Marta Isabel Sánchez Ordóñez · 1. Relaciones predador-presa 1.1 En primer lugar estudiamos la comunidad de invertebrados y su efecto sobre la comunidad de limícolas, analizando

1

Dpto. de Biología Aplicada Dpto. de Biología Ambiental y Salud PúblicaESTACIÓN BIOLÓGICA DE DOÑANA UNIVERSIDAD DE HUELVA(CSIC)

TESIS DOCTORAL

RELACIONES ECOLÓGICAS ENTRE LIMÍCOLAS E INVERTEBRADOS ENLAS SALINAS DE LAS MARISMAS DEL ODIEL

Memoria presentada por Marta Isabel Sánchez Ordóñez para optar al grado deDoctora en Biología

Directores

Dr. Andy J. Green Dr. Eloy M. Castellanos VerdugoCientífico Titular Profesor Titular de UniversidadEstación Biológica de Doñana (CSIC) Universidad de Huelva

2005

2

3

A mis padres

4

INDICE

5

INDICE

Introducción general 9

CAPÍTULO 1 19

Seasonal variation in the diet of the Redshank Tringa totanus in the Odiel

Marshes, south-west Spain : a comparison of faecal and pellet analysis.

Marta I. Sánchez, Andy J. Green and Eloy M. Castellanos.

Bird Study 52: 210-216.

CAPÍTULO 2 35

Spatial and temporal fluctuations in use by shorebirds and in availability of

chironomid prey in the Odiel saltpans, south-west Spain.

Marta I. Sánchez, Andy J. Green and Eloy M. Castellanos.

Hydrobiologia 567: 329-340.

CAPÍTULO 3 61

Seasonal and spatial variation in the aquatic invertebrate community at the

Odiel salt pans (SW Spain) and their implications for migratory waders.

Marta I. Sánchez, Andy J. Green & Eloy M. Castellanos.

Archive für Hydrobiologie 166: 199-223.

INDICE

6

CAPÍTULO 4 93

Shorebird predation affects abundance and size distribution of benthic

chironomids in saltpans: an exclosure experiment.

Marta I. Sánchez, Andy J. Green and Raquel Alejandre.

Journal of North American Benthological Society 25(1): 9-18.

CAPÍTULO 5 117

Cestodes from Artemia parthenogenetica (Crustacea, Branchiopoda) in the

Odiel Marshes, Spain: a systematic survey of cysticercoids.

Boyko B. Georgiev, Marta I. Sánchez, Andy J. Green, Pavel N. Nikolov,

Gergana P. Vasileva and Radka S. Mavrodieva.

Acta Parasitologica 50 (2): 105-117.

CAPÍTULO 6 145

Passive internal transport of brine shrimps and seeds by migratory waders in

the Odiel marshes, south-west Spain.

Marta I. Sánchez, Andy J. Green, Francisco Amat & Eloy M. Castellanos.

Marine Biology 151: 1407-1415.

Journal of Avian Biology 37: 201-206.

INDICE

7

CAPÍTULO 7 171

Dispersal of invasive and native brine shrimps Artemia (Anostraca) via

waterbirds.

Andy J. Green, Marta I. Sánchez, Francisco Amat, Jordi Figuerola, Francisco

Hontoria, Olga Ruiz, Francisco Hortas.

Limnology and Oceanography, 50 (2): 737-742.

Apéndices 187

Discusión general 195

Conclusiones 215

Agradecimientos 219

8

INTRODUCCIÓN

9

INTRODUCCIÓN

Justificación de la tesis

Las marismas del Odiel (Huelva, 37º17'N 06º55'W) se encuentran en uno de los más

productivos y diversos estuarios del suroeste de Europa (Castellanos et al. 1998). Su

elevado valor ecológico le ha llevado a su protección como Paraje Natural, Sitio Ramsar

(Bernués 1998), Reserva de la Biosfera y Zona de Especial Protección para las aves en

la Unión Europea. Miles de limícolas que siguen la ruta migratoria del Atlántico Este

dependen de las marismas del Odiel, utilizándolas para descansar y reponer energía año

tras año.

La existencia de hábitats alternativos de alimentación en las zonas estuarinas

reviste una gran importancia para el mantenimiento de las poblaciones de limícolas, ya

que los fangos intermareales donde se alimentan, sólo están disponibles una parte del

ciclo mareal, permaneciendo anegados durante la marea alta. Además su importancia es

aún mayor teniendo en cuenta la degradación y pérdida de hábitats naturales a que

estamos asistiendo durante las últimas décadas, y que afecta de forma especialmente

intensa al litoral, amenazando las poblaciones de limícolas de todo el mundo.

Las salinas costeras son humedales manejados por el hombre que ofrecen unas

condiciones excepcionales para la alimentación y descanso de las aves acuáticas, tanto

migradoras como sedentarias (Pérez-Hurtado & Hortas 1991, Velásquez 1993, Masero

& Pérez-Hurtado 2001). Se estima que aproximadamente la mitad de los 500.000

limícolas migratorios e invernantes que utilizan el Mediterráneo durante los meses de

invierno, se encuentran en salinas (Sadoul et al. 1998), jugando un papel importantísimo

en el ciclo anual y conservación de las poblaciones de estos migradores de largas

distancias.

La producción de sal es, probablemente, la industria con mayor tradición en la

cuenca Mediterránea, modelando el paisaje de toda la costa a lo largo de los siglos. De

las 7158 has que ocupan las marismas del Odiel, 1174 están transformadas en salinas,

siendo la unidad ambiental de la marisma que mayor capacidad de carga soporta. Sin

embargo, junto con los humedales costeros, las salinas continúan desapareciendo en la

región Mediterránea debido a la intensa presión antrópica. La actividad salinera en

INTRODUCCIÓN

10

general ha ido reduciéndose considerablemente, mientras la tradicional prácticamente ha

desaparecido, abandonándose sus instalaciones o utilizándose para otros fines.

La situación actual de crisis ambiental que afecta especialmente a los humedales

costeros demanda de estudios multidisciplinares que incluyan aspectos que relacionen

tanto la dinámica de las aves, como la de sus invertebrados presa y los factores

ambientales que les afectan. En definitiva, estudios que permitan comprender mejor las

relaciones ecológicas de los sistemas acuáticos con el fin de poder compensar el

impacto de la actividad humana sobre los ecosistemas naturales y artificiales. Los

sistemas manejados por el hombre, ofrecen una oportunidad excepcional para

desarrollar este tipo de estudios, ya que con una adecuada gestión es posible aumentar

con creces su valor ecológico de cara a las aves (Rehfisch 1994).

Existen diversos estudios en nuestra región sobre la autoecología de las aves

limicolas. No es prioritario, por tanto, repetir dichos estudios en Odiel, ya que no caben

esperar diferencias sustanciales. En cambio, es prioritario profundizar en el

conocimiento de las comunidades de invertebrados y de sus interacciones con la de las

aves limícolas, estudios aún escasos en la región Mediterránea. El funcionamiento del

ecosistema acuático sólo puede ser entendido cuando se toma en su conjunto. Hurlbert y

colaboradores (1983) demostraron la fuerte influencia que ejercían las aves acuáticas en

general sobre la estructura de los ecosistemas acuáticos, acuñando el término de

ornitolimnología para aunar estas dos disciplinas (ornitología y limnología). Los

limícolas en particular presentan requerimientos energéticos extremadamente altos y

altísimas tasas de alimentación (Nagy 2001), pudiendo ejercer, en un breve periodo de

tiempo, profundos efectos sobre la dinámica de las poblaciones de presa (y especies

asociadas a través de efectos indirectos) y condicionando, en último término, la

dinámica del ecosistema acuático completo. Además de las relaciones predador-presa,

las relaciones parásito-hospedador también juegan un papel clave en estructurar las

comunidades animales. Existe una gran diversidad de caminos por los que los parásitos

pueden afectar a sus hospedadores. La mayoría de los cestodos (Brown et al. 2001)

(muchos de ellos parásitos de limícolas y otras aves acuáticas) utilizan hospedadores

intermedios a los que "manipulan", induciendo cambios en el fenotipo y en el

comportamiento, con el fin de alcanzar a sus hospedadores definitivos. La estrategia de

infección por transmisión trófica, por la que el parásito incrementa la susceptibilidad a

la predación, puede ocasionar profundos efectos sobre las poblaciones del hospedador

intermedio. Además los individuos parasitados pueden considerarse como nuevos

INTRODUCCIÓN

11

organismos en el ecosistema, involucrados en nuevas interacciones con otras especies

(Thomas et al, 1997). Otro tipo de relación interespecífica entre limícolas e

invertebrados, es la dispersión de invertebrados por las aves. Los procesos de dispersión

son centrales para mantener la biodiversidad de los ecosistemas acuáticos y explicar los

patrones de distribución y composición de las comunidades. Las aves acuáticas se han

considerado como uno de los principales vectores de dispersión de propágulos de

invertebrados y plantas acuáticas (Darwin 1859). Sin embargo, las evidencias sobre la

implicación real de las aves en la dispersión son escasas.

Esta tesis pretende ampliar el conocimiento del funcionamiento de los

ecosistemas de salinas, a través del estudio de las relaciones ecológicas entre limícolas e

invertebrados en unas salinas de importancia internacional para las aves acuáticas.

Ahondar en el conocimiento básico de los sistemas artificiales, más susceptibles a la

gestión aplicada a través de estudios como el presente, ayudará a diseñar medidas para

compensar de alguna manera el impacto sobre las poblaciones de limícolas que supone

la pérdida de hábitats costeros en todo el mundo.

Presentación de la tesis

El objetivo de esta tesis lo abordamos desde tres puntos de vista.

1. Relaciones predador-presa

1.1 En primer lugar estudiamos la comunidad de invertebrados y su efecto sobre la

comunidad de limícolas, analizando los requerimientos tróficos, y determinando la

distribución y abundancia de aves e invertebrados (capítulos 1, 2, 3).

1.2. Una vez analizado el efecto de los recursos sobre la distribución y abundancia de

limícolas, abordamos el estudio desde el punto de vista del impacto de las aves sobre

sus recursos, determinando las consecuencias de la predación sobre la densidad,

biomasa y tamaño de presas (capítulo 4).

INTRODUCCIÓN

12

2. Relaciones parásito-hospedador

Realizamos una primera aproximación a las relaciones parásito-hospedador en el

sistema Cestodos-Artemia-limícolas, describiendo las diferentes especies de cestodos de

la población de Artemia del Odiel (capitulo 5). Este constituye el estudio más completo

realizado hasta la fecha en nuestro país.

3. Dispersión de propágulos

Finalmente analizamos la capacidad de dispersión de propágulos por los limícolas,

evaluando su implicación en la distribución de especies y estructura de las comunidades

de invertebrados acuáticos y plantas (capítulos 6 y 7).

El estudio de la dieta ha sido y continúa siendo uno de los primeros pasos en el estudio

de la ecología básica de las especies. Para los ecólogos de comunidades, la dieta juega

un papel central en determinar la dinámica de competencia entre especies (Mittelbach &

Osemberg 1994), interacciones predador-presa (Sih et al. 1985) e interacciones

indirectas de la comunidad (Wootton 1993). La forma en que los limícolas explotan los

recursos constituye una primera aproximación para comprender e interpretar su

distribución y abundancia. Por eso, en el capítulo 1 tratamos de identificar las presas

más importantes y su variación a lo largo del año, tomando como modelo el Archibebe

común Tringa totanus. Para ello comparamos los resultados de dos metodologías no

invasivas: análisis de excrementos y egagrópilas. Elegimos esta especie debido a su

abundancia en la zona de estudio y a la facilidad para recoger las muestras. Las larvas y

pupas de Chironomus salinarius, constituyeron la base de la dieta del Archibebe y

probablemente de otras especies durante el paso migratorio de primavera, por lo que en

el capítulo 2, abordamos la dinámica estacional de la comunidad de limícolas en

relación a la dinámica poblacional de su principal presa. De los quironómidos,

invertebrados dominantes en el sedimento, pasamos al estudio de los invertebrados de la

columna de agua, y en el capítulo 3 analizamos la distribución y dinámica estacional de

las diferentes especies (con especial atención a Artemia), así como su relación con las

características químicas del agua y otros factores. Chironomus salinarius en el bentos, y

Artemia parthenogenetica en la columna de agua son los principales invertebrados

INTRODUCCIÓN

13

presentes en la salina y constituyen las dos especies clave que unen los niveles de

detritus y fitoplancton con los niveles tróficos superiores, especialmente las aves.

Una vez estudiado cómo la distribución y abundancia de los recursos influye en

el establecimiento de la dinámica poblacional de los limícolas, analizamos cómo

simultáneamente la comunidad de aves altera las características de las poblaciones de

presa. La predación juega un papel fundamental estructurando las comunidades de

invertebrados bentónicos (Thrush 1999). Así, tras comprobar en los capítulos 1 y 2 la

importancia de los quironómidos para los limícolas, evaluamos, en el capítulo 4, el

efecto de la predación sobre la abundancia, biomasa y tamaño de larvas durante el paso

migratorio de primavera mediante un experimento de exclusión de predadores.

Como parte del estudio de la comunidad de invertebrados, una vez comprobada

la elevada tasa de parasitismo en la zona, y dada la importancia de las interaciones

ecologicas entre estos organismos y las aves, incluimos la descripción y abundancia de

las especies de metacestodos de Artemia (parásitos de limícolas y otras aves acuáticas)

(capítulo 5). La infección a través de Artemia supone una interesante estrategia de

encuentro entre el parásito y el hospedador, ya que la presencia de cisticercoides de

ciertas especies de cestodo está acompañada de cambios en la pigmentación y

modificaciones en el comportamiento de la artemia parasitada, lo cual pensamos,

aumenta la susceptibilidad a la predación, favoreciendo la infección del hospedador

final (Gabrion 1982, Amat 1991, datos propios sin publicar).

Los procesos de dispersión a larga distancia pueden tener efectos sobre la

diversidad y flujo genético entre poblaciones (Bohonak 1999, Freeland et al. 2000,

Figuerola et al. 2005), y afectar a la distribución de especies y estructura de las

comunidades. A pesar de que las aves han sido consideradas como importantes vectores

de dispersión de organismos acuáticos, son escasas las evidencias empíricas de dicho

transporte (Figuerola & Green 2002), particularmente para el caso de los limícolas,

migradores de largas distancias y con un elevado potencial como dispersantes. En el

primer capítulo encontramos que una gran cantidad de propágulos (especialmente

quistes de Artemia y semillas) eran consumidos por los limícolas. Por ello en el

capítulo 6 cuantificamos en detalle el número de propágulos intactos y su viabilidad

tras la ingestión en diferentes especies de limícolas y estaciones del año. Analizamos el

efecto de la dieta y otros factores sobre la probabilidad de ingestión de propágulos y la

supervivencia a la ingestión. El hecho de que las aves puedan actuar de manera efectiva

en la dispersión de organismos, plantea la cuestión, hasta ahora inexplorada, de cuál es

INTRODUCCIÓN

14

la implicación de las aves en un problema de actualidad que amenaza la diversidad a

nivel global como es la expansión de especies exóticas. En el capítulo 7 evaluamos la

capacidad de los limícolas para dispersar Artemia franciscana, una especie invasora

originaria de Norte América, que está desplazando a una gran velocidad a sus

congéneres autóctonos (la especie sexual A. salina, y las líneas partenogenéticas

diploides y tetraploides) poniendo en grave peligro sus poblaciones y aquellas especies

estrechamente asociadas como sus parásitos (Amat et al. 2005). Hasta el momento no se

ha encontrado A. franciscana en el Odiel, donde existe una población partenogenética

principalmente diploide con un pequeño porcentaje tetraploide (Amat et al. 2005). Sin

embargo su vulnerabilidad es evidente teniendo en cuenta que ya ha sido registrada en

los humedales ibéricos más importantes para los limícolas en la ruta migratoria del

Atlántico Este, siendo la única especie presente en Portugal y habiéndose localizado ya

en la bahía de Cádiz. Los resultados de este capítulo permiten valuar el riesgo que

supone el que los limícolas estén transportando quistes de esta especie invasora, de la

bahía de Cádiz o el Algarve, hacia Odiel.

De esta forma hemos abordado el estudio de las relaciones ecológicas entre

limícolas e invertebrados. Por un lado estudiando la distribución y abundancia de

predadores y presas, y analizando los efectos directos que unos ejercen sobre los otros;

por otro lado con una primera aproximación a las relaciones entre limícolas y

metacestodos; y finalmente estudiando las consecuencias derivadas de la predación de

formas de resistencia o propágulos que pueden ser dispersados por las aves.

Los capítulos de esta tesis están presentados en forma de artículos, algunos de los cuales

han sido ya aceptados y otros están enviados pendientes de aceptación. Por esa razón

resulta inevitable el solapamiento de algunas frases de los apartados de introducción y

métodos. No obstante los datos y resultados que se presentan en cada capítulo son

originales y específicos de cada uno.

Bibliografía

Amat, F., Illescas, M.P. & Fernandez, J. 1991. Brine shrimp Artemia parasitized by

Flamigolepis liguloides (Cestoda, Hymenolepidadae) cysticercoids in Spanish

Mediterranean salterns. Vie Milieu 41 (4): 237-244.

INTRODUCCIÓN

15

Amat, F., Hontoria, F., Ruiz, O., Green, A.J., Sánchez, M.I., Figuerola, J. & Hortas, F.

2005. The American brine shrimp Artemia franciscana as an exotic invasive species in

the Western Mediterranean. Biological Invasions 7: 37-47.

Bernués, M., ed., 1998. Humedales Españoles inscritos en la Lista del Convenio de

Ramsar. Ministerio de Medio Ambiente, Organismo Autónomo de Parques Nacionales,

Madrid.

Bohonak, A. J. 1999. Dispersal, gene flow and population structure. Quarterly Review

of Biology 74: 21-45.

Brown, S. P., Renaud, F., Guégan, J.-F. & Thomas, F. 2001. Evolution of trophic

transmission in parasites: the need to reach a mating place?. Journal of Evolutionary

Biology 14: 815-820.

Castellanos, E. M., Heredia, C., Figueroa M. E. & Davy, A. J. 1998. Tiller dynamics of

Spartina maritima in successional and non-successional mediterranean salt marsh.

Plant. Ecology 137: 213-225.

Darwin, C. 1859. On the Origin of Species by Means of Natural Selection. Murray,

London.

Figuerola, J. & Green, A. J. 2002. Dispersal of aquatic organisms by waterbirds: a

review of past research and priorities for future studies. Freshwater Biology 47: 483-

494.

Figuerola, J., Green, A.J. & Michot, T.C. 2005. Invertebrate eggs can fly: Evidence of

waterfowl mediated gene-flow in aquatic invertebrates. American Naturalist: en prensa.

Freeland, J. R., Romualdi, C. & Okamura, B. 2000. Gene flow and genetic diversity: a

comparison of freshwater bryozoan populations in Europe and North America. Heredity

85: 498-508.

INTRODUCCIÓN

16

Gabrion, C., MacDonald-Crivelli, G. & Boy, V. 1982. Dynamique des populations

larvaires du cestode Flamingolepis liguloides dans une population d'Artemia en

Camargue. Acta Oecologica 3(2): 273-293.

Hulbert, S. H. & Chang, C. C. Y. 1983. Ornitholimnology: Effects of grazing by the

andean flamingo (Phoenicoparrus andinus). Proceeding of the Natural Academy of

Science 8: 4766-4769.

Masero, J. A. & Pérez-Hurtado, A. 2001. Importance of the supratidal habitas for

maintaining overwintering shorebirds populations: how redshank use tidal mudflats and

adjacent saltworks in southern Europe. The Condor 103: 21-30.

Mittelbach, G.G. & Osenberg, C.W. 1994. Using foraging Theory to study trophic

interactions. Pages 45-59 in D.J. Stouder, K.L. Fresh, and R.J. Feller (eds.), Theory and

application in fish feeding ecology, Belle W. Baruch Library in Marine Sciences, no.

18, University of South Carolina Press, Columbia, South Carolina.

Nagy, K. A. 2001. Food requirements of wild animals: Predictive equations for free-

living mammals, reptiles, and birds. Nutrition Abstracts and Reviews, Series B 71, 21R-

31R.

Pérez-Hurtado, A. & Hortas, F. 1991. Information about the habitat use of salines and

fishponds by wintering waders in Cádiz Bay. Wader Study Group Bulletin 66:48-53.

Rehfisch, M. M. 1994. Man-made lagoons and how their attractiveness to wader might

be increased by manipulating the biomass of an insect benthos. Journal of Applied

Ecology 31: 383-401.

Sadoul, N., Walmsley, J.G. & Charpentier, B. 1998. Salinas and Nature Conservation.

Conservation of Mediterranean Wetlands No.9, Tour du Valat, Arles (France), 96 pages,

(In French & English)

INTRODUCCIÓN

17

Sih, A., Crowley, P., McPeek, M., Petranka, J. & Strohmeier, K. 1985. Predation,

competition and prey communities: a review of field experiments. Annual Rev. Ecol.

Systematics 16: 269–311.

Thrush, S. F. 1999. Complex role of predators in structuring soft-sediment

macrobenthic communities: Implications of changes in spatial scale for experimental

studies. Australian Journal of Ecology 24: 344-354.

Velásquez, C. R. 1992. Managing artificial saltpans as a waterbirds habitat: species'

responses to water level manipulation. Colonial Waterbirds 15 (1): 43-55.

Wootton, J. T 1993. Indirect effects and habitat use in an intertidal community:

interaction chains and interaction modifications. American Naturalist 141(1): 71-89.

18

CHAPTER 1 THE DIET OF REDSHANK

19

CHAPTER 1

SEASONAL VARIATION IN THE DIET OF THE REDSHANK TRINGA

TOTANUS IN THE ODIEL MARSHES, SOUTH-WEST SPAIN: A

COMPARISON OF FAECAL AND PELLET ANALYSIS

MARTA I. SÁNCHEZ1, 2, ANDY J. GREEN1 AND ELOY M. CASTELLANOS2

1Departamento de Biología Aplicada, Estación Biológica de Doñana, Avenida de María Luisa s/n,

Pabellón del Perú, 41013 Sevilla, Spain2Departamento de Biología Ambiental y Salud Pública, Facultad de Ciencias Experimentales,

Universidad de Huelva, Campus de El Carmen, Avda. Fuerzas Armadas s/n, 21071 Huelva, Spain

Key words: Tringa totanus, diet, faecal analysis, pellet analysis, Odiel marshes,

migration

Running head: The diet of Redshank

Bird Study 52: 210-216.


20

ABSTRACT

Capsule Redshank diet from southern Europe during migration shows spatial and

seasonal variations

Aims To assess seasonal variation in Redshank diet at a major passage site, and to

compare data derived from analysing pellets or faeces.

Methods At the Odiel marshes in 2001, pellets from spring migration (39), autumn

migration (121) and midwinter (15) were analysed, together with faecal samples from

autumn (84).

Results The abundance of different invertebrate groups in pellets varied between

seasons. In spring, Chironomus salinarius pupae and larvae dominated by volume,

followed by Ephydridae larvae and the beetle Paracymus aenus. Polychaetes and

molluscs dominated in autumn, and isopods in midwinter. In autumn, chironomid

larvae, Mesembryanthemum nodiflorum seeds and Artemia cysts were relatively more

abundant in faeces, whereas polychaetes, isopods, molluscs and cestode cysticercoids

were more abundant in pellets. Harder and/or larger items were thus relatively more

abundant in pellets than faeces. Pellet analysis gave more emphasis to mudflat prey, and

faeces to saltpan prey.

Conclusion Pellet and faecal analysis give different results for wader diet, and it is

useful to combine the two methods. However, they show significant correlations both in

diet range and rank abundance of prey items. Redshank diet shows much seasonal and

spatial variation in southern Europe.


21

INTRODUCTION

The Odiel Marshes in south-west Spain are protected as a Biosphere Reserve, Ramsar

site, EU-SPA and Natural Park owing to their importance for migratory waterbirds.

They are internationally important for six species of waders including the nominal

subspecies of the Redshank Tringa totanus totanus. Here we present a detailed study of

the diet of Redshank Tringa totanus totanus using the saltpan complex within the Odiel

Marshes. This subspecies is under decline at a flyway scale (Wetlands International

2002). Loss of habitat on passage sites may be one cause of decline since energetic

requirements are particularly high during migration (Recher 1966, Davis & Smith 1998,

Pfister et al. 1998).

We compare Redshank diet during spring migration, autumn migration and

winter and assess the relative importance of prey items from salt pans and from

surrounding tidal mudflats in each period. Salt pans are known as important foraging

habitat for waders on the Atlantic coast (Velasquez et al. 1991, Masero 2003) although

no one has previously assessed their relative importance at different times of the annual

cycle. Ours is the first study of Redshank diet in southern Europe during migration.

Previous studies during migration were carred out in northern Europe where different

prey are available (Goss-Custard & Jones 1976, Goss-Custard et al. 1977).

Many species of shoreirds produce pellets containing the indigestible hard parts

of their prey. Shorebirds known to produce pellets include Turnstone Arenaria interpres

(Jones 1975), Grey plover Pluvialis squatarola (Goss-Custard & Jones 1976), Curlew

Numenius arquata (Pérez-Hurtado et al. 1997) and Common Sandpiper Actitis

hypoleucos (Arcas 2001).

Comparisons of diet between wader species, seasons or sites (e.g. Pérez-Hurtado

et al.1997, Kalejta 1993) are complicated by the use of various methods with distinct

biases (analyses of stomach contents, faeces, pellets or direct observations). We

compare the data on Redshank diet provided by analyses of faeces and pellets. Although

several authors have pointed out the need to combine the results of different methods

(Duffy & Jackson 1986, Jenni et al. 1989, Arcas 1999, Harris & Wanless 1993,

Sheiffarth 2001), as far as we are aware ours is the first study to make a detailed

comparison of methods used at the same time and place. Previous comparisons of

analyses of faeces and pellets have focussed only on the size of prey items of a given


22

type (Goss-Custard et al. 1977, Dekinga & Piersma 1993), whereas we focus on the

whole range of prey items including their relative abundance and overall diversity.

STUDY AREA

The Odiel Marshes (37º17'N 06º55'W) is an estuarine complex formed at the mouths of

the rivers Odiel and Tinto, south-west Spain. They contain 6,000 ha of intertidal

mudflats and 1,185 ha of salt pans. During weekly counts throughout 2001 of our study

area which occupied 27% of the total area of salt pans, we counted up to 20,775 waders

including up to 2,170 Redshank. Numbers of Redshank peak during spring and autumn

migration.

METHODS

Samples were collected at different times of the year in 2001 from Redshank at a high

tide roost. We collected 15 pellets from midwinter Jan 13. We then collected a total of

39 pellets during spring migration from late February to late April. Likewise, 121

pellets were collected during autumn migration. In autumn, we collected 84 faecal

samples. To avoid repeated sampling from the same individuals, we only collected fresh

samples and did not collect more than one sample where several were found within 20

cm of each other. We only took samples when a monospecific group of Redshank were

observed with a telescope at the sampling spot for at least 30 minutes.

Samples were stored at 5 ºC. Prior to analysis, they were rehydrated and

separated in water, then observed with a binocular microscope in a Petri dish. Prey

items were identified using suitable keys (see Sánchez et al. 2000). The volume of each

diet component was estimated as a proportion of the total sample volume, using the

following seven categories of relative abundance: absent (assigned the rank of 1 for

non-parametric analysis), <10 % (rank of 2), 10-25 % (3), 26-50 % (4), 51-75 % (5), 76-

90% (6) and >90 % (7).

The percentage of individual samples in which each food item was recorded (i.e.

the percentage occurrence, PO) was calculated for faecal and pellet samples for each

season. Differences in PO of different diet components in pellets between the three


23

seasons (spring, autumn and winter) were analysed using Kruskal-Wallis tests

employing Statistica 5.5 (StatSoft 1999). Only food items with a PO > 10% in at least

one season were analysed, and P values were Bonferroni corrected to avoid type I

errors. Mann-Whitney U tests were used as post-hoc tests to determine significant

differences between each season. The numbers of readily countable items (Artemia

cysts (eggs), seeds and cestode cysticercoids) in pellets were compared between seasons

in the same way. All these items were counted, including those that were partially

digested.

The relative abundance of prey items and number of countable items were

compared for faecal samples and pellets collected in autumn in a similar way, using

Bonferroni-corrected Mann-Whitney U tests and analyzing those items with a PO of >

10% in at least one type of sample. In order to compare the diversity of faecal and pellet

contents, we compared the number of prey items recorded in each with Mann-Whitney

U tests. To assess the similarity and repeatability of pellet and faecal contents, we

calculated the average abundance for each prey item (i.e. the average of the seven ranks

defined above), and compared these average ranks with a Spearman's rank correlation.

We excluded green plant material (mainly Chenopodiacea) from our comparison of

pellets and faeces because, in the case of faeces (but not pellets), we were unable to

distinguish reliably between material excreted and material that had become stuck to the

faeces after excretion.

In order to compare the relative abundance of chironomid larvae and pupae in

pellets for a given season, we used a sign test for each season.

RESULTS

Seasonal differences in pellet contents

The prey item recorded most often in spring pellets was chironomid C. salinarius pupae

(74% of pellets). The other prey items occurring in more than 50% of pellets were (in

order of decreasing frequency) the beetle Ochthebius corrugatus, Artemia cysts, C.

salinarius larvae, unidentified Coleoptera and green plant material (Table 1). In winter

pellets, Isopoda were the most frequently recorded prey (67% of pellets), followed by

unidentified Coleoptera, polychaetes and green plant material (Table 1). In autumn

pellets, polychaetes occurred most frequently (84%) followed by molluscs (Table 1).


24

Chironomid pupae also dominated in spring pellets by volume (Table 1),

representing >90% of the volume in 13 (33%) of the samples (n = 39). The next most

important prey by volume were Ephydridae larvae and the beetle Paracymus aenus,

representing >50% of volume in 6 (15%) and 5 (13%) samples respectively. In winter

pellets, isopods were dominant, representing >90% of the volume in 6 (40%) of the

samples (n = 15). In autumn pellets, polychaetes were dominant, representing >90% of

the volume in 29 (24%) of the samples (n = 121), followed by molluscs that made up

>50% of the volume of 18 (15%) samples. Grit was unusually abundant in autumn

pellets, constituting >50% of the volume of 11 (9%) samples.

There were significant differences between seasons in the abundance of different

food items in pellets as follows (Table 1). Sonchus oleraceus seeds, chironomid larvae

and pupae, Stratiomyidae larvae, Ephydridae pupae, O. corrugatus, P. aenus and

Artemia cysts were more abundant in spring than in autumn. Chironomid larvae and

pupae and Artemia cysts were more abundant in spring than in winter. Isopods,

polychaetes and molluscs were more abundant in autumn than in spring. Artemia cysts,

polychaetes and molluscs were more abundant in autumn than in winter. Isopods were

more abundant in winter than in the other two seasons, and polychaetes were more

abundant in winter than in spring (Table 1).

We also found significant differences between seasons in the numbers of

countable items recorded in pellets: S. oleraceus seeds were more abundant in spring

than in autumn (U = 1633.5, p < 0.05). Artemia cysts were more numerous in spring

than the other two seasons (autumn: U = 1670, p < 0.05; winter: U = 562.5, p < 0.05),

and more numerous in autumn than in winter (U = 105, p < 0.01). There were no

significant differences between seasons in numbers of Arthrocnemum macrostachyum

seeds, Mesembryanthemum nodiflorum seeds or cestode cysticercoids (Flamingolepis

liguloides)

As measured by categories of volumetric abundance, chironomid pupae were

relatively more abundant in spring pellets than chironomid larvae (Sign test, z = 2.04, p

< 0.05), whereas differences in their relative abundance for other seasons (Table 1) were

not significant.

Differences between pellets and faeces in autumn

The diversity of items varied, with more classes of food items being recorded in pellets

(46) than in faeces (38). Suaeda seeds, S. oleraceus seeds and nematodes, were only


25

found in faeces, while Salicornia seeds, bryozoan statoblasts, adult chironomids,

Dolichopodidae, Stratiomyidae, Ephydridae larvae, P. aenus, decapods, cirripeds and

gastropods were only found in pellets.

Polychaetes were the most frequently recorded prey items in both pellets and

faecal samples (Table 1). However, whereas molluscs were the next most important

items for pellets, Artemia cysts and O. notabilis beetles were the next most important

for faeces (Table 1).

Polychaetes were dominant in volumetric terms in both pellets (see above) and

faeces, representing >90% of the volume of 13 (15%) of faecal samples (n = 84).

Unlike pellets, in which molluscs and grit were abundant (see above), the next most

important items in faeces were chironomid larvae, M. nodiflorum seeds and Artemia

cysts (Table 1) which made up >50% of the volume of 8 (10%), 9 (11%) and 2 (2%)

samples respectively.

There were significant differences in the relative abundance of different

components between pellets and faeces. Chironomid larvae were more abundant in

faeces, whereas isopods, polychaetes, molluscs and grit were more abundant in pellets

(Table 1). Amongst countable items, M. nodiflorum seeds and Artemia cysts were more

abundant in faeces (U = 3828, p < 0.01; U = 3824.5, p < 0.01, ), whereas the number of

cestode cysticercoids was higher in pellets (U = 2714, p < 0.01).

The relative abundance of different components in pellets and faeces showed a

highly significant correlation (rs = 0.56, p < 0.01, n = 48) when comparing mean values

of volumetric ranks (see methods). The correlation for percent occurrence of each item

was equally significant (rs = 0.51, p < 0.01). However, each method gave a different

ranking to the abundance of different items (Table 2).

DISCUSSION

The diet of waders can be very variable at different moments of the tidal cycle, and

waders can excrete various pellets differing in composition during the course of one

cycle (Goss-Custard & Jones 1976, Worral 1984). In our study area, it was not possible

to collect samples throughout the tidal cycle, as only roost sites used at high tide were

accessible for sample collection. However, we collected all samples from the same

place and at the same point of the tidal cycle, to make them comparable.


26

Many studies of Redshank diet have been carried out in northern Europe, where

different prey items are available (e.g. Goss-Custard 1970, Goss-Custard & Jones

1976). In southern Europe, the few previous studies have been made in winter. In faeces

from the Tagus Estuary in Portugal, the gastropod Hydrobia ulvae was dominant

(Moreira 1996). In pellets collected from salt pans in Cádiz Bay, the dominant prey

were diptera, Coleoptera and H. ulvae (Perez-Hurtado et al. 1997). In our winter pellet

samples, isopods, Coleoptera and polychaetes were dominant.

The contrast between our study and that in the Tagus may be explained by the

different composition of faeces and pellets (see below). The difference with the Cádiz

Bay study is probably explained by different management of salt pans. In Cádiz, the

depth of many evaporation ponds is reduced in autumn and winter (Masero & Pérez-

Hurtado 2001), allowing waders to feed efficiently on chironomids and other prey in the

ponds. In Odiel 90 km to the north-west, there are no drawdowns in this period, and

almost all the ponds remain too deep in winter for Redshank, forcing them to feed in the

mudflats on isopods and polychaetes.

Based on pellet composition, insect prey from salt pans were most important in

the diet in spring, whereas prey from mudflats (polychaetes, molluscs, isopods) were

most important in autumn and winter, suggesting a switch in habitat use (Table 1). The

relative value of the salt pans and mudflats as foraging habitat changes between seasons,

since fluctuations in available biomass in these two habitats are asynchronous (Masero

et al. 1999, Masero & Pérez-Hurtado 2001). Thus, in mudflats in Cádiz the biomass of

available prey dropped by a third from February to March (Masero et al. 1999), when

the biomass of chironomid prey in the Odiel salt pans increased. Furthermore, in Odiel

the depth in salt pans is reduced in spring, favouring their use by waders (Sánchez et al.

unpublished). Hence it is not surprising that the relative importance of mudflat and salt

pan species in Redshank diet changed between seasons. Our data confirm the

complementary importance of both habitat types for Redshank in the Iberian peninsula

at different times of the year (see Masero et al. 1999).

In benthic samples from the salt pans, we have found the density of chironomid

pupae to be low compared with that of larvae (one pupa per 125-162 larvae in January,

March and September 2001). Thus, the greater relative abundance of C. salinarius

pupae in Redshank pellets suggests the pupae are taken from the surface during

emergence events (Armitage et al. 1995), as supported by observations of feeding


27

behavior. Pupae were most abundant in pellets in spring, when we recorded the highest

density of pupae in the sediments (64 m-2, Sánchez et al. unpublished).

The consumption of seeds by waders is common, but the reasons for this are

poorly understood (Green et al. 2002). In our study, S. oleraceus seeds were consumed

mainly in spring, coinciding with the time of seed production (Valdés et al. 1987).

Artemia cysts were most abundant in autumn and spring, the peak in cyst production

being in summer (Martínez 1989).

As we have shown, the use of different methods to study wader diet can produce

important differences in results and their associated biases. We found pellets to contain

a greater diversity of prey items than faeces, but all items recorded in more than 7.5% of

general samples were also recorded in pellets and vice versa. Thus, both kinds of

samples include all major prey items, contrary to previous suggestions (Goss-Custard et

al. 1977, Worrall 1984). Although most items are present, those found in faeces are

often more fragmented and harder to identify. Moreby (1988) suggested that the value

of faecal analysis is limited by poor detectability following digestion. In our study,

almost all arthropod groups left detectable hard parts in faeces, although their

identification is time-consuming.

These two methods give different, but correlated, indices of the relative

importance of prey, with polychaetes, isopods and mollusca being particularly abundant

in pellets, and chironomids in faeces. These differences appear to be related to

differences in digestibility, with harder items being more represented in pellets. They

also appear related to size, with smaller items being excreted as faeces. Thus, M.

nodiflorum seeds (<1 mm diameter) and Artemia cysts (<0.3 mm) were more frequent in

faeces, despite their hardness. For a given mollusc species, individuals in pellets are

larger than those in faeces (Goss-Custard et al. 1977, Dekinga & Piersma 1993). At

Cádiz Bay, the selection of prey size by Redshank was studied by comparing the size of

prey in faeces with those available (Masero & Pérez-Hurtado 2001). Our results suggest

it is important to compare both faecal and pellet contents in such studies, since they

represent different fractions of the prey sizes consumed. They also suggest that studies

comparing the diet of wader species that use faecal analysis for some species and pellet

analysis for others (e.g. Pérez-Hurtado et al. 1997) may produce misleading results.

In general, pellets give more representation to larger, harder prey items

consumed on mudflats and faeces to softer, smaller items consumed in the salt pans.We

used the average volumetric category for each prey item to rank them in importance for


28

faeces and pellets (Table 2). The sum of these rankings gives a more reliable indication

of the diet composition (Duffy & Jackson 1986). This suggests that polychaetes were

most important in autumn diet, followed by Artemia cysts and O. notabilis (Table 2).

Nevertheless, even this combined method is biased owing to different digestibilities and

detectabilities between prey items. Indeed, to confirm that a combined method is more

representative than either pellets or faeces alone, it would be necessary to undertake

controlled experiments with known ingesta composition. Polychaetes were ranked first

in both sample types because they contained both hard, indigestible mandibles (recorded

mainly in pellets) and fine chaetae (detectable in faeces).

Although pellet and faecal analysis produce different results and combination of

both methods is the best option, the abundances and frequencies of prey items recorded

in the two methods are highly correlated. Thus, either method provides a useful and

related assessment of redshank diet.

ACNOWLEDGEMENTS

The first author was supported by a PhD grant from the Ministerio de Ciencia y

Tecnología. The Consejería de Medio Ambiente, Junta de Andalucía and Aragonesas

Industrias y Energía S.A. provided permission to work in the salt pans. Juan Carlos

Rubio, Director of The Odiel Marshes Natural Park, provided logistical support

and advice. Carmen Elisa Sainz-Cantero helped with invertebrate identifications. Niall

Burton and Will Cresswell provided helpful comments on the manuscript.

REFERENCES

Arcas, J. 2001. Predation of Common Sandpiper Actitis hypoleucos on Orchestia

gammarellus (pallas 1766), (Crustacea: Amphipoda): problems in assessing its diet

from pellet and dropping analysis. Wader Study Group Bull. 94: 31-33.

Armitage, P., Cranston, P.S. & Pinder, L.C.V. (eds). 1995. The Chironomidae: the

biology and ecology of non-biting midges. London: Chapman & Hall.


29

Davis, C. A. & Smith, L. M. 1998. Ecology and management of migrant shorebirds in

the playa lakes region of texas. Wildlife monoraph. 140: 1-45.

Dekinga, A. & Piersma, T. 1993. Reconstructing diet composition on basis of faeces in

a mollusc-eating wader, the Knot Calidris canutus. Bird Study 40: 144-156.

Duffy, D. C. & Jackson, S. 1986. Diet studies of seabirds: a review of methods.

Colonial Waterbirds 9: 1-17.

Goss-Custard, J. D. 1970. The responses of redshank (Tringa totanus (L.)) to spatial

variations in the density of their prey. J. Anim. Ecol. 39: 101-113.

Goss-Custard, J. D.,& Jones, R. E. 1976. The diets of redshank and curlew. Bird Study

23: 233-43.

Goss-Custard, J. D, Jones, R. E. & Newbery, P. E. 1977. The ecology of the Wash. I.

Distribution and diet of wading bierds (Charadrii). J. Appl. Ecol. 14:681-700.

Green, A. J., Figuerola, J. & Sánchez, M. I. 2002. Implications of waterbirds ecology

for the dispersal of aquatic organisms. Acta Oecol. 23: 177-189.

Harris, M. P. & Wanless, S. 1993. The diet of Shags Phalacrocorax aristotelis during

the chick-rearing period assessed by three methods. Bird Study 40: 135-139.

Jenni, L., Reutiman, P. & Jenni-Eiermann, S. 1989. Recognizability of different food

types in faeces and in alimentary flushes of Sylvia warblers. Ibis 132: 445-453.

Jones, R. E. 1975. Food of Turnstones in the Wash. British Birds 68: 339-341.

Kalejta, B. 1993. Diets of shorebirds at the Berg River Estuary, South Africa: spatial

and temporal variation. Ostrich 64: 123-133.

Martínez, A. 1989. Estudio de la biología de Artemia en una salina y de su influencia

sobre la calidad de la sal. Tesis, Univ. Sevilla. 388 p.


30

Masero, J. A., Pérez-González, M., Basadre, M. & Otero-Saavedra, M. 1999.

Foodsupply for waders (Aves: Charadrii) in an estuarine area in the Bay of Cádiz (SW

Iberian Peninsula). Acta Oecol. 20 (4): 429-434.

Masero, J. A. & Pérez-Hurtado, A. 2001. Importance of the supratidal habitats for


adyacent saltworks in southern Europe. The Condor 103: 21-30.

Masero, J. A. 2003. Assessing alternative anthropogenic habitats for conserving

waterbirds: salinas as buffer areas against the impact of natural habitat loss for

shorebirds. Biodiversity and Conservation 12: 1157-1173.

Moreby, S. J. 1988. An aid to the identification of arthropods fragments in the faeces of

gamebird chicks (Galliformes). Ibis 130: 519-526.

Moreira, F. 1996. Diet and feeding behaviour of grey plovers Pluvialis squatarola and

redshank Tringa totanus in a southern european estuary. Ardeola 43 (2): 145-156.

Pérez-Hurtado, A., Goss-Custard, J. D. & García, F. 1997. The diet of wintering waders

in Cádiz, southwest Spain. Bird Study 44: 45-52.

Recher, H. F. 1966. Some aspects of the ecology of migrant shorebirds. Ecology 47:

393-407.

Sánchez, M. I., Green A. J. & Dolz, J. C. 2000. The diets of the White headed duck

Oxyura leucocephala, Ruddy duck O. jamaicensis and their hybrids from Spain. Bird Sty

47 (3): 275-285.

Scheiffarth, G. 2001. The diet of bar-tailed godwits Limosa lapponiaca in the Wadden

Sea: combining visual observation and faeces analyses. Ardea 89 (3): 481-494.

StatSoft. 1999. Statistica 5.5. Tusla, OK: StatSoft Inc.


31

Valdés, B., Talavera, S. & Fernández-Galiano, E. (Eds.) 1987. Flora Vascular de

Andalucía Occidental. Vol I-III. Ed. Ketres. Barcelona.

Velasquez, C. R., Kalejta, B. & Hockey, P. A. R. 1991. Seasonal abundance, habitat

selection and energy consumption of waterbirds at the Berg River Estuary, South Africa.

Ostrich 62: 109-123.

Wetlands International 2002. Waterbird Population Estimates – Third Edition. Wetlands

International Global Sries No. 12, Wageningen, The Netherlands.

Worral, D. H. 1984. Diet of Dunlin Calidris alpina en the Severn Estuary. Bird Study

31: 203-212.

Table 1. Contents of Redshank pellets and faeces, showing the percentage occurrence of each food item (PO) and the percentage of samples in

which each item represented more than 10% of the sample volume (V > 10%).Autumn Autumn (A) Spring (S) Winter (W)faeces, n = 84 pellet, n = 121 pellet, n = 39 pellet, n = 15

PREY Habitat PO V > 10% PO V > 10% PO V > 10% PO V > 10% U H PCSGREEN PLANT MATERIAL 26 7 36 1 51 3 53 ...Angiospermae S 24 7 33 1 51 3 53 ...Algae 5 ... 4 ... ... ... ... ...SEEDS 32 19 12 ... 46 3 17 ...Arthrocnemum macrostachyum M/S 4 1 3 ... 13 ... 13 ... 5.8Salicornia ... ... 2 ... ... ... ... ...Suaeda 1 ... ... ... ... ... ... ...Sonchus oleraceus S 1 ... ... ... 31 3 ... ... 44.6**** S>AMesembryanthemum nodiflorum S 26 18 2 ... 8 ... ... ... 3820**Unidentified seeds 4 1 6 ... ... ... 13 ... 4.2INVERTEBRATES 100 90 100 100 100 100 100 100Bryozoan statoblast ... ... 1 ... ... ... ... ...Chironomus salinarius (L) S 35 18 7 1 59 10 ... ... 3615.5**** 58.6**** S>A,WChironomus salinarius (P) S 33 15 10 5 74 62 7 ... 3889** 77.6**** S>A,WChironomus salinarius ... ... 1 ... ... ... ... ...Dolychopodidae (L) ... ... 2 ... ... ... ... ...Stratiomyidae (L) S ... ... 4 ... 26 ... 7 ... 16.4**** S>AEphydridae (L) ... ... 1 ... 8 5 7 7Ephydridae (P) S 1 1 11 2 28 18 27 20 4766.5 13.8*** S>AUnidentified Diptera (L) 4 1 6 ... 3 ... ... ...Unidentified Diptera (P) 10 1 1 ... ... ... ... ..Ochthebius notabilis (A) S 54 10 26 11 23 ... 33 7 3848.5** 0.70Ochthebius corrugatus (A) S 12 1 16 ... 67 ... 40 ... 4898.5 37.8**** S>AOchthebius (L) 6 ... 1 ... ... ... ... ...Paracymus aenus (A) S ... ... 1 ... 41 21 20 13 48.1**** S>AUnidentified Coleoptera (A) 42 12 33 2 54 3 60 20 4511.5 8.9*Formicidae (A) 6 2 17 3 26 ... 33 7 4553.5 3.10Unidentified Hymenoptera (A) 1 ... 1 ... ... ... 13 ... 13.1***Corixidae (A) 4 ... 1 ... ... ... 13 ... 13.1***Unidentified Insecta (A) 14 1 14 ... 18 3 13 ... 5061.5 0.40

33

Artemia parthenogenetica 17 12 2 1 ... ... ... ... 4348.5A. parthenogenetica cyst S 61 8 38 4 64 ... ... ... 3900** 17.6**** S,A>WFlamingolepis liguloides cysticercoids 23 1 27 2 28 3 ... ... 4844 5.4Isopoda M 4 ... 45 6 3 3 67 60 2985**** 33.2**** W,A>SAnphipoda 1 ... 12 3 ... ... ... ... 4510.5 7.2*Decapoda ... ... 7 ... 3 ... 27 7 9.7**Cirripedia ... ... 1 ... ... ... ... ...Ostracoda M 1 ... 20 ... ... ... ... ... 4134.5* 12.3**Unidentified Crustacea 1 ... 3 ... 10 ... ... ... 4.02Araneida 7 ... 10 ... 21 ... 7 ... 3.5Acarina 2 ... 1 ... ... ... ... ...Polychaeta M 77 39 84 72 8 3 60 20 3466.5**** 71.1**** A,W>SForaminifera M 1 ... 18 ... ... ... ... ... 4218.5* 11.1**Gastropoda ... ... 7 2 8 ... 13 7 0.50Unidentified Mollusca (shells) M 18 4 74 37 15 3 20 ... 1999.5**** 48.9**** A>W,SCestoda 4 1 2 ... ... ... ... ...Nematoda 1 ... ... ... ... ... ... ...Unidentified invertebrate 8 2 7 ... ... ... ... ... 5027.5Invertebrate eggs 11 1 10 ... 13 ... ... ... 5.7FISH 12 4 22 2 3 3 7 ... 4584 9*GRIT M/S 1 ... 56 18 23 18 47 ... 2275.5**** 9.1*OTHERS ... ... 7 ... 3 ... 7 7

L, larvae; P, pupae; A, adults. S, prey from saltpan; M, prey from tidal mudflats. Others, nylon line and other artificial objects. Seasonal

differences for pellets were tested with Kruskal-Wallis tests, H. Differences between faeces and pellets in autumn were tested with Mann-

Whitney U post-hoc tests, U. *P < 0.05, **P < 0.01 without Bonferroni correction, ***P < 0.05 after Bonferroni correction, ****P < 0.01 after

Bonferroni correction. PCS, summary of significant Pair wise Seasonal Comparisons for pellet composition with Mann-Whitney U post-hoc

tests.


34

Table 2. Order of importance (by ranks) of different prey items found in autumn faeces

and pellets, and when combining both methods. The ranks were based on the mean

values of seven volumetric categories (see methods). Combined ranks were based on the

means of the two means for pellets and faeces. L, larvae; P, pupae; A, adults.

pellet faeces pellet+faecesPolychaeta 1 1 1A. parthenogenetica cyst 5 3 2Ochthebius notabilis (A) 4 5 3Unidentified Mollusca (shells) 2 9.5 4Mesembryanthemum nodiflorum 8 4 5Unidentified Coleoptera (A) 7 7 6Flamingolepis liguloides cysticercoids 9 9.5 7Chironomus salinarius (P) 13.5 6 8Fish 10 11 9Chironomus salinarius (L) 21 2 10Isopoda 6 21 11.5Formicidae (A) 11 16 11.5Ochthebius corrugatus (A) 16.5 13 12Grit 3 27 13Invertebrate eggs 19.5 12 14Artemia parthenogenetica 26 8 15Ephydridae (P) 13.5 21 16Araneida 19.5 17 17Arthrocnemum macrostachyum 23 14 18Ostracoda 12 27 19Foraminifera 15 27 20Cestoda 28 15 21Anphipoda 16.5 27 22.5Algae 24.5 19 22.5Ochthebius (L) 33.5 18 24Gastropoda 18 35.5 25Corixidae (A) 33.5 21 26Acarina 33.5 23 27Decapoda 22 35.5 28Stratiomyidae (L) 24.5 35.5 29Salicornia 28 35.5 30.5Dolychopodidae (L) 28 35.5 30.5Suaeda 39 27 35.5Sonchus oleraceus 39 27 35.5Nematoda 39 27 35.5Bryozoan statoblast 33.5 35.5 35.5Chironomus salinarius (A) 33.5 35.5 35.5Ephydridae (L) 33.5 35.5 35.5Paracymus aenus (A) 33.5 35.5 35.5Cirripedia 33.5 35.5 35.5

CHAPTER 2 SPATIAL AND TEMPORAL FLUCTUATIONS IN SHOREBIRDS AND CHIRONOMIDS

35

CHAPTER 2

SPATIAL AND TEMPORAL FLUCTUATIONS IN USE BY SHOREBIRDS AND

IN AVAILABILITY OF CHIRONOMID PREY IN THE ODIEL SALTPANS,

SOUTH-WEST SPAIN

MARTA I. SÁNCHEZ1,2, ANDY J. GREEN1 AND ELOY M. CASTELLANOS2



Universidad de Huelva, Campus de El Carmen, Avda. Fuerzas Armadas s/n 21071 Huelva, Spain

Key words: saltpans, Chironomus salinarius, salinity, shorebirds, foraging habitat,

Odiel marshes

Running head: Spatial and temporal fluctuation in shorebirds and chironomids



36

ABSTRACT

We studied the seasonal variation in abundance and distribution of shorebirds and

chironomid Chironomus salinarius larvae in both traditional and industrial salines in the

Odiel marshes, south-west Spain, in 2001. We selected 12 ponds that were

representative of the different phases of the salt production process. The benthic

chironomids were sampled in each pond every two months, and the birds were counted

weekly. Chironomid larvae were most abundant in spring and autumn, and in the ponds

of lower salinity. The density of larvae averaged 7023 ± 392 m-2 (± s.e) over the six

sampling events. Shorebirds were always more abundant at high tide than at low tide,

and were especially abundant during the spring and autumn migration periods when up

to 20,775 birds were counted. A total of 24 species were recorded, six of which were

present in internationally important numbers. The salines were especially important as

foraging and roosting habitat during migration. The percentage of birds that were

feeding in the ponds was positively correlated with the abundance of chironomid larvae

at accessible depths. The number of feeding birds was also higher in ponds with more

chironomid larvae available. Despite more intensive management, industrial salines

held higher densities of birds and a similar abundance of chironomids when compared

with traditional salines.


37

INTRODUCTION

Many shorebirds or waders (Charadrii) are long distance migrants which migrate

thousands of kilometres between breeding and wintering sites and are heavily dependent

on passage sites along the flyways, where they can rest and refuel (Alexander et al.

1996, Iverson et al. 1996). Natural and artificial coastal wetlands tend to be highly

productive and are a vital habitat for these birds (Velasquez 1992, Masero 1999), which

are very sensitive to habitat change (Alexander et al. 1996). In recent decades, many

coastal wetlands have been destroyed or transformed, resulting in major impacts on

shorebird populations (Goss-Custard et al. 1977a,b, Goss-Custard & Moser 1988).

Artificial wetlands such as salines can provide important foraging habitats for

shorebirds, especially at high tide when intertidal marshes are flooded and inaccessible

(Pérez-Hurtado & Hortas 1991). Salt production via the circulation of sea water through

a system of ponds in salines is an ancient activity in the Mediterranean region and other

warm coastal areas (Britton & Johnson 1987). Aquatic invertebrates in saltpans

represent abundant prey for shorebirds (Velasquez 1992), although there are relatively

few invertebrate taxa owing to the extreme salinities. Amongst these taxa, chironomid

larvae are particularly important (Velasquez 1992, Pérez-Hurtado et al. 1997).

The Odiel marshes in south-west of the Iberian peninsula are one of the most

diverse and productive coastal marsh systems in southern Europe (Castellanos et al.

1998). They are situated on the East Atlantic flyway (Smit & Piersma 1989), and their

importance for waterbirds has led to their protection as a Natural Park, Ramsar site

(Bernués 1998) and Biosphere Reserve. The salines represent 16% of the surface area of

the marshes, and are an important feeding and roosting area for shorebirds. During the

migration periods and in winter, flocks of over 12,000 birds can regularly be observed.

However, there are no previous studies of the shorebird and invertebrate communities

in the salines, or of shorebird habitat use.

In this study, we describe the seasonal variation in the abundance and

distribution of shorebirds in the Odiel salines, as well as in the abundance and

distribution of one of their principle prey species, the chironomid Chironomus

salinarius. We assess the differences between saltpans of different salinities and

between industrial salines (with an intensive salt production process) and traditional

salines in their value for shorebirds and their prey. We test whether or not chironomid

abundance predicts the abundance of foraging shorebirds in space and time. We


38

consider what changes to current management may increase the shorebird carrying

capacity of our study site.

STUDY AREA

The Odiel marshes (37º17'N 06º55'W), found in the combined estuaries of the Tinto and

Odiel rivers, are tidal marshes with a total surface area of 7,185 ha. The salines occupy

1,185 ha, of which c.1,118 is an industrial saline complex and c.56 ha is a traditional

saline (Figure 1).

In both kinds of salines, sea water is introduced via a tidal canal to a complex of

large and deep ponds which act as reservoirs (the primary evaporation ponds PEPs),

where the salinity is relatively low and the diversity of invertebrates relatively high.

From there, the water circulates into a system of shallower ponds (the secondary

evaporation ponds SEPs) with intermediate salinity and where the invertebrate

community is dominated by C. salinarius in the benthos and the brine shrimp Artemia

parthenogenetica in the water column. These two species are the most abundant prey

for waterbirds in the salines. C. salinarius is actually a species complex requiring

further taxonomic study (Armitage et al. 1995). The brine shrimp population is

dominated by the diploid form of A. parthenogenetica (F. Amat pers. comm., Amat et

al. 1995).

From the secondary evaporation ponds, the water passes to the pre-

crystallization ponds (PCPs), a group of shallow and hypersaline ponds where water is

stored until it approaches saturation point. Finally, the water then enters the

crystallization ponds (CPs), where the salt precipitates and is harvested (Figure 1). The

abundance of invertebrates in these last two classes of ponds fluctuates according to the

salt concentration.

METHODS

Invertebrate sampling

Twelve ponds covering the range of salinities were selected and sampled throughout

2001, including nine in the industrial salines and three in the traditional salines (Figure


39

1). The ponds included two PEPs, six SEPs, two PCPs and two CPs. The traditional

saline is more a labyrinth of canals than a complex of ponds, through which the water

flows continuously. We did not sample the traditional PEP, which was too deep for

shorebirds.

The benthos was sampled every two months, selecting four points at random

from each pond within the depth range accessible to shorebirds (0-20 cm, Ntiamoa-

Baidu et al. 1998). At each point, three core samples were taken to a depth of 3 cm with

a 19.6 cm2 corer. The salinity was measured at the same time using a densometer. In the

laboratory, the sample was filtered through 0.5 and 0.1 mm sieves. In order to separate

chironomids from the sediments retained in the sieves, we floated them in saturated salt

solution collected from a CP. The larvae and pupae were collected from the surface and

preserved in 70% ethanol.

Shorebird counts

In each of the study ponds we counted the number of shorebirds of each species that

were feeding and resting one day each week, using a 20-60x telescope. On each day, we

carried out two counts of three hours duration, one centred around high tide and the

other around low tide. The ponds were always counted at the same time (by choosing to

count on the day that high or low tide occurred around 09:00 h) and always following

the same route between ponds.

Calculation of the available surface area

In most of the ponds, only shallow areas of 0-20 cm around the edge and around islands

are available to foraging shorebirds. The accessible surface area varies with fluctuations

in the overall water level, which were monitored by recorded depth at a reference point

in each pond at the time of conducting surveys. The depth profile of each pond was

established by conducting various transects, and the surface area accessible for foraging

at the time of survey was estimated via image analysis (Sigma Scanpro, version 4.0).

Statistical analysis

The abundance of chironomid larvae and pupae and of shorebirds were analyzed using

generalized linear models (GLMs) following GENMOD procedure in SAS (v. 8.2, SAS

Institute 2000). POND and MONTH were included as fixed factors. POND had 12


40

levels for the shorebird model, but only nine for the chironomid models as three ponds

where chironomids were absent in all but one month were removed. MONTH had 12

levels for shorebirds and six for chironomids. For the shorebird model, we also include

TIDE as a fixed factor of two levels (low or high). Owing to overdispersion observed in

the data for larvae and shorebirds, we used a negative binomial error distribution (Bliss

& Fisher, 1953, Kopocinski et al. 1998), log link function and type III tests. Such a

model for pupae did not converge owing to the high proportion of zeros, so we

conducted a non-parametric analysis using ranks in GENMOD with an identity link

(RANK procedure in SAS).

The deviance of each fitted GLM model is analogous to the residual sum of

squares in ordinary linear regression. The reduction in deviance compared to the null

model is used to assess the contribution of the model to the explanation of the variance

in the data set. The significance of the reduction in deviance can be estimated by

comparison with the distribution of the chi-square statistic, with degrees of freedom df

equal to the change in df compared to the null model. Post-hoc differences between two

levels of a factor were tested with the Wald chi-square test for differences between

least-squares means (SAS Institute Inc. 1997).

Spearman’s rank correlations were conducted between the proportion of

shorebirds that were feeding and the number of chironomid larvae available, and

between salinity and the density of larvae.

RESULTS

Abundance of chironomids

Chironomid larvae were present in the sediments all year round but with a marked

pattern in seasonal abundance (Figure 2), with the first and strongest peak in May (mean

of 11,835 ± 1,470 larvae m-2, mean ± s.e., n = 804) and a second peak in November

(9,933 ± 1,063 m-2). The same patterns were observed for large and small larvae

collected from the 0.5 and 0.1 mm sieves, although the larger larvae were relatively

more abundant when the total number of larvae was lower (Figure 2). Larvae were only

recorded in the CPs and the industrial PCP during one month (November for the

traditional CP, March for the other two ponds). In the CPs, this was probably owing to

the compacted sandy nature of the sediments and the effects of salt crystallization, while


41

in the PCP it was owing to incrustations of gypsum salts on the sediment surface which

made it impossible for larvae to enter the sediments and construct their tubes.

The annual average density of chironomid larvae varied between different

ponds, with the highest densities recorded in PEPs and SEPs and the lowest in the CPs

(Table 1). This suggests a gradual decrease in larval density during the evaporation and

crystallization process, as confirmed by a significant negative correlation between

average salinity in each pond and the average larval density (rs = -0.713, P = 0.008, n =

12). There were no consistent differences between industrial and traditional salines in

the abundance of larvae for a given salinity type (Table 1).

GLMs revealed highly significant effects of POND, MONTH and the

POND*MONTH interaction on the abundance of benthic chironomid larvae (Table 2).

Thus, there were strong differences in abundance between different ponds and times of

the year, while seasonal differences varied between ponds (as shown by differences

between ponds in the month when abundance peaked, Table 1). Similar results were

recorded for GLMs analysing the numbers of large (those retained on a 0.5mm sieve)

and small (those retained on a 0.1 mm sieves) larvae (Table 2). In a GLM for benthic

pupae, the POND*MONTH interaction was highly significant (Table 2), indicating that

seasonal patterns in chironomid emergence were not consistent between ponds.

Abundance of shorebirds

A total of 24 shorebird species were recorded in the study area, the most abundant

being dunlins (Calidris alpine (L.)), black-tailed godwits (Limosa limosa (L.)), and

curlew sandpiper (Calidris ferruginea (Pontoppidan)) (Table 3). The highest count was

of 20,775 birds in April. The total numbers of shorebirds showed a strong seasonal

pattern, with peak counts in April and August coinciding with the pre and post-breeding

migration periods (Figure 4). The post-breeding migration was the stronger and more

prolonged, with high counts being recorded from July to September. In contrast, the pre-

breeding migration was only marked during April (Figure 4). The number of birds

recorded in the salines was always higher at high than at low tide (Figure 4). The

proportion of birds that remained at low tide varied between seasons, and was higher

during spring migration than autumn migration (Figure 4).

In a GLM, there were highly significant main effects of POND, MONTH and

TIDE on the number of shorebirds (Table 2). All two way interactions were also highly


42

significant. Thus, we recorded strong spatial, temporal and tidal effects on the

distribution of shorebirds that interacted in a complex way.

The seasonal patterns in abundance varied greatly between shorebird species.

Amongst the more abundant species (Table 3), dunlins (Calidris alpina) and little stints

(Calidris minuta (Leisler)) showed only a strong spring migration. In contrast, black-

tailed godwits (Limosa limosa), curlew sandpipers (Calidris ferruginea), redshanks

(Tringa tetanus (L.)), avocets (Recurvirostra avosetta (L.)), kentish plovers (Charadrius

alexandrinus (L.)) and black-winged stilts (Himantopus himantopus (L.)) showed only a

strong autumn migration. Ringed plovers (Charadrius hiaticula (L.)) and grey plovers

(Pluvialis squatarola (L.)) showed both a strong spring and a strong autumn migration.

Relationship between the abundance of chironomids and of shorebirds

When we analyze the use of the salines by shorebirds throughout the year, we find that

the proportion of birds observed feeding was strongly correlated with the availability of

chironomid larvae (Rs = 0.88, p = 0.033, n = 6). The proportion of birds feeding was

highest in May when the abundance of chironomid larvae was greatest (Figure 5). There

was also a strong positive correlation between the number of larvae available in each

pond and the number of feeding shorebirds (using annual means, Rs = 0.66, p = 0.019, n

= 12).

In the traditional salines, a higher proportion of the surface area of ponds is

available for foraging (25% on average, compared with 10% for industrial salines, Table

1). Nevertheless, the highest densities of shorebirds were recorded in the industrial

salines, where the peaks corresponding to the spring and autumn migrations were very

pronounced (Figure 6). In the traditional salines the highest density was recorded during

the winter period in February (Figure 6). On average, 8.5% of available foraging habitat

was found in the traditional salines (Table 1). Thus, 15 of 16 shorebird species with

average counts of more than 10 individuals made more use of the industrial salines for

foraging than would be expected at random (Table 3).

DISCUSSION

Chironomus salinarius is a chironomid species (complex) that is particularly tolerant of

high salinities (Armitage 1995) and is often recorded as the only benthic invertebrate


43

species. In the nearby Cádiz Bay, this species is thought to have about five generations a

year (Arias & Drake 1995). Such multivoltinism is characteristic of latitudes such as

those of southern Spain, where high temperatures allow high growth rate (Huryn 1990).

The presence of pupae throughout the annual cycle in our samples confirms that this

species must have many generations a year, since the benthic pupal stage lasts a few

days at most (Armitage 1995).

For the salines as a whole, we observed a marked seasonality in the abundance

of larvae of this species, with peaks in May and November. This pattern contrasts with

those observed in marine soft-bottom habitats in temperate regions, where abundance of

invertebrates peaks in winter and spring (Service & Feller 1992). Our results are also

different to those found in a tidal lagoon in Cádiz bay, where the abundance of larvae

peaked from late summer until winter (Drake & Arias 1995). This difference is probably

related to the differences in habitat and management of water levels at each site (see

Drake & Arias 1995). Differences in management of water levels between individual

ponds are also likely to explain the strong POND*MONTH interaction we observed at

our site.

In our study site, the main salinity gradient is spatial between ponds rather than

temporal between months of different temperatures (as observed in other Mediterranean

aquatic systems), and we recorded a negative correlation between the salinity of each

pond and the larval density. Thus, high salinities allow C. salinarius to monopolise

benthic resources, but further increases appear to reduce growth and/or survival rates.

Similar effects of extreme salinities on the density of chironomid larvae have been

recorded in other wetland types (Galat et al. 1988, Hammer et al. 1990). A negative

correlation between larval abundance and salinity was also recorded in Cádiz Bay

(Arias & Drake 1994), but it confounds spatial and temporal variation and is hard to

interpret.

Chironomid larvae are one of the principal food items of shorebirds in the

coastal wetlands of southern Europe (Pérez-Hurtado et al. 1997), unlike northern

Europe where polychaetes, gastropods and bivalves tend to be dominant prey (Goss-

Custard 1977a, Worral 1984, Durell & Kelly 1990, but see Rehfisch 1994). The high

density and availability of Chironomus salinarius (together with that of Artemia) makes

the Odiel salines an important foraging habitat for shorebirds, especially during the

migration periods. During autumn passage, most birds used the salines only at high tide

when the tidal marshes were unavailable, leaving to feed in the tidal marshes at low


44

tide. Similar results have been reported in salines elsewhere (Masero et al. 2000).

However, during spring passage we found most shorebirds to remain in the salines at

low tide, suggesting that the salines provided a relatively better foraging habitat in

spring than in autumn. The strong relationship detected between changes in the density

of chironomid larvae and in the numbers of feeding shorebirds suggests that the birds

make decisions about feeding in salines and which pond to feed in based largely on the

availability of chironomids. Previous studies in other regions have shown that

shorebirds respond to variation in prey density, with a positive correlation between prey

density and bird density (Goss-Custard 1970, Goss-Custard et al. 1977a, Goss-Custard

et al. 1991, Velasquez 1992). The same pattern appears to occur in the Odiel salines.

Further evidence that changes in the density of chironomid prey determine these

changes in the use of salines by foraging shorebirds comes from a diet study of the

redshank (Sánchez et al. in press). Pellets collected in spring 2001 were dominated by

invertebrate prey from the salines (chironomid larvae and pupae, and Coleoptera), while

those collected in autumn were dominated by prey from the tidal marshes (isopods,

bivalves and polychaetes). Of 39 pellets collected in spring, 59 % contained chironomid

larvae, compared to only 6.6 % of 121 pellets from the autumn (χ2 = 48.47, P < 0.001).

Nevertheless, the abundance of chironomid pupae in these pellets (found in 74.4 % of

pellets in spring and 9.9% in autumn, χ2 = 60.93, P < 0.001), despite the relative rarity

of pupae in the benthos suggests that, as well as feeding on chironomids in the

sediments, shorebirds also take pupae as they come to the surface and before the adults

have had time to emerge.

Changes in use of the salines by shorebirds will also depend on fluctuations in the

availability of prey in tidal marshes between seasons. We have no data to assess how prey

abundance in tidal marshes differed between spring and autumn. Between the spring and

autumn migrations, we also observed an important shift in the composition of the shorebird

community, e.g. with relatively more dunlins and little stints in spring and more black-

tailed godwits, avocets and black-winged stilts in autumn. Given the differences in diet and

habitat use by these species, these changes may also have influenced the relative increase

in the use of salines for foraging during the spring migration.

We found that the more traditional manner of salt production did not produce a higher

availability of chironomid prey, and did not provide a preferred habitat for waders. Waders

were found at a higher density in the industrial salines, perhaps because these ponds were

larger than traditional ones (Figure 1) and thus permit more effective vigilance against


45

predators (Cayford 1993), and because the traditional salines suffered more disturbance

from a road.

The extent to which a high production of chironomid larvae is translated into a

good foraging habitat for shorebirds depends largely on appropriate management of

water levels (Velasquez 1992, Rehfisch 1994). Smaller shorebird species are those that

are most limited in the depth range where they can feed, and also those most dependent

on alternative, artificial habitats such as salines since their low body mass and high

metabolic rate requires them to feed practically all day round (Goss-Custard et al.

1977b, Fasola & Canova 1993). Some species such as avocets are less limited by water

depth, as they also feed on Artemia in deeper parts of our study site (by swimming and

taking brine shrimps close to the surface).

The high ecological and conservation value of the Odiel salines is obvious given

the numbers and diversity of birds that it supports. For six different species, our partial

counts sometimes exceeded the 1% threshold for the flyway population used to identify

wetlands of international importance for a given species (Table 3). The salines offer both a

good food supply and disturbance-free areas for resting, two key factors that determine the

habitat use by shorebirds (Goss-Custard 1969). Nevertheless, the quality of the habitat as a

foraging area could be increased by changing management practices to increase the

accessibility of chironomids to shorebirds, and particularly by using drawdowns to

increase access to deeper areas during the migration periods. At the moment, the

proportion of chironomid production that is consumed by birds is relatively low compared

to lagoons managed specifically for birds (Rehfisch 1994). In most of the ponds, the

majority of the benthos is inaccessible to shorebirds throughout the year (Table 1). On the

other hand, in those areas of our study ponds where shorebirds are able to feed, exclosure

experiments show that shorebird predation has a “top-down” effect in regulating the

density of chironomid larvae (authors, unpublished data). This suggests that the foraging

intake of a shorebird feeding at a given moment is likely to be limited by shorebird use of

the site in the previous weeks, and underlines the potential benefits of drawdowns so that

the Odiel salines can provide a more efficient refuelling site for a larger number of

migratory shorebirds.


46

ACKNOWLEDGEMENTS

The first author was supported by a phd grant from the Ministerio de Ciencia y

Tecnología. Consejería de Medio Ambiente, Junta de Andalucía and Aragonesas

Industrias y Energía S.A. provided permission to work in the salines. Juan Carlos

Rubio, Director of the Odiel Marshes Natural Park, provided logistical support

and advice. Claudine de le Court, José Manuel Sayago and Enrique Urbina also

provided helpful advice. Raquel Alejandre and Carlos Roldán helped with field work.

REFERENCES

Alexander, S. A., Hobson, K.A., Gratto-Trevor, C.L. & Diamond, A.W. 1996.

Conventional and isotopic determinations of shorebird diet at an inland stopover: the

importance of invertebrates and Potamogeton pectinatus tubers. Canadian Journal of

Zoology 74: 1057-1068.

Amat, F., Barata, C., Hontoria, F., Navarro, J.C. & Varó, I. 1995. Biogeography of the

genus Artemia (Crustacea, Branchipoda, Anostraca) in Spain. International Journal of

Salt Lake Research 3: 175-190.

Arias, A. M. & Drake, P. 1994. Structure and production of the benthic

macroinvertebrate community in a shallow lagoon in the Bay of Cádiz. Marine Ecology

Progress Series. 115: 151-167.

Armitage, P., Cranston, P.S. & Pinder, L.C.V. eds. 1995. The Chironomidae: the


Bernués, M., ed. 1998. Humedales Españoles inscritos en la Lista del Convenio de


Madrid.

Bliss, C.I. & Fisher, R.A. 1953. Fitting the negative binomial distribution to biological

data, with a note on the efficient fitting. Biometrics 9: 174-200.


47

Britton, R. H. & Johnson, A.R. 1987. An ecological account of a Mediterranean salina:

the Salin de Giraud, Camargue (S. France). Biological Conservation 42: 185-230.

Castellanos, E. M., Heredia, C., Figueroa, M.E. & Davy, A.J. 1998. Tiller dynamics

of Spartina maritima in successional and non-successional mediterranean salt marsh.

Plant Ecology 137: 213-225.

Cayford, J. 1993. Wader disturbance: a theoretical overview. Wader Study Group Bull.

68: 3-5.

Drake, P. & Arias, A.M. 1995. Distribution and production of Chironomus salinarius

(Diptera: Chironomidae) in a shallow coastal lagoon in the Bay of Cádiz. Hydrobiologia

299: 195-206.

Durell, S. E. A. Le V. & Kelly, C.P. 1990. Diets of dunlin Calidris alpina and grey

plover Pluvialis squatarola on the Wash as determined by dropping analysis. Bird

Study 37: 44-47.

Goss-Custard, J. D. 1969. The winter feeding ecology of the Redshank Tringa totanus.

Ibis 111: 338-356.

Goss-Custard, J. D. 1970. The responses of Redshank (Tringa totanus (L.)) to spatial

variations in the density of their prey. Journal of Animal Ecology. 39: 91-113.

Goss-Custard, J. D., Jones, R.E. & Newbery, P.E. 1977 a. The ecology of the Wash. I.

Distribution and diet of wading birds (Charadrii). Journal of Applied Ecology 14: 681-

700.

Goss-Custard, J. D., Jenyon, R.A., Jones, R.E., Newbery, P.E. & Williams, R. Le B.

1977 b. The ecology of the Wash. II. Seasonal variation in the feeding conditions of

wading birds (Charadrii). Journal of Applied Ecology 14: 701-719.


48

Goss-Custard, J. D. & Moser, M.E. 1988. Rates of change in the numbers of Dunlin

Calidris alpina, wintering in british estuaries in relation to the spread of Spartina

Anglica. Journal of Applied Ecology 25: 95-109.

Goss-Custard, J. D., Warwick, R.M., Kirby, R., McGrorty, S., Clarke, R.T., Pearson, B.,

Rispin, W.E., Dit Durell, S. E. A. Le V. & Rose, J.R. 1991. Towards predicting

wading bird densities from predicted prey densities in a post-barrage Severn estuary.

Journal of Applied Ecology 28: 1004-1026.

Fasola, M. & Canova, L. 1993. Diel activity of resident and inmigrant waterbirds at

Lake Turkana, Kenya. Ibis 135: 442-450.

Galat, D. L., Coleman, M. & Robinson, R. 1988. Experimental effects of elevated

salinity on three benthic invertebrates in Pyramid Lake, Nevada. Hydrobiologia 158:

133-144.

Hammer, U. T., Sheard, J.S. & Kranabetter, J. 1990. Distribution and abundance of

littoral benthic fauna in Canadian prairie saline lakes. Hydrobiologia 197: 173-192.

Iverson, G. C., Warnock, S. E., Butler, R.W., Bishop, M.A. & Warnock, N. 1996.

Spring migration of western sandpipers along the Pacific Coast of North America: A

telemetry study. Condor 98: 10-21.

Huryn, A. D. 1990. Growth and voltinism of lotic midge larvae: patterns across an

Appalachian Mountain basin. Limnology and Oceanography 35: 339-351.

Kopocinski B., Lonc, E. & Modrzejewska, M. 1998. Fitting a bivariate negative

binomial model to the distribution of bird lice (Phthraptera, Mallophaga) parasitizing

the pheasant (Phasianus colchicus L.). Acta Parasitologica 43: 81- 85.

Masero, J. A., Pérez-González, M., Basadre, M. & Otero-Saavedra, M. 1999. Food

supply for waders (Aves: Charadrii) in an estuarine area in the Bay of Cádiz (SW

Iberian Peninsula). Acta Oecologica 20: 429-434.


49

Ntiamoa-Baidu, Y., Piersma, T., Wiersma, P., Poot, M., Battley, P. & Gordon, C.

1998. Water depth selection, daily feeding routines and diets of waterbirds in coastal

lagoons in Ghana. Ibis 140: 89-103.

Pérez-Hurtado, A. & Hortas, F. 1991. Information about the habitat use of salines and

fishponds by wintering waders in Cádiz Bay. Wader Study Group Bulletin 66: 48-53.

Pérez-Hurtado, A., Goss-Custard, J.D. & García, F. 1997. The diet of wintering waders

in Cádiz Bay, southwest Spain. Bird Study 44: 45-52.



Ecology 31: 383-401.

Sánchez M. I, Green, A.J. & Castellanos, E.M. 2005. In press. Seasonal variation in

the diet of the Redshank Tringa totanus in the Odiel Marshes, south-west Spain: a

comparison of faecal and pellet analysis. Bird Study.

SAS Institute Inc. (1997) SAS/STAT Software, changes and enhancements through

Release 6.12. Cary, NC.

SAS Institute Inc. (2000) SAS/STAT software, User’s Guide. Cary, NC. Service, S.

K. & Feller, R.J. 1992. Long-term trends of subtidal macrobenthos in North Inlet,

South Carolina. Hydrobiologia 231: 13–40.

Smit, C. J. & Piersma, T. 1989. Numbers, midwinter distribution, and migration of

wader populations using the East Atlantic Flyway. Pages 24-63 in Flyways and reserve

networks for water birds (H. Boyd and J.-Y. Pirot, Ed.). International Waterfowl and

Wetlands Research Bureau, Slimbridge.

Velasquez, C. R. 1992. Managing artificial saltpans as a waterbirds habitat: species'

responses to water level manipulation. Colonial Waterbirds: 15(1): 43-55.


50

Worral, D. H. 1984. Diet of the Dunlin Calidris alpina in the Severn estuary. Bird

Study 31: 203-212.


51

Table 1. Annual average salinity, size and density of chironomid larvae recorded in

individual saltpans in Odiel, 2001. See methods for details of pond type and how

available surface area was calculated. I = industrial, T = traditional.

Pond Pond type ManageSalinity(g/cm3)

Total surfacearea (m2)

Availablesurface (m2)

Chironomiddensity (m-2,mean ± s.e.)

Peak density(month)

1 PEP I 1.03 842200 550 ± 400 3984 ± 1267 8068 (nov)2 PEP I 1.04 790700 5787 ± 1229 20396 ± 5688 44671(may)3 SEP I 1.09 196100 48039 ± 5948 2505 ± 951 5817(mar)4 SEP I 1.09 171900 37165 ± 14641 5846 ± 1814 13461 (mar)5 SEP I 1.07 176300 34215 ± 2168 9193 ± 2806 19873 (nov)6 SEP I 1.07 512200 43552 ± 3170 13234 ± 3144 25223 (nov)7 SEP I 1.07 179100 13434 ± 951 10389 ± 3564 23524(nov)8 PCP I 1.1 164500 77109 ± 969 106 ± 106 637(mar)9 CP I 1.16 90253 51828 ± 40623 64 ± 44 255(mar)10 SEP T 1.04 41938 8943 ± 0 6723 ± 3563 24331(nov)11 PCP T 1.08 28092 6154 ± 0 5761 ± 2619 17028(may)12 CP T 1.09 37579 11428 ± 0 226 ± 226 1359 (nov)


52

Table 2. Summary of GLM models testing the effects of POND, MONTH and TIDE on the

abundance of Chironomidae larvae and pupae and shorebirds in the Odiel salines. For larvae,

totals are modelled as well as the number of larvae retained on 0.5 mm and 0.1 mm sieves. Main

effects shown are those observed when interactions are not included in the model. D =

percentage of additional deviance explained by the final model in comparison to null models.

See methods for more details.

Effect df Chi-Square pTotal larvae pond 8 71.84 < 0.0001

month 5 67.56 < 0.0001N = 216D = 4.92% pond*month 40 99.65 < 0.00010.5 Larvae pond 8 44.6 < 0.0001

month 5 27.25 < 0.0001N = 216D = 4.32% pond*month 40 97.25 < 0.00010.1 Larvae pond 8 85.6 < 0.0001

month 5 101.03 < 0.0001N = 216D = 8.30% pond*month 40 129.48 < 0.0001Pupae pond 8 10.25 0.2477

month 5 6.25 0.2823N = 216D = 35.03% pond*month 40 76.94 0.0004Shorebirds pond 11 671.93 < 0.0001

month 11 213.73 < 0.0001tide 1 84.24 < 0.0001pond*month 121 758.19 < 0.0001pond*tide 11 71.4 < 0.0001

N = 467D = 14.99%

month*tide 11 44.65 < 0.0001

Table 3. Shorebird species recorded in our study site and their abundance. Note figures refer to counts made in the ponds selected for our study

(i.e. only part of the salines). * highest counts exceeded the 1% threshold for the flyway population used to identify wetlands of international

importance for a given species (Delany & Scott 2002).

Common name Latin name Mean count ± s.e. Range of countsMean number offeeding birds ± s.e.

Range of feedingbirds

% feeding intraditional salinas

dunlin* Calidris alpina 1750 ± 453 17-15689 803 ± 262 7-1068 0.3black-tailed godwit* Limosa limosa 1209 ± 254 0-6684 232 ± 56 0-1309 9.5curlew sandpiper Calidris ferruginaea 925 ± 220 0-5567 459 ± 122 0-3214 0.0redshank Tringa totanus 693 ± 91 5-2170 366 ± 50 2-1310 5.3ringed plover* Charadrius hiaticula 332 ± 67 0-1780 97 ± 32 0-1182 0.2avocet* Recurvirostra avosetta 309 ± 43 2-1155 42 ± 9 0-227 1.4grey plover Pluvialis squatarola 253 ± 56 0-1313 30 ± 14 0-480 0.2kentish plover* Charadrius alexandrinus 179 ± 50 0-1561 72 ± 30 0-1180 0.0Little stint Calidris minuta 166 ± 36 0-888 143 ± 29 0-885 1.0black-winged stilt* Himantopus himantopus 148 ± 33 0-817 97 ± 23 0-572 5.3bar-tailed godwit Limosa lapponica 68 ± 21 0-754 16 ± 13 0-507 0.0sanderling Calidris alba 63 ± 14 0-364 47 ± 11 0-332 0.2curlew Numenius arquata 58 ± 22 0-830 0.07 ± 0.05 0-2 0.0spotted redshank Tringa erythropus 43 ± 12 0-351 28 ± 10 0-306 5.0greenshank Tringa nebularia 25 ± 6 0-202 15 ± 4 0-162 6.3turnstone Arenaria interpres 17 ± 5 0-143 10 ± 3 0-101 4.6red knot Calidris canutus 7.32 ± 4.43 0-166 2.97 ± 1.47 0-47 0.0whimbrel Numenius phaeopus 5.5 ± 3.07 0-121 0.07 ± 0.05 0-2 0.0oystercatcher Haematopus ostralegus 3.95 ± 0.79 0-17 0 0 ....ruff Philomachus pugnax 3.1 ± 1.79 0-71 2.2 ± 1.22 0-48 21.6little ringed plover Charadrius dubius 2.07 ± 1.70 0-68 1.65 ± 1.57 0-63 0.0common sandpiper Actitis hypoleucos 0.25 ± 0.08 0-2 0.2 ± 0.07 0-2 75.0marsh sandpiper Tringa stagnatilis 0.15 ± 0.10 0-4 0.02 ± 0.02 0-1 0.0red-necked phalarope Phalaropus lobatus 0.1 ± 0.06 0-2 0.1 ± 0.06 0-2 0.0Total 6253 ± 758 53-20775 2466 ± 368 49-13438 2.2

CHAPTER 2 SPATIAL AND TEMPORAL FLUCTUATION IN SHOREBIRDS AND CHIRONOMIDS

54

Figure 1. The location of the Odiel salines in SW Spain, and of our study ponds in the

traditional and industrial salines.

Spain

1

OdielRiver

0 1000 m

2

3

6

5

4

87

9

111210

PEPsSEPPCPCP

IndustrialTraditional

Road


55

Figure 2. Seasonal variation in the density of chironomid larvae (mean ± s.e., n = 144

for each month), in the Odiel saltpans in 2001.

Ja Ma My Jl Se No

Chi

rono

mid

larv

ae (N

º. m

-2)

0

2000

4000

6000

8000

10000

12000

14000

Larvae (0.5)Larvae (0.1)Total larvae


56

Figure 3. Seasonal variation in the density of chironomid pupae (mean ± s.e., n = 144

for each month), in the Odiel saltpans in 2001.

Ja Ma My Jl Se No

Chi

rono

mid

pup

ae (N

º. m

-2)

0

20

40

60

80

100


57

Figure 4. Total number of shorebirds counted (mean ± s.e., n = 4 counts for each month)

in our 12 study ponds in the Odiel saltpans at high and low tide, 2001.

Ja Fe Ma Ap My Jn Jl Au Se Oc No De

Nº o

f sho

rebi

rds

0

2000

4000

6000

8000

10000

12000

14000

16000High tideLow tide


58

Figure 5. Fluctuations in the percentage of shorebirds that are feeding, compared with

changes in the number of chironomid larvae available. Using the bird census conducted

at that date closest to the date of benthic sampling, we calculated the proportion of birds

present at high tide that were feeding. Using densities of larvae and the surface area of

ponds that were available to shorebirds (i.e.≤ 20 cm), we estimated the number of larvae

available.

ja ma my jl se no

% o

f fee

ding

sho

rebi

rds

0

10

20

30

40

50

Chi

rono

mid

larv

ae

2.0e+7

4.0e+7

6.0e+7

8.0e+7

1.0e+8

1.2e+8

1.4e+8

1.6e+8

1.8e+8

2.0e+8

2.2e+8

2.4e+8

% of feeding shorebirds at high tideChironomid larvae


59

Figure 6. Average monthly density of shorebirds recorded at high tide in industrial and

traditional salines. Densities were recorded based on the total surface area of salines.

Ja Fe Ma Ap My Jn Jl Au Se Oc No De

Sho

rebi

rds

(Nº .

1000

has

-1)

0

20000

40000

60000

80000IndustrialTraditional

60

CHAPTER 3 AQUATIC INVERTEBRATE COMMUNITY IN THE ODIEL SALT PANS

61

CHAPTER 3

SEASONAL AND SPATIAL VARIATION IN THE AQUATIC INVERTEBRATE

COMMUNITY AT THE ODIEL SALT PANS (SW SPAIN) AND THEIR

IMPLICATIONS FOR MIGRATORY WADERS

MARTA I. SÁNCHEZ1,2, ANDY J. GREEN1 AND ELOY M. CASTELLANOS2

1Departamento de Biología Aplicada, Estación Biológica de Doñana, Avenida de María Luisa s/n, Pabellón del

Perú, 41013 Sevilla, Spain2Departamento de Biología Ambiental y Salud Pública, Facultad de Ciencias Experimentales, Universidad de

Huelva. Campus de El Carmen. Avda. Fuerzas Armadas s/n 21071 Huelva, Spain

Key words: Artemia parthenogenetica, Cletocamptus retrogressus, diversity, saltpans,

salinity, shorebirds, species richness, Odiel marshes, wind

Running head: Aquatic invertebrate community in the Odiel salt pans

Archive für Hydrobiologie 166: 199-223.


62

ABSTRACT

We studied the seasonal variation in abundance and distribution of invertebrates in the water

column in both traditional and industrial salines in the Odiel marshes, south-west Spain, in

2001. We selected 12 ponds that were representative of the different stages of salt production.

Every two months, the invertebrates were sampled at four points in each pond within the 0-20

cm depth range used by foraging waders. We identified 41 taxa in our samples, including 30

aquatic and nine terrestrial metazoan invertebrates. Invertebrate species richness and diversity

decreased significantly with pond salinity, whereas an index of biomass (sample volume)

showed a non-significant increase. Overall, Artemia parthenogenetica constituted 67.6% of

the total number of invertebrates sampled, and 95.5% of the biomass. The copepod

Cletocamptus retrogressus was next in importance, representing 31.1% of the total number of

invertebrates sampled, and 0.6% of the biomass. Invertebrate biomass and dominance of

Artemia was highest in September and lowest in November. There was significant spatial and

temporal variation in abundance for all major aquatic taxa, and for a given pond and month

the depth, distance to shoreline and fetch (wind effects) all had an important partial effect on

invertebrate distribution. Ordination methods showed a strong relationship between variation

in community structure and in water chemistry (redox potential, pH and salinity), with a clear

separation between ponds with fish and submerged macrophytes and other ponds. The more

intensively managed industrial salines held lower densities and biomass of invertebrates than

traditional salines, perhaps owing to greater protection from wind or inputs of detritus. The

beetle Ochthebius corrugatus was abundant in the traditional ponds but absent from the

industrial ones. The number of feeding waders using each pond was strongly correlated with

the total available biomass of invertebrates.


63

INTRODUCTION

Salt pans are widespread, artificial hypersaline habitats that are of great importance for

migratory waterbirds owing to the high productivity and predictability in time and space, as

well as their shallow depth (Britton & Johnson 1987, Warnock et al. 2002). In recent decades,

many salt pans have been abandoned or transformed into other uses, leading to the loss of

their waterbird and invertebrate populations. This has been accompanied by a tendency to

convert small saltpans managed with traditional methods to larger, intensive pans where

heavy machinery is used to extract salt (Sadoul et al. 1998, Consejería de Medio Ambiente

2004).

With the exception of the extensive literature on brine shrimps Artemia (Abatzopoulos

et al. 2002), there are few studies of the invertebrate community in salt pans on which

waterbirds depend. Only a handful of studies have described temporal and spatial fluctuations

in the invertebrate community in salt pans or attempted to identify determinants of

invertebrate abundance (Carpelan 1957, Britton & Johnson 1987, Williams 1998). None of

these studies have compared the effects of different management practices found within a

given saltpan complex.

The Odiel saltpans in south-west Spain hold internationally important numbers of

migratory waders (Sánchez et al. in press a). In this study we describe the seasonal changes in

abundance and distribution of different invertebrate taxa in the water column of saltpans of

different salinities during an annual cycle. We analyze the relationship between this variation

in abundance and variation in depth, wind effects and water chemistry. We compare the

abundance of invertebrate taxa between small, traditional saltpans and larger industrial ones.

We consider how fluctuations in time and space in the available invertebrate biomass affects

the abundance of migratory waders. Finally, we consider the implications of our findings for

management practices.

STUDY AREA

The Odiel marshes (37º17'N 06º55'W), located in the combined estuaries of the Tinto and

Odiel rivers, are tidal marshes with a total surface area of 7,185 ha. Salt pans occupy 1,185

ha, of which 1,118 ha are industrial salt pans and 56 ha are traditional salt pans (Figure 1). In

the industrial complex, sea water is pumped along a series of ponds and salinity increases via


64

evaporation until the final crystallizing pans are reached. Sea water first circulates through

eight primary evaporation ponds, followed by 11 secondary evaporation ponds and 12 pre-

crystallization ponds, ending up in 11crystallization ponds where the salt precipitates. Annual

production is 160,000 metric tonnes. A single harvest is extracted between early summer and

late autumn.

The traditional salt pans consist of a circuit of canals following a similar progression,

with approximately 30% of the surface area being dykes, 30% primary evaporation ponds,

30% secondary evaporation ponds and 10% crystallizers. Water circulates via gravity via

sluice gates. From three to five successive harvests are obtained between spring and early

autumn.

In primary evaporation ponds, the submerged macrophytes Ruppia cirrhosa and

Althenia filiformis are present, as well as Chaetomorpha and Enteromorpha algae. As salinity

increases, the biomass of primary producers becomes dominated by microalgae. At the

highest salinities, the chlorophyte Dunaliella salina is dominant.

METHODS


Samples were collected throughout 2001. Nine industrial ponds (I1-I9) were selected,

reflecting the whole salinity range (Fig. 1). Three traditional ponds (T1-T3) were selected

excluding the primary evaporation ponds which were too deep to be attractive to waders.

Every two months (in January, March, May, July, September and November), four points

were sampled per pond, filtering 20 l of water throughout a plankton net of 47 m. The exact

points varied each month and were selected at random from the depth range of 5-20 cm

accessible to waders (Ntiamoa-Baidu et al. 1998). Depth and hence the position of areas

sampled varied seasonally. At each point, we also took three sediment samples to study the

benthos (data presented in Sánchez et al. in press a). In each pond, we measured the salinity

(with a densometer), temperature, pH and redox potential in one point. At all four points we

measured the depth, distance to the nearest shoreline and fetch (upwind distance to shoreline,

according to the prevailing winds).

The samples were stored in 70% alcohol and the total sample volume was measured

by displacement as an estimate of biomass. The percentage of total volume that was made up

by each taxon was estimated visually. Usually a single species made up over 90% of the


65

sample. Taxa were identified to the lowest posible taxonomic level using various keys (Giusti

& Pezzoli 1980, Richoux 1982, Nieser et al. 1994, Arias & Drake 1997, Alonso 1998). Numbers

of each taxon were counted, and their individual volume was estimated based on linear

measurements of average sized individuals and using the volume for a simple geometric

figure. For example, the volume of chironomid larvae was estimated using the formula for a

cylinder. By multiplying this individual volume by the number of each taxon in a sample, we

obtained a second estimate of the total volume of each taxon. We used the average of our two

measures in our analyses.

We include all the invertebrates recorded in our samples in this study, including

terrestrial species owing to their potential value as prey items for waders. The copepod

Cletocamptus retrogressus and Chironomus salinarius larvae were abundant in the water

column, despite the fact that both species are principally benthic. We also included egg stages

such as Artemia cysts, which are important food items for waders (Sánchez et al. in press b).

Shorebird counts

In each of the study ponds we counted the number of shorebirds of each species that were

feeding and resting one day each week, using a 20-60x telescope. On each day, we carried out

a count of three hours duration around high tide when the densities of waders are highest

(Sánchez et al. in press b). The ponds were always counted at about the same time of day and

always following the same route between ponds.

Calculation of the available surface area

In most of the ponds, only shallow areas of 0-20 cm around the edge and around islands are

available to foraging shorebirds. The accessible surface area varies with fluctuations in the

overall water level, which were monitored by recorded depth at a reference point in each pond

at the time of invertebrate sampling. The depth profile of each pond was established by

conducting various transects, and the surface area accessible for foraging at the time of survey

was estimated via image analysis (Sigma Scanpro, version 4.0).


We used generalized linear models (GLMs) following GENMOD procedure in SAS (v. 8.2,

SAS Institute 2000) to analyze the spatial and temporal variation in the abundance and

estimated biomass (volume) of those invertebrate taxa present in more than 10% of samples.

These were Artemia parthenogenetica (diploid strain) and their cysts, Cletocamptus


66

retrogressus, other unidentified harpacticoid copepods, adults of the coleopteran Ochthebius

notabilis, Ochthebius spp. larvae, Chironomus salinarius larvae and unidentified planarians.

Owing to overdispersion observed in the data, we used a negative binomial error distribution

(Bliss & Fisher 1953, Kopocinski et al. 1998), log link function and type III tests to analyze

abundance. To analyze volume we could find no suitable transformation owing to the high

proportion of zeros. We thus only analysed non-zero data (log transformed), with an identity

link and normal error distribution.

Pond and month were included as fixed factors in all models. We also included depth,

distance to shoreline and fetch. While controlling for pond and month, there were significant

partial correlations between fetch and distance (r = 0.13, p = 0.017) and also between depth

and distance (r = 0.23, p < 0.001), but these r values are too low to cause problems of

multicollinearity (Graham 2003).

The deviance of each fitted GLM model is analogous to the residual sum of squares in

ordinary linear regression. The reduction in deviance compared to the null model is used to

assess the contribution of the model to the explanation of the variance in the data set. The

significance of the reduction in deviance for the models for abundance was estimated by

comparison with chi-square distribution, with degrees of freedom df equal to the change in df

compared to the null model. Significance for the models of volume was derived from F tests

(Crawley 1993). Post-hoc least-squares means tests (SAS Institute Inc., 1997) were used to

compare differences between pairs of traditional and industrial ponds with similar salinities

(two secondary evaporation ponds, two precrystallization ponds and two crystallization

ponds).

In order to test the relationship between different measures of abundance of waders

and of their invertebrate prey in different months or ponds, we calculated the Pearson

correlation coefficient (after confirming that both variables had a normal distribution, using a

Kolmogorov-Smirnov test). We related invertebrate abundance and available surface area on

the days of invertebrate sampling to the data from wader counts carried out on the nearest date

(1-2 days before or after sampling). Similarly, we calculated the correlation coefficient

between the salinity of different ponds and overall measures of invertebrate volume, species

richness and diversity (using the Shannon-Weiner index, Krebs 1989). In cases where

variables did not meet assumptions of normality, we present the Spearman non-parametric

correlation coefficient. In order to determine the partial effects of invertebrate volume and the

area accessible to waders in each pond on the number of waders, we used a multiple

regression with log transformed variables.


67

We used multivariate ordination methods and cluster analysis to assess the degree of

association between ponds in relation to their water chemistry, invertebrate community and

the interaction between the two. Principal Components Analysis (PCA) was used to plot

ponds against their chemistry data (log transformed). The affinities between ponds based on

the abundance of different invertebrate taxa were established using MDS (non-metric

multidimensional scaling) together with a cluster analysis using the UPGMA (unweighted

pair group method using arithmetic averages) method. The significance of the ordination in

the MDS) was tested using the Kruskal stress coefficient (Kruskal & Wish 1978).

We used a Canonical Correspondence Analysis (CCA) to explore the relationship

between water chemistry and the invertebrate community. The resulting ordination of the

ponds using this method is directly related to the values of the chemistry data (Ter Braak

1990). Data on the abundance of invertebrates was square root transformed and the Bray-

Curtis similarity index (Sánchez-Moyano et al. 2000) was used to establish the similarity

between ponds based on their invertebrate fauna. These multivariate methods were carried out

using the PRIMER (Plymouth Routines In Multivariate Ecological Analysis) programme

(Clarke & Gorley, 2001) and the PC-ORD programme (McCune & Mefford 1997).

RESULTS

We identified a total of 41 taxa including 30 aquatic metazoan invertebrates, one

foraminiferan, one fish and nine terrestrial invertebrates (Appendix 1). The most diverse

group were the insects with representatives of four orders and nine families (excluding

terrestrial species). The number of aquatic taxa recorded in each pond was negatively

correlated with average salinity (r = -0.78, p = 0.003, n = 12), and varied from four at pond I8

to 27 at pond I1. The Shannon-Weiner diversity index (based on total abundances of each

taxon) showed a significant negative correlation with the salinity of each pond (r = -0.65, p =

0.0217, n = 12) and varied between 0.002 (pond I8) and 1.693 (I2). Similarly, an alternative

diversity index based on the overall volume of each taxon showed a strong negative

correlation with salinity (r = -0.70, p = 0.01) and varied from 0.004 (pond I8) to 1.835 (I1). In

least saline ponds, the fauna included the exotic fish Fundulus heteroclitus and the Copepoda,

Gastropoda, Oligochaeta and Corixidae were abundant. Ponds of intermediate salinity were

dominated by Artemia, C. salinarius larvae and Ochthebius beetles (Fig. 2). In ponds of

highest salinity Artemia were dominant (Fig. 2). Artemia and C. salinarius larvae were the


68

only taxa recorded in all the ponds (Appendix 1). The number of aquatic taxa recorded in a

given month ranged from 19 in January to 26 in March.

Water chemistry varied between ponds (Table 1). Overall, salinity varied from 21 g/l

(pond I1 in November) to 231 g/l (T3 in May). Temperatures varied from 8 oC (November) to

37.5 oC (May), pH varied from 7.09 (May) to 9.49 (January) and redox potential from -8.46

(May) to 175.5 (January).

Analyses of abundance

The average annual abundance (number/l) of Artemia and C. salinarius larvae reached a

maximum in ponds of intermediate salinity (around 70-80 g/l, Fig. 2). The total abundance of

aquatic invertebrates (excluding eggs) peaked in March (767.7 ± 121.2 individuals/l, media ±

se) and reached a minimum in November (9.1 ± 1.3 individuals/l). The relative abundance of

the main taxa varied between months (Table 2). Artemia were numerically dominant in all

months except March, when 80% of individuals were Cletocamptus retrogressus (Table 2).

Overall, Artemia constituted 67.6% of the total number of aquatic invertebrates sampled

(excluding cysts).

There were significant differences between ponds and months in the abundance of all

major taxa (Table 3). Depth had a significant partial effect on C. salinarius larvae, Artemia

cysts and planarians, with abundance increasing at lower depths (Table 3). Distance to

shoreline showed a significant negative correlation with abundance of Artemia cysts,

Ochthebius larvae and O. notabilis adults (Table 3). Abundance of C. salinarius larvae and

Artemia cysts also showed a significant positive correlation with fetch (Table 3). Post-hoc

tests showed that, for all taxa, there were significant differences between at least one of the

three pairs of industrial and traditional ponds (Table 4). In 12 of 13 significant differences,

invertebrates were more abundant in traditional than in industrial ponds, the exception being

for unidentified harpacticoids (Table 4, Fig. 2).

Analyses of volume

There was a marked seasonal pattern in the volume of invertebrates in the water column, with

a maximum in September (0.10 ± 0.04 cm3 of invertebrates / l of water, mean ± s.e., n = 40)

and minimum in November (0.004 ± 0.0009, n = 40). The relative volume of the different

taxa varied between seasons (Table 2). Artemia always constituted over 70% of invertebrate

volume, except in november when corixids and chironomid larvae were particularly important


69

(Table 5). Overall, Artemia constituted 95.5% of the total volume of aquatic invertebrates

(excluding cysts).

Invertebrate volume was highest in high salinity ponds, with a maximum of 0.14 ±

0.01 cm3of invertebrates / l of water (mean ± s.e., n = 24) in I8 where 99.6 % was constituted

by Artemia. The lowest value (0.003 ± 0.0003) was recorded in low salinity I1 where only

0.7% were Artemia. There was a positive but non-significant correlation between average

invertebrate volume per sample and the salinity of each pond (r = 0.43 , n = 12, p = 0.16).

There were significant differences between both ponds and months in the volume of

Artemia, Cletocampus, C. salinarius larvae and planaria (Table 5). With the exception of

Ochthebius notabilis, there were significant differences between ponds for the volume of

other taxa analyzed (Table 5). Depth had a significant partial effect on volume for four taxa,

with a positive correlation for Artemia and a negative one for Cletocamptus, Ochthebius

larvae and C. salinarius larvae (Table 5). The volume of Artemia cysts showed a significant

negative correlation with depth to shoreline and positive one with fetch (Table 5). The volume

of C. salinarius larvae also showed a significant positive correlation with fetch. Post-hoc tests

showed that, for six taxa, there were significant differences between at least one of the three

pairs of industrial and traditional ponds (Table 4). In seven of eight significant differences, the

volume of invertebrates was greater in traditional than in industrial ponds, the exception being

for unidentified harpacticoids (Table 4).

Relationship between abundance of invertebrates and waders

The numbers of waders using the Odiel saltpans underwent a marked seasonal variation (Fig.

3=, with counts for the 12 study ponds at high tide peaking in september (9701 ± 1782, mean

± s.e., n = 4) with a minimum in march (2338 ± 547, n = 4). There was no significant

relationship between the total number of waders and the food abundance (estimated as mean

volume of invertebrates per litre in our samples, rs = 0.37, p = 0.50, n = 6). Likewise there

was no significant relationship with the number of feeding waders (rs = 0.09, p = 0.92) or

when considering the density of waders in the shallow 0-20 cm area available for feeding (rs

= 0.31, p = 0.56, n = 6).

In contrast, there was some evidence that birds distributed themselves between ponds

in relation to differences in the volume of prey present in the water column. There was a

strong relationship between the mean number of feeding waders in a given pond and the total

food available (estimated as the mean volume of invertebrates per sample multiplied by the

total area in the accessible depth range of 0-20 cm, rs = 0.71, p = 0.008, n = 12). However, it


70

is the combination of both differences between ponds in the available surface area and in the

mean food abundance that appears to cause this relationship. On their own, neither the

available surface area (rs = 0.52, p = 0.08, n = 12) nor food abundance (rs = 0.50, p = 0.09)

showed a significant correlation with the number of feeding waders in each pond. Similarly,

when these two variables were considered together as two predictors in a multiple regression

with the number of feeding birds as the dependent variable, each had a positive but non-

significant partial effect.

Ordination analyses

Water chemistry

The first axis of a PCA explained 50.6% of the total variance in chemistry data, and was

significantly positively correlated with salinity (r = 0.66, p < 0.025, n = 12) and negatively

correlated with pH (r = -0.52, p < 0.05, n = 12). The second axis explained 27.9% of the

variance and was positively correlated with temperature (r = 0.69, p < 0.01, n = 12) and redox

potential (r = 0.64, p < 0.025, n = 12). Thus PC1 ordered the ponds mainly in relation to the

salinity gradient and PC2 mainly in relation to temperature and redox potential.

Invertebrate community

The MDS analysis separated the ponds of lowest salinity from the others (Fig. 5). These three

ponds (T1, I1 and I2) had submerged macrophytes and fish and a distinctive invertebrate

fauna (see above). The ordination of ponds in space based on the invertebrate community is

consistent with the results of the PCA (Fig. 4), although in the PCA a gradient is observed

rather than a clear separation as seen in the MDS.

Relationship between water chemistry and the invertebrate community

The relationship between the invertebrate community and environmental variables is

described by the CCA (Fig. 6, Table 6). The first axis explained 43.4% of the variance and

was determined mainly by salinity, but also by pH. Thus the most saline ponds were grouped

on the positive side of the axis and the least saline ones on the negative side. The second axis,

which only explained 1.2% of the variance, was correlated with salinity and redox potential.

Taxa characteristic of more saline ponds were C. salinarius, Artemia and Ochthebius (see also

above). Invertebrates associated with the positive extreme of the second axis, such as the

oligochaetes, gastropods and amphipods dominated the ponds of high redox potential and low


71

pH, whereas the Berosus beetles, corixids and the dipteran Nemotelus were characteristic of

ponds with higher pH and lower redox potential (Fig. 6).

DISCUSSION

Ours is the most detailed study to date of spatio-temporal variation in the invertebrate

community in active saltworks of the Mediterranean region. We have recorded a marked

variation in the structure of the invertebrate community and in the abundance of different taxa

in space and time in the Odiel salt pans. As observed in saltpans in Mediterranean France

(Britton & Johnson 1987), we recorded a drop in species richness and diversity as salinity

increases between ponds. We observed a greater diversity of taxa than that recorded in

abandoned saltworks in which the spatial salinity gradient has been lost (Thiéry & Puente

2002). Most of the invertebrates we recorded in our study are important prey for waders

(Durell & Kelly 1990, Kalejta 1993, Pérez-Hurtado et al. 1997). Even the terrestrial

invertebrates we detected in our samples are consumed by waders in an opportunistic manner

(Sánchez et al. in press b).

We found salinity to be the most important chemical factor determining the structure

of the invertebrate community along a spatial gradient. The strong effect of salinity on

primary producers and invertebrates is well known (Hart et al. 1998, López-González et al.

1998, Tripp & Collazo 2003, Thiéry & Puente 2002). However, it is often unclear whether

the distribution of a given taxon is limited directly by salinity itself, or by its association with

other factors such as pH or the influence of salinity on other taxa (especially predators, prey

species or food plants, Williams 1998). Our ordination analyses suggest that the presence of

fish and submerged plants has a dramatic influence on invertebrate community structure.

Submerged macrophytes were only found in the ponds of lowest salinity, and this may be one

reason why the invertebrate community in these ponds was particularly diverse since these

plants have a profound effect on community structure and provide a substrate for many

macroinvertebrates (Wolfram et al. 1999, Weatherhead & James 2001).

Fish were also limited to the ponds of lowest salinity, and these predators also have

profound effects on invertebrate species composition and size distribution (Jeppesen et al.

1997, Hart et al. 1998, Zimmer et al. 2001, Marklund 2002). In Odiel, the fish together with

other predators (notably corixids and copepods) may be directly responsible for the absence

of Artemia from the ponds of lowest salinity (Williams 1998). The number of predatory taxa


72

was highest in these ponds, whereas the importance of detritivorous taxa such as C. salinarius

and Ochthebius was highest in ponds of intermediate salinity, in which Artemia were

dominant. In ponds of highest salinity, the food web was at its simplest and Artemia were

highly dominant amongst invertebrates. Despite the drop in species diversity and richness

with salinity, total biomass tended to increase. The relative importance of planktonic food

chains compared to benthic ones appears to increase at extreme salinities, as also suggested

by the negative correlation amongst ponds between salinity and densities of benthic

chironomid larvae (Sánchez et al. in press a).

Most of the taxa we recorded peaked in abundance in spring or summer, in contrast to

studies of benthic chironomids at Odiel and nearby Cádiz Bay, which suggest maximum

abundance in winter (Arias & Drake 1994, Sánchez et al. in press a). The seasonal

fluctuations in abundance of chironomid larvae in the water column observed in this study

were not synchronous with those recorded in benthic samples. Numbers and biomass in the

water column peaked in March and were at a minimum in January, whereas benthic density

peaked in May and was at a minimum in September. This may be a consequence of weather

conditions at the time of sampling. Strong winds and associated turbulence in these shallow

ponds are likely to release more larvae from sediments and make them available to birds

searching visually in the shallows. Our results suggest that a combination of sampling in

sediments and in the water column is essential to obtain a good measure of the availability of

chironomid prey to birds. Verkuil et al. (1993) previously suggested that the density of brine

shrimps in hypersaline lagoons in the Ukraine was strongly influenced by wind direction. We

found no evidence for that for Artemia adults, but a strong influence of fetch on the

abundance of their cysts and of chironomid larvae, both of which are important food items for

waders and other waterbirds (Green et al. 2002, Sánchez et al. in press b).

Verkuil et al. (1993) found that the density of Artemia increased with depth and

suggested that longer legged wader species were thus better able to exploit this food resource.

We also found that the biomass of Artemia increased with depth, supporting their suggestion.

In contrast, we found that the abundance and/or biomass of Artemia cysts, chironomid larvae,

Cletocamptus, Ochthebius and planarians is greatest at the shallowest depths over the 0-20 cm

range studied, suggesting that shorter legged birds may be better able to feed on these

resources. However, the chironomid larvae we have recorded in the water column are less

numerous than those remaining in the sediments, and benthic larval abundance increased with

depth in our ponds (authors, unpublished data).


73

The community structure and its relationship with the salinity gradient was generally

similar in the traditional and industrial salt pans. The most striking exception is the absence of

Ochthebius corrugatus in the industrial ponds, and its abundance in traditional ponds.

Invertebrate abundance was generally higher in the traditional ponds, and the reasons for this

are unclear. We could find no evidence for a difference in water chemistry between these two

complexes. The smaller pond size (Fig. 1) and greater relative surface area of dykes in

traditional ponds may be an important factor. The dykes provide greater protection from wind

and, unlike in industrial ponds, are covered in natural vegetation in traditional ponds

providing more detritus and potentially boosting productivity. Traditional ponds are used by

lower densities of feeding waders than industrial ponds (Sánchez et al. in press a). Thus,

reduced predation of invertebrates in traditional ponds may be another factor contributing to

our results. Most waterbirds tend to avoid smaller wetlands surrounded by steep banks

making it hard for them to detect predators from a distance (Green 1998). Although

traditional salt pans are not especially important for waders, they clearly provide a better

habitat for invertebrates as well as plants and some passerines.

Artemia are clearly the most important prey for waders in the water column of salt

pans at Odiel (Sánchez et al. in press b) and elsewhere (Masero & Pérez-Hurtado 2001). We

observed a crash in their abundance during winter months as expected, since many studies

have described the disappearance of adults at cold temperatures and their persistence as

resistant cysts (Amat et al. 1991, Wayne 2001, Thiéry & Puente 2002). In the Camargue,

waterbirds move into areas of lower salinity in winter to look for alternative prey when

Artemia populations in high salinity ponds have collapsed (Britton & Johnson 1987). In

Odiel, waders do not show such a switch to low salinity ponds in winter, although these ponds

do become important in winter for ducks and coot. Alternatively, waders may increase their

dependence on benthic chironomids or on marine prey such as polychaetes and bivalves from

tidal areas at this time (Masero et al. 2000, Sánchez et al. in press b).

We have used volumetric measures as an index of biomass of Artemia and other

invertebrates. However, for a given volume, Artemia are likely to have a lower energetic

value to waders than chironomid larvae and many other prey (Rubega & Inouye 1994,

Ludwing & Naegel 2002). Some wader species (notably Red-necked Phalaropes Phalaropus

lobatus and Broad-billed Sandpiper Limicola falcinellus) appear unable to retain their body

weight when fed exclusively on Artemia (Rubega & Inouye 1994, Verkuil et al. 2003). In

contrast, Artemia have been shown to be profitable prey for Dunlin Calidris alpina and

Curlew Sandpipers C. ferruginea (Verkuil et al. 2003). At Odiel, Black-tailed Godwit Limosa


74

limosa are major predators of Artemia. Of 42 droppings collected in August 2001, in 38

Artemia represented over 90% of sample volume (authors, unpublished data).

We have found no evidence that the seasonal changes in the numbers of waders at the

Odiel salt pans are related to changes in abundance and availability of invertebrates in the

water column. They are more likely to be related to other factors determining the timing of

migratory movements. In contrast, we have found evidence that the use of different ponds by

feeding birds is positively related to both the size of the shallow area of 0-20 cm where prey

are accessible, and to the density of prey in that area. Our failure to find stronger evidence is

probably related to the error in our measurements of abundance of such patchily distributed

invertebrate prey, to the importance of benthic prey considered elsewhere (Sánchez et al. in

press a) and to our collective analysis of different wader species varying in dietary

preferences. Use of different ponds is likely to be related to other factors such as pond shape

and size, levels of disturbance (Cayford 1993) and distance to alternative habitats (e.g. tidal

areas used for foraging at low tide, Masero et al. 2000, Luís et al. 2002). Waders may also

take into account the high osmoregulatory costs of feeding in the most saline ponds (Purdue

& Haines 1977, Wollheim & Lovvorn 1995) when making decisions about pond use. We

would need to study a much larger number of ponds to tease apart the importance of so many

different variables.

ACKNOWLEDGEMENTS

The first author was supported by a phd grant from the Ministerio de Ciencia y Tecnología

and an I3P postgraduate grant from the Consejo Superior de Investigaciones Científicas.

Consejería de Medio Ambiente, Junta de Andalucía and Aragonesas Industrias y Energía S.A.

provided permission to work in the salines. Juan Carlos Rubio, Director of the Odiel Marshes

Natural Park, provided logistical support and advice. José Manuel Guerra helped with

statistical analysis. Claudine de le Court, Jordi Figuerola, José Manuel Sayago and Enrique

Urbina also provided helpful advice. Francisco Amat, Dagmar Frisch and and Carmen Elisa

sainz helped to identify invertebrate taxa. Raquel Alejandre and Carlos Roldán helped with

field work.


75

REFERENCES

Abatzopoulos, T. J., J. A. Beardmore, J. S. Clegg, and P. Sorgeloos. 2002. Artemia: Basic and

Applied Biology. Kluwer Academic.

Arias, A. M. & Drake P. 1994. Structure and production of the benthic macroinvertebrate

community in a shallow lagoon in the Bay of Cádiz. Marine Ecology Progress Series 115:

151-167.

Alonso, M. 1998. Crustacea (Branchiopoda). In Fauna Ibérica, Vol 7. Museo Nacional de

Ciencias Naturales y Consejo Superior de Investigaciones Científicas, Madrid.

Amat, F., Hontoria, F., Navarro, J. C., Gozalbo, A. & Varó, I. 1991. Bioecología de Artemia

(Crustacea, Branchiopoda) en la Laguna de la Mata, Torrevieja, Alicante. Instituto de

Acuicultura de Torre de la Sal (CSIC).

Arias, A. M. & Drake, P. 1997. Fauna acuática de las salinas del Parque Natural Bahía de Cádiz.

EGMASA. Consejería de Madio ambiente. Junta de Andalucía.

Bliss, C.I. & Fisher, R.A. 1953. Fitting the negative binomial distribution to biological data,

with a note on the efficient fitting. Biometrics 9: 174-200.

Britton, R. H. & Johnson A. R. 1987. An ecological account of a Mediterranean salina: the

Salin de Giraud, Camargue (S. France). Biological Conservation 42: 185-230.

Carpelan, L. H. 1957. Hydrobiology of the Alviso SALT ponds. Ecology 38: 382-385.

Cayford, J. 1993. Wader disturbance: a theoretical overview. Wader Study Group Bulletin 68:

3-5.

Clarke, K. R. & Gorley, R. N. 2001. Primer (Plymouth Routines In Multivariate Ecological

Research). Vol 5: User Manual/Tutorial. PRIMER-E Ltd., Plymouth, 91 pp.


76

Crawley M.J. 1993. GLIM for Ecologist, Blackwell Scientific Publications, Cambridge.

Durell, S. E. A. Le V. & Kelly C. P. 1990. Diets of dunlin Calidris alpina and grey plover

Pluvialis squatarola on the Wash as determined by dropping analysis. Bird Study 37: 44-47.

Giusti, F. & Pezzoli, E. 1980. Gasteropodi, 2 (Gastropoda: Prosobranchia: Hydrobioidea,

Pyrguloidea). In Guide per il riconoscimento delle specie animali delle acque interne italiane,

Vol 8. Consiglio Nazionale delle Ricerche, Verona.

Graham, M.H. 2003. Confronting multicollinearity in ecological multiple regression. Ecology

84 (11): 2809-2815.

Green, A. J. 1998. Habitat selection by the Marbled Teal Marmaronetta angustirostris,

Ferruginous Duck Aythya nyroca and other ducks in the Göksu Delta, Turkey in late summer.

Revue d’Ecologie, Terre et Vie 53: 225-243.

Green, A. J., Figuerola, J. & Sánchez, M. I. 2002. Implications of waterbird ecology for the

dispersal of aquatic organisms. Acta Oecologica 23: 177-189.

Hart. C. M., González, M. R. Simpson, E. P. & Hurlbert, S. H. 1998. Salinity and fish effects

on Salton Sea microecosystems: zooplancton and nekton. Hydrobiologia 381: 129-152.

Jeppesen, E., Lauridsen, T., Mitchell, S. & Burns, C. W. 1997. Do planktivorous fish structure

the zooplankton communities in New Zealand lakes?. New Zealand Journal of Marine

Freshwater Research. 31: 163-173.

Kalejta, B. 1993. Diets of shorebirds at the Berg River Estuary, South Africa: spatial and

temporal variation. Ostrich 64: 123-133.

Krebs, C. J. 1989. Ecological methodology. Harper and Row, Publishers. New York. 654 pp.

Kruskal, J.B. & Wish, M. 1978. Multidimensional Scaling. Beverly Hills, CA: Sage

Publications.


77

López-González, P. J., Guerrero F. & Castro M. M. 1998. Seasonal fluctuations in the plankton

community in a hypersaline temporary lake (Honda, southern Spain). Hydrobiologia 6: 353-371.

Luís, A., Goss-Custard, J. D. & Moreira, M. H. 2002. The feeding strategy of the dunlin

(Calidris alpina, L.) in artificial and non-artificial habitats in Ria de Aveiro, Portugal.

Hydrobilogia 475/476: 335-343.

Marklund, O., Sandsten, H., Hansson, L. & Blindow, I. 2002. Effect of waterfowl and fish on

submerged vegetation and macroinvertebrates. Freshwater Biology 47: 2049-2059.

Masero, J. A., Pérez-Hurtado, A., Castro, M. & Arroyo, G.M. 2000. Complementary use of

intertidal mudflats and adjacent salinas by foraging waders. Ardea 88 (2): 177-191.

Masero, J. A. & Pérez-Hurtado, A. 2001. Importance of the supratidal habitats for


adyacent saltworks in southern Europe. The Condor 103: 21-30.

McCune, B. & Mefford, M. J. 1997. PC-ORD. Multivariate Analysis of Ecological Data. Mjm

Software design, Gleneden Beach, 47 pp.

Nieser, N., Baena, M., Martinez-Aviles, J. & Millan, A. 1994. Claves para la identificación de

los heterópteros acuáticos (nepomorpha & gerromorpha) de la Península Ibérica. Asociación

Española de Limnología, Publicación No. 5, Madrid.

Ntiamoa-Baidu, Y., Piersma, T., Wiersma, P., Poot, M., Battley, P. & Gordon, C. 1998. Water

depth selection, daily feeding routines and diets of waterbirds in coastal lagoons in Ghana.

Ibis 140:89-103.

Pérez-Hurtado, A., Goss-Custard, J. D. & García, F. 1997. The diet of wintering waders in

Cádiz Bay, southwest Spain. Bird Study 44: 45-52.

Purdue, J. R. & Haines H. 1977. Salt water tolerance and water turnover in the Snowy Plover.

Auk 94:248-255.


78

Richoux, P. 1982. Coléoptères aquatiques. In Introduction pratique à la systématique des

organismes des eaux continentales francaises, Vol 2. Association Francaise de Limnologie, Paris.

Rubega, M. A. & Inouye, C. 1994. Prey switching in Red necked Phalaropes Phalaropus

lobatus: feeding limitations, the functional response and water managementat Mono Lake,

California, USA. Biologial Conservation 70: 205-210.

Sadoul, N., Walmsley, J. & Charpentier, B. 1998. Salinas and nature conservation.

Conservation of Mediterranean Wetlands Nº 9. Tour du Valat. Arles. France.

Sánchez, M. I., Green, A. J. & Castellanos, E. M. 2005. In press a. Spatial and temporal

fluctuations in use by shorebirds and in availability of chironomid prey in the Odiel saltpans,

south-west Spain. Hydrobiologia.

Sánchez, M. I., Green, A. J. & Castellanos, E. M. 2005. In press b. Seasonal variation in the

diet of the Redshank Tringa totanus in the Odiel Marshes, south-west Spain: a comparison of

faecal and pellet analysis. Bird Study.

Sánchez-Mollano, J. E., Estacio, F. J., García-Adiego, E. M. & García Gómez, J. C. 2000.

The molluscan epifauna of the alga Halopteris scoparia in Southern Spain as a bioindicator of

coastal environmental conditions. Journal of molluscan studies 66: 431-448.

SAS Institute Inc. 1997. SAS/STAT Software, changes and enhancements through

Release 6.12. Cary, NC.

SAS Institute Inc. 2000. SAS/STAT software, User’s Guide. Cary, NC.

Ter Braak, C. J. F. 1990. Inerpreting canonical correlation analysis through biplots of

structure correlations and weights. Psychometrika 55: 519-531.

Thiéry A. & Puente L. 2002. Crustacean assemblage and environmental characteristics of a

man-made solar saltwork in southern France, with emphasis on anostracan (Branchiopoda)

population dynamics. Hydrobiologia 486: 191-200.


79

Tripp, K. J. & Collazo, J. A. 2003. Density and distribution of water boatmen and brine

shrimp at a major shorebird wintering area in Puerto Rico. Wetlands Ecology and

Management 11: 331-341.

Verkuil, Y., Van der Have, T. M., Van der Winden, J. & Chernichko, I. I. 2003. Habitat use

and diet selection of northward migrating waders in the Sivash (Ukraine): the use of Brine

shrimp Artemia salina in a variable saline lagoon complex. Ardea 91 (1): 71-83.

Warnock, N., Page, G. W., Ruhlen, T. D., Nur, N., Takekawa J. Y. & Hanson, J. T. 2002.

Management and conservation of San Francisco Bay salt ponds: effects of pond salinity, area,

tide, and season on Pacific Fly-way waterbirds. Waterbirds 25: 79-92.

Weatherhead, M. A. & James, M. R. 2001. Distribution of macroinvertebrates in relation to

physical and biological variables in the littoral zone of nine New Zealand lakes.


Williams, W.D. 1998. Salinity as a determinant of the structure of biological communities in

salt lakes. Hydrobiologia 381: 191-201.

Wolfram, G., Donabaum, K., Schagerl, M. and Kowark, V. A. 1999. The zoobenthic

community of shallow salt pans in Austria -preliminary results on phenology and the impact

of salinity on benthic invertebrates. Hydrobiologia 408/409: 193-202.

Wollheim, W. M. & Lovvorn, J. R. 1995. Salinity effects on macroinvertebrate assemblages and

waterbird food webs in shallow lakes of the Wyoming High Plains. Hydrobiologia 310: 207-223.

Zimmer, K. D., Hanson, M. A. & Butler, M. G. 2000. Factors influencing invertebrate

communities in prairie wetlands: a multivariate approach. Canadian Journal of Fisheries Aquatic

Sciences 57: 76-85.


80

Table 1. Physico-chemical measurements from different ponds in the Odiel salt pans, showing

media ± se and the range (in parentheses). N = 6 for each pond with data taken from one point

every two months from January to November 2001.

Pond Salinity Temperature pH Redox potentialI1 25.12 ± 0.62 (21.19 - 29.28) 21.25 ± 1.56 (13 - 31) 8.41 ± 0.03 (8.20 - 8.64) 82.45 ± 2.46 (70.90 - 103.30)T1 44.10 ± 3.14 (27.00 - 67.54) 19.48 ± 1.13 (12.8 - 25) 8.61 ± 0.06 (8.27 - 9.01) 70.96 ± 3.47 (50.30 - 103.30)I2 44.64 ±1.23 (35.28 - 54.08) 19.17 ± 1.38 (11.5 - 29.5) 8.79 ± 0.04 (8.50 - 9.19) 61.82 ± 2.49 (53.10 - 84.30)I3 66.16 ± 2.82 (46.41 - 83.57) 20.5 ± 1.23 (15 - 31.5) 8.22 ± 0.03 (8.01 - 8.52) 79.70 ± 4.09 (51.60 - 113.50)I4 66.86 ± 2.76 (43.45 - 86.17) 22.67 ± 1.00 (17 - 31) 8.14 ± 0.10 (7.09 - 8.63) 85.83 ± 6.78 (47.10 -141.06)I5 71.22 ± 2.62 (54.82 - 86.92) 23.58 ± 1.25 (15 - 30) 8.26 ± 0.05 (8.01 - 8.66) 58.16 ± 2.80 (34.40 - 73.40)T2 80.46 ± 6.45 (47.06 - 113.50) 23.57 ± 2.24 (10 - 34.4) 8.38 ± 0.07 (7.97 - 8.94) 51.12 ± 6.00 (2.85 - 73.50)I6 86.01 ± 4.65 (58.65 - 118.93) 22.5 ± 1.28 (13 - 29.5) 8.26 ± 0.09 (7.69 - 8.78) 63.56 ± 4.46 (23.43 - 89.60)I7 90.27 ± 5.28 (58.88 - 125.26) 23.75 ± 1.87 (12 - 36) 8.35 ± 0.08 (7.80 - 8.89) 44.01 ± 2.06 (31.30 - 60.80)I8 103.79 ± 8.30 (44.18 - 148.19) 19.42 ± 0.94 (12 - 25.5) 8.57 ± 0.11 (8.02 - 9.49) 47.36 ± 3.15 (29.50 - 68.60)I9 117.72 ± 19.43 (37.43 - 221.28) 21.12 ± 1.46 (15.5 - 30) 8.00 ± 0.07 (7.50 - 8.21) 65.59 ± 17.96 (-8.46 - 175.5)T3 125.77 ± 22.86 (56.21 - 231.23) 20.17 ± 3.79 (8 - 37.5) 7.87 ± 0.12 (7.35 - 8.27) 38.13 ± 7.06 (17.70 - 70.90)

Table 2. Relative abundance (% of individuals or of sample volume) of the major invertebrate taxa recorded in the Odiel saltpans in 2001. L =

larvae. Cletocamptus = Cletocamptus retrogressus. Harpacticoida = unidentified harpacticoids (excluding Cletocamptus). Ochthebius L = larvae

of both O. notabilis and O. corrugatus. Total volume is given in ml.

Artemia Cysts Cletocamptus Harpacticoida O. notabilis Ochthebius L Chironomus L Planaria S. stagnalis Others Total ___________ ____________ ____________ ___________ ___________ ____________ ___________ ___________ ___________ __________ ____________month % ind % S vol % ind % S vol % ind % S vol % ind % S vol % ind % S vol % ind % S vol % ind % S vol % ind % S vol % ind % S vol % ind % S vol sum ind sum volJanuary 52.6 72.55 41.1 7.24 2.91 1.5 0 0 0.03 0.41 0.04 0.15 0.02 1.4 0 0 0 0 3.35 16.74 111333 5.16March 14.9 83.57 3.73 0.56 80.5 7.04 0 0 0 0.04 0.01 0.05 0.07 5.31 0 0 0 0 0.77 3.43 233502 12.73May 76.5 94.34 20.2 0.55 1.23 0.12 0.03 0 0.04 0.1 0.09 0.05 0.12 1.56 0.06 0.02 0 0.06 1.75 3.22 161210 48.75July 74.5 96.92 23.2 0.38 0.51 0.03 0.28 0.01 0.18 0.27 0.13 0.05 0.1 0.79 0.35 0.07 0.07 1.24 0.72 0.25 131550 65.4September 63.2 97.78 33 0.6 3.06 0.2 0.1 0 0.12 0.2 0.05 0.02 0.07 0.59 0.04 0.01 0.01 0.3 0.37 0.29 183737 82.06November 29 49.9 57.6 2.22 8.11 1.1 0.57 0.05 0.22 0.79 0.23 0.19 0.73 13.83 0.04 0.02 0.59 25.14 2.97 6.75 20625 4.37


82

Table 3. Results of generalized linear models analyzing effects of pond, month, depth,

distance to shoreline and fetch on the abundance (counts) of those aquatic invertebrates

recorded in more than 10% of samples. A negative binomial error distribution and log

link were used. D = percentage of deviance explained by the model in comparision with

the null model. k = aggregation parameter of the negative binomial distribution, where

the variance of the dependent variable y = µ+k*µ2, and µ is the mean of y. For the

estimates for each pond and month and other details of the models, see Appendix 2.

Effect DF Estimate Chi-Square PArtemia parthenogenetica Pond 11 218.58 < 0.0001N = 264 Month 5 81.78 < 0.0001D = 8.01 Depth 1 0.01 0.35 0.5514K = 1.27 Shore distance 1 -0.002 0.2 0.6575 Fetch 1 -0.0001 0.03 0.8527Artemia Cysts Pond 11 195.31 < 0.0001N = 264 Month 5 22.66 0.0004D = 9.81 Depth 1 -0.06 6.77 0.0093K = 1.30 Shore distance 1 -0.01 7.58 0.0059

Fetch 1 0.002 5.55 0.0184Cletocamptus retrogressus Pond 11 213.16 < 0.0001N = 264 Month 5 68.17 < 0.0001D = 10 Depth 1 -0.04 2.52 0.1124K = 1.03 Shore distance 1 -0.01 1.79 0.1814 Fetch 1 -0.0004 0.13 0.7149Other Harpacticoida Pond 11 64.24 < 0.0001N = 264 Month 5 46.56 < 0.0001D = 9.50 Depth 1 -0.1 3.12 0.0772K = 0.33 Shore distance 1 -0.01 0.08 0.7788

Fetch 1 -0.001 0.31 0.5763Ochthebius notabilis Pond 11 84.21 < 0.0001N = 264 Month 5 16.76 0.005D = 3.05 Depth 1 0.04 1.24 0.2661K = 0.51 Shore distance 1 -0.04 9.61 0.0019 Fetch 1 -0.003 3.79 0.0515Ochthebius spp. larva Pond 11 105.51 < 0.0001N = 264 Month 5 34.03 < 0.0001D = 2.50 Depth 1 -0.06 3.55 0.0637K = 0.63 Shore distance 1 -0.02 6.06 0.0138

Fetch 1 -0.001 0.27 0.6022Chironomus salinarius larva Pond 11 135.07 < 0.0001N = 264 Month 5 54.24 < 0.0001D = 0.04 Depth 1 -0.04 5.58 0.0182K = 0.97 Shore distance 1 -0.01 3.64 0.0565 Fetch 1 0.002 8.44 0.0037Planaria Pond 11 97.24 < 0.0001N = 264 Month 5 92.85 < 0.0001D = 5.06 Depth 1 -0.07 4.49 0.034K = 0.53 Shore distance 1 0.0001 0 0.9911 Fetch 1 0.0003 0.1 0.7548


83

Table 4. Summary of post-hoc tests of the differences between industrial (I2, I6, I9) and

traditional (T1, T2, T3) salt pans of comparable salinities, as calculated from

generalized linear models analyzing effects of pond, month, depth, distance to shoreline

and fetch. The estimate for the traditional pond was subtracted from that for the

industrial pond, and (-) indicates that numbers of biomass were greater in the traditional

pond. * p < 0.05, ** p < 0.01, *** p < 0.001

Abundance (counts) Biomass (volume) I2-T1 I6-T2 I9-T3 I2-T1 I6-T2 I9-T3Artemia parthenogenetica (-) * (-) (-) *** (-) (-) (-) **Artemia cysts (-) *** (+) (-) *** (-) *** (-) (-) **Cletocamptus retrogressus (-) *** (-) *** (-) *** (-) *** (-) *** (-)Other harpacticoida (+) ** (-) (-) (+) * ... ...Ochthebius notabilis (-) (-) ** (-) … (-) ...Ochthebius spp. larva (-) *** (-) (-) (-) (-) ...Chironomus salinarius larva (+) (-) ** (-) (-) (-) ** (-)Planaria (-) *** (-) ** (+) (-) ** (-) ...


84

Table 5. Results of generalized linear models analyzing effects of pond, month, depth, distance

to shoreline and fetch on the volume (ml/l biomass index) of aquatic invertebrates recorded in

more than 10% of samples. A normal error distribution and identity link were used. Zero values

were excluded from these analyses, owing to problems of model convergence. D = percentage

of deviance explained by the model in comparision with the null model. DFn = DF for

numerator, DFd = DF for denominator. For the estimates for each pond and month and other

details of the models, see Appendix 3.

Effect DFn DFd Estimate F PArtemia parthenogenetica Pond 11 196 15.54 < 0.0001N = 216 Month 5 196 32.55 < 0.0001D = 64.05 Depth 1 196 0.03 8.5 0.0036

Shore distance 1 196 -0.002 1.46 0.2262 Fetch 1 196 -0.0002 0.34 0.5606Artemia Cysts Pond 11 223 14.83 < 0.0001N = 243 Month 5 223 1.98 0.0776D = 47.19 Depth 1 223 -0.01 0.66 0.4169

Shore distance 1 223 -0.004 4.11 0.0425Fetch 1 223 0.001 3.92 0.0479

Cletocamptus retrogressus Pond 10 139 11.73 < 0.0001N = 158 Month 5 139 6.48 < 0.0001D = 55.75 Depth 1 139 -0.03 6.23 0.0126

Shore distance 1 139 -0.0001 1.76 0.1846 Fetch 1 139 0.6308 0.07 0.7887Other Harpacticoida Pond 7 27 2.88 0.0052N = 41 Month 3 27 0.65 0.5831D = 51.48 Depth 1 27 -0.003 0.1 0.7544

Shore distance 1 27 -0.001 0.03 0.8601Fetch 1 27 0.4751 1.17 0.2785

Ochthebius notabilis Pond 8 43 1.1 0.358N = 60 Month 5 43 1.64 0.1464D = 48.37 Depth 1 43 0.01 0.15 0.6966

Shore distance 1 43 -0.01 0.36 0.5486 Fetch 1 43 -0.0003 0.09 0.7648Ochthebius spp. larva Pond 10 57 2.45 0.0063N = 76 Month 5 57 1.29 0.2653D = 58.10 Depth 1 57 -0.03 11.62 0.0007

Shore distance 1 57 0.005 2.37 0.1238Fetch 1 57 -0.0002 0.33 0.5674

Chironomus salinarius larva Pond 11 111 5.62 < 0.0001N = 131 Month 5 111 3.02 0.0099D = 44.94 Depth 1 111 -0.02 10.51 0.0012

Shore distance 1 111 -0.001 1.01 0.3147 Fetch 1 111 0.001 11.01 0.0009Planaria Pond 8 49 2.62 0.0072N = 65 Month 4 49 4.34 0.0016D = 45.07 Depth 1 49 -0.02 2.41 0.1203

Shore distance 1 49 -0.001 0.15 0.6976 Fetch 1 49 0.0004 0.76 0.3845

Table 6. Results of a Canonical Correspondence Analysis (CCA) between measures of

invertebrate abundance and water chemistry. See text for details.


85

* p < 0.05, ** p < 0.01, *** p < 0.001

Axis 1 Axis 2

Species-environment correlation 0.95 0.18

Percentage of species variance 43.4 1.2

Correlation with environmental variables

Salinity 0.82*** -0.52*

Temperature 0.5 0.41

pH -0.75** -0.28

Redox -0.28 0.53*


86

Figure 1. Location of the Odiel salt pans and of the ponds included in the present study

(industrial ponds I1-I9 and traditional ponds T1-T3) . Arrows indicate the circulation

routes for water between input from the sea and salt precipitation.

Spain

I1

OdielRiver

0 1000 m

I2

I7

I4

I5

I6

I9

T3T1IndustrialTraditional

Road

Water circulation

I3T2I8


87

Figure 2. Abundance of the major invertebrate taxa (numbers per litre, mean ± s.e.) in

different ponds in relation to their salinity. Mean salinity in each pond increases from

left to right.

Cletocamptus retrogressus

0

300

400

500

600

700

0

20

40

60

80

100

120

140

160

Other Harpacticoida

Artemiaparthenogenetica

0

100

200

300

400

0.0

0.4

0.6

0.8

1.0

1.2

1.4 Ochthebiusnotabilis

0.0

0.2

0.4

0.6

0.8

Ochthebius spp. larvae

0.0

0.1

0.2

0.3

0.4

0.5

Chironomus salinaris larvae

0.0

0.2

0.4

0.6

0.8

Planaria

I1 T1 I2 I3 I4 I5 T2 I6 I7 I8 I9 T30.0

0.20.8

1.0

1.2

1.4Total invertebrates

I1 T1 I2 I3 I4 I5 T2 I6 I7 I8 I9 T30

100

200

300

400

Industrial pondTraditional pondSalinity

Inve

rtebr

ate

abun

danc

e (li

tres

-1, m

ean

± s.

e.)

Salin

ity (g

/l)

0.15

15


88

Figure 3. Seasonal changes in the abundance of waders in our 12 study ponds, in the

mean volume of invertebrates available in the same area (cm3 of invertebrates per l of

water), and in mean salinity. Wader densities were calculated from the high tide count

made closest to the date of invertebrate sampling.

months

Ja Ma My Jl Se No

avai

labl

e in

verte

brat

e vo

lum

e (c

m3/

l)

0.00

0.02

0.04

0.06

0.08

0.10

0.12

Nº o

f sho

rebi

rds

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

salin

ity (g

/l)

40

50

60

70

80

90

100

110invertebratesshorebirdssalinity


89

Figure 4. Principal Components Analysis of the physico-chemical data from each pond.

Solid circles represent ponds without fish or macrophytes; open squares represent ponds

with fish and submerged macrophytes.

-2,0 -1,0 0 1,0 2,0

Axis 1

-2,0

-1,0

0

1,0

2,0

Axis 2I9

T3

I1

I4

I5I3 I6

I7

I2

T1

I8

T2


90

Figure 5. Cluster analysis (UPGMA method) and MDS ordination of the ponds

according to the abundance of different invertebrate taxa. Solid circles represent ponds

without fish or macrophytes; open squares represent ponds with fish and submerged

macrophytes.

100

80

60

40

20

Similarity (%)

I1 I2 T1 I9 T3 I5 T2 I3 I8 I4 I6 I7

I9

T3

I1

I4

I5I3 I6

I7I2

T1

I8

T2

Stress: 0,01


91

Figure 6. Graphical representation of the ponds, invertebrate taxa and environmental

variables with respect to the first two axes of a Canonical Correspondence Analysis

(CCA). Solid circles represent ponds without fish or macrophytes; open squares

represent ponds with fish and submerged macrophytes. Taxa associated with the upper

part of axis 2 are those associated with high redox potential and low pH (see text). L =

larvae, P = Pupae.

I9

T

I1

I4

I5I3

I6I7I2

T1

I8

T2

Artemia parthenogenetica

Cletocamptus retrogressus

Calanoida

Harpacticoida

Ochthebius notabilisOchthebius corrugatus

Chironomus salinarius L

Chironomus salinarius P

Halocladius spp. L

Collembola

Dolychopodidae L

Sirphydae L

Decapoda

Ostracoda

Amphipoda

Fish

NematodaHydrobia ulvae

Hydrobia ulvae L

Paracymus aenus

Ephydra spp. L

Ephydra spp. P

PlanariaDytiscidae L

Bryozoa

Sigara stagnalis

Foraminiphera

Hydrophyllus spp. L

Berosus spinosus

Oligochaeta

Nemotelus spp. L salinity

temperature

ph

redox

Axis 1

Axis 2

Ochthebius spp. L

92

CHAPTER 4 EFFECT OF SHOREBIRD PREDATION ON CHIRONOMIDS

93

CHAPTER 4

SHOREBIRD PREDATION AFFECTS ABUNDANCE AND SIZE

DISTRIBUTION OF BENTHIC CHIRONOMIDS IN SALTPANS: AN

EXCLOSURE EXPERIMENT

MARTA I. SÁNCHEZ1, ANDY J. GREEN1 AND RAQUEL ALEJANDRE


Pabellón del Perú, 41013 Sevilla, Spain

Key words: shorebirds, predation, exclosures, Chironomus salinarius, size distribution,

saltpans, Odiel marshes, top down regulation

Running head: Effect of shorebird predation on Chironomids

Journal of North American Benthological Society 25(1): 9-18.


94

ABSTRACT

Chironomus salinarius larvae are the dominant benthic organisms in the saltpans of the

Odiel marshes, south-west Spain, an area internationally important for migratory

shorebirds. Using predator exclosures, we examined the effects of shorebirds on

chironomid larvae during spring migration. During the first experimental period from 12

February to 26 March 2002, wader densities were relatively low, peaking at 4 feeding

birds ha-1. During the second period from 19 March to 3 May, densities were higher,

peaking at 32 feeding birds ha-1. Owing to predation by shorebirds, the change in larval

density and biomass from the beginning to the end of each period was significantly

higher in exclosures than in controls. Predation decreased larval density by 35 % during

period 1 and 32 % during period 2, and decreased biomass by 37 % during period 1 and

49 % during period 2. Predation also changed prey size distribution, reducing the

proportion of larger larvae. These effects did not vary significantly between periods,

suggesting that increased predation in period 2 was compensated by higher larval

growth or recruitment. Variation in exclosure effects was not related to variation in the

initial larval density or biomass in a given patch. Predator-prey interactions are complex

in saltpans, with evidence of simultaneous top-down and bottom-up regulation.


95

INTRODUCTION

It is generally recognised that predation plays a fundamental role in the structuring of

benthic invertebrate communities (Thrush 1999). Chironomidae larvae are extremely

abundant or dominant in the benthos in a variety of aquatic ecosystems, and their

abundance and size distribution is often regulated by insect or fish predators (Armitage

et al. 1995, Kornijów 1997, Batzer 1998). Benthic midge larvae are also major prey for

a variety of waterbirds (Krapu & Reinecke 1992, Rehfisch 1994), yet there is very little

information on the influence that avian predators have on chironomid communities

(Armitage et al. 1995). Only one study has shown an influence of avian predation on

larval abundance (Székely & Bamberger 1992), and no study has shown an influence on

larval size.

Hypersaline systems such as saltpans have relatively simple food webs (Britton

& Johnson 1987), making them ideal for studies of predator-prey interactions. More

than 20,000 shorebirds use saltpans within the Odiel Marshes in south-west Spain

during migration. Chironomid larvae dominate the benthos and are a major prey of

shorebirds (Sánchez et al. in press a, in press b). Shorebirds are episodic predators likely

to have major effects on invertebrate prey owing to their high foraging intake and

energy demand and their formation of large concentrations at stopover and wintering

sites (Wilson 1991, Masero & Pérez-Hurtado 2001, Kvist & Lindström 2003),

especially during migration when they may double their body mass in 20 days (Hicklin

& Smith 1984). They have been shown to reduce benthic prey abundance (Weber &

Haig 1997, Sutherland et al. 2000) and influence prey size (Botto et al. 1998) in

intertidal ecosystems. However, as far as we are aware, their effect on prey abundance

or size in saltpans has not previously been studied.

In this study we investigate the impact of predation by shorebirds on the

population of benthic chironomids in the Odiel saltpans. Using exclosure experiments,

we compare the effects of shorebirds on the density, biomass and size of larvae at two

different bird densities at the beginning and peak of spring migration. We test the

hypotheses that exclosure effects are more pronounced at the time of higher bird

density, and in patches with higher prey density.


96

STUDY SITE

The study was carried out in spring 2002 within the 1,185 ha of saltpans in the Odiel

Marshes (37º17'N 06º55'W) at the mouths of the rivers Odiel and Tinto. The Odiel

Marshes in south-west Spain are protected as a Natural Park by the regional government

and as a UNESCO Biosphere Reserve, wetland of international importance under the

Ramsar Convention and Specially Protected Area of the European Union (Bernués

1998). The saltpans support internationally important numbers of Black-winged stilt

Himantopus himantopus, Kentish plover Charadrius alexandrinus, Avocet

Recurvirostra avosetta, Ringed plover Charadrius hiaticula, Black-tailed godwit

Limosa limosa and Dunlin Calidris alpina (Sánchez et al., in press a).

The saltpans include a gradient of salinity in relation to salt production, and this

study was conducted in the secondary evaporation ponds that provide good access and

high abundance of chironomids to shorebirds (see Sánchez et al. in press a for details).

Chironomus salinarius Kieffer was the only chironomid species recorded in the

benthos. This is a multivoltine species estimated to have an average of five generations

a year in another coastal system 100 km away in south-west Spain (Drake & Arias

1995). C. salinarius is a widespread species complex requiring further taxonomic study

(Armitage et al. 1995). There were no fish, crabs or shrimps, and shorebirds were the

only abundant predators in the area studied. Flamingos and other waterbirds did not use

the areas selected for exclosures during the experiment. None of the other invertebrates

recorded in the wetlands are known to predate on C. salinarius larvae (authors,

unpublished data).

METHODS

Exclosure experiments were conducted over two different periods. The first (period 1)

coincided with the beginning of spring migration (Fig. 1) when exclosures were in place

from 32 to 38 days between 12/02/02 and 26/03/02. Period 2 included the peak of

migration (Fig. 1) when exclosures were in place from 37 to 46 days between 19/03/02

and 03/05/02. These dates were selected on the basis of regular censuses of shorebirds

during the same months in 2001.


97

Exclosures

In each period, we installed 20 exclosures paired together with 20 controls. Iron poles of

5 mm diameter were used to mark out squares of 1 x 1 m for both exclosures and

controls. Exclosures included 15 mm PVC mesh netting spread over across a PVC

square frame placed on top of polystyrene floats. These floats were threaded onto the

iron poles so that the net could move up or down but was always suspended about 5 cm

above the water level. Thus, exclosures were not affected by changes in water level

associated with rainfall and management for salt production. Neither did the net touch

the water surface or affect water movement. We deliberately did not place netting

around the sides of the square to avoid influencing water velocity or sedimentation

(Sutherland et al. 2000). There was no algal growth and shading effects were close to

zero.

Five pairs of controls and exclosures were established in each of four ponds,

placing them always in the 0-20 cm water depth range used by feeding shorebirds. Pairs

were separated from each other by 20-30 m. Within each pair, the control and exclosure

were placed at the same water depth, separated by 5 m. The two structures varied

slightly in their relative position to the pond edge, but the decision as to which went

where was taken at random.


At the beginning and end of each period, three samples of benthos were taken at random

from within each exclosure and control, using a corer of 5 cm diameter to sample the

top 3 cm of soft sediments overlying an impenetrable clay bed. Sampling at the

beginning and end of each period was carried out in different halves (chosen at random)

of the 1 x 1 m plot to avoid resampling disturbed sediments. Each sample was washed

in sieves of 0.5 mm and 0.1 mm within 24 h of collection. Invertebrates were then

separated from remaining sediments by flotation in saturated brine collected from

crystallisation ponds. They were counted and stored in 70 % alcohol.

At the end of period 1, exclosures and controls were moved to new positions 2 m

away. Slight overlap between the two periods (Fig. 1) is due to the delay between

sampling and moving the exclosures and controls in the first pond and doing the same in

the last pond. In a given pond, all the exclosures and controls were established and/or

sampled on the same day. Owing to the time required to extract invertebrates from the

samples en vivo in the laboratory, it was not possible to sample all the ponds together.


98

We removed two of the exclosure-control pairs from analyses for period 2 because they

dried out due to a drop in water level.

Chironomid larvae retained on the 0.5 mm sieve were later measured to the

nearest 10 m using an image analyzer. Body length data were transformed into

estimated dry mass using the regression equation of Smock (1980). We did not estimate

dry mass of the smaller larvae retained on the 0.1 mm sieve as it was relatively

insignificant. In order to demonstrate this, we compared the actual (not estimated) dry

weight of larvae retained on different sieves in 24 samples taken at different times and

in different periods.

Censuses of shorebirds

We carried out monthly censuses of the study area of 123 ha from January to May 2002,

recording the total number of individuals and the number feeding for each shorebird

species. Censuses were carried out at high tide when tidal flats are unavailable as

alternative foraging habitat and wader densities in the saltpans are highest (Sánchez et

al., in press a).


Data from the three benthic samples taken within each exclosure or control were pooled

before analysis. We used Generalized mixed Linear Models (GLMs, McCullagh &

Nelder 1989) to analyze the effects of treatment (exclosure or control) and experimental

period on the abundance, estimated biomass and size distribution of chironomid larvae.

Pond and exclosure-control pairs nested within ponds were included as random factors,

using the GLIMMIX macro (SAS Institute 1996). For abundance and biomass we used

an identity link and normal error distribution, transforming the dependent variable to

overcome heteroscedasticity. We analyzed the change in density of larvae (number

counted in the three samples) from the beginning to the end of an exclusion period as

log10 (final number + 1) - log10 (initial number + 1)) and the change in estimated

biomass (mg in the three samples) in the same way. For size distribution we conducted a

logistic regression using the number of larvae retained on the 0.5 mm sieve (numerator)

and the total number retained on the 0.5 mm and 0.1 mm sieves (denominator) as the

dependent variable (Crawley 1993). We used data from the samples collected at the end

of each exclusion period, with a logit link and binomial error distribution. We also

analyzed the effect of treatment and period on the mean size of larvae retained on the


99

0.5 mm sieve at the end of each period, using an identity link and normal error

distribution.

Tests on the effects of each predictor were performed using F-statistics. The

deviance of each fitted GLM is analogous to the residual sum of squares in ordinary

linear regression. The reduction in deviance compared to the null model is used to

assess the explicative power of the model (Crawley 1993). We initially considered the

water depth at each exclosure-control pair as an additional predictor (results not shown).

However, we found no evidence of a depth*treatment interaction as would be expected

if predation effects varied with depth. Period*treatment interactions were not significant

and were excluded from final models.

We used similar GLMs to test whether the exclosure effect in a given patch

depended on the abundance or biomass of larvae in that patch at the beginning of an

experiment. We used the difference between the abundance in an exclosure and its

paired control at the end of a period as the dependent variable (transformed as log10

([final exclosure + 1] / [final control + 1])), and the average abundance for each pair at

the beginning of a period as a predictor. Period was a fixed factor and pond was a

random factor. We used an identity link and a normal error distribution. We repeated

this analysis using biomass data instead of abundance. The interactions between initial

abundance or biomass and period were not significant.

We used matched-paired t tests to analyze the change in the abundance and

biomass of larvae in controls between the beginning and end of each period.

RESULTS

In period 1, C. salinarius larvae and small numbers of C. salinarius pupae were the only

invertebrates recorded (Table 1). In period 2, small numbers of Ochthebius beetles were

also recorded (Table 1).

Density of larvae

At the beginning of each experiment, there was no difference between controls and

exclosures in the density of larvae (Fig. 2). At the end of period 1, 35.0 % more

chironomid larvae were found in exclosures than in controls (Fig. 2a). At the end of

period 2, 32.2 % more larvae were found in exclosures. However, in spite of these


100

predation effects, the density of larvae increased in controls throughout the study (Fig.

2a). This increase was not significant between the beginning and end of period 1 (t = -

0.95, p = 0.35), but was highly significant between the beginning and end of period 2 (t

= -3.69, p = 0.002).

We found a significant increase in larval density in exclosures compared to

controls (Table 2). There was a significant effect of period on the change in density over

time, which was greater in period 2 (Table 2). The period*treatment interaction was not

significant (F1,48 = 0.78, P = 0.38).

Biomass of larvae

The dry mass of small larvae retained on the 0.1 mm sieve was relatively unimportant,

representing only 3.8 ± 1.8% (mean ± s.e., n = 24) of total mass from the 0.1 mm and

0.5 sieves combined. At the beginning of each experiment, there was no difference

between controls and exclosures in the estimated biomass of larger larvae retained on

the 0.5 mm sieve (Fig. 2). At the end of period 1, the estimated biomass was 36.7 %

higher in exclosures than in controls (Fig. 2b). At the end of period 2, biomass was 48.9

% higher in exclosures. In controls, biomass decreased significantly during the course of

period 1 (t = 2.37, p = 0.028) but underwent a non-significant increase during period 2 (t

= -2.00, p = 0.062; Fig. 2b).

We found a highly significant increase in larval biomass in exclosures compared

to controls (Table 2). There was a significant effect of period on the change in biomass

over time, which was greater in period 2 (Table 2). The period*treatment interaction

was not significant (F1,48 = 0.03, P = 0.86).

Size distribution of larvae

The larvae retained on the 0.5 mm sieve showed a bimodal size distribution

indicating the presence of third and fourth (final) instars (Fig. 3). Although mean larval

size in exclosures was slightly larger than in controls, this difference was not significant

(using data from the end of periods 1 and 2, F1,48 = 2.12, P = 0.15). Mean size was

significantly larger at the end of period 1 than period 2 (F1,48 = 51.96, P < 0.0001). The

period*treatment interaction for mean size was not significant (F1,47 = 0.27, P = 0.6).

At the end of period 1, the proportion of larvae retained on the 0.5 mm sieve was

43.9 ± 5.5 % (mean ± s.e.) for controls and 52.1 ± 5.6 % for exclosures. At the end of

period 2, the proportion was 47.8 ± 6.1 % for controls and 50.8 ± 5.8 % for exclosures.


101

The fraction of larvae retained on the 0.5 mm sieve was significantly higher in

exclosures than in controls, but did not differ significantly between periods (Table 2).

The period*treatment interaction was not significant (F1,48 = 0.82, P = 0.37).

Effect of initial larval abundance on predation effects

While controlling for period and pond, we found no effect of initial larval density on the

difference between exclosures and controls at the end of experimental periods (r = -

0.02, F1,31 = 0.03, p=0.85). Similarly, we found no significant effect of initial larval

biomass on the difference between exclosures and controls at the end of experimental

periods (r = -0.19, F1,31 = 1.30, p=0.26).

Shorebirds numbers

During period 1, numbers of migrating shorebirds in the study area were relatively low

with a maximum count of 3116, of which 13.9 % were feeding. Numbers increased

during period 2 to a maximum of 4835, of which a much higher 80.7 % were feeding

(Fig. 1, Table 3). There was also a shift in the composition of the wader community. In

total, 18 species were recorded, all of which were feeding in the saltpans and 12 of

which were observed feeding inside our controls (Table 3). Redshank, Avocets, Black-

tailed godwits were relatively abundant during period 1, whereas Curlew sandpipers

Calidris ferruginea and Dunlins were dominant during period 2 (Table 3). Of 3904

birds recorded feeding at the end of period 2 (30 April), 53 % were Curlew sandpipers

and 43 % Dunlin.

DISCUSSION

We have recorded strong effects of predator exclosure on the density, size distribution

and especially biomass of chironomid larvae,. We observed no effects of our exclosures

on the environment and, in the absence of other predators, our results can only be

explained by predation by shorebirds. Our experiments may have underestimated the

effects of shorebirds because there may have been a net emigration of mobile benthic

chironomid larvae away from our small exclosures where densities were elevated

(Wilson 1989).


102

Other authors have shown that shorebirds have high intake rates and have

suggested that they are likely to consume a significant fraction of the production of

benthic chironomid larvae (Rehfisch 1994, Masero & Pérez-Hurtado 2001).

Nevertheless, without exclosure experiments such as ours it is not apparent that such

predation leads to a reduction in the density or standing crop of larvae, since loss via

predation may potentially be compensated for by other effects. In particular, reduced

interspecific competition leading to lower mortality rates or higher growth rates in

midge larvae (Armitage et al. 1995). Indeed, experiments excluding shorebirds or other

avian predators have not always produced detectable effects on chironomids (Smith et

al. 1986, Ashley et al. 2000) or other benthic invertebrates (Raffaelli & Milne 1987,

Wilson 1991, Lopes et al. 2000). As far as we know, the only other study in which

avian predation has been shown to reduce the density of midge larvae is one from an

inland oxbow lake (Székely & Bamberger 1992).

Given our absence of data on chironomid productivity and on feedback

processes, our results from the exclosure experiments can not be translated into a

precise estimate of the proportion of production or standing crop consumed by

shorebirds (Mitchell & Wass 1996). Nevertheless, they demonstrate a top-down control

of chironomids by shorebirds and indicate that shorebirds have an important influence

on chironomid dynamics in this system, at least during spring migration.

The highly significant effect of wader predation on larval biomass was due to a

significant decrease both in larval abundance and in the proportion of large larvae.

Shorebirds appear to have selected larger larvae retained on a 0.5 mm sieve (mean

length ± s.e. = 8.09 ± 0.04 mm, range = 1.55-16.78 mm) at the expense of smaller ones

retained on a 0.1 mm sieve (mainly first and second instars). As far as we are aware, this

is the first evidence of size selection of chironomid larvae by shorebirds, and the first

direct evidence that avian predation changes the size distribution of chironomid prey.

Size selection of invertebrates by shorebirds is well documented (Goss-Custard 1977,

Worral 1984, Dierschke et al. 1999), but in the only other study in which it led to a

measurable change in the size distribution of prey the mean size of polychaetes was

increased by wader predation (Botto et al. 1998). In other studies, size selection by

shorebirds did not influence the size distribution of their prey (Weber & Haig 1997).

Thus, whilst we found no significant evidence that wader predation affects the size

range of those chironomid larvae retained on a 0.5 mm sieve, prey size selection might

still occur in this range.


103

Some authors have attributed size selection of benthic invertebrates by

vertebrate predators to the vertical stratification of sizes within the sediments,

suggesting that smaller polychaetes or chironomids were selected because they were

closer to the surface (Hershey 1985, Botto et al. 1998). We doubt that larger larvae were

selected in our study because they were more accessible. We only sampled 3 cm of

sediments, all of which were accessible to most of the wader species. The smallest first

instars are likely to be particularly close to the surface since they are the principal stage

for dispersal through the water column (Armitage et al. 1995). On the other hand,

Redshank at Odiel consume more chironomid pupae than expected from their low

density in the benthos, presumably because they become particularly accessible when

preparing for emergence (Sánchez et al. in press b).

We found no evidence that the impact of wader predation was related to water

column depth over the 0-20 cm range studied, perhaps owing to the range of wader

species present and their partitioning between different depths (Ntiamoa-Baidu et al.

1998). Neither did we find that exclosure effects were related to initial biomass or

density. This suggests that shorebirds did not concentrate larval predation in patches

where prey abundance was highest. Different wader species may respond differently to

variation in depth or larval size when foraging optimally, and perhaps the pooling of

their effects means that there is no optimal foraging response detectable at a community

level. At the higher scale of pond selection, shorebirds do select those saltpans

providing higher densities of larvae (Sánchez et al. in press a).

Although the density of foraging shorebirds was much higher in period 2 than in

period 1, this was not reflected in significant differences between periods in the

observed exclosure effects. Thus, detectable predation effects were not greater as

predator density increased (contrary to Botto et al. 1998) and even relatively low

densities of shorebirds produce detectable effects. Indeed, in areas exposed to predation

the biomass decreased significantly during period 1 but increased during period 2.

Higher predation rates in period 2 may have been compensated for by higher larval

recruitment or productivity. Both may have increased between early February and late

April in association with increased water temperature (Drake & Arias 1995). Our results

show that the effects that predation has on benthic prey populations does not only

depend on the intensity of predation, but also on its timing in relation to prey life cycles.

A relatively small amount of consumption at a moment of low recruitment or growth


104

may have more consequences for prey populations than much greater consumption at a

time of peak productivity (Mitchell & Wass 1996).

Our results indicate that the availability of chironomid prey to a migratory wader

depends on the absolute time of its arrival, since there is significant variation in larval

abundance with date. Furthermore, it also depends on the number of other individuals

arriving beforehand, since earlier predation reduces prey availability at a given moment.

The foraging intake rates of shorebirds have previously been shown to decrease as the

density of benthic midge larvae decreases (Székely & Bamberger 1992). Thus,

decisions that shorebirds take about when to migrate may be influenced by prey

depletion at stopover sites such as we have recorded in the Odiel saltpans (Schneider &

Harrington 1981). These changes in prey density and intake rates may also influence the

time birds need to spend at stopover sites such as Odiel marshes to reach the body

condition required to continue on to breeding sites. Any delay in arrival at breeding sites

is likely to reduce breeding success (Sandercock et al. 1999). Since all wader species in

our study site feed on chironomid larvae (pers. obs.) and most were seen feeding inside

our experimental controls (Table 3), our results can not be attributed to a single predator

species and suggest that interspecific competition may be an important process in this

system.

Elsewhere we present evidence that changes in the abundance of C. salinarius in

Odiel saltpans during the course of the annual cycle partly determines the numbers of

shorebirds feeding in saltpans instead of alternative intertidal habitats (Sánchez et al. in

press a). Differences in density of midge larvae also explain differences in wader

density between individual ponds (Sánchez et al. in press a). Similarly, changes in the

importance of C. salinarius in wader diet is related to fluctuations in their abundance

(Sánchez et al. in press b). Thus, a localized predator-prey relationship appears to exist

at Odiel in which both bottom-up control and top-down control are important in the

regulation and structuring of predator and prey populations in saltpans. In a given pond,

the abundance of midge larvae undergoes top-down control by shorebirds, but also

exerts a bottom-up control on the abundance of shorebirds.


105

ACKNOWLEDGEMENTS


Tecnología and a postgraduate grant from the Consejo superior de Investigaciones

Científicas. Consejería de Medio Ambiente, Junta de Andalucía and Aragonesas

Industrias y Energía S.A. provided permission to work in the salines. J.C. Rubio,

Director of the Odiel Marshes Natural Park, provided logistical support

and advice. E. Castellanos, C. de le Court, J.M. Sayago and E. Urbina also provided

helpful advice. J.A. Amat, J. Figuerola and J.A. Masero helped us to improve an earlier

version of this manuscript.

REFERENCES

Armitage, P., Cranston, P.S. & Pinder, L.C.V. eds. 1995. The Chironomidae: the


Ashley, M., Robinson, J.A., Oring, L.W., & VinyardY, G.A. 2000. Dipteran standing

stock biomass and effects of aquatic bird predation at a constructed wetland. Wetlands

20(1): 84-90.

Batzer, D.P. 1998. Trophic interactions among detritus, benthic midges, and predatory

fish in a freshwater marsh. Ecology 79: 1688-1698.

Botto, F., Iribarne, O.O. & Martínez, M.M. 1998. The effect of migratory shorebirds on

the benthic species of three southwestern atlantic argentinean estuaries. Estuaries

21(4B): 700-709.

Britton, R. H. & Johnson, A.R. 1987. An ecological account of a Mediterranean salina:


Crawley, M.J. 1993. GLIM for Ecologist, Blackwell Scientific Publications,

Cambridge.


106

Dierschke, V. K. J. & Rippe, H. 1999. Feeding ecology of dunlins Calidris alpina

staging in the southern Baltic Sea, 2. Spatial and temporal variations in the harvestable

fraction of their favourite prey Hediste diversicolor. Journal of Sea Research 42: 65-82.

Drake, P. & Arias, A.M. 1995. Distribution and production of Chironomus salinarius

(Diptera: Chironomidae) in a shallow coastal lagoon in the Bay of Cádiz. Hydrobiologia

299: 195-206.

Goss-Custard, J. D. 1977. Optimal foraging and the size selection of worms by

redshank, Tringa totanus, in the field. Animal Behaviour 25: 10-29.

Hershey, A. E. 1985. Effects of predatory sculpin on the chironomid communities in an

arctic lake. Ecology 66 (4): 1131-1138.

Hicklin, P. W. & Smith, P.C. 1984. Selection of foraging sites and invertebrate prey by

Semipalmated Sandpipers, Calidris pusilla (Pallas), in Minas Basin, Bay of Fundy.

Canadian Journal of Zoology 62: 2201-2210.

Kornijów, R. 1997. The impact of predation by perch on the size-structure of

Chironomus larvae - the role of vertical distribution of the prey in the bottom sediments,

and habitat complexity. Hydrobiologia 341/343: 207-213.

Krapu, G.L. & Reinecke, K.J. 1992. Foraging ecology and nutrition. Pp. 1-29 in: Batt,

B.D.J, Afton, A.D., Anderson, M.G., Ankney, C.D., Johnson, D.H., Kadlec, J.A. &

Krapu, G.L., eds. Ecology and Management of Breeding Waterfowl. University of

Minnesota Press, Minneapolis and London.

Kvist, A. & Lindström, A. 2003. Gluttony in migratory waders – unprecedented energy

assimilation rates in vertebrates. Oikos 103: 397-402.

Lopes, R. J., Pardal, M.A. & Marques, J.C. 2000. Impact of macroalgal blooms and

wader predation on intertidal macroinvertebrates: experimental evidence from the

Mondego estuary (Portugal). Journal of Experimental Marine Biology and Ecology 249:

165-179.


107

Masero, J. A. & Pérz-Hurtado, A. 2001. Importance of the supratidal habitas for


adjacent saltworks in southern Europe. The Condor 103: 21-30.

McCullan, & Nelder. 1989. Generalized Linear Models, 2nd edn. Chapman & Hall,

London.

Mitchell, S. F. & Wass, R.T. 1996. Quantifying herbivory: grazing consumption and

interaction strength. Oikos 76: 573-576.

Ntiamoa-Baidu, Y., Piersma, T., Wiersma, P., Poot, M., Battley, P. & Gordon, C. 1998.

Water depth selection, daily feeding routines and diets of waterbirds in coastal lagoons

in Ghana. Ibis 140: 89-103.

Raffaelli, D. & Milne, H. 1987. An experimental investigation of the effects of

shorebirds and flatfish predation on estuarine invertebrates. Estuarine Coastal Shelf

Science 24: 1-13.



Ecology 31: 383-401.

Sánchez, M. I., Green, A.J. & Castellanos, E.M. In press a. Spatial and temporal

fluctuations in use by shorebirds and in availability of chironomid prey in the Odiel

saltpans, south-west Spain. Hydrobiologia.

Sánchez, M. I., Green, A.J. & Castellanos, E.M. In press b. Seasonal variation in the

diet of the Redshank Tringa totanus in the Odiel Marshes, south-west Spain: a

comparison of faecal and pellet analysis. Bird Study.

Sandercock, B., Lank, D.B. & Cooke, F. 1999. Seasonal declines in the fecundity of

arctic-breeding sandpipers: different tactics in two species with an invariant clutch size.

Journal of Avian Biology 30: 460-468.


108

SAS Institute Inc. 1996. SAS System for Mixed Models. Cary, NC

Schneider, D. C. & Harrington, B.A. 1981. Timing of shorebird migration in relation to

prey depletion. The Auk 98: 801-811.

Smith, L. M., Vangilder, L.D., Hoppe, R.T., Morreale, S.J. & Brisbin, I.L. 1986. Effect

of diving ducks on benthic food resources during winter in South Carolina, U.S.A.

Wildfowl 37: 136-141.

Smock, L. A. 1980. Relationships between body size and biomass of aquatic insects.

Freshwater Biology 10: 375-383.

Sutherland, T. F., Shepherd, P.C.F. & Elner, R.W. 2000. Predation on meiofaunal and

macrofaunal invertebrates by western sandpipers (Calidris mauri): evidence for dual

foraging modes. Marine Biology 137: 983-993.

Székely, T. & Bamberger, Z. 1992. Predation of waders (Charadrii) on prey

populations: an exclosure experiment. Journal of Animal Ecology 61: 447-456.

Thrush, S. F. 1999. Complex role of predators in structuring soft-sediment

macrobenthic communities: Implications of changes in spatial scale for experimental

studies. Australian Journal of Ecology 24: 344-354.

Weber, L. M. & Haig, S.M. 1997. Shorebird-prey interactions in South Carolina coastal

soft sediments. Canadian Journal of Zoology 75: 245-252.

Wilson, W. H., JR. 1989. Predation and the mediation of intraspecific competition in an

infaunal community in the Bay of Fundy. Journal of Experimental Marine Biology and

Ecology 132: 221-245.

Wilson, W. H. 1991. The foraging ecology of migratory shorebirds in marine soft-

sediment communities: the effects of episodic predation on prey populations. American

Zoologist 34: 840-848.


109

Worrall, D. H. 1984. Diet of the Dunlin Calidris alpina in the Severn Estuary. Bird

Study 31: 203-212.


110

Table 1. Invertebrates recorded in benthic samples, their percentage of occurrence (PO,

i.e. the proportion of samples in which they were recorded) and the percentage of total

individuals (PI) that they represented.

Period I (n = 240) Period II (n = 216) PO (%) PI (%) PO (%) PI (%)C. Salinarius larvae 100 98.3 100 99C. Salinarius pupae 19.2 1.7 21.3 0.92Ochthebius spp. 0 0 0.9 0.026


111

Table 2. Summary of Generalized Linear Models testing the partial effects of treatment

(exclosure or control) and period on the density, biomass and size distribution of C.

salinarius larvae recorded in experimental plots (n = 76). Density is the change in total

numbers of larvae recorded in each 1 m2 plot from the beginning to the end of the

experiment. Biomass is the change in estimated dry mass of larvae retained on a 0.5 mm

sieve from the beginning to the end of the experiment. Size is the number of larvae

retained on a 0.5 mm sieve as a proportion of the total. Pond and exclosure-control pair

nested within pond were included as random factors using the GLIMMIX macro. See

methods for more details. Estimates given are those for controls and period 1.

Exclosures and period 2 were aliased.

Effect Estimate SE F1,49 pDensity Constant 0.33 0.08 D = 16.00 Treatment 5.83 0.019

Control -0.15 0.06Period 4.7 0.035

I -0.15 0.07 Biomass Constant 0.47 0.1D = 44.80 Treatment 17.86 0.0001

Control -0.25 0.06Period 41.42 <0.0001

I -0.44 0.07 Frec 0.5 Constant 0.12 0.33D = 14.34 Treatment 5.02 0.03

C -0.34 0.15Period 2.64 0.11

I -0.33 0.2

D = percentage of deviance explained by the final model compared to the null model.


112

Table 3. Total numbers of shorebirds (and number feeding) counted from January to

May 2002 in the study area of 123 ha.

30-january 28-february 30-march 30-april 29-mayTringa nebularia* 12 (4) 17 (1) 19 (5) 14 (11) 0 (0)Limosa limosa* 164 (146) 36 (3) 0 (0) 0 (0) 0 (0)Limosa lapponica 0 (0) 214 (1) 80 (0) 61 (14) 59 (2)Tringa totanus* 106 (65) 214 (73) 325 (155) 200 (106) 77 (26)Tringa erythropus* 0 (0) 1 (1) 53 (22) 4 (4) 0 (0)Actitis hypoleucos 0 (0) 0 (0) 8 (0) 1 (1) 0 (0)Recurvirostra avosetta* (101) 10 145 (126) 8 (7) 2 (2) 2 (2)Calidris alpina* 9 (9) 16 (13) 5 (5) 2149 (1684) 321 (265)Calidris canutus 0 (0) 0 (0) 0 (0) 13 (0) 44 (38)Calidris minuta* 3 (3) 3 (3) 3 (1) 1 (1) 0 (0)Calidris alba* 12 (12) 14 (9) 0 (0) 17 (14) 141 (131)Calidris ferruginea* 0 (0) 1951 (165) 6 (6) 2306 (2052) 9 (9)Charadrius hiaticula* 0 (0) 0 (0) 0 (0) 12 (8) 25 (17)Charadrius alexandrinus 1 (1) 0 (0) 0 (0) 0 (0) 0 (0)Pluvialis squatarola* 0 (0) 364 (0) 131 (0) 48 (5) 29 (8)Himantopus himantopus* 31 (15) 139 (35) 250 (118) 0 (0) 3 (3)Arenaria interpres 1 (0) 2 (2) 0 (0) 7 (2) 1 (1)Philomachus pugnax 0 (0) 2 (2) 0 (0) 0 (0) 0 (0)TOTAL 440 (265) 3118 (434) 880 (319) 4835 (3904) 711 (502)

* species seen feeding inside the control plots.


113

Figure 1. Numbers of shorebirds feeding in the study area in relation to the timing of the

exclosure experiments.

Date

30-1 28-2 30-3 30-4 29-5

Nº o

f fee

ding

sho

rebi

rds

0

1000

2000

3000

4000

5000

PeriodI Period II


114

Figure 2. a) Density and b) estimated biomass of C. salinarius larvae (mean ± s.e.) at

the beginning and end of exclosure experiments.

Period I Period II

Initial Final Initial Final0

5000

10000

15000

20000

25000

30000

35000

Chi

rono

mid

larv

ae d

ensi

ty (

nº/m

2)

Control Exclosure

0

20

40

60

80

Period I Period II

Initial Final Initial Final

Chi

rono

mid

larv

ae b

iom

ass

(mg)

Control Exclosure


115

Figure 3. Size frequency distribution of C. salinarius larvae at the beginning and end of

exclosure experiments. Only larvae retained on a 0.5 mm sieve are included. Initial data

are pooled for controls and exclosures. Length category 1 = 0-1mm, etc.

Initial

Category of length (mm)

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

% o

f obs

erva

tions

0

5

10

15

20

25

Final Control

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 190

5

10

15

20

25

Final Exclosure

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 190

5

10

15

20

25

Period I

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 190

5

10

15

20

25

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 190

5

10

15

20

25

Initial

Final Control

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 190

5

10

15

20

25

Final Exclosure

Period II

116

CHAPTER 5 CESTODES FROM ARTEMIA PARTHENOGENETICA

117

CHAPTER 5

CESTODES FROM ARTEMIA PARTHENOGENETICA (CRUSTACEA,

BRANCHIOPODA) IN THE ODIEL MARSHES, SPAIN: A SYSTEMATIC

SURVEY OF CYSTICERCOIDS

BOYKO B. GEORGIEV1,2*, MARTA I. SÁNCHEZ3, ANDY J. GREEN3, PAVEL

N. NIKOLOV1, GERGANA P. VASILEVA1 & & RADKA S. MAVRODIEVA1

1Central Laboratory of General Ecology, Bulgarian Academy of Sciences, 2 Gagarin Street, 1113 Sofia,

Bulgaria.2Department of Zoology, The Natural History Museum, Cromwell Road, London SW7 5BD, U.K.

3Estación Biológica de Doñana (CSIC), Avda. María Luisa s/n, 41013 Sevilla, Spain.

Key words: Artemia, Cyclophyllidea, cysticercoids, life cycles, intermediate host

Running head: Cysticercoids from Artemia parthenogenetica

Acta Parasitologica 50 (2): 105-117.


118

ABSTRACT

A total of 3,300 specimens of brine shrimps Artemia parthenogenetica from the Odiel

Marshes, Huelva Province, SW Spain, were studied during several seasons of 2002 and

2003 for the presence of cestode infections. Cysticercoids were found in 26.8% of brine

shrimps. Eight cestode species were recorded, i.e. Hymenolepididae: Flamingolepis

liguloides (adults parasitic in flamingos) with prevalence (P) 18.5%, mean intensity

(MI) 1.48 and mean abundance (MA) 0.28; F. flamingo (adults parasitic in flamingos),

P 0.9%, MI 1.03, MA 0.01; Confluaria podicipina (adults parasitic in grebes), P 6.5%,

MI 1.42, MA 0.09; Wardium stellorae (adults parasitic in gulls), P 0.2%, MI 1.00, MA

0.002; Dilepididae: Eurycestus avoceti (adults parasitic in waders and flamingos), P

2.7%, MI 1.08, MA 0.03; Anomotaenia sp., probably A. microphallos (adults parasitic

in waders), P 0.8%, MI 1.04, MA 0.01; A. tringae (adults parasitic in waders), P 2.2%,

MI 1.01, MA 0.02; Progynotaeniidae: Gynandrotaenia stammeri (adults parasitic in

flamingos), P 0.6%, MI 1.00, MA 0.01. The cysticercoids are described and

accompanying illustrations are presented. This study provides the first record of

Anomotaenia tringae in an intermediate host and the first records of C. podicipina, E.

avoceti, A. tringae and G. stammeri in Spain.


119

INTRODUCTION

Previous studies revealed the role of brine shrimps of the genus Artemia Leach, 1819

(Crustacea, Branchiopoda) as intermediate host in the life cycles of cestodes parasitising

aquatic birds. The most detailed studies were carried out at Tengiz Lake, Kazakhstan

(Maksimova 1973, 1976, 1977, 1981, 1986, 1987, 1988, 1989, 1991, Gvozdev and

Maksimova 1979) and in the Camargue, France (Gabrion and MacDonald 1980, Thiéry

et al. 1990, Robert and Gabrion 1991); in those places, cysticercoids of 11 and 6 avian

cestode species, respectively, were found in brine shrimps. In Spain, only two cestode

species, Flamingolepis liguloides (Gervais, 1874) and Wardium stellorae (Deblock,

Biguet et Capron, 1960), were found in Artemia at the Mediterranean coast (Amat et al.

1991a, 1991b, Varó et al. 2000).

The present study was carried out at one of the largest saltworks in Spain, which

is also a site of major importance for waterbirds (Bernués 1998, Sánchez et al. in press).

It is situated in the estuary of the Odiel and Tinto Rivers, Huelva Province, SW Spain,

near the Atlantic coast of Andalusia. The aim of this article is to characterise the species

composition of the avian cestodes utilising crustaceans of the genus Artemia as

intermediate hosts in the Odiel Marshes. It also provides the taxonomic framework for

ecological studies being carried out on brine shrimps and their cestode parasites (results

to be reported elsewhere).

MATERIALS AND METHODS

The identification of the brine shrimps was based on the work by Abatzopoulos et al.

(2002) and recent studies at the same site (Amat et al. 2005). In the Mediterranean

Region, the native species of Artemia are the bisexual A. salina Leach, 1819 (synonym

A. tunisiana Bowen et Sterling, 1978) together with a heterogeneous group of

parthenogenetic populations known under the binomen A. parthenogenetica Bowen et

Sterling, 1978 (see Abatzopoulos et al. 2002). A. parthenogenetica only has been

recorded in the Odiel Saltworks (Amat et al. 2005).

Samples, each containing 500 specimens of brine shrimps, were collected from

an evaporation pond (E-18) at the Odiel Saltworks in October and December of 2002


120

and in February, April, June and August of 2003. In addition, a sample of 200

specimens collected in October, 2002 and a sample of 100 specimens collected in

November, 2003 were examined. In total, 3,300 specimens of brine shrimps were

studied for the presence of cysticercoids. The collected brine shrimps were killed by

heating to 80ºC and preserved in 70% ethanol. They were mounted in temporary

glycerol mounts and examined under a stereomicroscope or compound microscope after

gentle pressure on the coverslip. If the identification of the cysticercoids recorded was

not possible at this stage, whole infected brine shrimps or isolated cysticercoids were

prepared as permanent mounts in Berlese’s medium in order to facilitate observations

on the morphology of rostellar hooks. Data on the prevalence, intensity and abundance

are based on these samples.

Another sample of 380 specimens collected from the same and neighbouring

ponds (E-11, E-18 and E-17) were examined in July, 2004 in order to provide further

information on the morphology of the cysticercoids, so as to complete that gathered on

the basis of the fixed materials. The brine shrimps were examined alive under a

stereomicroscope and a compound microscope. After observations of the cysticercoids

in situ, each infected specimen was gently pressed under the coverslip and the

observations continued on isolated cysticercoids. These were carried out initially in

hypertonic conditions (in water from ponds) and later in hypotonic conditions (in tap

water). The latter was found to provide longer survival of the cysticercoids (for about 1

hour) ex situ and, in some cases (Eurycestus avoceti, Anomotaenia microphallos and

Gynandrotaenia stammeri), to provoke their excystation.

In July, 2004, samples of other invertebrates abundant in the same ponds were

examined. These were adult coleopterans of the families Hydraenidae (Ochthebius

corrugatus Rosenhauer, 1856 – 23 specimens and O. notabilis Rosenhauer, 1856 – 55

specimens) and Dytiscidae (Nebrioporus ceresyi (Aubé, 1836) – 70 specimens,

chironomid larvae tentatively identified as Chironomus salinarius Kieffer in

Thienemann, 1915 – 424 specimens, and harpacticoid copepods Cletocamptus

retrogressus Shmankevich, 1875 – 160 specimens.

Voucher specimens of cysticercoids found are deposited in the collection of The

Natural History Museum, London (BMNH).

The metrical data in the following descriptions are based on specimens mounted

in Berlese’s medium only. They are given as the range, with the mean and the number

of measurements taken in parentheses. The measurements are in micrometres except


121

where otherwise stated. The infection descriptors (prevalence, intensity and abundance)

are after the definitions by Bush et al. (1997).

RESULTS

Infection characteristics

Of 3,300 specimens of Artemia studied, 886 specimens (26.8%) were infected with

cysticercoids. The intensity of the infection was 1-13 cysticercoids. The total values of

the mean abundance and the mean intensity were 1.71 and 0.44, respectively.

Cysticercoids of 8 cyclophyllidean species belonging to three families were recorded in

the brine shrimps examined. Among them, Flamingolepis liguloides was the most

prevalent and abundant species (Table 1).

The majority of the infected brine shrimps contained cysticercoids of one cestode

species; 171 specimens (19.3% of the infected shrimps) harboured more than one. Two

cestode species were recorded from 135 shrimps, three species from 32 specimens, four

species from three specimens and one shrimp contained cysticercoids of 5 tapeworm

species (Table 2).

No cysticercoids were recorded in other invertebrate species studied.

Systematic survey of cysticercoids

Family Hymenolepididae

Flamingolepis liguloides (Gervais, 1847)

Voucher specimens: BMNH 2005.2.23.1-5 (5 slides).

Description: Cyst oval (Figs. 9, 10), 560-810 x 372-597 (671 x 479, n = 14). Calcareous

corpuscles numerous. Scolex 358-771 x 339-490 (614 x 423, n = 14). Suckers round or

oval, with diameter 181-288 (233, n = 18). Rostellum elongate, 446-485 x 116-149 (465

x 125, n = 14), with strongly developed circular musculature. Rostellar hooks

skrjabinoid, 8 (n = 15) in number, 186-201 (189, n = 15) long (Figs. 1, 11); length of

blade 105-117 (110, n = 15). Ratio of length of blade / total hook length 0.56-0.60

(0.58, n = 15). When rostellum contracted, blades of hooks are posteriorly directed.


122

Cercomer elongate (Fig. 10), with length comparable with that of cyst, 620-810 x 45-68

(710 x 55, n = 8).

Remarks: The present material corresponds to the cysticercoids from Artemia sp.

identified by Robert and Gabrion (1991) as F. liguloides, and from Artemia sp.

(originally identified as A. salina) and Branchinella spinosa (M. Milne-Edwards)

reported by Maksimova (1973, 1989) as F. dolguschini Gvozdev et Maksimova, 1968.

The taxonomy of Flamingolepis Spasskii et Spasskaya, 1954 is poorly known. Amat et

al. (1991b) believed that F. dolguschini is a synonym of F. liguloides. However, the

adult cestodes described by Maksimova (1989) under these two names clearly differed

from each other. Nevertheless, the affiliation of these cysticercoids from France, Spain

and Kazakhstan, all with 8 skrjabinoid hooks 180-190 μm long, to a single species is

justified by their morphology. Pending a revision of the hymenolepidids specific to

flamingos, we follow Robert and Gabrion (1991) and accept F. liguloides as the most

probable identification of these cysticercoids.

Flamingolepis flamingo (Skrjabin, 1914)


Description: Cyst oval, thick-walled (Figs. 12, 13), 168-270 x 126-207 (231 x 177, n =

14). Calcareous corpuscles numerous, concentrated mostly in anterior part of cyst.

Scolex 141-225 x 108-183 (182 x 145, n = 12). Suckers oval, with diameter 45-75 (66,

n = 16). Rostellum highly elongate, with strong circular musculature, 108-180 x 33-48

(119 x 43, n = 5) when contracted or 177-225 x 33-39 (206 x 36, n = 3) when extended.

Rostellar hooks skrjabinoid, 8 (n = 16) in number, 55-61 (57, n = 15) long (Figs. 2, 14);

length of blade 28-30 (29, n = 15). Ratio of length of blade / total hook length 0.47-0.53

(0.50, n = 15). When rostellum contracted, blades of hooks are posteriorly directed.

Cercomer highly elongate (Figs. 12, 13), 6.5-7.4 mm (n = 3) long, 8-12 (10, n = 14)

wide.

Remarks: F. flamingo is a specific parasite of flamingos in Eurasia (Maksimova 1989).

Robert and Gabrion (1991) found its cysticercoid in Artemia sp. in France. The present

material is in agreement with their description.

Confluaria podicipina (Szymanski, 1905)



123

Description: Cyst oval, thin-walled (Fig. 19), 93-147 x 47-87 (121 x 73, n = 12), often

with distinct transverse striations. Calcareous corpuscles numerous, concentrated in

anterior and in posterior part of cyst (Figs. 18, 19). Scolex 72-104 x 38-72 (92 x 59, n =

12). Suckers slightly oval, with diameter 26-32 (28, n = 10). Rostellum 42-47 (46, n =

7). Rostellar hooks 10 in number, aploparaksoid (Fig. 3), 21-24 (22, n = 12) long; when

rostellum contracted, blades of hooks anteriorly directed. Cercomer very large. Anterior

part of cercomer forms additional envelope surrounding cyst, thus forming external

capsule (Figs. 17, 18) measuring 195-255 x 135-204 (215 x 152, n = 9). Posterior part

of cercomer highly coiled, densely packed in thin membranous envelope (Fig. 16),

which surrounds entire cysticercoid.

Remarks: C. podicipina is a parasite of grebes (Podicipedidae) in the Holarctic Region

(see Vasileva et al. 2000 for a survey). Its cysticercoid was described from Artemia in

Kazakhstan (Maksimova 1981, 1989). We were not able to measure the length of the

cercomer exactly because of its coiled configuration; however, it seems to be close to

the length reported by Maksimova (1981), i.e. 14.5-16.5 cm. We also confirm that the

cysticercoid rostellar hook of this species corresponds to the hook part referred to as the

‘refractive particle’ (Vasileva et al. 2000) or ‘tip hooklet’ (Maksimova 1981, 1989) of

adults, and therefore that hooks do grow in the final host.

Our observations on living specimens lead us to an interpretation of the structure of the

cysticercoid differing from that by Maksimova (1981). She believed that the external

wall of the cyst consisted of 6 layers. In contrast, we consider that only the ‘internal

cyst’ is homologous to the cysts of other hymenolepidid cysticercoids, while the

external 3 layers described by Maksimova (1981) are a modification of the anterior part

of the cercomer, forming an additional protective envelope surrounding the cyst. When

placed in hypotonic conditions, the real cyst is readily detached from the bottom of the

cavity formed by the anterior part of the cercomer (Figs. 18, 19); simultaneously, a

distinct orifice can be seen at the anterior end of the ‘external cyst’ (Fig. 18).

Wardium stellorae (Deblock, Biguet et Capron, 1960)

Synonyms: Hymenolepis stellorae Deblock, Biguet et Capron, 1960; Aploparaksis

parafilum Gąsowska, 1932 sensu Maksimova (1973).


Description: Cyst oval to lemon-shaped (Fig. 20), 189-213 x 153-189 (199 x 166, n =

5). Scolex 123-132 x 105-123 (126 x 111, n = 5). Rostellum oval, 39-45 x 25-30 (41 x


124

27, n = 5). When rostellum is contracted blades of rostellar hooks are directed

anteriorly. Suckers slightly oval, with diameter 45-57 (50, n = 7). Rostellar hooks10 in

number, 23 (n = 4) long, aploparaksoid, with distinct handle; blade considerably longer

than guard (Figs. 4, 20). Entire cercomer was not isolated; length of cercomer exceeds

400 (judging from fragmented cercomer), width 25-39.

Remarks: W. stellorae was described as an intestinal parasite of Larus ridibundus L. on

the French Atlantic coast (Deblock et al. 1960). Further studies recorded it from L.

genei Breme in the Ukraine and Kazakhstan and from L. californicus Lawrence in North

America (data summarised by Maksimova 1986, 1989). Its cysticercoid was reported

from brine shrimps from Kazakhstan (Maksimova 1986), France (Robert and Gabrion

1991) and the Mediterranean coast of Spain (Varó et al. 2000). The rostellar hooks of

the present material correspond with the previous descriptions of adults (Deblock et al.

1960, Maksimova 1986, 1989) and cysticercoids (Maksimova 1986, Robert and

Gabrion 1991). Maksimova (1986) mentioned that the cysticercoids from Artemia

identified by her (Maksimova 1973) as Aploparaksis parafilum Gąsowska, 1932 should

also be considered as belonging to W. stellorae.

Recently, Bondarenko and Kontrimavichus (2004a) erected the genus

Branchiopodataenia for a group of species from gulls and previously referred to

Wardium Mayhew, 1925. They used the structure of the female copulatory ducts as a

main distinguishing character and showed that the cysticercoids of these species

develop in branchiopods. None of the cysticercoids described by Bondarenko and

Kontrimavichus (2004b) has rostellar hooks identical to those of the present material. B.

anaticapicirra Bondarenko et Kontrimavichus, 2004, a parasite of Arctic gulls, has

rostellar hooks 18-22 long but with the blade only slightly larger than the guard. The

cysticercoid of B. arctowskii (Jarecka et Ostas, 1984), described from Antarctic gulls

and recorded at various places in the Holarctic Region, has hooks 15-16 long. B.

haldemani (Shiller, 1951) and B. pacifica (Spassky et Jurpalova, 1969) from Arctic

gulls are characterised by rostellar hooks 12-13 and 16-18 long, respectively

(Bondarenko and Kontrimavichus 2004b). Thus, W. stellorae is a distinct form,

differing from Branchiopodataenia spp. Although its cysticercoid develops in

branchiopods, the morphology of the adult worms (Deblock et al. 1960) does not

correspond to the diagnosis of Branchiopodataenia. Therefore, we prefer to consider it

as belonging to Wardium.


125

Family Dilepididae

Eurycestus avoceti Clark, 1954


Description: Outer capsule oval or almost spherical (Fig. 21), brownish, with granular

contents, 240-330 x 230-279 (295 x 255, n = 9). When compressed, outer capsule

breaks down into cercomer fragments (Fig. 22). Cyst oval (Fig. 23), 141-228 x 99-163

(182 x 137, n = 12). Calcareous corpuscles numerous. Scolex rounded (Fig. 23), 111-

150 x 81-114 (135 x 98, n = 8). Suckers round or slightly oval, with diameter 26-36 (30,

n = 15); their anterior margin armed with hooklets, 6-8 long, arranged in 1-2 layers

(Figs. 5c, 24); number of hooklets on 1 sucker 9-17 (n = 18): 9 (n = 1), 10 (n = 2), 11 (n

= 6), 12 (n = 1), 13 (n = 3), 14 (n = 1), 16 (n = 3) and 17 (n = 1). Rhynchus very long

(Figs. 25, 26). Rostellar hooks 14 (n = 1) or 16 (n = 12), in 2 rows (Figs. 25, 26).

Anterior hooks 16-18 (17, n = 12) long, blade 3-4 long (Fig. 5a). Posterior hooks 15-16

(n = 10) long, blade 3 long (Fig. 5b).

Remarks: Adults of E. avoceti are parasites of charadriiform birds, mostly Recurvirostra

spp., throughout the Holarctic Region (Baer 1968, Spasskaya and Spasskii 1978,

Maksimova 1991). In Kazakhstan, this species was also recorded in 23.3% of flamingos

studied (Maksimova 1991). The present identification is based on a correspondence

between the cysticercoids found and previous descriptions of adults (Baer 1968,

Maksimova 1991) and cysticercoids (Gabrion and MacDonald 1980, Maksimova 1991).

However, there are some substantial differences in the lengths of the posterior rostellar

hooks and the number of the sucker hooklets (Table 3), which suggest that cysticercoids

from France, Spain and Kazakhstan may represent more than one species. Until now, E.

avoceti is the only species of the genus that has been recorded in the Palaearctic Region.

Two further species occur in North America (Burt 1979) but their scoleces have not

been described. Furthermore, the two Palaearctic species of Paraliga Belopol’skaya et

Kulachkova, 1973, which includes dilepidids from charadriiform birds (Deblock and

Rosé 1962, Belopol’skaya and Kulachkova 1973), are also characterised by a similar

scolex armament (Table 3). Therefore, although the identification of our material as E.

avoceti is the most probable based on a comparison with Baer’s (1968) description, it is

somewhat provisional.

Anomotaenia sp. (cf. A. microphallos (Krabbe, 1869))



126

Description: Outer capsule spherical to slightly oval (Figs. 27, 28), brownish, with

granular contents, 228-316 x 192-283 (272 x 233, n = 11). Cyst oval (Figs. 28, 29), 186-

209 x 121-150 (194 x 136, n = 12). Calcareous corpuscles numerous, concentrated

mostly in anterior and posterior part of cyst. Scolex rounded (Figs. 29, 30), 135-165 x

105-120 (150 x 109, n = 8). Suckers slightly oval (Fig. 30), with diameter 48-60 (54, n

= 14). Rostellum elongate (Fig. 30), 65-85 x 25-36 (76 x 31, n = 7). Rostellar hooks 26

(n = 1), 28 (n = 8) or 30 (n = 2), form irregular crown (Fig. 29). Anterior hooks situated

one by one; 2 or 3 posterior hooks situated between each 2 anterior hooks; sometimes,

posterior hooks at different level, thus creating impression of crown consisting of three

rows. Anterior hooks with guard almost perpendicular to hook axis (Fig. 6a), 12-13 (n =

10) long; posterior hooks with guard almost parallel to blade (Fig. 6b), 11-12 (n = 10)

long.

Remarks: The number of rostellar hooks of the present material (26-30) suggests its

affiliation with the family Dilepididae. The only dilepidid species from Palaearctic

aquatic birds characterised by numerous (more than 20) rostellar hooks with a length of

about and less than 15 μm is Anomotaenia microphallos (for surveys of Palaearctic

dilepidids from charadriiform birds and from fish-eating birds, see Spasskaya and

Spasskii 1978 and Ryzhikov et al. 1985, respectively). This species was originally

described from Vanellus vanellus (L.) in Germany as having a crown of 24 hooks with a

length of 14-16 μm (anterior) and 12-14 μm (posterior) (Krabbe 1869). Further studies

added to the range of its final hosts waders of the genera Charadrius L., Calidris

Merrem, Gallinago Brisson, Philomachus Merrem and Tringa L. (see Spasskaya and

Spasskii 1978). However, none of the descriptions of adult cestodes of this species

surveyed by Spasskaya and Spasskii (1978) reported more than 24 hooks. Therefore, we

prefer to retain the identification at the generic level only.

Anomotaenia tringae (Burt, 1940)


Description: Outer capsule oval (Fig. 31), yellowish-brown, with granular contents,

249-316 x 218-297 (279 x 236, n = 12). Cyst oval (Fig. 32), 107-139 x 84-102 (131 x

93, n = 14). Calcareous corpuscles numerous, concentrated mostly in anterior part of

cyst. Scolex rounded, 76-85 x 66-72 (82 x 70, n = 9). Suckers indistinct in available

mounts in Berlese’s medium, obviously with weak musculature. Rostellum 62-66 x 23-

27 (n = 3). Rostellar hooks 18 (n = 6) or 20 (n = 4), in 2 rows forming compact crown


127

(Fig. 32); blade slightly shorter than handle (Fig. 7a, b); anterior hooks 20-21 (n = 9)

long; posterior hooks 19-20 (n = 8) long.

Remarks: This species was originally described from Tringa glareola L. and T. totanus

L. in Sri Lanka (Burt 1940). Subsequently, it was recorded from charadriiform birds of

the genera Tringa, Calidris, Charadrius, Gallinago, Philomachus and Vanellus Brisson

throughout the Old World, from Iceland to Zambia and from Kamchatka to Borneo

(Spasskaya and Spasskii 1978). The present study is the first record of A. tringae in an

intermediate host. Our identification is based on a comparison of the number, size and

the shape of rostellar hooks of the cysticercoids with those of the adult worms (syntypes

from Tringa glareola, British Museum (Natural History) Collection no. 1983.7.11.16).

The original description reported 18 rostellar hooks with a length of 17-18 μm “in what

is practically a single row, but they show a slight alteration in arrangement into two

levels but they are all similar in shape and size” (Burt 1940). Our observations show

that the rostellar hooks of the syntypes studied are 18 in number, 21-22 μm long and

have a similar shape (Fig. 7c, d) to those of the cysticercoids from Spain (Fig. 7a, b).

Family Progynotaeniidae

Gynandrotaenia stammeri Fuhrmann, 1936


Description: Cyst oval or lemon-shaped (Fig. 34), 142-192 x 95-170 (166 x 124, n =

17). Two layers distinguished in cyst wall: external layer hyaline, 9-11 thick; internal

layer finely striated, 5-7 thick. Calcareous corpuscles 70-80. Scolex 113-147 x 89-121

(133 x 103, n = 14). Suckers oval, with diameter 36-45 (39, n = 62); margins armed

with densely arranged spines (Figs. 35, 36). Invaginable anterior part of scolex (termed

‘proscolex’, see Fuhrmann 1936) distinct (Fig. 36), covered with small spines.

Rostellum 65-74 x 25-34 (70 x 29, n = 11). Rostellar hooks 6 (n = 17), 40-46 (43, n =

15) long (Figs. 8, 35, 36); length of blade 13-15 (13.5, n = 15). Ratio length of blade /

total hook length 0.28-0.35 (0.31, n = 15). Cercomer highly elongate, convoluted, much

larger than cyst (Fig. 33), 2.2-2.5 mm long and 47-78 wide (n=3).

Remarks: The morphology of these cysticercoids is in agreement with the previous

descriptions (Gvozdev and Maksimova 1979, Robert and Gabrion 1991). Adults of G.

stammeri have been recorded from Phoenicopterus roseus Pallas and P. minor Geoffroy

Saint-Hilaire in Europe (Fuhrmann 1936a, 1936b, Robert and Gabrion 1991),

Kazakhstan (Gvozdev and Maksimova 1979) and Kenya (Jones and Khalil 1980).


128

Cysticercoids were reported from Artemia from Kazakhstan (Gvozdev and Maksimova

1979) and France (Robert and Gabrion 1991).

Robert and Gabrion (1991) cited the paper by Gvozdev and Maksimova (1979)

as a record of cysticercoids of G. stammeri in an ostracod species. However, the only

intermediate host species mentioned in the latter article is Artemia.

DISCUSSION

According to the terminology proposed by Chervy (2002), the cysticercoids of

Flamingolepis liguloides, F. flamingo, Wardium stellorae and Gynandrotaenia

stammeri belong to the group of the cercocysticercoids, while those of Eurycestus

avoceti, Anomotaenia tringae and Anomotaenia sp. are considered monocysticercoids.

The cysticercoid of Confluaria podicipina is close to the modification termed

“ramicysticercoid” but its cercomer is not branching. This suggests the necessity of

further improvement of the terminology proposed by Chervy (2002).

The following 13 cyclophyllidean cestode species were previously known to use

brine shrimps of the genus Artemia as intermediate host in their life cycles:

Hymenolepididae (10 species): Confluaria podicipina (see Maksimova 1981, 1989),

Fimbriarioides tadornae Maksimova, 1976 (see Maksimova 1976, 1989),

Flamingolepis liguloides (synonym F. dolguschini) (see Maksimova 1973, 1989, Robert

and Gabrion 1991, Thiéry et al. 1990, Amat et al. 1991a, 1991b, Varó et al., 2000), F.

caroli (Parona, 1887) (see Robert and Gabrion 1991), F. flamingo (see Robert and

Gabrion 1991), F. tengizi (see Maksimova 1973, 1989), Hymenolepis californicus

Young, 1950 (see Young 1952), Wardium fusa (Krabbe, 1869) (see Maksimova 1987,

1989), W. gvozdevi Maksimova, 1988 (see Maksimova 1988, 1989) and W. stellorae

(Deblock, Biguet et Capron, 1960) (see Maksimova 1986, Robert and Gabrion 1991,

Varó et al., 2000).

Dilepididae (2 species): Eurycestus avoceti (see Gabrion and MacDonald 1980, Robert

and Gabrion 1991, Maksimova 1991) and Anomolepis averini (Spasskii et Yurpalova,

1967) (see Maksimova 1977).

Progynotaeniidae (1 species): Gynandrotaenia stammeri (see Gvozdev and Maksimova

1979, Robert and Gabrion 1991).


129

For 6 of these species, i.e., C. podicipina, F. liguloides, F. flamingo, W.

stellorae, E. avoceti and G. stammeri, our study confirms the role of brine shrimps as an

intermediate host in their life cycle. In addition, Anomotaenia tringae is recorded for the

first time in its intermediate host. Therefore, brine shrimps participate in the

transmission of the cestode parasites of flamingos, waders, grebes and gulls at the Odiel

Marshes.

Robert and Gabrion (1991) examined more than 64,000 brine shrimps in the

Camargue, France, and recorded a prevalence of cestode infection of between 1.56 and

5.12% in various seasons; the values of the mean abundance for the species in common

with these in our study are: F. liguloides 0.0506, F. flamingo 0.000743, W. stellorae

0.000077, G. stammeri 0.000031 and E. avoceti 0.00091. The data obtained for the

prevalence of cysticercoids in the Odiel Marshes (Table 1) are 6 times higher, and those

for the mean abundance are between 5 and 322 times higher. These differences suggest

much higher rates of cestode transmission compared to those recorded in the Camargue.

In contrast to the high infection rates in Artemia, no cysticercoids were found in any of

the other 5 invertebrate species studied from the same water body.

The present study is the first record of C. podicipina, E. avoceti, A. tringae and

G. stammeri in Spain.

ACKNOWLEDGEMENTS

We are grateful to Professor F. Amat (Instituto de Acuacultura de Torre de la Sal, CSIC,

Castellón, Spain), Professor V.V. Kornyushin (Institute of Zoology, National Academy

of Sciences, Kiev, Ukraine) and Dr V.V. Tkach (University of North Dakota, Grand

Forks, USA) for the stimulating comments and useful suggestions in the course of the

present study, and to Dr D.I. Gibson (The Natural History Museum, London) for

reading the manuscript. This study was carried out in the framework of a co-operative

project between the Bulgarian Academy of Sciences and the Consejo Superior de

Investigationes Cientificas with the title Ecological aspects of the role of brine shrimps

in the transmission of avian cestodes.


130

REFERENCES

Abatzopoulos T.J., Beardmore J.A., Clegg J.S. & Sorgeloos P. 2002. Artemia: Basic

and Applied Biology. Kluwer Academic Publishers, Dordrecht.

Amat F., Gozalbo A., Navarro J.C., Hontoria F. & Varó I. 1991a. Some aspects of

Artemia biology affected by cestode parasitism. Hydrobiologia, 212, 39-44.

Amat F., Hontoria F., Ruiz O., Green A.J., Sánchez M.I., Figuerola J. & Hortas F. In

press. The American brine shrimp Artemia franciscana as an exotic invasive species in

the Western Mediterranean. Biological Invasions. In press.

Amat F., Illescas M.P. & Fernandez J. 1991b. Brine shrimp Artemia parasitized by

Flamingolepis liguloides (Cestoda, Hymenolepididae) cysticercoids in Spanish

Mediterranean salterns. Quantitative aspects. Vie et Milieu, 41, 237-244.

Baer J.G. 1968. Eurycestus avoceti Clark 1954 (Cestode cylophyllidien) parasite de

l’avocette en Camargue. Vie et Milieu, 19, 189-198.

Belopol’skaya M.M. & Kulachkova V.G. 1973. Systematic position of Amoebotaenia

oophorae Belopolskaia 1971 (Cestoda: Cyclophyllidea). Parazitologiya, 7, 551-552. (In

Russian).

Bernués, M. (Ed.) 1998. Humedales Españoles inscritos en la Lista del Convenio de


Madrid.

Bondarenko S.K. & Kontrimavichus V.L. 2004a. On Branchiopodataenia n.g., parasitic

in gulls, and its type-species, B. anaticapicirra n. sp. (Cestoda: Hymenolepididae).

Systematic Parasitology, 57, 119-133.


131

Bondarenko S.K. & Kontrimavichus V.L. 2004b. Life-cycles of cestodes of the genus

Branchiopodataenia Bondarenko & Kontrimavichus, 2004 (Cestoda: Hymenolepididae)

from gulls in Chukotka. Systematic Parasitology, 57, 191-197.

Burt D.R.R. 1940. New species of cestodes from Charadriiformes, Ardeiformes, and

Pelecaniformes in Ceylon. Ceylon Journal of Science, Section B, Zoology and Geology,

22, 1-63.

Burt D.R.R. 1979. New cestodes of the genus Eurycestus Clark, 1954 from the avocet

Recurvirostra americana Gmelin, 1788. Zoological Journal of the Linnean Society, 65,

71-82.

Bush A.O., Lafferty K.D., Lotz J.M. & Shostak A.W. 1997. Parasitology meets ecology

on its own terms: Margolis et al. revisited. Journal of Parasitology, 83, 575-583.

Chervy L. 2002. The terminology of larval cestodes or metacestodes. Systematic

Parasitology, 52, 1-33.

Deblock S., Biguet J. & Capron A. 1960. Contribution a l’étude des cestodes de Lari

des côtes de France. I. Le genre Hymenolepis. Annales de Parasitologie Humaine et

Comparée, 35, 538-574.

Deblock S. & Rosé F. 1962. Liga celermatus – nouveau dilépididé de charadriiforme

des Côtes du Nord de la France. Bulletin de la Société Zoologique de France, 87, 600-

608.

Fuhrmann O. 1936a. Un singulier tenia d’oiseau Gynandrotaenia stammeri n. g. n. sp.

Annales de Parasitologie Humaine et Comparée, 14, 261-271.

Fuhrmann O. 1936b. Gynandrotaenia stammeri nov. gen. nov. spec. Revue Suisse de

Zoologie, 43, 517-518.


132

Gabrion C. & MacDonald G. 1980. Artemia sp. (Crustacé, Anostracé), hôte

intermédiaire d’Eurycestus avoceti Clark, 1954 (Cestode Cyclophyllide) parasite de

l’avocette en Camargue. Annales de Parasitologie Humaine et Comparée, 55, 327-331.

Gvozdev E.V. & Maksimova E.P. 1979. Morphology and developmental cycle of the

cestode Gynandrotaenia stammeri (Cestoidea: Cyclophyllidea) parasitic in flamingo.

Parazitologiya, 13, 56-60. (In Russian).

Jones A. & Khalil L.F. 1980. The helminth parasites of the lesser flamingo,

Phoeniconaias minor (Geoffroy), from lake Nakuru, Kenya, including a new cestode,

Phoenicolepis nakurensis n. g., n. sp. Systematic Parasitology, 2, 61-76.

Krabbe H. 1869. Bidtrag til Kundskab om Fuglenes Baendelorme. Kjøbehavn, B.L.

Bogtrykkeri.

Maksimova A.P. 1973. Branchiopod crustaceans as intermediate hosts of cestodes of the

family Hymenolepididae. Parazitologiya, 4, 349-352. (In Russian).

Maksimova A.P. 1976. A new cestode, Fimbriarioides tadornae sp. n., from Tadorna

tadorna and its development in the intermediate host. Parazitologiya, 10, 17-24. (In

Russian).

Maksimova A.P. 1977. Branchiopods as intermediate hosts of the cestode Anomolepis

averini (Spassky et Yurpalova, 1967) (Cestoda: Dilepididae). Parazitologiya, 11, 77-79.

(In Russian).

Maksimova A.P. 1981. Morphology and life cycle of the cestode Confluaria podicipina

(Cestoda: Hymenolepididae). Parazitologiya, 15, 325-331. (In Russian).

Maksimova A.P. 1986. On the morphology and biology of the cestode Wardium

stellorae (Cestoda: Hymenolepididae). Parazitologiya, 20, 487-491. (In Russian).

Maksimova A.P. 1987. On the morphology and the life cycle of the cestode Wardium

fusa (Cestoda: Hymenolepididae). Parazitologiya, 21, 157-159. (In Russian).


133

Maksimova A.P. 1988. A new cestode, Wardium gvozdevi sp. n. (Cestoda:

Hymenolepididae), and its biology. Folia Parasitologica, 35, 217-222.

Maksimova A.P. 1989. Hymenolepidid cestodes of aquatic birds in Kazakhstan.

Izdatel’stvo Nauka, Alma-Ata. (In Russian).

Maksimova A.P. 1991. On the ecology and biology of Eurycestus avoceti (Cestoda:

Dilepididae). Parazitologiya, 25, 73-76. (In Russian).

Robert F. & Gabrion C. 1991. Cestodoses de l’Avifaune Camarguaise. Rôle d’Artemia

(Crustacea, Anostraca) et stratégies de recontre hôte-parasite. Annales de Parasitologie

Humaine et Comparée, 66, 226-235.

Ryzhikov K.M., Ryšavý B., Khokhlova I.G., Tolkacheva L.M. & Kornyushin V.V.

1985. Helminths of fish-eating birds of the Palaearctic Region II. Cestoda and

Acanthocephales. Academia, Prague.

Sánchez M.I., Green A.J. & Castellanos E.M. In press. Spatial and temporal fluctuations

in use by shorebirds and in availability of chironomid prey in the Odiel Saltpans,

southwest Spain. Hydrobiologia.

Spasskaya L.P. & Spasskii A.A. 1978. Cestodes of birds in the USSR. Dilepididae of

aquatic birds. Izdatel’stvo Nauka, Moscow. (In Russian).

Thiéry A., Robert F. & Gabrion C. 1990. Distribution des populations d’Artemia et de

leur parasite Flamingolepis liguloides (Cestoda, Cyclophyllidea), dans les salins du

littoral méditerranéen français. Canadian Journal of Zoology, 68, 2199-2204.

Varó I., Taylor A.C., Navarro J.C., & Amat F. 2000. Effect of parasitism on respiration

rates of adults of different Artemia strains from Spain. Parasitology Research, 86, 772-

774.


134

Vasileva G.P., Georgiev B.B. & Genov T. 2000. Palaearctic species of the genus

Confluaria Ablasov (Cestoda, Hymenolepididae): redescriptions of C. podicipina

(Szymanski, 1905) and C. furcifera (Krabbe, 1869), description of C. pseudofurcifera n.

sp., a key and final comments. Systematic Parasitology, 45, 109-130.

Young R.T. 1952. The larva of Hymenolepis californicus in the brine shrimp (Artemia

salina). Journal of the Washington Academy of Sciences, 42, 385-388.


135

Table 1. Infection of brine shrimps with cysticercoids in the Odiel Marshes, Spain.

IntensityCestode species Prevalence(%) Range Mean ± SD

Meanabundance ±

SD

HymenolepididaeFlamingolepis liguloides 18.5 1-8 1.48 ± 0.89 0.28 ± 0.01Flamingolepis flamingo 0.9 1-2 1.03 ± 0.18 0.01 ± 0.10Confluaria podicipina 6.5 1-7 1.42 ± 0.82 0.09 ± 0.40Wardium stellorae 0.2 1-1 1.00 ± 0.00 0.002 ± 0.043

DilepididaeEurycestus avoceti 2.7 1-3 1.08 ± 0.31 0.03 ± 0.18Anomotaenia sp. (cf. A.microphallos)

0.8 1-2 1.04 ± 0.19 0.01 ± 0.10

Anomotaenia tringae 2.2 1-2 1.01 ± 0.12 0.02 ± 0.15

ProgynotaeniidaeGynandrotaenia stammeri 0.6 1-1 1.00 ± 0.00 0.01 ± 0.08

Cysticercoids, total 26.8 1-13 1.71 ± 1.28 0.44 ± 0.99

Table 2. Multiple species infections of brine shrimps with cysticercoids in the Odiel Marshes, Spain.

Number of cestode species participating in the multiple infectionCestode species2 3 4 5

Flamingolepis liguloides + + + + + + + + + + + + + + + + + + +Flamingolepis flamingo + + + + + + + + + + +Confluaria podicipina + + + + + + + + + + + + + + + +Wardium stellorae + + +Eurycestus avoceti + + + + + + + + + +Anomotaenia sp. (cf. A.microphallos)

+ + + + + +

Anomotaenia tringae + + + + + + + +Gynandrotaenia stammeri + + + + + + +Number of specimens ofArtemia infected by thiscombination of species

10 41 29 3 7 8 1 1 1 1 1 2 3 3 19 2 1 2 5 2 1 1 7 8 3 4 1 1 1 1 1

137

Table 3. Morphological characteristics of the scolex armament of adult cestodes and cysticercoids of Eurycestus avoceti and species with similarscoleces.

Rostellar hooks Sucker hookletsLength

Species Locality SourceNumber

Anterior PosteriorNumber Length

Eurycestus avoceti adult cestodes France Baer (1968) 14-16 14-16* 14-16* 10-14 5-6

Kazakhstan Maksimova (1991) 16 16-18 10-12 30-32 5-6 cysticercoids France Gabrion & Mac Donald (1980) 16 18 12 15-16 7

Kazakhstan Maksimova (1991) 16 16-18 10-12 30-32 4-5Spain Present study 14-16 16-18 15-16 9-17 6-8

Paraliga oophorae adult cestodes Russia

(White Sea)Belopol’skaya & Kulachkova(1973)

16 18 14 34-37 4

Paraliga celermatus adult cestodes France Deblock & Rosé (1962) 16 24-26* 24-26* 25-30 4-5

*Lengths of anterior rostellar hooks and length of posterior rostellar hooks not given separately.


138

Figs 1-8. Rostellar hooks and sucker hooklets of the cysticercoids found. 1.

Flamingolepis liguloides. 2. F. flamingo. 3. Confluaria podicipina. 4. Wardium

stellorae. 5. Eurycestus avoceti – a, anterior hook; b, posterior hook; c, sucker hooklets.

6. Anomotaenia sp., cf A. microphallos – a, anterior hook; b, posterior hooks. 7. A.

tringae. – a, b, rostellar hooks of a metacestode from a brine shrimp; c, d, rostellar

hooks of a syntype specimen from Tringa glareola from Sri Lanka. 8. Gynandrotaenia

stammeri.

20 µm

1

50µm

8

5a 5b 5c

6a 6b 7a 7b 7c 7d

2

43


139

Figs 9-14. Flamingolepis spp. 9-11. F. liguloides. 9. Metacestode in situ. 10. General

view of isolated metacestode, phase contrast. 11. Rostellar hooks. 12-14. F. flamingo.

12. General view of isolated metacestode (C, cercomer). 13. Cyst of metacestode, phase

contrast. 14. Rostellar hooks, phase contrast.

200 µm 200 µm

200 µm100 µm

100 µm 50 µm

9 10

12

1413

11

C

C


140

Figs 15-20. Confluaria podicipina and Wardium stellorae. 15-19. C. podicipina. 15.

Metacestode in situ; compare the size with that of the metacestode of F. liguloides (C,

cercomer). 16. Highly coiled and densely packed cercomer (C) in the body cavity of the

host. Inset: Portion of the cercomer, phase contrast. Note the thin membranous envelope

(arrow). 17. Isolated metacestode. 18. Isolated metacestode under hypotonic conditions.

Note the anterior orifice of the external capsule (arrow). 19. Metacestode, mounted in

Berlese’s medium. 20. W. stellorae, general view of the metacestode, mounted in

Berlese’s medium, phase contrast. Inset: Rostellar hooks, phase contrast.

200 µm 200 µm

100 µm

50 µm 100 µm

15 16

18

2019

17 200 µm

100 µm

25 µm

C

C

C


141

Figs 21-26. Eurycestus avoceti. 21. Metacestode in situ. 22. Isolated metacestode with

outer capsule breaking down into cercomer fragments. 23. Isolated cyst. 24. Anterior

end of isolated cyst. Note protruding rhynchus and sucker hooklets. 25. Isolated

metacestode with entirely protruded rhynchus. Inset: Rostellum. 26. Metacestode

excysted under hypotonic conditions. Inset: Rostellar hooks, phase contrast.

200 µm21

100 µm23

200 µm25 26

50 µm24

200 µm22

50 µm

200 µm

50 µm


142

Figs 27-32. Anomotaenia spp. 27-30. Anomotaenia sp., cf. A. microphallos. 27.

Metacestode in situ; compare the size with that of the metacestode of F. liguloides. 28.

Isolated and slightly flattened metacestode. 29. Isolated cyst. Inset: Rostellar hooks,

phase contrast. 30. Anterior part of isolated cyst. 31-32. A. tringae. 31. Isolated

metacestode. 32. Metacestode squashed in Berlese’s medium; note the granular contents

of the external capsule. Inset: Rostellar hooks, phase contrast.

200 µm27

100 µm29

200 µm31 32

50 µm30

200 µm28

100 µm

25 µm

25 µm


143

Figs 33-36. Gynandrotaenia stammeri. 33. Metacestode, general view (C, cercomer).

34. Cyst. 35. Armament of rostellum and suckers, mounted in Berlese’s medium. 36.

Cyst, mounted in Berlese’s medium, phase contrast. Note invaginated proscolex

(arrows).

200 µm 100 µm

50 µm50 µm

33

3635

34

144

CHAPTER 6 PROPAGULE DISPERSAL BY WADERS

145

CHAPTER 6

PASSIVE INTERNAL TRANSPORT OF BRINE SHRIMPS AND SEEDS BY

MIGRATORY WADERS IN THE ODIEL MARSHES, SOUTH-WEST SPAIN

MARTA I. SÁNCHEZ1,2, ANDY J. GREEN1 , FRANCISCO AMAT3 & ELOY M.

CASTELLANOS2



Universidad de Huelva, Campus de El Carmen, Avda. Fuerzas Armadas s/n 21071 Huelva, Spain3Instituto de Acuicultura de Torre de la Sal, 12595 Ribera de Cabanes, Spain

Key words: Brine shrimp dispersal, seeds dispersal, seed germination, cysts eclosion,

waders

Running head: Propagule dispersal by waders

Finally excinded in two:

“Internal transport of brine shrimps by migratory waders: dispersal probabilities depend on diet

and season”. Marine Biology 151: 1407-1415.

“Internal transport of seeds by migratory waders in the Odiel marshes, south-west Spain:

consequences for long-distance dispersal”. Journal of Avian Biology 37: 201-206.


146

ABSTRACT

Waders (Charadriiformes) undergo particularly long migratory flights, making them

ideal vectors for long-distance dispersal. We present a study of dispersal of plant seeds

and brine shrimp Artemia cysts by migratory waders in the Odiel saltworks in south-

west Spain. Viable cysts of Artemia parthenogenetica were abundant in the faeces and

pellets of Redshank Tringa totanus, Spotted Redshank Tringa erythropus, and Black-

tailed Godwit Limosa limosa during spring and autumn migrations, but were absent

during winter. Most cysts were ingested within Artemia adults. The proportion of cysts

destroyed during digestion decreased as the total number ingested increased, and

increased when accompanied by harder food items. More intact cysts were present in

Redshank faeces than in their pellets, but cysts extracted from pellets were more likely

to hatch. Viable seeds of Mesembryanthemum nodiflorum (Aizoaceae), Sonchus

oleraceus (Asteraceae) and Arthrocnemum macrostachyum (Chenopodiaceae) were

frequent in pellets and faeces, with 11 other seed types recorded at low density. This is

the first field study to demonstrate transport of viable seeds by waders. More intact M.

nodiflorum seeds were present in Redshank faeces than in their pellets, but seeds

extracted from pellets were more likely to germinate. More cysts and S. oleraceus seeds

were transported per Redshank pellet in spring, but more Redshank migrated through

the area in autumn. The distribution of the plants and Artemia transported are consistent

with an important role of long-distance dispersal by waders in their biogeography and

metapopulation biology. M. nodiflorum and S. oleraceus are introduced weeds in the

Americas and Australasia, and dispersal by birds may contribute to their rapid spread.

Although S. oleraceus is generally thought to be wind-dispersed, birds may be

responsible for longer distance dispersal events.


147

INTRODUCTION

The Charadriiformes (waders, gulls and terns) include species undergoing the longest

migrations of any animal (del Hoyo et al. 1996). Darwin (1859) proposed that migratory

waterbirds played a major role in the dispersal of aquatic invertebrates and plant seeds,

and waders are likely to be particularly important owing to their long-distance

migrations. Seeds lacking a fleshy fruit are often consumed by waders (see Green et al.

2002 for review) and studies in captivity suggest they can retain viability following gut

passage (see Figuerola & Green 2002a for review). Proctor (1962) demonstrated

transport of viable Chara oospores by waders in Texas. However, as far as we know,

dispersal of viable angiosperm seeds by waders has never been demonstrated in the

field. Field evidence of dispersal of invertebrates by waterbirds is also very limited, but

Green et al. (2005) recently showed that viable brine shrimp Artemia (Anostraca) cysts

(resistant eggs) are dispersed by waders on autumn migration.

In this study we document passive internal dispersal of seeds and Artemia cysts

by waders migrating through the Odiel marshes in south-west Spain, quantifying

numbers of intact propagules and assessing their viability. We quantify levels of

dispersal by different wader species at different times of the year, comparing the

potential for long-distance dispersal during spring and autumn migrations. We

investigate the role of diet and other factors in determining dispersal rates, especially the

rates of ingestion of propagules and their survival during the digestive process.

METHODS

A total of 140 faecal samples and 272 pellets were collected from 2001 to 2003 from

waders in the saltworks of the Odiel marshes in Huelva province in south-west Spain

(37°17'N 06°55'W, area 7158 ha, Fig. 1), a site of international importance for

migratory waders migrating through the East Atlantic flyway (Sánchez et al. in press a,

Stroud et al. 2004). For the purpose of seasonal comparison, sampling dates of 18/3/01,

23/4/01 and 7/4/03 were considered as spring, from 22 July to 21 August as summer,

from 5 September to 18 October as autumn and 13/1/01 plus 28/2/01as winter. Spring

samples coincided with the beginning of the northwards wader migration and summer

and autumn samples with the southwards migration (Hortas 1997, Sánchez et al. in


148

press a). Pellets and faeces were collected from Redshank Tringa totanus, pellets from

Spotted Redshank Tringa erythropus, and faeces from the Black-tailed Godwit Limosa

limosa, a species that does not regurgitate pellets. These species were selected because

of their abundance in the study area and because their faecal and pellet samples are

relatively large, making it easier to avoid soil contamination. Redshank and Black-tailed

godwit are two of the four most numerous waders at this site (Sánchez et al. in press a).

We collected fresh faeces and pellets from roost sites on dykes in traditional

saltworks used by monospecific flocks at high tide. Collection points were observed for

30 minutes prior to sampling and only fresh samples were taken to ensure accurate

species identification. In spring 2003, 52 pellet samples were collected from a mixed

flock of the two Tringa species and were only used for quantification of seed transport.

Each sample of excreta was carefully separated from the soil (discarding that part in

contact with soil) and placed in a tube. Given the number of birds present we are

confident that each sample was from a different bird.

The samples were stored at 5oC, until the time of analysis (a lapse of one week

for samples collected in 2003, and of six to 12 months for the rest). Seeds and Artemia

cysts were extracted by washing them in a 0.04 mm sieve and flotation in hypersaline

brine. The seeds and cysts were counted, washed in distilled water then dried for 48 h at

40oC, followed by storage at 5oC prior to hatching or germination experiments. The

composition of the rest of the faecal and pellet samples was also identified, assessing

the proportion of sample volume made up of different food items using the following

categories: 0%, <10%, 11-25%, 26-50%, 51-75%, 76-90% and >90% of total volume

(see Sánchez et al. in press b for details).

Hatching and germination experiments

Not all samples were stored so as to retain their viability, so that hatching or

germination was tested for a subset of samples. Cysts were then incubated in diluted

filtered sea water (25 g L -1) at 26ºC and under continuous illumination for at least 48 h.

Hatched nauplii were counted and transferred into 60 cm3 vessels and cultured in 70 g L-1 filtered brine (seawater plus crude sea salt), on a diet of live Dunaliella salina and

Tetraselmis suecica. They were maintained at 24ºC, under aeration on a 12D:12L

photoperiod. The medium was monitored and renewed every two days. Resulting adults

were identified to species morphologically (Amat et al. 2004).


149

Extracted seeds were placed on moist filter paper in a petri dish and watered

regularly with distilled water. They were placed inside a WTBbinder KBW 400

germination chamber at 22 oC on a 12D:12L photoperiod, on 2/10/03. They were checked

daily for germination for 30 days. One sample of 84 seeds from a Redshank pellet was

attacked by mould and removed from the experiment. Some seedlings were cultivated to

enable species identification.


Owing to the high proportion of zeros in the sample and severe problems of over-

dispersion and model convergence, we were unable to conduct satisfactory parametric

analyses of the numbers of intact cysts or seeds in samples in relation to differences

between season, wader species or sample type (pellet or faeces). We therefore used non-

parametric Kruskal-Wallis and Mann-Whitney U tests employing Statistica 5.5 (StatSoft

1999).

We used generalized linear models (GLMs) following the GENMOD procedure

in SAS (v. 8.2, SAS Institute 2000) to analyze the effects of season, wader species,

sample type (faeces or pellet) and diet on the percentage of cysts in a sample that were

broken, and the percentage of intact cysts that hatched. All these variables were

included as fixed factors in the models. Diet was included as a factor of three levels

according to whether the majority of the sample was made up of soft items (e.g.

Chironomid larvae or Artemia), hard items (e.g. bivalve shells or grit) or items of

intermediate hardness (e.g. Coleoptera, Ephydridae larvae or polychaetes). We used a

normal error distribution and identity link, and a square root transformation for the

dependent variable to overcome heteroscedasticity. There were insufficient data to

conduct GLMs of the percentage of broken seeds, or of the percentage of intact seeds

that germinated. In order to test for a relationship between the proportion of cysts

broken or hatching and the total number of cysts in a sample, we conducted GLMs in

which the total number of cysts (log transformed) was included as an additional

predictor variable. The deviance of each fitted GLM model is analogous to the residual

sum of squares in ordinary linear regression. The reduction in deviance compared to the

null model is used to assess the contribution of the model to the explanation of the

variance in the data set. The significance of the reduction in deviance for the models of

cyst breakage or hatching was derived from F tests, scaled to control for over-dispersion

(Crawley 1993).


150

Artemia cysts ingested can potentially be those present in the sediments, those in

the water column, those forming floating masess on the surface or those situated in the

ovisac within female adults. In the latter case, a significant correlation would be

expected between the abundance of adult Artemia and of cysts in different samples. In

order to test for this, we calculated Spearman rank correlations between the abundance

of Artemia adults (using the seven volumetric categories defined above) and the total

number of cysts recorded in Redshank and godwit samples.

Wader counts and migration routes

In 12 different salt ponds representing 27.6 % of the 1174 ha of saltworks, we counted

the number of shorebirds of each species that were feeding and resting one day each

week throughout 2001, using a 20-60x telescope. On each day, we carried out a count of

three hours duration around high tide when the densities of waders are highest (Sánchez

et al. in press a). The ponds were always counted at about the same time of day and

always following the same route between ponds.

Since 1992 there has been an intensive ringing operation at the Odiel saltworks

coinciding with the autumn wader migration. A smaller number of birds have also been

ringed in a second locality in Huelva province, Doñana National Park (37°07'N

06°28'W) found 50 km from Odiel. We present details of recoveries to shed light on the

likely direction and distance of long-distance dispersal events.

RESULTS

Artemia cysts

Intact Artemia cysts were recorded in the droppings and/or pellets of all three wader

species and in all seasons except winter (Table 1). Cysts were most abundant in godwit

faeces in summer 2001, with an average of 222 cysts per dropping.. They were also very

abundant in Redshank pellet in autumn, with an average of 189 cysts per dropping . A

mean of 31 cysts were recorded in Spotted Redshank pellets in spring (Table 1).

Overall, 85.3 % of cysts (n = 57729) recorded in samples were intact.

For Redshank pellets, there were highly significant differences between seasons

in the number of intact cysts per sample (Kruskall-Wallis test, n = 211, H = 44.16, df =

2, p < 0.0001). Post-hoc tests showed that the number of cysts in spring was


151

significantly higher than in autumn (Mann-Whitney U test, U = 1247, p < 0.0001) and

in winter (U = 94.5, p < 0.0001) and higher in autumn than in winter (U = 1302, p =

0.021). The real difference between spring and autumn was in the proportion of pellets

containing cysts (χ21 = 34.7, P < 0.0001). For Redshank pellet samples with at least one

intact cyst, there was no difference in the number of cysts between spring (median =

5.5) and autumn (median = 3, U = 392.0, p = 0.95).

Amongst Redshank samples from autumn, the median number of intact cysts

was significantly higher in faecal samples than in pellets (U = 4997, p = 0.0001). In

contrast, the maximum and mean were higher in pellets (Table 1). Amongst spring

pellets, numbers transported by Redshank were higher than those transported by Spotted

Redshank, and the difference was marginally significant (U = 85.00, p = 0.051)

In an analysis of the proportion of cysts that were destroyed during gut passage,

there were no significant effects of season or sample type (Table 2). However, diet had

a highly significant effect, the proportion of cysts destroyed increasing with the

hardness of other food items in the sample (Table 2). With the proportion of cysts

destroyed (arcsine transformed) as the dependent variable and species, season, sample

type (pellet or faeces) and total number of cysts (log transformed) as predictor variables,

there was a negative and significant correlation between the number of cysts and the

proportion destroyed (F1,169 = 19.12, r = -0.35, P < 0.0001).

For both faecal and pellet samples, there was a strong relationship between the

total number of cysts recorded in each sample and the proportion of the volume of that

sample made up by Artemia adults (godwit faeces, rs = 0.783, P < 0.0001, n = 56;

Redshank faeces, rs = 0.424, P < 0.0001, n = 84). for Redshank pellets, rs = 0.237, P <

0.0001, n = 211).

The viability of intact cysts was determined for a smaller number of samples.

Viability was confirmed for samples from all three wader species, and for samples

collected in spring, summer or autumn (Table 3). Viability was highest for Redshank

pellets in autumn, with an average of 19% of cysts in each sample hatching (Table 3).

There were no significant differences between wader species or seasons in the

proportion of cysts hatching, but there was a significant effect of sample type, with a

higher proportion of cysts hatching in Redshank pellets than in Redshank droppings

(Table 2). The partial effect of sample type remained significant when diet was added to

the model of Table 2 (F 1, 65 = 4.84, P = 0.031 ), but diet had no influence on the

proportion of cysts hatching (F2, 65 = 0.79, P = 0.46). Likewise, for godwit droppings


152

there was no correlation between the proportion of cysts hatching and diet hardness (n =

41, rs = -0.255, P = 0.11).

With the proportion of cysts hatching as the dependent variable and species,

season, sample type and total number of cysts as predictor variables, there was a

negative but non-significant correlation between the number of cysts and the proportion

hatching (F1, 67 = 0.83, r =-0.11, P = 0.37).

Seeds

Intact angiosperm seeds were recorded in the droppings and/or pellets of all three wader

species and in all seasons (Table 4). Overall, they were recorded in 21.8 % of samples

with 3.19 ± 0.94 intact seeds per sample (mean ± s.e. for those samples with at least one

seed, n = 90). In total, 74.2 % of all seeds recorded were intact (Table 4).

A total of 14 seed species were recorded. The Slender-leaved Iceplant

Mesembryanthemum nodiflorum L. (Aizoaceae) was particularly abundant in Redshank

faeces in autumn. The Common Sowthistle Sonchus oleraceus L. (Asteraceae) was

particularly abundant in spring pellets of Redshank and Spotted Redshank.

Arthrocnemum macrostachyum (Moric.) K.Koch (Chenopodiaceae) was the only

species recorded in samples from all wader species (Table 4). All three plant species are

abundant on the dykes of Odiel salt pans. Other intact seeds recorded were Suaeda spp.

(1 seed), Salicornia spp. (2 seeds) and 9 unidentified taxa (one seed of each).

The maximum number of intact seeds in one sample was 83 S. oleraceus seeds

in a Redshank pellet collected on 18/3/01 (Table 4). There were highly significant

differences between seasons in the abundance of intact S. oleraceus seeds in Redshank

pellets (Kruskal-Wallis test, H = 68.22, 2df, n = 211, p < 0.0001). They were only

recorded in spring, when they were significantly more abundant than autumn (Mann-

Whitney U test, U = 1648.5, p < 0.0001) or winter (U = 220.5, p = 0.0021). A similar

and marginally significant seasonal trend was recorded in the abundance of intact A.

macrostachyum seeds (H = 5.82, 2 df, p = 0.055). In autumn, intact M. nodiflorum seeds

were significantly more abundant in Redshank faeces than in their pellets (Mann-

Whitney U test, U = 5654, p < 0.0001). There were no significant differences in the

abundances of intact seeds of the three species between spring pellets of Redshank or

Spotted Redshank (Mann-Whitney U tests, P > 0.5).

The proportion of all seeds that were intact was 41.0 % for A. macrostachyum,

54.3 % for M. nodiflorum and 84.3 % for S. oleraceus (Table 4). There were insufficient


153

data to allow multivariate analyses of differences in the proportion of seeds broken

between wader species or seasons, or of the relationship between the numbers of seeds

in a given sample and the proportion broken.

Overall, 45.5 % of A. macrostachyum, 23.5 % of M. nodiflorum and 76.3 % of S.

oleraceus seeds germinated (Table 5). One unidentified seed also germinated. Amongst

autumn Redshank samples, M. nodiflorum seeds from pellets were significantly more

likely to germinate than those from faeces (Fisher exact test, p= 0.0385, selecting one

seed at random from three samples with more than one seed to avoid pseudoreplication).

Wader abundance and migration routes

Wader species differed in their phenology during migration through the Odiel marshes

(Fig. 2). Redshank had a smaller peak during spring migration in April, and a larger one

between July and October during autumn migration. Spotted Redshank numbers peaked

in October with a lower peak in April. Godwit peaked from July to August at the

beginning of autumn migration.

Ringing recoveries show that Redshank and godwit stopping at the Odiel

marshes share a common flyway and come from as far north as Sweden (Fig. 1). Colour

ringed birds from the Icelandic breeding population of Black-tailed godwit have also

been recorded at Odiel (T. Gunnarsson pers. comm.). There were 70 recoveries of

Redshank ringed at Odiel, 60 from the Odiel marshes, 5 from the Netherlands, 1 from

Belgium, 3 from France and one from Sweden. There were three recoveries of godwits

from the Netherlands. From birds ringed elsewhere in Huelva province, there were six

recoveries of Redshank, three from France, one from Portugal, one from Germany and

one from the Netherlands. There were five recoveries of godwits, one from Germany,

three from the Netherlands and one from Seville. There were no recoveries of Spotted

Redshank.

DISCUSSION

We have found Black-tailed Godwits, Redshank and Spotted Redshank to be effective

dispersers of A. parthenogenetica and three plant species at the Odiel marshes, where

thousands of waders pass through on migration, especially in autumn (Fig. 2) when on

their way to winter in West Africa (Wetlands International 2002, Stroud et al. 2004). The


154

East Atlantic Flyway, used by an estimated 15.5 million shorebirds (including 440,000

Redshank, 200,000 Black-tailed Godwit, and 50,000 Spotted Redshank, Stroud et al.

2004), is likely to be a major means of long-distance dispersal of plants and

invertebrates. Artemia cysts and seeds are also likely to be transported internally by some

of the other 24 migratory shorebird species using the East Atlantic Flyway (Stroud et al.

2004), as well as by other birds that use salt pans (Green et al. 2002). Viable cysts and

seeds are also likely to be transported externally on the feathers and feet of waterbirds

(Figuerola & Green 2002b).

The birds we studied included a mixture of birds that had just completed or were

just about to commence long-distance movements, and birds that were making

movements between feeding and roosting sites within our study sites. The average time

spent by shorebirds at stopover sites during migration varies from one to 50 days and is

unknown for sites in south-west Iberia (Hortas 1997). One of two icelandic godwits

seen at Odiel on autumn migration was still there 36 days later (T. G. Gunnarsson, J. A.

Gill & P. M. Potts, unpublished data). Most of the propagules transported in faeces or

pellets by waders at Odiel are likely to be transported short distances of less than 10 km,

and it is currently impossible to assess what proportion would undergo long-distance

dispersal. Godwits and other shorebirds regularly move between saltpans and other

wetlands separated by up to 20 km while at stopover sites (Farmer & Parent 1997, P.M.

Potts, pers. comm.), thus facilitating propagule dispersal between different parts of a

wetland complex. Most major passage and wintering sites on the East Atlantic Flyway

contain saltmarshes and/or saltworks, facilitating the long distance dispersal of Artemia

and plants.

Shorebirds move rapidly between coastal wetlands, and distances between

stopovers can easily exceed 1000 km (Iverson et al. 1996, Pennycuick & Battley 2003).

Black-tailed Godwits, Redshank and other waders are thought to fly non-stop between

Odiel and northern Europe (Beintema & Drost 1986). Redshank and godwits fly at 56-60

km h-1 (Welham 1994). In a laboratory study, the modal retention time (much less than

the mean) of viable Artemia cysts defecated by another shorebird, the Killdeer

Charadrius vociferus (size intermediate between redshank and godwit), was 90 min,

and the maximum was 26 h (Proctor et al. 1967). This suggests that, during migration,

the maximum dispersal distance of viable cysts would be c.1500 km. Laboratory studies

of gut passage of intact seeds suggests that they may be transported even longer

distances. In a study of 13 seed and three wader species, maximum retention time varied


155

from 3 h to 14 days (Proctor 1968). There is evidence that species such as godwits that

do not regurgitate compacted pellets from the gizzard can regurgitate seeds on their own

from the gizzard after extraordinary intervals (Proctor 1968). The modal retention time

of Amaranthus and Sagittaria seeds defecated by Killdeer was found to be 1-2 h

(deVlaming & Proctor 1968).

In a previous study at Cadiz Bay and Castro Marim in the south-west of the

Iberian peninsula during autumn migration, Redshank were shown to disperse A.

parthenogenetica, and Black-tailed Godwits and Redshank were both shown to be

effective dispersers of the alien A. franciscana (Green et al. 2005). However, ours is the

first study to demonstrate Artemia dispersal by waterbirds during spring migration.

Although the number of Artemia cysts per sample was higher at Odiel, their viability was

lower than at Cadiz Bay and Castro Marim. This is probably a consequence of the longer

storage time of our samples between collection in the field and hatching (up to 15 months)

which is likely to have reduced viability. A noteable contradiction between these two

studies is that Green et al. (2005) found that the proportion of cysts hatching was higher

for Redshank faeces than for pellets, whereas we found the opposite. Both results were

weakly significant, illustrating the noise and variation in space and time that can be

observed when studying dispersal in the field. Cysts in pellets do not necessarily have the

same origin as those in faeces (e.g. cysts consumed within female Artemia and those

ingested separately may be processed differently), and this may potentially explain

differences in viability. Whereas the mode of ingestion of Artemia cysts has previously

been open to speculation (Green et al. 2005), we found strong evidence that most cysts

recorded in faeces or pellets were ingested within adult females. The absence of cysts

from winter pellets is consistent with such a pattern. Cyst production peaks in spring

and summer and Artemia adults almost disappear in winter, although cysts remain

reasonably abundant in the water column and in the sediments (Martínez 1989, authors

unpublished data). Studies in captivity (Horne 1966, Charalambidou et al. in press)

suggest that gut passage itself does not affect the viability of those A. franciscana cysts that

survive intact, but their viability was reduced by gut passage in flamingos when they were

mixed with sand (MacDonald 1980).

It has often been suggested that dietary variation is likely to influence the

survival rates of propagules ingested by different bird species or individuals (e.g.

Figuerola & Green 2002a), but ours is the first field study to show that diet influences

the survival rate of ingested invertebrate propagules. We found that Artemia cyst


156

survival was directly related to the hardness of food items, but other aspects are also

likely to be important. Charalambidou et al. (in press) found that survival of A.

franciscana cysts following gut passage through Mallards Anas platyrhnchos was lower

for birds fed on animal-based pellets than those fed on grains, and suggested that this was

due to higher concentrations of enzymes for digesting animal matter. We also found that,

as the number of cysts in a sample increased, the proportion that were destroyed

decreased. This indicates a positive relationship between the quantity and quality

components (sensu Schupp 1993) of dispersal in this system, with individuals that

consummed more cysts expelling a higher proportion of viable cysts. Such a

relationship has also been reported for seed dispersal by waterfowl (Figuerola et al.

2002a).

We found that M. nodiflorum seeds and Artemia cysts were more abundant in

faeces than in pellets. This is likely to further long distance dispersal, as propagules are

likely to be retained longer when expelled in faeces rather than in pellets (Nogales et al.

2001). In contrast, M. nodiflorum seeds extracted from pellets were more likely to

germinate. This is evidence that longer retention times involved in passage through the

intestines reduces viability. However, Nogales et al. (2001) found no difference in the

viability of seeds extracted from pellets or droppings of gulls.

Ours is the first field study to demonstrate transport of viable angiosperm seeds

by waders. Overall, we recorded seeds (whether intact or not) in 22% of our samples.

This is lower than the frequency of seeds recorded in faeces or pellets for most species

of waders wintering in Cadiz Bay (Pérez-Hurtado et al. 1997). The abundance of seeds

recorded in many other studies of wader diet (Green et al. 2002, Montaltil et al. 2003)

suggests that waders play a widespread and important role in the dispersal of many plant

species. Darwin (1859) recognised that dispersal of propagules by migratory birds such

as waders plays a vital role in explaining the distribution of species. However, he

underestimated the capacity of seeds lacking a fleshy fruit to survive passage through

the digestive tract, and imagined that long-distance dispersal by internal transport may

require a role for birds of prey in extracting seeds from a bird’s crop before they enter

the gizzard. Our study shows that waders can disperse seeds over great distances

without the need for such rare predation events.

The native distribution of dispersed species included in this study

(Triantaphyliidis et al. 1998, Tutin et al. 2001) is consistent with a role of long-distance

dispersal by waders to and from sites such as the Odiel marshes. S. oleraceus seeds


157

were transported more readily in spring, coinciding with the timing of seed production

(Valdés et al. 1987). Taking into account the frequencies of viable propagules recorded

in different seasons, and the greater numbers of birds moving through the Odiel

saltworks at different times, our data suggest that there should be greater long distance

dispersal of S. oleraceus seeds northwards from Odiel, and greater dispersal of M.

nodiflorum and A. macrostachyum seeds in a southerly direction. S. oleraceus is

widespread across the range of ringing recoveries (Fig. 1) as far north as Sweden. It has

also been recorded in Iceland where it is considered an alien (A. Gardarsson, pers.

comm.), and is native down to southern Africa which is visited by some Redshank and

other waders breeding in Europe (Wetlands International 2002). M. nodiflorum is

considered a Mediterranean species but is also native in southern Africa. A.

macrostachyum is native across the Mediterranean and down to Mauritania, a major

wintering ground for waders passing through Odiel (Stroud et al. 2004).

Parthenogenetic brine shrimp populations are native down to South Africa and

up to North-west France (Triantaphyliidis et al. 1998). It is difficult to predict what

directionality there may be in effective dispersal of A. parthenogenetica cysts by

waders, as whilst they were present in a higher proportion of Redshank pellets in spring

(76%) than in autumn (30%), there were many more Redshank and godwits passing

through Odiel on autumn migration. Godwit samples from the beginning of the

southwards migration period had the most cysts. Stopover times and the probabilities

that waders move between suitable habitats for the dispersed species are likely to differ

between spring and autumn migration, and to vary between wader species, but there are

currently no data on this issue. Furthermore, there are some wader species at Odiel

(Dunlin Calidris alpina and Little stint Calidris minuta) that are much more abundant

during spring than autumn migration, and it is possible that these species are also good

dispersers of plants or Artemia.

S. oleraceus and M. nodiflorum are also widespread invasive exotics in the

Americas, Australia and in Pacific Islands (Villaseñor et al. 2004,

http://plants.usda.gov). Our study suggests that dispersal by waders and other migratory

birds may play a previously unrecognised role in the spread of these invasive weeds. S.

oleraceus seeds have been recorded in bird droppings in the introduced range (McGrath

& Bass 1999). However, S. oleraceus is generally assumed to be a wind-dispersed

species (Jakobsson & Eriksson 2003). Our study support the proposal of Higgins et al.


158

(2003) that birds can sometimes be the main means of long-distance dispersal for seeds

that appear morphologically adapted for wind dispersal.

In summary dispersal via waders is likely to have a significant role in the

distribution and metapopulation biology of plants and aquatic invertebrates. More

detailed studies are required before the true significance of this dispersal can be

established and compared to that of other vectors such as wind, water or man. In

particular, more data are required on the role of wader species not included in this study,

and on the probabilities that propagules are dispersed to suitable habitats (see Nathan et

al. 2003, Green & Figuerola 2005 for further discussion).

ACKNOWLEDGEMENTS


Tecnología and an I3P postgraduate grant from the Consejo Superior de Investigaciones

Científicas. Consejería de Medio Ambiente, Junta de Andalucía and Aragonesas

Industrias y Energía S.A. provided permission to work in the salines. J.C. Rubio

provided logistical support and advice. P. García Murillo helped to identify the plants.

F. Palomares helped with field work. J. Elmberg helped with a literature search. J.

Figuerola provided valuable comments on the manuscript. Ringing data were provided

by the Oficina de Especies Migratorias, Ministerio de Medio Ambiente.

REFERENCES

Amat, F., Hontoria, F., Ruiz, O., Green, A. J., Sánchez, M. I., Figuerola, J. & Hortas, F.



Beintema, A.J. & Drost, N. 1986. Migration of the Black-tailed Godwit. Gerfaut 76: 37-

62.


159

Charalambidou, I., Santamaría, L., Jansen, C. & Nolet, B. Digestive plasticity in ducks

modulates dispersal probabilities of aquatic plants and crustaceans. Functional Ecology.

In press

Crawley M.J. 1993. GLIM for ecologists, Blackwell Scientific Publications, Cambridge.

Darwin, C. 1859. On the origin of species by means of natural selection. John Murray.

del Hoyo, J., Elliott, A. & Sargatal, J., eds. 1996. Handbook of the Birds of the World

vol. 3. Lynx Edicions, Barcelona.

Figuerola, J. & Green, A. J. 2002a. Dispersal of aquatic organisms by waterbirds: a


494.

Figuerola, J. & Green, A. J. 2002b. How frequent is external transport of seeds and

invertebrate eggs by waterbirds? A study in Doñana, SW Spain. Archiv für

Hydrobiologie 155: 557-565.

Figuerola, J., Green, A. J. & Santamaría, L. 2002. Comparative dispersal effectiveness

of wigeongrass seeds by waterfowl wintering in south-west Spain: quantitative and

qualitative aspects. Journal of Ecology 90: 989-1001.

Green, A. J., Figuerola, J. & Sánchez, M. I. 2002. Implications of waterbird ecology for

the dispersal of aquatic organisms. Acta Oecologica 23: 177-189.

Green A. J., Sánchez, M. I., Amat, F., Figuerola, J., Hontoria, F & Hortas, F. 2005.

Dispersal of invasive and native brine shrimp Artemia (Anostraca) via waterbirds.

Limnology and Oceanography 50: 737-742.

Higgins, S.I., Nathan, R. & Cain, M.L. 2003. Are long-distance dispersal events in

plants usually caused by nonstandard means of dispersal?. Ecology 84 (8): 1945-1956.


160

Horne FR. 1966. The effect of digestive en-zymes on the hatchability of Artemia salina

eggs. Transactions of the American Microscopical Society 85: 271-74

Hortas, F. 1997. Migración de aves limícolas en el Suroeste Ibérico, vía de vuelo del

Mediterráneo Occidental y Africa, p. 77-116. In A. Barbosa [ed.], Las aves limícolas en

España. Dirección General de Conservación de la Naturaleza, Ministerio de Medio

Ambiente.

Iverson, G. C., Warnock, S. E., Butler, R. W., Bishop, M. A. & Warnock, N. 1996.

Spring migration of western sandpipers along the Pacific Coast of North America: A

telemetry study. Condor 98: 10-21.

Jakobsson, A. & Eriksson, O. 2003. Trade-offs between dispersal and competitive

ability: a comparative study of wind-dispersed Asteraceae forbs. Evolutionary Ecology

17 (3): 233-246.

MacDonald, G. H. 1980. The use of Artemia cysts as food by the flamingo

(Phoenicopterus ruber roseus) and the shelduck (Tadorna tadorna), p. 97-104. In

Persoone, G., Sorgeloos, P., Roels, O. & Jaspers, E. [eds.], The brine shrimp Artemia.

Ecology, culturing, use in aquaculture, Vol. 3. Universa Press.

McGrath, R.J. & Bass, D. 1999. Seed dispersal by Emus on the New South Wales

north-east coast. Emu 99: 248-252.

Martínez, A. 1989. Estudio de la biología de Artemia en una salina y de su influencia

sobre la calidad de la sal. Tesis, Univ. Sevilla. 388 p.

Montalti D., Arambarri A., Soave, G. E., Darrieu, C. A. & Camperi, A. 2003. Seeds in

the diet of the White-Rumped Sandpiper Calidris fuscicollis in Argentina. Waterbirds

26 (2): 166-168.


161

Nogales, M., Medina, F.M., Quilis, V. & González-Rodríguez, M. 2001. Ecological and

biogeographical implications of Yellow-Legged Gulls (Larus cachinnans Pallas) as

seed dispersers of Rubia fruticosa Ait. (Rubiaceae) in the Canary Islands. Journal of

Biogeography 28: 1137-1145.

Pennycuick, C. J. & Battley, P. F. 2003. Burning the engine: a time-marching

computation of fat and protein compsuption in a 5420-Km non-stop flight by great

knots, Calidris tenuirostris. Oikos 103: 323-332.

Pérez-Hurtado, A., Goss-Custard, J. D. & García, F. 1997. The diet of wintering waders

in Cádiz Bay, southwest Spain. Bird Study 44: 45-52.

Proctor, V. W. 1962. Viability of Chara oospores taken from migratory water birds.

Ecology 43 (3): 528-529.

Proctor, V. W. 1968. Long-distance dispersal of seeds by retention in digestive tract of

birds. Science 160: 321-322.

Proctor, V. W., Malone, C. R. & deVlaming, V. L. 1967. Dispersal of aquatic

organisms: viability of disseminules recovered from the intestinal tract of captive

Killdeer. Ecology 48: 672-676.

Sánchez, M. I., Green, A. J. & Castellanos, E. M. 2005. Spatial and temporal

fluctuations in use by shorebirds and in availability of chironomid prey in the Odiel

saltpans, south-west Spain. Hydrobiologia. In press a.

Sánchez, M. I., Green, A. J. & Castellanos, E. M. 2005. Seasonal variation in the diet of

the Redshank Tringa totanus in the Odiel Marshes, south-west Spain: a comparison of

faecal and pellet analysis. Bird Study. In press b.

SAS Institute Inc. 2000. SAS/STAT software, User’s Guide. Cary, NC.

Schupp, E. W. 1993. Quantity, quality and the effectiveness of seed dispersal by

animals. Vegetatio 107/108: 15-29.


162

StatSoft 1999. Statistica 5.5. Tusla, OK: StatSoft Inc.

Stroud, D. A., Davidson, N. C., West, R., Scott, D. A., Haanstra, L., Thorup, O., Ganter,

B. & Delany, S. 2004. Status of migratory wader populations in Africa and Western

Eurasia in the 1990s. International Wader Studies 15: 1-259.

Triantaphyliidis, G.V., Abatzopoulos, T.J. & Sorgeloos, P. 1998. Review of the

biogeography of the genus Artemia (Crustacea, Anostraca). Journal of Biogeography

25: 213-226.

Tutin, T.G., Heywood, V. H., Burges, N. A., Valentine, D. H., Walters, S. M. &

Webb, D. A. 2001. Flora Europaea. Cambridge University Press.

Valdés, B., Talavera, S. & Fernández-Galiano, E. (Eds.). 1987. Flora Vascular de

Andalucía Occidental. Vol I-III. Ed. Ketres. Barcelona.

Villaseñor, J. V. & Espinosa-García, F. J. 2004. The alien flowering plants of Mexico.

Diversity and Distributions 10: 113-123.

Wetlands International. 2002. Waterbird Population Estimates – Third Edition.

Wetlands International Global Series No. 12, Wageningen, The Netherlands.


163

Table 1. Number of samples with Artemia cysts, number of intact cysts per sample and

proportion of cysts that were broken. Each sample contained only a small fraction of the

cysts excreted in a 24 h period by a given bird.

Nº intact cysts per sample_______________________

% broken cysts________________

n With cysts With intact cysts mean ± se range mean ± se rangeTringa totanus drop. Aut-01 84 51 38 98.1 ± 49.4 (0, 3339) 41.2 ± 5.5 (0, 100)

pell. Spr-01 33 25 24 50.7 ± 36.3 (0, 1187) 14.8 ± 5.9 (0, 100)Aut-01 157 48 33 189.3 ± 105.0 (0, 15445) 41.1 ± 6.3 (0, 100)Win-01 21 0 0 0 ± 0 (0, 0) - -

Limosa limosa drop. Sum-01 42 41 41 221.6 ± 35.4 (0, 980) 0 (0, 0)Sum-02 14 5 3 0.3 ± 0.2 (0, 2) 40 ± 24.5 (0, 100)

Tringa erythropus pell. Spr-01 9 5 3 31.4 ± 31.1 (0, 280) 40 ± 24.5 (0, 100)Total 360 175 142 136.8 ± 47.6 (0, 15445) 27.7 ± 3.0 (0, 100)


164

Table 2. Generalized linear models of the effects of season, sample type and diet on the

proportion of Artemia cysts that were broken and the proportion of intact cysts that

hatched in samples from godwits and Redshank. The dependent variable was arcsine

transformed, and an identity link and normal error distribution were used. Summer

(godwit) samples, faecal samples and diet of medium hardness were aliased. Spring and

autumn samples were from Redshank. Samples from spotted redshank were excluded

from the models owing to lack of data.

Effect Estimate SE DFN DFD F P% Broken cysts Constant 0.43 0.13 N = 170 Season 1 164 0.45 0.5018D = 27.23 Autumn 0.19 0.12

Spring 0.10 0.17Summer 0 -Sample type 1 164 0.83 0.3648faeces 0 -pellet -0.10 0.11Diet 2 164 13.29 < 0.0001soft -0.38 0.11medium 0 -hard 0.35 0.12

% Cysts hatching Constant 0.07 0.04 N = 71 Season 1 67 1.94 0.1681D = 10.00 Autumn -0.03 0.08

Spring -0.31 0.22Summer 0 -Sample type 1 67 5.50 0.022Faeces 0 -

pellet 0.24 0.10

D = percentage of deviance explained by the final model compared to the null model.


165

Table 3. Number of Artemia nauplii hatched and the proportion of cysts that hatched.

Sample sizes are lower than in Table 1 because not all samples were stored in such a

way as to conserve viable cysts.

Nº nauplii per sample* % cysts hatching ________________ ____________ Nº of samples With intact cysts With nauplii mean ± se range mean ± se rangeTringa totanus drop. Aut-01 58 14 3 0.4 ± 0.2 (0, 3) 3.9 ± 3.5 (0, 50)

pell. Spr-01 5 2 0 0 0 0 0Aut-01 133 13 6 3 ± 1.6 (1, 16) 19.2 ± 10.1 (0, 100)Win-01 21 0 0

Limosa limosa drop. Sum-01 42 41 26 9.7 ± 2.3 (0, 70) 7.4 ± 1.6 (0, 40)Sum-02 14 0 0

Tringa erythropus pell. Spr-01 9 2 1 2 ± 2 (0, 4) 2.3 ± 2.3 (0, 4.6)Total 282 72 36 8.5 ± 1.4 (0, 70) 8.5 ± 2.2 (0, 100)

* For samples with at least one intact cyst. A total of 304 adult Artemia were reared

from nauplii and all were diploid A. parthenogenetica.

Table 4. Abundance of A. macrostachyum, M. nodiflorum, S. oleraceus and other unidentified seeds in wader droppings and pellets from the

Odiel saltpans, listing the numbers of samples with seeds (including those broken) and intact seeds. The total number of seeds in those samples

are given in parentheses. Each sample contains only a small fraction of the seeds excreted in a 24 h period by a given bird.

A. macrostachyum

_____________M. nodiflorum____________

S. oleraceus____________

Others____________

Total______________

N Seeds Intact Seeds Intact Seeds Intact Seeds Intact Seeds IntactTringa totanus drop. Aut-01 84 3 (3) 2 (2) 22 (29) 13 (16) 1 (1) 1 (1) 2 (2) 2 (2) 25 (35) 16 (21)

pell. Spr-01 33 5 (10) 3 (4) 1 (2) 1 (2) 12 (155) 12 (144) 0 (0) 0 (0) 16 (167) 15 (150)Aut-01 157 6 (7) 3 (3) 2 (2) 2 (2) 0 (0) 0 (0) 8 (7) 7 (8) 15 (16) 12 (13)Win-01 21 2 (1) 0 (0) 2 (2) 1 (1) 0 (0) 0 (0) 1 (1) 0 (0) 5 (4) 1 (1)

Limosa limosa drop. Sum-01 42 2 (2) 1 (1) 0 (0) 0 (0) 0 (0) 0 (0) 1 (1) 0 (0) 3 (3) 1 (1)drop. Sum-02 14 1 (1) 0 (0) 0 (0) 0 (0) 1 (1) 0 (0) 1 (1) 0 (0) 2 (3) 0 (0)

Tringa erythropus pell. Spr-01 9 4 (6) 2 (2) 0 (0) 0 (0) 6 (10) 5 (7) 0 (0) 0 (0) 6 (16) 5 (9)Tringa spp. pell. Spr-03 52 5 (8) 4 (4) 0 (0) 0 (0) 15 (31) 8 (13) 1 (1) 1 (1) 18 (40) 12 (18)Total 412 28 (38) 15 (16) 27 (35) 17 (21) 35 (198) 26 (165) 14 (13) 10 (11) 90 (284) 62 (213)

167

Table 5. Germination of A. macrostachyum, M. nodiflorum, S. oleraceus and other unidentified seeds in wader droppings and pellets from the

Odiel saltpans, listing numbers of samples with intact seeds placed for germination, and with at least one seed germinating. The total numbers of

seeds involved are listed in parentheses. Sample sizes are lower than Table 4 because not all samples were stored so to conserve viable seeds.

A. macrostachyum

______________M. nodiflorum

______________S. oleraceus

______________Others

_____________Total

_______________N Intact Germinated Intact Germinated Intact Germinated Intact Germinated Intact Germinated

Tringa totanus drop. Aut-01 84 1 (1) 1 (1) 11 (14) 1 (1) 0 0 0 0 12 (15) 2 (2)pell. Spr-01 33 2 (2) 0 0 0 11 (60) 11 (54) 0 0 13 (62) 11 (54)

Aut-01 157 2 (2) 1 (1) 2 (2) 2 (2) 0 0 3 (3) 0 7 (7) 3 (3)Win-01 21 0 0 1 (1) 1 (1) 0 0 0 0 1 (1) 1 (1)

Limosa limosa drop. Sum-01 42 0 0 0 0 0 0 0 0 0 0drop. Sum-02 14 0 0 0 0 0 0 0 0 0 0

Tringa erythropus pell. Spr-01 9 2 (2) 1 (1) 0 0 5 (7) 3 (4) 0 0 5 (9) 3 (5)Tringa spp. pell. Spr-03 52 4 (4) 2 (2) 0 0 8 (13) 8 (13) 1 (1) 1 (1) 12 (18) 11 (16)Total 412 11 (11) 5 (5) 14 (17) 4 (4) 24 (80) 22 (71) 4 (4) 1 (1) 50 (112) 31 (81)


168

Figure 1. Distribution of recoveries of Redshank and Black-tailed Godwit ringed at the

Odiel marshes and adjacent parts of Huelva province from 1992 to 2003, with recovery

distances. Local recoveries are not shown. In total, there were 16 recoveries of

Redshank and seven of godwits from outside Spain (see text).

0 500 1000 Km

1

4

6 7 8 109 12

16

1. Odiel Marshes2. Los Palacios (Seville): 54 Km3. Algarve (Portugal): 133 Km4. Charente Maritime (France): 1065

Km5. Vendeé (France): 1136 Km6. Somme (France): 1605 Km7. Pas de Calais (France): 1605 Km8. West-Vlaanderen (Belgium): 1703

Km9. Zeeland (Holland): 1809 Km10. Gelderland (Holland): 1921 Km11. Overijssel (Holland): 1947 Km12. Noord-Holland (Holland): 1964 Km13. Griend (Holland): 2009 Km14. Hannover (Germany): 2100 Km15. Schleswig-Holstein (Germany): 2272

Km

3

5

1113

1415

2

0ºW 10ºW 20ºW 30ºW

35ºN

50ºN

45ºN

40ºN

55ºN

60ºN

Tringa totanusLimosa limosaT. totanus and L. limosa


169

Figure 2. Seasonal fluctuations in the abundance of waders at the Odiel saltpans in

2001. Numbers shown represent counts at high tide in 27.6 % of the total area of

saltpans.

Tringa erythropus

Months

Ja Fe Ma Ap My Jn Jl Au Se Oc No De0

50

100

150

200

250

300

Limosa limosa

1000

2000

3000

4000

5000

6000

Tringa totanus

200

400

600

800

1000

1200

1400

1600

1800

2000N

º of s

hore

bird

s

170

CHAPTER 7 DISPERSAL OF BRINE SHRIMPS VIA BIRDS

171

CHAPTER 7

DISPERSAL OF INVASIVE AND NATIVE BRINE SHRIMPS ARTEMIA

(ANOSTRACA) VIA WATERBIRDS

ANDY J. GREEN1,5, MARTA I. SÁNCHEZ1,2, FRANCISCO AMAT3, JORDI

FIGUEROLA1, FRANCISCO HONTORIA3, OLGA RUIZ3, FRANCISCO

HORTAS4

1Estación Biológica de Doñana, Avda. María Luisa s/n, 41013 Sevilla, Spain2Departamento de Biología Ambiental y Salud Pública, Universidad de Huelva, Avda Fuerzas Armadas

s/n, 21071 Huelva, Spain3Instituto de Acuicultura de Torre de la Sal, 12595 Ribera de Cabanes, Spain

4Departamento de Biología, Universidad de Cadiz, Apartado 40, 11510 Puerto Real, Spain

Key words: Artemia franciscana, Artemia parthenogenetica, cysts viability, waders,

dispersal

Running head: Dispersal of brine shrimps via birds.

Limnology and Oceanography, 50 (2): 737-742.


172

ABSTRACT

North American brine shrimps Artemia franciscana have been exported worldwide

since the 1950s for use in the aquarium trade and fish farming. Aquaculture is

expanding along the Mediterranean coast, leading to the release of A. franciscana into

native Artemia populations. A. franciscana was first detected in 1981 in Portugal, and

has since spread to saltworks along the East Atlantic flyway used by shorebirds. Once

A. franciscana becomes established in a locality, native Artemia tend to disappear. To

test whether migratory shorebirds can disperse invasive and native Artemia between

wetlands, we extracted Artemia cysts from faeces and pellets collected at Castro Marim

(Portugal) and Cadiz Bay (Spain) during southwards migration. We found that large

numbers of viable eggs of A. franciscana and native A. parthenogenetica were dispersed

by Redshank Tringa totanus, Black-tailed Godwit Limosa limosa, and other shorebirds

migrating through the Iberian Peninsula. This is the most extensive field demonstration

to date that invertebrates can disperse readily via gut passage through birds.


173

INTRODUCTION

Invasion by non-native species is second only to habitat loss as a threat to global

biodiversity, has a huge economic impact, and has had its greatest impact in aquatic

ecosystems (Ruiz et al. 1999, Mooney & Cleland 2001). Non-native aquatic

invertebrates are typically moved between continents by man, e.g., in the ballast of

ships or intentionally for aquaculture or fisheries purposes (Leppäkoski et al. 2002,

Bailey et al. 2003). Once established, they are generally assumed to disperse using their

own active mechanisms, via ocean or river currents, or via intraregional boat traffic

(Wasson et al. 2001). The role of migratory birds in spreading exotic invertebrates has

not been fully assessed, despite the importance of understanding the dispersal

mechanisms of invasive species so that their spread can be predicted. Some authors

reviewing dispersal mechanisms of aquatic invasives have made no mention of a

potential role for birds (Leppäkoski et al. 2002).

Darwin attributed much importance to the role of migratory waterbirds as

dispersers of invertebrates ("The wide distribution of fresh-water plants and of the lower

animals I believe mainly depends on the wide dispersal of their seeds and eggs by

animals, more especially by fresh-water birds, which have great powers of flight, and

naturally travel from one piece of water to another", Darwin 1859). He suggested that

shorebirds have a particularly important role, and proposed that alien snails extended

their range by dispersing on the feathers or feet of birds (Darwin 1859). Various

laboratory studies have since demonstrated that propagules can survive passage through

the digestive system (Figuerola & Green 2002a). However, few studies have

demonstrated such dispersal in the field (Proctor 1964, Figuerola & Green 2002a).

Owing to similar morphology, until the 1980s all populations of brine shrimps

Artemia (Crustacea, Anostraca) were considered strains of a single species, but six

sexual species are now recognised together with a heterogeneous group of

parthenogenetic populations under the binomen A. parthenogenetica (Abatzopoulos et

al. 2002). The sexual A. franciscana is widespread in the Americas. In the

Mediterranean region, the native species are the sexual A. salina (formerly A. tunisiana)

together with A. parthenogenetica (mainly diploid and tetraploid populations,

Abatzopoulos et al. 2002). Since the 1950s, A. franciscana cysts have been

commercially exported worldwide from San Francisco Bay and Great Salt Lake, U.S.A.

for use in aquaculture and in the aquarium and pet trade (Abatzopoulos et al. 2002). In


174

the Mediterranean region, many traditional saltworks where salt production has become

unprofitable are being transformed into aquaculture facilities, leading to the release of

A. franciscana into hatcheries and its escape into habitats with native Artemia

populations (Amat et al. 2004). Given its use in the pet trade and even microscope sets,

indiscriminate releases into other saline wetlands suitable for Artemia are also possible.

Between 1993 and 2003, Artemia cysts (resting eggs) from numerous saltworks in the

West Mediterranean were collected and their species composition determined (Amat et

al. 2004, authors unpubl.). By then, A. franciscana was the only species recorded in

Portugal, where it is found in areas both with and without aquaculture. It is also found in

Cadiz Bay (Spain), two French sites and one in Morocco. A. franciscana was first

detected in 1981 in the Portuguese Algarve (Amat et al. 2004) and has since spread to

all the Iberian sites considered to be most important for shorebirds along the East

Atlantic Flyway: Tejo, Aveiro, Faro and Sado in Portugal and Cadiz Bay (Boyd & Pirot

1989, Amat et al. 2004). To date, there has been no spread detected to the many

saltworks east of Gibraltar outside the flyway (Fig. 1). At Aveiro, only A.

parthenogenetica was recorded in 1985, but by 1991, it had been replaced by A.

franciscana.

In this study, we assessed the ability of shorebirds (the most abundant waterbirds

in coastal saltworks) to disperse A. franciscana and A. parthenogenetica by sampling

two areas where A. franciscana has been recorded: Castro Marim (Portugal, 37°12'N

07°26'W) and Cadiz Bay (36°27'N 06°11'W). Both are situated on the Atlantic coast of the

south-western part of the Iberian Peninsula (Fig. 1) and are listed as wetlands of

international importance under the Ramsar Convention

(http://www.wetlands.org/RSDB/default.htm). Both sites hold saltworks and tens of

thousands of shorebirds during migration periods (Hortas 1997).

MATERIALS AND METHODS

We collected fresh faeces and pellets from roost sites used by monospecific flocks in

saltworks at Castro Marim on 23 July and Cadiz Bay from 22 July to 03 August 2002.

Several shorebirds, including the Redshank Tringa totanus, regurgitate pellets after

digestion. The dates of sampling coincided with the beginning of the southwards

(autumn) shorebird migration (Hortas 1997). Each sample of faeces or pellet was


175

carefully separated from the soil (discarding that part in contact with soil) and placed in

a tube. Given the number of birds present and the freshness of the samples, we are

confident that each sample was from a different bird. The samples were stored at 5oC,

and within a month, Artemia cysts were extracted by washing them in a 0.04 mm sieve

and resuspension in hypersaline brine, in which intact cysts float. The cysts were

counted, washed in distilled water then dried for 48 h at 40oC. These cysts were then

incubated in diluted filtered sea water (25 g L -1) at 26ºC and under continuous

illumination for 48 h. Hatched nauplii were counted and transferred into 60 cm3 vessels

and cultured in 70 g L -1 filtered brine (seawater plus crude sea salt), on a diet of live

Dunaliella salina and Tetraselmis suecica. They were maintained at 24ºC, under

aeration on a 12D:12L photoperiod. The medium was monitored and renewed every two

days. Resulting adults were identified morphologically (Amat et al. 2004).

We also collected thousands of cysts from saltpans adjacent to sites where we

collected faeces, determining hatching success, culturing adults and identifying them in

the same way as for cysts extracted from faeces. We also established hatching success

of A. parthenogenetica cysts collected in January 2002 from saltworks at Sanlúcar de

Barrameda (36°50'N 06°20'W), 30 km from Cadiz Bay.

The numbers of cysts transported and their hatchability were compared between

bird species, between localities, and between faeces and pellets using non-parametric

Mann-Whitney U-tests. Initial attempts to use parametric statistics were abandoned owing

to overdispersion and the high proportion of zeros in the data.

RESULTS

In both localities, we found all shorebird species studied to be transporting viable Artemia

cysts during autumn migration (Tables 1,2). At Castro Marim, Redshank and Black-tailed

Godwit Limosa limosa were transporting large numbers of viable A. franciscana cysts, and

a small number of A. parthenogenetica. At Cadiz Bay, Redshank and Dunlin Calidris

alpina were transporting viable A. parthenogenetica cysts, with the occasional presence of

A. franciscana.

At Castro Marim, Redshank faeces held more Artemia cysts than Redshank pellets

(Mann-Whitney U-tests, n = 36, 31, U = 250, p = 0.0001) or godwit faeces (n = 36, 30, U

= 161, p < 0.0001). The proportion of cysts that hatched was also higher for Redshank


176

faeces than for godwit faeces (n = 36, 27, U = 142.5, p < 0.0001) or Redshank pellets (n =

36, 30, U = 379.5, p = 0.039). Numbers of cysts recorded at Cadiz Bay were fewer than the

number at Castro Marim for Redshank pellets (n = 31, 55, U = 141.5, p < 0.0001), but the

proportion of cysts that hatched was higher at Cadiz Bay (n = 30, 25, U = 101, p < 0.0001).

However, the total number of viable cysts transported per pellet was more than five times

higher for A. franciscana in Castro Marim than for A. parthenogenetica in Cadiz Bay.

Amongst cyst samples taken directly from saltpans, we found 100% A.

franciscana at both Castro Marim and Cadiz Bay. Hatching success of A. franciscana at

Castro Marim was 74.1% (n = 17,400). Hatching success of A. parthenogenetica at

Sanlúcar de Barrameda was 50.2% (n = 15,200).

DISCUSSION

Redshank and godwits were effective dispersers of A. franciscana. Thousands of

individuals of both species migrate through each of our study sites, and move thousands of

kilometres between breeding and wintering sites (Hortas 1997). An estimated 15.5 million

shorebirds (including 440,000 Redshank, 200,000 Black-tailed Godwit, and 2,340,000

Dunlin) migrate through the East Atlantic Flyway (Stroud et al. 2004), where most major

passage and wintering sites contain saltworks. Banding data shows that some birds stop

at more than one saltwork complex as they migrate along the Iberian Atlantic coast

(details available from first author).

The birds we studied included a mixture of birds that had just completed or were

just about to commence long-distance movements, and birds that were making

movements between feeding and roosting sites within our study sites. The average time

spent by shorebirds at stopover sites during migration is unknown for our study sites,

making it impossible to assess what proportion of the cysts extracted from faeces or

pellets were transported from long distances. However, the poor match between the

species composition of Artemia cysts extracted and those from adjacent saltpans shows

that the birds were transporting cysts between different parts of wetland complexes.

Redshank, godwits and other shorebirds regularly move between saltpans separated by

up to 20 km within stopover sites (Potts, P. M., pers. comm.), favouring rapid Artemia

dispersal within and between nearby saltworks. Dominance of A. parthenogenetica in

bird samples from Cadiz (Table 1) probably reflects feeding in saltpans that had been


177

dry since A. franciscana invaded, allowing native Artemia to hatch from cysts in

sediments upon reflooding (cysts can remain viable for decades in sediments,

Abatzopoulos et al. 2002). There are hundreds of pans in Cadiz Bay, and it was not

possible to know exactly where the birds had been feeding when we collected faeces.

Redshank and godwits fly at 56-60 km h-1 (Welham 1994). In a laboratory study, the

modal retention time (much less than the mean) of viable A. franciscana cysts defecated

by another shorebird, the Killdeer Charadrius vociferus (size intermediate between

redshank and godwit), was found to be 90 min, and the maximum was 26 h (Proctor et

al. 1967). This suggests that, during migration, the maximum dispersal distance of

viable cysts would be c.1500 km. Studies of gut passage of viable A. franciscana cysts

in other birds suggests that the maximum retention time increases with body size,

ranging from 3 h in a canary Serinus canarius to at least 38 h in a Shelduck Tadorna

tadorna (Proctor 1964, MacDonald 1980).

Radio and satellite tracking shows that shorebirds move rapidly between coastal

wetlands, and distances between stopovers can exceed 1000 km (Iverson et al. 1996). A.

franciscana cysts are also likely to be transported internally by some of the other 24

migratory shorebird species using the East Atlantic Flyway (Stroud et al. 2004), as well

as other birds that consume Artemia and move frequently between wetlands. Artemia cysts

are consumed by Greater Flamingos Phoenicopterus ruber and Shelduck in the wild

(MacDonald 1980). Viable cysts are also likely to be transported externally on the feathers

and feet of waterbirds (Figuerola & Green 2002b). We do not expect that such dispersal

processes are restricted to these study sites or to autumn migration. In a study of Redshank

diet at the Odiel marshes in Spain (37°17'N 06°55'W), cysts were similarly abundant in

pellets in both spring and autumn migration (Sánchez et al. in press).

We observed much variation between waterbird species and localities in the

number of intact Artemia cysts transported, and in their viability (Tables 1, 2). This

variation is likely to be related to differences between bird populations in habitat use, diet,

gut morphology and/or retention time (Figuerola & Green 2002a). Faeces of both

Redshank and godwits contained remains of adult Artemia, and they are likely to ingest

cysts that are still inside adult Artemia (some of which may have been immature and thus

less viable, see Bohonak & Whiteman 1999), as well as those loose on the water surface or

in sediments. The greater viability of Redshank cysts recovered from faeces rather than

pellets furthers long distance dispersal, as cysts are likely to be retained longer when

expelled as faeces than when regurgitated (Nogales et al. 2001). Cysts expelled in pellets


178

may be less viable because they are ground together with grit and hard prey items in the

gizzard (Sánchez et al. in press). Adding grit to cysts fed to flamingos reduces their

viability after gut passage (MacDonald 1980).

At Castro Marim, the hatching success of A. franciscana cysts taken from bird

samples was much lower than for cysts collected in adjacent saltpans. Hatching success of

cysts from Redshank pellets at Cadiz Bay was higher than for cysts from the closest A.

parthenogenetica site we could find at Sanlúcar de Barrameda. Such comparisons are

difficult to interpret as floating cysts collected at the edge of saltpans may not be the ones

ingested by the birds and may have different viabilities. Gut passage through the Greater

Flamingo and Shelduck has been shown experimentally to reduce hatching success of A.

parthenogenetica cysts (MacDonald 1980).

Although there are few data available on the timing of arrival of A. franciscana at

different locations, and on the timing and location of introductions by man, the current

distribution of this exotic in the Iberian Peninsula is consistent with expansion via

shorebirds. Of nine saltworks studied on the main East Atlantic Flyway west of Gibraltar

since 1993 (Amat et al. 2004, authors unpubl.), A. franciscana is dominant at seven of

them (Fig. 1). Yet this species has not been detected in any of the five saltworks studied to

the east of Gibraltar and off the flyway (Fisher exact test, one-tailed p = 0.0105).

Ours is the most extensive field demonstration to date that aquatic metazoans

can disperse readily via gut passage through birds. Recently, ducks and coots were

shown to transport numerous invertebrate eggs of unknown viability (Figuerola et al.

2003). In a previous field demonstration of internal transport, Proctor (1964) found

unknown numbers of viable Cladocera and Ostracoda in three ducks. There are increasing

numbers of exotic crustaceans and bryozoans observed in aquatic systems (Leppäkoski et

al. 2002), and many are likely to have the ability to disperse as resistant eggs via birds

(Figuerola & Green 2002a). The conservation of migratory waterbirds is essential to

maintain connectivity and indigenous invertebrate biodiversity in the world’s wetlands

(Amezaga et al. 2002). Nevertheless, their capacity to disperse invasive species within

and between continents makes the needs to control the importation and release of exotic

species at a global level all the more urgent.

There is evidence of competitive exclusion of native Artemia by an exotic

species. The data available suggest that, once A. franciscana is detected amongst native

Artemia in existing populations, native Artemia disappear within a few years (Amat et

al. 2004). A. franciscana outcompetes A. parthenogenetica and A. salina within two or


179

three generations under laboratory conditions using few individuals (Abatzopoulos et al.

2002). We expect viable A. franciscana propagules to eventually reach all coastal

saltworks in the Mediterranean region via birds, but there may be conditions under

which invasion can be resisted, as in other zooplankton communities (Havel & Shurin

2004). Preventing the expansion of aquaculture into protected coastal areas still holding

native Artemia populations (see Abatzopoulos et al. 2002 for inventory) is likely to be

the most effective measure to facilitate their conservation in the short term.

Limnological effects of birds are extremely complex (Daborn et al. 1993). In

their seminal paper on ornitholimnology, Hurlbert & Chang (1983) showed how

waterbirds may affect the dynamics of invertebrate populations and, indirectly, of entire

aquatic ecosystems via predation effects. We suggest they may also do so via dispersal

effects.

ACKNOWLEDGMENTS

We are indebted to C. de le Court, N. Grade and E. Moreno for help collecting samples.

This study has been partially funded by the Spanish R+D National Plan (projects

AGL2001-4582 and BOS2003-02846). M. I. S. was supported by a grant from the

Ministry of Science and Technology. O. R. was supported by a fellowship from the

Ministry of Education and Culture. Comments by P. Jordano, C. Rico, L. Santamaría,

K. Schwenk, and D. M. Wilkinson helped to improve the manuscript. P. M. Potts

provided valuable information

REFERENCES

Abatzopoulos, T. J., Beardmore, J. A., Clegg, J. S. & Sorgeloos, P. 2002. Artemia:

Basic and Applied Biology. Kluwer Academic.

Amat, F., Hontoria, F., Ruiz, O., Green, A. J., Sánchez, M.I., Figuerola, J. & Hortas, F.

2004.


180

The American brine shrimp Artemia franciscana as an exotic invasive species in the

Western Mediterranean. Biol. Invasions. In press.

Amezaga, J. M., Santamaría, L. & Green, A.J. 2002. Biotic wetland connectivity-

supporting a new approach for wetland management policy. Acta Oecol. 23: 213-222.

Bailey, S. A., Duggan, I.C. & Van Overdijk, C. D. A. 2003. Viability of invertebrate

diapausing eggs collected from residual ballast sediment. Limnol. Oceanogr. 48: 1701-

1710.

Bohonak, A. J. & Whiteman, H. H. 1999. Dispersal of the fairy shrimp Branchinecta

coloradensis (Anostraca): Effects of hydroperiod and salamanders. Limnol. Oceanogr.

44: 487-493.

Boyd, H. & Pirot, J. Y. 1989. Flyways and reserve networks for water birds.

International Waterfowl and Wetlands Research Bureau.

Daborn, G. R., Amos, C.L., Brylinsky, M., Christian, H., Drapeau, G., W. Faas, R.,

Grant, J., Long, B., Paterson, D.M., Perillo, G.M.E. & Piccolo, M. C. 1993. An

ecological cascade effect: migratory birds affect stability of intertidal sediments.

Limnol. Oceanogr. 38: 225-231.

Darwin, C. 1859. On the origin of species by means of natural selection. John Murray.

Figuerola, J. & A. J. Green. 2002a. Dispersal of aquatic organisms by waterbirds: a

review of past research and priorities for future studies. Freshwater Biol. 47: 483-494.

Figuerola, J. & Green, A.J. 2002b. How frequent is external transport of seeds and

invertebrate eggs by waterbirds? A study in Doñana, SW Spain. Arch. Hydrobiol. 155:

557-565.

Figuerola, J., Green, A.J. & Santamaria, L. 2003. Passive internal transport of aquatic

organisms by waterfowl in Doñana, south-west Spain. Global Ecol. Biogeogr. 12: 427-

436.


181

Havel, J. E. & Shurin, J.B. 2004. Mechanisms, effects, and scales of dispersal in

freshwater zooplankton. Limnol. Oceanogr. 49: 1229-1238.

Hortas, F. 1997. Migración de aves limícolas en el Suroeste Ibérico, vía de vuelo del

Mediterráneo Occidental y Africa, p. 77-116. In A. Barbosa [ed.], Las aves limícolas en

España. Dirección General de Conservación de la Naturaleza, Ministerio de Medio

Ambiente.

Hurlbert, S. H. & Chang, C.C.Y. 1983. Ornitholimnology: Effects of grazing by the

Andean flamingo (Phoenicoparrus andinus). Proc. Natl. Acad. Sci. USA 80: 4766-

4769.

Iverson, G. C., Warnock, S.E., Butler, R.W., Bishop, M.A. & Warnock, N. 1996. Spring

migration of western sandpipers along the Pacific Coast of North America: A telemetry

study. Condor 98: 10-21.

Leppäkoski, E., Gollasch, S. & Olenin, S. 2002. Invasive aquatic species of Europe-

distribution, impacts and management. Kluwer Academic.

MacDonald, G. H. 1980. The use of Artemia cysts as food by the flamingo

(Phoenicopterus ruber roseus) and the shelduck (Tadorna tadorna), p. 97-104. In G.

Persoone, P. Sorgeloos, O. Roels and E. Jaspers [eds.], The brine shrimp Artemia.

Ecology, culturing, use in aquaculture, Vol. 3. Universa Press.

Mooney, H. A. & Cleland, E.E. 2001. The evolutionary impact of invasive species.

Proc. Natl. Acad. Sci. USA 98: 5446-5451.

Proctor, V. W. 1964. Viability of crustacean eggs recovered from ducks. Ecology 45:

656-658.

Proctor, V. W., Malone, C.R. & deVlaming, V.L. 1967. Dispersal of aquatic organisms:

viability of disseminules recovered from the intestinal tract of captive Killdeer. Ecology

48: 672-676.


182

Ruiz, G. M., Fofonoff, P., Hines, A.H. & Grosholz, E.D. 1999. Non-indigenous species

as stressors in estuarine and marine communities: assessing invasion impacts and

interactions. Limnol. Oceanogr. 44: 950-972.

Sánchez, M. I., Green, A.J. & Castellanos, E.M. In press. Seasonal variation in the diet

of the Redshank Tringa totanus in the Odiel Marshes, south-west Spain: a comparison

of faecal and pellet analysis. Bird Study.

Stroud, D. A., Davidson, N.C., West, R., Scott, D.A., Haanstra, L., Thorup, O., Ganter,

B. & Delany,S. 2004. Status of migratory wader populations in Africa and Western

Eurasia in the 1990s. International Wader Studies 15: 1-259.

Wasson, K., Zabin, C.J., Bedinger, L., Diaz, M.C. & Pearse, J.S. 2001. Biological

invasions of estuaries without international shipping: the importance of intraregional

transport. Biol. Cons. 102: 143-153.

Welham, C. V. J. 1994. Flight speeds of migrating birds: a test of maximum range

speed predictions from three aerodynamic equations. Behav. Ecol. 5: 1-8.


183

Table 1. Number of samples with intact Artemia cysts, number that produced nauplius

larvae, and number of adult Artemia reared. Not all nauplii hatched from cysts survived

to the adult stage that permitted species identification, thus the numbers of samples

containing each taxon may be underestimated. Each sample contains only a small

fraction of the cysts excreted in a 24 h period by a given bird.

Total With cysts With nauplii With A.franciscana

With nativeArtemia*

Total A.franciscana†

Total nativeArtemia*

Castro Marim Redshank faeces 36 36 36 35 7 522 10 Redshank pellets 31 30 26 23 4 180 4 Godwit faeces 30 27 18 12 7 24 10Cadiz Bay Redshank pellets 55 25 24 1 15 1 40 Dunlin faeces 103 8 1 0 1 0 1

* diploid A. parthenogenetica females.

† 52% were males.


184

Table 2. Number of intact Artemia cysts collected, numbers of nauplii hatched and the

proportion of cysts that hatched.

* For samples with at least one cyst (see Table 1 for n).

No. cysts per sample No. nauplii per sample* % cysts hatching

_____________________ ______________________ mean ± SE

No. ofsamples

mean ± SE Range mean ± SE range Castro Marim Redshank faeces 36 65.0 ± 13.1 6 - 379 18.0 ± 3.3 ene-98 33.3 ± 2.5 Redshank pellets 31 26.3 ± 6.9 0 - 160 7.5 ± 2.7 0 - 62 25.8 ± 4.2 Godwit faeces 30 27.8 ± 14.9 0 - 447 1.6 ± 0.4 0 - 8 12.4 ± 2.7Cadiz Bay Redshank pellets 55 1.7 ± 0.3 0 - 13 2.5 ± 0.4 0 - 8 68.8 ± 5.9 Dunlin faeces 103 0.1 ± 0.04 0 - 2 0.1 ± 0.1 0 - 1 12.5 ± 12.5


185

Fig. 1. Castro Marim (left insert) and Cadiz Bay (right) on a map of the Iberian

Peninsula, showing the current distribution of A. franciscana (black squares) and native

Artemia (white dots) in relation to the East Atlantic Flyway (shaded after Stroud et al.

2004). Hatched areas within inserts represent saltworks. Other Artemia sites with no

information on species composition since 1992 are not shown.

0 2 4 km

0 1 2Atlantic Ocean

AtlanticOcean

42º00 N

37º00 N

1º00 W

0 2 4 km

186

APPENDIXES

187

APPENDIXES

APPENDIXES

188

APPENDIXES

189

Appendix 1 (Chapter 3). Taxonomic classification of invertebrate taxa showing spatial

distribution in its presence. Cl: Class; O: Order; F: Family; sF: Subfamily; L: larva; P:

pupa; A: adult; I1-I9: industrial ponds; T1-T3: traditional ponds.

Invertebrates Ponds

Cl. Phylactolaemata (Bryozoa)

O. Plumatellida

F. Plumatellidae

Plumatella spp. I1, I6

Cl . Annelida

O. Oligochaeta I1

Cl. Cestoda

O Cyclophillidea

F. Hymenolepididae

Flamingolepis liguloides (Cysticercoids) I1, I2, I3, I4, I5, I6, I7, I8, I9, T1,

T2, T3

Cl. Nematoda I1, I2, I7, T2

Cl. Turbellaria I1, I2, I3, I4, I5, I6, I7, T1, T2

Cl. Gastropoda

O Mesogastropoda

F. Hydrobiidae

Hydrobia ulvae (L, A) I1, T1, T2

Other Gastropoda I1

Cl. Copepoda

APPENDIXES

190

O. Calanoida (L, A) I1, I2, T1

O. Harpacticoida

F. Cletodidae

Cletocamptus retrogressus (L, A) I1, I2, I3, I4, I5, I6, I7, I9, T1, T2,

T3

Other Harpacticoida (L, A) I1, I2, I3, I4, I7, T1, T2, T3

Cl. Ostracoda I1, I2, I4

Cl. Malacostraca

O. Amphipoda

F. Gammaridae I1

Other Malacostraca I1, I4, T1

Cl. Branchiopoda

O. Anostraca

F. Artemiidae

Artemia parthenogenetica (L, A) I1, I2, I3, I4, I5, I6, I7, I8, I9, T1,

T2, T3

Cl. Insecta

O. Heteroptera

F. Corixidae

Sigara stagnalis (L, A) I1, I2, T1, T3

O. Coleoptera

F. Dytiscidae (L) I3, I5, I6, I7, T1

F. Hydrophilidae

Berosus spinosus (A) I6

Paracymus aenas (A) I1, T1, T2

Hydrophilus spp. (L) I1, I2, T1

F. Hydraenidae

Ochthebius notabilis (L, A) I1, I2, I3, I4, I5, I6, I7, I8, T1, T2,

T3

Ochthebius corrugatus (L, A) T1, T2, T3

APPENDIXES

191

O. Diptera

F. Chironomidae

sF. Orthocladiinae

Halocladius spp. (L) I1, I2

sF. Chironominae

Tr. Chironomini

Chironomus type salinarius (L, P, A) I1, I2, I3, I4, I5, I6, I7, I8, I9, T1,

T2, T3

Other Chironomidae (L) I3, T1

F. Ephydridae

Ephydra spp. (L, P) I3, I1, I2, I4, I6, I9, T1, T2

F. Stratiomyidae

Nemotelus spp. (L) I2, T1, T2

F. Dolychopodidae (L) I2, I5, I6, T1, T2

F. Syrphidae (L) T2

Other Diptera (L) I1, I2, I3, I6, I7, I8, T1, T2, T3

O. Collembola I1, I7, T1, T2, T3

Terrestrial Invertebrates

O. Hymenoptera I5, I7, T1, T2

O. Hemiptera I1, I4, I6, T1, T2

O. Mallophaga I1, T1

O. Diptera I3, I9

O. Coleoptera I2, T1, T3

Other insecta I3, I5, I6, I8, I9, T1, T2, T3

O. Araneida I1, I2, I4, I6, T1, T3

O. Acarina I1, I2, I3, I6, I7, T1, T2, T3

Vertebrates

Cl. Ostheichthyes

O. Ciprinodontiforme

APPENDIXES

192

F. Fundulidae

Fundulus heteroclitus I1, I2, T1

Foraminiferans

Cl. Granuloreticulosea (Protozoa) I2, I3

Appendix 2 (Chapter 3). Details of the GLMs of abundance (counts) summarised in Table 3, showing the estimates and their standard errors for

each parameter, and thus the relative abundance for each month and pond.

A.parthenogenetica Artemia cyst C. retrogressus

OtherHarpacticoida O. notabilis

Ochthebius spp.larva C. salinarius larva Planaria

_________________ ______________ ______________ _________________ _______________ _________________ _______________ ______________Parameter Estimate SE Estimate SE Estimate SE Estimate SE Estimate SE Estimate SE Estimate SE Estimate SEIntercept 5.75 0.58 7.62 0.47 6.55 0.62 1.6 0.88 0.9 0.71 1 0.58 3.33 0.38 -0.4 0.65I9 -1.45 0.6 -3.16 0.51 -8.21 1.04 -15.52 1174.59 -24.51 34634 -24.7 28978.98 -4.71 0.83 -16.99 1825.34T3 2.62 0.58 -0.75 0.55 -3.01 0.72 -0.96 1.11 -0.6 0.67 1.63 0.62 -3.19 0.6 -17.33 1907.08I1 -8.88 0.63 -7.78 0.5 -2.3 0.52 3.49 0.78 -27 42369 -5.38 1.13 -3.22 0.44 -2.34 0.81I4 -0.77 0.57 -2.49 0.47 -3.24 0.66 -1.1 1.41 -2.6 0.91 -2.79 0.76 -2.14 0.42 0.07 0.61I5 -0.3 0.56 -1.75 0.47 -5.39 0.61 -16.3 978.23 -3 0.7 -3.44 0.75 -2.15 0.39 -0.28 0.61I3 -0.63 0.56 -0.61 0.46 -4.97 0.6 -2.1 1.42 -2 0.7 -2.46 0.67 -1.39 0.37 0.02 0.58I6 -0.74 0.57 -0.02 0.5 -5.01 0.6 -16.48 1068.64 -2.2 0.74 -0.78 0.61 -1.03 0.37 -2.58 0.88I7 -0.62 0.55 -1.38 0.48 -3.45 0.59 -0.43 1.1041 -1.72 0.71 -1.56 0.64 -1.97 0.38 -1.33 0.66I2 -5.33 0.61 -4.04 0.51 -2.09 0.55 3.07 0.79 -26.24 35832 -4.45 0.9 -3.12 0.45 -1.27 0.65T1 -4.11 0.54 -2.22 0.43 1.23 0.55 0.7 0.76 -0.34 0.49 -0.67 0.5 -3.6 0.48 2.34 0.51I8 -0.51 0.59 -0.82 0.45 -22.48 706.5 -16.16 1002.16 -4.04 1.21 -3.26 0.88 -3.85 0.51 -18.17 1396.07T2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0January -0.09 0.46 -0.08 0.34 -0.99 0.5 -19.31 794.6 0.19 0.64 -0.3 0.56 -2.36 0.4 -17.01 1056.43March 1.75 0.41 -0.21 0.36 2.25 0.53 -19.26 847.31 0.12 0.79 2.18 0.63 0.52 0.3 -1.75 0.83May 2.85 0.37 0.72 0.36 -0.46 0.51 -2.3 0.76 1 0.61 2.36 0.54 0.003 0.3 1.5 0.54July 2.55 0.38 0.31 0.36 -0.66 0.48 -0.65 0.67 2.11 0.62 2.21 0.53 -0.25 0.31 2.82 0.52September 3.27 0.41 1.14 0.36 -0.95 0.49 -1.01 0.74 1.68 0.61 2.07 0.55 -0.22 0.31 1.37 0.55November 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Depth 0.01 0.02 -0.06 0.02 -0.04 0.03 -0.09 0.05 0.04 0.04 -0.06 0.03 -0.04 0.02 -0.07 0.03Shore distance -0.002 0.004 -0.01 0.004 -0.01 0.01 -0.01 0.02 -0.04 0.02 -0.02 0.01 -0.01 0.004 0.0001 0.007Fetch -0.0001 0.0008 0.002 0.0008 -0.0004 0.001 -0.001 0.002 -0.003 0.002 -0.001 0.001 0.001 0.001 0.0003 0.001

Appendix 3 (Chapter 3). Details of the GLMs of volume (biomass) summarised in Table 4, showing the estimates and their standard errors for

each parameter, and thus the relative abundance for each month and pond. Zero values were excluded from these models, hence the removal of

some ponds and months from some models.A.

partenogenética Artemia cyst C. retrogressusOther

Harpacticoida O. notabilisOchthebius spp.

larvaC. salinarius

larva Planaria ________________ _____________ ______________ _______________ _______________ _________________ ______________ ______________Parameter Estimate SE Estimate SE Estimate SE Estimate SE Estimate SE Estimate SE Estimate SE Estimate SEIntercept -1.84 0.21 -2.55 0.21 -2.3 0.22 -4.4 0.28 -2.64 0.23 -2.83 0.18 -1.07 0.13 -3.77 0.23I9 -0.25 0.23 -1.19 0.24 -1.74 0.49 ... ... ... ... ... ... -0.94 0.26 ... ...T3 0.53 0.29 -0.46 0.25 -1.02 0.33 -0.2 0.4 -0.05 0.22 0.31 0.18 -0.82 0.22 ... ...I1 -2.41 0.37 -2.33 0.3 -0.81 0.2 1 0.24 ... ... -0.77 0.33 -0.8 0.16 -0.13 0.33I4 -0.23 0.22 -0.65 0.22 -1.13 0.23 0.45 0.73 -0.42 0.35 -0.52 0.2 -0.81 0.13 0.04 0.24I5 0.07 0.21 -0.47 0.22 -1.62 0.25 ... ... -0.35 0.34 -0.75 0.19 -0.79 0.13 -0.17 0.21I3 0.06 0.21 0.03 0.21 -1.43 0.24 0.03 0.87 -0.46 0.24 -0.36 0.18 -0.55 0.12 0.09 0.21I6 -0.05 0.22 -0.11 0.22 -1.19 0.28 ... ... -0.51 0.27 -0.17 0.17 -0.37 0.13 -0.16 0.35I7 0.08 0.21 -0.22 0.22 -0.92 0.24 -0.43 0.6 -0.32 0.25 -0.29 0.15 -0.63 0.13 -0.32 0.22I2 -1.84 0.28 -1.77 0.22 -0.7 0.21 0.77 0.27 ... ... -0.53 0.33 -0.81 0.17 -0.04 0.26T1 -1.44 0.24 -0.78 0.21 0.44 0.2 0.24 0.26 0.26 0.16 0.06 0.13 -0.56 0.18 0.62 0.19I8 0.4 0.22 -0.26 0.22 ... ... ... ... -0.72 0.53 -0.29 0.29 -0.81 0.22 ... ...T2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0January -0.26 0.15 -0.09 0.15 -0.01 0.19 ... ... -0.17 0.2 -0.11 0.18 -0.35 0.17 ... ...March 0.88 0.17 -0.08 0.16 0.68 0.18 ... ... 0.05 0.3 0.23 0.2 0.15 0.11 -0.48 0.35May 0.99 0.15 0.19 0.14 -0.13 0.17 0.38 0.42 0.16 0.21 0.14 0.17 -0.05 0.1 0.26 0.18July 1.21 0.16 0.17 0.15 -0.25 0.18 0.04 0.21 0.44 0.22 0.38 0.18 -0.18 0.1 0.59 0.17September 1.4 0.16 0.31 0.15 -0.25 0.19 -0.26 0.24 0.35 0.21 0.27 0.18 -0.2 0.11 0.23 0.18November 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Depth 0.03 0.01 -0.01 0.01 -0.03 0.01 -0.01 0.02 0.006 0.01 -0.03 0.01 -0.02 0.01 -0.02 0.01Shore distance -0.002 0.002 -0.004 0.002 -0.003 0.002 -0.003 0.01 -0.005 0.01 -0.005 0.003 -0.001 0.001 -0.001 0.003Fetch -0.0002 0.0003 0.001 0.0003 0.0001 4.00E-04 -0.001 0.001 -0.0003 0.001 -0.0002 0.0003 0.001 0.0002 0.0004 0.0004

DISCUSIÓN GENERAL

195

DISCUSIÓN GENERAL

Los resultados de esta tesis se discutieron en detalle en cada capítulo, por lo que me

limitaré en esta sección a presentar una visión general de las aportaciones de este trabajo

al conocimiento de los ecosistemas de salinas (a través de las relaciones entre

invertebrados y limícolas), la aplicación de los resultados obtenidos a la gestión, así

como a plantear aquellas cuestiones que considero importantes para desarrollar en el

futuro.

Aportaciones de la tesis

Esta tesis confirma la importancia de las salinas como hábitat de alimentación y

descanso para los limícolas (ver Velásquez y Hockey 1992, Masero 2003) y destaca el

valor de las salinas del Odiel a nivel internacional para al menos 6 especies de limícolas.

Dado que sólo se censó el 27% de la superficie de la salina, es de esperar que la zona

alcance importancia internacional para un mayor número de especies. Con más de

20.000 aves censadas durante los pasos migratorios en la superficie estudiada,

demostramos que las salinas del Odiel constituyen una de las zonas más importantes

para los limícolas en la ruta migratoria de Atlántico Este (ver Stroud et al. 2004), junto

con Doñana y la Bahía de Cádiz (Martí y del Moral 2002).

La calidad de un hábitat se define en gran medida por la disponibilidad de

recursos (para el caso de los limícolas por la disponibilidad de invertebrados). En el

Odiel dominaron los quironómidos (Chironomus salinarius) y las artemias (Artemia

parthenogenetica), donde constituyen recursos complementarios, alcanzando sus

máximas densidades en momentos diferentes del ciclo anual. Mientras que a finales de

primavera (mayo) la densidad de ambos es elevada, a comienzos del periodo invernal

(noviembre) Artemia prácticamente ha desaparecido, siendo muy abundantes las larvas

de quironómido. La salinidad parece ser el principal condicionante de la presencia y

abundancia de invertebrados, tanto a nivel espacial como temporal. No obstante otros

factores correlacionados con la salinidad podrían estar implicados (como la presencia de

macrófitos o de predadores, ver Williams 1998), siendo necesarios experimentos

detallados para poder discernir entre uno y otros efectos.

Comprobamos la importancia relativa de la salina tradicional e industrial para las

comunidades de limícolas e invertebrados. La salina industrial destacó especialmente

DISCUSIÓN GENERAL

196

desde el punto de vista de las aves, mientras que la tradicional lo hizo desde el punto de

vista de los invertebrados, con una alta productividad y especies únicas, como

Ochthebius corragatus (abundante y ampliamente distribuido) que nunca se registraron

en la salina industrial.

Comprobamos el valor de combinar el análisis de excrementos y egagrópilas

para evaluar la dieta de los limícolas. En base a dicho análisis realizamos el estudio de

la dieta del Archibebe común (Tringa totanus) más detallado hasta el momento.

Observamos un cambio en la dieta entre el paso migratorio de primavera y el de otoño,

con predominio de presas típicas de la salina en primavera y especies típicas de los

fangos intermareales en otoño, lo que sugiere un cambio en el uso del hábitat entre

estaciones.

Hemos visto cómo las relaciones entre limícolas e invertebrados son complejas,

existiendo diversas vías por las que aves e invertebrados interaccionan y afectan

mutuamente. En primer lugar consideramos las relaciones predador-presa. Los

quironómidos, invertebrados dominantes en el bentos, jugaron un papel clave en el

establecimiento de la dinámica de la comunidad de limícolas, condicionando cambios

en la dieta y en el uso de las salinas a lo largo del ciclo anual. Durante el paso

migratorio de primavera los limícolas dependen en gran medida de larvas y pupas de

quironómido, ejerciendo una fuerte influencia sobre la dinámica de la población de

presas, incluso a densidades de aves relativamente bajas. Sin embargo no encontramos

relación entre abundancia de aves (y uso del hábitat) y abundancia de invertebrados de

la columna de agua a lo largo del tiempo, a diferencia de lo observado para las larvas de

quironómido, lo que sugiere que éstos no están implicados en la fenología de los

movimientos migratorios de estas aves. Artemia fue el metazooplancton dominante en

la mayor parte del proceso de producción de sal, tanto en términos de abundancia como

de biomasa, ocurriendo sólo de forma ocasional en las balsas de primera evaporación,

donde son predadas por la comunidad de peces (Britton y Johnson 1987). No obstante,

de cara a las aves, no sólo es importante la cantidad, sino la calidad del recurso. En este

sentido el valor nutritivo de Artemia es bajo en comparación con el de otras presas

como las larvas de díptero (Rubega & Inouye 1994), lo que apoya la hipótesis de que en

las salinas del Odiel son las larvas de quironómido las que juegan el papel más

importante en las interacciones predador-presa. No obstante, es posible que la escasa

respuesta de las aves a los cambios en la abundancia de Artemia se deba al hecho de

encontrarse las máximas densidades en las balsas de elevada salinidad, donde

DISCUSIÓN GENERAL

197

alimentarse puede conllevar elevados costes por osmorregulación (Purdue & Haines

1977). Sin embargo, encontramos evidencias de que los limícolas seleccionaron las

balsas con mayor biomasa de invertebrados, tanto para el caso de los quironómidos

como de Artemia. Es posible que la correlación entre aves e invertebrados estuviera

enmascarada al tratar la comunidad de limícolas en su conjunto, mezclando especies

con diferentes preferencias de dieta. Por ejemplo, existe una gran variabilidad

interespecífica en la capacidad de los limícolas para utilizar Artemia como recurso.

Mientras que unas especies son capaces de compensar la baja calidad de un recurso,

consumiendo mayores cantidades (Berkuil et al 2003), otras se encuentran

fisiológicamente limitadas (Rubega, 1994). Por ejemplo se ha comprobado que una

dieta basada en Artemia es rentable para los Correlimos común (Calidris alpina) y

zarapitín (C. ferruginea) (con una tasa de ganancia de masa corporal de más de 0.5 g/dia

en condiciones óptimas) pero no para el Correlimos falcinelo (Limicola falcinellus)

(Berkuil et al (2003). Es posible que la relación relativamente débil encontrada entre la

disponibilidad de invertebrados planctónicos y la abundancia de aves se deba a un error

en el diseño del muestreo. Monitorizar las poblaciones de Artemia es complicado

debido a su distribución altamente contagiosa (Lenz, 1980) y heterogénea en las balsas,

influida entre otros factores por el viento, la intensidad de la luz y la temperatura,

factores altamente cambiantes, que determinan la disponibilidad de recursos y por tanto

la producción de biomasa de Artemia (Naegel & Rodríguez-A 2002). Por tanto, para

obtener una estima adecuada con un error estándar suficientemente pequeño es

necesario muestrear una gran cantidad de puntos usando un procedimiento de muestreo

estratificado (Baert et al. 2002). Un muestreo a menor escala, comparando la

abundancia de presas entre puntos con mayor y menor densidad de limícolas, dentro de

una balsa, sería más adecuado para contestar esta pregunta; no obstante, esta no fue la

prioridad de la tesis.

Un segundo nivel de interacciones es el que se refiere a las relaciones parásito-

hospedador. En esta tesis abordamos una primera aproximación al estudio de las

interacciones del sistema cestodos-Artemia-limícolas, describiendo la comunidad de

metacestodos que portaba la población de Artemia parthenogenetica del Odiel. Se trata

del estudio más completo de este grupo de parásitos realizado en la Península Ibérica

hasta la fecha. Nuestros resultados revelaron una alta tasa de parasitismo por cestodos,

así como una elevada diversidad, con un total de 8 especies identificadas. Destacó

Flamingolepis liguloides (cuyos adultos parasitan flamencos) por su elevada

DISCUSIÓN GENERAL

198

prevalencia, y Anomotaenia tringae (parásito de limícolas) por ser el primer registro en

hospedador intermedio.

Existen evidencias de que los parásitos juegan un papel fundamental en la

dinámica de poblaciones y estructuración de las comunidades animales (Price et al

1986, Dunn & Dick 1998). Los parásitos puede además afectar las interacciones

predador-presa. Muchas especies de parásitos, como la mayoría de los cestodos, poseen

complejos ciclos de vida que involucran la transmisión trófica de un hospedador a otro

por consumo un hospedador intermedio. Algunas tienen la capacidad de alterar el

comportamiento de sus hospedadores intermedios y hacerlos más vulnerables a la

predación por el hospedador definitivo (Bakker et al. 1997, Fuller et al. 2003). El

consumo de artemias parasitadas por los limícolas es un hecho habitual en el Odiel,

como pone de manifiesto la presencia de cisticercoides en los excrementos y

egagrópilas de archibebes. Diversos trabajos muestran que las artemias parasitadas por

Flamingolepis liguloides (Cestoda: Ciclophillidea) presentan un color rojo intenso y

ocupan un micro hábitat diferente al de los individuos no parasitados (Gabrion et al.

1982, Amat 1991, datos propios sin publicar). Datos propios no publicados sugieren que

los cambios morfológicos provocados por la infección aumenta el riesgo de predación

por los limícolas. Cuando el costo energético del parasitismo es moderado, y la captura

de la presa está facilitada por los parásitos, el consumo de las presas infectadas puede

ser beneficioso, lo que explica que en muchos casos no exista una presión selectiva para

evitar las presas infectadas (Lafferty 1992). Sin embargo, es probable que en Odiel los

limícolas se estén alimentando de artemias parasitadas por Flamingolepis liguloides u

otros parásitos no específicos de limícolas, de manera que ello no represente ningún

coste. En el caso opuesto, un predador podría reducir los efectos negativos del

parasitismo cambiando su dieta óptima (Lozano 1991). El papel de los parásitos ha sido

enormemente ignorada en la Teoría del forrajeo óptimo, por la cual los animales

maximizan su "fitness" maximizando la tasa a la que consumen los recursos. Sin

embargo existen ejemplos en la literatura que contradicen dicha teoría (Norris 1999), en

los que la maximización de la tasa de ingesta de energía conlleva costos derivados de

una mayor exposición a parásitos.

Otra importante vía por la que las aves son capaces de afectar la estructura de las

comunidades de invertebrados es mediante la dispersión. Aunque se ha asumido que las

aves son importantes vectores de dispersión de propágulos de animales y plantas, son

escasos los estudios empíricos que demuestren dicha capacidad (Figuerola & Green

DISCUSIÓN GENERAL

199

2002, Bohonak & Jenkins 2003). En esta tesis se presenta el trabajo más exhaustivo

realizado en el campo hasta la fecha que demuestra que los invertebrados pueden ser

dispersados vía tracto digestivo de las aves. Los limícolas transportaron quistes viables

de Artemia, tanto en el paso migratorio de primavera como en el paso de otoño.

Además, demostramos por primera vez en el campo que los limícolas son capaces de

dispersar semillas de angiospermas viables. Los limícolas son migradores de largas

distancias, implicados en los procesos de dispersión a larga distancia. La distribución

nativa de Artemia y de semillas de angiospermas registradas en este estudio es

consistente con dicha hipótesis. Investigamos la implicación de diversos factores en la

viabilidad de los propágulos, como por ejemplo la dieta, la forma de excreción

(mediante egagrópilas o excrementos) o el número total de propágulos ingeridos,

encontrando una mayor viabilidad con dietas blandas y a mayor número de quistes

ingeridos. Aunque se ha sugerido que las variaciones en la dieta pueden influir la tasa de

supervivencia de los propágulos, esta es la primera vez que se demuestra en el campo su

implicación real. La relación de la viabilidad con la forma de excreción no estuvo clara,

hallando resultados opuestos en dos trabajos diferentes. No obstante los resultados

fueron débilmente significativos, lo que sugiere que otras variables pudieran estar

creando extra variabilidad, confundiendo los resultados. Este tipo de problemas son

inherentes a los estudios de campo, donde se tiene un escaso o nulo control sobre el

sistema.

La implicación de los limícolas en la dispersión de especies exóticas es otro

resultado interesante de esta tesis. El hallazgo de semillas de Sonchus oleraceus y

Mesembryanthemum nodiflorum en los excrementos y egagrópilas de los limícolas,

sugiere que las aves podrían contribuir a la dispersión y expansión de estas especies en

sus rangos, introducidas en América y Australia. Especialmente importante y revelador

fue el hallazgo de quistes de Artemia franciscana viables en las egagrópilas y

excrementos de los limícolas. A. franciscana es una especie invasora procedente de

Norte América que está desplazando a gran velocidad a las especies autóctonas (Amat

et al. 2005). En España existe una especie sexual, A. salina, y una línea partenogenética,

reconocida como A. parthenogenetica, con poblaciones tanto diploides como

tetrapliodes. En el Odiel coexiste una población mayoritariamente dipliode con una

pequeña proporción de individuos tetrapliodes. Actualmente A franciscana es la única

especie presente en Portugal y ya ha alcanzado las salinas de la bahía de Cádiz. Existen

evidencias de exclusión competitiva de las artemias nativas por A franciscana (Browne

DISCUSIÓN GENERAL

200

1980, Browne & Halanych 1989). La llegada de esta especie introducida al Odiel a

través de los limícolas es inevitable. No obstante su efecto sobre las poblaciones nativas

del Odiel y otras partes del mediterráneo no está claro, ya que existen mecanismos por

los que la invasión puede ser resistida. A pesar de su alta capacidad competitiva en el

laboratorio, los propágulos de A. franciscana dispersados a nuevas áreas podrían no

establecerse debido al "efecto de prioridad" de las especies autóctonas que ocupan el

hábitat en grandes densidades (Jenkins & Buikema 1998). Las poblaciones de Artemia

establecidas tienen bancos extremadamente abundantes de propágulos, con los que

deben competir los quites de las invasoras. Se ha demostrado experimentalmente que las

comunidades de zooplancton son capaces de resistir a la invasión por nuevas especies

(Shurin 2000). Además cabe esperar que las poblaciones nativas de Artemia presenten

resistencia a la invasión, ya que los machos de A. franciscana no muestran preferencia

para aparearse con las hembras de su propia especie, al menos en el laboratorio (Browne

1980). Así, es improbable que A. franciscana pueda reproducirse cuando la especie

nativa es mucho más abundante. Sin embargo, los datos disponibles sobre su

distribución actual sugieren que, una vez llega a ser suficientemente abundante para ser

detectada entre las elevadas densidades de Artemia nativa de la población preexistente,

rápidamente la desplaza (Amat et al. en prensa). Si un lugar ocupado por Artemia está

cerca de un lugar infectado y hay un alto grado de conectividad vía movimiento de aves,

la tasa de inmigración de propágulos invasores podría ser suficiente para, en poco

tiempo, superar la resistencia a la invasión. Sin embargo, dada la actual incertidumbre

sobre la biología de la invasión por A. franciscana, prevenir la expansión de la

acuicultura en el Odiel, salinas de Sanlúcar, Cabo de Gata y otras áreas costeras

protegidas que aún albergan poblaciones nativas de Artemia, es probablemente la

medida más efectiva para facilitar su conservación.

Con esta tesis hemos profundizado en el conocimiento de las relaciones entre

limícolas e invertebrados explorando distintos niveles de dichas interacciones. Durante

el estudio se han suscitado muchas preguntas, unas las hemos abordado y muchas

quedan aún por resolver. Ahondar en las cuestiones planteadas permitirá entender mejor

las complejas relaciones ecológicas que se dan en los ecosistemas acuáticos y ayudará a

conservar tan amenazados hábitas.

DISCUSIÓN GENERAL

201

Gestión y conservación

Como se indicó en la introducción, la pérdida de hábitats es actualmente el principal

problema ambiental a nivel global, estando especialmente afectados los sistemas

acuáticos (Sadoul et al. 1998, Wilcove et al. 1998). Los humedales artificiales se han

propuesto como alternativa a dicha pérdida de hábitat (Velasquez 1992; Masero et al.

1999), ya que con un adecuado manejo pueden llegar a ser tan productivos como los

sistemas naturales e igualmente atractivos para las aves (Davison y Evans, 1986;

Rehfisch, 1994). En este sentido, la información recogida en esta tesis puede ser de una

gran utilidad para el manejo y gestión de las salinas del Odiel, y otros humedales de

similares características.

Manejo del agua. Los limícolas dependen de la elevada productividad de quironómidos

en la salina, especialmente durante el paso migratorio de primavera, cuando basan su

dieta fundamentalmente en larvas y pupas de Chironomus salinarius. C. salinarius

dominó en el bentos a lo largo de la mayor parte del proceso de cristalización,

presentando un marcado patrón estacional con picos de abundancia en primavera (1470

larvas/m2) y en otoño (1063 larvas/m2). No obstante, una gran parte de la productividad

no está accesible a las aves debido a los niveles de agua de las balsas. En nuestro área

de estudio los limícolas explotaron profundidades de hasta 20 cm (salvo las avocetas,

que se alimentaron a 25 cm nadando), variando según las especies. En promedio sólo un

10% de la superficie del conjunto de las balsas estudiadas fue menor de 20 cm,

quedando la mayor parte inaccesible a las aves. Una reducción del 16% de la superficie

accesible, observada durante el paso de otoño respecto al paso de primavera,

acompañado de una mayor afluencia de aves y de forma más prolongada en el tiempo,

podría justificar parcialmente el menor uso de la salina en ese periodo, aunque otros

factores podrían estar implicados (como cambios estacionales en los beneficios de

alimentarse en los fangos intermareales, o de Artemia por diferencias en la prevalencia

de parásitos, cambios en su distribución, tamaño, etc). En cualquier caso, un descenso

general de los niveles de agua durante ambos periodos migratorios, que garantizase el

acceso a una superficie somera más amplia, aumentaría enormemente el valor de las

salinas del Odiel como hábitat de forrajeo para los limícolas. Como ya se comentó en la

discusión, la predación por limícolas ejerce una regulación "top-down" sobre los

quironómidos durante el paso de primavera, afectando a diversos parámetros

DISCUSIÓN GENERAL

202

poblacionales, como la abundancia, biomasa y distribución de tamaños de larvas. Este

efecto es significativo incluso a densidades de aves relativamente bajas, lo que sugiere

que la tasa de ingesta de las aves esté limitada por el uso que hayan hecho previamente

las primeras en llegar. Por tanto proponemos que el descenso general de los niveles de

agua vaya acompañado de un descenso paulatino (drawdown) en los niveles de algunas

balsas, por ejemplo con una velocidad de 1 cm por día en los momentos máximos de

paso (aproximadamente entre el 15 y el 30 de abril y entre el 1 de agosto y 15 de

septiembre). Dicho drawdown hace que haya continuamente nuevas zonas de

sedimentos disponibles para las aves de manera que puedan acceder a altas densidades

de larvas que no han sido consumidas días anteriores. El descenso gradual también evita

la mortandad de larvas si grandes superficies quedaran completamente drenadas

(Rehfisch 1994). El manejo debería favorecer igualmente el acceso a Artemia, más

abundantes en las zonas profundas (menos accesibles) como muestran nuestros

resultados Un descenso de los niveles de agua pondría disponible un gran número de

ellas al quedar concentradas en la superficie somera. Las balsas más adecuadas para

manejar los niveles de agua son las de segunda evaporación. De todo el conjunto de la

salina es la zona que reúne mejores condiciones para la alimentación de los limícolas.

Son balsas en general someras, con suaves orillas, de grandes superficies y los niveles

de perturbación antrópica son bajos. La salinidad oscila en torno a 45 y 125 g/l, lo que

promueve altas densidades tanto de quironómidos como de artemias. No obstante el

óptimo varía entre ambos, y mientras que los quironómidos alcanzan las máximas

densidades (hasta 24.000 larvas/m2) al principio del rango (entre 45 y 80 g/l), Artemia

lo hace al final (hasta 400 individuos/l por encima de 80 g/l,). Las balsas de segunda

evaporación son además las que menos perturbación pueden crear desde el punto de

vista de los intereses industriales de la salina. En este sentido, tanto las balsas de

primera evaporación, que actúan de reservorio de agua con que se nutre el resto de la

salina, como la zona de cristalización, que requiere de un fino manejo que garantice el

precipitado del cloruro sódico lo más puro posible, constituyen las áreas más sensibles

al manejo de agua de cara a la producción de sal y por tanto son las balsas menos

susceptibles al manejo del agua de cara a las aves.

En conclusión, ligeras modificaciones en el manejo del agua de las salinas,

especialmente de ciertas balsas, a través de las medidas que proponemos, incrementaría

notablemente la proporción de la producción de quironómidos, artemias y otros

DISCUSIÓN GENERAL

203

invertebrados consumida por las aves, cabiendo esperar un efecto positivo sobre sus

tamaños poblacionales a lo largo de su ruta migratoria (Sutherland 1996, 1998).

Suavizado de las pendientes de las orillas. Las labores de mantenimiento y construcción

de muros en las salinas son actividades prácticamente constantes. En las salinas

industriales como la que estudiamos, con una gran superficie, las actividades de

reparación requieren, la mayor parte de las veces, de maquinaria pesada. El trasiego de

estas máquinas por las orillas, destroza con frecuencia las escasas zonas someras que

tienen los limícolas para alimentarse, promoviendo orillas con pendientes abruptas

difíciles de utilizar para una gran cantidad de aves, especialmente las de pequeño

tamaño. Entre la fauna de limícolas existen especies de muy variado tamaño, con

diferentes longitudes de picos y patas, y estrategias de forrajeo. Por ejemplo, en nuestro

área de estudio las avocetas explotaron las mayores profundidades, observándose

nadando a profundidades de hasta 25 cm, mientras que los pequeños correlimos lo

hicieron entre 1 y 3 cm de profundidad. Por tanto para garantizar un hábitat adecuado

para todas las especies es necesario el mantenimiento de orillas con suaves pendientes

que permitan ser explotadas por toda la comunidad de limícolas. Una alternativa es la

construcción de islas de pendientes suaves en medio de cada balsa, de manera que las

aves puedan alimentarse en sus bordes en lugar de en la orilla de la propia balsa.

Además ello constituye una medida eficaz contra predadores terrestres como los gatos.

Las escasas islas que existen en actualmente en la salina industrial (por ejemplo en E11

(I3)) son utilizadas por un gran número de limícolas tanto para descansar como para

alimentarse en sus bordes, lo que demuestra la eficacia real las elevaciones del terreno

en el interior de las balsas y cuya construcción mejoraría notablemente el hábitat para

los limícolas.

Control del uso de especies invasoras. Nuestros resultados demostraron que especies

exóticas como Artemia franciscana pueden ser eficazmente dispersadas vía limícolas.

La invasión por especies exóticas es, después de la pérdida de hábitat, el principal

problema que afecta a la diversidad a nivel global (Wilcove et al. 1998, Money &

Cleland 2001). Una vez que esta especie es detectada entre las artemias nativas, estas

últimas tienden a desaparecer en sólo unos pocos años (Amat et al. 2005). Los quistes

de A. franciscana se comercializan para acuicultura en todo el mundo. El abandono de

salinas y cambio de usos hacia actividades piscícolas está favoreciendo la dispersión de

DISCUSIÓN GENERAL

204

A. franciscana a una gran velocidad a lo largo de la costa mediterránea. A diferencia de

la Bahía de Cádiz, con una gran tradición en acuicultura, y donde ya ha llegado A.

franciscana, en el Odiel no ha existido tradición alguna y hasta el momento sólo la

autóctona A. parthenogenetica ha sido registrada. No obstante, la llegada de individuos

de zonas próximas como el Algarve en Portugal o la Bahía de Cádiz es inminente. Sin

embargo, especialmente preocupante es la transformación, en Odiel, de una antiguas

salinas abandonadas, en piscifactorías, justamente al otro lado del canal mareal que

suministra agua a la salina estudiada, lo que destaca la necesidad de evitar el uso de A.

franciscana como fuente de alimento para los peces. Dado que los limícolas se mueven

regularmente entre zonas de descanso y alimentación distancias de hasta 20 km, el paso

de aves de un complejo a otro sería extremadamente fácil. Además, es altamente

probable las artemias exóticas escapen de las instalaciones de la piscifactoría,

contaminando directamente las balsas de la salina. La expansión de A. franciscana no

sólo amenaza las poblaciones nativas de Artemia, sino también la comunidad de

parásitos asociada, muy diversa en las salinas estudiadas, y prácticamente ausentes en

ejemplares de A. franciscana analizados (datos propios sin publicar). También

ignoramos el efecto que podría tener para los limícolas, el cambio de artemias

autóctonas a exóticas. Si, como sugieren nuestras observaciones, las aves están

predando en gran medida sobre artemias parasitadas (especialmente con cestodos

parásirtos de flamencos), dado que la prevalencia de parásitos en la especie exótica

parece ser muy escasa, sería probable que la tasa de ingestión de los limícolas se viera

reducida considerablemente tras el cambio a un sistema dominado por A. franciscana.

En conclusión, prevenir la expansión de la acuicultura en los espacios protegidos, donde

existen poblaciones de Artemia nativas, es probablemente la medida más adecuada para

frenar su expansión y conservar la biodiversidad de los sistemas de salinas.

Por otro lado, las poblaciones de especies autóctonas de invertebrados y plantas, en

humedales como las salinas de Odiel, están en parte sostenidas a través de la dispersión

por limícolas y otras aves que mantienen la conectividad entre poblaciones de distintos

humedales (Amezaga et al. 2002). Esta es una razón más de la necesidad de conservar la

salinas del Odiel, y evitar su abandono o cambios de uso que impliquen la

transformación a otro tipo de hábitat como supondría por ejemplo la acuicultura. En tal

caso cabría esperar un impacto negativo no sólo sobre las poblaciones de aves,

invertebrados y plantas en la propia salina sino también en otras partes de la ruta

DISCUSIÓN GENERAL

205

migratoria del Atlántico Este, al perderse una pieza fundamental en el conjunto de

humedales que mantiene la biodiversidad de plantas e invertebrados a través de su

conectividad vía vectores de dispersión como las aves. En definitiva, la necesidad de

conservar las salinas del Odiel trasciende a las fronteras de Andalucía y España, siendo

fundamental para mantener las poblaciones, por ejemplo de Artemia autóctonas o de

Arthrocnemum en países colindantes.

Importancia de las salinas tradiciones. En la región mediterránea los humedales

costeros y las salinas continúan desapareciendo. La actividad tradicional prácticamente

ha sido abandonada, o transformada en otros usos. En las marismas del Odiel aún se

mantiene el uso tradicional coexistiendo con el industrial, que habría que hacer un

esfuerzo en conservar. Los resultados de nuestro estudio muestran que ambos manejos,

industrial y tradicional, no son equivalentes en cuanto a la comunidad de invertebrados

y uso por las aves que uno y otro favorecen. Al contrario de lo que podría esperarse por

su manejo menos intensivo, el uso tradicional no proporciona un hábitat preferido para

los limícolas, siendo la salina industrial, de grandes superficies, la que albergara

mayores densidades de aves. No obstante, el manejo tradicional, proporciona un tipo de

hábitat distinto, con mayor diversidad y biomasa de invertebrados acuáticos, mayor

diversidad de plantas y probablemente de aves e invertebrados terrestres. El uso

tradicional favorece la diversidad biológica por su integración en el entorno y

adaptación a los ciclos naturales. Por todo ello, consideramos que su conservación en su

estado actual es una prioridad para el espacio protegido. La combinación de salinas

industriales y tradicionales aumenta la biodiversidad en el espacio, lo que puede ser un

buen modelo de gestión para otras zonas mediterráneas con salinas activas.

Prioridades para el futuro

A lo largo de esta tesis hemos abordado diversas cuestiones relacionadas con las

relaciones ecológicas entre limícolas e invertebrados. No obstante también se han

suscitado muchas cuestiones que habrá que abordar en futuros estudios. Es prioridad

seguir profundizando en el trabajo iniciado en esta tesis, en cada unos de los temas

tratados. A continuación se proponen las líneas de trabajo que se consideran prioritarias.

DISCUSIÓN GENERAL

206

1. En primer lugar proponemos la necesidad de realizar censos más completos que

incluyan toda la salina para, evaluar su importancia a escala global, y poder determinar,

por ejemplo, qué especies cumplen el criterio RAMSAR que concede importancia

internacional a aquellos humedales que albergan el 1% de la población de una especie

2. Son necesarios estudios que evalúen los cambios en el uso de las salinas, en base a las

fluctuaciones en la disponibilidad de presas en los fangos intermareales. Al igual que

hemos hecho para las salinas, habrá que analizar cómo cambia la abundancia y

estructura de las comunidades de invertebrados entre primavera y otoño en los caños,

principal hábitat de alimentación intermareal para los limícolas durante la marea baja

(Velásquez et al. 1991, Kalejta 1993). Asimismo habría que evaluar, mediante

experimentos de exclusión, el efecto de la predación sobre las comunidades de

invertebrados de los fangos intermareales. Sería interesante repetir también los

experimentos de exclusión durante el paso migratorio de otoño en la salina. De nuevo,

diferencias en el uso de las salinas entre primavera y otoño podrían estar ocasionados

por diferencias en la abundancia de las diferentes especies de limícolas entre ambos

periodos, que utilizan de forma diferencial los recursos.

3. Proponemos igualmente ahondar en las relaciones tróficas entre invertebrados y

limícolas, con estudios detallados analizando las respuestas de las especies más

importantes de limícolas (como las agujas o los archibebes), a los cambios de sus

principales presas (tanto entre balsas como a una escala más fina dentro de una balsa).

Para ello será necesario identificar las presas más importantes para las diferentes

especies, con más estudios de dieta. Una forma alternativa al análisis de egagrópilas y

excrementos, con un gran potencial, es el uso de isótopos estables. Estas técnicas

permiten evaluar la dieta de las aves analizando la composición isotópica los tejidos del

consumidor (íntimamente relacionado con la dieta), evitando el sesgo inherente al

análisis de excremetos y egagrópilas (Duffy & Jackson 1986). No obstante, estas

técnicas también son susceptibles al sesgo, por ejemplo en el caso de no tener estimas

del tiempo empleado alimentándose en distintas zonas dentro de un área con una alta

variabilidad isotópica (Alexander et al. 1996). No obstante es una técnica potente que

bien utilizada puede aportar una valiosísima información. Además presenta otras

ventajas adicionales, ya que debido a diferencias en las tasas metabólicas entre tejidos,

el análisis de isótopos estables puede indicar la dieta en una variedad de periodos que va

DISCUSIÓN GENERAL

207

desde unos días (análisis de hígados), a varias semanas (análisis del músculo), y hasta

años (análisis de huesos) (Hobson & Clark 1992). Además de profundizar en las

relaciones tróficas entre limícolas e invertebrados, proponemos estudios que ahonden en

las relaciones tróficas dentro de la comunidad de invertebrados.

4. En esta tesis se realizó una primera aproximación a las relaciones parásito-

hospedador. Son necesarios estudios que ahonden en las relaciones cestodos-Artemia-

limícolas, realizando experimentos que permitan ampliar el conocimiento del efecto del

parasitismo sobre Artemia. Parece ser que ciertas especies de cestodos provocan

cambios en la coloración de Artemia (adquiriendo un intenso color rojo) y modifican su

micro distribución, entre otras cosas (Amat et al. 1991, Gabrion et al. 1982). No

obstante es muy escasa la información que existe al respecto. En un segundo nivel será

necesario evaluar si los cambios provocados por los parásitos aumentan la

susceptibilidad a la predación por el hospedador final. Experimentos con aves en

cautividad permitirán testar dicha hipótesis, además de otras relacionadas con la

selección de presas parasitadas. Por ejemplo, si los limícolas pueden o no distinguir

entre artemias con cestodos parásitos de ellos mismos (por ejemplo Anomotaenia

tringae) y cestodos específicos de otra aves (como Flamingolepis loguloides, parásito

de flamencos).

Un paso más en el estudio de las relaciones parásito-hospedador, será descifrar

las interacciones entre parásitos que infectan simultáneamente un mismo individuo así

como el efecto que tiene la infección mixta sobre el comportamiento y morfología de

Artemia, y en última instancia el efecto sobre la selección de presas por las aves. De

especial interés, y que aún no ha sido evaluado, es el papel que pudieran desempeñar los

parásitos en las relaciones de competencia entre A. franciscana y las especies

autóctonas. Se ha demostrado la implicación de los parásitos en los procesos de

invasiones biológicas, mediando en la competencia entre especies autóctonas e

invasoras (Hudson & Greenman, 1998, Bauer et al. 2000). La tasa de parasitismo de

especies exóticas a menudo es muy baja en comparación con las especies nativas (Dunn

& Dick 1998), bien por una menor susceptibilidad de las especies invasoras a la

infección por parásitos autóctonos, bien porque en el proceso de invasión se

seleccionaron individuos resistentes. Ello conferiría ventajas competitivas sobre las

especies autóctonas, facilitando la invasión (Settle & Wilson 1990). Por otro lado, el

efecto de una especie de parásito puede variar según el hospedador al que parasite,

DISCUSIÓN GENERAL

208

como demostraron Bauer et al. (2000) al comprobar una menor eficacia del parásito

acantocéfalo Pomphorhynchus laévis cuando infectaba una especie invasora de un

anfípodo gammárido que infectaba a la autóctona. Los parásitos tienen una gran

influencia sobre la comunidad de invertebrados a través de muy variados efectos, que

incluyen relaciones predador-presa, competencia interespecífica etc. No obstante la

mayoría de estudios de ecología de poblaciones dejan al margen el papel de los

parásitos en la estructuración y dinámica de las comunidades animales, que habría que

empezar a tener en cuenta.

5. En cuanto a la dispersión de propágulos por limícolas son necesarios estudios más

amplios, para evaluar la direccionalidad de la dispersión, incluyendo un mayor número

de muestras entre otoño y primavera, y un mayor número especies de limícolas.

Estudios de la dispersión de semillas de Sonchus oleraceus y Mesembryanthemum

nodiflorum en su rango introducido ayudarían a esclarecer el papel de las aves en su

expansión. Habría que profundizar en el estudio de los movimientos de los limícolas

entre el Odiel y otras localidades, para evaluar la probabilidad de dispersión a sitios y

hábitats adecuados. A este respecto un enorme potencial tiene el seguimiento por radio-

telemetría (via satélite) utilizado ya en numerosos estudios con limícolas y que está

proporcionando excelentes resultados (Takekawa et al. 2002).

Referencias

Alexander, S. A., Hobson, K. A., Gratto-Trevor, C. L. & Diamond, A. W. 1996.

Conventional and isotopic determinations of shorebird diet at an inland stopover: the

importance of invertebrates and Potamogeton pectinatus tubers. Canadian Journal of

Zoology 74: 1057-1068.

Amat, F., Illescas, M.P. & Fernandez, J. 1991. Brine shrimp Artemia parasitized by

Flamigolepis liguloides (Cestoda, Hymenolepidadae) cysticercoids in Spanish

Mediterranean salterns. Vie Milieu 41 (4): 237-244.

DISCUSIÓN GENERAL

209

Amat, F., Hontoria, F., Ruiz, O., Green, A. J., Sánchez, M. I., Figuerola, J. & Hortas F.



Amezaga, J.M., Santamaría, L. & Green, A.J. 2002. Biotic wetland connectivity-

supporting a new approach for wetland policy. Acta Oecologica 23 (3): 213-222.

Bakker, T. C. M., Mazzi, D. & Zala, S. 1997. Parasite-induced changes in behavior and

color make Gammarus pulex more prone to fish predation. Ecology 78 (4): 1098-1104.

Baert, P., Nguyen Thi, N. A., Burch, A. & Sorgeloos, P. 2002. The use of Artemia

biomass sampling to predict cyst yields in culture ponds. Hydrobiologia 477: 149-153.

Bauer, A., Trouvé, S., Grégoire, A., Bollache, L. & Cézilly, F. 2000. Differential

influence of Pomphorhynchus laevis (Acanthocephala) on the behaviour of native and

invader gammarid species. International Journal for Parasitology 30: 1453-1457.

Bohonak, A. J. & Whiteman, H. H. 1999. Dispersal of the fairy shrimp Branchinecta

coloradensis (Anostraca): Effects of hydroperiod and salamanders. Limnology and

Oceanography 44: 487-493.

Britton, R. H. & Johnson, A. R. 1987. An ecological account of a Mediterranean salina:


Browne, R. A. 1980. Competition experiments between parthenogenetic and sexual

strains of the brine shrimp, Artemia salina. Ecology 61 (3): 471-474.

Browne, R. A. & Halanych, K. M. 1989. Competition between sexual and

Parthenogenetic Artemia: a re-evaluation (Branchiopoda, Anostraca). Crustaceana 57

(1): 57-71.

Duffy, D. C. & Jackson, S. 1986. Diet studies of seabirds: a review of methods.

Colonial Waterbirds 9: 1-17.

DISCUSIÓN GENERAL

210

Dunn, A. M. & Dick, J. T. A. 1998. Parasitism and epibiosis in native and non-native

gammarids in freshwater in Ireland. Ecography 21:593-598.

Figuerola, J. & Green. A. J. 2002. Dispersal of aquatic organisms by waterbirds: a


494.

Fuller, C. A., Rock, P. & Philips, T. 2003. Behavior, Color changes, and predation risk

induced by Acanthocaphalan parasitism in the caribbean termite Nasutitermes

acajutlae. Caribbean Journal of Science 39 (1): 128-135.

Gabrion, C., MacDonald-Crivelli, G. & Boy, V. 1982. Dynamique des populations

larvaires du cestode Flamingolepis liguloides dans une population d'Artemia en

Camargue. Acta Oecologica 3 (2): 273-293.

Hobson, K. A. & Clark, R. G. 1992. Assessing avian diets using stable isotopes I:

Turnover of 13C in tisúes. Condor 94: 181-188.

Hudson, P. & Greenman, J. 1998. Competition mediated by parasites: biological and

theoretical progress. Trends in Ecology and Evolution 13: 387-390

Jenkins, D. G. & Buikema, A. L., Jr. 1998. Do similar communities develop in similar

sites? a test with zooplankton structure and function. Ecological Monographs 68 (3):

421-443.

Kalejta, B. 1993. Diets of shorebirds at the Berg River Estuary, South Africa: spatial

and temporal variation. Ostrich 64: 123-133.

Lafferty, K. D. 1992. Foraging on prey that are modified by parasites. The American

Naturalist 140: 854-867.

Lenz, P. H., 1980. Ecology of an álcali-adapted variety of Artemia from Mono Lake

California, USA. In Persoone, G., Soorgelos, P., Roels, O. & Jaspers, E. (eds), The

Brine Shrimp Artemia, Vol 3. Universa Press. Wetteren, Belgium: 79-96.

DISCUSIÓN GENERAL

211

Lozano, G. A. 1991. Optimal foraging theory: a posible role for parasites. Oikos 60 (3):

391-395.

Martí, R., & del Moral, J.C. 2002. La invernada de aves acuáticas en España. Dirección

General de Conservación de la Naturaleza-SEO/BirdLife. Organismo Autónomo

Parques Nacionales, Ministerio de Medio Ambiente. Madrid.

Masero, J. A., Pérez-González, M., Basadre, M. & Otero-Saavedra, M. 1999. Food

supply for waders (Aves: Charadrii) in an estuarine area in the Bay of Cádiz (SW

Iberian Peninsula). Acta Oecologica 20: 429-434.

Masero, J. A. 2003. Assessing alternative anthropogenic habitats for conserving

waterbirds: salinas as buffer areas against the impact of natural habitat loss for

shorebirds. Biodiversity and Conservation 12: 1157-1173.

Mooney, H. A. & Cleland, E. E. 2001. The evolutionary impact of invasive species.

Proceedings of National Academy of Science, USA. 98 (10): 5446-5451.

Naegel, L. C. A. & Rodríguez-A, S. 2002. Ecological observations and biomass

proximate composition of the brine shrimp Artemia (Crustacea: Anostraca) from

Pichilingue, Baja California Sur, México. Hydrobiologia 486: 185-190.

Norris, K. 1999. A trade-off between energy intake and exposure to parasites in

oystercatchers feeding on a bivalve mollusc. Proceeding of the Royal Society of London

266: 1703-1709.

Prize, P. W., Westoby, M., Rice B, Fritz, R. S., Thompson, J. N. & Mobly, C. 1986.

Parasite mediation in ecological interactions. Annual Review of Ecology Systematic 17:

487-505.

DISCUSIÓN GENERAL

212

Purdue, J.R. & Haines, H. 1977. Salt water tolerance and water turnover in the snowy

plover. Auk 94: 248-255.

Rehfisch, M. M., 1994. Man-made lagoons and how their attractiveness to wader might


Ecology 31: 383-401.

Rubega, M. A. & Inouye, C. 1994. Prey switching in Red necked Phalaropes

Phalaropus lobatus: feeding limitations, the functional response and water

managementat Mono Lake, California, USA. Biologial Conservation 70: 205-210.

Sadoul, N., Walmsley, J. & Charpentier, B. 1998. Salinas and nature conservation.

Conservation of Mediterranean Wetlands Nº 9. Tour du Valat. Arles. France.

Settle, W., & Wilson, L. T. 1990. Invasion by the variegated leafhopper and biotic

interactions: parasitism, competition, and apparent competition. Ecology 71: 1461–

1470.

Shurin, J. B. 2000. Dispersal limitation, invasion resistance, and the structure of pond

zooplankton communities. Ecology 81 (11): 3074-3086.

Stroud, D. A., Davidson, N. C., West, R., Scott, D. A., Haanstra, L., Thorup, O., Ganter,

B. & Delany, S. (compilars on behalf of the International Wader Study Group). 2004.

Status of migratory waders populations in Africa and Western Eurasia in the 1990s.

International Wader Studies 15: 1-259.

Sutherland, W.J. 1996. Predicting the consequences of habitat loss for migratory

populations. Proceedings of the Royal Society London Series B. 263: 1325-1327.

Sutherland, W.J. 1998. The effect of local change in habitat quality on populations of

migratory species. Journal of Animal Ecology 35: 418-421.

Takekawa, J. Y., Warnock, N., Martinelli, G. M. Miles, A. K. & Tsao, D. T. 2002.

Waterbird use of bayland wetlands in the San Francisco Bay Estuary: Movements of

DISCUSIÓN GENERAL

213

long-billed Dowitchers during the winter. Waterbirds 25 (Special Publication 2): 93-

105.

Velásquez, C. R. Kalejta, B. K. & Hockey, P. A. R. Seasonal abundance, habitat

selection and energy consumtion of waterbirds at the Berg River Estuary, South Africa.

Ostrich 62: 109-123.

Velásquez, C. R., 1992. Managing artificial saltpans as a waterbirds habitat: species

responses to water level manipulation. Colonial Waterbirds 15(1): 43-55.

Velásquez, C. R. & Hokey, P. A. R. 1992. The importance of supratidal foraging

habitats for waders at a south temperate estuary. Ardea 80: 243-253.

Verkuil, Y., Van der Have, T. M., Van der Winden, J. & Chernichko, I. I. 2003. Habitat

use and diet selection of northward migrating waders in the Sivash (Ukraine): the use of

Brine shrimp Artemia salina in a variable saline lagoon complex. Ardea 91 (1): 71-83.

Wilcove, D.S., Rothstein, D., Dubow, J. Phillips, &. Losos, A. E . 1998. Quantifying

threats to imperiled species in the United States. BioScience 48:607–615.

Williams, D. D. Williams, N. E. 1998. Aquatic insects in an estuarine environment:

densities, distribution and salinity tolerance. Freshwater Biology 39: 411-421.

214

CONCLUSIONES

215

CONCLUSIONES

1. Las salinas del Odiel sustenta de manera regular más de 20.000 limícolas, albergando

el 1% de los individuos de la población de 6 especies de limícolas, criterio que estipula

el convenio RAMSAR para considerar un humedal de importancia internacional para

las aves acuáticas.

2. La comunidad de invertebrados de las salinas del Odiel está dominada por larvas de

Chironomus salinarius (en el bentos) y Artemia parthenogenética (en la columna de

agua), salvo en las balsas de primera evaporación donde la diversidad de invertebrados

es alta (con 27 especies diferentes registradas).

3. Entre los factores químicos, la salinidad fue el más importante condicionante de la

abundancia y diversidad de invertebrados, determinando la estructura de la comunidad a

lo largo del gradiente espacial. No obstante el efecto de la salinidad difirió para los

invertebrados planctónicos y bentónicos, estando favorecidas las cadenas tróficas

planctónicas frente a las bentonicas con el aumento de la salinidad. Entre los factores

biológicos destacaron la presencia de peces y macrófitos.

4. El estudio de la dieta de los limícolas rinde resultados diferentes para el análisis de

egagrópilas y excrementos, pero ambos están correlacionados en la abundancia y rango

de presa consumidos. Los dos métodos son complementarios y una combinación de

ambos es la opción más adecuada para describir la dieta de los limícolas.

5. Los cambios estacionales en la abundancia de quironómidos, determinan cambios en

la dieta y uso de las salinas por los limícolas a lo largo del ciclo anual. Durante el paso

migratorio de primavera, los Archibebes comunes se alimentaron preferentemente en las

salinas, consumiendo larvas y pupas de Chironomus salinarius; en el paso de otoño

utilizaron los fangos intermareales, alimentándose fundamentalmente de poliquetos y

moluscos.

6. Las variaciones espaciales en la abundancia de invertebrados y superficie disponible,

de forma conjunta, explicaron diferencias en la densidad de aves. A pesar del manejo

menos intensivo de las salinas tradicionales y la mayor biomasa de invertebrados que

CONCLUSIONES

216

sustentaron, las salinas industriales albergaron mayores densidades de aves.

7. Los limícolas ejercieron una importante influencia sobre la dinámica de larvas de

quironómidos en las salinas el Odiel. La exclusión de aves mediante cercados produjo

un fuerte efecto sobre la densidad, biomasa y distribución de tamaño de larvas.

8. Los experimentos de exclusión mostraron que el efecto de la predación no fue más

intenso a mayor densidad de aves, existiendo un profundo efecto incluso a baja

densidades de predadores. Por tanto la llegada de los primeros limícolas afecta la tasa de

ingesta de los migrantes más tardíos. La intensidad del efecto de la predación no sólo

depende de la densidad de predadores sino del momento en que tiene lugar la predación

en relación al los ciclos de productividad de las presas.

9. Las relaciones predador-presa en las salinas del Odiel son complejas, existiendo

evidencias de un doble control "top down" y "bottom up" en la regulación y

estructuración de las poblaciones de predador y presa.

10. La población de Artemia parthenogenetica del Odiel mostró una diversa comunidad

de metacestodos con un total de 8 especies identificadas, 4 de ellas parásitos de

flamencos, 3 de limícolas, 1 de zampullines y 1 de láridos. Destacaron Flaminolepis

liguloides (parásito de flamencos) por su elevada prevalencia, Anomotaenia tringae

(parásito de limícolas) por ser el primer registro en hospedador intermedio y Confluaria

podicipina (parásito de zampullines), Euricestus avoceti (parásito de limícolas y

flamencos) y Gynandrotaenia stammeri (parásitos de flamencos) por registrarse por

primera vez en España.

11. Los limícolas son eficaces vectores de dispersión de propágulos de invertebrados y

plantas. En el Odiel encontramos que Archibebes comues, Archibebes oscuros y Agujas

colinegras transportaron quistes de Artemia parthenogenetica y semillas de Sonchus

oleraceus, Mesembryanthemum nodiflorum y Arthrocnemum macrostachyum viables, lo

que sugiere una importante implicación de los limícolas en la distribución de estas

especies y flujo génico entre poblaciones.

CONCLUSIONES

217

12. El transporte y viabilidad de quistes y semillas varió con la localidad, la estación, el

tipo de muestra (excrementos o egagrópilas) y la especie de limícola, lo cual tiene

implicaciones sobre la probabilidad y direccionalidad de la dispersión.

13. La dieta tuvo una fuerte influencia sobre la tasa de supervivencia de los quistes de

Artemia, observándose una mayor cantidad de quistes intactos a medida que disminuyó

la dureza de la dieta. Otros factores como forma de excreción de los propágulos

(mediante excrementos o egagrópilas) o el número de popágulos ingeridos también

afectó a la viabilidad de los mismos, encontrándose en este último caso una correlación

positiva entre los componentes de cantidad y calidad de la dispersión (a mayor ingesta

de quistes, mayor proporción de quistes viables).

14. Los limícolas tienen una importante función en la dispersión de especies exóticas,

tanto de invertebrados (Artemia franciscana) como de plantas (Sonchus oleraceus y

Mesembryanthemum nodiflorum). La expansión de Artemia franciscana vía limícolas

pone en serio peligro las poblaciones de Artemia autóctonas en toda Europa, así como a

las especies estrechamente asociadas a ella, como sus parásitos.

218

AGRADECIMIENTOS

219

AGRADECIMIENTOS

No podría poner punto y final al trabajo de estos últimos años sin dar las gracias a todas

las personas que han colaborado de una u otra manera haciendo posible la realización de

esta tesis y por su puesto haciendo mucho más ameno el trabajo. Ellos han sido

imprescindibles durante estos años, y a todos les dedico mi esfuerzo.

En primer lugar me gustaría dar las gracias a mis directores de tesis, Andy y

Eloy, por haberme ofrecido la oportunidad de trabajar con ellos. Con Andy me inicié en

este apasionante mundo y de él he aprendido muchísimo. Nada más acabar la carrera me

ofreció trabajar con la dieta de la Malvasía y accedí sin pensarlo dos veces. Aquel fue

mi primer contacto con la investigación, que acabó de reafirmar mi interés por las aves

acuáticas y los ecosistemas acuáticos en general, y del que guardo un entrañable

recuerdo. Andy confió en mí desde el primer momento (mucho más que yo misma!), y

espero no haberle defraudado. Trabajar con él ha sido todo un privilegio para mí,

tremendamente estimulante y enriquecedor. Le agradezco su implicación, dedicación y

asesoramiento con mi trabajo, y su amistad.

Con Eloy aprendí muchas cosas sobre el funcionamiento de las marismas del

Odiel, despertando en mí el interés y la pasión que ahora siento por la marisma. Le

agradezco su ayuda e implicación, inestimables, durante el arranque y primera fase de

mi tesis. De esa época guardo muy buenos recuerdos.

A la Estación Biológica de Doñana y a la Universidad de Huelva agradezco el

haberme acogido, permitiéndome trabajar en sus centros y poniendo a mi alcance todos

los medios que necesité.

Juan Carlos Rubio, Director Conservador del Paraje Natural Marismas del Odiel,

defendió desde el principio mi proyecto de tesis, proporcionándonos los permisos para

trabajar en la salina y poniendo todos los medios del paraje a nuestra disposición. Le

agradezco el interés mostrado en todo momento por mi trabajo y su colaboración.

Gracias a Javier Aranda, director de las Salinas, por permitirme trabajar en las

salinas, poniendo todos los medios de ARAGONESAS a mi alcance.

La mayor parte de mi tesis ha sido sustentada económicamente por una beca FPI

de Acción MIT financiada por el Ministerio de Ciencia y Tecnología. Durante el último

año disfruté de una beca I3P de postgrado del CSIC.

K. Schwenk, D. M. Wilkinson, N. Burton, W. Cresswell, J. Figuerola, P.

Jordano, C. Rico, L Santamaría, F. Amat, V.V. Kornyushin y V.V. Tkach, aportaron

AGRADECIMIENTOS

220

interesantes comentarios y sugerencias a los manuscritos. Pablo García Murillo y Carlos

Luque ayudaron con la identificación de semillas y Dagmar Frisch, Carmen Elisa Sanz

y Francisco Amat con la identificación de invertebrados.

Los datos de anillamiento fueron proporcionados por la Oficina de Especies

Migratorias (Ministerio de Medio Ambiente).

Durante estos años he tenido la suerte de convivir con gente de una gran calidad

científica y humana. Claudine me ofreció su ayuda y colaboración durante toda mi tesis,

desde el primer momento cuando me sugirió la posibilidad de optar a la beca que he

disfrutando durante estos años. Gracias por tu interés y por tu amistad. Los guardas de

las marismas del Odiel, especialmente José Manuel Sayago y Enrique Urbina, me

ayudaron a reducir el inmenso mundo que suponía para mí la identificación de los

limícolas. Con ellos aprendí a distinguir los correlimos zarapitines de los comunes en

invierno, las aguja colinegras de las colipintas, y muchas cosas más. Ramona, Gerardo y

las niñas me cuidaron como a una más de la familia. Sin esos colacaos calentitos, las

noches de trabajo hubieran sido muy distintas!. Gracias, Jeni, por tu ayuda, "pescando"

quironómidos. Y gracias por supuesto por vuestra grata compañía, vuestro cariño y

hospitalidad, que nunca olvidaré. Antonio el Topo, Juan Luis Benítez y Benito López

Iñiguez me enseñaron el funcionamiento de la salina y el manejo del agua, y me

facilitaron amablemente toda la información y datos que necesité. Los trabajadores de la

salina, "El Lobo" y compañía, también me prestaron su ayuda siempre que la necesité.

Los estudiantes de Ambientales de Huelva, especialmente Jorge, Eduard, Cristina,

María José, Jacinto y Carlos, me echaron buena mano con los muestreos, censos y

demás labores del campo. Con ellos compartí muy buenos momentos, además de tortilla

de patatas y gazpacho!. Gracias a mis compañeros de grupo Cristina, Héctor y Jordi; a

Esther y Patricia del laboratorio acuático; a Maria del Mar y a Raquel, que me ayudaron

de una u otra forma con el trabajo. Especialmente quiero dar las gracias a Cristina, a

quien he acudido infinidad de veces, siempre dispuesta a echarme una mano y

apoyándome cuando más lo he necesitado. Gracias guapa, y ahora a por la tuya!. Hugo

y Esther me han ayudado muchísimo, más de lo que ellos piensan. Con ellos he

compartido risas y lágrimas (a partir de ahora prometo compartir sólo risas!). Gracias

por vuestra amistad y complicidad. Espero no perderos de vista a ninguno. Raquel,

Carlos, Elvira, Mari Carmen, Paz, Gema, Jose, Juan, Nico (espero no dejarme a

ninguno) no dudaron en ponerse las botas o los prismáticos cada vez que lo necesité.

Gracias a todos por vuestra ayuda y amistad. Nuria y Josefina también estuvieron

AGRADECIMIENTOS

221

siempre conmigo, en los momentos más difíciles, y en los mejores también. Gracias a

Luis por tantos "chutes" compartidos de energía. Emilio no dudó en echarme un cable

cada vez que lo necesité, por sólo unas cervecillas!. La "limilla que corre" te está

tremendamente agradecida (y su padre también). De Miguel Delibes nunca me faltaron

sus ánimos, su cariño y su amistad. Gracias a Manolo Carrión, mi artista revelación

preferido, por recordarme casi todas las mañanas que "la prisa mata" y por sacarme cada

día, sin excepción, una sonrisa (o una carcajada). Con Paco compartí gran parte del

tiempo de la tesis y aprendí mucho de su experiencia; me echó buena mano en el campo

(de lo que guardo un bonito recuerdo) y con mis infinitas dudas estadísticas. El

responsable de tan preciosa portada, el artista, es Francis, que no dudó en cederme uno

de sus maravillosos dibujos. Gracias!!!. Manolo me ha ayudado mucho con su cariño y

amistad, y ha hecho mucho más llevadero el tirón final.

Finalmente, mi más sincero agradecimiento a mi familia por su apoyo

incondicional, por haber confiado en mi en todo momento y animado a seguir siempre

mi camino. A mis padres y a mi abuelo les debo mi amor y respeto por los "bichos" y el

campo.

A todos quisiera tener cerca y no perderos de vista, al menos, en los próximos 50 años!

Documents

Marta Isabel Sánchez Ordóñez · 1. Relaciones predador-presa 1.1 En primer lugar estudiamos la comunidad de invertebrados y su efecto sobre la comunidad de limícolas, analizando