, VITOR; RAPOSEIRO, PEDRO AND ANA CRISTINA COSTA. Benthic diatoms and macroinvertebrates in theassessment of the ecological status of Azorean streams

329 FEIJOO, CLAUDIA; COMERMA, MARTA; MARCE, RAFAEL; GARCIA, JUAN CARLOS; BALAYLA, DAVID; NAVARRO,ENRIQUE AND JOAN ARMENGOL. Influence of phosphorus and irradiance on phytoplanktonic chlorophyll-a concen-tration and phosphorus contents at a diel scale in a Mediterranean reservoir

Limnetica, 27 (2): 195-210 (2008)Limnetica, 27 (2): 195-210 (2008)c© Asociacion Iberica de Limnologıa, Madrid. Spain. ISSN: 0213-8409

Contaminacion por nitrogeno en el sistema “rıo-acuıfero aluvial” de lacuenca del Jarama (Comunidad de Madrid, Espana) ¿Origen agrıcolao urbano?

M. Arauzo ∗,1, J. J. Martınez-Bastida 1,2 y M. Valladolid 3

1 Dpto. de Contaminacion Ambiental, Centro de Ciencias Medioambientales-CSIC, Serrano 115 dpdo. 28006Madrid, Espana.2 martinez.bastida@ccma.csic.es3 Dpto. de Biodiversidad y Biologıa Evolutiva, Museo Nacional de Ciencias Naturales-CSIC, Jose GutierrezAbascal 2, 28006 Madrid, Espana. E-mail: marval@mncn.csic.es2

∗ Autor responsable de la correspondencia: mercedes@ccma.csic.es2

Recibido: 22/9/07 Aceptado: 11/4/08

ABSTRACT

Nitrogen pollution in the “river-alluvial aquifer” system of the Jarama catchment (Comunidad de Madrid, Spain):Agricultural or urban origin?

A hydrochemical characterization of the “river-alluvial aquifer” system in the Quaternary deposits of the Jarama catchmenthas been performed, including the rivers Jarama, Henares, Manzanares, Tajuna and a part of Tajo, as well as the alluvialaquifer associated to the said fluvial net. The roles of agriculture and urban uses have been explored as possible sources ofnitrogen pollution in surface and underground water resources.Two sampling campaigns were performed, in March (at the end of the winter) and in August of 2005 (when irrigation waterdemand was at its highest), with water samples collected at 35 sampling stations (16 wells on the alluvial aquifer and 19 fluvialstations). The medium and low areas of the Manzanares, Jarama, Tajuna, and Tajo rivers did not meet the standards of qualityestablished by the Tajo Catchment Hydrologic, due to high levels of ammonia, QOD and. electric conductivity, attributableto the incorporation of water treatment effluents originating from urban areas, (in some areas, conductivity levels could beexplained by the geological context). The medium and low areas of the alluvial aquifer (in each subcatchment) showed veryhigh values of nitrate concentration and conductivity, making its use unsuitable for drinking and restricting it for agriculturalirrigation. A clear concordance was found between the spatial distribution of nitrate pollution in the alluvial aquifer and theagricultural irrigated areas. Using the N/P ratio as an indicator of the agricultural or urban origin it is interpreted that nitrogenpollution is mainly of urban origin in the rivers and of agricultural precedence in the alluvial aquifer (attributable to the badmanagement of fertilization and irrigation). The problem is compounded when water from nitrogen polluted rivers is usedfor irrigation, transferring nitrogen of fluvial origin into the alluvial aquifer with the irrigation return flows (urban nitrogen+ fertilizers nitrogen). On the other hand, it has been observed that the excess of irrigation reverts the natural dynamic ofthe aquifer, producing rises of the phreatic level in extensive areas during the summer. These results provide a scientificbasis to consider the declaration of the Quaternary alluvial deposits of the Jarama catchment as a Vulnerable Zone to nitratecontamination of agricultural origin, as established in the Directive 91/676/EEC.

Key words: “River-alluvial aquifer”, Jarama catchment, nitrogen, nitrate, ammonia, irrigated agriculture, urban effluents,Vulnerable Zones to nitrate pollution.

RESUMEN

Contaminacion por nitrogeno en el sistema “rıo-acuıfero aluvial” de la cuenca del Jarama (Comunidad de Madrid, Es-pana) ¿Origen agrıcola o urbano?

Se ha realizado una caracterizacion hidroquımica del sistema “rıo-acuıfero aluvial” situado en los depositos cuaternarios dela cuenca del Jarama, incluyendo los rıos Jarama, Henares, Manzanares, Tajuna y una parte del Tajo, ası como el acuıferoaluvial asociado a dicha red fluvial. Se ha explorado el papel de los usos agrıcolas y urbanos como fuentes potenciales decontaminacion por nitrogeno de los recursos hıdricos superficiales y subterraneos.

196 Arauzo et al.

Se realizaron dos campanas de muestreo, en marzo (final del invierno) y agosto de 2005 (cuando la demanda de agua parariego era maxima), con recogida de muestras de agua en 35 estaciones de muestreo (16 pozos sobre el acuıfero aluvial y19 estaciones fluviales). En los tramos medios y bajos de los rıos Manzanares, Jarama, Tajuna y Tajo no se alcanzaban losobjetivos de calidad establecidos en el Plan Hidrologico de la Cuenca del Tajo, debido a los elevados valores de amonio,DQO y conductividad electrica, atribuibles a la incorporacion de efluentes de depuradora procedentes de las areas urbanas(los niveles de conductividad pueden explicarse por el propio contexto geologico en algunas zonas). Los tramos medios ybajos del acuıfero aluvial (en cada subcuenca) mostraron valores muy elevados de nitrato y conductividad, imposibilitandosu uso para el abastecimiento y restringiendolo para el riego agrıcola. Se aprecia una clara concordancia entre la distribu-cion espacial de la contaminacion por nitrato en el acuıfero aluvial y las zonas agrıcolas de regadıo. Utilizando el cocienteN/P como indicador del origen agrıcola o urbano de la contaminacion por nitrogeno, se interpreta una procedencia prin-cipalmente urbana en los rıos y una procedencia agrıcola en el acuıfero aluvial (atribuible a las malas practicas de riegoy abonado). El problema se complica cuando los riegos se realizan con aguas fluviales contaminadas por nitrogeno, produ-ciendose un trasvase de nitrogeno de origen fluvial hacia el acuıfero aluvial con los retornos de riego (nitrogeno urbano +nitrogeno de los fertilizantes). Por otra parte, se ha observado que el riego en exceso invierte la dinamica hıdrica natural delacuıfero, produciendo ascensos en el nivel freatico en amplias zonas durante el verano. Estos resultados proporcionan unabase cientıfica para considerar la de claracion de los aluviales cuaternarios de la cuenca del Jarama como Zona Vulnerablea la contaminacion por nitrato de origen agrıcola, tal como se establece en la Directiva 91/676/CEE.

Palabras clave: “Rıo-acuıfero aluvial”, cuenca del Jarama, nitrogeno, nitrato, amonio, agricultura de regadıo, efluentesurbanos, Zonas Vulnerables a la contaminacion por nitratos.

INTRODUCCION

La contaminacion difusa tiende a adquirir cadavez mayor protagonismo en la degradacion de losrecursos hıdricos (Knapp, 2005), si bien, en terri-torios intensamente antropizados, con frecuenciano es facil identificar la procedencia de los conta-minantes en las masas de agua. Aspectos general-mente ignorados, como el estudio de las interac-ciones entre el rıo y su acuıfero aluvial asociado,o el papel de los usos del territorio en el deteriorode la calidad del agua, pueden proporcionar unainformacion esencial para la gestion sostenible delos recursos hıdricos a escala de cuenca.

Bajo las premisas de observacion a escala decuenca e interpretacion integral del sistema “rıo-acuıfero aluvial”, se ha realizado una caracteriza-cion hidroquımica del sistema “rıo-acuıfero alu-vial” situado en los depositos cuaternarios de lacuenca del rıo Jarama (Comunidad de Madrid,Espana), incluyendo el rıo Jarama, sus afluentesHenares, Manzanares y Tajuna, una parte del rıoTajo (a su paso por la Comunidad de Madrid),y el acuıfero aluvial asociado a dicha red flu-vial. Los objetivos principales de la investigacion

fueron determinar los niveles de contaminacionpor nitrogeno en los rıos y en el acuıfero aluvial,identificar las formas quımicas de nitrogeno do-minantes en cada parte del sistema rıo-acuıferoy explorar el papel de las principales fuentes denitrogeno en el proceso de contaminacion (lixi-viacion de fertilizantes en las zonas agrıcolas deregadıo y vertido de efluentes de depuradora pro-cedentes de las areas urbanas).

La cuenca del rıo Jarama es la mas extensay antropizada de la Comunidad de Madrid. Susrecursos hıdricos superficiales se ven sometidosa una intensa e insostenible demanda de aguay a una elevada carga contaminacion del origenurbano e industrial. Existe abundante informa-cion sobre el deficiente estado de calidad en lostramos medios y bajos del rıo Jarama y algu-nos de sus tributarios (Confederacion Hidrografi-ca del Tajo, 2005a), siendo el nitrogeno uno delos contaminantes de mayor presencia. De he-cho, en el Plan Hidrologico de la Cuenca del Tajo(R.D. 1664/1998; Orden 18236 de 13 de agostode 1999) se reconoce implıcitamente la imposi-bilidad de recuperacion de la calidad del agua enesos tramos, al excluir entre sus objetivos de ca-

Contaminacion por nitrogeno en la cuenca del Jarama 197

lidad los usos para abastecimiento, vida piscıcolay bano. Por otra parte, sobre los depositos alu-viales de la cuenca del Jarama se situan las de-nominadas Vegas de la Comunidad de Madrid(Vega del Jarama, del Henares, del Tajuna y delTajo), cuyas economıas se basan principalmen-te en la agricultura de regadıo. La infiltracion enel terreno de aguas con alto contenido en nitra-to, como resultado de una excesiva fertilizacionnitrogenada y unas practicas de riego poco opti-mizadas (Hall et al., 2001), contribuye al dete-rioro de los recursos hıdricos subterraneos (Ni-xon et al., 2000; Ball et al., 2005; Heathwaite etal., 2005; Abrantes et al., 2006). El Ministeriode Medio Ambiente (2001) senala que la cuencamedia-baja del rıo Jarama es una de las que so-portan mayores aportes globales de nitrogeno enEspana, debido a los usos agrıcolas.

El exceso de nitrato en las aguas destinadas aabastecimiento puede afectar a la salud humana(Varela, 1994; Morales-Suarez et al., 1995; San-dor et al., 2001; Forman, 2004; Thorpe & Shir-mohammadi, 2005) y contribuye al desarrollo deprocesos de eutrofizacion en las aguas superficia-les (FAO/CEPE, 1991; Neal & Jarvie, 2005). LaDirectiva 91/676/CEE regula la proteccion de lasaguas contra la contaminacion por nitrato de ori-gen agrıcola. En esta Directiva se define comoZona Vulnerable a aquella superficie del territo-rio cuya escorrentıa o filtracion afecte o puedaafectar a la contaminacion del agua por el nitra-to procedente de los fertilizantes, considerando-se aguas afectadas por la contaminacion aque-llas con contenido en nitrato superior a 50 mg/l,ası como las que manifiestan una tendencia quehaga prever la necesidad de medidas de protec-cion. La Directiva Marco del Agua fija los plazosde cumplimiento de los objetivos medioambien-tales en las Zonas Vulnerables y determina losprogramas de seguimiento de las masas de agua.Asimismo, adopta como unidad de planificacionpara el control de los procesos de contaminacionpuntual y difusa la cuenca fluvial.

Las prospecciones de masas de agua con-taminadas por nitrato en las que se ha inclui-do el territorio de la Comunidad de Madridse han circunscrito a los acuıferos del Tercia-rio, no habiendo dado lugar declaracion de Zo-

nas Vulnerables (Instituto Tecnologico Geomine-ro de Espana, 1998; Ministerio de Medio Am-biente, 2001; Confederacion Hidrografica del Ta-jo, 2005b). La naturaleza permeable del sustratoy la escasa profundidad del nivel freatico en losterrenos aluviales contribuyen a incrementar lavulnerabilidad de las masas de agua subyacentes.Sin embargo, a pesar de la combinacion de facto-res de riesgo que convierten a los acuıferos alu-viales de los depositos cuaternarios de la Comu-nidad de Madrid en altamente vulnerables, hastaahora no se habıa abordado su estudio. Navas etal. (1998) describen a la unidad cuaternaria co-mo “de vulnerabilidad muy alta, constituida pordepositos fluviales de gravas y arenas, y en me-nor proporcion limos y arcillas, con una zona nosaturada altamente permeable y nivel freatico amenos de 5 m de profundidad”, si bien no deta-llan los procesos de contaminacion que afectana los recursos hıdricos de los depositos cuaterna-rios. La carencia de informacion sobre el estadode calidad del acuıfero aluvial de la cuenca delJarama podrıa explicarse por la gran dificultad enla localizacion de puntos de muestreo, puesto queexisten ya pocos pozos accesibles (Las Vegas dela Comunidad deMadrid cuentan con sistemas deacequias para el riego, con agua procedente de lared fluvial, habiendo desaparecido la mayor partede los pozos en el area aluvial). De ahı que unode los objetivos parciales de este trabajo haya si-do establecer una primera red de muestreo para lacaracterizacion hidroquımica del acuıfero aluvialde la cuenca del Jarama.

Los elevados niveles de nitrogeno en el sis-tema “rıo-acuıfero aluvial” de la cuenca de Ja-rama precisa del esclarecimiento de las posi-bles fuentes de contaminacion y de sus efec-tos sobre cada parte del sistema. Sin embargo,en areas tan antropizadas como la que nos ocu-pa existe un importante grado de incertidum-bre en el diagnostico de las fuentes de conta-minacion que plantea diversos de interrogantes:¿Donde y cuando se produce la contaminacionpor nitrogeno? ¿En que formas quımicas? ¿Comoafecta a cada parte del sistema “rıo-acuıferoaluvial”? ¿Es atribuible a un origen urbano,agrıcola o mixto? ¿Donde ha de establecerseel lımite de las responsabilidades?

198 Arauzo et al.

Figura 1. Localizacion de las estaciones de muestreo en los rıos y en el acuıfero aluvial. En color gris claro: Extension del acuıferoaluvial sobre los depositos cuaternarios de la cuenca del Jarama (incluyendo una parte de la cuenca del Tajo). En color gris oscuro:Distribucion del area agrıcola de regadıo en la Comunidad de Madrid (fuente: MAPYA, 2005). Se muestra la situacion de los princi-pales nucleos urbanos (> 25 000 habitantes). Location of the sampling stations on the rivers and alluvial aquifer. Light grey colour:Area of the alluvial aquifer over the quaternary deposits of the Jarama catchment (including a part of the Tajo catchment). Dark greycolour: Distribution of the irrigated agriculture area in the Comunidad de Madrid (source: MAPYA, 2005). Main urban areas areshown (> 25 000 inhabitants).

MATERIAL Y METODOS

Area de estudio

El area de estudio queda delimitada por la super-ficie que ocupan los depositos aluviales del Cua-ternario de la Comunidad de Madrid (Espana) so-bre los que se situa el sistema “rıo-acuıfero alu-vial” de la cuenca del rıo Jarama, incluyendo elrıo Jarama y sus afluentes Henares, Manzanares

y Tajuna, una parte del rıo Tajo (a su paso por laComunidad de Madrid) y el acuıfero aluvial aso-ciado a dicha red fluvial.

El acuıfero aluvial es de tipo libre y su exten-sion queda definida por la propia superficie alu-vial (1480 km2). Su sustrato geologico lo consti-tuyen gravas poligenicas, arenas y limos (prime-ras terrazas aluviales) y arenas, limos arenososy cantos (fondos de valle), todos ellos depositosde alta permeabilidad (ITGME-CAM, 1988). Los

Contaminacion por nitrogeno en la cuenca del Jarama 199

perfiles edaficos dominantes son de tipo Fluvi-sol y Luvisol, tambien muy permeables (Guerra yMonturiol, 1970; Monturiol y Alcala 1990). De-bido a su probable desconexion con el acuıferoaluvial principal, las masas de agua de menor en-tidad situadas en las terrazas medias y altas nohan sido incluidas en el estudio.

En la figura 1 se muestra un mapa de la redfluvial y se delimita la extension de los depositoscuaternarios, en los que se encuentra el acuıferoaluvial. Las estaciones de muestreo de la red flu-vial se denominaron RJno, RHno, RMno, Rtnno

y RTno, mientras que los pozos para el mues-tro del acuıfero aluvial se nombraron como PJno,PHno, PMno, Ptnno y PTno, asignandose una nu-meracion creciente desde las zonas de cabecerahasta los tramos bajos y utilizando los siguien-tes codigos de cuenca: Jarama (J), Henares (H),Manzanares (M), Tajuna (tn) y Tajo (T). Tambiense representan las superficies destinadas a agri-cultura de regadıo en la Comunidad de Madrid(MAPYA, 2005), denominadas localmente Vegasdel Jarama, el Henares, el Tajuna y el Tajo, y sesituan los principales nucleos urbanos. Cabe des-tacar que solo el area metropolitana de Madridcuenta con una poblacion de mas de 5 800 000habitantes, sobrepasandose en Alcala de Hena-res, Torrejon, Coslada, Guadalajara y Aranjuezlos 40 000 habitantes. Segun la localizacion delos principales nucleos urbanos (Fig. 1), los rıosque potencialmente podrıan verse mas afectadospor el vertido de efluentes de depuradora serıanel Manzanares, el Jarama, el Henares y el Tajo(aguas abajo de su confluencia con el Jarama).

En la Tabla 1 se muestran algunos parametrosde localizacion de las estaciones de muestreo y seespecifican los objetivos de calidad establecidosen el Plan Hidrologico de la Cuenca del Tajo paracada tramo de los rıos y del acuıfero aluvial.

Metodologıa

Del 6 al 8 de marzo y del 1 al 3 de agosto de2005 se realizaron dos campanas de muestreo,con recogida de muestras de agua en 19 esta-ciones fluviales (rıos Jarama, Henares, Manzana-res, Tajuna y Tajo) y en 16 pozos situados en el

acuıfero aluvial (Fig. 1). La eleccion de los pe-riodos de muestreo responde a la necesidad deexplorar los efectos de la agricultura de regadıosobre la calidad del agua y la dinamica hıdricaen el sistema rıo-acuıfero: periodo final del in-vierno (tras la recarga invernal, momento en el seesperaba que la superficie freatica se encontrarseen su nivel mas alto) y periodo de verano (des-pues de la fertilizacion y durante la maxima de-manda de agua para riego, con extracciones porbombeo desde los cauces fluviales).

En cada campana se realizaron medidas in si-tu de temperatura del agua, porcentaje de satu-racion de oxıgeno, pH y conductividad electri-ca en todas las estaciones de muestreo (rıos yacuıfero). Para ello se utilizo un sistema multi-parametrico portatil conectado a multisonda y kitpH/redox modelo YSI 556. Las medidas de coor-denadas UTM y altitud se realizaron medianteun GPS modelo Garmin GPS 12. Se efectuaronanotaciones sobre los usos del suelo en cada pun-to de muestreo (zona natural, uso terciario, erial,agricultura de regadıo, tipo de cultivo, tipo de rie-go, etc.). En los pozos tambien se midio la pro-fundidad del nivel freatico, mediante un Hidro-nivel Meyer. Se extrajeron muestras de agua detipo simple en los rıos, e integradas cada dos me-tros de profundidad en los pozos. Las muestras deagua de pozo se recogieron mediante una botellamuestreadora de apertura horizontal de 2.5 l, obien usando un Hidronivel Meyer dotado de bo-tella tomamuestras. Sobre cada muestra se ana-lizaron los siguientes parametros: Nitrogeno to-tal, nitrato, nitrito, amonio, fosforo total, sulfatos,carbonatos, bicarbonatos, cloruros, calcio, pota-sio, magnesio, sodio, sılice y demanda quımicade oxıgeno (DQO). Los aniones se determinaronmediante cromatografıa ionica, mientras que pa-ra el analisis de cationes se utilizo la espectro-metrıa de emision ICP-AES (plasma de acopla-miento inductivo). Se realizaron clasificacionesdel agua segun la dureza y segun el contenidoionico (diagrama de Piper). La DQO se deter-mino mediante la tecnica del dicromato potasi-co (APHA, 1998). El nitrogeno total y el fosfo-ro total por colorimetrıa, el primero mediante elmetodo 2.6-dimetil fenol (Lange, 1998) y el se-gundo mediante el metodo del fosfomolibdeno

200 Arauzo et al.

Tabla 1. Parametros de localizacion de las estaciones de muestreo, objetivos de calidad establecidos en el Plan Hidrologico de laCuenca del Tajo y clasificaciones del agua segun la dureza y el diagrama de Piper. Location parameters of the sampling stations,quality objectives established in the Tajo Catchment Hydrologic Plan and water classifications based on the water hardness and thePiper diagram.

Estacion Enclave Coordenadas UTM Altitud Objetivos de calidad Dureza Diagrama de Piper

(m) Plan Hidrologico ◦ F Clasificacion Clasificacion

RH1 Rıo Henares 30T 0490257 4521152 700 A3 050 dura sulfatada calcicaRH2 Rıo Henares 30T 0480489 4491619 685 A3 060 muy dura sulfatada calcicaRH3 Rıo Henares 30T 0480458 4491621 622 — 049 dura sulfatada calcicaRH4 Rıo Henares 30T 0461100 4476681 566 — 048 dura sulfatada calcica

RM1 Rıo Manzanares 30T 0424998 4510670 910 A2 001 muy blanda bicarbonatada sodicaRM2 Rıo Manzanares 30T 0436721 4479229 664 — 010 blanda bicarbonatada sodicaRM3 Rıo Manzanares 30T 0455431 4463738 540 — 036 dura bicarbonatada sodica

Rtn1 Rıo Tajuna 30T 0499474 4491742 755 A2 031 medianam. dura bicarbonatada calcicaRtn2 Rıo Tajuna 30T 0451511 4432779 511 A3 074 muy dura sulfatada calcicaRtn3 Rıo Tajuna 30T 0451211 4442648 510 A3 078 muy dura sulfatada calcica

RJ1 Rıo Jarama 30T 0460729 4524400 916 A2 022 medianam. dulce bicarbonatada calcicaRJ2 Rıo Jarama 30T 0451866 4492768 603 A2 037 dura sulfatada calcicaRJ3 Rıo Jarama 30T 0456864 4471829 551 — 036 dura sulfatada calcicaRJ4 Rıo Jarama 30T 0458842 4464220 533 — 049 dura sulfatada calcicaRJ5 Rıo Jarama 30T 0454076 4454079 520 — 039 dura sulfatada calcicaRJ6 Rıo Jarama 30T 0448153 4437763 507 — 045 dura sulfatada calcica

RT1 Rıo Tajo 30T 0499442 4450505 730 A2 057 muy dura sulfatada calcicaRT2 Rıo Tajo 30T 0451507 4432708 514 A2 093 muy dura sulfatada calcicaRT3 Rıo Tajo 30S 0436933 4423813 491 A2 059 muy dura sulf. clor. calcica-sodica

PH1 Acuıfero aluvial, Henares 30T 0489886 4520641 706 A3 030 medianam. dura sulfatada calcicaPH2 Acuıfero aluvial, Henares 30T 0483281 4498618 652 A3 075 muy dura sulfatada calcicaPH3 Acuıfero aluvial, Henares 30T 0478599 4493268 634 — 058 muy dura sulf. bic. cal. magnesicaPH4 Acuıfero aluvial, Henares 30T 0475736 4485694 599 — 091 muy dura sulf. calcico magnesicaPH5 Acuıfero aluvial, Henares 30T 0461997 4476072 576 — 168 muy dura sulf. calcico-magnesica

PM1 Acuıfero aluvial, Manzanares 30T 0453886 4464873 546 — 178 muy dura sulfatada calcica

Ptn1 Acuıfero aluvial, Tajuna 30T 0472644 4453911 578 A3 131 muy dura sulfatada calcicaPtn2 Acuıfero aluvial, Tajuna 30T 0459527 4447349 540 A3 104 muy dura sulfatada calcicaPtn3 Acuıfero aluvial, Tajuna 30T 0452402 4443247 506 A3 108 muy dura sulfatada calcica

PJ1 Acuıfero aluvial, Jarama 30T 0452814 4491800 587 A2 056 muy dura sulfatada calcicaPJ2 Acuıfero aluvial, Jarama 30T 0458922 4463226 537 A2 054 muy dura sulfatada calcicaPJ3 Acuıfero aluvial, Jarama 30T 0451776 4451772 510 — 067 muy dura sulfatada calcicaPJ4 Acuıfero aluvial, Jarama 30T 0447148 4436826 487 — 203 muy dura clor. sulfatada sodica

PT1 Acuıfero aluvial, Tajo 30T 0484078 4438055 548 A2 163 muy dura sulfatada calcicaPT2 Acuıfero aluvial, Tajo 30T 0453997 4434530 501 A2 115 muy dura sulfatada calcicaPT3 Acuıfero aluvial, Tajo 30S 0429397 4420288 475 A2 086 muy dura sulfatada calcica

(Murphy & Riley, 1962). Para la determinaciondel amonio se uso el metodo colorimetrico delindofenol azul (Lange, 1998).

El cociente N/P se calculo a partir de las me-didas de nitrogeno total y de fosforo total, y seutilizo como un indicador del origen agrıcola ourbano de la contaminacion por nitrogeno: Enaguas naturales no contaminadas, o cuando sesospecha que la procedencia del enriquecimien-to en nitrogeno es de origen urbano (efluentes de

depuradora) los valores de cociente N/P se en-cuentran en torno a 12-16 (San Diego-McGlongeet al., 2000). Valores mas elevados (que puedensuperar en uno y en dos ordenes de magnitud losanteriores) podrıan indicar contaminacion de ori-gen agrıcola. Esto se debe a que en suelos basi-cos o muy ricos en calcio, como los del areade estudio (Guerra y Monturiol, 1970; Monturioly Alcala 1990), o bien cuando el agua de rie-go es calcarea (Tabla 1), el fosforo pasa rapida-

Contaminacion por nitrogeno en la cuenca del Jarama 201

mente a ser insoluble, a diferencia del nitrato, yno presenta problemas de lixiviacion (College ofAgricultural Sciences, Agricultural Research andCooperative Extension, 2001).

La variable denominada “superficie de re-gadıo aguas arriba de cada pozo” se elaboro su-perponiendo el mapa de las superficies destina-das a agricultura de regadıo (MAPYA, 2005)y los puntos de muestreo situados sobre el acuıfe-ro aluvial (Fig. 1), y estimando la distancia li-neal maxima (en km) indicadora de la extensionagrıcola aguas arriba de cada pozo, a fin explo-rar la relacion entre la superficie dedicada al re-gadıo en el area de influencia de cada punto demuestreo del acuıfero y los niveles de contami-nacion por nitrato en el mismo.

Los mapas de isolıneas de concentracion denitrato en el acuıfero aluvial se elaboraron a partirde las concentraciones de nitrato obtenidas encada punto, trazando isolıneas por interpolaciontriangular e interpretativa, teniendo en cuenta elsentido del flujo del agua. En losmapas de isolıneasde nitrato tambien representa la distribucion delcontenido ennitrogeno total en los rıos de la cuenca.

Los datos de precitacion se obtuvieron de unaestacion meteorologica modelo Vantage Pro Plus,situada en la zona media del area de estudio (Fin-

ca Experimental La Poveda CCMA-CSIC). Lainformacion geologica y edafica se recopilo apartir de ITGME-CAM (1988), Guerra y Mon-turiol (1970) y Monturiol y Alcala (1990).

RESULTADOS Y DISCUSION

La distribucion del territorio destinado a la agri-cultura de regadıo en la Comunidad de Madrid(MAPYA, 2005) presenta un apreciable solapa-miento con los tramos medios y bajos del acuıfe-ro aluvial situado en los depositos cuaternarios(Fig. 1). Durante la campana de marzo se ob-servo la dominancia de rastrojeras de cereal deinvierno, mientras que en el verano el cultivoprincipal era el maız, seguido de hortıcolas yotras forrajeras. Cultivos arboreos, vinedos, zo-nas de recuperacion de soto, eriales y areas desti-nadas a uso terciario, coexistıan con los cultivosherbaceos, aunque en menor extension. Durantela campana de agosto se constato el uso generali-zado del riego por inundacion en los cultivos demaız, alfalfa y arboreos de las zonas correspon-dientes a las estaciones PH2, PH3, PJ3, PT1 yPT2, mientras que en el resto de la superficie de-dicada al regadıo se aplicaba riego por aspersion.

Figura 2. Nivel freatico del acuıfero aluvial al final del invierno (tras la recarga invernal) y durante el verano de 2005 (periodo demaxima demanda de agua para riego agrıcola). Las flechas senalan los puntos del acuıfero en los que se produjeron ascensos del nivelfreatico durante el verano. Phreatic level of the alluvial aquifer at the end of winter (after winter recharge) and during the summer of2005 (period of highest water demand for agricultural irrigation).The arrows show the points of the aquifer where higher phreaticlevels occurred during the summer.

202 Arauzo et al.

La escasa profundidad del nivel freatico en elarea de estudio (valor medio anual ± desviacionestandar: 4.3 ± 1.9 m; Fig. 2) y la elevada per-meabilidad del suelo (mayoritariamente de tipoFluvisol, Luvisol) y del sustrato litologico (gra-vas arenas y limos en fondos de valle y prime-ras terrazas aluviales), confieren una alta vulne-rabilidad al acuıfero aluvial (ya referida con an-terioridad por Navas et al., 1998). Las variacio-nes invierno-verano del nivel freatico revelan unasituacion singular, observandose una clara inver-sion en la dinamica natural de recarga del acuıfe-ro en amplias zonas del mismo (Fig. 2). El nota-ble ascenso del nivel freatico durante agosto de2005 en los puntos PH3, Ptn3, PJ3, PJ4, PT1,PT2 y PT3 respecto a los niveles de marzo, podrıaexplicarse por las recargas procedentes de los re-tornos de riego (Arauzo et al. 2007; Martınez-Bastida et al. 2008), puesto que apenas se regis-

traron precipitaciones durante ese periodo (apor-te total por lluvia: 29 mm). En los puntos en losque se practica el riego por inundacion, el ascen-so estival del nivel freatico fue significativamen-te superior con relacion al resto de los puntos delacuıfero (test t de Student: t = 3.74, p < 0.01).La procedencia mayoritariamente fluvial de lasaguas de riego en la cuenca del Jarama, en uncontexto de practicas de riego poco optimizadasy alta permeabilidad de la zona no saturada, fa-vorecio el trasvase artificial de agua desde la redfluvial al acuıfero aluvial, alterando la dinamicahıdrica natural en el sistema rıo-acuıfero.

La composicion ionica del agua en la red flu-vial y en el acuıfero aluvial se representa en eldiagrama de Piper de la figura 3. En la cuenca do-mina el agua sulfatada-calcica (con ciertas varia-ciones en concordancia con el contexto geologi-co local), si bien el rıo Manzanares presenta un

Figura 3. Diagrama de Piper. Piper diagram.

Contaminacion por nitrogeno en la cuenca del Jarama 203

Figura 4. Temperatura del agua, conductividad electrica, pH, porcentaje de saturacion de oxıgeno y DQO en los rıos y en el acuıferoaluvial al final del invierno y durante el verano de 2005.Water temperature, electric conductivity, pH, percentage of oxygen saturationand QOD in the rivers and alluvial aquifer at the end of the winter and during the summer of 2005.

204 Arauzo et al.

agua de tipo bicarbonatada sodica y las cabece-ras del Jarama y el Tajuna de tipo bicarbonatadacalcica (Tabla 1). La clasificacion segun la dure-za del agua muestra tipologıas que varıan de duraa muy dura, tanto en los rıos como en el acuıfero,excluyendo las zonas de cabecera y tramo mediodel rıo Manzanares, de tipo muy blanda a blanda(Tabla 1). En las figuras 4 y 5 se representan losparametros de calidad analizados en la red fluvialy en el acuıfero aluvial. En general, se observa unbuen nivel de calidad del agua en las zonas altasdel sistema rıo-acuıfero. Sin embargo, en los pun-tos RM3, Rtn2, Rtn3, RJ3, RJ4, RJ5, RJ6, RT2 yRT3 de la red fluvial no se alcanzan los objeti-vos de calidad establecidos en el Plan Hidrologi-co de la Cuenca del Tajo (R.D. 1664/1998; Orden18236 de 13 de agosto de 1999). Se observan va-lores muy elevados de amonio y DQO en el tramobajo del rıo Manzanares, el tramo medio-bajo delJarama y el tramo del Tajo posterior la incorpora-cion del Jarama, atribuibles a la incorporacion deefluentes de aguas residuales tratadas, proceden-tes de las principales nucleos urbanos (Fig. 1).En el tramo bajo del rıo Manzanares tambiense excede el lımite fijado para el fosforo (igual-mente explicable por los aportes urbanos). Lostramos medio del Henares y medio-bajo del Ta-juna, localizados en el area de influencia de faciesmargo-yesıferas del Mioceno, presentan valoresde conductividad electrica por encima del valorde referencia. Respecto a la calidad del agua enel acuıfero aluvial, la conductividad y la concen-tracion de nitrato tambien exceden los lımites es-tablecidos como objetivos de calidad en el PlanHidrologico en una parte considerable de su ex-tension (Fig. 4 y 5), lo cual imposibilita su usocomo reserva para abastecimiento y confiere ungrado de restriccion para el uso en riego agrıcolade moderado a severo (Ayers & Westcot, 1985).

En la red fluvial el nitrogeno total pre-sento una distribucion creciente desde las zonasde cabecera hacia los tramos bajos de cada sub-cuenca (Fig. 5). En el acuıfero aluvial, sin em-bargo, no se aprecio siempre un patron similar,apareciendo areas con mayores concentracionesde nitrogeno en los tramos medios de las sub-cuencas del Henares y el Tajuna (Fig. 5). Las di-ferentes formas quımicas del nitrogeno tampoco

mostraron una distribucion uniforme en el siste-ma rıo-acuıfero. Al ser la zona no saturada unsistema oxidante abierto, la forma quımica do-minante en todo el acuıfero aluvial es el nitrato(Fig. 5). Sin embargo, en los tramos medios y/obajos de los rıos Manzanares, Jarama, Henares yTajo (despues de la incorporacion del Jarama) elamonio es con frecuencia mas abundante que elnitrato y las concentraciones de nitrito son gene-ralmente muy elevadas (Fig. 5), en concordanciacon los valores maximos de DQO y fosforo totaly los valores mas bajos en saturacion de oxıgeno(Fig. 4 y 5), parametros todos ellos que indicanla incorporacion de aguas residuales urbanas conun mayor o menor grado de tratamiento.

En la figura 6 se muestran los mapas de distri-bucion del contenido en nitrato en el acuıfero alu-vial de la cuenca del Jarama y parte de la cuencadel Tajo (a su paso por la Comunidad de Madrid)durante las dos campanas de muestreo. Se esti-ma que el 36% del area total del acuıfero aluvialpresento concentraciones de nitrato superiores a50 mg/l (lımite maximo establecido por la Direc-tiva 91/676/CEE), el 25% entre 25 y 50 mg/l (laDirectiva 75/440/CEE establece un lımite guıa de25 mg/l), y unicamente en el 39% se registraronconcentraciones inferiores a 25 mg/l. La distribu-cion espacial del nitrato no presento incrementosimportantes entre el final del invierno y el periodoestival, salvo en el tramo medio del Tajuna, el tra-mo bajo del Jarama y el tramo del Tajo posteriora la incorporacion del Jarama, con concentracio-nes de nitrato entre 75 y 125 mg/l. Arauzo et al.(2006a y b) observaron una dinamica invierno-verano en la concentracion de nitrato mucho masintensa en los acuıferos aluviales de la cuencadel Oja-Tiron (La Rioja-Castilla y Leon) expli-cable, en ese caso, por la alta tasa de recargade los mismos al final del invierno, que permitıacierta recuperacion temporal.

A diferencia del acuıfero aluvial, el nitrogenoen los rıos de la cuenca del Jarama aparece ba-jo distintas formas quımicas (Fig. 5), de ahı queen la figura 6 se opte por la representacion comonitrogeno total. Ya se ha mencionado el probableorigen urbano del nitrogeno en los tramos conta-minados de los rıos (Fig. 6), solo cabe precisarque no se aprecian diferencias importantes en los

Contaminacion por nitrogeno en la cuenca del Jarama 205

Figura 5. Nitrogeno total, nitrato, nitrito, amonio, fosforo total y cociente N:P en los rıos y en el acuıfero aluvial al final delinvierno y durante el verano de 2005. Total nitrogen, nitrate, nitrite, ammonia, total phosphorous and N:P ratio in the rivers andalluvial aquifer at the end of the winter and during the summer of 2005.

206 Arauzo et al.

Marzo de 2005 Agosto de 2005

0-25

25-50

50-75

75-100

100-12510 Km

0-6

6-12

12-18

18-24

Figura 6. Distribucion del contenido en nitrato en el acuıfero aluvial y del contenido en nitrogeno total en los rıos, al final delinvierno y durante el verano de 2005. Nitrate content distribution in the alluvial aquifer and total nitrogen content in the rivers, atthe end of the winter and during the summer of 2005.

niveles de contaminacion fluvial por nitrogenoentre el invierno el verano.

En areas tan antropizadas como la que nosocupa existe un amplio grado de incertidumbreen el diagnostico de las fuentes de contamina-cion difusa de los recursos hıdricos. Una prime-ra aproximacion visual nos ha permitido apreciarel solapamiento entre la distribucion de las areas

destinadas al regadıo en la Comunidad de Ma-drid (Fig. 1) y las zonas que presentan maximasconcentraciones de nitrato en el acuıfero aluvial(Fig. 6). Las correlaciones significativas entre laconcentracion de nitrato en el acuıfero y la exten-sion dedicada al regadıo aguas arriba de cada po-zo, y entre la concentracion de nitrato y el cocien-te N/P en todas las estaciones (Tabla 2) vienen

Contaminacion por nitrogeno en la cuenca del Jarama 207

Tabla 2. Correlaciones de Pearson entre la concentracion denitrato en las estaciones de muestreo del acuıfero y la superficiede regadıo aguas arriba de cada pozo, y entre la concentracionde nitrato y el cociente N/P en todas las estaciones de muestreo,durante las campanas de marzo y agosto de 2005. *: Signifi-cacion estadıstica para p < 0.05; ***: Significacion estadısti-ca para p < 0.001. Pearson correlations between the nitrateconcentration in the sampling stations of the aquifer and theupstream irrigation area for each well, and between the nitrateconcentration and the N/P ratio in all sampling stations, duringMarch and August 2005 campaigns. *: Statistical significancefor p < 0.05; ***: Statistical significance for p < 0.001.

[NO−3 ] [NO−

3 ]

marzo 2005 agosto 2005

Superficie de regadıoaguas arriba de cada pozo(estaciones de acuıfero)

0.61∗ (n = 16) 0.49∗ (n = 16)

N/P (estaciones fluviales yde acuıfero)

0.59∗∗∗ (n = 35) 0.60∗∗∗ (n = 27)

a reforzar esta hipotesis. El cociente N/P se hautilizado en este caso como un indicador del ori-gen agrıcola o urbano del nitrogeno (vease Me-todologıa), prescindiendo de su uso comun co-mo indicador del factor limitante para el desa-rrollo de procesos de eutrofizacion. En la ma-yorıa de los suelos espanoles el fosforo se en-cuentra en forma de fosfatos tricalcicos, insolu-bles en agua, que lentamente pasan a la solu-cion del suelo. La escasa solubilidad del fosfo-ro en este tipo de suelos confiere al cociente N/Pun interesante valor como indicador de la exis-tencia de procesos de lixiviacion de nitrato pro-cedente de los fertilizantes hacia las masas deagua de la zona saturada. En la figura 5 se ob-servan valores bajos en el cociente N/P en losrıos y en los tramos altos del acuıfero, corres-pondientes a zonas no contaminadas o a zonascontaminadas por nitrogeno de procedencia ur-bana. El acuıfero aluvial, sin embargo, presentavalores muy elevados en los tramos medios y/obajos de cada una de las subcuencas (Fig. 5), enconcordancia con las areas destinadas al regadıo.

En terminos generales puede decirse que lacontaminacion por nitrato del acuıfero aluvial esatribuible a la lixiviacion de los fertilizantes, de-bido a las malas practicas de riego y abonado, alsustrato muy permeable y a la escasa profundi-dad del nivel freatico. Debe excluirse la zona delacuıfero en el tramo bajo del Manzanares, conescasa dedicacion a la agricultura, niveles muy

elevados de contaminacion fluvial y cociente N/Pmoderado, datos que apuntan mas bien a un ori-gen urbano del nitrogeno. En el tramo medio-bajo del Jarama y el tramo del Tajo posteriora la incorporacion del Jarama debe considerar-se que el nitrogeno del acuıfero aluvial procedede una fuente difusa de tipo mixto (efecto adi-tivo agrıcola y urbano; Fig. 6). En estos tramosel riego en exceso con agua fluvial contaminadapor nitrogeno de origen urbano, genera retornosde riego en los que se suma la carga de nitrogenolixiviado procedente de los fertilizantes y la delagua de riego (Esteller, 2002).

A pesar de la complejidad del area de estudiodebido a la intensa presion antropica, estos resul-tados proporcionan una base cientıfica preliminarpara considerar la posible declaracion de los alu-viales cuaternarios de la cuenca del Jarama comoZona Vulnerable a la contaminacion por nitratode origen agrıcola en la Comunidad de Madrid,tal como indica la Directiva 91/676/CEE. La Di-rectiva Marco del Agua no solo establece comouno de sus objetivos el conocimiento de los pro-cesos de interaccion entre las aguas superficia-les y subterraneas a escala de cuenca, sino tam-bien el estudio de los flujos de los contaminantesentre las distintas partes del ciclo hidrologico ydel impacto sobre los ecosistemas acuaticos y te-rrestres. Iniciar la aplicacion de “buenas practicasagrıcolas” en las Vegas de la Comunidad de Ma-drid y generalizar la depuracion terciaria en lasareas urbanas de la cuenca, son iniciativas esen-ciales para el cumplimiento la normativa euro-pea respecto a la concentracion de nitrogeno enlas aguas del sistema “rıo-acuıfero aluvial” de lacuenca del Jarama.

CONCLUSIONES

1. Se ha realizado una caracterizacion hidro-quımica del sistema “rıo-acuıfero aluvial”situado en los depositos cuaternarios dela cuenca del Jarama (Comunidad de Ma-drid, Espana), incluyendo el rıo Jarama, susafluentes Henares, Manzanares y Tajuna,una parte del rıo Tajo (a su paso por laComunidad de Madrid), y el acuıfero alu-

208 Arauzo et al.

vial asociado a dicha red fluvial. Se handeterminado los niveles de contaminacionpor nitrogeno, se han identificado las formasquımicas de nitrogeno dominantes en cadaparte del sistema rıo-acuıfero y se ha evalua-do el papel de los usos agrıcolas y urbanoscomo fuente de nitrogeno en el proceso decontaminacion del agua.

2. La naturaleza permeable de los sustratos alu-viales, la escasa profundidad del nivel freati-co y las labores agrıcolas de fertilizacion yriego en exceso, son los factores principalesque determinan el desarrollo de los procesoslixiviacion de nitrato. En el area de estudioconfluyen estos factores de riesgo, lo cualconfiere un alto grado de vulnerabilidad parasus recursos hıdricos.

3. La procedencia mayoritariamente fluvial delas aguas de riego favorecio el trasvase deagua desde la red fluvial al acuıfero alu-vial, alterandose la dinamica hıdrica natu-ral del sistema rıo-acuıfero. Las variacionesinvierno-verano del nivel freatico revelaronuna inversion en la dinamica natural de re-carga del acuıfero, observandose un ascensode nivel en algunas zonas durante el veranorelacionado con el riego.

4. En la red fluvial el nitrogeno total pre-sento una distribucion creciente desde las zo-nas de cabecera hacia los tramos bajos decada subcuenca. En el acuıfero aluvial no seaprecio un patron similar, existiendo una cla-ra concordancia con la distribucion de las zo-nas dedicadas a regadıo.

5. La forma quımica de nitrogeno dominanteen el acuıfero aluvial fue el nitrato (de origenagrıcola). En los tramos medios y/o bajos delos rıos Manzanares, Jarama, Henares y Ta-jo (despues de la incorporacion del Jarama)el amonio fue con frecuencia mas abundanteque el nitrato, y el nitrito presento concen-traciones elevadas (de origen urbano).

6. A partir de los mapas de distribucion delcontenido en nitrato en el acuıfero aluvial seestimaque el 36%delmismopresento concen-traciones superiores a 50 mg/l, el 25% entre25 y 50 mg/l, y el 39% inferiores a 25 mg/l.

7. El cociente N/P, utilizado como indicadordel origen agrıcola o urbano del nitrogeno,presento valores bajos en los rıos y en lostramos altos del acuıfero (zonas no conta-minadas, o contaminadas por nitrogeno deprocedencia urbana). En el acuıfero alu-vial se registraron valores muy elevados enlos tramos medios y/o bajos de cada unade las subcuencas, en concordancia con lasareas destinadas a regadıo. El origen mix-to (agrıcola y urbano) del nitrogeno en elacuıfero, en tramo medio-bajo del Jarama yen la zona del Tajo posterior a la incorpo-racion del Jarama, se explica por el efec-to aditivo de la carga de nitrogeno lixivia-do de los fertilizantes y del nitrogeno delagua de riego (procedente de tramos flu-viales contaminados por efluentes urbanos).

8. Los resultados de este trabajo proporcionanuna base cientıfica para considerar la posi-ble declaracion de los aluviales cuaternariosde la cuenca del Jarama como Zona Vulnera-ble a la contaminacion por nitrato de origenagrıcola en la Comunidad de Madrid, tal co-mo se establece en la Directiva 91/676/CEE.

AGRADECIMIENTOS

Este Proyecto de Investigacion ha sido financia-do por la Consejerıa de Educacion de la Comu-nidad de Madrid, el Fondo Europeo para Desa-rrollo Regional y el Fondo Social Europeo (Ref.:GR/AMB/0745/2004). La Consejerıa de Educa-cion de la Comunidad de Madrid y el Fondo So-cial Europeo han colaborado con la dotacion deuna beca de Formacion de Personal Investigador.

BIBLIOGRAFIA

ABRANTES, N., R. PEREIRA & F. GONCALVES.2006. First step for an ecological risk assessmentto evaluate the impact of diffuse pollution in lakeVela (Portugal). Environ. Monitoring Assessment,117: 411-431.

APHA 1998. Standard methods for the examinationof water and wastewater. 20th Ed. American Public

Contaminacion por nitrogeno en la cuenca del Jarama 209

Health Association, American Water Works Asso-ciation and Water Environment Federation. USA.1134 pp.

ARAUZO,M., J. J. MARTINEZ-BASTIDAy M. VA-LLADOLID. 2006a. Dinamica espacio-temporaldel contenido en nitrato de las aguas superficialesy subterraneas en la cuenca del rıo Oja (La Rioja,Espana): vulnerabilidad del acuıfero aluvial. Lim-netica, 25: 753-762.

ARAUZO, M., M. VALLADOLID y J. J. MAR-TINEZ-BASTIDA. 2006b. Contaminacion por ni-trato en los acuıferos aluviales del Oja y el Tiron.Una aproximacion a escala de cuenca. CCMA-CSIC. Madrid. 59 pp.

ARAUZO, M., J. A. DIEZ, J. J. MARTINEZ-BAS-TIDA, M. VALLADOLID y P. HERNAIZ. 2007.Comparacion de un metodo directo y un metodoindirecto para la estimacion del drenaje y el ba-lance hıdrico en la zona no saturada. En: Estudiosde la zona no saturada, Vol. VIII, ZNS’07. J. V.Giraldez y F. J. Jimenez (eds.): 77-82. Universidadde Cordoba, CSIC, Junta de Andalucıa. Cordoba.

AYERS, R. S. & D. W. WESTCOT. 1985.Waterqua-lity for agriculture. FAO Irrigation and Drainage.Paper, n.o 29. Food and Agriculture Organization oftheUnitedNations.Rome.174pp.

BALL, D. F., A. M. MACDONALD & A. LILLY.2005. Agriculture and diffuse pollution: ground-water nitrate vulnerable zones in Scotland. ScottishJ. Geology, 41: 61-68.

COLLEGE OF AGRICULTURAL SCIENCES,AGRICULTURAL RESEARCH AND COOPE-RATIVE EXTENSION. 2001. Managing phos-phorus for agriculture and the environment. ThePennsylvania State University, USA. 16 pp.

CONFEDERACION HIDROGRAFICA DEL TA-JO. 2005a. Redes de control. Calidad de agua.www.chtajo.es/redes/calidad.htm.

CONFEDERACION HIDROGRAFICA DEL TAJO.2005b. Informe anual de las aguas subterraneas dela Confederacion Hidrografica del Tajo, ano 2005.Confederacion Hidrografica del Tajo, Madrid. Me-moria de 23 pp. + 10 Anexos.

ESTELLER, M. V. 2002. Vulnerabilidad de acuıfe-ros frente al uso de aguas residuales y lodos enagricultura. Revista Latino-Americana de Hidro-geologıa, 2: 103-113.

NEAL, C. & H. P. JARVIE. 2005. Agriculture, com-munity, river eutrophication and the water frame-work directive. Hydrol. Processes, 19: 1895-1901.

FAO/CEPE. 1991. Legislation and Measures for theSolving of Environmental Problems Resulting fromAgricultural Practices (With Particular Referen-ce to Soil, Air and Water), Their Economic Con-sequences and Impact on Agrarian Structuresand Farm Rationalization. Agri/Agrarian Structu-res and Farm Rationalization, Report No 7. UnitedNations publication, Geneve. 53 pp.

FORMAN, D. 2004. Commentary: Nitrites, nitratesand nitrosation as causes of brain cancer in chil-dren: epidemiological challenges. Internat. J. Epi-demiology, 33: 1216-1218.

GERRA, A. y F. MONTURIOL. 1970. Mapa desuelos de la provincia de Guadalajara. Esca-la 1:250.000. Instituto Nacional de Edafologıa yAgrobiologıa Jose Marıa Albareda, CSIC, Madrid.Memoria de 89 pp. + 1 mapa.

HEATHWAITE, A. L., P. F. QUINN & C. J. M. HE-WETT. 2005. Modelling and managing criticalsource areas of diffuse pollution from agriculturalland using flow connectivity simulation. J. Hydrol.,304: 446-461.

HALL, M. D., M. J. SHAFFER, R. M. WASKOM &J. A.DELGADO. 2001. Regional nitrate leachingvariability: What makes a difference in northeas-tern Colorado. J. Am. Water Res. Ass., 37: 130-150.

ITGME-CAM 1988. Atlas geocientıfico del medionatural de la Comunidad de Madrid. Ed: InstitutoTecnologico Geominero de Espana y Comunidadde Madrid. Madrid. Memoria de 83 pp. + 7 mapas.

INSTITUTO TECNOLOGICO GEOMINERO DEESPANA. 1998. Mapa de contenido en nitra-to de las aguas subterraneas en Espana. Escala1:1.500.000. Ministerio de Medio Ambiente. Ma-drid. Memoria de 45 pp. + 1 mapa.

KNAPP, M. F. 2005. Diffuse pollution threats togroundwater: a UK water company perspective.Quarterly J. Engin. Geol. Hydrogeol., 38: 39-51.

LANGE. 1998. Handbook of photometrical operationanalysis. LASA/lP/CADAS/ISIS. Germany. 320 pp.

MAPYA 2005.Mapa de cultivos y aprovechamientosde Espana. Escala 1:50.000. Ministerio de Agri-cultura, Pesca y Alimentacion. www.mapa.es

MARTINEZ-BASTIDA, J. J., M. ARAUZO, J. A.DIEZ, M. VALLADOLID y P. HERNAIZ. 2008.The influence of irrigation on soil water dynamicsand nitrate leaching. En: 15th N Workshop: To-wards a better efficiency in N use: (en prensa). Uni-versidad de Lleida, RUENA. Lleida.

210 Arauzo et al.

MINISTERIO DE MEDIO AMBIENTE. 2001. Ca-racterizacion de las fuentes agrarias de contami-nacion de las aguas por nitratos. Secretarıa deestado de Aguas y Costas, Direccion General deObras Hidraulicas y Calidad de Aguas. Madrid.Memoria de 151 pp. + 11 mapas.

MONTURIOL, F. y L. ALCALA. 1990.Mapa de ca-pacidad potencial de uso agrıcola de la Comuni-dad de Madrid. Escala 1:200.000. Consejerıa deAgricultura y Cooperacion, Comunidad de Madridy Consejo Superior de Investigaciones Cientıficas.Madrid. Memoria de 3 pp. + 1 mapa.

MORALES-SUAREZ, M. M., A. LLOPIS-GONZA-LEZ & M. L. TEJERIZO-PEREZ.] 1995. Impactof nitrates in drinking-water on cancer mortality inValencia, Spain. Europ. J. Epidemiol., 11: 15-21.

MURPHY, J. & J. P. RILEY. 1962. A modified singlesolution method for the determination of phospha-te in natural waters. Anal. Chim. Acta, 27: 31-36.

NAVAS, E., M. GARCIA, R. LLAMAS y J. VRBA.1998. La vulnerabilidad de las aguas subterraneasen el Plan Regional de la Comunidad Autonomade Madrid. En: Jornadas sobre la contaminacionde las aguas subterraneas: un problema pendiente:429-436. AIH-GE. Valencia.

NIXON, S. C., T. J. LACK, D. T. E. HUNT, C. LA-LLANA & A. F. BOSCHET. 2000. ¿Es sosteni-ble el uso del agua? Situacion, perspectivas y pro-blemas. Informe de evaluacion ambiental, AgenciaEuropea de Medio Ambiente. Copenhague. 36 pp.

SAN DIEGO-MCGLONE, M. L., V. S. SMITH & V.F. NICOLAS. 2000. Stoichiometric interpretationsof C:N:P ratios in organic waste material. MarinePoll. Bulletin, 40: 325-330.

SANDOR, J., I. KISS, O. FARKAS & I. EMBER.2001. Association between gastric cancer morta-lity and nitrate content of drinking water: Ecologi-cal study on small area inequalities. Europ. J. Epi-demiol., 17: 443-447.

THORPE, N. & A. SHIRMOHAMMADI A. 2005.Herbicides and nitrates in groundwater of Mary-land and childhood cancers: A geographic infor-mation systems approach. J. Environ. Sci. andHealth Part C-Environmental Carcinogenesis &Ecotoxicology Reviews, 23: 261-278.

VARELA, M. 1994. Nitrate contamination of drin-king water source. International Workshop ofhealth aspects of nitrate and its metabolite (par-ticularly nitrite). Council of Europe. RIVM. Bilt-hoven, 8-10 noviembre 1994.

Limnetica, 27 (2): 211-226 (2008)Limnetica, 27 (2): 211-226 (2008)c© Asociacion Iberica de Limnologıa, Madrid. Spain. ISSN: 0213-8409

Seguimiento de la calidad de un embalse de abastecimiento de aguapotable segun las directrices de la Directiva Marco (embalse delAnarbe. Cuenca Norte)

Henar Fraile 1, Jose Manuel Leonardo 1, Begona G. de Bikuna 1 e Itziar Larumbe 2

1 Anbiotek S.L. Ribera de Axpe 11, B-201 48950 Erandio.2 AGASA Po de Errotaburu 1-6a 20018 Donostia2

∗ Autor responsable de la correspondencia: henar@anbiotek.com2

Recibido: 2/11/06 Aceptado: 11/4/08

ABSTRACT

Monitoring of the quality of a drinking water supply reservoir according to the Water Framework Directive (Anarbereservoir, North basin)

The Anarbe reservoir is located in the Iberic-Macaronesian region and is a water body provisionally identified as heavilymodified, comparable to ‘lake’, that according to the “A” classification system of the WFD is “siliceous, lowlands, big andvery deep”, and according to the “B” system is ¨lowlands, northern, warm monomictic and of acidic waters”. Two annualcycles of studies have been completed (from May of 2004 to January of 2006) looking to the main biological, physico-chemical and hydromorphological parameters. The trophic state is oligotrophic for most of the parameters, except for the totalphosphorus, that classifies it as mesotrophic. It does not present algal blooms and the average oxygen concentration in thehypolimnion in the zone of the dam during the stratification is 4.5 mg O2/l. The percentage of variation of the water volumeof the reservoir respective to the maximum volume varies between 40 and 33% for the years 2004 and 2005, respectively. Itshows a good ecological potential and could be considered as a reference within its group.

Key words: Reservoir, trophic state, ecological potential.

RESUMEN

Seguimiento de la calidad de un embalse de abastecimiento de agua potable segun las directrices de la Directiva Marco(embalse del Anarbe. Cuenca Norte)

El embalse del Anarbe se encuentra en la Region iberico-macaronesica y es una masa de agua identificada provisionalmentecomo muy modificada, asimilable a ‘lago’, que segun el Sistema A de clasificacion de la DMA es del tipo ‘silıceo, tierrasbajas, grande y muy profundo’ y segun el Sistema B; ‘tierras bajas, septentrional, monomıctico calido y de aguas acidas’.Se han completado dos ciclos anuales de estudio (desde mayo de 2004 hasta enero de 2006) con los principales parametrosbiologicos, fisicoquımicos e hidromorfologicos. Su estado trofico es de oligotrofia para la mayorıa de parametros, exceptopara el fosforo total, que lo clasifica como mesotrofico. No presenta blooms algales y la concentracion media de oxıgeno enel hipolimnion de la zona de la presa durante la estratificacion es de 4.5 mg O2/l. El porcentaje de variacion del volumen delembalse con respecto al volumen maximo varıa entre el 40 y el 33%, para los anos 2004 y 2005, respectivamente. Presentaun buen potencial ecologico y podrıa ser considerado como de referencia para su tipologıa.

Palabras clave: Embalse, estado trofico, potencial ecologico.

212 Fraile et al.

INTRODUCCION

Uno de los elementos mas novedosos de la Di-rectiva marco del agua (2000/60/CE) (DOCE,2000), en adelante DMA, es la concepcion deltermino ‘estado ecologico’ que se utiliza para ex-presar la calidad de la estructura y funcionamien-to de los ecosistemas acuaticos asociados a lasaguas superficiales. En el caso de masas de aguaartificiales o muy modificadas se utiliza eltermino ‘potencial ecologico’, que en el mejorde los casos se corresponde con un potencialecologico maximo (MPE), ya que las condicionesnaturales muy buenas u optimas se han perdido.

Para definir el estado o potencial ecologico esnecesario considerar tres criterios basicos (WFDCIS No 4, 2003; WFD CIS No 10, 2003; WFDCIS No 13, 2005):

– Identificar las masas de agua superficiales,encuadrandolas dentro de una de las categorıasdefinidas en la DMA (rıos, lagos, aguas detransicion y aguas costeras) o bien comomasasdeaguaartificialesomuymodificadas,

– Tipificar dichas masas de agua segun el sis-tema A o bien el B de la DMA y establecerunas condiciones de referencia acordes a lascaracterısticas de las mismas,

– Definir el estado o potencial ecologico co-mo una expresion integrada de la diferenciaexistente entre los valores de los indicadoresbiologicos, fisicoquımicos e hidromorfologi-cos evaluados, frente a los valores que, paraesos mismos indicadores, se han establecidoen las condiciones de referencia.

El principal objetivo de la DMA es lograr al-canzar un buen estado o potencial ecologico enlas masas de agua para el horizonte del ano 2015.

La DMA establece que las masas de agua uti-lizadas para la captacion de agua destinada alconsumo humano que proporcionen un prome-dio de mas de 100 m3 diarios deberan ser contro-ladas mediante un seguimiento de sus indicado-res biologicos, hidromorfologicos y fisicoquımi-cos segun se describe en su Anexo V. En este con-texto el organismo gestor del embalse del Anar-

be (AGASA) ha promovido desde la primaverade 2004 hasta la fecha, la realizacion de una seriede controles realizados de manera conjunta por laempresa Anbiotek S.L. y el laboratorio de AGA-SA. En este trabajo se resume el estudio de dosciclos anuales en el embalse del Anarbe desdeun punto de vista limnologico encaminado haciala definicion de su estado trofico, pero incluyen-do ademas como elemento novedoso una apro-ximacion metodologica al calculo del potencialecologico siguiendo las directrices de la DMA.

AREA DE ESTUDIO

El embalse del Anarbe esta situado en el NEde la costa cantabrica (UTM 30TWN914852).Se localiza en la cuenca del rıo Anarbe a160 msnm (Unidad Hidrologica del Urumea,Cuenca Intercomunitaria Norte), entre el Te-rritorio Historico de Gipuzkoa (C.A.P.V.) y laComunidad Foral de Navarra.

La cuenca vertiente del embalse se caracte-riza por su relieve abrupto, con arroyos encaja-dos en profundos valles, sin apenas llanuras deinundacion y con caracter torrencial (Basoinsa,1994). Se localiza en una zona de climatologıatemplada-oceanica, hiperhumeda, con tempera-turas suaves y una amplitud termica pequena (La-rumbe, 1991). En la cuenca predominan los sus-tratos silıceos, con alternancia de pizarras y grau-vacas, que favorecen la impermeabilidad del te-rreno. Hoy en dıa, es una zona de baja actividadantropica, aunque en los siglos pasados la explo-tacion forestal, carboneo y minerıa deterioraronnotablemente los bosques. La red hidrologica dela cuenca es compleja con numerosos arroyos yconducciones o derivaciones tanto para abasteci-miento de la comarca de Donostia como para laproduccion de electricidad en centrales como lasde Berdabio, Okilegi y Anarbe (Basoinsa, 1994).

El embalse se construyo en la decada de 1970y tiene como principal uso el suministro de aguapotable a una poblacion mancomunada de masde 350 000 habitantes.

El area de estudio corresponde al propio embal-se del Anarbe, donde se han muestreado trimes-tralmente tres estaciones a lo largo de su eje lon-

El embalse de Anarbe 213

Figura 1. Area de estudio. Study area.

gitudinal (presaWN9147285302, centro y cola). Sulocalizacion aparece reflejadaen la figura1.

MATERIALES Y METODOS

Se ha visitado el embalse del Anarbe con unafrecuencia trimestral desde mayo de 2004 hastaenero de 2006, realizando perfiles verticales deTo, oxıgeno, pH y conductividad (con una sondade profundidad HIDROLAB) en tres estacionesde su eje longitudinal (Fig. 1). La transparenciase ha estimado mediante la profundidad de vi-sion del disco de Secchi y se han recogido mues-tras de agua a tres profundidades (mediante bo-tella oceanografica Van Dorn) para las determi-naciones analıticas de nutrientes generales y me-tales, pigmentos y microbiologıa (APHA, 1992).En la zona de la presa tambien se han recogidomuestras de fitoplancton en primavera y veranode 2004, y primavera, verano y otono de 2005(preservadas con Lugol y cuantificadas de acuer-do con la tecnica de Utermohl (1958).

En el ano 2005 se incluyo el estudio delzooplancton en la zona de la presa (en primavera,verano y otono), y del bentos profundo del embalseen sus tres estaciones del eje longitudinal (enprimavera y otono). El zooplancton se capturo conredes de 200 y 80μm para diferenciar mesozoo-plancton ymicrozooplancton. Se realizo un barridovertical desde 30 metros de profundidad (muestra

cualitativa) y se fijo con formol al 4% hasta suidentificacion. Para la recogida de las muestrasde bentos profundo se ha seguido la Norma ISO9391:1993 (E) mediante una unica extraccion porpunto con una draga modelo Petite Ponar Grab de0.023m2 de superficie demuestreo.

Los datos morfometricos e hidrologicos delembalse han sido facilitados por AGASA.

Estado trofico

La concentracion de fosforo total en un embal-se es un parametro crucial en la eutrofizacion, yaque suele ser el elemento que limita el crecimien-to de las algas. La biomasa algal es un indica-dor de respuesta trofica y se suelen utilizar dosparametros como estimadores. Uno es la densi-dad celular (no celulas/ml) y el otro es la concen-tracion de clorofila a (μg/l) en la zona fotica, tan-to valores medios como maximos anuales. Otroparametro relacionado con la biomasa algal, esla transparencia medida como la profundidad ala que deja de verse el disco de Secchi (valoresmedios y mınimos anuales).

Existen numerosos ındices para estimar el es-tado trofico de lagos y embalses; los de mas am-plia utilizacion y por lo tanto, los mas contrasta-dos son los siguientes: OCDE (1982); Margalef(1983) y EPA (1976).

Otro metodo para evaluar el estado trofico esel ındice de Carlson (1974) que utiliza como va-riables los valores medios anuales de la profun-didad de vision del disco de Secchi, la concen-tracion superficial media anual de fosforo total yde clorofila a. Este ındice es un valor que puedevariar entre oligotrofia (<35); mesotrofia (35-55);eutrofia (55-70) e hipereutrofia (>70).

Potencial ecologico

El termino potencial ecologico expresa la calidadde la estructura y funcionamiento de los ecosiste-mas acuaticos asociados a una masa de agua ar-tificial o muy modificada. Dado que este tipo demasas de agua presentan en su esencia una mo-dificacion importante, hablamos de ‘maximo po-tencial ecologico’ (MPE) para designar la mejorsituacion de calidad posible. Las condiciones de

214 Fraile et al.

referencia o MPE de estas masas de agua muymodificadas deben de aproximarse a las condi-ciones inalteradas de la categorıa de masa deagua superficial mas estrechamente comparable.En el caso de un embalse esta masa de agua es unlago. Dado que estamos hablando de masas deagua modificadas por un uso humano, las altera-ciones derivadas de este uso no pueden ser obvia-das o eliminadas. Ası, el abastecimiento de aguapotable, regadıo o cualquier otro aprovechamien-to del agua retenida en un embalse, provoca unaoscilacion en el volumen almacenado que, dentrode unos lımites, consideramos algo inevitable.

La variacion del volumen del embalse es elprincipal parametro hidromorfologico que pue-de condicionar la calidad del agua embalsada. Eneste trabajo proponemos que para una zona conun regimen de lluvias importantes y dentro de latipologıa correspondiente al embalse del Anar-be, una variacion del volumen mensual de has-ta el 20% del maximo puede ser acorde con elMPE para las caracterısticas hidromorfologicas.Para este embalse una perdida del 20% del vo-lumen maximo almacenado supone un descensode nivel de 8 metros, respecto a su cota maxi-ma. Si el porcentaje mensual maximo de varia-cion supera el 20%, no se alcanzara el MPE.Sin embargo, analizando los elementos de cali-dad biologicos y fisicoquımicos podremos defi-nir provisionalmente la clasificacion del poten-cial ecologico como bueno, moderado, deficienteo malo (WFD CIS No 13, 2005).

El grado de eutrofizacion que se correspondacon elMPE tambien dependera de su tipologıa.Ası,dentro de la tipologıa donde se enmarca el embalsedel Anarbe, proponemos que un estado oligotroficose corresponda con un buen potencial ecologico;a un sistema mesotrofico le correspondera unpotencial ecologico moderado; a uno eutrofico,un potencial ecologico deficiente y a un mediohipereutrofico, unpotencial ecologicomalo.

Hasta la fecha aun no existe una metodologıaclara y consensuada para el establecimiento delpotencial ecologico. Se han realizado importan-tes trabajos de aproximacion intentando seguirlas directrices de la DMA, tanto por parte dela Confederacion Hidrografica del Ebro (Infraes-tructura & Ecologıa, 2003; 2006) o la Agencia

Catalana del Agua (2003; 2006), como por partedel Gobierno Vasco (2002).

En este documento se realiza una aproxima-cion al calculo del potencial ecologico acordecon los rangos y criterios seguidos en el docu-mento de la Agencia Catalana del Agua (2006)y de la Confederacion Hidrografica del Ebro (In-fraestructura & Ecologıa, 2006), y basados en losindicadores que se han considerado en el estudiodel embalse del Anarbe (Tabla 1). Para esta ti-pologıa de embalse, los valores de los distintosparametros que separan las condiciones de oli-gotrofia de la mesotrofia se han considerado co-mo el lımite entre las condiciones buenas y mo-deradas. Ası, una media anual de densidad al-gal total superior a 2000 cel/ml indicarıa condi-ciones moderadas y mas de 15 000 cel/ml, de-ficientes o malas (EPA, 1976). Para valorar lamedia y el maximo anual de clorofila a (μg/l)de la zona fotica, ası como la media y el mıni-mo anual de la profundidad de vision del discode Secchi (m) y la media anual de la concentra-cion de fosforo total (μgP/l) se utilizan los ran-gos propuestos por la OCDE (1982). La Orga-nizacion Mundial de la Salud (Chorus & Bar-tram, 1999) senala 105 cel/ml como la cantidadde cianobacterias con moderadas probabilidadesde efectos adversos para la salud, por lo queproponemos este valor como el lımite entre lascondiciones deficientes y malas; y 2 000 cel/mlde cianobacterias (valores maximos anuales), co-mo el lımite entre el bueno y moderado (tam-bien utilizado por la CHE, Infraestructura & Eco-logıa, 2006). Los valores propuestos por JRC(1992) se utilizan para valorar las condicionesde oxigenacion del embalse. Ası, una concen-tracion media de oxıgeno hipolimnetico duran-te el periodo de estratificacion de 6 mg O2/l esel lımite entre el bueno y moderado; siendo ellımite entre moderado y deficiente, 4 mg O2/l y2 mg O2/l, entre deficiente y malo.

En primer lugar se realiza una valoracion decada parametro por separado, puntuando de ma-nera sencilla la categorıa optima con un 5, labuena con un 4, moderada con un 3, deficientecon un 2 y mala con un 1. La media aritmeti-ca de los distintos parametros de un mismo ele-mento nos dara la valoracion de cada elemen-

El embalse de Anarbe 215

Tabla 1. Indicadores y valoracion de los parametros considerados para estimar el potencial ecologico. Propuesta por Anbiotek S.L.hasta que se aprueben unos rangos definitivos. Indicators and appraisal of the parameters considered to estimate the ecologicalpotential. Proposal of Anbiotek until some final ranks are approved.

Valoracion de los parametros

Optimo Bueno Moderado Deficiente MaloPuntuacion 5 4 3 2 1

Indicador Elemento Parametro

Biologico Ref.

Fitoplancton

Densidad algal total,media anual (cel/ml)

EPA, 1976 < 1 000 1 000-2 000 2 000-15 000 > 15 000 > 15000

Chl a, media anual fotica (μg/l) OCDE, 1982 < 1 1 − 2.5 2.5 − 8 8 − 25 > 25Chl a maxima anual (μg/l) OCDE, 1982 < 2.5 2.5 − 8 8 − 25 25 − 75 > 75Cianobacterias, maximo

anual (cel/ml)Chorus & Bartram

(1999)< 500 500 − 2 · 103 2 · 103 − 2 · 104 2 · 104 − 105 > 105

Fisico-quımico

Transparencia

Secchi, media anual (m) OCDE, 1982 > 12 12 − 6 6 − 3 3 − 1.5 < 1.5Secchi, mınimo anual (m) OCDE, 1982 > 6 6 − 3 3 − 1.5 1.5 − 0.7 < 0.7

Oxigenacion

Oxıgeno hipolimnetico medio duranteestratificacion (mg/l)

JRC, 1992 > 8 8 − 6 6 − 4 4 − 2 < 2

Nutrientes

Fosforo total, media anual (μg P/l) OCDE, 1982 < 4 4 − 10 10 − 35 35 − 100 > 100

APROXIMACION AL POTENCIAL

ECOLOGICO

La valoracion del peor de los indicadores(biologico o fisicoquımico)

to considerado. Ası hay elementos (E) compues-tos por un solo parametro (P) como Enutrientes(= P fosforo total anual) oEoxigenacion (= P oxı-geno hipolimnetico) y otros, compuestos por doso mas, como Efitoplancton (PChl media, PChlmax, PDensidad algal, PCianofıceas cel/ml), oEtransparencia (Psecchi medio, Psecchi min).

Siguiendo la metodologıa descrita por laAgencia Catalana del Agua (2003) para obtener elvalor de los indicadores fisicoquımicos se calculala media de sus elementos; en cambio el valorde los indicadores biologicos se obtendra consi-derando la valoracion del peor de sus elementos.Finalmente el valor provisional del potencialecologico se obtendra del indicador (biologico ofisicoquımico) con una peor valoracion ecologica.

La Agencia Catalana del Agua cuenta con ma-sas de agua de referencia para cada tipo de em-balse, por lo que pueden estimar el grado de des-

viacion del potencial ecologico maximo. Sin em-bargo, en la Cuenca Norte no hay establecidascondiciones de referencia para embalses. Por ellola clasificacion provisional del potencial ecologi-co para los embalses estara representada por elpeor de los valores de los indicadores (biologi-cos o fisicoquımicos).

RESULTADOS Y DISCUSION

Caracterizacion y tipificacion del embalse delAnarbe

El embalse del Anarbe se localiza en la Regioniberico-macaronesica y ha sido designado provi-sionalmente como ‘masa de agua muy modifica-da’ asimilable a ‘lago’. Segun el Sistema A declasificacion de la DMA se corresponde a un tipo‘silıceo, tierras bajas, grande y muy profundo’;

216 Fraile et al.

A

B

Superficie (m2)

Capacidad(hm3)

Cota(m.s.n.m.)

Cota(m.s.n.m.)

Figura 2. A.- Curva hipsografica (cota-superficie). B.- Rela-cion cota-capacidad en el embalse del Anarbe. Fuente: AGA-SA. A.- Hipsographic curve(level-surface). B.- level-capacityrelation for the Anarbe reservoir. Source: AGASA.

y en base al Sistema B corresponderıa a un tipo‘tierras bajas, septentrional, monomıctico calidoy de aguas acidas’.

Hasta la fecha no hay establecida nin-guna masa de agua de referencia con estascaracterısticas en la Cuenca Norte.

La masa de agua natural mas similar al embalsedel Anarbe en la Penınsula es la lago de Sanabria(Zamora), clasificado como ‘silıceo, alto, grandey muy profundo’ segun el Sistema A. En cuantoal Sistema B, el lago de Sanabria es ‘de mediamontana, monomıctico calido y de aguas acidas’.La diferencia es la altitud a la que se encuentranambas masas de agua: el embalse se localiza amenosde200mdealtitudy el lago amas de800m.

Indicadores hidromorfologicos

La morfometrıa de un embalse y la importan-cia relativa de su cuenca vertiente determinanen gran medida sus caracterısticas fisicoquımicasy biologicas (Hutchinson, 1957; Wetzel, 1981;Hakanson, 1981; Catalan, 1987).

Tabla 2. Parametros morfometricos del embalse de Anarbe.Morphometric parameters of the Anarbe reservoir.

PARAMETROS MORFOMETRICOS

Superficie de la cuenca total (ha) 6 900Superficie del embalse (ha) 201% Cuenca ocupado por embalse Anarbe 2.91%Longitud total del embalse (m) 7 400Anchura maxima (m) 674Profundidad maxima (m) 65Profundidad media (m) 21.74Perımetro en nivel maximo (m) (pista perimetral) 20 000Cota a maximo nivel embalsado (msnm) 160Cota(s) de la toma(s) de agua principal(es) (msnm) 122 y 110

Capacidad maxima (hm3) 43.7

Capacidad util (hm3) 37.3Profundidad relativa (Zr)% 4.06Zm:Zmax 0.33Desarrollo de volumen (Dv) 1.00Desarrollo del litoral (DL) 3.98Ac/V m−1 1.58Ac/A 34.33

La variacion del area y volumen, con relacion ala profundidad se puede observar en la figura 2.Los parametros e ındices del embalse del Anarbe(Tabla 2) senalan que la cubeta tiene una formaasimilable a un cono invertido con una gran pen-diente en sus margenes (relacion Zm:Zmax iguala 0.33 y Dv de 1) y una profundidad relativa quefavorece la estabilidad durante la estratificaciontermica (Zr de 4.06%). Su forma es sinuosa congran superficie de contacto entre el embalse y elmedio terrestre circundante (DL igual a 3.98). Larelacion entre la cuenca vertiente y el volumeno el area del embalse (Ac/V y Ac/A) es bastanteelevada e indica que la masa de agua es sensi-ble a presentar problemas de eutrofia, debido a lagran repercusion que cualquier modificacion enlos usos del suelo de la cuenca pueden producirsobre el estado trofico del embalse.

El volumen maximo del embalse se alcanzaen invierno y el mınimo en el mes de noviembre(Fig. 3). Las aportaciones maximas al embalsepor el rıo Anarbe y sus afluentes se producen en-tre diciembre y febrero, coincidiendo con la epo-ca de lluvias mas importantes; ademas hay otropico importante con el deshielo de primavera enabril. El caudal de salida mayor se produce en fe-brero y en abril; por otra parte, el uso para abas-

El embalse de Anarbe 217

Figura 3. A.- Variacion del volumen embalsado (hm3) en elembalse del Anarbe (enero 2004-diciembre 2005). B.- Caudalde entrada (Q, l/s). C.- Caudal de salida mensual (Q, l/s). A.-Variation of the dammed volume (hm3) in the Anarbe reservoir(January 2004-December 2005). B.- Entry volume (Q, l/s). C.-Monthly exit volume (Q, l/s).

Tabla 3. Principales parametros hidraulicos del embalse delAnarbe.Main hydraulic parameters of the Anarbe reservoir.

PARAMETROS HIDRAULICOS

Ano 2004 Ano 2005

Volumen medio (hm3) 30.39 31.12Superficie media (ha) 142.35 143.53Profundidad media (m) 21.34 21.68

Entradas (hm3/ano) 69.31 66.79

Salidas (hm3/ano) 68.33 60.09Tiempo de retencion hidraulico Tw (anos) 0.44 0.46

Tasa de renovacion D (anos−1) 2.3 2.14

tecimiento de este embalse cuenta con una con-cesion de 1500 l/s que es la salida, mas o menosconstante, en el resto de los meses.

El tiempo de retencion hidraulico determinael tiempo de que dispone un determinado procesopara llevarse a cabo en el embalse (por ejemplo,el crecimiento del plancton). Los tiempos de resi-dencia hidraulica de los embalses espanoles osci-lan entre 0.5 y 50 anos (Margalef, 1983) y el em-balse del Anarbe se encuentra entre los que tienenun menor tiempo de residencia (Tw de 0.4 anos),lo que implica que se renueva algo mas de 2 vecesen un ano (Tabla 3). El porcentaje de variaciondel volumen con respecto al volumen maximo esmayor en el mes de noviembre, siendo del 40%en 2004 y del 33% en 2005 (Tabla 4).

Tabla 4. Porcentaje de variacion del volumen con respecto alvolumen maximo en el embalse del Anarbe. Volume variationpercentage respective to the maximum volume in the Anarbereservoir.

Meses % Variacionvolumen 2004

% Variacionvolumen 2005

Enero 1.11 9.97Febrero 2.65 0.00Marzo 0.50 0.92Abril 1.33 1.56Mayo 0.00 1.06Junio 5.31 4.40Julio 13.60 15.29Agosto 22.25 22.89Septiembre 31.01 28.35Octubre 38.89 33.25Noviembre 40.80 33.36Diciembre 34.66 8.99

Variacionmaxima anual 40% 33.3%

218 Fraile et al.

Figura 4. Perfiles verticales de temperatura (A y B) y% saturacion de oxıgeno (C y D) en la presa del embalse del Anarbe. Verticaltemperature profiles (A and B) and percentage of oxygen saturation (C and D) in the dam of the Anarbe reservoir.

Indicadores fisicoquımicos

El embalse del Anarbe se clasifica como mo-nomıctico calido y presenta en la zona mas pro-funda de la presa una estratificacion termica quecomienza en primavera y finaliza en diciembre(Fig. 4). La termoclina estacional se localiza enverano entre los 25 y 30 m de profundidad, exis-tiendo tambien una termoclina secundaria en elepilimnion entre los 10 y 15 m, que se forma de-bido al intenso calentamiento superficial resulta-do del ciclo nictemeral en verano (Fig. 4).

El embalse del Anarbe no presenta problemasgraves de anoxia en el periodo 2004/2006. Sinembargo, en la presa y en el centro del embal-se en verano se observa un mınimo relativo deoxıgeno asociado a la termoclina (tanto estacio-nal como secundaria); y en otono e incluso eninvierno, estos mınimos de oxıgeno se mantie-

nen en la presa, asociados a la cubeta profun-da en la que la mezcla aun no se ha completado(Fig. 4). Hay que senalar que los mınimos relati-vos de oxıgeno son mas acusados en el periodo2005/2006 que en el periodo anual 2004/2005.La termoclina actua como una ‘barrera’ sobre laque tienden a acumularse microorganismos, ma-teria organica e inorganica disuelta y sales mine-rales que contribuyen al aumento relativo de laconductividad (Fig. 5) y a la disminucion relativade oxıgeno. La concentracion media de oxıgenohipolimnetico durante el periodo de estratifi-cacion es de 4.5 mg O2/l.

El embalse del Anarbe tiene su maxima trans-parencia en la zona de la presa, donde la capafotica media alcanza entre los 17 y los 19 mde profundidad. Los valores medios y mınimosanuales de la profundidad de vision del disco deSecchi en la zona de la presa indican condiciones

El embalse de Anarbe 219

Figura 5. Perfiles verticales de pH (A y B) y conductividad (μS/cm) (C y D) en la presa del embalse del Anarbe. Vertical pH (Aand B) and conductivity (μS/cm) (C y D) profiles in the dam of the Anarbe reservoir.

de oligotrofia. La transparencia disminuye pro-gresivamente hacia la cola, como corresponde alcaracter mas fluvial de esta zona.

Las aguas del embalse son acidas, con los va-lores de pH mas bajos en la presa, sobre todo enverano y en el hipolimnion (Fig. 5). Tienen unamineralizacion debil y los valores medios anua-les indican una composicion ionica bicarbonata-da clorurada calcico sodica (Tabla 5).

De las forma del nitrogeno, el nitrato es lamas abundante, pero en concentraciones muy ba-jas (entre 1.5 y 3.5 mgNO3/l). La peor calidad seda en la presa (en el periodo 2005/2006) cuan-do se obtiene una calidad tipo A2 por la concen-tracion de amonio y una calidad tipo ciprinıcolapara la vida piscıcola segun los niveles de nitrito(R.D. 927/1988, de 29 de julio).

Las concentraciones de fosforo en el embalseson inferiores en el ciclo anual 2005/2006, que en

el anterior (Tabla 5); sin embargo en ambos pe-riodos se encuentran dentro del rango de la me-sotrofia. El aluminio y el sılice presentan concen-traciones muy homogeneas en el embalse.

Los metales hierro y manganeso presentan lasconcentraciones puntuales mas elevadas cuandoel embalse esta en su cota mas baja, esto es enverano y otono. El hierro no presenta problemaspara la calidad de las aguas; si bien, el mangane-so tiene una concentracion media en la presa queimplica un nivel de tratamiento de tipo A3 pa-ra produccion de agua potable (R.D. 927/1988,de 29 de julio). Los niveles de manganeso en elsustrato pizarroso y granıtico de la cuenca delAnarbe son elevados (Basoinsa, 1994) y su mo-vilidad hacia la fase lıquida, dado el pH acido delas aguas, es tambien alto.

El contenido de materia organica, determina-do a partir de la DBO5 y DQO es bajo y no pre-

220 Fraile et al.

Tabla 5. Valores medios de las principales variables fisicoquımicas y de la clorofila a para cada estacion de muestreo del embalse(presa, centro y cola) desde mayo de 2004 a febrero de 2005 (2004/2005) y desde junio de 2005 a enero de 2006 (2005/2006).Entre parentesis se senala la desviacion estandar (DS) y el mınimo y maximo (min-max) de cada variable. Average values of the mainphysicochemical variables and of the chlorophyll-a for each sampling station of the reservoir (dam, centre and tail) from May of 2004to February of 2005 (2004/2005) and from June of 2005 to January of 2006 (2005/2006). Standard deviation (SD) and minimum andmaximum (min-max) for each variable are shown between brackets.

PRESA CENTRO COLA

Periodo 2004-05 2005-06 2004-05 2005-06 2004-05 2005-06

Prof. Secchi (m) 7.40 (2.6)(4.9-10.5)

8.50 (2.6)(5.80-11.00)

6.28 (1.7)(4.60-8.50)

6.90 (1.1)(5.5-8.1)

5.70 (2.3)(4.40-9.10)

5.78 (1.4)(4.50-7.50)

Chl a (μg/l) 0.76 (0.7)(0.06-2.36)

0.27 (0.2)(0.01-0.55)

1.10 (1.4)(0.15-4.46)

0.84 (0.9)(0.04-2.04)

2.25 (2.4)(0.41-8.7)

0.49 (0.5)(0-1.25)

Ca++ (meq/l) 0.41 (0.1)(0.31-0.64)

0.36 (0.08)(0.29-0.52)

0.37 (0.05)(0.28-0.44)

0.36 (0.05)(0.30-0.42)

0.39 (0.07)(0.30-0.50)

0.37 (0.05)(0.31-0.45)

Mg++ (meq/l) 0.17 (0.13)(0.05-0.42)

0.09 (0.01)(0.06-0.10)

0.18 (0.2)(0.01-0.51)

0.09 (0.01)(0.07-0.11)

0.13 (0.08)(0.08-0.37)

0.09 (0.01)(0.07-0.11)

Na+ (meq/l) 0.29 (0.07)(0.24-0.48)

0.23 (0.02)(0.21-0.26)

0.29 (0.05)(0.23-0.35)

0.23 (0.02)(0.21-0.26)

0.28 (0.05)(0.22-0.35)

0.23 (0.02)(0.21-0.26)

K+ (meq/l) 0.03 (0.01)(0.03-0.04)

0.02 (0.00)(0.01-0.03)

0.03 (0.01)(0.03-0.05)

0.02 (0.00)(0.01-0.03)

0.03 (0.00)(0.02-0.03)

0.02 (0.00)(0.01-0.03)

HCO−3 (meq/l) 0.39 (0.11)

(0.27-0.59)0.42 (0.09)(0.33-0.64)

0.38 (0.08)(0.27-0.50)

0.41 (0.05)(0.32-0.48)

0.40 (0.08)(0.30-0.51)

0.43 (0.04)(0.36-0.49)

SO−4 (meq/l) 0.12 (0.04)

(0.07-0.17)0.09 (0.05)(0.05-0.21)

0.13 (0.05)(0.09-0.19)

0.10 (0.06)(0.06-0.23)

0.11 (0.03)(0.07-0.16)

0.10 (0.05)(0.06-0.21)

Cl− (meq/l) 0.14 (0.05)(0.08-0.20)

0.18 (0.02)(0.17-0.22)

0.14 (0.04)(0.08-0.18)

0.18 (0.02)(0.14-0.21)

0.13 (0.03)(0.10-0.16)

0.19 (0.02)(0.17-0.21)

NH+4 (mg/l) 0.03 (0.03)

(0-0.09)0.11 (0.16)(0-0.38)

0.04 (0.04)(0-0.09)

0.01 (0.01)(0-0.02)

0.04 (0.04)(0-0.12)

0.05 (0.07)(0-0.23)

NO−2 (mg/l) 0.01 (0.01)

(0-0.03)0.03 (0.05)(0-0.19)

0.01 (0.01)(0-0.02)

0.02 (0.01)(0.01-0.03)

0.01 (0.01)(0-0.03)

0.02 (0.00)(0.01-0.02)

NO−3 (mg/l) 2.46 (0.59)

(1.08-3.04)2.47 (0.68)(1.30-3.36)

2.59 (0.49)(1.89-3.24)

2.47 (0.54)(1.61-3.10)

2.28 (0.43)(1.6-2.85)

2.31 (0.62)(1.47-3.20)

NTK (mg/l) 0.40 (0.5)(0-1.12)

0.17 (0.33)(0-1.00)

0.28 (0.35)(0-0.96)

0.25 (0.58)(0-2.00)

0.86 (0.96)(0-3.00)

1.38 (3.21)(0-10.00)

P-PO3−4 (μgP/l) 3.83 (4.53)

(0-10.00)1.58 (2.62)(0-8.70)

3.48 (4.29)(0-10.00)

0.96 (1.07)(0-2.60)

3.81 (4.41)(0-10.00)

1.13 (1.31)(0-3.90)

P Total (μgP/l) 27.01 (37.5)(0.1-127.14)

11.82 (8.36)(0-31.00)

26.90 (29.4)(0.1-101.00)

27.92 (36.2)(6.5-132.40)

37.30 (39.5)(9.6-132.85)

22.35 (25.5)(3.10-92.70)

SiO2 (mg/l) 12.46 (1.5)(10.37-15.6)

11.23 (2.4)(7.54-14.52)

11.61 (2.34)(8.06-16.40)

11.06 (4.01)(3.71-17.61)

11.83 (1.7)(8.16-14.90)

11.29 (3.0)(6.47-15.29)

Aluminio (mg/l) 0.01 (0.01)(0-0.04)

0.02 (0.01)(0.01-0.05)

0.03 (0.04)(0-0.12)

0.03 (0.01)(0.01-0.06)

0.02 (0.02)(0-0.06)

0.03 (0.01)(0.01-0.05)

Fe total (μg/l) 96.58 (143.4)(13-531)

51.75 (39.9)(17-164)

115.83 (83.8)(52-362)

167.0 (165.5)(39-554)

127.9 (114.1)(35-433)

111.5 (75.3)(35-286)

Fe disuelto (μg/l) 28.33 (19.4)(5-71)

34.08 (33.7)(8-136)

84.11 (89.7)(25-307)

73.50 (63.5)(30-249)

90.44 (95.8)(28-329)

88.00 (66.1)(29-232)

Mn total (μg/l) 109.9 (281.4)(7-1000)

153.1 (403.0)(6-1420)

67.92 (87.2)(7-258)

52.67 (102.3)(6-374)

45.42 (48.1)(11-173)

79.75 (152.1)(6-410)

Mn disuelto (μg/l) 96.21 (253.5)(4-897.50)

96.17 (246.1)(0.50-860)

58.33 (80.5)(7-233)

29.08 (54.7)(0-197)

33.50 (37.1)(7-117)

45.58 (115.3)(0.50-405)

El embalse de Anarbe 221

senta problemas ya que se encuentra por debajode los lımites mas exigentes de calidad recomen-dados para produccion de agua potable.

Indicadores biologicos

En los muestreos de verano, tanto de 2004 como de2005, se observo la presencia bastante abundante deun pequeno cnidario, probablementeCraspedacus-

ta sowerbii, que corresponde a medusas de aguadulce, cuya presencia ya se ha observado en otrosembalses de la penınsula (Perez-Bote et al., 2005;Alonso, comunicacion personal).

Los valores de clorofila a son muy bajos, acor-des con la baja densidad fitoplanctonica encon-trada (inferior a 2000 cel/ml) e indicativos decondiciones de oligotrofia (Tabla 5).

En el ano 2004 se estudio el fitoplancton en

Tabla 6. Densidad (cel/ml) del fitoplancton en el epilimnion de la presa del Anarbe. Phytoplankton density (cell/ml) in the epilim-nion of the dam of the Anarbe reservoir.

PRESA DEL ANARBETAXONES 18/5/04 8/11/04 23/6/05 14/9/05 8/11/05CianofıceasChroococcus sp. 15.15Coelosphaerium pusillum 60.60Cyanogranis sp. 1.00 30.30Oscillatoria sp. 1.00 1.00Synechococcus sp. 196.95ClorofıceasBotryococcus braunii 19 160.90Chlamydomonas sp. 3 30.30Chlorella sp. 2 30.30 1.00Chlorolobion lunulatum 15.15Clorococal indet. 30.30 1.00Closterium lunula 1 15.15Coenoccocus sp. 30.30Coenochloris piscinalis 1.00Dictyosphaerium pulchellum 121.20Eutetramorus fottii 173 145.75Hyaloraphidium curvatum 1.00 15.15Oocystis lacustris 60.60 1.00 7 15.15Pandorina morum 8 30.30Planktosphaeria gelatinosa 45.45Radiocystis sp. 45.45Scenedesmus armatus 30.30Scenedesmus disciformis 16Scenedesmus dispar 2 15.15Scenedesmus ellipticus 30.30Tetrastrum komarekii 2.00 15.15Crisofitos

BacilariofıceasCyclotella sp. 2.00 60.60Cyclotella cf. stelligera 85.75 15.15Cyclotella meneghiniana 3 45.45Fragilaria sp. 15.15Navicula sp. 1.00 1 15.15Nitzschia amphibia 15.15Tabellaria flocculosa 2

XantofıceasCharaciopsis saccata 15.15

DinofıceasPeridinium sp. 45.45 1.00 84 166.65 1.00EuglenofıceasTrachelomonas volvocina 30.30 1.00

Densidad (cel/ml) 185.80 209.80 325 836.90 488.80

222 Fraile et al.

primavera y otono (Tabla 6). En la presa, el gru-po mas abundante corresponde a las clorofıceas(Oocystis lacustris) en primavera; y a las dia-tomeas (Cyclotella sp.) en otono, con presenciapuntual en esta campana de cianofıceas.

En el ano 2005 se estudio el fitoplancton enprimavera, verano y otono (Tabla 6). Los clorofi-tos y dinofitos son los principales representantesdel fitoplancton en primavera; en verano, el grupoclaramente dominante son las algas clorofıceas;mientras que en otono se produce un incrementonotable en la presencia de cianofıceas de pequenotamano (Synechococcus sp. y Coelosphaeriumpusillum) que pasa a ser el grupo mas abundante.

El zooplancton se ha estudiado en la zona dela presa del embalse en primavera y otono de2005. En general, en cada campana de muestreose encuentra una comunidad bien representadaformada principalmente por una o dos especiesde Daphnia o Ceriodaphnia; una o dosespecies de copepodos en menor abundancia;y rotıferos, con mayor diversidad pero menorabundancia. Estan presentes especies de aguaspoco mineralizadas, como los cladoceros Ce-riodaphnia quadrangula, Daphnia longispina,

D. pulicaria, D. galeata, Bosmina longirostriso D. cucullata, esta ultima poco frecuente;ası como la especie Ploesoma hudsoni, rotıferoencontrado tambien en otros embalses del NWde la Penınsula. En los embalses, la mayor tasade renovacion frente a los lagos favorece a lasespecies de rapida multiplicacion (rotıferos ycladoceros) frente a los copepodos.

El bentos profundo del embalse del Anarbe seha estudiado en primavera y otono de 2005 en lastres estaciones de su eje longitudinal (Tabla 7). Losoligoquetos son el grupomas representado, seguidode los quironomidos. La relacion entre ambos gru-pos indica la clara dominancia de los oligoquetos;si bien su importancia relativa va disminuyendohacia la cola. La presencia de esta fauna bentonicarefleja el buen estado del embalse, que mantieneorganismos propios de ambientes profundos en losque la diversidad es baja y las condiciones deoxıgeno restrictivas. La comunidad es propia desistemas oligotroficos con una dependencia mayorde los aportes de origen aloctono que de la pro-duccion propia del embalse. Este aporte nutricionales favorecido por la fluctuacion del nivel del aguaacompanada de procesos de sedimentacion rapida.

Tabla 7. Densidad (ind/m2) de macroinvertrbrados del bentos profundo en el embalse del Anarbe. Relacion Q/O: quironomi-dos/oligoquetos. Density (ind/m2) of the deep benthic macroinvertebrates in the Anarbe reservoir. Q/O relation: Quironomi-dae/Oligochaeta.

PRESA CENTRO COLATAXONES 23/6/05 8/11/05 23/6/05 8/11/05 23/6/05 8/11/05

NematodaMermithidae 87AnnelidaOligochaeta 1087 1391 1783 2174 1174 1261Helobdella sp. 43 43HydrachnidaeHydrachna sp. 43 87CrustaceaOstracoda 43 522Insecta (Chironomidae)Tanypodinae 130 391 87 130 174Tanytarsini 87 43 43Chironomus plumosus 174 956MolluscaPhysidae 43Pisidium sp. 130 43 217 43

TOTAL (ind/m2) 1260 1434 2434 2956 1781 2564

Relacion Q/O 0.12 0 0.27 0.04 0.29 0.93

El embalse de Anarbe 223

Tabla 8. Valoracion del estado trofico en el embalse del Anarbe. Desde mayo de 2004 a febrero de 2005 (2004/2005) y desde juniode 2005 a enero de 2006 (2005/2006). Appraisal of the trophic state of the Anarbe reservoir. From May of 2004 to February of 2005(2004/2005) and from June of 2005 to January of 2006 (2005/2006).

2004/05 2005/06

OCDE (1982)PT (μg/l) 27.01 Mesotrofico 11.82 Mesotroficomedia fotica Chl a (μg/l) 0.90 Ultraoligotrofico 0.54 UltraoligotroficoChl a max (μg/l) 2.40 Ultraoligotrofico 0.56 UltraoligotroficoSecchi (m) 7.40 Oligotrofico 8.50 OligotroficoSecchi min (m) 4.90 Oligotrofico 5.80 Oligotrofico

EPA (1976)PT (μg/l) 27.01 Eutrofico 11.82 Mesotroficono cel algales/ml 90.90 Oligotrofico 275 OligotroficoChl a max (μg/l) 2.40 Oligotrofico 0.56 Oligotrofico

Margalef (1983)PT (μg/l) 27.01 Eutrofia avanzada 11.82 Eutrofia moderadano cel algales/ml 90.90 Eutrofia moderada 275 Eutrofia moderadamedia fotica Chl a (μg/l) 0.90 Eutrofia moderada 0.54 Eutrofia moderadaSecchi (m) 7.40 Eutrofia moderada 8.50 Eutrofia moderada

Carlson (1974)TSI secchi 31.20 Oligotrofico 29.16 OligotroficoTSI Chl a fotica 29.80 Oligotrofico 24.55 OligotroficoTSI PT fotico 55.60 Eutrofico 39.75 MesotroficoTSI medio 38.87 Mesotrofico 31.15 Oligotrofico

Ası, la comunidad trofica aparece dominada por losrecolectores a lo largo de todo el eje longitudinaldel embalse, si bien los detritıvoros, oligoquetosfundamentalmente, pierden peso respecto a losfiltradores a medida que desciende la profundidadhacia la cola. Entre estos ultimos destacan losquironomidos Tanytarsini y Chironomus plumosusy el molusco bivalvo Pisidium sp. Cabe mencionartambien la presencia significativa de depredadorescomoesel casodelquironomidoTanypodinae.

La microbiologıa de las aguas del embalseno presenta problemas, como corresponde a unacuenca poco humanizada. La densidad de coli-formes totales es indicativa en todo el eje del em-balse de una calidad de agua tipo A2 (entre 50-5000 ufc/100 ml) para produccion de agua pota-ble (R.D. 927/1988, de 29 de julio).

Estado trofico

La estima del estado trofico del embalse se harealizado con los valores medios anuales de lazona de la presa, ya que son las condiciones massimilares a un lago y el estado trofico de esta zo-

na es el que va a afectar directamente a la calidaddel agua de abastecimiento.

El estado trofico del embalse de Anarbe es engeneral bueno, indicando niveles de oligotrofiapara la mayor parte de los ındices exceptopara las concentraciones de fosforo total, quesegun el ındice considerado se incluye en lacategorıa de mesotrofia o eutrofia (Tabla 8). Laconcentracion media de fosforo total es menoren el ciclo anual 2005/2006, que en el periodoanterior, aun ası el nivel medio de fosforo totalindica una ligera mesotrofia.

Estos resultados son coincidentes con los pre-sentados en el informe de Basoinsa (1994) quesenalan la misma problematica trofica debida alfosforo en los anos 1991, 1992 y 1993.

Potencial ecologico

El mayor porcentaje de variacion del volumenrespecto al volumen maximo supera el lımitepropuesto para el MPE en los dos anosestudiados. Por lo tanto, las caracterısticashidromorfologicas, en este caso la variacion del

224 Fraile et al.

Tabla 9. Valoracion de los diversos indicadores analizados en el embalse de Anarbe para la estima del Potencial Ecologico;A: (enero 2004-febrero 2005) y B: (junio 2005-enero 2006). Se consideran los parametros medios en la zona de la presa del em-balse. Appraisal of the indicators analyzed in the Anarbe reservoir to estimate the ecological potential; A: (January 2004- February2005) and B: (June 2005- January 2006). The average parameters in the zone of the dam of the reservoir are considered.

Anarbe Valoracion

ParametroValor

Parametro Elemento IndicadorPotencial Ecologico

provisionalA B

Densidad algal total, media anual (cel/ml) 90.90 275 5

5 5

3.3

Chl a. media anual fotica (μg/l) 0.90 0.54 5Chl a maxima anual (μg/l) 2.40 0.56 5Cianobacterias, maximo anual (cel/ml) 0.00 303 5

Secchi. media anual (m) 7.40 8.50 44

3.3Secchi. mınimo anual (m) 4.90 5.80 4Oxıgeno hipolimneticomedio durante estratificacion (mg/l) 4.52 4.58 3 3Fosforo total, media anual (μgP/l) 27.01 11.82 3 3

volumen por encima de un 20% del maximo,supone una alteracion hidromorfologica queimpide alcanzar el MPE. La oscilacion del volu-men es inherente al uso para el que fue construidoel embalse del Anarbe y no impide que pueda te-ner un buen potencial ecologico, sino unicamenteque este no sera el maximo posible.

La asignacion provisional del potencialecologico se realiza teniendo en cuenta la meto-dologıa descrita anteriormente y se resume en laTabla 9. La aproximacion al calculo del poten-cial ecologico del embalse del Anarbe lo senalacomo Bueno, aunque con una valoracion proxi-ma a Moderado, debido a que la concentra-cion media anual de fosforo total se encuentradentro del rango de la mesotrofia y a que lacantidad de oxıgeno hipolimnetico se encuentratambien dentro de la clase moderada. La pun-tuacion final en esta primera aproximacion alcalculo del potencial ecologico del embalse delAnarbe es de 3.3, que podrıa ser consideradacomo de referencia para su tipo.

AGRADECIMIENTOS

Agradecemos el trabajo realizado por EncarnaZafra y la Dra. Marina Aboal de la Universidadde Murcia por la identificacion de las especies fi-toplanctonicas; y a Santiago Robles de la empre-sa CIMERA S.A. y al Dr. Miguel Alonso por la

identificacion del zooplancton. Gracias tambienal personal del laboratorio de AGASA por faci-litar todos los medios necesarios para la realiza-cion de los muestreos de campo, ası como al per-sonal de la presa del Anarbe por el apoyo pres-tado. Finalmente, los trabajos de campo no hu-bieran sido posibles sin la colaboracion de JesusAngel Arrate ni de Eva Lopez (Anbiotek S.L.).

BIBLIOGRAFIA

AGENCIA CATALANA DE L’AIGUA. 2003. Ca-racteritzacio i propostes d’estudi dels embassa-ments catalans segons la Directiva 2000/60/CE delParlament Europeu. 212 pp.

AGENCIA CATALANADE L’AIGUA. 2006. Proto-col d’avaluacio del potencial ecologic dels embas-saments. ECOEM. 47 pp.

APHA-AWWA-WPCF. 1992.Metodos normalizadospara el analisis de aguas potables y residuales.17th Edicion. Ed. Dıas de Santos, S.A. Madrid.1 550 pp.

BASOINSA, S. L. 1994. Estudio de propuesta deusos de la cuenca del embalse de Anarbe para sal-vaguardar las caracterısticas de calidad del aguay vida util del embalse. Fase I. 2a. Hidrologıa.Mancomunidad de Aguas del Anarbe-Anarbekouren mankomunitatea. 29 pp.

CARLSON, R. E. 1974. A trophic state index for la-kes. Limnol. Oceanogr., 23: 361-369.

CATALAN, J. 1987. Limnologia de l’estany Re-

El embalse de Anarbe 225

do (Pirineu Central). Tesis Doctoral. Universidadde Barcelona. 230 pp.

CHORUS, I. & J. BARTRAM. 1999. Toxic Ciano-bacteria in water. A guide to their public healthconsequences, monitoring and management. WordHealth Organization (WHO). London. E & FHSpon. 416 pp.

DOCE. 2000. DIRECTIVA 2000/60/CE del Parlamen-to Europeo y del Consejo de 23 de octubre de 2000.DiarioOficial de lasComunidadesEuropeas, Serie Lno 327, de22deDiciembrede2000. 72pp.

GOBIERNO VASCO. 2002. Caracterizacion de lasmasas de aguas superficiales de la CAPV. Tomo 3.Embalses. Dpto. Ordenacion del Territorio y Me-dio Ambiente. Direccion de Aguas. 197 pp.

HAKANSON, L. 1981. A manual of Lake Morpho-metry. Springer-Verlag Ed. Berlin. 78 pp.

HUTCHINSON, G. E. 1957. A treatise on limnology.Vol. I. Geography, physics and chemistry. JohnWi-ley & Sons, New York, NY. 1015 pp.

INFRAESTRUCTURA & ECOLOGIA S.L. 2003.Seguimiento de la calidad de las aguas en em-balses de zona sensibles. Documento de Sıntesis.MMA. CHE. Comisarıa de aguas. 45 pp.

INFRAESTRUCTURA & ECOLOGIA S.L. 2006.Memoria. Ejecucion de trabajos relacionados conlos requisitos de la Directiva Marco (2000/60/CE)en el ambito de la Confederacion Hidrograficadel Ebro referidos a: Elaboracion del registrode zonas protegidas, Determinacion del potencialecologico en embalses, Desarrollo de programasespecıficos de investigacion. MMA. CHE. Direc-cion General del Agua. 88 pp.

INTERNATIONALSTANDARD, ISO 9391:1993 (E).Water quality-Sampling in deep waters for macro-invertebrates-Guidance on the use of colonization,qualitative andquantitative samplers. 13pp.

JRC Report EUR 14563. PREMAZZI, G. & CHIA-DUANNI, G. 1992. Ecological quality of Surfa-ce Water. Comision of the European Communities,124 pp.

LARUMBE, M. I. 1991. Estudio de las variables que

influyen sobre la calidad de las aguas del embal-se de Anarbe. Tesis Doctoral. Universidad del PaısVasco/Euskal Herriko Unibertsitatea. Fac. Quımi-cas. Donostia-San Sebastian. 258 pp.

MARGALEF, R. 1983. Limnologıa. Ed. Omega. Bar-celona. 1010 pp.

OCDE. 1982. Eutrophisation des eaux. Metodes desurveillance, d’evaluationet de lutte.Paris. 164 pp.

PEREZ-BOTE, J. L., R. MORAN, R. ROSO & A. J.ROMERO. 2005. Craspedacusta sowerbii, la me-dusa que vino de Asia. Quercus, 236: 22-25.

REAL DECRETO 927/1988, de 29 de julio. Regla-mento de la Administracion Publica del Agua y dela Planificacion Hidrologica, en desarrollo de lostıtulos II y III de la Ley de Aguas. B.O.E. num.209: 26412-16425, de 31 de agosto de 1988.

U. S. EPA.1976. Quality Criteria for water. The redbook. Washington, D.C. (EPA 440/9-76-023, July,1976). 534 pp.

UTERMOHL, H. 1958. Zur Vervollkommnung derquantitativen Phytoplankton-Methodik. Mitt. Int.Ver. Theor. Angew. Limnol., 9: 1-38.

WETZEL, R. G. 1981. Limnologıa. Ed. Omega. Bar-celona. 679 pp.

WFD CIS Guidance Document No 4. 2003. Identifi-cation and Designation of Artificial and HeavilyModified Waterbodies. Published by the Directo-rate General Environment of the European Co-mision, Brussels, ISBN No 92-894-5124-6, ISSNNo 1725-107.

WFD CIS Guidance Document No 10. 2003. Riversand Lakes-Typology, Reference Conditions andClassification Systems. Published by the Directo-rate General Environment of the European Co-mision, Brussels, ISBN No 92-894-5614-0, ISSNNo 1725-1087.

WFD CIS Guidance Document No 13. 2005. OverallApproach to the Classification of Ecological Statusand Ecological Potential. Published by the Direc-torate General Environment of the European Co-mision, Brussels, ISBN N 92-894-6968-4, ISSNNo 1725-1087.

Limnetica, 27 (2): x-xx (2008)Limnetica, 27 (2): 227-238 (2008)c© Asociacion Iberica de Limnologıa, Madrid. Spain. ISSN: 0213-8409

Distribution patterns of Hydropsychids and Rhyacophilids species(Trichoptera) in a not regulated Mediterranean river (SW Spain)

Antonio Ruiz Garcıa ∗,1 and Manuel Ferreras-Romero 1

Departamento de Sistemas Fısicos, Quımicos y Naturales. Universidad Pablo de Olavide. 41013 Sevilla, Spain.2

∗ Corresponding author: aruigar@upo.es2

Received: 7/8/07 Accepted: 28/5/08

ABSTRACT

Distribution patterns of Hydropsychids and Rhyacophilids species (Trichoptera) in a non-regulated Mediterraneanriver (SW Spain)

This paper investigates the longitudinal ordination of the Hydropsychidae and Rhyacophilidae species present in the highbasin of the Hozgarganta River. The ordination of the Hydropsychids species present in the basin show three distributionpatterns: Diplectrona felix and Hydropsyche infernalis are confined to the head streams; H. siltalai and H. iberomaroccana aredistributed throughout the entire study zone, although the second one does not reach the highest sections of the basin; finally,H. lobata occupies the lowest sections, specially in the main axis of the river. The two Rhyacophila species studied show alsoa well differentiated distribution. R. fonticola is more abundant in the gorges and high sections, whereas R. munda prefers theriverbed of the main river, although it’s also found in some tributaries. Even in these intermediate sections, the segregationbetween the two species is almost perfect.The water permanence in the riverbeds influences the distribution of some species. We have found thatD. felix andH. infernalisinhabit the permanent sections of the stream heads whereas H. iberomaroccana significantly dominates the intermittent ones.H. siltalai, H. lobata, R. fonticola, and R. munda do not show a significant preference for any type of riverbed (permanent,intermittent, or ephemeral). It is interesting to highlight the survival of some H. iberomaroccana larvae in isolated poolsduring the summer. We suggest the possibility that these larvae survive thanks to the existence of a weak underground flow ofsubsurface origin between adjacent pools.

Key words: Hydropsychidae, Rhyacophilidae, distribution, seasonality, Hozgarganta river basin.

RESUMEN

Patrones de distribucion de especies de Hydropsychidae y Rhyacophilidae (Trichoptera) en un rıo mediterraneo no regu-lado (Suroeste de la Penınsula Iberica)

En el presente trabajo se investiga la ordenacion longitudinal de las especies de Hydropsychidae y Rhyacophilidae presentesen la cuenca alta del rıo Hozgarganta. La ordenacion de las especies de Hydropsychidae presentes en la cuenca muestra trespatrones de distribucion: Diplectrona felix e Hydropsyche infernalis estan confinadas en los arroyos de cabecera; H. siltalaie H. iberomaroccana estan distribuidas por toda la zona de estudio, aunque la segunda no alcanza los tramos mas altos dela cuenca; por ultimo, H. lobata ocupa los tramos mas bajos, especialmente en el eje principal del rıo. Las dos especies deRhyacophila estudiadas tambien muestran una distribucion bien diferenciada. R. fonticola es mas abundante en las gargantasy tramos altos, mientras que R. munda prefiere el cauce del rıo principal, aunque tambien se encuentra presente en algunostributarios. Incluso en estos tramos intermedios la segregacion entre ambas especies es casi perfecta.La permanencia del agua en los cauces influye en la distribucion de algunas especies. Hemos encontrado que D. felix eH. infernalis habitan los tramos permanentes de cabecera, mientras que H. iberomaroccana domina de forma significativa enlos intermitentes. H. siltalai, H. lobata, R. fonticola y R. munda no muestran una preferencia significativa por algun tipo decauce (permanente, intermitente o efımero). Es interesante resaltar la supervivencia de algunas larvas de H. iberomaroccanaen pozas aisladas durante el verano. Sugerimos la posibilidad de que estas larvas sobrevivan gracias a la existencia de undebil flujo subterraneo de origen freatico entre pozas adyacentes.

Palabras clave: Hydropsychidae, Rhyacophilidae, distribucion, estacionalidad, cuenca rıo Hozgarganta.

228 Ruiz & Ferreras-Romero

INTRODUCTION

Both Hydropsychids and Rhyacophilids are twoof the most diverse families of caddisflies in theIberian Peninsula, in addition to the Hydroptili-dae (Gonzalez et al., 1987) and they are wides-pread in all Spanish catchments.

The ecological differences between both fa-milies are a consequence of the way of using thesilk (Mackay &Wiggins, 1979). The Rhyacophi-lids have free-living larvae that only use the silkbefore pupation, whereas the Hydropsychidae aresedentary and net-building along with other filter-feeders, therefore they perform an important fun-ction in the treatment of the organic matter in flu-vial ecosystems (Basaguren & Orive, 1991). Inthe Mediterranean rivers they are, together withthe Hydroptilidae, the most important family ofcaddisflies due to their abundance and species’richness (Giudicelli et al, 1985).

Many studies have pointed out a longitudinalreplacement of Hydropsychidae species (Garcıade Jalon, 1986, Camargo, 1992, Gallardo-Mayenco et al., 1998). The environmental fac-tors change along the watercourse and the spe-cies occupying a particular position in the lon-gitudinal sequence seem to have appropriate sui-tes of physiological and behavioural characteris-tics (Edington & Hildrew, 1995).

Mediterranean climatic regions are distributedaround the world, and seasonality and variabilityin rainfall is its main characteristic. Although theseasonal precipitation pattern is highly predicta-ble in Mediterranean climatic areas, annual rain-fall can vary largely from year to year (Gasith &Resh, 1999). As a consequence of this climaticpattern, flow irregularity is one of the most im-portant features of the Mediterranean rivers (Giu-dicelli et al., 1985), with a seasonal pattern ofhigh discharge during the wet period, which isfollowed by low discharge in the summer. Al-though, in some areas, they can be dry.

Aquatic insects that live in temporary pondshave developed a series of strategies that allowthem to survive a periodic loss of habitat (Wig-gins et al, 1980). Several species of Trichopterainhabiting temporary streams have developedspecial adaptations. This is also the case of

Stenophylax species, which are well adapted toreplace other limnephilids in temporary waters:all species are univoltine; the eggs are enclosedby a gelatinous matrix that allows them to sur-vive long periods of time out of water; adult orovarian diapause is present; the burrowing larvaecan use the humid substratum during low-waterperiods, and the pupal cases are buried verticallyin the substratum. The Rhyacophilids larvae areprobably the caddisflies that are most restrictedto conditions of high current-speed (Edington &Hildrew, 1995), occupying, generally, headwa-ters. Hydropsychids larvae are rheophilic inhabi-tants of stream rifles (Edington & Hildrew, 1973;Fuller & Mackay, 1980; Osborn & Herricks,1987; Bonada, 2003). Because of its high disper-sion capacity, the survival of both Rhyacophilidsand Hydropsychids species during the dry periodin Mediterranean streams may involve a highcapacity for the reconstruction of colonies fromnearby permanent sources (Resh, 1982; 1992).

Though the longitudinal distribution of theHydropsychids has been largely studied, we thinkthat studies on the longitudinal zonation of en-demic or restricted distribution range species, li-ke some present in this catchment, are necessary.The aim of this study was to search for the res-ponse of Hydropsychid and Rhyacophilid speciespresent in the river, to the spatial-temporal gra-dients, as well as to establish their relationshipwith the water permanence.

METHODS

Study area and sampling methods

The Hozgarganta River basin covers a surface of245 km2 and it extends in a NW - SE direction(Fig. 1). The system drains the Eastern slopes ofthe mountain range of the Aljibe and after 55 kmthe watercourse joins the River Guadiaro, near itsmouth (Blanco et al., 1991).

The study area included all of the riverbasins within the limit of the Los AlcornocalesNatural Park. The river is born at 160 m a. s. l.,at the confluence of the Pasada Blanca canyonand the La Sauceda gorge. The first one flows

Distribution of Hydropsychids and Rhyacophilids in a not regulated Mediterranean river 229

Figure 1. Study area. Localizacion del area de estudio.

through calcareous lands and the second oneflows through the Aljibe unit (dominated bysandstone with marl and clay).

The study area is covered by a well preservedMediterranean forest, in which the cork oak(Quercus suber) dominates. Other commonspecies are the Andalusian gall oak (Quercuscanariensis) and the wild olive (Olea europaea).In the humid and shaded canyons there areremains of a subtropical forest, a relict from thetertiary age (e.g. Laurus nobilis, Rhododendrumponticum) and an arboreal stratum of alder-trees(Alnus glutinosa), willows (Salix atrocinerea,Salix pedicellata), and ash-trees (Fraxinus an-gustifolia) (Jurado-Dona, 2002). In this basin,the land is mostly used for cork oak forestexploitation and extensive ranching.

The flows in successive years show the ty-pical irregularity of the Mediterranean ri-vers, with dry and wet alternating periods

(Fig. 2a). The average annual flow in the period1980/99 was 1050 m3 s−1. In figure 2b we cansee that the year when the study was performed(1997) was a wet period.

Another important hydrological feature is theseasonality (Table 1). The maximum registeredannual flows are in autumn and early winter,whereas in summer the superficial currentdisappears, and the water is limited to isolatedpools. The average annual period in whichthere is no superficial flow is 2.65 months(range: 0-6 months, n = 20 years).

Nineteen sampling sites were chosen, fourin the river’s main channel and the rest on thetributaries (Fig. 1).

From November 1996 through December1997 samplings were carried out in November(1996), February, March, April, June, August,October, and December of 1997. In June, Eca andEre sampling sites (Table 1) were dry and there-

230 Ruiz & Ferreras-Romero

Table 1. Mean values of the physical-chemical parameters measured in the Hozgarganta basin. Cond: conductivity; Cl−: chloride;Alk: alkalinity; O2: dissolved oxygen; Temp: temperature; Altit: altitude; Season: seasonality *: 1. permanent flow; 2. ephemeral;3. intermittent. Inv: invaluable. Valores medios de los parametros fısico-quımicos medidos en la Cuenca del rıo Hozgarganta. Cond:conductividad; Cl−: cloruros; Alk: alcalinidad; O2: oxıgeno disuelto; Temp: temperatura; Altit: altitud; season: estacionalidad *:1. flujo permanente; 2. efımero; 3. intermitente. Inv: valor no detectado.

No Cod. Sampling site CondμS/cm

Cl−mg/l

Alkmeq/l

O2

mg/lTemp◦C

Altit a. s. l. Order Season * NO−3

mg/lPO=

4

μgat/lNH+

4

mg/l

1 E1 Diego Duro 0630 122,0 2.1 6.7 16.9 155 4 3 0.2 0.8 0.022 Ecn Puente las Canillas 0639 126,0 1.9 8.5 20.5 145 4 3 0.01 0.4 0.063 Eher Herrunbroso 0428 083,0 1.9 9.5 19.1 110 4 2 inv 0.2 inv4 E2 Jimena de la Frontera 0249 053,0 1.5 9.3 20.1 035 4 2 inv 0.2 0.015 Epbm Ga Pasada Blanca 1 0368 053,0 2.6 5.1 18.1 250 3 3 0.02 0.4 0,016 Epbh Ga Pasada Blanca 2 0433 061,0 3.0 5.1 18.2 245 3 3 0.01 0.2 0.037 Eo Garganta de la Balsa 0168 049,0 2.1 7.3 16.8 145 1 3 0.005 0.3 0.028 Ero Ga Pasada Llana 0148 027,0 1.4 8.7 16.0 500 2 3 0.007 0.2 0.029 Eca Ao Carnero 0123 020,0 1.2 7.8 10.9 360 1 2 inv inv inv

10 Esa Garganta de la Sauceda 1605 490,0 1.9 7.4 18.6 230 3 3 0.005 inv 0.0211 Edi Ao Reinoso (Diego Duro) 0095 020,0 1.2 7.7 15.1 320 1 2 inv inv inv12 Ere Ao Reinoso 0168 025,0 1.7 8.4 11.8 160 1 2 inv inv inv13 Emo Garganta de Moracha 0180 024,0 1.9 9.7 15.1 170 2 2 inv inv inv14 Ehu Garganta del Huevo 0142 026,0 1.3 9.6 14.4 230 2 2 inv inv 0.0115 Ega Garganta de Gamero 0326 080,0 2.5 9.7 19.7 060 2 2 inv 0.5 0.0216 Evi Ao de las Vinas 0355 048,0 2.4 9.3 19.9 050 2 3 0.006 0.3 0.0117 Em Canuto del Moro 009 001.6 1.2 7.5 12.3 880 1 1 inv inv inv18 Emc C. Molino de las Cuevas 012 002.1 1.7 7.8 12.1 800 1 1 inv inv inv19 Eml Canuto del Moral 009 001.7 1.3 8.8 12.4 700 1 1 inv inv inv

fore could not be sampled. In August there wasonly superficial water in E1, Ecn, Epbm, Epbh,Eo, Ero, Esa, and Evi sampling sites (Table 1). InDecember the sampling sites Epbm, Epbh, Ega,and Evi were inaccessible after autumn rains.

A kick net of 0.3×0.3 m opening and 0.5 mmmesh size and the same unit of effort in allthe localities (three replicates per site) was usedfor the extraction of the macroinvertebrates. Allhabitats were sampled along a 50 m riverbedstretch. The samples were fixed using 70% al-cohol and later identified to species level inthe laboratory. Five Hydropsychidae (Diplec-trona felix, Hydropsyche infernalis, H. Siltalai,H. Lobata, and H. iberomaroccana) and twoRhyacophilidae (Rhyacophila munda and Rhya-cophila fonticola) species were identified. Be-cause of the taxonomic difficulties in identif-ying H. Iberomaroccana larvae, this study wascomplemented with adult records.

For the physical-chemical analysis, samplesin February (winter), June (spring), August(summer), and December (autumn) were taken.The average values of the analysed parametersare shown in Table 1.

a)

b)

33

Volu

me

(m/s

eg)

3Volu

me

(m/s

eg)

Figure 2. a) Registered average annual flows in twenty yearsin the water gauge station of Jimena de la Frontera. Dischargevalues × 1 000. Caudal anual medio registrado durante veinteanos en la estacion de aforo de Jimena de la Frontera. Cau-dal × 1 000. b) Comparison of the monthly average dischar-ges (Qmean) of the temporal series 1981-99 with the registe-red ones in 1997 (Q97), measures in the water gauge station ofJimena de la Frontera. Elaborated from the database of Con-federacion Hidrografica del Sur de Espana. Comparacion delcaudal medio mensual (Qmean) de la serie temporal 1981-99con los registrados en 1997 (Q97), medidos en la estacion deaforo de Jimena de la Frontera. Elaborado a partir de la basede datos de la Confederacion Hidrografica del Sur de Espana.

Distribution of Hydropsychids and Rhyacophilids in a not regulated Mediterranean river 231

Data analysis

For the physical-chemical characterisation of thebasin a Principal Component Analysis (PCA) onphysical-chemical parameters × localities datamatrix was performed. To establish the environ-mental features of the river system, the four mostexplanatory variables in a previous work in thearea (Ruiz et al., 2006) were selected: altitude,conductivity, dissolved oxygen, and temperature.

The ordination method used was selected ba-sed on the length of the gradient calculated byDetrended Correspondence Analysis (DCA) (Bo-nada et al., 2005). Since the first DCA axis

has a gradient length of 5.2 standard deviationunits, the use of a unimodal ordination techniquewas justified. To study the relationship betweenenvironmental variables and species, a Canoni-cal Correspondence Analysis (CCA) with seaso-nal abundance data was performed. The signifi-cance of CCA axes was tested with the Mon-te Carlo permutation test (999 unrestricted per-mutations, P < 0.005). All variables used we-re log-transformed to achieve normality. All theanalyses were performed with the PCORD pro-gram. To test for the influence of water per-manence (seasonality) on the species’ distribu-tion a Kruskal-Wallis test was used. To test

Figure 3. Classification by Principal Component Analysis based on four selected environmental variables (altitude, conductivity,dissolved oxygen and temperature) of sampling sites of the Hozgarganta catchment. Clasificacion de las localidades muestreadas enla Cuenca del rıo Hozgarganta mediante un Analisis de Componentes Principales basado en cuatro variables (altitud, conductividad,oxıgeno disuelto y temperatura).

232 Ruiz & Ferreras-Romero

the thermal preferences of species an adjustedaverage temperature (AAT) was calculated (Ga-llardo-Mayenco et al., 1998).

RESULTS

The first PCA axis accounted for 52.9% of thevariance and shows sampling sites ordered ac-cording to altitude and water conductivity, repre-senting a geomorphologic and water permanencegradient, where headstream sites (Em, Eml andEmc) have permanent conditions (Fig. 3). Thenegative end of this axis was associated to spring

samples of downstream sites (E2, Ecn, E1 andEsa), where high values of both conductivity andtemperature were registered. The second PCAaxis explained 28.2% of the variance (the cumu-lative variance was 81.1%). This axis distributedthe sampling sites according to the concentrationof dissolved oxygen (Fig. 3). We can see that thesummer samples from E1 and Epbh are separatedfrom others due to the low value of the dissolvedoxygen. Seasonality was of little importance inexplaining the sites’ distribution.

The CCA performed from seasonal data sho-wed that the total variance (“inertia”) was 2.75;in the procedure three canonical axes were ob-

Figure 4. Projection of the two first axis of the CCA with seasonal data. Symbol code categories of season (winter (w); spring(sp); summer (s) and autumn (a)) are showed at the end of the acronyms of the sites. Symbol code categories of the variable waterpermanence (Perm) as follow: 1. permanent; 2. ephemeral; 3. intermittent. + symbol show species. Species codes are explained inTable 2. Sites codes are explained in Table 1. Proyeccion de los dos primeros ejes de un CCA con datos estacionales. El codigo delas estaciones del ano (invierno (w); primavera (sp); verano (s) y otono (a)) aparece al final de los acronimos de las localidadesde muestreo. El codigo de las categorıas de la variable permanencia del agua en el cauce (Perm) es el siguiente: 1. permanente;2. efımera; 3. intermitente. El sımbolo + representa a las especies. El codigo de las especie es el mismo que aparece en la Tabla 2.El codigo de las localidades de muestreo aparece en la Tabla 1.

Distribution of Hydropsychids and Rhyacophilids in a not regulated Mediterranean river 233

Table 2. Seasonality influence of the channels (permanent/intermittent/ ephemeral) in the species distribution calculatedwith the Kruskal-Wallis test. Significant p-values (P < 0.005)are show in bold. Influencia de la estacionalidad de los cursosde agua (permanentes/ intermitentes/efımeros) en la distribu-cion de las especies, calculada con el test de Kruskal-Wallis.Los valores significativos del ajuste (P < 0.005) aparecen ennegrita.

species code k-w test p-value

D. felix Dfe 17.84 0.000H. infernalis Hin 12.04 0.002H. siltalai Hsi 4.1 0.128

H. iberomaroccana Hib 9.87 0.007H. lobata Hlo 5.2 0.073

R. fonticola Rfo 2.5 0.275

R. munda Rmu 4.58 0.100

tained. The Monte Carlo test indicated that onlythe first two axes were significant ( p = 0.005).The first axis explained 28.8% of the varian-ce; the second one 6.7%. The cumulative va-riance was 35.4%. We obtained that the varia-bles strongly related to the first axis were altitude(r = −0.91) and conductivity (r = 0.88), and thetemperature was weakly related with the secondaxis (r = 0.18). The percentage of unexplainedvariance was 64.6%.

Figure 4 shows sites and species distributed inthe plane formed by axes 1 and 2. Axis 1 placedthe species from left to right according to altitude.D. felix and H. infernalis occupied headstreams;R. fonticola and H. siltalai inhabited middlestreams and the others were downstream species.

The second factor that determined species’distribution was water temperature and conduc-tivity. The low half of axis 2 (Fig. 4) groupedmost of the high conductivity sites, whereas thetop half of this axis discriminated sampling siteswith high temperature. Table 2 and figure 5 showthe species significantly related ( p < 0.05) toflow permanence. D. felix and H. infernalis weresignificantly present in permanent conditions,H. iberomaroccana inhabited intermittent sites,while R. munda, R. fonticola, H. siltalai, and H.lobata did not show a preference to a particularhabitat. R. fonticola was mainly associated withthe Ero site (intermittent conditions), thoughit also occupied headwaters (permanent condi-

Figure 5. Box-plot showing the abundance distribution of thethree significantly selected species in the kruskal-Wallis test ac-cording to water permanence in the channels. Code of the Xaxis: 1. permanent; 2. ephemeral and 3. intermittent. Distribu-cion de la abundancia de las tres especies significativamenteseleccionadas por el test de Kruskal-Wallis en funcion de lapermanencia del agua en los cauces. Eje X: 1. permanente;2. efımero; 3. intermitente.

tions); H. siltalai dominated the winter samplefrom the Emo site (ephemeral conditions).On the other hand, H. lobata occupied lowreaches, in both intermittent and ephemeralsites (Evi, Ega and E2 stations).

D. felix and H. infernalis were present in co-ol water conditions (12.1 and 12.8◦ C of AAT,respectively); H. siltalai and H. iberomarocca-

234 Ruiz & Ferreras-Romero

na were found in sites of moderate temperature(17.7◦ C and 17.38◦ C of AAT, respectively); and,H. lobata preferred warm conditions (19.4◦ C ofAAT). On the other hand, the Rhyacophilids spe-cies showed a similar pattern of thermal prefe-rences, where R. fonticola occupied the cool wa-ter sites (12.45◦ C of AAT), while R. munda wasabundant in warm waters (18.01◦ C of AAT).

DISCUSSION

Numerous studies have pointed out the replacementof species of Hydropsychidae along the watercourse(Camargo, 1992; Garcıa de Jalon, 1986; Verneaux& Fassel, 1976; Gallardo-Mayenco et al., 1998).This succession may be a consequence of differentfeeding habits (Voelz & Ward, 1992), metabolicneeds (Roux et al., 1992), and both differentialcompetitive ability and different mesh size of thespecies (Tachetetal., 1992).

In Mediterranean streams a strong environ-mental gradient between headwater and loweraltitude streams has been observed (Gallardo-Mayenco et al., 1998). In our study, the geomor-phological gradient is the most important factorexplaining the distribution pattern of Hydropsy-chid species. In the high Hozgarganta basin wehave obtained an upstream assemblage, compo-sed of D. felix and H. infernalis; H. lobata in-habits lower reaches in the main river, while H.iberomaroccana and H. siltalai are widespreadthroughout the study area, thoughH. iberomaroc-cana does not reach upstream reaches (e.g. Canu-tos del Moral, Molino de las Cuevas y del Moro).

Rhyacophila larvae are probably the caddisflythat are most restricted to conditions of high cu-rrent speed. Their commitment to fast flow ratesreflects both the distribution of their food supplyand their physiological limitations (Edington &Hildrew, 1995). Their distribution is restricted,almost exclusively, to the headwaters (Tachet etal., 2002). In our case, both Rhyacophila speciesdisplay a differentiated distribution pattern. R.munda is a wide spectrum ecological species,inhabiting from cool upstream habitats to warmstream ones and the great lower rivers (Garcıa deJalon & Gonzalez del Tanago, 1986). In the Hoz-

garganta river, R. munda inhabits the principalchannel and it is displaced at the headwaters byR. fonticola, a species characteristic of springs(Giudicelli & Dakki, 1984) and upstreams (Ruizet al., 2001). A similar result was obtainedby Garcıa de Jalon & Gonzalez del Tanago(1986) in the Guadalteba river, where R. pascoeidisplaces R. munda at the headwaters. In thesame way, Fernandez-Alaez et al. (2002) foundR. terpsichore in the upper reaches of the Boezariver, which disappeared in the lower reaches,where ubiquitous species like R. meridionalisand R. relicta appeared.

Another characteristic of the Mediterraneanrivers is the irregularity of their flow, maximumin autumn and spring and minimum in summer(Giudicelli et al., 1985) and its high inter-annualvariability (McElravy et al., 1989), that may im-ply an inter-annual variability in the conditionsof temporality of one site (Del Rosario & Resh,2000). This determines that the temporary riversare widely distributed by the climates of Medite-rranean type (Gasith & Resh, 1999).

The duration of the dry period has been re-cognised as an important factor explaining thebiological diversity in these rivers (Williams &Hynes, 1976, Abell, 1984, Williams, 1996, Bona-da, 2003, Ruegg & Robinson, 2004). Overall, thetemporary rivers have fewer species than the per-manent ones (Del Rosario & Resh, 2000, Bona-da, 2003, Arab et al., 2004, Ruegg & Robinson,2004). In this sense, Bonada (2003) suggests thatthe low taxonomical richness in ephemeral con-ditions, and the high difference in richness frompermanent sites, would suggest a slow recoveryfrom the last dry period.

Hydropsychids typically build their nets inrapidly-flowing waters (Edington & Hildrew,1995; Tachet et al., 2002) and in mediterraneanrivers, they are indicators of riffles conditions,along with Rhyacophilids, (Bonada, 2003). Ourstudy agrees with these results, because D. fe-lix and H. infernalis are restricted to the ups-tream reaches; H. siltalai. R. munda and R. fon-ticola either inhabit intermittent streams or areindifferent to water permanence. In this case,their life cycle is adapted to these environmen-tal conditions, finishing their larval growth in the

Distribution of Hydropsychids and Rhyacophilids in a not regulated Mediterranean river 235

wet period (winter-spring). However, H. ibero-maroccana and H. lobata displayed populationsthroughout the year, despite the fact that theyprefer temporary conditions. Our results are inagreement with those by Gallardo-Mayenco etal. (1998) that, in a study of Hydropsychid spe-cies in the Guadaira and Guadalete river basins,point out the opportunistic behaviour of H. ibe-romaroccana (as H. punica), specially due toits capacity to colonise and thrive in temporaryhabitats. In the Hozgarganta river these speciesspend the dry season (summer) living in isola-ted pools, where the absence of superficial flow isnoticeable. Numerous studies highlighted the ef-fects of water velocity on the distribution of Hy-dropsychidae species. Overall, downstream spe-cies seem to be more tolerant to low velocityconditions than upstream species (Tachet et al,1992). Hildrew and Edington (1979) showed thatH. pellucidula, a sister species to H. iberomaroc-cana, was indifferent or more abundant in lowvelocity sites. On the other hand, the interrup-tion of the superficial flow does not necessarilyimply that of an underground flow between adja-cent pools that might remain connected throughthe aquifer. We suggest that H. iberomaroccanamight live in a wide range of current speeds andthis allows it to survive in underground-fed pools.

The coexistence of several Hydropsychidspecies is facilitated by microhabitat partition(Czachorowski, 1989; Harding, 1997), wheresubstrate size and type is important in modu-lating species’ micro distribution. However,multiple factors operate synergistically overseveral spatial scales and thus, influence the dis-tribution of the Hydropsychid species (Fairchild& Holomuzki, 2002) in addition to temporalsegregation in their life histories (Recasens &Puig, 1987). Hydrological disturbances mayalso facilitate the coexistence of Hydropsychidspecies (Resh et al., 1990). In our case, D. felixand H. infernalis were two coexistent speciesin permanent headstreams. Our data suggeststhat coexistence is possible, at least partially,by temporal segregation, because D. felix wasdominant in the spring-summer season while H.infernalis was very scarce in summer.

ACKNOWLEDGEMENTS

We thank Tony Herrera, Juan C. Salamanca andFrancisco Cano-Villegas for their assistance withthe field work and two anonymous reviewers fortheir valuable comments and improvements onthe manuscript.

REFERENCES

ABELL, D. L. 1984. Benthic invertebrates in someCalifornia intermittent stream. In: Jain, S. & Moy-le, P. (eds.). Vernal pools and intermittent streams.Institute of Ecology. Publication no 28. UC Davis.Davis, CA. 46-60 pp.

ARAB, A., S. LEK, A. LOUNACI & Y. S. PARK.2004. Spatial and temporal patterns of benthicinvertebrate communities in an intermittent river(North Africa). Ann. Limnol.-Int. J. Lim., 40: 317-327.

BASAGUREN, A. & E. ORIVE. 1991. Los insectostricopteros como indicadores de calidad del aguaen los rıos de Vizcaya. Cuenca del Nervion. Kobie,20: 29-50.

BLANCO, R., J. CLAVERO, A. CUELLO, T. MA-RANON & J. A. SEISDEDOS. 1991. Guıas natu-ralısticas de la provincia de Cadiz III. Sierras delAljibe y del Campo de Gibraltar. 315 pp.

BONADA, N. 2003. Ecology of the macroinvertebra-te communities in mediterranean rivers at differentscales and organisation levels. Tesis de Doctorado.Universidad de Barcelona. 355 pp.

BONADA, N., C. ZAMORA-MUNOZ, M. RIERA-DEVALL & N. PRAT. 2005. Ecological and his-torical filters constraining spatial caddisfly distri-bution in Mediterranean rivers. Freshwat. Biol.,50: 781-797.

CAMARGO, J. A. 1992. Changes in a Hydropsychidguild downstream from a eutrophic impoundment.Hydrobiologia, 239: 25-32.

CZACHOROWSKI, S. 1989. Differentiation of thehabitats of Hydropsychidae larvae (Insecta: Tri-choptera) in the pasleka river as a result of avoidan-ce of trophic competition. Pol. Arch. Hydrobiol.,36: 123-132.

DEL ROSARIO, R. B. & V. R. RESH. 2000. Inverte-brates in intermittent and perennial streams: Is thehyporheic zone a refuge from drying? J. N. Am.Benthol. Soc., 19: 680-696.

236 Ruiz & Ferreras-Romero

EDINGTON, J. M. & A. G. HILDREW. 1973. Expe-rimental observations relating to the distribution ofnet-spinning Trichoptera in streams. Verein. Theor.Internat. Ang. Limnol., 18: 1549-1558.

EDINGTON, J. M. & A. G. HILDREW. 1995. Case-less caddis larvae of the British Isles. A key withecological notes. Freshwater Biological Associa-tion. 134 pp.

FAIRCHILD, M. P. & J. R. HOLOMUZKI. 2002.Spatial variability and assemblage structure ofstream hydropsychid caddisflies. J. N. Am. Bent-hol. Soc., 21: 576-588.

FERNANDEZ-ALAEZ, C., J. DE SOTO, M. FER-NANDEZ-ALAEZ & F. GARCIA CRIADO.2002. Spatial structure of the caddisfly (Insec-ta, Trichoptera) communities in a river basin inNW Spain affected by coal mining.Hydrobiologia,487: 193-205.

FULLER, R. L. & R. J. MACKAY. 1980. Field andlaboratory studies of net-spinning activity by Hy-dropsyche larvae (Trichoptera: Hydropsychidae).Can. J. Zool,. 58: 2006-2014.

GALLARDO-MAYENCO, A., J. PRESA & J. TO-JA. 1998. Spatio-temporal distribution and eco-logical preferences of coexisting Hydropsychidspecies (Trichoptera) in two mediterranean ri-ver basins (S. Spain). Internat. Rev. Hydrobiol.,83(2): 123-134.

GARCIA DE JALON, D. & M. GONZALEZ DELTANAGO. 1986. Ephemeroptera, Plecoptera yTrichoptera de los principales rıos de Malaga. IISimp. del agua en Andalucıa: 331-346.

GARCIA DE JALON, D. 1986. Los Hydropsychidae(Trichoptera) de la cuenca del Duero. Boln. Asoc.Esp. Ent., 10: 127-238.

GASITH, A. & V. H. RESH. 1999. Stream in medite-rranean climate regions: abiotic influences and bio-tic responses to predictable seasonal events. Annu.Rev. Ecol. Sist., 30: 51-81.

GIUDICELLI, J. & M. DAKKI. 1984. Les sourcesdu Moyen Atlas et du Rif (Maroc): faunistique(description de deux especes nouvelles de Tri-chopteres), ecologie, interet biogeografique. Bij-dragen tot de Dierkunden, 54(1): 83-100.

GIUDICELLI, J., M. DAKKI & A. DIA. 1985. Ca-racteristiques abiotiques et hydrobiologiques deseaux courantes mediterraneenes. Verth. Internat.Verein. Limnol., 22: 2094-2101.

GONZALEZ, M. A., D. GARCIA DE JALON & L.W. DA TERRA. 1987. Faunistic studies on IberianTrichoptera: a historical survey and present state

of knowledge. In: Proc. of the 5th Int. Symp. onTrichoptera. M. Bournaud & H. Tachet (eds.): 85-90. Dr. W. Junk Publisers, Dordrecht.

HARDING, J. S. 1997. Strategies for coexistence intow species of New Zealand Hydropsychidae (Tri-choptera). Hydrobiologia, 350: 25-33.

HILDREW, A. G. & J. M. EDINGTON. 1979. Fac-tors facilitating the coexistence of Hydropsychidcaddis larvae (Trichoptera) in the same river sys-tem. Journal of Animal Ecology, 48: 557-576.

JURADO-DONA, V. 2002. Los bosques de la sie-rra del Aljibe y del Campo de Gibraltar (Cadiz yMalaga). Ecologıa, transformaciones historicas ygestion forestal. Consejerıa del Medio Ambiente.Junta de Andalucıa. 219 pp.

MACKAY, R. J. & G. B. WIGGINS. 1979. Ecologi-cal diversity in Trichoptera. Ann. Rev. Entomol.,24: 185-208.

McELRAVY,E.P., G. A. LAMBERTI& V. H. RESH.1989. Year-to-year variation in the aquatic macro-invertebrate fauna of northern California stream. J.N. Am. Benthol. Soc., 8(1): 51-63.

OSBORNE, L. L. & E. E. HERRICKS. 1987. Micro-habitat characteristics of Hydropsyche (Trichopte-ra: Hydropsychidae) and the importance of bodysize. J. N. Am. Benthol. Soc. 6: 115-124.

PRENDA, J. & A. GALLARDO-MAYENCO. 1999.Distribution patterns, species assemblages and ha-bitat selection of the stoneflies (Plecoptera) fromtwo mediterranean river basins in Southern Spain.Internat Rev. Hydrobiol., 84(6): 595-608.

RECASENS, L. & M. A. PUIG. 1987. Life cyclesand growth patterns of trichoptera in the Mata-rrana, a karstic river. In: Proc. of the 5th Int.Symp. on Trichoptera. M. Bournaud & H. Tachet(eds.): 247-251. Dr. W. Junk Publishers, Dordre-cht: 247-251.

RESH, V. H. 1982. Age structure alteration in a cad-disfly population after habitat loss and recovery.Oikos, 38: 280-284.

RESH, V. H. 1992. Year to year changes in the agestructure of a caddisfly population following lossand recovery of a springbrook habitat. Ecography,15: 314-317.

RESH, V. H., J. K. JACKSON & E. P. McELRAVY.1990. Disturbance, annual variability and loticbenthos: examples for a California stream influen-ces by a mediterranean climate. Mem. Ist. Ital.Idrobiol., 47: 309-329.

ROUX, C., H. TACHET, M. BOURNAUD & B. CE-LLOT. 1992. Stream continuum and metabolic rate

Distribution of Hydropsychids and Rhyacophilids in a not regulated Mediterranean river 237

in the larvae of five species of Hydropsyche (Tri-choptera). Ecography, 15: 70-76.

RUEGG, J. & C. T. ROBINSON. 2004. Comparisonof macroinvertebrate assemblages of permanentand temporary streams in an Alpine flood plain,Switzerland. Arch. Hydrobiol., 161: 489-510.

RUIZ, A., SALAMANCA-OCANA, J. C. & M.FERRERAS-ROMERO. 2001. Fauna de tricopte-ros (Insecta: Trichoptera) de cursos de agua quedrenan canutos del Parque Natural Los Alcornoca-les (Sur de Espana). Boln. Asoc. Esp. Ent., 25(3-4): 105-120.

RUIZ GARCIA, A., A. F. HERRERA GRAO & M.FERRERAS-ROMERO. 2006. Distribution of Tri-choptera communities in the Hozgarganta catch-ment. (Los Alcornocales Natural Park SW Spain).Internat. Rev. Hydrobiol., 91(1): 71-85.

TACHET H., J. P. PIERROT, C. ROUX & M.BOURNAUD. 1992. Net-building behaviour of sixHydropsyche species (Trichoptera) in relation tocurrent velocity and distribution along the RhoneRiver. J. N. Am. Benthol. Soc., 11: 350-365.

TACHET, H., P. RICHOUX, M. BOURNAUD & P.

USSEGLIO POLATERA. 2002. Invertebres d’eaudouce. Systematique, biologie, ecologie. CNRSeditions. 587 pp.

VERNEAUX, J. & B. FASSEL. 1976. Larves dugenre Hydropsyche (Trich., Hydropsychidae). Ta-xonomie, donnees biologiques et ecologiques.Annls. Limnol., 12(1): 7-16.

VOELZ, N. A. & J. V. WARD. 1992. Feedings ha-bits and food resources of filter-feeding Trichopte-ra in a regulated mountain stream. Hydrobiologia,231: 187-196.

WIGGINS, G. B., R. J. MACKAY & I. M. SMITH.1980. Evolutionary and ecological strategies ofanimals in annual temporary pools. Arch. Hydro-biol. Suppl., 58: 97-206.

WILLIAMS, D. D. & H. B. N. HYNES. 1976. Theecology of temporary streams. I. The fauna oftow Canadian streams. Int. Rev. ges. Hydrobiol.,61: 761-787.

WILLIAMS, D. D. 1996. Environmental constraintsin temporary waters and their consequences for theinsect fauna. J. N. Am. Benthol. Soc., 15(4): 634-650.

Limnetica, 27 (2): x-xx (2008)Limnetica, 27 (2): 239-250 (2008)c© Asociacion Iberica de Limnologıa, Madrid. Spain. ISSN: 0213-8409

Macroinvertebrates assemblages on reed beds, with special attentionto Chironomidae (Diptera), in Mediterranean shallow lakes

Maria Sahuquillo ∗,1, Maria Rosa Miracle 1, Maria Rieradevall 2 and Riyzard Kornijow 3

1 Dept. de Microbiologia i Ecologia. Universitat de Valencia. 46100 Burjassot (Valencia) Spain2 Dept. de Ecologia. Universitat de Barcelona. 08007 Barcelona. Spain3 Dept. Hydrobiol. Ichthyobiol. University of Agriculture in Lublin. 20-950 Lublin, Akademicka 13, Poland.2

∗ Corresponding author: maria.sahuquillo@uv.es2

Received: 15/11/06 Accepted: 13/5/08

ABSTRACT

Macroinvertebrates assemblages on reed beds, with special attention to Chironomidae (Diptera), in Mediterraneanshallow lakes

Macroinvertebrates associated to reed-beds (Phragmites australis) in six shallow natural water bodies along the 220 km ofcoast of the Comunidad Valenciana (Spain) were studied. These sites were selected to reflect different trophic states, but also,and due to the natural variability of mediterranean wetlands, they greatly differ in salinity and hydroperiod. To unify the sam-pling, reed bed was chosen to provide data from a habitat common to all wetlands, including the most eutrophic ones wheresubmerged macrophytes have disappeared due to water turbidity. Individual submerged stems of Phragmites australis weresampled along with the surrounding water. The animal density found refers to the available stem surface area for colonization.Forty-one taxa were recorded in total, finding Chironomidae to be the most important group, quantitatively and qualitatively.In freshwater sites it was observed an increase in macroinvertebrate’s density at higher trophic states. Nevertheless each stu-died region had a different fauna. The PCA analysis with macroinvertebrate groups distinguished three types of environment:freshwaters (characterized by swimming insect larvae, collectors and predators, oligochaetes and Orthocladiinae), saline wa-ters (characterized by crustaceans and Chironominae) and the spring pool, which shares both taxa. Chironomids were paidspecial attention for being the most abundant. A DCA analysis based on the relative abundance of Chironomids reveals sali-nity as the main characteristic responsible for its distribution, but trophic state and hydrological regime were also shown to beimportant factors.

Key words: Macroinvertebrates, Phragmites australis, chironomids, Mediterranean wetlands.

RESUMEN

Comunidades de macroinvertebrados asociados al carrizo en humedales mediterraneos, con especial atencion a los Chiro-nomidae (Diptera)

Se han estudiado los macroinvertebrados asociados a la vegetacion de carrizo (Phragmites australis) en seis lagunas some-ras a lo largo de los 220 km de costa de la Comunidad Valenciana (Espana). Las lagunas se eligieron de manera que secontemplasen diferentes estados troficos, pero ademas y reflejando la natural variabilidad de los humedales mediterraneos,presentan importantes diferencias en cuanto al hidroperiodo y la salinidad. Para unificar el muestreo, se eligio el carrizopor ser un elemento comun en todos los humedales, incluso los mas eutroficos en los que los macrofitos sumergidos handesaparecido debido a la turbidez del agua. Para cada muestra se toma individualmente la parte sumergida de una planta dePhragmites australis junto con el agua circundante. La densidad de animales encontrados se refiere a superficie colonizabledel tallo de la planta. En total se han encontrado 41 taxones, siendo Chironomiidae el grupo mas importante tanto cuanti-tativa como cualitativamente. En las lagunas de agua dulce, se observo un aumento de la densidad de macroinvertebradosen los niveles troficos mayores. Sin embargo cada zona de estudio tiene una fauna diferente. El analisis PCA de los gruposde macroinvertebrados diferencia tres tipos de ambientes: aguas dulces (caracterizados por larvas nadadoras de insectosrecolectores y depredadores, oligoquetos y orthocladinos), aguas salobres (caracterizadas por la presencia de crustaceos yChironomiinae) y el “ullal” o surgencia de agua que comparte taxones de los dos grupos anteriores. Se ha prestado especial

240 Sahuquillo et al.

atencion al grupo de quironomidos por ser los mas abundantes. Un analisis DCA basado en la abundancia relativa de lasespecies de quironomidos, muestra la salinidad como la principal caracterıstica responsable de su distribucion, siendo luegoel estado trofico y el regimen hidrologico factores tambien importantes.

Palabras clave: Macroinvertebrados, Phragmites australis, quironomidos, humedales mediterraneos.

INTRODUCTION

Reed beds are the most common and dominanttype of vegetation on the littoral zone of mostshallow lakes, even the most eutrophic ones whe-re submerged macrophytes could have disappea-red due to increased turbidity. They representan essential part in wetlands and lake littoraland play a crucial role as shelter and substra-ta for food resources for many organisms, frommacroinvertebrates to fish and birds. In addi-tion,their litter supports an important detritivo-rous biocenosis (Varga, 2001). Although they ha-ve undoubted importance, the biological commu-nity associated to reed stems in Mediterraneanwetlands is scarcely known.

In addition, its simple architecture enablesthe colonisable surface area to be measuredeasily. Consequently, submerged reed stems arean interesting habitat to measure quantitative andcomparable variables of the biological commu-nity for different types of lakes. For this reason,the study of their macroinvertebrate communityhas been proposed as a useful tool in the de-termination of the ecological status of wetlandsand shallow lakes (Kornijow & Kairesalo, 1994).The present paper is a further development ofthe data collected for a pan-European project(ECOFRAME, Moss et al., 2003) which attem-pted to test a classification system for shallowlakes in the field according to the requirementsof the European Water Framework Directive(WFD). Among a wide range of biologicalvariables several macroinvertebrates indexeswere tested in this project, including some basedon macroinvertebrates of reed beds.

This study presents the quantitative and quali-tative results for the macroinvertebrates associa-

ted with emergent macrophytes (Phragmites aus-tralis) in six shallow water bodies at the Medite-rranean coast of the Iberian Peninsula. Since chi-ronomids were the dominant macroinvertebrategroup, they were studied in more detail and at thefinest possible taxonomic level.

STUDY SITES

Macroinvertebrates associated to reeds weresampled in six shallow water bodies locatedin protected wetlands along the 220 km of theMediterranean coast of Valencia, Spain in Juneand July 2001. The same sites were studiedfor the mentioned multidisciplinary projectECOFRAME and are described elsewhere(Moss et al., 2003; Blanco et al., 2003;Sahuquillo et al., 2007; Poquet et al., 2008).

All sites were very shallow (less than 3 m indepth) and they were chosen due to their oligo-hypereutrophic gradient: Albufera being hyper-eutrophic, Cap de Terme eutrophic, Hondo me-sotrophic, and the rest oligo-mesotrophic (Xere-sa, Baldovı and Cabanes). In addition, and re-flecting the natural variability in the Mediterra-nean wetlands, the sites differed in other environ-mental characteristics. All sites are small shallowlakes, except the large Albufera lagoon; the na-mes used here correspond with the general na-mes of the wetland where they are located. Twosites had brackish waters: Hondo and Cabanes,the former one mainly due to evaporation and soilcharacteristics and the latter to marine influen-ce. With respect to hydrology, Baldovı is a springfed pond, which has a high subterranean karsticwater flow of constant temperature and mineralcomposition, with some marine influence due to

Macroinvertebrates assemblages on reed beds, in Mediterranean shallow lakes 241

sea closeness. The flow is also relatively impor-tant in Cap de Terme, excavated for drainage totransform the marshland into agricultural fields;and in the Albufera lagoon (Vicente & Miracle,1982). Both sites received important agricultu-ral inflows and Albufera also received industrialand domestic waters (only partly purified). Ca-banes is an excavated pond for peat extractionwith permanent water and without important in-flows. Xeresa and Hondo represent hydrologica-lly fluctuant systems because they are shallowand have reduced water inflows that determinetheir semi-permanent character; they can dry upat the end of summer during very dry years. Allsites have extensive reed belts.

Table 1 summarizes the most important fea-tures of the studied sites and the values of somelimnological variables measured simultaneouslyduring the samplings of reeds stems.

METHODS

Macroinvertebrate sampling

Invertebrates from submerged reed (Phragmitesaustralis) stems were sampled in four areas ateach one of the study sites from June 28 to July3, 2001. The areas were chosen randomly across

sites and stems were taken with a boat fromthe water edge of each reed bed. The samplingmethod involves collecting each individual reedstem and surrounding water by means of atransparent plastic tube (5 cm internal diameterand 75 cm high) with a cutting devise and a filterat the bottom (Kornijow & Kairesalo, 1994).After cutting the aerial part of the stem, the tubewas lowered into the water surrounding the stem,the top of the tube was closed with a cork and thestem cut at the lower base, then the stem floatedand the tube was closed with a 250 μm sievecup, which allowed us to pick up the individualstem and filter the surrounding water. Five stemswere collected separately within each area butpooled together to give one sample per area.Right after sampling the larger individuals wereindividually picked out and preserved in 70%ethanol. The stems were then rinsed thoroughlywith water and this water together with the restof the sample was filtered through a sieve ofa smaller size (100 μm) to collect all the otheranimals, which were preserved in the sameway. Macroinvertebrates were manually sorted,identified and counted under a stereoscope.

The length and diameter of each stemcollected was measured to calculate theirtotal area and express the density of animalsper m2 of stem surface area.

Table 1. Main characteristics of studied sites during sampling dates in 2001. Sites are ordered from North to South. Caracterısticasprincipales de los lugares de estudio durante el muestreo ordenados por latitud.

UTMXY

Area(ha)

Conductivity(μS/cm)

Secchi(m)

Chl-a(μg/l)

Temp(◦ C)

Date

CABANES (Charca del Pozo) 31T 2619134452142

2.3 11 840 Bottom 1.8 24.2 2 jul

ALBUFERA lagoon 30S 7305244356507

2200 2230 0.37 169.7 28.3 28 jun

BALDOVI 30S 7315774347997

0.5 3050 Bottom 1.2 30.6 1 jul

CAP DE TERME 30S 7399774326566

3.5 2660 0.4 86.5 29.5 29 jun

XERESA (Rini) 30S 7432604321856

0.1 2590 Bottom 5.7 11.4 29 jun

HONDO (Charca Sud-Oeste) 30S 6969254229649

11.0 18 480 Bottom 35.5 31.7 3 jul

242 Sahuquillo et al.

Macroinvertebrate and Chironomidaeidentification

In all the samples, invertebrates were identified tothe following taxonomic levels: Turbellaria, Nai-didae, other Oligochaeta, Gastropoda, Amphipo-da, Isopoda, other Crustacean, Hydracarida, Zy-goptera, Anisoptera, Ephemeroptera, Heteropte-ra, Lepidoptera, Trichoptera, Tanypodinae, Ort-hocladiinae, Chironominae, other Chironomidae,and Ceratopogonidae, following the ECOFRA-ME scheme that looked for a balance between theinformation required for a quality index and thetime needed for identification and counting.In two samples per site, individuals werecounted at species level and identified whenpossible, with special attention to chironomids.Chironomid larval muscle mass and tissues thatcould have made identification difficult wereremoved with a KOH solution warmed to 85◦ Cfor 20 minutes, They were then dehydratedin 70% and 96% ethanol successively, andmounted in permanent slides in Euparalmedia. Chironomids were identified at the lo-west taxonomical level possible following Hir-venoja (1981), Wiederholm (1983), and Schmid(1993). Pupal skins were used to validate larvalidentifications (Langton, 1991). Coding namesin the figures presented followed Schnell et al.(1999). At the same time direct Chironomid gutcontent observations were performed.

Data analysis

Density differences between sites were testedusing one-way ANOVA. Number of taxa (ta-xon richness) and H’ diversity index (Shannon &Wiener) were calculated with samples analyzedat the species level.

A Principal component analysis (PCA) wasapplied to standardized data consisting ofarcsen √ transformed proportions of macroinver-tebrate groups. The number of variables was re-duced by grouping taxa at a high taxonomic le-vel for the non-predators: Oligochaeta, Gastro-poda Crustaceans, Hydracharida, Ephemeropte-ra, Trichoptera, Orthocladiinae, and Chironomi-nae. Due to the importance of the feeding ha-

bits, predator taxa were treated separately in threegroups: non-dipteran insect predators (Zygop-tera, Anisoptera, Heteroptera, Lepidoptera, andthe Ecnomus genus within Trichoptera), dipteranpredators (Ceratopogidae and Tanypodinae), andsmall predators (hydra and microturbellarians).

RESULTS

In the six studied sites, densities of animals re-corded on the reed ranged between 260 and39 000 ind/m2 of stem surface area (Fig. 1a). In-dividual stems sampled had about 0.04-0.13 m2

surface area and these values can be used toestimate numbers per stem. In freshwater si-tes, density differences according to trophic sta-te categories were highly significant (0ne wayANOVA resulted significantly different d.f. = 3,F = 5.069, p < 0.01). Densities increased witheutrophy; the more eutrophic sites, Albuferaand Cap de Terme, had clearly higher den-sities of macroinvertebrates than the oligotro-phic sites (Baldovı and Xeresa) (Fig. 1a). Den-sities were generally lower in brackish watersthan in the freshwater sites.

A relation was found between increasing den-sity and percentage of chironomids but onlywhen similar systems were compared (Fig. 1b).In permanent freshwater aquatic systems, like Al-bufera, Cap de Terme and Baldovı, the sites withhigher eutrophication had higher macroinverte-brate densities mainly due to higher chironomiddensities. In the brackish sites the highest densityis also related to the highest chironomid propor-tion. Predators were more important in freshwa-ter and less eutrophic sites (Fig. 1c).

In all samples and sites, chironomids were thedominant group. The total dipteran larvae com-prised between 60 and 100% of the total indivi-duals found (Fig. 2). Within dipterans all indivi-duals belonged to Chironomidae, except in Xe-resa where a few Ceratopogonidae (0.5% relati-ve abundance) were recorded. In freshwaters si-tes, non dipteran insects and oligochaetes werealso important, the former being more abundantin Cap de Terme and Baldovı, and the latter in

Macroinvertebrates assemblages on reed beds, in Mediterranean shallow lakes 243

Figure 1. (a) Densities of macroinvertebrates associated to reed stems (Phragmites australis) (box-plots) and planktonicchlorophyll-a (dots and dash line), (b) Relative abundance of chironomids (%) and (c) Relative abundance of predators (%) insix Mediterranean water bodies ordered with decreasing trophic status. White plots: freshwater sites, striped ones: brackish sites.Box-plots with mean = dashed line, median= solid line, upper limit 75th percentile and lower limit 25th percentile, n = 4 samplesper site with five stems per sample. Sites are ordered with increasing trophic status, from eutrophic (on the left) to oligotrophicfor each group of sites. (a) Densidades de macroinvertebrados asociados al carrizo (Phragmites australis) (box-plots) y clorofila-aplanctonica (puntos y lınea discontinua), (b) Abundancia relativa de quironomidos (%) y (c) Abundancia relativa de macroinverte-brados depredadores (%) en seis lagunas mediterraneas separadas en dos grupos, aguas dulces en blanco y salobres en rayado, ysituadas en orden decreciente de eutrofia dentro de cada grupo. Box-plots representan: la media= lınea discontinua, la mediana =lınea continua, los percentiles 25 y 75 = lımites de la caja; 4 muestras por laguna con 5 canas por muestra. ALB, Albufera; TER,Cap de Terme; XER, Xeresa; BAL, Baldovı; HON, Hondo and CBN, Cabanes.

Albufera. In brackish sites, the second group inabundance was crustacea. Nevertheless, none ofthese groups reached relative densities over 20%.The freshwater spring pool Baldovı shares cha-racteristic taxa with both types of waters, due toits high chlorine levels and marine proximity.

Overall, forty-one taxa were identified (Table2 and Table 3). Species richness per site rangedfrom 6 to 15, being higher in the permanent andnot hypertrophic freshwater sites; Cap de Ter-me and Baldovı with 15 and 14 taxa, respecti-vely, were the highest. The highest diversity va-lues were also recorded on these sites (Fig. 3).

A PCA analysis applied to the proportionsof macroinvertebrate groups distinguished three

Figure 2. Relative abundance (%) of main macroinvertebrategroups recorded in P. australis stems in sixMediterranean waterbodies. Freshwater sites in white bars and saline sites in stripedbars. Abundancia relativa (%) de los principales grupos de ma-croinvertebrados recolectados en carrizos (P. australis) en seislagunas mediterraneas. Histogramas blancos: lagunas de aguadulce; rayados: salobres.

244 Sahuquillo et al.

Table 2. Macroinvertebrate taxa associated to reed stems in six Mediterranean water bodies in 28 June-3 July 2001. The meanabundance (ind/m2 stem surface area) calculated over four samples per site (5 stems per sample). Taxones de macroinvertebradosasociados al carrizo en seis lagunas Mediterraneas. Se indica la abundancia media (ind/m2 de superficie de cana) de cuatro muestraspor sitio (5 canas por muestra).

Albufera Cap Terme Xeresa Baldovı Hondo Cabanes

Hydra 9Microturbellarians 26

OLIGOCHAETAChaetogaster limnaei Baer 96Chaetogaster sp. 524Nais sp. 668 91 15 17Pristina sp. 16Aelosoma sp. 20

GASTROPODAPhysella acuta Draparnaud 113 163Melanopsis tricarinata Bruguiere 115Theodoxus fluviatilis (Linnaeus) 13Ferrissia wautieri Mirolli 96Semisalsa sp. 29Valvata piscinalis Muller 8Planorbidae 5

CRUSTACEAGammarus aequicaudaMartynov 215 5Echinogammarus pacaudi Hubault & Ruffo 3Lekanesphaera hookeri Leach 75 165 328Dugastella valentina Ferrer Galdiano 2Palaemonetes zariquieyi Sollaud 4Heterotanais oerstedi Kroyer 3

HYDRACHNIDA 526 232

ODONATAIschnura elegans (Van der Linden) 1098Other Coenagrionidae 37 12Sympetrum fonscolombei (Selys) 51Other Anisoptera 3

EFEMEROPTERACaenis luctuosa (Burmeister) 284Cloeon dipterum (Linnaeus) 139

HETEROPTERAPlea minutissima Leach 22Micronecta sp. 10 4

LEPIDOPTERALepidoptera 2 4

TRICHOPTERAEcnomus sp. 804Other Trichoptera 27

CERATOPOGONIDAE 39

ORTHOCLADIINAE 24680 5962 8147 706 6 2602

CHIRONOMINAE 1083 3171 102 2 2524 2729

TANYPODINAE 4 349

Macroinvertebrates assemblages on reed beds, in Mediterranean shallow lakes 245

Figure 3. (a) Number of species for total macroinvertebrate community and for Chironomidae and (b) H′ diversity of total macro-invertebrate community and of Chironomidae associated to reed stems in six Mediterranean water bodies. Histograms represent meanof two samples per site. In (a) black = Chironomidae and white or striped = other macroinvertebrates. Striped histograms correspondto brackish and unstriped to freshwater localities. (a) Numero total de especies de macroinvertebrados y (b) de quironomidos y (b)H′ diversidad del total de macroinvertebrados y de la familia de los Chironomidae encontrados en el carrizo en seis lagunas medi-terraneas. Los histogramas representan la media de dos muestras por laguna. En a) el fondo negro indica quironomidos y en blancoo rayado otros macroinvertebrados. Histogramas blancos: lagunas de agua dulce; rayados: salobres.

kinds of environment: freshwater, brackish wa-ters and the spring pool (Fig. 4). The first twoaxes explained 63% of the variation (38% and25% first and second axes respectively). Themain variability is due to salinity. Brackishwater bodies are positioned on the right sideas well as Chironominae, crustaceans andgastropods, whereas Orthocladiinae togetherwith big insects (Ephemeroptera, Odonata) andHydracharida, more characteristic of freshwaterbodies, are placed on the opposite side. However,Albufera is placed in the centre and the springpool Baldovı is positioned on the positiveside because of its abundant crustacean fauna.The second axis is again related to salinityand separates mainly the brackish water bodies

(with higher negative scores) from Baldovı (withhigher positive scores). Chironominae and Or-thocladiinae are also on the opposite side of thesecond axis, but Oligochaeta and Gastropodahave high positive values on this axis.

Chironomid assemblages are of special inter-est because they are the most abundant group inthis habitat and, in the present study, they had dis-tinct species compositions characterizing the dif-ferent sites (Table 3). Orthocladiinae dominatedin freshwater sites while Chironominae were do-minant in the brackish water ones, but belongingto the Chironomini tribe in Hondo, and to Tani-tarsini in Cabanes. In permanent freshwater bo-dies Cricotopus sylvestris was dominant and wasnot present in any of the brackish water sites.

246 Sahuquillo et al.

a) b)

Figure 4. Principal components analysis (PCA) applied to the proportions of different groups of macroinvertebrates collected onreed stems in six Mediterranean water bodies (4 samples per water body. (a) Loadings of macroinvertebrate taxa and (b) sample-scoresgrouped by water-bodies are represented in the space of the first two principal components. Analisis de componentes principales(PCA) aplicado a las proporciones de grupos de macroinvertebrados asociados al carrizo en seis lagunas mediterraneas (4 muestraspor sitio). Representacion de los dos primeros ejes para los resultados de los taxones de (a) macroinvertebrados y (b) muestrasagrupadas por lagunas.

However, it was not observed in the less eutro-phic and semipermanent Xeresa site where Cri-cotopus flavocinctus dominated, the latter specieswas also important in Cabanes, the more oligo-trophic of the brackish water sites. The accom-

panying secondary species were different in eachof the studied water bodies (Table 3). The DCA,applied to the relative abundances of chironomidspecies, indicated that salinity was the most im-portant feature responsible for chironomid spe-

Table 3. Chironomid species associated to reed stems in six Mediterranean water bodies sampled in 28 June-3 July 2001. The meanabundance (ind/m2 stem surface area) calculated over two samples per site (5 stems per sample). Feeding mechanism (P = predator;CG = collector-gatherer) and dominant food resource are also indicated after gut content examination. Especies de quironomidosasociadas al carrizo en seis lagunas mediterraneas muestreadas entre el 28 de junio y el 3 de julio de 2001. Se indica la abundanciamedia (ind/m2 de superficie de cana) de dos muestras por laguna (5 canas por muestra), ası como los mecanismos de alimentacion(P = depredator; CG = recolector-ramoneador) y la principal fuente de alimento tras la observacion directa del contenido intestinal.

AlbuferaCapTerme

Xeresa Baldovı Hondo CabanesFeeding

mechanism(dominant food)

TANYPODINAEAblabesmyia monilis (Linnaeus) 508 P (C. flavocinctus)

ORTHOCLADIINAECorynoneura 6 CG (diatoms)Cricotopus flavocinctus (Kieffer) 214 13381 5 4707 CG (diatoms)Cricotopus (Isocladius) sylvestris (Fabricius) 19271 8247 1169 CG (detritus)

CHIRONOMINIChironomus gr. luridus 4992 CG (detritus)Chironomus/Einfeldia 3760 CG (diatoms)Dicrotendipes pallidicornis (Goetghebuer) 2058 771 CG (diatoms)Parachironomus 7 3002 5 P

TANYTARSINIParatanytarsus 7 CG (detritus)Tanytarsus 18 14 4916 CG (diatoms)

Macroinvertebrates assemblages on reed beds, in Mediterranean shallow lakes 247

Figure 5. Detrended correspondence analysis (DCA) appliedto the proportions of chironomid species associated to reed sam-ples in sixMediterranean water bodies. Byplot of axis 1 and 2 ofspecies and samples (two per water body). Analisis de corres-pondencias Detrended (DCA) de las proporciones de especiesde quironomidos en el carrizo en seis lagunas mediterraneas(dos muestras por sitio). Representacion de casos y variablessobre los ejes 1 y 2. Water bodies (Lagunas): ALB, Albufera;TER, Cap de Terme; XER, Xeresa; BAL, Baldovı; HON, Hon-do and CBN, Cabanes. Species (Especies): Abla mon: Abla-besmyia monilis; Cric fla: Cricotopus flavocinctus; Cric syl:Cricotopus (Isocladius) sylvestris; Chirglur: Chironomus gr. lu-ridus; Chir-Ein: Chironomus/Einfeldia; Dicr pal: Dicrotendipespallidicornis; Paracind: Parachironomus; Paratind: Paratanytar-sus; Tanyind: Tanytarsus.

cies distribution, whereas trophic state had a lessapparent role (Fig. 5). Tanytarsini and Einfeldiawere placed on the positive extreme of the firstaxis with the brackish water sites, whereas Crico-topus sylvestris and Chironomus gr. luridus wereplaced near the origin where the permanent fresh-water sites also came grouped together. The se-cond axis of DCA separates Xeresa with a morefluctuating water level (which can dry up in someyears) from the rest of the sites, thus indicatingthe importance of hydrological regime.

DISCUSSION

The sampling method applied proved suitable tostudy animal composition attached to reed stems.However, some species inhabiting vegetationbeds are good swimmers and probably escapedfrom this trap, therefore their presence could ha-ve been underestimated. The macroinvertebratecommunity associated with plants was studied

using a hand net in the same sites the year before(Sahuquillo et al., 2007). The number of taxarecorded was higher using the hand net amongemergent plants, mainly reed stems, as well assubmerged ones (54 taxa), rather than with themethod used in the present study. Moreover,in the present study more than 90% of theindividuals corresponded to chironomids; whilein hand net samples this group represented about40%. The difference in composition in the pre-sent study was mainly due to the scarce presenceof big gastropod taxa and good swimmers, suchas crustaceans or coleopterans. Several studiesin small ponds support that reed beds presentsimilar macroinvertebrate assemblages and spe-cies richness as other plant beds (Dvorak & Best,1982; Van de Meutter et al., 2005), althoughrelative abundance could differ. Garcıa-Criado &Trigal (2005) found a higher relative abundanceof naidid oligochaetes and Orthocladiinae inemergent macrophytes. The reduced number ofspecies recorded in our study could be relatedto the small sample size of the collector, whichdoes not collect the surrounding free water.The most delicate part of the method is themanipulation when cutting the aerial part of thestem at the surface of the water, as some species,like big gastropods, can fall down or swim away.In this respect the precaution taken by Oertlyet al. (1995), with a comparable method forTypha latifolia, was to cut this aerial part the daybefore sampling (or some time before) so thatswimming species could resettle. Taking thismeasure should improve the method. Howeverthe aim of this study was not to describe thewhole lake’s macroinvertebrate community butquantify the populations attached to reed stems.

Despite the differences in community com-position among sites, the density of individualsresponded to eutrophication with the number ofindividuals increasing in line with trophic sta-te, as expected, but this was only clear in thefreshwater sites. On comparing lakes of diffe-rent trophic levels, most studies found a den-sity increase with trophic status (Kornijow, 1989,Pieczynska et al., 1999). Moreover, Pieczynskaet al. (1999) also pointed out a chironomid in-crease in line with macroinvertebrate density in-

248 Sahuquillo et al.

crease. In our results, in freshwater permanent si-tes (ALB, TER, BAL), trophic state also showeda clear direct relationship with both density andrelative abundance of chironomide. However, inthe two brackish sites this relationship wasnot observed. In this case the communitycomposition was quite different: Hondo wasdominated by the larger Chironomus/Einfeldiawhile in Cabanes the most frequent chironomidwere minute individuals of C. flavocintus.Therefore total biomass was indeed lower inthe oligotrophic site. The increase in eutrophi-cation had repercussions firstly on an increasein density, mainly due to chironomids andoligochaetes that feed mostly on algae or/anddetritus, and secondly, in a decrease in thenumber of taxa with loss of predators. In Capde Terme, despite high chlorophyll-a concen-trations, an important biomass of submergedmacrophytes is still maintained, which supportsa wide range of taxa including many predators(Zygoptera and Trichoptera being abundant).With respect to Oligochaeta, commonly usedfor pollution surveillance (Learner et al, 1978),a finer taxonomic level could improve theassessment of the ecological status, since someopportunistic species may rapidly increasein abundance over a relatively short periodof time, like Nais does, and are abundantin hypertrophic sites (like Albufera), whileother species living in oligotrophic sites mayhave a more complex functional relationship,such as Chaetogaster limnei, which livesin association with gastropods in Baldovı.

Each studied wetland had a quite distinctfauna. The sites differed broadly in environ-mental characteristics, such as salinity andhydrology and this was reflected in communitycomposition. In the present work, as in otherswhere study sites differed in salinity conditions(Arnolds & Ormerod, 1997; Bayo, 2001), themain ecological factor responsible for speciescomposition was salinity. This was reflected inthe PCA analysis, which clearly separated salinefrom freshwater sites on the first two axes.In permanent freshwater sites (Xeresa and Capde Terme) the community was composed ofswimming insect larvae (both predators, such as

Odonata, and collectors, like Ephemeroptera),whereas in Albufera and Baldovi Oligochaetaand Gastropods were more important.

Chironomids, the most abundant invertebratesin the samples, had distinct species assemblageswhich could be related to environmental qualityassessment. The species found are known asepiphytes and have widespread distribution(Armitage et al., 1995). C. sylvestris had beenconsidered tolerant to eutrophication (Verneauxet al., 2004) and chironomid stratigraphy studiesrelate this species with high total phosphoruslevels (Langdon et al., 2006). High densities ofthis species were found in the most eutrophicsites. In the hypertrophic Albufera C. sylvestriswas abundant altogether with Chironomus gr.luridus. In Cap de Terme, which is a eutrophiclake but still has a good macrophyte coverage,although C. sylvestris was abundant the chirono-mid community was more diverse. Meanwhile,in the oligotrophic permanent freshwater si-te, Baldovı, this species was also present butin much lower densities, indicating its broadtolerance to environmental conditions.

Xeresa, which is a semi-permanent freshwa-ter site of low trophic state, was differentia-ted by the presence of the tanypod Ablabes-myia monilis, a predator that has been obser-ved with remains of Cricotopus flavocinctus intheir guts. Moreover, in chironomid stratigraphystudies Langdon et al. (2006) related this spe-cies with low total phosphorus levels. Other stu-dies also found the Tanypodinae family to be agood quality indicator (Trigal, 2006), althoughsome deep benthonic species can tolerate lowdissolved oxygen levels, like Procladius (Prat &Rieradevall, 1995). In a mesocosm experimentalstudy performed in a close-by pond in the sa-me protected area, the addition of nutrients re-duced Tanypodinae abundance and enhanced Or-tocladiine development in a year of low waterlevel, while in a year with higher water levels,nutrient enhanced the growth of a Chironomuspopulation (Miracle et al., 2006).

The brackish sites differ in their water che-mistry and trophic state. Cabanes, near the seashore, may be considered as a truly brackish si-te, because its ion composition is characteris-

Macroinvertebrates assemblages on reed beds, in Mediterranean shallow lakes 249

tic of diluted marine water, while Hondo has ahigher sulphate/chloride ratio than Cabanes. Thesouthern most and warmest site, Hondo, receivesquite saline inflows from irrigation run off and,high evaporative rates in this water body, con-centrates these salts even more. They also differin trophic state, Cabanes being the more oligo-trophic. Although both sites share the same crus-tacean species, the chironomid community wasquite different as we commented previously.

The sampling method used proved suitable forobtaining quantitative and comparable data of in-vertebrate abundance in a variety of shallow la-kes, despite their clear or turbid state. Further-more, the macroinvertebrate densities can be re-lated to trophic status when the study sites aresimilar in both hydrology and salinity. The spe-cies composition of the macroinvertebrate com-munities in reed-beds reflects the different envi-ronmental features of the lakes studied, especia-lly chironomids, which are the main group cha-racteristic of this habitat. A study with the samemethodology (Canedo-Arguelles & Rieradevall,2007) in other coastal shallow lakes found thatChironomids in reed-beds were the most abun-dant taxa, averaging 79% of total density and re-presenting 44% of total taxa richness.

REFERENCES

ARMITAGE, P., P. S. CRANSTON & L. C. V. PIN-DER (eds.). 1995. The Chironomidae. Chapman &Hall. 572 pp.

ARNOLD, S. L. & S. J. ORMEROD. 1997. Aquaticmacroinvertebrates and environmental gradients inPhragmites reedswamps: implications for conser-vation. Aquatic Conservations: Marine and Fresh-water ecosystems, 7: 153-163.

BAYO, M., M. ORTEGA, P. LANGTON y J. J. CA-SAS. 2001. Evaluacion ecologica de humedales yla Directiva marco del Agua: sobre el valor indica-dor de las comunidades de dıpteros quironomidosen los humedales litorales de la provincia de Al-merıa. Actas V Simposio sobre el Agua en Anda-lucıa, Almerıa: 375-384.

BLANCO, S., S. ROMO, M. J. VILLENA & S.MARTINEZ. 2003. Fish communities and food

web interactions in six shallow Mediterranean la-kes. Hydrobiologia, 506-509: 473-480.

CANEDO-ARGUELLES, M. 2006. Macroinverte-brate communities associated to Phragmites aus-tralis in 11 shallow lakes of Llobregat’s river delta(Catalonia, Spain). Diploma de Estudios Avan-zados en Ecologıa Fundamental y Aplicada. Uni-versidad de Barcelona. 30 pp.

DVORAK, J. & E. P. H. BEST. 1982. Macro-invertebrate communities associated with the ma-crophytes of La Vechten: structural and functionalrelationships. Hydrobiologia, 95: 115-126.

GARCIA-CRIADO, F. & C. TRIGAL. 2005. Com-parison of several techniques for sampling macro-invertebrates in different habitats of a North Ibe-rian pond. Hydrobiologia, 545: 103-115.

HIRVENOJA, M. 1973. Revision der Gattung Crico-topus van derWulp und ihrer Verwandten (Diptera,Chironomidae). Ann. Zool. Fennici, 10: 1-363.

KORNIJOW, R. 1989. Macrofauna of elodeids oftwo lakes of different trophy. I. Relationships bet-ween plants and structure of fauna colonizingthem. Ekologia Polska, 37 (1-2): 31-38.

KORNIJOW, R. & T. KAIRESALO. 1994. A simpleapparatus for sampling epiphytic communities as-sociated with emergent macrophytes. Hydrobiolo-gia, 294: 141-143

LANGDON, P. G., Z. RUIZ, K. P. BRODERSEN &I. D. L. FOSTER. 2006. Assessing lake eutrophi-cation using chironomids: understanding the natu-re of community response in different lake types.Freshwat. Biol., 51: 562-577.

LANGTON, P. H. 1991. A Key to pupal exuviae ofWest Palaearctic Chironomidae. P. H. Langton(Ed.). Cambridgeshire. 386 pp.

LEARNER, M. A., G. LOCHHEAD & B. D. HU-GHES. 1978. A review of the biology of BritishNaididae with emphasis on the lotic environment.Freshwat. Biol., 8: 357-375.

MIRACLE, M. R., B. MOSS, E. VICENTE, S. RO-MO, J. RUEDA, E. BECARES, C. FERNANDEZ-ALAEZ, M. FERNANDEZ-ALAEZ, J. HIE-TALA, T. KAIRESALO, K. VAKKILAINEN,D. STEPHEN, L. A. HANSSON & M. GY-LLSTROM. 2006. Response of macroinvertebra-tes to experimental nutrient and fish additions inEuropean localities at different latitudes. Limneti-ca, 25 (1-2): 585-612.

MOSS, B., D. STEPHEN, C. ALVAREZ, E. BECA-RES, W. VAN DE BUND, S. E. COLLINGS,

250 Sahuquillo et al.

E. VAN DONK, E. DE EYTO, T. FELDMANN,C. FERNANDEZ-ALAEZ, M. FERNANDEZ-ALAEZ, R. J. M. FRANKEN, F. GARCIA–CRIADO, E. M. GROSS,M. GYLLSTROM, L. A.HANSSON, K. IRVINE, A. JARVALT, J. P. JEN-SEN, E. JEPPESEN, T. KAIRESALO, R. KOR-NIJOW, T. KRAUSE, H. KUNNAP, A. LAAS, E.LILL, B. LORENS, H. LUUP, M. R. MIRACLE,P. NOGES, T. NOGES, M. NYKANEN, I. OTT,W. PECZULA, E. PEETERS, G. PHILLIS, S. RO-MO, V. RUSSELL, J. SALUJOE, M. SCHEF-FER, K. SIEWERTSEN, H. SMAL, C. TESCH,H. TIMM, L. TUVIKENE, I. TONNO, T. VIRRO,E. VICENTE & D. WILSON. 2003. The determi-nation of ecological status in shallow lakes- a tes-ted system (ECOFRAME) for implementation ofthe European Water Framework Directive. Aqua-tic Conservation: Marine and Freshwater Ecosys-tems, 13: 507-549.

OERTLY, B. & J. B. LACHAVANNE. 1995. The ef-fects of shoot age on colonization of an emergentmacrophyte (Typha latifolia) by macroinvertebra-tes. Freshwat. Biol., 34: 421-431.

PIECZYNSKA, E., A. KOŁODZIEJCZYK & J. I.RYBAK. 1999. The response of littoral invertebra-tes to eutrophication-linked changes in plant com-munities. Hydrobiologia, 391: 9-21.

POQUET, J. M., F. MEZQUITA, J. RUEDA &M. R.MIRACLE. 2008. Loss of Ostracoda biodiversityin Western Mediterranean wetlands. Aquatic Con-servation: Mar. Freshw. Ecosyst., 18: 280-296.

PRAT, N. & M. RIERADEVALL. 1995. Life Cycleand Production of Chironomidae from the Kars-tic Lake Banyoles (NE Spain). Freshwat. Biol.,33: 511-524.

SAHUQUILLO, M., J. M. POQUET, J. RUEDA &M. R. MIRACLE. 2007. Macroinvertebrate com-

munities in sediment and plants in coastal Medite-rranean water bodies (Central Iberian Peninsula).Ann. Limnol.-Int. J. Lim., 43 (2): 117-130.

SCHMID, P. E. 1993. A key to the larval Chirono-midae and their instars from Austrian Danube re-gion streams and rivers with particular referenceto a numerical taxonomic approach. Part I. Diame-sinae, Prodiamesinae and Orthocladiinae. Wasserund Abwasser Supplement, 3/93: 1-514.

SCHNELL, O. A., M. RIERADEVALL, I. GRANA-DOS & O. HANSSEN. 1999. A Chironomid taxacoding system for use in ecological and palaeoe-cological databases. NIVA REPORTNo. 3710-97,Molar Project Manual, Annex B. 19 pp.

TRIGAL, C. 2006. Evaluacion del estado ecologicode las lagunas esteparias de la Depresion del Due-ro: ¿Son los macroinvertebrados buenos indicado-res?. Tesis Doctoral. Universidad de Leon. 242 pp.

VAN DE MEUTTER, F., R. STOKS & L. DEMEESTER. 2005. The effect of turbidity state andmicrohabitat on macroinvertebrate assemblages:a pilot study of six shallow lakes. Hydrobiologia,542: 379-390.

VARGA, I. 2001. Macroinvertebrates in reed litter.Int. Rev. Hydrobiol., 86: 573-583.

VERNEAUX, V., J. VERNEAUX, A. SCHMITT, C.LOVY, & J.C. LAMBERT. 2004. The Lake Bio-tic Index (LBI) : an applied method for assessingthe biological quality of lakes using macrobent-hos ; the Lake Chalain (French Jura) as an exam-ple. Ann. Limnol. Int. J. Lim., 40 (1): 1-9.

VICENTE, E. &M. R.MIRACLE. 1992. The coastallagoon Albufera de Valencia: An ecosystem understress. Limnetica, 8: 87-100.

WIEDERHOLM, T. 1983. Chironomidae of the Ho-larctic region. Keys and diagnoses. Part 1. Larvae.Ent. Scand. Suppl., 18: 1-457.

Limnetica, 27 (2): x-xx (2008)Limnetica, 27 (2): 251-258 (2008)c© Asociacion Iberica de Limnologıa, Madrid. Spain. ISSN: 0213-8409

Prevalence and intensity of black-spot disease in fish community froma subtropical stream (Santa Lucıa river basin, Uruguay)

Franco Teixeira-de Mello ∗,1,2 and Gabriela Eguren 1

1 Grupo de Investigacion en Ecotoxicologıa y Quımica Ambiental, Facultad de Ciencias, Igua 4225. CP. 11400,Universidad de la Republica, Montevideo, Uruguay.2 Grupo de Investigacion en Ecologıa Basica y Aplicada, Asociacion Civil I+D, Igua 4225, CP 11400, Montevi-deo, Uruguay.2

∗ Corresponding author: frantei@fcien.edu.uy2

Received: 26/11/06 Accepted: 2/6/08

ABSTRACT

Prevalence and intensity of black-spot disease in a fish community from a subtropical stream (Santa Lucıa river basin,Uruguay)

The objective of the present study was to identify fish species susceptible to black spot disease, the temporal variation inintensity prevalence and abundance of black spot, and its correlation with environmental and host parameters. In order to doso, several sampling campaigns were conducted bimonthly during one year, along the main course of an agricultural watershed.A total of twenty-seven fish species were collected, out of which only two, Astyanax aff. fasciatus (“mojarra”) andOligosarcusjenynsii (“dientudo”) presented black spot disease. In A. aff. fasciatus, 63% of black spot prevalence was registered and didnot show significant seasonal changes during the sampling period, nor did the abundance and intensity. Regarding possiblerelations between the disease and environmental and host parameters, a higher intensity of black spot infection was observedin the ventral region of the fish and allometric coefficient (WT − LS, relationship) values did not differ significantly betweenparasitized and non-parasitized fish groups.

Key words: Black-spot, metacercariae digeneans, Astyanax aff. fasciatus, subtropical stream, Santa Lucıa river basin.

RESUMEN

Prevalencia e intensidad de la enfermedad del punto negro en la comunidad de peces en un arroyo subtropical (cuenca delrıo Santa Lucıa, Uruguay)

El objetivo del presente estudio fue identificar las especies susceptibles a la enfermedad del punto negro, la variacion tempo-ral de su intensidad, prevalencia y abundancia; ası como su relacion con parametros ambientales y del hospedador. Para ello,se realizaron muestreos bimensualmente durante un ano, en el curso principal de una cuenca agrıcola suburbana. Fueron co-lectadas un total de veintisiete especies, de las cuales en solo dos, Astyanax aff. fasciatus (“mojarra”) y Oligosarcus jenynsii(“dientudo”), se observo la enfermedad del punto negro. En A. aff. fasciatus se registro una prevalencia de puntos negros del63% y la misma no presento diferencias significativas durante el perıodo de muestreo, al igual que su abundancia e inten-sidad. En cuanto a las posibles relaciones de la enfermedad con parametros ambientales y del hospedador, se observo unamayor intensidad de la infeccion en la region ventral de los peces y no se encontraron diferencias significativas en los valoresdel coeficiente alometrico (relacion WT − LS) entre los peces parasitados y no parasitados.

Palabras clave: Punto negro, metacercarias digenea, Astyanax aff. fasciatus, arroyo subtropical, cuenca del rıo Santa Lucıa.

252 Teixeira-de Mello & Eguren

INTRODUCTION

The black-spot disease, caused by encystmentof metacercariae of digenic trematodes, pro-duces a response easily recognizable in thefish for the accumulation of melanin aroundthe encysted metacercariae.

The life cycle of these parasites includes twointermediate hosts, the first stage (miriacida) in-fects snails, and the second intermediaries are fi-shes. When the cercariae stage comes out fromthe snails, it massively attacks the fish, enteringthrough the skin, muscles and fins of the fishesremaining there in the metacercariae stage untilthe infested host is eaten by other fish or water-fowl. In the digestive tract of the final host, theadult flukes reproduce. The eggs are released viaexcreta and pass to the water where the mira-ciadia emerges, penetrates the snails and starts anew cycle (Olsen, 1974; de Kinkelin et al., 1985).Freshwater digenean’s life cycle is affected bytemperature, seasonal changes in prevalence andabundance related to enhance cercarial emergen-ce with an increase in water temperature are re-ported (Chubb, 1979). To our knowledge, thereare no studies in Uruguay about the black-spot di-sease of freshwater fish; for this reason the prin-

cipal trematodes and its specificity in terms of in-termediate hosts (snails and fishes) are unknown.

Several digenic trematode species have beenreported as pathogenic for secondary interme-diary fish hosts. The infection may be lethal,especially during the first months of the fishlife cycle (Lucky, 1970). Both in aquacultu-re and natural populations, most parasite-fishhost systems present negative or neutral impactsof parasites on the host fish survival, growthand condition (Lyayman and Sadkovskaya, 1952;Harrison and Hadley, 1982; Baker and Bulow,1985; Sindermann, 1987). However, an increa-sed growth of parasitized fish have been infor-med in several studies, especially related to ces-todes parasites (Arnott et al. 2000; Museth, 2001;Loot et al., 2002). In digenean larvae infectedfishes, increased growth has only been recor-ded in the Cyprinidae Phoxinus phoxinus (Lin-naeus, 1758), with this effect disappearing inhigh infections (Ballabeni, 1994).

These parasites can promote different ef-fects in different fish species with consequencesat the population level. In this sense, the aimof the present study was to identify fish speciessusceptible to black spot disease, its temporalvariation regarding intensity, prevalence, abun-

Figure 1. Map of the study area, showed the four sampling sites of Canada del Dragon stream, Santa Lucıa river basin. Mapa delarea de estudio, se muestran los cuatros sitios de muestreo en el Arroyo Canada del Dragon, Cuenca del rıo Santa Lucıa.

Black-spot disease in fish community from subtropical stream 253

dance, and its correlation with environmentaland host parameters.

MATERIAL AND METHODS

The study was carried out in an agricultu-ral watershed located in the lower-basin ofSanta Lucıa River (Fig. 1), the most im-portant water resource for human consump-tion in Uruguay (DNH, 1999). The watershed,Canada del Dragon (34◦47′07′′ S, 56◦14′56′′Wto 34◦42′15′′ S, 56◦18′34′′W), covers 14.7 km2

and approximately 50% is covered by deciduousfruit-tree plantations. The main course is 13.7 kmlong, and according to Strahler (1986) criteria,was classified as a third order stream.

Six sampling campaigns were conducted fromJuly 2002 to June 2003, in four sites along thestream to ensure a high number of environments(riffles and pools, vegetated and non-vegetatedzones), in order to capture a representative num-ber of the species presented in the Canada delDragon stream. Fish were captured using elec-tro fishing (Sachs Elektrofischfanggerate GmbH,Type FEG 1000), and the sampling effort in eachsite was 15 electric pulses along 100 meters. Allfish were classified taxonomically and only theindividuals that presented black spot disease we-re transported to the laboratory. These fish we-re sacrificed with an overdose of anesthesia (so-lution of 2-Phenoxy-ethanol, 1 ml L−1), fixedin formalin (10%), preserved in alcohol (70%),

and stored at the Vertebrate Collection of Fa-cultad de Ciencias, Universidad de la Republi-ca (Uruguay) (Institutional code ZVC-P). Sam-ples of fish skin with black spot disease we-re wholly mounted and the metacercariae, sto-red in the Helminthological Collection from theabove mentioned Facultad de Ciencias.

Water parameters such as pH, dissolvedoxygen concentration (mg L−1), conductivity(μS cm−1) and temperature (◦ C) were measuredin situ. Water samples were collected and analy-zed for alkalinity (mg CaCo3 L−1), organic mat-ter and total suspended solids (mg L−1) (APHA,1995). The temporal variation of water physico-chemical parameters was analyzed using the non-parametric Kruskal-Wallis test (Zar, 1999).

In the species that presented black spot di-sease, prevalence (percentage of fish with blackspot disease), mean abundance (mean number ofblack spots considering all fish examined) andmean intensity (mean number of black spots perfish infected) were calculated following Bush etal., (1997). Temporal variation of prevalence wastested using Proportion tests, whereas changesin abundance and intensity were tested using thenon-parametric Kruskal-Wallis test (Zar, 1999).

The relationship between the number of para-sites and environment (water physico-chemistryparameters) and host parameters (fish total, stan-dard length, and weight) were evaluated throughthe non-parametric Spearman correlation test(Sokal and Rohlf, 1979). To determine infectionpreference, each organism was subdivided in four

Figure 2. Astyanax aff. fasciatus, picture showing the four regions of the body employed for the black-spot analysis. Arrow indicatesthe black spot. Astyanax aff. fasciatus, fotografıa mostrando las cuatro regiones del cuerpo empleadas para el analisis de los puntosnegros. La flecha indica un punto negro.

254 Teixeira-de Mello & Eguren

anatomical regions (Fig. 2) and the number ofblack spots in each section was compared usingWilcoxon matched-paired test (Zar, 1999).

We compared the growth type of parasitizedand non-parasitized fish (n = 107 and 57, res-pectively, for A. aff. fasciatus) estimating the pa-rameters of the potential relationship WT = a SbL,whereWT is total weight and SL is standard length(Ricker, 1973), by a linear regression after da-ta logarithmic transformation. We tested whetherfish growth was statistically different betweengroups, by t-tests with α = 0.05 (Mayrat, 1970;Ricker, 1973; Ricker, 1975).

RESULTS

A total of 2235 fish were collected, corres-ponding to 27 species and five Orders. Onlytwo species, of the Characidae family, presen-ted black spot disease: Astyanax aff. fasciatus(Cuvier, 1819) (“mojarra”) (n = 164) and Oli-gosarcus jenynsii (Gunther, 1864) (“dientudo”)(n = 2). One hundred and five individuals ofA. aff. fasciatus (prevalence = 63%, mean in-tensity = 7) and both individuals of O. jenynsiipresented black spot disease. Therefore, thetemporal variation and its correlation with envi-ronmental and host’s parameters were analyzedonly for A. aff. fasciatus.

The results for prevalence, mean abundance,and intensity of black spot disease in A. aff. fas-ciatus, is shown in Table 1. The Kruskal-Wallistest did not show significant differences for abun-dance temporal variations (H5,150 = 6.85; p = 0.23)

Table 1. Summary statistics of black spot disease in Astya-nax aff. fasciatus (prevalence, abundance and intensity) fromCanada del Dragon stream, SD, standard deviation, n, numberof individuals captured. Resumen estadıstico de la enfermedaddel punto negro en Astyanax aff. fasciatus (prevalencia, abun-dancia e intensidad) del arroyo Canada del Dragon, SD, desvıoestandar, n, numero de individuos capturados.

Prevalence Abundance IntensitySamplingDate

% infectedfish (n)

Mean± SD

RangeMean± SD

Range

July 66.6 (9) 6.1 ± 8.4 0-23 9.8 ± 8.9 1-23September 66.6 (18) 1.8 ± 2.7 0-8 2.8 ± 2.2 1-8November 63.1 (65) 3.6 ± 6.7 0-33 5.8 ± 7.7 1-33February 60.4 (48) 3.9 ± 8.3 0-52 6.6 ± 9.8 1-52April 77.8 (18) 10.5 ± 14.7 0-51 13.7 ± 15.5 1-51June 83.3 (6) 7.3 ± 8.1 0-21 8.8 ± 8.1 1-21

and intensity (H5,104 = 9.36; p = 0.10). The Pro-portion test did not show significant differences forprevalence’s temporal variation ( p > 0.05).

The physico-chemical parameters analyzed(Tables 2 and 3) showed a significant temporalvariation (Kruskal Wallis test, p < 0.05), but didnot show a significant correlation with black spotintensity (Spearman r, p < 0.05).

We found no significant correlation betweenintensity of the disease and host’s parameters(standard and total length, and weight). However,we found significant differences in the intensity ofblack spot between sections of the fish, particularlyventral (median = 2) and dorsal region (median = 1)(Wilcoxon test, t = 1561.5, p < 0.05).

The regressions for the two fish groups (para-sitized and non-parasitized) were:

1) Parasitized fishes:

LogWT = log−1.44(±0.03)+2.85(±0.04) log SL

Table 2. Summary of the mean pH, conductivity (C μ S cm−1), temperature (◦ C), and dissolved oxygen (DO mg l−1). SD = standarddeviation (mean and SD values based on 10 replicate collected from 4 sities, total n = 40), from Canada del Dragon stream. Resumende los valores medios para el pH, conductividad (C μ S cm−1), temperatura (◦ C), y oxıgeno disuelto (DO mg l−1). SD = desvıoestandar (valores de la media y SD basados en 10 replicas colectadas en 4 sitios, total n = 40), en el arroyo Canada del Dragon.

Sampling Date n pH C μ S cm−1 T◦ C DO mg l−1

mean SD mean SD mean SD mean SD

July 40 8.0 0.3 692.1 150.4 11.0 1.0 6.8 0.7September 40 7.6 0.2 593.0 79.5 16.6 1.1 6.1 0.4November 40 7.7 0.2 706.9 144.8 20.6 0.9 4.8 0.6February 40 7.3 0.3 457.2 77.7 24.4 0.6 4.0 0.5April 40 7.6 0.4 465.1 92.8 18.3 0.7 7.7 0.9June 40 7.5 0.3 423.2 147.8 10.9 0.4 8.9 0.8

Black-spot disease in fish community from subtropical stream 255

Table 3. Summary of the mean alkalinity (Alk, mg CaCo3 l−1), suspended total solids (STS, mg l−1), and suspended organic matter(SOM, mg l−1). SD = standard deviation (mean and SD values based on 3 replicate samples collected from 4 sities, total n = 12), fromCanada del Dragon stream. Resumen de los valores medios para la alcalinidad (Alk, mg CaCo3 l−1), solidos totales en suspension(STS, mg l−1), y materia organica en suspension (SOM, mg l−1). SD = desvıo estandar (valores de la media y SD basados en 3replicas colectadas en 4 sitios, total n = 12), en el arroyo Canada del Dragon.

Sampling Date n Alk (mg CaCo3 l−1) STS (mg l−1) SOM (mg l−1)mean SD mean SD mean SD

July 12 61.7 2.3 35.9 14.5 5.7 1.4September 12 182.1 50.5 12.7 7.7 5.4 2.5November 12 174.5 46.9 21.7 8.5 3.7 1.2February 12 138.4 34.9 15.7 5.4 2.7 1.6April 12 151.6 36.4 27.2 12.7 3.3 1.4June 12 118.6 54.75 16.4 11.4 3.7 3.6

2) Non-parasitized fishes:

LogWT = log−1.44(± 0.04)+2.84(± 0.06) log SL

Standard errors are shown in parenthesis.The allometric coefficient values did notdiffer significantly between the fish groups(Student test, p < 0.05).

DISCUSSION

From a total of twenty-seven species, only two,A. aff. fasciatus and O. jenynsii, were infectedwith digenic trematodes, which could indicate ahigh specificity for the secondary host (de Kin-kelin et al., 1985) and/or ecological characteris-tics of these fish that make them more susceptible(Ondrackova et al., 2002). However, other spe-cies present in Canada del Dragon stream withecological characteristics similar to A. aff. fas-ciatus, such as other small non-migratory Cha-raciformes, (“mojarras”) Cheirodon interruptus(Jenyns, 1842), Bryconamericus iheringii (Bou-lenger, 1887), and Hyphessobrycon luetkenii(Boulenger, 1887), did not present black spot di-sease. On the other hand, in other streams andcreeks of the Santa Lucıa River basin Arro-yo Colorado, Canada del Colorado, Canada delas Conchillas, Arroyo Las Piedras, and Canadadel Juncal, (unpublished data), we have detec-ted black spots in species collected in this studysuch as “madrecita” Cnesterodon decemmacula-tus (Jenyns, 1842), “overito” Jenynsia multiden-tata (Jenyns, 1842) (Cyprinodontiformes), and

“mojarra” B. iheringii (Characiformes). Thesedifferences can be due to the presence of diffe-rent parasites in the streams sampled, or to envi-ronmental characteristics causing differential in-fection susceptibility in fish species. The low cat-ches of O. jenynsii prevented the evaluation ofthe prevalence and intensity of the black spot di-sease. Considering that O. jenynsii is a commonspecies in our freshwater ecosystems (Ringueletet al., 1967), we believe that its susceptibility toblack spot disease should be further evaluated.

The prevalence, abundance, and intensity ofblack spot disease in A. aff. fasciatus, did notshow seasonal changes. On the contrary, Chubb(1979) found seasonal changes in prevalence andabundance, related to cercarial emergence withhigher water temperatures. Nevertheless, the sea-sonal changes in black spot abundance and preva-lence can be masked as old infections (Donges,1964). We found the greatest intensity of blackspots in the ventral region of the fish. The diffe-rential distribution of metacercariae, may be af-fected by behavior and habitat of hosts (snailsand fish) (Ondrackova et al., 2002). In this sense,Ringuelet (1975) described the Tetragonopteri-des, to which A. aff. fasciatus belongs, as a groupof little fish that inhabit pools and vegetated zo-nes. However, we have often captured A. aff. fas-ciatus, O. jenynsii and different snail species (i.e.Heleobia sp. and Pomacea sp.) in riffles and non-vegetated zones in the Canada del Dragon stream.We hypothesize that the presence of potentially-infected snails over bare sediment, can explainthe spatial pattern of the infection on the fish: the

256 Teixeira-de Mello & Eguren

parasites would be released from the bottom ofthe water column and reach the ventral region ofthe fish easier than other sections of the animals.

We found no differences in growth, betweeninfected and non infected fish, which indicatesa neutral effect of this parasite on Astyanax aff.fasciatus, at least over this variable. As it doesnot affect growth, it is also possible that it doesnot cause an effect at the population level in theWT − LS relationship.

Our results suggest that, a high susceptibilityof Astyanax aff. fasciatus to black spot disease,and/or a high specificity of the parasite occursin the Canada del Dragon stream. This is notreflected in several host’s parameters (total andstandard length and weight), which only showhigher infection in the ventral part of the fish.The infected fish did not seem affected, at leastconcerning their growth.

Future studies will be addressed to determinewhether the black spot disease detected in diffe-rent fish species from the Santa Lucia River basinis caused by the same species of Digenea. Thiswould help us to understand whether the patternsof infection in this basin are due to a species-specific relationship between the parasite and thehost, or due to environmental characteristics re-lated to water quality.

ACKNOWLEDGEMENTS

Authors wish to thank Alejandro D’Anatro, Es-teban Charbonier, Marcelo Loureiro, and LucıaBoccardi for technical assistance during data co-llection. Also, we would like to thank RodrigoPonce de Leon, Marcelo Loureiro, Carlos Igle-sias, and Mariana Meerhoff for the critical revi-sion of the manuscript. We are also grateful tothe anonymous referees which helped to improveearlier versions of this manuscript.

REFERENCES

APHA. 1995. Standard methods for the examinationof water and wastewater. APHA/AWWA/WPCF.Whashington, 1268 pp.

ARNOTT, S. A., I. BARBER & F. A. HUNTING-FORD. 2000. Parasite-associated growth enhance-ment in a fish-cestode system. Proceedings of theRoyal Society of London Series B, 267: 657-663.

BAKER, S. C. & F. J. BULOW. 1985. Effects ofBlack-Spot Disease on the Condition of Stonero-llers Campostoma anomalum. American MidlandNaturalist, 114: 198-199.

BALLABENI, P. 1994. Experimental differences inmortality patterns between European minnows,Phoxinus phoxinus, infected with sympatric orallopatric trematodes, Diplostomum phoxini. J.Fish Biol., 45: 257-267

BUSH, A. O., K. D. LAFFERTY, J. M. LOTZ & A.W. SHOSTAK. 1997. Parasitology meets ecologyon its own terms: Margolis et al. revisited. TheJournal of Parasitology, 83: 575-583.

CHUBB, J. C. 1979. Seasonal occurrence of hel-minths in freshwater fishes. Part II. Trematoda. Ad-vances in Parasitology, 17: 141-313.

De KINKELIN, P., C. MICHEL & P. GHITTINO.1985. Tratado de las enfermedades de los pe-ces. Editorial ACRIBIA S. A. Espana, Zaragosa.368 pp.

DNH, Division recursos hıdricos, 1998. Aprovecha-miento de los recursos hıdricos superficiales inven-tario nacional 1997-1998. Informe tecnico, 96 pp.

DONGES, J. 1964. The life cycle of Posthodiplosto-mum cuticola (v. Nordmann 1832) Dubois 1936(Trematoda, Diplostomatidae). Zeitschrift fur Pa-rasitenkunde, 24: 160-248.

HARRISON, E. J. & W. F. HADLEY. 1982. Possi-ble effects of black-spot disease on northern pike.Trans. Am. Fish. Soc., 111: 106-109.

LOOT, G., R. POULIN, S. LEK & J. F. GUEGAN.2002. The differential effects of Ligula intestinalis(L.) plerocercoids on host growth in three naturalpopulations of roach, Rutilus rutilus (L.). Ecologyof Freshwater Fish, 11: 168-177.

LUCKY, Z. 1970. Pathological changes with postho-diplostomosis of fish fry. Acta Veteterinaria Bru-nensis, [Suppl] 1: 51-66.

LYAYMAN, E. M. & O. D. SADKOVSKAYA. 1952.The black-spot disease of carps and treatment. TrNauch Issl Inst Prud Oz Rech Ryb Khoz USSR,8: 108-116.

MAYRAT, A. 1970. Allometrie et taxinomie. Revuede Statistique Appliquee, 18: 47-58.

MUSETH, J. 2001. Efects of Ligula intestinalis onhabitat use, predation risk and catchability in Eu-ropean minnows. J. Fish Biol., 59: 1 070-1 080.

Black-spot disease in fish community from subtropical stream 257

OLSEN, O. W. 1974. Animal parasites, their life cy-cles and ecology. University Park Press, Baltimore.562 pp.

ONDRACKOVA, M., P. JURAJDA &M. GELNAR.2002. The distribution of Posthodiplostomum cu-ticola metacercariae in young-of-the-year cyprinidfishes. J. Fish Biol., 60: 1 355-1 357.

RICKER, W. E. 1973. Linear regressions in fisheryresearch. Journal of the Fisheries Research Boardof Canada, 30: 409-434.

RICKER, W. E. 1975. Computation and interpreta-tion of biological statistics of fish populations. Bu-lletin of the Fisheries Research Board of Canada,191: 1-382.

RINGUELET, R. A., R. H. ARAMBURU & A. AR-MONIA. 1967. Los peces argentinos de agua dul-ce. Comision de Investigacion Cientıfica, BuenosAires. La Plata. 602 pp.

SINDERMANN, C. J. 1987. Effects of parasites onfish populations: practical considerations. Interna-tional Journal of Parasitology, 17: 371-382.

SOKAL, R. R. & F. J. ROHLF. 1979. Biometrıa:Principios y metodos estadısticos en investigacionbiologica. Editorial H. Blume, Madrid. 832 pp.

STRAHLER, A. N. 1986. Geografıa fısica. Ed.Omega, S.A., Barcelona. 767 pp.

ZAR, J. H. 1999. Bioestadistical Analysis. 3th edi-tion, New Jersey: Prentice Hall Inc. 663 pp.

Limnetica, 27 (2): x-xx (2008)Limnetica, 27 (2): 259-272 (2008)c© Asociacion Iberica de Limnologıa, Madrid. Spain. ISSN: 0213-8409

Seasonal changes of benthic communities in a temporary stream ofIbiza (Balearic Islands)

Liliana Garcıa ∗, Cristina Delgado and Isabel Pardo

Departamento de Ecologıa y Biologıa Animal, Universidad de Vigo, E-36330 Vigo, Espana2

∗ Corresponding author: lilizar@uvigo.es2

Received: 17/1/08 Accepted: 17/6/08

ABSTRACT

Seasonal changes in benthic communities in a temporary stream of Ibiza (Balearic Islands)

Seasonal changes in benthic communities (diatoms and invertebrates) in a temporary stream of the Ibiza island (Balearic Is-lands, Spain) were studied. The physico-chemical parameters, diatom quality indices and some invertebrate metrics were usedto describe and identify the observed temporal changes in benthic communities. A total of 43 diatom taxa and 51 invertebratetaxa were identified. Only 4 diatom species appeared in all the samples: Achnanthidium minutissimum, Diploneis oblongella,Navicula veneta and Nitzschia inconspicua. The invertebrate community was dominated by the orders Diptera, Oligochaetaand Gastropoda along the sampling period. Changes in the physico-chemical parameters of the water and hydrological eventsdetermined the structure of the benthic communities in this temporary stream.

Key words: Diatoms, Ibiza, invertebrates, Mediterranean island, temporary streams.

RESUMEN

Cambios estacionales de las comunidades bentonicas en un rıo temporal de Ibiza (Islas Baleares)

Se han estudiado los cambios estacionales de las comunidades bentonicas (diatomeas e invertebrados) en un arroyo temporalde la isla de Ibiza (Islas Baleares, Espana). Los parametros fısico-quımicos, ındices diatomologicos de calidad y algunosmetricos de invertebrados se han utilizado para describir e identificar los cambios temporales que se observan en las comu-nidades bentonicas. Se han identificado un total de 43 taxones de diatomeas y 51 taxones de invertebrados. Unicamente 4especies aparecen en todas las muestras: Achnanthidium minutissimum, Diploneis oblongella, Navicula veneta y Nitzschiainconspicua. La comunidad de invertebrados estuvo dominada por los ordenes Diptera, Oligochaeta y Gastropoda a lo largodel perıodo de muestreo. Los cambios en los parametros fısico-quımicos del agua y los eventos hidrologicos determinaron laestructura de la comunidad bentonica en este rıo temporal.

Palabras clave: Diatomeas, Ibiza, invertebrados, isla mediterranea, rıos temporales.

INTRODUCTION

The Mediterranean climate of the Balearic Is-lands is characterized by irregular precipitationsthroughout the year. This includes heavy rainsthat generally take place during the autumn andspring months with the driest conditions prevai-ling in the summer (Pardo & Alvarez, 2007),

when temperature and light intensity are higher.In the Mediterranean streams as well as in othertemporary systems, the differences in rainfall in-duce a periodically and predictable seasonal se-quence of floods and droughts (Towns, 1985; Sa-bater & Armengol, 1986; Resh et al., 1988; Poff,1992; Flecker & Feifarek, 1994; Romanı & Sa-bater, 1997; Gasith & Resh, 1999; Bonada et al.,

260 Garcıa et al.

2000; Lake, 2000; Bravo et al., 2001; Lake, 2003;Morais et al., 2004). These involve changes in thephysico-chemical parameters and on the commu-nity structure because of the disrupting processes(Gasith & Resh, 1999; Lake, 2000).

The composition of benthic algal communi-ties can be explained by the variations in the mi-neral content of the water (Sabater & Sabater,1988; Sabater, 1989) and in response to the an-nual variation in the magnitude of temperature,light and herbivorism (Alvarez & Pardo, 2006),whereas invertebrate assemblages seem to bemainly influenced by the temporal and spatial va-riations in resources (Poff & Ward, 1989). Otheraspects, such as geomorphological factors (Stout,1981, 1982), substratum size (Death, 1996) andriparian vegetation could influence (Sabater etal., 1998) on the benthic communities.

In the last twenty years, several studies de-veloped in the Balearic Islands have related wa-ter chemistry to algal communities (Moya etal., 1991; Moya et al., 1993; Llobera & Fe-rriol, 1994) and invertebrate assemblages (Garcıa

Aviles, 1990). Even more integrative studies ha-ve focused on the stream mentioned above (abio-tic and biotic components) (Alvarez & Pardo,2005; Alvarez & Pardo, 2006; Pardo & Alvarez,2006) and carried out on the island of Majorca,while the island of Ibiza has received much lessattention (Margalef, 1951).

The objective of this study was to analyse theexisting seasonal changes in the composition andabundance of benthic diatoms and invertebratesin a temporary stream of the Ibiza island, as wellas to identify the influence of environmental fac-tors in such changes.

MATERIAL AND METHODS

Study area

The island of Ibiza is located in the western partof the Mediterranean Sea (Fig. 1) having an areaof 570 km2. It is the warmest island of the Balea-ric archipelago, with a semiarid climate. The ave-

Figure 1. The Cas Berris stream and location of the sampling site. Torrente de Cas Berris y localizacion de la estacion de muestreo.

Benthic communities in a temporary stream 261

rage temperature ranges between 16 to 19.6◦ C,the humidity varies between 15-95% and the pre-cipitation average is low, with a monthly annualranging from 3.4 to 116.6 mm during the studyyear. Geologically, the island is calcareous (li-mestone and conglomerates) and topographica-lly has two small mountain ranges. These run ina SE-NE direction where the highest mountainis Sa Talasa Puig (486.7 m a.s.l.). The local geo-morphology may play a major role in the hydro-logy dynamics and the fluvial system in the SantJosep basin (SW of the island), which has severaltemporary streams of reduced length.

One of the few temporary streams of the is-land with a significant water length period, is theCas Berris stream (UTM 352342 4308124) lo-cated close to the town of Sant Josep. Its wa-tershed area is 18.4 km2 and the altitude is ap-proximately 100 m a.s.l. Calcite-dolomite bedro-ck covers most of the Cas Berris stream bedwhich is dominated by accumulations of sand, li-me, stones and boulders along the stream. Ourarea is characterised by typical rural land use andthe vegetation is dominated by Pinus halepensis,Arundo donax, tall herbs and shrubs.

Sampling, water chemistry andhydromorphological characterisation

Samples were taken over the period of a water cy-cle: November (22/11/2005), March (04/03/2006)and May (14/05/2006). Sampling included acollection of biotic communities (epilithic diatomsand benthic invertebrates) and abiotic featuresin each data sampling. Field data sheets werecompleted with different aspects of the riparianadjacent areas, landuses andhuman impacts.

We selected a 100 m stretch, where water flo-wed over the sampling period for approximately7 months. Rainfall data was registered from ameteorological station (83730 (LEIB), Es Codo-la) and accumulated for 15 days before the da-te of every sampling. The values ranged between73.10 mm in November to 36.10 mm in Marchand 17.30 mm in May. The stream width was0.7 m and the maximum depth was 15 cm du-ring the studied period. Water samples for che-mical analysis were collected from running water

and environmental factors such as temperature,pH, dissolved oxygen, electric conductivity of thewater, and water flow were measured in situ withportable instruments calibrated in the field. Thetemperature and the oxygen were measured withan oxymeter “WTW 197”, the water conductivitywas measured at 25◦ C with an Orion Model 115,the pH with a Termo Orion 290+ and the waterflow was measured three times with a current me-ter Probe in one transect of the stream. Standardmethods for chemical water analysis were ca-rried out according to APHA protocols (APHA,1989) and comprised the following nutrients andions: calcium (Ca2+), magnesium (Mg2+), potas-sium (K+), sodium (Na+), nitrates (NO−

3 ), silica(SiO2), phosphates (PO3−

4 ) and sulphates (SO2−4 ).

Benthic samples of chlorophyll a (Chl a) weretaken from stones, stored on ice and frozen untilprocessed. In the laboratory, these samples werefiltered using the glass-fibber filters. Later, theywere ground in 90% acetone. Chlorophyll con-centration was determined by extraction during48 h at 4◦ C in the dark. After extraction, chlo-rophyll a was measured spectrophotometrically(Hitachi Model U-2001 UV/Visible Spectropho-tometer) and corrected for degradation productsusing the equations given by Lorenzen (1967).

Hydromorphological features (taken, in thefield, along the stretch and 500 meters upstream)were registered in the CARAVAGGIO softwa-re in May 2006. This software is a version ofthe River Habitat Survey (RHS) used to evalua-te the hydromorphological quality (Buffagni etal., 2002, 2004) being also able to derive in-formation on its local and hydrologic character.According to the results, our stream is compo-sed by diverse habitats (Habitat Quality Assess-ment, HQA = 37), the water flow is slow andvaries with riffles and pools (Lentic-lotic RiverDescriptor, LRD = 45) and the whole reach isgently modified by the anthropogenic effect (Ha-bitat Modification Score, HMS = 26). The ripa-rian vegetation cover is mainly due to the exis-tence of reeds, shrubs, grasses (50%) and co-nifers (20%) occupying the upper part of thebasin. Harvest, due to cleaning activities, isperformed every year in the banks to preventdebris dam accumulation along the stream.

262 Garcıa et al.

Community assemblage

Epilithic diatoms were collected from naturalsubstrates (stones) following the European norms(Kelly et al., 1998; European Committee forStandardization, 2004; AFNOR, 2003). The sam-ples were preserved with formaldehyde solu-tion (4%) immediately after collection. After-wards, samples were treated to obtain a sus-pension with the clean frustules. Organic mat-ter was eliminated with hydrogen peroxide anddiluted HCl was added to remove the calciumcarbonate (Renberg, 1990). Finally, after disti-lled water rinsing, permanent slides were moun-ted with Naphrax R©. Diatoms were identified tospecies level using light microscopes OlympusBX40; at least 400 valves were identified andcounted from each slide (Prygiel & Coste, 1993).The diatoms were identified at the lowest ta-xonomical level according to the following aut-hors: Coste (1982); Krammer & Lange-Bertalot(1985, 1986, 1988, 1991a, 1991b, 2000); Kram-mer (1997a, 1997b, 2002); Lange-Bertalot (1993,1999, 2001); Lange-Bertalot & Krammer (1989);Lange-Bertalot & Moser (1994).

Diatom abundance data were introduced inthe Omnidia v.4.2 software (Lecointe et al.,1993) that calculates different indices, in order toanalyse seasonal changes in water quality. Eachindex differs in the number of used species andthe sensitivity values of the taxa that have beenjudged after compiling available literature infor-mation (Prygiel & Coste, 1993; Van Dam et al.,1994). We studied four indices: Specific Pollu-tion Sensitivity Index: IPS (Coste, 1982); Biolo-gic Diatom Index: IBD (Lenoir & Coste, 1996);Trophic Diatom Index: TDI (Kelly & Witton,1995) and European Index: CEE (Descy & Coste,1989). We selected these indices due to their wi-despread use in Spain, Portugal and other Euro-pean countries. Final values of these indices we-re transformed in water, quality estimates rangingbetween 1 (worse quality) and 20 (best quality).After adjustment by linear relation, five qualitycategories, established by the WFD (EuropeanUnion, 2000), can be defined by the values of the-se indices: high (> 17), good (13-17), moderate(9-13), poor (5-9) and bad (< 5). The Shannon-

Wiener Diversity Index was calculated as the sumover all the species in a sample.

Invertebrates were collected following amulti-habitat procedure (adapted from EPA, Bar-bour et al., 1999) in which, 20 sampling units(with a kicknet of 500 μm), that correspond toan area of 2.5 m2, were sampled. Samples we-re preserved in plastic bags in the field with et-hanol (70%), carried to the laboratory and sto-red until their treatment. Then, samples were wa-shed under tap water in three different fractions(5 mm, 0.5mm and 0.1 mm) and specimens werecounted and identified under 40X magnification(Olympus U-TV1X) up to the lowest possibleidentification level (except for some Diptera, Oli-gochaeta and Hydrachnidia). Sub-sampling wasmade, when necessary, to obtain a representa-tive fraction of the total community (Wrona etal., 1982). Assignment of taxa to functional fee-ding groups followed the classification of Cum-mins and Merrit (1996) and Alvarez (2004). Thecomposition and abundance data of the inver-tebrate community were introduced in the AS-TERICS program (software v.3.01) to calcula-te several indices (i.e. richness, Shannon-WienerIndex, EPT taxa) which were used to examinechanges in invertebrate community structure.

Table 1. Seasonal values of the physico-chemical parame-ters measured in Cas Berris stream. Valores estacionales de losparametros fısico-quımicos medidos en el torrente de Cas Be-rris.

Physico-chemicalparameters 22/11/2005 04/03/2006 14/05/2006

Water Temperature (◦ C) 15.9 15.9 19.6

Dissolved oxygen (mg L−1) 9.2 9.2 8.4pH 7.9 7.6 7.9

Conductivity (μS cm−1) 2375.0 1875.0 2179.0

Water flow (L s−1) 1.1 18.6 1.17

Chl a (mg cm2 L−1) 0.16 5.30 2.21

Ca2+ (mg L−1) 199.40 117.10 118.40

K+ (mg L−1) 5.17 5.19 3.28

Mg2+ (mg L−1) 84.79 39.33 43.43

Na+ (mg L−1) 301.60 159.90 160.90

SO2−4 (mg L−1) 185.52 133.95 130.33

PO3−4 (mg L−1) 0.32 0.04 0.12

NO−3 (mg L−1) 16.33 19.33 25.77

S (mg L−1) 6.19 24.77 41.93

SiO2 (mg L−1) 4.30 6.91 6.85

Benthic communities in a temporary stream 263

Table 2. List and abundance percentage of diatoms taxa identified in the Cas Berris stream during the sampling period. One asterisk:new taxa for Ibiza island and two asterisk new taxa for Balearic diatom flora. Listado y porcentaje de abundancia de las diatomeasidentificadas en el torrente de Cas Berris durante el periodo de muestreo. Un asterisco: las nuevas aportaciones a la flora de la islade Ibiza y dos asteriscos las nuevas aportaciones a la flora de las Islas Baleares.

Code Taxa 22/11/2005 04/03/2006 14/05/2006

ATHE **Achnanthidium thermale Rabenhorst 0.00 9.63 12.72ADMI Achnanthidium minutissimum (Kutzing) Czarnecki 15.67 72.25 55.22ASP1 Achnanthidium sp. 0.00 1.83 1.53AMMO **Amphora montana Krasske 0.00 0.00 0.25ANOR *Amphora normanii Rabenhorst 2.21 0.00 0.00APED Amphora pediculus (Kutzing) Grunow 0.22 0.00 0.25AMPS Amphora species 0.00 0.00 0.25BVIT *Brachysira vitrea (Grunow) Ross in Hartley 0.00 0.00 1.53CPUL Caloneis pulchraMessikommer 0.22 0.00 0.00CALS Caloneis species 0.44 0.00 0.00CMIC Cymbella microcephala Grunow 0.00 0.69 4.83DKUE **Denticula kuetzingii Grunow var. kuetzingii 0.44 0.00 0.00DSUB **Denticula subtilis Grunow 0.66 0.00 0.00DTEN Denticula tenuis Kutzing 0.00 0.00 0.00DOBL Diploneis oblonguella (Naegelii) Cleve-Euler 1.99 1.15 2.29DOVA Diploneis ovalis (Hilse) Cleve 0.22 0.00 0.51FCRP **Fragilaria capucina Desm. var. rumpens 0.22 0.00 0.51FRAS Fragilaria species 0.00 0.00 1.78FUAC Fragilaria ulna var. acus (Kutzing) Lange-Bertalot 0.00 0.00 0.25GCLA **Gomphonema clavatum Ehrenberg 0.00 1.61 0.00GOMP **Gomphonema dichotomum Kutzing 0.00 2.52 13.23GGRA *Gomphonema gracile Ehrenberg 0.22 0.00 0.00GPUM **Gomphonema pumilum (Grunow) Reichar. & Lange-Bert. 1.10 0.23 0.00GROS *Gomphonema rosenstockianum Lange-Bert. & Reicha. 0.00 0.46 0.00NSIT *Grunowia tabellaria (Grunow) Rabenhorst 0.00 0.23 0.00LMTP **Luticola muticopsis (Van Heurck) D.G. Mann 0.22 0.00 0.00NCIN *Navicula cincta (Ehrenberg) Ralfs in Pritchard 9.93 0.23 0.00NCTO **Navicula cryptotenelloides Lange-Bertalot 0.00 0.00 1.02NCRY Navicula crytocephala Kutzing 0.44 0.00 0.00NAVI NAVICULA J.B.M. Bory de Sant Vincent 0.22 0.00 0.51NLAN Navicula lanceolata (Agardh) Ehrenberg 0.00 0.00 0.25NVEN **Navicula veneta Kutzing 16.11 0.23 1.78NIFR **Nitzschia frustulum (Kutzing) Grunow var. frustulum 1.99 0.23 0.00NINC **Nitzschia inconspicua Grunow 41.06 7.80 0.51NREC **Nitzschia recta Hantzsch in Rabenhorst 0.44 0.00 0.00NIVI *Nitzschia vitrea Norman 1.77 0.00 0.00PSCA **Pinnularia subcapitata Gregory 0.22 0.00 0.00PLFR **Planothidium frequent Lange-Bert.) Round & Bukh. 0.00 0.46 0.00SSTM **Sellaphora stroemii (Hustedt) Mann 0.00 0.00 0.76SOVI Surirella ovalis Brebisson 0.22 0.00 0.00TAPI Tryblionella apiculata Gregory 0.22 0.00 0.00TDEB **Tryblionella debilis Arnott ex O’Meara 3.53 0.00 0.00UBIC *Ulnaria biceps (F.T. Kutzing) Compere 0.00 0.46 9.41

A hierarchical clustering and an ordination bynon-metric multidimensional scaling (MDS) wasmade with the abundance data from both bio-tic communities. These analyses were carried outwith the software PRIMER 6, previous log (x+1)transformation of the abundances.

RESULTS

Physico-chemical and hydromorphological

The water chemistry and hydromorphologicalcharacteristics of the stream channel were as-

264 Garcıa et al.

Table 3. Quality and diversity indices of benthic diatom com-munity of Cas Berris stream in each sample. Indices de calidady diversidad de la comunidad de diatomeas bentonicas del to-rrente de Cas Berris en cada muestra.

Diatom Indices 22/11/2005 04/03/2006 14/05/2006

IPS 9.2 17.3 15.3

IBD 6.9 14.2 14.3

TDI 8.0 16.6 15.4

CEE 9.2 15.6 16.6

Shannon-Weaver Diversity 2.8 01.6 02.6

sessed during the three different seasons usedto characterise the Cas Berris torrent (Table 1).Water temperature ranged annually between15.95-19.60◦ C, indicating little variation duringthe year, with the highest value in May. Oxygenconcentration did not vary too much between sea-sons (8.37-9.24 mg L−1). The pH ranged between7.62 and 7.99 during the study period. Water con-ductivity was considered as a representative va-riable of the total ionic strength; it was high in allseasons but it had the highest value in November(2 375 μS cm−1). Nitrate values were also highbetween 16.33-25.77 mg L−1, which is related tothe groundwater input of the nutrients probablyfrom agricultural adjacent areas. Ca2+ is domi-nant in Balearic running waters because of thedissolution of the calcareous substrata and it ran-ged between 117.1-199.4 mg L−1. The chl a andthe silica had the lowest values in November.

Diatoms

A total of 43 taxa belonging to 19 genera, we-re identified over the year (Table 2). Of these ta-xa, 26 were first cited in the island of Ibiza and18 of them in the Balearic Islands, after revisionof the latest literature updated by Aboal et al.(2003). Only four species appeared in all seasons:Achnanthidium minutissimum, with percentagesover 15% in all samples, Diploneis oblongella,Navicula veneta and Nitzschia inconspicua. Thediatom community in Cas Berris stream was do-minated in November by Nitzschia inconspicua(41.06%) together with Achnanthidium minutis-simum (15.67%) and Navicula veneta (16.11%).Taxa as Amphora normanii (2.21%), Navicula

cincta (9.93%) and Tryblionella debilis (3.53%)also characterised the community in this season,but in lower percentages. In March, Achnanthi-dium minutissimum was the most abundant spe-cies (72.25%) with a strong decrease of Nitzs-chia inconspicua (7.8%) and the apparition ofAchnanthidium thermale (9.63%), and of Gom-phonema dichotomum (2.52%). The percentagesof Achnanthidium thermale (12.72%), Gompho-nema dichotomum (13.23%) and Ulnaria biceps(9.41%) increased towards May, slightly redu-cing the relative Achnanthidium minutissimumabundance (55.22%).

Diatom quality indices

The values of the diatom indices: IPS, IBD, CEE,TDI and the Shannon-Wiener diversity index cal-culated with the software Omnidia v.4.2 showedthat the values of the different quality indicesgreatly varied between the three sampling pe-riods, with values ranging from 6.9 to 17.3.

We observed that the values of all indices arelowest in November (Table 3), but in contrast, thediversity is the highest, with a value of 2.77. InMarch, the IPS and TDI had the highest valuesbut the diversity was the lowest (Table 3).

Invertebrates

A total of 51 invertebrate taxa were recordedin the Cas Berris torrent, although those thatcontributed less than 1% to the total abundancewere eliminated, resulting in a total of 21 taxa(Table4). The invertebrate community in the streamover the sampling periodwas dominated byDiptera(69.8% from the total, distributed between 5 taxa),Gastropoda (9.74%distributed between 4 taxa) andOligochaeta (8.6%).The rest of the communitywascomposed by Ephemeroptera, Ostracoda, Acari,Odonata, andTrichoptera taxa.

The invertebrate composition changed sea-sonally; Oligochaeta and Gastropoda were wellrepresented during November, when the lowestbenthic abundance occurred (Table 5). In Marchand May, the community was dominated by Dip-tera, mostly Chironomidae and species such asCloeon dipterum, Hydroptila sp. and Oxyethira

Benthic communities in a temporary stream 265

Table 4. List and percentage of total abundance of benthic invertebrates identified in the Cas Berris stream during the samplingperiod. The table only includes those taxa with a percentage higher than 1% of the total abundance. Listado y porcentaje de laabundancia total de los invertebrados bentonicos identificados en el torrente de Cas Berris durante el periodo de muestreo. La tablasolo incluye aquellos taxones que presentaron un porcentaje superior al 1% de la abundancia total.

ORDER TAXANAME Shortcode 22/11/2005 04/03/2006 14/05/2006

ACARI Hydrachnidia Gen. sp. HYDRGESP 0.00 0.00 3.07COLEOPTERA Dryops sp.Lv. DRYOPSSP 0.09 1.18 0.05DIPTERA Ceratopogoninae Gen. sp. CENAGEN 0.09 4.42 0.00

Muscidae Gen. sp. MUSCGEN 0.00 0.00 1.49Orthocladiinae Gen. sp. ORTINAEG 4.76 15.60 6.71Simulium sp.Lv. SIMULISP 0.00 19.03 0.19Tanypodinae Gen. sp. TANNAEGE 1.43 4.73 9.48Tanytarsini Gen. sp. TANINIGE 0.00 20.31 59.00

EPHEMEROPTERA Caenis luctuosa CAENLUCT 1.28 0.91 2.05Cloeon dipterum CLOEDIPT 0.00 0.00 3.55

GASTROPODA Gyraulus laevis GYRALAEV 6.90 0.59 0.00Gyraulus sp. GYRASP 0.00 0.00 1.49Lymnaea (Galba) truncatula GALBTRUN 5.56 14.71 0.00Physella acuta PHYSACUT 2.23 4.42 0.88Pseudamnicola spirata PSEUSP 1.90 0.00 0.00

NEMATODA Nematoda Gen. sp. NEMATOGE 1.90 0.00 0.74ODONATA Libellulidae Gen. sp. LIBEGEN 0.00 0.00 1.20OLIGOCHAETA Enchytraeidae Gen. sp. ENCHYGEN 68.75 11.24 0.74OSTRACODA Ostracoda Gen. sp. OSTRGEN 1.90 1.18 3.77TRICHOPTERA Hydroptila sp. HYTILASP 0.00 0.00 1.07

Oxyethira sp. OXYESP 0.00 0.00 1.95

sp., which only appeared in May (Table 5). Cae-nis luctuosa was more abundant during May, inparallel with the highest values of EPT index. Ta-xa richness increased from November (22 taxa)to March (25 taxa) and May (31 taxa). Higher di-versity (H′) was observed in March, although itwas similar between seasons (Table 5).

Total collectors were the most important tro-phic group over the studied period, although theproportions changed between seasons (Table 5).Collector-gatherers were best represented in No-vember while collector-filterers in March andMay. Predators increased towards summer andscrapers were more important in other sea-sons. Shredders had the smallest representation(3.08%) in the community.

The ordination analyses showed the distribu-tion of the benthic species along the year. Inthe plot we can differentiate three groups ofspecies with a 60% of similarity in their ta-xa composition (Fig. 2), representing the seaso-nal changes identified with the succession of thebenthic communities (diatoms and invertebrates)from November to March and May.

DISCUSSION

Although the data analysed in this study waslimited to one small temporary stream, the in-formation supplied in this article adds to thecurrent knowledge on the aquatic communitiesof the Balearic Islands, the hydromorphologi-cal and physico-chemical factors, and the ben-thic communities of the Ibiza Island.

Table 5. Seasonal values of some metrics calculated for thebenthic invertebrate community of Cas Berris stream in eachsample. Valores estacionales de algunos metricos calculadospara la comunidad de invertebrados bentonicos del torrente deCas Berris en cada muestra.

Indices 22/11/2005 04/03/2006 14/05/2006

Abundance 3363 21760 34388Richness 22 25 31Diversity (S-W Index) 1.36 2.14 1.72EPT (%) 5.33 8.91 20.69% Collectors (undetermined) 1.94 1.18 3.80% Collector-Filterers 0.00 39.65 59.64% Collector-Gatherers 77.11 28.10 13.44% Predators 2.64 9.56 16.58% Scrapers 17.98 19.73 5.58% Shredders 0.33 1.78 0.97

266 Garcıa et al.

Figure 2. Multidimensional scaling of the biotic abundances (diatoms and invertebrates) represented in the Cas Berris stream.Es-calamiento multidimensional de las abundancias de la biota (diatomeas e invertebrados) representadas en el torrente de Cas Berris.

There were major changes in the diatom compo-sition in the Cas Berris stream along the year, re-lated to small variations in the physico-chemicalconditions and temporal succession. In Novem-ber, the water chemistry was characterised withthe highest values of Ca2+, Na+, Mg2+, SO2−

4 ,and PO3−

4 and the diatom community was cha-racterised by a high abundance of Nitzschia in-conspicua, that appeared together with Achnan-thidium minutissimum, Navicula veneta, Navicu-la cincta, and Tryblionella debilis. Nitzschia in-conspicua is abundant in waters of medium tohigh electrolytic conditions and tolerates a highorganic loading. This species is tolerant to thepollution (Goma, 2004), which confers low va-lues to the quality indices in November, definedbetween the poor and moderate classes.

The values of some nutrients were reduced inMarch possibly due to the strong increase of dis-charge that could induce a high development ofAchnanthidium minutissimum sensu lato whichwas the dominant taxa. This species is one ofthe most frequent diatoms occurring in freshwa-

ter benthic samples (Krammer & Lange-Bertalot,1991), being considered an ubiquitous species(Van Dam et al., 1994) and a cosmopolitan pio-neer in disturbed environments (Passy & Bode,2004). This species dominated the community inMarch together with the presence of Nitzschia in-conspicua and Achnanthidium thermale, influen-cing the lowest values of diversity.

The species Achnanthidium thermale and itsvarieties is particularly abundant in the North ofAfrica. In Europe it is a frequent inhabitant of mi-neral and thermal sources (Hustedt, 1930, 1959;Lange-Bertalot & Krammer, 1989; Krammer &Lange-Bertalot, 1991) and alkaline waters withhigh conductivity and rich in Ca2+, Na+, K+ andSO2−

4 (Krammer & Lange-Bertalot, 1991; Coste& Ector, 2000). The presence of this species inCas Berris stream is uncommon because it didnot appear in the other Balearic Islands (Pardoet al., 2007), but it is possible that the high va-lues of water temperature, conductivity, Na+ andCa2+ that exist in Cas Berris stream favoured itspresence. Its abundance increased in May toge-

Benthic communities in a temporary stream 267

ther with Ulnaria biceps and Gomphonema di-chotomum; this season was characterised by ha-ving the highest NO−

3 levels and water temperatu-res, due to the high temperatures. The abundanceof A. minutissimum was reduced in May possi-bly because the community was in the process tostabilising, increasing the number of species andthe diversity. The values of the indices IPS, TDIand CEE had similar values in March and May.Their values ranged between 15.3 and 17.3; onlyCEE had values lower than 15, but according tothem, the status of the water quality in Cas Berrisstream was between good and high.

Particularly interesting are the 26 new diatomrecords found for the Ibiza flora and the 18 forthe Balearic Islands; this great amount of newcited taxa can be explained by the lack of im-portant taxonomical studies in the island. Seve-ral common taxa in the European flora have notbe previously cited in Ibiza due to a lack of stu-dies or because some of them are taxa recentlydescribed or separated from older complex spe-cies such as Brachysira vitrea,Grunowia tabella-ria, Sellaphora stroemii and Tryblionella debilis.In the future, we expect to explore the existenceof other new species after revision with electro-nic microscope, because, some species as Diplo-neis oblongella may vary towards: D. fontane-lla or D. separanda (Werum & Lange-Bertalot,2004) or in the case of Cymbella microcephala,can be different species of the genus Encyonopsisas: E. microcephala, E. minuta, E. krammerii orE. subminuta (Krammer, 1997b).

We have observed an increase in invertebratetaxa richness and a temporal replacement of speciesin the Cas Berris stream along the year, although intemporary systems, sudden increases or decreasesin the flow can break the benthic successionand, consequently, may imply changes in benthiccommunities (Lake, 2003). The time elapsed afterthe resumption of water played an important rolein the colonisation of invertebrates, as well as thepermanence of the water in temporary systems(Alvarez & Pardo, 2007) increasing the complexityin the richness of the community.

Aquatic invertebrates have adapted their lifecycles in response to environmental factors suchas temperature, food, habitat, photoperiod (Peran

et al., 1999), and discharge (Flecker & Feifa-rek, 1994; Lake, 2003). The invertebrate com-munity in the studied stream changed seasonal-ly. As expected after the autumn rainfall, thelowest abundance occurred in November. Pos-terior increases in periphyton biomass (measu-red as chlorophyll a) may have promoted thecolonisation by scrapers (molluscs). The availa-bility of food may determine the existence ofthe invertebrates along the water period. In thisstream, dominant functional feeding groups arecollectors indicating a dominance of fine orga-nic matter along the year, followed in representa-tion by scrapers revealing the importance of pe-riphyton as an alternative food source in autumnand spring for these systems.

Oligochaeta and Diptera are characteristic ta-xa in spring habitats of Majorca (Alvarez, 2004)and they appear in high percentages in ourstream; this may be due to their capacity to re-sist and recover from extreme conditions (La-ke, 2003). Cloeon dipterum is a common spe-cies in this study and in other temporary streamsof the Balearic Islands (Pardo et al., 2007) be-cause its preference to inhabit in pools andstreams close to being dry (Belfiore, 1983, Stu-demann et al., 1992). We also note the pre-sence of other common inhabitants of tempo-rary streams, like Caenis luctuosa with flexi-ble life cycles and highly tolerant to changingconditions (Peran et al., 1999).

We observed, for example, that some Gas-tropoda (Physella acuta and Lymnaea (Galba)truncatula) had a negative relation with dischar-ge (Table 1), indicating that their abundance isreduced in May. Both species are cosmopolitanand well-represented in Ibiza and in other tem-porary streams of the Balearic Islands (Pardo etal., 2007). Other taxa such as Libellulidae sp.,Cloeon dipterum, Hydroptila sp. and Oxyethi-ra sp have an opposite response, appearing onlyat the beginning of the dry period and increa-sing as a consequence the EPT index. In Marchand May the community was dominated by op-portunistic taxa, mostly Chironomidae whichare species well adapted with a short life cy-cle, and resistant to changes in water conditions(Williams, 1996, Langton & Casas, 1999).

268 Garcıa et al.

In general, temporary streams are systems withunique characteristic. In these systems the im-portant and predictable flow changes are themain disturbance determining biotic communi-ties, where sudden changes in water chemis-try and water flow may determine the bioticcommunity; although there are several abioticfactors that can influence diatom communities(Leira & Sabater, 2005) and invertebrate com-munities (Peran et al., 1999; Lake, 2003). Itseems that the succession of the diatom commu-nity is driven by flow thus influencing physico-chemical local changes and nutrient contents.Meanwhile, the invertebrate community seemsto be more influenced by the seasonal cons-traints in water availability and related lengthof the water period influencing life cycles com-pletion as well as other environmental factorssuch as temperature and food resources.

ACKNOWLEDGEMENTS

This paper complements some of the resultsobtained in the project relating to the appli-cation of the Water Framework Directive inthe Balearic Islands. The financial support forthis study has been facilitated by the BalearGovernment (Spain) and this also included thesupport of the University of Vigo (Spain). Weare grateful to Mar Dominguez for the help inthe chemical analysis of the water.

REFERENCES

ABOAL, M., M. ALVAREZ-COBELAS, J. CAM-BRA & L. ECTOR. 2003. Floristic list of the nonmarine diatoms (Bacillariophyceae) of Iberian Pe-ninsula, Balearic Islands and Canary Islands. Dia-tom Monographs, vol. 4. Scientific Books Koe-nigstein (Germany). 639 pp.

AFNOR. 2003. Norme francaise NF EN 13946. Qua-lite de l’eau -Guide pour l’echantillonnage en rou-tine et le pretraitement des diatomees benthiquesde rivieres. Association Francaise de Normalisa-tion, 18 pp.

ALVAREZ, M. 2004. Ecologıa de los rıos tempora-les de la isla de Mallorca. Ph.D. Thesis. Universityof Vigo. 186 pp.

ALVAREZ, M. & I. PARDO. 2005. Life historyand production of Agapetus quadratus (Trichop-tera: Glossosomatidae) in a temporary spring-fedstream. Freshwat. Biol., 50: 930-943.

ALVAREZ, M. & I. PARDO. 2006. Factors contro-lling epilithon biomass in a temporary, karsticstream: the interaction between substratum andgrazing. J. North Am. Benthol. Soc., 26(2): 207-220.

ALVAREZ, M. & I. PARDO. 2007. Do temporarystreams of Mediterranean islands have a distinctmacroinvertebrate community? The case of Major-ca. Fundamental and Applied Limnology. Archivfur Hydrobiology, 168(1): 55-70.

APHA. 1989. Standard methods for the examina-tion of the water and wastewater. 18th Ed. Ameri-can Public Health Association. Washington, D.C.874 pp.

BARBOUR, M. T., J. GERRITSEN, B. D. SNY-DER, & J. B. STRIBLING, 1999. Rapid Bioas-sessment Protocols for Use in Streams and Wadea-ble Rivers: Periphyton, Benthic Macroinvertebra-tes and Fish, Second edition. EPA 841-B-99-002.U.S. Environmental Protection Agency; Office ofWater; Washington, D. C. 339 pp.

BELFIORE, C. 1983. Guide per il riconoscimentodelle specie animali delle acque interne italiane24.Efemerotteri. Consiglio Nazionale delle Ricer-che. 113 pp.

BONADA, N., M. RIERADEVALL y N. PRAT.2000. Temporalidad y contaminacion como clavespara interpretar la biodiversidad de macroinverte-brados en un arroyo mediterraneo (Riera de SantCugat, Barcelona). Limnetica, 18: 81-90.

BRAVO, R., M. C. SORIGUER, N. VILLAR & J. A.HERNANDO. 2001. The dynamics of fish popula-tions in the Palancar stream, a small tributary of theriver Guadalquivir, Spain. Acta Oecologica, 22: 9-20.

BUFFAGNI A., S. ERBA,R.BALESTRINI,M.CAZ-ZOLA, J. L.KEMP,R. PAGNOTTA,R.ALBER,E.BIELLI, C. BELFIORE & D. HERING, 2002. De-finizione della qualita ecologica nei fiumi europeisulla base dei macroinvertebrati bentonici: il Pro-getto EUAQEM. Acta Biologica, 78(1): 181-193.

BUFFAGNI A., S. ERBA, D. ARMANINI, D. DEMARTINI & S. SOMARE. 2004. Aspetti idromor-fologici e carattere lenticolotico dei fiumi medi-

Benthic communities in a temporary stream 269

terranei: River Habitat Survey e descrittore LRD.In ‘Classificazione ecologica e carattere lentico-lotico in fiumi mediterranei’. Quaderni Istituto diRecerca sulle Acque, Roma 122: 41-63.

COSTE, M. 1982. Etude des methodes biologiquesd’appreciation quantitative de la qualite des eaux.Rapport Q.E. Lyon, Agence de l’Eau Rhone-Mediterannee-Corse-Cemagref, Lyon, 218 pp.

COSTE, M. et L. ECTOR. 2000. Diatomees invasi-ves exotiques ou rares en France: principales ob-servations effectuees au cours des dernieres decen-nies.SystematicsGeographyofPlants, 70: 373-400.

CUMMINS, W. K. & R. W. MERRITT, 1996. Eco-logy and Distribution of Aquatic Insects. In: Aqua-tic Insects of North American, 3rd ed. K. W. Cum-mins & R. W. Merritt (eds.): 74-86. Dubuque, Io-wa, USA.

DEATH, R. G. 1996. The effect of patch disturban-ce on stream invertebrate community structure:the influence of disturbance history. Oecologia,108: 567-576.

DESCY, J. P. & COSTE, M. 1989. Application d’unnouvel indice diatomique (indice CEE 88) auReseau National de Bassin Rhone-Mediterrannee-Corse. Rapport A.F.B. Rhone-Mediterrannee-Corse, 86 pp.

EUROPEAN COMMITTEE FOR STANDARIZA-TION. 2004. Water quality - Guidance standardfor the identification, enumeration and interpre-tation of benthic diatom samples from runningwaters. European Standard EN 14407. EuropeanCommittee for Standardization, Brussels, 12 pp.

EUROPEAN UNION. 2000.Directive 2000/60/EC ofthe European Parliament and of the Council of 23October 2000. Establishing a framework for com-munity action in the field of water policy. OfficialJournal of the European Communities, 327: 1-72.

FLECKER, A. S. & B. FEIFAREK. 1994. Disturban-ce and the temporal variability of invertebrate as-semblages in two Andean streams. Freshwat. Biol.,31: 131-142.

GARCIA-AVILES, J. 1990. Insectos acuaticos deBaleares (Odonata, Ephemeroptera, Heteroptera,Plecoptera y Coleoptera). Ph. D. Thesis. Univer-sity Complutense of Madrid. 690 pp.

GASITH, A. & V. H. RESH. 1999. Streams in Medi-terranean climate regions: Abiotic influences andbiotic responses to predictable seasonal events.Ann. Rev. Ecol. Syst., 30: 51-81.

GOMA, J. 2004. Les diatomees bentoniques de laTordera: diversitat I utilitzacio com a indicadors

de la qualitat biologica de l’aigua. Diputacion deBarcelona, 2004, pp. 37-45.

HUSTEDT, F. 1930. Die Kieselalgen Deutschlands,Osterreichs und der Schweiz. In: L. Rabenhorst’sKryptogamen-Flora von Deutschlands, Osterrei-chs und der Schweiz, 7(1). 920 pp.

HUSTEDT, F. 1959. Die Kieselalgen Deutschland,Osterreich und der Schweiz. In: L. RabenhorstsKryptogamen-Flora von Deutchland, Osterreichund der Schweiz, 7(2). 845 pp.

KELLY, M. G. & B. A. WHITTON. 1995. The Tro-phic Diatom Index: A new index for monitoring eu-trophication in rivers. J. Appl. Phycol., 7: 433-444.

KELLY, M. G., A. CAZAUBON, E. CORING, A.DELL’UOMO, L. ECTOR, B. GOLDSMITH, H.GUASCH, J. HURLIMANN, A. JARLMAN, B.KAWECKA, J. KWANDRANS, R. LAUGASTE,E. A. LINDSTROM, M. LEITAO, P. MARVAN,J. PADISAK, E. PIPP, J. PRYGIEL, E. ROTT, S.SABATER, H. VAN DAM & J. VIZINET. 1998.Recommendations for the routine sampling of dia-toms for water quality assessments in Europe. J,Appl. Phycol., 10: 215-224.

KRAMMER, K. 1997a. Die cymbelloiden Diato-meen. Eine Monographie der weltweit bekanntenTaxa. Allgemeines und Encyonema Part. Teil 1.In: Bibliotheca Diatomologica, vol. 36. J. Kram-mer (eds.). Berlin-Stuttgart. 382 pp.

KRAMMER, K. 1997b. Die cymbelloiden Diato-meen. Eine Monographie der weltweit bekanntenTaxa. Encyonema Part., Encyonopsis und Cym-bellopsis. Teil 2. In: Bibliotheca Diatomologi-ca, vol. 37. J. Krammer (eds.). Berlin- Stuttgart.469 pp.

KRAMMER, K. 2002. Cymbella. In: Diatoms of Eu-rope. Vol. 3. H. Lange-Bertalot (ed.). Koenigstein.A. R. G. Gantner Verlag K. G. 584 pp.

KRAMMER, K. & H. LANGE-BERTALOT. 1985.Naviculaceae. Neue und wenig bekannte Taxa,neue Kombinationen und Synonyme sowie Bemer-kungen zu einigen Gattungen. In: Bibliotheca Dia-tomologica, vol. 9. 230 pp.

KRAMMER, K. & H. LANGE-BERTALOT. 1986.Bacillariophyceae 1 Teil: Naviculaceae. In:Subwasserflora von Mitteleuropa. Band 2/1. Ettl,H., J. Gerloff, H. Heynig, & D. Mollenhauer (eds).Gustav Fischer Verlag, Stuttgart-Jena. 876 pp.

KRAMMER, K. & H. LANGE-BERTALOT. 1988.Bacillariophyceae 2 Teil: Bacillariaceae, Epithe-miaceae, Surirellaceae. In: Subwasserflora vonMitteleuropa. Band 2/2. Ettl, H., J. Gerloff, H.

270 Garcıa et al.

Heynig, & D. Mollenhauer (eds). Gustav FischerVerlag, Stuttgart-Jena. 596 pp

KRAMMER, K. & H. LANGE-BERTALOT. 1991a.Bacillariophyceae 3 Teil: Centrales, Fragilaria-ceae, Eunotiaceae. In: Subwasserflora von Mitte-leuropa. Band 2/3. Ettl, H., J. Gerloff, H. Heynig,& D. Mollenhauer (eds). Gustav Fischer Verlag,Stuttgart-Jena. 576 pp.

KRAMMER, K. & H. LANGE-BERTALOT. 1991b.Bacillariophyceae 4 Teil: Achnanthaceae. In:Subwasserflora von Mitteleuropa. Band 2/4. Ettl,H., J. Gerloff, H. Heynig, & D. Mollenhauer (eds).Gustav Fischer Verlag, Stuttgart-Jena. 437 pp.

KRAMMER, K. & H. LANGE-BERTALOT. 2000.2ndedition. Bacillariophyceae 3. Teil: Centrales,Fragilariaceae, Eunotiaceae. In: Subwasserfloravon Mitteleuropa. H. Ettl, J. Gerloff, H. Heynig& D. Mollenhauer (Ed.), Spektrum AkademischerVerlag, Heidelberg & Berlin, 2/3. 576 pp.

LAKE, P. S. 2000. Disturbance, patchiness, and di-versity in streams. J. North Am. Benthol. Soc.,16: 221-238.

LAKE, P. S. 2003. Ecological effects of perturbationby drought in flowing waters. Freshwat. Biol.,48: 1161-1172.

LANGE-BERTALOT, H. 1993. 85 neue Taxa unduber 100 weitere neu definierte Taxa erganzend zurSusswasserflora von Mitteleuropa. In: BibliothecaDiatomologica. Vol. 27. J. Krammer (eds.). Berlin-Stuttgart. 454 pp.

LANGE-BERTALOT, H. 1999. Neue Kombinatio-nen von Taxa aus Achnanthes (sensu lato). In:Iconographia Diatomologica. Lange-Bertalot, H.(ed.). 6: 278-291.

LANGE-BERTALOT, H. 2001. Navicula sensu stric-to. 10 Genera separated from Navicula sensu la-to. Frustulia. In: Diatoms of Europe. Vol. 2. H.Lange-Bertalot. (ed.). A. R. G. Gantner Verlag K.G. 526 pp.

LANGE-BERTALOT, H. und K. KRAMMER. 1989.ACHNANTHES eine Monographie der Gattung.Mit definition der Gattung Cocconeis und Na-chtragen zu den Naviculaceae. In: Bibliotheca Dia-tomologica, vol. 18. J. Krammer (eds.). Berlin-Stuttgart. Berlin. 393 pp.

LANGE-BERTALOT, H. & G. MOSER. 1994. Bra-chysira. Monographie der Gattung. In: BibliotecaDiatomologica. Vol. 29. J. Cramer. (ed.). Berlin-Stuttgart. 212 pp.

LANGTON, P. & J. CASAS. 1999.Changes in chiro-nomid assemblage composition in two Mediterra-nean mountain streams over a period of extreme hy-drological conditions. Hydrobiologia, 390: 37-49.

LECOINTE, C., M. COSTE &, J. PRYGIEL. 1993.“«OMNIDIA» software for taxonomy, calculationof diatom indices and inventories management”.Hydrobiologia, 269: 509-513.

LEIRA, M. & S. SABATER. 2005. Diatom assem-blages distribution in Catalan rivers, NE Spain, inrelation to chemical and physiographical factors.Water Research, 39(1): 73-82.

LENOIR, A. & M. COSTE. 1996. Development of apractical diatom index of overall water quality ap-plicable to the French National Water Board Net-work. In: Use of Algae for Monitoring Rivers II.B. A. Whitton & E. Rott (eds.): 29-43. UniversitatInnsbruck, Innsbruck.

LLOBERA, M. & A. FERRIOL. 1994. Seasonal fluc-tuations in the chemistry of limestone springs fromthe Tramuntana Mountain Range (Mallorca, Spain).Verh. Internat. Verein. Limnol., 25: 1445-1448.

LORENZEN, C. J. 1967. Determination of chlo-rophyll and phaeopigments: spectrophotometricequations. Limno. Oceanograph., 12: 342-346.

MARGALEF, R. 1951. Materiales para la hidrobio-logıa de la Isla de Ibiza. Publicaciones del Institutode Biologıa Aplicada, 8: 5-70.

MORAIS, M., P. PINTO, P. GUILHERME, J. RO-SADO& I. ANTUNES. 2004. Assessment of tem-porary streams: the robustness of metric and multi-metric indices under different hydrological condi-tions. Hydrobiologia, 516: 229-249.

MOYA, G., A. FERRIOL & M. LLOBERA. 1991.Chlorophyll contents in the freshwater vegetationof Serra de Tramuntana springs (Majorca, Spain).Verh. Internat. Verein. Limnol., 24: 2002-2006.

MOYA, G., BENNASAR, C. FRAU, L. GARCIA,M. GOMEZ & G. RAMON. 1993. Long termchanges (after twelve years) in the compositionof phytoplankton communities in the Gorg Blaureservoir (Majorca, Spain). Verh. Internat. Verein.Limnol., 25: 1257-1261.

PARDO, I. & M. ALVAREZ. 2006. Comparison ofresource and consumer dynamics inAtlantic andMe-diterranean streams. Limnetica, 25(1-2): 271-286.

PARDO, I., M. ALVAREZ, C. DELGADO, L.GARCIA, P. LUCENA. 2007: Implementacion dela DMA en Baleares: evaluacion de la calidadambiental de las masas de agua epicontinentalesutilizando indicadores e ındices biologicos. Tomo

Benthic communities in a temporary stream 271

I: Torrentes (Informe Tecnico). Universidad deVigo. 302 pp.

PASSY, S. I. & R. W. BODE. 2004. Diatom ModelAffinity (DMA), a new index for water quality as-sessment. Hydrobiologia, 524: 241-251.

PERAN, A., J. VELASCO & A. MILLAN. 1999. Li-fe cycle and secondary production of Caenis luc-tuosa (Ephemeroptera) in a semiarid stream (Sout-heast Spain). Hydrobiologia, 400: 187-194.

POFF, N. L. & J. V. WARD. 1989. Implications ofstreamflow variability and predictability for lo-tic community structure: a regional analysis ofstreamflow patterns. Can. J. Fish. Aquat. Sci.,46: 1 805-1 818.

POFF, N. L. 1992.Why disturbances can be predicta-ble: a perspective on the definition of disturbance instreams. J. North Am. Benthol. Soc., 11(1): 86-92.

PRYGIEL, J. & M. COSTE. 1993: Utilisation des in-dices diatomiques pour la mesure de la qualite deseaux du bassin Artois-Picardie: bilan et perspecti-ves. Annals of Limnology, 29(3-4): 255-267.

RENBERG, I. 1990. A procedure for preparing largesets of diatom slides from sediment cores. J. Pa-leolimnol., 4: 87-90.

RESH, V. H., A.V. BROWN, A. P. COVICH, M. E.GURTZ, H. W. LI, G. W.MINSHALL, S. R.REICE, A. L. SHELDON, J. B. WALLACE & R.WISSMAR. 1988. The role of disturbance in streamecology. J. North Am. Benthol. Soc., 7: 433-455.

ROMANI, A. M. & S. SABATER. 1997. Metabo-lism recovery of a stromatolitic biofilm afterdrought in a Mediterranean stream. Arch. Hydro-biol., 140: 261-271.

SABATER, S. 1989. Encrusting algal assemblagesin a Mediterranean river basin. Arch. Hydrobiol.,114: 555-573.

SABATER, F. & J. ARMENGOL. 1986. Chemicalcharacterizationof theTer river.Limnetica, 2: 75-84.

SABATER, S. & F. SABATER. 1988. Diatom as-semblages in the river Ter. Arch. Hydrobiolol.

111(3): 397-408.SABATER, S., A. BUTTURINI, I. MUNOZ, A. RO-

MANI, J. WRAY& F. SABATER. 1998. Effects ofremoval of riparian vegetation on algae and hetero-trophs in aMediterranean stream. Journal of Aqua-tic Ecosystem Stress and Recovery, 6: 129-140.

STOUT, R. J. 1981. How abiotic factors affect thedistribution of two species of tropical preda-ceous aquatic bugs (family: Naucoridae). Ecology,62: 1 170-1 178.

STOUT, R. J. 1982. Effects of a harsh environmenton the life history patterns of two species of tropi-cal aquatic Hemiptera (Family: Naucoridae). Eco-logy, 63: 75-83.

STUDEMANN, D., P. LANDOLT, M. SARTORI, D.HEFTI & I. TOMKA. 1992. Ephemeroptera. So-ciete entomologique Suisse. Switzerland. 175 pp.

TOWNS, D. R. 1985. Limnological characteristics ofa South Australian intermittent stream, Brown HillCreek. Australian Journal of Marine and Freshwa-ter Research, 36: 821-837.

VAN DAM, H., MERTENS, A. & J. SINKELDAM.1994. A coded checklist and ecological indica-tor values of freshwater diatoms from The Net-herlands. Netherlands Journal of Aquatic Ecology,28: 117-133.

WERUM, M. &H. LANGE-BERTALOT. 2004. Dia-toms in springs from central Europe and elsewhereunder the influence of hydrogeology and anthro-pogenic impacts. In: Iconografia Diatomologica,vol. 13. H. Lange-Bertalot (ed.). A. R. G. GarnerVerlag K.G. 417 pp.

WILLIAMS, D. D. 1996. Environmental constrainsin temporary freshwaters and their consequencesfor the insect fauna. J. North Am. Benthol. Soc.,15: 634-650.

WRONA, F. J., J. M. CULP & R. W. DAVIES. 1982.Macroinvertebrate subsampling: a simplified ap-paratus and approach. Can. J. Fish. Aquat. Sci.,39: 1 051-1 054.

Limnetica, 27 (2): x-xx (2008)Limnetica, 27 (2): 273-284 (2008)c© Asociacion Iberica de Limnologıa, Madrid. Spain. ISSN: 0213-8409

Identification of the main factors in structuring rotifer communityassemblages in ponds of Donana National Park using the amino acidcomposition of the species

Castor Guisande1,∗, Carlos Granado-Lorencio2, Julia Toja 2 and David Leon2

1 Edificio de Ciencias, Universidad de Vigo, Campus Lagoas-Marcosende, 36200 Vigo, Spain2 Departamento de Biologıa Vegetal y Ecologıa, Facultad de Biologıa, Universidad de Sevilla, 41080-Sevilla,Spain2

∗ Corresponding author: castor@uvigo.es2

Received: 26/11/07 Accepted: 31/3/08

ABSTRACT

Identification of the main factors in structuring rotifer community assemblages in the ponds of Donana National Parkusing the amino acid composition of the species

The use of the role of the species in the habitat (the niche), as an alternative to systematics for studying the processes thatdetermine which and how many species live in a specific habitat (community assembly), is an approach that has been limitedby the difficulties in the characterization of the niche. The aim of this study was to identify the determining factors in rotifers’assemblage in several ponds, using the amino acid composition (AAC) of the species as a fingerprint of the differential nicheusage. We found that species with a similar AAC and, hence, with a similar trophic niche, spatially co-exist, showing thattrophic-niche differentiation is not a main factor in structuring these lentic associations. The negative relationship betweenamino acid separation and spatial overlap among rotifer species can be considered as evidence that habitat filtering (abioticfactors) is the factor responsible for the assemblages.

Key words: Amino acids, zooplankton, niche, community.

RESUMEN

Identificacion de los factores mas importantes en la estructuracion de las asociaciones de rotıferos utilizando la composi-cion de aminoacidos de las especies

La utilizacion del papel de las especies en el habitat (el nicho), como alternativa a la aproximacion sistematica, en el estudiode los procesos que determinan cuales y cuantas especies pueden coexistir en un determinado espacio (ensamble de comuni-dad), se ha visto limitada por las dificultades que existen para caracterizar el nicho. El objetivo de este estudio es identificarlos factores determinantes de la asociacion de rotıferos en distintas lagunas, utilizando la composicion de aminoacidos (CAA)de las especies como un marcador del distinto uso del nicho. Encontramos que las especies con similar CAA, y por lo tanto,nicho trofico, coexisten espacialmente, lo que demuestra que la diferenciacion trofica no es un factor estructurador de estasasociaciones lenıticas. La relacion negativa entre la separacion de aminoacidos y el solapamiento especial entre las especiesde rotıferos se puede considerar como una evidencia de que el filtro ambiental (factores abioticos) es el factor responsable delas asociaciones. Los resultados muestran que la salinidad y conductividad son las variables mas importantes.

Palabras clave: Aminoacidos, zooplancton, nicho, comunidad.

274 Guisande et al.

INTRODUCTION

Many studies have been undertaken on the sub-ject of the processes that determine which andhow many rotifer species live in a specific habi-tat (community assembly). Predation (Gilbert &Williamson, 1978; Stemberger & Evans, 1984;Williamson & Butler, 1986; Green, 2001; Die-guez & Gilbert, 2002), interference from cla-docerans (Gilbert, 1988; MacIsaac & Gilbert,1989; Pace & Vaque, 1994; Arvola & Salonen,2001; Nandini et al., 2002), and food availa-bility (Stelzer, 2001; Duggan et al., 2002), aswell as other abiotic factors such as tempera-ture, salinity or acidification (Guisande & To-ja, 1988, Arnott & Vanni, 1993; Devetter, 1998),have often been cited as relevant factors indetermining rotifer assemblages.

The studies mentioned above are mainly ba-sed on systematics. A different approach is touse biochemical fingerprints that provide infor-mation about the adaptation of the species tothe habitat. The identification of common adap-tations of co-occurring species may, in turn,allow the identification of the main factors instructuring community assemblages.

The amino acid composition (AAC) of zoo-plankton species is species-specific (Guisande etal., 2002; 2003; Boechat & Adrian, 2005) andremains relatively constant despite different nu-tritional supplies (Guisande, Maneiro & Rivei-ro, 1999; Guisande et al., 2000; Helland et al.,2002; Boechat & Adrian, 2005). Moreover, pro-tein expression can be directly related to stress.Zooplankton species use proteins to adapt tochanging habitat conditions (Kimmel & Bradley,2001). This indicates that AAC may be a finger-print of the adaptation of the species to abioticconditions. Therefore, AAC may provide a natu-ral tag for the adaptation of each species to itshabitat; in other words, it provides informationabout the niche of the species.

The AAC of field populations of cladoce-rans and copepods has been studied (Guisan-de, Maneiro & Riveiro, 1999; Guisande et al.,2000; Guisande et al., 2002; 2003; Helland et al.,2002), but that of field rotifer populations has not.The aim of this study was to identify the main

factors in structuring the assembly of rotifers inponds, using the AAC of the species as an indi-cator of the differential use of the niche.

MATERIALS AND METHODS

Plankton collection

From the 2nd to the 4th of July and from the 9th tothe 11th of November of 2004 zooplankton wascollected from nine ponds in the Donana Natio-nal Park (SW Spain). To estimate zooplanktonabundance, one qualitative pelagic sample wastaken using a 40 μm net by horizontal hauls and,depending on the size of the pond, between 1and 3 quantitative samples were taken using a5 l bottle. Each sample was concentrated witha 20 μm mesh net and preserved in 100 ml of4% formaldehyde solution.

For the amino acid analysis of the roti-fers, live animals were collected using a 35 μmnet by horizontal hauls and kept cold duringthe fieldwork. They were isolated and prepa-red for analysis on return to the field lab within1 hour after collection.

At the same time and place where the zoo-plankton samples were collected, between 1 and3 quantitative samples of phytoplankton were ta-ken at 0.5 m. The samples were preserved withLugol solution. The abundance of each specieswas determined by Utermohl’s method using5 or 20 ml depending on plankton abundance.The census was halted when 200 counting units(cells, colonies or filaments) of the most abun-dant species had been reached.

Physical and chemical variables

At a depth of 0.5 m pH, conductivity, tempe-rature, ammonia (NH+

4 ), nitrate (NO−3 ), nitrite

(NO−2 ), silicate (SiO

−4 ) and phosphate (PO

3−4 ) we-

re analyzed at the same place where the planktonsamples were collected. Filtered water (0.45 μm)was used for analyzing nutrients with an auto-analyzer BRAN + LUEBBE AAIII.

The chlorophyll was analyzed by filtering thewater through fiber glass (Watman GF/C) filters,

Rotifer community assemblages 275

and the photosynthetic pigments were extractedin darkness in cold methanol. The Talling andDriver formula (Vollenweider, 1969) was used tocalculate chlorophyll a concentrations.

Analysis of amino acids

When the abundance of the rotifer species wassufficient, individuals were isolated from thesample. In order to establish a common amountof total material, amino acid analysis was per-formed on samples containing 15-20 rotifersof each species per vial, and there were 2 to9 replicates of each species per pond. Aminoacids were measured by high-performance liquidchromatography (HPLC) using an Alliance sys-tem, a 474 scanning fluorescence detector, and a15 × 3.9 Nova-Pak C18 column (VanWandelen &Cohen, 1997). The amino acids were hydrolyzedat 114 ◦ C with ClH 6N. With this method someamino acids such as cysteine and methionine aredegraded and, hence, cannot be measured. Aminoacid standard H NCI0180 PIERCE was used foridentification and quantification.

Estimation of spatial overlap among species

TheMorosita index (CH) was used to measure thespatial overlap or degree of co-occurrence amongrotifer species:

CH =

2n∑i=1

pijpik

n∑i=1

p2ij +n∑

ii=1p2ik

(1)

where p is the proportion of the abundance of thespecies j and k in the sample i from the total abun-dance of each species in all the samples, and n isthe number of samples. A higher Morisita indexindicates a greater spatial overlap.

Optimum of each rotifer species for thephysical and chemical variables

The weighted mean for the physical and chemi-cal variables measured in the ponds (Table 1) wasused as an indicator of the optimal temperature,

conductivity, pH, etc., for each rotifer species.The chemical and physical variables were

standardized using the following equation:

Vip =x −Minp

Maxp −Minp(2)

where Vip is the standardized value of sample x ofeach physical or chemical variable p in each pondi, and Min and Max are the minimum and maxi-mum values of the physical or chemical variablep, respectively, considering all ponds.

Standardized means of the physical and che-mical variables measured in the ponds (Table 1),weighted for rotifer, were used as indicators ofoptimal temperature, conductivity, pH, etc., foreach rotifer species, as follows:

−xps=

n∑i=1

wisVip

n∑i=1

wis

(3)

where Vip is the standardized value of the physi-cal or chemical variable p (temperature, pH, theconcentration of ammonia, etc.) in pond i, wis isthe abundance of the rotifer species sin pond i,and n is the number of ponds.

To obtain a graphic representation of the opti-mum for each rotifer species, taking into accountthe weighted means of all physical and chemicalvariables, a polar coordinate system was used toposition each rotifer species in the diagram. Thecoordinates of this polar plot were calculated bythe following equations:

Xs =

n∑p=1

∣∣∣(Vp)∣∣∣ cos (α π

180

)

Ys =

n∑p=1

∣∣∣(Vp)∣∣∣ sin (α π

180

) (4)

where X and Y are the positions in polar plot ofthe species s, Vp is the standardized value for thephysical or chemical variable p for species s, α isthe arbitrary angle assigned to variable p, and n

276 Guisande et al.

Table 1. Mean values of physical and chemical variables during the sampling period in the ponds. The number of samples wasthree in Santa Olalla and Dulce, two in Sopeton and one sample in the rest of ponds for each sampling period. Valores medios de lasvariables fısicas y quımicas durante el periodo de muestreo en las distintas lagunas. El numero de muestras fueron 3 en Santa Olallay Dulces, dos en Sopeton y una en las restantes, en cada periodo de muestreo.

Pond Latitude Longitude

NO−3

(μM)

NO−2

(μM)

PO3−4

(μM)

NH+4

(μM)

SiO−4

(μM)

Temperature

(◦ C)Conductivity

(μS cm−1) pH

Santa Olalla 36◦58′50.6′′ 6◦28′55.4′′ 0.20 0.26 0.56 0.70 265.68 28.5 1 563 8.4Dulce 36◦58′44.9′′ 6◦29′2.3′′ 0.08 0.38 0.79 0.60 179.22 28.4 747 8.4Taraje 36◦59′16.6′′ 6◦29′46.1′′ 0.05 0.89 1.06 0.92 91.70 27.5 905 7.63

Zahillo 36◦59′3.1′′ 6◦29′15.8′′ 0.05 0.70 0.37 0.65 106.20 26.4 419 7.72Toro 36◦59′10.4′′ 6◦30′22.0′′ 0.42 0.30 0.26 1.23 189.04 23.9 1 034 7.64Acebuche 37◦02′55.0′′ 6◦34′1.5′′ 0.30 0.84 0.62 2.08 118.00 32.6 217 7.42Ojillo 37◦00′20.8′′ 6◦30′26.7′′ 0.04 0.23 0.08 0.49 18.16 31.3 371 7.32

Las Verdes 36◦57′27.3′′ 6◦26′58.7′′ 0.05 1.09 0.32 1.14 62.78 29.1 495 6.08Sopeton 36◦59′03′′ 6◦27′48.5′′ 0.14 0.24 1.73 0.85 310.47 32.3 1 162 7.17

is the number of physical and chemical variables.As the number of chemical and physical varia-bles was 8 (Table 1), the α of the first variablep was 45◦ (360◦/8), the α of the second varia-ble was 90◦, the third 135◦, etc.

Amino acid discrimination among species

To show that the AAC consistently discrimina-tes between rotifer species, a discriminant analy-sis was carried out. Discriminant analysis is apattern-recognition method that helps to separatetwo or more groups from data provided for seve-ral variables. This type of analysis of the AAC ofspecies has been successfully used in the discri-mination of zooplankton species (Guisande et al.,2000; Guisande et al., 2002; 2003).

Differences in the AAC and the optimum forthe physical and chemical variables amongspecies

Average distance (Djk) was used to determine theseparation in the AAC and the optimum for phy-sical and chemical variables among species:

Djk =

√√√√ n∑i=1

(Xji − Xki

)2n

(5)

where X is the mean of the scores for axis i (spe-cies centroid) of the rotifer species jand k, obtai-ned from the discriminant analysis performed onthe AAC or the weighted mean of each physicalor chemical variable estimated using equation 3;

Table 2. Mean percentage of the abundance of each phytoplankton taxon. Due to degradation of the sample there is not informationof the phytoplankton community of the pond Toro. Porcentaje medio de la abundancia de cada taxon del fitoplancton. Debido a ladegradacion de la muestra no existe informacion de la laguna de Toro.

Pond Cyanophyta Dinophyta Cryptophyta Euglenophyta Diatom Chlorophyta

Totalabundance

(cell ml−1)

Chlorophyll aconcentration

(μg l−1)

% richness % richness % richness % richness % richness % richnessSanta Olalla 97.2 5 0.0 0 00.5 1 0.1 1 00.4 5 01.9 9 50 416 205.2

Dulce 29.0 6 0.1 1 03.3 3 5.7 5 22.5 11 39.3 23 0 9753 101.2Taraje 31.4 8 0.0 0 04.9 1 4.4 5 19.1 2 40.2 11 0 5916 231.3Zahillo 75.2 8 0.0 0 00.0 0 3.1 3 09.3 6 12.4 5 0 1870 043.0

Acebuche 22.0 3 0.0 0 01.1 2 1.8 3 01.6 4 73.5 9 0 6735 220.5Ojillo 20.1 4 0.0 0 62.3 2 0.6 3 04.4 6 12.5 11 24 766 623.3

Las Verdes 29.0 7 0.0 0 01.2 3 6.2 5 18.8 4 44.9 14 0 9889 344.4Sopeton 52.2 5 0.0 0 02.2 3 5.2 12 04.0 3 36.3 21 24 746 194.8

Rotifer community assemblages 277

Figure 1. Mean abundance of the rotifer species in the pondsin Donana National Park over the period studied. Testudinellapatina (T.p.), Poyarthra sp. (P.sp.), Platyas quadricornis (P.q.),Notholca acuminata (N.a.), Lophocaris oxyternon (L.o.), Le-padella patella (L.p.), Lecane quadridentata (L.q.), L. lunaris(L.l.), L. luna (L.n.), L. bulla (L.b.), Keratella quadrata (K.q.),K. tropica (K.t.), Hexarthra sp. (H.sp.), Filinia terminalis (F.t.),Euchlanis sp. (E.sp.), Cephalodella gibba (C.g.), Brachionusquadridentatus (B.q.), B. plicatilis (B.p.), B. leydigi (B.l.), B.falcatus (B.f.), B. calyciflorus (B.c.), B. bidentata (B.b.), B. an-gularis (B.a) and Asplanchna sp. (A.sp.). Abundancia media delos rotıferos en las distintas lagunas del Parque Nacional deDonana a lo largo del periodo de estudio. Testudinella pati-na (T.p.), Poyarthra sp. (P.sp.), Platyas quadricornis (P.q.), Not-holca acuminata (N.a.), Lophocaris oxyternon (L.o.), Lepadellapatella (L.p.), Lecane quadridentata (L.q.), L. lunaris (L.l.), L.luna (L.n.), L. bulla (L.b.), Keratella quadrata (K.q.), K. tropica(K.t.), Hexarthra sp. (H.sp.), Filinia terminalis (F.t.), Euchlanissp. (E.sp.), Cephalodella gibba (C.g.), Brachionus quadridenta-tus (B.q.), B. plicatilis (B.p.), B. leydigi (B.l.), B. falcatus (B.f.),B. calyciflorus (B.c.), B. bidentata (B.b.), B. angularis (B.a) andAsplanchna sp. (A.sp.).

and n is either the number of axes of the discrimi-nant analysis, or the number of ponds, when com-paring the weighted means. A higher amino acid

overlap and a higher similarity in the optimumof the physical and chemical variables, for eachrotifer species, are obtained when Dik is smaller.

RESULTS

Limnological characteristics of the ponds

The concentration of nutrients, NO−3 and PO3−

4were low in all the ponds (Table 1). The maindifferences between ponds were in the concentra-tion of (SiO−

4 ) and conductivity. The ponds Ojillo,Las Verdes, Acebuche and Zahillo were at the lo-wer end of the ranges for conductivity and SiO−

4 .Santa Olalla, Toro and Sopeton had the highestvalues for conductivity and SiO−

4 .

Phytoplankton community

Phytoplankton abundance was high in most ofthe ponds (Table 2). Cyanophyta and chlorophy-ta were the most abundant phytoplankton groups(Table 2), with the exception of the Ojillo pondwhere the cryptophyta was the dominant taxa.Species richness ranged from 21 in Santa Olallaand Acebuche to 49 in Dulce.

Figure 2. Relationship between rotifer and cyanophyta abun-dances in the ponds. Relacion entre la abundancia de rotıferosy cianofıtas en las lagunas.

278 Guisande et al.

Figure 3. Relationship between rotifer and phytoplankton ri-chness in the ponds. Relacion entre la riqueza de rotıferos yfitoplancton en las lagunas.

Zooplankton community

The zooplankton community was dominated byrotifers. The dominant rotifer species in theponds were Keratella tropica and species belon-ging to the genus Brachionus: B. plicatilis, B. fal-

catus, B. calyciflorus and B. angularis (Fig. 1).There was a significant positive correlation bet-ween total rotifer density and the abundance ofcyanobacteria (Fig. 2, regression slope differentfrom zero, p = 0.007).

Species richness of the rotifer community ran-ged from 5 (in Santa Olalla and Zahillo) to 10species in Sopeton (Fig. 1). Rotifer richness washigher in those ponds with greater phytoplanktonrichness (Fig. 3, regression slope different fromzero, p = 0.006). The distribution of rotifer spe-cies according to physical and chemical variablesis shown in figure 4. The differences in the op-timum for each physical and chemical variableamong species can be seen in this graph.

Zooplankton crustaceans were representedby small species: the copepods Acanthocy-clops kieferi and Copidodiaptomus numidicus,the cladocerans Alonella nana, Chydorussphaericus, Macrothrix hirsuticornis and Moi-na brachiata. The exception was the largecladoceran Simocephalus vetulus, but this spe-cies is more benthic than planktonic. Moreover,the abundance of all zooplankton crustaceanswas never higher than 1 ind l−1, except for A.kieferi in one sample from Santa Olalla, wherea figure of 6 1 ind l−1 was reached.

Table 3. Amino acid composition (mean ± SD weight percentage of total amino acids yield) of rotifer species collected fromthe ponds. Amino acid abbreviations: ASP-aspartic acid; SER-serine; GLU-glutamic acid; GLY-glycine; HIS-histidine; ARG-arginine; THR-threonine; ALA-alanine; PRO-proline; TYR-tyrosine; VAL-valine; LYS-lysine; ILE-isoleucine; LEU-leucine; PHE-phenylalanine. Species abbreviation as in Figure 1. Composicion de aminoacidos (media ± SD) de los rotıferos recolectados en laslagunas. Las abreviaciones corresponden a: ASP-acido aspartico; SER-serina; GLU-acido glutamico; GLY-glicina; HIS-histamina;ARG-arginina; THR-treonina; ALA-alanina; PRO-prolina; TYR-tirosina; VAL-valina; LYS-lisina; ILE-isoleucina; LEU-leucina;PHE-fenilalanina. Las abreviaciones de las especies se encuentran en la Figura 1.

A.sp. B.f . B.a. B.c. B.p. F.t. P.sp. K.t.

ASP 10.2 ± 1.6 6.7 ± 1.0 7.4 ± 2.4 9.4 ± 2.7 8.8 ± 0.6 7.8 ± 4.9 6.9 ± 0.0 7.5 ± 1.8SER 9.8 ± 2.1 10.0 ± 1.5 10.9 ± 2.0 11.5 ± 2.0 10.0 ± 2.3 11.3 ± 2.5 10.9 ± 2.4 10.8 ± 2.9GLU 7.2 ± 1.6 5.6 ± 0.8 6.4 ± 2.0 7.4 ± 2.1 8.4 ± 1.3 6.2 ± 1.9 5.6 ± 0.5 6.3 ± 2.2GLY 8.8 ± 0.6 11.9 ± 1.5 15.0 ± 2.6 11.4 ± 1.9 11.7 ± 3.5 17.4 ± 1.3 12.4 ± 0.8 12.7 ± 2.2HIS 1.7 ± 0.1 1.8 ± 0.1 1.9 ± 0.3 2.0 ± 0.7 2.1 ± 0.3 1.8 ± 0.0 2.0 ± 0.1 1.9 ± 0.5ARG 5.4 ± 3.2 2.4 ± 0.7 4.5 ± 2.3 2.8 ± 1.2 2.2 ± 0.0 6.9 ± 0.0 6.2 ± 3.5 2.8 ± 0.6THR 5.2 ± 0.6 4.8 ± 0.5 4.5 ± 0.6 5.2 ± 0.5 4.6 ± 0.5 4.8 ± 1.3 5.4 ± 0.1 4.4 ± 0.5ALA 10.5 ± 1.1 14.1 ± 1.6 12.0 ± 2.4 10.4 ± 1.3 8.1 ± 4.5 11.0 ± 3.6 9.7 ± 1.4 14.9 ± 2.3PRO 5.3 ± 0.7 7.6 ± 0.5 5.4 ± 0.9 5.7 ± 1.9 7.0 ± 0.3 3.8 ± 0.0 4.6 ± 0.6 6.2 ± 2.6TYR 3.1 ± 0.7 3.4 ± 0.3 3.8 ± 0.8 3.0 ± 0.7 4.2 ± 1.9 4.2 ± 1.1 4.1 ± 0.1 4.3 ± 0.7VAL 5.6 ± 0.5 7.6 ± 0.5 6.5 ± 1.8 6.8 ± 0.8 7.2 ± 0.8 5.4 ± 0.7 5.9 ± 0.0 6.3 ± 0.9LYS 7.2 ± 0.9 4.2 ± 0.5 4.5 ± 1.0 5.6 ± 0.6 4.9 ± 0.6 3.8 ± 0.8 5.8 ± 0.2 3.9 ± 0.6ILE 5.9 ± 0.8 5.7 ± 0.2 5.1 ± 0.8 5.7 ± 0.6 6.0 ± 0.2 5.1 ± 0.6 6.2 ± 0.4 5.3 ± 0.8LEU 8.9 ± 0.4 9.5 ± 0.3 7.9 ± 0.8 8.5 ± 1.1 9.7 ± 0.8 6.6 ± 0.6 8.8 ± 0.6 8.5 ± 1.1PHE 5.1 ± 0.3 4.7 ± 0.5 4.5 ± 0.5 4.6 ± 0.7 5.3 ± 0.1 4.0 ± 0.6 5.5 ± 0.8 4.0 ± 0.4

Rotifer community assemblages 279

Figure 4. Polar diagram of the weighted means of the speciesfor the physical and chemical variables measured in the ponds.Species abbreviations as in figure 1.Diagrama polar de los taxaen el espacio definido por las variables fısicas y quımicas. Lasabreviaciones de las especies como en la figura 1.

Amino acid composition of the rotifer species

The AAC of the species is shown in Table 3.Discriminant analysis of the AAC of the speciesindicated that, despite the intraspecific variation,94.2% of cases were correctly classified (Table 4and Fig. 5). Individually, six species were correctlyclassified in >92% of the cases (Table 4). Therewas onlymisclassification betweenB. plicatilis andB. calyciflorus, which may be due partly to an errorduring the isolation of the rotifers, because thesetwo species are very similar and, hence, difficultto distinguish under the binocular microscope.Rotifer species mainly differed in the proportionof lysine, valine andglycine (Table 5).

Figure 5. Plots of the first two discriminant function for theamino acids of the rotifer species (upper plot) and the mean ±SD of the scores (lower plot). Species abbreviations as in figure1. Espacio definido por las dos primeras funciones discrimi-nantes para los aminiacidos de las especies (arriba) y la media±DS de los ejes (abajo). Las abreviaciones como en la figura 1.

Table 4. Results of a discriminant analysis show the percent of rotifer species correctly classified from the original data accordingto the amino acids of each species. Species abbreviations as in Figure 1. Los resultados del Analisis discriminante muestran elporcentaje de especies de rotıferos correctamente clasificados a partir de los datos correspondientes a la frecuencia de aminoacidosde las especies. Las abreviaturas de las especies se muestran en la Figura 1.

True groups Predicted groups

A.sp. B.f . B.a. B.c. B.p. F.t. P.sp. K.t.

A.sp. 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0B.f . 0.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0B.a. 0.0 0.0 92.3 0.0 0.0 0.0 7.7 0.0B.c. 0.0 0.0 0.0 88.9 11.1 0.0 0.0 0.0B.p. 0.0 0.0 0.0 33.3 66.7 0.0 0.0 0.0F.t. 0.0 0.0 0.0 0.0 0.0 100.0 0.0 0.0P.sp. 0.0 0.0 0.0 0.0 0.0 0.0 100.0 0.0K.t. 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0

280 Guisande et al.

Table 5. Structure matrix of the discriminant analysis per-formed on the amino acids of the rotifer species indicatingthe intra-groups correlations between the discriminant varia-bles and the discriminant functions (* significant correlations)and the percentage of variance explained by each discrimina-te function. Amino acid abbreviations as in Table 3. Matriz delanalisis discriminante de los aminoacidos de las especies indi-cando las correlaciones intragrupos entre variables y funcionesdiscriminantes (* correlaciones significativas) y el porcentajede la varianza explicada por cada funcion discriminante. Lasabreviaciones de los aminoacidos se muestran en la Tabla 3.

Function I(55.1%)

Function II(27.0%)

Function III(9.2%)

ASP −0.135 0.033* 0.184*SER 0.003 0.020* −0.085*GLU −0.085 −0.037* 0.042*GLY 0.156 0.114* −0.723*HIS −0.008 −0.064* −0.025*ARG −0.078 0.260* −0.243*THR −0.146 0.030* 0.173*ALA 0.283 0.015* 0.219*PRO 0.074 −0.226* 0.225*TYR 0.151 0.093* −0.119*VAL 0.036 −.252* 0.003*LYS −0.432 0.076* 0.442*ILE −0.084 −0.066* 0.238*LEU −0.028 −0.268* 0.467*PHE −0.223 −0.127* 0.041*

It is interesting to point out that, with the excep-tion of Brachionus falcatus, discrimination of therotifer species by their AAC (Fig. 5) has a simi-lar pattern to the polar distribution of the spe-cies according to their weighted means for phy-sical and chemical variables (Fig. 4). Therefo-re, those species with a similar AAC have si-milar preferences for the physical and chemicalvariables, indicating that the AAC of each spe-cies is a biochemical fingerprint of the adaptationof the species to the habitat.

There was a negative relationship betweenamino acid separation and spatial overlap amongrotifer species (Fig. 6), indicating that those spe-cies with a similar AAC have a higher spa-tial overlap. As each zooplankton species pairwas not independent from the others, a boot-strap method was used to evaluate the statis-tical significance of this relationship (Davison& Hinkley, 1997). Regression was recalculated1000 times using random series in which only50% of the pairs’ abundance data were used tocalculate spatial niche overlap. In all cases, the

slope of the regression was both negative andsignificantly different from zero.

From all the physical and chemical variablesmeasured in the ponds (Table 1), there was onlya significant relationship between amino acid

Aminoacidseparation

Spatial overlap

Figure 6. Relationship between spatial overlap (Morisita in-dex) and amino acid separation (Euclidean distance using thescores of the discrimiant analysis) among zooplankton species(upper plot) and the mean ± SD of the values (lower plot) forthe intervals of spatial overlap 0-0.2, 0.2-0.4, 0.4-0.6, 0.6-0.8and 0.8-1. Relaciones entre el solapamiento especial (indice deMorisita) y separacion entre aminoacidos (distancia Euclideausando los valores del analisis discriminante) entre las especiesdel zooplancton (arriba) y la media ± DS de los valores (aba-jo) para los intervalos de solapamiento espacial 0-0.2, 0.2-0.4,0.4-0.6, 0.6-0.8 y 0.8-1.

Rotifer community assemblages 281Speciesseparationaccordingto

silicates

Amino acid separation

Figure 7. Relationship between the separation of species ac-cording to their weighted mean for silicates (Euclidean distan-ce) and amino acid separation (Euclidean distance using thescores of the discrimiant analysis) among zooplankton species(upper plot) and the mean ± SD of the values (lower plot) forthe intervals of amino acid separation 1-2, 2-3, 3-4 and 4-5.Relaciones entre la separacion de especies de acuerdo a su pe-so medio para silicatos (distancia Euclidea) y separacion deaminoacidos (arriba) y la media ±DS de los valores (abajo)para los intervalos de separacion de los aminoacidos 1-2, 2-3,3-4 y 4-5.

separation and species separation according totheir weighted means for silicates (Fig. 7) andconductivity (figure not shown because there isa clear relationship between the concentration ofsilicates and conductivity). For silicates the re-gression was recalculated 1000 times using ran-

dom series in which only 50% of the pairs ofamino acids data were used to calculate aminoacid separation. In 99.1% of cases the slopeof the regression was both positive and signi-ficantly different from zero.

DISCUSSION

This study shows that rotifer community assem-bly in the ponds is mainly governed by salinity.The effects of predation and interference compe-tition from large crustacean zooplankton, as wellas exploitative competition are less important.

In our study, the lack of effect of large crusta-cean zooplankton on rotifers may be partly due tothe high abundance of cyanobacteria in most ofthe ponds. The highest densities of rotifers havebeen reported when there were heavy cyanobac-terial blooms (Geng et al., 2005). We also foundthat rotifer abundance was higher in those pondswith a high amount of cyanophyta. Rotifers arepreyed on by large crustacean zooplankton (Gil-bert & Williamson, 1978; Stemberger & Evans,1984; Williamson & Butler, 1986; Dieguez &Gilbert, 2002), or are susceptible to mechanicalinterference from them (Gilbert, 1988; MacIsaac& Gilbert, 1989; Arvola & Salonen, 2001;Nandini et al., 2002). However, large copepodsand cladocerans are negatively affected by cya-nobacterial blooms (Gliwicz & Lampert, 1990,Lauren-Maata et al., 1997). It is therefore likelythat the occurrence of cyanobacterial blooms inthe ponds caused a shift within the dominantcrustacean zooplankton from larger species tosmaller ones and, thereby, weakened the negativeinteraction between crustaceans and rotifers(Geng et al., 2005). However, it is necessary topoint out that there is a significant correlationbetween conductivity and cyanobacteria abun-dance ( p = 0.013). Therefore it is not possibleto reject the idea that the potential relationshipbetween cyanobacteria abundance and rotifers isdue to that both are governed by other factors,as for instance conductivity.

Food limitation in rotifers can be conside-red in terms of the minimal food level for re-production (Stemberger & Gilbert, 1985; 1987).

282 Guisande et al.

Threshold food concentrations, for which po-pulation growth rate is zero, range from0.06 μg ml−1 dry mass, for small species (such asKeratella cochlearis), to 0.38 for medium-sizedspecies (such as Brachionus calyciflorus), and to0.6 for large species (such as Asplanchna prio-donta) μg ml−1 dry mass (Stemberger & Gilbert,1985; 1987; Guisande & Mazuelos, 1991). Themaximum growth rate for medium-sized speciesis achieved at around 10 μg ml−1 dry mass (Gui-sande & Mazuelos, 1991). When we conside-red dry mass to be a chlorophyll a ratio of ap-proximately 100, in nearly all the ponds the drymass was higher than 10 μg ml−1 (Table 2), wellabove the food concentration needed to achie-ve maximum growth rates. The only exceptionwas Zahillo, where the estimated dry mass was4.3 μg ml−1. When merely considering phyto-plankton, and not other food resources for roti-fers such as bacteria, it would appear that rotiferswere not food limited in the ponds.

In a study carried out in 29 Pyrenean oli-gotrophic lakes, where zooplankton species we-re food limited (Guisande et al. (2003)), a po-sitive relationship between amino acid separa-tion and spatial overlap in cyclopoid and cla-docera species was seen. For these oligotrophiclakes, exploitative competition was, by drivingco-evolutionary histories either at present or inthe past, a significant factor in structuring thezooplankton communities. However, we foundthe opposite pattern in the ponds (Fig. 5). Spe-cies with a similar AAC, and hence with a si-milar trophic niche, spatially co-exist, suppor-ting the view that trophic-niche differentiationwas not the main factor in structuring rotiferassemblages in the ponds.

The fact that exploitative competition does notplay a key role in determining the assemblage ofthe rotifer community in the ponds means thatmost of the species were not food limited, butthis does not mean that the effect of food sup-ply was unimportant. The positive relationshipbetween the richness of phytoplankton and roti-fer communities indicates that a higher diversityof food resources favors rotifer richness, which,in turn, may be explained by a trophic-nichedifferentiation among rotifer species.

The high similarity in the AAC among co-occurring species observed in our study could beinterpreted as evidence for a predominant role forhabitat filtering, indicating the relevant role ofabiotic factors on the assemblages of the rotifercommunity in the ponds. The positive relation-ship between amino acid separation and speciesseparation according to their weighted mean forthe concentration of silicates indicates that sali-nity was the main abiotic factor in structuring theassembly of the rotifer community in the ponds.

We have demonstrated that the AAC of zoo-plankton species, in addition to being species-specific and a good indicator of the trophic ni-che of the species (Guisande et al., 2002; 2003;McClelland & Montoya, 2002; Boechat &Adrian, 2005), is also a good indicator of theadaptation of the species to the abiotic conditionsof the habitat. Therefore, the AAC is a good indi-cator of the ecological niche of zooplankton spe-cies and, hence, a good tool to elucidate how zoo-plankton communities assemble from a regionalpool of species. The approach of studies usingthe role of species in the habitat (the niche) is im-portant in ecology, but it has been limited by ob-vious difficulties in the characterization of the ni-che. The possibility of showing the ecological ni-che of the species by using the AAC would allowthe study of the ecological diversity, which is avaluable approach in ecology, rather than the ta-xonomic diversity of zooplankton communities,and possibly of other taxonomic groups.

ACKNOWLEDGEMENTS

This research was supported by the FundacionRamon Areces.

REFERENCES

ARNOTT, S. E. & M. J. VANNI. 1993. Zooplanktonassemblages in fishless bog lakes: Influence of bio-tic and abiotic factors. Ecology, 74: 2361-2380.

ARVOLA, L. & K. SALONEN. 2001. Plankton com-munity of a polyhumic lake with and withoutDaphnia longispina (Cladocera). Hydrobiologia,445: 141-150.

Rotifer community assemblages 283

BOECHAT, I. G. & R. ADRIAN. 2005. Biochemicalcomposition of algivorous freshwater ciliates: Youare not what you eat. FEMS Microbiology. Eco-logy, 53: 393-400.

DAVISON, A. C. & D. V. HINKLEY. 1997. Boots-trap Methods and Their Applications. CambridgeUniversity Press, New York, 320 pp.

DEVETTER, M. 1998. Influence of environmentalfactors on the rotifer assemblage in an artificial la-ke. Hydrobiologia, 387-388: 171-178.

DIEGUEZ, M. C. & J. J. GILBERT. 2002. Suppres-sion of the rotifer Polyarthra remata by the omni-vorous copepod Tropocyclops extensus: predationor competition. J. Plankton Res., 24: 359-369.

DUGGAN, I. C., J. D. GREEN& R. J. SHIEL. 2002.Distribution of rotifer assemblages in North Island,New Zealand, lakes: relationships to environmen-tal and historical factors. Freshwat. Biol., 47: 195-206.

GENG, H., P. XIE., D. DENG & Q. ZHOU. 2005.The rotifer assemblage in a shallow, eutrophic Chi-nese lake and its relationships with cyanobacterialblooms and crustacean zooplankton. J. Freshwat.Ecol., 20: 93-100.

GILBERT, J. J. & C. E. WILLIAMSON. 1978.Predator-prey behavior and its effect on rotifer sur-vival in associations ofMesocyclops edax, Asplan-chna girodi, Plyarthra vulgaris, and Keratella co-chlearis. Oecologia, 37: 13-22

GILBERT, J. J. 1988. Susceptibilities of ten rotiferspecies to interference from Daphnia pulex. Eco-logy, 69: 1 826-1 838.

GLIWICZ, Z. M. &W. LAMPERT. 1990. Food thres-holds in Daphnia species in the absence and pre-sence of blue-green filaments. Ecology, 71: 691-702.

GREEN, J. 2001.Variability and instability of plank-tonic rotifer associations in Lesotho, southernAfrica. Hydrobiologia, 446/447: 187-194.

GUISANDE, C. & J. TOJA. 1988. The dynamics ofvarious species of the genus Brachionus (ROTA-TORIA) in the Guadalquivir river. Archiv fur Hy-drobiologie, 112: 579-595.

GUISANDE, C. & N. MAZUELOS. 1991. Repro-ductive pattern of Brachionus calyciflorus Pallasat different food concentrations. J. Plankton Res.,13: 279-286.

GUISANDE, C., I. MANEIRO& I. RIVEIRO. 1999.Homeostasis in the essential amino acid composi-tion of the marine copepod Euterpina acutifrons.Limnol. Oceanogr., 44: 691-696.

GUISANDE, C., I. RIVEIRO& I. MANEIRO. 2000.Comparison between the amino acid compositionof females, eggs and food to determine the relati-ve importance of food quantity and food quality oncopepod reproduction. Marine Ecology Progressi-ve Series, 202: 135-142.

GUISANDE, C., I. MANEIRO, I. RIVEIRO, A. BA-RREIRO & Y. PAZOS. 2002. Estimation of cope-pod trophic-niche in the field using amino acidsand marker pigments. Marine Ecology Progressi-ve Series, 239: 147-156.

GUISANDE, C., F. BARTUMEUS, M. VENTURA& J. CATALAN. 2003. Role of food partitioning instructuring the zooplankton community in moun-tain lakes. Oecologia, 136: 627-634.

KIMMEL, D. G. & B. P. BRADLEY. 2001. Specificprotein responses in the calanoid copepod Euryte-mora affinis (Poppe, 1880) to salinity and tempera-ture variation. Journal Experimental Marine Bio-logy and Ecology, 266: 135-149.

HELLAND, S., J. C. NEJSTGAARD, H. J. FYHN, J.K. EGGE & U. BMSTEDT. 2002. Effects of star-vation, season, and diet on the free amino acid andprotein content of Calanus finmarchicus females.Marine Biology, 143: 297-306.

LAUREN-MAATA, C., O. KLEIVEN & J. KIRI-VANTA. 1997. Horizontal distributions of Daph-nia pulex in response to toxic and non-toxic algalextracts. J. Plankton Res., 19: 141-148.

MCCLELLAND, J. W. & J. P. MONTOYA. 2002.Trophic relationships and the nitrogen isotopiccomposition of amino acids in plankton. Ecology,83: 2173-2180.

MACISAAC, H. J. & J. J. GILBERT. 1989. Compe-tition between rotifers and cladocerans of differentsizes. Oecologia, 81: 295-301.

PACE, M. L. & D. VAQUE. 1994. The importanceof Daphnia in determining mortality rates of pro-tozoans and rotifers in lakes. Limnol. Oceanogr.,39: 985-996.

STELZER, C. P. 2001. Resource limitation and re-productive effort in a planktonic rotifer. Ecology,82: 2521-2533.

STEMBERGER, R. S. & S. EVANS. 1984. Rotiferseasonal succession and copepod predation in lakeMichigan. Journal Great Lakes Research, 10: 417-428.

STEMBERGER, P. L. & J. J. GILBERT. 1985. Bodysize, food concentration and population growth inplanktonic rotifers. Ecology, 66: 1151-1159.

284 Guisande et al.

STEMBERGER, P. L. & J. J. GILBERT. 1987. Roti-fer threshold food concentrations and the size effi-ciency hypothesis. Ecology, 68: 181-187.

VAN WANDELEN, C. H. & S. A. COHEN. 1997.Using quaternary high-performance liquid chroma-tography eluent systems for separating 6-ami-noquinolyl-N-hydroxysuccinimidyl carbamate-de-rivatized amino acid mixtures. Journal Chromato-

graphy,763:11-22.VOLLENWEIDER, R. A. 1969. Primary production

in aquatic environments. I.B.P. Handbook n 12,Blackwell Sci. Publ., Oxford 250 pp.

WILLIAMSON, C. E. & N. M. BUTLER. 1986. Pre-dation on rotifers by the suspension-feeding cala-noid copepod Diaptomus pallidus. Limnol. Ocea-nogr., 31: 393-4.

Limnetica, 27 (2): x-xx (2008)Limnetica, 27 (2): 285-298 (2008)c© Asociacion Iberica de Limnologıa, Madrid. Spain. ISSN: 0213-8409

Influence of iron and nitrate concentration in water on aquaticColeoptera community structure: Application to the Avia River(Ourense, NW. Spain)

Marta Fernandez-Dıaz, Cesar Joao Benetti ∗ and Josefina Garrido

Department of Ecology and Animal Biology, Faculty of Biology. University of Vigo, Campus Lagoas-Marcosende, 36310, Vigo, Spain2

∗ Corresponding author: cjbenetti@uvigo.es2

Received: 26/3/08 Accepted: 8/7/08

ABSTRACT

Influence of iron and nitrate concentration in water on aquatic Coleoptera community structure: Application to theAvia river (Ourense, NW. Spain)

In this work the influence of the concentration of dissolved iron and nitrate in the water on the structure of the communityof aquatic Coleoptera of a river in northwest Spain is analyzed. A total of 45 species and subspecies of aquatic Coleoptera(Adephaga and Polyphaga) were collected between March and November 1999 from 10 sites distributed along the Avia river(Ourense, N.W. Spain). Also, 24 physical and chemical parameters of water were measured and a CCA was made to evaluatewhich of these factors had the greatest influence on the studied fauna. At the same time an ecological analysis was made usingthe distribution of species and subspecies in relation to two water parameters (concentration of nitrate and dissolved iron). Thestudy of distribution in relation to these factors by means of elaborating their ecological profiles suggested indicator speciesfor each of these parameters. Seventeen species were found to be indicators for at least one of the parameters analyzed. Thespecies Deronectes costipennis gignouxi, Enichocerus legionensis, Hydraena testacea, Megasternum obscurum, Nebrioporusdepressus elegans and Stenelmis canaliculata are the best indicators of these parameters, if we consider only taxa collected inclass 4, that is, with high values of iron and nitrate.

Key words: Aquatic Coleoptera, dissolved iron, nitrate, indicator species, fluvial ecosystem.

RESUMEN

Influencia de la concentracion de hierro y nitrato en el agua sobre la estructura de una comunidad de Coleoptera acuati-cos: Aplicacion al rıo Avia (Ourense, NO de Espana)

En este trabajo se analiza la influencia de la concentracion de hierro disuelto y nitrato en el agua sobre la estructura de lacomunidad de coleopteros acuaticos de un rıo en el noroeste de Espana. Un total de 45 especies y subespecies de Coleopteraacuaticos (Adephaga y Polyphaga) fueron recolectados entre marzo y noviembre de 1999 en 10 puntos distribuidos a lo largodel rıo Avia (Ourense, NW Espana). Tambien se tomaron medidas de 24 de parametros fısicos y quımicos del agua y se hizoun CCA para evaluar cual de estos factores fue el de mayor influencia en la fauna estudiada. A partir de eso, realizamosun analisis ecologico de la distribucion de la presencia de especies y subespecies en relacion con dos factores quımicos delagua (concentracion de nitrato y hierro disuelto). El estudio de la distribucion en relacion con estos factores a traves dela elaboracion de sus perfiles ecologicos senalo las especies indicadoras para cada uno de estos parametros, de acuerdoa su informacion recıproca de especie-factor. Diecisiete especies demostraron ser indicadoras de por lo menos unos de losparametros analizados. Las especies Deronectes costipennis gignouxi, Enichocerus legionensis, Hydraena testacea, Megas-ternum obscurum, Nebrioporus depressus elegans y Stenelmis canaliculata son las mejores indicadoras de estos parametrossi tenemos en cuenta solo los taxones recogidos en la clase 4, es decir, con altos valores de hierro y nitrato.

Palabras clave: Coleoptera acuaticos, hierro disuelto, nitrato, especies indicadoras, ecosistema fluvial.

286 Fernandez-Dıaz et al.

INTRODUCTION

Fluvial ecosystems are of special interest be-cause of their high biological richness and di-versity. Communities of aquatic organisms havedeveloped adaptations that allow them to pros-per in these environments, which at the same ti-me, make them very vulnerable to possible al-terations in streams. In this sense, human acti-vity often causes severe ecological damage to ri-ver systems. These disturbances produce altera-tions in the chemical composition of water, and inthe structure of the communities of organisms li-ving in this environment (Somolders et al., 1999;Smolders et al., 2003; Oller & Goitia, 2005).

According to Nummelin et al. (2007) diffe-rent groups of insects are excellent indicators ofcontamination by heavy metals. For Beasley &Kneale (2003) macroinvertebrates, among themdifferent groups of Coleoptera, are good indica-tors of high iron concentrations. A detailed studyof the autoecology of different species of aqua-tic beetles shows that many of them have qui-te narrow ecological requirements and are veryuseful as bioindicators in determining the charac-teristics of aquatic environments (Castella et al.,1984; Flechtner, 1986; Davis et al., 1987; Fosteret al., 1990; Foster et al., 1992; Eyre et al., 1992;Eyre et al., 1993; Collinson et al., 1995; Riberaet al. 1995a; Ribera et al. 1995b; Moreno et al.,1997). Other authors (Eyre & Foster, 1989; Ey-re & Rushton, 1989) consider them as the mostuseful group in assessing environmental qualityand change in static water. Brancucci (1980) andRibera & Foster (1992) state that beetles can beconsidered good indicators of macro and microenvironmental characteristics of the aquatic envi-ronment in which they live.

Aquatic beetles are abundant in different fresh-water habitats and show a high species richness ofaround 600 species and subspecies in the IberianPeninsula (Ribera, 2000). In the first studies ofwater quality they remained very much in thebackground, but have now become more importantas indicators (Garcıa-Criado & Fernandez-Alaez,2001), since most of their life cycle occurs inaquatic environments. Jeffries (1988) shows thatcommunities of aquatic beetles are a reflection of

macroinvertebrate communities as a whole, andHeuss (1989) affirms that some Coleoptera candenote as many degrees of contamination as levelsof dissolved oxygen. In the Iberian Peninsula,different studies on fluvial networks (Garcıa deJalon & Gonzalez del Tanago, 1982; Gonzalez delTanago, 1986; Gonzalez et al., 1986a; Gonzalezet al., 1986b; Prat et al., 1983; Prat et al., 1984;Alba-Tercedor et al., 1986; Palau&Palomes, 1986;Alba-Tercedor & Jimenez, 1987; Cortes, 1989;Ribera&Foster, 1992), point out Coleopterawithinbenthic communities as important bioindicators ofpollution and evaluators of good water quality. Inthis sense the most abundant families in rivers areHydraenidae and Elmidae, considered as the bestindicators of environmental impacts (Dıaz, 1991;Garcıa-Criado, 1999; Garcıa-Criado & Fernandez-Alaez, 1995; Garcıa-Criado & Fernandez-Alaez,2001;Garcıa-Criado et al., 1999).

Different studies have used ecological profi-les to designate indicator species (Petraglia &Tomaselli, 2003; Bonada et. al. 2004; Mitov &Stoyanov, 2005). There are different studies onthe autoecology of several species of these Co-leoptera (Puig 1983, Sainz-Cantero et al., 1987;Sainz-Cantero et al., 1988; Gil et al., 1990;Dıaz ,1991; Garrido et al., 1994), and othersthat evaluate the sensitivity of many species todifferent ecological factors by ecological profi-le analysis and calculation of reciprocal informa-tion (Valladares et al., 1990; Garrido et al., 1994;Garcıa-Criado & Fernandez-Alaez, 1995). Thestudy shows the preferences of several species ofColeoptera within the ranges of values found forseveral chemical factors of water.

METHODS

The study area is the Avia river in the southof the Galician Community, in the north-northwest of Ourense Province, NW Spain (Fig. 1).It is surrounded by a set of massive mountainranges forming the border with the provincesof Pontevedra and Lugo. Along its course itis fed by numerous streams forming a riverof 643 Km2 surface. The eastern part of theriver flows on schistose substrate and mainly on

Influence of iron and nitrate concentration in water on aquatic Coleoptera 287

Figure 1. Location of the study area: Avia river with the sampling sites, in the Minho river basin. Localizacion del area de estudio:rıo Avia, con los puntos de muestreo, en la cuenca del rio Mino.

granite substrate in the west. The river Avia islocated in the Mino river basin.

Samples were taken seasonally during an an-nual cycle (fromMarch 1999 to November 1999)and were collected from ten sites (A1-A10) so weinitially had forty samples designated with a co-de indicating the site (Ax) and season (Wi, Sp,Su, Au) when the sample was taken. The sam-ples were taken with an entomological water net(0.5 μm mesh), along 10 m for ten minutes. Thisis one of the more efficient methods of samplingto evaluate the specific diversity and structures ofthe community (Garcıa-Criado & Trigal, 2005).Also, different physical and chemical parameterswere measured, of which pH, temperature, totaldissolved salts (TDS), dissolved oxygen and con-ductivity were measured in situ.

Laboratory analysis of the water samples in-cluded tests for silica, phosphate, chlorine, am-monium, sulphate, nitrate, nitrite and the concen-tration of ions (including metals) dissolved in the

water were obtained. The determination of tra-ce elements was made with ICP-MS technique.The measured ions were: calcium, magnesium,sodium, potassium, iron, manganese, copper,chromium, aluminium, lead, zinc and cadmium.Some of the environmental variables did not sur-pass the detection limit, for example, chromium,cadmium, lead, zinc, phosphates and sulphates.These do not appear in the results (Table 1) andwere not considered in this paper.

Canonical correspondence analysis (CCA)was used to analyse species-environment rela-tionships (see Ter Braak & Van Tongeren, 1995)in order to identify environmental (abiotic) fac-tors potentially influencing aquatic Coleopteraassemblages. The factors analyzed were potas-sium, sodium, aluminium, manganese, iron andnitrate. The statistical significance of ordinationaxes 1 and 2 was determined using a Monte Car-lo permutation test. Rare species (represented byjust one individual) were not down-weighted thus

288 Fernandez-Dıaz et al.

Table 1. Physical and chemical variables measured in Avia river during period march- november, 1999. Means and ranges (mini-mum and maximum) values are represented. Variables fısicas y quımicas medidas en el rıo Avia durante el perıodo marzo- noviembrede 1999. Valores medios y rangos (mınimo y maximo) son representados.

RangeParameter Mean Maximum Minimum

Calcium mg/l 00.65 002.08 0.17Magnesium mg/l 00.60 001.66 0.23Potassium mg/l 01.97 008.99 0.22Sodium mg/l 04.34 013.55 0.39Manganese μg/l 07.16 060.27 0.77copper μg/l 02.06 098.70 < 0.5Iron μg/l 57.56 484.33 4.56Aluminium μg/l 27.26 189.65 4.22Silica mg/l 02.19 097.40 < 0.2Chlorine mg/l 09.10 021.75 < 5Amonium mg/l 00.23 000.45 < 0.05Nitrite mg/l 00.14 000.28 < 0.01Nitrate mg/l 05.39 011.20 < 2.2Temperature ◦C 11.92 019.40 7.43pH 06.92 008.17 6.13Conductivity μS/cm 36.71 076.70 19.6O2 mg/l 09.69 012.16 7.07O2 % 94.64 111.00 76.3TDS mg/l 15.25 034.00 9.00

all occurrences were included in the analysis. Thedata for the ecological profiles were analysed asfollows: firstly, the observed range of values foreach habitat parameter was divided into a seriesof intervals, as homogeneously as possible. Thesamples were assigned according to the valuesobtained for each parameter, thus enabling us todetermine the number of samples in the differentclasses. The expression of Daget, Godron & Gui-llerm (1972) was used to determine water quality.

First the data was shown in the form of an ab-solute frequency profile (number of presences ofeach species in each class of the parameter). Ho-wever, the total number of samples included ineach class of parameter had to be considered, aswell as the frequency of each species in the set ofsamples. This way, differences between the fea-tures of rare and frequent species were correc-ted. For this reason, we presented the profile ofcorrected frequencies that considered the relati-ve average frequency of presences. The values ofthis profile were calculated using the expressionof Daget & Godron (1982). We then determinedwhich of the studied species were more closelylinked to the different factors, both positively and

negatively, and which could be considered as in-dicators, thus allowing the analysis to be conti-nued with a smaller group of species. The infor-mation provided by a species in relation to thedifferent factors was estimated from the ecologi-cal features of presence and absence, named “Re-ciprocal information” and was taken as a measureof the indicator value of the species in relation tothe physicochemical parameters considered. Fora species E and iron levels L, such informationis expressed as I (L; E) and defined from the ex-pression mentioned in Godron (1968). Finally, tosummarise the information given by each pro-file in a single value we calculated barycentre“G”. This can be taken as a measurement of theoptimum ecological degree of the species. Theexpression used for this calculation is mentio-ned in Daget & Godron, 1982.

RESULTS

A total of 6 250 specimens of aquatic Coleoptera,belonging to 45 taxa, were identified in 10 sam-pling sites in the Avia river, all of which are listed

Influence of iron and nitrate concentration in water on aquatic Coleoptera 289

Table 2. Total species studied in Avia river (march-november, 1999) and respective values of corrected frequencies, mutual infor-mation and barycentres. Total de especies estudiadas en el rıo Avia (marzo-noviembre, 1999) y respectivos valores de frecuenciascorregidas, informacion mutua y baricentro.

Dissolved Iron NitrateCorrected frequencies Mutual inf Barycentre Corrected frequencies Mutual inf Barycentre

Species C 1 C 2 C 3 C 4 C 1 C 2 C 3 C 4

HaliplidaeHaliplus lineatocollis 0 2.571 0 0 0.039 2 2.571 0 0 0 0.039 1

DytiscidaeDeronectes costipennis gignouxi 0 2.571 0 0 0.039 2 0 0 0 7.2 0.083 4Deronectes ferrugineus 0 2.571 0 0 0.079 2 0 0 2.571 3.6 0.094 3.583Graptodytes fractus 0 2.571 0 0 0.039 2 0 0 5.143 0 0.068 3Graptodytes ignotus 0 1.714 1.714 0 0.069 2.5 0.857 1.2 1.714 0 0.024 2.227Hydroporus nigrita 0 0 2.571 4.5 0.104 3.636 0 1.8 0 3.6 0.079 3.333Nebrioporus depressus elegans 0 0 0 9 0.093 4 0 0 5.143 0 0.068 3Nebrioporus carinatus 0.818 1.285 0 2.25 0.049 2.846 0 0.9 2.571 1.8 0.106 3.171Oreodytes sanmarkii alienus 0 2.571 0 0 0.039 2 0 0 5.143 0 0.068 3Stictonectes epipleuricus 1.454 1.142 0.571 0 0.072 1.721 0.857 2 0.571 0 0.127 1.917Stictonectes lepidus 0 2.571 0 0 0.122 2 0 1.2 1.714 2.4 0.068 3.226Stictotarsus bertrandi 1.227 1.285 0.642 0 0.055 1.815 0.321 2.25 1.286 0 0.174 2.25

HydrophilidaeAnacaena globulus 1.09 1.285 0.857 0 0.034 1.928 0.429 1.2 2.571 0 0.114 2.51Megasternum obscurum 0 0 0 9 0.093 4 0 0 0 7.2 0.083 4

HydrochidaeHydrochus angustatus 0 2.571 0 0 0.079 2 0 0 2.571 3.6 0.093 3.583

HelophoridaeHelophorus flavipes 3.272 0 0 0 0.049 1 0 3.6 0 0 0.051 2

HydraenidaeEnichocerus legionensis 0 2.571 0 0 0.039 2 0 0 0 7.2 0.083 4Hydraena barrosi 1.636 1.285 0 0 0.1 1.44 0.429 0.6 3.429 0 0.184 2.673Hydraena brachymera 0.909 1.142 0.857 1 0.01 2.498 0.714 1 1.714 0.8 0.11 2.615Hydraena corinna 1.309 0.514 2.057 0 0.06 2.193 1.543 0.72 1.029 0 0.044 1.844Hydraena hispanica 1.006 1.384 0.791 0 0.098 1.932 0.791 0.554 2.374 0.554 0.19 2.63Hydraena iberica 1.5 0.964 0.428 0.75 0.274 2.118 0.75 1.35 1.286 0.6 0.149 2.435Hydraena inapicipalpis 2.013 0.395 1.186 0 0.264 1.77 0.198 1.938 1.582 0.554 0.263 2.583Hydraena lusitana 1.636 1.071 0.428 0 0.134 1.615 0.857 1.5 1.286 0 0.101 2.118Hydraena sharpi 1.402 1.102 0.734 0 0.109 1.794 0.989 1.385 1.187 0 0.1 2.056Hydraena stussineri 1.963 1.028 0 0 0.093 1.344 0.514 0.72 3.086 0 0.115 2.595Hydraena testacea 0 0 0 9 0.093 4 0 0 0 7.2 0.083 4Hydraena unca 0 0 5.142 0 0.068 3 0 3.6 0 0 0.051 2

ElmidaeDupophilus brevis 0.818 1.028 1.028 1.35 0.022 2.689 0.771 0.9 1.8 0.72 0.195 2.589Elmis aenea 0.654 1.028 1.371 1.2 0.036 2.733 0.857 1.2 1.029 0.96 0.008 2.517Elmis maugetii maugetii 1.272 1.142 0.857 0 0.136 1.873 0.857 1 1.429 0.8 0.036 2.531Elmis perezi 2.181 0.857 0 0 0.06 1.282 0 1.2 3.429 0 0.118 2.741Elmis rioloides 0.872 1.2 1.028 0.6 0.02 2.366 0.857 1.2 0.686 1.44 0.034 2.648Esolus angustatus 1.227 1.607 0 0 0.14 1.567 0.964 0.9 1.929 0 0.081 2.254Esolus parallelepipedus 1.033 1.218 0.812 0.473 0.047 2.205 0.947 0.758 1.353 1.137 0.029 2.639Limnius opacus 1.309 0.514 0 3.6 0.117 3.086 1.029 0 2.057 1.44 0.077 2.864Limnius perrisi carinatus 1.09 1.142 0.857 0.5 0.031 2.213 1 1.2 1.429 0 0.174 2.118Limnius volckmari 0.981 0.514 1.542 1.8 0.061 2.86 0.771 1.08 1.543 0.72 0.024 2.538Normandia nitens 0 1.285 0 4.5 0.075 3.556 0 0 5.143 0 0.143 3Oulimnius bertrandi 1.09 1.142 0.761 0.666 0.069 2.274 0.762 1.333 1.143 0.8 0.062 2.491Oulimnius troglodytes 0 0.964 1.928 2.25 0.17 3.25 1.286 0 0.643 2.7 0.179 3.028Oulimnius tuberculatus perezi 1.09 1.428 0.571 0 0.072 1.832 0.857 0.8 1.714 0.8 0.027 2.589Stenelmis canaliculata 0 0 0 9 0.093 4 0 0 0 7.2 0.083 4

DryopidaeDryops luridus 0 0.857 3.428 0 0.102 2.8 1.714 1.2 0 0 0.054 1.412

290 Fernandez-Dıaz et al.

in Table 2. Also, 24 physical and chemical va-riables were measured; five of them gave resultsbelow the detection limits and are not includedin this analysis (Table 1).

According to the CCA the most significantaxis is 1, with a significance of 0.362. Theparameters were very significantly correlated(r = 0.91 between iron and manganese; r = 0.49between manganese and nitrate; r = 0.44 bet-ween iron and nitrate) and also showed highcorrelation with Axis 1 (iron: r = 0.61; man-ganese: r = 0.39; nitrate: 0.31). The canonicalcorrespondence analysis showed that the ma-jority of species responded negatively to iron,nitrate and manganese (Fig. 2). However, thejoint CCA of species clearly separated a subsetof five aquatic beetle species along the ironenvironmental axis (Hydraena testacea, Me-gasternum obscurum, Stenelmis canaliculata,

Nebrioporus depressus elegans and Oulimniustroglodytes), two species along the nitrateaxis (Enicocerus legionensis and Deronectescostipennis gignouxi) and only one species(Hydroporus nigrita) along the manganese axis(Fig. 2). After making a multivariate analysiswe selected two variables with significantvalues and variations that could influence thedistribution or location of the species, thesefactors were concentration of dissolved ironand nitrates in the water. The CCA indicatesthat the manganese concentration is one ofthe most significant elements in these rivers.However, ecological profiles were not made asthere was no significant correlation with anyof the species (corrected frequency high andbarycentre above 2). According to Nummelin etal. (2007) the best bioindicators of this metalare water striders and not water beetles.

8.0

_ 0.4

_0.4

1.0

Figure 2. Canonical Correspondence Analysis (CCA) based on the water beetles species in the Avia river. Analisis de Correspon-dencias Canonicas (CCA) basado en las especies de escarabajos acuaticos en el rıo Avia.

Influence of iron and nitrate concentration in water on aquatic Coleoptera 291

a) Hydraena testacea b) Hydraena unca

c) Hydroporus nigrita

i) Stenelmis canaliculata

h) Oulimnius troglodytesg) Normandia nitens

f) Nebrioporus depressus eleganse) Megasternum obscurum

d) Limnius opacus

Figure 3. (a-i) Ecological profiles for the species from sites with a high concentration of dissolved iron. (a-i) Perfiles ecologicos delas especies de los puntos con una concentracion alta de hierro disuelto.

292 Fernandez-Dıaz et al.

a) Deronectes costipennis gignouxi b) Deronectes ferrugineus

c) Enichocerus legionensis d) Graptodytes fractus

e) Hydraena testacea f) Hydrochus angustatus

g) Hydroporus nigrita h) Megasternum obscurum

Figure 4. (a-o) Ecological profiles for the species from sites with high concentration of nitrate. (a-o) Perfiles ecologicos de lasespecies de los puntos con una concentracion alta de nitrato.

Influence of iron and nitrate concentration in water on aquatic Coleoptera 293

i) Nebrioporus carinatus j) Nebrioporus depressus elegans

k) Normandia nitens l) Oreodytes sanmarkii alienus

m) Oulimnius troglodytes n) Stenelmis canaliculata

o) Stictonectes lepidus

Figure 4. (continuation)

294 Fernandez-Dıaz et al.

The results of the ecological profiles appearin Table 2 and show the corrected frequenciesand indicator value for each species (reciprocalspecies-describer information) and the barycen-tre for each of the two variables studied. With re-gard to species preferences for dissolved iron andnitrate concentrations, we only consider the spe-cies with high reciprocal information values foreach parameter and barycentre equal to 4, (Figs. 3and 4). The response of the species most closelylinked (those giving more information) to each ofthe surveyed variables is shown below:

Iron: The observed sampling quality value forthis variable was 0.932, which is quite high, indi-cating that it was sufficiently sampled. The ran-ge of values for this parameter (4.56-484.3 μg/l)was divided into four intervals (C1: less 25 μg/l;C2: 25-49.9 μg/l; C3: 50-74.9 μg/l; C4: more75 μg/l), depending on the number of the sam-ples. Thus, the last interval was more extensi-ve due to the low number of samples. The mostrepresentative species for this variable (barycen-tre equal or superior to 3) are indicated in figu-re 3. These species are characteristic of sites witha high concentration of dissolved iron and werefound fundamentally in classes 3 and 4.

Nitrate: A sampling quality value of 0.949,higher than that for iron, was obtained, indicatingthat it was also sufficiently sampled. The nitra-te values were between 2.2-11.2 mg/l. Due to theexistence of a great number of samples below thedetection limit, the interval < 2.2 mg/l was crea-ted to include those samples with very low nitra-tes values. So this range was divided into fourintervals (C1: less 2.2 mg/l; C2: 2.2-5.29 mg/l;C3: 5.30-6.49 mg/l; C4: more 6.5 mg/l), was wellas iron, depending on the number of the sam-ples. According to their profiles of distributionin these intervals, the most representative spe-cies for this variable (barycentre equal or supe-rior to 3) are indicated in figure 4. As in thecase of iron, we can also state that, for nitra-te, these species are characteristic of sites withhigh concentrations of nitrate and were foundfundamentally in classes 3 and 4.

DISCUSSION

The results obtained in this paper contribute datato the study of the autoecology of species andto how physical and chemical variables of theenvironment influence the habitats and distri-bution of aquatic Coleoptera. In order to reachfinal conclusions on the ecological preferencesof each species, the effect of polluting agents andtolerance to their different levels, more studiesin other rivers are necessary. The objective ofthis study was to verify whether iron and nitrateinfluence the distribution of species of aqua-tic Coleoptera and result in the establishmentof communities more or less tolerant to highconcentrations of these variables. We have toconsider that many other local characteristicscan also influence the distribution of the speciesin greater or smaller measure.

When interpreting the results, it should betaken into account that in this river the highconcentrations of iron and nitrate, producedby discharge of fertilizers and industrial waste,were localized, so it was difficult to distinguishwhether the distribution of a certain taxa de-pended on the chemical characteristics of thewater. Also, we must not forget that this researchonly deals with lotic environments. Therefore,specimens peculiar to lenthic habitats, such asHydraena testacea, have different ecologicalpreferences in lenthic sites.

Of all 45 species and subspecies collected, 17proved to be indicators for at least one of the pa-rameters analyzed and collected in classes 3 and4 (Figs. 3 and 4). This number is reduced to 6 ifwe consider only taxa collected in class 4, whichare the best indicators of these parameters. The6 species are: Deronectes costipennis gignou-xi, Enichocerus legionensis, Hydraena testacea,Megasternum obscurum, Nebrioporus depressuselegans and Stenelmis canaliculata. These taxaare likewise ones with the greatest average indi-cator value, thus giving more global informationabout the variables under consideration.

With regard to the values for iron in the groupof species found in areas with high concentra-tions, we observed differences when comparedwith results obtained by other authors. This is the

Influence of iron and nitrate concentration in water on aquatic Coleoptera 295

case of Hydraena testacea which, in this study,was collected in areas with high iron concen-trations, in contrast to data provided by Dıaz(1991) in the Landro river (Lugo, NW Spain),in a study on the Hydraenidae family. In thestudy carried out by Paz (1993) on Adephaga,in the same river, Hydroporus nigrita was foundin practically all the samplings sites, with bothhigh and low iron concentrations. However, inour study it was mainly found close to high con-centrations of this metal. Differences with Dıaz(1991) and Paz (1993) are due to the low con-centrations of these parameters in the Landroriver and therefore the analyzed species werenot indicators in these studies.

Another species found exclusively in siteswith high iron values was Nebrioporus depres-sus elegans. There is no significant informationon this species in other works, though Lorenz(2004) indicates that it is a negative indicatorfor freshwaters in Germany. There is little infor-mation available in other studies on the relationbetween the studied species and iron. Therefore,the results obtained here do not have a compa-rison point and should be considered with pre-caution when stating that these species are indi-cators of iron in freshwater bodies. Extensive li-terature supports the fact that high iron concen-trations have a negative effect on the fauna ofaquatic ecosystems (Blasco et al., 1999; Blascoet al., 2000), something also stated in our study,since most of the studied species do not occurin sites with high concentrations of this metal.According to Beasley and Kneale (2003) spe-cies of Coleoptera, specially Hydrophilidae andDytiscidae, are the best bioindicators of iron inaquatic systems, agreeing with the results obtai-ned in our study, though we have also indicatedspecies of Hydraenidae and Elmidae as excellentindicators of this metal. (Fig. 2).

In the case of the second parameter, nitrate,we found that the studied species have antago-nistic behaviours in other studies. Thus, Enicoce-rus legionensis is classified in our study as beingdistributed in areas of high concentration, in con-trast to Dıaz (1991) who, in a study in the Landroriver (Lugo, NW Spain), states that it is sensiti-ve to organic contamination as it was only pre-

sent in two sampling sites with very clean water.However Garcıa-Criado (1999) found the speciesin sites affected by coal mining in Leon (Spain),which could be indicative of tolerance to con-tamination. As in the case of iron, we do nothave extensive comparative data for nitrate andtherefore cannot affirm that these species are in-dicative of high nitrate levels, although the col-lected data are a strong indication of this rela-tion in the studied rivers. According to differentauthors (Blasco et al., 2000; Ortiz et al., 2005)high nitrate concentrations in the water make thesurvival of many species difficult, in benefit ofother opportunist species that are more tolerantto this chemical parameter. This was also detec-ted by Oliveira & Cortes (2005) who emphasi-ze the importance of measuring nitrate concen-trations to evaluate fluvial systems correctly. Inour work only a few species were tolerant to highnitrate values and the same occurred with iron,indicating possible contamination in certain sites.

It should be pointed out that iron and ni-trate occur naturally in fluvial systems, thoughhigh levels could indicate localized pollution.The levels in the Avia river were not excessi-vely high, however, both of them showed peaksthat may be indicative of a certain level of po-llution. The peaks observed in the points A5and A10 in spring and summer and are dueto discharge of fertilizers (agricultural areas) orindustrial waste (light industries).

The CCA analysis, corroborated by barycen-tre values, shows clearly and briefly what we ha-ve discussed so far. Deronectes costipennis gig-nouxi and Enichocerus legionensis stand out be-cause of their tolerance to relatively high levelsof nitrate. Oreodytes sanmarkii alienus and Stic-tonectes lepidus exhibit more intermediate posi-tions, though also a certain degree of tolerance.Hydraena testacea, Megasternum obscurum andStenelmis canaliculata stand out because of theirtolerance to high levels of dissolved iron. Nebrio-porus depressus elegans andOulimnius troglody-tes exhibit more intermediate positions, but alsoshow a high degree of tolerance. At the opposi-te end are most of the studied species, presentat sites with low values for both analysed para-meters. According to the ecological profiles of

296 Fernandez-Dıaz et al.

the studied species in relation to their respon-se to the analysed parameters (iron and nitra-te) it can be stated that three of them (Hydrae-na testacea, Megasternum obscurum and Stenel-mis canaliculata) were found exclusively in siteswith high concentrations of both iron and nitra-te. This could indicate that they are tolerant spe-cies, though also opportunist, since they occupythe place of other more sensitive and less tolerantspecies not found in these sites.

REFERENCES

ALBA-TERCEDOR, J. y F. JIMENEZ-MILLAN.1987. Evaluacion de las variaciones estacionalesde la calidad de las aguas del rıo Guadalfeo ba-sada en el estudio de las comunidades de macro-invertebrados acuaticos y de los factores fısico-quımicos. Proyecto Lucdeme III, ICONA. Madrid.174 pp.

ALBA-TERCEDOR, J., I. GUISASOLA y A. SAN-CHEZ-ORTEGA. 1986. Variaciones estacionalesde las caracterısticas fısico-quımicas y de la cali-dad biologica de las aguas del R. Guadalfeo (Gra-nada). In: El agua en Andalucıa, I. Universidad deGranada (ed.): 235-247. Universidad de Granada,Granada.

BEASLEY, G. & P. KNEALE. 2003. Investigatingthe influence of heavy metals on macroinvertebra-te assemblages using Partial Canonical Correspon-dence Analysis. Hydrobiology and Earth SystemSciences, 7: 221-233.

BLASCO, J., A. M. ARIAS & V. SAENZ. 1999.Heavy metals in organisms of the River Guadal-quivir estuary: possible incident of the Aznalcollardisaster. Sci. Total Environ., 242: 249-259.

BLASCO, J., V. SAENZ & A. GOMEZ-PARRA.2000. Heavy metal fluxes at the sediment-waterinterface of three coastal ecosystems from south-west of the Iberian Peninsula. Sci. Total Environ.,247: 189-199.

BONADA, N., C. ZAMORA-MUNOZ, M. RIERA-DEVALL & N. PRAT. 2004. Ecological profiles ofcaddisfly larvae in Mediterranean streams: impli-cations for bioassessment methods. EnvironmentalPollution, 132: 509-521.

BRANCUCCI, M. 1980. Observations on the eco-logy of the Dytiscidae (Coleoptera) in the watersystem connected to the south coast of Neuchatel

lake (Switzerland). Mitteilungen der Schweizeris-chen Entomologischen Gesellschaft, 53: 365-378.

CASTELLA, E., M. RICHARDOT-COULE, C.ROUX& P. RICHOUX. 1984. Macroinvertebratesas “describers” of morphological and hydrologicaltypes of aquatic ecosystems abandoned by RhoneRiver. Hydrobiologia, 119: 219-225.

COLLINSON, N. H., J. BIGGS., A. CORFIELD.,M. J. HODSON., D. WALKER., M. WHITFIELD& P. J. WILLIAMS. 1995. Temporary and perma-nent ponds: An assessment of the effects of dryingout on the conservation value of aquatic macroin-vertebrate communities. Biological Conservation,74: 125-133.

CORTES, R. M. V. 1989. Biotipologıa de ecosiste-mas loticos do Nordeste de Portugal. Ph.D. Thesis,Universidad de Tras os Montes e Alto Douro, VilaReal, Portugal, 305 pp.

DAGET, P. et M. GODRON. 1982. Analyse frequen-tielle de l’Ecologie des especes dans les commu-nautes. Coll. d’Ecologie. Masson, Parıs, 163 pp.

DAGET, P., M. GODRON et J. M. GUILLERM.1972. Profils ecologiques et information mutuelleentre especes et facteurs ecologiques. In: Grund-fragen und Methoden in der Pflanzensoziologie.Van der Maarel (ed.): 191-202. La Haya.

DAVIS, J. A., S. W. ROLLS & S. A. BALLA. 1987.The role of the Odonata and aquatic Coleoptera asindicators of environmental quality in wetlands. In:The Role of Invertebrates in Conservation and Bio-logical Survey. Majer, J. D. (ed.): 31-42. WesternAustralian Department of Conservation and LandManagement Report.

DIAZ, J. A. 1991. Estudio taxonomico y ecologicode los Hydraenidae (Coleoptera) de la cuenca delrıo Landro (Lugo, Espana). Ph.D. Thesis. Univer-sidad de Santiago de Compostela. 515 pp.

EYRE, M. D. & G. N. FOSTER. 1989. A compari-son of aquatic Heteroptera and Coleoptera commu-nities as a basis for environmental and conserva-tion assessments in static water sites. Journal Ap-plied Entomology, 108: 355-362.

EYRE, M. D. & S. P. RUSHTON. 1989. Quantifica-tion of conservation criteria using invertebrates.Journal of Applied Ecology, 26: 159-171.

EYRE, M. D., G. N. FOSTER & A. G. YOUNG.1993. Relationships between water-beetle distribu-tions and climatic variables: A possible index formonitoring global climatic change. Arch. Hydro-biol., 127: 437-450.

Influence of iron and nitrate concentration in water on aquatic Coleoptera 297

EYRE, M. D., R. CARR, R. P. MCBLANE & G. N.FOSTER. 1992. The effects of varying site-waterduration on the distribution of water beetle as-semblages, adults and larvae (Coleoptera: Halipli-dae, Dytiscidae, Hydrophilidae). Arch. Hydrobiol.,124: 281-291.

FLECHTNER, G. 1986. Association analysis of wa-ter-beetles communities (Coleoptera, Dytiscidae etHaliplidae).Entomologica Basiliensia, 11: 297-308.

FOSTER, G. N., A. P. FOSTER., M. D. EYRE& D.T. BILTON. 1990. Classification of waterbeetles assemblages in arable fenland and rankingof sites in relation to conservation value. Freshwat.Biol., 22: 343-354.

FOSTER, G. N., B. H. NELSON., D. T. BILTON., D.A. LOTT., R. MERRIT, R. S. WEYL &M. D. EY-RE. 1992. A classification and evaluation of Irishwater beetle assemblages. Aquatic Conservation:Marine and Freshwater Ecosystems, 2: 185-208.

GARCIA-CRIADO, F. 1999. Impacto de la minerıadel carbon sobre Hydraenidae y Elmidae (Coleop-tera) en la cuenca del rıo Sil (Leon, Espana).Ph.D. Thesis. Universidad de Leon, 281 pp.

GARCIA-CRIADO,F. &M. FERNANDEZ-ALAEZ.1995. Aquatic Coleoptera (Hydraenidae and Elmi-dae) as indicators of the chemical characteristicsof water in the Orbigo River basin (N-W Spain).Annals of Limnology, 31: 185-199.

GARCIA-CRIADO,F. &M. FERNANDEZ-ALAEZ.2001. Hydraenidae and Elmidae assemblages (Co-leoptera) from a Spanish river basin: good indi-cators of coal mining pollution? Arch. Hydrobiol.,150: 641-660.

GARCIA-CRIADO,F., C. FERNANDEZ-ALAEZ &M. FERNANDEZ-ALAEZ. 1999.Enviromental va-riables influencing the distribution of Hydraenidaeand Elmidae assemblages (Coleoptera) in a mode-ratly-polluted river basin in north-western Spain.European Journal of Entomology, 96: 37-44.

GARCIA-CRIADO, F. & C. TRIGAL. 2005. Com-parison of several techniques for sampling macro-invertebrates in different habitats of a North Ibe-rian pond. Hydrobiologia, 545: 103-115.

GARCIA DE JALON, D. & M. GONZALEZ DELTANAGO. 1982. Introduccion a una zoologıa delmacrobentos en los rıos de la sierra de Guadarra-ma. Boletın de la Estacion Central de Ecologıa,11: 63-71.

GARCIA DE JALON, D. & M. GONZALEZ DELTANAGO. 1986. Metodos biologicos para el es-tudio de la calidad de las aguas. Aplicacion a la

cuenca del Duero. ICONA- Monografıas, 45. Ma-drid. 244 pp.

GARRIDO, J., J. A. DIAZ & J. A. REGIL. 1994. Co-leopteros acuaticos de Extremadura II. (Hydrae-nidae, Hydrochidae, Helophoridae, Hydrophilidae,Dryopidae y Elmidae). Boletın de la AsociacionEspanola de Entomologıa, 18: 113-133.

GARRIDO, J., M. FERNANDEZ-ALAEZ & J. A.REGIL. 1994.Geographical distribution of Adepha-ga and Polyphaga (Coleoptera) in the CantabrianMountains (Spain): Specific richness and analysis ofthe altitude factor. Arch. Hydrobiol., 131: 353-380.

GIL, E., C. MONTES., A. MILLAN & A. G. SO-LER. 1990. Los coleopteros acuaticos (Hydraeni-dae & Elmidae) de la cuenca del rıo Segura (S.E.Espana). Anales de Biologıa, 16: 23-31.

GODRON, M. 1968. Quelques applications de la no-tion de frequence en ecologie vegetale (reco-uvrement, information mutuelle entre especes etfacteurs ecologiques, echantillonnage). OecologiaPlantarum, 3: 185-212.

GONZALEZ, F., A. MIRANDA., J. R. ALONSO &M. ABELLA. 1986a. Estudio comparativo de lacalidad del agua de dos rıos Asturianos. Limneti-ca, 2: 217-223.

GONZALEZ, F., A. MIRANDA., J. R. ALONSO &M.ABELLA.1986b.La calidaddel agua en la cuencadel rıoPilona (Asturias). Limnetica, 2: 225-234.

HEUSS, K. 1989. Water beetles as indicators of wa-ter quality. The Balfour-Browne Club Newsletter,44: 7-12.

JEFFRIES, M. 1988. Do water beetles communi-ties reflect the wider freshwater community? TheBalfour-Browne Club Newsletter, 42: 14-17.

LORENZ, A. 2004. Mid-sized Mountain Streams.Typology, Assessment and Reliability of Samplingand Assessment Methods. Ph.D. Thesis. Univer-sitat Duisburg-Essen, Germany. 146 pp.

MITOV, P. G. & I. L. STOYANOV. 2005. Ecologicalprofiles or harvestmen (Arachnida, Opiliones) fromVitosha Mountain (Bulgaria): a mixed modelling ap-proach using GAMs. (generalized additive models).The Journal of Arachnology, 33: 256-268.

MORENO, J. L., A. MILLAN., M. L. SUAREZ.,M. R. VIDAL-ABARCA & J. VELASCO. 1997.Aquatic Coleoptera and Heteroptera assemblagesin waterbodies from ephemeral coastal streams(“ramblas”) of south-eastern Spain. Arch. Hydro-biol., 141: 93-107.

NUMMELIN, M., M. LODENIUS, E. TULISALO,H. HIRVONEN, & T. ALANKO. 2007. Predatory

298 Fernandez-Dıaz et al.

insects as bioindicators of heavy metal pollution.Environmental Pollution, 145: 339-347.

OLIVEIRA, S. V. & R. M. V. CORTES. 2005.A bio-logically relevant habitat condition index for streamsin northern Portugal. Aquatic Conservation: MarineandFreshwaterEcosystems, 15: 189-210.

OLLER, C. & E. GOITIA. 2005. Benthic macroin-vertebrate and heavy metals in the Pilcomayo Ri-ver (Tarija, Bolivia). Revista Boliviana de Eco-logıa, 18: 17-32.

ORTIZ, J. D., E. MARTI, & M. A. PUIG. 2005. Re-covery of the macroinvertebrate community belowa wastewater treatment plant input in a Mediterra-nean stream. Hydrobiologia, 545: 289-302.

PALAU, A. & A. PALOMES. 1986. Los macroinver-tebrados bentonicos como elementos de juicio parala evaluacion de la calidad biologica del rıo Segre(Lerida, N.E. Espana). Limnetica, 2: 205-215.

PAZ, C. de, 1993. Hydradephaga (Coleoptera) de lacuenca del rıo Landro (NW Penınsula Iberica). Es-tudio faunıstico y ecologico. Ph.D. Thesis. Univer-sidad de Santiago de Compostela. 381 pp.

PETRAGLIA, A. & M. TOMASELLI. 2003. Ecolo-gical profiles of wetland plant species in the nort-hern Apennines (N. Italy). Journal of Limnology,62(1): 71-78.

PRAT, N. F., G. GONZALEZ, X. MILLET & M.A.PUIG. 1983. Prediccio i control de la qualitat deles aigues dels rius Besos i Llobregat. II. El pobla-ment faunistic i la seva relacio amb la qualitat deles aigues. Monografies, Diputacio de Barcelona,9: 164.

PRAT, N. F., M. A. PUIG, G. GONZALEZ, M. F.TORT &M. ESTRADA. 1984. Llobregat. In: Eco-logy of European Rivers. B.A. Whitton, (ed.): 527-552. Blackwell Scientific Publications. Oxford.

PUIG, M. A. 1983. Distribucion y ecologıa de las es-pecies de Elmidae (Coleoptera: Dryopiodea) de lacuenca del rıo Llobregat. Boletin de la AsociacionEspanola de Entomologıa, 7: 211-219.

RIBERA, I. 2000. Biogeography and conservation ofIberian water beettles. Biological Conservation,92: 131-150.

RIBERA, I. & G. N. FOSTER. 1992. Uso de los co-leopteros acuaticos como indicadores biologicos(Coleoptera). Elytron, 6: 61-75.

RIBERA, I., J. ISART & J. A. REGIL. 1995a. Au-toecologıa de algunas especies de Hydradephaga(Coleoptera) de los Pirineos. I. Gyrinidae, Halipli-dae, Noteridae e Hygrobiidae. Zoologica baetica,6: 33-58.

RIBERA, I., J. ISART & J. A. REGIL. 1995b. Au-toecologıa de algunas especies de Hydradephaga(Coleoptera) de los Pirineos. II. Dytiscidae. Zoolo-gica baetica, 6: 59-104.

SAINZ-CANTERO, C. E., A. SANCHEZ-ORTEGA& J. ALBA-TERCEDOR. 1987. Distribucion y au-toecologıa de Hydraenidae (Col.) en Sierra Nevada(Espana). Boln. Asoc. Esp. Ent., 11: 355-365.

SAINZ-CANTERO, C. E., C. ZAMORA-MUNOZ&J. ALBA-TERCEDOR. 1988. Coleopteros acua-ticos del rıo Monachil (Sierra Nevada, Espana).Elytron, 2: 97-106.

SMOLDERS, A. J. P., G. VAN HENGSTUM, J.LOERMANS, A. MONTES, H. RIZO e I. CAS-TILLO. 1999. Efectos de la Contaminacion Mi-nera sobre la Composicion de la MacrofaunaBentonica en el Rıo Pilcomayo. Revista Bolivianade Ecologıa, 6: 229-237.

SMOLDERS, A. J. P., A. A. C. LOCK, G. VANDERVELDE, R. I. MEDINA HOYOS & J. G. M.ROELOFS. 2003. Effects of Mining Activitieson Heavy Metal Concentrations in Water, Sedi-ment and Macroinvertebrates in Different Reachesof the Pilcomayo River, South America. Archiv.of Environmental Contamination and Toxicology,44: 314-323.

TER BRAAK, C. J. F. & O. F. R. VANTONGEREN.1995.Data Analysis in Community and LandscapeEcology. Cambridge University Press: Cambridge.299 pp.

VALLADARES, L. F., C. FERNANDEZ-ALAEZ &M. FERNANDEZ-ALAEZ. 1990. Influence of al-titude in the distribution of the aquatic Hydrophi-loidea (Coleoptera) in the province of Leon (NWSpain). Limnetica, 6: 79-86.

Limnetica, 27 (2): x-xx (2008)Limnetica, 27 (2): 299-316 (2008)c© Asociacion Iberica de Limnologıa, Madrid. Spain. ISSN: 0213-8409

SALINDEX: A macroinvertebrate index for assessing the ecologicalstatus of saline “ramblas” from SE of the Iberian Peninsula

Cayetano Gutierrez-Canovas ∗, Josefa Velasco Garcıa and Andres Millan Sanchez

Departamento de Ecologıa e Hidrologıa. Universidad de Murcia. Campus de Espinardo. 30100 Murcia.2

∗ Corresponding author: cayeguti@um.es2

Received: 5/12/07 Accepted: 8/7/08

ABSTRACT

SALINDEX: Amacroinvertebrate index for the ecological status assessment of saline “ramblas” from SE of the IberianPeninsula

A new index (SALINDEX) for assessing the ecological status of saline “ramblas” from the southeast of the Iberian Peninsulais presented. It was applied in the Protected Area of Ajauque Wetland and Rambla Salada (Murcia). The index is based on fourmetrics related with macroinvertebrate community: the richness of macroinvertebrate families (FR), the ratio between richnessof Coleoptera and Hemiptera species (C/H) and the presence and abundance of species that act as indicators of a healthy (I+)and impaired (I-) status (including exotic species). Metric scores were calculated with regard to reference condition valuesfor each of the 7 ecotypes established according to salinity and water flow velocity gradients. The four metrics that comprisethe index have the same weight, although the three first (FR, C/H, I+) score positively and the last one (I-) scores negatively.An ecological status class was assigned to each resulting index value (–1, 0, 1, 2, 3) according with the criteria set out in theWater Framework Directive. SALINDEX is based in ecological patterns typical of inland saline waters ecosystems (e.g., lowertaxonomic richness, predominace of Coleoptera and Hemiptera species, presence of halotolerant/halophilic species, etc.),and was able to detect anthropogenic perturbations, such as eutrophication and salinity drop processes, and any subsequentrecovery. In contrast, the IBMWP index, commonly used for the ecological evaluation of Iberian streams, was unsuited tosaline streams, and showed a negative correlation with the SALINDEX. The index presented here is the first tool specificallydeveloped for assessing saline ramblas from the Iberian southeast and, although some weaknesses were observed, these can becorrected by improving ecotypes definition, increasing the number of reference stations and complementing with biologicalinformation from inland saline waters ecosystems for its application at peninsular level.

Key words: Ecological status, aquatic macroinvertebrates, saline “ramblas”, Water Framework Directive, functional indica-tors.

RESUMEN

SALINDEX: Un ındice de macroinvertebrados para la evaluacion del estado ecologico de ramblas salinas en el sureste dela penınsula Iberica

Se presenta un nuevo ındice (SALINDEX) para la evaluacion del estado ecologico de las ramblas salinas en el sureste iberico,el cual ha sido aplicado al Paisaje Protegido del Humedal de Ajauque y Rambla Salada (Murcia). El ındice se basa en cuatrometricas de las comunidades de macroinvertebrados: riqueza de familias (RF), relacion entre la riqueza de especies decoleopteros y hemıpteros (C/H), presencia y abundancia de especies indicadoras de naturalidad (I+) y presencia de especiesindicadoras de alteracion (I-), incluidas las especies exoticas. La valoracion de cada metrica se ha realizado atendiendo alas condiciones de referencia para los 7 ecotipos establecidos segun el gradiente de salinidad y la velocidad del flujo de agua.Las cuatro metricas que componen el ındice tienen la misma ponderacion, aunque las tres primeras (RF, C/H, I+) puntuande forma positiva si se cumplen las condiciones establecidas y la ultima (I-) puntua de forma negativa. A cada posible valorresultante del ındice (–1, 0, 1, 2 o 3) se le ha asignado una categorıa de estado ecologico (malo, deficiente, moderado,bueno o muy bueno) en concordancia con las establecidas por la Directiva Marco de Agua. SALINDEX esta disenado sobreciertos patrones ecologicos caracterısticos de los ecosistemas salinos de interior (menor riqueza taxonomica, dominio decoleopteros y hemıpteros, presencia de especies halotolerantes y halofilas etc.), siendo capaz de detectar perturbaciones deorigen antropico, como procesos de dilucion y eutrofizacion, y la posterior recuperacion del sistema. En cambio, el IBMWP,

300 Gutierrez-Canovas et al.

uno de los ındices mas empleado para la evaluacion del estado ecologico de los rıos ibericos, resulto poco apropiado para rıossalinos, mostrando una correlacion negativa con SALINDEX. El ındice aquı presentado, constituye la primera herramientaespecıficamente disenada para evaluar ramblas salinas en el sur-sureste iberico y, aunque presenta algunas debilidades,estas pueden ser subsanables en un futuro a traves de una mejor definicion de ecotipos, de un mayor numero de estaciones dereferencia y de completar la informacion biologica para su aplicacion en los ecosistemas salinos a nivel peninsular.

Palabras clave: Estado ecologico, macroinvertebrados acuaticos, ramblas salinas, Directiva Marco de Agua, indicadoresfuncionales.

INTRODUCTION

Based on the EU Water Framework Directive(WFD, Directive 2000/60/EC) the ecological sta-tus assessment of streams and wetlands is an im-portant tool for the management, conservationand restoration of continental water bodies. Sincethe 1960’s, many methods have been developed,including the bioassessment of streams usingmacroinvertebrates (Rosemberg & Resh, 1993;Metcalfe-Smith, 1994; Barbour et al., 1999, Bo-nada et al., 2006). One of the most widely-usedindices for determining the water quality of Spa-nish streams is the IBWMP (Iberian Biologi-cal Monitoring Working Party) (Alba-Tercedor etal., 2002), an Iberian adaptation of the BMWP(Armitage et al., 1983) and is based on macroin-vertebrate differential tolerance to pollution. Ho-wever, as many others, this index was designedfor freshwater streams and is not appropriate forsaline streams, which are frequent in Mediterra-nean semiarid regions of the Iberian Peninsula:for example those that occur in the Segura, Gua-dalquivir, Jucar and Ebro basins.

In Iberian Southeast, saline “ramblas”, thatare watercourses with specific geomorphologicalfeatures that make them different from all othertemporary streams (Gomez et al., 2005), and as-sociated wetlands are very common in marly se-dimentary basins (Triassic and Miocene). Theyare characterised by hydrological extremes thatresult in drying and flooding, together with highsalt concentrations that determine the composi-tion of the community adapted to these extremeconditions (Gomez et al., 2005).

The GUADALMED project (Prat, 2002)developed a system of structural indicators of

ecological status, called PRECE (Protocolo Rapi-do de Evaluacion de la Calidad Ecologica) thatcomprises the IBMWP and another two indices(Jaimez-Cuellar et al., 2002). This protocolincludes saline streams in the “ramblas” ecotypewithout taking into account the heterogeneity ofthis group, which embraces a broad gradient ofsalinity (from very low to until hypersaline waterswith values close to 300 g/L) and temporality.

The described indices do not work in sali-ne systems due to the particularity of the sa-line biota, which includes halotolerant, halo-philic and halobiont species adapted to osmo-tic stress (Williams, 1981; Williams y Feltmate,1992). These communities show a lower taxono-mic richness and, unlike in freshwater systemsthe orders Diptera, Coleoptera and Hemipterashow great richness and abundance (Ward, 1992;Millan et al., 2001). For a wide salinity gra-dient (3-300 g/L), the relation between numberof taxa and salinity is negative (Williams et al.,1990). In Mediterrenean areas, freshening andeutrophication caused by an intensive agricultu-ral hydric surplus are the most significant impactson aquatic saline ecosystems. These impacts in-crease the number of macroinvertebrate speciesand families, reducing halophilic and halobionttaxa (Velasco et al., 2006).

Most indices for wetland assessment were de-veloped based on the macroinvertebrates found incoastal Almerıa, SE Spain (Ortega et al., 2004) andbased on crustaceans and insects from Catalonia(Boix et al., 2005). However, these indices did notembrace the athalassohaline wetlands associatedto drainage network systems, which are typicalof the Iberian Southeast (Ramırez-Dıaz, 1992;Vidal-Abarca et al., 2001;Gomez et al., 2005).

New index for saline “ramblas” assessment 301

Table 1. Ecotypes according to saline habitat classes as func-tion of current of velocity and salinity. Ecotipos resultantes dela clasificacion de los habitats salinos en funcion de la veloci-dad del flujo y de la salinidad.

Ecotype Velocity of flow Salinity

1 Lotic waters Hiposaline (∼3-∼20)2 Lentic waters Hiposaline (∼3-∼20)3 Lotic waters Mesosaline (∼20-∼40)4 Lentic waters Mesosaline (∼20-∼40)5 Lotic waters α−hypersaline (∼40-∼100)6 Lentic waters α−hypersaline (∼40-∼100)7 Lotic waters β−hypersaline (∼100-∼140)

Bearing in mind the singularity of aquatic sa-line taxa and the lack of appropriate tools fortheir ecological status assessment, the objectivesof this paper were: 1) To design an index basedon the macroinvertebrate community for asses-sing “ramblas” and wetlands in the Iberian Sout-heast; 2) To apply the index in the Protected Areaof Wetland of Ajauque and Rambla Salada; 3) Toassess its response in the face of dilution stressand 4) To analyse the relationships between theindex and environmental, biotic variables (struc-tural and functional) and the IBMWP.

MATERIALS AND METHODS

Development of SALINDEX

To develop this index we have followed the cri-teria for multimetric index design provided byPaulsen et al. (1991) and Barbour et al. (1995,1999) with some modifications. Firstly, the salineaquatic environments occurring in marly basins

of the Iberian Southeast were classified. Subse-quently, a group of metrics (parameters that res-pond to human stressors in a foreseeable way)were selected and scoring criteria for each wereestablished. Finally, the metrics were integratedin the index (SALINDEX) and the overall sco-ring system was established.

Aquatic environments classification

Seven ecotypes were established based on thetwo parameters that determine the distributionand the abundance of the species: salt concen-tration (from hyposaline to hypersaline waters)and flow velocity (lentic or lotic waters) (Ta-ble 1). The characterization of the aquatic en-vironments was made following the frameworkfor thalassic waters (Montes & Martino, 1987).Temporality was not considered as an importantvariable since all the studied water bodies arepermanent or semi-permanent.

Metric selection

The metrics were selected according to a biblio-graphic revision of ecological status indicatorsin saline aquatic ecosystems and their respon-se to human stressors (Table 2). The richnessof aquatic macroinvertebrate families (FR) andthe coefficient of Coleoptera/Hemiptera species(C/H) have been used previously in another mul-timetric index (Ortega et al., 2004). In addition, afurther two metrics were included: presence andabundance of Coleoptera and Hemiptera speciesthat indicate a good health (I+) and stressed (I-)status, including allocthonous species.

Table 2. Metrics selected for saline ecosystems and their predictable response to dilution (D) and eutrophication (E). ▲: increasing;▼: decreasing; V: variable. Metricas seleccionadas para ambientes salinos y su respuesta esperada ante las perturbaciones pordilucion (D) y por eutrofizacion (E). ▲: incremento; ▼: disminucion; V: variable.

Expectableresponse

Indicator Label Previous studies D E

Family richness FR Williams et al., 1990; Ortega et al., 2004 ▲ V

Coleoptera/Hemiptera coefficient C/H Greewood y Wood, 2003; Ortega et al., 2004; Valladares et al., 2004; ▼ ▼

Indicator species of naturality I+ Millan et al., 1996; 2002; Sanchez-Fernandez et al., 2003 ▼ ▼

Indicator species of degradation I- Sanchez-Fernandez et al., 2003 ▲ ▲

302 Gutierrez-Canovas et al.

Figure 1. Location of the study area, delimitation of Protected Area of Ajauque Wetland and Rambla Salada and sampling stations.Localizacion del area de estudio, Paisaje Protegido de Ajauque y Rambla Salada, y ubicacion de las estaciones de muestreo.

Criteria establishment for metric scoring

Reference stations were selected from theDIVERSAL database, elaborated by the EcologıaAcuatica research group (Universidad deMurcia),containing information about aquatic macroinver-tebrates from numerous saline environments ofthe Iberian Southeast (Granada, Cordoba, Jaen,Almerıa, Albacete, Murcia and Alicante) collec-ted since early 80’s and ongoing. This databaseincludes information of collected individuals atfamily level with the exception of Coleopteraand Hemiptera, which are at species level. Thereference stations were selected according to:a) dominance of natural vegetation and drylandfarming; b) keeping historical salinity values.

The metric scoring criteria were obtainedfrom the fauna collected in reference stations, fo-llowing different procedures for each metric. Thescoring range in the different habitats of the me-

trics richness of macroinvertebrate families (FR)and the Coleoptera/Hemiptera coefficient (C/H)was established, basically, from the 25th and 75th

metric values, after adjustment to avoid overlap-ping between ecotypes. A Kruskal-Wallis non-parametric test was made to detect significant dif-ferences between ecotypes values.

To obtain the Coleoptera and Hemipteraspecies with the best health-indicator capacityan IndVal analysis (Dufrene & Legendre, 1997)was performed using the software PC-ORD4.20. This analysis estimates the indicator value(IV) with regard to the relative abundanceand the occurring frequency of each speciein each previously defined ecotype. The finaldecision about which species are the besthealth-indicators was made taking into accountour taxonomical experience and the availablebibliography. Invasive allocthonous specieswere included as impairment-indicators.

New index for saline “ramblas” assessment 303

Table 3. Sampling sites used for SALINDEX application.Estaciones de muestreo empleadas en la aplicacion deSALINDEX.

A1 Pool on the headwater of the Rambla de AjauqueA4 Rambla de Ajauque downstream of the confluence

with the Sanel WetlandA5 Derramadores wetlandA7 Rambla de Ajauque downstream of the Cabecicos Ne-

grosA8 Rambla de Ajauque downstream of the diversion

channelA9c Rambla de Ajauque at Ajauque wetlandA9h Ajauque wetlandS0 Rambla Salada headwaterS2s Hypersaline spring in Rambla SaladaS2c Rambla Salada downstream of the diversion channelS3 Rambla Salada at Finca de las SalinasS4a Rambla Salada upstream of the confluence with the

Rambla de AjauqueS4d Rambla Salada downstream of the confluence with the

Rambla de AjauqueS5 Rambla Salada upstream of the Santomera Reservoir

Index application

Study area

The Protected Area of Ajauque Wetland andRambla Salada is located in the north-east of theProvince of Murcia (Fig. 1) and is a representa-tive example of the natural ecosystems compo-sed of saline “ramblas” and associated wetlands.This area is protected by environmental lawsat regional level (declared Paisaje Protegido byLey 4/92 de Ordenacion del Territorio), at Eu-ropean level (Special Protection Area and Spe-cial Area of Conservation) due to the singula-rity and integrity of this saline zone and the pre-sence of communities and species of commu-nity interest. The natural vegetation of the ba-sin is halophilic, nitrophilic, gypsyc and ther-mophilic Mediterranean shrubs, although muchis dedicated to irrigated crops.

The climate in the basin is semiarid with amean annual temperature around 18◦ C and meanannual precipitation of below 300 mm, concen-trated in autumn and spring. The study area islocated in the Neogene Abanilla-Fortuna basinwhere gypsiferous marls from the late Mioce-ne are abundant. Badlands are very common as

a consequence of high erosion rates on marlswhich shaped a complex drainage network com-posed of numerous intermittent water coursesthat converge in two main streams, the Ramblaof Ajauque and Rambla Salada, that flow to theSantomera reservoir. Associated to these coursesare many wetlands and saline soils in the floodplain. On the other hand, the water bodies arehighly mineralized as a result of lithology and cli-matic aridity, salinity ranging from 3 to 180 g/L.However, this feature has been modified by afreshening process due to the freshwater inputsfrom surrounding irrigation crops (Ballester etal., 2003) and losses from the Tagus-Segura di-version channel. For instance the salinity of theRambla Salada has fallen from 100 g/L in theearly 1980’s (Vidal-Abarca, 1985) to an averageof 33 g/L in recent years (Velasco et al., 2006).The minimum salinity value was recorded on 2October 2003 when it fell to 3.5 g/L as a result ofthe diversion channel being emptied for repairs.

Assessment of the ecological status of theaquatic ecosystems in the protected area ofAjaque-Rambla Salada

The proposed index for monitoring and contro-lling the ecological status of the Protected Areaof Ajauque-Rambla Salada was applied in 15sampling stations (Table 3 and Fig. 1) which re-present the diversity of the aquatic habitats in thearea according to salinity, lotic or lentic watersand the degree of impairment.

We also included the station at the headwaterof Rambla Salada (S0), even though it was loca-ted outside the protected area, since it is a goodexample of the reference status of ecotype 3 (lo-tic mesosaline) and belongs, of course, to the sa-me basin. The sampling was made on late spring(30/05/2005).

Protocol for index application

Macroinvertebrate collection involved multihabi-tat samping in a 100 m reach following the pro-tocol provided by Jaimez-Cuellar et al. (2002).The collected material was stored in 70% ethanoland, then, a preliminary sorting was made, which

304 Gutierrez-Canovas et al.

was confirmed in the laboratory using taxonomickeys [see Nieser et al. (1994), for Hemiptera, Ri-bera et al. (1998) for a recompilation of the ta-xonomic keys used for Coleoptera and Tachet etal. (2000) for the rest]. Coleoptera and Hemip-tera individuals were sorted at species level whi-le the other orders were classified at family le-vel. Finally, abundance classes were defined asthe number of individuals collected in the reach:rare if occurs less than 3 individuals, common ifbetween 3 and 15 and abundant if more than 15.

For the index, we followed the macroinverte-brate sampling protocol described above. Whenthe fauna composition was known, values ob-tained were compared with scoring criteria foreach metric specified in Table 5. The metrics FR,C/H, I+ scored positively if all the criteria we-re fulfilled. The family richness or the Coleopte-ra/Hemiptera coefficient values, respectively, hadto be inside the score range. The presence ofthe all health-indicator species (I+) added onepoint to the total score. While the occurrence ofany impairment-indicator specie, reduced the to-tal score by one point.

At last, to obtain the global score it is ne-cessary to add all the individual metrics sco-red with no weighting. For each possible result(−1, 0, 1, 2 o 3) an ecological status category(bad, poor, moderate, good or high) exists accor-ding to the classes defined by WFD. The leastperturbed status is represented by the classesgood and high represents the objective for all thewater bodies according to the WFD.

Figure 2. Variation of salinity and discharge during 2003-05due to freshwater inputs to Rambla Salada. Variacion de la sa-linidad y el caudal durante el periodo 2003-05 debido a lasentradas de agua dulce en Rambla Salada.

Index assessment

Data obtained in the intensive survey of theRambla Salada (station S3, Rambla Salada atFinca de las Salinas, lotic hypesaline, ecotype5) during two years were used to analyse theresponse to large changes in salinity and therelationship with other environmental and bio-tic (structural and functional) variables, togetherwith the index IBMWP. Sampling was perfor-med with an almost nearly bimonthly frequencyfrom 10/04/2003 to 07/04/2005. This study iden-tified two different periods according to distur-bance and recovery processes observed: the firstwas disturbed by large inputs of freshwater fromthe diversion channel (from April 2003 to Octo-ber 2003) while during the second period (fromDecember 2003 to May 2005) salinity recovered(Fig. 2). The highest discharges were found whi-le diversion channel was emptied leading to thelowest salinity value (2 October 2003)

The aquatic macroinvertebrates were sam-pled following the previously mentioned proto-col. Using the obtained biological data SALIN-DEX, the IBMWP score was calculated (Alba-Tercedor et al., 2002) using original bounda-ries (labelled as IBMWP-o) (Alba-Tercedor &Sanchez-Ortega, 1988) and the ramblas ecoty-pe boundaries (labelled as IBMWP-r) proposedby the Guadalmed proyect (Alba-Tercedor et al.,2002). In some cases, when the index scoredid not exceed by more than 5 unities the qua-lity class boundary (e.g., between moderate andgood), it was classified as an intermediate class(e.g., moderate-good class), as recommended inJaimez-Cuellar et al. (2002).

On each date, discharge was estimated frommeasurements of depth and current velocityalong a cross-section of the run. Conductivitywas measured in the morning with an ECme-ter (TetraConR 325) that automatically calcula-tes salinity. Stream metabolism rates were mea-sured using an open-system, single-station ap-proach (Odum, 1956). Water temperature anddissolved oxygen were measured in situ at 15 mi-nute intervals over 24 hours on each date usinga multi-parameter recorder (WTW, MultiLineP4). Reaeration coefficient (KS) and ecosystem

New index for saline “ramblas” assessment 305

respiration (ER) were calculated following thenight-time regression method (Thyssen & Ke-lly, 1985) using River Metabolism Estimatorv. 1.2, an MS Excel spreadsheet available athttp://www.cawthron.org.nz that also estima-tes gross primary production (GPP). In this me-thod the reaeration coefficient is obtained fromthe slope of the linear regression between thenight-time rate of change of stream dissolvedoxygen versus the saturation deficit.

Moreover, chlorophyll a, total suspended so-lids and water dissolved nutrients (nitrite, ni-trate, ammonium and orthophosphate), organicmatter in sediment and macroinvertebrate den-sity were calculated using the methods describedin Velasco et al. (2006).

The mean density of each taxon was multi-plied by the relative habitat area in order to ob-tain macroinvertebrate abundance in the reach.Benthic macroinvertebrate biomass was obtained

from length-mass equations available for the sa-me or nearest taxon from saline streams (Mo-reno, 2002; Barahona et al., 2005) or from thegeneral equations for macroinvertebrate families(Smock, 1980; Benke et al., 1999).

Data analyses

All the variables were transformed to normalisedistribution and equalise variance, with the excep-tion of SALINDEX. The relationships betweenthe SALINDEX score and C/H coefficient score,FR score, the IBMWP score and the environ-mental and biotic variables were studied throughSpearman correlations. Furthermore, a one-wayANOVA analysis was carried out to detect mea-ningful differences between impaired dates (bad,poor ormoderate ecological status) and healthy da-tes (good or high). All analyses were conducted byStatistica 6.0 (StatSoft, 2001) software.

Richness of families

Coleoptera/Hemiptera coefficient

Median

25 %-75 %

Range withoutoutliers

Outliers

Figure 3. Boxplot of the metrics family richness and Coleoptera/Hemiptera coefficient for lotic and lentic ecotypes. Boxplot de lasmetricas riqueza de familias y relacion coleopteros/hemıpteros para ecotipos loticos y lenıticos.

306 Gutierrez-Canovas et al.

Table 4. Results of IndVal analysis, showing the indicator va-lue (IV) for each ecotype of species with p-value ≤ 0.1. Resulta-do del analisis IndVal, mostrando el valor indicador (IV) paracada ecotipo de las especies con p-valor ≤ 0,1.

Specie Ecotype IV p

Helophorus fulgidicollis 1 16.7 0.082Herophydrus musicus 2 18.6 0.055Laccophilus minutus 2 20.1 0.052Enochrus falcarius 5 70.7 0.001Nebrioporus baeticus 5 26.7 0.063Ochthebius cuprescens 5 29.9 0.032Parcymus aeneus 5 31.3 0.013Enochrus bicolor 6 22.1 0.053Ochthebius notabilis 6 35.5 0.010Ochthebius glaber 7 65.1 0.001

RESULTS

Development of the SALINDEX

The richness of macroinvertebrate families (FR)decreased when salinity increased, althoughecotypes 3 and 5 (lotic mesosaline and hypersali-ne) showed no differences (Fig. 3). For close sali-nities, the lentic habitats exhibited a lower familyrichness than the lotic habitats. Meaningful dif-ferences were detected between ecotypes for thisvariable (Kruskal-Wallis, H = 25; p = 0.0003).

The Coleoptera/Hemiptera richness coefficient(C/H) increased with salinity, although forsimilar salinities this coefficient is usually lowerin lentic habitats (Fig. 3). No Hemiptera specieswere observed in the ecotype 7 (lotic hyper-saline) due to their lower tolerance to salinity.Significant differences were found in the C/Hcoefficient values (Kruskal-Wallis, H = 17.2;p = 0.0042) between ecotypes.

The species with the highest indicator valuesand p ≤ 0.05 for the different ecotypes areshown in Table 4. Exceptionally, other specieswith p-value ranging between 0.05 and 0.10were included as health-indicators dependingon their ecological preferences (Millan et al.,1996; 2002) and their distribution in the Murciaregion (Sanchez-Fernandez et al., 2003). All thehealth-indicator species were Coleoptera Enochrusfalcarius and Ochthebius glaber being the specieswith the highest degree of habitat specificity forhypersaline lotic waters ( p ≤ 0.001). However, theIndVal analysis did not identity anyhealth-indicatorspecies for ecotypes 3 and 4 (lotic and lenticmesosaline, respectively). Therefore, halophilictaxawere chosen to complete thismetric because ofto their large tolerance to osmotic changes (Millanet al., 1996; 2002; Sanchez-Fernandez et al.,

Table 5. Scoring criteria for each metric and ecotype. See Table 2 for metrics abbreviations. A: > 15 individuals, C: 4-15 individuals,R: < 4 individuals. Condiciones de puntuacion para cada metrica y para cada ecotipo. Ver Tabla 2 para las abreviaturas de lasmetricas. A: > 15 individuos, C: de 4-15 individuos, R: < 4 individuos.

EcotypeMetrics

FR C/H I+ I-

1 10-15 4/1 - 7/1Ochthebius cuprescensHelophorus fulgidicollis

AC Laccophilus hyalinus

2 4-7 2/1 - 3/1Herophydrus musicusLaccophilus minutus

AA Sphaeroma serratum

3 4-10 7/1 - 8/1 Ochthebius cuprescens A

4 4-6 3/1 - 4/1 Enochrus bicolor A Gammarus aequicauda

Ochthebius cuprescens A > RNebrioporus baeticus A Micronecta scholtzi

5 4-9 > 8/1 Enochrus falcarius AParacymus aeneus C Hydroglyphus geminus

Enochrus bicolor A6 1-4 > 4/1 Ochthebius notabilis A Yola bicarinata

7 1-4 NH Ochthebius glaber A

New index for saline “ramblas” assessment 307

Table 6. Results obtained from the application of SALINDEX for each sampling station in the Protected Area of Ajauque-RamblaSalada. Sampling sites abbreviations in Table 3. Resultados obtenidos de la aplicacion de SALINDEX a cada estacion de muestreodel Paisaje Protegido de Ajauque-Rambla Salada. Abreviaturas de las estaciones en la Tabla 3.

Station Ecotype Salinity (g/L)Metrics score

SALINDEX score Ecological statusFR C/H I+ I-

A1 2 009.5 0 0 0 −0 0 PoorA4 4 020.3 0 0 0 −0 0 PoorA5 4 036.3 1 0 0 −0 1 ModerateA7 1 005.5 1 0 0 −0 1 ModerateA8 1 004.5 0 0 0 −0 0 PoorA9h 4 048.5 1 0 1 −0 2 GoodA9c 1 010.5 1 0 0 −1 0 PoorS0 3 030.0 1 1 1 −0 3 HighS2s 7 157.0 1 1 1 −0 3 HighS2d 1 011.0 0 1 0 −0 1 ModerateS3 5 063.0 1 1 1 −0 3 HighS4a 5 039.5 1 0 0 −1 0 PoorS4d 3 025.5 1 0 0 −1 0 PoorS5 2 026.0 1 0 0 −1 0 Poor

2003; Greewood & Wood, 2003), e.g. Ochthebiuscuprescens andEnochrus bicolor.

The selected impairment-indicator species(Table 5) were brackish Crustacean, such as Gam-mnarus aequicauda and Shaeroma serratum, andColeoptera (Laccophilus hyalinus, Hydroglyphusgeminus, Yola bicarinata and the HemipteraMicronecta scholtzi) all of which commonly foundin freshwater and/or eutrophicwaters.

The scoring criteria for each metric and eachecotype are shown in Table 5.

Application of the SALINDEX

Ecological status of the aquatic ecosystems in theprotected area of Ajauque-Rambla Salada

In general, stations with poor or moderateecological status dominated in the protected area.We only found four stations without clear signsof impairment: one with a good ecological statusand three with a high ecological status (Table 6).Rambla Salada showed the best preserved aquatichabitats, with the exception of the reach influencedby irrigated crops or by waters from the Ramblaof Ajauque. In contrast, the aquatic habitats of theAjauque sub-basin were more polluted and onlyone station (Humedal de Ajauque) showed goodecological status. On the other hand, it is important

to highlight the good conservation status (highecological status) of the Rambla Salada headwater(station Cabecera de Rambla Salada, S0) whichis just outside the protected area. Moreover, nostationswith a bad ecological statuswere found.

The family richness (FR) was the metric thatscored most frequently. C/H and I+ appear to bemore selective and, generally, they only scoredwhen the ecological status was good or high. Themetric I- scored in just four stations because ofthe presence of the brackish species Gammarusaequicauda and Sphaeroma serratum.

Assessment of the SALINDEX

The results of the application of the two indi-ces is presented in Table 7: the index propo-sed in this paper (SALINDEX) and the IBMWP,with the original (IBMWP-o) and ramblas ecoty-pe (IBMWP-r) scoring boundaries. SALINDEXclassified the ecological status on four occasionsas poor, three times as moderate and four timesas good. It was able to detect the most intensivefreshening process in October 2003, classifyingit as poor. In addition, a good ecological statuswas assigned by SALINDEX to most of the sam-pling dates following repair of the channel (fromDecember 2003 to April 2005). In contrast, the

308 Gutierrez-Canovas et al.

Table 7. Results obtained from the application of SALINDEX and IBMWP (IBMWP-o: original classes; IBMWP-r: ecotype ram-blas classes) in Finca de las Salinas (Rambla Salada). Resultados de la aplicacion de SALINDEX e IBMWP (IBMWP-o: acotacionesoriginales; IBMWP-r: acotaciones ecotipo ramblas) en la Finca de las Salinas (Rambla Salada).

DateMetrics score

SALINDEX IBMWP score IBMWP-t IBMWP-rFR C/H I+ I-

10/04/2003 0 0 0 0 Poor 33 Poor Good12/06/2003 0 1 0 0 Moderate 38 Poor/Mode Good28/07/2003 0 1 0 0 Moderate 42 Moderate High02/10/2003 0 0 0 0 Poor 61 Poor/Good High11/12/2003 1 1 0 0 Good 26 Poor Good24/03/2004 0 1 0 0 Moderate 28 Poor Good03/06/2004 0 0 0 0 Poor 30 Poor Good29/07/2004 0 0 0 0 Poor 38 Poor/Mode Good14/10/2004 1 1 0 0 Good 29 Poor Good13/01/2005 1 1 0 0 Good 27 Poor Good07/04/2005 1 1 0 0 Good 24 Poor Good

isolated dilution events were not detected as inthe case of April 2005. At last, a good ecologicalstatus was always found when metric FR scored.

Comparing the results obtained with SALIN-DEX with those obtained with IBMWP-o, sixdates were classified with the same ecologicalstatus, two of them being intermediate classes.

The results of the IBMWP-o yielded seven dateswith poor ecological status (twice as poor-moderateintermediate class), one date with a moderateand one date with moderate-good class, the lastcoinciding with the freshening process (October2003). After first time the channel was repaired,salinity recovered pre-dilution levels, although

Table 8. Spearman correlations between SALINDEX, metrics FR, C/H and IBMWP and the environmental and biotic variables(* at p < 0.05 and ** at p < 0.01). Correlaciones de Spearman entre SALINDEX, RF, C/H e IBMWP y las variables ambientales ybioticas estudiadas (* a p < 0.05 y ** a p < 0.01).

Variables SALINDEX FR C/H IBMWP

Water temperature −0.61* −0.66** −0.12* 0.79**Dissolved oxygen 0.64* 0.72** 0.17* −0.56**Oxygen saturation 0.07* 0.06** 0.29* 0.12**Conductivity 0.10* 0.12** 0.12* −0.20**Discharge −0.10* 0.00** −0.23* 0.13**Benthonic organic matter 0.37* 0.24** 0.17* −0.65**Chlorophyll a in water −0.57* −0.72** −0.17* 0.37**Particulated organic matter 0.37* 0.48** −0.12* −0.35**Total suspended solids 0.30* 0.36** −0.12* −0.29**NO3−N 0.03* −0.06** −0.12* 0.04**NO2−N 0.10* 0.18** 0.00* −0.26**NH4−N 0.13* 0.30** 0.23* −0.20**PO4−P 0.49* 0.48** 0.09* −0.12**GPP −0.81* −0.84** −0.29* 0.75**ER −0.74* −0.84** −0.17* 0.78**Biomass of Cladophora glomerata −0.43* −0.36** −0.45* 0.30**Biomass of Enteromorpha instestinalis −0.20* −0.14** −0.07* −0.05**Biomass of epipelon 0.24* 0.24** 0.40* −0.38**Biomass of Ruppia maritima −0.44* −0.48** −0.41* 0.51**Richness of families −0.75* −0.85** −0.32* 0.86**Biomass of macroinvertebrates −0.57* −0.66** −0.23* 0.90**SALINDEX — 0.89** 0.64* −0.73**

New index for saline “ramblas” assessment 309

IBMWP-o did not detect the recovery in theecological status, all the dates being classified aspoor, with the exception of one date. On the otherhand, IBMWP-r, assessed the ecological status asgood on nine dates and as high on two dates, one ofthose being October 2003 which contrast with thephysic-chemical changes observed.

The SALINDEX scores showed a meaningfulnegative correlationwith temperature, FR score andmetabolic rates (GPP and ER), and a significant po-sitive correlation with dissolved oxygen (Table 8).In contrast, the IBMWP score (IBMWP-o andIBMWP-r) presented an opposite pattern andwas positively correlated with temperature, GPP,ER, FR score and macroinvertebrate biomass andnegatively with organic matter in the sediments.Hence, the correlation between both index scoreswas negative. Only the C/H score was positivelycorrelatedwith theSALINDEXscore.

Meaningful differences were observed bet-ween healthy (good or high) and impaired (bad,poor or moderate) dates for some variables.Indeed, temperature (F = 8.6; p ≤ 0.05), Chlo-rophyll a (F = 8.6; p ≤ 0.05), GPP (F = 16.6;p ≤ 0.01), ER (F = 6.6; p ≤ 0.05), family rich-ness (F = 15.1; p ≤ 0.01) and macroinvertebra-te biomass (F = 6.5; p ≤ 0.05) were significantlygreater on perturbed dates while mean dissolvedoxygen values (F = 6.8; p ≤ 0.05) were substan-tially lower on impaired dates.

DISCUSSION

Methodological questions

The development of an index for assessing eco-logical status is not exempt from questions in-herent to a given methodology, such as the dif-ficulty in defining ecotypes and establishing re-ference stations (Reynoldson et al., 2000; Bo-nada et al., 2002; Nijboer et al., 2004) sinceaquatic ecosystems have long suffered stress inthe Mediterranean Basin (Bonada et al., 2002;Sanchez-Montoya et al., 2005; 2007). In this pa-per, we have used the general classification sys-tem for thalassic waters to classify the saline en-vironments with regard to the salinity gradient

Figure 4. Hypothetical response of Coleoptera/Hemipteracoefficient to salinity changes. Respuesta hipotetica de la re-lacion coleopteros/hemıpteros frente a cambios en la salinidad.

described by Montes & Martino (1987). This clas-sification is based on shifts in algal communities(Hammer et al., 1983) and other organisms insaline lakes (Hammer, 1986) as a result of salinitychanges. However, the discontinuities along thesalinity gradient that would determine quantifiablechanges in the macroinvertebrate communitiesof saline streams is not well defined due tothe high degree of tolerance of their biota tophysical-chemical changes and the lackof scientificpapers that study the response of community tosalinity variations (Bunn & Davies, 1992, Nielsenet al., 2003, Pinder et al., 2005, Velasco et al.,2006). For instance, Velasco et al. (2006) in a studyanalysing the shifts in the composition and structureof the community in the face of salinity variations,carried out in Rambla Salada, found substantialdifferences when salinity surpassed 75 g/L. Belowthis level, the differences observed in communitystructure were greater between habitats than thosebetween different levels of salinity. The results ofthe IndVal analysis seems to support this hypothesissince it did not detect any health-indicator taxa formesosaline habitats (ecotypes 3 and 4), probablybecause the eurihaline response of many speciesthat can live in a wide rage of salinity. This suggestthat the upper boundary for mesosaline ecotypescould be higher than 40 g/L although further studiesare required to improve the classification of salineaquatic ecosystems in the Iberian Peninsula, which

310 Gutierrez-Canovas et al.

would improve the robustness of the index andits area of application. The meaningful differencesfound in the coefficient values for the familyrichness and the Coleoptera/Hemiptera species bet-ween the proposed ecotypes appear to support theproposed classification, at least for the study area.

Response of the different metrics

Richness of families

The richness of families metric responded linea-lly in the fact of salinity changes, decreasing assalinity increased. In the Rambla Salada, whenthe FR metric scored (when richness of familieswas within the metric scoring range for referen-ce sites for each ecotype), the ecological statusclass was, at least, good. Moreover, the SALIN-DEX score and the richness of families value we-re negatively correlated. The positive correlationfound between the FR metric score and Chlo-rophyll a in water (impairment indicator) sup-ports the usefulness of this metric. Boix et al.(2005) also found a negative correlation betweenthe RIC index, based on the richness of Crusta-ceans and Insect taxa, and the Chlorophyll a anda positive correlation with dissolved oxygen. TheFRmetric is easy to use and does not require hightaxonomic knowledge. To identify macroinverte-brate families, the graphic taxonomic keys of Ta-chet et al. (2000) are recommended.

Coleoptera/Hemiptera ratio

The C/H coefficient increased with salinity, sincethe richness of Coleoptera species decreased toa lesser extent than Hemiptera species. Neverthe-less, for salinity levels above 70 g/L, the C/H ten-ded to show a non-lineal response with regard tosalinity (Fig. 4). In the 70-100 range this coeffi-cient decreased because only one Hemiptera spe-cie, Sigara selecta, tolerates this degree of sali-nity (Velasco et al., 2006) and only a small onColeoptera species are able to support these con-ditions. Above 100 g/L the C/H coefficient tren-ded to increase again because Hemiptera speciesdisappeared. This metric appears to be inappro-priate for assessing hypersaline habitats (ecoty-

pes 5, 6 and 7) and we recommended it be exclu-ded from in the index for these ecotypes.

Indicator species

Adding health indicator species to SALINDEXcompletes and improves the information given bythe FR metric which is quite demanding since itonly scores when all the health-indicator speciesfor the ecotype occur in sufficient abundance.On the other hand, the inclusion of impairment-indicator species in the index, including alloc-thonous species, lowers the score in the case ofthe occurrence of any of such species, reflec-ting their impact in the native community. If the-se species are present, the index cannot classifythe site as high. Thus, the occurrence of exoticspecies is a criterion that can be used to discardsites in the reference station selection process(Barbour et al., 1999; Sanchez-Montoya et al.,2005; Kennard et al., 2006).

For their application both metrics require hightaxonomic resolution which must be considereda disadvantage. However, the low species rich-ness in saline habitats and the use of a graphicguide with photographs and distinguishing fea-tures of the indicator species may facilitate theiruse. Besides, the use of Coleoptera and Hemipte-ra at family level has no indicator value in sa-line habitats due to different habitat preferen-ces of the species to which belong these fami-lies, there being as many halophilic species asopportunistic species (e.g., Hydraenidae, Dytis-cidade, Hydrophilidae, Corixidae). Indeed, thetaxonomic resolution employed in the metricsI+ and I- seems to be sufficient.

Response of the SALINDEX to anthropicperturbation

The application of SALINDEX in the RamblaSalada at Finca de las Salinas during the period2003-05, identified the dates when the systemsuffered a great dilution as a result of the massi-ve input of freshwater from the Tagus-Segura di-version channel accident and confirmed the sys-tem’s recovery after its repair. A clear response

New index for saline “ramblas” assessment 311

to perturbation was observed from the negativecorrelation of the SALINDEX score with GPPand ER. The freshwater inputs together with nu-trient enrichment produced an increase in fila-mentous algae (Velasco et al., 2006) and, in turn,an increase in metabolic rates (GPP and ER) andthe richness and biomass of macroinvertebrates.The metabolic rates, as other functional varia-bles, seemed to respond clearly to anthropogenicperturbations (Gessner & Chauvert, 2002; Younget al., 2004) reflecting shifts in ecosystem func-tioning, and could therefore be considered use-ful indicators for saline habitat assessment. How-ever, the index appears to be less sensitive in de-tecting light perturbations, as occurred in April2005 when freshwater input was short. Lastly, theindex score was positively correlated with dissol-ved oxygen which indicates good quality water.

Comparison with the IBWMP

The inverse relation found between the SALIN-DEX and IBMWP was to be expected becau-se of the differing composition and structure ofthe macroinvertebrate communities in freshwa-ter and saline streams, together with the negativeresponse of the family richness to salinity. Theinitial SALINDEX hypothesis is that, in a natu-ral habitat, saline environments exhibit lower ta-xonomic richness than analogous freshwater ha-bitats and a hypothetical perturbation would in-crease the taxonomic richness due to saline stressreduction. In contrast, the IBMWP is based onthe inverse hypothesis and, in this case, the res-ponse of the community to pollution would be anincrease in tolerant species (low score) and a de-creasing in intolerant species (high score). Themaximum IBMWP scores were obtained whenthere were a lot of intolerant families, such asthose belonging to Plecoptera, Trichoptera andEphemeroptera (Ward, 1992; Williams & Felma-te 1992; Vivas et al., 2002) which do not ap-pear in saline streams, or do so intermittently.Other Odonata families, such as Aeshnidae,Coenagrionidae or Libellulidae, which com-monly occur in lentic habitats of low minerali-zation (Vivas et al., 2002), score quite high in thementioned index (Alba-Tercedor et al., 2002) al-

though their presence in saline “ramblas” couldpoint to freshwater inputs. Nevertheless, in sali-ne streams and wetlands the dominant taxa areDiptera, Coleoptera and Hemiptera (Ward, 1992)whose score in the IBMWP, in general, is low.Thus, if the IBMWP is applied to hypersaline“ramblas” with salinity above 100 g/L, where thereference community is composed of one or twoColeoptera species and some Diptera species, thescore will always be low, indicating impairment.Indeed, in the present contribution, the IBMWPreached its maximum score in Rambla Saladaduring the long dilution period, classifying theecological status as moderate-good using originalboundaries and high using the “ramblas” ecoty-pe boundaries. Thus, the IBMWP was not ableto detect a clear dilution perturbation or recoveryof the system after freshwater input cessation.Such results agree with the negative relation bet-ween the IBMWP score and conductivity foundby Alba-Tercedor et al. (2002).

The SALINDEX and IBMWP indices weregenerally related with the same environmentalvariables but in an opposing way. For examplea seasonal influence was observed in the sco-res of the two indices, temperature being co-rrelated negatively with the SALINDEX scoreand positively with IBMWP score. This couldhave been due, partially, to their relationshipwith variables highly dependent on tempera-ture o radiation such as the metabolic rates(Velasco et al., 2003) or the macroinvertebratebiomass (Barahona et al., 2005).

Given the results obtained, the IBMWP can-not be considered suitable for assessing the eco-logical status of saline streams, whether usingoriginal boundaries or the proposed “ramblas”ecotype. This ecotype has shown great variabi-lity in its macroinvertebrate communities and in-cludes water bodies from a wide range of minera-lization and water permanence, for which reasonwe recommend its division into subgroups for thecorrect application of WFD (Sanchez-Montoyaet al., 2007). In the same way, other indices de-veloped for freshwater streams, such as the indexof canopy vegetation quality (QBR, Qualitat delBosc de Ribera), have shown similar problemswhen applied to “ramblas” (Suarez et al., 2002).

312 Gutierrez-Canovas et al.

Application area and future perspectives forSALINDEX

The application of SALINDEX in the diffe-rent aquatic ecosystems in the protected area ofAjauque-Rambla Salada identified the best pre-served reaches and hydrological sectors, in addi-tion to the most perturbed areas showing, gene-rally, a better ecological status the aquatic habi-tats placed on Rambla Salada in comparison withthose located on Ajauque. The index assigneda good or high ecological status to the stationssurrounded by dryland farming (mainly cereals)and/or natural gypsic and halophilic vegetation,where freshwater and nutrient inputs were irrele-vant. On the contrary, the stations where pertur-bation was detected were influenced by irrigationcrops or organic pollution.

In the beginning, SALINDEX was designedas a component of an ecological indicator sys-tem for monitoring and controlling the ecolo-gical status of the Protected Area of AjauqueWetland and Rambla Salada (Project Interreg IIIB Espacio MEDOC MedWet/Regiones). Such asystem included, moreover, other indices or me-trics based on hydrochemical features, vegeta-tion, land uses, terrestrial invertebrates and aqua-tic and terrestrial vertebrates for the assessmentof terrestrial and aquatic ecosystems of the pro-tected area (Alvarez y Gomez, 2005). Never-theless, the area to which SALINDEX couldbe applied includes the Iberian Southeast where“ramblas” and associated wetlands are the mostfrequent saline habitat.

In conclusion, SALINDEX is designed in ac-cordance with ecological patterns typical of in-land saline ecosystems (low taxonomic richness,predominance of Diptera, Coleoptera and He-miptera, presence of halotolerant and halophi-lic species) and is the first tool specifically ela-borated for these kinds of habitats. Finally, al-though some weaknesses were detected in the in-dex and now are being studied by the authors ofthis paper, SALINDEX was able to detect themost important dilution processes and the sys-tem’s recovery to date, then, it seems the mostsuitable tool for assessing the ecological status ofsaline “ramblas” and wetlands.

ACKNOWLEDGEMENTS

We thank Pedro Abellan, David Sanchez, Jose Ba-rahona, Juan Hernandez, Mar Ruiz y RocıoAlcantara for their assistance in field sampling andsample processing and Arturo Elosegui for usefulsuggestions. Thanks also toMiguel Angel Nunez ofthe Centro de Interpretacion de Rambla Salada fortheir generous support and assistance in the field andto J. Faustino Martınez, Director of the ProtectedArea of Humedal de Ajauque and Rambla Saladafor the permitting of sampling in the study area.Thanks also to Jorge Alvarez for his help in figuredesign. This researchwas supported by the researchproject BOS2002-00702 (Spanish Investigation,Development and InnovationProgram)andpartiallyby a predoctoral grant from the Seneca Foundation,ScienceandTechnologyAgencyofMurciaRegion.

REFERENCES

ALBA-TERCEDOR, J. y A. SANCHEZ-ORTEGA.1988. Un metodo rapido y simple para evaluar lacalidad biologica de las aguas corrientes basado enel de Hellawell (1978). Limnetica, 4: 51-56.

ALBA-TERCEDOR, J., P. JAIMEZ-CUELLAR, M.ALVAREZ, J. AVILES, N. BONADA, J. CA-SAS, A. MELLADO, M. ORTEGA, I. PARDO,N. PRAT, M. RIERADEVALL, S. ROBLES, C.E. SAINZ-CANTERO, A. SANCHEZ-ORTEGA,M. L. SUAREZ, M. TORO, M. R. VIDALABAR-CA, S. VIVAS y C. ZAMORA-MUNOZ. 2002.Caracterizacion del estado ecologico de rıos me-diterraneos ibericos mediante el ındice IBMWP(= BMWP′). Limnetica, 21: 175-185.

ALVAREZ, J. y R. GOMEZ (Coord.). 2005.Diseno eimplementacion del plan de seguimiento del Espa-cio Natural Protegido Humedal de Ajauque y Ram-bla Salada. Direccion General del Medio Naturalde la Comunidad Autonoma de Murcia. Murcia.427 pp.

ARMITAGE, P. D., D. MOSS, J. F. WRIGHT & M.T. FURSE. 1983. The performance of a new bio-logical water quality score system based on ma-croinvertebrates over a wide range of unpollutedrunning-water sites. Water Res., 17: 333-347.

BALLESTER, R., M. R. VIDAL-ABARCA, M. A.ESTEVE y M. L. SUAREZ. 2003. Los humeda-

New index for saline “ramblas” assessment 313

les de la Region de Murcia. Humedales y ramblasde la Region de Murcia. Consejerıa de Agricultura,Agua y Medio Ambiente. Murcia. 137 pp.

BARAHONA, J., A. MILLAN & J. VELASCO.2005. Population dynamics, growth and produc-tion of Sigara selecta (Fieber, 1848) (Hemiptera,Corixidae) in a Mediterranean hypersaline stream.Freshwat. Biol., 50: 2101-2113.

BARBOUR, M. T., J. B STRIBLING & J. R. KARR.1995. Multimetric approach for establishing bio-criteria and measuring biological condition. In:Biological assessment and criteria. Tools for wa-ter resource planning and decision making. W.S.Davis and T.P. Simon (eds.): 63-77. Lewis Publis-hers, Boca Raton, Florida.

BARBOUR, M. T., J. GERRITSEN, B. D. SNYDER& J. B. STRIBLING. 1999. Rapid bioassesmentprotocols for use in streams and wadeable rivers:Periphyton, Bentic Macroinvertebrates and Fish,2nd ed. EPA 841-B-99-002. USEPA, Office of Wa-ter, Washington DC. 339 pp.

BENKE, A. C., A. D. HURYN, L. A. SMOCK& J. B.WALLACE. 1999. Length–mass relationships forfreshwater macroinvertebrates in North Americawith particular reference to the southeastern UnitedStates. J.N.Am.Benthol. Soc.,18: 308-343.

BOIX, D., S. GASCON, J. SALA, M. MARTINOY,J. GIFRE & X. D. QUINTANA. 2005.Anew indexof water quality assessment in Mediterraneanwetlands based on crustacean and insect assembla-ges: the case of Catalunya (NE Iberian peninsula).Aquat.Conserv.-Mar.Freshw.Ecosyst.,15: 635-651.

BONADA, N., N. PRAT, A. MUNNE, M. RIERA-DEVALL, J. ALBA-TERCEDOR, M. ALVAREZ,J. AVILES, J. CASAS, P. JAIMEZ-CUELLAR,A. MELLADO, G. MOYA, I. PARDO, S. RO-BLES, G. RAMON, M. L. SUAREZ, M. TO-RO, M. R. VIDAL-ABARCA, S. VIVAS y C.ZAMORA-MUNOZ. 2002. Ensayo de una tipo-logıa de las cuencas mediterraneas del proyectoGUADALMED siguiendo las directrices de la Di-rectiva Marco del Agua. Limnetica, 21: 77-78.

BONADA, N., N. PRAT, V. H. RESH & B. STATZ-NER. 2006. Developments in aquatic insect bio-monitoring: a comparative analysis of recent ap-proaches. Annu. Rev. Entomol., 51: 495-523.

BUNN, S. E. & P. M. DAVIES. 1992. Communitystructure of the macroinvertebrate fauna and wa-ter quality of a saline river system in south-westernAustralia. Hydrobiologia, 248(2): 143-160.

DUFRENE, M. & P. LEGENDRE. 1997. Species as-semblage and indicator species: the need for aflexible asymmetrical approach. Ecol. Monogr.,67: 345-366.

GESSNER, M. O. & E. CHAUVET. 2002. A case forusing litter breakdown to assess functional streamintegrity. Ecol. Appl., 12(2): 498-510.

GOMEZ R., I. HURTADO, M. L. SUAREZ & M. R.VIDAL-ABARCA. 2005. Ramblas in south-eastSpain: threatened and valuable ecosystems. Aquat.Conserv., 15: 387-402.

GREENWOOD, M. T. & P. J. WOOD. 2003. Ef-fects of seasonal variation in salinity on a popula-tion of Enochrus bicolor Fabricius 1792 (Coleop-tera: Hydrophilidae) and implications for otherbeetles of conservation interest. Aquat. Conserv.-Mar. Freshw. Ecosyst., 13: 21-34.

HAMMER, U. T., J. SHAMES & R. C. HAYNES.1983. The distribution and abundance of algae insaline lakes of Saskatchewan, Canada. Hydrobio-logia, 16: 1-26.

HAMMER, U. T. 1986. Saline lakes: distribution anduses. In: Evaluating saline waters in a plains envi-ronment. D. T. Waite (ed.): 1-22. Universidad deRegina.

JAIMEZ-CUELLAR, P., S. VIVAS, N. BONADA,S. ROBLES, A. MELLADO, M. ALVAREZ, J.AVILES, J. CASAS, M. ORTEGA, I. PARDO,N. PRAT, M. RIERADEVALL, C.E. SAINZ-CANTERO, A. SANCHEZ-ORTEGA, M. L.SUAREZ, M. TORO, M. R. VIDAL-ABARCA,C. ZAMORA-MUNOZ y J. ALBA-TERCEDOR.2002. Protocolo GUADALMED (PRECE). Limne-tica, 21: 187-204.

KENNARD, M. J., B. D. HARCH, B. J. PUSEY &A. H. ARTHINGTON. 2006. Accurately definingthe reference condition for summary biotic me-trics: a comparison of four approaches. Hydrobio-logia, 572: 151-170.

METCALFE-SMITH, J. L. 1994. Biological water-quality assessment of rivers: use of macroinverte-brate communities. In: The rivers handbook. Vo-lume 2. Hydrological and ecological principles.P. Calow and G. E. Petts (eds.): 144-170. Black-well Science, Oxford, UK.

MILLAN, A., J. VELASCO, M. R. VIDAL-ABAR-CA, M. L. SUAREZ y L. RAMIREZ-DIAZ. 1996.Distribucion espacial de los Adephaga acuaticos(Coleoptera) en la cuenca del rıo Segura (SE dela Penınsula Iberica). Limnetica, 12: 13-29.

314 Gutierrez-Canovas et al.

MILLAN, A., J. L. MORENO y J. VELASCO. 2001.Estudo faunıstico y ecologico de los coleopteros yheteropteros acuaticos de las lagunas de Albacete(Alboraj, Los Patos, Ojos de Villaverde, Ontalafiay Petrola). Sabuco, 1: 43-94.

MILLAN, A., J. L. MORENO y J. VELASCO. 2002.Coleopteros y Heteropteros de la provincia de Al-bacete: Estudio faunıstico y ecologico. Instituto deEstudios Albacetenses. Albacete. 180 pp.

MONTES, C. y P. MARTINO. 1987. Las aguas sali-nas espanolas. En: Bases cientıficas para la protec-cion de humedales de Espana: 95-145. Real Aca-demia de Ciencias Exactas, Fısicas y Naturales.Madrid.

MORENO, J. L. 2002. Habitats, recursos troficos yestructura de la comunidad de macroinvertebra-dos bentonicos en un arroyo salino del Sures-te Iberico (Rambla del Reventon). Tesis doctoral,Universidad de Murcia. 273 pp.

NIELSEN, D. L., M. A. BROCK, K. CROSSLE, K.HARRIS, M. HEALEY & I. JAROSINSKI. 2003.Does salinity influence aquatic plant and zooplank-ton communities developing from the sediment oftwo wetlands. Freshwat. Biol., 48: 2 214-2 223.

NIESER, N., M. BAENA, J. MARTINEZ-AVILES yA. MILLAN. 1994. Claves para la identificacionde los Heteropteros acuaticos (Nepomorpha y Ge-rromorpha) de la Penınsula Iberica (con notas so-bre las especies de las islas Azores, Baleares, Ca-narias y Madeira). Asociacion Espanola de Lim-nologıa, 5. Barcelona. 112 pp.

NIJBOER, R. C., R. K. JOHNSON, P. F. M VER-DONSCHOT, M. SOMMERHAUSER & A. BUF-FAGNI. 2004. Establishing reference conditionsfor European streams.Hydrobiologia, 516: 93-107.

ORTEGA, M., J. VELASCO, A. MILLAN, & C.GUERRERO. 2004. An Ecological Integrity Indexfor Littoral Wetlands in Agricultural Catchmentsof Semiarid Mediterranean Regions. Environ. Ma-nag., 33: 412-430.

PAULSEN, S. G., D. P. LARSEN, P. R. KAUF-MANN, T. R. WHITTIER, J. R. BAKER, D.PECK, J. MCGUE, R. M. HUGHES, D. MC-MULLEN, D. STEVENS, J. L. STODDARD, J.LAZORCHAK, W. KINNEY, A. R. SELLE, &R. HJORT. 1991. EMAP-surface waters moni-toring and research strategy, fiscal year 1991.EPA-600-3-91-002. USEPA, Office of Researchand Development, Washington, D.C. and Environ-mental Research Laboratory, Corvallis, Oregon.183 pp.

PINDER, A. M., S. A. HALSE, J. M. MCRAE & R.J. SHIELD. 2005. Ocurrence of aquatic invertebra-tes of the weatbelt region of Western Australia inrelation to salinity. Hydrobiologia, 543: 1-24.

PRAT, N. 2002. El Proyecto GUADALMED. Limne-tica, 21: 1.

RAMIREZ-DIAZ, L. (coord.). 1992. Los Humedalesde la Region de Murcia. Caja de Ahorros del Me-diterraneo, Murcia. 109 pp.

REYNOLDSON, T. B. & J. F. WRIGHT. 2000. Thereference condition: problems and solutions. In:Assessing the Biological Quality of Fresh Wa-ters: RIVPACS and Other Techniques. J. F. Wright,D. W. Sutcliffe &M. T. Furse (eds.): 293-303. Am-bleside UK: Freshw. Biol. Assoc.

RIBERA, I., C. HERNANDO & P. AGUILERA.1998. An annotated checklist of the Iberian waterbeetles (coleoptera). Zapateri, 8: 43-111.

ROSENBERG, D. M. & V. H. RESH. 1993. Fresh-water Biomonitoring and Benthic Macroinverte-brates. Chapman and Hall, New York. 488 pp.

SANCHEZ-FERNANDEZ, D., P. ABELLAN, J.VELASCO y A. MILLAN. 2003. Los coleopterosacuaticos de la Region de Murcia. Catalogofaunıstico y areas prioritarias de conservacion.Monografıas SEA, 10. Zaragoza. 71 pp.

SANCHEZ-MONTOYA, M. M., M. L. SUAREZ yM. R. VIDAL-ABARCA. 2005. Propuesta de cri-terios para la seleccion de estaciones de referenciaen rıos mediterraneos en el contexto de la DirectivaMarco de Agua. Tecnologıa del Agua, 162: 42-52.

SANCHEZ-MONTOYA, M. M., T. PUNTI, M. L.SUAREZ, M. R. VIDAL-ABARCA, M. RIE-RADEVALL, J. M. POQUET, C. ZAMORA-MUNOZ, S. ROBLES, M. ALVAREZ, J. ALBA-TERCEDOR, M. TORO, A. M. PUJANTE, A.MUNNE & N. PRAT. 2007. Concordance bet-ween ecotypes and macroinvertebrate assembla-ges in Mediterranean streams. Freshwat. Biol.,52(11): 2 240-2 255.

SMOCK, L. A. 1980. Relationships between body si-ze and biomass of aquatic insects. Freshwat. Biol.,10: 375-383.

STATSOFT, INC. 2001. STATISTICA (data analysissoftware system), version 6.0. www.statsoft.com.

SUAREZ, M. L., M. R. VIDAL-ABARCA, M. M.SANCHEZ-MONTOYA, J. ALBA-TERCEDOR,M. ALVAREZ, J. AVILES, N. BONADA, J. CA-SAS, P. JAIMEZ-CUELLAR, A. MUNNE, I.PARDO, N. PRAT,M. RIERADEVALL,M. J. SA-LINAS, M. TORO y S. VIVAS. 2002. Las riberas

New index for saline “ramblas” assessment 315

de los rıos mediterraneos y su calidad: el uso delındice QBR. Limnetica, 21: 135-148.

TACHET, H., RICHOUX, P., BOURNAUD, M. et P.USSEGLIO-POLATERA. 2000. Invertebres d’eaudouce. Systematique, biologie, ecologie. CNRSEd. Paris. 588 pp.

THYSSEN, N. & M. G. KELLY. 1985 Water-air ex-change of carbon dioxide and oxygen in a river:Measurement and comparison of rates. Arch. Hy-drobiol., 105: 219-228.

VALLADARES, L. F. y D. MIGUELEZ. 2004. Fau-na actual de Coleopteros y Heteropteros de la tur-bera fosil de Espinosa de Cerrato (Palencia). Bo-letın de la Asociacion espanola de Entomologıa,28(1-2): 71-88.

VELASCO, J., A. MILLAN, M. R.VIDAL-ABAR-CA, M. L. SUAREZ, C. GUERRERO, & M. OR-TEGA. 2003. Macrophytic, epipelic and epilithicprimary production in a semiarid Mediterraneanstream. Freshwat. Biol., 48: 1 408-1 420.

VELASCO, J., A. MILLAN, J. HERNANDEZ, C.GUTIERREZ-CANOVAS, P. ABELLAN, D.SANCHEZ-FERNANDEZ & M. RUIZ. 2006.Response of biotic communities to salinity chan-ges in a Mediterranean hypersaline stream. SalineSystems, 2: 12.

VIDAL-ABARCA, M. R. 1985. Las aguas superfi-ciales de la cuenca del rıo Segura (SE. De Es-pana). Caracterizacion fısico-quımica en relacional medio fısico y humano. Tesis doctoral, Univer-sidad de Murcia. 789 pp.

VIDAL-ABARCA, M. R., M. A. ESTEVE y M. L.SUAREZ. 2001. Revision y actualizacion del Inven-tario Regional de Zonas Humedas. Consejerıa deAgricultura, Agua y Medio Ambiente, ComunidadAutonoma de la Region de Murcia, Murcia. 127 pp.

VIVAS, S., J. CASAS, I. PARDO, S. ROBLES, N.BONADA, A. MELLADO, N. PRAT, J. ALBA-TERCEDOR, M. ALVAREZ, M.M. BAYO, P.JAIMEZ-CUELLAR, M.L. SUAREZ, M. TORO,M.R. VIDAL-ABARCA, C. ZAMORA-MUNOZ,N. PRAT y G. MOYA. 2002. Aproximacion multi-variante en la exploracion de la tolerancia ambien-tal de las familias de macroinvertebrados de losrıos mediterraneos del proyecto GUADALMED.Limnetica, 21: 149-173.

WARD, J. V. 1992. Aquatic Insect Ecology. Biologyand Habitat. John Wiley & Songs, Toronto. 456 pp.

WILLIAMS, W. D. 1981. The limnology of saline la-kes in western Victoria. Hydrobiologia, 82: 233-259.

WILLIAMS, D. D. & B. W. FELTMATE. 1992.Aquatic Insect. CAB International, Oxon. 358 pp.

WILLIAMS, W. D., A. J. BOULTON & R. G.TAAFFE. 1990. Salinity as a determinant of saltlake fauna: a question of scale. Hydrobiologia,197: 257-266.

YOUNG, R. G., C. R. TOWNSEND & C. MATT-HAEI. 2004. Functional Indicators of RiverEcosystem Health-An interim guide for use in NewZealand. Cawthron, Dunedin, New Zealand (Re-port No 870). 60 pp.

Limnetica, 27 (2): x-xx (2008)Limnetica, 27 (2): 317-328 (2008)c© Asociacion Iberica de Limnologıa, Madrid. Spain. ISSN: 0213-8409

Benthic diatoms and macroinvertebrates in the assessment of theecological status of Azorean streams

Vıtor Goncalves 1, Pedro Raposeiro 1,∗ and Ana Cristina Costa 1

1 CIBIO-Polo Acores e Departamento de Biologia, Universidade dos Acores, R. Mae de Deus, 13A, 9501-801Ponta Delgada, Portugal2

∗ Corresponding author: raposeiro@uac.pt2

Received: 17/10/07 Accepted: 18/7/08

ABSTRACT

Benthic diatoms and macroinvertebrates in the assessment of the ecological status of Azorean streams

To meet the Water Framework Directive goals all freshwater ecosystems have to be categorized by 2015. This paper analyzesthe benthic diatoms and the macroinvertebrates of streams in two islands (Sao Miguel and Santa Maria) of the Azoreanarchipelago. Differences were found between the epilithic diatom communities located upstream (forested areas) and down-stream (urban and agricultural) of the water’s courses., The diatom-based indices revealed that 10% of the sites presenteda good/excellent state of conservation. The macroinvertebrates indices confirmed the generally poor ecological condition ofthese streams. However, these results must be interpreted with care due to the low biodiversity and heavy rainfall conditionsin these island streams. The existence of these specialized conditions requires a different approach to correctly evaluate theecological quality of the archipelago streams.

Key words: Water framework directive, benthic diatoms, benthic macroinvertebrates, ecological quality indices, islands,Azores.

RESUMEN

Diatomeas bentonicas y macroinvertebrados en la valoracion del estado ecologico de los rıos de las Azores

Para cumplir la Directiva Marco del Agua todos los ecosistemas de agua dulce deben ser categorizados antes de 2015. En elpresente trabajo se analizan las diatomeas bentonicas y los macroinvertebrados en los rıos de dos islas (San Miguel y SantaMarıa) del archipielago de las Azores. Se encontraron diferencias entre las comunidades de diatomeas epilıticas localizadasen la parte superior (areas forestales) e inferior (areas urbanas y agrıcolas) de los cursos de agua. Los ındices basados enlas diatomeas revelaron que un 10% de los lugares presentaban un estado de conservacion bueno/excelente. Los ındices demacroinvertebrados confirmaron las condiciones ecologicas generalmente pobres de estos rıos. Sin embargo, estos resultadosdeben ser interpretados con cuidado debido a la baja biodiversidad y el caracter torrencial de los rıos insulares. La existenciade estas condiciones especiales requiere un enfoque diferente para evaluar correctamente la calidad ecologica de los rıos delarchipielago.

Palabras clave: Directiva marco del agua, diatomeas bentonicas, macroinvertebrados bentonicos, ındices de calidad ecologi-ca, islas, Azores.

INTRODUCTION

The Water Framework Directive (WFD) (Euro-pean Parliament & The Council of the European

Union, 2000) introduced the concept of ecologi-cal status for the assessment of aquatic ecosys-tems, putting the emphasis on biological end-points to assessing environmental health.

318 Goncalves et al.

The ecological status concept overcomes the li-mitations of assessing ecological conditions ba-sed on physico-chemical measurements. Bioticindicators integrate environmental changes overa long time period, whereas physical-chemicalmeasurements report anthropogenic pressures atthe time of sampling (Cranston et al., 1996;Giller & Malmqvist, 1998; US EPA, 2002).

The monitoring programmes, established forthe purpose of estimating values of the biologi-cal quality elements for each surface category,may utilise particular species or groups of specieswhich are representative of the quality element asa whole (European Parliament & The Council ofthe European Union, 2000). The comparison overtime of community conditions in pristine controlareas –reference conditions– with those of siteswhose ecological status is under assessment isthe principle under biomonitoring. The ecologi-cal status should be expressed as quality ratiosrepresenting the relationship between the valuesof the biological parameters observed for a givensurface water body and the values for the sameparameters in the reference conditions applicableto that body (European Parliament & The Coun-cil of the European Union, 2000).

Fish could be an important element in aqua-tic ecosystems, but their presence in Azorean in-land waters is mostly limited to lentic systemsand exclusively due to human introductions. Inthe Azores, microalgae (planktonic and benthic)and benthic macroinvertebrates are the key ele-ments in monitoring the Azorean streams; in par-ticular, diatoms (Kolkwitz &Marsson, 1908), ha-ve been used to assess the ecological quality ofrunning waters. Diatoms respond quickly to en-vironmental changes in rivers and streams (Ste-venson & Pan, 1999) and are excellent indicatorsof local conditions (Chasse, 1997). Many waterquality assessment methods with diatoms havebeen developed (Prygiel & Coste, 1993), inclu-ding several diatom indices (Descy, 1979; Cema-greff, 1982; Sladecek, 1986; Leclercq & Maquet,1987; Coste &Ayphassorho, 1991; Descy &Cos-te, 1991; Kelly & Whitton, 1995; Prygiel & Cos-te, 2000) that were shown to be one of the mosteffective tools for the evaluation of the ecologicalstatus of rivers (Eloranta & Soininen, 2002).

Benthic macroinvertebrates are common inha-bitants of freshwater systems around the globeand are among the most sensitive elements ofaquatic biota. They are an important food sour-ce for fishes, and participate actively in decom-position processes influencing nutrient recyclingfor primary production. Macroinvertebrates havebeen widely used as water quality bioindicators(e.g. Rosenberg & Resh, 1993; Metcalfe-Smith,1994; Alba-Tercedor, 1996). The advantages ofusing this fauna component are mostly relatedto abundance and/or presence in different aqua-tic habitats, taxa selective sensitivities to a va-riety of pollutants and contamination levels, rela-tively long life cycles, and short re-colonizationperiods (Pujante, 1997). Moreover, the collec-tion and identification of benthic macroinverte-brates is relatively easy and there are standardprotocols for most used groups for quality indi-ces (e.g. BWMP, ASPT, BBI). Several benthicmacrofauna-based indices have been proposedand developed with the objective of environmen-tal quality assessment of aquatic ecosystems (Ro-senberg & Resh, 1993). However, the knowled-ge of freshwater benthic diatoms and macroin-vertebrates, especially in Azorean streams, is inits infancy and mainly historical (Archer, 1874;Guerne, 1887; Barrois, 1896). Recently, their ro-le in the streams ecology has been addressed(e.g. Goncalves et al., 2005).

The aim of the present study was to evaluatethe effectiveness of the methodologies applied inIberian lotic systems to island streams.

MATERIAL AND METHODS

Study Area

The Azores, located at 36◦55′43′′-39◦43′2′′ North,and 24◦46′15′′-31◦16′15′′ West, is an oceanicarchipelago comprising nine islands and severalislets (Fig. 1). Being 1 300 km apart from thenearest continental coast (Cabo da Roca-Portugal),and 1 900 km from the American Continent,it is the most remote Macaronesian archipelago.Due to their oceanic situation and volcanicorigin, the freshwater systems of the Macaronesian

Ecological status of azorean streams 319

Figure 1. Location of the sampling sites in the Azorean archipelago. Localizacion de los puntos de muestreo en el archipielago delas Azores.

Islands differ strongly from continental systemsin watershed morphology and on their bioticassemblage’s composition (Hughes&Malmqvist,2005). Insular watersheds, originated fromvolcanic processes, are characteristically smalland very steep (Hughes, 2003; Smith et al.,2003). Streams drop dramatically in altitude overa very short distance and are similar to continentalheadwater streams, being narrow, straight andshallow with turbulent, torrential, and oftenseasonal flow. Substrates are coarse comprisingbedrock, boulders, cobbles, and sand. The waterchemical composition results mainly from threeprocesses: i) atmospheric inputs by oceanic rains,ii)weatheringof the volcanic rocks of the drainagebasin, and iii) inflows of more concentratedthermal spring waters (Louvat & Allegre, 1998).According to the geologicalmapofSaoMiguel Is-

land (Zbyszewski et al., 1958, 1959), the streamssampled in the present study belong to watershedsofbasaltic, trachyticor intermediate rocks.

Twenty five sites, selected from 11 catch-ments north and south of Sao Miguel and San-ta Maria islands (Table 1), covered all perma-nent freshwater lotic systems of these islands,and included a range of water quality from‘pristine’ to highly organically polluted or suf-fering from extreme anthropogenic disturbance.The sites were sampled during 2003 (Octoberand November) and 2004 (March and April),from upstream to downstream along a 100 mlong reach. At each site physical and chemi-cal measurements of the stream’s water weretaken simultaneously with the biological sam-ples by the Instituto de Inovacao Tecnologi-ca dos Acores (INOVA). The results of the

320 Goncalves et al.

Table 1. Sampling sites and their respective codes. Estacionesde muestreo y sus respectivos codigos.

Island Stream Local Code

Sao Miguel

GuilhermeUpstream RGU1Downstream RGU2

CaldeiroesMiddlestream-Lenho RC1Middlestream-Caldeiroes RC2Downstream RC3

Salga Downstream RSG1

GrandeUpstream RG1Midllestream RG2Downstream RG3

Seca Downstream RSC1Praia Upstream RP1

Quente

Upstream RQ1Midllestream-Promineral RQ2Midllestream-Central RQ3Downstream RQ4

Pelames Downstream RPL1

Povoacao

Midllestream Central RPV1Midllestream West RPV2Midllestream East RPV3Downstream RPV4

Faial da TerraUpstream RTF1Downstream RFT2

Santa Maria Sao FranciscoUpstream RSF1Midllestream RSF2Downstream RSF3

physical-chemical analysis are presented else-where (INOVA, 2005; Cymbron et al., 2005).The Ribeira Seca (RSC) stream showed a strongheavy metal (chromium, arsenic, and zinc) conta-mination. On the other hand, the Ribeira Quente(RQ) stream had high levels of phosphorus (so-luble, inorganic and total), ammonia, biochemi-cal oxygen demand (BOD), total suspended so-lids, and high counts of bacteria. Ribeira de SaoFrancisco (RSF), in Santa Maria Island, recor-ded high levels of sulphates, zinc, chloride, andlow BOD values. The other streams were very si-milar in physical-chemical conditions; however,slightly higher levels of iron and manganese we-re found in Ribeira Grande (RG) and in RQ whencompared with the remaining streams studied.

Methods

Sampling was carried out in two different timeperiods, one in the fall of 2003 (between Octoberand November) and another during the spring of2004 (March and April).

The diatom sampling was carried out accordingto European recommendations (Kelly et al.,1998). Epilithic diatoms were collected by bru-shing at least five stones per sampling site, eachtime, and preserved in situ in 4% formalin (v/v).Samples were treated as described by Germain(1981) using warm nitric acid and mounted inNaphrax c©. Relative abundance of each taxon wasdetermined after counting at least 400 valves ineach sample (Prygel & Coste, 2000), using phasecontrast microscopy at maximum magnification(total magnification = 1000×). The diatoms wereidentified to the lowest taxon possible usingstandard identification keys (Krammer & Lange-Bertalot, 1986, 1988, 1991a, 1991b, 2000).

Aquatic invertebrate larvae were collected fromsubstrata as well as algae, mosses, macrophytes,and leaf litter by kick sampling using a hand net(500μmmesh) over a 3 min period. The sampleswere than preserved in situ with 96% alcohol. Inaddition, submersed stones from each site werebrushed to obtain the epilithic organisms. At thelaboratory, the organismswere sorted and identifiedto the family following available taxonomicliterature since it is the taxonomic level used todetermine faunistic indices (Garcia, 1987; Nieseret al., 1994; Tachet et al., 1994; Gonzalez, 1997).

Non-metric multidimensional scaling was ap-plied to the data for samples trend identification. ASIMPER analysis was used to identify individualcontributions to sample grouping. An ANOSIMtest was applied to check for differences betweensamples using the software package PRIMER(Clarke &Warwick, 2001). Pearson’s Correlationsbetween biological indices and environmentalvariableswere calculatedusingSPSSvs.15.

To evaluate water quality several diatom basedbiotic indices (IPS-Indice de polluo-sensibilite-,IDG-Indice diatomique generique-, TDI-Trophicdiatom index- and IBD-Indice Biologique dia-tomees-) were calculated using the softwareOMNIDIA vs.4.2 (Lecointe et al., 1993) and theIBWMP (Iberian Biological Monitoring WorkingParty) as adapted by Alba-Tercedor (1996) wasapplied to themacroinvertebrates data.

Diatoms based indices values are classified accor-ding to five quality levels: very clean waters (higherthan 16), clean waters with slight signals of stress

Ecological status of azorean streams 321

Figure 2. Ordination diagrams (nMDS) of diatoms from the studied sites with superimposed abundances of species contributing themost. Site codes as in table 1; a- refer to spring samples and b- to summer samples. Diagramas de ordenacion (nMDS) de diatomeasde las estaciones de estudio con superposicion de abundancias de las especies con mayor contribucion. Codigos como en la tabla 1;a- muestreos de Primavera y b- muestreos de Verano.

(between 13 and 16), contaminated waters (between9 and 12), very contaminated waters (between 5 and8), and extremely contaminated waters (below 5).

The IBWMP index was scaled according to refe-rence conditions found in this type of rivers andfive quality levels were established. These qua-

322 Goncalves et al.

lity levels include very clean waters (values over120), waters with stress signals (values between61 and 100), contaminated waters (values bet-ween 36 and 61), very contaminated waters (va-lues between 16 and 35), and extremely contami-nated waters (values below 15).

RESULTS

A total of 139 diatoms taxa belonging to 45 ge-nera were identified. Diatom communities weredominated by Nitzschia and Navicula species.

The non-metric multidimensional scaling(nMDS) performed on diatoms species data(Fig. 2) revealed a clear separation of RG1 andRQ2 sites. In the nMDS diagram, a clear gra-dient between upstream and downstream siteswas evident. Diatom site distribution, accordingto the assemblages, follows the same pattern asthe physical-chemical data assemblages obtainedby PCA (Cymbron et al. 2005).

According to the SIMPER analysis (Table 2),Pinnularia subcapitata, Eolimna minima, Am-phora pediculus, Achnanthidium minutissimum,and Eunotia exigua predominate in the upstream

sites. The RG1 site (upstream site of Ribei-ra Grande) is the best example of this group,showing a great dominance of P. subcapitata,E. exigua, and E. minima. In middle and down-stream sites Nitzschia species are dominant, withNitzschia frustulum, N. abbreviata, N. amphi-bia and N. palaea being particularly abundant.Other characteristic species of these sites are Na-vicula gregaria, Mayamaea atomus var. permi-tis and Luticola goeoppertiana.

For most streams, the values obtained for dia-tom indices (IPS, IDG, TDI and IBD) indicatea poor to moderate water quality (Fig 3). Morethan 50% of the samples indicate bad water qua-lity, and only 7% and 3% correspond respecti-vely to a good and better water quality. There-fore, results indicate high contamination at thestudied streams which was generally corrobora-ted by physical-chemical data (Cymbron et al.,2005). Nevertheless, highest levels of diatom in-dices were reported in sites with low nutrientload and low microbiological contamination.IPS and IBD present highly significant correla-tions ( p < 0.001) with soluble reactive phosp-horus (SRP), inorganic phosphorus (Pinorg), to-tal phosphorus (TP), and temperature. They also

Table 2. Average abundance of benthic diatom at the upstream and downstream sites, and their contribution for the dissimilarityfound between sites. Media de la abundancia de diatomeas bentonicas en las estaciones situadas en la cabecera y en el curso bajo,y contribucion para la diferenciacion entre estaciones.

TaxaMean Abundance Dissimilarity Contribution Acumulated

Reference Impact Mean (%) Contribution (%)

Nitzschia frustulum 02.19 18.41 09.00 10.30 10.30Pinnularia subcapitata 17.72 00.01 08.86 10.14 20.43Eolimna minima 17.15 02.51 08.14 09.31 29.74Nitzschia abbreviatta 02.02 14.12 07.32 08.37 38.11Navicula gregaria 02.59 11.73 05.60 06.41 44.52Amphora pediculus 09.96 00.85 05.07 05.80 50.32Achnanthes minutissima 08.58 02.94 04.89 05.59 55.91Nitzschia amphibia 01.38 07.06 03.35 03.84 59.75Eunotia exigua 05.81 00.00 02.90 03.32 63.07Synedra ulna 03.06 03.78 02.89 03.31 66.38Mayamacea atomus var. permitis 00.00 05.38 02.69 03.08 69.46Nitzschia palea 01.97 05.07 02.32 02.65 72.11Gomphonema clavatum 04.56 00.11 02.30 02.64 74.74Navicula goeppertiana 00.00 04.55 02.27 02.60 77.34Navicula reichardtiana 03.52 00.68 01.71 01.95 79.30Gomphonema parvulum 02.30 02.83 01.67 01.91 81.21

Ecological status of azorean streams 323

Figure 3. Results for the ecological quality for four diatomindices: IPS (Indice de polluo-sensibilite), IDG (Indice diato-mique generique), TDI (Trophic diatom index), IBD (IndiceBiologique diatomees) and IBMWP (Iberian Biological Moni-toring Working Party). Resultados de la calidad ecologica paracuatro ındices de diatomeas: IPS (Indice de sensibilidad a lapolucion), IDG (Indice generalizado de diatomeas), TDI (Indi-ce trofico de diatomeas), IBD (Indice Biologico de diatomeas)y IBMWP (Iberian Biological Monitoring Working Party).

correlate very significantly ( p < 0.01) with ni-trate and nitrite concentrations, conductivity, aswell as with coliforms and streptococcus counts.In spite of the positive response given by the-se indices, some limitations were detected inthe discrimination of the most severe conditions(e.g. RQ2).

The ordination obtained by nMDS on the ma-croinvertebrates data clearly separates site RG1from all the others (Fig. 4). Overall, downstreamsites tend to be placed towards one side of thediagram and closer to each other than to theupstream ones that presented higher heteroge-neity in their faunistic composition. Chironomi-dae, Oligochaeta, Simulidade, Hydroptilidae, Hi-

dracarina, and Psycodidae were the taxa that con-tributed the most for the observed similarity with-in, and between the up, middle, and downstreamsite categories, as revealed by the SIMPER analy-sis (Table 3). By superimposing these taxa abun-dances on the nMDS diagram (Fig. 4) it becomesclear that Simulidae and Hidracarina are moreabundant in Faial da Terra, Guillherme, and Cal-deiroes streams. Hydroptilidae are more abun-dant in Autumm at Ribeira Grande, Caldeiroes,Faial da Terra, and Povoacao (RPV1 e RPV2)streams. The most abundant macroinvertebratesare Chironomidae which are absent only at Ri-beira Grande (RG1a) and Ribeira da Povoacao(RPV3a). Oligochaeta are found mostly at RQand in downstream sites elsewhere. A small por-tion of upstream and downstream sites dissimi-larity (5.33%) is due to Psycodidae taxa. Thesetaxa are found mostly in downstream Povoacao,upstream Ribeira Grande, and Faial da Terra.

DISCUSSION

The freshwater Azorean streams revealed a po-lluted condition or a poor ecological water qua-lity when standard indices are applied to biolo-gical data. In fact, organic pollution mostly re-lated with agriculture and livestock practices butalso from urban origins, conditions the physical-chemical features of these ecosystems and it wasreflected on its biota composition.

In relation to microalgae, their occurrence inhigh ecological quality or reference conditionswas seldom the case. Actually, the most abundant

Table 3. Average abundance of benthic macroinvertebrate at upstream and downstream sites, and their contribution for the dissi-milarity found between the sites. Media de la abundancia de los macroinvertebrados bentonicos en las estaciones de cabecera y delcurso bajo, y contribucion para la diferenciacion entre estaciones.

TaxaMean Abundance Dissimilarity Contribution Acumulated

Reference Impact mean (%) Contribution (%)

Chironomidae 6.68 8.06 10.73 22.04 22.04

Oligochaetas 0.49 2.67 08.55 17.57 39.61

Hydroptilidae 1.66 1.07 07.77 15.97 55.58

Simulidae 2.19 1.11 07.23 14.86 70.44

Hidracarina 1.73 1.90 07.16 14.71 85.15

Psycodidae 0.40 0.59 02.69 05.53 90.69

324 Goncalves et al.

Figure 4. Ordination diagrams (nMDS) of macroinvertebrates in the studied sites with superimposed abundances for the mostcontributing species. Site codes as in table 1; a- refer to spring samples and b- to summer samples. Diagramas de ordenacion (nMDS)de macronivertebrados en las estaciones de estudio con superposicion de abundancias de las especies con mayor contribucion.Codigos como en la tabla 1; a- reportase a Primavera e b- a muestreos de Verano.

genera in the studied streams were Nitzchia andNaviculawhose species are usually associated withpolluted ecosystems (Rumeau&Coste, 1988).

The usage of benthic diatoms revealed to be agood instrument to access ecological stream qua-lity and has potential for application in routine

Ecological status of azorean streams 325

monitoring programs in the Azores, particularlybecause benthic diatoms seem to strongly corre-late with organic contamination, as diatom indi-ces significantly correlate with physical-chemicalparameters for these systems.

In relation to the benthic macroinvertebratefauna, freshwater streams from Sao Miguel andSanta Maria had poor taxa richness when com-pared to continental locations. Insular conditionswere determinant for this character of the faunaproviding extra difficulty to data interpretation.Nevertheless, aquatic macroinvertebrate fauna inthese streams was dominated by Diptera simi-larly to what is reported by literature for other lo-tic communities elsewhere (e.g. Barnes & Mann,1991; Tachet et al., 1994). Oligochaeta are fa-voured by organic enrichment (e.g. Rosenberg &Resh, 1993) and therefore were found in the siteswhere this condition was more obvious (RQ), aswell as in downstream sites.

The upstream conditions revealed heterogeneityin macroinvertebrate fauna composition compromi-sing their choice as reference sites for the Azores.

The difficulty to establish the reference condi-tions using macroinvertebrate fauna does not seemtobe related togeneral degradation, as demonstratedby the physical-chemical and microalgae data, butto the particularities of these insular freshwaterecosystems reflected in low values for IBMWPindex. Therefore, it might be necessary to adapt anddevelop a new benthic macroinvertebrate faunisticbased index for monitoring purposes in the Azores,as happened in Madeira (e.g. Hughes, 2001). Thenext step towards this end would be to improvetaxonomic expertise to species level in order todevelopa speciesbased index.

The poor results/performance revealed by theapplication of the currently used macroinvertebrateindices derive from particularities of local fauna,namely those related with its distinct insularcharacter. In fact, insular freshwater fauna parti-cularities, such as its lower diversity in relationwith continental systems (Smith et al., 2003) anddominance of active disperses as insects (Bilton etal., 2001), result from the interplay of complexbiological and geological processes (Smith et al.,2003). In fact, dispersal oceanic barriers, localabiotic factors, and hydromorphological factors

(e.g. highly seasonal and torrential flow regime)affect the success and establishment of colonizers(Poff, 1997;Malmqvist, 2002).

Moreover, the lower spatial heterogeneity andhabitat diversity at these systems, when compa-red with larger continental ones, limit the poolof invertebrates able to colonize and inhabit them(Malmqvist, 2002). Nevertheless, the results he-rein presented demonstrate the importance of thestudy of macroinvertebrate fauna for water qua-lity assessment purposes in the Azores. How-ever, the need to develop locally adapted proto-cols was demonstrated. Therefore, urging ecolo-gical research has to be carried out not only todevelop a reliable classification and local speci-fic protocols, but also to prevent the environmen-tal degradation threat already present in some ofthese unique and vulnerable freshwater macaro-nesian systems. In fact, Malmqvist et al. (1993)and Nilsson et al. (1998) clearly demonstratedthe increased destruction of freshwater systemsand species extinction in Macaronesia due to ha-bitat reduction and environmental degradation.

ACKNOWLEDGEMENTS

This study was funded by Direccao Regionaldo Ordenamento e Recursos Hidrıcos (SecretariaRegional da Ambiente e Mar). We thank Jose Ur-bano Tavares for the field work, to Helena Mar-ques and Vera Malhao for laboratory work. Wealso want to thank Luz Paramio and David Mar-tinez for the translation into Spanish as well asto Prof. Malcolm B. Jones and Dr. Paula Aguiarfor their help with the English language and use-ful comments on the manuscript. We also liketo thank Instituto de Inovacao Tecnologica dosAcores for the physical-chemical data.

REFERENCES

ALBA-TERCEDOR, J. 1996. Macroinvertebradosacuaticos y calidad de las aguas de los rios. Actasdel IV Simposio del agua en Andalucia (SIAGA),2: 203-213.

326 Goncalves et al.

ARCHER, W. 1874. Notes on some collections madefrom Furnas Lake, Azores, containing algae and afew other organisms. The Journal of the LinneanSociety Botany, 4: 328-340.

BARNES, R. S. K. & K. H. MANN. 1991. Funda-mentals of Aquatic Systems. 2ond ed. Oxford: Bla-ckwell Scientific Publications. 280 pp.

BARROIS, T. 1896. Recherches sur la Faune desEaux Douces des Acores. Lille: Imprimerie L. Da-nel. 16 pp.

BILTON, D. T., J. R. FREELAND & B. OKAMU-RA. 2001. Dispersal in freshwater invertebrates.Ann. Rev. Ecol. Syst., 32: 159-181.

CEMAGREFF 1982. Etude des methodes biologi-ques d’appreciation quantitative de la qualite deseaux. Agence financiere de Bassin Rhone-Medite-rranee-Corse, Pierre. Benite, Lyon. 218 pp.

CHASSE, R. 1997. Periphyton. In: Technical Eva-luation of Monitoring Methods Using Macrophy-tes, Phytoplankton and Periphyton to Assess theImpacts of Mine Effluents on the Aquatic Environ-ment. L. Cattaneo, A. St.-Cyr, R. Chasse & C. G. J.Fraikin (eds.): 102-119. Canada Centre for Mine-ral and Energy Technology. Ontario.

CLARKE, K. R. & R. M. WARWICK 2001. Chan-ge in marine communities: an approach to statis-tical analysis and interpretation. PRIMER-E. Ply-mouth. 172 pp.

COSTE, M. et H. AYPHASSORHO. 1991. Etude dela qualite des eaux du Bassin Artois-Picardie al‘aide des communautes de diatomees benthiques(Application des indices diatomiques). RapportCemagref. Bordeaux-Agence de l‘Eau Artois. Pi-cardie, Douai. 227 pp.

CRANSTON, P. S., P. FAIRWEATHER & G.CLARKE. 1996. Biological indicators of waterquality. In: Indicators of Catchment Health: atechnical perpective J. Walker & D. J. Reuter(eds.), 143-154. CSIRO. Melbourne.

CYMBRON, R., D. PACHECO, V. GONCALVES,C. CABRAL, J. V. CRUZ, P. RAPOSEIRO, A.COSTA, H. MARQUES, M. DOMINGOS, J. C.NUNES e R. COUTINHO. 2005. Monitorizacaoda qualidade das aguas interiores das ilhas de San-ta Maria e Sao Miguel da Regiao Autonoma dosAcores. In: Tecnicas e metodos para a gestao sus-tentavel da agua na Macaronesia. ITC-InstitutoTecnologico de Canarias & Cabildo de Lanzarote(eds.), 295-334. Islas Canarias.

DESCY, J. P. 1979. A new approach to water qua-lity estimation using diatoms. Beih. Nova Hedw,64: 305-323.

DESCY, J. & M. COSTE. 1991. A test of methodsfor assessing water quality based on diatoms. Verh.Internat. Verein. Limnol., 24: 2112-2116.

ELORANTA, P. & SOININEN, J. 2002. Ecologicalstatus of some Finnish rivers evaluated using benthicdiatomcommunities. J.Appl. Phycol., 14(1): 1-7.

EUROPEAN PARLIAMENT & THE COUNCIL OFTHE EUROPEAN UNION. 2000. Directive2000/60/EC of the European Parliament and of theCouncil establishing a framework for the Com-munity action in the field of water policy. OfficialJournal of the European Communities, 327: 1-72.

GARCIA, J. G. 1987. Claves de identificacion delos tubelarios de las aguas continentales de laPenınsula Iberica e Islas Baleares. Barcelona.Asociacion Espanola de Limnologia. 34 pp.

GERMAIN, H. 1981. Flore des diatomees. Diato-mophycees d’eaux douces et saumatres du massifArmoricain et des contrees voisines de l’Europeoccidentale. Collection “Faunes et flores actue-lles”. Boubee editions. Paris. 444 pp.

GILLER, P. S. & B. MALMQVIST. 1998. The Bio-logy of Streams and Rivers. Oxford UniversityPress. Oxford. VIII + 296 pp.

GONCALVES, V., P. RAPOSEIRO & A. C. COS-TA. 2005. The role of benthic diatoms and macro-invertebrates in the assessment of the ecologicalstatus of Azorean freshwaters. Internacional sym-posium on Assessing the ecological status of ri-vers, lakes and transitional waters: 34-35.

GONZALEZ, G. 1997. Claves para la identificacionde las larvas y pupas de los simulideos (Diptera)de la Penınsula Iberica. Barcelona. Asociacion Es-panola de Limnologia. 77 pp.

GUERNE, M. J. 1887. La faune des eaux douce desAcores et le transport des animaux a grande distan-ce par l’eitermediaire des oiseaux. Comptes Rendude la Societe de Biologie, 8: 580-584.

HUGHES, S. J. 2003. A study of the freshwater ma-croinvertebrate fauna of Madeira and their appli-cation in a regional ecological monitoring system.PhD thesis. University of London, UK. 328 pp.

HUGHES, S. J. &M. T. FURSE. 2001. Developmentof a biotic score for the assessment of the ecolo-gical quality of the rivers and streams of Madei-ra. Arquipelago-Life and Marine Sciences, Supple-ment 2: 19-32.

Ecological status of azorean streams 327

HUGHES, S. J. & B. MALMQVIST. 2005. Atlan-tic Island freshwater ecosystems: challenges andconsiderations following the EUWater FrameworkDirective. Hydrobiologia, 544: 289-297.

INOVA 2005.Monitorizacao da Qualidade de AguasSuperficiais e Subterraneas nas ilhas de Sao Mi-guel e Santa Maria-Caracterizacao Fısico-Quımicae Microbiologica. INOVA. Ponta Delgada. 32 pp.

KELLY, M., A. CAZAUBON, E. CORING, A.DELL’UOMO, L. ECTOR, B. GOLDSMITH, H.GUASCH, J. HURLIMANN, A. JARLMAN, B.KAWECKA, J. KWANDRANS, R. LAUGASTE,E.-A. LINDSTRRM, M. LEITAO, P. MARVAN,J. PADISAK, E. PIPP, J. PRYGIEL, E. ROTT, S.SABATER, H. VAN DAM & J. VIZINET. 1998.Recommendations for the routine sampling of dia-toms for water quality assessments in Europe. J.Appl. Phycol., 10: 215-224.

KELLY, M. & B. WHITTON. 1995. The TrophicDiatom Index: a new index for monitoring eutro-phication in rivers. J. Appl. Phycol., 7: 433-444.

KOLKWITZ, R. und M. MARSSON. 1908. Okolo-gie der pflanzlichen Saprobien. Berichte der Deuts-chen Botanischen Gesellschaft, 26 A: 505-519.

KRAMMER, K. & H. LANGE-BERTALOT. 1986.Bacillariophyceae 1/2 Naviculaceae. In: Susswas-serflora von Mitteleuropa. H. Ettl (ed.), 876 pp.G. Fisher Verlag. Jena.

KRAMMER, K. & H. LANGE-BERTALOT. 1988.Bacillariophyceae 2/2 Bacillariaceae, Epithemi-aceae, Surirellaceae. In:Susswasserflora vonMittele-uropa. H. Ettl (ed.), 576 pp. G. Fisher Verlag. Jena.

KRAMMER, K. & H. LANGE-BERTALOT. 1991a.Bacillariophyceae. 2/3: Centrales, Fragilariaceae,Eunotiaceae. In: Susswasserflora von Mitteleuro-pa. H. Ettl (ed.), 576 pp. G. Fisher Verlag. Jena.

KRAMMER, K. & H. LANGE-BERTALOT. 1991b.Bacillariophyceae. 2/4: Achnanthaceae. In: Suss-wasserflora vonMitteleuropa.H. Ettl (ed.), 576 pp.G. Fisher Verlag. Jena.

KRAMMER, K. & H. LANGE-BERTALOT. 2000.Bacillariophyceae. 2/5: English and French trans-lation of the keys. In: Susswasserflora von Mittele-uropa.H. Ettl (ed.), 576 pp. G. Fisher Verlag. Jena.

LECLERCQ, L. & B. MAQUET. 1987. Two newchemical and diatomic indicators of the qualityof running water. Application to the Samson andits tributaries (basin of the Belgian Meuse). Com-parison with other chemical indicators, biocenoseand diatomic indicators. Inst. Roy. Sc. Nat. Belg.,38: 113.

LECOINTE, C., M. COSTE & J. PRYGIEL. 1993.Omnidia: software for taxonomy, calculation ofdiatom indices and inventories management. Hy-drobiologia, 269-270: 1 509-513.

LOUVAT P. & ALLEGRE C. J. 1998. Riverine ero-sion rates on Sao Miguel volcanic island, Azoresarchipelago. Chemical Geology, 148: 177-200.

MALMQVIST, B. 2002. Aquatic invertebrates in ri-verine landscapes. Freshwat. Biol., 47: 4 679-694.

MALMQVIST, B., A. N. NILSSON, M. BAEZ, P. D.ARMITAGE & J. BLACKBURN. 1993. Streammacroinvertebrate communities in the island of Te-nerife. Arch. Hydrobiol., 128: 2 209-235.

METCALFE-SMITH, J. 1994. Biological water-quality assessment of rivers. Use of macroinver-tebrates communities. In: The rivers handbook.P. Calow & G. Petts (eds.), 144-169. BlackwellScientific Publications. Oxford.

NIESER, N., M. BAENA, J. MARTINEZ-AVILES yA. MILLAN. 1994. Claves para la identificacionde los heteropters acuaticos (nepomorpha & ge-rromorpha) de la Penınsula Iberica-Con notas so-bre las especies de las Islas Azores, Baleares, Ca-narias y Madeira. Barcelona: Asociacion Espanolade Limnologia. 111 pp.

NILSSON, A. N., B. MALMQVIST, M. BAEZ, J.H. BLACKBURN & P. D. ARMITAGE. 1998.Stream insects and gastropods in the island of GranCanaria (Spain). Annls Limmol., 34:4 413-435.

PRYGIEL, J. & M. COSTE. 1993. The assessmentof water quality in the Artois-Picardie water basin(France) by the use of diatom indices. Hydrobiolo-gia, 269/279: 343-349.

PRYGIEL, J. et M. COSTE. 2000.Guide Methodolo-gique pour la mise en oeuvre de l’Indice Biologi-que Diatomees. Agences de l’Eau. France. 134 pp.

POFF, N. 1997. Landscape filters and species traits:towards mechanistic understanding and predictionin stream ecology. Journal of the North AmericanBenthological Society, 16: 391-409.

PUJANTE, A. 1997. Los artropodos como bioindica-dores de la calidad de las aguas. Boletin de la So-ciedad Entomologica Aragonesa, 20: 277-284.

ROSENBERG, D. M. & VINCENT H. R. 1993. In-troduction to freshwater biomonitoring and benthicmacroinvertebrates. In: Freshwater Biomonitoringand Benthic Macroinvertebrates. D. M. Rosenberg&H. R. Vincent (eds.).: 1-9. Kluwer. London.

RUMEAU, A. et M. COSTE. 1988. Initiation a lasystematique des diatomees d’eau douce. Bull. Fr.Peche Piscic, 309: 1-69.

328 Goncalves et al.

SLADECEK,V. 1986.Diatoms as indicators of organicpollution. Acta Hydrochim Hydrobiol, 14: 555-566.

SMITH, G. C., A.P. COVICH&A.M.D. BRASHER.2003. An Ecological Perspective on the Biodi-versity of Tropical Island Streams. BioScience,53: 1 048-1 051.

STEVENSON, R. J. & Y. PAN. 1999. Assessing en-vironmental conditions in rivers and streams withdiatoms. In: The diatoms: applications for the en-vironmental and earth sciences. 11-40. CambridgeUniversity Press. Cambridge.

TACHET, H., M. BOURNAND et P. RICHOUX.1994. Introduction a l’etude des macroinvertebresdes eaux douces. Lyon: Universite Claude Bernard.155 pp.

U.S. EPA. 2002. Methods for Evaluating WetlandCondition: Introduction to Wetland BiologicalAs-sessment. Office of Water, U.S. EnvironmentalProtection Agency, Washington, DC. EPA-822-R-02-014. 35 pp.

ZBYSZEWSKI, G., ALMEIDA F. M., O. V. FE-RREIRA e C. T. ASSUNCAO. 1958. CartaGeologica de Portugal na escala de 1:50 000.Notıcia explicativa da Folha B, S. Miguel (Acores).Servicos Geologicos de Portugal. Lisboa. 37 pp.

ZBYSZEWSKI, G., ALMEIDA F. M., O. V. FE-RREIRA e C. T. ASSUNCAO. 1959. CartaGeologica de Portugal na escala de 1:50 000.Notıcia explicativa da Folha A, S. Miguel (Acores).Servicos Geologicos de Portugal. Lisboa. 22 pp.

Limnetica, 27 (2): x-xx (2008)Limnetica, 27 (2): 329-342 (2008)c© Asociacion Iberica de Limnologıa, Madrid. Spain. ISSN: 0213-8409

Influence of phosphorus and irradiance on phytoplanktonicchlorophyll-a concentration and phosphorus contents at a diel scalein a Mediterranean reservoir

Claudia Feijoo 1,2,∗, Marta Comerma 1, Rafael Marce 1, Juan Carlos Garcıa 3, David Balayla 1,Enrique Navarro 1 & Joan Armengol 1

1 Fluvial Dynamics and Hydraulic Engineering (FLUMEN), Departament d’Ecologia, Universitat de Barcelona,Diagonal, 645, 08028 Barcelona, Spain.2 Programa de Investigaciones en Ecologıa Acuatica, Departamento de Ciencias Basicas, Universidad Nacionalde Lujan, C.C. 221, (6700) Lujan, Argentina.3 Aigues Ter-Llobregat, Barcelona, Spain.2

∗ Corresponding author: clasife@coopenetlujan.com.ar2

Received: 2/6/08 Accepted: 5/8/08

ABSTRACT

Influence of phosphorus and irradiance on phytoplanktonic chlorophyll-a concentration and phosphorus contents at adiel scale in a Mediterranean reservoir

Phosphorus concentration in the water is one of the main factors regulating phytoplankton biomass and productivity in inlandwater bodies. However, phosphorus uptake by algae could not cause immediate growth, because other factors (light and othernutrient availability) may limit production. Nonetheless, “luxurious” phosphorus uptake (i.e. phosphorus uptake beyond thealgal requirements) has been observed, and it has been interpreted as storage to use in situations of low nutrient availability.Thus, the assessment of the effect of phosphorus supply on algal growth is not straightforward, especially at very short timescales.In this study, we analyze the relationships between environmental phosphorus levels and internal phosphorus and chlorophyll-acontents in phytoplankton at a diel scale in a Mediterranean reservoir, considering the different algal (intracellular andmembrane-associated) and water phosphorus pools. We also evaluated the influence of light on these relationships by samplingat two water depths with different irradiance levels. Our hypothesis is that chlorophyll-a and intracellular phosphorus contentsin phytoplankton are both influenced by ambient phosphorus and irradiance levels, which are complementary resources foralgae as the nutrient-light hypothesis proposes.Phosphorus concentration and relative contribution of each phosphorus fraction was similar at both sampling depths. Totalphosphorus concentration was dominated by the particulate pool (70%), and dissolved inorganic phosphate represented onlyone third of the dissolved pool. Total phosphorus content and the relative contribution of the different pools in algal cellswere similar at both sampling depths. Intracellular phosphorus pool was on average 77% of the total nutrient content, whilephosphorus associated to membranes represented the remaining of the total pool. Mean intracellular SRP concentration was36% of the intracellular phosphorus content. None of the algal phosphorus pools showed significant correlations with un-derwater light levels at both sampling depths. Total chlorophyll-a concentration showed no significant correlations with theseveral water phosphorus pools at both sampling depths, because each algal group responded differently to environmentalphosphorus levels, and these responses also varied with depth. We found significant relationships between the intracellularphosphorus content and the different phosphorus pools in water, but this almost ‘automatic’ response of algae to phosphoruslevels in water was not reflected in changes in the chlorophyll-a content, at least within the 3-day time frame of this study.A possible explanation is that processes associated to nutrient uptake and biomass construction operate over different timescales. The results of this work emphasize the complexity of the links between environmental phosphorus concentration andphytoplanktonic phosphorus content and biomass, and the importance of scale in analysing such relationships.

Key words: Internal algal phosphorus, phytoplankton biomass, phosphorus, wind regime, reservoir, diel cycle.

330 Feijoo et al.

RESUMEN

Influencia del fosforo y de la irradiancia en la concentracion de clorofila y contenido de fosforo fitoplanctonico a escaladiaria en un embalse mediterraneo

La concentracion de fosforo en el agua es uno de los factores principales que regulan la biomasa y produccion del fitoplanc-ton. Sin embargo, la asimilacion de fosforo por las algas puede no causar un crecimiento inmediato, ya que otros factores(luz u otros nutrientes) pueden limitar la produccion. Ademas, se ha observado la asimilacion “lujuriosa” de fosforo (esdecir, asimilacion que va mas alla de los requerimientos algales), lo que se ha interpretado como reserva a ser utilizada enmomentos de deficiencia de este nutriente. Por lo tanto, la caracterizacion del efecto del fosforo en el crecimiento algal no esinmediata, especialmente en escalas de tiempo muy cortas.En este estudio analizamos las relaciones entre el fosforo ambiental, el fosforo interno y la concentracion de clorofila-a fito-planctonica a escala diaria en un embalse mediterraneo, considerando las diferentes fracciones de fosforo algal (intracelulary asociado a membranas) y en el agua. Tambien evaluamos la influencia de la luz en estas relaciones muestreando a dosprofundidades con diferente nivel de irradiacion. Nuestra hipotesis es que la clorofila-a y el fosforo intracelular estan influen-ciados por el fosforo ambiental y los niveles de irradiacion, tal y como propone la hipotesis nutriente-luz.La concentracion de fosforo y la contribucion relativa de cada fraccion fueron similares en ambas profundidades. La concen-tracion de fosforo total estuvo dominada por la fraccion particulada (70%), y el SRP represento solo un tercio de la fracciondisuelta. En contenido de fosforo total y la contribucion relativa de las diferentes fracciones en las algas fue similar en ambasprofundidades. El fosforo intracelular represento el 77% del contenido total de nutriente, mientras que el fosforo asociadoa membranas represento el resto. La concentracion promedio de SRP intracelular fue el 36% del contenido intracelular defosforo. Ninguno de las fracciones de fosforo en las algas mostro correlacion significativa con los niveles de irradiancia enninguna de las dos profundidades. La concentracion de clorofila-a no mostro ninguna correlacion con las fracciones de fosfo-ro en ninguna de las dos profundidades, ya que cada grupo algal respondio de forma diferente a los niveles ambientales defosforo, y estas respuestas tambien variaron con la profundidad. Encontramos correlaciones significativas entre el contenidode fosforo intracelular y las diferentes fracciones de fosforo en el agua, pero esta respuesta casi “automatica” de las algasa los niveles de fosforo no se reflejo en cambios en el contenido de clorofila-a, al menos en el marco de los tres dıas queduro el estudio. Una posible explicacion es que los procesos asociados a la asimilacion de nutrientes y aquellos asociados ala sıntesis de biomasa operan en escalas de tiempo diferentes. Los resultados de este estudio destacan la complejidad de lasrelaciones entre la concentracion de fosforo ambiental, el contenido de fosforo de las algas y la produccion de biomasa, y laimportancia de la escala en el analisis de estas relaciones.

Palabras clave: Fosforo algal interno, biomasa fitoplanctonica, fosforo, regimen de viento, embalse, ciclo diario.

INTRODUCTION

Phosphorus concentration in the water is one ofthe main factors regulating phytoplankton bio-mass and productivity in inland water bodies. It iswell known that activity, structure, and biomassof the phytoplankton community in the epilim-nion may change in response to changes in thenutrient regime (Komarkova & Hejzlar, 1996).Even though dissolved inorganic phosphate is themost readily available form of this element for al-gal growth, its concentration in epilimnetic wateris often insufficient to satisfy phytoplanktonic de-mand (Wetzel, 2001). Nevertheless, the phospho-rus pool is rapidly regenerated by several mecha-

nisms (Dodds, 1983; Nalewajco & Lean, 1980),which included enzymatic hydrolysis of dissol-ved organic compounds by phosphatases relea-sed by microorganisms (Boavida &Heath, 1984).Therefore it is important to consider all phospho-rus fractions when we wish to quantify the wholeenvironmental phosphorus availability.

Phytoplankton growth rate is not only relatedto the availability of phosphorus and other nu-trients, but also to its own physiological pheno-mena (uptake and assimilation processes consi-dered separately). Thus, growth rate can be rela-ted to the average internal nutrient content thatis limiting growth (Droop, 1983). Nevertheless,after covering their nutrient requirements, algae

Influence of phosphorus and irradiance on chlorophyll-a at a diel scale 331

can continue absorbing nutrients in a processknown as ‘luxury uptake’, and the excess storedcan be used in periods of low nutrient availabi-lity in water. Algal phosphorus uptake may al-so be influenced by irradiance level. The light-nutrient hypothesis (Sterner et al. 1997) propo-ses that high light or low phosphorus supply canlead to phytoplankton biomass that will be car-bon rich and phosphorus poor, increasing the C:Palgal ratio. Within the framework of this hypothe-sis, Dickman et al. (2006) experimentally analy-zed the response of phytoplankton stoichiometryto light and nutrients in reservoirs of varying pro-ductivity. They found that the light-nutrient hy-pothesis was supported and that the effects oflight and nutrients on phytoplankton stoichiome-try were strongly interactive. They proposed thatlight and nutrients might serve as complementaryresources (Tilman 1982); under limitation by oneresource, phytoplankton may use the other, moreavailable resource to partially compensate for thelack of the limiting factor. Consequently, phyto-plankton may take up excess nutrients to partia-lly compensate low light levels, resulting in lowC:nutrient ratios. Higher uptake of nutrients mayallow phytoplankton to build cellular machineryneeded for increasing light absorption (Klaus-meier et al. 2004, Dickman et al. 2006).

The phosphorus content of algal cells hasbeen partitioned into intracellular and membra-ne-associated phosphorus pools, which can con-tribute differently to the total cellular phospho-rus content. The relative contribution of eachpool varies under different algal growth. Sanudo-Wilhelmy et al. (2004) observed that surface-adsorbed phosphorus was higher in senescentthan in exponential growing cultures, and sug-gested that phosphorus uptake by phytoplank-ton could be a two-step kinetic process: first,phosphorus adsorption to cell surface caused byan inorganic scavenging process, followed byincorporation into the intracellular pool, whichwould reflect the algal nutritional status mo-re accurately than the phosphorus associated tomembranes. This study highlights the importan-ce of considering all phosphorus fractions of phy-toplankton to quantify its response to changesin environmental phosphorus levels.

The processes of nutrient uptake and storage byphytoplankton can operate at small time scales.Turnover time of inorganic phosphorus variesfrom minutes or seconds to hours or days, depen-ding on the trophic state of the water, the pre-dominant phosphorus pool (particulate or solu-ble), and time of year; the velocity of the pro-cess is facilitated by the fact that no redox chan-ges are necessary for algal uptake and metabo-lism (Harris 1986, Nalewajko & Lean 1980). Ina broader context, significant and often repeata-ble variations in phytoplankton biology are ofteninduced, and commonly observed, on time sca-les shorter than a day (Prezelin, 1992); therefore,they are not well documented in phytoplanktonstudies at the seasonal scale. However, few stu-dies have investigated phytoplankton spatial andtemporal dynamics at a diel scale, and they ge-nerally lasted only one day (Patterson & Wilson,1995; Barbosa et al., 1989; Aleya, 1991).

In this study, we analyze the relationshipsbetween environmental phosphorus levels andinternal phosphorus and chlorophyll-a contentsin phytoplankton at a diel scale in a Medite-rranean reservoir, considering the different al-gal and water phosphorus pools. We also evalua-ted the influence of light on these relationshipsby sampling at two water depths with differentirradiance levels. Our hypothesis is that chloro-phyll-a and intracellular phosphorus contentsin phytoplankton are both influenced by am-bient phosphorus and irradiance levels, which arecomplementary resources for algae as the nu-trient-light hypothesis proposes. Consequently,our predictions are that:

a) At higher water depth and consequently, lo-wer amounts of light available, phytoplank-ton phosphorus content will be higher thanat lower depth.

b) Water phosphorus content will show a positi-ve relationship with intracellular phosphorusbut not with membrane-associated phospho-rus of phytoplankton.

c) Chlorophyll-a concentration will be positivelyrelated to water phosphorus content, and thisrelationship will be stronger at lower depth.

332 Feijoo et al.

To test our hypothesis, daily changes in waterand algae P contents and chlorophyll-a con-centrations were intensively monitored at twodifferent water depths during three diel cyclesin Sau Reservoir (Spain).

MATERIALS AND METHODS

Sau Reservoir (NE Spain) is a eutrophic and mo-nomictic canyon-shaped reservoir that suppliesdrinking water to Barcelona. The Ter River, themain inflow to Sau Reservoir, is heavily polluted,exhibiting high concentrations of soluble reacti-ve phosphorus and ammonia (Vidal & Om, 1993;Armengol et al., 1994). The annual pattern ofthermal stratification in summer (from April toSeptember) and vertical mixing in winter hasbeen observed in Sau since 1963 (Vidal and Om,1993; Comerma, 2003). During summer, whenthe thermal stability of the reservoir water co-lumn is high, the river is colder and denser thanreservoir’s surface waters and river water pro-gressively sinks towards thermocline depth withlittle mixing with the surrounding reservoir water(Armengol et al., 1999).

The study was conducted in summer between11th and 14th September 2001. Samples were ta-ken every two hours from a station located about1 000 m from the dam.

Temperature, pH, conductivity, dissolved oxy-gen concentration, and turbidity profiles weredrawn using a multiparametric probe Turo Wa-ter Analyzer. Photosynthetically active radiation(PAR) was measured at different depths witha LI-COR Underwater Radiation Sensor. Chlo-rophyll-aconcentrationprofileswere estimatedwitha submersible fluorescent FluoroProbe that can dis-criminate among the main phytoplanktonic groups(Chlorophyceae, Cyanobacteria, Cryptophyceae,and Bacillariophyceae). Previous to the study,we calibrated the device with algae of differentgroups cultured from autochthonous populations; inaddition, themethod’s accuracy has also been testedin other studies, which showed that fluorescencemeasurements were in good agreement withHPLC and microscopic count data (Beutler et al.,2002;Leboulangeretal., 2002).

For the estimation of the different water and algalphosphorus pools, water samples were taken at2 m-depth and at 6 m-depth with a 5-L capacitySchindler-Patalas hydrographic bottle. Sampleswere filtered in the field through a 53 μm net toeliminate zooplankton, and transported in plasticbottles in the dark to a nearby laboratory, wherethey were immediately analyzed.

Unfiltered water was used to determine totalphosphorus in reservoir water. Samples werepreviously digested (Grasshoff et al., 1983), andanalyzed for reactive phosphorus concentrationby the ascorbic acid method (APHA, 1992).Filtered water (Whatmann GF/F glass fiberfilters pre-combusted at 450◦ C) was analysed fordissolved inorganic phosphate (this phosphorusfraction was considered as soluble reactivephosphorus (SRP)) and total dissolved phos-phorus (Grasshoff et al., 1983; APHA, 1992).Particulate phosphorus (PP) in water was estima-ted as the difference between total phosphorusand total dissolved phosphorus, while dissolvedorganic phosphorus (DOP) was the differencebetween total dissolved phosphorus and SRP.Considering that the study was focused on surfa-ce layers during the stratification period we didnot measure sedimentary phosphorus fractions.

Different internal algal phosphorus forms we-re estimated by filtering water samples throughpre-combusted GF/F filters, and boiling filters inwater at 99◦ C during one hour to assure the brea-king of cells and the release of the intracellularnutrient pool (Thorensen et al., 1982). The re-sulting extract was then passed through a pre-combusted GF/F filter. This filter was digested(Grasshoff et al., 1983), and particulate phos-phorus (i.e. the phosphorus linked to cell mem-branes or MP) was estimated by the ascorbic acidmethod. The concentrations of soluble phospho-rus fractions of the filtrate were then estimatedby the methods indicated above. From the filtra-te we measured the total dissolved phosphorus(assumed to be the total intracellular phosphorus,IP) and the internal dissolved inorganic phospha-te (measured as reactive phosphorus (SRPi)).

From the raw data obtained with the Fluoro-Probe (vertical concentration profiles expressed inμgchl-aL−1), we calculated total chlorophyll-a

Influence of phosphorus and irradiance on chlorophyll-a at a diel scale 333

Figure 1. Direction and velocity of the wind and thermocline depth during the sampling days in Sau Reservoir. Positive values ofwind velocity indicate that wind blew from the dam wall to the river, while negative values indicate the opposite direction. Direcciony velocidad del viento y profundidad de la termoclina durante los dıas de muestreo en el embalse de Sau. Valores positivos de lavelocidad del viento indican que el viento soplo desde la presa al rıo, mientras que valores negativos indican la direccion opuesta.

concentration in awater columnof 7m-depth (fromthe water surface to the bottom of the epilimnion),obtaining the chlorophyll-a content per unit surface(mgchl-am−2) at each samplingoccasion.

To determine thermocline depth, we establi-shed the depth at which the maximum tempera-ture difference was observed for each tempera-ture profile recorded in all sampling occasions(N = 34). Then, we estimated water temperatu-re at this depth by linear interpolation down eachprofile, and averaged all the estimated temperatu-res to obtain a value of 19.51◦ C. The depth of the19.51◦ C isotherm was determined by linear in-terpolation for each sampling point, and this wasused as a representation of thermocline depth.

Wedderburn number (W) (Imberger & Patter-son, 1990) was also calculated for each samplingoccasion. This is an index of mixing potentialand the degree of tilting of a thermocline in len-tic environments, which combines the effects ofstability, wind forcing, and mixed-layer aspectratio into a single parameter.

Meteorological data were recorded by a Camp-bell Scientific Station situated 200 m from theshoreline and 12 m above the water surface whenthe reservoir is at maximum capacity.

Nonparametric Spearman rank correlationswere applied to explore the relationships betweenvariables.

RESULTS

Weather conditions and physico-chemicalcharacterization of the system

Weather was stable during the study period. Airtemperature increased along the sampling dayswithhigher values between 16:00 and 18:00h, whileatmospheric pressure decreased. Maximum solarradiation was recorded at 14:00h. Wind regimeshowed a diel regular pattern (Fig. 1): during day,wind blew from the river towards the dam (W-SW)while it blew in the opposite direction during thenight (E-NE), as a light seabreeze.

334 Feijoo et al.

A

B

Figure 2. Vertical profiles of (A) temperature (◦ C), and(B) conductivity (μS cm−1) in Sau Reservoir during the sam-pling days. Perfiles verticals de (A) temperature (◦ C) y (B) con-ductividad (μS cm−1) en el embalse de Sau durante los dıas demuestreo.

Wedderburn number (W) estimated for each sam-pling occasion was always higher than 10 witha mean value of 591, indicating the lack of mi-xing, high thermal stability, and a clear stra-tification profile in Sau Reservoir during thestudy. Thermocline was situated between 6 and

9 m-depth and oscillated in relation to wind(r = −0.49, P < 0.01) (Fig. 1 and 2A). Therefo-re, when the wind blew towards the river (posi-tive wind values), thermocline moved upwards,while it deepened when the wind blew in the op-posite direction. Highest water temperatures we-re 23-24◦ C and were recorded during the af-ternoon (13:00-20:00 h) (Fig. 2A). Oxygen satu-ration was 20-130%, and its profile was simi-lar to that of water temperature. Unlike tempe-rature and oxygen saturation, conductivity didnot show a daily pattern. Largest conductivi-ties, which evidence the presence of Ter Ri-ver waters (Armengol et al. 1999), were ob-served at 7 m depth in 29 of the 34 samplingoccasions (Fig. 2B), implying an interflow ri-ver circulation at the bottom of the epilimnionthat was not affected by thermocline oscillation.

As expected, underwater light levels variedalong the day at both sampling depths, and theywere one order of magnitude higher at 2 m-depth than at 6 m-depth, with mean values of83.80 μmol of photons m−2 seg−1 and 2.52 μmolof photons m−2 seg−1, respectively.

Phosphorus in water and in algal cells

Mean water phosphate concentration was higherat 6 m than at 2 m-depth, because of the presen-ce of two phosphate concentration peaks at thehigher depth (Fig. 3). But when these peaks we-re excluded from the analysis, concentration andrelative contribution of each phosphorus fractionwas similar at both sampling depths (Table 1).Total phosphorus concentration was dominatedby the particulate pool (� 70%), and SRP repre-sented only one third of the dissolved pool. SRPconcentration at 2 m-depth increased during themorning and showed lower values during the af-ternoon and night, but this pattern was not noti-ceable at 6 m-depth (Fig. 3).

SRP concentration at 6 m-depth showed nosignificant correlation with thermocline depth,but was related to temperature, conductivityand dissolved oxygen concentrations (r-valueswere, respectively, −0.37, 0.43, and −0.40, withP < 0.05). These relationships were not signifi-cant at 2 m-depth. Consequently, high phosphate

Influence of phosphorus and irradiance on chlorophyll-a at a diel scale 335

CONCENTRATIO

Nμmol

L−1

CONCENTRATIO

Nμmol

L−1

Figure 3. Dissolved inorganic phosphate concentrations (as SRP) in water and within algal cells at both sampling depths in SauReservoir during the sampling days. Note the difference of scale on the Y-axis. Concentraciones de fosfato inorganico disuelto (comoSRP) en el agua y dentro de las algas a las dos profundidades del embalse de Sau durante los dıas de muestreo.

levels in the base of the epilimnion were relatedto the lower temperature and oxygen concentra-tion and the higher conductivity that characteri-zed Ter River waters.

Total chlorophyll-a content per unit surfa-ce was higher during the night and showed astrong decrease in the morning (8:00-12:00 h)

Table 1. Mean concentration in μmol L−1 of different phosp-horus fractions in water (SRP: soluble reactive phosphorus;DOP: dissolved organic phosphorus; PP: particulate phospho-rus) and algae (IP: intracellular phosphorus; MP: phosphorusassociated to membranes) in Sau Reservoir at two water depthsduring the sampling days. N = 34, except for phosphorus con-centration at 6 m-depth where N = 27, because the values oftwo peaks recorded at this depth were not included. Concentra-cion media en μmol L−1 de las diferentes fracciones de fosforoen el agua (SRP: fosforo reactivo soluble; DOP: fosforo organi-co disuelto; PP: fosforo particulado) y en las algas (IP: fosforointracellular; MP: fosforo asociado a membranes) en el embal-se de Sau a dos profundidades en los dıas del muestreo. N = 34,excepto para la concentracion de fosforo a 6 m de profundidaden la que N = 27, debido a que los dos picos medidos a estaprofundidad no fueron incluidos.

2 m-depth 6 m-depthconcentration % concentration %

Water SRP 0.093 007.53 0.073 006.63DOP 0.256 021.40 0.265 023.33PP 0.881 071.07 0.795 070.04Total 1.239 100.00 1.135 100.00

Algae IP 0.581 078.00 0.534 077.00MP 0.167 022.00 0.160 023.00Total 0.748 100.00 0.694 100.00

(Fig. 4). Among all considered variables (air andwater temperature, atmospheric pressure, solarradiance, wind velocity, and Wedderburn num-ber), chlorophyll-a concentration per unit surfa-ce only showed significant relationships with wa-ter temperature (r = 0.39, p < 0.05) and Wedder-burn number (r = 0.49, p < 0.01).

Table 2. Spearman rank correlations between concentrationsof different phosphorus pools in water and in algal cells at thetwo sampling depths in Sau Reservoir. SRP: soluble reactivephosphorus; DOP: dissolved organic phosphorus; PP: particula-te phosphorus; SRPi: intracellular SRP; IP: intracellular phosp-horus; MP: phosphorus associated to membranes. ***: P <0.001; **: P < 0.01; *: P < 0.05; ns: not significant. Corre-laciones de rango de Spearman entre concentraciones de losdiferentes grupos de fosforo en el agua y en las algas a las dosprofundidades de muestreo en el embalse de Sau. SRP: fosfororeactivo soluble; DOP: fosforo organico disuelto; PP: fosfo-ro particulado; SRPi: SRP intracelular; IP: fosforo intracelu-lar; MP: fosforo asociado a las membranes. ***: P < 0.001;**: P < 0.01; *: P < 0.05; ns: no significativo.

Algae Water 2 m-depth 6 m-depth

SRPi SRP 0.50 *** 0.69 ***DOP 0.35 *** 0.39 ***PP 0.59 *** 0.69 ***

IP SRP 0.51 *** 0.43 ***DOP 0.45 *** 0.37 ***PP 0.47 *** 0.72 ***

MP SRP 0.27 ns 0.08 nsDOP 0.51 *** 0.34 ***PP 0.45 *** 0.26 ns

336 Feijoo et al.

Sau 11-14 of September 2001

Figure 4. Chlorophyll-a concentration per unit of area during the sampling days in Sau Reservoir. Positive values of wind velocityindicate that wind blew from the dam wall to the river, while negative values indicate the opposite direction. Concentracion declorofila-a por unidad de superficie durante los dıas de muestreo en el embalse de Sau. Valores positivos en la velocidad del vientoindican que el viento soplo desde la presa al rıo, mientras que valores negativos indican la direccion opuesta.

Total phosphorus content and the relative contri-bution of the different pools in algal cells we-re similar at both sampling depths; consequently,phytoplankton community showed no differencesin its phosphorus concentration when comparingsamples from the upper water layer with thosefrom the bottom of the epilimnion. Intracellularphosphorus pool (IP) was on average 77% of thetotal nutrient content, while phosphorus associa-ted to membranes (MP) represented the remai-ning total pool. Mean SRPi concentration was0.2 μmol L−1; i.e., 36% of the total intracellu-lar phosphorus content. None of the algal phos-phorus pools showed significant correlations withunderwater light levels at both sampling depths.

We found positive and highly significant corre-lations among the different phosphorus pools inwater and the intracellular dissolved forms of algalphosphorus (Table2), but not significant relationsresulted between SRP in water and phosphorus as-sociated to algalmembranes at both sampled layers.

Chlorophyll-a concentration andenvironmental phosphorus levels

Chlorophyll-a profiles showed that phytoplank-ton community was dominated by Chlorophy-ceae (1.99-34.56 μg L−1), while abundances of

the other algal groups (Cryptophyceae, Cya-nobacteria and Bacillariophyceae) were lower(0.00-15.06 μg L−1). Epilimnetic chlorophyll-aconcentrations were higher during the night anddecreased during the morning, and this patternwas consistently observed during the three sam-pling days. However, the contribution of the dif-ferent algal groups to these changes was diffe-rent: decline of chlorophyll-a concentration du-ring the day was related to a decrease of Ch-lorophyceae, while Cyanobacteria, Cryptophy-ceae and Bacillarophyceae either maintainedsimilar abundances or increased.Total chlorophyll-a concentration (in μg L−1)showed no significant correlations with the se-veral water phosphorus pools at both samplingdepths, because each algal group responds dif-ferently to environmental phosphorus levels, andthese responses also varied with depth (Fig. 5).At 2 m-depth, chlorophyll-a concentration ofChlorophyceae and Cyanobacteria were signi-ficant and negatively related to water dissol-ved inorganic phosphate (SRP) (r = −0.42 withP < 0.01 and R = −0.41 with P < 0.05, respec-tively), while Cryptophyceae showed a positiverelation (r = 0.44, P < 0.05) and diatoms werenot significantly related. Most of these significantcorrelations were not maintained at 6 m-depth,

Influence of phosphorus and irradiance on chlorophyll-a at a diel scale 337

Figure 5. Relationships between water SRP concentration (μmol L−1) and chlorophyll-a concentration (μg L−1) for the differentalgal groups at both sampling depths in Sau Reservoir. Linear equations fitted to the points are also indicated. Note the difference ofscale on the Y-axis. Relaciones entre concentraciones de SRP y clorofila-a de los diferentes grupos algales en las dos profundidadesde muestreo en el embalse de Sau. Se indican tambien las ecuaciones lineales de ajuste de los puntos.

where only Crytophyceae were significantly re-lated to water SRP (R = 0.40, P < 0.05). Whenthe dissolved organic forms of phosphorus in wa-ters were considered, significant correlations bet-ween DOP concentration and chlorophyll-a con-tent were observed for Cyanobacteria and Cry-ptophyceae (R = −0.41 and R = 0.44, respecti-vely, with P < 0.05) at 2 m-depth, but not forChlorophyceae and diatoms. Our results indica-te that there was not a direct relation betweenwater phosphorus level and total phytoplank-ton abundance due to the different response ofeach algal group, and that significant relation-ships between both variables were discernible at2 m-depth but not at 6 m-depth.

Concerning irradiance levels, only chlo-rophyll-a concentration of Cyanobacteria andCryptophyceae showed significant relationswith underwater light levels at 2 m-depth(R-values were, respectively, −0.61 and 0.50,with P < 0.05). Chlorophyll-a contents of allalgal groups were not significantly related toirradiance levels at 6 m-depth.

DISCUSSION

Sau Reservoir was well stratified during the studyand the water column showed a high thermal sta-

bility. The wind regimen coincided with a dailydisplacement of metalimnion depth, with ther-mocline rising during the night when the windblowed towards the river and with the opposi-te pattern during the day. According to the epi-limnetic water circulation model of Sau Reser-voir proposed by Armengol et al. (2005), ther-mocline oscillation generate internal seiches thatare associated with advective movements of sur-ficial water masses, which are alternatively dis-placed from end to end of the reservoir. The mo-del has been validated by several studies made inSau Reservoir where interactions between wind,seiches and superficial advective movements we-re demonstrated (Vidal et al. 2005, Marce et al.2007). In addition to this wind-induced water cir-culation, we observed higher conductivities at thebottom of the epilimnion that indicate the exis-tence of an interflow river circulation moving atan approximately constant depth (� 7 m), with-out being affected by thermocline tilting. Thispattern of river circulation has been consistentlyobserved at the end of summer for many yearsin Sau Reservoir (Armengol et al. 1999). The in-terflow circulation further increases thermal sta-bility of the water column and injects phospho-rus into the epilimnion due to the high nutrientload of Ter River (Vidal & Om, 1993; Armengolet al., 1994; Armengol et al. 1999).

338 Feijoo et al.

Some authors have documented the input ofphosphorus from the hypolimnion of lakes asso-ciated to storm events; in some cases this internalload constitutes the main phosphorus entrain-ment to the epilimnion (Larsen et al., 1981; So-ranno et al., 1997). Storm events with high windvelocity transfer kinetic energy to water, erodingthe base of the metalimnion and injecting phos-phorus from deeper waters to the epilimniom.Phosphorus entrainment is not only increased bythe number and strength of storm fronts but bythe weakness of stratification. In our study, werecorded two peaks of phosphate concentration atthe bottom of the epilimnion during the night andthe morning of the first two sampling days. Con-sidering the high thermal stability of the watercolumn and the lack of storms during the study,we cannot attribute the occurrence of these peaksto phosphorus entrainment from the hypolimniondue to thermocline erosion. We observed that dis-solved inorganic phosphate levels at the bottomof the epilimnion related positively with conduc-tivity and negatively with water temperature anddissolved oxygen concentration, suggesting thatphosphate peaks observed at 6 m-depth originatefrom the interflow river circulation.

Phosphorus is present in algal cells in seve-ral chemical forms. Storage forms are represen-ted by ortophosphates and polyphosphates, whilethere are four main organic phosphorus fractions:RNA, DNA, and lipid- and ester-phosphorus.Polyphosphates are much more important as sto-rage compounds than phosphates and regulateintracellular levels of ATP, ADP, and phospha-tes. They are synthesized rapidly transferringphosphate from ATP but are gradually degra-ded upon resumption of growth and nucleic acidsynthesis (Nalewajko & Lean 1980). In additionto this intracellular phosphorus pool, Sanudo-Wilhelmy et al. (2004) reported the existence ofsurface-adsorbed phosphorus in different mari-ne algal species. In the centric diatom Thalas-siosira weissflogii, surface-adsorbed phosphoruspool (90%) in senescent individuals was con-siderably higher than the amount measured inthe exponential growth phase (30%), sugges-ting that phytoplankton have developed mecha-nisms to access at least some of the surface-

adsorbed pool (Sanudo-Wilhelmy et al. 2004).In our study, phosphates represented 36% ofthe intracellular pool at both sampling depths,while phosphorus associated to membranes was23% of the total pool. Thus, our results suggestthat the phytoplankton community was activelygrowing and not limited by the intracellular ac-cumulated phosphorus. This result might, how-ever, be slightly overestimated because of the fil-ter used in the SRP determinations (GF/F filterinstead of 0.45μm filters).

The total phosphorus in a cell fluctuates withchanges in the phosphorus supply (Nalewajko &Lean 1980). In summer, phosphorus turnover ti-mes are short and algal cells can replenish theirquotas of nutrients very fast, surviving in pe-riods of nutrient starvation. This capacity allowsthem to integrate and damp the external fluctua-tions in nutrient availability (Harris, 1986). Wefound significant relationships between the in-tracellular phosphorus content and the differentphosphorus pools in water which could be ex-plained by this storage capacity. However, phos-phorus associated to membranes was not signi-ficantly related to dissolved inorganic phospha-te, giving some evidence to the hypothesis ofSanudo-Wilhelmy et al. (2004) that phosphorusadsorption to cell surface is caused by an inor-ganic process rather by active biogenic uptake.Consequently, this pool will not adequately re-flect the nutritional status of phytoplankton, a factthat must be taken into account in studies wherealgal stoichiometry is analysed.

The almost ‘automatic’ response of algae tophosphorus levels in water was not reflected inchanges in the chlorophyll-a content of the phy-toplankton community in Sau Reservoir, at leastwithin the 3-day time frame of this study, giventhat total chlorophyll-a concentration (in μg L−1)was not significantly related to water phosphoruslevels. The lack of relationship between phyto-plankton biomass and water phosphate concen-tration was also observed by Viner (1989) in anAustralian reservoir, where no increase in algalbiomass was observed after six days of fertili-zation with ammonium phosphate. Furthermore,Payne et al. (1988) demonstrated that phosphorusaddition to lake water produces luxury phospho-

Influence of phosphorus and irradiance on chlorophyll-a at a diel scale 339

rus uptake by the phytoplankton community butno algal response by the way of chlorophyll-aproduction. A possible explanation is that pro-cesses associated to nutrient uptake and biomassconstruction operate over different time scales.The turnover of the dissolved inorganic phospha-te pool in summer may be measured in minu-tes or seconds, while the algal growth processmay be estimated over scales of days and weeks(Harris, 1986). So, changes in phosphorus wa-ter concentration will be not reflected in the algalchlorophyll-a content at a diel scale.

Our data also indicate that chlorophyll-aconcentration of each algal group showed a dif-ferent relationship with dissolved inorganic phos-phate concentration, and that these relationshipswere significant at 2 m-depth but not at 6 m-depth (Fig. 5). Most studies have focused on theeffects of nutrient addition on the whole phyto-planktonic community through experimental ma-nipulation of the N:P ratio, but the influence onthe different algal groups has been less analy-sed. It has been observed that the decrease inN:P ratio produces cyanobacterial blooming whi-le the increase in N:P ratio most often do notcause the acceleration of Bacillarophyceae bio-mass, instead it leads to the maximum biomassof Chlorophyceae (Nalewajco & Lean, 1980; Le-vich, 1996; Levich & Bulgakov, 1992). We ha-ve not estimated N:P ratios in this study; there-fore, comparisons of our data with those of thebibliography are not possible. But, even thougha significant correlation between variables doesnot always imply a causal relationship, our re-sults suggest that algal response to environmen-tal phosphate concentration is variable amongtaxonomic groups and that underwater light le-vels somewhat mediate this response. This hy-pothesis should motivate further research becau-se of its possible consequences on eutrophicationprocesses in lentic environments.

Epilimnetic chlorophyll-a concentration es-timated per unit surface showed a diel pattern(Fig. 4) that could not be linked to sedimentationlosses, giving that the lowest concentrations va-lues were attained in moments with less stability.The lack of a significant relationship betweenchlorophyll-a concentration and solar irradiance

also suggests that variations in the phytoplank-ton growth rate cannot explain the observedvariability. A diel cycle was also observed fordissolved inorganic phosphate concentration at2 m-depth, with higher values in the morningand lower ones in the afternoon and the night.As it was said above, our data indicate theexistence of a thermocline oscillation associatedto wind regimen. Previous studies demonstratedthat this is a common pattern in Sau Reservoirand that it is linked to advective surface watermovements (Vidal et al. 2005, Marce et al.2007). Consequently, we hypothesized that thediel variation of dissolved inorganic phosphateand chlorophyll-a was produced by advectionof water masses with different phytoplanktondensity and phosphate concentration. Similartemporal and spatial dynamics of phytoplanktonmaxima have been observed in other reservoirsand have been attributed to horizontal patchinessand advective movements (Viner, 1989; Patter-son & Wilson, 1995). However, our results arenot conclusive to validate this hypothesis.

The light-nutrient hypothesis states thatphytoplankton C:nutrient ratios are driven by theratio of available light and nutrients, and it wassupported by studies using different approaches.For example, Sterner et al. (1997) analysed thehypothesis integrating data from the epilimnionof a broad range of lakes, while Dickman etal. (2006) experimentally quantified the effectsof light and nutrients on intact phytoplanktonassemblages from reservoirs of varying produc-tivity. Here, we compared phosphorus content inphytoplankton assemblages situated at differentdepths and, as a result, receiving different un-derwater light level. We found no differences inalgal phosphorus concentration at both samplingdepths and no significant relationships betweenirradiance and the different algal phosphoruspools. Consequently, the prediction that phos-phorus content will be greater at higher depthis not supported by our data. It must be takeninto account that the response of phytoplanktonstoichiometry to light and nutrients can be highlyvariable. Dickman et al. (2006) suggested thathighly diverse assemblages may exhibit greaterflexibility in their response than assemblages

340 Feijoo et al.

with lower diversity. Klausmeier et al. (2004)proposed that relative algal phosphorus contentmight vary depending on whether phytoplanktonis under exponential growth or at competitiveequilibrium under different limiting resources(nitrogen, phosphorus, and light). In addition tothis, within a phytoplanktonic assemblage eachspecies has distinct growth and mortality ratesand nutrient storage capacity. All these factorscould contribute to the lack of effects of light onalgal nutrient content in this study.

The results of this work emphasize (1) thecomplexity of the links between environmen-tal phosphate concentration and phytoplanktonicphosphorus content and biomass, and (2) the im-portance of scale in analysing such relationships.The response of algae to water phosphorus le-vels also seems to be mediated by light, not withrespect to the intracellular phosphorus content–as proposed by the light:nutrient hypothesis, butturning relationships between biomass and en-vironmental phosphorus more ‘apparent’. More-over, algal nutrient content may be influenced notonly by environmental light and nutrient levelsbut also by several factors, including assemblagediversity and characteristics of the phytoplankto-nic populations that are variable among species.Future studies on the relationships between algaeand environmental nutrient levels should consi-der phytoplankton composition to quantify thedifferential response of each algal taxon.

ACKNOWLEDGEMENTS

The authors wish to thank Marıa de los AngelesGallegos for her assistance with laboratory andfieldwork and Adonis Giorgi for its helpful com-ments on the manuscript. The limnological studyof Sau Reservoir is part of a long-term studysupported by Aigues Ter-Llobregat (ATLL). Thisproject has been partially funded by the Ministe-rio de Ciencia y Tecnologıa (Projects REN2001-2185-CO2-O2 and CGL2004-05503-CO2-01).The postdoctoral staying of C. Feijoo at theUniversitat de Barcelona was partially suppor-ted by the Universidad Nacional de Lujan andATLL. Rafael Marce is granted of the Minis-

terio de Educacion, Cultura y Deportes. Thecontribution of the Comision Interministerial deCiencia y Tecnologıa (CIRIT) of the Generalitatde Catalunya is greatly appreciated.

REFERENCES

ALEYA, L. 1991. Ecophysiological significance ofthe diel biochemical changes of particulatescoupled with metabolic and environmental para-meters in two trophically different lakes. Arch. Hy-drobiol., 120: 403-432.

ARMENGOL, J., J. TOJA & A. VIDAL. 1994. Sea-sonal rhythm and secular changes in Spanish re-servoirs. In: Limnology now: A paradigm of plane-tary problems. Margalef, R. (ed.): 237-253. Else-vier Science, Amsterdam.

ARMENGOL, J., J. C. GARCIA, M. COMERMA,M. ROMERO, J. DOLZ, M. ROURA, B. H. HAN,A. VIDAL & K. SIMEK. 1999. Longitudinal pro-cesses in canyon type reservoirs: The case of Sau(N.E. Spain). In: Theoretical Reservoir Ecologyand its Applications. Tundisi, J. G. & M. Straskra-ba (eds.): 313-345. Brazilian Academy of Sciencesand Backjhuys Publishers, Leiden.

ARMENGOL, J., E. NAVARRO, L. CAPUTO & J.DOLZ. 2005. Viento, estabilidad termica y ondasinternas: bases para una gestion de los embalses acorto plazo. In: Tercer Taller Internacional de Eu-trofizacion de lagos y embalses. I. Vila & J. Pizarro(eds.): 29-50. CYTED XVIIB, Santiago. Chile.

APHA. 1992. Standard methods for the examinationof water and wastewater. 18th Ed. APHA, Wa-shington D.C. 1 100 pp.

BARBOSA, F. A. R., J. G. TUNDISI & R. HENRY.1989. Diel variations in a shallow tropical Brazi-lian lake II. Primary production, photosynthetic ef-ficiency and chlorophyll-a content. Arch. Hydro-biol., 116: 435-448.

BEUTLER M., K. W. WILTSHIRE, B. MEYER, C.MOLDAENKE, C. LURING, M. MEYERHO-FER, U. P. HANSEN & H. DAU. 2002. A fluo-rometric method for the differentiation of algal po-pulations in vivo and in situ. Photosynthesis Re-search, 72: 39-53.

BOAVIDA, M. J. & R. T. HEATH. 1984. Are thephosphatases released by Daphnia magna compo-nents of its food? Limnol. Oceanogr., 29: 641-645.

COMERMA, M. 2003. Processos heterotrofics mi-

Influence of phosphorus and irradiance on chlorophyll-a at a diel scale 341

crobians a l’embassament de Sau. Doctoral thesis,University of Barcelona. 192 pp.

DICKMAN, E. M., M. J. VANNI & M. J. HOR-GAN. 2006. Interactive effects of light and nu-trients on phytoplankton stoichiometry.Oecologia,149: 676-689.

DODDS, W. K. 1993. What controls levels of dissol-ved phosphate and ammonium in surface waters?Aquatic Sciences, 55: 132-142.

DROOP, M. R. 1983. 25 years of algal growth kinetics.Apersonal view.BotanicaMarina, 26: 99-112.

GRASSHOFF, K., M. ERHARDT & K. KREM-LING. 1983. Methods of seawater analysis. 2ond

Ed. Wiley-VCH, Weinheim. 419 pp.HARRIS, G. P. 1986. Phytoplankton Ecology. Chap-

man and Hall, London. 384 pp.IMBERGER, J. & J. C. PATTERSON. 1990. Physical

limnology. In: Advances in Applied Mechanics vol.27. T.Wu(ed.).: 303-475.AcademicPress,Boston.

KLAUSMEIER, C. A., E. LITCHMAN, T. DAU-FRESNE & S. A. LEVIN. 2004. Optimal nitro-gen to phosphorus stoichiometry of phytoplankton.Nature, 429: 171-174.

KOMARKOVA, J. & J. HEJZLAR. 1996. Summermaxima of phytoplankton in the Iımov Reservoirin relation to hydrologic parameters and phospho-rus loading. Arch. Hydrobiol., 136: 217-236.

LARSEN, D. P., D. W. SCHULTS & K. W. MA-LUEG. 1981. Summer internal phosphorus sup-plies in Shagawa Lake, Minnesota. Limnol. Ocea-nogr., 26: 740-753.

LEBOULANGER C., U. DORIGO, S. JACQUET, B.LE BERRE, G. PAOLINI &. J. F. HUMBERT.2002. Application of a submersible spectrofluo-rometer for rapid monitoring of freshwater cya-nobacterial blooms: a case study. Aquat. Microb.Ecol., 30: 83-89.

LEVICH, A. P. 1996. The role of nitrogen-phospho-rus ratio in selecting for dominance of phytoplank-ton by cyanobacteria or green algae and its applica-tion to reservoir management. Journal of AquaticEcosystem Health, 5: 55-61.

LEVICH, A. P. & N. G. BULGAKOV. 1992. Regu-lation of species and size composition in phyto-plankton communities in situ by N:P ratio. RussianJournal of Aquatic Ecology., 1: 149-158.

MARCE, R., C. FEIJOO, E. NAVARRO, J. OR-DONEZ, J. GOMA & J. ARMENGOL. 2007. Inter-action betweenwind-induced seiches and convectivecooling governs algal distribution in a canyon-shapedreservoir.Freshwat.Biol., 52(7): 1336-1352.

NALEWAJKO, C. & D. R. S. LEAN. 1980. Phos-phorus. In: The physiological ecology of phy-toplankton. I. Morris (ed.).: 235-258. BlackwellScientific Publications, Oxford.

PATTERSON, G. & K. K. WILSON. 1995. The in-fluence of the diel climatic cycle on the depth-timedistribution of phytoplankton and photosynthesisin a shallow equatorial lake (Lake Baringo, Ken-ya). Hydrobiologia, 304: 1-8.

PAYNE, G. W., E. WHITE, S. E. PICKMERE, & P.H.WOODS. 1988. Algal responses to nitrogen andphosphorus additions in four central North Island,New Zealand lakewaters. Verh. Internat. Verein.Limnol., 23: 602-606.

PREZELIN, B. B. 1992. Diel periodicity in phyto-plankton productivity. Hydrobiologia, 238: 1-35.

SAUDO-WILHELMY,S.A., A. TOVAR SANCHEZ,F. X. FU, D. G. CAPONE, E. J. CARPENTER &D. A. HUTCHINS. 2004. The impact of surface-adsorbed phosphorus on phytoplankton Redfieldstoichiometry. Nature, 432: 897-901.

SORANNO, P. A., S. R. CARPENTER & R. C.LATHROP. 1997. Internal phosphorus loading inLake Mendota: response to external loads andweather. Can. J. Fish. Aquat. Sci., 54: 1 883-1 893.

STERNER, R. W., J. J. ELSER, E. J. FEE, S. J.GUILFORD & T. H.CHRZANOWSKI. 1997. Thelight:nutrient ratio in lakes: tha balance of energyand materials affects ecosystem structure and pro-cess. The American Naturalist, 150(6): 663-684.

THORENSEN, S. S., Q. DORTCH & S. I. AHMED.1982. Comparison of methods for extracting intra-cellular pools of inorganic nitrogen from marinephytoplankton. J. Plankton Res., 4: 695-704.

TILMAN, D. 1982. Resource Competition and Com-munity structure. Princeton University Press, Prin-ceton, N. J. 296 pp.

VIDAL, A. &J. OM. 1993. The eutrophication processin Sau Reservoir (NE Spain): A long term study.Verh. Internat. Verein. Limnol.,25: 1247-1256.

VIDAL, J., X. CASAMITJANA, J. COLOMER & T.SERRA. 2005. The internal wave field in Sau re-servoir: Observation and modeling of a third verti-cal mode. Limnol. Oceanogr., 50: 1 326-1 333.

VINER,A.B. 1989. Fluorescence and carbon fixationby phosphorus deficient phytoplankton in a WesternAustralian reservoir.Arch.Hydrobiol.,115:1-19.

WETZEL, R. G. 2001. Limnology: Lake and RiverEcosystems. 3rd Ed. Academic Press, San Diego.850 pp.

2001. Limnology: Lake and River Ecosystems.3rd Ed. Academic Press, San Diego. 850 pp.

INSTRUCCIONES PARA LOS AUTORES

Ambito de publicacion de LimneticaLimnetica publica artıculos originales de investigacion sobre la ecologıa

de las aguas continentales. El ambito de publicacion de Limnetica incluyela ecologıa de rıos, lagos, embalses, lagunas costeras, zonas humedas, bio-geoquımica, paleolimnologıa, desarrollo de metodologıas, taxonomıa, biogeo-grafıa y todos los aspectos de la ecologıa acuatica continental teorica yaplicada como gestion y conservacion, evaluacion de impactos, ecotoxicologıay contaminacion. Por este motivo Limnetica aceptara para su publicacionartıculos cientıficos presentando avances del estado del conocimiento, dedesarrollo tecnologico ası como los que resulten de aplicaciones practicasnovedosas en las especialidades de interes de la revista.

Presentacion de manuscritosLos autores interesados deberan enviar un manuscrito vıa e-mail al

Editor de la revista (jarmengol@ub.edu). Tambien se podra enviar por correoordinario un original mas dos copias en papel y una en soporte magnetico(CD, DVD o similares) al Editor de la revista. La copia en soporte informaticoestandarizado incluira un fichero unico con el texto, las tablas y las figurassegun las presentes normas de publicacion. Este fichero estara en un editorde textos compatible con un ordenador tipo PC (preferentemente Word, oprograma compatible).

Tanto los originales impresos como las copias en soporte magnetico serealizaran en hojas tipo A-4 y estaran escritos a doble espacio. Los artıculosno sobrepasaran las 6 000 palabras en el texto ni 25 hojas impresas (incluidasfiguras y tablas) y podran estar escritos en castellano o en ingles. Excepcio-nalmente, y previa consulta con el Editor, se podran presentar manuscritos demayor longitud que versen sobre revisiones generales, sistematica de grupostaxonomicos amplios o estudios regionales comparativos de un mismo tipo deecosistemas acuaticos. No se aceptaran aquellos trabajos que no cumplan laspresentes instrucciones de publicacion.

El Comite Editorial de Limnetica decidira sobre la publicacion o no delos trabajos recibidos, e informara de ello a los autores. Los autores recibiranantes de su publicacion una copia de las primeras pruebas de imprenta de sutrabajo para su correccion. Una vez publicado el trabajo, el autor responsablede la publicacion recibira una copia en formato pdf.

Estructura del manuscritoTodas las palabras en MAYUSCULAS se acentuaran tanto en el TITULO

como en los apartados (INTRODUCCION, etc.).La primera pagina del manuscrito ha de contener los siguientes apartados:• Tıtulo en mayusculas.• Relacion de autores con indicacion del autor responsable a efectos de la• correspondencia. Se debera indicar expresamente la direccion e-mail de• dicho autor.• Direccion postal completa de los autores.• Tıtulo abreviado.La segunda pagina incluira el Resumen en castellano, palabras clave, el

Abstract en ingles y keywords. Tanto el Resumen como el Abstract no deberansobrepasar las 400 palabras y deberan incluir el tıtulo del trabajo en el idiomacorrespondiente.

Las siguientes paginas se ordenaran en apartados que se estructuraran alestilo cientıfico. Los apartados y parrafos del texto comenzaran sin sangrado.Se acentuaran las mayusculas en todos los casos.

Los apartados se escribiran sin numerar y se escalaran segun el siguienteformato:

Apartado primario.- Mayusculas y en negrita (INTRODUCCION).Apartado secundario.-Minusculas y en negrita.Apartado terciario.- Italica.Apartado de cuarto nivel.- Letra normal subrayada.Siguientes niveles.- numericos (1), (1.1), (1.1.1), etc.Las Tablas constituyen una de las partes mas costosas en tiempo y

presupuesto por lo que se ruega se preparen procurando ocupar el mınimoespacio posible. Las tablas pueden tener la anchura de una columna (8 cm) odos columnas (16 cm) y su longitud no puede exceder de 25 cm. Se incluiranal final del manuscrito y tendran numeracion arabiga. En el texto siempre secitaran de forma completa (p.e. Segun se puede ver en la Tabla 6... etc. o,Los datos (Tabla 6) indican que... etc.) y nunca en forma abreviada –Tab. 6 otab. 6. Las leyendas de las tablas se presentaran en castellano e ingles y seincluiran en el mismo apartado que el texto de las figuras. No deberan usarselıneas verticales y los encabezamientos de las columnas deberan ser breves. Se

prestara particular atencion en no publicar tablas que dupliquen informacionque ya esta en forma de figuras.

Las figuras tendran numeracion arabiga con el texto explicativo en el pie.El texto incluira la version en castellano y en ingles. Las figuras pueden ir atres tipos de caja 8, 12.5 y 16 cm. Los autores procuraran que los originalestengan el tamano de letra y el grueso de lınea necesario para que al reducirsepuedan ser interpretables y legibles. No se aceptaran figuras que no cumplaneste requisito.

Los pies de figura, junto con los encabezamientos de las tablas, estaran enuna pagina aparte situada inmediatamente despues de la bibliografıa y antesde las tablas y figuras.

Las citas de las figuras en el texto se haran de forma completa y enminuscula cuando se inserte dentro del texto de un parrafo (p.e. En la figura1 se indica la situacion de los puntos de muestreo). Por el contrario, secitara de forma abreviada y en mayuscula cuando este entre parentesis y norelacionada directamente con el texto del parrafo [p.e. Las muestras se hanrecogido en cinco estaciones distribuidas a lo largo del rıo (Fig. 1) y con unaperiodicidad mensual]. Las figuras y fotografıas en color solo se aceptaran deforma excepcional y previa consulta con el Editor.

Las unidades se expresaran preferiblemente en el Sistema Internacional(S.I.) con los sımbolos en forma abreviada cuando vayan precedidos de unaexpresion numerica. Cuando se exprese un valor como combinacion de dosunidades estas se indicaran con el signo aritmetico correspondiente p.e. m/s,mol/ m3, ind/l, pero para mas de dos unidades se usaran exponentes, p.e. m/s,mol/m3 ind/l, pero para mas de dos unidades se usaran exponentes, p.e. mgCm−2 h−1 μmol m−2 s−1.

Las cantidades con decimales se expresaran con un punto (4.36), los milescon cuatro numeros sin ninguna separacion o sımbolo (4392) y para valoresiguales o superiores a las decenas de mil se intercalaran blancos separando losmiles (13 723 o 132 437). Siempre que sea posible se indicaran los numeroscon notacion exponencial decimal con el mınimo posible de decimales (13.7 ·103, 13.2 · 104).

La BIBLIOGRAFIA se ordenara al final del texto, alfabeticamente ycronologicamente para cada autor, segun las pautas siguientes:

• Revistas:RUEDA, F. J., E. MORENO-OSTOS & J. ARMENGOL. 2006. The

residence time of river water in reservoirs. Ecological Modelling, 191: 260-275.

GRACA M. A. S. & C. CANHOTO. Leaf litter processing in low orderstreams. Limnetica, 25(1-2): 1-10.

RECHE, I., E. PULIDO-VILLENA, R. MORALES-BAQUERO & E. O.CASAMAYOR. 2005. Does ecosystem size determine aquatic bacterial rich-ness? Ecology, 86: 1715-1722.

• Libro:KALFF, J. 2002. Limnology. Prentice Hall. NJ. USA. 592 pp.• Capıtulo de libro:IMBODEN, D. M. 1998. The influence of Biogeochemical Processes on

the Physics of Lakes. In: Physical Processes in Lakes and Oceans. J. Iberger(ed.): 591-612. American Geophysical Union. Washington. USA.

• Congresos:GEORGE, D. G. 2006. Using airborne remote sensing to study the mixing

characteristics of lakes ans reservoirs.10th European Workshop on PhysicalProcesses in Natural Waters. June 26-28, 2006. Granada, Spain: 2001-207.

• Informes:DOLZ, J. & E. VELASCO. 1990. Analisis cualitativo de la hidrologıa

superficial de las cuencas vertientes a la marisma del Parque Nacional deDonana (Informe Tecnico). Universidad Politecnica de Cataluna. 152 pp.

• Tesis y Maestrias:MORENO-OSTOS, E. 2004. Spatial dynamics of phytoplankton in El

Gergal reservoir (Seville, Spain). Ph.D. Thesis. University of Granada. 354 pp.THOMPSON, K. L. 2000. Winter mixing dynamics and deep mixing in

Lake Tahoe. Master’s Thesis, University of California, Davis. 125 pp.En el manuscrito se listaran unicamente los trabajos citados en el texto; en

este, las referencias se haran en minusculas (Kalff, 2002; Dolz & Velasco,1991; Rueda et al., 2006). En ningun caso se aceptaran como referenciastrabajos no publicados (p.e. en preparacion) o aun no aceptados (p.e. enviado).Sı se podran incluir citas de trabajos aceptados para su publicacion (enprensa). Se recuerda la conveniencia de reducir al maximo las referenciasbibliograficas de difıcil consulta como informes, resumenes a congresos, etc.

INSTRUCTIONS FOR AUTHORS

ScopeLimnetica publishes original research papers on ecology of continental

waters. Its scope includes ecology of rivers, lakes, reservoirs, lagoons andwetlands, biogeochemistry, paleolimnology, development of new methods,taxonomy, biogeography, and all aspects of theoretical and applied continentalaquatic ecology, like management and conservation, impact assessment,ecotoxicology and pollution. Limnetica will accept for publication scientificpapers presenting advances in knowledge or technological development, aswell as papers derived from new practical approaches on the topics coveredby the journal.

Manuscript presentationManuscripts must be submitted by e-mail to the journal Editor (jarmen-

gol@ub.edu). Manuscripts also can be sent to the Editor by regular mail (ori-ginal plus two hard copies and one digital copy. The digital copy must includea file with text, tables and figures following the present instructions, made withPC-compatible text-edition software (MSWord, Wordperfect, etc.).

Both hard and digital copies will be typed at double space on A-4 sheets.Papers can be written in Spanish or English, and must not exceed 6000 wordsof text nor 25 printed pages (figures and tables included). Exceptionally, andafter consulting the Editor, longer manuscripts can be published for generalreviews, systematics of broad taxonomic groups, or regional comparativestudies of one kind of aquatic ecosystems. Papers that do not follow the presentinstructions will be rejected.

Limnetica’s Editorial Board will decide whether to publish or not thereceived manuscripts, and will tell their decision to the authors. Prior topublication, authors will get galley proofs to be corrected. When the paperhas been published, the leading author will get a copy in pdf format.

Manuscript structureFor manuscripts in Spanish, words in UPPER CASE will be accentuated

when convenient, both in the title and section headings (INTRODUCCION,etc.).

The first page must include:• Title in upper case.• List of authors detailing the corresponding author, whose e-mail address• must be shown.• Complete postal address of authors.• Running title.The second page will include Abstract and key words, both in English and

Spanish. Abstracts must start with the title and not exceed 400 words.Following pages must be structured in sections following the scientific

style. Section headings and text will have no left indent. Upper case words inSpanish will be accentuated.

Sections and subsections will not be numbered, and must adjust to thefollowing format:

Main section.- Bold, upper case (INTRODUCTION).2nd-level section.- Bold, lower case.3rd-level section.- Italics.4th-level section.- Plain text, underlined.Lower-level sections.- They will go numbered (1), (1.1), (1.1.1), etc.Tables are one of the most costly parts, both in terms of time and money;

therefore, they must be drawn as compact as possible. Tables can be 1-column(8 cm) or 2-column (16 cm) wide, and their length cannot exceed 25 cm. Theywill be included at the end of the manuscript and numbered in Arabic numbers.In the text they will be written in complete form (e.g., as can be seen in Table6. . . , or Data (Table 6) show that. . . ), never in abbreviated form (neither Tab.6 nor tab. 6). Table captions will be written in both English and Spanish, andwill be included in the text in the same section than Figure legends. No verticallines can be drawn in tables, and column headings must be short. No table willbe published that shows information presented in figures.

Figures will have Arabic numbers, and legends will go below, both inEnglish and Spanish. Figures can fit three box-sizes: 8 cm, 12.5 cm, or 16 cm.

Authors must make sure that font size and line thickness can be easily readafter reduction, otherwise figures will be rejected.

Figure legends and table captions will go in a page after Literature Citedand before Tables and Figures.

Figure calls must be made in complete, lower case form when in the text(e.g., Location of sampling sites is shown in figure 1), in abbreviated, uppercase when going in a parenthesis and not directly related to the text [e.g.,Samples were taken monthly at five sites along the river (Fig. 1)]. The Editorwill accept to publish colour figures and photographs only exceptionally andwhen explicitly requested.

Units must be expressed preferably following the International System(I.S.), with abbreviated symbols when preceded by numeric expressions.Values combining two units must be expressed with the correspondingarithmetic sign, like m/s, mol/m3, ind/l, but when there are more than twounits exponentials must be used, like in mgC m−2 h−1, μmol m−2 s−1.

Decimal numbers will be expressed with a dot (4.36), thousands with 4digits, with no blank space or symbols (4392), and figures over ten thousandwill have blank space markings (13 723 or 132 437). Whenever possible thescientific notation will be used, with the smallest possible number of decimals(13.7·103, 13.2·104).

BIBLIOGRAPHY will be after the text, in alphabetic order, chronologi-cally for each author, and adhere to the following style:

• Journals:RUEDA, F. J., E. MORENO-OSTOS & J. ARMENGOL. 2006. The

residence time of river water in reservoirs. Ecological Modelling, 191: 260-275.

GRACAM. A. S. & CRISTINA CANHOTO. Leaf litter processing in loworder streams. Limnetica, 25(1-2): 1-10.

RECHE, I., E. PULIDO-VILLENA, R. MORALES-BAQUERO & E. O.CASAMAYOR. 2005. Does ecosystem size determine aquatic bacterial rich-ness? Ecology, 86: 1715-1722.

• Books:KALFF, J. 2002. Limnology. Prentice Hall. NJ. USA. 592 pp.• Book chapters:IMBODEN, D. M. 1998. The influence of Biogeochemical Processes on

the Physics of Lakes. In: Physical Processes in Lakes and Oceans. J. Iberger(ed.): 591-612. American Geophysical Union. Washington. USA.

CASTRO, M., J. MARTIN-VIDE & S. ALONSO. 2005. El clima deEspana: pasado, presente y escenarios de clima para el siglo XXI. In:Evaluacion preliminar de los impactos en Espana por efecto del CambioClimatico. J. M. Moreno Rodrıguez (ed.): 113-146. Ministerio de MedioAmbiente.

• Conferences:GEORGE, D. G. 2006. Using airborne remote sensing to study the mixing

characteristics of lakes ans reservoirs.10th European Workshop on PhysicalProcesses in Natural Waters. June 26-28, 2006. Granada, Spain: 2001-207.

• Reports:DOLZ, J. & E. VELASCO. 1990. Analisis cualitativo de la hidrologıa

superficial de las cuencas vertientes a la marisma del Parque Nacional deDonana (Informe Tecnico). Universidad Politecnica de Cataluna. 152 pp.

• PhD and Master Dissertations:MORENO-OSTOS, E. 2004. Spatial dynamics of phytoplankton in El

Gergal reservoir (Seville, Spain). Ph.D. Thesis. University of Granada.354 pp.

THOMPSON, K. L. 2000. Winter mixing dynamics and deep mixing inLake Tahoe. Master’s Thesis, University of California, Davis. 125 pp.

The Bibliography will only contain papers cited in the text, where theymust go in lower case (Margalef, 1975; Wetzel & Likens, 1991; Riera et al.,1992). In no case will unpublished (e.g., in prep., submitted) papers be cited,unless they are accepted for publication (in press). References to works hardto get (reports, conference abstracts, etc.) must be limited to the minimumpossible.

ORDEN DE SUSCRIPCIÓN

NÚMEROS ATRASADOS

En el caso de que la persona o dirección a la que se factura seadiferente de la indicada anteriormente, utilice estos espacios:

Población PoblaciónPaís PaísC.P.

Tel. / Fax NIF / CIF

En el caso de que la persona o dirección a la que se factura seadiferente de la indicada anteriormente, utilice estos espacios:

Población PoblaciónPaís PaísC.P.

Tel. / Fax NIF / CIF

Formas de pago

Formas de pago Ejemplares atrasados

Suscripción anual (2 n )os

Por transferencia bancaria a: CC: 0075 0233 60 0600277602En este caso indicar “Suscripción LIMNETICA” en la transferencia y remitir el resguardo de lamisma junto con esta orden de suscripción.

Por transferencia bancaria a: CC: 0075 0233 60 0600277602En este caso indicar “Suscripción LIMNETICA” en la transferencia y remitir el resguardo de lamisma junto con esta orden de suscripción.

Por cheque bancario a nombre de la ASOCIACIÓN IBÉRICA DE LIMNOLOGÍA

Por cheque bancario a nombre de la ASOCIACIÓN IBÉRICA DE LIMNOLOGÍA

Mediante tarjeta VISA nº

Mediante tarjeta VISA nº

fecha de caducidad

fecha de caducidad

España: 50 Euros

Extranjero: 60 Euros

Indicar los números

4O EUROS/nº

Revista

LIM

NETIC

A

Secció

ndePublica

ciones

Asocia

ciónIbérica

deLim

nología

C/LosÁngeles,

33

46920Misla

ta(Valen

cia)

SPAIN

Revista

LIM

NETIC

A

Secció

ndePublica

ciones

Asocia

ciónIbérica

deLim

nología

C/LosÁngeles,

33

46920Misla

ta(Valen

cia)

SPAIN

Fran

queo

Fran

queo

Volumen 27 (2) Diciembre de 2008