comunidades de macroinvertebrados en un rio altoandino

Transcription

COMUNIDADES DE
MACROINVERTEBRADOS EN UN RIO
ALTOANDINO:
IMPORTANCIA DEL MICROHABITAT,
DINAMICA DE LA DERIVA, PAPEL DE LA
MATERIA ORGANICA Y RELEVANCIA DE LA
OVOPOSICION
BLANCA RIOS TOUMA
Departament d’Ecologia
UNIVERSITAT DE BARCELONA
Octubre 2008
Tesis Doctoral
Universitat de Barcelona
Facultat de Biología- Departament d’Ecologia
Programa de doctorado: Ecología. Bienio 2002-2004
COMUNIDADES DE MACROINVERTEBRADOS EN UN RIO
ALTOANDINO:
IMPORTANCIA DEL MICROHABITAT, DINAMICA DE LA
DERIVA, PAPEL DE LA MATERIA ORGANICA Y RELEVANCIA
DE LA OVOPOSICION
Memoria presentada por Blanca Patricia Ríos Touma para optar al
título de Doctor por la Universitat de Barcelona, bajo la dirección
del doctor Narcís Prat i Fornells
Blanca Patricia Ríos Touma
Barcelona, Octubre de 2008
El director de la Tesis:
Dr. Narcís Prat i Fornells
Catedràtic d’Ecologia
Facultat de Biología (UB)
A mis padres
Índice
Agradecimientos...................................................................................
I
Prefacio..................................................................................................
V
Introducción..........................................................................................
Los Ríos Tropicales Altoaninos..................................................
El microhábitat...........................................................................
Deriva y recolonización..............................................................
Importancia de la materia orgánica............................................
La ovoposición……………………………………………………...
Objetivos................................................................................................
Chapter 1 Macroinvertebrate assemblages at an Andean high
altitude stream: Importance of microhabitat, flow and season…….
Introduction………………………………………………………….
Materials and Methods…………………………………………….
Results……………………………………………………………….
Discussion…………………………………………………………..
Chapter 2 Invertebrate drift and colonization processes in a
Tropical Andean Stream…………………………………………………..
Introduction………………………………………………………….
Results………………………………………………………………
Discussion……………………………………………………….....
Chapter 3 Leaf litter organic matter dynamics and associated
invertebrates in a high altitude tropical Andean stream…………….
Introduction………………………………………………………….
Materials and Methods………….…………………………………
Results……………………………………………………………….
Discussion…………………………………………………………..
Chapter 4 Oviposition of Aquatic Insects in a High Altitude
Tropical Stream…………………………………………………………......
Introduction……………………………………………………….....
Results………………………………………………………………
Discussion…………………………………………………………..
Conclusiones…………………………………………………………........
Resumen General……………………………………..............................
Bibliografía…………………………………………………………………..
ANEXO1 Traducción al kichwa del Prefacio, Introducción y
Conclusiones………………………………………………………………..
Kallariy (versión Kichwa del Prefacio)………………………………....
Kallari yuyal (versión Kichwa de la Introducción)......……………….
Ashalla shimikunawan puchukanchik (version Kichwa de las
conclusiones y resumen general)……………………………………….
VII
IX
XII
XIV
XV
XVII
XIX
1
3
6
10
18
29
31
34
37
47
55
57
59
63
71
75
77
80
83
92
99
105
119
137
139
140
150
Agradecimientos
Agradecimientos
Quiero agradecer a mi director de tesis, Narcís Prat, por todo el apoyo dado en
la concepción y transcurso de la tesis, sobre todo por haber valorado y apoyado
la idea de realizar la tesis sobre un río de los Andes. A todos los miembros del
grupo FEM y antiguos ECOBILLs por su interés, entusiasmo, apoyo y amistad
durante estos años: María Rieradevall, Mireia Vila-Escalé, Núria Bonada, Caro
Solá, Raúl Acosta, Cesc Múrria, Toni Munné, Núria Cid, Tura Punti, Rosa
Cassanovas, Rosa Andreu, Iraima Verkaik, Miguel Cañedo, Pau Fortuño, Laura
Puértolas, Teresa Vegas, y a los más nuevos Mia, Núria S., Christian, Isabelle
y Esther.
A Andrea Encalada quién a más de dirigir, apoyar, mejorar y enriquecer este
trabajo ha sido muy importante en todo el proceso de la tesis. Gracias Andrea
por todo el apoyo científico, moral y afectivo de estos años! Gracias también por
abrir las puertas del laboratorio de Ecología Acuática de la USFQ y emprender
la aventura de esta tesis de forma tan positiva. A todos los miembros del Lab.
de Ecología Acuática: Gaby, Fernanda, Daniel, Cristina, Diego, Verónica y
también a la doña Mary. Un agradecimiento especial a Esteban Suárez por la
ayuda en el diseño de los índices de caudal. También a Wills Flowers, quien
durante
su
visita
al
Laboratorio
de
Ecología
Acuática
identificó
los
Ephemeroptera.
A todas las personas que me ayudaron en el trabajo de campo y laboratorio. A
Adelaida Aigaje, por ser una gran ayudante, entusiasta colaboradora de campo
y acompañarme al río, durante los días de sol y los días de tormenta (cuando la
gente nos decía que estábamos locas). Al Pucho Andino, muchísimas gracias
por tu trabajo entusiasta y tu sentido del humor, tanto en el campo como en
laboratorio. A Carolina Arroyo, por su esfuerzo y dedicación en el laboratorio. A
Rodrigo por su ayuda en laboratorio y campo. Al Juanjo Vásconez, por ser un
amigo incondicional y ayudarme en la confección de parte del material de
campo con una precisión impresionante, por venir al campo a ayudarme, por
sobrevivir al accidente más grave que tuvimos en toda la fase de campo, de la
I
Agradecimientos
cuál aprendimos lecciones para toda la vida… A Pauli Viteri, Jaime Camacho y
Saskia Flores de EcoCiencia, por apoyar la idea de hacer la tesis en Oyacachi y
ayudarme dando mis primeros pasos en la comunidad. A todos los amigos que
colaboraron y me dieron ánimos en las etapas más duras y en las más felices
de esta tesis en especial a Xavier Cisneros, Pauli Viteri, Esteban Neumann,
Paola Carrera, Ana Troya, Lola Guarderas, Jorge Celi, Rafa Yánez, Paula
Orihuela, Juan Pablo Roggiero, José Hidalgo, Jaime Chaves, Oscar Pérez,
Fede Brown, Luisa Chaves, Vero Brown, Ana Bustamante, María Moreno de los
Ríos, Raúl Sstrambótica, Pau, Raúl Acosta, Iraima, Mireia y a Tiago Marques
por los consejos estadísticos.
A la comunidad de Oyacachi, por permitirme estudiar el río Piburja y el río
Oyacachi y compartir su vida en mis estancias en el pueblo de Oyacachi. En
especial a la Familia Aigaje, al Pato, María, Adelaida, Patricio, Saúl y Saskita
por la generosidad y el cariño.
Al Departament d’Ecologia en especial a todos los compañeros de Tercer Ciclo:
Silvia, Salva, Pere, Mary, Oriol, Dani, Izaskun, Ainhoa, Eusebi, Biel, Jaimito,
Enric, Tania, Laia, Julio, Gemma, Carles, Octavi, Neus, Xavi, Rafa, Elena,
Ernesto, Gerardo. A Marina, Anna María, los profesores del Departament y a
todos los que me estoy olvidando… A Francesc Oliva del Departament de
Bioestadística de la UB por la ayuda con la estadística.
De todos los agradecimientos el más grande para mi familia, Patricia, Carlos,
Carla, Paula, Ana María, Nadia, Ada, Pancho, Nelly, por ser el apoyo más
grande, constante y entusiasta de todos los que he tenido. Por hacer de
ayudantes, sicólogos, de sponsors, de hostal, centro de reuniones, guías
turísticos, chóferes, taxi (a veces autobús), chefs, solucionadores de problemas
de todo tipo y un millón de cosas más que han hecho por mí, pero sobre todo
gracias por confiar en mí, por todo el apoyo, el cariño, los mimos y por ser un
referente de valores y “equipo”. A toda la Gran Familia, encabezada por el
abuelito Nube, sus fantásticos hijos (incluida la Amparito), nietos y bisnietos y
las diversas sucursales de México, Valle de Napa, Chicago, Argentina y
II
Agradecimientos
Uruguay. A la familia de Uruguay, al Rolo, la Ale, el Nico, Walter, Cora y
tooooda la familia que es imposible de escribir en un solo párrafo. Pero, quiero
mencionar especialmente a los que vinieron a ayudarme a Oyacachi en algún
momento: Papá, Mamá, Paula, Carla, Rolo, Ale, Gaby, Chabelo, Santi y Tibi.
También a la Clemen, la China y el Marcelino que me echan cables desde el
cielo.
A todos los amigos que han sido una familia aquí en Barcelona: Malicha,
Margarita, Luis, Alicia, Ana, Vero, Sole, Ramiro, Mireira, Iraima, Raúl (que ha
sido también como un “hermano” en esto de estudiar ríos Andinos), Núria Cid,
Sandra, Kenny, Yolanda, Lluis, Esther, Nico, Chivo, Paz, Lorenzo, Diego,
Marina, Eva, Betty, Mery, Marta, Patxi, Luciano, Tura, Fiona, Joan y todos los
que me olvido. Les agradezco a todos por hacer mi vida en Barcelona más feliz,
siempre los voy a recordar con cariño.
Quiero agradecer especialmente al Pau, por ser mi gran amigo, por cuidarme
cuando estuve enferma, hacerme reír cuando estoy triste, por ayudarme y
acompañarme en esta etapa de la vida.
Al Eric, por cruzar el mundo para estar cerca y ayudarme con el impulso final
para terminar esta tesis. Gracias por el amor y por ser un auténtico bicho.
A la Agencia Española de Cooperación Internacional para el Desarrollo y a
Education For Nature Program de WWF por financiar este proyecto de Tesis.
Barcelona, Octubre del 2008
III
Prefacio
La concepción de la idea de esta tesis se remonta a 1999 durante mi primer
trabajo en ríos, cuando estudiaba las poblaciones de Gasterópodos acuáticos
en una gradiente altitudinal en la cordillera de los Andes del Ecuador. Fue ahí
cuando nacieron las primeras incógnitas acerca de los factores que afectan la
distribución de las especies. Estas incógnitas fueron incrementando con mi
trabajo de Diplomado en Estudios Avanzados en Ecología, el cuál realicé en
condiciones de referencia y contaminación en ríos del valle interandino en
Ecuador. Posteriormente tuvimos la oportunidad de realizar dos acciones
conjuntas sobre estos temas, con la Universidad San Francisco de Quito y la
Universidad La Molina de Lima, en ríos de los Andes de Ecuador y Perú, cuya
intención era la de crear en base a la experiencia en ríos Mediterráneos del
grupo FEM (basadas en el proyecto ECOBILL y GUADALMED I y II),
herramientas de evaluación de la contaminación y el estado ecológico de los
ríos
Andinos. Durante estas acciones, pudimos apreciar la
diversidad que
tienen los ríos en los Andes y también visualizar la incesante degradación que
sufren, ríos de páramo, ríos de bosque montano, ríos secos o ríos que en lugar
de aumentar el caudal río abajo su caudal disminuye o desaparece y cauces
que están llenos de basura.
No sabemos cuántas especies acuáticas han desaparecido ni las que existen, y
el conocimiento de cómo funcionan los ríos en los Andes es incipiente y por lo
tanto tampoco sabemos cómo restaurarlos y protegerlos. Esta tesis nace de la
necesidad de comprender que pasa en estos ríos, y es tan solo una
pequeñísima parte de todo lo que nos queda por estudiar y entender de los ríos
Andinos. Tuve la oportunidad de intentar contestar algunas de las preguntas de
cómo funcionan estos ríos en este trabajo de investigación, en el Río Piburja,
localizado en el Territorio Kichwa de Oyacachi, tierra de ríos, bosques y gente
maravillosa, cuya diversidad espectacular y riqueza cultural y natural están
lamentablemente amenazadas por las grandes ciudades, que explotan y
muchas veces despilfarran sus recursos acuáticos, y por la ambición
individualista de las sociedades alejadas de los ritmos y el respeto a la
naturaleza.
V
INTRODUCCIÓN
Introducción general
Los Ríos Tropicales Altoandinos
La diversidad de ríos a nivel mundial es enorme y el estudio los procesos
ecológicos, biogeoquímicos, hidrológicos y geomorfológicos han proporcionado
un marco unificador en la visión de las características estructurales y
funcionales con compendios y revisiones, que nos presentan de forma más
ordenada la estructura y funcionamiento de los ríos, así podemos citar: Allan
and Castillo (2007), Giller and Malmqvist (1998), Cushing et al.(2006). En todos
estos trabajos que describen patrones generales y características de los
ecosistemas fluviales, la mayoría de ejemplos corresponden a zonas templadas
donde el estudio de los ríos se remonta a varias décadas mientras que los
ejemplos de las zonas tropicales son escasos. Solo muy recientemente se ha
publicado un libro especialmente dedicado a los ríos tropicales (Dudgeon 2008).
Al contrario de Europa y Norte América donde la mayoría de estudios se ha
realizado en ríos pequeños, en América del Sur los estudios de ríos se han
enfatizado en los grandes ríos como el Orinoco o el Amazonas (Sioli 1984,
Covich 1988, Lewis et al. 1995), pese a que existen algunas excepciones de
estudios en los ríos de mayor altitud (Turcotte and Harper 1982b, a, Flecker
1992, Flecker and Feifarek 1994, Jacobsen et al. 1997, Jacobsen and Encalada
1998, Jacobsen and Terneus 2001, Jacobsen 2003, 2004, 2005, Allan et al.
2006, Jacobsen and Marín 2007), pero la información sobre las cabeceras y ríos
de bajo orden de montaña tropicales es en general escasa (Jacobsen 2008). Es
en estos ecosistemas (ríos tropicales altoandinos) donde se centra el trabajo de
esta tesis doctoral
Los Andes del Norte, ubicados desde el sur de Venezuela hasta la depresión de
Huancabamba al norte de Perú (Figura 1), son una de las regiones con mayor
actividad tectónica del continente, con una compleja climatología y topografía
que
ha
favorecido
la
presencia
de
distintas
formaciones
vegetales
caracterizadas por un alto endemismo (Myers et al. 2000). En consecuencia, los
ríos altoandinos tienen formas y apariencias muy variadas donde podemos
encontrar ríos con fuertes pendientes y cascadas a pequeñas salidas de laguna
de sustrato suave (Jacobsen 2008). Pese a que nuestra idea de ríos tropicales
IX
Introducción General
está más asociada a climas cálidos y vegetación exuberante, los ríos de altitud
de los trópicos tienen más similitudes con los ríos de altitud de zonas templadas
que con ríos tropicales de tierras bajas (Jacobsen 2008). El régimen estacional
constituye una de las diferencias principales entre los ríos de altitud de zonas
templadas con los tropicales. La estacionalidad en el trópico se mide como
cambios en el régimen de lluvias y no como cambios de temperatura. Al
contrario que en zonas de alta montaña templadas, la variación de la
temperatura en zonas altoandinas es diaria y está relacionada a la radiación
solar, las cuales provocan fluctuaciones diarias de temperatura que pueden
oscilar entre los -11 a los 25 ºC (Guhl 1989) con promedios comprendidos entre
los 2 y 10 ºC (Luteyn 1999). Los regímenes de precipitación en las zonas
altoandinas suelen ser constantes y los cambios entre las épocas secas y
lluviosas son menos drásticos que en las zonas tropicales bajas. La
precipitación puede variar entre 500 y 3000 mm/año con una humedad relativa
entre el 25 y el 100% (Luteyn 1999, Ortíz 2003, Jacobsen 2008).
Figura 1. Andes del Norte. El límite norte son los Andes de Venezuela y el
límite sur está marcado en el paralelo 6º S en la depresión de Huancabamba.
Modificado de:http://www.ucm.es/info/agrygfdp/Web/Aula%20de%20cartograf%EDa/amsurfis.jpg
X
Pese a su extraordinaria diversidad, los ríos tropicales altoandinos son
probablemente uno de los ecosistemas menos estudiados del mundo (Ward
1994, Allan et al. 2006, Jacobsen 2008). En una reciente revisión sobre los ríos
de altura tropicales Jacobsen (2008) pone de manifiesto el escaso conocimiento
de estos ríos, tanto sistemático como ecológico. Por lo que sabemos hasta el
momento la composición de las comunidades de macroinvertebrados y su
distribución espacial en los ríos de los trópicos está más ligada a los regímenes
de caudal y perturbaciones hidrológicas que a otros factores ambientales
(Flecker and Feifarek 1994, Jacobsen et al. 1997, Jacobsen 2005, 2008). Los
patrones de la variabilidad temporal en el caudal de los ríos tienen efectos
profundos en la estructura y funcionamiento de los ecosistemas así como en las
dinámicas de macroinvertebrados acuáticos (Resh et al. 1988, Poff and Ward
1990). Estos cambios originan, junto con la variabilidad del sustrato, la compleja
disposición espacial de los macroinvertebrados, así como provocan cambios en
la deriva y en la disponibilidad de los recursos, como la materia orgánica
alóctona (Turcotte and Harper 1982a, Benson and Pearson 1987, Pearson et al.
1989). A su vez, estos cambios tienen profundos efectos en el funcionamiento
de los ríos y en procesos relacionados con la persistencia de las comunidades
de macroinvertebrados, como es la colonización del sustrato del río.
Adicionalmente, se ha establecido que parte
de la persistencia de estas
comunidades de macroinvertebrados en un sistema que sufre perturbaciones
impredecibles de caudal, está atribuida al hecho de que mayoritariamente las
especies de insectos acuáticos en los trópicos son multivoltinas (Jackson &
Sweeney 1995) con lo cual el reclutamiento de nuevos individuos que en zonas
templadas se da una vez por año, en el trópico sería un proceso continuo.
Sin embargo el conocimiento de las consecuencias de estos distintos procesos
involucrados en la persistencia de las comunidades de macroinvertebrados de
ríos altoandinos es incipiente mientras que urge obtenerlo ya que los ríos de
altura en los trópicos son unos de los ecosistemas más amenazados por el
calentamiento global y las perturbaciones antrópicas (Bradley et al. 2006,
Jacobsen 2008).
XI
En este contexto, en la tesis me centro en cuatro aspectos fundamentales de la
persistencia de las comunidades de macroinvertebrados de ríos altoandinos,
contemplando las diferencias estacionales de caudal: 1) la disposición espacial
de las comunidades bentónicas (microhabitats); 2) la dinámica de la deriva y el
proceso de recolonización; 3) el papel de la materia orgánica alóctona y 4) la
ovoposición de los adultos en el río. Estos son aspectos poco o nada
estudiados en los ríos altoandinos.
El Microhabitat
Los cambios de caudal en los ríos son fundamentales ya que pueden generar
una perturbación que crea nuevos espacios, cambia la disponibilidad de los
recursos o su variabilidad temporal (Pickett and White 1985), y ello
generalmente provoca que la biota se organice en parches dentro del río.
Cuando las perturbaciones por caudal, como las crecidas o sequías, ocurren
varias veces en la vida de un organismo, a una escala temporal corta, los
efectos negativos de las perturbaciones se pueden compensar por movimientos
hacia zonas menos afectadas por el flujo del caudal
(Lancaster 2000). La
presencia de estas zonas, permite que los organismos tengan más
probabilidades de subsistir y puedan recolonizar las zonas que han sufrido los
efectos más severos, asegurando la persistencia local de las especies
(Lancaster and Hildrew 1993b, a, Lancaster and Belyea 1997, Townsend et al.
1997). A éstas áreas se las denomina “refugios” (Caswell and Cohen 1991,
Townsend and Hildrew 1994, Lancaster and Belyea 1997). La importancia que
tienen éstos para las comunidades acuáticas es difícil de estudiar ya que en
situaciones de caudal normal los refugios pueden no ser distinguibles de otras
zonas o microhabitats, ya que las fuerzas hidráulicas a las que están sometidas
los diferentes sustratos del río varían con la descarga (Lancaster and Hildrew
1993b, a).
A nivel de comunidad, en los ríos tropicales, la subsistencia a cambios de
caudal impredecibles y recurrentes, puede estar mediada por adaptaciones en
su historia de vida, como tiempos de desarrollo o estrategias reproductivas,
XII
como por ejemplo el multivoltinismo. Los tiempos de desarrollo de los estadíos
acuáticos también pueden influir, siendo favorecidos los organismos con
tiempos de desarrollo más rápido como Chironomidae, Simuliidae o Baetidae.
Sin embargo, en especies presentes en el neotrópico con tiempos de desarrollo
más largo que, por lo tanto estarían más expuestas a los efectos negativos de
las crecidas, adaptaciones de comportamiento, como el uso de refugios podría
ser clave para la persistencia de las mismas. Estas interacciones entre las
perturbaciones y la presencia o importancia de los “refugios” han sido
estudiadas en zonas templadas (Lancaster and Hildrew 1993b, Winterbottom et
al. 1997, Winterbottom 1997, Lytle 1999, Lancaster 2000, Lytle 2001) sin
embargo se desconoce su ocurrencia en el neotrópico, pese a que su existencia
podría ser crítica en el mantenimiento de las comunidades.
En síntesis, en los ríos altoandinos, que como dijimos en un principio sufren
cambios de caudal rápidos e impredecibles, que sumados a la topografía
variable
(con
fuertes
pendientes)
pueden
provocar
también
fuertes
perturbaciones en el sustrato, se desconocen los mecanismos usados por los
distintos taxa presentes para recuperar las poblaciones luego de las crecidas. A
escala espacial, las reducciones antropogénicas del caudal provocan cambios
en la dinámica hidrológica del río y por lo tanto de las comunidades de
macroinvertebrados. Por esta razón en este capítulo también incluimos datos de
un tramo afectado por la desviación de agua para la cría de truchas, en el cuál
el caudal base se reduce en un 50 porciento. En este contexto las preguntas
que intento responder en el primer capítulo son: ¿Cuál es el efecto de la
variación a corto plazo y estacional del caudal en la comunidad de
macroinvertebrados y qué importancia potencial tienen los refugios y las
características del microhabitat en la persistencia de las mismas? ¿Cuál es el
efecto de la reducción del caudal en la comunidad de macroinvertebrados?
Deriva y recolonización
La recolonización del sustrato, que sigue a las perturbaciones, es uno de los
procesos más importantes que estructura las comunidades (Boyero and Bosch
XIII
2004). Éste suele ser un proceso que depende principalmente de los individuos
que llegan con la deriva, pero también de las migraciones contracorriente de
algunos macroinvertebrados, las comunidades de parches adyacentes y del
reclutamiento por ovoposición
(Williams and Hynes 1976). Este proceso está
influenciado por las características del sustrato y recursos alimenticios
asociados así como la competencia, la depredación (Resetarits 1991, Mackay
1992, Resetarits 2001) y la recurrencia mayor o menor de las perturbaciones.
La composición de la deriva afecta profundamente la dinámica de las
comunidades bentónicas, por un lado disminuyendo la densidad de las especies
más susceptibles a derivar y por otro lado proveyendo continuamente
organismos para la colonización del bentos (Townsend and Hildrew 1976). En
zonas templadas el papel de la deriva en la recolonización es un proceso que
ha sido muy estudiado y documentado (Brittain and Eikeland 1988, Mackay
1992). En los trópicos se han realizado pocos estudios, algunos de ellos muy
recientes, que han revelado que la deriva en estos sitos no es estacional y está
más ligada a cambios de caudal (Pringle and Ramirez 1998, Ramirez and
Pringle 2001, Jacobsen and Bojsen 2002, Rodriguez-Barrios et al. 2007).
De los estudios recientes sobre deriva en los trópicos sabemos que a nivel local
la recolonización depende de la migración de los parches adyacentes, pero a
nivel de secciones de río depende de la deriva de zonas más lejanas (Boyero
and DeLope 2002, Boyero and Bosch 2004, Melo and Froehlich 2004). Sin
embargo la mayoría de estos estudios se han realizado solo en una estación,
generalmente la de menor lluvia, por lo tanto los cambios de composición de la
deriva asociados al caudal y su influencia en la recolonización no ha sido
abordada. Tampoco se sabe que taxa son más proclives a derivar. Además
ríos tropicales que no poseen poblaciones de peces locales, pero que sin
embargo poseen especies de peces introducidas (p.e. varias especies de
truchas), han demostrado carecer de periodicidad en la deriva (Flecker 1992,
Jacobsen and Bojsen 2002, Jacobsen 2008).
Las interrogantes respecto a la deriva y a la recolonización, en el contexto de
los ríos de alta montaña tropicales con perturbaciones de caudal impredecibles
XIV
y a su vez temperatura media de 10 ºC, son muchas. Entre otras preguntas que
hemos intentado contestar en el capítulo dos de esta tesis están: ¿Hay
periodicidad en la deriva? ¿Cuál es la variación estacional en la composición de
la deriva y que taxa está siendo más propenso a ser arrastrado por la misma?
¿Cómo afectan los patrones de deriva en la recolonización del sustrato? ¿Se
diferencia la recolonización entre estaciones y microhábitats distintos?
Importancia de la materia orgánica.
La entrada de materia orgánica alóctona en forma de hojarasca constituye la
principal fuente de energía y nutrientes en los ríos de cabecera (Siccama et al.
1970, Anderson and Sedell 1979, Wallace et al. 1997, Graça 2001). Parte de
esta energía es aprovechada por los consumidores y otra es exportada río
abajo por acción de la corriente (Likens et al. 1970, Fisher and Likens 1973) y
por ello existe vínculo trófico entre las comunidades de los ríos de cabecera y
los tramos más bajos de los ríos que está mediado por los descomponedores y
detritívoros (Graça 2001). En zonas templadas la interacción entre las entradas,
los descomponedores y los detritívoros y los factores físicos ha sido
ampliamente estudiada (Anderson and Sedell 1979, Suberkropp and Wallace
1992, Suberkropp and Chauvet 1995, Wallace and Webster 1996, Wallace et al.
1997, Graça 2001, Graça et al. 2001b).
Al contrario, en los ecosistemas
tropicales la información cuantitativa respecto a entradas, almacenamiento y
ciclos de estas entradas se desconocen pese a que usualmente los ríos
tropicales drenan áreas con densas vegetaciones y las entradas alóctonas
probablemente son la principal fuente de energía (Graça et al. 2001a, ColónGaud et al. 2008). Debido a la escasa estacionalidad de los ríos tropicales (en
términos de temperatura), estas entradas son potencialmente continuas durante
todo el año, al contrario que en ríos de zonas templadas que reciben la mayor
parte de entradas en otoño. Sin embargo, en el trópico y en los estudios
realizados hasta el momento, los máximos de entrada de materia orgánica
están asociados a tormentas (Covich 1988), o al comienzo (Gonçalves et al.
2006) o final de la estación de lluvias (Afonso et al. 2000).
XV
La capacidad de los ríos de retener estas entradas está relacionada a la
morfología y heterogeneidad espacial del canal del río (Speaker et al. 1984,
Lamberti et al. 1989) y en general los ríos más angostos de fondo rugoso son
más retienen más material (Mathooko et al. 2001). Debido a la inestabilidad del
caudal, en los ríos tropicales la acumulación de la material alóctona en el fondo
del río no solo está regulada por las entradas del bosque adyacente sino
también por la magnitud y variabilidad de la descarga (Pearson et al. 1989). En
situaciones de caudales altos, la capacidad retentiva disminuye y aumenta el
movimiento y redistribución de la hojarasca en el fondo del río (Larrañaga et al.
2003). El patrón general en el trópico es que mayores cantidades de hojarasca
se acumulen durante las épocas secas (Covich 1988), mientras que la cantidad
2
acumulada anual puede ser muy variable p.e de 35 g AFDM/m (Friberg et al.
2
1997) a más de 1000 g AFDM/m (Colón-Gaud et al. 2008).
Los estudios de zonas templadas han demostrado que la relación de esta
materia alóctona acumulada en el río puede ser usada por la comunidad de
macroinvertebrados como refugio (Palmer et al. 1996), o como recurso
alimenticio por parte de los trituradores (Wallace and Webster 1996, Graça
2001, Graça et al. 2001b). Al contrario que en zonas templadas, en el trópico los
trituradores al parecer no son una parte importante de la comunidad (Ramirez
and Pringle 1998, Mathuriau and Chauvet 2002, Mathuriau et al. 2008), pero
existe alguna evidencia de que en los ríos tropicales de montaña pueden tener
más relevancia (Cheshire et al. 2005). Además el análisis de contenido del
tracto digestivo de insectos acuáticos en ríos tropicales han demostrado que la
mayoría de especies explota dos niveles tróficos, o se alimentan de más de un
tipo de recurso (Wantzen et al. 2005, Tomanova et al. 2006, Wantzen and
Wagner 2006). En este sentido, la importancia de la materia orgánica alóctona
(hojarasca) como recurso alimenticio en los ríos tropicales de montaña es
incierta, y probablemente ésta dependa de los recursos disponibles a nivel local,
ya que al parecer los macroinvertebrados tropicales parecen tener hábitos
alimenticios flexibles (Covich 1988).
XVI
Para abordar este tema en el contexto de los ríos altoandinos, las preguntas
que intentamos responder en el capítulo tres fueron: ¿Cuál es la importancia de
las entradas de materia alóctona en un río altoandino? ¿Hay algún efecto
estacional en el transporte, retención de ésta materia orgánica y qué implica
para las comunidades de macroinvertebrados? ¿Qué taxa se están alimentando
de este recurso?
La Ovoposición
La ovoposición tiene extrema importancia en los ecosistemas acuáticos, ya que
establece el tamaño inicial de las poblaciones, y además constituye uno de los
principales procesos de recolonización (Williams and Hynes 1976, Encalada
and Peckarsky 2006). En ríos con caudales que fluctúan ampliamente de forma
impredecible, como la mayoría de ríos tropicales, la deriva catastrófica y la
mortalidad de las larvas puede ser muy importante. En este contexto, el
reclutamiento de nuevos individuos por ovoposición podría ser un proceso
fundamental para la estabilidad y resiliencia de éstas comunidades.
El papel del reclutamiento en las comunidades acuáticas es poco conocida
(Encalada et al. en prep.), y la poca información que existe sobre ovoposición
está restringida al hemisferio norte (Hinton 1981, Brittain 1989, Merritt and
Cummins 1996). Sin embargo, es de esperar que en zonas templadas, sea solo
importante durante los meses de primavera y verano, cuando los insectos que
han emergido para la reproducción vuelven al agua a depositar sus huevos. La
estrategia del univoltinismo favorece que los estadios aéreos ocurran cuando
las condiciones para la reproducción y ovoposición son más favorables. Las
zonas tropicales por el contrario carecen de esta marcada estacionalidad en la
temperatura y en este contexto, el multivoltinismo es una estrategia que puede
ser altamente beneficiosa para la persistencia de las comunidades. Esta
estrategia ha sido reportada para varios insectos tropicales (Collier and Smith
1995, Jackson and Sweeney 1995) principalmente de altitudes bajas.
XVII
Por el momento los datos de historia de vida y ovoposición de insectos andinos
de alta montaña son inexistentes (Jacobsen 2008). Sin embargo, conocer este
aspecto es fundamental para entender la persistencia de las comunidades en
las condiciones ambientales más restrictivas de la alta montaña, como es la
temperatura media más baja (~10ºC en Ecuador) y donde la saturación de
oxígeno está afectada por la altitud (Jacobsen 2008).
En este contexto, las interrogantes en este tema fueron: ¿Cuál es la importancia
del reclutamiento por ovoposición en ríos de altitud? ¿Hay
efecto de la
estacionalidad hidrológica? ¿Qué relación tienen los patrones observados en
las formas acuáticas (bentos y deriva) con los observados en los estadios
adultos? En el cuarto capítulo de esta tesis intentamos responder a estas
preguntas.
XVIII
Objetivos
El objetivo general de esta tesis es entender como los procesos que operan en
los ríos altoandinos influencian en la persistencia de las comunidades de
macroinvertebrados. Las diferentes incógnitas que nos hemos planteado se
tratan en cuatro capítulos independientes, que tratan de responder a la pregunta
más general planteada desde diversas aproximaciones. Los capítulos están
escritos en inglés y de tal forma que se aproximen al formato que tendrán en su
posible futura publicación.
Los objetivos de cada capítulo son:
1.1 Caracterizar el régimen hidrológico en un río tropical altoandino en varios
días consecutivos durante dos estaciones diferentes (seca y húmeda).
1.2 Determinar si existe variación en la comunidad de macroinvertebrados
entre estaciones, microhábitats (rápidos vs. refugios) o tramos (caudal
alterado vs. caudal natural).
1.3 Determinar la relación entre la comunidad de macroinvertebrados y la
estabilidad del caudal.
2.1 Describir la deriva diaria y estacional.
2.2 Estudiar las diferencias estacionales en la deriva y estimar la propensión de
los distintos taxa de macroinvertebrados a derivar en cada época.
2.3 Describir el proceso de colonización a corto plazo (1 a 7 días) en
condiciones de caudal base.
2.4 Estudiar la dinámica de la recolonización a medio plazo (1 mes) en cada
época y entre épocas.
2.5 Determinar la importancia del microhábitat en el proceso de recolonización.
3.1 Describir las entradas anuales de hojarasca en un río tropical altoandino.
3.2 Definir si hay diferencias estacionales en el transporte y retención de la
Materia orgánica particulada gruesa (MOPG).
XIX
3.3 Estudiar las diferencias espaciales y estacionales de la materia orgánica
alóctona acumulada en el río
y sus potenciales relaciones con la
composición de la comunidad de macroinvertebrados.
3.4 Evaluar si los recursos alóctonos son importantes en la dieta de los
macroinvertebrados del río mediante el análisis de contenidos intestinales.
4.1 Estudiar las diferencias estacionales en la riqueza y abundancia de adultos
de insectos acuáticos volando y ovopositando.
4.2 Evaluar si existen diferencias entre el número de masas huevos y la
identidad de éstas y su morfología entre las dos épocas estudiadas.
4.3 Estudiar si existe algún tipo de preferencia de las hembras en el sustrato
usado para ovopositar.
4.5 Relacionar la identidad de las larvas presentes en el río con la de los
adultos encontrados en las mismas estaciones.
XX
CHAPTER 1:
Macroinvertebrate assemblages at an
Andean high altitude stream: Importance
of microhabitat, flow and season
CHAPTER 1: Importance of microhabitat, flow and season
Introduction
The water discharge in fluvial ecosystems, varies widely in terms of its quantity,
quality, timing, and temporal variability (Allan and Castillo 2007), and is largely
determined by natural variations in climate, vegetation, geology, and terrain (Poff
et al. 1997, Allan and Castillo 2007). At the same time, the patterns of temporal
variability in river flows have profound impacts on the structure and function of
fluvial ecosystems. But in addition to its ecosystem level effects, variations in
water discharge are also part of the natural disturbance patterns of streams,
having a key influence on the structure and dynamics of biological communities
(Resh et al. 1988, Poff and Ward 1990). From this point of view, the occurrence
of disturbances is fundamental in terms of creating “open spaces” and changes
in resource availability or influencing the temporal dynamics (Pickett and White
1985) that often result in patchy distributions of stream biota. Moreover, extreme
events like floods are also a primary selective pressure for adaptation because
of the high mortality rates that they can inflict in some aquatic taxa (Lytle and
Poff 2004).
The spatial and temporal variation in disturbance regimes in streams determines
which mechanism operates to avoid the effects of disturbance on biological
communities. At large multigenerational scales, external recruitment is the main
mechanism that maintain populations, while at smaller scales, during the lifetime
of an individual, organisms may avoid the negative effects of disturbance by
moving between patches not affected by the disturbance and ensuring the local
persistence of the species (Lancaster 2000). In such situations, and specially if
flow disturbances occur at time intervals of less than one generation, the
persistence of one species might depend on the availability of areas of the river
where physical characteristics provide the organisms with a greater probability of
subsisting
through
periods
of
severe
conditions,
facilitating
further
recolonization of patches where the effects of disturbance have been more
severe (Lancaster and Hildrew 1993b, a, Lancaster and Belyea 1997, Townsend
et al. 1997). These areas that are less susceptible to the impacts of disturbance
are called “refugia”. Their existence has been suggested by empirical and
3
theoretical evidence that emphasize the importance of temporal and spatial
patterns of habitat heterogeneity (Caswell and Cohen 1991, Townsend and
Hildrew 1994, Lancaster and Belyea 1997). However, the importance of such
refugia in stream communities is difficult to assess because, in situations of low
or basal flow, a refugium patch might be undistinguishable from other patches,
as the hydraulic forces vary with the discharge (Lancaster and Hildrew 1993b,
a).
In Neotropical rivers, severe and unpredictable hydraulic changes are a major
source of disturbance and play an important role structuring aquatic
communities (Flecker and Feifarek 1994). At the community level, the net effect
of these disturbances is mediated by natural history attributes of the species
such as reproductive strategies and developmental times. For example,
multivoltinism among many neotropical species is a major factor influencing the
persistence of aquatic macroinvertebrate communities. Similarly, developmental
times of aquatic stages vary widely among different taxa, determining either less
susceptibility to hydraulic disturbance for macroinvertebrates with faster
developmental times (e.g. Chironomidae, Baetidae), or development of
morphological
or
behavioral
adaptations
for
organisms
with
longer
developmental times such as Trichoptera or Plecoptera. Even though the
influence of these life history strategies in tropical streams is poorly understood,
they might be critical in terms of determining the ability of a species to withstand
the effects of disturbance or to avoid them through the use of refugia. Although
previous studies have addressed similar interactions in temperate (Lancaster
and Hildrew 1993b, Winterbottom et al. 1997, Winterbottom 1997, Lancaster
2000) or desert areas (Lytle 1999, Lytle 2001) little is known about their
occurrence in tropical regions, where other factors maybe influencing the use of
refugia at different scales.
Tropical high-altitude streams in the Andean region are characterized by rapid
flow changes that are not easily predictable from seasonal variations in
precipitation. Moreover, the occurrence of spates might be a crucial form of
disturbance in these ecosystems, as the topography of the region already
4
determines fast flow regimes, turbulence, and high substrate instability. The
importance of this type of disturbance in tropical mountain streams was already
suggested by Jacobsen (2005), who reported high temporal variability in faunastone relationships related to channel stability in the Ecuadorian Andes.
However, how taxa persist after spates and the consequences of anthropogenic
flow regulation of these high flow-variable streams remain poorly studied.
In this context, our aim is to provide a further insight into the effects of short-term
(hours and days) and mid-term (seasonal) variations in flow, on the
macroinvertebrate communities of a high altitude tropical stream, assessing the
potential importance of refugia on the persistence of these communities. More
specifically, we intended to: 1) characterize the hydrological regime in a high
altitude tropical stream in several consecutive days during two different seasons
(dry vs. wet); 2) determine whether macroinvertebrate community composition
differs between seasons (dry vs. wet), microhabitats (exposed areas [rapids] vs
potential refugia) and reaches (normal vs. anthropogenic reduced flow; and 3)
determine whether there is a relationship between macroinvertebrate community
composition and flow stability. In addition comparing a section of stream with
minor human disturbance, with a contiguous section which flow has been
reduced, we may assess the importance of human induced disturbance on the
relative importance of refugia in determining community structure and dynamics
in this ecosystem.
Study area
We performed the study at the Piburja stream (3300 m a.s.l.), a first order
stream in the Oyacachi river basin, situated at the Cayambe-Coca Ecological
Reserve (RECAY) in Ecuador (0º12’ 44.70”S, 78º 04’ 40.27”W). Oyacachi basin
is located at the slopes of the Andean Range and is covered by several
vegetation types from high altitude tropical grasslands (páramos) to tropical rain
forest (~4000 to 1800 m). This watershed has a western-eastern slope, and is
part of the Amazon basin (Terneus & Vásconez, 2003). The Oyacachi river
starts in the upper slopes of the cordillera, running through páramo grasslands
dominated by Calamagrostis spp. and Neurolepis sp., and mixed forests of
5
Polylepis sp. and Gynoxis sp.; in its middle ranges, the river flows through an
ever-green forest dominated by Alnus acuminata in the river shores and in the
disturbed areas (Baez et al., 1999). Total precipitation in the town of Oyacachi is
~ 1600 mm/year, and daily mean temperature is ~10 ºC. Pluviometric records
(~10 years) from nearby localities show lower rainfall from December to
February compared to a more wet period from May to September. Precipitation
is 2 or 3 times greater than the potential evapotranspiration, which makes this
area very humid throughout the year (Skov 1999).
We studied two 70m-long reaches at Piburja stream: The pristine reach had a
minimum width of 1.6 m during the dry season, and a maximum of 4.0 m during
the Wet season. The surrounding vegetation is an evergreen forest dominated
by Andean alder (Alnus acuminata) and a dense understory with high diversity of
Melastomataceae, Asteraceae, and Rosaceae. The disturbed reach is located
250 meters downstream from pristine reach. Its discharge is reduced to 50% of
the upstream water at basal flow due to an anthropogenic diversion for trout
rearing pools. Parts of this area have been transformed to pastures, but the
stream shores still preserve the forest cover similar to that of the undisturbed
reach. During spates, the reduction of flow is negligible. Flow measurements
during sampling period are provided in the results section.
Materials and Methods
To describe and characterize stream flow, aquatic invertebrate communities,
and the relationship between them, we measured these parameters at three
different scales: 1. Temporal scale (dry and wet season); 2. Reach scale
(pristine and disturbed section) and 3. Within the reach, microhabitat scale
(rapids and refugia).
Flow and environmental variables
To describe and characterize flow stability, we sampled discharge during two
hydrologically distinct periods: wet and dry. The wet season was studied from
6
April through May 2006, and the dry season from January through February
2007.
Flow was measured using a Global water flow probe model FP201 making a
minimum of two daily measurements during the sampling periods in a fixed
transect at each reach. Detailed (every ten minutes) flow variation was
measured using a GW Water Logger model WL16 located at the pristine
transect, which was calibrated using data from the fixed transect. In addition to
instantaneous flow we calculate an index of flow stability (INEST), by dividing
the cumulative flow of 12, 24, and 48 hours before sampling by the variance of
the flow at the same time intervals.
We measured conductivity, pH, temperature, and oxygen a minimum of 20 times
during each sampling period (between 6h00 and 17h00), using a 4-Star Orion
pH/conductivity portable meter and a WTW oxygen meter. The proportion of
benthic substrates (i.e. boulders, cobbles, gravel or sand) was measured at ten
cross-section transects at each reach, in addition to depth and water velocity
which were measured every 20 cm along each of these transects.
The substrate movement was measured by the movement of three big boulders,
and 12 painted cobbles at each site. During the sampling period boulders did not
move at any spate event, while cobbles moved only moderately in the same
reach, with movements raging from less than 10 m at the dry season, to a
maximum of 20 m during the wet season. These data suggest that no large
catastrophic floods, eroding the substrate, occurred during our sampling periods.
Macroinvertebrates
To understand the relationship between flow measurements, microhabitat and
macroinvertebrate community composition we used a small Surber sampler
(14x14 cm). This sampling device (opposed to the typical Surber sample of
20x20 cm) allowed us to discriminate between the two microhabitats: fast flow
areas (rapids) and areas with lower or null water current velocity (potential
refugia). We sampled on five different dates at each season, with an interval of
7
four to six days between sampling occasions. To avoid damage of sites within
our reach we always sampled moving upstream. At wet season we sampled
before, during, and after spates, guided by the level of a rain gauge located at
Oyacachi town. We collected six samples at each microhabitat condition (rapid
or refugium) during the wet season, and ten during the dry season, with a total of
30 and 50 samples in the wet and dry season respectively, for each microhabitat
condition (rapid vs. refugium) and reach (pristine vs. disturbed). In order to
distinguish between rapids and refugia, we measured mean velocity at each
Surber location. Species accumulation plots suggested that there was not effect
in taxa richness or macroinvertebrate density collected using six or ten samples,
therefore data can be compared directly.
Samples were preserved in 10% formalin and taken to the laboratory for
identification. All samples were analyzed under a stereomicroscope to
differentiate
small
invertebrates,
especially
small
dipterans.
For
macroinvertebrate identification we used the keys of: Holzenthal and Flint
(1995), Merritt and Cummins (1996), Roldán (1996), Fernandez & Dominguez
(2001) and Dominguez et al. (2006).
Data treatment
We calculated descriptive statistics for all physico-chemical variables and used
the test of Kruskal-Wallis to detect significant differences among reaches and
seasons. This analysis was necessary because not all the environmental
variables satisfied the homogeneity of variance requirement even when
transformed.
We identified invertebrates to genus level when possible (and subfamily level for
Chironomidae), and calculated several community metrics: richness (S),
abundance, density, Fisher’s alpha diversity index, and Shannon-Weiner
diversity index (H’). Additionally, with some invertebrate groups we conducted
individual presence/absence analyses, or measured changes in their density in
the different treatments.
8
Macroinvertebrate densities for each treatment were calculated excluding rare
taxa (present in <10% of samples and <0.5 % of total density) and then log(x+1)
transformed prior analysis. With this data matrix, we performed a Bray-Curtis
similarity index to construct a NMDS (Non-metric Multi-Dimensional Scaling) to
compare communities among microhabitat, reaches and seasons.
We used a General Linear Model (GLM) to assess differences in community
metrics and taxa densities in microhabitats at both, space (reach) and time
(season) scales. Fixed categorical factors in the GLM were Season (wet and
dry), Reach (disturbed and undisturbed) and microhabitat condition (rapids and
refugia). Total taxa (S), total abundance (N), Fisher’s α and Shannon (H’)
2
diversity indexes and taxa densities (ind/m ) were analyzed as dependent
variables. GLM was performed with pooled data (two seasons together) and for
each season separately (GLM was performed only with reach and microhabitat
condition as fixed factors). This was done to identify if the factors that are more
related to the community metrics differed between seasons.
To analyze the relationship between macroinvertebrates, stream flow, and the
index of flow stability 12 hours before sampling, we performed correlations and
linear regression analyses with all variables log transformed. Flow stability
indexes after 24 and 48 hours were not included in the analysis after preliminary
tests showed that flow stability 12 hours before sampling were more associated
with taxa densities and also that this index was highly correlated with the 24 and
48 hours before sampling flow stability indexes. We also made Logistic binary
regression analysis, to relate flow and flow stability to taxa presence and to
obtain the probabilities of taxa occurrence in different flow conditions.
Bonferroni correction of p-values was applied to all statistical test, but corrected
values were analyzed with caution due to the unnecessarily conservative
restrictions of these corrections (Gotelli & Ellison, 2004) in multiple tests.
Statistical analyses were performed using Statistica 6, Primer 6.0 and SPSS 14
statistical software.
9
Results
Flow stability and hydrological regimes.
The hydrological regime was clearly different between wet and dry periods (Fig.
1), with large differences in mean flow, as well as in its minimum and maximum
values, at both reaches and hydrological periods studied (Table 1). Flow was
significantly higher in the wet than in the dry season, and significantly higher in
the pristine than disturbed reaches (Appendix 1).
“Rapid zones” were
significantly faster than “refugia zones” (Kruskal Wallis H= 91.37, test p < .001)
and the velocity measures at each microhabitat type where similar at both
seasons (Fig. 1).
Table 1. Mean, minimum and maximum values of flow for the pristine and the
disturbed reaches during wet and dry season at Piburja stream, Ecuador.
Reach
season
Mean flow
(l/s)
95% Confidence
intervals.
Pristine
Dry
Wet
Dry
Wet
35,15
92,65
16,66
43,66
± 0,54
± 0,92
± 0,23
± 0,85
Disturbed
Minimum Maximum
6,49
48,17
2,01
12,86
117,30
355,78
52,59
400,90
The mean current velocity at the reach scale, was significantly higher in wet
season in both reaches. Depth presented differences at disturbed reach
between seasons, with more depth at the wet season, and also between
reaches at both seasons with deeper measurements at the pristine reach
(Appendix 1).
Seasonal changes of environmental conditions
Temperature varied significantly between seasons, and between reaches during
the dry season. Water temperature was significantly higher during the wet
season, compared to the dry season, and also was consistently higher in the
disturbed reach, compared to the pristine site (Appendix 1). Only at wet season
the temperature between reaches did not vary significantly. The pH and
conductivity was different between seasons, but not between reaches, being
lower at wet season. Although mean pH values were significantly lower at wet
10
season, (Kruskal-Wallis analysis (Appendix 1)), there was a high overlap of
values of pH. Finally the substrate composition structure did not differ between
reaches (R=-0.009, ANOSYM analysis).
1a
WET
DRY
400
flow l/s
300
200
Prisitine
Disturbed
100
0
1b4000
April 06
May 06
Jan 07
Feb 07
INEST 12
3000
2000
1000
0
Disturbed
Pristine
2
suber velocity m/s
Pristine
1c 2
1
1
0
Disturbed
0
RAP REP RAD
RED
RAP
REP RAD RED
Figure 1. Hydrological changes between seasons, reaches and microhabitats at
Piburja stream. 1a) changes in flow; 1b) Changes in flow stability INEST 12
th
th
hours before sampling (Box plots, with 75 and 25 percentile); 1c) differences
on near bed velocity between at each Surber sample site (microhabitats). RAP:
Rapids pristine site; REP: Refugia pristine site; RAD: Rapids disturbed site;
RED: Refugia disturbed site (Box plots, with percentile 75 and 25).
11
Factors explaining the macroinvertebrate community changes
The results of MDS (Non-metric Multi-Dimensional Scaling) with taxa densities
using Bray Curtis similarity index, showed that macroinvertebrate communities
differ between seasons (Fig. 2). Season is the main factor explaining the strong
differentiation of community assemblages, independent from the reach (Fig. 2a).
At the dry season, the community was more homogeneous (in terms of
composition) compared with the wet season. Additionally, microhabitat (rapid or
refugium) was an important factor determining communities at each season (Fig.
2b), while reach was not an important factor determining the community
composition.
2a
2b
Figure 2. MDS using Bray Curtis similarity index with taxa densities
(transformed to log x+1) of invertebrates sampled in Piburja stream Oyacachi.
Dry and Wet seasons are plotted with squares (dry) and rhombus (wet) and
samples are identified with a letter as p (pristine reach), d (disturbed reach) and
in 2a, ra (rapid microhabitat) or re (refugium microhabitat) in 2b.
12
Seasonal changes of macroinvertebrates biodiversity and abundance
Temporal scale
We performed several GLM analyses to test if macroinvertebrate biodiversity
and abundance (S, N, H’, Fishers’α and taxa densities) differed between
analyzed factors (season, reach, microhabitat). Season was the most important
factor
generating
differences
both
in
community
metrics
and
in
macroinvertebrate densities for the majority of taxa being higher in dry season
(Fig. 3a, Appendix 1). The exceptions were Macrelmis and Baetodes (Fig. 3f)
that presented the opposed pattern being denser at the wet season. Few taxa
did not suffer any seasonal significant density variation. The complete results of
the effects of the factors: season, reach and microhabitat on the taxa densities
(log transformed) and community metrics of Piburja stream in the General Linear
Model can be found at the Appendix 2.
Reach scale
We did not found any difference on biodiversity between reaches, but
differences in densities of certain taxa were found (Table 2, Fig. 3b & 3c). The
taxa Hyalella, Tanytarsini, Oligochaeta and Tricladida were significantly denser
at the disturbed section of the stream (pattern of Fig. 3c) while Cochliopsyche,
Ochrotrichia, and Claudioperla shown a density increase at the pristine reach
(pattern of Fig. 3b). These tendencies were maintained during the dry season,
where Cocliopsyche and Nectopsyche were also denser at the pristine reach.
On the other hand, Tanytarsini, Andesiops, Baetidae, Oligochaeta and Tricladida
had higher densities at the disturbed site in the dry season (pattern of Fig. 3c).
During the wet season Oligochaeta and Tricladida showed a significant
difference of density, being higher at the disturbed site (pattern of Fig. 3c).
13
Figure 3. Relation of community metrics with season, reach and microhabitat of
invertebrates sampled at Piburja stream in Oyacachi, Ecuador. These graphs
show all the different patterns that we found for different taxa and for measured
metrics. The complete list of taxa and community metrics that followed these
patters are shown in table 2. All values are means (± 1 SE). a. metrics that
show an increase at dry season. b. metrics that show an increase at dry season
and at the pristine reach. c. metrics that show an increase at dry season and at
the disturbed reach. d. metrics that show an increase at dry season and in
refugia. e. metrics that show an increase at dry season and in rapids. f.
Baetodes increasing density at wet season and rapids (same pattern of
Macrelmis).
14
Microhabitat scale
Total abundance of Prionocyphon, Hyalella, Ceratopogonidae, Tanytarsini,
Cocliopsyche and Nectopsyche were higher at refugia (pattern of Fig. 3d) while
Simuliidae, Leptohyphes, Hydrobiosidae, and Ochrotrichia showed higher
densities at rapids. Baetodes (Table 2, Fig. 3f) and Baetidae show this increase
in rapids. This tendency generally remains during either wet or dry season.
During the dry season Heterelmis, Prionocyphon, Chinoronominae and
Hydracarina also showed significant higher densities at refugia (pattern of Fig.
3d). At wet season Shannon index was significantly higher in rapids.
Table 2. List of metrics and taxa that follow any of the different patterns
described in Figure 3, related to stream, reach or microhabitat characteristics in
the General Linear Model analysis.
TAXA / community
metrics
Fisher’s α
H’
Heterelmis
Neocylloepus
Macrelmis
Total Elmidae
Prionocyphon
Probezzia
Atrichopogon
Diamesinae
Podonominae
Tanytarsini
Chelifera
Psychodidae
Limoniidae
Simuliidae
Andesiops
Baetodes
Leptohyphes
Hyalella
Hydracarina
Oligochaeta
Andiperlodes
Claudioperla
Total Gripopterygidae
Hydrobiosidae
Cocliopsyche
Ochrotrichia
Nectopsyche
Tricladida
Corresponding
Figure 3
3a
3a
3a
3a
3f
3a
3d
3a, 3d
3a, 3d
3a
3a
3c
3a
3a
3a
3e
3a
3f
3a, 3e
3c, 3d
3a
3a, 3c
3a
3b
3a
3e
3b, 3d
3a, 3b, 3e
3a, 3d
3a, 3c
15
The relation of flow and flow stability with macroinvertebrate fauna
For the majority of the taxa, densities were inversely correlated with flow
independent from season. Only Macrelmis showed a positive correlation with
flow (Appendix 3). Andiperlodes and Hydrobiosidae, also showed significant
positive correlations with flow stability (12 hours before sampling). We did not
find a predictive function between the taxa densities and the flow or flow stability
suggested by the significant lack of fit of the test for Linear and polynomial
regression models. This was probably because of high densities found on the
dry season and the lack of a clear pattern in wet season
Logistic binary (presence- absence) regressions (Table 3) revealed a negative
slope (B) for 13 taxa, which means a decrease of the probability of the presence
of these taxa when flow increases (Fig. 4). All taxa that showed this relationship
with flow were also significantly denser at the dry season. Only Andiperlodes
had positive slope with flow stability (12 hours before sampling), and Tricladida
showed
a
decrease
of
presence
while
the
stability
increases.
Figure 4. Probability of presence of Diamesinae at Piburja stream with flow,
predicted by logistic binary regression. This same pattern of decreasing
probability of presence as flow increase, were followed by the 13 taxa shown in
table 3.
16
Table 3. Summary of Logistic Binary regression with forward LR Likelihood ratio
test of taxa presence or absence between flow and flow stability (only dependent
significant metrics are included) at Pibujra Stream, Ecuador. B= values for the
logistic regression equation for predicting the taxa presence from the
independent variables (flow and flow stability). The prediction equation is
log(p/1-p) = b0 + b1*x1 + b2*x2 ; S.E. = standard errors associated with the
coefficients; where b0= Constant. Wald and P.= Wald chi-square value and 2tailed p-value used in testing the null hypothesis that the coefficient (parameter)
is 0. Exp(B) = the odds ratios for the predictors.
Variables in the logistic equation
B
S.E. Wald df
P
Taxa
Heterelmis
Neocylloepus
Elmidae
Diamesinae
Podonominae
Simuliidae
Andesiops
Exp(B)
flow
-0.02
0.01
6.075
1
0.014
0.983
Constant
2.032
0.67
9.12
1
0.003
7.628
flow
-0.05
0.01
10.45
1
0.001
0.955
Constant
2.772
0.84
10.96
1
0.001
15.983
flow
-0.02
0.01
3.762
1
0.052
0.983
Constant
3.557
1.13
9.952
1
0.002
35.068
flow
-0.03
0.01
11.02
1
0.001
0.966
Constant
2.368
0.73
10.47
1
0.001
10.673
flow
-0.02
0.01
8.673
1
0.003
0.976
Constant
1.618
0.62
6.833
1
0.009
5.042
flow
-0.02
0.01
5.334
1
0.021
0.985
Constant
2.072
0.68
9.188
1
0.002
7.942
flow
-0.02
0.01
5.351
1
0.021
0.985
Constant
1.29
0.58
5.024
1
0.025
3.633
Hydracarina
flow
Constant
-0.02
3.144
0.01
0.96
4.959
10.8
1
1
0.026
0.001
0.982
23.198
Andiperlodes
inest12
Constant
0.001
-1.91
0
0.52
6.167
13.44
1
1
0.013
0
1.001
0.148
Cochliopsyche
flow
Constant
-0.02
1.549
0.01
0.6
7.171
6.591
1
1
0.007
0.01
0.981
4.706
Hydrobiosidae
flow
-0.03
0.01
7.544
1
0.006
0.975
Ochrotrichia
Constant
flow
3.663
-0.02
1.09
0.01
11.37
7.201
1
1
0.001
0.007
38.978
0.981
Constant
2.217
0.7
9.973
1
0.002
9.179
flow
-0.05
0.001
5.055
0.01
10.98
1
0.001
0.956
0
1.54
5.682
10.71
1
1
0.017
0.001
0.999
156.77
Tricladida
inest12
Constant
17
Discussion
High altitude Andean streams had been considered aseasonal streams
(Jacobsen 2005). At Piburja stream, although temperature showed variations
monthly changes in tropical and specifically Andean streams are often narrow.
The few previous studies of Andean streams, showed a temperature dial
variation (~ 5 ºC) larger than monthly variation (~1 ºC) in contrast to rainfall that
show great seasonal changes, and therefore hydrological differences as those
found at Piburja stream (Flecker and Feifarek 1994, Jacobsen and Encalada
1998, Jacobsen and Marín 2007, Perez and Segnini 2007). Although we lack of
a complete 24 hour water temperature records, monthly mean air temperature in
the area had a difference of less than ~1 ºC, for this reason we can not assume
that differences present on the structure and composition of the benthic
community are due to temperature variations between seasons. However at
Piburja stream, flow and the related variables: flow stability, velocity and depth,
clearly had a seasonal and reach variation during the two sample periods and
are good candidates to explain differences in macroinvertebrate community and
structure.
If seasonal, hydrological related events are key drivers in this Andean stream,
large seasonal changes on the benthic community should be expected. In fact,
at Piburja stream we found important seasonal changes in all macroinvertebrate
parameters measured: community composition, species richness, number of
individuals, taxa density and diversity. Most part of taxa were present at the two
seasons but densities of most of them were higher at dry season, the most
stable hydrologic period. This increase of abundance and diversity at dry periods
is reported in almost every previous study in Andean streams (Flecker and
Feifarek 1994, Jacobsen and Encalada 1998, Jacobsen and Marín 2007, Perez
and Segnini 2007).
Only two taxa Baetodes and Macrelmis were denser at the wet season. This
could be related to a more availability of rapid areas at high flow conditions
although contradictory information may be found in the literature. Thus, while
18
Macrelmis has been reported as a riffle inhabitant in Brazilian tropical streams,
with biomass peaks at larvae or adult instars at summer (Passos et al. 2003), in
Colombian rivers, the same taxa showed no relation of the abundance when
related to wet or dry seasons. On the other hand, Baetodes is known to be an
organism that prefer fast flow areas (Zuñiga de Cardoso et al. 1997), and
Baetids in general enter in drift actively as a feeding behavior (Kohler and
McPeek 1989, Peckarsky 1996). Probably these two taxa are adapted to
increase densities at wet season due to more fast flow area availability and this
could explain its higher densities on the stream at this season.
A further question arises from these clear differences in density for most taxa
between the dry and the wet season in Piburja stream: Might there be
differences in density due to synchronized life cycles? Although there are
evidences of synchronized life cycles of taxa in other tropical streams (Jackson
and Sweeney 1995), and thus synchronicity in life cycles of some aquatic
insects maybe present in some tropical areas with strong environmental
seasonal changes in response to more favorable developmental conditions,
most tropical aquatic insects show multivoltine life cycles or are present during
all year (Turcotte and Harper 1982, Jackson and Sweeney 1995). But, some
studies had shown that Elmids, like Macrelmis, have long life-cycles in rivers,
that could last 4 years (Steedman and Anderson 1985) and no seasonal
patterns in density were present (Suren and Jowett 2006). Unfortunately we do
not collect data during a complete year and therefore we can not conclude if
there is a preferred reproductive season for the taxa present at Piburja stream
that were more abundant in one of the seasons. As we found most part of taxa
at both seasons, and thus the multivoltine strategy seems to be the rule. Anyway
much more research is needed to define if all taxa in tropical streams really had
multivoltinism and asynchronic life cycles, especially in those taxa with longer
developmental times that may be more vulnerable to natural disturbances.
For hydrologically disturbed environments is expected that the fauna is randomly
organized (Death 2004). In fact, Jacobsen (2005), with cobble stones samples
in a monthly year study at Ecuadorian Andean rivers, found a random
19
distribution of the macroinvertebrate community, with a positive relation with
current velocity and number of macroinvertebrate families. Contrarily, in
Venezuelan highland Andean streams Perez and Segnini (2007) found clearly
differentiated benthic communities in Surber samples in high velocity and low
velocity habitats, taken samples every 15 days during a year. At Piburja stream
microhabitat condition related to flow velocity (rapid, refugium) was an important
factor differentiating the community at dry season while at the wet season, as
expected for disturbed habitats, fauna were more randomly organized indicating
that seasonal differences are important and related to flow conditions. The
differences found in this study with the previous studies could be related to the
sampling frequency. Short term sampling, as the present study, are needed to
visualize changes or patterns of distribution of fauna at river bed with high
unstable flow and velocity condition, because changes occur in short periods of
time (hours or days). Anyway in streams is expected that flow heterogeneity and
differences in composition and abundance of benthic fauna between habitat
patches arise after the spates (Jacobsen 2005) and therefore habitat
preferences of aquatic organism should became more evident at the dry season
as is the case at Piburja stream.
Despite the large flow seasonal differences in Piburja stream, certain taxa
maintained their preferences for rapids (Simuliidae, Leptohyphes, Baetodes and
Ochrotrichia) during the both seasons. For these taxa there is a clear evidence
of habitat preference that can be mediated by specific adaptative traits that allow
them to avoid the effects of flow increase and the consequent disturbance
events during the wet season. On the other hand, taxa like Nectopsyche,
Prionocyphon, Probezzia and Hyalella had greater densities at refugia areas. In
general taxa that were more abundant at refugia had higher densities at dry
season.
The importance of refugia in streams has been reported in several studies
(Lancaster and Hildrew 1993b, a, Palmer et al. 1995, Palmer et al. 1996,
Lancaster and Belyea 1997, Winterbottom 1997, Lancaster 1999, Matthaei et al.
1999, Lancaster 2000, Matthaei et al. 2000, Gjerlov et al. 2003, Melo et al. 2003,
20
Bergey 2005, Hose et al. 2007). We found that the species that permanently
prefer rapid areas do not increase the density (p.e. Simuliidae) at refugia in the
wet season, therefore is seems that refugia searching behavior is not present in
these taxa despite the fact that some individuals of these taxa were present, but
in lower densities at refugia in both seasons (probably by passive arrival). In
other studies species that prefer fast flowing areas, like Deleatidium (mayfly) had
shown microbistributional changes in response to flow increases, to avoid drift
(Holomuzki and Biggs 2000). Although this behavior may be present in the
mayflies of Piburja stream we do not detect it during our study.
Another
possibility is that the flow events that we recorded during both seasons were not
strong enough to reveal the behavior of refugium search to avoid spate effects.
On the other hand for species found in higher densities at refugia or low flow
areas during both seasons, like Nectopsyche, may have a permanent behavior
to avoid catastrophical drift, and increase the odds to persist. This behavior was
also present in studies for other crawling caddisflies, Potamophylax latipennis
showed a permanent preferences for refugium under controlled laboratory
conditions in response to fully turbulent flow (Lancaster et al. 2006).
Community composition is not altered by the water extraction at Piburja stream,
and both reaches showed similar responses at community level. Differences in
community structure at reach scale were not detected, and only some density
differences in certain taxa where found. Most part of Trichoptera, and Plecoptera
showed a density increase in the pristine reach, while Baetidae, Oligochaeta and
Tricladida increased density at the disturbed site. As it was part of the same
river, most that taxa should arrive through drift to the lower reach, but the fact
that some taxa decrease in density at the disturbed reach, while other taxa
increased at the reach that suffered water extraction, suggest that these taxa are
sensitive to water reduction. Therefore the effect of water extraction in this
stream was not a change on species richness but the lowering of densities of
more intolerant taxa (e.g. Claudioperla). A future issue in the study of the effect
of water extraction in these ecosystems may be if this water reduction in altered
areas produces changes in community structure or species lost at long term.
This will be crucial, also for the understanding of the implications of water
21
extractions in these Andean areas, where the pressure to aquatic ecosystems is
increasing due to flow reductions related to glacier retreats produced by the
climate change (Bradley et al. 2006).
22
APPENDIX1. Kruskal-Wallis Test for Seasonal and Reach differences of Flow and Physico-chemical variables of Piburja
stream, Ecuador. H= Kruskal-Wallis test; R= rainy season; Dy= Dry season; P=Pristine Reach; D= Disturbed Reach;
*=p<0.05;**=p<0.01, N.S = Non significant differences .
SEASONS
flow l/s
R
Mean
H
127.39
22.4
Dy 28.74
INEST 12h00
R
479.98
Dy 1731.51
INEST 24h00
INEST 48h00
R
velocity cm/s
depth cm
pH
temperature
ºC
conductivity
µS/cm
P
8.49
3.48
N.S
547.67
R
28.67
4.62
27.6
R>Dy
R
2
Dy 10.79
N.S
R
11
6.91
R
15.5
Dy 8.60
R>Dy
R
24.8
79.22
131.01
0.00**
0.00**
Dy>R
Dy
0.03*
0.157
Dy 7.14
9.49
0.062
Dy>R
Dy 10.96
12.28
0.004**
Dy>R
1837.61
R
0.00**
R>Dy
Dy 1309.40
Dy 1837.88
Rainy
0.00**
0.00**
Dy>R
Dry
Mean
H
P
P
71.92
2.56
0.11
D
71.05
N.S
P
976.63
1.55
D
1466.62
N.S
P
1622.55
2.85
D
1437.48
N.S
P
1483.70
2.42
D
1000.24
N.S
P
16.97
0.64
D
20.73
N.S
P
13.22
3.65
D
9.21
N.S
P
7.07
0.02
D
7.01
N.S
P
8.79
2.05
D
9.22
N.S
P
111.10
106.04
0.28
D
0.21
0.09
0.12
REACHES
Mean
H
P
50.15
8.69
D
38.84
P>D
P
1523.74
0.18
D
1551.34
N.S
P
1591.05
1.12
D
1193.40
N.S
P
1622.66
0.94
D
1631.00
N.S
P
0.00**
0.67
0.29
0.33
P
13.44
3.05
D
14.19
N.S
0.06
P
11.36
32.01
D
10.68
P>D
0.89
P
7.10
0.79
D
7.11
N.S
P
8.70
11.46
D
8.78
D>P
P
121.99
123.70
0.09
0.42
0.15
0.59
N.S
D
0.08
0.00**
0.37
0.00**
0.77
N.S
Mean
H
P
66.40
7,089
D
41.51
P>D
P
1087.96
0
D
1797.97
NS
P
1853.32
0
D
880.13
N.S
P
1703.13
0
D
1141.61
N.S
Pristine reach
P
0.01**
0.96
0.96
0.96
P
13.56
2.1
D
19.12
N.S
P
13.88
31.9
D
7.42
P>D
P
7.10
0.59
D
7.05
N.S
P
8.56
11.3
D
9.23
D>P
P
117.67
0.07
0.795
115.81
N.S
D
0.147
0.00**
0.443
0.000**
Mean
H
R
71.92
16.4
Dy
50.15
R>Dy
R
976.63
4.64
Dy
1523.74
Dy>R
R
1622.55
0.86
Dy
1591.05
N.S
R
1483.70
1.42
Dy
1622.66
N.S
Disturbed reach
P
0.00**
Mean
H
P
71.05
9.2
0.002**
Dy 38.84
0.03*
0.353
R
0.233
1466.62
3.4
N.S
R
1437.48
R
1000.24
Dy 1631.00
R
16.97
19.4
13.44
R>Dy
R
13.22
0.36
Dy
11.36
N.S
R
7.07
6.69
Dy
7.10
N.S
R
8.79
12.2
Dy
8.70
R>Dy
R
111.10
14.3 0.002**
121.99
Dy>R
0.00**
0.551
R
20.73
0.000**
0.064
3
0.085
N.S
4
0.04*
Dy>R
8.5
0.003**
Dy 14.19
R>Dy
R
4.6
9.21
Dy 10.68
0.009
R>Dy
Dy 1551.34
Dy 1193.40
Dy
Dy
R
R
7.01
0.031*
R>Dy
4.4
0.04*
Dy 7.11
Dy>R
R
5.3
9.22
.022*
Dy 8.78
R>Dy
R
11
Dy
106.04
123.70
.001**
Dy>R
23
APPENDIX 2. F and P values of the General Linear Model performed for the
community metrics of Piburja Stream. Fixed factors: Season (wet and dry),
reach (pristine and disturbed), microhabitat (rapids and refugia) and the
respective interactions are presented. Highlighted cells correspond to significant
P-values after Bonferroni corrections.
dependent
variable
S
N
Fisher’s α
Intercept
F 1475.985
season
reach
season
season*
reach*
*reach
mhabitat
mhabitat
mhabitat
reach*
mhabitat
*season
110.420
0.374
1.635
0.696
4.232
0.250
P
0.000
0.000
0.545
0.210
0.410
0.048
0.620
0.956
F
211.039
20.621
0.887
4.091
0.220
4.178
0.270
0.439
P
0.000
0.000
0.353
0.052
0.642
0.049
0.607
0.512
F 1222.428
85.062
0.022
0.370
0.400
1.541
0.178
0.090
P
0.003
0.000
0.000
0.883
0.547
0.532
0.223
0.676
0.766
H’
F 2486.628
52.353
0.708
2.159
4.732
5.126
0.169
0.087
P
0.000
0.000
0.406
0.152
0.037
0.030
0.683
0.769
Heterelmis
F
101.482
4.773
1.414
2.589
0.011
5.399
7.993
4.265
P
0.000
0.036
0.243
0.117
0.918
0.027
0.008
0.047
Pharceonus
F
33.651
33.651
0.290
0.738
0.290
0.738
2.003
2.003
P
0.000
0.000
0.594
0.397
0.594
0.397
0.167
0.167
F
51.254
29.961
4.926
1.430
0.285
3.090
14.015
10.124
P
0.000
0.000
0.034
0.240
0.597
0.088
0.001
0.003
F
180.880
160.381
3.470
2.688
1.162
0.730
1.864
0.337
P
0.000
0.000
0.072
0.111
0.289
0.399
0.182
0.566
Elmidae
F
101.232
87.709
0.014
0.282
0.332
1.507
0.017
0.319
P
0.000
0.000
0.905
0.599
0.569
0.229
0.896
0.576
Prionocyphon
F
33.259
33.259
0.521
15.035
0.521
15.035
0.203
0.203
Macrelmis
Neocylloepus
P
0.000
0.000
0.476
0.000
0.476
0.000
0.656
0.656
F
70.104
12.639
0.146
6.896
0.983
6.615
0.013
0.244
P
0.000
0.001
0.705
0.013
0.329
0.015
0.909
0.624
Staphilinidae
F
21.933
21.933
0.040
1.696
0.040
1.696
0.634
0.634
P
0.000
0.000
0.843
0.202
0.843
0.202
0.432
0.432
Hyalella
F
39.322
3.978
4.483
17.997
0.232
6.792
1.986
0.052
P
0.000
0.055
0.042
0.000
0.634
0.014
0.168
0.822
Probezzia
F
405.594
405.594
8.982
17.965
8.982
17.965
0.911
0.911
P
0.000
0.000
0.005
0.000
0.005
0.000
0.347
0.347
Atrichopogon
F
33.470
33.470
0.395
2.943
0.395
2.943
0.035
0.035
P
0.000
0.000
0.534
0.096
0.534
0.096
0.854
0.854
F
411.222
9.428
2.740
16.161
2.632
0.517
0.190
0.366
P
0.000
0.004
0.108
0.000
0.115
0.477
0.666
0.549
Chironominae
F
196.453
1.813
2.447
0.617
1.605
3.553
0.172
0.134
P
0.000
0.188
0.128
0.438
0.214
0.069
0.681
0.717
Diamesinae
F
266.443
235.600
1.188
0.329
0.014
0.160
0.028
1.299
P
0.000
0.000
0.284
0.570
0.908
0.692
0.869
0.263
Scirtidae
Ceratopogonidae
24
dependent
variable
Orthocladiinae
Podonominae
Tanypodinae
Tanytarsini
Dolichopodidae
Chelifera
Intercept
season
reach
season
season*
reach*
*reach
mhabit
mhabit
mhabit
reach*
mhabit
*season
F 3082.041
2.295
1.736
0.664
0.000
3.275
0.007
0.463
P
0.000
0.140
0.197
0.421
0.995
0.080
0.936
0.501
F
105.409
74.412
0.001
0.045
2.579
3.429
2.579
0.001
P
0.000
0.000
0.972
0.834
0.118
0.073
0.118
0.972
F
92.571
0.015
0.934
1.919
0.029
0.414
0.131
0.834
P
0.000
0.903
0.341
0.176
0.867
0.524
0.720
0.368
F
334.337
11.813
7.440
8.947
0.251
0.166
0.019
0.215
P
0.000
0.002
0.010
0.005
0.620
0.686
0.891
0.646
F
25.129
25.129
0.798
0.013
0.798
0.013
3.574
3.574
P
0.000
0.000
0.378
0.909
0.378
0.909
0.068
0.068
F
370.547
370.547
0.148
0.289
0.148
0.289
0.004
0.004
P
0.000
0.000
0.703
0.595
0.703
0.595
0.949
0.949
Limoniidae
F
76.959
76.959
2.207
3.301
2.207
3.301
0.804
0.804
P
0.000
0.000
0.147
0.079
0.147
0.079
0.377
0.377
Maurinia
F
16.971
11.390
2.328
0.337
0.610
1.755
0.668
0.005
Simuliidae
Stratiomyidae
Tipula
Tipulidae
P
0.000
0.002
0.137
0.566
0.440
0.195
0.420
0.943
F
131.875
10.712
0.131
28.145
0.039
0.069
0.010
0.005
P
0.000
0.003
0.720
0.000
0.845
0.795
0.921
0.946
F
34.903
3.315
0.099
3.939
0.837
0.571
1.449
0.309
P
0.000
0.078
0.755
0.056
0.367
0.455
0.237
0.582
F
40.232
40.232
0.120
7.911
0.120
7.911
0.005
0.005
P
0.000
0.000
0.732
0.008
0.732
0.008
0.943
0.943
F
54.918
1.126
0.021
10.629
0.248
0.750
0.252
0.001
P
0.000
0.297
0.886
0.003
0.622
0.393
0.619
0.971
Andesiops
F
63.223
9.525
3.584
0.913
7.822
1.887
1.220
0.013
P
0.000
0.004
0.067
0.347
0.009
0.179
0.278
0.910
Baetodes
F
417.727
6.841
0.034
43.137
0.009
0.133
0.437
1.244
P
0.000
0.013
0.855
0.000
0.925
0.718
0.513
0.273
F
272.131
13.572
0.122
5.719
8.606
0.419
0.467
0.116
Leptohyphes
Planorbiidae
Hydracarina
Ostracoda
Oligochaeta
Andiperlodes
P
0.000
0.001
0.729
0.023
0.006
0.522
0.499
0.736
F
17.589
1.333
0.127
3.015
0.127
0.165
0.458
1.064
P
0.000
0.257
0.723
0.092
0.723
0.688
0.503
0.310
F
263.414
32.143
3.754
0.698
2.052
1.225
0.009
0.000
P
0.000
0.000
0.062
0.410
0.162
0.277
0.924
0.989
F
27.504
27.504
0.502
1.862
0.502
1.862
0.502
0.502
P
0.000
0.000
0.182
0.484
0.182
0.484
0.484
F
646.437
4.688
0.484
44.09
1
2.630
4.370
2.298
0.695
1.138
P
0.000
0.038
0.000
0.115
0.045
0.139
0.411
0.294
F
14.228
9.765
1.527
1.062
0.347
2.814
0.217
0.033
P
0.001
0.004
0.225
0.311
0.560
0.103
0.644
0.858
25
dependent
variable
Claudioperla
Gripopterygidae
Cochliopsyche
Contulma
Intercept
season
reach
season
season*
reach*
*reach
mhabitat
mhabita
mhabitat
reach*
mhabitat
*season
F
32.425
0.198
6.088
0.821
2.437
0.025
2.000
2.469
P
0.000
0.659
0.019
0.372
0.128
0.876
0.167
0.126
F
18.754
18.754
2.049
0.936
2.049
0.936
6.397
6.397
P
0.000
0.000
0.162
0.341
0.162
0.341
0.017
0.017
F
68.319
14.986
7.306
5.762
6.029
3.040
0.238
0.541
P
0.000
0.001
0.011
0.022
0.020
0.091
0.629
0.467
F
85.988
85.988
1.483
0.500
1.483
0.500
0.441
0.441
P
0.000
0.000
0.232
0.485
0.232
0.485
0.512
0.512
Hydrobiosidae
F
56.498
0.030
0.636
15.274
1.482
0.615
2.806
1.201
P
0.000
0.864
0.431
0.000
0.232
0.439
0.104
0.281
Ochrotrichia
F
167.344
42.191
9.482
18.602
0.076
0.418
1.102
0.075
Nectopsyche
Leptoceridae
Tricladia
P
0.000
0.000
0.004
0.000
0.785
0.522
0.302
0.786
F
282.710
10.338
2.542
7.770
1.343
4.977
0.769
0.025
P
0.000
0.003
0.121
0.009
0.255
0.033
0.387
0.876
F
282.863
10.577
2.614
7.665
1.399
4.899
0.820
0.016
P
0.000
0.003
0.009
0.246
0.034
0.372
0.900
F
97.846
15.322
0.116
15.46
7
0.023
0.015
1.167
0.005
1.348
P
0.000
0.000
0.000
0.881
0.903
0.288
0.944
0.254
26
APPENDIX 3. Pearson’s correlations, r(X,Y),between taxa densities and flow
and flow stability (all variables log transformed) at Piburja stream, Ecuador. r², t
and p values are also presented.
inest12log
Heterelmis
Macrelmis
Neocylloepus
Elmidae Total
Scirtidae
Hyalella
Probezzia
Chironominae
Diamesinae
Orthocladiinae
Podonominae
Tanypodinae
Tanytarsini
Chelifera
Maurinia
Simuliidae
Stratiomyidae
Tipulidae
Andesiops
Baetodes
Leptohyphes
Planorbiidae
Hydracarina
Oligochaeta
Andiperlodes
Claudioperla
Cochliopsyche
Hydrobiosidae
Ochrotrichia
Nectopsyche
Tricladia
INES12log
r(X,Y)
0.186
-0.202
0.424
0.350
0.006
0.237
0.425
0.125
0.461
0.067
0.385
0.076
0.226
0.218
0.194
0.300
0.130
-0.132
0.431
0.058
0.279
0.212
0.322
-0.097
0.570
0.048
0.274
0.484
0.327
0.302
0.168
r²
t
p
0.034
0.041
0.180
0.122
0.000
0.056
0.180
0.016
0.213
0.004
0.148
0.006
0.051
0.048
0.038
0.090
0.017
0.018
0.186
0.003
0.078
0.045
0.104
0.009
0.325
0.002
0.075
0.234
0.107
0.091
0.028
1.164
-1.272
2.884
2.303
0.038
1.507
2.891
0.775
3.206
0.411
2.568
0.470
1.431
1.378
1.221
1.941
0.810
-0.824
2.945
0.361
1.791
1.335
2.097
-0.598
4.277
0.296
1.753
3.406
2.132
1.952
1.054
0.252
0.211
0.006
0.027
0.970
0.140
0.006
0.443
0.003
0.683
0.014
0.641
0.161
0.176
0.230
0.060
0.423
0.415
0.005
0.720
0.081
0.190
0.043
0.553
0.000
0.769
0.088
0.002
0.040
0.058
0.299
CaudalLOG
r(X,Y)
r²
-0.343
0.334
-0.703
-0.513
-0.276
-0.273
-0.237
-0.101
-0.729
-0.105
-0.557
-0.146
-0.480
-0.414
-0.219
-0.414
-0.019
0.030
-0.449
0.083
-0.361
-0.160
-0.592
-0.290
-0.414
0.145
-0.276
-0.600
-0.421
-0.337
-0.477
0.117
0.111
0.495
0.263
0.076
0.075
0.056
0.010
0.532
0.011
0.310
0.021
0.231
0.171
0.048
0.172
0.000
0.001
0.202
0.007
0.130
0.026
0.350
0.084
0.172
0.021
0.076
0.360
0.177
0.113
0.227
t
p
-2.248
2.181
-6.100
-3.685
-1.769
-1.750
-1.503
-0.625
-6.574
-0.650
-4.135
-0.908
-3.375
-2.801
-1.384
-2.807
-0.117
0.186
-3.097
0.516
-2.387
-1.002
-4.523
-1.868
-2.807
0.904
-1.767
-4.626
-2.860
-2.203
-3.341
0.030
0.035
0.000
0.001
0.085
0.088
0.141
0.536
0.000
0.520
0.000
0.370
0.002
0.008
0.174
0.008
0.907
0.854
0.004
0.609
0.022
0.323
0.000
0.069
0.008
0.372
0.085
0.000
0.007
0.034
0.002
27
CHAPTER 2:
Invertebrate drift and colonization
processes in a Tropical Andean
Stream
CHAPTER 2: Invertebrate drift and colonization processes
Introduction
Recolonization of substrate after disturbance is one of the most important
processes that structure the macroinvertebrate communities (Boyero and Bosch
2004).
This recovery is generally fast (Mackay 1992, Flecker and Feifarek
1994), and usually begins the first hour after disturbance (Boyero and DeLope
2002) and may be completed in a relatively short time (30 days) (Lake and
Schreiber 1991). Recolonization depends mainly on the individuals that arrive
with the drift, upstream migration within the water, communities in surrounding
patches and recruitment by oviposition (Williams and Hynes 1976). Substrate
characteristics, associated food sources, competition and predation can affect
colonization (Resetarits 1991, Mackay 1992, Resetarits 2001)
One of the main mechanisms of substrate recolonization is macroinvertebrate
drift and its composition may profoundly influence benthic community dynamics.
On one hand, reducing benthic density of the species more prone to drift than
others and thus affecting the local composition and abundance. On the other
hand, the continuous settling in the substrate of animals from drift plays an
important role in the colonization of benthos (Townsend and Hildrew 1976). The
study of drift, which is a recurrent topic of study in stream ecology, had
generated a lot of data but few clear conclusions about this important process. In
temperate areas, where the role of drift on benthic recolonization is well studied
and documented, (Brittain and Eikeland 1988, Mackay 1992) it has been
established that the great majority of colonizing fauna came from drift (Williams
and Hynes 1976). Although, benthic fauna from near undisturbed patches may
play an important role in the recolonization process (Lancaster and Belyea 1997,
Townsend et al. 1997). In tropical areas, recent studies show that drift diel
patterns are only present on rivers with fish while no diel patterns have been
reported in fish-less rivers (Flecker 1992, Pringle and Ramirez 1998, Jacobsen
and Bojsen 2002). In these areas, drift is generally aseasonal (Pringle and
Ramirez 1998, Ramirez and Pringle 2001, Jacobsen and Bojsen 2002,
Rodriguez-Barrios et al. 2007) and the drift density seems to be more dependent
on flow, increasing with high flow conditions (Turcotte and Harper 1982, Benson
31
and Pearson 1987). Although, some tropical streams show an increase during
dry seasons (Ramirez and Pringle 2001, Rodriguez-Barrios et al. 2007)
The role of drift in recolonization processes in tropical systems has been
recently addressed (Boyero and DeLope 2002, Boyero and Bosch 2004, Melo
and Froehlich 2004).
These studies find evidences that recolonization by
movement of invertebrates in tropical rivers varies among riffles but not among
sections of river and the individuals that recolonize the stones depend on both
the surrounding patches at the local level and the drift coming from upstream at
riffles and sections of river (Boyero and Bosch 2004, Melo and Froehlich 2004).
However, these studies have been developed only in one season and for a short
period of time, therefore seasonal differences on drift, that can cause differences
on the recolonization process, have not been addressed in previous studies.
Also, different species traits and life-history adaptations of benthic fauna, can
influence how and when certain taxa arrive to new or disturbed substrata. The
effect of microhabitat hydrological conditions on the recolonization process is still
unknown.
Besides these increasing on the knowledge of drift and recolonization of tropical
rivers, tropical Andean streams remain remarkably understudied (Allan et al.
2006). Streams in the Andean region are characterized by rapid flow changes
that are not easily predictable from seasonal variations in precipitation. This high
unpredictability on the occurrence of spates should have important effects on
drift and recolonization process in these streams. On the other hand mean
temperature in these rivers are low (~10ºC) and oxygen saturation is always
close to 70-80 % due to altitude (Jacobsen 2008). In this context, our aim is to
describe the diel and seasonal variation of drift in a high altitude Andean stream,
including the taxa propensity to enter in drift and the differences of this
propensity among seasons. Finally, the importance of drift for recolonization of
stones is studied and related to drift intensity and propensity in two different
hydrological seasons.
32
In this context our objectives are: 1) Describe diel and seasonal drift in a high
altitude Andean stream; 2) Study the seasonal differences in drift and estimate
the propensity of macroinvertebrate taxa to drift at each season; 3) Describe
short term colonization process (1 to 7 days) in basal flow conditions; 4) Study
stone colonization and dynamics within and between seasons; 5) Determine the
importance of microhabitat conditions in the recolonization process (fast velocity
areas vs. low velocity areas)
Study Area
We performed the study at the Piburja stream (3300 m a.s.l.), a first order
stream in the Oyacachi river basin, located at the Cayambe-Coca Ecological
Reserve (RECAY) in Ecuador (0º13’ S, 78º 03’ W). Oyacachi basin, includes
several vegetation types from paramos to tropical rain forest (~4000 to 1800 m
a.s.l.), has a western-eastern slope and is part of the Amazon basin (Terneus
and Vásconez 2004). The Oyacachi river starts in the Paramo (Polylepis and
mix forest), and continues through an ever-green forest (Alnus and ever-green
high forest (Báez et al. 1999). Pluviometric records (10 years) from nearby
localities show lower rainfall from December to February (dry season) compared
with May until September.
Total precipitation in Oyacachi town is ~ 1600
mm/year, and mean annual temperature ~10 ºC. Precipitation is 2 or 3 times
greater than the potential evapotranspiration, which makes very humid
conditions during the whole year (Skov 1999).
We studied a 70 m pristine reach of the stream located at 3300 m asl. The
stream width ranges from 1.60 m to 4.0 m. The surrounding vegetation is an
evergreen forest with Alnus acuminata trees and a high diversity of
Melastomataceae, Asteraceae and Rosaceae shrubs.
The stream physic-
chemical parameters measured (Table 1) showed seasonal differences in mean
flow and mean water velocity, which had been higher at wet season (Kruskalwallis test). More detailed physico- chemical characteristics of the reach can be
found at Chapter 1 of this thesis. We performed all surveys and experiments
during April and May of 2006 (the wet season) and January and February of
2007 (the dry season). The river has a population of rainbow trout, introduced
33
from a fish culture located downstream. Trout density is not known but is
presumably low (Aigaje family pers comm.).
Table 1. Main physico-chemical parameters during dry and wet seasons of
Pijurba streams, Oyacachi, Ecuador. * means significant change between
seasons according to Kruskal-Wallis analysis (Chapter 1).
Mean
Min
Max
Parameter
Wet
Dry
Wet
Dry
Wet
Dry
season season season season season season
Flow (l/s)*
150.79
36.26
68.49
11.42
262.90
121.64
Velocity
(cm/s)*
25.08
9.44
17.87
4.09
35.34
22.55
Depth cm
14.04
13.83
10.17
9.89
16.81
18.55
pH
6.90
6.59
6.38
7.17
7.32
7.70
Temperature
(°C)
9.20
8.32
8.30
6.70
10.00
9.20
Conductivity
(µS/cm)
78.69
131.59
53.50
67.70
120.70
162.10
Drift
To assess the diel dynamic of drift we made a 24 h sampling during low flow
conditions at the dry season in new moon. We took samples during four periods
in this day: 1) From 10h30 to 13h30; 2) From 16h30 to 19h30; 3) from 22h30 to
01h30 and 4) From 04h30-07h30. Four nets (250 µm mesh, 15 x 35 cm frame, 2
m long) were placed in the river for 3 hours at each of the four sampling periods.
Water level at each net mouth and average velocity was measured at the
beginning, at the middle and end of each sampling period. Each replicate was
stored separately and preserved with formalin 4%.
To identify seasonal changes in drift and its relationship with flow we sampled
four times at each season. During the wet season we sampled twice during
spates and twice during low flow conditions. At the dry season the four samples
were taken at basal flow conditions. The samples of low flow conditions from the
wet season as well as the samples of the dry season were taken at the same
hour of the day when spates occurred to minimize the error related to diel
34
patters on taxa drift. Four to six drift nets (same model as above) were placed in
the river for 20 minutes to 1 hour, depending on flow conditions at each of the
eight sampling events. Each replicate was stored separately and preserved with
formalin 4%. We made all drift density calculations following Allan & Russek
(1985). To assess the drift propensity of taxa, we applied the formula: drift
density/ benthic density (McIntosh et al. 2002) using mean values of drift density
and benthic density of each season. The benthos densities were taken from a
simultaneous study developed at the same time at the same reach (see chapter
1).
Recolonization of benthic substrate
We assessed short term recolonization, performing 7 day colonization
experiments during the dry season (basal flow conditions). At each experiment,
we placed in the first day 24 cleaned cobble stones with similar size and surface
in the stream. We marked the stones with white paint to recognize them, and
measure their turning or rolling. We used stones because they are considered
well defined and discrete habitats (Douglas and Lake 1994), appropriate for
colonization studies (Boulton et al. 1988). Twelve stones were placed in high
velocity areas (rapids) and twelve in low velocity areas (potential refugia). Four
stones from rapids and four from refugia were collected placing a 250 µm mesh
hand net downstream at 1, 3 and 7 days from the beginning of the experiment.
Surrounding velocity and stone area (maximum length x maximum width) were
measured for each stone. Stone turning was scored in 5 categories: 1: no
change; 2: less than 45º of inclination, 3: between 45º to 135º; 4: Close to 180º
and 5: upside down. This process was repeated 3 times during the dry season at
the same reach. The stones that we placed in the stream did not move
downstream during the experiment.
In a second experiment we wanted to know the recolonization dynamics at a
longer scale of time. For this purpose marked clean and dry cobbles were
placed in rapids and refugia areas in the whole reach (45 in the wet season and
60 in the dry season). Stones of both areas were taken on four dates at both
seasons (at days 7, 12-14, 17-19 and 22-25 after the stones were placed on the
35
river). The detailed characterization of rapids and refugia in this stream may be
found in the chapter 1. We recorded the surrounding velocity of each stone
together with the stone area (maximum length x maximum width) at the moment
of collection from the stream.
Data analysis
Variables were log(x+1) transformed when necessary to satisfy homogeneity of
variance assumptions. In addition to taxa densities, in all experiments we
calculate four community metrics: Richness (S), Total Density (N), Fisher’s α
and Shannon Weiner (H’) index.
To assess differences of the community parameters between in samples from
the four intervals sampled at the drift periodicity survey we used a Repeated
Measures Analysis of Variance (ANOVA). To assess differences of taxa
densities from the four intervals sampled at the drift periodicity survey we used
the Kruskal-Wallis analysis of variance (using the four intervals sampled as
factor). In the seasonal drift survey, drift community metrics were related to flow
using Spearman Correlations. To assess seasonal changes on the composition
of drift we applied the Non Metric Multidimentional Scaling using Bray-Curtis
similarities with drift densities log transformed. Also ANOSIM analysis with all
factors (seasons, colonization days and microhabitats) was performed.
In the short term (7 days) stone colonization experiment we used the KruskalWallis analysis of variance to asses differences in community metrics and taxa
densities considering three factors: colonization days, stone movement and
microhabitat. In the mid-term stone colonization experiment, we used a two way
ANOVA with the factors: colonization days and microhabitats, to assess
differences in community metrics. Tuckey post hoc comparison was used to
determine significant differences among colonization days. Differences of taxa
densities in recolonized stones with the two factors (colonization days and
microhabitat) were analysed by a Kruskal-Wallis analysis of variance. Bonferroni
progressive correction of p-values was applied to all multiple test. The analyses
were performed using SPSS 14 and Primer 6 statistical software.
36
Results
1. Drift diel periodicity.
At community level we did not find significant differences in community metrics
between hours during the diel experiment performed at the dry season (Table 2).
Total density, Richness, Fishers’ α, and Shannon- Weiner indexes did not differ
among hour periods, and the slight rise detected at dusk time was not
significant.
Table 2. Summary of the Repeated Measures ANOVA of community metrics
measured during the diel experiment at Piburja stream, Ecuador( df.=3; at four
time intervals: 10h30 - 13h30;16h30-19h30;22h30-01h30;04h30-07h30)
Community metrics
F
p
S
1.567
0.264
N
1.204
0.363
Fisher
0.697
0.577
H'(loge)
0.945
0.459
From the 28 taxa that were common on drift samples in the diel experiment only
the ephemeropterans Baetodes and Leptohyphes showed significant differences
among periods (Fig. 1) according to the Kruskal-wallis analysis of variance
(Table 3). Higher drift densities at the period that included dusk (between 17h30
and 19h30) are found although a large variability between replicates was
present.
60
45
40
50
35
40
25
Leptohyphes
Baetodes
30
20
15
10
5
0
22h30 - 01h30
10h30 - 13h30
04h30-07h30
16h30 - 19h30
Time of the day
30
20
10
0
Mean
±SE
10h30 - 13h30
22h30 - 01h30
16h30 - 19h30
04h30-07h30
Time of the day
Mean
±SE
Figure 1. Diel changes in Baetodes and Leptohyphes drift density
3
(Individuals./100m ) at base flow conditions at Piburja stream, Ecuador. Mean
value is marked by squares and bars are 1 SE.
37
Table 3. Kruskal Wallis Test for macroinvertebrate drift densities
3
(Individuals/100m ) at the four sampling periods. * significant p values after the
bonferroni progressive correction.
Kruskal-Wallis Test
Taxa
X²
Mean drift density at each time of the day
df
P
10h30-3h30
16h30-9h30
22h30-1h30
04h30-h30
10.711
3
0.013
0.386
26.469
0.000
0.824
Baetodes
9.075
3
0.028
14.372
29.647
6.329
6.895
Chironominae
6.790
3
0.079
4.919
43.211
17.453
6.895
Ochrotrichia
5.481
3
0.140
11.210
17.579
3.237
6.189
Nectopsyche
5.412
3
0.144
19.188
65.013
30.973
27.641
Leptohyphes
Oligochaeta
5.186
3
0.159
2.718
36.495
2.487
1.799
Simuliidae
5.009
3
0.171
54.730
100.768
44.617
46.970
Andiperlodes
4.604
3
0.203
0.000
8.545
0.000
0.412
Contulma
4.420
3
0.220
0.000
8.859
1.849
0.824
Stratiomyidae
3.918
3
0.270
7.679
1.440
2.487
4.141
Prionocyphon
3.616
3
0.306
0.000
1.042
4.264
3.349
21.051
Tanytarsini
3.363
3
0.339
18.553
61.584
16.742
Tipula
3.169
3
0.366
2.718
9.108
0.781
1.106
Orthocladiinae
2.934
3
0.402
51.737
72.468
29.932
45.165
Hydracarina
2.596
3
0.458
39.337
43.205
22.578
31.394
Atopsyche
2.120
3
0.548
1.158
9.504
1.244
1.583
Staphylinidae
2.121
3
0.548
5.167
12.655
5.264
9.983
Blepharoceridae
1.775
3
0.620
7.120
29.828
7.255
15.425
Clognia
1.473
3
0.688
3.514
2.233
0.318
2.937
Lampirydae
1.420
3
0.701
2.160
0.438
1.244
3.415
Diamesinae
1.403
3
0.705
2.857
2.127
0.636
3.349
Psychodidae
1.370
3
0.713
5.115
3.110
0.781
3.631
Atrichopogon
1.274
3
0.735
3.393
0.479
2.054
1.387
Elodes
0.773
3
0.856
2.201
1.356
2.917
6.633
2. Changes in density and composition of drift related to season and
flow conditions.
Although community metrics (Richness, total drift density, Shannon and Fisher’s
α index) did not showed significant variation among seasons (Table 4), richness
and Fisher’s α are highly correlated with flow (Rho Spearman: 0.76; p=0.031
and Rho Spearman: 0.83; p=0.010 respectively). However strong community
composition differences were found among seasons and the NMDS analysis
clearly shows this pattern (Fig. 2). The ANOSYM also confirms this high relation
between the community composition and the season (Global R=0.322 p=0.001).
38
Table 4. Mean values of community metrics of macroinvertebrate drift and
summary of ANOVA for seasonal differences among these metrics at Piburja
Stream.
Community Metrics
S
total drift density
Fisher
H'(loge)
Mean
WET
27.75
600.48
6.30
2.19
Mean
DRY
± 8.22
± 169.65
± 2.34
± 0.21
26.00
596.35
5.67
2.29
df
± 6.98
± 201.91
± 1.38
± 0.27
7
7
7
7
F
P
0.105
0.756
0.001
0.214
0.331
0.976
0.660
0.586
A similar pattern of seasonal change in community composition was also found
in benthic Surber samples from the same site in the same seasons (Chapter 1:
page 13, and Appendix 1 of this chapter). Comparing the composition of benthos
and drift samples, we found that from the 52 taxa present in drift samples at the
wet season 17 were absent from the benthos samples, and only Tricladia and
Hirudinea were present in benthos Surber samples but not in drift samples. On
the other hand, at dry season 28 taxa were found in the drift and 50 in benthos
Surber samples. Seven taxa from the benthos were absent in drift samples and
just three coleopteran taxa were present in drift but not on benthos samples
(Appendix 1).
Figure 2. Non Metric Multidimensional Scaling of the four drift sampings in the
dry (inverted triangles) and the wet season (triangles). Each triangle is the
average of the 4 to 6 replicates of each sampling date. Samples from the dry
and the wet seasons are clearly differentiated because of the different taxa
composition.
39
In general drift propensity values are higher at wet season compared to the dry
season (Fig 3, Appendix 1). For taxa that are present in both seasons, the
general pattern is an increase of the drift propensity in the wet season (Fig 3,
Appendix 1). This is especially true for Podonominae, Diamesinae, Muscidae,
Hydracarina, Ochotrichia, Simuliidae, Pyralidae and Planorbiidae that were the
taxa that presented higher drift propensity values, but only at the wet season
because, at the dry season all taxa had small drift propensity values.
Stratiomydae, Maurinia and Tipula are the only taxa that presented more drift
propensity at the dry season, although absolute values of all of them are very
small.
0 .16
Drift Propensity
w et drift p ro p en s ity
d ry d rift p rop e ns ity
0 .14
0 .06
0 .05
0 .04
0 .03
0 .02
0 .01
Tipula
Scirtes
Tanypodinae
Hyalella
Chelifera
Nectopsyche
Baetodes
Nematomorpha
Hydrobiosidae
Maruina
Scirtidae
Andiperlodes
Stratiomyidae
Pyralidae
Planorbiidae
Simuliidae
Ochrotrichia
Muscidae
Hydracarina
Diamesinae
Podonominae
0 .00
Figure 3. Mean drift propensity at wet and dry seasons for taxa present at
Piburja stream, Ecuador.
At dry season, drift and benthos composition and taxa densities had similarities
and differences in the dominant taxa (Table 5, Appendix1).
Thus,
Orthocladiinae and Nectopsyche were the most common taxa (up to >30% of
the abundance), both in benthos and drift, while Simuliidae which was abundant
in drift (13%) is only the 3% of the individuals of benthos Surber samples.
Reciprocally
other
abundant
taxa
in
benthos
Surber
samples
like
Ceratopogonidae (8%) and Elmidae (5%) were less important in drift or almost
absent.
40
In the wet season (Table 5), Orthocladiinae and Baetodes were the dominant
taxa in benthos (>50%) and drift (>40%). While some taxa like Hydracarina,
Simuliidae and Ochotrichia were part of the abundant drifting taxa, other such as
Chironominae and Ceratopogonidae are dominant in Surber samples.
The
principal difference between seasons is that Baetodes is one of the dominant
taxa at wet season but become less important in terms of dominance at the dry
season. On the other hand, Nectopsyche, dominant during the dry season
become less important at wet season. Interestingly Podonominae present higher
densities in drift at the wet season but in benthos at dry season. As stated
previously (Fig. 3) this taxa has the highest propensity to enter in drift at wet
season.
Table 5. Macroinvertebrate dominance (as % of total density) of drift, Surber and
stone samples at Piburja stream during wet and dry seasons. Only taxa
representing more than 1% of total density were included in the table.
taxa
drift
Surber
Orthocladiinae
20.30 36.59
Baetodes
19.75 12.92
Simuliidae
10.22
2.44
Hydracarina
9.93
1.63
Ochrotrichia
5.15
1.08
Podonominae
4.47
0.18
Nectopsyche
4.43
4.16
Chironominae
3.76 11.02
Tanypodinae
1.88
2.62
Chelifera
1.63
1.72
Stratiomyidae
1.60
0.54
Tanytarsini
1.30
2.26
Ceratopogonidae
1.13
6.41
Maruina
1.12
0.45
Muscidae
1.07
0.15
Planorbiidae
1.02
0.27
Total Scirtidae
0.93
0.45
Hydrobiosidae
0.88
0.54
Blephariceridae
0.85 Diamesinae
0.83
0.09
Oligochaeta
0.79
4.25
Tipula
0.64
0.90
2.08
Total Elmidae
0.61
Leptohyphes
0.58
1.26
Limonidae
0.55 Cochliopsyche
0.29
0.63
Andiperlodes
0.29
0.09
WET
stones
high
velocity
18.04
30.69
20.80
0.42
2.39
0.82
7.18
0.82
4.67
2.05
0.39
0.55
4.09
0.42
0.87
1.68
1.50
0.82
stones
low
velocity
46.81
7.76
0.64
6.64
2.97
0.87
15.87
3.29
1.05
0.71
2.39
1.33
4.12
2.01
drift
Surber
21.88 20.00
2.99
2.01
12.75
3.35
8.44
5.93
4.47
3.60
0.09
0.62
12.95 11.31
6.35
7.19
0.37
0.84
2.27
1.81
2.89
0.35
1.15
3.35
2.61
8.41
3.14
0.50
0.04
0.11
0.88
0.12
1.01
0.89
0.73
1.27
0.72
0.09
1.19
2.78
2.28
4.24
2.75
0.37
0.20
5.28
0.10
4.17
1.78
0.47
0.32
1.74
0.55
1.01
DRY
stones
high
velocity
18.51
7.17
17.15
4.4
4.06
0.15
11.04
5.68
0.13
0.98
0.14
9.02
0.95
0.3
2.55
1.5
2.36
0.14
1.79
6.01
0.72
1.47
1.18
stones
low
velocity
18.47
7.17
5.40
6.11
2.74
2.35
23.92
5.19
0.37
0.29
7.16
2.18
0.21
0.43
0.52
0.47
1.56
0.46
5.47
0.35
3.55
1.30
41
3. Short term recolonization of stones at basal flow conditions
In the 3 seven-day colonization experiment (performed only at the dry season) a
total of 42 taxa were found, 27 after the first day of colonization, 30 after 3 days
and 38 after 7 days. From these 42 taxa, 8 arrive at the end of colonization
experiment and only one (Tricladia) appear on the first day and disappear the
following days.
According to the Kruskal-wallis analysis of variance (Table 6) there were higher
invertebrate density and richness at the seventh day than any other day.
Additionally, densities of Chironominae, Orthocladiinae, Tanytarsini, and
Leptohyphes, were significantly higher at day 7 (Table 6). As expected, there
st
th
were more differences detected between the 1 and the 7 day, than between
rd
st
the 3 and the 1 day.
No movement of stones was detected during the experiment, only turning of
some of them in the same place.
Stone turning was only important for
Podonominae and Hydrobiosidae that showed at significant reduction of density
when the stones roll or turned at least 180º. At community level there was no
effect of stone movement on the community metrics.
Richness, total density and Fisher’s alpha diversity were significantly higher in
rocks located in rapids than those in refugia areas. Also, the densities of
Simuliidae, Baetodes and Hydrobiosidae were higher at rapids while only
Cocliopsyche had higher density at slow flow areas.
42
Table 6. Kruskal-wallis analysis of variance for community metrics and taxa
densities (rare taxa excluded) between colonization days (1,3 or 7 days), stonemovement (from 1 to 5 in order of increase movement), and mirohabitat (high vs.
low current areas) . P significant values (after Bonferrroni correction) are in bold
letters.
Colonization days
P
X²
df
TAXA
S
N
Fisher
H'(loge)
Elmidae
Prionocyphon
Chironominae
Orthocladiinae
Podonominae
Tanytarsini
Chelifera
Limonidae
Maruina
Psychodidae
Simuliidae
Stratiomyidae
Baetodes
Leptohyphes
Hydracarina
Oligochaeta
Contulma
Cochliopsyche
Hydrobiosidae
Hydroptilidae
Nectopsyche
8.045
9.67
6.976
5.285
0.318
1.134
10.125
22.023
3.932
7.507
6.788
2.062
2.086
1.046
0.001
5.111
0.813
10.399
1.755
0.938
5.178
4.171
3.903
0.832
3.244
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
stone movement
X²
df
P
0.018 6.924
0.008 4.501
0.031 8.295
0.071 7.304
0.853 0.929
0.567 5.118
0.006 3.564
0.000 6.965
0.140 17.872
0.023 2.230
0.034 1.169
0.357 9.555
0.352 10.605
0.593 6.232
0.999 8.175
0.078 3.632
0.666 5.440
0.006 7.804
0.416 5.467
0.626 8.182
0.075 3.586
0.124 1.278
0.142 14.010
0.660 2.398
0.198 1.691
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
0.140
0.342
0.081
0.121
0.920
0.275
0.468
0.138
0.001
0.693
0.883
0.049
0.031
0.182
0.085
0.458
0.245
0.099
0.243
0.085
0.465
0.865
0.007
0.663
0.792
microhabitat
X²
df
P
7.161
7.928
6.479
3.394
1.348
0.001
2.816
1.697
3.717
2.131
4.653
1.162
1.072
0.715
33.497
0.923
30.593
3.132
0.020
0.011
0.002
7.737
10.863
3.350
3.683
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.007
0.005
0.011
0.065
0.246
0.972
0.093
0.193
0.054
0.144
0.031
0.281
0.300
0.398
0.000
0.337
0.000
0.077
0.887
0.918
0.967
0.005
0.001
0.067
0.055
4. Mid-term experiment of recolonization and composition of stone
fauna. Effect of local hydrological conditions and seasons.
Data of this experiment come from the marked cobles located in high velocity
and low velocity areas during dry and wet seasons, and were sampled from 7
days up to 25 days with 5-7 days intervals.
Seasons
From the 41 taxa registered at both seasons, few but abundant taxa showed
significant seasonal changes (Kruskal-Wallis analysis of variance, Appendix 2;
Appendix3).
Thus,
Chironominae,
Tanytarsini
and
Leptohyphes
were
43
significantly more abundant at dry season, and Tanypodinae was more
abundant in wet season. At community level, the Non Metric Multidimensional
Scaling (NMDS) did not detect significant differences in the composition of the
community between seasons.
Colonization days
Using the NMDS, no differences for community composition were found
between colonization days (7 to 25 days) during both seasons. Only at wet
season the two-way ANOVA performed found significant differences (Table 7)
between days in total density. The Tukey post hoc comparisons detect that at
th
th
the day 17 total density was significantly higher than day 12 (p= 0.029) and
th
day 22 (p=0.013) . We did not found significant differences of the other
community metrics with the two-way ANOVA performed with colonization days
and microhabitats (Table 7) at each season. From all the taxa analysed, only
Orthocladiinae showed a significant relationship with colonization days
(Appendix 2, Kruskal-Wallis test), being higher at day 14 and 25 at the dry
season. No relationship was found between colonization days and density of
different taxa at the dry season. The relation between taxa density and days was
weak in the wet season, and again only Orthocladiinae showed a significant
increase at day 17 (Kruskal-Wallis test Appendix 2).
Microhabitat
Community metrics differed among microhabitat types (Table 7) with higher
richness, density and diversity at stones located in rapid areas, independent
form colonization days. The taxa that showed significant microhabitat preference
were those that are more abundant at rapid areas (Kruskal-Wallis analysis,
Appendix 2).
At the dry season only total density showed a significant increase at stones
located in rapids compared to those in refugia (Table 7). This pattern is opposite
to benthos Surber samples (Chapter 1), were the low velocity samples showed
denser and more diverse communities (the same was true for the short term
colonization experiment). At dry season, 39 taxa were found in the stones, from
44
which 35 were present on rapid stones and 31 on slow flow stones. Several taxa
showed significant differences among microhabitats. While Nectopsyche
represents the 24% of the density of taxa present in slow flow areas, Simuliidae
was the opposite, being 17% of taxa density in stones located in rapids (Table
5). When comparing the abundance of different taxa in stones with Surber and
drift samples, Orthocladiinae and Nectopsyche had similar percentages of total
densities at both drift and benthos Surber samples and therefore colonizers may
arrive from both sources to the stones (Table 5, Fig. 4).
However the
dominance of Simuliidae in stones located in rapids could only be related to drift
because of its low density in benthos Surber samples. In this low flow season
there is a markedly decrease on Baetodes densities in both drift and benthos
samples but this taxa still represented an important part of the individuals of
stone fauna (7% of the density of both refugium and rapid located stones).
At the wet season several community metrics showed a significant increase at
stones located in rapids (Table 7), except total density that did not differ
significantly among microhabitats. For example densities of Tanypodinae,
Simuliidae and Baetodes increased at rapids. Despite the lack of differences on
total density during the wet season, mean density of many taxa differed among
stones located in rapids and stone located in low velocity areas (Table 5). More
than 50 % of the density of stones located in rapids was from Baetodes and
Simuliidae, while almost 47% of density in low velocity areas was from
Orthocladiinae, followed by Nectopsyche (16%). In drift and benthos Surber
samples, in this season Orthocladiinae and Baetodes were dominant taxa,
therefore colonization of new stones can come from both sources. Simuliidae, as
in dry season should came mainly from drift (Fig. 4). One interesting fact is that
Nectopsyche, in terms of percentage of abundance is less important in the wet
season in drift and benthos Surber samples, but at the stones located in low
velocity areas this taxa present the highest percentages of density.
45
Figure 4. Diagram that summarizes the colonization process in stones on
Piburja Stream, using season and microhabitat factors. Taxa which mean
density is higher in any of the combinations is indicated. Taxa in bold had
significant differences with the factors (Appendix 2, Kruskal Wallis test) In
addition if taxa had higher abundance in drift or benthos Surber samples is
indicated by ˜ or * respectively.
DRY
Stratiomydae˜
Chironominae
Tanytarsini*
Maurinia
Nectopsyche*
Limoniidae˜
Leptohyphes*
Elmidae*
Diamesinae*
Ochotrichia
Simuliidae˜
Hydracarina
Oligochaeta
Andiperlodes*
Hydrobiosidae
Ceratopogonidae*
Podonominae
Baetodes
Chelifera
Orthocladiinae
Tanypodinae
Scirtidae
Blephariceriidae
Cocliopsyche
Tipula˜
WET
LOW
VELOCITY
LOW / HIGH
VELOCITY
HIGH
VELOCITY
Table 7. Summary of the two way ANOVA of community metrics related to days
of permanence of the stone on the river and microhabitat at Piburja stream,
Ecuador. Data is presented for each season (dry and wet season) and for the
two seasons together (Global).D= colonization days, M=microhabitat (rapid or
refugia), D*M=interaction between the two factors.
Community
metrics
Density
Richness
d
Fisher alpha
H'
46
D
0.003
0.049
0.139
0.147
0.114
Global
M
0.001
0.000
0.002
0.003
0.002
D*M
0.450
0.856
0.967
0.953
0.781
Dry season
D
M
0.004
0.214
0.515
0.333
0.313
0.286
0.378
0.311
0.487
0.307
D*M
0.400
0.653
0.715
0.765
0.629
Wet Season
D
M
D*M
0.023 0.036 0.288
0.157 0.000 0.509
0.171 0.000 0.358
0.183 0.000 0.329
0.373 0.000 0.304
Discussion
We found that overall; there was no difference in invertebrates drift though out
the day, this was true for all the community metrics tested and also for the
density of most taxa density. This result differs with data coming from temperate
areas (Allan 1984, Allan 1987, Allan et al. 1988) where night drift densities can
be up to ten times greater. However, it coincides with previous studies in tropical
streams in which diel periodicity in drift has not been found either (Turcotte and
Harper 1982, Pringle and Ramirez 1998, Jacobsen and Bojsen 2002,
Rodriguez-Barrios et al. 2007). Despite this, two ephemeropterans, Baetodes
and Leptohyphes, showed a clear diel periodicity. This can be related to the
presence of the introduced rainbow trout as reported to high Andean rivers
(Flecker 1992) and agrees with studies that demonstrated that some
Ephemertotera larvae on fish-less streams did not showed diel periodicity in drift,
after contact with fish was established, they maintain nocturnal drift behaviour
even when returned to a fishless habitat (Allan 1978, Allan et al. 1986, Flecker
1992, McIntosh and Townsend 1994, Peckarsky and McIntosh 1998, McIntosh
et al. 2002). Flecker (1992) found this pattern only for Baetis and not for
Baetodes in a high Andean stream from Venezuela, where trout was introduced,
while Turcotte & Harper(1982)
and Jacobsen (2002) did not found diel
periodicity in Baetodes or Baetis in a high Andean streams in Ecuador. We
found this pattern for the most abundant ephemeropterans of the stream (which
suppose a significantly high percentage of the total community density). As our
study is limited in time, and more surveys in several consecutive days with full
and new moon in both seasons are needed to understand the diel periodicity of
macroinvertebrate fauna in any stream, we cannot assure that this pattern is
repetitive in our stream. Unfortunately the introduced rainbow trout is
generalized in Andean rivers from Ecuador and is difficult to find streams without
trout to assess the natural condition of fish-less Andean streams. From our
results, only ephemeropterans seem to have a “fixed” evolved behavioural
response to avoid fish predation in these streams.
47
Like previous studies (Turcotte and Harper 1982, Ramirez and Pringle 2001,
Rodriguez-Barrios et al. 2007) community metrics in drift did not change
between seasons. However, unlike Rodriguez-Barrios(2007) and
Ramirez &
Pringle (2001) we found an increase, and not a decrease, in drift richness and
diversity with higher flow conditions, a pattern reported previously by Turcotte
and Harper (1982). Despite this lack of differences among seasons in
community metrics, community composition had strong differences among
seasons because of the important changes in the dominance of several
abundant taxa. This results are otherwise to those that previously reported
aseasonality on drift densities of tropical streams (Brittain and Eikeland 1988)
and although some evidence of this seasonality has been previously reported
(Boyero and Bosch 2002) this is the first time that differences are reported on
composition among drift and benthos in two different seasons, based on several
consecutive sampling at each season. This seasonality on drift should affect
both benthic and stone recolonization and composition.
Drift propensity clearly differs among seasons and thus behavioural and
catastrophic drift could be differentiated because a great majority of taxa showed
high drift propensity values at wet season. All these taxa had higher benthic
densities at the dry season except Pyralidae, and higher drift densities at wet
season. Among the dominant taxa present in both drift and benthos, only
Simuliidae and Hydracarina seem to suffer catastrophical drift during the wet
season. These differences in drift propensity values for taxa between seasons
suggest that the high values observed at the wet season, are result of a
continuous drag of benthic community during the this season. Dominant taxa,
like Nectopsyche and Baetodes, that had similar drift propensity values among
seasons showed seasonality on benthic Surber densities, indicating that
catastrophic drift is not related to the changes in abundances between seasons.
This is interesting in terms of the strategies that different taxa are using to
maintain their populations and avoid catastrophic drift. We mention previously
(Chapter 1) that Baetodes show high benthic densities at wet season and in
rapids, which could be related to higher availability of fast flow benthos areas in
this season. Contrary, Nectopsyche, show a clear preference of refugium areas
48
in both seasons, being less exposed to be washed out. This preference of low
velocity areas for this taxa could be an indicator of a behavioural trait to avoid
drift that has been reported to other crawling caddisflies (Lancaster et al. 2006)
which permanently avoid high velocity areas and reinforce the results of
microhabitat preferences for the taxa found in Chapter 1.
Interestingly in
Nectopsyche even if the greatest Surber densities in benthos are found at dry
season and during the wet season on recolonized stones located at slow
velocity areas the abundance is even higher than drift or benthos Surber
samples at this season. Therefore we may conclude that this taxa is actively
searching refugium, and definitely prefer low velocity areas.
As expected for tropical areas, recolonization of stones was stabilized after 7
days, and the most common taxa showed clear differences between
microhabitats. Values of community metrics (richness, density, H’, Fishers’ α)
were higher at stones located in high velocity, as has been reported for stonefauna experiments at Australia (Downes et al. 1995) and also at high Andean
streams of Ecuador (Jacobsen 2005). Degree of upturn was only important to
few taxa as Podonominae and Simuliidae, that were strongly affected. This can
also explain the great drift propensity values of these taxa during the wet season
were more stone up turn is expected. Thus, colonization days and microhabitat
are at community level, and for several dominant taxa, the factors that explain
composition of recolonized stones.
Microhabitat played a very important role differentiating communities in both
short-term
and mid-term
recolonization
experiments. The response of
community composition to small scale environmental changes has been
observed in temperate aquatic biota and may be related to many factors
including to predator avoidance (Resetarits 1991, Resetarits 2001, Binckley and
Resetarits 2005). In our study there is a general tendency of richer and more
diverse community at stones located in rapid areas maybe are more related to
abiotic factors and the continuous influence of drift. Dynamics on stone fauna at
mid term (~25 days) varied between seasons, showing and strong influence of
both drift and the actual benthic communities in the composition of either low or
49
high velocity located stones. While Orthocladiinae was an important component
of drift and benthos in both seasons, other taxa had different percentage of
mean benthos and drift densities, therefore differences may be due to the fact
that colonizing taxa preferentially came from one source or other. Thus
Simuliidae showed high drift propensity rates at both seasons and is an
important component of drift diversity, and as a consequence colonizing
Simuliidae on stones located at rapids should came from drift.
With the present study we were able to synthesize a general view of the
seasonal processes of drift and dispersion of macroinvertebrate taxa and the
consequences of these processes in colonization of new habitat in a high
altitude Andean stream. We still have to understand the details of the population
dynamics, especially with taxa that present or could present more longevity of
the aquatic instars and of taxa that are fully aquatic. Fill the gap of information
on the biology of aquatic insects in neotropical, and specially the Andean areas,
is of primary importance to address this and other ecological questions and
enhance the understanding of this aquatic ecosystems.
50
APPENDIX 1. Mean values of drift and benthos densities and drift propensity of
taxa present at both seasons at Piburja stream, Ecuador.
Drift density m3
taxa
WET
DRY
Heterelmis
0.014
0.004
Total Elmidae
0.035
0.011
Elodes
0.002
0.017
-
Benthos density
m2
WET
5.952
DRY
Drift propensity
WET
DRY
12.245
0.002
0.000
19.558 108.673
0.002
0.000
5.357
0.001
0.003
2.278
1.276
0.005
0.000
4.252
18.367
0.013
0.003
1.701
1.822
Scirtes
0.011
Total Scirtidae
0.054
0.056
Hyalella
0.009
0.004
4.082
0.005
0.001
Ceratopogonidae
0.065
0.146
60.374 172.959
0.001
0.001
Chironominae
0.217
0.355 103.741 147.959
0.002
0.002
Diamesinae
0.048
0.067
57.143
0.057
0.001
Orthocladiinae
1.171
1.223 344.388 411.224
0.003
0.003
Podonominae
0.258
0.005
1.701
12.755
0.152
0.000
Tanypodinae
0.109
0.021
24.660
17.347
0.004
0.001
Tanytarsini
0.075
0.064
21.259
68.878
0.004
0.001
Chelifera
0.094
0.127
16.156
37.245
0.006
0.003
Muscidae
0.062
0.002
1.367
2.296
0.045
0.001
Maruina
0.065
0.175
4.252
10.204
0.015
0.017
Simuliidae
0.590
0.712
22.959
68.878
0.026
0.010
Stratiomyidae
0.092
0.161
5.102
7.143
0.018
0.023
Tipula
0.037
0.154
8.503
7.653
0.004
0.020
Andesiops
0.007
0.020
7.653
7.143
0.001
0.003
Baetodes
1.139
0.167 121.599
41.327
0.009
0.004
Leptohyphes
0.034
0.005
11.905
85.714
0.003
0.000
Planorbiidae
0.059
0.049
2.551
2.551
0.023
0.019
Hydracarina
0.573
0.471
15.306 121.939
Pyralidae
0.056
0.014
Nematomorpha
0.019
0.024
Oligochaeta
0.045
0.128
Andiperlodes
0.017
0.031
Claudioperla
0.007
-
Anacroneuria
0.001
-
0.850
0.037
0.004
1.786
0.025
0.008
2.733
5.102
0.007
0.005
39.966
87.245
0.001
0.001
0.850
20.833
0.020
0.001
8.503
14.881
0.001
0.000
0.456
0.510
0.003
0.000
2.278
Cochliopsyche
0.017
0.018
5.952
35.714
0.003
0.001
Hydrobiosidae
0.051
0.041
5.102
26.020
0.010
0.002
Ochrotrichia
0.297
0.250
10.204
73.980
0.029
0.003
Nectopsyche
Tricladida
0.256
-
0.724
0.004
39.116 232.653
1.701
11.224
0.007
0.000
0.003
0.000
51
APPENDIX 2. Summary of Kruskal-Wallis p-values for macroinvertebrate density for: season, microhabitats and colonization
days at Piburja stream, Ecuador. Pooled data is marked as GLOBAL and at each season the p-values for the factors:
colonization days and microhabitats.
GLOBAL
DRY SEASON
COLONIZATION
SEASON
TAXA
p
DAYS
MICROHABITATS
p
WET SEASON
COLONIZATION
p
DAYS
COLONIZATION
MICROHABITATS
p
p
DAYS
MICROHABITATS
p
p
Elmidae
0.130
0.309
0.730
0.298
0.56
-
-
Georissidae
0.388
0.407
0.376
0.405
0.307
-
-
Elodes
0.388
0.407
0.376
0.405
0.307
-
-
Prionocyphon
0.219
0.118
0.208
0.114
0.144
0.409
0.166
Staphylinidae
0.247
0.407
0.376
1.000
1.000
0.409
0.470
Hyalella
0.820
0.587
0.208
0.405
0.307
0.409
0.470
Ostracoda
0.388
0.407
0.258
0.405
0.328
-
-
Blephariceridae
0.247
0.407
0.258
1.000
1.000
0.409
0.166
Ceratopogonidae
0.124
0.958
0.985
0.298
0.547
0.409
0.166
Chironominae
0.001
0.438
0.033
0.095
0.324
0.530
0.047
Diamesinae
0.018
0.160
0.118
0.134
0.223
-
-
Orthocladiinae
0.008
0.072
0.172
0.006
0.112
0.013
0.375
Podonominae
0.492
0.160
0.464
0.505
0.676
0.114
0.604
Tanypodinae
0.000
0.756
0.043
1.000
1.000
0.672
0.002
Tanytarsini
0.000
0.588
0.917
0.454
0.365
1.000
1.000
Chelifera
0.437
0.705
0.027
0.251
0.043
0.815
0.162
Limonidae
0.079
0.110
0.206
0.099
0.333
-
-
Clognia
0.130
0.253
0.688
0.256
0.498
-
-
52
GLOBAL
DRY SEASON
COLONIZATION
SEASON
DAYS
WET SEASON
COLONIZATION
MICROHABITATS
DAYS
COLONIZATION
MICROHABITATS
DAYS
MICROHABITATS
Maruina
0.219
0.587
0.847
0.581
1.000
-
-
Simuliidae
0.044
0.356
0.000
0.383
0.000
0.193
0.000
Stratiomyidae
0.219
0.545
0.208
0.545
0.144
-
-
Tipula
1.000
0.345
1.000
1.000
1.000
-
-
Andesiops
0.397
0.251
0.910
0.581
0.144
0.260
0.198
Baetodes
0.222
0.240
0.000
0.217
0.003
0.538
0.000
Leptohyphes
0.002
0.475
0.009
0.292
0.052
0.906
0.219
Hydracarina
0.087
0.119
0.849
0.432
0.632
0.289
0.131
Pyralidae
0.388
0.407
0.376
0.405
0.307
-
-
Nematoda
0.820
0.545
0.875
0.351
0.328
0.409
0.470
Oligochaeta
0.615
0.196
0.627
0.700
0.974
0.048
0.571
Andiperlodes
0.645
0.446
0.909
0.502
0.976
0.086
0.958
Claudioperla
0.388
0.407
0.258
0.405
0.328
-
-
Contulma
0.955
0.482
0.848
0.256
0.498
0.114
0.635
Cochliopsyche
0.942
0.014
0.859
0.085
0.532
0.157
0.654
Atopsyche
-
-
-
0.493
0.152
-
-
Cailloma
-
-
-
0.329
0.011
-
-
Hydrobiosidae
0.208
0.169
0.002
-
-
0.277
0.198
Smicridea
0.388
0.407
0.258
0.405
0.328
-
-
Ochrotrichia
0.782
0.448
0.027
0.460
0.029
0.805
0.383
Nectopsyche
0.221
0.095
0.486
0.251
0.674
0.250
0.151
Tricladida
0.740
0.309
0.047
0.114
0.162
0.409
0.166
53
CHAPTER 3:
Leaf litter organic matter dynamics and
associated invertebrates in a high
altitude tropical Andean stream
CHAPTER 3: Leaf litter organic matter dynamics and associated invertebrates
Introduction
Inputs of leaf litter from riparian vegetation comprise a significant source of
energy and nutrients for headwater streams (Siccama et al. 1970, Anderson and
Sedell 1979, Wallace et al. 1997, Graça 2001). Some studies have evaluated
the importance of litter dynamics throughout the ecosystem using energy
budgets (Gosz et al. 1972, Likens 1972, Fisher and Likens 1973, Meyer et al.
1981). In a forested stream, nearly all available energy comes from the forest. In
temperate areas, the interactions between the inputs (leaves), the decomposers,
the detritivores and the physical factors have been widely studied (Anderson and
Sedell 1979, Suberkropp and Wallace 1992, Suberkropp and Chauvet 1995,
Wallace and Webster 1996, Wallace et al. 1997, Graça 2001, Graça et al.
2001b). The general pattern is that an important part of this energy is not used
by consumers in the headwater streams, but is exported downstream by the
current (Likens et al. 1970, Fisher and Likens 1973) and the trophic link between
headwaters, which receive the greatest input, and the lower parts are mediated
by decomposers and detritivores (Graça 2001).
For tropical streams, there is a lack of quantitative information on inputs, storage
and cycling of allochthonous inputs, even though these streams often drain
heavily forested catchments and the litter inputs should be a major energy
source (Graça et al. 2001a, Colón-Gaud et al. 2008). In the context of the low
seasonality patterns of tropical high altitude regions, litter inputs may enter to
tropical streams throughout the year, unlike in most temperate streams, where
most allochthonous materials enter in the fall. Therefore, understanding the
potential importance of this continuous entrance is essential for comprehending
the functioning of tropical streams. However, despite these continuous inputs,
peak litterfall in tropical streams is often associated with storm events (Covich
1988), or with the beginning (Gonçalves et al. 2006) or the end of the wet
season (Afonso et al. 2000).
Channel morphology and habitat heterogeneity are key factors determining the
capacity of streams to retain leaf litter (Speaker et al. 1984, Lamberti et al.
57
1989). In general, narrow rough-bottomed streams are more retentive
(Mathooko et al. 2001). In tropical streams, litter accumulation and distribution in
stream beds are regulated not only by inputs from the adjacent forest and
channel morphology but also by the magnitude and variability of discharge
(Pearson et al. 1989). Therefore, temporal changes in stream hydrology (such
as high discharge events) result in greater hydraulic power, thus decreasing
retention capacity (Larrañaga et al. 2003) and increasing the movement and
redistribution of leaf litter. The general pattern is that greater amounts of
standing stock are accumulated during the dry season (Covich 1988). However,
the amount of standing stock accumulated in tropical streams (on a yearly basis)
2
ranges widely, from 35 g AFDM/m (Friberg et al. 1997) to more than 1000 g
2
AFDM/m (Colón-Gaud et al. 2008).
The relation of organic matter standing stock and macroinvertebrates is well
documented for temperate areas, but the study of this relation has been rare for
tropical streams (Mathuriau et al. 2008). In general, leaf litter can be used by
macroinvertebrates as refugium (Palmer et al. 1996) or as a food source
(Wallace and Webster 1996, Graca 2001, Graca et al. 2001b). Contrary to
temperate streams, shredders in tropical streams seem to be an unimportant
part of the community (Ramirez and Pringle 1998, Mathuriau and Chauvet 2002,
Mathuriau et al. 2008). However, in contrast to the pattern found in the lowland
tropical streams, new evidence shows that in tropical mountain streams
shredders may be important (Cheshire et al. 2005). On the other hand, gut
content analysis from neotropical detritivorous insects has shown that each taxa
may have at least two trophic species traits (Wantzen et al. 2005, Wantzen and
Wagner 2006) or that each taxa is feeding on more than one source (Tomanova
et al. 2006). Therefore, the importance of leaf litter as a food source remains
unclear for tropical mountain streams. Its documented importance would
probably change
with
the
availability
of
feeding
sources,
because
macroinvertebrate taxa of tropical areas seem to have highly flexible feeding
habits (Covich 1988).
58
With this study, we want to assess the importance of leaf litter for the
macroinvertebrate communities of an Andean forested stream, and our specific
objectives are: 1) Describe the annual litter inputs in a high altitude Andean
stream; 2) define whether there are seasonal differences in the transport and
retention of Coarse Particulate Organic Matter (CPOM); 3) study whether there
are differences in litter standing stock between seasons and the potential
relationship of this stock with macroinvertebrate community assemblages; and
4) assess whether litter is an important food source for the most abundant
invertebrate taxa based on gut content analysis.
Study area
We performed the study at the Piburja stream, a first order stream in the
Oyacachi river basin, located at the Cayambe-Coca Ecological Reserve
(RECAY) in Ecuador (0º13’ S, 78º 03’ W) at 3300 m asl. We studied a 70 m
reach with a minimum width of 1.60 m and a maximum of 4.0 m. The
surrounding vegetation was an evergreen forest with Alnus acuminata trees.
Pluviometric records (10 years) from nearby localities show lower rainfall from
December to February (dry season) compared, with May to September. Rainfall
ranged between 32.56 and 244.11 mm per month during the year of the study.
Total precipitation in Oyacachi town is ~ 1600 mm/year, and mean temperature
is ~10ºC. Precipitation is probably 2 or 3 times greater than potential
evapotranspiration, which implies very humid conditions during the whole year
(Skov, 2000). The most important seasonal changes in the physico-chemical
characteristics of the stream are a significant increase in flow and water velocity
during the wet season. A detailed description of the physical-chemical
characteristics of the stream can be found in Chapter 1.
In order to assess the litter input, we sampled the stream once a month during
one year. Movement, retention and standing stock were sampled in two different
seasons, wet (April-May 2006) and dry (January-February 2007). Stream
hydrology varied accordingly with this seasonality.
59
Annual litter input
To understand vertical litter input, we estimated it for a 70m reach of one first
order stream which (as described previously) is surrounded by a high altitude
riparian forest. Following the methodology of Elosegi and Pozo (2005), we
2
placed 8 plastic litter-fall traps, each with an area of 0.082 m . The traps were
placed randomly, and were suspended over the stream channel and banks
using ropes. Every month, we collected the material in the traps, and then air
dried and preserved the material in paper mesh bags for weighing later in the
laboratory. Subsequently, samples where dried at 60ºC for 2 days and weighed.
Samples were then ignited in a muffle furnace oven at 500ºC for 4 hours, to
2
determine ash-free dry mass per m per day of each sample. Litter input was
then related to monthly precipitation using Pearson correlation. We transformed
the mean day rainfall of each month into two categorical variables by subtracting
the lowest value from the highest value and dividing by 2. In this way, values of
less than 3.98 mm rain/day would be considered “dry periods” and values
greater than this value would be considered “wet periods”. We used ANOVA
analysis to assess differences between rainfall periods and litter input. The
differences between mean monthly litter inputs were analyzed using ANOVA
with Fisher’s least significant difference (LSD) post hoc analysis. Litter input
values were log transformed to achieve homogeneity of variance.
Litter movement
To understand litter movement through the system, we placed four to six drift
2
nets (495 cm of cross-sectional area) in the stream for 20 minutes (wet season)
and 1 hour (dry season), four times during each sampling period. Using a Global
water flow probe, we measured the average water velocity at the net mouth at
the beginning and at the end of each sampling time. All the material in the net
was collected and taken to the laboratory. At the lab, all material greater than 1
mm was dried at 60ºC for 2 days and weighed. Afterwards, all samples were
incinerated at 500ºC for 4 hours to determine ash-free dry mass (AFDM) per m
3
3
of each sample. The AFDM/m values were log transformed and correlated with
60
water flow using Pearson’s correlation. In addition, ANOVA was used to assess
seasonal differences in the litter movements.
Litter retention
We chose two species of plants: Alnus acuminata and Eucalyptus globulus, to
assess differences in the retentiveness of flexible and hard leaves, respectively.
We choose Alnus acuminata because it is a common tree in the riparian forest of
the Piburja stream, and Eucalyptus globulus because it is a common introduced
species in Andean forests. In addition, the retention of Eucalyptus leaves has
been previously assessed and constitutes a good leaf reference (Pozo et al.
1997, Canhoto and Graça 1998, Bañuelos et al. 2004). We collected a hundred
air-dried recently fallen leaves of each species and marked them with a small
waterproof ink mark. The night before the experiment, we soaked the leaves
overnight to give them neutral buoyancy.
We placed a stop net of 2 mm mesh size and released the leaves one by one 40
metres upstream of the net. We allowed the stream to disperse the leaves for
one hour. After one hour, we made a multiple point collection (Elosegi, 2005);
we counted the leaves that had arrived at the net, and went upstream to
measure the distance travelled by each retained leaf and the habitat type where
it was retained. We measured discharge before releasing the leaves and after
collecting all the leaves retained. We conducted this procedure once during each
of the two hydrological seasons (wet and dry). To obtain the instantaneous
retention rate and the average travel distance of each leaf type during each
season (Elosegi, 2005), we plotted the number of released leaves in transport to
the distance travelled. Data were then fitted to the exponential decay model
k*d
(Young et al. 1978), Ld=L0*e , were L0 is total leaves recovered, Ld the number
of leaves still in transport at distance d (40m), and k the instantaneous retention
rate. The average travel distance was 1/k. Finally, we also recorded the
percentage of leaves of each species retained at different substrates and
habitats of the stream to evaluate their importance in the retention process.
61
Standing stock
2
We took Surber samples (196 cm , 250 µm mesh size) on five dates in each
season, with an interval of four to six days between samplings. To avoid
damaging our own sites within our reach we always sampled moving forward in
the upstream direction. At the lab, we removed all macroinvertebrates from each
sample and then filtered all the material to select the CPOM (>1 mm). CPOM
was dried at 60ºC for 2 days, and weighed. Afterwards, these samples were
2
incinerated at 500ºC for 4 hours to determine ash-free dry mass per m of each
sample.
Macroinvertebrates were identified to the lowest taxonomic level possible
(usually genera, but some small individuals were identified to family level).
Oligochaeta, Tricladida, Hydracarina and Ostracoda, were identified to order and
Chironomidae to sub-families. We used Merrit & Cummins (2007) and
Tomanova (2006) to assign macroinvertebrates to functional feeding groups.
The community metrics used to analyse community patterns were total density,
richness, Shannon Weiner index, and Fisher’s alpha index.
To analyse the differences in organic matter standing stock between seasons,
we used the Mann-Whitney U test. This non-parametric test was used because
after transformation the data was did not achieve a normal distribution. We used
Spearman correlations to relate the standing stock, functional feeding groups
and community metrics.
Gut content analysis
The digestive tube was dissected and mounted on a plate with a solution of 1:1
glycerine and lactic acid. The walls of the digestive tubes were broken using
pliers to leave all the material in the plate. A plaque was made for each
individual, and 10 random visual fields of 100x to 400x magnification were
analyzed for each plaque, in order to classify the occupancy rate of each food
category: a) CPOM (> 1 mm), b) Fine particulate organic matter (< 1 mm,
FPOM), c) diatoms or d) chitin. The data was expressed as a percentage of
each category of food per visual field. To analyze seasonal differences in food
62
preferences, we performed an ANOVA. The percentages were previously
arcsine transformed.
We analysed five individuals of each taxa, from each season. Only the
shredders (classified using Merrit and Cummings, 2007 and Tomanova, 2006)
and the most abundant taxa were analysed. After a preliminary analysis, the
following taxa were included: Baetodes, Chironominae, Claudioperla, Hyallela,
Leptohyphes,
Nectopsyche,
Oligochaeta,
Orthocladiinae,
Prionocyphon,
Simuliidae and Tipula. Although Tanytarsini, Hydracarina and Ochotrichia were
also important but due to the size of the individuals (too small), we were not able
to analyse the gut content with the applied methodology. Probezzia
(Ceratopogonidae) showed an empty gut in all individuals analysed and we do
not know if this was due to the gut content analysis methodology or to the
preservation methodology (formalin 4%). All statistical analyses were performed
with Statistica 6.0 and SPSS 15.
Results
1. Annual litter input
2
Mean monthly g AFDM/m /day ranged from 0.12 on July 2006 to 1.50 on
2
February 2007. The annual mean litterfall value was 0.59 ±0.42 g AFDM/m /day.
2
-1
The total annual litter input was 216.07 g AFDM/m / y . The litter input was not
2
correlated with mean daily rainfall, but the highest values of AFDM/m /day
coincided with lower rainfall values and the lowest input values coincided with
the highest rainfall values (Fig 1).
The one way ANOVA (Table 1) of rainfall periods showed that there were no
2
significant differences in AFDM/m /day between the two rainfall periods.
However, the ANOVA between the months and the litter input showed significant
differences (F =2,22, df = 11, p =0 . 022). The main differences were between
the period of May to August 2006, with lower litterfall values, and the period of
February to April 2007, with higher litterfall values (Fig. 1, Table 2).
63
2.2
10
Plot 1
Col 26 vs Col 27
2.0
9
1.8
8
7
1.4
1
1.2
6
2
5
1.0
4
0.8
rain (mm/day)
2
AFDM/m /day
1.6
3
0.6
0.4
2
0.2
1
0.0
0
6
6
6
6
6
7
7
6
6
7
7
6
y 0 ju n 0 ju l 0 g o 0 e p 0 o c t 0 o v 0 e c 0 ja n 0 e b 0 a r 0 p r 0
f
a
s
n
d
a
m
ma
2
Figure 1. Monthly gAFDM/m /day of litterfall and mean mm rain/day at the
Piburja Stream, Ecuador. Arrows mark: 1) the lowest litterfall value, which
coincided with the highest rainfall value, and 2) the highest litter fall value, which
coincided with the lowest rainfall value.
Table 1. Analysis of Variance (ANOVA) table to test differences of litter input
2
(gAFDM/m /day) during different rainfall periods (dry period = values <3.98 mm
rain/day; rainy period = values >3.98 mm rain/day), at the Piburja stream,
Ecuador.
SS
D.F
MS
F
p
28.648
1
28.648
44.088
0.000
rainfall periods
0.532
1
0.532
0.819
0.368
Error
50.684
78
0.650
Intercept
64
Table 2. Summary of the multiple comparison procedure to determine significant
differences (p<0.05) in monthly litter input (gAFDM/m2/day) at the Piburja
stream, Ecuador. The method used to discriminate among the means is Fisher's
least significant difference (LSD) procedure.
comparisons
Sig.
May
06Feb
07
*
May
06Apr
07
*
Jun
06Feb0
7
*
Jun
06Apr
07
*
Jul
06Feb
07
*
Jul
06Mar
07
*
Jul
06Apr
07
*
Ago
06Feb
07
*
Ago
06Apr
07
*
Nov
06Feb
07
*
Jan
07Feb
07
*
Difference
-0.55
-0.44
-0.56
-0.45
-0.64
-0.42
-0.53
-0.61
-0.50
-0.46
-0.38
+/- Limits
0.35
0.38
0.36
0.39
0.36
0.37
0.39
0.40
0.42
0.40
0.36
2. Litter movement and retention
a. Litter movement
3
Mean values of AFDM/m collected in drift nets were slightly higher during the
dry season than during the wet season, but the ANOVA (F=0,11; df=1; p=0.75)
showed that differences between seasons were not significant (Fig 2). There
3
was no correlation between flow and AFDM/m (Pearson correlation: r=0,466
p=0,244).
b. Leaf retention
During the dry season, no leaves of the two species arrived at the net. In
contrast, during the wet season, some leaves were retained at the net that was
located at 40 m, implying a faster downstream transport during the wet season.
Almost 30% of Eucalyptus leaves were in movement (arrived at the net) and
most alder leaves were retained in the first 25 meters of the section. The
average travel distances (ATD) during the wet season were 41.7 m (Alder) and
454.5m (Eucalyptus); during the dry season, ATD values were 24.4 m (Alder)
and 8.7 m (Eucalyptus).
65
0.22
0.20
g AFDM/m3
0.18
0.16
0.14
0.12
0.10
0.08
0.06
WET
M ean
±SE
DRY
season
Figure 2. Mean confidence intervals of the CPOM in transport (g AFDM/m3)
during the wet and the dry seasons at the Piburja stream, Ecuador.
Stones and high velocity areas were the abiotic structures and habitat that
retained Eucalyptus and Alder leaves during the wet season (Table 3). During
the dry season, Eucalyptus leaves were caught up by stones (50%) and Alder
leaves in detritus (55.17%). During this season, both kinds of leaves were found
mainly in shallow habitats. Overall, there was a higher retention of both types of
leaves during the dry season than during the wet season.
Table 3. Percentage of Alder and Eucalyptus leaves retained at different stream
substrates, depths and velocities at Piburja stream, Ecuador during the wet and
dry seasons.
Alder
RETAINING STRUCTURE
Gravel and pebbles
Cobbles and Boulders
Natural barriers (debris dams)
TOTAL
DEPTH/VELOCITY
shallow/low
deep/low
shallow/high
deep/high
TOTAL
66
Eucalyptus
DRY
WET
DRY
WET
15
30
55
100
0
73
27
100
11
50
39
100
0
77
23
100
25.84
12.36
35.96
25.84
100
11.39
5.06
12.66
70.89
100
40.66
21.98
37.36
0
100
10.34
3.45
77.59
8.62
100
100
% of leaves in transport
alder wet
alder dry
80
eucalyptus wet
eucalyptus dry
60
40
20
0
0
5
10
15
20
25
distance traveled (m)
30
35
40
Figure 3. Downstream decrease in the number of leaves transported in Piburja
stream, expressed as a percentage of Eucalyptus and Alder leaves in transport.
During the dry season, both leaf species were retained in the first 25 m. During
the wet season, Alder leaves presented a higher retention compared to
Eucalyptus leaves.
3. Organic matter standing stock and relation with invertebrate assemblages
Standing stock of benthic particulate organic matter (CBOM) showed significant
differences between seasons (Fig 4a). The amounts of standing stock were
significantly higher during the dry season (Mann-Whitney: U=1149.0, p=0.000).
Diversity, density and richness of macroinvertebrates were positively correlated
with the standing stock (Table 4) when data was pooled (wet and dry season
2
together). During the wet season, g AFDM/m was significantly correlated with
richness and diversity, while, during the dry season, no significant relationship
was found between the amount of standing stock and shredder abundance or
other community metrics.
For the pooled data, all functional feeding groups (FFG) found in the stream had
a strong correlation with the standing stock, except the scrapers (Table 4). We
found the same pattern when we analyzed the wet season data alone (Table 4),
but no correlations were found during the dry season. In both seasons, the
gathering collectors were the most abundant FFG (Fig 5). Density of all FFG
was significantly higher during the dry season as compared to the wet season
67
(Table 5). Orthocladiinae (collector gatherers) were the most abundant taxa in
both seasons (17 and 34 % of total abundance in dry and wet seasons,
respectively). In the dry season, Nectopsyche (10%), Ceratopogonidae (7%) and
Chironominae (6%) were also dominant taxa. In the wet season, Orthocladiinae
(34%), Baetodes (12%), Chironominae (10%) and Ceratopogonidae (6%) were
the dominant taxa.
b
a
Figure 4. Comparison of (a) standing stock (as CBOM) and (b) shredder density
between seasons at Piburja stream, Ecuador. The increase of standing stock (a)
in the dry season is similar to the increase of shredder density (b).
Table 4. Summary of Spearman’s correlation analysis of standing stock (g
2
AFMD/m ) and community parameters and shredders density at the Piburja
stream, Ecuador during both seasons studied.
Community
Parameters/
Functional Feeding
Groups
Two seasons
(wet + dry)
Spearman
p
R
Richness (S)
Density
Shannon Index (H')
Fishers'α
gathering collectors
filtering collectors
shredders
scrapers
predator
0.835
0.761
0.827
0.681
0.308
0.426
0.293
0.155
0.326
68
0.000
0.000
0.000
0.001
0.000
0.000
0.000
0.053
0.000
WET
Spearman
R
0.769
0.588
0.745
0.164
0.273
0.531
0.421
0.232
0.432
p
0.009
0.074
0.013
0.651
0.042
0.000
0.001
0.085
0.001
DRY
Spearman
R
-0.018
0.261
0.030
-0.176
-0.013
0.167
0.112
-0.045
-0.014
p
0.960
0.467
0.934
0.627
0.895
0.096
0.269
0.656
0.886
2
Table 5. Density (ind/m ) changes in functional feeding groups between
seasons at the Piburja Stream, Ecuador. *= Significant differences between
seasons with the Kruskal-Wallis analysis of variance.
Wet
Dry
Mean
Mean
S.E
S.E
gathering collectors*
395.408
94.045
712.755
82.713
filtering collectors*
29.085
225.000
33.008
84.184
shredders*
166.667
50.992
446.429
151.891
scrappers*
42.767
174.490
32.605
140.306
predator*
119.048
41.619
401.020
70.006
Table 6. F and p values of the one way ANOVA of feeding preferences between
the wet and the dry season based on the % gut content of several taxa and
shredders (arcsine transformed values) at the Piburja stream.
Taxa
df.
F
p
Baetodes
Claudioperla
Hyalella
Nectopsyche
Oligochaeta
1
1
1
1
1
1
0.668
2.104
3.897
0.533
3.491
1.402
0.025
0.437
0.185
0.084
0.486
0.099
0.270
0.877
Prionocyphon
Simuliidae
1
4. Invertebrate gut content analysis
Fine Particulate Organic Matter (FPOM) and CPOM were the most consumed
resources by all the taxa analysed (Fig 6). Feeding preferences did not change
with seasons (Table 6). Diatoms and chitin were only found occasionally and in
small amounts (Fig 6.) Based on the literature, we hypothesized that Hyallela,
Chironominae, Tipula Claudioperla, Nectopsyche and Prionocyphon were
shredders. This was generally confirmed by the gut content analysis, but these
taxa were also feeding on FPOM. Chironominae was feeding on only FPOM,
because we did not find CPOM in their guts. On the other hand, Baetodes and
Oligochaeta, considered as collectors were also feeding on CPOM in important
percentages. Baetodes gut analysis showed that CPOM was more important
than FPOM in the gut content analysis. Based on the gut content of the taxa
analysed, which are representative of the 43.65% of the total abundance during
the dry season and 60.73% of the total abundance during the wet season, we
69
showed that CPOM and FPOM are the main food resources of the community
(FPOM 65% and CPOM 35%).
predator
5%
scrappers
20%
predator
21%
scrappers
5%
shredders
5%
filtering
collectors
5%
gathering
collectors
64%
Funtional Feeding Groups Dry
Season
shredders
4%
gathering
collectors
60%
filtering
collectors
11%
Functional Feeding Groups Wet
Season
Figure 5. Percentage of density of the macroinvertebrate functional feeding
groups at the Piburja stream (Ecuador) during dry (a) and wet (b) seasons.
Figure 6. Percentage of gut content for the most abundant taxa and some taxa
considered shredders at Piburja stream in Ecuador. Preferences of the four
categories are identified by colours at each season.
70
Discussion
This high altitude tropical stream receives permanent but irregular litter input
through out the year. In contrast to other tropical streams at lower altitudes
(Colón-Gaud et al. 2008), we did not find a strong relationship between rainfall
and litter input. Although there is some pattern of increase of litter input during
the drier months, this pattern was not significantly different. Interestingly, there
are large significant differences in litter input between months; for example, the
2
amount received during July was 0.125 g AFDM/m /day and during February
2
was 1.50 g AFDM/m /day. Humidity in Oyacachi is high throughout the year
(Skov 1999), and we can thus assume that the vegetation is not suffering
extreme changes in moisture (water scarcity) like other tropical forests.
Therefore, leaf loss is less predictable and this could weaken the relationship
between rainfall and litterfall. The wind probably plays a more important role in
litter input dynamics, as is suggested for some Brazilian streams (Afonso et al.
2000). Total annual inputs were more similar to temperate or Mediterranean
streams than to tropical steams (see values cited in Abelho ( 2001)). These
lower litter inputs, compared to the values of lower parts of the Andean ranges,
could be related to the harsher climatic conditions in the highlands. Although our
primary objective in this study was not to assess the variation of litter inputs with
altitude, this is an important issue for the understanding the ecology of tropical
highland rivers, which seem to behave differently from the lowland tropical
rivers.
The amount of litter transported in the water column was not significantly
different between the dry and the wet seasons but, as the present study was
done during only two seasons, more detailed study over time is needed to
understand CPOM transport during the year. However, the amount of litter
retained was higher during the dry season, potentially because of greater
obstruction of rocks and other structures that appear when there is less water in
the stream. Retention of the two species tested, varied between seasons, with a
clear tendency of alder leaves to accumulate in leaf packs (debris dams) in the
dry season and rocks in the wet season. Eucalyptus leaves were retained in
71
rocks in both seasons. These differences could be attributed to the leaf species
characteristics, as hard leaves (eucalyptus) are more difficult to retain and are
easily transported (Canhoto and Graça 1998). The retaining structures can vary
largely in rivers within the same geographical area that may differ in spatial
heterogeneity. This is especially true when abrupt changes of slope are present,
as in the Andean ranges. In the Piburja stream, as in other studies of tropical
(Mathooko et al. 2001) and Mediterranean (Canhoto and Graça 1998)
headwater streams, these structures were mainly debris dams and rocks.
Due to the high retention rate, we also found a higher accumulation of organic
matter (or standing stock) in the stream during the dry season, with respect to
the wet season. This increase in the dry season is also reported for Australian
forested streams (Bunn 1986). However, no relation was found between litter
inputs and standing stock, a pattern previously reported for Iberian streams
(González and Pozo 1996). Interestingly, invertebrates followed the resource;
the more organic matter found, the higher the diversity and abundance of
invertebrates. During the dry season, when there are lower discharge levels
(Chapter 1), and therefore lower disturbance and hydrologic stress, biotic
interactions could be more intense. Consequently, abundance and richness of
different groups might vary due to competition or predation effects.
During the wet season, even though the standing stocks decreased dramatically,
the abundance of different feeding groups (except scrapers) showed strong
correlations with benthic particulate organic matter. This indicates that the
community is tracking the source when is scarce. Our functional feeding group
classification, based on literature, produced results similar to previous studies in
tropical streams (Ramirez and Pringle 1998, Mathuriau and Chauvet 2002,
Wantzen et al. 2005, Gonçalves et al. 2006, Tomanova et al. 2006, Wantzen
and Wagner 2006, Colón-Gaud et al. 2008, Mathuriau et al. 2008) with a small
increase in the importance of shredders. Taking into account the gut content (we
analyzed taxa that represented almost 50% of the total density), we showed that
CPOM constitutes at least the 35% of the feeding sources of all taxa analyzed. If
we compare this value with the 4% or 5% of the organisms that are classified as
72
shredders using the literature, we see that, by using this classification of
functional feeding groups, we are not getting a clear picture of the taxa that
consume CPOM in our stream. Previous works have shown that the functional
feeding groups in neotropical rivers are difficult to assign because taxa
commonly use at least two trophic levels (Tomanova et al. 2006, Wantzen and
Wagner 2006, Tomanova et al. 2007). Even if the community structure of the
Piburja stream was generally similar to those reported in previous work on
Andean rivers (Turcotte and Harper 1982, Flecker and Feifarek 1994, Jacobsen
and Encalada 1998, Pringle and Ramirez 1998, Ramirez and Pringle 1998), we
found important differences when comparing the gut content analysis of Piburja
stream with those of insects in Bolivia Andean ranges (Tomanova et al. 2006).
This suggests that the taxa are feeding on the resources available under the
local conditions, making functional feeding groups a not very useful approach for
neotropical high altitude stream communities. This is because of the generalist
tendency (Covich 1988) and great adaptability to local conditions of the most
abundant taxa. Our findings differ from previous work on Andean streams
(Mathuriau et al. 2008), which analyzed the data based on the functional feeding
groups defined by Merritt et al. (1996) for North-American fauna, and did not find
any relationship with detritus and shredders or collectors. We show here that the
functional feeding groups defined by the literature do not reflect the real
resources consumed in this Andean stream. The differences in altitude and
riparian characteristics could also generate important differences in the
functioning and feeding preferences of macroinvertebrate communities of
different Andean streams.
In conclusion, we show that most of the invertebrates in the stream feed on
organic matter, either CPOM or FPOM, and this suggests that the Piburja
stream is a heterotrophic system. Primary production, principally by diatoms and
aquatic plants might not be very important in this system. This is contrary to
most parts of highland Ecuadorian streams, which are generally autotrophic
systems with a high presence of macrophytes, due to the absence of forest at
high altitudes (Jacobsen and Terneus 2001). The presence of riparian forest,
and the light limitations produced by this vegetation, regulates the ecosystem
73
functioning in this stream and also regulates the community assemblage.
Although the presence of forest in high altitude streams of the Andean ranges is
not common, it can be found in some parts of the northern (Rios 2004) and
central Andes (Acosta 2005) up to 3900 m a.s.l. It is likely that the presence of
riparian vegetation, in addition to the constraints of altitude (such as temperature
and
low
oxygen
pressure)
(Jacobsen
2003,
2008),
regulates
the
macroinvertebrate assemblages and functioning of Andean high altitude
streams. However, Andean headwater streams are highly diverse and largely
unknown (Allan et al. 2006) and with a lack of
complete information is
impossible to make generalizations about the functioning of these ecosystems.
Because of this diversity, we expect a large variety of conditions for functional
ecology in these tropical streams, even if the general structure of the
macroinvertebrate community is similar between sites.
74
CHAPTER 4:
Oviposition of Aquatic Insects
in a High Altitude Tropical
Stream
CHAPTER 4:Oviposition of Aquatic Insects
Introduction
In streams the permanence of populations of aquatic insects depends of the
recruiment of larval populations from the egg masses deposited by adults,
especially after disturbance (Williams and Hynes 1976). Thus, oviposition
(together with drift, see Chapter 2) is of great importance because it establishes
the initial size of populations (Encalada and Peckarsky 2006). It has been
suggested that at large multigenerational scales, external recruitment, compared
to drift or recolonization from hyporheos, is the main mechanism that maintain
populations (Lancaster and Belyea 1997). However, despite the key importance
of oviposition few studies in lotic population dynamics address adult and egg life
stages. Moreover studies that address recolonization of stream substrate are
focused more on benthic larvae arriving from drift than in oviposition studies
(e.g.: Benson and Pearson 1987, Boulton et al. 1988, Boyero and DeLope 2002,
Boyero and Bosch 2004).
Measuring recruitment is an especially challenging problem for organisms with
complex life cycles and high mobile dispersal stages, like aquatic insects
(Peckarsky et al. 2000). In most insects, as in all the animals that lack of
parental care, females can increase their fitness by ovipositing in habitats that
minimize egg and larval mortality and increase larval growth (Rausher 1979,
Hinton 1981, Binckley and Resetarits 2008). Thus, while specialized oviposition
behaviors have been described mainly for terrestrial insects oviposition in
aquatic insects are generally described as non-selective (Hinton 1981). Although
in many aquatic taxa non-specialized reproductive behaviors seem to be the
rule, specialized oviposition behaviors can be found in some of them, including:
Diptera (Gillett 1971, Baba and Takaoka 1989, 1991), Ephemeroptera
(Peckarsky et al. 2000, Encalada and Peckarsky 2006, 2007), Trichoptera
(Lancaster et al. 2003, Reich and Downes 2003a, Reich and Downes 2003b) or
Coleoptera (Binckley and Resetarits 2008).
77
In general, information of reproductive strategies of aquatic insects
is
geographically restricted and is only available for some species of the northern
hemisphere (Hinton 1981, Brittain 1989, Merritt and Cummins 1996). Detailed
information of egg and embryonic development related to environmental
conditions have been mainly generated for temperate species although some
relevant information from neotropical species is available (Jackson and
Sweeney 1995). However, information on selective oviposition and eggs
characteristics of aquatic insects is particularly scarce for southern Hemisphere
taxa (Reich 2004).
For lotic habitats that are highly variable and unpredictable in flow conditions,
hydrological disturbance might be a key component for larval mortality originated
by catastrophic drift. In these ecosystems one important process for community
permanence may be recruitment by oviposition, but the importance of this
mechanism in lotic ecosystems is virtually unknown (but see Encalada et al in
prep). Due to seasonal constrains, one might expect that in high latitude
streams, recruitment by oviposition is only important during spring and summer
months, when insects have emerged out of the water to reproduce and return to
lay their eggs. Certainly, univoltinism is common life history strategy for insects
living in lotic systems in cold and temperate zones and consequently adult
stages can cope with more favorable environmental conditions for flying and
reproduction. In contrast, in tropical lotic systems where annual seasonality is
not the rule, but daily changes in temperature are always present (Guhl 1989,
Luteyn 1999, Jacobsen 2008), a multivoltinism strategy would be more
beneficial for the maintenance of population throughout the time. In fact, the
research made until now have reported multivoltinsm in several taxa of aquatic
insects from the low lands of the tropics (Flowers 1987, Jackson and Sweeney
1995).
As Andean tropical streams are very dynamic hydrologically and water
discharge fluctuations have an important role structuring aquatic communities
(Flecker and Feifarek 1994, Jacobsen 2005, 2008), we should expect that
78
hydrological changes might have a key role in population dynamics.. Therefore,
recruitment by oviposition might be more pivotal in these tropical systems in
which there is substantial short term discharge fluctuations, most of which might
be unpredictable. But the knowledge on how the hydraulic changes affects flying
seasons, mating and ovipositing behaviors is inexistent.
The objective of our study is to understand the importance of the reproductive
strategies of aquatic insects in a high altitude tropical stream. Since this
ecosystem
presents
a strong seasonal component
and consequently,
conspicuous high and low hydrologic periods, we want specifically to analyze if
there are differences between these periods in: 1) richness and abundance of
adult forms of aquatic insects flying and ovipositing, 2) number of eggs
(oviposition pattern) and egg mass identity and morphology and 3) substrate
preferences by ovipositing females. Finally, we want to determine, for both
seasons, if there a relationship between the identity of larvae present in the
stream (using data from chapters 1 and 2) with that of adults flying at the same
time.
Study area
We performed the study on a pristine reach of 100 meters length of the Piburja
stream (3300 m asl), surrounded by an evergreen forest with Alnus acuminata
trees and a high diversity of Melastomataceae, Asteraceae and Rosaceae
shrubs. This headwater stream of the Oyacachi River Basin is, located at the
Cayambe-Coca Ecological Reserve (RECAY) in Ecuador (0º13’ S, 78º 03’ W).
Pluviometric records (10 years) from nearby localities show lower rainfall from
December to February (dry season) compared, with May until September. Total
precipitation in Oyacachi town is ~ 1600 mm/year, and mean air temperature is
~10 ºC. Precipitation is 2 or 3 times greater than the potential evapotranspiration
at this area, which makes very humid conditions during the whole year (Skov
1999). The stream width ranges from 1.60 m to 4.0 m. Seasonal differences in
flow, mean water velocity, were found during the studying periods, with higher
discharges at wet season. More detailed physico-chemical characteristics of the
79
reach can be found at Chapter 1. We performed all surveys (eggs, larvae and
adult sampling) during May of 2006 (the wet season) and February of 2007 (the
dry season).
Materials and methods
Adult sampling
To estimate the relative abundance of flying and/or ovipositing females we
placed two kinds of sticky traps. Both traps consisted in transparent acetate
sheets (210x 297 mm) with tree pest adhesive at both sides (Encalada and
Peckarsky 2007). The first type was the flying trap, and 18 of these sticky traps
were suspended in three nylon transects at two different altitudes above the
stream surface. At each transect three acetate sheets were suspended, across
the river from nylon lines located five to eight cm above the water surface, and
three at 80 to 120 . The other type of trap was a platform trap, which were
placed horizontally three cm above the water surface. These traps were made of
dark plywood (to simulate the river bed color) with the acetate sheets above, and
anchored using iron rebar. Twenty four of this platform traps were placed at
random locations on the 100 meters transect. All traps were deployed on 23, 24
and 25 of July of 2006 (wet season) and from 19 to 22 of February of 2007 (dry
season). Traps were changed two times daily at 0700 and 1800 hours. In total,
we had data for two days and two nights at each season.
At the laboratory each acetate sheet was soaked on kerosene dissolvent to
enable the separation of trapped individuals. All individuals were placed on
ethanol 75º and identified to lowest possible taxonomic level. The lack of
appropriate identification keys for adults of aquatic insects of this geographic
area plus the damage suffered by the individuals (due to the trapping method)
were determinants in the identification process. Thus, we maintained the
identifications to family level for most taxa, although in some cases we were able
to identify until genera level. The taxonomic keys used were: Domínguez et al.
1992, Merritt and Cummins 1996, Fernández and Domínguez 2001, Dominguez
80
et al. 2006. We identified the sexes when possible and for all gravid females we
recorded the percentage of abdomen filled with eggs. For all individuals of each
taxa that sex was not possible to determine we applied the Fisher's principle
(Fisher 1930) assuming a sex ratio of 1:1. Several gravid females from each
family were dissected and the eggs were counted and described.
Egg masses sampling
To understand oviposition patterns in this high altitude stream, we sampled for
egg masses in three different occasions (wet 2006, dry 2007 and wet 2007). At
each occasion, we sampled 85 rocks at random places on the same 100 m
transect were the sticky traps were placed. Each rock was categorized as either
submerged or emergent and the longest length and width axis were recorded.
The surrounding water velocity of each rock was also recorded using a Global
water flow probe model FP201. All egg masses found were counted and
reference samples were collected and transported to the laboratory for
incubation in containers with river water for further identification of hatching
larvae. We also took reference egg masses in ethanol at 75º for description and
egg counting.
Larvae samples
During the same sampling periods, we collected six to ten benthic insect
samples in five occasions, at the same stream reach using a Surber net of 14 x
14 cm2, (data from Chapter 1). Also, four to six drift samples (data from Chapter
2) were collected at the same seasons at four occasions, using a net of 250 µm
mesh, and 15 x 35 cm frame. The nets were placed in the river for 20 minutes to
1 hour, depending on flow conditions at each of the eight sampling events. Each
replicate was store separately and preserved with formalin 4%. Samples were
taken to the laboratory for further sorting and identification.
Insects were
preserved in ethanol and identified to the lowest possible taxonomic level.
81
Data analysis
To analyze seasonal changes in taxa richness and relative abundance of flying
adults we used a three way ANOVA analysis between the proportion of the taxa
arcsine transformed and considering the factors: season, behavior (flying [sticky]
or ovipositing [plataforms] traps) and diel periodicity (day or night). The effects of
the interaction of the factors were not included in the model due to the low
significance of the factors and the low number of degrees of freedom of the
model. To analyze the seasonal changes in males and females ratios (response
variable) we performed one way ANOVA, with sex as factor. All proportions were
arcsine transformed.
To analyze if relative abundance of gravid females (arcsine transformed) was
different between seasons (wet or dry), diel periods (day or night) and behaviors
(flying or ovipositing traps), we performed general linear model procedure for a
three-way ANOVA. Also, in this analysis we did not include the effects of the
interaction of the factors because of the low significance of the factors and the
low number of degrees of freedom of the model. Due to the small numbers of
individuals collected of some taxa we made the analysis only with those taxa
that had a total abundance of more than 0.5 % of the individuals collected at
least in one season.
To assess seasonal differences in the number and density of egg masses, we
conducted a Kruskal-Wallis analysis, due to the small number of egg masses
found. Since, there were no significant differences between seasons of each
taxa, a logistic multinomial regression was conducted with pooled data to
understand which factors are the most important to explain the presence of an
egg mass of each taxa at a rock. The logistic multinomial regression was
performed
with
the
factors:
rock
area,
surrounding
velocity,
and
emergence/submerged.
To determine which physical characteristics of the rock may influence the
abundance of egg masses (number of egg masses/rock) we performed a
82
negative binomial regression with pooled data. The factors in the model were:
rock emergence (categorical), surrounding velocity and area (continuous factor).
Following Reich & Downes ( 2003a), we use the number of egg masses per rock
for the analysis instead of the density of egg masses per rock area, because we
did not have evidence of a consistent relationship
between egg mass
abundance and rock size across a wide range of rock dimensions. The negative
binomial regression allowed us to analyze the count of egg masses with
categorical and continuous factor. A best subset selection of the factors was
done to produce the best possible model.
As explained above, we did not
separate the data from each season due to the small number of egg masses
collected for each taxon. We made the negative binomial regression only for
most abundant taxa.
To analyze the relationship between the larvae present in the stream with that
of adult stages we performed a Spearman correlation analysis with the
proportions of the taxa found at each type of sample (aerial, drift, benthos). Also,
we use a Non Metric Multidimentional Scaling (NMDS) with the Bray Curtis
Similarity, obtained from the proportions of taxa at each type of sample (arcsine
transformated) to address the same issue. For all statistical analysis we used
SPSS 15, Primer 6 and Statgraphics Centurion statistical software.
Results
Aerial stages of aquatic insects
We found a total 822 adults of terrestrial invertebrates (from 13 orders) and 2383
adults of aquatic insects in the sticky traps, of five orders: Ephemeroptera,
Trichoptera, Coleoptera, Plecoptera and Diptera corresponding to 28 families
(Appendix 1). Dipterans were the 88.8% of the total individuals collected.
Eighteen taxa (at family level) were found at both seasons while five were found
only at the wet season and three were exclusive of the dry season. Richness
and the relative abundance of most taxa was not significantly different between
seasons, behaviors or diel period. Only adults of Ephydridae (Diptera) had
83
significant higher abundance at the wet season and those of Ochrotrichia
(Trichoptera) at the dry season (Table 1).
The
proportion
of
males
and
females
was
significantly different
for
Ceratopogonidae and Chironomidae presenting greater proportions of females
during both seasons (Table 2), while Baetidae and Ochrotrichia had a significant
greater proportion of females only at the dry season. For the other taxa present
at both seasons not significant differences were founded between sex ratios.
During both sampling periods we found females with eggs for a total of 15
different taxa (13 in the dry season and 14 in the wet season). Females of
Chironomidae family were the taxa most commonly found with eggs. A
description of the eggs found gravid females of most abundant taxa is detailed in
Table 3. We did not find significant differences in the proportion of females with
eggs between seasons, diel periods or different behaviors (sticky vs. platform
traps) of the different taxa. The only exception was Nectopsyche, in which we
found significant greater proportions of females in traps placed at night time
(Table 4) suggesting a nocturnal behavior.
Oviposition patterns
Only a small number of rocks had egg masses, 10% at dry season while at wet
seasons of 2006 and 2007 the proportion was 18 and 13 %, respectively. We
found a total of 147 egg masses of five types (Table 5) in rocks. Most of them
were from Baetidae and Chironomidae with 91 and 41 egg masses respectively
together with few egg masses of Ceratopogonidae, Hydrobiosidae and
Nectopsyche were also found (Table 5). With the pooled data, we did not found
significant differences in both, number of egg masses or masses per rock area,
between seasons (Kruskal-Wallis test: H ( 2, N= 257) =2.620 p =0.270).
Due to the small numbers of egg masses recorded we only made the statistical
analysis, to test differences in seasonality, for Baetidae and Chironomidae. We
did not found significant seasonal differences in the presence and abundance of
egg masses (as direct counting) both for Chironomidae (Kruskal-Wallis ANOVA:
84
H (2, N= 257) =1.438011 p =0.4872; H (2, N= 257) =1.337896 p =0.5122) and
Baetidae (Kruskal-Wallis ANOVA: H (2, N= 257) =5.311986 p =0.0702; H ( 2, N=
257) =5.340174 p =0.0692).
Table 1. Summary of the three way ANOVA (d.f=7) of adults of aquatic insects
between seasons, behaviors and diel period at Piburja stream, Ecuador. pvalues significant after Bonferonni progressive correction are in bold.
Taxa
richness
Elmidae
Staphilinidae
Blepharoceridae
Ceratopogonidae
Chironomidae
Dolichopodidae
Empididae
Ephydridae
Muscidae
Psychodidae
Simuliidae
Tabanidae
Tipulidae
Baetidae
Atopsyche
Ochotrichia
Nectopsyche
F/p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
season
1.200
0.335
0.612
0.478
2.168
0.215
2.639
0.180
2.014
0.229
2.251
0.208
6.068
0.069
1.242
0.327
18.319
0.013
1.640
0.270
4.521
0.101
0.161
0.709
1.934
0.237
1.453
0.295
0.665
0.461
0.288
0.620
19.776
0.011
8.544
0.043
Behavior
0.000
1.000
0.219
0.664
5.062
0.088
0.081
0.790
7.092
0.056
1.667
0.266
17.568
0.014
3.109
0.153
4.061
0.114
1.104
0.353
0.836
0.412
0.004
0.951
0.679
0.456
3.686
0.127
0.113
0.753
0.944
0.386
0.309
0.608
0.688
0.453
Diel
period
0.533
0.506
0.022
0.888
0.262
0.636
0.324
0.600
0.773
0.429
0.407
0.558
1.339
0.312
0.000
0.990
1.952
0.235
2.494
0.189
1.747
0.257
4.451
0.103
0.227
0.659
7.756
0.050
0.163
0.707
2.398
0.196
0.882
0.401
14.211
0.020
85
Table 2. Summary of the one way ANOVA of adult sex proportions (arcsine
transformed) in the two sampled seasons at Piburja stream, Oyacachi, Ecuador.
Only taxa that presented significant differences are presented, in all significant
cases females were more abundant.
Taxa
d.f
DRY
SS
MS
F
p
WET
SS
MS
F
p
Ceratopogonidae
1
0.2136
0.2136
35.2373
0.001
0.1757
0.1757
7.2957
0.035
Chironomidae
1
2.1684
2.1684
333.729
0.000
1.9749
1.9749
10409.541
0.000
Baetidae
1
2.2757
2.2757
23.3990
0.003
1.4479
1.4479
4.1220
0.089
Ochotrichia
1
0.7132
0.7132
47.3120
0.000
1.4597
1.4597
3.8007
0.099
Table 3. Number and characteristics of eggs found inside the abdomen of gravid
females. Females were collected attached to sticky traps located above stream
sections at Piburja stream, Oyacachi, Ecuador.
form
max.
diameter
(mm)
color
taxa
N
number
of
eggs/female
Chironomidae
406
31-1000
elliptical
0.1
white-yellowish
Ceratopogonidae
39
60-300
spherical
0.1
white
Simuliidae
26
100-1000
elliptical
0.12
white-yellowish
Tipulidae
23
21-105
spherical
0.1
white-yellowish
Hydroptilidae
10
40-100
ovoid
0.12
white
Baetidae
7
15 - 150
spherical
0.08
light yellow
Dolichopodidae
5
300-1000
elliptical
0.05
yellow
Tabanidae
5
29-145
elliptical
0.24
brown-yellowish
Nectopsyche
(Leptoceridae)
4
19-40
spherical
0.2
white
Hydrobiosidae
4
200-300
spherical
0.05
white
None of the rock attributes used as predictors of the presence of egg masses
showed a significant relationship for Chironomidae and Baetidae. The logistic
regression model adequately fits the data of the predictors and the
presence/absence of egg masses both for Chironomidae (X2 =8.131, df=8,
p=0.421) and Baetidae (X2 = 6.574, df=8, p=0.583).
86
Table 4. Summary of the p-values of the 3 way ANOVA (d.f=7) performed using
the proportion (arcsine transformed) of gravid females found at each season,
behaviors, or diel period, at Piburja stream, Ecuador.
Psychodidae
Simuliidae
Tabanidae
Tipulidae
Baetidae
Atopsyche
Cailloma
Ochotrichia
Nectopsyche
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
F
p
Intercept
3.571
0.132
6.291
0.066
2.944
0.161
4.613
0.098
6.391
0.065
2.841
0.167
1.000
0.374
5.443
0.080
18.235
0.013
season
3.571
0.132
1.373
0.306
0.366
0.578
0.003
0.957
0.014
0.910
0.476
0.528
1.000
0.374
1.575
0.278
3.615
0.130
14
behavior
3.571
0.132
6.291
0.066
0.558
0.497
0.418
0.553
0.014
0.910
0.037
0.857
1.000
0.374
0.114
0.752
0.128
0.738
diel period
0.143
0.725
2.506
0.189
0.133
0.734
1.034
0.367
0.362
0.580
0.646
0.467
1.000
0.374
0.014
0.910
18.235
0.013
emergent
submerged
12
Egg masses
10
8
6
4
2
0
Baetidae
Baetidae
Chironomidae Chironomidae
Figure 1. Mean number for egg masses present in emergent or submerged
rocks at Piburja stream (Ecuador) for the two most abundant families. Error bars
represent one standard error of the mean.
Most part of egg masses were located at emergent rocks (Figure 1), and in
contrast to the absence of relation with the rock attributes and the presence of
egg masses, the number of egg masses per rock showed significant relationship
87
with rock attributes for the family Baetidae. The best predictors for Baetidae egg
mass abundance in the Negative bionomial regression were rock emergence
and rock area (Table 6, Figure 2). Despite this significant relation, the explained
percentages of the deviance by the model was only 32,85 % for Baetidae (and a
scarce 2,36 % for Chironomidae). The equations of the models for each taxa
were:
Baetidae egg mass abundance = exp(-4,07531 + 48,2988*area + 0,83398* rock
emergence)
Chironomidae egg mass abundance = exp(-3,21588 + 18,4374*area + 1,0932*
rock emergence)
2
Were area is m , velocity is m*s-1 and rock emergence= emergent/submerged
Plot of Fitted Model Chironomidae
with 95.0% confidence limits
12
Egg Masses
Emergent rock
9
6
3
0
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
Area
Plot of Fitted Model Baetidae
with 95.0% confidence limits
50
Emergent Rock
Egg masses
40
30
20
10
0
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
Area
Figure 2. Negative Binomial regression model predicting the egg mass
2
abundance for Baetidae and Chironomidae. Area in squared meters (m ).
88
Table 5. Description of egg masses found in rocks at Piburja stream (Ecuadorian Andes) during the dry season (2007) and wet
seasons (2006 and 2007).
Egg size
(max. egg
diameter;
mm)
Number
of eggs
per mass
Most frequent
substrate
# egg
masses
Season
Description
Baetidae
91
dry &wet
yellowish egg masses
with yellow white
circular eggs
0.08
31-62
Nectopsyche
4
wet
Transparent egg mass
encapsulated with clear
yellow eggs forming
lines. Ovoid eggs
0.10
120-360
submerged
median rocks
Hydrobiosidae
5
dry &wet
transparent spherical
egg mass with circular
white eggs
0.05
236-380
Large emergent
rocks
small larvae (31) also
founded at masses with
a size of larvae with
sizes around 0.2 mm
Diamesinae
(Chironomidae)
13
dry &wet
Transparent egg mass
encapsulated with clear
yellow eggs forming
lines. Ovoid eggs
0.18-0.20
172-1062
Large emergent
rocks
small larvae (840) were
found in one mass. Each
larvae of 0.53 mm long
Chironomidae
28
dry &wet
transparent
encapsulated with white
eggs forming a worm
like mass
0.10-0.20
310-999
Large emergent
rocks
small larvae (87) were
found in one mass. Each
larvae of 0.55 mm long
Ceratopogonidae
6
wet
brownish egg mass.
Circular eggs.
0.10
120-280
Median emergent
rocks
Taxa
Observations
Large emergent
rocks
89
Table 6. Results of the negative binomial regression (d.f.=2), assessing whether
rock attributes were significant predictors of egg mass abundance for
Chironomidae and Baetidae. When the variable was not included in the model is
marked by (-).Bold p values are significant after Bonferroni correction.
Prediction of abundance of egg masses
Rock attributes
Chironomidae
Baetidae
X2
p
X2
p
rock emergence
-2,497
1.000
7,963
0.005
Surrounding water
velocity
-
-
-
-
0,674
0.412
149,283
0.000
rock area
Comparison of the faunal composition from the aquatic and aerial stages
We have compared the percentage of individuals in the aerial samples with the
percentage of importance of the same taxa in drift and benthos Surber samples
in both seasons sampled (Table 7). The Chironomidae, are the most important
family for all types of samples (data from Rios, chapter 1 and 2), together with
Ceratopogonidae which are also important in both aerial and benthic samples.
The most important difference between the aerial and aquatic samples is that
Baetidae (wet season) and Nectopsyche (dry season) are underrepresented in
the aerial samples, because the dominance of Diptera in the aerial traps. The
Spearman correlations between the proportions of taxa in aquatic samples (drift
of benthos Surber samples) with the proportion of adults were not significant.
Only at the dry season we found a significant correlation, but with a low R value,
between the proportion of taxa samples in drift with adults samples (R=0.444,
p=0.0263). This absence of relation was clearly shown by the NMDS analysis
(Fig. 3.).
90
Table 7. Percentage of abundance (adults) and density (drift and benthos) of
aquatic and aerial stages of benthic fauna at Piburja stream (Ecuador) during
the wet (2006) and the dry season (2007). The most important value for each
category is in bold and the second and third values, in order of importance are
in italic.
WET
DRY
taxa
adults
drift
benthos
adults
drift
benthos
Chironomidae
48.66
32.54
Ceratopogonidae
12.61
1.13
52.76
32.07
31.03
34.78
6.41
16.43
2.61
Tipulidae
7.78
0.64
8.41
0.9
4.03
2.75
0.37
0.25
Dolichopodidae
6.08
0.03
-
19.67
3.8
Staphilinidae
4.38
-
-
8.21
-
-
Simuliidae
4.29
10.22
2.44
4.74
12.75
3.35
Muscidae
2.29
1.07
0.15
6.44
0.04
0.11
Chelifera
1.61
1.63
1.72
1.11
2.27
1.81
Hydrobiosidae
1.03
0.88
0.54
1.52
0.73
1.27
Tabanidae
0.98
0.07
0.05
0.39
-
-
Nectopsyche
0.81
4.43
4.16
0.16
12.95
11.31
Baetidae
0.63
20.04
13.01
1.03
3.54
3.02
Blephariceridae
0.54
0.85
-
-
0.72
0.09
Blepharoceridae
0.54
0.85
-
-
0.72
0.09
Ochrotrichia
0.36
5.15
1.08
2.06
4.47
3.6
Maruina
0.36
1.12
0.45
5.61
3.14
0.5
Total Elmidae
0.36
0.61
2.08
0.16
0.2
5.28
Anomalopsychidae
0.27
0.07
-
-
-
1.02
Ephydridae
0.16
5.72
0.04
1.88
0.4
0.04
Leptohyphes
0.09
0.58
1.26
0.16
0.1
4.17
Lampyridae
0.09
0.2
0.05
0.08
0.24
-
Haliplidae
0.09
-
-
-
-
0.02
Culicidae
0.09
-
-
-
-
Syrphidae
0.09
-
-
-
-
Stratiomyidae
-
1.6
0.54
0.32
2.89
Cochliopsyche
-
0.29
0.63
0.08
0.32
0.35
1.74
Gripopterygidae
-
0.42
0.99
0.16
0.55
1.73
Leptophlebiidae
-
-
-
0.08
-
-
91
Figure 3. Non Metric Multidimentional Scaling based on the Bray Curtis
Similarity of the proportion of taxa present at drift, benthos and adult samples of
Piburja Stream, Ecuador.
Discussion
Studies on oviposition and the relationship between the benthic fauna and the
adults present in a stream are scarce and mainly from temperate streams (Ward
and Cummins 1978, Allan and Flecker 1989, Peckarsky and McIntosh 1998,
Peckarsky et al. 2000, Wissinger et al. 2003, Encalada and Peckarsky 2006,
Petersen et al. 2006, Encalada and Peckarsky 2007, Encalada and Peckarsky in
prep.). The few studies on tropical streams show that continuous emergence
and long flight period seems to be the rule with dominance of Diptera and
variable sex ratio (Wolf et al. 1988, Flint 1991). Our results showed that in the
tropical mountain Piburja stream the richness and relative abundance of adults
do not present differences between the sampled seasons, except for Ephydridae
and Ochrotrichia (but only few individuals of these two taxa were collected). Also
that dipterans were the dominant order always, that, for most abundant taxa,
gravid females were present and in similar proportions at both seasons, while
92
egg masses, were scarce in contrast to the available data from other
geographical areas. The diversity of the egg masses were also low and again
did not showed seasonal changes. Only a relationship between the rock area,
the emergent rocks and the number of Baetidae egg masses was observed. In
summary, our data agree with other studies on tropical areas, and seasonality is
not present in the structure and composition of adult aquatic insects.
However, some previous results disagree with this general pattern, for example
seasonality has been reported for some Ephemeroptera taxa in tropical streams
(Sweeney et al. 1995), or in another study in New Zealand, at least the 50 % of
the species Trichoptera adults were found only six of the twelve months of the
year (Smith et al. 2002). In our case we found strong seasonal differences on
density of Baetodes and Nectopsyche larvae (chapter 1) and therefore a similar
seasonality to that of the New Zealand streams could be present in Piburja
stream at least for these taxa. However, with the adult sampling focused only in
a week at each season, we are not able to define if the taxa is present or not
during the whole year. Seasonal differences in larval densities of these two taxa
could be related to post recruitment control of local population by processes
operating in the larval stage (Peckarsky et al. 2000) more than factors related to
adult presence.
Chironomidae are the dominant taxa in our adult samples, as in other studies
on Andean areas (Wolf et al. 1988, Flint 1991), while the number of
Ephemeroptera, Plecoptera and Trichoptera (EPT) was low. This was coincident
with a previous study that found that the total EPT individuals were less than a
hundred after one year of monthly continuous sampling, using sticky traps
(Winterbourn and Crowe 2001). Despite the conspicuousness of Chironomidae,
behavioral patterns of mating and oviposition are generally not studied while
several studies exist for Ephemeroptera (Sweeney et al. 1995, Peckarsky et al.
2000, Encalada and Peckarsky 2006, 2007), Trichoptera (Komzak 2001,
Lancaster et al. 2003, Reich and Downes 2003b) and Plecoptera (Petersen et
al. 2006). Data on mating and ovipositing behavior of Chironomidae is scarce,
93
and only descriptions of egg masses are known (Branch 1928, Hinton 1981,
Nolte 1993). Due to the importance of Chironomidae more studies are needed
about reproductive strategies of these aquatic insects.
Sex ratio reported in previous studies of tropical streams is highly variable. For
example Winterbourn et al. (2007) found a higher proportion males of hidroptilids
while Wolf et al. ( 1988) report a higher proportion of females in the same family.
In our case, Ceratopogonidae and Chironomidae presented a higher proportion
of females and this pattern has been also reported for emergence in Colombian
Andean rivers (Wolf et al. 1988). The same pattern was found for Baetidae and
Ochrotrichia, but due to the few individuals collected this result must be
confirmed in future. The proportion of females with eggs did not differ between
seasons at least for the most abundant taxa present in the stream. This is
coincident with the hypothesis of multivoltinism as the main strategy to maintain
macroinvertebrate populations in this stream and is coincident with previous
evidence for Neotropical but lowland rivers (Jackson and Sweeney 1995). The
small numbers of non-dipterans aquatic insects trapped restrict our conclusions,
but as they were present in both seasons we can suppose that they are able to
reproduce any time of the year. This does not mean that there is not a preferred
season for oviposition of some taxa, but longer and more frequent trapping times
is needed to found seasonal patterns. Again our benthic data (Chapter 1 and 2)
suggested that for some taxa (Nectopsyche, Baetodes) seasonal changes in
density may also indicate changes in adult emergence. Alternatively differences
on egg survival may explain differences in benthic density while the number of
adults and females with eggs are similar between seasons.
The types of egg masses found coincide with the literature (Hickin 1968, Hinton
1981, Reich 2004) for the same families, but few types and egg masses were
found compared to from Australian rivers (Reich and Downes 2003a, Reich
2004, Encalada and Peckarsky 2007) and in temperate mountain streams
(Encalada and Peckarsky 2007). Unlike the above mentioned studies, there is
not a high season of reproduction and therefore this indicates again the
94
aseasonality of reproductive traits in this stream. But our sampling method was
focused in egg masses in rocks, and a greater diversity of oviposition sites might
be present at the stream which could change our results. For example several
aquatic insects just release the eggs when flying above the water (Hinton 1981,
Encalada and Peckarsky 2007). Another interesting result is that the great
majority of egg masses were from Baetidae, in contrast to the adult patters
where Chironomidae dominates. One explanation for this fact may be that the
presence of egg masses is not always related to the presence or emergence of
adults of the same taxa in the same site. For example Peckarsky et al. (2000)
found that Baetis eggs appeared before any adults had emerged at that site.
Evidences of a non random distribution on egg masses and specialized
behaviors for aquatic taxa have been previously documented (Gillett 1971,
Reich and Downes 2003a, Reich and Downes 2003b, Encalada and Peckarsky
2006). The fact that almost all egg masses found at Piburja stream were on
emergent large rocks suggest that this is the preferred substrate for oviposition,
although we were able to analyze only the relation of the egg masses of
Baetidae and Chironomidae with rock characteristics. The same tendency was
previously described for Baetis (Peckarsky et al. 2000) and in Australian rivers
for Trichoptera and Diptera (Reich and Downes 2003a, Reich 2004). This fact
might be attributed to an increase hydrological stability of larger substrates that
may reduce egg mortality (Peckarsky et al. 2000). Since strong unpredictable
changes in flow can occur at the Piburja stream, choosing more stable substrate
can have important consequences preventing egg mortality. This is especially
important for Baetidae, that presented few gravid females compared to those of
Chironomidae, and thus selective oviposition that increases egg survival
appears as very important for population persistence of this taxa in this stream.
Another fact that may explain differences in adult and larvae dominance may be
hatching times. The short period of hatching of dipterans (Jackson and Sweeney
1995) could be an advantage for the survival of eggs and prevent egg stranding
(Jackson and Sweeney 1995, Reich 2004). Taxa with longer developmental
95
times as Ephemeroptera, Plecoptera and Trichoptera (Jackson and Sweeney
1995, Reich 2004), would need more specialization on ovipositing site selection
to increase egg survival compared to Diptera. Also hatching times are
temperature dependent for most part of these taxa (Brittain 1982, Brittain 1989)
and is expected that developmental times last more at high altitude forested
tropical streams, like Piburja, where the maximum registered temperature is 10
ºC. On the other hand the importance of diapause for certain insects described
on temperate streams is probably absent for mountain Andean streams, but the
importance of this mechanism has not been studied.
The mechanisms involved and the implications in population dynamics remain
unknown and different alternative explanations may be given for our results
(oviposition behavior, adult maturity, adult dispersion, etc). This study was a first
attempt to describe oviposition patterns in an Andean tropical mountain stream,
and we provide evidence on multivoltinism as a generalized strategy, which is
coincident with other studies in tropical areas. Also oviposition site selection for
one taxa (Baetidae) was described. Moreover a more detailed taxonomic study
that may determine differences at species level and much more time intensive
research is needed to understand the importance of reproductive strategies in
the population dynamics of highland altitude tropical streams.
96
APPENDIX 1. Total individuals of adults collected in the sticky traps during the
dry and the wet seasons. Individuals are separated by sex (females with or
without eggs).
DRY
wet
Gravid
Gravid
TAXA
female females male female females male
Elmidae
1
1
3
1
Haliplidae
1
Lampyridae
1
1
Staphilinidae
53
51
24
25
Blepharoceridae
2
3
1
Ceratopogonidae
91
22
95
61
17
63
Chironomidae
212
148
46
197
258
89
Culicidae
1
Dolichopodidae
124
3
122
33
2
33
Empididae
8
6
9
9
Ephydridae
1
1
13
8
Muscidae
14
15
36
1
35
Psychodidae
32
8
31
3
1
Simuliidae
17
24
19
27
2
19
Stratyiomydae
2
2
Syrphidae
1
Tabanidae
3
1
1
5
4
2
Tipulidae
16
8
27
25
15
47
Baetidae
7
4
2
2
3
2
Leptohyphidae
1
1
1
Leptophlebiidae
1
Gripopterygidae
1
1
1
1
1
Anomalopsychidae
Atopsyche
4
1
6
3
2
6
Cailloma
2
2
1
2
Helicopsychidae
1
Ochotrichia
9
9
8
2
1
1
Nectopsyche
2
4
2
3
Total general
599
231
436
455
312
350
97
CONCLUSIONES
Conclusiones
Conclusiones y Resumen General
Luego de abordar en profundidad los efectos de los cambios de caudal sobre la
estructura de la comunidad de macroinvertebrados, estudiándolos durante dos
épocas hidrológicamente distintas en el río Piburja y los mecanismos que
pueden explicarlos, en cuatro capítulos diferentes, volvemos a las preguntas
planteadas en un inicio para establecer las conclusiones de esta tesis.
¿Cuál es el efecto de la variación a corto plazo y estacional del caudal en
la comunidad de macroinvertebrados y qué importancia potencial tienen
los refugios y las características del microhabitat en la persistencia de las
mismas? ¿Cuál es el efecto de la reducción del caudal en la comunidad
de macroinvertebrados?
La variación estacional de caudal en el río Piburja fue importante. Las
comunidades de macroinvertebrados mostraron un patrón de variación que se
corresponde con la variabilidad estacional del caudal. Encontramos una mayor
riqueza, diversidad y densidad de la mayoría de las especies en la época seca,
cuando los caudales del río son más estables. La excepción fue Baetodes, un
taxa que fue dominante en la época de lluvias y su densidad tuvo una marcada
disminución en la época seca. Estas diferencias podrían estar relacionadas a
una mayor disponibilidad de zonas rápidas en el río durante esta época, que en
el caso de Baetodes son las que más utiliza. La diferenciación de comunidades
por microhábitats fue más marcada en la época seca mientras que en la época
de lluvias la disposición espacial de los distintos taxa en el río fue más bien
aleatoria. No encontramos evidencias del uso de zonas con menor stress
hidrológico o refugios, en la época de caudal inestable (lluvias). Sin embargo
ciertos taxa (p.e. Nectopsyche) tuvieron una tendencia constante a permanecer
en zonas con poca corriente (de refugio), lo cual podría ser un comportamiento
para evitar permanentemente zonas donde aumente la probabilidad de ser
arrastrados por la corriente.
101
Conclusiones
En general la respuesta de la comunidad del tramo de caudal regulado y el
tramo prístino a los cambios estacionales de caudal fueron similares. Pese a
esto, la reducción antropogénica del caudal produjo la disminución de ciertos
taxa (Ochrotrichia, Claudioperla), en general las más sensibles a la reducción
de oxígeno y por el contrario favoreció el aumento de la densidad de ciertos
taxa más resistentes (Oligochaeta, Tricladida).
¿Hay periodicidad diaria en la deriva? ¿Cuál es la variación estacional en
la composición de la deriva y que taxa está siendo más propenso a ser
arrastrado por la misma? ¿Cómo afectan los patrones de deriva en la
recolonización del sustrato? ¿Se diferencia la recolonización entre
estaciones y microhábitats distintos?
En general la comunidad no presentó una tendencia a derivar más en horas
cercanas al atardecer como se ha visto en ríos de montañas templados. Solo
dos taxa de efemerópteros (Baetodes y Leptohyphes) presentaron este
comportamiento, que puede estar relacionado a la presencia de la especie
introducida de trucha Oncorhynchus mykiss.
Por otra parte, la deriva tuvo una fuerte variación estacional, siguiendo un
patrón contrario al del bentos, ya que en general las densidades de la deriva de
la mayoría de taxa aumentaron en la época de lluvias y disminuyeron en la
época seca, provocando cambios estacionales en la estructura de la
comunidad. Esto sugiere un aumento de la deriva accidental por aumento de
caudal en la época de lluvias, que también fue corroborada con la mayor
propensión a derivar de la mayoría de taxa en ésta época.
La recolonización de piedras estuvo claramente influenciada por los cambios en
la composición estacional de la deriva. Además, aquellas piedras localizadas en
zonas rápidas tuvieron una composición taxonómica distinta de las encontradas
en zonas lentas, con una tendencia general a tener mayor riqueza, diversidad y
densidad en las primeras. El aumento de la densidad de Nectopsyche durante
la época de lluvias en piedras localizadas en zonas lentas en comparación con
102
Conclusiones
las muestras Surber del bentos, sugiere que este taxa busca activamente éstas
zonas donde los efectos del flujo del agua son menores.
En resumen la deriva tiene variaciones estacionales importantes y la
recolonización está influenciada por estos cambios y por las condiciones del
microhábitat.
¿Cuál es la importancia de las entradas de materia alóctona en un río
altoandino? ¿Hay algún efecto estacional en el transporte, retención de
ésta
materia
orgánica
y qué
implica
para
las
comunidades
de
macroinvertebrados? ¿Qué taxa se están alimentando de este recurso?
El río Piburja demostró ser un sistema heterotrófico. Las entradas de materia
orgánica al río fueron constantes, pero no tuvieron una clara relación con el
régimen de lluvias. Sin embargo las mayores entradas se produjeron en meses
en que las lluvias fueron menos intensas y viceversa. Pese a que el transporte
total no tuvo diferencias claras entre épocas, la retención aumentó
considerablemente en la época seca.
La materia orgánica acumulada tuvo relación con directa con la riqueza,
diversidad y densidad de macroinvertebrados, especialmente en la época de
lluvias cuando este recurso fue más escaso. Los recursos consumidos en
mayor cantidad por la comunidad (de acuerdo con los contenidos intestinales),
independientemente del grupo funcional alimenticio adjudicado por la literatura,
fueron la materia orgánica particulada gruesa y fina en las dos épocas
estudiadas. Esto sugiere que la comunidad en este río ha adaptado sus hábitos
alimenticios a los recursos disponibles, y que por lo tanto puede ser que los
mismos taxa en otras localidades consuman otros recursos. Esto reafirmaría el
hecho de que en general, las comunidades de insectos acuáticos neotropicales
tienen hábitos alimenticios flexibles.
103
Conclusiones
¿Cuál es la importancia del reclutamiento por ovoposición en ríos de
altitud? ¿Hay
efecto estacional? ¿Qué relación tienen los patrones
observados en las formas acuáticas (bentos y deriva) con los observados
en los estadios adultos?
Encontramos muchos adultos de insectos acuáticos durante las dos épocas, la
mayoría de ellos dípteros. La riqueza y abundancia relativa de adultos no
presentó cambios significativos entre épocas, a excepción de dos taxa poco
abundantes (Ephydridae and Ochrotrichia). En general encontramos muchos
individuos volando y pocos ovopositando (pocas hembras con huevos, pocas
masas de huevos) y esto fue similar en las dos épocas. Presumiblemente existe
moltivoltinismo y la reproducción se daría durante todo el año, pero
necesitaríamos datos por lo menos durante todo un año para realizar
conclusiones más contundentes sobre este patrón. Encontramos poca
diversidad y pocas masas de huevos (y muchas de ellas de Baetidae) en rocas
durante las dos épocas, y la mayoría de ellas estaba en rocas grandes
emergentes, probablemente debido a que son un sustrato más estable. Debido
a que muchos taxa solo expulsan sus huevos sobre el agua, nuestro método de
ubicación de masas de huevos excluye a otros taxa, cuyos huevos podrían
encontrarse en otros sustratos del río.
No hubo una clara relación entre la diversidad y abundancia relativa de adultos
y las formas acuáticas (bentos o deriva). La principal similitud constituye la
importancia que tiene tanto en el ambiente acuático como en el aire la familia
Chironomidae. Por otro lado la principal diferencia constituye que las larvas de
Baetodes, muy abundante en la época de lluvias, y las de Nectopsyche, muy
abundante en la época seca, estuvieron poco representadas en las muestras de
adultos. El tema de la ovoposición y su importancia en el reclutamiento es el
que quizá deja más interrogantes. Es mucho lo que desconocemos sobre la
historia de vida de los insectos acuáticos altoandinos mientras que su
investigación es primordial para el estudio de las dinámicas poblacionales.
104
RESUMEN GENERAL
Resumen General
Capítulo 1. Comunidades de macroinvertebrados en un río Tropical
Altoandino: La importancia del microhábitat, el caudal y la estacionalidad
En este capítulo nos centramos en la importancia del microhábitat, el caudal y la
estacionalidad en la estructuración de las comunidades de macroinvertebrados
en el río. Los ríos altoandinos sufren cambios de caudal rápidos e impredecibles
que sumados a la topografía variable propia de la zona (con fuertes pendientes)
pueden provocar fuertes perturbaciones en el sustrato. Se desconocen en esta
área los mecanismos usados por los distintos taxa presentes para recuperar las
poblaciones luego de las crecidas. A escala espacial, las reducciones
antropogénicas del caudal provocan cambios en la dinámica hidrológica del río
y por lo tanto de las comunidades de macroinvertebrados. Por esta razón en
este capítulo también incluimos datos de un tramo afectado por la desviación de
agua para la cría de truchas, en el cuál el caudal base se reduce en un 50 por
ciento. En este contexto las preguntas que se intentan responder en el primer
capítulo son: ¿Cuál es el efecto de la variación a corto plazo y estacional del
caudal en la comunidad de macroinvertebrados y qué importancia potencial
tienen los refugios y las características del microhabitat en la persistencia de
las mismas? ¿Cuál es el efecto de la reducción del caudal en la comunidad de
macroinvertebrados?
Para contestar estas preguntas realizamos muestreos Surber en cinco
ocasiones durante cada época. En cada ocasión tomamos de 6 a 10 muestras
Surber en dos tipos de microhábitat: zonas de fuerte corriente o rápidos y zonas
con menores fuerzas hidrológicas a las que denominamos refugios. En cada
Surber se midió la velocidad cercana al sustrato y la profundidad. Durante toda
la época de muestreo colocamos un medidor en continuo de la profundidad a
partir del cual pudimos obtener el caudal cada 10 minutos. Estos datos de
caudal fueron relacionados a la composición y abundancia de las comunidades
de macroinvertebrados presentes.
La variación estacional de caudal en el río Piburja fue importante. La época
seca tuvo un caudal menor y mucho más estable comparado con la época de
107
Resumen General
lluvias, donde el caudal además de ser significativamente mayor que en la
época seca, tuvo variaciones importantes en el tiempo (Fig 1. Capítulo 1). Las
comunidades de macroinvertebrados mostraron un patrón de variación que se
corresponde con la variabilidad estacional del caudal, en consecuencia las
comunidades estuvieron estructuradas de forma distinta en las dos épocas (Fig
2, Capítulo 1). Encontramos una mayor riqueza, diversidad y densidad de la
mayoría de las especies en la época seca, cuando los caudales del río son más
estables (Fig 3a, Capítulo 1). En general los estudios previos realizados en los
Andes muestran también este incremento en las métricas de la comunidad
durante las épocas secas o de menos lluvias (Flecker and Feifarek 1994,
Jacobsen and Encalada 1998, Jacobsen and Marín 2007, Perez and Segnini
2007). La excepción a esta tendencia general fueron dos taxa Baetodes y
Macrelmis, especialmente el primero que fue dominante en la época de lluvias
mientras que su densidad tuvo una marcada disminución en la época seca (Fig
3f, Capítulo 1). Estas diferencias podrían estar relacionadas a una mayor
disponibilidad de zonas rápidas en el río durante esta época, que en el caso de
estos dos taxa son las que más utilizan (Kohler and McPeek 1989, Peckarsky
1996, Zuñiga de Cardoso et al. 1997).
Además de la variación estacional en las comunidades, también encontramos
una diferenciación de éstas por microhábitats. Esta variación fue más marcada
en la época seca mientras que en la época de lluvias la disposición espacial de
los distintos taxa en el río fue más bien aleatoria (Fig. 2 y 3, Capítulo 1). Ciertos
taxa como Simuliidae (Fig. 3e, Capítulo 1) tuvieron una preferencia por zonas
de rápidos en las dos épocas y no encontramos para ellos evidencias del uso
de zonas con menor stress hidrológico o refugios, en la época de caudal
inestable (lluvias).
Por otra parte, otros taxa como Nectopsyche (Fig. 3d,
Capítulo 1) tuvieron una tendencia constante a permanecer en zonas con poca
corriente (de refugio), lo cual podría ser un comportamiento para evitar
permanentemente zonas donde aumente la probabilidad de ser arrastrados por
la corriente. Este comportamiento ha sido evidenciado en condiciones
controladas de caudal (en laboratorio) en otros tricópteros (Lancaster et al.
2006)
108
Resumen General
En general la respuesta de la comunidad del tramo de caudal regulado y el
tramo prístino a los cambios estacionales de caudal fueron similares. Pese a
esto, la reducción antropogénica del caudal produjo la disminución de la
abundancia de ciertos taxa (Ochrotrichia, Claudioperla), en general las más
sensibles a la reducción de oxígeno y por el contrario favoreció el aumento de la
densidad
de
los
taxa
más
tolerantes
(Oligochaeta,
Tricladida).
109
Resumen General
Capítulo 2. Los procesos de deriva y recolonización en un río Tropical
Altoandino.
En este capítulo tratamos de dos procesos fundamentales en el mantenimiento
de las comunidades de macroinvertebrados de los ríos la deriva y la
recolonización. De los estudios recientes sobre deriva en los trópicos sabemos
que a nivel local la recolonización depende de la migración de los parches
adyacentes, pero a nivel de secciones de río puede depender de la deriva de
zonas más lejanas (Boyero and DeLope 2002, Boyero and Bosch 2004, Melo
and Froehlich 2004). Sin embargo la mayoría de estos estudios se han
realizado solo en una estación del año, generalmente la de menor lluvia, por lo
tanto los cambios de composición de la deriva asociados al caudal y su
influencia en la recolonización no ha sido abordada. Tampoco se sabe que taxa
son más proclives a derivar. Además ríos tropicales que no poseen poblaciones
de peces locales, pero que sin embargo poseen especies de peces introducidas
(p.e. varias especies de truchas), han demostrado carecer de periodicidad en la
deriva (Flecker 1992, Jacobsen and Bojsen 2002, Jacobsen 2008).
Las interrogantes respecto a la deriva y a la recolonización, en el contexto de
los ríos de alta montaña tropicales con perturbaciones de caudal impredecibles
y a su vez temperatura media de 10 º C, son muchas. Entre otras preguntas que
hemos intentado contestar en el capítulo dos de esta tesis están: ¿Hay
periodicidad en la deriva? ¿Cuál es la variación estacional en la composición de
la deriva y que taxa está siendo más propenso a ser arrastrado por la misma?
¿Cómo afectan los patrones de deriva en la recolonización del sustrato? ¿Se
diferencia la recolonización entre estaciones y microhábitats distintos?
Para contestar las preguntas relacionadas a la deriva realizamos dos tipos de
muestreo. El primero fue un muestreo de 24 horas en caudal base, en luna
nueva (en la época seca), con el que queríamos describir el patrón diario en la
deriva del comportamiento. Con el segundo tipo de muestreo, realizado en las
dos épocas en cuatro ocasiones en cada época a horas similares, queríamos
estudiar las variaciones estacionales y relacionadas al caudal en la composición
110
Resumen General
de la deriva. En estos muestreos se usaron de cuatro a seis réplicas en cada
evento de muestreo. Para hacer los cálculos de propensión a la derivar de los
distintos taxa se relacionó su densidad promedio por época en la deriva con la
densidad promedio presentes en las muestras Surber tomadas durante los
mismos periodos (Capítulo 1).
La recolonización fue estudiada a dos escalas temporales: una corta, tomando
en cuenta los 7 días en que según los estudios previos los procesos de
recolonización se completan en el trópico; y una escala temporal a medio plazo
(hasta 25 días) donde se pueden apreciar las variaciones a partir de una
comunidad establecida y relacionarla con los cambios de caudal y situación
hidrológica del microhábitat. Para esto utilizamos piedras previamente lavadas y
marcadas y las colocamos en los dos microhábitats hidrológicos (zonas rápidas
y zonas lentas o de refugio). El proceso de recolonización fue analizado y
relacionado con la composición de la deriva y del bentos.
Al contrario que en los ríos de montaña de las zonas templadas (Allan 1984,
Allan 1987, Allan et al. 1988), la comunidad en general, no presentó una
tendencia a derivar más en horas cercanas al atardecer. Solo dos
efemerópteros (Baetodes y Leptohyphes) presentaron este comportamiento
(Fig. 1, Capítulo 2), que puede estar relacionado a la presencia de la especie
introducida de trucha Oncorhynchus mykiss. Este comportamiento ha sido
ampliamente estudiado en efemerópteros de zonas templadas, donde se ha
visto que el contacto con peces produce una mayor deriva en horas del
atardecer (Allan 1978, Allan et al. 1986, McIntosh and Townsend 1994,
Peckarsky and McIntosh 1998, McIntosh et al. 2002). También ha sido
observado en Baetis en los Andes de Venezuela (Flecker 1992).
La deriva tuvo una fuerte variación estacional, siguiendo un patrón contrario al
del bentos, ya que en general las densidades de la deriva de la mayoría de taxa
aumentaron en la época de lluvias y disminuyeron en la época seca,
provocando cambios estacionales en la estructura de la comunidad (Fig. 2,
Capítulo 2). Esto sugiere un aumento de la deriva catastrófica o accidental por
111
Resumen General
aumento de caudal en algunas ocasiones en la época de lluvias, que también
fue corroborada con la mayor propensión a derivar de la mayoría de taxa en
ésta época (Fig. 3, Capítulo 3). Este patrón de cambio, es contrario a los
encontrados en otros ríos tropicales (Brittain and Eikeland 1988).
La recolonización de las piedras en los experimentos realizados estuvo
claramente influenciada por los cambios en la composición estacional de la
deriva. Además aquellas piedras localizadas en zonas rápidas tuvieron una
composición taxonómica distinta de las encontradas en zonas lentas, con una
tendencia general a tener mayor riqueza, diversidad y densidad en las primeras
(Tabla 7, Capítulo 2). Este patrón de aumento de las métricas de la comunidad
en las piedras localizadas en las zonas rápidas ha sido previamente reportada
en ríos de Australia (Downes et al. 1995) y de Ecuador (Jacobsen, 1995). El
aumento de la densidad de Nectopsyche durante la época de lluvias en piedras
localizadas en zonas lentas en comparación con las muestras Surber del
bentos, sugiere que este taxa busca activamente éstas zonas donde los efectos
del flujo del agua son menores. En resumen se puede decir que en el río
Piburja, la deriva tiene variaciones estacionales importantes y la recolonización
está influenciada por los cambios hidrológicos que se producen y por las
condiciones del microhábitat.
112
Resumen General
Capítulo 3. Dinámica de la hojarasca y los macroinvertebrados asociados
en un río tropical Altoandino.
La entrada de materia orgánica alóctona en forma de hojarasca constituye la
principal fuente de energía y nutrientes en los ríos de cabecera (Siccama et al.
1970, Anderson and Sedell 1979, Wallace et al. 1997, Graça 2001). Parte de
esta energía es aprovechada por los consumidores y otra es exportada río
abajo por acción de la corriente (Likens et al. 1970, Fisher and Likens 1973) y
por ello existe un vínculo trófico entre las comunidades de los ríos de cabecera
y los tramos más bajos de los ríos que está mediado por los descomponedores
y detritívoros (Graça 2001). En los ecosistemas tropicales la información
cuantitativa respecto a entradas, almacenamiento y ciclos de estas entradas se
desconocen pese a que usualmente los ríos tropicales drenan áreas con
vegetaciones densas y las entradas alóctonas probablemente son la principal
fuente de energía (Graça et al. 2001a, Colón-Gaud et al. 2008). Debido a la
escasa estacionalidad de los ríos tropicales (en términos de temperatura), estas
entradas son potencialmente continuas durante todo el año, al contrario que en
ríos de zonas templadas que reciben la mayor parte de entradas en otoño. Sin
embargo, en el trópico y en los estudios realizados hasta el momento, los
máximos de entrada de materia orgánica están asociados a tormentas (Covich
1988), o al comienzo (Gonçalves et al. 2006) o final de la estación de lluvias
(Afonso et al. 2000).
La capacidad de los ríos de retener estas entradas está relacionada a la
morfología y heterogeneidad espacial del canal del río (Speaker et al. 1984,
Lamberti et al. 1989) y está regulada no solo por las entradas del bosque
adyacente sino también por la magnitud y variabilidad de la descarga (Pearson
et al. 1989). En situaciones de caudales altos, la capacidad retentiva disminuye
y aumenta el movimiento y redistribución de la hojarasca en el fondo del río
(Larrañaga et al. 2003). El patrón general en el trópico es que mayores
cantidades de hojarasca se acumulen durante las épocas secas (Covich 1988).
Los estudios de zonas templadas han demostrado que la relación de esta
materia alóctona acumulada en el río puede ser usada por la comunidad de
113
Resumen General
macroinvertebrados como refugio (Palmer et al. 1996), o como recurso
alimenticio por parte de los trituradores (Wallace and Webster 1996, Graça
2001, Graça et al. 2001b). Al contrario que en zonas templadas, en el trópico los
trituradores al parecer no son una parte importante de la comunidad (Ramirez
and Pringle 1998, Mathuriau and Chauvet 2002, Mathuriau et al. 2008), pero
existe alguna evidencia de que en los ríos tropicales de montaña pueden tener
más relevancia (Cheshire et al. 2005). Además el análisis de contenido del
tracto digestivo de insectos acuáticos en ríos tropicales han demostrado que la
mayoría de especies explota dos niveles tróficos, o se alimentan de más de un
tipo de recurso (Wantzen et al. 2005, Tomanova et al. 2006, Wantzen and
Wagner 2006). En este sentido, la importancia de la materia orgánica alóctona
(hojarasca) como recurso alimenticio en los ríos tropicales de montaña es
incierta, y probablemente ésta dependa de los recursos disponibles a nivel local,
ya que al parecer los macroinvertebrados tropicales parecen tener hábitos
alimenticios flexibles (Covich 1988).
Para abordar este tema en el contexto de los ríos altoandinos, las preguntas
que intentamos responder en el capítulo tres fueron: ¿Cuál es la importancia de
las entradas de materia alóctona en un río altoandino? ¿Hay algún efecto
estacional en el transporte, retención de ésta materia orgánica y qué implica
para las comunidades de macroinvertebrados? ¿Qué taxa se están alimentando
de este recurso?
La hojarasca proveniente del bosque de ribera fue medida durante todo un año,
por medio de canastas plásticas rectangulares (ocho) dispuestas sobre el canal
del río. El material acumulado en las canastas fue colectado mensualmente. La
retención, transporte y materia acumulada en el río fueron medidos durante las
dos épocas hidrológicas distintas.
Las entradas de materia orgánica al río fueron constantes, pero no tuvieron una
clara relación con el régimen de lluvias, aunque la mayor entrada se produjero
en meses en que las lluvias fueron menos intensas y viceversa (Fig. 1, Capítulo
3). Pese a que el transporte total no tuvo diferencias claras entre épocas, la
114
Resumen General
retención aumentó considerablemente en la época seca (Fig.2 y Fig. 3, Capítulo
3).
La materia orgánica acumulada tuvo relación con directa con la riqueza,
diversidad, densidad total y de grupos funcionales alimenticios, especialmente
en la época de lluvias cuando este recurso fue más escaso (Tabla 5, Capítulo
3), y por otro lado, cuando durante la época seca, cuando encontramos más
materia orgánica acumulada en el río, hubo un incremento en la densidad de
todos los grupos funcionales alimenticios (Tabla 6, Capítulo 3).
Los recursos consumidos en mayor cantidad por la comunidad (de acuerdo con
los contenidos intestinales), independientemente del grupo funcional alimenticio
adjudicado por la literatura, fueron la materia orgánica particulada gruesa y fina
en las dos épocas estudiadas (Fig.6, Capítulo 3). Las preferencias alimenticias
no se diferenciaron en las dos épocas. Estos resultados sugieren que la
comunidad en este río ha adaptado sus hábitos alimenticios a los recursos
disponibles, y que por lo tanto puede ser que los mismos taxa en otras
localidades consuman otros recursos. Esto reafirmaría el hecho de que en
general, las comunidades de insectos acuáticos neotropicales tienen hábitos
alimenticios flexibles.
115
Resumen General
Capítulo 4. Ovoposición de los Insectos acuáticos en un río tropical
altoandino.
Este capítulo se centra en la Ovoposición de los insectos acuáticos del río
Piburja. El papel del reclutamiento en las comunidades acuáticas es poco
conocida (Encalada et al. en prep.), y la poca información que existe sobre
ovoposición está restringida al hemisferio norte (Hinton 1981, Brittain 1989,
Merritt and Cummins 1996). Sin embargo, es de esperar que en zonas
templadas, sea solo importante durante los meses de primavera y verano,
cuando los insectos que han emergido para la reproducción vuelven al agua a
depositar sus huevos. La estrategia del univoltinismo favorece que los estadios
aéreos ocurran cuando las condiciones para la reproducción y ovoposición son
más favorables. Las zonas tropicales por el contrario carecen de esta marcada
estacionalidad en la temperatura y en este contexto, el multivoltinismo es una
estrategia que puede ser altamente beneficiosa para la persistencia de las
comunidades. Esta estrategia ha sido reportada para varios insectos tropicales
(Collier and Smith 1995, Jackson and Sweeney 1995) principalmente de
altitudes bajas. Por el momento los datos de historia de vida y ovoposición de
insectos andinos de alta montaña son inexistentes (Jacobsen 2008). Sin
embargo, conocer este aspecto es fundamental para entender la persistencia de
las comunidades en las condiciones ambientales más restrictivas de la alta
montaña, como es la temperatura media más baja (~10ºC en Ecuador) y donde
la saturación de oxígeno está afectada por la altitud (Jacobsen 2008). En este
contexto, las interrogantes en este tema fueron: ¿Cuál es la importancia del
reclutamiento por ovoposición en ríos de altitud? ¿Hay
efecto de la
estacionalidad hidrológica? ¿Qué relación tienen los patrones observados en
las formas acuáticas (bentos y deriva) con los observados en los estadios
adultos? En el cuarto capítulo de esta tesis intentamos responder a estas
preguntas.
Para contestar estas preguntas realizamos muestreos durante la época seca y
la época de lluvias. En los muestreos empleamos dos tipos de trampas con
pegamento: las trampas de vuelo, hojas de acetato con pegamento que estaban
116
Resumen General
suspendidas en transectos transversales en el río (en total 18 trampas) y
trampas de ovoposición que consistían en hojas de acetato
dispuestas en
plataformas de plywood negro (en total 24 trampas). Estas fueron colectadas
durante dos días y dos noches en cada época. También muestreamos huevos
en las rocas del río (85 en cada época) y de cada roca registramos la velocidad
circundante, el área y si estaba sumergida en el agua o emergía de la misma.
Encontramos muchos adultos de insectos acuáticos durante las dos épocas, la
mayoría de ellos dípteros. La riqueza y abundancia relativa de adultos no
presentó cambios significativos entre épocas (Tabla 1, Capítulo 4), a excepción
de dos taxa poco abundantes (Ephydridae and Ochrotrichia). En general
encontramos muchos individuos volando y pocos ovopositando (pocas hembras
con huevos, pocas masas de huevos) y esto fue similar en las dos épocas.
Presumiblemente existe multivoltinismo y la reproducción se daría durante todo
el año, pero necesitaríamos datos por lo menos durante todo un año para
obtener conclusiones más contundentes sobre este patrón.
Encontramos poca diversidad y pocas masas de huevos en rocas durante las
dos épocas, y la mayoría de ellas estaba en rocas grandes emergentes (Tabla
5, Capítulo 4), probablemente debido a que son un sustrato más estable. Las
masas de huevos de Baetidae fueron las más comunes, y éstas presentaron
una relación significativa de la abundancia de masas de huevos con el área de
la roca y el hecho de que ésta sea emergente (Tabla 6, Capítulo 4). Debido a
que muchos taxa expulsan sus huevos sobre el agua, nuestro método de
ubicación de masas de huevos excluye a aquellos taxa, cuyos huevos no han
sido depositados en rocas directamente.
No hubo una clara relación entre la diversidad de adultos y las formas acuáticas
(Tabla 7, Fig.3 Capítulo 4). La principal similitud constituye la importancia que
tiene tanto en el ambiente acuático como en el aire la familia Chironomidae. Por
otro lado la principal diferencia constituye que Baetodes, muy abundante en la
época de lluvias, y Nectopsyche, muy abundante en la época seca, estuvieron
poco representadas en las muestras de adultos. El tema de la ovoposición y su
117
Resumen General
importancia en el reclutamiento es el que quizá deja más interrogantes. Es
mucho lo que desconocemos sobre la historia de vida de los insectos acuáticos
altoandinos y su investigación primordial para el estudio de las dinámicas
poblacionales.
118
BIBLIOGRAFIA
Bibliografía
Abelho, M. 2001. From Litterfall to Breakdown in Streams: A Review. The
Scientific World Journal 1:656-680.
Acosta, R. 2005. Caracterización de la Comunidad de Macroinvertebrados
Bentónicos de la Cuenca Altoandina del río Cañete (Lima, Perú). Pages
93 pp. in Trabajo de Investigación del programa de Doctorado y
Diplomado
en
Estudios
Avanzados
en
Ecología.Universitat
de
Barcelona., Barcelona, España.
Afonso, A. A. d. O., R. Henry, R. Rodella, and C. S. Maimoni. 2000.
Allochthonous matter input in two different stretches of a headstream
(Itatinga, São Paulo, Brazil). Brazilian Archives of Biology and
Technology 43:335-343.
Allan, J. 1987. Macroinvertebrate drift in a Rocky Mountain stream.
Hydrobiologia 144:261-268.
Allan, J. D. 1978. Trout predation and the size composition of stream drift.
Limnology and Oceanography 23:1231-1237.
Allan, J. D. 1984. The size composition of invertebrate drift in a Rocky Mountain
stream. Oikos 43:68-76.
Allan, J. D., A. S. Flecker, and N. L. Mcclintock. 1986. Diel epibenthic activity of
mayfly nymphs,and its nonconcordance with behavioral drift. Limnology
and Oceanography 31:1057-1065.
Allan, J. D., and M. M. Castillo. 2007. Stream Ecology: Structure and function of
running waters, 2nd ed. edition. Springer.
Allan, J. D., and A. S. Flecker. 1989. The mating biology of a mass-swarming
mayfly. Animal Behaviour 37:361-371.
Allan, J. D., A. S. Flecker, S. Segnini, D. C. Taphorn, E. Sokol, and G. W. Kling.
2006. Limnology of Andean piedmont rivers of Venezuela. Journal of the
North American Benthological Society 25:66-81.
Allan, J. D., G. N. Herbst, R. Ortal, and Y. Regev. 1988. Invertebrate drift in the
Dan River, Israel. Hydrobiologia 160:155-163.
Allan, J. D., and E. Russek. 1985. The Quantification of Stream Drift. Canadian
Journal of Fisheries and Aquatic Sciences 42:210-215.
121
Bibliografía
Anderson,
N.
H.,
and
J.
Macroinvertebrates
in
R.
Sedell.
Stream
1979.
Detritus
Ecosystems.
Processing
Annual
Review
by
of
Entomology 24:351-377.
Baba, M., and H. Takaoka. 1989. Oviposition sites and the number of larval
instars of two mountain blackfly species, Simulium japonicum and S.
rufibasis (Diptera: Simuliidae). Japanese Journal of Sanitary Zoology
40:307-313.
Baba, M., and H. Takaoka. 1991. Oviposition Habits of a Univoltine Blackfly,
Prosimulium-Kiotoense (Diptera, Simuliidae), in Kyushu, Japan. Medical
and Veterinary Entomology 5:351-357.
Báez, S., F. Cuesta, and M. Peralvo. 1999. Caracterización vegetal de la cuenca
alta
del
Río
Oyacachi,
Reserva
Ecológica
Cayambe-Coca.
in
EcoCiencia, Quito, Ecuador.
Bañuelos, R., S. Larrañaga, A. Elosegi, and J. Pozo. 2004. Effects of eucalyptus
plantations on CPOM dynamics in headwater streams: a manipulative
approach. Archiv für Hydrobiologie 159:211-228.
Benson, L. J., and R. G. Pearson. 1987. The role of drift and effect of season on
macroinvertebrate colonization of implanted substrata in a tropical
Australian stream. Freshwater Biology 18:109-116.
Bergey, E. A. 2005. How protective are refuges? Quantifying algal protection in
rock crevices. Freshwater Biology 50:1163-1177.
Binckley, C. A., and W. J. Resetarits. 2005. Habitat selection determines
abundance, richness and species composition of beetles in aquatic
communities. Biology Letters 1:370-374.
Binckley, C. A., and W. J. Resetarits. 2008. Oviposition behavior partitions
aquatic landscapes along predation and nutrient gradients. Behavioral
Ecology 19:552-557.
Boulton, A. J., G. M. Spangaro, and P. S. Lake. 1988. Macroinvertebrate
Distribution and Recolonization on Stones Subjected to Varying
Degrees of Disturbance - an Experimental Approach. Archiv Fur
Hydrobiologie 113:551-576.
122
Bibliografía
Boyero, L., and J. Bosch. 2002. Spatial and
temporal variation of
macroinvertebrate drift in two neotropical streams. Biotropica 34:567574.
Boyero, L., and J. Bosch. 2004. Multiscale spatial variation of stone
recolonization by macroinvertebrates in a Costa Rican stream. Journal
of Tropical Ecology 20:85-95.
Boyero, L., and J. L. DeLope. 2002. Short-term recolonization of stones in a
tropical island stream. Marine and Freshwater Research 53:993-998.
Bradley, R. S., M. Vuille, H. F. Diaz, and W. Vergara. 2006. CLIMATE CHANGE:
Threats to Water Supplies in the Tropical Andes. Science 312:17551756.
Branch, H. E. 1928. Description and identification of some Chironomid egg
masses. Annals of the Entomological Society of America 2:566-570.
Brittain, J. E. 1982. Biology of mayflies. Ann. Rev. Entomol. 27:119--147.
Brittain, J. E. 1989. Life history strategies in Ephemeroptera and Plecoptera.
Series Entomologica (Dordrecht) 44.
Brittain, J. E., and T. J. Eikeland. 1988. Invertebrate drift
A review.
Bunn, S. E. 1986. Origin and fate of organic matter in Australian upland streams.
Pages pp. 277–291. in P. a. W. de Dekker, W.D., Eds., editor.
Limnology in Australia. CSIRO Australia, Melbourne.
Canhoto, C., and M. A. S. Graça. 1998. Leaf retention: a comparative study
between two stream categories and leaf types. Verh. Internat. Verein.
Limnol. 26:990-993.
Caswell, H., and J. E. Cohen. 1991. Communities in patchy environments: a
model of disturbance, competition, and heterogeneity. Ecological
Studies.
Cheshire, K., L. Boyero, and R. G. Pearson. 2005. Food webs in tropical
Australian streams: shredders are not scarce. Freshwater Biology
50:748-769.
Collier, K. J., and B. J. Smith. 1995. Sticky trapping of adult mayflies, stoneflies
and caddisflies alongside three contrasting streams near Hamilton, New
Zealand. New Zealand Natural Sciences 22:1-9.
123
Bibliografía
Colón-Gaud, C., S. Peterson, M. Whiles, S. Kilham, K. Lips, and C. Pringle.
2008. Allochthonous litter inputs, organic matter standing stocks, and
organic seston dynamics in upland Panamanian streams: potential
effects of larval amphibians on organic matter dynamics. Hydrobiologia
603:301-312.
Covich, A. P. 1988. Geographical and historical comparisons of neotropical
streams - biotic diversity and detrital processing in highly variable
habitats. Journal of the North American Benthological Society 7:361386.
Cushing, C., K. Cummins, and G. Minshall. 2006. River and Stream Ecosystems
of the World. UNIV CALIFORNIA PRESS.
Death, R. G. 2004. Patterns of spatial resource use in lotic invertebrate
assemblages. Hydrobiologia 513:171-182.
Domínguez, E., M. D. Hubbard, and W. L. P. B. Acuática, editors. 1992. Clave
para ninfas y adultos de las familias y géneros de Ephemeroptera
(Insecta) sudamericanos. Instituto de Limnología «Dr. Raúl A.
Ringuelet» (UNLP-CONICET), La Plata.
Dominguez, E., C. Molineri, M. L. Pescador, M. D. Hubbard, and C. Nieto. 2006.
Ephemeroptera of South America. in Ephemeroptera of South America.
Pensoft Publishers.
Douglas, M., and P. S. Lake. 1994. Species Richness of Stream Stones - an
Investigation
of
the
Mechanisms
Generating
the
Species-Area
Relationship. Oikos 69:387-396.
Downes, B. J., P. S. Lake, and E. S. G. Schreiber. 1995. Habitat structure and
invertebrate assemblages on stream stones: A multivariate view from
the riffle. Austral Ecology 20:502-514.
Dudgeon, D. 2008. Tropical Stream Ecology. Academic Press., San Diego.
Encalada, A., and B. Peckarsky. 2006. Selective oviposition of the mayfly Baetis
bicaudatus. Oecologia 148:526-537.
Encalada, A., and B. Peckarsky. 2007. A comparative study of the costs of
alternative mayfly oviposition behaviors. Behavioral Ecology and
Sociobiology 61:1437-1448.
124
Bibliografía
Encalada, A., and B. Peckarsky. in prep. The influence of recruitment on the
population dynamics of the mayfly Baetis bicaudatus in fish and fishless
streams. Submitted to Ecology.
Fernández, H. R., and E. Domínguez, editors. 2001. Guía para la determinación
de los artrópodos bentónicos sudamericanos. Editorial Universitaria de
Tucumán, Tucumán.
Fisher, R. A. 1930. The Genetical Theory of Natural Selection. Clarendon Press,
Oxford.
Fisher, S. G., and G. E. Likens. 1973. Energy Flow in Bear Brook, New
Hampshire: An Integrative Approach to Stream Ecosystem Metabolism.
Ecological Monographs 43:421-439.
Flecker, A. S. 1992. Fish Predation and The Evolution of Invertebrate Drift
Periodicity: Evidence from Neotropical Streams. Ecology 73:438-448.
Flecker, A. S., and B. Feifarek. 1994. Disturbance and the Temporal Variability
of Invertebrate Assemblages in 2 Andean Streams. Freshwater Biology
31:131-142.
Flint, O. S. J. 1991. Studies of Neotropical Caddisflies Xlv. The Taxonomy
Phenology and Faunistics of the Trichoptera of Antioquia Colombia.
Smithsonian Contributions to Zoology:I-113.
Flowers, R. W. 1987. The Adult Stage of 3 Central-American Baetodes
(Ephemeroptera, Baetidae) with Notes on the Genus. Aquatic Insects
9:1-10.
Friberg, N., M. J. Winterbourn, K. A. Shearer, and S. E. Larsen. 1997. Benthic
communities of forest streams in the South Island, New Zealand: Effects
of forest type and location. Archiv Fur Hydrobiologie 138:289-306.
Giller, P. S., and B. Malmqvist. 1998. The biology of streams and rivers. in The
biology of streams and rivers. Oxford University Press.
Gillett, J. D. 1971. Mosquitos. Weidenfeld and Nicolson, London.
Gjerlov, C., A. G. Hildrew, and J. I. Jones. 2003. Mobility of stream invertebrates
in relation to disturbance and refugia: a test of habitat templet theory.
Journal of the North American Benthological Society 22:207-223.
125
Bibliografía
Gonçalves, J. F., J. S. Franca, and M. Callisto. 2006. Dynamics of allochthonous
organic matter in a tropical Brazilian headstream. Brazilian Archives of
Biology and Technology 49:967-973.
Gonçalves, J. F., J. S. França, A. O. Medeiros, C. A. Rosa, and M. Callisto.
2006. Leaf Breakdown in a Tropical Stream. International Review of
Hydrobiology 91:164-177.
González, E., and J. Pozo. 1996. Longitudinal and temporal patterns of benthic
coarse particulate organic matter in the Agüera stream (northern Spain).
Aquat. Sci. 58, 355.
Gosz, J. R., G. E. Likens, and F. H. Bormann. 1972. Nutrient Content of Litter
Fall on the Hubbard Brook Experimental Forest, New Hampshire.
Ecology 53:770-784.
Graça, M. A. S. 2001. The role of invertebrates on leaf litter decomposition in
streams - A review. International Review of Hydrobiology 86:383-393.
Graça, M. A. S., C. Cressa, M. O. Gessner, M. J. Feio, K. A. Callies, and C.
Barrios. 2001a. Food quality, feeding preferences, survival and growth
of shredders from temperate and tropical streams. Freshwater Biology
46:947-957.
Graça, M. A. S., R. C. F. Ferreira, and C. N. Coimbra. 2001b. Litter processing
along a stream gradient: the role of invertebrates and decomposers.
Guhl, E. 1989. Efectos geoecológicos en la dinámica de la vegetación
antropogénica en los páramos y sus consecuencias biogeográficas en
los Andes ecuatoriales. Ecológica 1:40-45.
Hickin, N. E. 1968. Caddis Larvae: Larvae of the British Trichoptera. Hutchinson
of London.
Hinton, H. E. 1981. Biology of insect eggs. Volume I, Volume II, Volume III.
Pergammon Press., Oxford UK.
Holomuzki, J. R., and B. J. F. Biggs. 2000. Taxon-specific responses to highflow disturbance in streams:implications for population persistence.
126
Bibliografía
Holzenthal, R. W., and O. S. Flint, Jr. 1995. Studies of neotropical caddisflies, LI:
Systematics of the neotropical caddisfly genus Contulma (Trichoptera:
Anomalopsychidae). Smithsonian Contributions to Zoology 0:I-59.
Hose, G. C., T. Walter, and A. J. Brooks. 2007. Short-term colonisation by
macroinvertebrates of cobbles in main channel and inundated stream
bank habitats. Hydrobiologia 592:513-522.
Jackson, J. K., and B. W. Sweeney. 1995. Egg and Larval Development Times
for 35 Species of Tropical Stream Insects from Costa-Rica. Journal of
the North American Benthological Society 14:115-130.
Jacobsen, D. 2003. Altitudinal changes in diversity of macroinvertebrates from
small streams in the Ecuadorian Andes. Archiv Fur Hydrobiologie
158:145-167.
Jacobsen, D. 2004. Contrasting patterns in local and zonal family richness of
stream invertebrates along an Andean altitudinal gradient. Freshwater
Biology 49:1293-1305.
Jacobsen, D. 2005. Temporally variable macroinvertebrate-stone relationships in
streams. Hydrobiologia 544:201-214.
Jacobsen, D. 2008. Tropical High-Altitude Streams. Pages 219-256 in D.
Dudgeon, editor. Tropical Stream Ecology. Elsevier Inc. Academic
Press., San Diego.
Jacobsen, D., and B. Bojsen. 2002. Macroinvertebrate drift in Amazon streams
in relation to riparian forest cover and fish fauna. Archiv Fur
Jacobsen, D., and A. Encalada. 1998. The macroinvertebrate fauna of
Ecuadorian highland streams in the wet and dry season. Archiv Fur
Jacobsen, D., and R. Marín. 2007. Bolivian Altiplano streams with low richness
of macroinvertebrates and large diel fluctuations in temperature and
dissolved oxygen. Aquatic Ecology.
Jacobsen, D., R. Schultz, and A. Encalada. 1997. Structure and diversity of
stream invertebrate assemblages: the influence of temperature with
altitude and latitude. Freshwater Biology 38:247-261.
127
Bibliografía
Jacobsen, D., and E. Terneus. 2001. Aquatic macrophytes in cool aseasonal
and seasonal streams: a comparison between Ecuadorian highland and
Danish lowland streams. Aquatic Botany 71:281-295.
Kohler, S. L., and M. A. McPeek. 1989. Predation Risk and The Foraging
Behavior of Competing Stream Insects. Ecology 70: 1811-1825.
Komzak, P. 2001. The spatio-temporal diversity in caddisfly communities
(Trichoptera, Insecta) of the Oslava and Chvojnice streams (Czech
Republic). Scripta Biology (Brno) 27:63-85.
Lake, P. S., and E. S. G. Schreiber. 1991. Colonization of stones and recovery
from disturbance: an experimental study along a river. Internationale
Vereinigung
fuer
Theoretische
und
Angewandte
Limnologie
Verhandlungen 24.
Lamberti, G., S. Gregory, L. Ashkenas, R. Wildman, and A. Steinman. 1989.
Influence of channel geomorphology on retention of dissolved and
paticualte matter in a Cascade Mountain Stream. USDA Forest Service
Gen. Tech. Rep.PSW-110.:33-39.
Lancaster, J. 1999. Small-scale movements of lotic macroinvertebrates with
variations in flow. Freshwater Biology 41:605-619.
Lancaster, J. 2000. Geometric scaling of microhabitat patches and their efficacy
as refugia during disturbance. Journal of Animal Ecology 69:442-457.
Lancaster, J., and L. R. Belyea. 1997. Nested hierarchies and scaledependence of mechanisms of flow refugium use. Journal of the North
American Benthological Society 16:221-238.
Lancaster, J., T. Buffin-Belanger, I. Reid, and S. Rice. 2006. Flow- and
substratum-mediated movement by a stream insect. Freshwater Biology
51:1053-1069.
Lancaster, J., B. J. Downes, and P. Reich. 2003. Linking landscape patterns of
resource distribution with models of aggregation in ovipositing stream
insects. Journal of Animal Ecology 72:969-978.
Lancaster, J., and A. G. Hildrew. 1993a. Characterizing in-Stream Flow Refugia.
Canadian Journal of Fisheries and Aquatic Sciences 50:1663-1675.
128
Bibliografía
Lancaster, J., and A. G. Hildrew. 1993b. Flow Refugia and the Microdistribution
of
Lotic
Macroinvertebrates.
Journal
of
the
North
American
Benthological Society 12:385-393.
Larrañaga, S., J. R. Díez, A. Elosegi, and J. Pozo. 2003. Leaf retention in
streams of the Agüera basin (northern Spain). Aquatic Sciences Research Across Boundaries 65:158-166.
Lewis, W. M., Jr., S. K. Hamilton, and J. F. Saunders, III. 1995. Rivers of
northern South America. Pages 219-256 in Ecosystems of the World;
River and stream ecosystems. Elsevier Science Publishers B.V.;
Elsevier Science Publishing Co. Inc.
Likens, G. E. 1972. Eutrophication and aquatic ecosystems. in G. E. Likens,
editor. Nutrients and eutrophication. Special Symposia Vol. I Amer. soc.
Limnol. Oceanogr. Amer. soc. Limnol. Oceanogr, Lawrence, Kansas.
Likens, G. E., F. H. Bormann, N. M. Johnson, D. W. Fisher, and R. S. Pierce.
1970. Effects of Forest Cutting and Herbicide Treatment on Nutrient
Budgets in the Hubbard Brook Watershed-Ecosystem. Ecological
Monographs 40:23-47.
Luteyn, J. L. 1999. Paramos: A checklist of plant diversity, Geographic
Distribution and Botanical Literature. Memoirs of The New York
Botanical Garden 84:1-278.
Lytle, D. A. 1999. Use of rainfall cues by Abedus herberti (Hemiptera :
Belostomatidae): A mechanism for avoiding flash floods. Journal of
Insect Behavior 12:1-12.
Lytle, D. A. 2001. Disturbance Regimes and Life-History Evolution. The
American Naturalist 157:525-536.
Lytle, D. A., and N. L. Poff. 2004. Adaptation to natural flow regimes. Trends in
Ecology & Evolution 19:94-100.
Mackay, R. J. 1992. Colonization by Lotic Macroinvertebrates - a Review of
Processes and Patterns. Canadian Journal of Fisheries and Aquatic
Sciences 49:617-628.
Mathooko, J. M., G.O.Morara, and M.Leichtfried. 2001. Leaf litter transport and
retention in a tropical Rift Valley stream: an experimental approach.
129
Bibliografía
Mathuriau, C., and E. Chauvet. 2002. Breakdown of leaf litter in a neotropical
stream. Journal of the North American Benthological Society 21:384396.
Mathuriau, C., A. G. B. Thomas, and E. Chauvet. 2008. Seasonal dynamics of
benthic detritus and associated macroinvertebrate communities in a
neotropical stream. Fundamental and Applied Limnology 171.
Matthaei, C. D., C. J. Arbuckle, and C. R. Townsend. 2000. Stable surface
stones as refugia for invertebrates during disturbance in a New Zealand
stream. Journal of the North American Benthological Society 19:82-93.
Matthaei, C. H. r. D., K. A. Peacock, and C. R. Townsend. 1999. Scour and fill
patterns in a New Zealand stream and potential implications for
invertebrate refugia. Freshwater Biology 42:41-57.
McIntosh, A. R., B. L. Peckarsky, and B. W. Taylor. 2002. The influence of
predatory fish on mayfly drift: extrapolating from experiments to nature.
Freshwater Biology 47:1497-1513.
McIntosh, A. R., and C. R. Townsend. 1994. Interpopulation Variation in Mayfly
Antipredator Tactics: Differential Effects of Contrasting Predatory Fish.
Ecology 75: 2078-2090.
Melo, A. S., and C. G. Froehlich. 2004. Colonization by macroinvertebrates of
experimentally disturbed stones in three tropical streams differing in
size. International Review of Hydrobiology 89:317-325.
Melo, A. S., D. K. Niyogi, C. D. Matthaei, and C. R. Townsend. 2003.
Resistance, resilience, and patchiness of invertebrate assemblages in
native tussock and pasture streams in New Zealand after a hydrological
disturbance. Canadian Journal of Fisheries and Aquatic Sciences
60:731-739.
Merritt, R., and K. W. Cummins. 1996. An introduction to the aquatic insects of
North America. Third edition. in An introduction to the aquatic insects of
North America. Third edition. Kendall/Hunt Publishing Company.
Meyer, J. L., G. E. Likens, and J. Sloane. 1981. Phosphorus, Nitrogen, and
Organic-Carbon Flux in a Headwater Stream. Archiv Fur Hydrobiologie
91:28-44.
130
Bibliografía
Myers, N., R. A. Mittermeier, C. G. Mittermeier, G. A. B. D. Fonseca, and J.
Kent. 2000. Biodiversity hotspots for conservation priorities. Nature
403:853-858.
Nolte, U. 1993. Egg Masses of Chironomidae (Diptera) - a Review, Including
New
Observations
and
a
Preliminary
Key.
Entomologica
Scandinavica:1-75.
Ortíz, D. 2003. Ecuador. in R. P. S. P. M. Hofstede, editor. Los Páramos del
Mundo. Proyecto Atlas mundial de los Páramos. GlobalPeatland
Initiative/NC-UICN/EcoCiencia, Quito.
Palmer, M. A., P. Arensburger, P. S. Botts, C. C. Hakenkamp, and J. W. Reid.
1995.
Disturbance
and
the
Community
Structure
of
Stream
Invertebrates - Patch-Specific Effects and the Role of Refugia.
Palmer, M. A., P. Arensburger, A. P. Martin, and D. W. Denman. 1996.
Disturbance and patch specific responses: The interactive effects of
woody debris and floods on lotic invertebrates. Oecologia 105:247-257.
Passos, M. I. S., J. L. Nessimian, and L. F. M. Dorvillé. 2003. Life strategies in
an elmid (Insecta:Coleoptera: Elmidae) community from a first order
stream in the Atlantic Forest southeastern Brazil. Acta Limnol. Bras.
15:29-36.
Pearson, R. G., R. K. Tobin, R. E. W. Smith, and L. J. Benson. 1989. Standing
Crop and Processing of Rainforest Litter in a Tropical Australian Stream.
Archiv Für Hydrobiologie 115:481-498.
Peckarsky, B. L. 1996. Alternative Predator Avoidance Syndromes of StreamDwelling Mayfly Larvae. Ecology 77: 1888-1905.
Peckarsky, B. L., and A. R. McIntosh. 1998. Fitness and community
consequences of avoiding multiple predators. Oecologia 113:565-576.
Peckarsky, B. L., B. W. Taylor, and C. C. Caudill. 2000. Hydrologic and
behavioral constraints on oviposition of stream insects: implications for
adult dispersal. Oecologia 125:186-200.
Perez, B., and S. Segnini. 2007. Variación espacial de la composición y
diversidad de géneros de Ephemeroptera (Insecta) en un río tropical
altiandino. Ecotropicos 16(2):45-63 2003.
131
Bibliografía
Petersen, I., Z. Masters, S. J. Ormerod, and A. G. Hildrew. 2006. Sex ratio and
maturity indicate the local dispersal and mortality of adult stoneflies.
Pickett, S. T. A., and P. S. White. 1985. Patch Dynamics a Synthesis. Pages
371-384 in Pickett, S. T. A. And P. S. White (Ed.). The Ecology of
Natural Disturbance and Patch Dynamics. Xv+472p. Academic Press,
Inc., Publishers: Orlando, Fla., USA; London, England. Illus. Maps.
Poff, N., and J. Ward. 1990. Physical habitat template of lotic systems: Recovery
in the context of historical pattern of spatiotemporal heterogeneity.
Environmental Management 14:629-645.
Poff, N. L., J. D. Allan, M. B. Bain, J. R. Karr, K. L. Prestegaard, B. D. Richter, R.
E. Sparks, and J. C. Stromberg. 1997. The natural flow regime.
Bioscience 47:769-784.
Pozo, J., E. Gonzalez, J. R. Diez, J. Molinero, and A. Elosegui. 1997. Inputs of
Particulate
Organic
Matter
to
Streams
with
Different
Riparian
Vegetation. Journal of the North American Benthological Society
16:602-611
Pringle, C. M., and A. Ramirez. 1998. Use of both benthic and drift sampling
techniques to assess tropical stream invertebrate communities along an
altitudinal gradient, Costa Rica. Freshwater Biology 39:359-373.
Ramirez, A., and C. M. Pringle. 1998. Structure and Production of a Benthic
Insect Assemblage in a Neotropical Stream. Journal of the North
Ramirez, A., and C. M. Pringle. 2001. Spatial and temporal patterns of
invertebrate drift in streams draining a Neotropical landscape.
Rausher, M. D. 1979. Larval Habitat Suitability and Oviposition Preference in
Three Related Butterflies. Ecology 60:503-511.
Reich, P. 2004. Patterns of composition and abundance in macroinvertebrate
egg masses from temperate Australian streams. Marine and Freshwater
Research 55:39-56.
132
Bibliografía
Reich, P., and B. J. Downes. 2003a. The distribution of aquatic invertebrate egg
masses in relation to physical characteristics of oviposition sites at two
Victorian upland streams. Freshwater Biology 48:1497-1513.
Reich, P., and B. J. Downes. 2003b. Experimental evidence for physical cues
involved in oviposition site selection of lotic hydrobiosid caddis flies.
Oecologia 136:465-475.
Resetarits, W. 2001. Colonization under threat of predation: avoidance of fish by
an aquatic beetle, Tropisternus lateralis (Coleoptera: Hydrophilidae).
Oecologia 129:155-160.
Resetarits, W. J., Jr. 1991. Ecological Interactions Among Predators in
Experimental Stream Communities. Ecology 72:1782-1793
Resh, V. H., A. V. Brown, A. P. Covich, M. E. Gurtz, H. W. Li, G. W. Minshall, S.
R. Reice, A. L. Sheldon, J. B. Wallace, and R. C. Wissmar. 1988. The
Role of Disturbance in Stream Ecology. Journal of the North American
Rios, B. 2004. Las comunidades de macroinvertebrados bentónicos de dos
cuencas altoandinas del Ecuador. Trabajo de Investigación del
programa de Doctorado y Diplomado en Estudios Avanzados en
Ecología. Pages 51 pp. in U. d. Barcelona., editor., Barcelona, España.
Rodriguez-Barrios, J., R. Ospina-Torres, and J. D. Gutierrez 2007. Density and
biomass of drifting aquatic macroinvertebrates in a tropical mountain
creek at Bogotá, Colombia. caldasia 29:397-412.
Siccama, T. G., F. H. Bormann, and G. E. Likens. 1970. The Hubbard Brook
Ecosystem Study: Productivity, Nutrients, and Phytosociology of the
Herbaceous Layer. Ecological Monographs 40:389-402.
Sioli, H. 1984. The Amazon: limnology and landscape ecology of a mighty
tropical river and its basin. Dr. W. Junk, The Hague Netherlands.
Skov, F. 1999. Ambiente Físico. Pages 120 in D. H. B. Pedersen, editor.
Oyacachi- La Gente y la Biodiversidad. Centro para la Investigación de
la Diversidad Cultural y Biológica de los Bosques Pluviales Andinos
(DIVA), Dinamarca y Ediciones Abya Yala, Ecuador, Quito.
Smith, B. J., K. J. Collier, and N. J. Halliday. 2002. Composition and flight
periodicity of adult caddisflies in New Zealand hill-country catchments of
133
Bibliografía
contrasting land use. New Zealand Journal of Marine and Freshwater
Research 36:863-878.
Speaker, R., K. Moore, and S. Gregory. 1984. Analysis of the process of
retention of organic matter in stream ecosystems. Verh Internat Verein
Limnol 22:1835-1841.
Steedman, R. J., and N. H. Anderson. 1985. Life-History and Ecological Role of
the Xylophagous Aquatic Beetle, Lara-Avara Leconte (Dryopoidea,
Elmidae). Freshwater Biology 15:535-546.
Suberkropp, K., and E. Chauvet. 1995. Regulation of Leaf Breakdown by Fungi
in Streams: Influences of Water Chemistry. Ecology 76:1433-1445
Suberkropp, K., and J. B. Wallace. 1992. Aquatic Hyphomycetes in InsecticideTreated and Untreated Streams. Journal of the North American
Suren, A. M., and I. G. Jowett. 2006. Effects of floods versus low flows on
invertebrates in a New Zealand gravel-bed river. Freshwater Biology
51:2207-2227.
Sweeney, B. W., J. K. Jackson, and D. H. Funk. 1995. Semivoltinism, Seasonal
Emergence, and Adult Size Variation in a Tropical Stream Mayfly
(Euthyplocia-Hecuba). Journal of the North American Benthological
Society 14:131-146.
Terneus, E., and J. J. Vásconez. 2004. Preliminary characterization of the
aquatic ecosystems in Oyacachi and Muertepungu Basins. Lyonia
6:177-198.
Tomanova, S., E. Goitia, and J. Helešic. 2006. Trophic Levels and Functional
Feeding Groups
of
Macroinvertebrates
in Neotropical Streams.
Tomanova, S., P. A. Tedesco, M. Campero, P. A. Van Damme, N. Moya, and T.
Oberdorff.
2007.
Longitudinal
and
altitudinal
changes
of
macroinvertebrate functional feeding groups in neotropical streams: a
test of the River Continuum Concept. Fundamental and Applied
Limnology / Archiv für Hydrobiologie 170:233-241.
134
Bibliografía
Townsend, C. R., and A. G. Hildrew. 1976. Field Experiments on the Drifting,
Colonization and Continuous Redistribution of Stream Benthos. The
Journal of Animal Ecology 45: 759-772.
Townsend, C. R., and A. G. Hildrew. 1994. Species Traits in Relation to a
Habitat Templet for River Systems. Freshwater Biology 31:265-275.
Townsend, C. R., M. R. Scarsbrook, and S. Doledec. 1997. Quantifying
disturbance in streams: alternative measures of disturbance in relation
to macroinvertebrate species traits and species richness. Journal of the
North American Benthological Society 16:531-544.
Turcotte, P., and P. P. Harper. 1982a. Drift patterns in a high Andean stream.
Turcotte, P., and P. P. Harper. 1982b. The Macroinvertebrate Fauna of a Small
Andean Stream. Freshwater Biology 12:411-419.
Wallace, J. B., S. L. Eggert, J. L. Meyer, and J. R. Webster. 1997. Multiple
Trophic Levels of a Forest Stream Linked to Terrestrial Litter Inputs.
Science 277:102-104.
Wallace, J. B., and J. R. Webster. 1996. The Role of Macroinvertebrates in
Stream Ecosystem Function. Annual Review of Entomology 41:115-139.
Wantzen, K. M., F. R. Rosa, C. O. Neves, and C. N. da Cunha. 2005. Leaf litter
addition experiments in riparian ponds with different connectivity to a
Cerrado stream in Mato Grosso, Brazil. Amazoniana-Limnologia Et
Oecologia Regionalis Systemae Fluminis Amazonas 18:387-396.
Wantzen, K. M., and R. Wagner. 2006. Detritus processing by invertebrate
shredders: a neotropical-temperate comparison. Journal of the North
Ward, G. M., and K. W. Cummins. 1978. Life-History and Growth-Pattern of
Paratendipes-Albimanus in a Michigan Headwater Stream (DipteraChironomidae). Annals of the Entomological Society of America 71:272284.
135
Bibliografía
Ward, J. V. 1994. Ecology of Alpine Streams. Freshwater Biology 32:277-294.
Williams, D. D., and H. B. N. Hynes. 1976. Recolonization Mechanisms of
Stream Benthos. Oikos 27:265-272.
Winterbottom, J. H., S. E. Orton, and A. G. Hildrew. 1997. Field experiments on
the mobility of benthic invertebrates in a southern English stream.
Winterbottom, J.H., A.G. Hildrew and J. Lancaster. 1997. Field experiments on
flow refugia in streams. Freshwater Biology 37, 569-580:11.
Winterbourn, M. J., W. L. Chadderton, S. A. Entrekin, J. L. Tank, and J. S.
Harding. 2007. Distribution and dispersal of adult stream insects in a
heterogeneous montane environment. Fundamental and Applied
Limnology 168:127-135.
Wissinger, S. A., W. S. Brown, and J. E. Jannot. 2003. Caddisfly life histories
along permanence gradients in high-altitude wetlands in Colorado
(USA). Freshwater Biology 48:255-270.
Wolf, E. M., U. Matthias, and G. Roldan P. 1988. Estudio del desarrollo de los
insectos acuaticos, su emergencia y ecologia en tres ecosistemas
diferentes en el departamento de Antioquia. Actualidades Biologicas
(Medellin) 17:2-27.
Young, S. A., W. P. Kovalak, and K. A. DelSignore. 1978. distances travelled by
autumn leaves introduced into a woodland stream. American Midland
Naturalist 100:217-222.
Zuñiga de Cardoso, M. C., A. M. Rojas, and S. Mosquera. 1997. Biological
Aspects of Ephemeroptera in rivers of Southwestern Colombia (South
America).
Ephemeroptera
Systematics:261-268.
136
&
Plecoptera:
Biology-Ecology-
Anti Hawa Mayukunapi Hatun Tullu Illak
Kuru Wiwa Ayllukunamanta: Imapi
Kawsaymanta, Imashina Kuyurimanta,
Ima Wanuta Llukchimanta, Ima Shina
Lulunta Churaymantapash
Kallariy 1
Kayta killkanaka 1999 watapimi kallarirka, chaypimi wakin mayukunapi rikushpa
rurakurkani. Chaypimi wakin mayukunapi kawsak, allpapi kawsak wiwakunata
urku hawa urachacana kuskapi rikurkani. Chaytami Ecuador anti hawapi
rikurkani.
ruraypakka
Chaypi rikushpami wakin yuyaykuna shamurka, chaywanmi ñuka
ima
shina
kay
wiwakuna
mayukunapi
kawsakta
riksirkani.
Kaykunatami sumak kawsaypi hatun diplomado nishkata rurakushpa alli
yacharkani. Chaypimiima shina urku mayukunata mapayachinakuymanta
riksirkanchik. Chaypak katimi chay hawallata ishkay ruraykunata charirkanchik
San Francisco de Quito Sumak Yachay Wasiwan shinallata La Molina de Lima
Sumak Yachay Wasiwampash kay Ecuador, Perú mamallaktakunapi. Kay
yachaykunawanka Anti urku shinallata Grupo FEM Meditrerráneo charik urku
mayukunapi mapayachi hawamantami riksirkanchik shinallata (proyecto ECOBIL
y GUADALMED I y II) charishkakunatapash. Kaypimi ima shina kay mayukuna
kakta, shinallata ima wiwakunalla kaypi tiyakta riksirkanchik Shinallata ima shina
Anti mayukuna mapayashka kakta, yacharkanchik. Shinallata urku mayukuna
hawapash, chakishka mayukunatapash, ashalla yakuwan mayukunatapash
riksirkanchik shina wakin mayukunaka ashalla yakuwan kaymanta mapakunata
shitakpi chakiktapash yacharkanchik. Mana yachanchikchu mashna wiwakuna
chinkashkata, imakunalla tiyaktapash ima shina mayukuna kaktapash mana alli
yachanchikchu, shinallata ima shina wakllishka kakpi allichinatapash mana
yachanchikchu. Kay killkayka imata kay mayukuna tukunakuktami riman kayka
ashakullami kan kay mayukunata yachay hawa rimaymantaka.
Shuk kutinka kay mayukuna ima shina kakta rimankapak charirkanimi kay
Piburja mayupi wakin taripaykunata rurashka kaymanta kay mayuka Oyacachi
kichwa runakunapak allpata yallin, kaypika sumak sachakuna, sumak runakuna
kawsan shinapash kay sumak tiyashkakunaka llakipimi kan hatun llaktakuna
mapayachishpa kanakukpi shinallata yaku ukupi tiyashka wiwakunata ushashka
apinakukpi shinallata
kay
hatun
llaktamanta
runakuna
pachamamawan
phiñarishka kawsanakushkamanta.
1
Traducción de español a kichwa: Luis Enrique Cachiguango
139
Anti hawapi wacharishka kunuk mayukuna
Allpa mamapi tiyashka mayukunaka tawkapachami, shinallata allpa-kawsaychakllisinchita, yakukunata, allpa-samikunata rikushpaka ñami allillata rikuyta
ushanchik imakunatalla mayukuna charikta, ima shina kakta kaykunawan
ashtawan kay mayukuna ima shinapacha kawsanakukta yachankapak, shina
kaypika rikunchik Allan & Castillo 2007, Giller & Malmqvist 1998, Cushing et al.
2006. Paykunapak rurashkapika ima shina mayu yakukuna kakta, imakunawan
pakta kawsaktapash rikuchin, chaykunapika chiri-kunuk allpakunata yallik
mayukunapi
tawka
watakunapi
yachaykunata
rurashkata
rikuchin
kutin
shinapash kunuk allpakunata yallik mayukunapi yachaykunata rurashkaka
wakinllami tiyan. Kay hawaka chayrakmi shak kamu llukshishka (Dudgeon
2008). Shinapash Uray Europa, Uray América allpakunapika
tukuylla
yachaykunatami uchilla mayukunapi rurarishka, kutin wichay américa allpapika
yachaykunataka Orinoco, Amazonas (Sioli 1984; Covich 1988; Lewis et al.
1995) hatun mayukunapimi rurarishka, shinapash kayka mana tukuyllachu kan,
tiyanrakmi wakin yachaykunata rurashka urku hawa allpakunapi (Turcotte &
Harper 19982b; a; Flecker 1002; Flecker & Feibarek 1994; Jacobsen et al.
1997 ; Jaconsen & Encalada 1998 ; Jacobsen&Terneus 2001 ; Jacobsen 2003,
2004, 2005 ; Allan et al. 2006; Jacobsen&Marín 2007). Kaykunapika mayu
wacharimanta shinallata uchilla kunuk allpakunata yallik mayukunamantaka
mana rimanchu (Jacobsen 2008), kaykunapika anti hawamanta kunuk allpata
yallik mayukunatami rimasha ninchik kay sumak yachay killkapika.
Uray
antikunapika,
Venezuelamanta
wichayman
Huancabamba
Perú
wichaykaman (Fig. 1) allpa mama kuyunalla allpakuna kan, kaypika urkukuna,
pampakuna,
chirikunuypash
samikunami
tiyan,
shinallata
kaypika
yura
unkuykunapash tiyan (Myers et al. 2000). chaymantami anti hawa mayukunaka
paykunapak kaytaka shikan shikanta charin, chaymi ñukanchikka sinchi kallpak
mayukunata, phakchakunata, uchilla kuchakunamanta llukshik mayukunata
charinchik (Jacobsen 2008). Shinapash kunuk mayukunata rimashpaka kunuk
allpakunata yallik achka yurakunata chariktami rimanchik shinami urku
hawamanta kunuk mayukunaka chirilla allpata yallik mayukunawan paypura kan
140
kutin kaykunaka shikanmi kan kunuk allpakunata yallik mayukunawan
(Jabobsen 2008). Kay mayukunapaka pachakunami shukta shukta tikrachin,
shina Kunuk pachakunapika mana rikunachu allpa kunukyakpi, tamiyakuna
mirakpimi rikuna. Kutin chiri-kunuklla allpakunapika punchan punchanmi
shukman tikran kaykunaka kutin intiwanmi shina tukun, chaymi urkukunapika
tamiyan chay allpakunapika 11º mana kashpaka 25º chayan (Guhl 1989).
Shinallata 2º mana kashpaka 10º kunukka chayan (Luteyn 1999). Tamuyapash
kay anti hawapika kutin kutinmi urman chaymi kay allpakunapika mana achka
tiyanchu tamiya pachakuna shinallata rupay pachakunapash kutin uku
allpakunapika mana shinachu kan. kaypika watapi 500 mana kashpaka 3.000
mm yakuta tamiyanllami. Chashna kaymantami 25 mana kashpaka patsakpi
patsakkaman tsapakyanlla (Luteyn 1999; Ortíz 2003; Jacobsen 2008).
Shuyu 1. Uray Anti. Urayman saywaka Venezuela antikunami, kutin wichayman
saywaka 6º ishkantin wichayman sirik Huancabambapimi
(http://www.ucm.es/info/agrygfdp/web/aula%20de%20cartograf%EDa/amsurfis.jpg
141
Achka sami kawsaykunata charikpipash anti hawamanta mayukunataka mana
yachaykunata rurashka tiyanchu (Ward 1994; Allan et al. 2006; Jacobsen 2008).
Chayraklla urku hawa mayukunata taripashka yachaypika Jacobsen (2008)
churanmi kashna mayukunamantaka ima shina kakta, imakuna tiyaktapash
mana yachanchikchu nishpa. Kunankaman
yachaypika hatun tullu illak
wiwakunaka llashak kunuk allpa yakukunapimi kawsan shinallata kay yakukuna
tamiyashpa
mirashkakunapipash
tiyan
Pachamamapi
tiyashkakuna
kawsaypipash rikuna tiyan (Flecker&Feifarek 1994; Jacobsen et al. 1997,
Jacobsen 2005, 2008). Chaymantami pachamama shikan kaykunata charikpi,
shinallata
yakukunapash
pishiyakun
mirakun
kakpika
tullu
illak
kuru
wiwakunapash chaypimi mirashpa kawsay ushan (Resh et el. 1988; Poff &Ward
1990) Yakukuna mirakun pishiyakun kaymanta, shinallata Pachamama shikan
shikan kaykunawanka tullu illak wiwakunapash shikan mirarikunata charin
shinallata mana chay yaku ukupi tiyakkunapash shikan shikan tukun
(Turcotte&Harper 1982a; Benson&Pearson et al. 1989). Shinallata kay shuk
tukuymantaka mayu yakukunapash tukuymi shuk shuk tikran, shinami tullu illak
kuru wiwakunapash shikan mirarikunata charin, chaytaka ima shina mayu ukupi
ima shina miraripi rikunchik. Cahypak hawa kashna shuk shuk kaymantaka kay
tukku illak kuru wiwakunaka sinchipacha tikraykunatami charin yakukuna
wakimpika mirakun, wakinpika chakikun kaymanta, kaykunaka chay yaku ukupi
kawsak kurukunapash shikan shikan kaymantami shina tukun multivoltina
kaymanta (Jackson & Sweeney 1995) Shina kaymantaka chirilla kuskakunapi
watapi shuklla wiñayta charik wiwakunaka kunuk allpapika kutin kutin mirarin.
Shinapash kay anti hawapi wiwakuna kashna mirarinakushpapash kaykunataka
ima shinapacha mirarinakuktapash mana alli riksinchichu, chaymantami kaytaka
utkalla riksina urmanchik kay wiwakuna allpa mama kunukyakuymanta kay
shina kunukyashpaka kay yaku ukupi tiyak wiwakunaka tukuyllami wañunka
(Bradley et al. 2006 Jacobsen 2008).
Kashna rikushpaka kay killkaypika chusku yachaytami ashtawan rimasha nini,
kaykunaka ima shina hatun tullu illak wiwakuna Anti hawa urkukunapi ima shina
tamiya pachakunapi, rupay pachakunapi kawsaktami rikuchisha nini, yaku miray
142
pachapi: 1) Ima shina kay wiwa aylluna yaku uku allpapi kawsaymanta; 2) Ima
shina mirarishpa kawsaymanta; 3) Ima shina yaku ukupi tiyakkuna paykunapak
mikuna kaymanta; kutin 4) Ima shina hatun wiwakuna yaku ukupi lulunta
churashpa mirarimantapash. Kaykunataka pipash mana yachakuykunata
rurashkachu hawa anti urku mayukunapika.
Kawsaypak kuskamanta
Mayukunapi
yakukuna
miray
pishiyaykunaka
allipachami
kan
mushuk
kawsaykuna tiyarichun shinashpami mushuk mikunakuna yaku ukupi tiyarin,
wakimpika chay pachakunapaklla mikunakuna tiyarin (Pickett & White 1985),
chaymantami yaku ukupi tiyak wiwakuna yurakunapash paykuna usharishkapi
mirarita ushan. Shinallata yakukuna mirakpi pishiyakpi, ima kashpaka kutin kutin
yaku hatarishpa, yaku chakishpaka chaypi kawsak wiwakunaka llakitami charin
mirarinkapak, shina kaymantami kay wiwakunaka maypi mana yakukuna
llakichiklla kuskakunaman ashtawan rishpa kawsan (Lancaster 2000). Kashna
alli kuskakuna tiyashpaka ninan allimi, chaypimi wiwakunaka ashtawan alli
kawsashpa mirarita ushan shinallata chaymanta ashtawan mirashpa katita
ushan, shinami paypurakuna alli mirarishpa chay kuskapi huntayta ushan
(Lancaster&Hildrew 1993b, a; Lancaster &Belyea 1997; Townsend et al. 1997).
Kashna sumak kuskakunatami “minkarinalla” (Caswell & Cohen 1991;
Townsend &Hildrew 1994; Lancaster & Belyea 1997). Kashna kuskakuna
imakunatalla
wiwakuna
mikushpa
kawsachun
chariytaka
mana
alli
yachanchikchu kaytaka wakimpi yaku hatarishkapika mana alli yachay
ushanallachu kan yaku hatarishpa tukuyta apashpa rinlla kaymanta, shinallata
yaku hatarikpika tukuy yaku ukupi tiyashkakunami kayta chayta kuyurinkuna
(Lancaster & Hildrew 1993b, a).
Shinallata
yacharinchu
kashna
ima
wiwa
shina
ayllukunapika
yaku
hatarikpi
kunuk
allpakunapika
kaktaka
wakimpika
mana
alli
kaykunaka
wiwakunapak kawsaypika paykuna hatunyanakuypi mana kashpaka paykuna
wachanakuypi, paykuna multivoltismo kaypimi kanka yuyanchik chaypa
hawapash yakukuna llashaklla, pishiklla kashpapash kay wiwakuna miranataka
143
rikunata charin yuyanchik, wakimpika kashnashpami ñapash hatunyashpa
mirarita yanapan yuyanchik kay Chironomidae, Simuliidae, Baetidae nishka
wiwakunata,
shinapash
kunuklla
allpakunapi
tiyak
mayukunapika
kay
wiwakunaka llashak pachakunatami charin hatunyashpa mirarinkapakka,
chaymantami kaykunapika kay wiwakunaka ashtawan llakikunata charin
yakukuna hatarishpa tukuyta apakpika, chaymantami kaypi kawsak wiwakunaka
minkarinalla kuskakunata charin yaku ama apachun pakakunkapak. Kashna
yakukuna hatarikpi ama apay tukunkapak kuskakunapi pakakuytaka chirilla
allpapi tiyak mayukunapimi yachaykunata rurashka (Lancaster & Hildrew 1993b;
Winterbotton et al.1997; Winterbotton 1997; Lancaster 2000; Lytle 2001),
shinapash mana alli yachanchikchu ima shina kaykunaka kunuk allpa
mayukunapi kaktaka, kaykunapika kay wiwakunapakka hatu hatiriyka ninan
llakimi kanka yuyanchik.
Puchukaypika, anti hawa mayukunapika, ima shina kallaripi ninakurkanchik
yakukunaka kutin kutinmi hatarin, chakimpash, chaypak hawa allpakunapash
pata pukru kaymanta (phakchakuna tiyaymanta)
Sinchipacha yaku kuyurikunatami kay allpakunaka charin, chaymantami mana
alli yachanchik ima sami taxa yurakuna chay yaku ukupi tiyaktapash kay
yakukuna ama tukuyta apashpa richun. Kutin ashtawan nishpaka yakukuna
hatarishpaka ninantami tukuy ima tiyashkata shukman tikrachin chaymantami
tullu illak kuru wiwakunaka ima shinapash kashka mayukunapi kawsana tukun,
chaymantami kay killkapipash churashkanchik ima shina yakukunata allichina
trucha challuwakunata wiñachinkapak, chaypakka yaku kallpakuytaka 50%
kanatami pishiyachina kan, chaymanta shukniki yachaypika kay tapuykunatami
kutichinkapak munani: Ima shinata unaypi, ñapash yaku hatarikunapika tullu illak
wiwakunaka kawsan, shinallata ima shinata kay wiwakunaka pakakushpa mana
yakupi apay tukun, chaypak hawapash ima shinatak paykunaka yaku ukupi
kawsay ushan. Ima shinata kay tullu illak wiwakunaka yakukuna chakikpika
kawsayta ushan.
144
Chakchurishpa kutin mirarimanta
Alli kuskata maskashpa ima shina kay wiwakuna kutin mirashpa katimantaka
hatun allipacha yuyanchikmi kayta yachayka (Boyero & Bosch 2004). Kayka
mashna, maykan wiwakunalla yaku hatarishka hipa chayakkunawanmi paktarin,
chay hawapash wakin purik wiwakunawampash paktarin kay purik wiwakunaka
yaku hawaman purin, shinallata paykuna purishkapika lulunkunata churashpa
purin (Williams & Hynes 1976). Kashna kaytaka Yaku ukupi allpakunami
yanapan, shinallata chay allpakunapi mikunakunami yanapan chaypika maykan
mishak mikukmi ashtawan yallin (Resetarits 1991; Mackay 1002; Resentarits
2001) shinallata chaypakka haku hatarikunapashmi rikunata charin. Kaypika
purik wiwakunapash rikunata charin shuk kuskapilla kawsak wiwakuna
mirarichun kay wiwakunaka wakinkunaka mikunata tukuchin, chayka uchilla
wiwakunaka wañushpa tukurinllami, kutin wakinkunaka utkalla mishashpa
paykunapak mikuna kuskakunapika huntan (Townsend & Hildrew 1976). Chirilla
allpa mayukunapika purik wiwakunata, shinallata ima shina mikunakunata
allichiktaka achkami yachaykunata rurashka tiyan (Brittain & Eikeland 1988;
Mackay
1992),
kunuk
allpa
mayukunapika
wakinlla
yachaykunatami
rurashkanchik, wakinkunaka kunampilla rurashkami kan, chaypimi kay purin
wiwakunaka mana chay kuskapilla kawsan, ashtawankarin yakukuna hataripi
ashtawan puriktaka yachanchik (Pringle & Ramirez 1998; Ramirez & Pringle
2001; Jacobsen & Bojsen 2002; Rrodriguez-Barrios et al. 2007).
Kaynanilla yachaykunata rurashkapika kunuk allpa mayukunapika kutin yaku
hatarishka hipa wiwakuna mirarinkapakka shuk purik wiwakuna tiyakpillami
ushan, shinallata ima mikunakuna tiyakpimi mirarin chaymantami karuta puriklla
wiwakuna ashtawan alli mirayta ushan (Boyero & DeLope 2002; Boyero &
Bosch 2004; Melo & Froehlich 2004). Shinapash kay yachaykunataka mana
achka
rurashka kanchu,
ashalla tamiyakuna pachapilla ima shina kaktalla
rurashka tiyan, chaymantami yaku hatarikpi ima shina wiwakuna kutin
mirarimantaka mana imapash tiyan, shinallata mana yachanchikchu ima
wiwakuna ashtawan purikkuna kaktapash Shinallata mana yacharinchu
imamanta
kunuk
allpa mayukuna challuwakunata
mana charsihpapash
145
runakuna churashka challuwakunata charin (tawka trucha challuwakunamanta
shina ninchik), kaykunapika rikushkanchikmi mana kutin kutin yakukuna
hatishpa llakichikta (Flecker 1992; Jacobsen & Bojsen 2002; Jacobsen 2008).
Yakukuna
hatarishpa
wiwakuna
chakchushka
hipa
ima
shina
kutin
mirarimantaka tawka rimaykunami tiyan, ima shina yaku hatarishpa 10º chiri
yakukunapi
kawsashkamanta.
Kay
ishkay
niki
killkaypika
kaykunatami
kutichinkapak mushkanchik: Ima pachakunapitak yaku hatarishpa wiwakunata
chakchurin. Ima sami mikunakunata tiyan kay chakchurishka wiwakuna
mikuchunka. Ima wiwakunata ashtawan yaku hatarikpika apay tukun. Ima
shinatak chakchurishka kashpapash wiwakunaka kutinlla tantarishpa mirarin.
Ima
shinatak
wiwakunaka
mushuk
mikunakunawan,
mushuk
kawsak
kuskakunapika yacharin.
Mikunakuna yaku ukupi tiyaymanta
Yurakunamanta phitirishka phankakuna taku ukuman urmashkami hatun alli
mikunakuna kan kay wiwakunapakka mayukuna kallarikuk kuskakunapika
(Siccama et al. 1970; Anderson & Sedell 1979; Wallace et al. 1997; Graca
2001). Kay mikuykunataka wiwakunami mikun, kutin ashaka yaku apakpi urkuta
uraymanmi uriyakun (Likens et al. 1970; Fisher & Likens 1973) chaymantami
urku hawa mayukunapi, shinallata uray kunuk allpa mayukunapi kawsay
wiwakunaka paypura watarishka shina kan chaypimi ima tiyashkatapash
mikushpa wanuman tikranchin (Graca 2001). Chirilla allpakunapika kay ismuchik
wiwakuna,
shinallata
mikushpa
wanuyachik
wiwakunataka
achka
yachaykunatami rurashka kan (Anderson & Sedell 1979; Suberkropp & Wallace
1992; Suberkropp & Chauvet 1995; Wallace & Webster 1996; Wallace et al.
1997; Graça 2001; Graça et al. 2001b).
Kutin shinallata kunuk allpa
mayukunapi kawsak wiwakunataka mana alli riksinchikchu chaymantami mana
yachanchik ima shina mikunakunata tantachikta, ima shina pachakunapi
mirariktapash, kaytaka kashna mayukunapika achka yurakuna urmashpa yakupi
tiyakpipash mana yachanchikchu. Yuyanchikmi kaypi kawsak wiwakunaka yaku
ukupi tiyak mikunakunatami mikushpa kawsanka yuyanchik (Graca et al. 2001a;
146
Colón-Gaud et al. 2008). Kunuk allpa mayukunaka achka yurakunatami
yakuman urmashkataka charin, kayka kutin kutinmi urmanakun tukuylla watapi.
Kutin chirilla allpa mayukunaka yurakuna chakinakuy pachapillami kaytaka
charin shinapash kay kunuk allpa mayukunapi yachaykunata rurashpaka,
wiwakunapakka ashtawan mikunaka wayrakuna, sinchi tamiyakuna urmashpalla
mikunata mirachikta yachanchik (Covich 1988), tamiyakuna kallarikpi mana
kashpaka tamiyakuna tukurinakukpi (Gonçalves et al. 2006; Afonso et al. 2000)
Kay
mayukuna
urmashka
phankakunata
charinkapakka
paykuna
ima
allpakunata, ima shina kaykunamantami rikuna kan shinashpallami imatak
chariktapash yachanchik río (Speaker et al. 1984; Lamberti et al. 1989) Mana
chay sirilla yakukuna kaymantaka yaku ukupi allpakuna shakra kashpallami
ashtan kaypakka alli kanka ninchik (Mathooko et al. 2001). Yakukunakutin kutin
hatarinakukpi, kunuk allpa mayukunapika urmashka phankakuna miraytaka
mana achka kirukuna tiyaymantallaka kanchu, kaypika yakukuna mirashpa
phankakunata pashkakunapashmi kan, shinami yaku ukupika kay phankakuna
mirashpa tukuylla yaku uku allpami rakirin (Larrañaga et al. 2003). Kunuk
allpakunapika urmashka phankakunaka chakishka pachakunapimi ashtawan
tiyan (Covich 1988) chaymanta watapika 35 g AFDM/m2 (Friberg et al. 1997) a
más de 1000 g AFDM/m2 tukunkakaman mirarin (Colón-Gaud et al. 2008).
Kay chiri akkpa mayukunapi yachayta rurashpaka kay yaku ukupi tiyak
phankakunaka chay yakupi tiyak wiwakuna ama yakupi apay tukunkapak
pakakunkapakmi mutsurin ninchik (Palmer et al. 1996), mana kashpaka chay
phankakunata mikuk wiwakunapak mikunami tukun (Wallace & Webster 1996;
Graça 2001; Graça et al. 2001b). Shina yuyakpika kutin kunuk allpa
mayukunapika kay phankakunata mikuk wiwakunaka ninan mutsurishka
wiwakunami kan kay mayukunata allichinkapak (Ramirez & Pringle 1998;
Mathuriau & Chauvet 2002; Mathuriau et al. 2008), shina kashpapash kunuk
allpapi tiyak mayukuna urku hawapi wacharikushpaka kashna wiwakuna tiyayka
may hatun mutsurishkami kan (Cheshire et al. 2005). Chaypak hawa kay
wiwakunapak wiksakunaka kunuk allpa mayukunapika ashatawan sinchikunami
kan shukta shukta mikunakunata mikunakuymanta (Wantzen et al. 2005;
147
Tomanova et al. 2006; Wantzen & Wagner 2006). Kaypi rikushpaka, kay
yakuman urmashka khupakunaka ninan mutsurishka mikunakunami kan kay
yaku ukupi kawsak wiwakunapakka, shinapash urku hawa mayukunapi kawsak
wiwakunaka maymanta mikuktapash mana alli yacharinchu, imapash chaypillata
tiyashka
mikunakunatami
kikushpa
kawsanka
yuyanchik
kunuk
allpa
mayukunapi tiyal tullu illak wiwakuna imatapash mikukta rikuymantami shina
yuyanchik. (Covich 1988).
Anti hawa mayukunapi kashna kakta rimankapakka kay tapuykunatami
kutichinkapak
munanchik:
Imamantatak
anti
hawa
mayukunapika
yura
phankakuna urmayka mutsurishka kanka. Imapash rikunata charinchu kay
phankakunata apankapak yakukuna hatariyka shinallata kay yaku hatarishpaka
ima shinata wiwakunataka llakichin chakchushpa. Ima wiwakunata kay
phankakunawanka kawsan
Lulunkunata churaymanta
Yaku ukupi kawsak wiwakuna lulunkunata churayka tukuylla wiwakunapimi
rikunata charin chayllami wiwakuna miranakun mana miranakuktapash, ima
shina kutin miranakuktapash yacharin (Williams & Hynes 1976; Encalada &
Peckarsky 2006). Kutin kutin yaku hatarik mayukunapika
ima shinami kan
kunuk allpa mayukuna, kaykunapika yaku hatarika ninan llakimi kan achka wiwa
lulunkunata apashpa tukuchiklla kaymanta, chaymantami may mutsurishka kan
ashtawan wiwakuna lulunkunata churayka, shinashpallami kay wiwakunaka
tukuylla yaku ukupi tiyashka wiwa ayllukunataka kawsayta kushpa katin.
Ima shina ashtawan kay wiwakuna mirarishpa ashtawan lulunkunata churashpa
katitaka mana yachanchikchu, (shinapash taripana kanchik Encalada et al. en
prep.), kaypika
wiwakuna lulun churaymantaka uray allpakunapakllami
rimashka kan (Hinton 1981; Brittain 1989; Merritt & Cummins 1996). shinapash
shuyanami kanchik chirilla allpakunapika sisay pachakunapi, phanka urmay
pachakunapilla kashna kaktaka chay pachakunapika kurukunaka yakumanta
llukshishpa lulunta churankapaklla yakuman tikran univoltismo nishka, shuk
148
kutinlla lulun churaytaka wiwakunaka tukuy ima tiyashkakunapash alli kakpilla
churan mana shina kakpika mana imatapash churanchu, kutin kunuk allpa
mayukunapika mana tiyanchu kashna pachakuna chaymantami kaypi kawsak
wiwakunaka tawka kutin lulunkunataka churan shinashpami ima pachakuna
kakpipash llakikunata yallishpa kawsashpa katinlla kashna ruraytaka kunuk
allpakunapi kawsak tawka kurukunata rikushpami shina ninchik (Collier & Smith
1995; Jackson & Sweeney 1995) ashtawanka kunukpacha allpakunapi.
Kunampillaka anti hawa mayukunapi wiwakuna ima shina kawsan, ima shina
lulunkunata churan, chaykunataka mana yachanchikchu, imatapash mana kay
hawaka
charinchikchu
(Jacobsen
2008).
shinapash
kayta
yachanaka
mutsurishkapachami kan ima shina urku hawa mayukunapi wiwakuna
mikunakuna illashpapash kawsayta ushayta yachayka kutin wakinkunaka
chirikunapipash kawsan (~10ºC en Ecuador) shinallata samaypash
hawapi
kaymanta ashallapi kawsan (Jacobsen 2008).
Kaypi rikushpaka kay hawaka kaykunatami tapurishkanchik: Urku hawa
mayukunapika wiwakunaka ima shinata mirarinkapakka lulunkunata churan.
Tamiya pachakunaka imatatak rikunata charin. Imatatak rikunata charin wawa
wiwakuna hatun wiwakunawan yaku hatarishpa chakchukpika Chusku niki
yachaypimi kay tapuykunataka kutichinkapajk munanchik.
Kay killkapika ima shinami wiwakunaka anti hawa mayukunapi kawsayta ushan
shinallata ima shinatak kay wiwakunaka yanapan yaku ukupi kawsak
ayllukunataka.
Tukuy
kay
tapuykunataka
chusku
yachaykunapimi
churashkanchik, chaykunapimi kutichinkapak munashkanchik ima shinatak kay
wiwakunaka kashna llaki mayu ukukunapi kawsayta ushaktaka. Shinallata inglés
shimipi killkashka yachaykunataka ñukanchik shimi shinallata churashkanchik
kaykunatapash mincha punchakunami kamuman tikrachichun
149
Ashalla shimikunawan puchukanchik
Yakukuna hatarishpa ima shina chay yaku ukupi kawsak wiwakunata, shinallata
chay yaku ukupi tiyak ayllukunata yanapakta rimankapakka, Piburja mayupimi
ishkay hatun pachakunata rikushpa yachakushkanchik shinallata kaykunataka
chusku
yachaykunapimi
churashkanchik.
Kaypika
kallaripi
churashka
tapuykunata kutin churanchik kay killkapa puchukaykunata churankapak.
Ima shinatak yakukuna hatarishpaka yaku ukupi kawsak wiwakunataka
chakchun, shinallata kay hatarikunaka ima shinatak kay wiwakunataka
pakakushpa
mirarichun
yanapan,
chay
hawa
ima
shinatak
kay
wiwakunapak kawsak allpakunaka. Imatatak kay wiwakunaka yaku
chakishpa pishiyakpika kawsashpa katin.
Piburja mayupi shikan pachakunata rikuyka allipachami karka. Kaypika tullu illak
hatun
wiwakunaka
paykunapash
ima
allichirishpa
shina
pachakuna
kawsanakurka
kakpipash
shinallata
chaymanllatami
yakukuna
chakishpa
pishiyakpimi tukuy pachakunapi yalli ashtawan wiwakunata tarirkanchik
yakukuna mana hatarishpa kakpi, shinapash Baetodes nishkakunami tamiya
pachakunapika ashtawan tiyarka, katin kaykunaka rupay pachakuanpika
tutkurirkallami. kashna kaykunaka ñukanchik yuyaypika yakukuna ashtawan
kallpanakuywanmi rikunata charinka yuyanchik kay Baetodes nishkakunawan
rikushpaka
llashak
yakukunapi
ashtawan
purita
ushan.
Wiwakunapak
ayllukunataka chakishka pachakunapimi ashtawan alli rikuy usharkanchik kutin
tamiya
pachakunapika
wakinkunatallami
rikurkanchik.
Tamiyakunawan
yakukuna hatarishkakunapika mana kay wiwakuna pakakuna llaki illak
kuskakunataka tarirkanchikchu shinapash wakin
kashna
yakulla
pachakunapika
yaku
ashalla
Nectopsyche nishkakunaka
kallpakun
kuskakunapimi
yallinakurka, kaykunata rikushpami kay wiwakunaka maypi ama yaku apanalla
kuskakunata maskashpa kawsayta ushakta yacharkanchik.
150
Chashna rikushpaka ashalla kallpakuk yakuwan kutin paykunapak ñawpamanta
kawsak kuskakunawan chimpapurashpaka ima shina wiwakuna chakchiriyka
shinallatami karka, shinapash yaku pishiyashpaka wakin
(Ochrotrichia,
Claudioperla), nishkakunatapashmi pishiyachirka, kaykunaka ñapashmi samay
pishiyaktaka hapirka, shina kashpapash shukkuna tukurinakukpipash kutin
ashtawan sinchi yurakunami mirarishpa katirka (Oligochaeta, Tricladida).
nishkakuna.
Tukuy punchakunachu yakukuna mirakun pishiyakun. Ima pachakunapitak
yakukunaka shuk shuk kan, shinallata ima yura phankakunata ashtawan
yakuwan apay tukun. Ima shinatak wiwakunaka
yaku hatarikpika
chakchurin, kutin ima shinatak kutin mirarin kay mirarikunaka imatatak
rikunata charin paykunapak kawsak kuskakunawanka.
Kaykunata rikushpaka yaku ukupi kawsak ayllukunaka chishi pachakunapillami
ashata tikraykuna rikurirka ima shina chirilla urkukunapi rikushkanchik. Kutin
ishkay sami yurakunallami (Baetodes y Leptohyphes) shukman tikraytaka
rikuchishka. Kashna kayta shikan trucha challuwakunata Oncorhynchus mykiss.
churashka kaymantami kanka yuyanchik.
Kutin shukpi rikushpaka wiwakuna wañuyka pachakuna tikraypimi rikurirka, chay
pachakunapika yaku ukupi tiyak mikunakunaka mana wañurkachu kaytaka tukuy
yaku ukupi tiyak ayllukunata rikushpami shina ninchik, shina nishkakuna mirarka
tamiya pachakunapi kutin pishiyarka chakishka pachakunapi. Kayta rikushpami
tamiya pachapi yakukuna hatarikpaka achka wiwakunami wañun ninchik,
kaytaka shinallata rikurkanchik kashna pachakunapi yura phankakunapash
mirakta.
Yaku ukupi tiyak rumikunapi wiwakuna kutin kawsankapak chayaktaka
rikurkanchikmi kashka yakukuna hatarishka hipa shina wakin yakukuna sinchi
kallpakuk kuskapi tiyak rumikunapika shikan yura phankakunatami tarirkanchik
ashalla kallpakuk kuskakunapi tiyak phankakunawan rikunkapakka, shina
kallarikunapika ashtawan alli kakta rikurkanchik kutin hipamanka mirashka kakta
151
rikurkanchik Nectopsyche nishkakunata tamiya pachakunapi, kaykunapika
ashalla yaku kallpakuk kuskakunapika ashtawanmi tiyarla sinchi kallpakuk
yakupi tiyan rumikunawan rikukpika chaymantami ninchik kay wiwakuna, kay
ayllukunaka ashalla kallpakuk yakukunatami maskashpa kawsan, chaypimi
phankalla kawsay ushan.
Ashalla
shimikunapi
nishpaka:
yakukuna
hatarishpami
wiwakunataka
chakchushpa mirachinkapak yanapan shinallata kay wiwa mirarikunaka
paykunapak kawsak kuskakunawampash rikunata charin.
Imatatak rikunata charin mana chaymanta kashpata mushuk mikunakuna
anti hawa mayukunapi yaykushpaka. Imapash alli tiyanchu kashna
mikunakuna yaykushpa chay yakukunapi, shinallata ima shinata llakichin
mana kashpaka yanapan chaypi kawsak wiwekunataka. Imakunatak mikun
chay mikunakunataka.
Piburja mayupika mana paykunallata mikunata rurayta ushakkunallami rikurirka,
kaypika kashna mikuykunaka kutin kutinmi yaykunakurka,
shinapash
kaykunaka
ashtawan
mana
tamiya
pachakunapilla
yaykurkachu,
shina
yaykurkuna tiyarka mana achka tamiyakuk pachapi, kutin wakimpika tamiyakuk
pachapipash yaykuykuna tiyarka chaymanta mana allipachaka yacharinchu ima
pachakunapipacha kashna yakukuna hatarishpa tukuyta apak pachakuna
tiyaktaka, chaymantami tukuy ima yaku ukupi tiyashkapash chakishka
pachakunapika mayukunapi huntarka.
Yaku ukupi tiyashkakunaka tukuypimi rikunata charirka ima pachakunapi
ashtawan yurakuna tiyan, imakunalla tiyan, ima shina tamiyakuypipash rikunata
charin. Kasha pachakunapika yaku ukupi kawsak ayllukunaka (paykunapak
chunchullikunata
rikushpaka)
paykunaka
kashnakunatami
mikun
nishpa
killkashka kamukunapi rikushpaka kashna yurakunatami ashtawan mikushka
karka, ishkantin yachayta ruurashka pachakunapika raku, hamtsi yura
phankakunatami mikushka karka. kaykunata rikushpami. Kay wiwa ayllukunaka
kay mayupika yaku ukupi ima tiyashkakunatami mikun kashka yuyanchik, shina
152
rikushpaka shuk kuskakunapipash shuk mikunakunata charinka yuyanchik.
Chaymantami yaku ukupi kawsak kurukunaka ima pachapi kashpapash ima
yaku ukupi tiyashkakunata mikunka yuyanchik.
Ima shinatak mirarinkapakka anti hawa mayukunapika wiwakunaka
lulunkunata churan. Kaypika pachakuna rikunata charinchu. Kaypakka
imatatak rikunata charin yaku ukupi kawsakkuna, shinallata yaku hatarikpi
chakchurik wiwakunaka, rukuyashka wiwakunawan rikushpaka.
Ishkantin pachakunapimi yakupi kawsak rukuyashka kurukunata tarishkanchik,
ashtawanka
ishkay
rikrayuk
kurukunata
tarishkanchik.
Rukuyashka
kurukunataka mana mirarirka, pishiyarikta rikushkanchikchu tamiya pachapi,
rupay
pachapipash,
phankakunaka
pishiyaktami
kallpariyanakukta,
tarishkanchik,
shinapash
(Ephydridae
rikushkanchik.
shinapash
(kaytaka
wakinlla
wakinlla
lulunyuk
and
Ochrotrichia).
Tarishkanchikmi
yura
achka
lulunkunata
churanakuktaka
warmikunata,
wakinlla
lulun
churashkakunata rikushpa ninchik). Kashna kayka ishkantin pachakunapimi
shinallata kashka. cahymantami tukuy pachakunapi lulunkunata churanka
ninchik
chayka tukuy watatami wacharishpa mirarinakusha yallinka ninchik,
shinapash kayta ninkapakka taripashkakunata mutsunchik shuk watatallapash
kaykunaka shinapacha kakta alli yachankapak. Kashna maskashpaka wakin
wiwakunatalla tarishkanchik wakinlla lulunkunatalla tarishkanchik ishkantin
pachakunapi, tarishpapash kay lulunkunaka yakumanta llukshishka hatun
rumikunapilla tiyarka. Kaytaka llakikunamanta kishpirinkapakmi shina ruranka
yuyanchik, wakin yura phankakunaka kurukunapak lulunkunataka yaku
hawaman llukchin. Ñukanchikka shuk sami yura phankakunata rikushkanchik,
chaypi tiyak lulunkunaka shuk yakukunapi tiyanka yuyanchik.
Mana alli rikuy ushashkanchikchu ruku yaku kurukunawan yaku ukupi tiyak
ayllukunawan ima shina rantimanta kawsaktapash. Ñukanchik yuyaypika ima
shina allpa hawapi kawsak ayllukuna rantimanta kawsankapak yanaparishpa
kawsan Chironomidae. Shinallatami yaku ukupipash yanaparishpa kawsanka
153
yuyanchik. Chay hawa shikan kaytaka Baetodes ayllumi tamiya pachapika
ashtawan tiyan, kutin Nectopsyche ayllumi rupay pachakunapika ashtawan tiyan
rukuyashka
kurukunata
rikushpaka.
kurukuna
lulun
churaymanta,
mirarimantapash rimashpaka ashtawan tapuykunatarkmi sakin yuyanchik.
Ashtakatami mana riksinchik anti hawa mayukunapi yaku kurukunapak
kawsaymantaka chaymantami taripaykunaka mutsurishkapacha kan ima shina
wiwa ayllukuna kutin kutin mirarishpa kawsanakukta yachankapak
154

comunidades de macroinvertebrados en un rio altoandino

Transcription

Similar documents

File

blessed virgin mary help of christians church

Forecast - Info

TODAY`S READINGS First Reading — When the wicked turn away

bâÜ _twç Éy _ÉâÜwxá V{âÜv - Our Lady of Lourdes

ST. PIUS V CATHOLIC CHURCH—May 22, 2016

St. Jerome Church - St. Jerome Catholic Church

Cuarteles - Portal de Turismo de Puente Genil

St. Joseph Catholic Church

Sts. Peter and Paul Catholic Church Sanctuary of St. Toribio Romo

The phylogeny of yuccas - Plant Ecology at Syracuse

PASTIS: an R package to facilitate phylogenetic assembly with soft

On the Phylogeny and Taxonomy of the Genus Uromastyx