Macroalgas Bentónicas De Los Rı́os De Castilla

UNIVERSIDAD DE CASTILLA -LA MANCHA ESCUELA TÉCNICA SUPERIOR DE INGENIEROS AGRÓNOMOS Y MONTES Departamento de Producción Vegetal y Tecnología Agraria CENTRO REGIONAL DE ESTUDIOS DEL AGUA Área de Limnología Aplicada e Hidrobiología

Macroalga Castilla - La Mancha. Bases paras el bentónicas estudio de de su los potencial rı́os de indicador.

Memoria presentada por Laura Monteagudo Canales para optar al grado de Doctor en Ciencia e Ingeniería Agrarias por la Universidad de Castilla - La Mancha.

Director: José Luis Moreno Alcaraz.

Albacete, 2015

FINANCIACIÓN

Los trabajos realizados en el desarrollo de esta Tesis Doctoral han sido financiados por los siguientes convenios y proyectos concedidos por el gobierno regional (Junta de Comunidades de Castilla-La Mancha):

• PREG01-0016. Convenio específico de colaboración entre la Consejería de Obras Públicas de la Junta de Comunidades de Castilla-La Mancha y la Universidad de Castilla-La Mancha en materia de calidad de aguas. • PREG06-027. Estudio del Estado Ecológico de las masas de agua de CLM: determinación de caudales ecológicos, estaciones de referencia y cursos naturales. • PO1109-0190-8090. Uso de la flora acuática como bioindicador de la calidad del agua de los ecosistemas fluviales de Castilla-La Mancha en aplicación de la Directiva marco del agua (2000/60/CE). • PPII10-0271-1349. Macrófitos acuáticos de los ríos de Castilla-La Mancha: condiciones de referencia, intercalibración de índices de calidad y evaluación del estado ecológico según la Directiva Marco del Agua (2000/60/CE).

A mi madre

Brillaste tanto que sólo con la estela que dejas tras de ti, iluminas todo mi camino. Nos vemos al final.

ÍNDICE

1. RESUMEN ...... 1 2. INTRODUCCIÓN ...... 7 3. OBJETIVOS ...... 15 3.1. Diversidad y distribución de las macroalgas bentónicas de Castilla-La Mancha ...... 17 3.2. Presiones antropogénicas en el área de estudio ...... 17 3.3. Relación entre la comunidad de cianobacterias y la calidad del agua ...... 17 4. PLAN DE TRABAJO Y METODOLOGÍA GENERAL ...... 19 5. ARTÍCULOS CIENTÍFICOS ...... 23 5.1. Lista de géneros de macroalgas fluviales recolectadas en Castilla-La Mancha durante el período 2001-2014...... 25 5.2. Descripción morfológica y ecología de algunas algas consideradas ‘raras’ en los ríos de Castilla – La Mancha...... 41 5.3. Sobre la presencia de Nostochopsis lobata Wood ex Bornet et Flahault en España: aspectos morfológicos, ecológicos y biogeográficos...... 53 5.4. Eutrofización fluvial: Comparación del impacto del regadío y secano a través de diferentes escalas espaciales...... 73 5.5. ¿Son las cianobacterias bentónicas indicadores de presiones antropogénicas en los sistemas fluviales? ...... 89 6. DISCUSIÓN GENERAL ...... 103 7. CONCLUSIONES ...... 111 7.1. Diversidad y distribución de las macroalgas bentónicas de Castilla-La Mancha ...... 113 7.2. Presiones antropogénicas en el área de estudio ...... 113 7.3. Relación entre la comunidad de cianobacterias y la calidad del agua ...... 114 8. BIBLIOGRAFÍA ...... 117 AGRADECIMIENTOS ...... 125

RESUMEN Nacimiento del Ojuelo, Munera (Albacete) Fotografía: José Luis Moreno Alcaraz

1 2 1. RESUMEN

Las especies indicadoras ofrecen, simplemente con su presencia, información acerca de las condiciones físicas y/o químicas del entorno que les rodea. Por tanto, la evaluación del potencial indicador de las macroalgas fluviales debe partir de una base sólida de conocimiento acerca de la diversidad de estos organismos y de cómo responden a las presiones antropogénicas que afectan a la calidad del agua.

En Castilla-La Mancha, la mayoría de estudios sobre algas se han centrado en la cuenca del Segura, una de las cinco que forman parte de este territorio. En el resto, los trabajos se centran principalmente en humedales bajo alguna figura de protección. Por otro lado, del hecho de que la actividad económica principal en la región sea la agricultura, se puede deducir que este uso de suelo es el primer causante de la eutrofización en ríos y arroyos. Sin embargo, no todos los tipos de agricultura afectan con la misma intensidad a las masas de agua, por lo que debería estudiarse independientemente el impacto de diferentes técnicas agrícolas como el regadío y el secano, entre otros. Otro factor a tener en cuenta en el análisis de las presiones es la escala espacial de trabajo, es decir, a qué escala se debe estudiar la influencia de los usos de suelo en la calidad del agua (p. ej. área total de drenaje, corredores de un determinado ancho, etc.).

El objetivo final de esta memoria es contribuir a sentar las bases y la metodología para el estudio del potencial indicador de las macroalgas. Para ello, se parte del estudio general de la diversidad y distribución de las macroalgas; del análisis de las presiones responsables de la eutrofización en ríos y arroyos en la región; para culminar con la aplicación y puesta a prueba de estos conocimientos en el estudio del potencial indicador de un grupo concreto de macroalgas, las cianobacterias bentónicas (Figura 1). Como resultado, se han generado cinco artículos científicos que recogen todas las aportaciones generadas en el desarrollo de esta Tesis Doctoral.

El primero de ellos (Artículo 5.1.) consiste en un listado de los géneros de macroalgas bentónicas recolectados en los ríos Castilla-La Mancha durante los

3 muestreos llevados a cabo por el Área de Limnología Aplicada e Hidrobiología del Centro Regional de Estudios del Agua, desde el año 2001 hasta 2014.

En los dos siguientes (Artículos 5.2. y 5.3.), se profundiza en los aspectos morfológicos, ecológicos y biogeográficos de algunas especies de interés a nivel europeo, nacional y/o regional que se han detectado en dichos muestreos: Nostochopsis lobata Wood ex Bornet et Flahault, Batrachospermum atrum (Hudson) Harvey, Chroothece richteriana Hansg, Oocardium stratum Nägeli, Tetrasporidium javanicum Möbius e Hydrurus foetidus (Villars) Trevisan.

En el cuarto artículo (5.4.) se recoge el análisis de las presiones antropogénicas que afectan a la calidad del agua en la región. De este trabajo se desprenden dos ideas principales: 1) el impacto de la agricultura de regadío en la calidad del agua es mayor que el de la agricultura de secano; 2) la escala espacial más adecuada para analizar la presión de los usos de suelo en el área de estudio es la correspondiente a 1km de radio aguas arriba de cada punto de muestreo. Además, se describe la metodología general para la elección de la escala espacial más adecuada en este tipo de estudios.

Para finalizar, en el último artículo (5.5.) se aplican las aportaciones de los trabajos previos para evaluar el potencial indicador de las cianobacterias. El enfoque estadístico elegido para este trabajo, ha permitido desentrañar la relación subyacente entre la comunidad de algas y las presiones humanas. Más allá de la conductividad, que aparece como el factor más determinante en la composición de especies de estas comunidades, las cianobacterias aportan información acerca de los niveles de nutrientes en el agua. De entre todas las especies incluidas en este estudio, cinco han mostrado tener capacidad indicadora de las presiones antropogénicas y podrían incluirse en los índices de macrófitos empleados en aplicación de la Directiva Marco del Agua (2000/60/CE): Nostoc verrucosum Vaucher ex Bornet et Flahault, Phormidium autumnale Gomont, Plectonema tomasinianum Bornet ex Gomont, Rivularia haematites C. Agardh ex Bornet et Flahault y Tolypothrix distorta Kützing ex Bornet et Flahault.

4 La contribución más importante de esta Tesis Doctoral reside en su interés global, ya que estas bases pueden ser aplicadas en la evaluación del potencial indicador de otros grupos de algas (rojas, verdes, etc.) y en otras zonas geográficas.

Figura 1. Estructura general de la presente Tesis Doctoral.

5 6

INTRODUCCIÓN Río Escabas, Priego (Cuenca) Fotografía: Laura Monteagudo Canales

8 2. INTRODUCCIÓN

La Directiva Marco del Agua (2000/60/CE) (DMA) establece un marco jurídico comunitario para proteger, regenerar y garantizar el uso sostenible del agua a largo plazo. En ella se insta a los Estados Miembros a alcanzar el “buen estado” de todas sus masas de agua antes de diciembre de 2015. El estado de una masa de agua es el grado de alteración que presenta respecto a sus condiciones naturales y viene determinado por sus estados químico y ecológico. En cuanto al estado ecológico, los Estados Miembros deben establecer sistemas de seguimiento de los elementos biológicos (invertebrados, diatomeas, macrófitos y peces, en el caso de los ríos) mediante la aplicación de métodos que permitan cuantificar la calidad en las masas de agua. Dichos métodos utilizan índices o métricas que clasifican finalmente el estado ecológico de una masa de agua en una de las 5 clases: muy bueno, bueno, moderado, deficiente y malo (Anexo V, DMA). Como consecuencia de la entrada en vigor de la DMA, han proliferado este tipo de métricas, siendo más populares las basadas en invertebrados y diatomeas que las basadas en macrófitos (Ferreira et al., 2005).

Con el término “macrófitos” se hace referencia a cualquier organismo autótrofo de tamaño macroscópico que se desarrolla en un ecosistema acuático. Desde un punto vista sistemático, es un grupo muy amplio y variado, abarcando organismos como plantas superiores, briófitos y algas (Cirujano et al., 2011). El hecho de que los índices basados en macrófitos ocupen un segundo plano entre el resto de métricas parece deberse principalmente a la baja precisión que han mostrado al aplicarse en diferentes regiones (Demars et al., 2012).

Las algas que desarrollan talos macroscópicos se denominan comúnmente “macroalgas” y son organismos exclusivamente acuáticos, íntimamente ligados a las condiciones ambientales del agua. A pesar de que la DMA las propone como indicador biológico del estado ecológico junto con el resto de macrófitos (Anexo V, DMA), la mayoría de macroalgas suelen estar poco representadas en los índices de macrófitos. Sin embargo, recientemente se está corrigiendo esta tendencia. A nivel nacional, se ha realizado un esfuerzo en generar diversos índices de macrófitos que

9 incluyen las macroalgas en sus listados de taxones indicadores, con el objetivo de ser aplicados en la evaluación del estado ecológico de los ríos peninsulares (Moreno et al., 2006, 2008; Suárez et al., 2005; Flor-Arnau et al., 2015).

Hay dos factores que pueden estar detrás de esta discriminación: la falta de conocimiento sobre las respuestas de estos organismos a las presiones antropogénicas (Thacker and Paul, 2001) y la complejidad de la identificación de algunos taxones a nivel de especie (Marquardt & Palinska, 2007). Cabe pensar, entonces, que es posible perfeccionar los índices de macrófitos identificando, e incorporando en ellos, aquellas especies de macroalgas que cumplan con los requisitos que debe cumplir un buen bioindicador. Estos serían, según Bellinger y Sigee (2010): aportar información sobre las condiciones físicas y/o químicas del entorno que les rodea, tener un rango ecológico más bien estrecho, una distribución geográfica amplia y que su identificación taxonómica sea fiable.

Como primer paso para identificar estos bioindicadores entre las macroalgas, es fundamental ahondar en el conocimiento de su diversidad y ecología, así como de las presiones que afectan a la calidad del agua que habitan. En lo que respecta a la flora algal, en los últimos años se ha ampliado el conocimiento sobre su diversidad y distribución geográfica como consecuencia de la aplicación de la DMA. Castilla-La Mancha cuenta con un escenario hidrogeológico muy diverso, estando bien representados los ríos silíceos y calcáreos, tramos altos y medios, climas húmedos y secos, ríos permanentes y temporales (Moreno et al., 2006). En estos ríos temporales, muy comunes en la zona mediterránea, es difícil que puedan desarrollarse plantas vasculares y briófitos verdaderamente acuáticos ya que gran parte del año están secos (Dodkins et al., 2012). Por el contrario, las algas sí tienen la capacidad de desarrollarse en estos cortos periodos de tiempo y además soportan condiciones de sequía siendo capaces de restablecerse rápidamente al restaurarse el caudal (Romani and Sabater, 1997; Robson, 2000, Robson et al., 2008).

Esta diversidad de ecosistemas acuáticos en la región favorece la existencia de una alta biodiversidad de la flora acuática en general, y de la flora algal en particular. Sin embargo, los estudios sobre algas en Castilla-La Mancha son muy escasos. De las

10 cinco cuencas principales que incluye la región, Tajo, Guadiana, Júcar, Guadalquivir y Segura, ésta última es la más intensamente estudiada desde el punto de vista de las algas fluviales (Aboal y Llimona, 1985; Aboal, 1988a, b, c; Aboal, 1989a, b, c; Sabater et al., 1989). En cuanto al resto de cuencas, las publicaciones se centran principalmente en los humedales de la región, destacando los trabajos de Aboal (1996), Álvarez-Cobelas (2007) y Cirujano y Medina (2002).

Con el objeto de contribuir al conocimiento de las macroalgas en la región, el Área de Limnología Aplicada e Hidrobiología del Centro Regional de Estudios del Agua ha venido realizando campañas extensivas de muestreo en los ríos de la región durante el período 2001-2014. Estos datos han constituido la base florística y ecológica en el desarrollo de la presente Tesis Doctoral. Desde un punto de vista aplicado, destaca la aportación por dicho equipo de un índice de macrófitos para la evaluación del estado ecológico de los ríos (IVAM, Índice de Vegetación Acuática Macroscópica; Moreno et al., 2006). Este índice recoge un listado de 36 géneros de macroalgas recolectados en la región hasta mitad de la década pasada, constituyendo un primer referente a escala regional.

En cuanto a las presiones antropogénicas que sufren los ecosistemas acuáticos, uno de los impactos que producen mayor detrimento en la calidad del agua es la eutrofización, consecuencia de la transformación de terrenos naturales a sistemas dominados por la actividad humana (Lund, 1967; Omernik et al., 1981; Smith, 2003). La eutrofización provoca cambios en la composición de especies y la proliferación de algas filamentosas, lo que conlleva el descenso del oxígeno disuelto, el empeoramiento de la calidad del agua y la pérdida de biodiversidad (Carpenter et al., 1998; Quinn, 1991; Smith et al., 1999).

La relación entre la eutrofización y las especies bioindicadoras puede analizarse a través de la concentración de nutrientes del agua, una medida directa, aunque puntual y esporádica; mientras que el análisis de los usos de suelo, representa una forma indirecta aunque posiblemente más precisa, ya que refleja el escenario real que genera el impacto a más largo plazo. Para analizar la relación entre usos de suelo y calidad del agua de forma adecuada, debe tenerse en cuenta que intervienen

11 numerosos factores a través de diferentes escalas espaciales y temporales (Frissell et al., 1986). De ahí que puedan surgir algunas cuestiones como: (1) qué clases de usos de suelo deben contabilizarse; (2) qué escala espacial de trabajo sería la más apropiada para ello.

En cuanto a la primera cuestión, los usos de suelo más comúnmente estudiados como causa de eutrofización de ríos y arroyos a nivel global, han sido el urbano y el agrícola, ya que son fuentes de contaminación puntual y difusa, respectivamente (Osborne and Wiley, 1988; Townsend et al., 1997; Schiller et al., 2008). En el caso de Castilla-La Mancha, prácticamente la mitad de su superficie está destinada a la agricultura (Ministerio de Agricultura, Alimentación y Medio Ambiente, 2014), por lo que la contaminación difusa es muy relevante. Sin embargo, no todos los tipos de agricultura tienen el mismo impacto. Existen trabajos previos en los que se comparan diferentes actividades agrícolas. Por ejemplo, Johnson et al. (1997) que compararon el cultivo en hileras con la agricultura de conservación (sin arado) o Lassaletta et al. (2009) que compararon entre cultivos permanentes, terreno arable y zonas mixtas en España. Según el Instituto Nacional de Estadística (www.ine.es) el regadío es más productivo que el secano debido al aporte artificial de agua y fertilizantes a los cultivos. Mediante esta técnica, aumenta el drenaje de compuestos de nitrógeno a las aguas superficiales, lo que la convierte en la principal causa de eutrofización de ríos en España (Cavero et al., 2003; Berzas et al., 2004; Álvarez- Cobelas et al., 2010). Por tanto, aunque en Castilla-La Mancha el regadío supone aproximadamente el 12% del total del suelo agrícola (Ministerio de Agricultura, Alimentación y Medio Ambiente, 2014), parece lógico diferenciarlo de otros tipos de agricultura más extensos pero con menor impacto potencial. Por todo ello, en el desarrollo de los trabajos que componen esta memoria se ha dividido el uso agrícola en tres subclases: regadío, secano y agricultura de bajo impacto (zonas agrícolas con una alta densidad de vegetación natural).

En cuanto a la segunda cuestión, hay autores que sugieren que debe emplearse la escala de cuenca de drenaje (Figura 2.a.) (Omernik et al., 1981; Richards y Host, 1994; Roth et al., 1996) ya que consideran que todo impacto aguas arriba de un punto tiene efecto en él; mientras otros argumentan que los usos de suelo más

12 cercanos tienen mayor influencia y, por tanto, deben emplearse escalas locales, como los corredores o zonas de influencia (Figura 2.b. y 2.c., respectivamente) (Harding et al., 1998; Nerbonne y Vondracek, 2001).

Figura 2. Representación de las diferentes escalas espaciales empleadas en el estudio de la influencia de los usos de suelo en la calidad del agua. a: subcuenca de drenaje; b: corredor longitudinal; c: zona de influencia (buffer radial).

En estudios más recientes, algunos autores optan por hacer una evaluación previa a través de varias escalas espaciales para determinar cuál es la más apropiada para el estudio en cuestión (p. ej. Chang, 2008; Tran et al., 2010). Este marco de trabajo ‘multi-escala’ es el que se ha considerado en el desarrollo de esta tesis. Así, se contabilizaron los usos de suelo que afectan a cada punto a través de diferentes escalas espaciales, de forma que los datos obtenidos variaron de una escala a otra. El análisis profundo de estos datos ha permitido detectar cuál es la escala más apropiada para el estudio del potencial indicador de las macroalgas bentónicas.

De entre todas las algas, las cianobacterias (o algas verde-azules) destacan por ser un grupo muy diverso, extendido y cuyas características fisiológicas exclusivas hacen que se puedan ver influenciadas por los nutrientes de manera distinta al resto de algas. Estos organismos son capaces de fijar nitrógeno atmosférico, lo que les permite vivir en condiciones de escasez de compuestos de nitrógeno disueltos. Por ello, algunos autores sugieren que su dependencia de nitrógeno en comparación con otros grupos de algas puede ser mucho menor (Larkum et al., 1988). Además, también existen diferencias entre cianobacterias con y sin heterocitos. Las primeras,

son capaces de fijar nitrógeno atmosférico (N2) en condiciones aerobias, mientras

13 que las segundas tienen esta capacidad limitada a condiciones anaerobias y/o de oscuridad (Potts, 1979; Lee, 2008). Loza et al. (2014) señalan que esta ventaja ecofisiológica puede ser el motivo de que las cianobacterias con heterocitos sean dominantes en ambientes pobres en compuestos de nitrógeno, y que las cianobacterias sin heterocitos tengan preferencia por zonas con altos niveles de los mismos. Con respecto al fósforo, algunas especies, como las del género Rivularia pueden sobrevivir en concentraciones limitantes de fósforo gracias a la actividad fosfatasa, la cual puede utilizarse como un buen indicador de condiciones oligotróficas (Mateo et al., 2010). Con todo ello, hay diversos estudios en todo el mundo que las relacionan con actividades humanas causantes de eutrofización en sistemas acuáticos (p.ej. Johansson, 1982; Jafari y Gunale, 2006; Parikh et al., 2006).

Todos estos factores ponen en evidencia que las cianobacterias son un grupo de organismos muy diverso, en el que encontramos especies propias tanto de sitios impactados como de sitios limpios. Por tanto, cabe esperar que algunas de estas especies sean buenos indicadores que puedan tenerse en cuenta en el desarrollo de índices biológicos de la calidad de agua. Para identificarlas, es esencial analizar la relación entre la comunidad de cianobacterias y el entorno, así como aplicar los requisitos que debe cumplir un buen bioindicador.

A través del caso concreto de las cianobacterias, este trabajo pretende sentar las bases generales del estudio del potencial indicador de las macroalgas en un marco de trabajo adecuado, de forma que sean aplicables al estudio de otros grupos de algas en el futuro.

OBJETIVOS Arroyo Alarconcillo, Ossa de Montiel (Albacete)

Fotografía: José Luis Moreno Alcaraz

16 3. OBJETIVOS

En este capítulo quedan recogidos los objetivos generales de esta Tesis Doctoral, agrupados según las líneas de trabajo planteadas en la estructura de la misma (Figura 1). Los objetivos específicos de cada publicación pueden ser consultados en su correspondiente capítulo.

3.1. Diversidad y distribución de las macroalgas bentónicas de Castilla-La Mancha:

. Conocer la diversidad y la distribución de las macroalgas bentónicas de Castilla-La Mancha (Artículos 5.1., 5.2. y 5.3.). . Describir la morfología y ecología de algunas especies consideradas ‘raras’ a nivel europeo o que suponen primeras citas en la región (Artículos 5.2. y 5.3.). . Revisar y comparar los aspectos morfológicos, ecológicos y la distribución geográfica de Nostochopsis lobata Wood ex Bornet et Flahault (Artículo 5.3.).

3.2. Presiones antropogénicas en el área de estudio:

. Demostrar que la escala espacial influye en los resultados de los estudios que relacionan los usos de suelo con la calidad del agua (Artículo 5.4.). . Determinar qué tipo de agricultura es responsable, en mayor medida, de la eutrofización de ríos y arroyos en la región, así como cuantificar su umbral de presión (Artículo 5.4.). . Proponer un protocolo general que ayude a la elección de la escala espacial más adecuada en estudios relacionados con el impacto de la contaminación difusa (Artículo 5.4.).

3.3. Relación entre la comunidad de cianobacterias y la calidad del agua:

. Analizar las variables ambientales que determinan la comunidad de cianobacterias bentónicas (Artículo 5.5.).

17 . Examinar las diferencias entre cianobacterias con y sin heterocitos (Artículo 5.5.). . Identificar qué especies son útiles como indicadoras de presiones antropogénicas en la zona de estudio (Artículo 5.5.).

PLAN DE TRABAJO Y METODOLOGÍA Río Sorbe, Galve de Sorbe (Guadalajara)

Fotografía: Laura Monteagudo Canales

20 4. PLAN DE TRABAJO Y METODOLOGÍA GENERAL

El trabajo desarrollado durante estos años puede agruparse en tres fases o líneas de estudio (Figura 3).

Figura 3. Esquema del plan de trabajo. Se relacionan los aspectos metodológicos y objetivos principales de cada una de las tres fases.

Las tres fases se abordaron siguiendo un orden cronológico. La primera fase comenzó en 2001, cuando se constituyó el equipo de Limnología Aplicada e Hidrobiología del Centro Regional de Estudios del Agua, años antes del comienzo de este proyecto doctoral. Los datos reunidos hasta el momento fueron incluidos y ampliados durante el desarrollo de la misma, hasta el año 2014. Durante los

21 muestreos se procedió a la recogida de las colonias de macroalgas en tramos de 100 m recorridos en zigzag, cubriendo la mayor variedad de hábitats posible. A su vez, también se tomaron muestras de agua para su análisis en laboratorio. Todas las macroalgas recogidas se identificaron, como mínimo, a nivel de género y se procedió a su conservación en formaldehido 3%. Con parte del material, se realizaron preparaciones permanentes con glicero-gelatina, pliegues o sobres en seco, y viales con gel de sílice. La colección se encuentra almacenada en el citado centro de investigación.

La fase referente al análisis de las presiones antropogénicas comenzó en 2010 y se llevó a cabo paralelamente a los muestreos de campo. En ella se analizaron capas vectoriales de usos de suelo (Corine Land Cover) y se establecieron las clases de usos de suelo a analizar: forestal, uso urbano, cultivos de regadío, cultivos de secano y terrenos agrícolas considerados de bajo impacto ambiental (pastos o cultivos abandonados con alta densidad de vegetación natural). El análisis del impacto de estas clases de suelo se analizó a través de diferentes escalas: área de drenaje, corredores, y zonas de influencia de 1 y 5 km de radio aguas arriba de los puntos de muestreo.

La última, fue la fase de aplicación de los conocimientos adquiridos y metodologías descritas en fases anteriores. Para ello se seleccionó un subgrupo de macroalgas, las cianobacterias, con buena representación territorial, y fueron identificadas a nivel de especie. Con este grupo ‘piloto’ se evaluó el efecto de diferentes variables ambientales en la composición de especies de la comunidad. Se aplicaron diferentes enfoques estadísticos para desentrañar las relaciones más sutiles, ya que a veces quedan enmascaradas bajo la influencia de factores que actúan a una mayor escala (p. ej conductividad).

La descripción detallada de la metodología empleada en cada procedimiento puede consultarse en la sección de material y métodos de los artículos en cuestión (Capítulo 5).

22 ARTÍCULOS CIENTÍFICOS Rivularia biasolettiana Fotografía: José Luis Moreno Alcaraz

24 5. ARTÍCULOS CIENTÍFICOS

5.1. Lista de géneros de macroalgas fluviales recolectadas en Castilla-La Mancha durante el período 2001-2014.

Título original: Lista de géneros de macroalgas fluviales recolectadas en Castilla-La Mancha durante el período 2001-2014 Autores: José Luis Moreno y Laura Monteagudo Estado: En revisión Fecha: 2015

RESUMEN

El presente trabajo recoge la lista de géneros de macroalgas fluviales de Castilla-La Mancha, resultado del muestreo de 178 estaciones situadas en ríos y arroyos a lo largo de la región, durante el periodo comprendido entre los años 2001 y 2014. Esta lista incluye un total de 64 géneros, todos ellos capaces de formar colonias o talos de tamaño macroscópico. También se presenta la localización geográfica precisa de los tramos fluviales donde se han recolectado, con indicación de la provincia y cuenca hidrológica a la que pertenecen. Este listado supone un primer acercamiento a la comunidad de macroalgas fluviales de la región, y puede servir como base para futuros estudios de calidad del agua, ficológicos y ecológicos.

26 Lista de géneros de macroalgas fluviales recolectadas en Castilla-La Mancha durante el período 2001-2014 José Luis Moreno y Laura Monteagudo Universidad de Castilla-La Mancha, Centro Regional de Estudios del Agua (CREA) Crta. de Las Peñas km. 3, Albacete 02071, España.

RESUMEN

El presente trabajo recoge la lista de géneros de macroalgas fluviales de Castilla-La Mancha, resultado del muestreo de 178 estaciones situadas en ríos y arroyos a lo largo de la región, durante el periodo comprendido entre los años 2001y 2014. Esta lista incluye un total de 64 géneros, todos ellos capaces de formar colonias o talos de tamaño macroscópico. También se presenta la localización geográfica precisa de los tramos fluviales donde se han recolectado, con indicación de la provincia y cuenca hidrológica a la que pertenecen. Este listado supone un primer acercamiento a la comunidad de macroalgas fluviales de la región, y puede servir como base para futuros estudios de calidad del agua, ficológicos y ecológicos.

1. INTRODUCCION posible papel como bioindicadores de calidad del agua. Las algas que desarrollan Castilla-La Mancha incluye en su territorio talos macroscópicos, denominadas cinco grandes cuencas hidrográficas: Tajo, comúnmente “macroalgas”, forman parte, Guadiana, Júcar, Segura y Guadalquivir. Es junto con briófitos y plantas vasculares, del una de las regiones más extensas del país y indicador biológico denominado cuenta con un escenario hidrogeológico muy “macrófitos” (Anexo V, DMA). Por ello, las diverso, estando bien representados los ríos macroalgas son habitualmente incluidas en silíceos y calcáreos, tramos altos y medios, los índices de macrófitos, índices que son climas húmedos y secos, ríos permanentes y utilizados en la evaluación del estado temporales (Moreno et al., 2006). Ello se ecológico de los ríos mediante la flora traduce en una alta biodiversidad de la flora acuática (p.ej. Moreno et al., 2006a; Flor- acuática en general y, de la flora algal en Arnau et al., 2015). Por tanto, ampliar la particular. Sin embargo, son muy escasos los información disponible sobre la estudios sobre algas que se han llevado a distribución, diversidad y ecología de las cabo en la región. Entre ellos, cabe destacar algas fluviales, es fundamental para poder los trabajos realizados por Aboal (1988a; desarrollar herramientas eficaces en 1988b; 1989a; 1989b) y Sabater et al. materia de bioindicación. (1989), centrados fundamentalmente en la cuenca del Segura, cuya cabecera se localiza Este trabajo pretende contribuir al en la provincia de Albacete. conocimiento de la diversidad de macroalgas en la región, proporcionando Como consecuencia de la entrada en vigor una lista de los géneros detectados durante de la Directiva Marco del Agua un periodo de trece años (2001-2014), junto (2000/60/CE), en adelante DMA, se ha con su localización geográfica precisa. La incrementado el interés por las algas y su información aportada también es de interés

27 a nivel peninsular, dado que son muy presentaron diversas morfologías escasos los estudios extensivos sobre (laminares, globulares, mechones, madejas macroalgas fluviales publicados hasta el filamentosas, masas gelatinosas) y formas momento. Por último, algunas especies con de vida (adheridas o tapizando diferentes mayor interés, bien por su rareza o bien por sustratos, enraizadas en sedimentos, ser primeras citas nacionales y/o flotantes). regionales, se han tratado de forma más Los muestreos se realizaron recorriendo extensa en dos trabajos previos (Moreno et tramos de 100 m en zigzag de orilla a orilla al., 2012; 2013). y río arriba, cubriendo toda la variedad de microhábitats y sustratos presentes. El material recogido fue refrigerado y 2. MATERIAL Y MÉTODOS trasladado al laboratorio para su 2.1. Área de estudio identificación. Las obras empleadas para ello fueron principalmente las siguientes: El régimen climático al que pertenece la Bornet and Flahault (1887), Geitler (1932), región es el mediterráneo-continentalizado, Bourrelly (1957; 1990), Desikachary marcado por inviernos fríos y veranos (1959), Aboal (1988a; 1988b; 1989a; calurosos con fuertes oscilaciones térmicas. 1989b), Komárek and Anagnostidis (2005) En cuanto a las precipitaciones, Eloranta and Kwandrans (2007), Ettl and encontramos desde zonas áridas como la Gartner (2009) y Eloranta (2011). El nivel llanura manchega y el sureste de la región taxonómico alcanzado ha sido el de género, con valores inferiores a los 300mm al año, ya que es el alcanzado habitualmente en la hasta zonas montañosas donde se superan aplicación de los índices de macrófitos los 1000mm (Fernández, 2000). utilizados en la evaluación del estado Geológicamente, se diferencian tres grandes ecológico de los ríos. Por otra parte, zonas: la zona oeste de naturaleza silícea, actualmente existe una gran inestabilidad formada por esquistos, gneises, pizarras y taxonómica en numerosos grupos de algas a granitos; la llanura central sedimentaria, nivel específico (e incluso genérico), debido abundante en calizas, arcillas y yesos; y la a la reciente aplicación de nuevas técnicas zona este, caracterizada por la presencia de de análisis genéticos. rocas calcáreas como calizas, dolomías, Posteriormente, el material fue fijado con margas, arcillas y conglomerados (Porras formaldehido 3% para su conservación. En Martín et al., 1985). Esta diversidad los casos de interés, también se realizaron climática, geológica y geográfica ha preparaciones permanentes con glicero- originado también una red hidrológica gelatina, pliegues o sobres en seco, y viales formada por ríos y arroyos de diversas con gel de sílice. La colección se encuentra tipologías (Moreno et al., 2006b). almacenada en el Centro Regional de 2.2. Diseño del muestreo y análisis del Estudios del Agua (Universidad de Castilla- material La Mancha), Laboratorio de Limnología Aplicada e Hidrobiología, Albacete. El presente estudio fue llevado a cabo en ríos y arroyos de Castilla-La Mancha, entre 2.3. Información geográfica los años 2001-2014. En total, se La localización de los puntos de muestreo se recolectaron macroalgas en 179 puntos de realizó en el entorno ArcGis 9, generando un muestreo (Figura 1), abarcando todo el mapa con cuadrícula superpuesta de gradiente climático y litológico presente en coordenadas UTM. El listado de géneros la región. En cada estación de muestreo, se incluye información sobre las provincias de recogieron a mano o con navaja las la región donde fueron recolectados, macroalgas detectadas, es decir, aquellas mientras que en el listado de puntos de algas capaces de desarrollar talos de tamaño muestreo se incluye el municipio y la cuenca macroscópico, visibles simple vista. Estas hidrológica a la que pertenecen.

28 Figura 1. Mapa de la zona de estudio dividida en cuadrículas UTM. Las líneas blancas marcan los límites provinciales (Ab, Albacete; Cu, Cuenca; CR, Ciudad Real; Gu, Guadalajara; To, Toledo).

3. RESULTADOS

El listado reúne un total de 64 géneros de macroalgas fluviales detectados en Castilla-La Mancha durante el periodo de estudio. Entre ellos aparecen; 27 representantes del filo Cyanophyta (=Cyanobacteria) (algas verdeazules); 1 del filo Bacillariophyta (diatomeas); 3 del filo Ochrophyta (algas ocres); 11 del filo Rhodophyta (algas rojas); 15 del filo Chlorophyta y 7 del filo Charophyta ambos pertenecientes al grupo de las llamadas algas verdes. Según la nomenclatura actualizada proporcionada por el proyecto www.algaebase.org.

Anabaena Bory de Saint-Vincent ex Bornet & Flahault - Ab, CR, To - 13, 17, 18, 19, 20, 21

Audouinella Bory de Saint-Vincent - Ab, CR, Cu, Gu - 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16

Bangia Lyngbye - Cu, Gu - 134, 140, 166, 173

29 Batrachospermum Roth - Ab, CR, Cu, Gu - 1, 3, 4, 5, 6, 7, 8, 13, 14, 16, 30, 40, 41, 42, 47, 56, 57, 59, 68, 81, 86, 87, 88, 89, 90, 91, 92, 99, 100, 103, 105, 109, 112, 118, 122, 125, 127, 128, 134, 177

Bulbochaete C.Agardh - Cu, Gu - 111, 128

Calothrix C.Agardh ex É.Bornet & C.Flahault – CR - 18

Chaetophora F.Schrank - Ab, CR, Cu, Gu - 112, 119, 86, 15, 111, 8, 4, 68, 42, 134, 5, 105, 11, 124, 155, 66, 82, 102, 108, 87, 89, 151, 79, 18, 148

Chamaesiphon A.Braun – Gu - 146, 148, 151, 153, 178, 179

Chara Linnaeus - Ab, CR, Cu, Gu, To - 3, 4, 6, 7, 13, 14, 15, 16, 17, 22, 29, 36, 42, 44, 45, 46, 53, 54, 56, 59, 64, 65, 77, 79, 85, 86, 87, 88, 102, 105, 108, 109, 112, 113, 116, 118, 119, 125, 128, 129, 134, 142, 144, 147, 155, 161, 166, 169, 173

Chroodactylon Hansgirg – Ab - 16

Chroothece Hansgirg - Ab, Gu - 7, 14, 118

Cladophora Kützing - Ab, CR, Cu, Gu, To - 1, 3, 4, 6, 7, 9, 10, 12, 13, 14, 16, 18, 21, 22, 23, 24, 25, 28, 30, 33, 35, 36, 37, 39, 40, 41, 44, 46, 50, 51, 52, 53, 54, 55, 56, 58, 61, 62, 63, 64, 65, 66, 68, 75, 77, 78, 79, 83, 85, 88, 89, 90, 91, 92, 93, 94, 95, 97, 98, 100, 101, 102, 106, 107, 108, 109, 110, 112, 113, 116, 117, 118, 120, 121, 123, 124, 125, 127, 128, 133, 134, 135, 136, 137, 138, 139, 140, 141, 146, 150, 154, 155, 156, 157, 160, 161, 163, 166, 168, 169, 171, 173, 174

Coleodesmium A.Borzì ex L.Geitler - CR, Gu - 2, 141, 148, 153

Compsopogon Montagne – Ab - 7

Cylindrospermum - F.T.Kützing ex É.Bornet & C.Flahault - Ab, CR - 7, 73, 108

Draparnaldia Bory de Saint-Vincent - Ab, CR, Cu, Gu - 2, 10, 11, 42, 66, 81, 82, 105, 146, 148, 149, 151, 152, 179

Enteromorpha Link – Ab - 107

Geitlerinema (Anagnostidis & Komárek) Anagnostidis – Ab - 54

Gloeocapsa Kützing - Ab, Gu - 94, 154

Gloeotrichia J.Agardh ex Bornet & Flahault – CR - 73

Heteroleibleinia (L.Geitler) L.Hoffmann - CR, Gu - 83, 179

Hildenbrandia Nardo - Ab, Cu, Gu, To - 4, 7, 57, 116, 133, 140, 146

30 Homeothrix (Thuret ex Bornet & Flahault) Kirchner – Gu - 179

Hyalotheca Ehrenberg ex Ralfs – Cu - 123

Hydrococcus Kützing – Ab - 7

Hydrocoleum Kützing ex Gomont – Ab – 87

Hydrodictyon Roth – CR – To - 18, 19, 20, 27, 32, 79, 84, 114, 132

Hydrurus C.Agardh – Gu - 141, 179

Klebsormidium P.C.Silva, Mattox & W.H.Blackwell – Ab - 89

Kyliniella Skuja – Ab - 108

Lemanea Bory - CR, Cu, Gu, To - 2, 5, 11, 66, 71, 79, 80, 82, 116, 125, 133, 141, 143, 148, 151, 153, 155, 178, 179

Leptolyngbya Anagnostidis & Komárek – To - 133, 161

Lyngbya C.Agardh ex Gomont - Ab, To - 13, 164, 168

Melosira C.Agardh - Ab, CR, Cu, Gu - 10, 18, 39, 55, 95, 107, 140, 149, 153, 154, 177

Microcoleus Desmazières ex Gomont - Cu, Gu, To - 19, 134, 140, 143

Microspora Thuret - Ab, CR, Cu, Gu, To - 1, 8, 14, 16, 17, 19, 40, 41, 42, 84, 85, 89, 114, 127, 128, 133, 172

Monostroma Thuret - Cu, Gu, CR - 83, 58, 124

Mougeotia C.Agardh - Ab, CR, Cu, Gu, To - 13, 14, 27, 34, 47, 54, 56, 66, 76, 145, 149, 153, 166, 169, 175

Nitella C.Agardh - Ab, CR - 13, 17, 69, 70, 76

Nostoc Vaucher ex Bornet & Flahault - Ab, CR, Cu, Gu, To - 1, 2, 3, 5, 6, 7, 11, 12, 15, 18, 19, 22, 28, 51, 54, 58, 62, 64, 66, 79, 84, 87, 89, 90, 95, 97, 99, 104, 108, 109, 116, 123, 125, 129, 141, 148, 151, 153, 166, 169, 178, 179

Nostochopsis H.C.Wood ex É.Bornet & C.Flahault – CR - 18

Oedogonium Link ex Hirn - Ab, CR, Cu, Gu, To - 5, 7, 8, 9, 11, 12, 13, 14, 15, 16, 17, 18, 19, 21, 34, 41, 42, 47, 50, 51, 56, 57, 64, 66, 70, 71, 76, 78, 79, 80, 82, 84, 85, 87, 88, 89, 91, 110, 111, 123, 128, 135, 139, 154, 155, 161, 163, 166, 168, 169, 177

Oocardium Nägeli – Ab - 14

31 Oscillatoria Vaucher ex Gomont - Ab, CR, Cu, Gu, To - 25, 44, 46, 72, 93, 123, 150, 164, 173, 175

Paralemanea (P.C.Silva) Vis & Sheath – To - 161

Phormidium Kützing ex Gomont - Ab, CR, Cu, Gu, To - 1, 3, 4, 6, 7, 10, 11, 12, 13, 14, 18, 21, 22, 27, 28, 29, 30, 31, 32, 33, 34, 36, 37, 38, 39, 40, 41, 42, 43, 48, 50, 51, 52, 54, 56, 57, 64, 67, 68, 70, 72, 81, 83, 87, 88, 89, 91, 92, 94, 96, 97, 98, 100, 102, 103, 106, 108, 116, 118, 125, 133, 139, 140, 142, 143, 148, 153, 155, 159, 161, 164, 177, 178, 179

Plectonema Thuret ex Gomont - Ab, Cu, Gu - 1, 13, 14, 30, 42, 56, 64, 87, 89, 93, 100, 104, 175

Pseudanabaena Lauterborn - Ab, Cu, Gu, To - 16, 21, 47, 159

Rhizoclonium Kützing - Ab, CR, Cu, Gu, To - 9, 21, 23, 25, 30, 32, 37, 39, 110, 114, 115, 120, 123, 126, 132, 135, 137, 145, 147, 157, 168

Rivularia C.Agardh ex É.Bornet & C.Flahault - Ab, CR, Cu, Gu - 3, 5, 13, 14, 34, 42, 43, 44, 45, 47, 53, 59, 60, 66, 68, 85, 86, 87, 88, 89, 91, 92, 99, 101, 102, 103, 104, 105, 108, 109, 112, 128, 129, 134, 142, 155, 166, 169, 173, 175

Schizothrix F.T.Kützing ex M.Gomont - Ab, Cu, Gu - 14, 15, 16, 56, 57, 88, 90, 92, 96, 100, 109, 134, 175

Scytonema C.Agardh ex É.Bornet & C.Flahault - Ab, CR, Cu, Gu - 2, 6, 34, 42, 53, 85, 88, 91, 102, 104, 128, 141, 153, 170

Spirogyra Link in Nees - Ab, CR, Cu, Gu, To - 1, 2, 3, 5, 7, 9, 10, 11, 12, 13, 14, 15, 17, 18, 19, 20, 21, 23, 24, 26, 27, 32, 33, 34, 36, 38, 39, 41, 42, 43, 45, 47, 51, 54, 56, 57, 58, 59, 60, 71, 75, 76, 77, 78, 79, 80, 81, 82, 84, 87, 90, 91, 93, 96, 97, 99, 100, 101, 102, 103, 104, 105, 108, 109, 113, 114, 115, 116, 119, 123, 124, 128, 131, 133, 135, 141, 142, 145, 147, 148, 149, 150, 151, 152, 153, 155, 161, 162, 166, 168, 169, 172, 173, 177, 178

Stigeoclonium Kützing - Ab, CR, Gu, To - 12, 16, 18, 42, 78, 95, 107, 116, 133, 135, 168, 171, 172, 178

Tetraspora Link ex Desvaux - Ab, CR, Cu, Gu, To - 18, 41, 68, 70, 71, 79, 82, 97, 149, 151, 162, 179

Tetrasporidium Möbius - CR, Gu - 2, 18, 70, 71, 149, 151

32 Thorea Bory – Ab - 7, 49

Tolypella (A.Braun) A.Braun – Gu - 142

Tolypothrix F.T.Kützing ex É.Bornet & C.Flahault - Ab, Cu, Gu - 6, 7, 11, 13, 42, 54, 56, 66, 86, 88, 92, 93, 100, 101, 103, 108, 116, 118, 122

Tribonema Derbès & Solier - Gu, To - 19, 21, 149, 153

Trichocoleus K.Anagnostidis – Gu - 175

Ulothrix Kützing - Ab, Cu, Gu - 1, 4, 11, 95, 148, 153

Vaucheria A.P.de Candolle - Ab, CR, Cu, Gu, To - 3, 6, 7, 9, 10, 12, 29, 30, 33, 36, 39, 41, 51, 52, 54, 55, 58, 62, 63, 68, 75, 80, 83, 89, 93, 94, 95, 97, 106, 110, 114, 120, 126, 127, 130, 138, 140, 143, 154, 157, 162, 176

Zygnema C.Agardh - Ab, CR, Cu, Gu - 5, 13, 14, 15, 17, 18, 26, 27, 32, 41, 42, 47, 53, 54, 56, 59, 60, 66, 71, 73, 76, 80, 82, 85, 87, 88, 91, 97, 99, 101, 103, 104, 105, 108, 109, 111, 113, 116, 119, 128, 141, 142, 144, 147, 153, 166, 169, 172

Agradecimientos

Este estudio ha sido realizado en el Centro Regional de Estudios del Agua (Universidad de Castilla-La Mancha, Albacete) y financiado por los siguientes convenios y proyectos concedidos por el gobierno regional (Junta de Comunidades de Castilla-La Mancha): PREG01-0016, PREG06-027, PO1109-0190-8090 y PPII10-0271-1349.

BIBLIOGRAFÍA

Aboal, M. 1988a. Aportación al conocimiento Aboal, M. 1989b. Aportación al conocimiento de las algas epicontinentales del de las algas epicontinentales del S.E. sudeste de España. III: Cianofíceas de España. V. Xantofíceas (Cyanophyceae Schaffner 1909). (Xanthophyceae P. Allorge ex Fritsch, Anales del Jardín Botánico de Madrid 1935). Boletim da Sociedade 45(1): 1-46. Broteriana 62: 239-248. Aboal, M. 1988b. Aportación al conocimiento Bornet, E. et Flahault, C. 1887. Revision des de las algas epicontinentales del S.E. Nostocacées hétérocystées contenues de España. VII. Clorofíceas dans les principaux herbiers de (Chlorophyceae Wille in Warming France.III. Ann. Sci. Nat. Bot., VII. 1884). Candollea 43: 521-548. Bourrelly, P. 1990. Les algues d'eau douce. Aboal, M. 1989a. Contribución al Initiation à la systématique. Tome I: conocimiento de las algas Les algues vertes. Société Nouvelle epicontinentales del SE de España II: des Éditions Boubée. Paris. Los rodófitos (Rhodophyceae). Lazaroa 11: 115-122.

33 Desikachary, T.V. 1959: Cyanophyta. der Schweiz.: 1-1196. Koeltz Scientific Monographs on Algae. Indian Council Books. Leipzig. of Agricultural Research, New Delhi. Komárek, J. and Anagnostidis, K. 2005. Eloranta, P. 2011. Rhodophyta and Süßwasserflora von Mitteleuropa 19/2. Phaeophyceae, Freshwater Flora of Cyanoprokaryota 2. Teil: Central Europe. Spektrum Oscillatoriales. Spektrum, Heidelberg. Akademischer Verlag, Heidelberg. Moreno, J.L., Navarro, C. and De las Heras, J. Eloranta, P. and Kwandrans, J. 2007. 2006. Abiotic ecotypes in south-central Freshwater red algae (Rhodophyta) : Spanish rivers: Reference conditions identification guide to European taxa, and pollution. Environmental particularly to those in Finland. pollution, 143, 388-396. Botanical Museum Finnish Museum Moreno, J.L., Aboal, M. and Monteagudo, L. of Natural History, Helsinki. 2012. On the presence of Nostochopsis Ettl, H. and Gartner, G. 2009. Süsswasserflora lobata Wood ex Bornet et Flahault in von Mitteleuropa 10. Chlorophyta.- 2. Spain: morphological, ecological and Tetrasporales, Chlorococcales, biogeographical aspects. Nova Gloeodendrales. Spektrum Hedwigia 95, 373–390. Akademischer Verlag. Heidelberg. Moreno, J.L., Monteagudo, L. and Aboal, M., Fernández, F. 2000. Los condicionantes 2013. Morphological description and climáticos del paisaje. En: Guía de Los ecology of some rare macroalgae in Espacios Naturales de Castilla-La south-central Spanish rivers (Castilla- Mancha. Junta de Comunidades de La Mancha Region). An. Jardín Castilla-La Mancha, Toledo. Botánico Madr. 70, 81–90. Flor-Arnau, N., Real, M., González, G., Porras Martín, J., Ruiz, C., Fernández, J.A., Cambra, J., Moreno, J.L., Solà, C. y Gómez de las Heras J. y Fabregat, V. Munné, A. 2015. Índice de Macrófitos 1985. Síntesis Hidrogeológica de Fluviales (IMF), una nueva Castilla-La Mancha. Instituto herramienta para evaluar el estado Geológico y Minero de España, ecológico de los ríos mediterráneos. Madrid. Limnetica 34 (1), 95-114. Sabater, S., Aboal, M. and Cambra, J. 1989. Geitler, L. 1932. Cyanophyceae von Europa. New Rhodophyceae records for the NE In: L. Rabenhorst (ed.), Kryptogamen- and SE Spanish continental waters. Flora von Deutschland, Österreich und Limnetica 5, 93–100.

ANEXO

Listado y localización geográfica de los puntos o estaciones de muestreo, expresada de la siguiente manera: Número de estación: Nombre del cuerpo de agua (Cuenca hidrográfica) Localidad, Municipio; Cuadrícula UTM 10x10 km; Altitud (m)

1: Río Horcajo (Guadalquivir) en El Horcajo, Alcaraz; 30SWH48; 1017 m 2: Río Estena (Guadiana) en Rocigalgo, Hontanar; 30SUJ68; 816 m 3: Río Cabriel (Júcar) en Salvacañete; 30TXK24; 1164 m 4: Río Guadazaón (Júcar) en Yemeda; 30SXK00; 852 m 5: Río Huécar (Júcar) en Palomera; 30TWK83; 1104 m 6: Río Júcar (Júcar) en Cuasiermas, Albacete; 30SXJ03; 623 m 7: Río Júcar (Júcar) en Las Mariquillas, Valdeganga; 30SXJ03; 607 m

34 8: Río Pesebre (Júcar) en Pesebre, Peñascosa; 30SWH58; 1148 m 9: Río Valdemembra (Júcar) en Villanueva de la Jara; 30SWJ96; 743 m 10: Río Bornova (Tajo) en San Andrés de Congosto; 30TVL93; 843 m 11: Río Sorbe (Tajo) en Valverde de los Arroyos, La Huerce; 30TVL85; 1060 m 12: Arroyo Alarconcillo (Guadiana) 200m aguas abajo del puente a Rochafrida, Ossa de Montiel; 30SWJ10; 865 m 13: Río Vado Blanco (Guadiana) en Osero, Villahermosa; 30SWJ10; 880 m 14: Arroyo Fuente de la Parra (Segura) en Royo-Odrea, Ayna; 30SWH76; 851 m 15: Río Zumeta (Segura) en Las Juntas, Yeste; 30SWH42; 749 m 16: Río Montemayor (Júcar) en los Batanes, Peñascosa; 30SWH68; 1036 m 17: Río Fresnedas (Guadalquivir) en Cortijo de Cobatillas, Calzada de Calatrava; 30SVH26; 540 m 18: Arroyo Nava del Rey (Guadalquivir) en Huertezuelas, Calzada de Calatrava; 30SVH26; 591 m 19: Río Tietar (Tajo) aguas arriba del embalse Rosarito, Oropesa (Dehesa del Verdugal); 30TUK14; 317 m 20: Río Tiétar (Tajo) en Parrillas; 30TUK24; 371 m 21: Río Tiétar (Tajo) en Palacio Rosarito, Oropesa (Dehesa del Verdugal); 30TTK94; 289 m 22: Arroyo de las Ánimas (Guadiana) en Sotuélamos, El Bonillo; 30SWJ31; 882 m 23: Río Azuer (Guadiana) en Carrizosa, Montiel; 30SVH99; 811 m 24: Río Arroyo del cortijo (Bañuelos) (Guadiana) en Cortijo de Abajo, Los Cortijos; 30SVJ15; 739 m 25: Río Bullaque (Guadiana) en Casa de Durán, Piedrabuena; 30SUJ91; 539 m 26: Río Bullaque (Guadiana) en El Torno, Porzuna; 30SUJ94; 597 m 27: Río Estena (Guadiana) en Boquerón del Estena, Navas de Estena; 30SUJ67; 630 m 28: Río Córcoles (Guadiana) en Cortijo del Santo, Munera; 30SWJ41; 928 m 29: Río Gigüela (Guadiana) en Naharros; 30TWK43; 917 m 30: Río Gigüela (Guadiana) en Segóbriga, Saelices; 30SWK11; 778 m 31: Río Gigüela (Guadiana) en Villar del Horno; 30TWK53; 996 m 32: Río Guadiana (Guadiana) en Las Hoces-Casa Majalahoz, Puebla de Don Rodrigo; 30SUJ53; 446 m 33: Río Jabalón (Guadiana) en Montiel; 30SWH18; 869 m 34: Río Pinilla (Guadiana) en Villahermosa; 30SWJ10; 890 m 35: Río Riánsares (Guadiana) en Huelves; 30TWK03; 810 m 36: Río Riánsares (Guadiana) en Paredes; 30TWK13; 852 m 37: Río Tirteafuera (Guadiana) en Abenójar; 30SUJ70; 589 m 38: Río Valdeazoques (Guadiana) en Estación de Chillón, Chillón; 30SUH38; 376 m 39: Río Záncara (Guadiana) en Zafra de Záncara; 30SWK31; 899 m

35 40: Río Záncara (Guadiana) en Zafra de Záncara; 30SWK31; 829 m 41: Río Arquillo (Júcar) en Laguna del Arquillo, Masegoso; 30SWH58; 988 m 42: Río Arquillo (Júcar) en Peñascosa; 30SWH57; 1230 m 43: Río Cabriel (Júcar) en Balneario Fuente Podrida, Villatoya; 30SXJ45; 400 m 44: Río Cabriel (Júcar) en Cardenete; 30SXK10; 812 m 45: Río Guadazaón (Júcar) en Valdemoro de la Sierra; 30TXK04; 1121 m 46: Río Jardín (Júcar) en merendero El Zarzalejo, Casas de Lázaro; 30SWH69; 849 m 47: Río Júcar (Júcar) La Losilla, Alarcón; 30SWJ77; 729 m 48: Río Júcar (Júcar) en Albaladejito, Cuenca (Los Llecos); 30TWK63; 905 m 49: Río Júcar (Júcar) en la piscifactoría de Bolinches, Valdeganga; 30SXJ13; 602 m 50: Río Júcar (Júcar) en La Parra de las Vegas; 30SWK61; 824 m 51: Río Júcar (Júcar) en Villalgordo del Júcar; 30SWJ75; 670 m 52: Río Júcar (Júcar) en Casas del Palancar, Villar de Olalla; 30SWK62; 868 m 53: Río Júcar (Júcar) en Alarcón; 30SWJ77; 749 m 54: Río Júcar (Júcar) en Fuensanta; 30SWJ84; 659 m 55: Río Júcar (Júcar) en Valdemeca, Huélamo; 30TXK06; 1217 m 56: Río Júcar (Júcar) en Presa La Losa, La Losa; 30SWJ75; 680 m 57: Río Júcar (Júcar) en Molino de los Frailes, Valdeganga; 30SXJ03; 609 m 58: Río Júcar (Júcar) en Molino Juan Romero, Cuenca (Los Llecos); 30TWK95; 1178 m 59: Río Júcar (Júcar) en La Noguera (Barranco Boca de la Hoz), Tébar; 30SWJ87; 715 m 60: Río Júcar (Júcar) en Rambla Rubielos, Picazo; 30SWJ76; 695 m 61: Río Júcar (Júcar) en Villalba de la Sierra, Cuenca (Los Llecos); 30TWK74; 929 m 62: Río Júcar (Júcar) en Venta de Juan Romero, Huélamo; 30TWK95; 1180 m 63: Río Laguna (Júcar) en Huerta del Marquesado; 30TXK14; 1250 m 64: Río Masegoso (Júcar) en Masegoso; 30SWH58; 1130 m 65: Río Ojos de Moya (Júcar) en Garaballa; 30SXK30; 906 m 66: Río Pedregoso (Júcar) en Molino de Juan Romero, Beamud; 30TWK95; 1194 m 67: Río Valdemembra (Júcar) en Tarazona de la Mancha; 30SWJ94; 682 m 68: Río Vencherque (Júcar) en Villar del Humo; 30SXK11; 970 m 69: Arroyo Aliseda (Guadalquivir) en Carretera a Solana del Pino, Fuencaliente; 30SUH86; 701 m 70: Arroyo de la Aliseda (Guadalquivir) en Ventillas (Pueblo), Fuencaliente; 30SUH86; 684 m 71: Río Cereceda (Guadalquivir) en Cueva Batanera, Fuencaliente; 30SUH85; 822 m 72: Arroyo Cereceda (Guadalquivir) en Fuencaliente; 30SUH85; 663 m 73: Arroyo Dañador (Guadalquivir) en Cortijo de la Cañada, Villamanrique; 30SWH05; 761 m

36 74: Arroyo Caballeros del Escorial (Guadalquivir) en Casas del Escorial, Brazatortas; 30SUH76; 749 m 75: Río Guadalmena (Guadalquivir) en Villapalacios; 30SWH27; 754 m 76: Arroyo de las Navas (Guadalquivir) en Cortijo de las Navas, Solana del Pino; 30SUH95; 671 m 77: Arroyo Ontavia (Guadalquivir) en Hoz de Terrinches, Montiel; 30SWH16; 687 m 78: Río Pradillo (Guadalquivir) en Fuencaliente; 30SUH84; 552 m 79: Río Rigüelo (Guadalquivir) en San Lorenzo de Calatrava; 30SVH25; 524 m 80: Río Robledillo (Guadalquivir) en Solanilla del Tamaral, Mestanza; 30SVH15; 421 m 81: Río Robledillo (Guadalquivir) en Solanilla del Tamaral, Mestanza; 30SVH15; 459 m 82: Arroyo Ruicastaño (Guadalquivir) en Viso del Marqués; 30SVH46; 740 m 83: Río Villanueva (Guadalquivir) en Barranco Zárcenas, Villanueva de la Fuente; 30SWH27; 820 m 84: Río Yeguas (Guadalquivir) en Fuencaliente; 30SUH84; 564 m 85: Río Bogarra (Segura) en Batán de Bogarra, Bogarra; 30SWH67; 844 m 86: Arroyo Fuenfría (Segura) en El Encebrico (Chorraeros), Paterna del Madera; 30SWH56; 1431 m 87: Arroyo de las Hoyas (Segura) en El Puerto (Umbría), Paterna del Madera; 30SWH56; 1281 m 88: Arroyo Madera (Segura) en Arguellite, Yeste; 30SWH54; 706 m 89: Río Madera-Endrinales (Segura) en área recreativa Fuentelisa, Paterna del Madera; 30SWH57; 1044 m 90: Río Mundo (Segura) en el aforo del molino, aguas arriba de la piscifactoría, Riópar; 30SWH56; 937 m 91: Río Mundo (Segura) en Aguas abajo del embalse, Isso; 30SWH96; 470 m 92: Río Mundo (Segura) en Puente de Isso, Hellín; 30SXH05; 438 m 93: Río Mundo (Segura) en Liétor; 30SWH96; 515 m 94: Río Mundo (Segura) en Agramón; 30SXH15; 360 m 95: Río Mundo (Segura) en Mesones (aguas abajo del camping), Molinicos; 30SWH56; 886 m 96: Río Mundo (Segura) en Las Minas, Hellín; 30SXH14; 307 m 97: Río Mundo (Segura) en Tavizna, Hellín; 30SXH05; 402 m 98: Río Segura (Segura) en Calasparra (Est.de Las Minas), Hellín; 30SXH14; 301 m 99: Río Segura (Segura) en Cortijo de Abajo (aguas abajo del aforo), Hellín; 30SXH04; 349 m 100: Río Segura (Segura) en Salida Embalse Cenajo, Hellín; 30SXH04; 350 m 101: Río Segura (Segura) en El Hondo (Maeso), Hellín; 30SXH14; 334 m 102: Río Segura (Segura) en La Donal, Yeste; 30SWH53; 647 m 103: Río Segura (Segura) en Puente de Híjar, Férez; 30SWH95; 460 m

37 104: Río Segura (Segura) en Salmerón; 30SXH14; 310 m 105: Río Taibilla (Segura) en Nerpio; 30SWH52; 1092 m 106: Río Taibilla (Segura) en Molino la Tercia, Nerpio; 30SWH52; 1159 m 107: Arroyo Tobarra (Segura) en Saladar de Cordovilla, Hellín; 30SXH26; 500 m 108: Río Tus (Segura) en Vado de Tus, Yeste; 30SWH44; 835 m 109: Río Vadillos (Segura) en Yeguarizas, Bogarra; 30SWH66; 834 m 110: Río de la Vega (Segura) en El Laminador, Riópar; 30SWH56; 926 m 111: Arroyo Cedrón (Tajo) en Ablanque; 30TWL63; 1113 m 112: Río Ablanquejo (Tajo) en Ablanque; 30TWL62; 1008 m 113: Río Ablanquejo (Tajo) en Huertahernando; 30TWL52; 914 m 114: Río Alberche (Tajo) en Escalona; 30TUK74; 415 m 115: Río Alberche (Tajo) en Talavera de la Reina; 30SUK42; 367 m 116: Río Bornova (Tajo) en Molino de la Oportuna, Villares de Jadraque; 30TVL95; 971 m 117: Río Bornova (Tajo) en San Andrés de Congosto, La Toba; 30TVL93; 831 m 118: Río Bullones (Tajo) en Escalera; 30TWL81; 1032 m 119: Río Cañamares (Tajo) en Naharros, La Minosa; 30TWL05; 995 m 120: Río Cabrillas (Tajo) en Megina; 30TWK99; 1192 m 121: Arroyo Cedron (Tajo) en La Guardia, Corral de Almaguer; 30SVK70; 694 m 122: Río Cuervo (Tajo) en el nacimiento del Río Cuervo, Cuenca (Los Llecos); 30TWK97; 1428 m 123: Río Cuervo (Tajo) en Puente de Vadillos, Cañizares; 30TWK78; 943 m 124: Río Dulce (Tajo) en Jodra de Pinares; 30TWL34; 1079 m 125: Río Dulce (Tajo) en La Pelegrina, Sigüenza; 30TWL34; 980 m 126: Río Dulce (Tajo) en Villaseca de Henares, Villaseca de Henares; 30TWL13; 831 m 127: Río Escabas (Tajo) en el Estrecho de Priego, Priego; 30TWK67; 862 m 128: Río Escabas (Tajo) en Fuente del Cayo, Cuenca (Los Llecos); 30TWK77; 971 m 129: Río Gallo (Tajo) en Castilnuevo; 30TWL91; 1080 m 130: Río Gallo (Tajo) en Ermita de la Hoz, Corduente; 30TWL82; 1017 m 131: Río Guadarrama (Tajo) en Bargas, Bargas; 30SVK02; 452 m 132: Río Guadarrama (Tajo) en Chozas de Canales; 30TVK14; 515 m 133: Río Gévalo (Tajo) en Buenasbodas, Sevilleja de la Jara; 30SUJ39; 569 m 134: Río Guadiela (Tajo) en Canalejas del Arroyo; 30TWK47; 720 m 135: Río Henares (Tajo) en Azuqueca de Henares; 30TVK78; 604 m 136: Río Henares (Tajo) en Fontanar, Guadalajara (Monte el Villar); 30TVL80; 663 m 137: Río Henares (Tajo) en Jadraque; 30TWL03; 802 m 138: Río Henares (Tajo) en Mojares, Sigüenza; 30TWL35; 1056 m 139: Río Henares (Tajo) aguas abajo de la depuradora Sigüenza, Sigüenza; 30TWL24; 986 m

38 140: Río Henares (Tajo) en Tórtola de Henares; 30TVL80; 667 m 141: Río de la Hoz (Tajo) en Camping de la Venta (ruinas), Cantalojas; 30TVL76; 1329 m 142: Río Hoz Seca (Tajo) en Orea; 30TXK08; 1507 m 143: Río Jarama (Tajo) en El Cardoso de la Sierra; 30TVL54; 1240 m 144: Río Jarama (Tajo) en Casa de Uceda; 30TVL62; 744 m 145: Río Jarama (Tajo) en Dehesa Nueva del Rey, Seseña; 30TVK43; 484 m 146: Río Jarama (Tajo) en Puente Medieval, Valdesotos; 30TVL73; 803 m 147: Río Jarama (Tajo) en Puebla de Valles, Tortuero; 30TVL73; 802 m 148: Río Lillas (Tajo) en Barranco de Carretas, Cantalojas; 30TVL66; 1429 m 149: Río Lillas (Tajo) en parking del Parque Tejera Negra, Cantalojas; 30TVL76; 1404 m 150: Río Pusa (Tajo) en Bernuy, Malpica de Tajo; 30SUK61; 386 m 151: Río Sonsaz (Tajo) en área recreativa del Parque Tejera Negra, Cantalojas; 30TVL75; 1446 m 152: Río Sonsaz (Tajo) en Cabezo de la Mesta, Cantalojas; 30TVL75; 1657 m 153: Río Sonsaz (Tajo) en Sierra del Ocejón, Cantalojas; 30TVL75; 1541 m 154: Río Sorbe (Tajo) aguas arriba de la unión del Valdecimbrio, Galve de Sorbe; 30TVL86; 1254 m 155: Río Sorbe (Tajo) en Beleña del Sorbe, Cogolludo; 30TVL83; 803 m 156: Río Sorbe (Tajo) en Muriel, Tamajón; 30TVL83; 857 m 157: Río Tajuña (Tajo) en Abánades; 30TWL42; 1026 m 158: Río Tajuña (Tajo) en Archilla, Abánades; 30TWL00; 768 m 159: Río Tajuña (Tajo) en Loranca de Tajuna; 30TVK97; 700 m 160: Río Tajuña (Tajo) en Masegoso de Tajuna, Cifuentes; 30TWL21; 884 m 161: Río Tiétar (Tajo) en La Iglesuela; 30TUK55; 442 m 162: Río Tajo (Tajo) en Albáreal del Tajo, Polan; 30SUK91; 432 m 163: Río Tajo (Tajo) en Algarga, Illana; 30TVK94; 559 m 164: Río Tajo (Tajo) en Santuario Ntra Sra Carrasca, Villarrubia de Santiago; 30TVK63; 528 m 165: Río Tajo (Tajo) en Dehesa Nueva del Rey, Seseña; 30TVK43; 487 m 166: Río Tajo (Tajo) en Los Cuchillos, Peñalén; 30TWL80; 1099 m 167: Río Tajo (Tajo) en Malpica del Tajo; 30SUK61; 390 m 168: Río Tajo (Tajo) en Mocejon, Toledo; 30SVK21; 465 m 169: Río Tajo (Tajo) en Peralejos de las Truchas, Peralejos de las Truchas; 30TWK89; 1124 m 170: Río Tajo (Tajo) en Sacedón; 30TWK18; 690 m 171: Río Tajo (Tajo) en Las Herencias, Talavera de la Reina; 30SUK31; 357 m 172: Río Tajo (Tajo) en Trillo; 30TWL30; 749 m 173: Río Tajo (Tajo) en Valtablado del Rio; 30TWL50; 777 m

39 174: Río Tajo (Tajo) en Las Vegas y San Antonio, La Pueblanueva; 30SUK52; 377 m 175: Río Tajo (Tajo) en Zorita de los Canes; 30TWK06; 595 m 176: Río Trabaque (Tajo) en Arcos de la Sierra; 30TWK76; 980 m 177: Arroyo Valdecimbrio (Tajo) en su desembocadura en el Sorbe, Galve de Sorbe; 30TVL86; 1272 m 178: Río Zarzas (Tajo) en la entrada del Parque Tejera Negra, Cantalojas; 30TVL76; 1381 m 179: Río Zarzas (Tajo) en Barranco de las Lagunas, Cantalojas; 30TVL66; 1564 m

40 5.2. Descripción morfológica y ecología de algunas algas consideradas ‘raras’ en los ríos de Castilla – La Mancha.

Título original: Morphological description and ecology of some rare macroalgae in south-central Spanish rivers (Castilla-La Mancha Region) Autores: José Luis Moreno, Laura Monteagudo y Marina Aboal Publicación: Publicado en Anales del Jardín Botánico de Madrid, 70(1). Impact Factor: 0.912 DOI: 10.3989/ajbm.2323 Fecha 2013

RESUMEN

El conocimiento sobre la biodiversidad algal de la región de Castilla-La Mancha, situada en la zona centro-sur de España, es escaso en comparación con el de otras regiones peninsulares. Sin embargo, la aplicación de la Directiva Marco del Agua (2000/60/CE) y la evaluación del estado ecológico de los ecosistemas acuáticos europeos, ha traído consigo un aumento en la frecuencia e intensidad en el muestreo de ríos, lagos y humedales. De esta forma, durante los últimos años, se han producido nuevos hallazgos en la región que han permitido ampliar el conocimiento de la biodiversidad de algas así como de la distribución geográfica de muchas de sus especies. En este trabajo se describen las condiciones ecológicas y las características morfológicas de cinco especies que pueden considerarse raras a nivel europeo: Nostochopsis lobata Wood ex Bornet et Flahault, Batrachospermum atrum (Hudson) Harvey, Chroothece richteriana Hansgirg, Oocardium stratum Nägeli and Tetrasporidium javanicum Möbius; y de una sexta especie, frecuente en España y Europa, pero que supone la primera cita para esta región, Hydrurus foetidus (Villars) Trevisan. Finalmente, se comparan las características morfológicas y ecológicas de las poblaciones estudiadas con otras citas Europeas.

42 Anales del Jardín Botánico de Madrid 70(1): 81-90, enero-junio 2013. ISSN: 0211-1322. doi: 10.3989/ajbm. 2323

Morphological description and ecology of some rare macroalgae in south-central Spanish rivers (Castilla-La Mancha Region)

Jose Luis Moreno Alcaraz1*, Laura Monteagudo Canales1 & Marina Aboal Sanjurjo2

1Centro Regional de Estudios del Agua, Universidad de Castilla-La Mancha, ctra. de Las Peñas km 3, E-02071 Albacete, Spain 2Departamento de Biología Vegetal, Universidad de Murcia, Campus de Espinardo, E-30100 Murcia, Spain [email protected]; [email protected]; [email protected]

Abstract Resumen Moreno Alcaraz, J.L., Canales Monteagudo, L. & Aboal Sanjurjo, M. 2013. Moreno Alcaraz, J.L., Canales Monteagudo, L. & Aboal Sanjurjo, M. 2013. Morphological description and ecology of some rare macroalgae in south- Descripción morfológica y ecología de algunas macroalgas fluviales de la central Spanish rivers (Castilla-La Mancha Region). Anales Jard. Bot. España centromeridional (Castilla-La Mancha). Anales Jard. Bot. Madrid Madrid 70(1): 81-90. 70(1): 81-90 (en inglés). The Castilla-La Mancha Region (south-central Spain) is scarcely studied in El conocimiento sobre la biodiversidad algal de la región de Castilla-La terms of freshwater algae. However, both the implementation of the Wa- Mancha, situada en la zona centro-sur de España, es escaso en compara- ter Framework Directive (2000/60/CE) and the evaluation of the ecological ción con el de otras regiones peninsulares. Sin embargo, la aplicación de la state of European aquatic ecosystems have increased the intensity and fre- Directiva Marco del Agua (2000/60/CE), y la evaluación del estado ecoló- quency of water body monitoring, including the rivers, lakes and wetlands gico de los ecosistemas acuáticos europeos, ha traído consigo un aumen- of this region. Thus, our knowledge on algal biodiversity and the geogra- to en la frecuencia e intensidad en el muestreo de ríos, lagos y humedales. phical distribution of many species is rapidly increasing. In this study we De esta forma, durante los últimos años se han producido nuevos hallaz- describe the occurrence, ecological conditions and morphological charac- gos en la región que han permitido ampliar el conocimiento de la biodi- teristics of five algal species which are rare at the European level: Nosto- versidad de algas así como de la distribución geográfica de muchas de sus chopsis lobata Wood ex Bornet & Flahault, Batrachospermum atrum (Hud- especies. En este trabajo se describen las condiciones ecológicas y las ca- son) Harvey, Chroothece richteriana Hansg., Oocardium stratum Nägeli racterísticas morfológicas de cinco especies que pueden considerarse raras and Tetrasporidium javanicum Möbius. In addition, we include Hydrurus a nivel europeo: Nostochopsis lobata Wood & Bornet & Flahault, Batra- foetidus (Vill.) Trev., a more common alga in Spain, since this is the first chospermum atrum (Hudson) Harvey, Chroothece richteriana Hansg., Oo- record for the region. Finally, we compare morphological and ecologi cal cardium stratum Nägeli y Tetrasporidium javanicum Möbius; y de una sex- characteristics of the studied populations with other European records. ta especie, frecuente en España y Europa, pero que supone la primera cita para esta región, Hydrurus foetidus (Vill.) Trev. Finalmente, se comparan las características morfológicas y ecológicas de las poblaciones estudiadas con otras citas Europeas. Keywords: Algae, stream, river, Spain, Cyanophyceae, Rodophyceae, Keywords: Algae, arroyo, río, España, Cyanophyceae, Rodophyceae, Chlorophyceae. Chlorophyceae.

INTRODUCTION “macrophytes and phytobenthos”. As a consequence, knowl- The Castilla-La Mancha administrative region (south-central edge on macroalgal biodiversity and the geographical distribu- Spain) remains as one of the most unknown of the Iberian Penin- tion of many species has rapidly increased in recent years. sula in relation to river algae diversity. Five main river basins In this study we describe the occurrence, ecological condi- are included in this region: Tajo, Júcar, Guadiana, Gua dalquivir tions and morphological characteristics of some uncommon al- and Segura, but only the last has been intensely surveyed with gal taxa. One of them has been the first record for Spain and the regards to algae (e.g. Aboal & Llimona, 1985; Aboal, 1988a-c, third for Europe: Nostochopsis lobata Wood ex Bornet & Fla- 1989a-c; Sabater & al., 1989; Aboal & al., 1996). The rest of re- hault; four additional species are hardly cited in Europe: Batra- gion is scarcely studied (e.g. Aboal, 1996; Álvarez & al., 2007) chospermum atrum (Hudson) Harvey, Chroothece richteriana although a species list of Charophytes in Castilla-La Mancha Hansg., Oocardium stratum Nägeli and Tetrasporidium java- focusing mainly on wetland areas has been published by Ciru- nicum Möbius; and finally, Hydrurus foetidus (Vill.) Trev., jano & Medina (2002). Additionally, a recent review on the sta- which has been collected in cold streams of some mountains of tus of river aquatic plants in this region (Moreno & al., 2011) Spain although our record is the first cite for the study area. provides an up to date regional catalogue of aquatic bryophyte All the taxa were found within the boundaries of the Au- species as well as a list of macroalgae genera. tonomous Community of Castilla-La Mancha (south-central The implementation of the Water Framework Directive Spain) (Fig. 1) which occupies an area of 79409 km2. This re- (2000/60/CE) and the evaluation of the ecological state of Euro- gion includes the upper and middle reaches of five large river pean aquatic ecosystems have increased the intensity and fre- basins: Tajo, Guadiana, Guadalquivir, Júcar and Segura. Land quency of monitoring of water bodies, including the rivers, uses are mainly agriculture (46 % of the regional area) and fo- lakes and wetlands of the study area. The assessment of the eco- rest (44 %). Regarding geology, three zones can be distin- logical state of rivers by applying the Water Framework Direc- guished: the western zone is rich in Precambrian siliceous tive (WFD) implies the monitoring of the biological element rocks (mostly quartzite, slate, shale, granite and gneiss); Meso-

* Corresponding author. 43 82 J.L. Moreno Alcaraz & al.

- - + concentration of nutrients (N-NO3 , N-NO2 , N-NH4 , and P- -3 PO4 ) was determined photometrically with MERCK Kits (Spectroquant®); ion chromatography was used to analyse chloride and sulphate; turbidity with a turbidimeter TN-100 (Eutech Instruments; infrared light); calcium and magnesium by complexometry (volumetry); sodium and potassium by atomic emission. All these parameters were analysed following standard procedures detailed in APHA (1998).

RESULTS Nostochopsis lobata Wood ex Bornet & Flahault (Fig. 2a-c) Some specimens of N. lobata were found at the Nava del Rey stream (Fig.1), a temperate temporary stream tributary of the Guadalquivir river. The stream was located in the southern part of Castilla-La Mancha (Province of Ciudad Real), and it flows over Palaeozoic siliceous metamorphic rocks (quartzites, slates and shales). The study site was located at medium altitude (590 m) at 9.5 km from the source with an upstream drainage area of 41.76 km2. N. lobata was found when the stream was at base Fig. 1. Map of the Iberian Peninsula showing the limits of the Castilla-La Mancha flow condition (June 2009). Two morphological forms were Region, the main rivers crossing the study area and the location of the species collected: compact globular specimens corresponding to young recorded. ♦ Hydrurus foetidus; ☐ Oocardium stratum; ● Nostochopsis lobata; ▲ colonies (Fig. 2a,b) were found attached to a moss stem where- ◯ ■ Chroothece richteriana; Batrachospermum atrum; Tetrasporidium javanicum; as free floating diffluent fragments coming from senescent Castilla-La Mancha Region. colonies were collected entangled with other macrophytes. Young colonies measured 0.5-3 cm in diameter while diffluent zoic calcareous rocks (limestone, dolomite, sandstone and con- fragments were about 0.1-1 cm; and its colour changed from glomerates) are dominant in the eastern area; and finally, Ter- brownish to bluish, respectively. The radial disposition of fila- tiary sedimentary fills are accumulated in the great central ments could be clearly observed in the outer part of the colonies, in transversal sections (Fig. 2b). Cells were isodia- plateau located at 700 m a.s.l. called “La Mancha”, where metrical to cylindrical, measuring from 2.0 µm up to 6.17 µm in clays, sandstones, gravels, stones, conglomerates, marls and diameter and length up to 7 µm. Heterocytes were predomi- gypsum are predominant (González & Vázquez, 2000). Moun- nantly lateral, sessile or pedicellate (Fig. 2c), measuring from tains are located mainly on the edges of the region and can 6.7 µm to 8.0 µm in diameter. Intercalary heterocytes were ra- reach more than 2000 m (2273 m, Pico del Lobo) re. The waters were oligo-mesotrophic, with low conductivity and alkalinity values (Table 1). The specimens were collected MATERIAL AND METHODS at marginal depositional habitats along with other aquatic As part of a regional river monitoring program, a stream macroflora such as Ranunculus peltatus Schrank, Cladophora reach of approximately 100 m long was deeply surveyed for sp., Spirogyra sp., Oedogonium sp., Zygnema sp. and Tetras- macroalgae, including all microhabitats present, e.g. riffles, poridium sp. pools, runs or stream margins as well as different kinds of sub- strata (sand, gravel, stones, aquatic vegetation, etc.). Macroal- Batrachospermum atrum (Hudson) Harvey (Fig. 2d-f) gal samples were collected in the river by hand and taken to the B. atrum was found at three study sites: the Vado Blanco laboratory. One part of the samples was fixed in 3 % formalde- stream, Cabriel stream and Júcar river (Fig. 1). The Vado Blan- hyde in the field whereas the other part was maintained fresh co stream connects two lagoons of the Ruidera lagoon complex until laboratory observations. Collected material was exam- made up of 15 karstic lagoons located on a high altitude cal- ined under a Leica M165C stereoscope and a light microscope careous plateau (around 900 m a.s.l.) where forest and irriga- OLYMPUS BX50. Glycerin-gelatine was used to make per- tion land are the main land uses. Nitrate concentration in manent slides which were used to take cell measurements. groundwater and surface water is high due to intensive agricul- Drawings were made with the help of macroscopic and micro- ture practises developed in the last decades (Berzas & al., scopic images taken by a Leica DFC 420 C camera and also by 2004). The Cabriel Stream is the main tributary of the Júcar direct observation of samples. Electric conductivity, pH and River and the study site was located at a high altitude near its dissolved oxygen were measured in situ using appropriate sen- source (1156 m a.s.l.) in a mountainous area where forest is sors (Multiline P4 WTW). Alkalinity was also obtained in the predominant. Groundwater inputs from springs and upwellings field using the sulphuric acid method (APHA, 1998). In addi- are important in this stream reach. The last study site was lo- tion, water samples were collected in polyethylene bottles (500 cated in a middle reach of the River Júcar at medium altitude ml) and were kept in the refrigerator at 4 ºC. All physico-chem- (600 m) in a wide valley dedicated to agricultural uses. The in- ical parameters were analyzed within 48 h after sampling. The fluence of a reservoir located 96.65 km upstream is very signi- 44 Rare macroalgae in south-central Spanish rivers 83

ficant due to flow regulation for agriculture. The river reach is also influenced by groundwater upwelling and seepages that 2.4 9.5 0.41 7.53 7.8 0.02 8 0 0.98 0.5 7.48 0.92 0.1 1564 12 16.61 <0.010 contribute to feed the river. The specimens were collected sub- foetidus Hydrurus (Vill.) Trev. merged in the stream margins, mainly over other macroalgae or entangled with them and exposed to a low current velocity. Thallus size ranged from 2.5-8 cm long and 60-120 µm wide, colour red-brownish, without mucilage (Fig. 2d and 2e), show- 3.5 9.11 8.3 n.d. 8.57 0.33 1.27

590 ing a regular cortication (Fig. 2f) along its central axis and 24.5 23.00 104.00 193 branches. The mean distance between nodes was of 250-525 µm and decreasing from the central axis base towards the api-

Möbius cal branches (up to 80 µm in branch tips). The whorls were very javanicum reduced, 100-200 µm diameter, non-adherent to the main axis, Tetrasporidium 1.5 7.85 2.24 10.00 7.46 0.08 70 0.112 <0.03 0.08 8.93 41.50 3.73 0.9 1.1 816 13.7 30 29 12.8215.5 15.44 and ring or pearl shaped (obconical). Primary fascicles were composed of 3-6 barrel-shaped cells of 3.3-6.8 µm in diameter. Secondary fascicles were composed of 1-3 cells, abundant and scattered along the internodes or concentrated in nodes, the api- cal cell of fascicles usually sharp-pointed. Gonimoblasts were hemispherical, 135-220 µm in diameter and 27-80 µm high, at- – 1 9.21 7.91 0.001 0.027 2.4 0.91

851 tached to the axis in internodes or nodes. Tricogyns shape was 25.19 11.5 31.05 10.73 19.47 55.17 Nägeli 284 547 clavate, 9.3-16.2 µm long and 6.2-8.3 µm wide, located mainly in internodes but also in nodes and axillas at the base of branches. Tricogyns usually included 1-2 round spermatia of 4.7-6 µm diameter attached to the tip. Regarding environmental conditions, it is important to highlight that the three locali- – 2 9.81 7.34 0.137 4.62 ties were under groundwater influence. Dissolved nutrients 1032 12.9 Hansg. showed low values except for one site (Vado Blanco stream, richteriana stratum Chroothece Oocardium Lagunas del Ruidera) located at an irrigation area with high levels of nitrate contents (Table 1). Accompanying aquatic macroflora included Anabaena sp., Batrachospermum gelati- 9.8 9.2 8.1 1.8 0.13 0.019 0 0.746 1.14 2.18

607 nosum (L.), Chara aspera C.L. Dethard. ex Willd. and Chara Júcar Bullones Fte. de la Parra Estena Nava del Rey Zarzas 25.39 53.1 15.6 13.19 174.76 23.6877.2 383.02 126.12 161.3 288 847 2470 hispida L. in Vado Blanco stream; Apium nodiflorum (L.) Lag., B. gelatinosum and Bryum sp. in Júcar river; and Apium nodiflorum, Audouinella sp. and B. gelatinosum in Cabriel stream. 6.9 7.98 0.38 0.029 0.007 2.85 0.44 6.71 0.64 Chroothece richteriana Hansg (Fig. 3a-c) (Hudson) 1156 15.5 41 13.5 85.52 167.65 313.66 67 212 577 Harvey C. richteriana was found in Bullones stream (Fig. 1), a head-

atrum water tributary of the River Tajo situated in a calcareous land- Batrachospermum scape in contact with marls and limestones. The drainage area of subcatchment was 137.97 km2, with 30 % of the area occupied by agriculture and 70 % by natural vegetation. The imme- 2.5 8.8 8 0.185 1.79 0.045 0.02 860 24.17 11.103 48.45 19.35 19.9 114.19 110.71 218.1 870 diate vicinity of the site was a semi-natural area mainly occu- calcareous calcareous calcareous calcareous calcareous siliceous siliceous siliceous 27/05/2009 30/03/2005 03/11/2009 29/05/2007 27/02/2012 13/05/2002 28/04/2010 26/03/2012 Vado Blanco Cabriel pied by Juniperus communis L. and J. thurifera L. The site was located at 1032 m a.s.l., 25 km downstream from the source. In general, colonies of this species were bright green, gelatinous and hemispheric in shape. Transversal sections revealed a strat- ified structure of the hemispherical colony (Fig. 3a): the surface was conformed by a continuous layer of cells on the top of 8.57 0.33 7.47 0.03 1.27 0.042 0.001 590 41.5 15.44 23 10 18.8 92 their radially-arranged stalks, with some parallel lines of sedi- & Flahault 09/06/2009

Nava del Rey ments and carbonate precipitates probably corresponding to

Wood ex. Bornet growth rings. C. richteriana was found growing epilithic, cov- Nostochopsis lobata ) 237 ) 6.84 -1

-1 ering around 80 % of most submerged rocks. Cells were cylin- ) -1

) drical and elongated, measuring from 5 µm to 9 µm in width -1 ) -1 ) ) -1 (6.9 µm average) and from 10 µm to 18 µm in length (15.9 µm ) ) -1 ) -1 ) -1 ) ) -1 -1 -1 ·s -1

3 average) (Fig. 3b,c). The cells were surrounded by a thick cell wall (from 2 to 6 µm wide) and disposed on the stalks (Fig. 3b).

Geographical and ecological data for the algae species studied. Inside some cells, a star-shaped chromatophore was visible (Fig. 3c). Regarding physico-chemical conditions (Table 1) discharge (m lithology (predominant)mean channel width (m) siliceous 3.5 altitude (m) date (dd/mm/yyyy) River/Stream conductivity 25º (μS·cm potassium (mg L Species magnessium (mg L sodium (mg L dissolved oxygen (mg·L pH T (ºC) chloride (mg L ammonium (mg L nitrate (mg L alcalinity (mg L calcium (mg L ortophosphate (mg L sulphate (mg L water was slightly saline due to the presence of marls alterna- Table 1. 45 84 J.L. Moreno Alcaraz & al.

Fig. 2. Nostochopsis lobata: a, macroscopic view; b, transversal section; c, mature filament with heterocytes (from Moreno & al., in press). Batrachos- permum atrum: d, macroscopic view; e, filament; f, detail of gonimoblast and cortication of the thallus. Scale: a, d = 50 mm; b = 1 mm; c, f = 50 μm; e = 500 μm. ting with limestones (I.T.G.E., 2000), with a high content of several centimetres thick on twigs and sticks (Fig. 3e). Under chloride and sulphate. An industry of salt extraction is located the microscope, a lateral vision of a colony (Fig. 3b) showed 8 km upstream (Salinas de Almallá) due to the presence of that this species formed gelatinous branched cylinders sur- evaporite outcorps. rounded by calcite with bright green cells at the top (Fig. 3g). Other aquatic taxa detected at the site were: Apium nodiflo- Cells were heart-shaped, showing a small median constriction rum, Veronica anagallis-aquatica L., Chara vulgaris sp., as in other desmids such as Cosmarium, and measured around Cladophora sp., Tetraspora sp., Vaucheria sp., Batrachosper- 13.5 µm in diameter (11-16 µm) (Fig. 3h). mum sp., Tolypothrix sp., and Phormidium sp. In some cases, colonies of Chroothece rupestris Hangs. were found over colonies of O. stratum. After using acetic acid Oocardium stratum Nägeli (Fig. 3d-i) to eliminate carbonate precipitates from Oocardium colonies, filaments of Rivularia sp. were observed inside them (Fig. 3i). The desmid zygnematal O. stratum was found in Fuente de These three algae species, together with the moss Didymodon la Parra stream (Fig. 1), a small tributary of the Mundo River tophaceus (Brid.) Lisa, were revealed to be the main biological (Segura river basin). This stream runs over calcareous rocks in builders of the global travertine structure. The accompanying a mountainous area covered by a Pinus pinaster Ait. forest. The aquatic macroflora included Callitriche stagnalis Scop., Vero- site was located at 851 m a.s.l. very close to the source (1.73 km nica anagallis-aquatica, Chara vulgaris, Batrachospermum downstream). The biodiversity and proliferation of macroalgae was low, due to the dense canopy reducing the entry of light. gelatinosum , Phormidium sp., Oedogonium sp., Cladophora O. stratum appeared on mosses, sticks and wet rocks in a small glomerata (L.) and Microspora sp. dam 3 m in height built to retain water from the source, where a travertine formation was extensive across the waterfall. To Tetrasporidium javanicum Möbius (Fig. 4a-e) the naked eye, colonies first formed hemispherical structures In the study area, this clorophyte was found at two sites. The like small grains a few millimetres long (Fig. 3d). However, first site was Estena stream (Fig. 1), a small tributary of the with the growth and the increase of carbonate precipitates, the Guadiana river running over siliceous rocks. The study site was colonies formed a lobulated layer of calcareous concretions located in a mountainous area of the Cabañeros National Park, 46 Rare macroalgae in south-central Spanish rivers 85

Fig. 3. Chroothece richteriana: a, transversal section of a colony; b, lateral view of pedunculated cells; c, cells with star-shaped chromatophores. Oocar- dium stratum: d, small colonies covering a moss branch; e, aged colony forming a thick lobulated layer; f, lateral microscopic view showing the calcite cylinders surrounding cell stalks; g, upper view of cells surrounded by calcite; h, detail of cell with stalk; i, colony of O. stratum together with Rivularia after carbonate removal. Scale: a, i = 500 μm; b, c, h = 20 μm; d = 5 mm; e = 1cm; f = 100 μm; g = 200 μm.

and 6.1 km downstream from the source. The drainage area (described in Nostochopsis lobata section), but different mor- was covered by a near natural forest of Quercus Mediterranan phological characteristics were observed. In this case, colonies maquia. T. javanicum appeared as small (environ 8 mm in were thin perforated sheets (Fig. 4d), the biggest one measur- length) free floating colonies forming a light green gelatinous ing 1.5 ϫ 2.1 cm. Cells were smaller and surrounded by con- tubular-net thallus (Fig. 4a). These small colonies probably centric gelatinous sheaths (Fig. 4e). The smaller cell size (from corresponded to fragments of a bigger colony. The ‘ribbons’ 3.3 µm to 5.7 µm in diameter) of this specimen matched with making up the thallus (Fig. 4b) measured 100-300 µm wide and the description of T. fottii Couté et Traccana included in the key presented revoluted margins. Cells were heterogeneously ar- of Ettl & Gärtner (2009), although for other authors Tetras- ranged in the mucilage and were spherical to slightly oval in poridium is a monospecific genus (see discussion). shape (Fig. 4c). A wide range of cell size was observed: from 5.1-10.7 µm in diameter, with 8.2 µm on average. Chloroplas- Hydrurus foetidus (Vill.) Trev. (Fig. 4f-h) ts were parietal with a single large pyrenoid (around 1.6 µm in diameter). Pseudo-flagella, typical of the genus Tetraspora, H. foetidus was found at Zarzas stream (Fig. 1), a headwater were absent. Environmental and physico-chemical data record- tributary of the Hoz stream which belongs to the largest Span- ed for this site indicated that the stream was mesotrophic, ish river basin, the Tajo river. The Zarzas stream runs through showing a very low mineralisation due to the siliceous litho- the Natural Park of Sierra Norte of Guadalajara, located in the logy (Table 1). The waters showed a carbonated character but northernmost part of the region. This area is mountainous and with a high proportion of chloride. Accompanying aquatic is covered by a forest of Pinus sylvestris L. with the presence of macroflora included Audouinella sp., Draparnaldia sp., Zyg - Fagus sylvatica L. The study site was located at an altitude of nema sp., Tolypothrix sp., Scytonema sp., Nostoc sp., and Chi- 1564 m, in one of the coldest areas of the region, 7.90 km loscyphus polyantos (L.) Corda. downstream from its source. The drainage area of subcatch- In addition, we collected Tetrasporidium in another ment is 26.90 km2. This alga was found as gold-brown tufts up siliceous oligo-mesotrophic stream, the Nava del Rey stream to 20 cm long attached to boulders and stones (Fig. 4f,g) some- 47 86 J.L. Moreno Alcaraz & al.

Fig. 4. Tetrasporidium javanicum: a, macroscopic view of a tubular-net colony (“ribbon-like”); b, detail of a part of the colony; c, detail of cells; d, macroscopic view of a laminar colony; e, detail of the colony showing cells surrounded by gelatinous sheaths. Hydrurus foetidus: f,g, macroscopic view of two thallus shapes; h, detail of cell arrangement within the mucilage. Scales: a = 2 mm; b = 100 μm; c, h = 20 μm; d = 500 μm; e = 50 μm; f, g = 1 mm. times covering 100 % of the substratum. Thalli were branched 1930-1932; Desikachary, 1959; Tiwari, 1978; Peerapornpisal and gelatinous. Cells embedded in the mucilage were peripher- & al., 2006). In Europe there are only third records from flow- ally arranged and oval to ellipsoidal in shape (Fig. 4h). Cell size ing warm waters: Banyuls-sur-Mer (France) (Frémy & Feld- was 5-8 µm wide (6.5 µm on average) and 7-13.5 µm long mann, 1934), Corsica (France) (Hoffmann, 1990), and L’Aqui - (10.5 µm on average) in. Regarding nutrients, waters were oli- la (Italy) (Del Grosso, 1977). The specimens of N. lobata col- go-mesotrophic (Table 1). Accompanying aquatic macroflora lected in Nava del Rey were the first recorded in Spain (Mo - was made up of the following taxa: Lemanea sp., Hildenbran- reno & al., in press). Morphologically, our material matched dria sp., Scapania undulata (Hedw.) P. Beauv., and Marchan- fairly well the detailed description of the first specimen report- tia polymorpha L. ed in Europe by Frémy & Feldmann (1934). In both cases N. lobata was found in small temporary streams running over DISCUSSION siliceous metamorphic rocks, located at low altitude and living in warm and oligotrophic waters, and it was collected in sum- N. lobata is included in the family Nostochopsaceae (Anag- mer. In terms of habitat, we found N. lobata growing epiphytic nostidis & Komarek, 1990; Komárek & al., 2003). This species (young) and free floating (senescent) even though it has been is characterized by a globular lobed thallus composed of unise- reported in different habitats: lakes (Tiwari, 1978; El Saied, riate and branched thricomes with intercalar or lateral hetero- 2007), running waters (Frémy & Feldmann, 1934; Palmer, cytes (pedicellate or sessile) (Bornet & Flahault, 1886-1888). 1941; Pandey & Pandey, 2008), thermal springs (Yoneda, N. lobata is widely distributed in tropical areas of the world and 1939), epiphytic (Sarma & Chapman, 1975), attached to rocks has been reported in several African countries (Frémy, 1929), (Hoffmann, 1990; Peerapornpisal, 2006), free floating (Wood, North and South America and Australia (Cáceres, 1973; Sarma 1872), on moist soils (Skinner & Entwisle, 2001; Aziz, 2008), & Chapman, 1975; Branco & al., 2001) and Asia (Geitler, and on humid walls (Skinner & Entwisle, 2001) among others. 48 Rare macroalgae in south-central Spanish rivers 87

However, the most extreme habitat where N. lobata has been ly it has been found growing in slightly saline streams of south- found is probably as cryptoendolithic in an arid climate (Weber eastern Spain (Murcia), mostly on subaerial conditions (un- & al., 1996). Current data point out that this species is dis- published data). This rare Rhodophyte included in the Por- tributed worldwide with an underlying tendency: it grows phyridiaceae family is characterized by mucilaginous colonies abundantly in tropical regions where its optimal climatic con- conformed by elliptical cells enclosed in a layered gelatinous ditions are stable for several months whereas it grows scarcely matrix (Eloranta & Kwandrans, 2007; Eloranta & al., 2011). in other regions only when conditions allowing its germination Additionally, a star-shaped chromatophore is clearly visible in- and growth are reached, probably for a few weeks in occasion- side the cells. Starmach (1977) and Eloranta & al. (2011) con- al years (Moreno & al., in press). This fact explains the diffi- sider cell size as a differential characteristic between the two culty in detecting this species in Europe, where it has always species of Chroothece present in Castilla-La Mancha: C. rich- been recorded only once per site. teriana measuring 6-10 µm wide and 15-18 µm long; C. ru- B. atrum belongs to subgenus Batrachospermum, which is pestris measuring 5-7 µm wide and 9-15 µm long. Additional- characterized by a carposporophyte multicellular, and to sec- ly, Sheath & Sherwood (2002) reported a cell size for C. rich- tion Setacea which is characterized by short carpogonium- teriana of 8-12 µm wide and 10-21 µm long (British Islands). bearing branches and whorls/fascicules (Eloranta & Kwan- Both studies match with the cell size of our colonies collected drans, 2007). B. atrum (section Setacea) is on divergent in Bullones stream (5-9 µm wide, 10-18 µm long). However, branches of molecular trees (Kumano, 2002) and constitutes a we noticed that according to our observations, the cell size distinct section, as proposed by Sheath & al. (1993). Goni- range of a Chroothece population can exceed that one reported moblasts form mamilliform or semispherical swellings on the from identification keys. On the other hand, the relatively high central axis and the trichogyn is clavate; the thallus is subge- water salinity of Bullones stream was in accordance with the latinous (Starmach, 1977). The habitus shows a compressed “more or less saline” environment described for C. richtheria - and compact thallus due to its short fascicules and whorls, re- na in the identification keys (Eloranta & al., 2011; Starmach, sembling a bamboo stem, similar to the thallus of the red algae 1977). Lemanea. According to Kumano (2002), B. atrum is distribut- Regarding Chroothece populations from Fuente de la Parra ed throughout Europe (UK, France, Belgium, Germany, (see Oocardium stratum section), they were identified as C. ru- Poland, Portugal and Sweden), Eastern Asia (China, Korea and pestris according to the identification keys criteria: lower cell Japan), Australia, New Zealand, South America (Brazil) and size, low water mineralisation, and also the environment “wet Africa (Angola), in addition to the USA (California and Texas; rocks”. In the same way, Margalef (1989) recorded C. rupes- Sheath & al, 1993). In Spain there is only one previous record tris from wet rocks of calcareous springs, described by him as from the Murcia Region (Aboal & al., 1995). The morphologi- “hygropetric environment”, a similar environment to that of cal characteristics match with other descriptions provided by Fuente de la Parra stream. other authors (Israelson, 1942; Sheath & al., 1993; Eloranta & Oocardium is a monospecific genus belonging to the family Kwandrans, 2007). B. atrum is an oligotrophic hard water Desmidiaceae, characterized by being mostly unicellular, ei- species (Sládecek, 1973; Rott & al., 1999; Eloranta & Kwan- ther solitary or in colonies, with cells divided into two com- drans, 1996, 2002). In Spain, Aboal & al. (1995) found this partments separated by an isthmus. West (1904) described species in calcareous springs and small streams with environ- Oocardium as the most extraordinary of all the genera of mental conditions similar to those described in this work. The Desmids usually occurring in large colonies. According to three sites where B. atrum was collected were calcareous and West (1904) and Bourrelly (1990), the colonies of O. stratum they showed groundwater influence associated to upwelling ar- are hemispherical in shape (1-2 mm in diameter), and are eas, springs or seepages. The trophic conditions in our sites formed by cells standing on more or less parallel, radiating were oligo-mesotrophic, with one site showing a high concen- strands of mucus which are encrusted in calcite. During the tration of dissolved nitrate due to its location within an inten- growing period, calcite crystals are formed by continuous de- sively irrigated cropland area (Álvarez-Cobelas & al., 2005). position by surrounding cells which, following division, calci- According to Eloranta & Kwandrans (2007) the species grows fy into slightly different directions upwards (Sanders & Rott, epiphytic or epilithic, but in our sites it was collected as epi- 2009). The colonies detected in Fuente de la Parra stream phytic, floating among other macrophytes close to river banks matched exactly with the above description and appeared joint- and avoiding high current velocity, more limnephilic than ed, forming a nodular crust covering substrates, similarly to rheophilic. According to Israelson (1942) and Aboal & al. colonies observed by Pentecost (1991). Regarding the environ- (1995) the species seem to occur only in spring, but in the study mental conditions, O. stratum seems to have very specific re- area the species was recorded both in spring and autumn, simi- quirements: it is only found in waterfalls and springs with cal- lar to the seasonal pattern frequently shown by the rest of the careous water (Margalef, 1983; Pentecost, 1991) usually being Batrachospermaceae in lotic waters (Aboal & al., 1995). The associated to deposits of travertine and tufa (Pentecost, 2001; fact that the species was collected only once per site despite Sanders & Rott, 2009; Bellinger & Sigee, 2010). Because of several visits may be due to the rapid development and short life these environmental restrictions, O. stratum is often considered cycle of the species or to the occasional interannual occurrence. as an unusual alga (e.g. Margalef, 1983; Bellinger & Sigee, After C. richteriana was described by Hansgirg (1884), few 2010) even though its presence has been detected worldwide: publications about this species have been released. C. richte- Switzerland (Nägeli, 1849) the United States of America riana has been reported in Britain (Sheath & Sherwood, 2002; (Prescott, 1981), Austria (Sanders & Rott, 2009), United King- Eloranta & al., 2011) and Spain (Eloranta & al., 2011). Recent- dom (West, 1904; Pentecost, 1991), Ireland (Carter, 1923), 49 88 J.L. Moreno Alcaraz & al.

Belgium (van Oye & Hubert, 1937), France (Kouwets, 1999), appropriate observation of the ultrastructure of cells was not Slovenia (Vrhovsek & al., 2006). Particularly in Spain, O. stra- possible and, like Entwisle & Skinner (2001), we could not tum has been previously reported in Barcelona (Margalef, conclusively identify our material as T. fottii. On the other 1955), Murcia (Aboal, 1991) and in Castilla-La Mancha hand, Calado & Rino (1992) suggest that the specific charac- (Ruidera Lakes) (Álvarez-Cobelas & al., 2007), although we teristics of T. fottii could be explained by the cytologic vari- think that it could be a common species in calcareous moun- ability within Tetrasporidium javanicum and, therefore, T. fot- tains and springs. tii should not be considered as an independent species. This T. javanicum is a clorophyte classified within the order idea is contemplated by other authors that classify Tetraspori- Tetrasporales, family Palmellopsidaceae (Ettl & Gärtner, dium as a monospecific genus (e.g. Bourrelly, 1990). Thus, we 2009). In the mature state the thallus is a thin, flat, more or less assigned our material to T. javanicum until more studies unra- circular sheet with numerous irregular perforations with a veling the taxonomy of this interesting genus are released. smooth or lobed margin (Iyengar, 1932). Nevertheless, the The chrysophyte Hydrurus foetidus is included in the order specimen collected in the Estena River was a tubular-net thal- Chromulinales, family Hydruraceae. The thallus is gelatinous, lus, probably as a result of the increase in the size perforation branched and attached to substrate reaching up to 30 cm long. with age. This fact supports the idea that ribbon-like and sheet- Cells are ellipsoid to subspherical and they are randomly ar- like thallus could represent plants of different ages (Entwisle & ranged in the mucilage (Rodriguez & Vergon, 1996). After col- Skinner, 2001). According to Ettl & Gärtner (2009) cells are lecting, it emanates a characteristic smell. This species is a cold oval to spherical in shape ranging from 6 µm to 15 µm in diam- water stenotherm also adapted to strong current velocity, being eter, slightly larger than our material (from 5.1 to 10.7). Re- an exclusive inhabitant of cold mountain streams worldwide garding distribution, this species is abundant in tropical areas (2-12 ºC) (Hieber & al., 2001; Wehr & Sheath, 2003; Kriz- within India (Iyengar, 1932), China (Hu & Wei, 2006), Hawai- manic & al., 2008). Under optimal conditions, H. foetidus be- ian Islands (Sherwood, 2004), Java, the United States, Australia comes dominant and may entirely cover submerged rocks, and Bangladesh (Entwisle & Skinner, 2001) being quite com- making them very slippery. In Europe it is a well-known inhab- mon in tropical areas (Entwistle & Skinner, 2000). In Europe, itant of alpine streams and those draining glaciated landscapes, T. javanicum has been reported in the Czech Republic, France, where it is described as a predominant or very common Portugal and Spain (Fott & al., 1965; Coute & Tracanna, 1981; macroalga by several authors (e.g. Hieber & al., 2001; Canto- Calado & Rino, 1992; Aboal & al., 1994, respectively). In nati & al., 2006; Krizmanic & al., 2008). In Spain, H. foetidus Spain, this species has been found in the Alicante Province has been reported in the Pyrenees (NE Spain) (Margalef, 1948; (Aboal & al., 1994), Cáceres Province (Marín-Murcia & Aboal, Llimona & al.,1985), Cantabrian mountains (N Spain) (Mar- 2007), Galicia (López-Rodríguez & Penalta-Rodríguez, 2007) galef, 1950) and Sierra Nevada (Granada, S Spain) (Sánchez- and, recently, in the Ebro river basin (NE Spain) (Tomás & al., Castillo, 1984) with this study being the first recordin for the in press). In almost all cases, the species is found growing in Castilla-La Mancha region. As for habitat conditions, H. foe- conditions of low current velocity, well oxygenated alkaline tidus is clearly rheophylic and stenothermic, disappearing when waters, high turbidity and nutrient enrichment due to agricul- water temperature rises above 10 ºC or 15 ºC according to Wehr tural practises. However, the material described in this work & Sheath (2003) and Starmach (1985), respectively. A combi- was found in low-mountain clear water streams. These condi- nation of hydrological factors and light conditions controls the tions are more similar to those described by Calado & Rino seasonal fluctuations of this macroalga (Cantonati & al., 2006). (1992) and López-Rodríguez & Penalta-Rodríguez (2007) for In general, this species is commonly found during winter and specimens collected in Portugal and Galicia, respectively. De- spring (Hieber & al., 2001). In Spain, it is a vernal species found spite the fact that the Estena stream is oligotrophic, T. javani - occasionally in summer at high altitudes (>2000 m) (Llimona & cum was found coinciding with a relatively high concentration al., 1985). Our material was collected in early spring at a high of nitrogen compounds (nitrite, nitrate and ammonium). This altitude (1564 m) when water temperature was 7.8 ºC. Both the fact could be due to the preference of T. javanicum for eutroph- cellular size (from 7 µm to 13.5 µm in length) and the thallus ic waters, as suggested by some authors (Aboal & al., 1994; length (up to 20 cm) of the specimens coincide with previous Marín-Murcia & Aboal, 2007; Tomás & al., in press) pointing literature which describes large specimens (up to 30 cm) with to this species as a potential good indicator of eutrophy. Our ob- cells measuring from 8 µm to 12 µm in length (Bourrelly, 1957; servations lead us to think that T. javanicum is more common Rodriguez & Vergon, 1996; Wehr & Sheath, 2003). than reported up to now in Spain and it will be collected more frequently in upcoming river macrophyte surveys. ACKNOWLEDGEMENTS Regarding T. fottii, the species was described by Couté & Tracanna (1981) from some specimens found in Breton ponds. This work was supported by the Junta de Comunidades de Castilla-La However, we were able to find only one other report on this Mancha, Consejería de Educación y Ciencia, through two regional research projects (PO1109-0190-8090 1349 and PPII10-0271-1349). species in literature, made by Neag & al. (2005) in the "Alexan- dru Borza" Botanical Garden, Cluj-Napoca, Romania. Ettl & REFERENCES Gärtner (2009) differenced T. fottii from other Tetrasporidium species according to the lateral position of the pyrenoid and its Aboal, M. 1988a. Zygnemataceae (Conjugales, Chlorophyceae) of the river Segura basin, southeastern Spain. Nova Hedwigia 47(3-4): 389-402. cell size (from 6.4 µm to 9.5 µm in diameter). However, the Aboal, M. 1988b. Aportación al conocimiento de las algas epicontinentales del material collected in Nava del Rey was slightly smaller, mea- sudeste de España. III: Cianofíceas (Cyanophyceae Schaffner 1909). Ana- suring from 3.3 µm to 5.7 µm in diameter. Unfortunately, an les del Jardín Botánico de Madrid 45(1): 1-46. 50 Rare macroalgae in south-central Spanish rivers 89

Aboal, M. 1988c. Aportación al conocimiento de las algas epicontinentales del Carter, N. 1924. A monograph of the British Desmiciacea. Volume V. Ray So- S.E. de España. VII. Clorofíceas (Chlorophyceae Wille in Warming 1884). ciety. London. Candollea 43: 521-548. Cirujano S. & L. Medina. 2002. Plantas acuáticas de las lagunas y humedales Aboal, M. 1989a. Contribución al conocimiento de las algas epicontinentales de Castilla-La Mancha. Real Jardín Botánico (CSIC) y Junta de Comuni- del SE de España II: Los rodófitos (Rhodophyceae). Lazaroa 11: 115-122. dades de Castilla-La Mancha. Madrid. Aboal, M. 1989b. Aportación al conocimiento de las algas epicontinentales del Couté, A. & Tracanna, B. 1981. Sur la présence en France de Tetrasporidium S.E. de España. V. Xantofíceas (Xanthophyceae P. Allorge ex Fritsch, javanicum Möbius (Chlorophyta, Euchlorophyceae, Tetrasporales) et sa 1935). Boletim da Sociedade Broteriana 62: 239-248. position systematique. Cryptogamie, Algologie 2(3): 209-219. Aboal, M. 1989c. Epilithic algal communities from River Segura Basin, South- Del Grosso, F. 1977. Nostochopsis Lobatus Wood: prima segnalazione per easetrn Spain. Archiv für Hydrobiologie 116(1): 113-124. l’Italia. Informatore Botanico Italiano 9: 285-286. Aboal, M. (1991). Aportación al conocimiento de las algas epicontinentales del Desikachary, T.V. 1959: Cyanophyta. Monographs on Algae. Indian Council S.E. de España. VIII: Desmidiáceas y Mesoteniáceas (DESMIDIACEAE, of Agricultural Research. New Delhi. MESOTAENIACEAE, ZYGOPHYCEAE Widder 1960). Acta Botanica El Saied, H.E. 2007. Molecular genetic monitoring of bacterial communities in Malacitana 16(2): 325-337 Manzala Lake, Egypt, based on 16S rRNA gene analysis. Egyptian Journal Aboal, M. & Llimona, X. 1985. Aportación al conocimiento de las algas del SE of Aquatic Research 33(3):179-194. de España. I. Caráceas (Characeae). Anales de Biología 6: 7-17. Eloranta, P. 2011. Süsswasserflora von Mitteleuropa, 7. Rhodophyta and Aboal, M., Puig, M.A., Sánchez-Godínez, A. & Soler, G. 1994. Algal stand- Phaeophyceae. Spektrum Akademischer Verlag. Heidelberg. ing-crop in some Mediterranean temporary rivers in southeastern Spain. Eloranta, P. & Kwandrans, J. 1996. Distribution and ecology of freshwater red Verhandlungen der Internationalen Vereinigung für Theoretische und algae (Rhodophyta) in some central Finnish rivers. Nordic Journal of Angewandte Limnologie 25: 1746-1750. Botany 16(1): 107-117. Aboal, M., López-Jiménez, E. & Asencio, A.D. 1995. Batrachospermum Eloranta, P. & Kwandrans, J. 2002. Notes on some interesting freshwater atrum (Huds.) Harv. (Batrachospermales, Rhodophyceae), novedad para la Rhodophyta from Finland. Archiv für Hydrobiologie. Algological studies flora algal epicontinental española. Anales del Jardín Botánico de Madrid 143: 95-109. 53(1): 121-123. Eloranta, P. & Kwandrans, J. 2007. Freshwater red algae (Rhodophyta): iden- Aboal, M. 1996. Comunidades fitobentónicas del Parque Nacional de Las Ta- tification guide to European taxa, particularly to those in Finland. Botani- blas de Daimiel. In: Álvarez, M. & Cirujano, S. (eds.), Las Tablas de Dai- cal Museum Finnish Museum of Natural History. Helsinki. miel. Ecología acuática y sociedad: 117-128. Madrid. Entwisle, T.J. & Skinner, S. 2001. Non-marine algae of Australia: 4. Floristic Aboal, M., Marín, J.P. & Nieva, A. 2006. Tetrasporidium javanicum Möbius survey of some colonial green macroalgae (Chlorophyta). Telopea 9(3): (Tetrasporales, Chlorophyta) ¿especie invasora? Resúmenes reunión cien- 725-739. tífica de la Sociedad Española de Ficología. Algas (Boletín de la Sociedad Ettl, H. & Gartner, G. 2009. Süsswasserflora von Mitteleuropa 10. Chloro- Española de Ficología) 37: 4. phyta. - 2. Tetrasporales, Chlorococcales, Gloeodendrales. Spektrum Álvarez-Cobelas, M., Cirujano, S., Montero, E., Rojo, C., Rodrigo, M.A., Piña, Akademischer Verlag. Heidelberg. E., Rodríguez, J.C., Soriano, O., Aboal, M., Marín, J.P. & Araujo, R. 2007. Fott, B., Nováková, M. & Kalina, T. 1965. Morphology, reproduction and oc- Ecología acuática y sociedad de las Lagunas de Ruidera. C.S.I.C. Ministe- currence of a tropical alga, Tetrasporidium javanicum Möbius (Chloro- rio de Educación y Ciencia. Madrid. phyceae). Preslia 37: 380-386. Álvarez-Cobelas, M., Rojo,C. & Angeler, D.G. 2005. Mediterranean limnolo- Frémy, P. & Feldmann, J. 1934. A propos de la présence en France du Nosto- gy: current status, gaps and the future. Journal of Limnology 64(2): 13-29. chopsis lobatus Wood. Bulletin de la Société Botanique de France 81(7/8): Anagnostidis, K. & Komarek, J. 1990. Modern approach to the classification 612-618. system of Cyanophytes. 5. Stigonematales. Algological Studies 59: 1-73. Gayral, P. 1951. Une rare Rhodophycee d’eau douce dans le lac d’Iseli (Haut. APHA. 1998. Standard methods for the examination of water and wastewater. Atlas Central): “Chroothece rupestris” Hangs. Bulletin de la Société des 20th Edition. American Public Health Association, American Water Works Sciences Naturelles du Maroc 31: 15-18. Association and Water Environment Association. Baltimore. Geitler, L. (1930-1932): Cyanophyceae von Europa. In: L. Rabenhorst (ed.), Aziz, A. 2008. Algal flora of Madhabkunda waterfall area in Maulvi Bazar, Kryptogamen-Flora von Deutschland, Österreich und der Schweiz.: 1- Bangladesh. III. New records of blue-greens and greens. Bangladesh Jour- 1196. Koeltz Scientific Books. Leipzig. nal of Botany 37(1): 43-48. González, A. & Vázquez, A. 2000. Las formas del relieve: llanuras, páramos y Bellinger, E.G. & Sigee, D.C. 2010. Freshwater algae: identification and use montañas. In: Guía de Castilla-La Mancha: 17-40. Junta de Comunidades as bioindicators. Wiley-Blackwell. Chichester, West Sussex. de Castilla-La Mancha. Toledo. Berzas, J.J., García, L.F., Martín-Álvarez, P.J. & Rodríguez, R.C. 2004. Qual- Hansgirg, A. 1886. Algarum aquae dulcis species novae. Oesterrichische Bo- ity assessment and chemometric evaluation of a fluvio-lacustrine system: tanische Zeitschrift 4: 109-111. Ruidera Pools Natural Park (Spain). Water, Air and Soil Pollution 155: Hieber, M., Robinson, C.T., Rushforth, S.R. & Uehlinger, U. 2001. Algal com- 269-289. munities associated with different alpine stream types. Artic, Antartic and Bornet, E. & Flahault, C. (1886-1888): Révision des Nostocacées heterocys- Alpine Research 33(4): 447-456. tées contenues dans les principaux herbiers de France. Annales des Scien- Hoffmann, L. 1990. Presence of Mastigocladopsis jogensis (Cyanophyceae, ces Naturelles, Botanique 7, Ser. 3: 323-381. Mastigocladopsiceae) in Corsica (France). Cryptogamie, Algologie 11(2): Bourrelly, P. 1957. Recherches sur les Chrysophycees: morphologie, phy- 219-224. logenie, systematique . Museum national d’histoire naturelle, Laboratoire Hu, H. & Wei, Y. 2006. The freshwater algae of China. Systematics, taxonomy de cryptogamie. Paris. and ecology. Science Press. Pekin. Bourrelly, P. 1990. Les algues d'eau douce. Initiation à la systématique. Tome Israelson, G. 1942. The freshwater florideae of Sweden:studies on their taxon- I: Les algues vertes. Société Nouvelle des Éditions Boubée. Paris. omy, ecology, and distribution. A.-B. Lundequistska Bokhandeln. Uppsala. Branco, L.H.Z., Necchi Jr., O., & Branco, C.C.Z. 2001. Ecological distribution Israelson, G. 1949. Two freshwater Protofloridae new to Sweden. Botaniska of Cyanophyceae in lotic ecosystems of São Paulo State. Revista Brasileira Notiser: 366-376 de Botânica 24(1): 99-108. I.T.G.E., 2000. Unidades hidrogeológicas de España. Mapa y datos básicos. Cáceres, E.J. (1973): Contribución al conocimiento de las cianófitas. I. Dos no- Instituto Tecnológico y Geominero de España, Madrid. vedades para la Argentina: Nostochopsis lobatus (Nostochopsidaceae) y Iyengar, M.O.P. 1932. Two Little-known Genera of Green Algae (Tetraspori- Rivularia rufescens (Rivulariaceae). Kutzingiana, 7: 165-179. dium and Ecballocystis). Annals of Botany 46(2): 191-192. Calado, A.J. & Rino, J. A. 1992. Observations and taxonomic considerations Komárek, J., King, H. & Komárková, J. 2003: Filamentous Cyanobacteria. In: on a Tetrasporidium (Chlorophyta, Tetrasporales) found in Portugal. Cryp- Wehr, J.D. &. Sheath, R (eds.), Freshwater algae of North America. Ecolo- togamie, Algologie 13(3): 157-167. gy and Classification: 117-196. San Diego. Cantonati, M., Gerecke, R. & Bertuzzi, E. 2006. Springs of the Alps – Sensi- Kouwets, F.A.C., 1999. A check-list of desmids (Chlorophyta, Zygnemaphy - tive ecosystems to environmental change: from biodiversity assessments to cea) of France. Muséum national d’histoire naturelle. Paris. long-term studies. Hydrobiologia 562(1): 59-96. Krizmanic, J., Subakov-Simic, G. & Karadzic, V. 2008. Supplementary notes 51 90 J.L. Moreno Alcaraz & al.

on the distribution of Hydrurus foetidus (Vill.) Trevisan (chrysophyta) in Rott, E., Pfister, P., van Dam, H., Pall, K., Pipp, E., Binder, N. & Ortler, K., Serbia. Archives of Biological Sciences 60(2): 13-14. 1999. Indikationslisten für Aufwuchsalgen. Teil 2: Trophieindikation und Kumano, S. 2002. Freshwater red algae of the world. Biopress. Bristol. 375 pp. autökologische Anmerkungen. Bundesministerium für Land- und Forst - Llimona, X., Ballesteros, E., Brugués, M., Comín, F.A., Cros, R.M., Molero, wirtschaft. Wien. J., Romero, J., Tomás, X. & Torrella, F. 1985. Història natural dels països Sabater, S, Aboal, M. & Cambra, J. 1989. Nuevas observaciones de rodofíceas catalans. 4, Plantes inferiors. Enciclopèdia Catalana. Barcelona. en aguas epicontinentales del NE y SE de España. Limnetica 5: 93-100. López-Rodríguez, M.C. & Penalta-Rodríguez, M. 2007. Freshwater algae in Sánchez-Castillo, P.M. 1984. Estudio ficológico de la vegetación reófila de Galician Central Macizo rivers (NW Spain) with new records for the Iberi- Sierra Nevada (Granada). Limnetica 1: 136-140. an Peninsula. Algological Studies 125: 57-77. Sarma, P. & Chapman, V.J. 1975. Occurrence of the genus Nostochopsis in Margalef, R. 1948. Flora, fauna y comunidades bióticas del Pirineo de la Cer- New Zealand. Nova Hedwigia 26: 305-308. daña. Monografías de la Estación de Estudios Pirenaicos Vol, 11. C.S.I.C. Sheath, R.G., Vis, M.L. & Cole, K.M. 1993. Distribution and systematics of Zaragoza. Batrahcospermum (Batrachospermales, Rhodophyta) in North America. 3. Margalef, R. 1950. Datos para la hidrobiología de la cordillera cantábrica, es- Section Setacea. Journal of Phycology 29(5): 719-725. pecialmente del macizo de los Picos de Europa. Publicaciones del Instituto Sheath, R.G. & Sherwood, A.R. 2002. Phylum Rhodophyta (Red Algae). In: de Biología aplicada 7: 37-76. John, D.M., Whitton, B.A. & Brook, A.J. (eds.), The freshwater algal flora Margalef, R. 1955. Materiales para una flora de las algas del NE de España. VI. of the British Isles. An identification guide to freshwater and terrestrial al- Desmidiales, Rhodophyceae. Collectanea Botanica 4(3): 319-330. gae: 123-143. Cambridge. Margalef, R. 1983. Limnología. Ediciones Omega, Barcelona. Sherwood, A.R. 2004. New records of freshwater macroalgae and diatoms Marín-Murcia, J.P. & Aboal, M. 2007. Estudio de las comunidades vegetales form Hawaian Islands. Bishop Museum Occasional Papers 79: 3-8. de los canales de Montijo y Lobón y de las condiciones que favorecen su de- Skinner, S. & Entwisle, T.J. 2001. Non-marine algae of Australia: 1. Survey of sarrollo (informe técnico). Confederación Hidrográfica del Guadiana. Mi- colonial gelatinous blue-green macroalgae (Cyanobacteria). Telopea 9(3): nisterio de Medio Ambiente. 573-599. Moreno, J.L., Ros, M.R. & De las Heras, J. 2011. La flora acuática de los ríos Sladecek, V. 1973. System of water quality from the biological point of view. y humedales de Castilla-La Mancha: Comunidades, hábitats y medidas de Archiv für Hydrobiologie, Ergebnisse der Limnologie 7: 1-218. conservación. In: Hernández, J.E. & Herranz, J.M. (eds.), Protección de la Starmach, K. 1977. Phaeophyta. Rhodophyta. Flora Stodkowodna Polslci, vol. diversidad vegetal y de los recursos fitogenéticos de Castilla-La Mancha: 14. Polska Academia Nauk, Instytut Botaniki.Warsaw. 277-315. Albacete. Tiwari, D.N. 1978. The heterocysts of the blue-green alga Nostochopsis loba- Moreno, J.L., Aboal, M. & Monteagudo, L. 2013. On the presence of Nosto- tus: Effects of cultural conditions. New Phytologist 81(3): 653-656. chopsis lobata Wood ex Bornet et Flahault in Spain: morphological, eco- Tomás, P., Moreno, J.L., Aboal, M., Oscoz, J., Durán, C. & Navarro, P. Distri- logical and biogeographical aspects. Nova Hedwigia 95: 373-390. bución de Tetrasporidium javanicum Möbius, 1893 (Chlorophyta: Tetras- Nägeli, C., 1849. Gattungen einzelliger Algen: physiologisch und systematisch porales) en la cuenca del río Ebro. Limnetica (in press). bearbeitet. Friedrich Schulthess. Zurich. Van Oye, P. & Hubert, B., 1937. Recherches sur les “crons” du jurassique bel- Neag, I., Momeu, L., Peterfi, L.S. 2005. Algal communities from some aquat- ge. Données générales. Biologische Jaarboek Dodonaea 4: 231-235. ic habitats of the “Alexandru Borza” Botanical Garden, Cluj-Napoca, Ro- Vrhovsek, D., Kosi, G., Krivograd-Klemencic, A., Smolar-Zvanut, N., Bernik, mania. Contribu ii Botanice 40: 153-162. T., Grbovic, J., Roser-Drev, A. & Remec-Rekar, S. 2006. Monograph on Palmer, C.M. 1941. Additional records for algae, including some of the less freshwater and terrestrial algae in Slovenia. Zalozba ZRC, ZRC SAZU. common forms. Butler University Botanical Studies 5(1): 224-234. Ljubljana. Pandey, U. & Pandey, J. 2008. Enhanced production of biomass, pigments and Weber, B., Wessels, D.C.J. & Budel, B. 1996. Biology and ecology of cryp- antioxidant capacity of a nutritionally important cyanobacterium Nosto- toendolithic cyanobacteria of a sandstone outcrop in the Northern Province, chopsis lobatus. Bioresource Technology, 99(10): 4520-4523. South Africa. Algological Studies 83: 565-579. Peerapornpisal, Y., Amornledpison, D., Rujjanawate, C., Ruangrit, K. & Kan- Wehr, J.D. & Sheath, R.G. 2003. Freshwater algae of North America: ecology janapothi, D. 2006. Two Endemic Species of Macroalgae in Nan River, and classification. Academic Press. Boston. Northern Thailand, as Therapeutic Agents. Science Asia 32(1): 71-76. West, G.S. 1904. A Treatise on the British Freshwater Algae. Cambridge Uni- Pentecost, A. 1991. A new and interesting site for the Calcite-encrusted desmid versity Press. Cambridge. Oocardium stratum Naeg. in the British Isles. British Phycological Journal Wood, H.C. 1872: A contribution to the history of the fresh-water algae of North 26: 297-301. America. Smithsonian Contributions to Knowledge, 241. Washington. Prescott, G.W., Croasdale, H.T., Vinyard, W.C., Bicudo, C. 1981. A synopsis Yoneda, Y. (1939): Studies on the thermal algae of Hokkaido (1). Acta Phyto- of North American desmids. Part II. Desmidiaceae: Placodermae. Section taxonomica et Geobotanica 8: 101-107. 3. University of Nebraska Press. Lincoln. Rodriguez, S. & Vergon, J.P. 1996. Guide pratique de détermination généri- Associate Editors: Javier Fuertes / Santos Cirujano que des algues macroscopiques d’eau douce. Ministère de l’environne - Received: 19-VI-2012 ment. Paris. Accepted: 7-IV-2013

52 5.3. Sobre la presencia de Nostochopsis lobata Wood ex Bornet et Flahault en España: aspectos morfológicos, ecológicos y biogeográficos.

Título original: On the presence of Nostochopsis lobata Wood ex Bornet et Flahault in Spain: morphological, ecological and biogeographical aspects Autores: José Luis Moreno, Marina Aboal y Laura Monteagudo Publicación: Publicado en Nova Hedwigia, 85(3-4). Impact Factor: 0.809 DOI: 10.1127/0029-5035/2012/0052 Fecha 2012

RESUMEN

Durante muestreo habitual en ríos de la región, se detectó un cianófito

consideradoel ‘raro’ a escala europea, Nostochopsis lobata Wood ex Bornet et Flahault, siendo la primera cita publicada para España. El ejemplar se recolectó en el Arroyo Nava del Rey, un arroyo temporal que discurre sobre rocas metamórficas en una zona de altitud media. En este trabajo se compara la morfología y ecología descrita para la especie en regiones y se revisa su distribución

biogeográfica. En cuanto a morfologíadiferentes y preferencias de hábitat, el ejemplar encontrado en el Arroyo Nava del Rey resultó ser similar a aquellos especímenes detectados en Europa (Francia, Italia y Portugal). Sin embargo, difería de aquellos descritos en zonas no europeas. En cuanto a su distribución geográfica, la revisión de la bibliografía y de la base de datos de Biodiversidad Global ( ) reveló

que N. lobata es una especie cosmopolita, ya que GBIF en todas las zonas climáticas a excepción de los polos (tropical setemplada, ha detectado árida y fría). Sin embargo, más allá de las zonas tropicales de forma mucho

más ocasional en el tiempo y en el espacio, ha aparecido detectada en la mayoría de los casos una única vez. Además, cabehabiendo destacar sido que en las zonas templadas la abundancia y tamaño de las colonias suele ser mucho menor, lo que las hace más difíciles de detectar en campo. Un estudio ecológico y genético más profundo permitiría comprender mejor la variabilidad de esta especie.

53 54 Nova Hedwigia Vol. 95 issue 3–4, 373–390 Article published online September 7, 2012

On the presence of Nostochopsis lobata Wood ex Bornet et Flahault in Spain: morphological, ecological and biogeographical aspects

Jose Luis Moreno1*, Marina Aboal2 and Laura Monteagudo1 1 Centro Regional de Estudios del Agua (CREA), Universidad de Castilla-La Mancha, Crtra. de las Peñas, km3, Albacete-02071, Spain 2 Departamento de Biología Vegetal, Universidad de Murcia, Campus Espinardo, Murcia, Spain

With 5 figures and 2 tables

Moreno, J.L., M. Aboal & L. Monteagudo 2012: On the presence of Nostochopsis lobata Wood ex Bornet et Flahault in Spain: morphological, ecological and biogeographical aspects. – Nova Hedwigia 95: 373–390. Abstract: During a general biological assessment in rivers of southern Spain, a rare Cyanophyte Nostochopsis lobata Wood ex Bornet et Flahault was detected for the first time in this country. The species was collected in a temporary stream located in low altitude metamorphic mountains (Nava del Rey stream). Morphological comparisons, as well as an ecological and biogeographical review of the species are provided in this work. N. lobata from Spain shares similar morphology and habitat preferences to specimens previously reported from Europe (France, Italy and Portugal). However, when compared with specimens collected from non European sites, differences were detected. In terms of geographical distribution, this study revealed that N. lobata is a cosmopolitan species, distributed worldwide throughout tropical, temperate, arid and cold climate zones, with the exception of the polar regions. The presence of this species in cold and dry climates was most restricted. However, outside the tropical climate zone, the species shows a very sparse occurrence in space and time, being collected in the majority of cases only once per site. Additionally, in temperate climates, the colony size is rather small, the abundance is much lower and the colonies are very hard to detect in the field. Further research on ecological and genetic aspects is needed in order to obtain comprehensive knowledge of this variable species. Key words: Blue-green algae, biogeography, environmental conditions, temporary stream, tropical species.

Introduction

The Cyanophyte genus Nostochopsis Wood ex Bornet et Flahault is included in the Nostochopsaceae family (Bourrelly 1985, Komárek et al. 2003) and, like most of the

*Corresponding author: Jose Luis Moreno; e-mail: [email protected]

55 genera of Stigonematales, is monospecific (Anagnostidis & Komárek 1990). A number of other Nostochopsis species have been described, including N. radians described by Bhâradwâja (1934) and also reported by Desikachary (1959) and N. wichmannii Weber van Bosse 1913 (Yoneda 1939, Hirano & Hirose 1977). However these species have subsequently been accepted as synonyms of N. lobata. Nostochopsis lobata was first described in 1869 by Wood (in Wood 1872) in the temperate climatic zone around Philadelphia, though traditionally it has been mainly considered a tropical species rarely found in temperate zones (Frémy & Feldmann 1934, Sarma & Chapman 1975, Anagnostidis and Komárek 1990). In fact, it is widely distributed in tropical areas worldwide, including Africa (Frémy 1929), South America (Cáceres 1973, Branco et al. 2001), Asia (Geitler 1930-1932, Watanabe & Komárek 1988, Peerapornpisal et al. 2006, Aziz 2008) and Oceania (Sarma & Chapman 1975, Sheath & Cole 1996). However, the number of records of N. lobata in climatic zones other than the tropics has increased in recent decades. Thus, according to Komárek (2003) the species is distributed within temperate, subtropical and tropical regions, but it is mainly confined within the tropics and subtropics (Anagnostidis and Komárek 1990). In Europe, there are only two published records of Nostochopsis lobata previous to this study: one from Banyuls sur Mer in France (Frémy & Feldmann 1934) and the other from L’Aquila in Italy (Del Grosso 1975). Hansgirg (1892) reported this species in Bohemia (Czechoslovakia) under the name of Nostochopsis lobatus var. stagnalis, but some authors have doubted the validity of this identification (Frémy and Feldmann 1934, Bourrely 1985, Anagnostidis and Komárek 1990). It might be noted that Hoffmann (1990) found a specimen in Corsica which had originally been assigned to Mastigocladopsis jogensis but subsequently identified as Nostochopsis sp. (Gugger & Hoffmann 2004). In addition, The Portuguese Algal Collection includes a strain of N. lobata (http://www.gbif.org/) but no publications describing this finding have been released. During a general monitoring study in rivers of south-central Spain, several specimens of N. lobata were found among the material collected in a temporary stream, Nava del Rey stream. The main objective of this study was to review and compare the morphology, ecological conditions and geographical distribution of this rare species.

Materials and methods

Sampling and processing: Since the year 2001 more than three hundred stream reaches in central Spain have been surveyed repeatedly for aquatic macrophytes (i.e. submerged and floating aquatic vegetation including macroalgae, bryophytes, pteridophytes and phanerogams). The stream reaches surveyed were 100m long. Macrophytes were collected by hand and fixed in the field with 4% formalin. This material was examined in the laboratory under a Leica M165C stereoscope and an Olympus BX50 light microscope, both equipped with a Leica DFC420C digital camera. Measurements and photomicrographs were taken and Glycerin-gelatin was used to make permanent slides.

Physico-chemical parameters: Physico-chemical analyses were performed and environmental conditions in the Nava del Rey stream were measured on four dates in order to cover seasonal changes in the aquatic environment (Table 1). Water flow (discharge) was calculated by integrating depth and current velocity sections (Platts et al., 1983) measured at regular distances along three random transects, and using a MiniAir© current meter. Electrical conductivity, pH and dissolved oxygen

56 were measured in situ using appropriate sensors (Multiline P4 WTW). Alkalinity was also obtained in field using the sulphuric acid method (APHA, 1998). One water sample per date was collected in polyethylene bottles (500 ml) after filtration using a syringe with a Swinnex filter holder (Millipore 0.7 µm) and was kept in the refrigerator at 4°C for analysis, always within 48 h after sampling.

- - + -3 Nutrient concentration (N-NO3 , N-NO2 , N-NH4 , and P-PO4 ) was determined photometrically with MERCK kits (Spectroquant©). Detection levels for nutrients were: 0.5 mg L-1 for nitrate, 0.05 mg L-1 for nitrite and 0.03 mg L-1 for ammonium and phosphate. In the case of nitrate analysed on the last date (Table 1), the detection level was decreased to 0.01 mg L-1 performing the cadmium reduction method (APHA, 1998). Ion chromatography was used to analyse chloride and sulphate; while total organic carbon (TOC) and inorganic carbon (IC) were analysed using infra-red spectrophotometry; turbidity with a turbidimeter TN-100 (Eutech Instruments; infrared light); calcium and magnesium by complexometry (volumetry); and sodium and potassium by atomic emission. All these parameters were analysed following standard procedures detailed in APHA (1998).

Biogeographycal approach: For the biogeographical study of this species, references included in both the scientific literature and the Global Biodiversity Information Facility (GBIF) were compiled and a data base was created. Finally, this data set was combined in ArcMap (ESRI), with updated Köppen-Geiger climate classification maps provided by the Food and Agriculture Organization of the United Nations (http://www.fao.org/sd/EIdirect/climate/EIsp0002.htm) and by Kottek et al. (2006), for the periods 1951 to 1976 and 1976 to 2000.

Results

Study site, ecological conditions and habitat: N. lobata was found among the material collected in the Nava del Rey stream sampling site (Fig. 1) (UTM: X = 429700, Y = 4265000) at medium altitude (590 m) (Fig. 2), a temporary warm stream located in the southern part of the Iberian Peninsula (Sierra Morena mountains, province of Ciudad Real, Autonomic Community of Castilla-La Mancha). On average, the stream was 3 m wide and 20 cm deep . The sampling site was located 9.5 km from the source (2nd order; upstream drainage area: 41.76 km2). The stream is a tributary of one of the largest rivers in the Iberian Peninsula, the Guadalquivir River. This stream runs over Palaeozoic siliceous metamorphic rocks (quartzites, slates and shales) (González & Vázquez 2000). The climate is Mediterranean semiarid with a mean annual precipitation of 600 mm (Fernández 2000). The main land use is extensive farming for cattle, and the landscape corresponds to Spanish "dehesa", a kind of managed Mediterranean maquia with Quercus ilex L. as the main tree species. The percentages of land uses in the drainage area were 81.23% for forestry, 18.62% for dry land agriculture and 0.15% for urban use. Spring-summer water temperature ranged from 18–27°C, which is typical of warm water, Mediterranean streams located at low altitudes in a semiarid climate. Discharge was low (0–70 L s-1) showing a seasonal flow pattern which varied from high spring flow to summer isolated pools; the wet stream bed ranged from 1 m to 6 m wide. Phy- sico-chemical characteristics of water included low conductivity values around 200 µS cm-1 due to the presence of siliceous rocks and pH values around neutral or slightly basic. According to its nutrient load, the water can be classified as oligo-mesotrophic waters (Table 1). Several specimens of N. lobata were found attached as epiphytes on a dead stem of moss as well as free floating fragments entangled with other macrophytes. The species

57 Table 1. Environmental features of the Nava del Rey Stream (Central Spain) on four dates. "-" no data.

Sampling date 09/06/2009 28/04/2010 05/08/2010 13/06/2011

Nostochopsis lobata present not detected not detected not detected flow conditions base flow high flow isolated pools base flow

Field measurements (in situ) wet channel width range (m) 1–4 1.5–6 1–3 1–5 discharge (L s-1) 30 70 0 40 conductivity 20º (µS cm-1) 237 192.70 262 180.8 dissolved oxygen (mg L-1) 6.84 9.11 6.41 10.16 dissolved oxygen (% saturation) 82.6 117.50 87.3 142.5 pH 7,47 8.30 7.27 9.33

T (ºC) 18.8 24.50 26.5 29.1 total alkalinity (mg L-1) - 104.00 92 104.0

Laboratory analysis

+ -1 ammonium (N-NH4 mg L )- < 0.03 0.03 0.042

- -1 nitrite (N-NO2 mg L ) - < 0.05 0.063 0.004

- -1 nitrate (N-NO3 mg L ) - < 0.50 0.34 0.33 ortophosphate -3 -1 - - 0.198 0.001 (P-PO4 mg L ) chloride (mg L-1) - 15.44 - 12.19 sulphate (mg L-1) - 41.50 - 30.17 potassium (mg L-1) - 1.27 - 2.02 sodium (mg L-1) - 8.57 - 8.67 calcium (mg L-1) - 23.00 - 25.97 magnesium (mg L-1) - 10.00 - 8.08

TOC (C mg L-1) - 2.44 - 7.69 turbidity (UNF) - 2.79 0.48 1.58

IC (C mg L-1) - 9.26 - 9.83

58 Accompanying aquatic flora Ranunculus R. peltatus R. peltatus R. peltatus (macrophytes) peltatus Tetrasporidium Hydrodyction Cladophora sp. javanicum H. reticulatum reticulatum Möbius Spirogyra sp. Cladophora sp. Cladophora sp. Cladophora sp.

Zygnema sp. Spirogyra sp. Oedogonium sp. Nostoc sp. Oedogonium Stigeoclonium Spirogyra sp. Spirogyra sp. sp. sp. Stigeoclonium Oedogonium sp. Zygnema sp. sp. Melosira sp. Anabaena sp.

Oscillatoria sp.

was associated with aquatic plants living in depositional habitats of pools or stream banks. Abundance in the field was very low since the species was not detected with the naked eye and only one small stem with Nostochopsis was found within a 100 m stream reach.

Associated flora: Nostochopsis lobata was found only in June 2009, when the stream showed its typical base flow conditions. The associated aquatic plants and macroalgae were represented mainly by Cladophora sp. mats along with other filamentous green algae such as Zygnematales and Oedogonium sp., and also irregularly dispersed stands of Ranunculus peltatus Shrank (Table 1). In April 2010 discharge was the highest, corresponding to spring flow, but N. lobata was not detected despite the high diversity of the aquatic vegetation. In August 2010 the stream was a mosaic of isolated pools nearly chocked with high biomass of chlorophytes Hydrodyction reticulatum (L.) Bory, Oedogonium sp. and Zygnematales, with colonies of Cyanophytes (Anabaena sp., Nostoc verrucosum (Vaucher) Bornet & Flahault) and stands of Ranunculus peltatus also dispersed along shallow pools. N. lobata was not recorded. Finally, we returned to the site in June 2011 (the same month that it had previously appeared) but the species was not detected (Table 1). In this last date the stream showed the base flow conditions with Cladophora sp. and Zygnematales as the predominant aquatic flora.

Morphology: Several compact young colonies were found attached to a small dead moss stem while numerous senescent fragments were found free floating in the sample. Young colonies were globular and irregularly lobed, measuring 0.5–3 cm in diameter (Table 2, Figs 3a, 3b and 4a). Additionally, numerous gelatinous diffluent fragments were found entangled with filamentous green algae such as Cladophora sp. or Zygnematales, and also in stands of R. peltatus (Figs 3c and 3d). The size of these fragments were smaller than young colonies, measuring around 0.1–1 cm. The colour of the colonies after formalin fixation was rusty-brownish in young specimens, due to ferric precipitates, and bluish in the case of free floating fragments (Fig. 3a, 3b, 3c and 3d).

59 Fig. 1. Location of study site: Nava del Rey Stream, southern Spain. In Guadalquivir River, water flows from east to west.

In young colonies as well as in senescent fragments, filaments were uniseriate with true branches and were embedded in mucilage which contained occasional filaments of Calothrix fusca Bornet. Two different regions were clearly observed in transversal section of young colonies (Fig. 4b). The internal region was characterized by an intricate mass of main trichomes (Fig. 3e) from which several branches emerged. Main trichomes were composed of barrel-shaped and sometimes torulose cells measuring from 1.8–4 µm in diameter and from 1.9–8.3 µm in length (Table 2; Figs 4c and 4d). In this region, branches were sinuous to spiral in arrangement, formed by cylindrical cells measuring from 1.4–2.9 µm in diameter and from 4.6–12 µm in length (Table 2). The layout of main trichomes was gradually radial in the external region of young colonies. In this case, cells composing main trichomes were slightly smaller (from 1.6–3.4 µm in diameter and from 2.7–6.9 µm in length) than those in the internal region but similar in shape (Table 2). The same pattern was observed when comparing cells from external branches with those from internal branches, the cell shape was similar but a decrease in size was observed, particularly in cell length, since measurement of cells from external branches was 1.3–2.6 µm in diameter and from 2.7–8.1 µm in length (Table 2). The layout of branches in the external region was slightly sinuous to straight (Fig. 3f). In both regions, T- type branching was predominant although some reverse V-shaped branches were also observed. Short branches frequently formed a right angle with main trichomes (Fig. 4d), while longer branches often formed an acute angle (Fig. 4c). Some variability in filament morphology was noted irrespective of the position inside the colony. On some occasions main trichomes, and branches which emerged from them,

60 Fig. 2. Study site: Nava del Rey Stream. a) General view. b) Habitat view.

were formed by almost isodiametrical cells measuring from 1.5–2.5 µm in diameter (Fig. 4d), probably representing the young and actively growing state of filaments. In senescent fragments, the cell-type of main trichomes was torulose and slightly elongated with a diameter range from 2.3–6.2 µm and a mean length of 7.2 µm (Fig. 4c). Branches were composed of cylindrical cells, longer than in young colonies, measuring up to 13 µm, with a diameter/length ratio of 1:5. In both young colonies and senescent fragments, filaments were gradually narrowed to the tip with end cells pointed, barrel or clavate shaped (Fig. 4f). Wide individual hyaline sheaths were clearly visible especially near the outer margin of the colony (Fig. 4e). Heterocytes were colourless and spherical to conical in shape, ranging from 3.5 µm– 8 µm in diameter and 3.2 µm–8 µm in length, being slightly bigger in senescent than in young specimens and more abundant in the central region. Intercalary heterocytes were rare (< 5%) while lateral ones were frequent. Both being sessile and stalked heterocytes were present (at the end of 1 to 4 celled branches) (Fig. 4c). Hormogonia were observed in the outer colony regions and were more abundant in senescent fragments. They comprised 2 to 10 isodiametrical cells which resulted from filament fragmentation (1:1 diameter/length ratio) (Fig. 4g).

Biogeography: The combined distributional records show that N. lobata has a largely worldwide distribution (Fig. 5). From a total of 96 sites included in the dataset, 52 (54%) were located within the tropics and 44 (45%) elsewhere. The northernmost and southernmost sites were located in Canada (53°35'12.04"N) and New Zealand (36°51'26"S) respectively. In terms of the Köppen-Geiger climatic zones, N. lobata has been recorded in 4 of the 5 main climate zones: tropical, temperate, dry and cold. Polar is the only climate in which this species has never been reported. As shown in Fig. 5, 89% of sites included in this study corresponded with tropical (46%) and temperate (42%) climates while only 10% corresponded with dry (5%) and cold (5%) climates. N. lobata was found in

61 Table 2. Morphological description and ecological observations of Nostochopsis lobata from Spain and other European specimens.

Reference and finding This study Frémy & Feldmann, 1934 Del Grosso, 1978 Hoffman, 1990 location Ciudad Real, Spain Banyuls sur Mer, France L’Aquila, Italy Corsica, France Life form senescent colony: young colony: solid globular thallus becoming spherical, subspherical or lobed similar to Nostoc or Rivularia; soft laminar solid, irregularly lobed, becoming hollow as it hollow when adult; colonies; hollow and irregularly lobed; fragments gets older; colour: brownish-rusty due to ferric colour: green-brownish colour: green-brownish colour: blue-green precipitate Colony diameter (cm) - 0.1–3 1–1.5 3–5 maximum 4

Main trichomes external zone internal zone external zone internal external zone internal - zone zone layout tangled radial tangled radial tangled radial tangled - cell shape torulose barrel-shaped barrel-shaped cylindri- isodiamtetrical cylindri- torulose; barrel-shaped to cal cal cylindrical cell diameter (µm) 2.3–6.2(3.4) 1.6–3.4(2.3) 1.8–4(2.6) 2–6 3–5 8 4–5 - cell length (µm) 4–13(7.2) 2.7–6.9(4.3) 1.91–8.3(5.1) 10–12 8 9–11 - cell ratio diameter:length 1:2.4 1:1.8 1:2 1:1 1:2.75 1:1 1:2 -

Branches T- and V- branching T- and V- branching T- branching T- and V- branching branches layout sinuous to spiral straight- sinuous-spiral sinuous - spiral helicoidal - sinuous cell form cylindrical cylindrical cylindrical cylindrical cylindrical cylindrical cell diameter (µm) 1.7–3.1(2.3) 1.3–2.6(1.8) 1.4–2.9(2) 2–3 2 2.0–4.8 cell length (µm) 7–16(11.2) 2.7–8.1(5.5) 4.6–12(8) 10 12 2.6–18.0 cell diameter:length 1:5 1:3 1:4 1:5 1:6 -

Trichomes ending gradually narrowed to the tip gradually narrowed to the tip gradually narrowed to the tip tapering and slightly pointed.

Heterocytes lateral, sessile and stalked; intercalary less than 5%; spherical to lateral, sessile and stalked; lateral, sessile and stalked; conical; colourless - intercalary rare; spherical to intercalary less than 5%; elongated; colourless spherical to ovate size (µm) 4–8(6.7) wide; 3.5–6.5(5) wide; 6–8 diameter 8 diameter 6.2–9.6 wide and 5.4–12.0 5–8(7.2) long 3.2–7.2(5) long long Sheath hyaline, sometimes visible, especially in the outer margin. pseudosheath and true sheath - hyaline, sometimes visible. 7–10 µm wide Mucilage (endophytic containing Calothrix fusca. containing Calothrix fusca - - species) Hormogonia 2–10 isodiametrical celled branches - 9–18 celled branches barrel-shaped cells branches size (µm) 2.2–3.2 wide , 2.2–3.9 long 3.4–4.2 wide and 2.4–3.8 long Ecological observations sampling date: 9 June; sampling date: 1 July; sampling date: 25 August; sampling date: September; water conditions: T = 26.5°C, pH = 7.27, oligo-mesotrophic waters; water temperature = 21°C; water conditions: T = 13°C, growing together with free floating (senescent) and epiphyte (young); associated to Nostoc linkia; pH = 7.4; Hildenbrandia rivularis; accompanying flora: filamentous green algae; attached to submerged rocks; associated with Cladophora attached to submerged rocks; slow water shallow stream slow water; glomerata; granitic region siliceous metamorphic rocks (shales, quartzites, slates) siliceous metamorphic rocks found at 1m depth in slow (shales) water calcareous stream

62 Reference and finding This study Frémy & Feldmann, 1934 Del Grosso, 1978 Hoffman, 1990 location Ciudad Real, Spain Banyuls sur Mer, France L’Aquila, Italy Corsica, France Life form senescent colony: young colony: solid globular thallus becoming spherical, subspherical or lobed similar to Nostoc or Rivularia; soft laminar solid, irregularly lobed, becoming hollow as it hollow when adult; colonies; hollow and irregularly lobed; fragments gets older; colour: brownish-rusty due to ferric colour: green-brownish colour: green-brownish colour: blue-green precipitate Colony diameter (cm) - 0.1–3 1–1.5 3–5 maximum 4

Trichomes ending gradually narrowed to the tip gradually narrowed to the tip gradually narrowed to the tip tapering and slightly pointed.

63 Fig. 3. Some images of Nostochopsis lobata. a) Habitus of a young colony. b) Fragment of a young colony. c) Fragments of a senescent colony. d) Young colony (left) and senescent fragments (right). e) General view of the inner of a young colony. f) Outer margin showing the layout of filaments (young colony).

15 of the 31 Köppen-Geiger sub-climatic classes, which were defined by temperature and precipitation conditions (Kottek et al 2006). Surprisingly, among the sites included in this study, the sub-climate with the highest percentage of appearance was temperate- fully humid with hot summers (25.00%) followed by tropical-winter dry (19.79%) and

64 Fig. 4. Nostochopsis lobata. a) Macroscopic view of a young colony; arrows represent the union point with a dead moss stem. b) Transversal section of a young colony; discontinuous line separates internal (tangled) and external (radial) layout of filaments. c) Mature filament with sessile and stalked heterocytes; branches are mostly in acute angle with main trichome. d) Filament in active growing; branches are in right angle with main trichome. e) Filaments with visible sheaths and formation of a T-type branch. Arrow indicates the septation of the branch-point cell. f) Different ends of filaments. g) Hormogones. Scale bar represents: 1500 µm for a); 500 µm for b); 20µm for c) and d); 5 µm for e), f) and g). tropical-fully humid (13.54%). N. lobata has been detected in all tropical sub-climate conditions. However, it has never been detected in 3 of the 9 temperate sub-climates: fully humid with cool summer, winter dry with cool summer and summer dry with

65 Fig. 5. World distribution of Nostochopsis lobata throughout main climates (Köppen-Geiger classification system): Tropical (black), Temperate (medium gray), Arid (light gray), Cold (striped) and Polar (white). Circles represent records supported by scientific publications; squares represent data obtained from the Global Biodiversity Information Facility (GBIF).

cool summer. Thus, temperate sub-climates in which N. lobata has not been detected are characterized by both cool summer and cold winter. The presence of this species in cold and dry climates was the most restricted. Only five references have been found corresponding with both climates. All cold-climate sites were located in North America while dry-climate sites were distributed among North America, Africa and Europe. Focusing its European distribution, N. lobata has been referred to from two different climates: temperate (France, Italy, and Portugal) and arid (Spain). In the case of Spain, the site was located in a sub-climate zone currently designated as steppe-cold arid (for the 1972–2000 period), thus constituting the first European arid zone in which N. lobata has been detected. However, it should be noted that a recent change in the climatic conditions in the Nava del Rey catchment was detected. The climate for the period 1950–1975 was classified as temperate according to the Köpen-Geiger classification.

Discussion

Morphology: Bhâradwâja (1934) described Nostochopsis radians as a new species of Nostochopsis differing from N. lobata by a much smaller, solid thallus, in the absence of special unseptate branches, and in the ends of the ultimate branches which were not club-shaped or clavate. Most recently, N. radians was reported from the Hawaiian Islands by Filkin et al (2003). These authors found both N. lobata and N. radians

66 growing in distant streams. They differentiated the two species by virtue of N. lobata colonies having hollow interiors, and increased branching towards the tip. Whereas N. radians appeared as a solid mat with branching that did not increase towards the tip. In terms of environmental conditions, both streams were similar in pH (8.1 and 8.5 respectively) and water temperature (22°C, 24°C respectively) although the stream where N. lobata was found was wider, deeper and with lower water conductivity (90 µS cm-1) than the stream with presence of N. radians (200 µS cm-1). Thus, even though this species is accepted by some authors (Desikachary 1959), differences between N. radians and N. lobata were not consistent enough to be considered a separate species, possibly due to the variability existing among different growth stages and environmental adaptations of Nostochopsis (Frémy and Feldmann 1934). Comparing our material with N. lobata specimens described worldwide, the greatest differences were found with those from New Zealand (Sarma & Chapman 1975), Australia (Skinner & Entwisle 2001), Thailand (Peerapornpisal et al 2006) and Bangladesh (Aziz 2008) with respect to heterocytes and branching. Sarma & Chapman (1975) and Aziz (2008) described branches formed by cells measuring up to 18 µm in length and emerging unilaterally whereas in the present study slightly smaller cells and both-direction branching were observed. Heterocytes described by Sarma & Chapman (1975) and Skinner & Entwisle (2001) showed a wide range of sizes reaching up to 12 µm in length, although such large heterocytes were not observed in the Spanish material from Nava del Rey. In contrast, Peerapornpisal (2006) described smaller spherical heterocytes ranging from 3–4 µm in diameter. According to Frémy & Feldmann (1934), Del Grosso (1975), Tiwari (1978) and Hoffmann (1990), intercalary heterocytes were rare, rounded and situated in main trichomes while Cáceres (1973) and Skinner & Entwisle (2001) described frequent intercalary heterocytes characterized by a cylindrical shape. Nevertheless, variability in frequency of intercalary heterocytes has been related with nitrogen content in culture (Tiwari 1978) with intercalary heterocytes suppressed in the presence of ammonium. Some discrepancies were found with respect to branching type. Various authors described T-branching ( Del Grosso 1978, Skinner & Entwisle 2001) while others also mentioned reverse V-shaped branching (Frémy and Feldmann 1934, Komárek 2003). However, a recent study by Gugger & Hoffmann (2004) revealed that V- or Y- shaped branches in Nostochopsis start out as typical T-branches and subsequent modifications in the branch-point cell could lead to a final V- or Y-shaped appearance. This observation supports the idea proposed by Begum et al. (1994) that Mastigocladopsis jogensis is a synonym of N. lobata since reverse V-shaped branching is not a stable characteristic and could be formed at a late stage of growth. A further comparison of our material with the other specimens collected within Europe (Frémy & Feldmann 1934, Del Grosso 1978, Hoffmann 1990) revealed few morphological differences. For instance, cells composing the main trichomes and branches in young colonies seemed to be slightly smaller and less elongated than those described by Frémy & Feldmann (1934) and Del Grosso (1978) while the opposite occurred for cells from senescent fragments (Table 2). Additionally, our material showed wider size ranges than other European specimens. This may be due to differences in

67 age and growth stage among the specimens observed. Frémy and Feldmann (1934), observed two different filament morphologies (Figs 4c and 4d). Filaments highly branched, formed by small isodiametrical cells (Fig. 4 d) which were observed forming spreading in small groups among young colonies. Therefore, these regions could be zones of active growth, supporting the hypothesis of Cáceres (1973) that growth in this species is produced by a lateral increase in the number of trichomes and not by centrifugal growth.

Habitat and environmental conditions: In terms of habitat, N. lobata shows a high degree of variability. It has been encountered in lakes (Tiwari 1977, El Saied 2007), running waters (Frémy & Feldmann 1934, Palmer 1941, Pandey & Pandey 2008), thermal springs (Yoneda 1939), as an epiphyte (Sarma & Chapman 1975, this study), attached to rocks (Hoffmann 1990, Peerapornpisal 2006), free floating (Wood 1872), on moist soils (Skinner & Entwisle 2001, Aziz 2008) and even on humid walls (Skinner & Entwisle 2001) among others. However the most extreme habitat where N. lobata has been found is probably as cryptoendolithic in an arid climate (Weber et al 1996). These authors found this species in South Africa growing together with Chroococcidiopsis sp. in the pore spaces inside sandstones. This is the only place where N. lobata has been found growing as cryptoendolithic and the strain isolated by the authors (SAG 2.97) has been used in geobiological (Büdel 1999, Büdel et al 2004) and genetic studies (Gugger & Hoffmann 2004). In this case, the polyphyly of true branching cyanophytes was confirmed as well as the possible need for including Westiellopsis, Fischerella and Nostochopsis in the same genus, due to sequence similarity above 95%. These genera show great morphological variability depending on growth stage and environmental conditions, thus the differentiation of these genera by means of traditional alpha taxonomy, based mainly on morphological characteristics, can lead to possible misidentifications. Polyphasic studies are necessary in order to avoid this problem. Additionally, genetic analyses are also necessary to test the monospecific character of the genus Nostochopsis. Focusing on European reports, there are some similarities in environmental conditions of sites: (1) in all cases the collection took place in summer in shallow warm waters, (2) all streams run over siliceous metamorphic rocks, except the Tirino stream (Italy) (Table 2). Thus, the underlying lithology could be a very important factor conditioning the presence of N. lobata. This is supported by the work of Branco et al. (2001) in Sao Paulo State (Brazil) who suggested that N .lobata shows a preference for low conductivity waters, although further research on ecological preferences of this species is required. With respect to the environmental conditions in Nava del Rey stream, N. lobata was not found in three additional site visits (Table 1). Comparing the two dates in June, some differences in pH, temperature, dissolved oxygen and conductivity were apparent. A narrow range of these ecological factors may define the restricted ecological niche where this species grows and, therefore, explain why this species and other Stigonematales are absent most of the time from its type localities (Hoffmann 1996). We can conclude that only occasionally the studied stream achieves the restricted ecological conditions required by N. lobata, during a short time in the summer.

68 Finally, it must be highlighted that the interest in nutritive and pharmacological properties of N. lobata is rapidly growing, especially in Asia (e.g. Peerapornpisal 2006, Pandey & Pandey 2008), which could increase the knowledge of this interesting species.

Biogeography: In terms of geographic distribution, N. lobata can be considered as a (sub-) cosmopolitan species since it is distributed worldwide, being found in tropical, temperate, arid and cold climatic zones, i.e., in all climates except polar (Fig. 5). However, some differences in form exists among climatic zones. For example, in several tropical zones as well as adjacent temperate sub-climatic zones, N. lobata grows as to macroscopic proportions (more than 10 cm after Peerapornpisal et al 2006) and is locally abundantly, and used as a local food and medical resource (Peerapornpisal et al 2006, Pandey & Pandey 2008). In contrast, in temperate zones, N. lobata grows as small spherical colonies which are difficult to detect in the field. Some examples include specimens from Philadelphia described by Wood (1872) measuring about 4 cm; Banyuls sur Mer (France) 1–1.5 cm according to Frémy & Feldmann (1934) or those from New Zealand described by Sarma & Chapman (1975) measuring up to 0.8 cm in diameter. This morphologic pattern is shown by some Cyanophytes which have many specialized types restricted to small defined habitats and localities but with morphologies that can be uniform in one area and variable in another (Komárek 2003). In the case of N. lobata from Spain, visits to the sampling site were repeated and the species was not found again. Most of the stigonematalean genera are monospecific and known only from their first discovered biotope (Anagnostidis & Komárek 1990). Furthermore, it has been stated that majority of stigonematalean species have worldwide distribution but usually occur in very limited and special conditions or under limited ecological events, that is, in disjunct areas, in spite of their wide geographical distribution (Anagnostidis & Komárek, 1990). For example, species associated to restricted niches or habitats like thermal springs typically show a mosaic-like geographical distribution reflecting the corresponding distribution of ecological niches (Hoffmann 1996). Thus, despite its worldwide distribution and its climatic tolerance, N. lobata may show an underlying tendency: it grows abundantly in tropical regions where its optimal climatic conditions are stable for several months whereas it grows scarcely in other regions only when conditions allowing its germination and growth are reached, probably for a few weeks in occasional years. This observation along with its small size and low abundance could also explain why this species is hard to detect in temperate sites where it has been previously found.

Acknowledgements

This study has been financed by Project PO1109-0190-8090 of the Ministry of Education and Science, Autonomic Community of Castilla-La Mancha.

69 References

ANAGNOSTIDIS, K. & J. KOMAREK 1990: Modern approach to the classification system of Cyanophytes. 5. Stigonematales. – Algol. Stud. 59: 1–73. APHA. 1998: Standard methods for the examination of water and wastewater. 20th Edition. – American Public Health Association, American Water Works Association and Water Environment Association, Baltimore. AZIZ, A. 2008: Algal flora of Madhabkunda waterfall area in Maulvi Bazar, Bangladesh. III. – New records of blue-greens and greens. – Bangladesh J. Bot. 37(1): 43–48. BEGUM, Z.N.T., R. AKHTER, A.K.M.N. ISLAM & A. AZIZ 1994: Taxonomy of the Stigonema- talean algae in culture. – In Phang S.M., L.Y. Kun, M.A. Borowitzka & B.A. Whitton (eds.): Algal Biotechnology in the Asia-Pacific Region: 257–262. Univ. Malaya. BHARADWAJA, Y. 1934: A new species of Nostochopsis (Nostochopsis radians sp. nov.). – New Phytol. 33(1): 1–7. BORNET, E. & C. FLAHAULT 1886–1888: Révision des Nostocacées heterocystées contenues dans les principaux herbiers de France. – Ann. Sci. Not. Bot., 7 Ser., 3: 323–381. BOURRELLY, P. 1985: Les algues d’eau douce. Initiation à la Systématique. Tome III: Les algues bleues et rouges, Les Eugléniens, Peridiniens et Cryptomonadines. – Société Nouvelle des Éditions Boubée, Paris. BRANCO, L.H.Z., O. NECCHI JR., & C.C.Z. BRANCO 2001: Ecological distribution of Cyanophyceae in lotic ecosystems of Sao Paulo State. – Rev. Brasil. Bot. 24(1): 99–108. Budel, B. 1999: Ecology and diversity of rock-inhabiting cyanobacteria in tropical regions. – European J. of Phycology 34: 361–370. Budel, B., B. Weber, M. Kuhl, H. Pfanz, D. Sultemeyer & D. Wessels 2004: Reshaping of sandstone surfaces by cryptoendolithic cyanobacteria: bioalkalization causes chemical weathering in arid landscapes. – Geobiology 2: 261–268. CÁCERES, E.J. 1973: Contribución al conocimiento de las cianófitas. I. Dos novedades para la Argentina: Nostochopsis lobatus (Nostochopsidaceae) y Rivularia rufescens (Rivulariaceae). – Kutzingiana 7: 165–179. DEL GROSSO, F. 1977: Nostochopsis lobatus Wood: prima segnalazione per l’Italia. – Informatore Botanico Italiano 9(3): 285–286. DESIKACHARY, T.V. 1959: Cyanophyta. – Monographs on Algae, Indian Council of Agricultural Research, New Delhi. pp. 686. Dominic, T.K. & P.V. Madhusoodanan 1999: Cyanobacteria from extreme acidic environments. – Current Science 77: 1021–1022. El Saied, H.E. 2007: Molecular genetic monitoring of bacterial communities in Manzala Lake, Egypt, based on 16S rRNA gene analysis. – Egyptian Journal of Aquatic Research 33: 179–194. FERNÁNDEZ, F. 2000: Los condicionantes climáticos del paisaje. – In: Guía de Castilla-La Mancha: 41–54. Junta de Comunidades de Castilla-La Mancha, Toledo, Spain. Filkin, N.R., A.R. Sherwood & m.L. Vis 2003: Macroalgae from 23 Stream Segments in the Hawaiian Islands. – Pacific Science 57: 421–431. FRÉMY, P. & J. FELDMANN 1934: A propos de la présence en Europe du Nostochopsis lobatus Wood. – Bull. Soc. Bot. France 81(7/8): 612–618. FRÉMY, P. 1929: Les Myxophycées de l’Afrique Équatoriale Française. –Arch. Bot. Mém. 2: 1–508.

70 GEITLER, L. 1930–1932: Cyanophyceae von Europa. Dr. L. Rabenhorst’s Kryptogamen-Flora von Deutschland, Österreich und der Schweiz. – Koeltz Scientific Books, Leipzig. GONZÁLEZ, A. & A. VÁZQUEZ 2000: Las formas del relieve: llanuras, páramos y montañas. – In: Guía de Castilla-La Mancha: 17–40. Junta de Comunidades de Castilla-La Mancha, Toledo, Spain. GUGGER, M.F. & L. HOFFMANN 2004: Polyphyly of trae branching cyanobacteria (Stigonematales). – Int. J. Syst. Evol. Microbiol. 52: 1577–1595. HANSGIRG, A. 1892: Prodromus der Algenflora von Böhmen. – Arch. Natur-Landesk. Böhm. 8: 1–268. HIRANO, M. & H. HIROSE 1977: Order Stigonematales. – In: HIROSE, H. & T. YAMAGUISHI (eds.): Illustrations of the Japanese freshwater algae: 141–151. Uchidarokakuo Publ. Co., Tokyo. HOFFMANN, L. 1990: Presence of Mastigocladopsis jogensis (Cyanophyceae, Mastigocladopsiceae) in Corsica (France). – Cryptogamie, Algol. 11(2): 219–224. HOFFMANN, L. 1996: Geographic distribution of freshwater blue-green algae. – In: KRISTIANSEN J. (ed.): Biogeography of Freshwater Algae: 33–40. Kluwer Academic Publishers, Belgium. KOMÁREK, J., H. KING & J. KOMÁRKOVÁ 2003: Filamentous Cyanobacteria. – In: WHER, J.D. & R.G. SHEATH, (eds.): Freshwater algae of North America. Ecology and Classification: 117–196. – Elsevier Science, Academic Press, San Diego. Kottek, M., J. Grieser, C. Beck, B. Rudolf & F. Rubel 2006: World Map of the Köppen- Geiger climate classification updated. – Meteorol. Z. 15: 259–263. NECCHI JR., O., C.C.Z. Branco & L.H.Z. Branco 2000: Distribution of stream macroalgae in Sao Paulo State, southeastern Brazil. – Algol. Stud. 97: 43–57. Palmer, C.M. 1941: Additional records for algae, including some of the less common forms. – Butler University Botanical Studies 5: 224–234. Pandey, U. & J. Pandey 2008: Enhanced production of biomass, pigments and antioxidant capacity of a nutritionally important cyanobacterium Nostochopsis lobatus. – Bioresource Technology 99: 4520–4523. PEERAPORNPISAL, Y., D. AMORNLEDPISON, C. RUJJANAWATE, K. RUANGRIT & D. KANJANAPOTHI 2006: Two endemic species of macroalgae in Nan River, Northern Thailand, as therapeutic agents. – Science Asia 32 Supplememt 1: 71–76. Peres, C.K., C.C.Z. Branco & R.A. Krupek 2008: Macroalgas de riachos da Serra da Prata, leste do Estado do Paraná, Sul do Brasil. – Acta Bot. Bras. 22. PLATTS, W.S., W.F. MEGAHAN & G.W. MINSHALL 1983: Methods for evaluating stream, riparian, and biotic conditions. – U.S Government Printing Office, Ogden, UT. SARMA, P. & V.J. CHAPMAN 1975: Occurrence of the genus Nostochopsis in New Zealand. – Nova Hedwigia 26: 305–308. SHEATH, R.G. & K.M. COLE 1996: Stream macroalgae of the Fiji Islands: A preliminary study. – Pacific Science 50: 46–54. Skinner, S. & T.J. Entwisle 2001: Non-marine algae of Australia: 1. Survey of colonial gelatinous blue-green macroalgae (Cyanobacteria). – Telopea 9(3): 573–599. TIWARI, D.N. 1978: The heterocysts of the blue-green alga Nostochopsis lobatus: effects of cultural conditions. – New Ohytol. 81: 653–656. WATANABE , M. & J. KOMÁREK 1988: Blue-green algae from Kathmandu. – WATANABE, M. & B.S. MALLA (eds.): Cryptogams of the Himalayas, I: Kathmandu valley: 1–13. Dep. Bot. Nat. Sci. Mus. Tsukuba, Japan.

71 WEBER, B., D.C. J. WESSELS & B. BÜDEL 1996: Biology and ecology of cryptoendolithoc cyanobacteria of a sandstone outcrop in the Northern Province, South Africa. – Algol. Stud. 83: 563–579. WOOD, H.C. 1872: A contribution to the history of the fresh-water algae of North America. – Smithsonian Contributions to Knowledge 241, Washington. YONEDA, Y. 1939: Studies on the thermal algae of Hokkaido (1). – Acta Phytotax. Geobot. 8: 101–107.

Manuscript received September 15, 2012; accepted February 2, 2012.

72 5.4. Eutrofización fluvial: Comparación del impacto del regadío y secano a través de diferentes escalas espaciales.

Título original: River eutrophication: Irrigated vs. Non-irrigated agriculture through different spatial scales Autores: Laura Monteagudo, José Luis Moreno y Félix Picazo Publicación: Publicado en Water Research, 46. Impact Factor: 4.655 DOI: 10.1016/j.watres.2012.02.035 Fecha 2012

RESUMEN

En este trabajo se analizó la relación entre los diferentes usos de suelo y la eutrofización de ríos y arroyos en la zona de estudio, haciendo especial hincapié en conocer cómo afecta la elección de la escala espacial de estudio a los resultados. Se calculó la composición porcentual de usos de suelo (agrícola, urbano y forestal) y se midió la concentración de nitrato, amonio y ortofosfato en 130 puntos de muestreo repartidos por la región. Además, para evaluar qué tipo de agricultura es responsable en mayor medida del aumento de nutrientes en el agua, se dividió el uso agrícola en tres subclases (regadío, secano y agricultura de bajo impacto). Posteriormente se hicieron correlaciones entre la concentración de nutrientes en el agua y los usos de suelo obtenidos a través de cuatro escalas espaciales diferentes: dos amplias (área de drenaje total y corredores de 100m de anchura) y dos locales (zonas de influencia de 1 y 5 kilómetros de radio). Los resultados mostraron que utilizar diferentes escalas espaciales puede producir resultados diferentes y, por tanto, puede llevar a conclusiones muy distintas. En el caso de escalas amplias, la autocorrelación espacial y la representación insuficiente de algunos usos de suelo produjeron resultados irreales. Por el contrario, con el uso de las escalas locales no surgió el inconveniente de la autocorrelación y los usos de suelo estuvieron representados de manera más uniforme. La escala de local de 1km radio aguas

73 arriba resultó ser la más apropiada para detectar eutrofización en el sistema fluvial de Castilla-La Mancha, con el regadío como principal causante del aporte de nitrato al agua. Además, se realizó un análisis de regresión segmentada mediante el cual se detectó un umbral del 50% de regadío en la escala de 1km de radio. Este resultado tiene una clara implicación en la gestión de la calidad del agua: para impedir que aumente la eutrofización de ríos y arroyos es necesario evitar que haya zonas con más de un 50% de regadío en 1 km de radio desde el cauce. Por último, se describió una metodología base para elegir la escala espacial más apropiada en el estudio de la eutrofización causada por contaminación difusa.

74 water research 46 (2012) 2759e2771

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/watres

River eutrophication: Irrigated vs. non-irrigated agriculture through different spatial scales

Laura Monteagudo a,*, Jose´ Luis Moreno a,Fe´lix Picazo b a University of Castilla-La Mancha, Regional Center of Water Research (CREA), Crtra. de Las Pen˜as km3, Albacete 02071, Spain b University of Murcia, Department of Ecology and Hydrology, Murcia, Spain article info abstract

Article history: The main objective of this study was to determine how spatial scale may affect the results Received 5 October 2011 when relating land use to nutrient enrichment of rivers and, secondly, to investigate which Received in revised form agricultural practices are more responsible for river eutrophication in the study area. Agri- 26 January 2012 culture was split into three subclasses (irrigated, non-irrigated and low-impact agriculture) Accepted 16 February 2012 which were correlated to stream nutrient concentration on four spatial scales: large scale Available online 28 February 2012 (drainage area of total subcatchment and 100 m wide subcatchment corridors) and local scale (5 and 1 km radius buffers). Nitrate, ammonium and orthophosphate concentrations Keywords: and land use composition (agriculture, urban and forest) were measured at 130 river reaches Stream in south-central Spain during the 2001e2009 period. Results suggested that different spatial Nitrate scales may lead to different conclusions. Spatial autocorrelation and the inadequate repre- Land use sentation of some land uses produced unreal results on large scales. Conversely, local scales Irrigation did not show data autocorrelation and agriculture subclasses were well represented. The Catchment local scale of 1 km buffer was the most appropriate to detect river eutrophication in central Spanish rivers, with irrigated cropland as the main cause of river pollution by nitrate. As regards river management, a threshold of 50% irrigated cropland within a 1 km radius buffer has been obtained using breakpoint regression analysis. This means that no more than 50% of irrigation croplands should be allowed near river banks in order to avoid river eutrophication. Finally, a methodological approach is proposed to choose the appropriate spatial scale when studying river eutrophication caused by diffuse pollution like agriculture. ª 2012 Elsevier Ltd. All rights reserved.

1. Introduction proliferation of filamentous algae which leads to a decrease in dissolved oxygen levels, water quality, fish death and often to Land use transformation from natural to human dominated a global loss of biodiversity (Carpenter et al., 1998; Quinn, 1991; systems has globally caused environmental impacts, making Smith et al., 1999). In order to plan effective prevention land use analysis a useful indicator of changes in stream measures against eutrophication, a deep understanding of the ecosystems (Meyer and Turner, 1994). One of the most relationship between land cover and stream conditions is extended impacts produced by landscape modification is needed. However, this relationship is characterized by eutrophication of rivers, lakes and marine ecosystems (Lund, numerous factors and environmental variables operating and 1967; Omernik et al., 1981; Smith, 2003). Eutrophication interacting through different scales, both spatial and temporal produces changes in community composition and the (Frissell et al., 1986). Thus, several questions may arise when

* Corresponding author. Tel.: þ34 967599200x2576; fax: þ34 967599269. E-mail address: [email protected] (L. Monteagudo). 0043-1354/$ e see front matter ª 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2012.02.035 75 2760 water research 46 (2012) 2759e2771

assessing the relationship between stream conditions and whereas others have argued that land uses located closer to surrounding landscape. For instance: 1) which land use classes the stream (i.e. reach scale, local buffers, riparian corridors) may be analysed? 2) which spatial scale/s would be the most are more important (Harding et al., 1998; Nerbonne and appropriate (e.g. catchment, riparian buffers, circular buffers, Vondracek, 2001). Recent studies suggest that a multiple among others)? and 3) how may this spatial perspective spatial scale approach could be a necessary step to identify the influence the interpretation of results? In this paper, we have appropriate framework (e.g. Chang, 2008; Tran et al., 2010). In studied the influence of land uses on river nutrient concen- a multi-scale approach, different spatial scales would provide tration focussing on different agricultural practices (Section different land use datasets and therefore different results. In 1.1). In addition, we have analysed how different spatial scales order to correctly interpret the results, some issues must be may affect the results and how to identify which scale is the taken into account to select the appropriate spatial scale: (1) most appropriate in this type of studies (Section 1.2). spatial autocorrelation between sites and (2) the lack of a wide range of values (%) for each land use class. Autocorrelated 1.1. Agricultural land use data violate the assumption of independence of most standard statistical procedures such as correlation analysis Regarding the first issue, the land uses most commonly (Legendre, 1993). In many cases, when working with large studied as causes of river eutrophication have been both scales (e.g. subcatchment drainage area or subcatchment agriculture and urban (Osborne and Wiley, 1988; Townsend riparian corridors) autocorrelation is caused by overlapping of et al., 1997; Von Schiller et al., 2008). Whereas urban efflu- subcatchment drainage areas of consecutive downstream ents cause point pollution, agriculture often causes the most sites. The effect of overlapping on statistical analysis is also important source of non-point or diffuse pollution worldwide. known as pseudoreplication (Hurlbert, 1984) and the conse- However, the level of eutrophication impact depends on the quence is an increase in the statistical power (higher corre- kind of agriculture, for instance irrigated cropland vs. non- lations than expected) but not in the biological or ecological irrigated. Irrigation allows crops to grow in permanently significance (Townsend et al., 1997). water-scarce or temporarily water-stressed environments, The second issue is a consequence of working with land use such as in Mediterranean countries and other semi-arid areas percentage values. Frequently, datasets include land use worldwide (World Water Assessment Programme, 2009). Since classes that cover a wide range across the impact gradient (i.e. the increase in population and demand for food, along with sites with low, medium and high percentages of agriculture) expected climate change over the next decades, suggest an while other land uses (e.g. urban) only reach low percentages increase in irrigation surface and water demand in arid and within subcatchments (for example 10e20% as maximum). In semi-arid areas (Tilman, 2001; World Water Assessment addition, as land uses are quantified in percentages of a surface, Programme, 2009), the control of diffuse pollution in this the higher the percentage value for one land use, the lower future scenario will be an important challenge, thus making percentage value for the remaining land uses. The statistical the identification of land uses responsible for freshwater consequence is a masking effect of classes reaching low cover eutrophication a real need. percentage by those more extended in the subcatchment. Few studies have been published comparing the eutro- In this study, we relate land uses to nutrient concentration phication impact caused by different agricultural land use in rivers through four different spatial scales (total drainage classes. For example Johnson et al. (1997) compared rowcrop area, 100 m wide corridors, and 5 and 1 km radius buffers vs. non-rowcrop agriculture in North America whereas upstream sites) in order to identify the most appropriate spatial Lassaletta et al. (2009) compared arable lands, permanent framework. The novelty of this research lies in testing new crops, pastures and heterogeneous areas in Spain. spatial scales (1 and 5 km upstream buffers) and comparing the According to the Spanish National Statistics Institute eutrophication effect caused by two kinds of agriculture (www.ine.es), irrigated agriculture is more intensive than common in Mediterranean countries: irrigated vs. non- non-irrigated in terms of production due to the artificial irrigated croplands. Hence, the main aims are: (1) to demon- supply of water and fertilizers to croplands. This technique strate that spatial scale affects the results of studies relating increases the export of nitrogen compounds to ground- and land use to stream condition; (2) to determine which type of surface waters, making irrigated agriculture the main cause of agriculture (irrigated vs. non-irrigated) is more responsible for eutrophication in Spanish rivers (A´ lvarez-Cobelas et al., 2010; river eutrophication and to define pressure thresholds (3) to Berzas et al., 2004; Cavero et al., 2003). propose a method to select the most adequate spatial scale to In this paper, we analyse the effect of some land uses assess the eutrophication impact by diffuse pollution. (agriculture, urban and forest) on river eutrophication, taking into account three classes of agricultural land use with different expected impact intensities: irrigated, non-irrigated 2. Material and methods and low-impact agriculture. 2.1. Study area 1.2. Spatial scale This study was carried out within the boundaries of the Regarding spatial scale, some authors have suggested that the Spanish administrative region of Castilla-La Mancha (south- influences of land uses on stream ecosystems and water central Spain). The study area includes the upper and middle quality must be analysed at the catchment scale (Omernik reaches of five large river basins: Tajo, Guadiana, Guadalquivir, et al., 1981; Richards and Host, 1994; Roth et al., 1996) Jućar and Segura. This region currently has the lowest 76 water research 46 (2012) 2759e2771 2761

population density in Spain (26.4 inhab/km2). Industry has springesummer season and 68 in autumnewinter season. noticeably increased during recent years but agriculture Sites were separated by a minimum distance of 5 km (Johnson continues to be the main economic activity. Land dedicated to et al., 1997). This distance avoided subcatchment overlapping agriculture covers almost half of the entire surface of the region when local scales were selected (see Section 2.4). (Ministerio de Medio Ambiente y Medio Rural y Marino, 2010), most of which is dedicated to extensive non-irrigated cropland 2.3. Chemical analyses (mainly cereals, vineyards and olive trees) (Table 1). Irrigated cropland (mainly maize and sunflowers) is usually located In order to compare different types of agriculture in terms of close to river banks. Due to water eutrophication by agricul- eutrophication impact, dissolved inorganic nutrients (nitrate, ture, six zones within the study area have been designated by ammonium and phosphate e as soluble reactive phospho- law as “vulnerable” with regard to nitrate pollution (Consejerıá rous, SRP) were measured since these are the main nutrient de Agricultura y Medio Ambiente, 1998, 2003; Consejerıá de compounds released by the inorganic fertilizers applied to Industria, Energıá y Medio Ambiente, 2009). Despite this, crops. Water samples were taken in 250 ml polyethylene groundwater and river eutrophication continue to be a serious bottles after filtration, and kept at 4 C in a cooler until anal- problem for water quality in the study area (A´ lvarez-Cobelas yses. Nutrient concentration was determined photometrically et al., 2010; Berzas et al., 2004; Moreno et al., 2006). with MERCK kits (Spectroquant ) within 48 h of sampling. The 1 þ 3 The climate in the study area is Mediterranean with values were expressed as mg L of N-NO3 , N-NH4 and P-PO4 . continental elements, exhibiting remarkable fluctuations in Detection levels were: 0.10 mg L 1 for nitrate and 0.01 mg L 1 daily and seasonal temperatures. Annual precipitation ranges for ammonium and phosphate. Confidence intervals for these from 300 mm on the plains (semi-arid) to more than 1000 mm commercial kits were: 0.07 for phosphates, 0.06 for in the mountains (humid), with rainfall mainly concentrated ammonium and 0.3 for nitrate (http://photometry.merck. in autumn (Fernańdez, 2000). As for geology, three zones can de). In the case of nitrate it was necessary to decrease the be distinguished: the western zone is rich in Precambrian detection level for unpolluted sites by performing the siliceous rocks (mostly quartzite, slate, shale, granite and cadmium reduction method to reach the lowest detection 1 gneiss); Mesozoic calcareous rocks (limestone, dolomite, level of 0.01 mg L N-NO3 (APHA, 1989). sandstone and conglomerates) are dominant in the eastern area; and finally, Tertiary sedimentary fills are accumulated in 2.4. Land use data the great central plateau located at 700 m a.s.l. called “La Mancha” where clays, sandstones, gravels, stones, conglom- All analyses and geographic data management were carried ª erates, marls and gypsum are abundant (Gonza´lez and out with ArcGIS 9.2 (Esri ). Land use percentage composition Va´zquez, 2000). Mountains are located mainly on the edges for each site was quantified with four different spatial scales: of the region and can reach 2000 m a.s.l. total drainage area of subcatchment upstream from each site (DA), 100 m wide subcatchment buffer corridors upstream 2.2. Survey design from each site (CD), and 1 and 5 km radius semicircular buffers oriented upstream (B1 and B5 respectively) (Fig. 2). Sites were chosen in order to reflect the widest possible The total upstream drainage area for each location (sub- gradient of land use and nutrient concentration. Nutrient catchment drainage area) was delineated based on data and concentration was measured at 130 sites during the images from the Spanish Spatial Data Infrastructure (IDEE; 2001e2009 period. In order to take seasonal differences into http://www.idee.es/show.do?to¼pideep_desarrollador_wms.es). account, samples were taken during two periods per year: Corridors were obtained by generating a 100 m wide buffer springesummer (SS) from March to September, and along the stream drainage net upstream from each site. autumnewinter (AW) from September to December. Some Finally, buffers of 1 and 5 km radii were generated from each sites subject to variations in nutrient concentration due to site, but surfaces draining downstream sites (after intersect- human impact and/or seasonal conditions were visited more ing with DA polygons) were removed (Fig. 2). As mentioned in than once. Thus, the total dataset was composed of 216 water Section 2.2, minimum distance between sites was 5 km. Thus, samples from 130 sites (Fig. 1), 148 were collected in the only local scales (B1 and B5) failed to produce overlapping areas on consecutive sites. Regarding the meaning of the term “buffer”, in literature it usually refers to riparian corridors created from longitudinal features (streams). However, here e Table 1 Estimated land use composition in the study we have used ArcGis terminology, which defines “buffers” area according to official data (Ministerio de Medio also as circular surfaces created around points (sites). Ambiente y Medio Rural y Marino, 2010). Land use data were obtained from the Corine Land Cover 2 km % (CLC, 2000) Project. Land cover classes characterized in CLC Total area 79,409 100 were reclassified into the following categories (Table 2): urban Urban 7941 10 and industrial land cover (URB); areas dedicated to agricultural

Agriculture 36,530 46 uses (AGR); and areas with natural or reforested vegetation Irrigated 5110 7 (FOR). Agricultural land use was also split into three classes: Non-irrigated 31,017 39 irrigated cropland (IRR), non-irrigated cropland (NIR) and

Forest 34,940 44 agricultural land uses expected to have a low impact on river ecosystems (LIA). 77 2762 water research 46 (2012) 2759e2771

Fig. 1 e Site distribution map showing the entire study area and main basin boundaries. Black arrows show the ﬂow direction.

2.5. Statistical analyses expected for randomly associated pairs of observations (Legendre and Legendre, 1998). Autocorrelation indicates lack Statistical analyses were performed using STATISTICA v.9 of independence among the observations and creates prob- ª ª (StatSoft ) and XLSTAT v.7.5.2 (Addinsoft ). Normality and lems when attempting to use statistical tests that require homogeneity of variance were tested using ShapiroeWilk and independence of the observations. In order to test for spatial Bartlett tests, respectively. These assumptions were not autocorrelation in our data, we applied the simple Mantel test satisfied and therefore, non-parametric analyses were which measures the correlation between two similarity or required. A KruskaleWallis test was applied to test statistical dissimilarity matrices (rM(AB)): a matrix (A) of distances differences between land use classes depending on the computed for the environmental variable tested, and a matrix selected spatial scale. (B) of geographic distances. Therefore, a correlation between A preliminary interpretation of eutrophy trends was matrices A and B measures the extent to which the variations carried out using correlation analysis between land use in the similarities of A correspond to the variations in B. Thus, percentages and dissolved nutrient concentrations, similarity matrices containing the Euclidean distance a common statistical approach in this type of studies. In this between observations were generated for the predictor vari- case, we calculated Spearman’s rank correlation coefficient able (land use composition through each spatial scale) and for each spatial scale dataset and for both springesummer also for the response variable (nutrient concentration). Each and autumnewinter seasons. matrix was then related with the geographic distance matrix Spatial autocorrelation is defined as the property of generated from UTM site coordinates. random variables which take values, at pairs of sites sepa- A ManneWhitney test was used to test for differences in rated a given distance, that are more similar (positive auto- nitrate concentration between two groups representing correlation) or less similar (negative autocorrelation) than different practices in agriculture: irrigated and non-irrigated 78 water research 46 (2012) 2759e2771 2763

Fig. 2 e Schematic representation of the spatial scales used in this study. Large scales: DA, total drainage area of upstream subcatchment; CD, 100 m wide upstream subcatchment corridor. Local scales: B5, 5 km upstream radius buffer; B1, 1 km upstream radius buffer. Subcatchment corridors in this ﬁgure are 500 m wide for an easier display.

cropland. Linear regression analyses between nitrate As for ammonium, 2.3% of the sites surpassed the maximum concentration and both groups of agriculture were performed. toxic levels for aquatic life (around 2.5 mg L 1; USEPA, 1999). Finally, a breakpoint regression analysis was applied in order These values were recorded in the Tajo River below the joining to identify any significant pressure threshold in the relation- of two tributaries collecting urban and industrial wastes from ship nitrate-irrigated agriculture. For this purpose, we used the huge metropolitan area of Madrid. Nitrate concentration SegReg software (Oosterbaan, 2010). never surpassed the limit established (50 mg L 1) by the Nitrate Directive (European Commission, 1991). Results for ShapiroeWilk (W) test ranged from 0.347 to 3. Results 0.960 ( p < 0.001, a ¼ 0.05) suggesting that samples (% of each land use class per scale) did not follow a normal distribution. c2 ¼ 3.1. Exploratory analysis In addition, Bartlett’s test was significant ( (23) 35.172, p < 0.0001, a ¼ 0.05) and therefore, the null hypothesis of Nutrient concentration ranged from <0.01 to 30 mg L 1 for homogeneity of variances was rejected. Thus, the non- < 1 þ nitrate (N-NO3 ), 0.01 to 5.5 mg L for ammonium (N-NH4 ) parametric Spearman’s rank correlation test between land < 1 3 and 0.01 to 12.0 mg L for orthophosphate (P-PO4 )(Table 3). use classes and nutrients was applied as a first exploration of 79 2764 water research 46 (2012) 2759e2771

Table 2 e Description of land uses analysed in the present study with their corresponding Corine Land Cover classes. Reclassiﬁcation code Description Corine Land Cover classes

Urban Urban and industrial land use 1. Artiﬁcial areas Agriculture Areas dedicated to agricultural uses 2. Agricultural areas

Irrigated agriculture Including crops, fruit trees and 212. Permanently irrigated land olive groves with a permanent 2212. Irrigated vineyards irrigation infrastructure 2222. Irrigated fruit trees and berry plantations 2232. Irrigated olive groves 2412. Annual crops associated with permanent crops on the same irrigated parcel 2422. Juxtaposition of irrigated small parcels of diverse annual crops, pasture and/or permanent crops 2423. Juxtaposition of irrigated and non-irrigated small parcels of diverse annual crops, pasture and/or permanent crops 2432. Land principally occupied by irrigated agriculture, with signiﬁcant areas of natural vegetation

Non-irrigated Cereals, legumes, fodder crops, 211. Non-irrigated arable land agriculture root crops, fallow land, fruit trees 2211. Non-irrigated vineyards and olive groves without 2221. Non-irrigated fruit trees and berry plantations irrigation infrastructures 2231. Non-irrigated olive groves 2411. Annual crops associated with permanent crops on the same non-irrigated parcel 2421. Juxtaposition of non-irrigated small parcels of diverseannual crops, pasture and/or permanent crops

Low-impact agriculture Pastures and abandoned agricultural 231. Pastures fields with dense grass cover and 2431. Land principally occupied by non-irrigated not under a rotation system agriculture, with significant areas of natural vegetation 2433. Land principally occupied by pastures, with significant areas of natural vegetation 244. Agro-forestry areas (annual crops or grazing land under the cover of forestry species)

Forest Areas with natural or 3. Forest and semi-natural areas reforested vegetation

the relationship between land use and river eutrophication. correlation with nitrate was observed on large scales The value of the correlation coefficient varied clearly (r ¼ 0.504, p < 0.001 for CD; r ¼ 0.491, p < 0.001 for DA) and for depending on spatial scale selected and, to a lower extent, on autumnewinter samples (rainy season), a logical effect stating the season (Table 4). In general, positive and high correlation that non-irrigated cropland is rain fed. values were obtained between agricultural land uses (AGR, Regarding ammonium, the highest correlation values were IRR, NIR) and nitrogen compounds, while phosphate concen- recorded for irrigated cropland on a large scale (DA) in trations were more correlated to urban land use percentage autumnewinter (r ¼ 0.510, p < 0.001). The same seasonal (URB). The influence of spatial scale on the land use defined as pattern but with a lower correlation value was observed for low-impact agriculture (LIA) was complex and difficult to the relationship ammoniumeurban use (r ¼ 0.459, p < 0.001) at interpret, probably due to the heterogeneity of land uses DA scale. For non-irrigated cropland and forest, the correla- compiled in this class. For this reason, this land use class was tion values with this nutrient were lower and less significant eliminated from statistical analyses. Natural or semi-natural (Table 4). land cover classified as forest (FOR) was strongly and nega- The correlation values between phosphate and land uses tively correlated with all nutrients irrespective of scale. were in general the lowest of all nutrients analysed, showing A global trend in relationship between nutrients and land the maximum value in the case of urban use for large scales use was observed: for large scales correlation values were (CD, DA) and in springesummer samples (r ¼ 0.357, p < 0.001 higher than for local scales (Table 4). However, the relation- and r ¼ 0.348, p < 0.001 respectively). ship between nitrate and irrigated cropland (IRR) showed the opposite trend: the highest correlation values were observed 3.2. Range of land use percentage by scale at 1 km buffer scale (r ¼ 0.451, p < 0.001) decreasing gradually towards larger scales. Regarding the relationship nitrate- As mentioned in Section 1.2, in a multi-scale approach irrigated cropland, the effect of seasonality was also remark- different scales would provide different land use datasets able with higher values in springesummer, coinciding with (Fig. 3). In order to demonstrate these differences statistically, the irrigation season in the study area when nitrate leaching is a KruskaleWallis test was performed. Results of this test enhanced. However, for non-irrigated cropland, the highest showing that observations (% of each land use class per 80 water research 46 (2012) 2759e2771 2765

and forest (H ¼ 19.741, p < 0.001). Therefore, percentages Table 3 e Mean values, standard deviation (SD) and range 3,516 L of nutrient concentration (mg L 1) measured during obtained for each land use class (URB, IRR, NIR, AGR, FOR) were 2001e2009. AW, autumnewinter season; SS, significantly different across the four spatial scales. As springesummer season. expected, all land uses showed a wide range of percentage n Mean SD Minimuma Maximum values at local scales (B1 and B5) and corridors (CD) (Fig. 3) except for urban land use, which never reached a value higher Total N-NO3 202 2.612 5.111 0.010 30.000 þ than 30% irrespective of scale. Intensive land uses near river N-NH4 213 0.166 0.515 0.010 5.500 3 banks such as irrigated croplands were also better represented P-PO4 156 0.467 1.570 0.010 12.000 at those scales (B1, B5 and CD) (Fig. 3b). In fact, when the largest AW N-NO3 68 1.793 2.325 0.010 10.900 þ scale (DA) was selected, only the most extensive land uses N-NH4 68 0.249 0.853 0.010 5.500 3 (non-irrigated agriculture, total agriculture and forest) showed P-PO4 68 0.149 0.324 0.010 1.380 a wide range of percentage values (Fig. 3cee, respectively). The SS N-NO3 134 31.000 6.009 0.010 30.000 þ lack of a wide range for some land uses may lead to a masking N-NH4 146 19.000 0.208 0.010 1.400 of those land uses which are poorly represented by those P-PO 3 88 77.000 2.038 0.010 12.000 4 better represented. Therefore, for the study area and the land a Detection limit by chemical analyses. uses analysed, the largest spatial scale tested (DA) does not seem to be the most appropriate for the study of river eutro- scale) did not come from the same distribution were as phication due to the poor representation of irrigated agricul- ¼ < follows: total agriculture (H3,516 9.493, p 0.05), irrigated ture (only low percentage values). These results demonstrate ¼ < cropland (H3,516 41.489, p 0.0001), non-irrigated cropland the influence of spatial scale on land use quantification and ¼ < ¼ < (H3,516 32.110, p 0.0001), urban (H3,516 83.796, p 0.0001) therefore on detecting river eutrophication.

Table 4 e Spearman’s rank correlation coefﬁcients for land use classes at different spatial scales and seasons. IRR, irrigated agriculture; NIR, non-irrigated agriculture; AGR, total agriculture; URB, urban; FOR, forest; B1, 1 km upstream radius buffer; B5, 5 km upstream radius buffer; CD, 100 m wide upstream subcatchment corridor; DA, total drainage area of upstream subcatchment; AW, autumnewinter; SS, springesummer. Signiﬁcation level (*<0.05, **<0.01 and ***<0.001) and sample size (n) are shown. Values in bold highlight the trend followed by the relationship nitrate-irrigated agriculture: higher correlation values at local scales decreasing towards large scales. B1 B5 CD DA

AW SS AW SS AW SS AW SS

0.445*** 0.451*** 0.351** 0.422*** 0.266* 0.292*** 0.243* 0.315*** N-NO3 IRR n ¼ 68 n ¼ 134 n ¼ 68 n ¼ 134 n ¼ 68 n ¼ 134 n ¼ 68 n ¼ 134 NIR 0.309* 0.484*** 0.319*** 0.504*** 0.433*** 0.491*** 0.449*** n ¼ 68 n ¼ 68 n ¼ 134 n ¼ 68 n ¼ 134 n ¼ 68 n ¼ 134 AGR 0.577*** 0.480*** 0.467*** 0.427*** 0.423*** 0.405*** 0.484*** 0.471*** n ¼ 68 n ¼ 134 n ¼ 68 n ¼ 134 n ¼ 68 n ¼ 134 n ¼ 68 n ¼ 134 URB 0.229** 0.349** 0.337*** 0.390*** 0.436*** 0.427*** n ¼ 134 n ¼ 68 n ¼ 134 n ¼ 134 n ¼ 68 n ¼ 134 FOR 0.550*** 0.485*** 0.470*** 0.445*** 0.339** 0.320*** 0.484*** 0.480*** n ¼ 68 n ¼ 134 n ¼ 68 n ¼ 134 n ¼ 68 n ¼ 134 n ¼ 68 n ¼ 134

þ N-NH4 IRR 0.252* 0.290*** 0.315** 0.387*** 0.418*** 0.345*** 0.510*** 0.398*** n ¼ 68 n ¼ 146 n ¼ 68 n ¼ 146 n ¼ 68 n ¼ 146 n ¼ 68 n ¼ 146 NIR 0.177* 0.207* 0.261** 0.270* 0.288*** n ¼ 146 n ¼ 146 n ¼ 146 n ¼ 68 n ¼ 146 AGR 0.259** 0.274* 0.309*** 0.309*** 0.279* 0.341*** n ¼ 146 n ¼ 68 n ¼ 146 n ¼ 146 n ¼ 68 n ¼ 146 URB 0.261* 0.306* 0.195* 0.335** 0.279*** 0.459*** 0.188* n ¼ 68 n ¼ 68 n ¼ 146 n ¼ 68 n ¼ 146 n ¼ 68 n ¼ 146 FOR 0.264** 0.282* 0.316*** 0.259* 0.406*** 0.297* 0.346*** n ¼ 146 n ¼ 68 n ¼ 146 n ¼ 68 n ¼ 146 n ¼ 68 n ¼ 146

3 P-PO4 IRR 0.251* 0.248* 0.270* n ¼ 88 n ¼ 68 n ¼ 88 NIR 0.281* 0.239* 0.303* 0.268* 0.335** 0.258* n ¼ 68 n ¼ 88 n ¼ 68 n ¼ 88 n ¼ 68 n ¼ 88 AGR 0.305** 0.297* 0.230* 0.332** 0.358** 0.289** n ¼ 88 n ¼ 68 n ¼ 88 n ¼ 68 n ¼ 68 n ¼ 88 URB 0.241* 0.357*** 0.348*** n ¼ 68 n ¼ 88 n ¼ 88 FOR 0.304** 0.300* 0.242* 0.312** 0.355** 0.321** n ¼ 88 n ¼ 68 n ¼ 88 n ¼ 68 n ¼ 68 n ¼ 88

81 2766 water research 46 (2012) 2759e2771

Fig. 3 e Arrangement of sites (n [ 130) by land use percentages through the four spatial scales analysed. Land use classes: a) urban, b) irrigated agriculture, c) non-irrigated agriculture, d) total agriculture, e) forest. Local scales: B1, 1 km radius upstream buffer; B5, 5 km radius buffer upstream from a site. Large scales: CD, 100 m wide corridor drainage area upstream from a site; DA, drainage area upstream from a site. Close rhombus represents the percentage of each land use class obtained at a particular spatial scale and for each site.

3.3. Spatial autocorrelation high correlation values can be due to pseudoreplication (Hurlbert, 1984). When sites located downstream show overlapping of drainage areas (nested subcatchments), a problem of spatial autocor- 3.4. Agricultural eutrophication: irrigated vs. non- relation may arise. This is more probable for large drainage irrigated cropland basins where each site shares a great part of its drainage area with the upstream preceding site (Fig. 2a and b). In these cases Taking into account the correlation analysis between nutrient geographical proximity of sites and the directional component concentration and land use (Table 4), and the problems derived of spatial autocorrelation can lead to an absence of indepen- from both the lack of a wide range of land use percentages and dence among data (Legendre and Legendre, 1998). Thus, land spatial autocorrelation among sites, attention was focused on use composition obtained at each spatial scale as well as nitrate level at the local scale. The criteria for making such nutrient concentration were checked for spatial autocorrela- a choice can be outlined as follows: urban land use only tion using the Mantel test. In the case of large scales, this test reached low percentage values and its effect could be masked found significant autocorrelation in land use composition by other land uses; forest land use did not contribute to river for total drainage area (DA: r ¼ 0.066, p < 0.05) and corridors eutrophication; large scales were not appropriate due to (CD: r ¼ 0.153, p < 0.001) while no autocorrelation was detec- spatial autocorrelation and also due to the narrow range in ted at local scales (buffers B1 and B5) for the response variable. percentage values of irrigated cropland (Fig. 3); and finally, Therefore, the underlying assumption of data independence nutrients other than nitrate showed the highest correlation might be violated and Spearman’s correlation coefficients values at large scales. Thus, the impact of eutrophication due must be taken with care for large scales (CD, AD) since the to irrigation practices became more evident and reliable for 82 water research 46 (2012) 2759e2771 2767

nitrate, at the most local scale of 1 km buffer (B1), and for springesummer data (Table 4). In order to check for significant differences in nitrate concentration between irrigated and non-irrigated agriculture, a non-parametric test of mean comparison (ManneWhitney) between these two groups was performed. One group (irrigated) included sites with more than 10% of irrigated land and 0% of non-irrigated cropland. The second group (non-irrigated cropland) included sites with more than 10% of non-irrigated cropland and 0% of irrigated land. In order to avoid influences other than agriculture, data from sites impaired by high point pollution were removed. The unilateral ManneWhitney test showed that the nitrate values Fig. 5 e Breakpoint regression analysis of nitrate from the irrigated cropland group were significantly higher concentration against percentage of irrigated agriculture. than those from the non-irrigated group (U ¼ 48.500, p < 0.01, Regression analyses were performed after data a ¼ 0.05). Such an eutrophication pattern can be illustrated in transformation of land use percentages (arcsin [(3/100)1/2]) Fig. 4. Linear regression analyses showed that an increase of and nitrate concentration (log( y D 1)). land dedicated to irrigated agriculture led to an increase in nitrate concentration (R2 ¼ 0.680, p < 0.0001, n ¼ 17). In the case of non-irrigated cropland, the relationship with nitrate was not significant. The performance of a breakpoint regression 50% of irrigated croplands followed by a sloping line explaining analysis showed a significant threshold of about 50% (Break- 74% of the variance and described by the function point ¼ 49.55, R2 ¼ 0.74, p < 0.01, n ¼ 17) of irrigated agriculture y ¼ 1.580x 0.679. These results imply a clear measure of (Fig. 5). Thus, the relationship between percentage of irrigated management: development of irrigated agriculture higher agriculture and nitrate concentrations may be represented by than 50% within a 1 km radius from the river banks leads to two trends: a non-significant horizontal segment for under river eutrophication in the study area.

4. Discussion

4.1. Inﬂuence of spatial scale

One of the main goals of this study was to demonstrate that spatial scale may affect results relating land uses to stream eutrophication. Each particular spatial scale entails some considerations that must be taken into account; otherwise results may be unreliable, giving way to misleading interpre- tations. The most important considerations are (1) the lack of a wide range of land use percentages to be analysed and (2) spatial autocorrelation among sites. Regarding the ﬁrst issue, a wide range of percentages for land use must be reached in order to relate these percentages to nutrient concentration. In this study, when land uses within the whole drainage area of subcatchments (DA) were taken into account, percentages of non-irrigated cropland ranged from 0 to 80%, while percentages of irrigated cropland hardly reached 20% (Fig. 3b and c). Thus, despite the fact that irrigation could be more important in terms of eutrophication, most extended land uses may mask the effect of irrigation cropland just because this land use class was poorly represented. Consequently, Spearman’s Fig. 4 e Irrigated vs. non-irrigated agriculture. Nitrate rank correlations were higher for non-irrigated than irrigated L L1 concentration (NO3 -N mg L ) against percentage of cropland, but to conclude that non-irrigated agriculture is agricultural land at the 1 km buffer scale (B1) and more inﬂuential in eutrophication than irrigation practices springesummer season. (a) Sites with more than 10% of would be a misleading and unreal interpretation. Therefore, irrigated agriculture and 0% of non-irrigated agriculture. results were scale-dependent in this study. This fact demon- (b) Sites with more than 10% of non-irrigated agriculture strates the need to consider multi-scale approaches for and 0% of irrigated agriculture. Regression analyses were unravelling the dynamics of water quality over space and time performed after data transformation of land use (Chang, 2008). percentages (arcsin [(3/100)1/2]) and nitrate concentration Regarding the spatial scales assessed in the present study, (log( y D 1)). those based on 100 m wide corridors (CD) and 1 km buffers 83 2768 water research 46 (2012) 2759e2771

(B1) provided the most complete range for irrigated cropland Other studies carried out in Spain (Moreno et al., 2006; (Fig. 3) mainly because this land use is usually located close to Lassaletta et al., 2009) also relate high nitrate concentration river banks. However, spatial autocorrelation was detected in in streams with high percentages of agriculture in the data from drainage areas (DA) and corridor (CD) scales (large drainage area by means of correlation analysis and regression scales) by means of a Mantel test. Thus, any statistical test (R2 ¼ 0.69, p < 0.001, n ¼ 30; r ¼ 0.632, p < 0.001, n ¼ 61, respec- requiring independence of data, such as correlation or tively). Similar correlation values were obtained by Dodds and regression analysis, would not be suitable (Legendre, 1993) Oakes (2008) between agriculture and nitrate when land use and should be applied only as exploratory analyses. In this was estimated for riparian corridors throughout the entire study, spatial autocorrelation detected on large scales (DA and watersheds (r ¼ 0.623, p < 0.05, n ¼ 68) and riparian corridors CD) was due to subcatchment overlapping which produced adjacent to the first-order streams (0.650, p < 0.05, n ¼ 68). pseudoreplication and enhanced correlation values. Some Buck et al. (2004) found a strong relationship between agri- researchers avoid autocorrelation and subcatchment over- culture and nitrate (r ¼ 0.67, p < 0.01, n ¼ 24e35) but only when lapping by selecting single sites in independent subcatch- first-order streams were excluded from the analysis. Other- ments (Lassaletta et al., 2009; Snyder et al., 2003) despite the wise, correlation values were not significant; probably fact that it is difficult to reach a satisfactory sample size when because the agricultural land in their study area was mainly this strategy is selected. The approach suggested in this study pastures (less intensive than agriculture in our study area) is to use those spatial scales which allow a satisfactory sample located at the middle and lower reaches of the catchment. size while, at the same time, providing no autocorrelated data. Some studies do suggest, however, that urban land use is A Mantel test indicated that data from 1 and 5 km radii a more important predictor of water quality than agriculture, buffers were not autocorrelated. In addition, ranges of land irrespective of the scale (Sliva and Williams, 2001; Chang, use percentages for these local scales were wider than for 2008). This result can be expected in highly urbanized areas large scales such as subcatchment drainage areas or sub- where urban surface is an important component of non-point catchment corridors. Therefore, both local scales can be pollution, as in the watersheds studied by Chang (2008). considered as appropriate for assessing land use effect on Nevertheless, Sliva and Williams (2001) point out that this river pollution by means of correlation analysis, except for result may have been influenced by point-sources associated urban land use which never reached more than 28%, as with urbanized areas in their study area. usually occurs in agricultural landscapes. Non-irrigated agriculture is in essence not intensive due to Even though the effect of urban land use on stream climatic constraints but, when irrigation is applied, farming eutrophication could be masked by the effects of better rep- practices tend to intensify: water consumption of crops during resented land uses, positive significant correlation values hot summers must be compensated by irrigation inputs relating urban land use to nitrogen compounds were found. associated to a continued use of N fertilizers exceeding The intensity of urban influence was higher in spring- optimum rates (Diez et al., 1997). In the study area, irrigated esummer, probably due to the decrease in water flow (Robles lands are located close to the rivers and, as in the case of et al., 2002) coupled to the increase in population and waste maize, they receive a quantity of fertilizers four-fold higher water effluents during summer holidays (National Statistics than non-irrigated cropland (Consejerıáde Industria, Energıá Institute, www.ine.es). Correlation values relating urban use y Medio Ambiente, 2010). This, of course, means that with phosphate level at local buffer scales were not signifi- nutrient inputs are higher and mobilized faster under irriga- cant, and this trend may be due to the fact that an important tion practices. When N fertilization exceeds that required for part of phosphate comes from effluents of urbanized or the crop, this surplus is considered a risk for river eutrophi- industrial zones (source point pollution) (Vega et al., 1998) that cation (Smith et al., 1999). Fertilization surpluses are often are not included in local buffers. In addition, a weak positive detected throughout Spain: e.g. Isidoro et al. (2006) described correlation was found between phosphate and agricultural an application of 398e453 kg-N/ha/year to irrigated maize uses supporting the idea that a part of the phosphate comes crops located in La Violada district (NE Spain). In areas from the leaching of crop fertilizers. The strong relationship vulnerable to nitrates located within the study area, N fertil- between agriculture and nitrate concentration detected at 1 ization limits have been established in 50e60 kg-N/ha/year for and 5 km buffer scales contrasted with the results of Strayer most common non-irrigated cropland, 80e100 kg-N/ha/year et al. (2003) who observed a weak influence of land uses for irrigated cereals, 200e210 kg-N/ha/year for maize, and within a 300 m radius buffer on the nitrate flux. This result 80e100 kg-N/ha/year for sunflower (Consejerıá de Industria, was probably obtained because the size of buffers might not Energıá y Medio Ambiente, 2010). These facts explain the include the surrounding agricultural land uses beyond high impact of irrigated agriculture in terms of eutrophication riparian vegetation. when a threshold of approximately 50% of irrigated agriculture is detected (Fig. 5). A threshold can be defined as the point 4.2. Agricultural eutrophication: irrigated vs. non- at which there is an abrupt change in an ecological quality, irrigated cropland property, or phenomenon or where small changes in a river can produce large responses in the ecosystem (Groffman et al., Results of a correlation analysis indicated that agriculture 2006). Thus, our results suggest that surpassing the threshold (AGR) measured at local scales (B1 and B5) was the most of 50% of irrigated agriculture in a radius of 1 km from a river important land use in terms of river eutrophication due to bank would lead to a rapid and important increase in nitrate nitrate (Spearman’s coefficient from 0.427 to 0.577 depending concentration and, therefore, to river eutrophication. In other on season and spatial scale, p < 0.001, n ¼ 68e134; Table 2). studies, Wang et al. (1997) observed a decline in habitat quality 84 water research 46 (2012) 2759e2771 2769

Fig. 6 e Methodological approach proposed to choose the appropriate spatial scale when studying river eutrophication caused by agriculture.

when 50% of agriculture was exceeded in both total catch- springesummer data, the relationship between nitrate and ment and 100-m corridors. These facts support the idea that both total agriculture and irrigation cropland was very similar. further research in predicting pressure thresholds is needed However, for autumnewinter data, correlation values due to the importance in ecosystem responses. Ecosystem between nitrate and total agriculture were higher than those recovery by means of restoration projects might require far reached when considering only irrigated cropland. This result more resources and time than avoiding entering the eutro- could be explained by the increase in the impact of non- phication state (Dodds et al., 2010). irrigated cropland in the autumnewinter season. Our results were clear and consistent: irrigation practices Summarizing, the fact that reduced land use such as irri- are more inﬂuential in nitrate export than non-irrigated gated cropland caused a higher impact than more extended agriculture as Scanlon et al. (2005) stated for groundwater land uses supports the idea that measurement of agricultural quality. Johnson et al. (1997) also studied two types of agri- intensity may be a better indicator of the impacts within culture, one more intensive (rowcrop agriculture) than the a river system than the percentage of differing land uses other (non-rowcrop agriculture). These authors found a strong (Harding et al., 1999). On the other hand, the variety of results relationship between nitrate and rowcrop agriculture obtained among different spatial perspectives throughout the (R2 ¼ 0.654, p < 0.0001, n ¼ 62) very similar to the relationship literature suggests that different land uses may have distinc- between nitrate and irrigated cropland described in the tive particularities and operate at different spatial scales. present study (R2 ¼ 0.680, p < 0.0001, n ¼ 17). Therefore, multi-scale approaches adapted to land use As regards seasonality, nitrate leaching in non-irrigated particularities could be the better choice when assessing their croplands was associated with the rainy season, while for inﬂuence on stream water quality. irrigated lands it increased during the dry season (irrigation season). The correlations reached between irrigated cropland and nitrate concentration for springesummer data were 5. Conclusions higher than for autumnewinter samples whereas, for non- irrigated croplands, the correlations were higher in the rainy (1) Irrigated cropland was determined as the main land use season. This seasonal nutrient pattern has also been recorded responsible for eutrophication in streams of south-central in northern Spanish basins (Isidoro et al., 2006) and is sup- Spanish rivers, suggesting that the intensity of agriculture ported by correlation values between nitrate and total agri- and its proximity to river banks are important factors to culture class (AGR): at local scales (B1 and B5) and for take into account. 85 2770 water research 46 (2012) 2759e2771

(2) A pressure threshold was found at 50% of irrigated agri- 06-0057 granted by the Junta de Comunidades de Castilla-La culture within a 1 km radius of river banks. Surpassing this Mancha. percentage value may lead to a rapid increase in nitrate concentration in streams and, therefore, to river references eutrophication. (3) The election of the spatial scale may have important influences on the final results and ecological interpretation of studies assessing the relationship between land uses A´ lvarez-Cobelas, M., Sańchez-Carrillo, S., Cirujano, S., and eutrophication in streams. Angeler, D.G., 2010. A story of the wetland water quality deterioration: salinization, pollution, eutrophication and (4) In the study area, the adequate spatial scale to assess the siltation. In: Sańchez-Carrillo, S., Angeler, D. (Eds.), Ecology of impact of non-irrigated and irrigated agriculture in stream Threatened Semi-Arid Wetlands: Long-Term Research in Las water quality was a 1 km radius buffer upstream from sites. Tablas de Daimiel. Wetlands: Ecology, Conservation and (5) Depending on the subject under study, the most appro- Management, vol. 2. Springer, Dordrecht, pp. 109e133. priate spatial scale may vary and it may even be possible to APHA, 1989. Standard Methods for the Examination of Water and find more than one adequate scale. As a new contribution, Wastewater. American Public Health Association, the following methodological approach may be outlined Washington, D.C. Berzas, J.J., Garcıá,L.F., Martıń-A´ lvarez, P.J., Rodrı´guez,R.C., 2004. when analysing relations between land use and eutro- Quality assessment and chemometric evaluation of a fluvio- phication indicators (Fig. 6): lacustrine system: Ruidera Pools Natural Park (Spain). Water, a) It is important to check for the spatial scale(s) that Air, & Soil Pollution 155, 269e289. provide(s) more complete percentage ranges for all Buck, O., Niyogi, D.K., Townsend, C.R., 2004. Scale-dependence of land uses under study. The wider the range of land use land use effects on water quality of streams in agricultural e percentages analysed, the more reliable the results catchments. Environmental Pollution 130, 287 299. Carpenter, S.R., Caraco, N.F., Correll, D.L., Howarth, R.W., obtained. Sharpley, A.N., Smith, V.H., 1998. Nonpoint pollution of b) Secondly, it is also necessary to check for autocorre- surface waters with phosphorous and nitrogen. Ecological lation of data when geographical distance between Applications 8, 559e568. sites is short (a few kilometres, sites located within the Cavero, J., Beltrań, A., Arague´s, R., 2003. Nitrate exported in same reach or segment), and always in the case of drainage waters of two sprinkler-irrigated watersheds. Journal nested subcatchments. of Environmental Quality 32, 916e926. c) The most appropriate spatial scale(s) to relate land Chang, H., 2008. Spatial analysis of water quality trends in the Han River basin, South Korea. Water Research 42, 3285e3304. uses to river eutrophication will be the one that Consejerıáde Agricultura y Medio Ambiente, 1998. Resolucioń 07/ provides the largest number of sites without gener- 08/1998 por la que se designan, en el a´mbito de la Comunidad ating data autocorrelation. Auto´ noma de Castilla-la Mancha, determinadas a´reas como zonas vulnerables a la contaminacio´ n de las aguas producidas This work suggests that a multiple perspective may por nitratos procedentes de fuentes agrarias. Official Gazette provide valuable information when relating land use to of Castilla-la Mancha. ´ stream eutrophication. This multiple perspective refers not Consejerıáde Agricultura y Medio Ambiente, 2003. Resolucion 10/ 02/2003 por la que se designan, en el a´mbito de la Comunidad only to the spatial scale but also the segregation of land use Auto´ noma de Castilla-la Mancha, determinadas a´reas como classes. In further research, more precise distinctions zonas vulnerables a la contaminacio´ n de las aguas producidas between land uses that are pooled into one land use class por nitratos procedentes de fuentes agrarias. Official Gazette (agriculture in general), together with more information of Castilla-la Mancha. about pressure thresholds, would provide interesting infor- Consejerıáde Industria, Energıáy Medio Ambiente, 2009. Orden mation about the effect of different types of agricultural de 21 de mayo de 2009 por la que se aprueba el mantenimiento de las zonas vulnerables designadas mediante las practices such as irrigated cropland, organic farming or resoluciones de 07/08/1998 y 10/02/2003 y se designa una precision agriculture, among others. Finally, as a main nueva denominada: Campo de Calatrava, en relacio´ nala precept for river management, irrigated cropland should not contaminacio´ n de las aguas por nitratos de origen agrario en be located close to river banks (<1 km) and, if allowed, la Comunidad Auto´ noma de Castilla-La Mancha. Official should generally not surpass the 50% threshold of Gazette of Castilla-la Mancha. irrigated vcroplands around rivers. In addition the develop- Consejerıáde Industria, Energıáy Medio Ambiente, 2010. Orden ment of good agricultural practices in terms of avoiding de 4 de febrero de 2010 sobre actuacioń aplicable a las zonas vulnerables a la contaminacio´ n por nitratos de origen agrario surpluses in fertilization would also help to prevent stream designadas en la Comunidad Auto´ noma de Castilla-La eutrophication. Mancha. Official Gazette of Castilla-la Mancha 32, pp. 6743e6759. Diez, J., Roman, R., Caballero, R., Caballero, A., 1997. Nitrate Acknowledgements leaching from soils under a maizeewheatemaize sequence, two irrigation schedules and three types of fertilisers. e We would like to thank all reviewers for their helpful Agriculture, Ecosystems & Environment 65, 189 199. Dodds, W.K., Oakes, R.M., 2008. Headwaters influences on comments, which improved the quality and understanding of downstream water quality. Environmental Management 41, this paper. Special thanks to Walter K. Dodds for his sugges- 367e377. tions and to Kathryn Ann Walsh for the review of English Dodds, W.K., Clements, W.H., Gido, K., Hilderbrand, R.H., language. This study was financed by the Regional Project PAI- King, R.S., 2010. Thresholds, breakpoints, and nonlinearity in 86 water research 46 (2012) 2759e2771 2771

freshwaters as related to management. Journal of the North Oosterbaan, R.J., 2010. SegReg Software. Available at: www. American Benthological Society 29, 988e997. waterlog.info/segreg.htm. European Commission, 1991. Council Directive (91/676/EEC) of 12 Osborne, L., Wiley, M., 1988. Empirical relationships between land December 1991 Concerning the Protection of Waters Against use/cover and stream water quality in an agricultural Pollution Caused by Nitrates from Agricultural Sources. watershed. Journal of Environmental Management 26, 9e27. Brussels, Belgium. Quinn, J., 1991. Guidelines for the Control of Undesirable Fernańdez, F., 2000. Los condicionantes clima´ticos del paisaje. In: Biological Growths in Water (Consultancy Report No. 6213/2). Guıáde los Espacios Naturales de Castilla-La Mancha. Junta de Water Quality Centre, Hamilton, New Zealand. Comunidades de Castilla-La Mancha, Toledo, pp. 41e54. Richards, C., Host, G., 1994. Examining land use influences on Frissell, C.A., Liss, W.J., Warren, C.E., Hurley, M.D., 1986. stream habitats and macroinvertebrates: a GIS approach. A hierarchical framework for stream habitat classification: Journal of American Water Resources Association 30, 729e738. viewing streams in a watershed context. Environmental Robles, S., Toro, M., Nun˜ o, C., Avile´s, J., Alba-Tercedor, J., Management 10 (2), 199e214. A´ lvarez, M., Bonada, N., Casas, J., Jaímez-Cue´llar, P., Gonza´lez, J.A., Va´zquez, A., 2000. Las formas del relieve: llanuras, Mellado, A., Munne´, A., Pardo, I., Prat, N., Sua´rez, M.L., Vidal- pa´ramos y montan˜ as. In: Guıáde los Espacios Naturales de Abarca, R., Vivas, S., Moya´, G., Ramon, G., 2002. Descripcio´ nde Castilla-La Mancha. Junta de Comunidades de Castilla-La las cuencas mediterrańeas seleccionadas en el proyecto Mancha, Toledo, pp. 17e40. GUADALMED. Limnetica 21, 35e61. Groffman, P.M., Baron, J.S., Blett, T., Gold, A.J., Goodman, I., Roth, N., Allan, J., Erikson, D.L., 1996. Landscape influences on Gunderson, L.H., Levinson, B.M., Palmer, M.A., Paerl, H.W., stream biotic integrity assessed at multiple spatial scales. Peterson, G.D., Poff, N.L., Rejeski, D.W., Reynolds, J.F., Landscape Ecology 11, 141e156. Turner, M.G., Weathers, K.C., Wiens, J., 2006. Ecological Scanlon, B.R., Reedy, R.C., Stonestrom, D.A., Prudic, D.E., thresholds: the key to successful environmental management Dennehy, K.F., 2005. Impact of land use and land cover change or an important concept with no practical application? on groundwater recharge and quality in the southwestern US. Ecosystems 9, 1e13. Global Change Biology 11, 1577e1593. Harding, J.S., Benfield, E.F., Bolstad, P.V., Helfman, G.S., Sliva, L., Williams, D.D., 2001. Buffer zone versus whole Jones, E.B.D., 1998. Stream biodiversity: the ghost of land use catchment approaches to studying land use impact on river past. Proceedings of the National Academy of Sciences of the water quality. Water Research 35, 3462e3472. United States of America 95, 14843e14847. Smith, V.H., 2003. Eutrophication of freshwater and coastal Harding, J.S., Young, R.G., Hayes, J.W., Shearer, K.A., Stark, J.D., marine ecosystems a global problem. Environmental Science 1999. Changes in agricultural intensity and river health along & Pollution Research 10, 126e139. a river continuum. Freshwater Biology 42, 345e357. Smith, V.H., Tilman, G.D., Nekola, J.C., 1999. Eutrophication: Hurlbert, S.H., 1984. Pseudoreplication and the design of impacts of excess nutrient inputs on freshwater, marine, and ecological field experiments. Ecological Monographs 54, 187. terrestrial ecosystems. Environmental Pollution 100, 179e196. Isidoro, D., Quı´lez,D., Aragu¨ e´s, R., 2006. Environmental impact of Snyder, C., Young, J., Villella, R., Lemarie´, D., 2003. Influences of irrigation in La Violada District (Spain). Journal of upland and riparian land use patterns on stream biotic Environmental Quality 35, 776. integrity. Landscape Ecology 18, 647e664. Johnson, L., Richards, C., Host, G., Arthur, J., 1997. Landscape Strayer, D.L., Beighley, R.E., Thompson, L.C., Brooks, S., influences on water chemistry in Midwestern stream Nilsson, C., Pinay, G., Naiman, R.J., 2003. Effects of land cover ecosystems. Freshwater Biology 37, 193e208. on stream ecosystems: roles of empirical models and scaling Lassaletta, L., Garcıá-Go´ mez, H., Gimeno, B.S., Rovira, J.V., 2009. issues. Ecosystems 6, 407e423. Agriculture-induced increase in nitrate concentrations in Tilman, D., 2001. Forecasting agriculturally driven global stream waters of a large Mediterranean catchment over 25 environmental change. Science 292, 281e284. years (1981e2005). The Science of The Total Environment 407, Townsend, C., Arbuckle, C., Crowl, T., Scarsbrook, M., 1997. The 6034e6043. relationship between land use and physicochemistry, food Legendre, P., 1993. Spatial autocorrelation: trouble or new resources and macroinvertebrate communities in tributaries paradigm? Ecology 74, 1659e1673. of the Taieri River, New Zealand: a hierarchically scaled Legendre, P., Legendre, L., 1998. Numerical ecology, second ed. In: approach. Freshwater Biology 37, 177e191. Developments in Environmental Modelling Elsevier, Tran, C.P., Bode, R.W., Smith, A.J., Kleppel, G.S., 2010. Land-use Amsterdam. proximity as a basis for assessing stream water quality in Lund, J.W., 1967. Eutrophication. Nature 214, 557e558. New York State (USA). Ecological Indicators 10, 727e733. Meyer, W., Turner, B.L., 1994. Changes in Land Use and Land United States Environmental Protection Agency (EPA), 1999. Cover: A Global Perspective. Cambridge University Press, National Recommended Water Quality Criteria. US Cambridge. Environmental Protection Agency, Office of Water, Office of Ministerio de Medio Ambiente y Medio Rural y Marino, 2010. Science and Technology, Washington, DC. Anuario de Estadı´stica2009. http://www.mapa.es/estadistica/ Vega, M., Pardo, R., Barrado, E., Debań, L., 1998. Assessment of pags/anuario/2009/Anuario_2009.pdf. seasonal and polluting effects on the quality of river water by Moreno, J., Navarro, C., De las Heras, J., 2006. Abiotic ecotypes in exploratory data analysis. Water Research 32, 3581e3592. south-central Spanish rivers: reference conditions and Von Schiller, D.D., Martı´,E., Riera, J.L., Ribot, M., Marks, J.C., pollution. Environmental Pollution 143, 388e396. Sabater, F., 2008. Influence of land use on stream ecosystem Nerbonne, B.A., Vondracek, B., 2001. Effects of local land use on function in a Mediterranean catchment. Freshwater Biology physical habitat, benthic macroinvertebrates, and fish in the 53, 2600e2612. Whitewater River, Minnesota, USA. Environmental Wang, L., Lyons, J., Kanehl, P., Gatti, R., 1997. Influences of Management 28, 87e99. watershed land use on habitat quality and biotic integrity in Omernik, J.M., Abernathy, A.R., Male, L.M., 1981. Stream nutrient Wisconsin streams. Fisheries 22, 6e12. levels and proximity of agricultural and forest land to streams: World Water Assessment Programme, 2009. The United Nations some relationships. Journal of Soil and Water Conservation 36, World Water Development Report 3: Water in a Changing 227e231. World. UNESCO, Paris. Earthscan, London.

87 88 5.5. ¿Son las cianobacterias bentónicas indicadores de presiones antropogénicas en los sistemas fluviales?

Título original: Could benthic freshwaters Cyanobacteria be biological indicators of anthropogenic pressures?

Autores: Laura Monteagudo y José Luis Moreno Publicación: Under Review en Ecological Indicators Impact Factor: 3.444 Fecha 2015

RESUMEN

El uso de las cianobacterias como indicadores biológicos de presiones antropogénicas y en la evaluación del estado ecológico de las masas de agua es bastante limitado en comparación con otros elementos biológicos. Esto puede deberse fundamentalmente a dos razones: (1) el conocimiento limitado sobre su respuesta a estas presiones; (2) la complejidad en la identificación de ciertas especies de cianobacterias para técnicos e investigadores. Sin embargo, las cianobacterias, como otras macroalgas, son organismos estrechamente ligados a las condiciones del agua y pueden ser potenciales indicadores de las presiones humanas, pudiendo llegar a reforzar los índices basados en macrófitos.

En este trabajo se analizó la variación en la comunidad de cianobacterias como respuesta a ocho variables ambientales (pH, conductividad, temperatura, altitud, nitrato, ortofosfato y dos usos de suelo: regadío y secano) en 85 estaciones de muestreo a lo largo de la región. Con ello, se evaluó la capacidad de las cianobacterias como indicadores de dos factores de estrés: el fosfato procedente de las aguas residuales urbanas y el nitrato derivado de los terrenos agrícolas.

Los análisis de modelos lineales realizados (Distance-based linear modelling) mostraron que las variables que mejor explicaban la variación en la composición de

89 especies fueron el pH y la conductividad. Además, se observaron diferencias entre la especies con y sin heterocitos. Mediante la técnica del ‘linkage tree’ o árbol de agrupación, se dividió la muestra en grupos biológicos diferentes atendiendo a diferencias en las variables ambientales por orden de influencia. Esto reveló que la variable ambiental más influyente en la variedad biológica entre grupos fue la conductividad. Comparando los nutrientes, el ortofosfato fue más determinante para la composición de especies que el nitrato. El análisis de porcentajes de similaridad mostró los porcentajes de contribución de cada especie a la similaridad total de cada grupo. Así, Coleodesmium wrangelii, Nostoc caeruleum y Phormidium fonticola resultaron ser las especies principales en los sitios con valores más bajos de conductividad (<30µS·cm-1), sin presiones antropogénicas. P. favosum fue la especie

3- - más común en los sitios con los valores más altos de ortofosfato (>0.15 mg P-PO4 ·l 1), es decir, aquellos afectados por vertidos urbanos. Por otro lado, en ambientes con

3- -1 valores bajos de ortofosfato (<0.02 mg P-PO4 ·l ) las especies del género Rivularia fueron las más comunes. Al contrario de lo esperado, una cianobacteria con heterocitos fue la más frecuente en sitios con los valores más elevados de nitrato (>2

- -1 mg N-NO3 ·l ), independientemente de si el origen de este compuesto era natural o procedente de la agricultura. En ambientes bajo presión agrícola, caracterizados con valores elevados de nitrato y con un alto porcentaje de regadío en su proximidad, una de las especies más comunes fue P. autumnale. Por último, N. verrucosum mostró una gran tolerancia a la mayoría de las variables ambientales medidas, exceptuando las derivadas de las presiones humanas.

De todos estos resultados se puede que algunas cianobacterias pueden

ser útiles en la bioindicación de lasdeducir presiones antropogénicas (p. ej. Nostoc verrucosum, Phormidium autumnale, Plectonema tomasinianum, Rivularia haematites, Tolypothrix distorta); mientras que otras ofrecen más información acerca de las condiciones físico-químicas naturales del agua en sitios no afectados por las actividades humanas (p. ej. N. caureuleum, Phormidiumm fonticola). Por tanto, ampliar el conocimiento sobre la relación entre comunidades de cianobacterias y otras macroalgas con el medio ambiente, en diferentes escenarios de impacto, podría ayudar a mejorar la precisión de los índices basados en macrófitos en un futuro.

90 Could benthic freshwater Cyanobacteria be biological indicators of anthropogenic pressures?

Laura Monteagudo* and José Luis Moreno University of Castilla-La Mancha, Regional Center of Water Research (CREA) Crta. de Las Peñas km. 3, Albacete 02071, Spain. *Corresponding author: Tel.: +34 967599200x2576;fax:+34 967599269; E-mail: [email protected]

ABSTRACT Cyanobacteria are hardly used as biological indicators of anthropogenic pressures or for assessing the ecological status of rivers, possibly for two main reasons: a) their response to anthropogenic pressures is often poorly known; b) reliably identifying cyanobacteria species is a challenge for technicians and researchers. However, some species can be good indicators of anthropogenic pressures and could, therefore, reinforce macrophyte methods to assess ecological status of rivers. We analysed variation in benthic cyanobacterial assemblages at 85 sites in (mainly agricultural) South-Central Spain as a response to eight environmental variables: pH, conductivity, temperature, altitude, nitrate, orthophosphate, irrigation land use and non-irrigation land use. We assessed the usefulness of cyanobacteria species as biological indicators of two human stressors: the high orthophosphate concentration that derives from urban wastewaters; the high nitrate concentration produced by agricultural land use. Distance‐based linear modelling showed that pH and conductivity defined the best model to explain variation in species composition. Differences in the response of heterocystous and non-heterocystous cyanobacteria were found. A linkage tree was used to generate divisive biotic samples clustering according to significant differences in environmental variables. Conductivity was the main environmental factor that contributed to differences between assemblages. Othophosphate was a more influential nutrient for community composition than nitrate. A similarity percentage analysis revealed that Coleodesmium wrangelii, Nostoc caeruleum and Phormidium fonticola were the main contributors at very low conductivity sites (30 µS·cm-1) with no human pressures. P. 3- -1 favosum was the commonest species at the sites with the highest orthophosphate levels (>0.15 mg P-PO4 · l ) impacted by urban wastewater. Rivularia species were dominant at the sites characterised by low 3- -1 orthophosphate values (<0.02 mg P-PO4 · l ), but not impacted by urban wastewater. Unexpectedly, a heterocystous cyanobacterium (R. haematites) became the most frequent species at the sites with highest nitrate - -1 values (> 2 mg N-NO3 ·l ), regardless of being impacted by agriculture or not. P. aumtumnale was present at the impacted sites with the highest orthophosphate and irrigated agriculture values. Finally, Nostoc verrucosum tolerated a wide range of environmental conditions, but avoided anthropogenic pressures, and was absent at the sites with high orthophosphate, nitrate or irrigation land use values. These results suggest that some cyanobacteria species (e.g. Nostoc verrucosum, Phormidium autumnale, Plectonema tomasinianum, Rivularia haematites, Tolypothrix distorta) could be useful tools for the bioindication of anthropogenic pressures, while others provide more information about natural physicochemical reference conditions (N. caeruleum, Phormidium fonticola). Further research into cyanobacteria and macroalgae assemblages in different impacted scenarios could help improve macrophyte indices. 1. INTRODUCTION those based on macroinvertebrates or diatoms (Ferreira et al., 2005). Demars et al. (2012) Since the Water Framework Directive (WFD) reviewed several macrophyte indices and came into force, EU Member States are highlighted the poor accuracy observed in this required to achieve a ‘good status’ in all surface metrics when applied to different regions. and groundwater bodies, which implies Some criticisms about macrophyte indices achieving a ‘good chemical’ and ‘good indicate that they are based on riverine plants ecological’ status. Regarding ecological status, prior to exclusive aquatic taxa and the usual Member States shall establish monitoring exclusion of benthic macroalgae (including systems to estimate the values of the biological cyanobacteria), despite these organisms being quality elements specified per surface water closely linked to water properties. Such category (WFD - Annexe V - 1.4.1.). This fact discrimination could be due to a poor has led to increasingly used metrics based on understanding of their response to different groups of organisms (e.g. anthropogenic pressures (Thacker and Paul, invertebrates, diatoms or macrophytes) to 2001), and technicians and researchers’ assess ecological status. In general, indices difficulty in reliably identifying several species based on macrophytes are less popular than (Marquardt and Palinska, 2007). Despite these

91 difficulties, studies worldwide have related heterocystous and non-heterocystous species; cyanobacteria to several human pressures (3) identify which species of cyanobacteria in responsible for eutrophication in freshwater the study area are suitable as indicators of ecosystems, such as agricultural loadings, human pressures. urban pollution or industrial discharges (e.g. 2. MATERIAL AND METHODS Johansson, 1982; Jafari and Gunale, 2006; Parikh et al., 2006). Although cyanobacteria 2.1. Study area and sampling design have been usually included in studies that have This study was carried out in Castilla-La related algal communities to environmental Mancha (South-Central Spain), a region marked factors (e.g. Potapova et al., 2005; Dell’Uomo by agriculture as its main anthropogenic land and Torrisi, 2009), some physiological features use. As semi-arid climatic conditions have established remarkable differences in the predominate in the study area, irrigation is way they are influenced by nutrients. Unlike needed for high-productivity crops. Thus other algae, cyanobacteria are able to fix 6.26% of agriculture is conducted under atmospheric nitrogen, which allows them to irrigation conditions (Spanish Survey of grow at low rates of dissolved nitrogen Surfaces and Crop Yields (ESYRCE), 2013), and compounds. Some authors have suggested that these cultures are usually located on the banks this characteristic may diminish their of streams and rivers. dependence on nitrogen availability in the water column (e.g. Larkum et al., 1988). In Sites were selected to cover the most complete addition, heterocysts, allow some gradient of environmental conditions, such as cyanobacteria to fix atmospheric nitrogen (N2) lithology and altitude, among others (Figure 1). under aerobic conditions, while N2 fixation is Regarding lithology, three main types were limited to anaerobic and (or) dark conditions defined: calcareous (Cretacic and Jurassic with non-heterocystous cyanobacteria (Potts, limestones), mixed (Quaternary sedimentary 1979; Lee, 2008). Loza et al. (2014) suggested valleys) and siliceous (granites, slates and that this ecophysiological advantage could shales). Reaches were waded in an upstream explain not only the dominance of zigzag pattern and sampling was addressed in heterocystous species at low levels of all habitat types. Thus the macroscopic combined nitrogen, but also the preference of cyanobacteria thalli that grew submerged or in non-heterocystous species for high levels of the splash zone were collected by hand and these compounds. With phosphorus, certain fixed with 4% formalin. Species determination cyanobacteria, such as Rivularia sp., are able to was carried out in the laboratory under a Leica live at phosphorus limitation because of M165C stereoscope and a light microscope phosphatase activity, which is a good indicator Olympus BX50. Glycerine-gelatine was used to of oligotrophic conditions (Mateo et al., 2010). make permanent slides. In this study, we considered only those species that All these facts evidence that cyanobacteria are predominated in macroscopic colonies. a large diverse group of organisms which includes freshwater species that grow at During sampling, pH and conductivity were unimpaired and impaired sites. Thus some of measured in situ with the appropriate sensors these species are expected to be suitable (Multiline P4 WTW). Phosphate (P-PO4 ), - + indicators to account for the development of nitrate (N-NO3 ) and ammonium (N-NH3−4 ) biological indices of water quality. In order to were determined photometrically with MERCK identify them, it is essential to examine the kits (Spectroquant®) in the laboratory. The relationship between the cyanobacteria percentage of four land uses (irrigated community and the surrounding environment agriculture, non-irrigated agriculture, urban conditions in depth, and to also apply and forestry) was quantified within a 1- requirements for good indicator species. Hence kilometre radius buffer upstream of each site, the main objectives of this study were to: (1) according to Monteagudo et al., 2012. analyse which variables determine the benthic Geographic data management was carried out cyanobacteria community in the study area; (2) with ArcGIS 9.2. examine the differences between

92 Figure 1. Study area showing land uses and location of sites. Site symbols correspond to three lithology categories: calcareous (circles), mixed (squares) and siliceous (triangles).

2.2. Statistical analyses the ‘Best’ procedure for the variables selection and the ‘AIC’ (“An Information Criterion”; Boxplots of the environmental variables for Akaike, 1973) criterion for model comparisons each species were performed with Statistica (Anderson et al., 2008). The criterion comes v.9. In order to explore the relationship from the likelihood theory and smaller AIC between the environmental variables and the values indicate a better model. Unlike the R2 cyanobacteria community, we performed the criterion, AIC is not influenced by the number DISTLM and LINKTREE routines implemented of predictor variables (Anderson et al., 2008). in Primer v6. Biological matrices were built Pseudo-F is the statistic for testing the general using the presence/absence data to avoid the null hypothesis of no relationship, which is an influence of factors such as hydrologic stability, analogue of Fisher’s F ratio used in traditional water clarity, grazers or light availability regression. In addition to overall species (Porter et al., 2008). composition, DISTLM was also performed for In order to analyse the relationship between both heterocystous and non-heterocystous cyanobacteria assemblage and the cyanobacteria separately as they were environmental variables, we used the distance- expected to show a different relation with based linear model routine (DISTLM). This nitrate. analysis performs ‘marginal tests’ for each For a further understanding as to how the single variable, and also tests all the possible cyanobacteria community changes in different combinations of variables to find the ‘best environmental scenarios, we used the linkage overall solution’; in other words, the best tree analysis (LINKTREE) included in Primer model to explain variation in biological data. v6. This classification analysis first identifies Among the model-building options, we selected

93 the best subset of environmental variables so 3. RESULTS that the Euclidean distances of the scaled environmental variables show a maximum 3.1. Recorded species correlation with community dissimilarities In this study, we considered only the benthic (biological Bray-Curtis similarity matrix); then species that formed macroscopic thalli, it uses these variables to describe how the best recorded at two sites, at least, with available assemblage samples are split into groups environmental data. Eighteen cyanobacteria (Clarke and Gorley, 2001). Thus while DISTLM species that belong to orders Nostocales and indicates which variables explain variation in Oscillatoriales were selected, including five the overall species composition, LINKTREE heterocystous genera (Coleodesmium Borzì ex combines the most influencing variables, Geitler, Nostoc Vaucher ex Bornet & Flahault, together with the biological assemblage Rivularia C.Agardh ex Bornet & Flahault, similarities among sites, and successively Scytonema C.Agardh ex Bornet & Flahault and divides all the samples into (statistically Tolypothrix Kützing ex Bornet & Flahault) and different) groups. The B% value obtained with four non-heterocystous genera (Microcoleus this analysis is an absolute measure of group Desmazières ex Gomont, Phormidium Kützing differences. Therefore, low B% values ex Gomont, Plectonema Thuret ex Gomont and correspond to the samples that come close Schizothrix Kützing ex Gomont) (Table 1). The together in a multidimensional scaling plot. As most frequent species were Nostoc verrucosum, divisions are carried out in an influential order Rivularia haematites and Phormidium favosum. of environmental variables, dissimilarity On the contrary, Microcoleus subtorulosus, M. between groups decreases in successive vaginatus, Phormidium uncinatum and divisions, as does the B% value. The Tolypothrix robusta were recorded only twice. significance of difference between groups is Certain genera (e.g. Leptolyngbya sp., measured by a simultaneous SIMPROF test Geitlerinema sp., Schizothrix sp. or Lyngbya sp.) (Clarke & Gorley, 2006). This routine performs were identified as secondary/occasional taxa in a series of “similarity profile” permutation tests macroscopic colonies, which were not taken to check if each split produces groups of sites into account in this study. Some taxa that that belong to the same assemblage or not (π formed macroscopic thalli, like statistic). For the purpose of interpreting Cylindrospermum sp., Anabaena sp. or biological differences between groups, we also Nostochopsis lobata Wood ex Bornet et Flahault applied a similarity percentage analysis (Moreno et al., 2012), were also recorded in the (SIMPER) to identify the indicator species study area, but were excluded from this study within each group by calculating its due to insufficient data. contribution based on similarity percentages.

Table 1. The species considered in this study and their frequency of appearance as number of sites (n).

Taxon n Sites Coleodesmium wrangelii Borzì ex Geitler 4 4 70 72 74 Microcoleus subtorulosus Gomont ex Gomont 2 71 77 Microcoleus vaginatus Gomont ex Gomont 2 69 79 Nostoc caeruleum Lyngbye ex Bornet & Flahault 5 4 70 72 73 76 Nostoc sphaericum Vaucher ex Bornet & Flahault 7 2 22 26 33 50 61 62 Nostoc verrucosum Vaucher ex Bornet & Flahault 26 1 4 15 18 21 24 28 31 32 35 36 37 38 42 44 45 52 57 60 64 65 66 77 80 81 85 Phormidium autumnale Gomont 10 5 6 22 24 35 43 48 51 68 79 Phormidium favosum Gomont 23 1 2 7 8 9 10 13 14 15 17 19 25 26 27 32 42 44 46 49 50 56 78 79 Phormidium fonticola Kützing 5 64 71 76 84 85 Phormidium retzii Kützing ex Gomont 6 5 11 20 34 59 64 Phormidium uncinatum Gomont ex Gomont 2 3 60 Plectonema tomasinianum Bornet ex Gomont 13 5 6 12 25 32 35 40 42 44 48 53 57 83 Rivularia biasolettiana Meneghini ex Bornet & Flahault 16 5 13 16 29 33 44 47 55 56 57 60 63 67 69 75 80 Rivularia haematites C.Agardh ex Bornet & Flahault 25 5 8 12 13 15 18 23 30 39 40 41 42 43 46 52 54 57 58 59 60 61 69 81 82 83 Scytonema hofmannii C.Agardh ex Bornet & Flahault 3 8 12 23 Scytonema myochrous C.Agardh ex Bornet & Flahault 4 4 39 57 67 Tolypothrix distorta Kützing ex Bornet & Flahault 15 5 12 24 25 26 33 41 43 47 54 56 57 60 64 76 Tolypothrix robusta N.L.Gardner 2 53 60

Figure 2. Boxplots of the environmental variables for the species considered herein. Species codes in Table 1.

95 3.2. Environmental ranges vaginatus and both Rivularia grew at the sites with the highest non-irrigated land use values. The environmental ranges for each species are compiled in Figure 2. In pH terms, the most 3.3. Distance-based linear model tolerant cyanobacterium was N. verrucosum Regarding species composition (Table 2.a.), (Figure 2.a.), which was able to grow from 6.65 marginal tests revealed that pH, conductivity, to 9.40 pH. Almost all the species were more altitude, nitrate and non-irrigated land use frequent at alkaline sites with pH > 8. However, were all statistically significant variables to , Coleodesmium wrangelii Microcoleus explain variation in species composition when subtorulosus and Nostoc caeruleum were stenoic species that were recorded mainly at the most explanatory variable, despite the acid-neutral values (pH < 7.5). According to proportionconsidered of alone variation (p≤0.05). explained Conductivity being waslow Figure 2.b., and were C. wrangelii N. caeruleum (5.6%). The best model was achieved by a also recorded at the lowest conductivity levels, combination of ‘pH’ and ‘conductivity’, which together with . All three Phormidium fonticola accounted for 9.3% of variability of the are characteristic species of siliceous mountain biological data cloud. streams. Overall, more than 75% of the species records corresponded to sites with Table 2. The distance-based linear model analysis results. P, p- -1 conductivity values below 900 µS·cm . The value; Prop. and R2, proportion of explained variation in the wider conductivity range was shown by biological data for each variable and for the model, respectively. Tolypothrix distorta and three Phormidium a) Species Composition species ( , and ). Marginal tests: Pseudo-F P Prop. P. retzii P. autumnale P. favosum pH 3.248 0.0004 0.038 Regarding temperature (Figure 2.c.), all the Temperature 1.164 0.3044 0.014 Conductivity 4.907 0.0001 0.056 species were found to be more frequent at 15- 3- P-PO4 1.285 0.1901 0.015 20ºC, and P. fonticola and N. caeruleum - N-NO3 3.618 0.0002 0.042 preferred the coldest streams. On the contrary, Altitude 2.385 0.0123 0.028 and grew at the Irrigated land use 1.566 0.1131 0.019 N.verrucosum P. favosum Non-irrigated land use 2.977 0.0018 0.035 warmest sites and also showed the widest Best overall solution: AIC R2 temperature range. Most species were more pH & Conductivity 703.96 0.093 frequent in medium-altitude mountain streams b) Heterocystous species composition (from 400 to 1,000 m a.s.l.), while C. wrangelii, Marginal tests: Pseudo-F P Prop. pH 8.401 0.0002 0.092 N. caeruleum and P. fonticola were recorded Temperature 0.601 0.6012 0.007 exclusively at sites above 800 m a.s.l., which Conductivity 2.690 0.0654 0.031 3- P-PO4 0.687 0.4896 0.008 constituted the typical assemblage in high - N-NO3 1.972 0.1470 0.023 mountain siliceous streams (>1,000 m altitude) Altitude 0.792 0.5054 0.009 (Figure 2.d.). As shown in Figure 2.e., all the Irrigated land use 1.078 0.3964 0.013 Non-irrigated land use 2.770 0.0567 0.032 species were tolerant to low nitrate 2 Best overall solution: AIC R concentrations. However, only N. verrucosum pH & Non-irrigated agriculture 473.07 0.124 and were found to be growing over P. favosum c) Non-Heterocystous species composition -1 -1 12 mgN-NO3 ·l . P. fonticola was recorded at Marginal tests: Pseudo-F P Prop. the lowest nitrate values and showed less pH 1.059 0.3770 0.013 Temperature 1.718 0.2090 0.028 tolerance to this nutrient. Once again (Figure Conductivity 1.183 0.0001 0.125 3- 2.f.), all the species were more frequent at low P-PO4 1.910 0.1649 0.022 - phosphate concentration levels; indeed N-NO3 2.312 0.1053 0.027 P. Altitude 2.683 0.0745 0.031 favosum appeared to be the most phosphate- Irrigated land use 8.788 0.0002 0.096 tolerant species. In general terms, all the Non-irrigated land use 1.957 0.1602 0.023 Best overall solution: AIC R2 species grew at sites with low agricultural land 3- Conductivity, P-PO4 & Irrigated land use 434.88 0.207 use percentages (from 0% to 20%), as shown in Figure 2.g. and 2.h. Tolerance to agricultural With heterocystous species (Table 2.b.), pH land use varied according to species and was the only statistically significant variable in agriculture type. For irrigated agriculture, the marginal tests and explained 9.2% of variation, most tolerant species were M. vaginatus, P. while the best model was that built with the pH autumnale and P. favosum, which were able to and non-irrigated land use combination, which grow in streams surrounded by more than accounted for 12.4% of variation. The best 80% of land used for irrigated agriculture; M. model for the non-heterocystous species explained 20.7% of variation, and was built

96 upon the conductivity, phosphate and irrigated Group H included the sites impacted by land use combination. intensive agriculture practices under irrigation, while groups I and J included the sites with a 3.4. Linkage tree and similarity - -1 nitrate concentration below 1.9 mg N-NO3 ·l percentage contribution - -1 and over 2.0 mg N-NO3 ·l , respectively. R. By starting with the whole group of samples, haematites was the best indicator species in the first split (1) of the classification these three groups, although it reached the dendrogram provided by LINKTREE (Figure 3) highest overall contribution percentage resulted in two groups based on conductivity (78.62%) in group J – a high nitrate values: group A, included very low conductivity concentration. sites (<30 µS·cm-1) and was separated from the 4. DISCUSSION other sites. Therefore, conductivity was the most influential variable, as suggested by the 4.1. Variables that influence the benthic DISTLM results for species composition (Table cyanobacteria assemblages in the study 2.a.). The indicator species of group A were area Coleodesmium wrangelii, Nostoc caeruleum and According to the results of this study, Phormidium fonticola, and they all equally conductivity was the most influential variable contributed (33.33%) to average within-group in the cyanobacteria species composition in the similarity. Conductivity also determined splits study area, and it was only when conductivity 2 and 5 by generating groups B and E, was not that relevant that the influence of respectively. Group B was characterised by low nutrients (phosphate and nitrate, in order of -1 conductivity values (from 43 to 142 µS·cm ), importance) became evident. Biggs (1990) also while group E included sites with medium described conductivity as the most important conductivity values (Figure 2.b.). The main variable for distinguishing periphyton species to contribute to both groups (B and E) community types, and pointed out that was N. verrucosum. The groups generated after conductivity integrates several higher-level group E (F to J) were characterised by medium- processes. Therefore, conductivity may mask -1 high conductivity values (>435 µS·cm ). After the influence of nutrients among other conductivity, the next influential variable was variables in the cyanobacteria species temperature. This variable determined splits 3 composition. For instance, the result of the and 7, which brought about groups C and G, distance-based linear models on the overall characterised by warm water sites (>20 ºC). species composition indicated that the pH and Therefore, the species that typified these conductivity combination was the best model groups were those that tolerated higher than to explain variation. Despite nitrate being average temperatures, mainly N. verrucosum, statistically significant in the marginal tests, Phormidium favosum, Rivularia biasolettiana both nutrients were excluded from this model. and Tolypothrix distorta (Figure 2.c.). According to this result, community The first nutrient in order of importance was composition would not be explained firstly by phosphate, which determined divisions 4 and nutrient concentration, as other studies have 6. The sites that were impacted by urban reported (e.g. Larkum et al., 1988; Thacker and wastewater gave the highest phosphate Paul, 2001). However, the LINKTREE routine 3- -1 concentration (>0.15 mg P-PO4 ·l ) and were helped reveal the underlying relationships included in group D, where P. favosum between the cyanobacteria community and contributed to the highest similarity environmental variables, which are percentage. It is noteworthy that the indicator summarised in Figure 4. Three main species in group D were all non-heterocystous conductivity groups were distinguished: very species, which could explain why the best low (<145 µS·cm-1), low (from 145 to 430 model explained why the non-heterocystous µS·cm-1) and medium-high (>430 µS·cm-1).The species composition included phosphate very low-conductivity sites corresponded to among the selected variables (Table 2.c.). Split the unimpaired freshwater streams located on 6 separated the low from the medium siliceous mountains (impaired sites are phosphate sites. Irrigation and nitrate had an expected to show higher conductivity values). influence only in the last classification steps Here the typical reference cyanobacteria once conductivity was medium-high and the community included Nostoc caeruleum, phosphate concentration was almost negligible. Coleodesmium wrangelii and Phormidium

97 Figure 3. Biotic clustering based on environmental variables and species contributions to groups. COND, -1 3- conductivity (µS·cm ); T, temperature (ºC); IRR, irrigation land use (%); P-PO4 , phosphate concentration 3- -1 - - -1 (mg P-PO4 ·l ); N-NO3 , nitrate concentration (mg N-NO3 ·l ); π: SIMPROF statistic; p: p-value of the SIMPROF test. Species codes in Table 1.

98 fonticola. This is in accordance with Kastovsky cyanobacteria considered herein. These et al. (2010), who reported these three species conductivity levels were recorded in middle in oligotrophic mountain waters. N. verrucosum reaches, as well as in calcareous mountains, was also a typical species at such sites despite and regardless of them being impacted or not. it showing better tolerance to conductivity in Under such conditions, the influence of the study area (Figure 2). variables other than conductivity arose. Thus the next three groups were described The sites included in the low-conductivity according to phosphate concentration (high- group were located at middle reaches, but had urban, medium and low) (Figure 4). The high- no remarkable anthropogenic influence. The urban-phosphate group (>0.15 mg P-PO 3-·l-1) community at these sites was dominated by N. 4 included the sites located downstream of the verrucosum, R. biasolettiana and Tolypothrix urbanised areas and small villages with poor distorta. Despite these species having been water treatment. The typical species in this often recorded in alkaline and mineralised group were mainly P. favosum, followed by P. waters (e.g. Aboal, 1988; Margalef, 1955), they autumnale and Plectonema tomasinianium. were found more frequently at the sites with Both Phormidium species have been previously lower conductivity values than the majority of described to be sewage- and organic pollution- species studied herein. Medium high- tolerant (Lukavský et al., 2006 and Barinova et conductivity values (>430 µS·cm-1) were the al., 2010, respectively). preferred range for the majority of

Figure 4. Summary scheme of the typical cyanobacteria assemblages according to the most relevant variables. The medium-phosphate group was verrucosum replaced P. autumnale when characterised by phosphate values within the compared with the high-phosphate group 3- -1 0.02-0.15 P-PO4 ·l range, which were (Figure 4). This fact supports the statement adequate nutrient conditions for almost all the that N. verrucosum is found mainly in habitats cyanobacteria species compiled herein. Here N. with a low nutrient content (Mollenhauer et al.,

99 1999). Finally, the low-phosphate group while P. favosum replaced P.autumnale. Finally, included a wide variety of sites with null or the low-nitrate group corresponded to the 3- -1 negligible values (<0.02 mg P-PO4 · l ). When oligotrophic sites represented mainly by R. the phosphate concentration was low, nitrate haematites, N. verrucosum and, to a lesser influenced the cyanobacteria assemblages. As extent, by P. favosum, according to the expected in nitrogen-fixing taxa, phosphate had percentage contribution values shown in a stronger impact than nitrate on community Figure 3 (Group I). This suggests that composition since the influence of nitrate was heterocystous species may prevail over non- perceptible only under low phosphate heterocystous species under low nitrate conditions. The dominant species at the low- conditions, as indicated for other regions (e.g. phosphate sites was Rivularia haematites, Stancheva et al., 2013). irrespectively of nitrate concentration. This 4.2. Suitability of species as indicators of could be explained by phosphatase activity, human pressures which would allow Rivularia to metabolise organic phosphorus and to tolerate According to Bellinger and Sigee (2010), given oligotrophic conditions (Sabater et al., 2000; their presence, good indicator species should Mateo et al., 2010). Differences in community provide information on the surrounding composition, depending on nitrate physical and/or chemical environment at a concentrations, were given by species other particular site, and are expected to generally than R. haematites, including heterocystous and have a narrow ecological range, a wide non-heterocystous species (Figure 4). The high geographical distribution and reliable agricultural-nitrate group included the sites identification. Following these criteria, not all located in intensive agricultural areas under cyanobacteria species are suitable as irrigation systems, which is considered the bioindicators of the human pressures in the most important origin of nitrate pollution in study area. For instance, Nostoc caeruleum, the study area (Monteagudo et al., 2012). A Coleodesmium wrangelli and Phormidium direct estimation of nitrate from a water fonticola only provided information on sample would provide a particular measure conductivity conditions; Rivularia biasolettiana that would correspond to a given time. Unlike appeared to be more related to temperature nitrate concentration, the percentage of land than to other variables; P. favosum was seen to used for irrigated agriculture would represent be tolerant to a wide range of nitrate and an indirect long-term nitrate measure, and phosphate concentrations; several species would more accurately reflect the real impact were infrequent in the study area. Conversely, scenario. Therefore, the typical species in this other species met the above-mentioned group are expected to be either nitrophilous- requirements and could be useful tools to or nitrate-tolerant. The characteristic develop biological indices, these being R. cyanobacteria assemblages in nitrogen-rich haematites, N. verrucosum, P. autumnale, sites influenced by irrigated agriculture Tolypothrix distorta and Plectonema included R. haematites, P. autumnale and T. tomasinianum. . This result agrees with Komárek and distorta N. verrucosum is a well-known worldwide Anagnostidis (2005), who described P. distributed species that is normally related to autumnale to be a nitrophilous species. low nutrient contents (Ennis, 1978; However, the same authors related T. distorta Mollenhauer et al., 1999). In our study area it with unpolluted waters, which suggests that T. was the most frequent species located at the distorta shows good tolerance and is able to sites characterised by low phosphate, nitrate grow in the absence and presence of nitrate, as and agriculture land use values. Thus the indicated by Ogawa and Carr (1969). In the results suggest that N. verrucosum is a sensitive medium-nitrate group, the concentration was species which could be a useful indicator of - -1 over 2.0 mg N-NO3 ·l , and the source of this absence of anthropogenic pressures. nutrient could be anthropogenic or natural, or even both. The sites in this group would be P. autumnale exclusively characterised sites located mainly on calcareous mountains with under anthropogenic pressure, which agrees natural nitrate inputs or in non-irrigated with other authors who have reported this agricultural areas where low fertilisation is species in polluted waters (Barinova et al., applied. The cyanobacteria assemblage in this 2010; Loza et al., 2013). Therefore, the group was also dominated by R. haematites, presence of this species could provide valuable

100 information about the water quality at a given phosphate had a stronger influence than site, providing it is not confused with P. nitrate. favosum, a morphologically similar species that (2) Nostoc verrucosum, Phormidium autumnale, is sometimes misidentified as P.autumnale, and Plectonema tomasinianum, Rivularia haematites vice versa (Komárek and Anagnostidis, 2005). and Tolypothrix distorta have been found to be P. tomasinianum has often been related to sensitive taxa to nutrient contents, and could oligotrophic conditions (Wehr and Sheath, be suitable indicators of anthropogenic 2003; Kastovsky et al., 2010), and less pressures. Therefore, they could be useful tools frequently to medium to high phosphorus for developing biological water quality indices. concentrations (Schneider and Melzer, 2003). (3) Regional studies on the ecological ranges In this study, appeared among P. tomasinianum and optima of species are desirable prior to the typical species in the cyanobacterial selecting those appropriate for bioindications. community of both high and medium phosphate concentrations, which suggests its (4) The suitability of some species as potential as an indicator of pollution in the bioindicators may vary from one different study area. This divergence of results may geographic region to another. suggest that the suitability of some species as (5) Further research into the indicator bioindicators may vary in different geographic potential of benthic cyanobacteria is desirable regions. to use them to improve the accuracy of R. haematites grew mainly at the sites with biological indices to assess the ecological status nitrate enrichment from both agricultural and of rivers according to the Water Framework natural sources, provided that the phosphate Directive. 3- -1 concentration was low (<0.02 mg P-PO4 ·l ). Previous references have also related this species with surrounding arable lands and Acknowledgements: nutrient enrichment (Johansson, 1982), and This study has been supported by several also with high N:P ratios (Stancheva et al., projects granted by the Regional Government 2013; Aboal et al., 2014). However, this species of Castilla-La Mancha (Spain): PO1109-0190- also appears under nitrate-reduced conditions. 8090, PPII10-0271-1349, PREG01-0016 and Therefore, it could be a good indicator of PREG06-027. absence of urban impacts in our study area, but not a good indicator of agricultural impact or nitrate enrichment given its high level of 6. REFERENCES tolerance to this nutrient. Aboal, M., 1988. Aportación al conocimiento de las References on T. distorta to low nutrient algas epicontinentales del Sudeste de España. III: contents (mainly total phosphorus or Cianofíceas (Cyanophyceae Schaffner 1909). An. phosphate) are frequent (e.g. Schneider and Jardín Botánico Madr. 45, 3–46. Melzer, 2003; Kastovsky et al., 2010; Mateo et Aboal, M., García-Fernández, M.E., Roldán, M., Whitton, al., 2010). Even the genus Tolypothrix is B.A., 2014. Ecology, morphology and physiology of described as being an inhabitant of clear Chroothece richteriana (Rhodophyta, unpolluted waters (Wehr and Sheath, 2003). Stylonematophyceae) in the highly calcareous Río However in our study area, T. distorta grew at Chícamo, south-east Spain. Eur. J. Phycol. 49, 83–96. sites with low phosphate concentrations, but in doi:10.1080/09670262.2014.893018 agricultural areas, and could be an indicator of Akaike H., 1973. Information theory as an extension of agricultural impact. This fact reinforces the the maximum likelihood principle. In: Petrov B.N. statement that the ecological ranges of species and Caski F. (eds.) Proceedings 2nd International should be studied regionally prior to Symposium of Information Theory. pp. 267-281. establishing the indicator potential of a Akademiai Kiado, Budapest. particular species, as suggested by Rott and Anderson, M., Gorley, R.N., Clarke, R.K., 2008. Schneider (2014). Permanova+ for Primer: Guide to Software and 5. CONCLUSIONS Statisticl Methods. Barinova, S., Tavassi, M., Glassman, H., Nevo, E., 2010. (1) In the study area, the most influential Algal indication of pollution in the Lower Jordan variable on species composition was River, Israel. Appl. Ecol. Environ. Res. 8, 19–38. conductivity. Among the considered nutrients,

101 Loza, V., Perona, E., Mateo, P., 2014. Specific responses identification and use as bioindicators. Wiley- to nitrogen and phosphorus enrichment in Bellinger,Blackwell, E.G., Chichester, Sigee, D.C., West 2010. Sussex Freshwater [u.a.]. algae : cyanobacteria: Factors influencing changes in species dominance along eutrophic gradients. Biggs, B.J.F., 1990. Periphyton communities and their Water Res. 48, 622–631. environments in New Zealand rivers. N. Z. J. Mar. doi:10.1016/j.watres.2013.10.014 Freshw. Res. 24, 367–386. doi:10.1080/00288330.1990.9516431 Lukavský, J., Moravcová, A., Nedbalová, L., Rauch, O., 2006. Phytobenthos and water quality of mountain Clarke, K.R., Gorley, R.N., 2001. PRIMER v6 PRIMER-E streams in the Bohemian Forest under the influence Ltd. Plymouth UK. of recreational activity. Biologia (Bratisl.) 61. Clarke K.R and Gorley R.N. 2006. PRIMER v6: User doi:10.2478/s11756-007-0069-5 manual/Tutorial. PRIMER-E Ltd, Plymouth. Margalef, R., 1955. Comunidades bióticas de las aguas Dell’Uomo, A., Torrisi, M., 2009. Freshwater algae and dulces del noroeste de España. Publicaciones Inst. their use for safeguarding the Mediterranean basin. Biol. Apl. 21, 5–85. Bocconea 23, 93–105. Marquardt, J., Palinska, K.A., 2007. Genotypic and Demars, B.O.L., Potts, J.M., TréMolièRes, M., ThiéBaut, phenotypic diversity of cyanobacteria assigned to G., Gougelin, N., Nordmann, V., 2012. River the genus Phormidium (Oscillatoriales) from macrophyte indices: not the Holy Grail!: River different habitats and geographical sites. Arch. macrophyte indices: an enticing mirage. Freshw. Microbiol. 187, 397–413. doi:10.1007/s00203-006- Biol. 57, 1745–1759. doi:10.1111/j.1365- 0204-7 2427.2012.02834.x Mateo, P., Berrendero, E., Perona, E., Loza, V., Whitton, Ennis, G.L., 1978. Algal indicators of water quality in B.A., 2010. Phosphatase activities of cyanobacteria the Flathead River Basin. J PhycolUnited States 14. as indicators of nutrient status in a Pyrenees river. Hydrobiologia 652, 255–268. doi:10.1007/s10750- Ferreira, M.T., Rodríguez-González, P.M., Aguiar, F.C., 010-0338-0 Albuquerque, A., 2005. Assessing biotic integrity in Iberian rivers: Development of a multimetric plant Mollenhauer, D., Bengtsson, R., Lindstrøm, E.-A., 1999. index. Ecol. Indic. 5, 137–149. Macroscopic cyanobacteria of the genus Nostoc: a doi:10.1016/j.ecolind.2005.01.001 neglected and endangered constituent of European inland aquatic biodiversity. Eur. J. Phycol. 34, 349– Jafari, N. G., & Gunale, V. R., 2006. Hydrobiological study 360. doi:10.1080/09670269910001736412 of algae of an urban freshwater river. J. Appl. Sci. Environ. Manag. 10 (2).153–158.Johansson, C., Monteagudo, L., Moreno, J.L., Picazo, F., 2012. River 1982. Attached algal vegetation in running waters of eutrophication: Irrigated vs. non-irrigated Jämtland, Sweden. agriculture through different spatial scales. Water Res. 46, 2759–2771. Kastovsky, J., Hauer, T., Komarek, J., Skacelova, O., 2010. doi:10.1016/j.watres.2012.02.035 The list of cyanobacterial species of the Czech Republic to the end of 2009. Fottea 245–249. Moreno, J.L., Aboal, M., Monteagudo, L., 2012. On the doi:10.5507/fot.2010.015 presence of Nostochopsis lobata Wood ex Bornet et Flahault in Spain: morphological, ecological and Komárek, J., Anagnostidis, K., 2005. Süßwasserflora von biogeographical aspects. Nova Hedwig. 95, 373– Mitteleuropa 19/2. Cyanoprokaryota 2. Teil: 390. doi:10.1127/0029-5035/2012/0052 Oscillatoriales. Spektrum, Heidelberg. Ogawa, R.E., Carr, J.F.C., 1969. The influence of nitrogen Larkum, A.W.D., Kennedy, I.R., Muller, W.J., 1988. on heterocyst production in blue-green algae. Nitrogen fixation on a coral reef. Mar. Biol. 98, 143– Limnol. Oceanogr. 14, 342–351. 155. doi:10.1007/BF00392669 doi:10.4319/lo.1969.14.3.0342 Lee, R. E. (2008). Phycology. Cambridge Parikh, A., Shah, V., Madamwar, D., 2006. University Press. Cyanobacterial Flora from Polluted Industrial Loza, V., Berrendero, E., Perona, E., Mateo, P., 2013. Effluents. Environ. Monit. Assess. 116, 91–102. Polyphasic characterization of benthic doi:10.1007/s10661-006-7229-x cyanobacterial diversity from biofilms of the Porter, S.D., Mueller, D.K., Spahr, N.E., Munn, M.D., Guadarrama river (Spain): morphological, Dubrovsky, N.M., 2008. Efficacy of algal metrics for molecular, and ecological approaches 1. J. Phycol. 49, assessing nutrient and organic enrichment in 282–297. doi:10.1111/jpy.12036 flowing waters. Freshw. Biol. 53, 1036–1054. doi:10.1111/j.1365-2427.2007.01951.x

102 DISCUSIÓN GENERAL Arroyo de Nava del Rey, Huertezuelas (Ciudad Real) Fotografía: José Luis Moreno Alcaraz

103 104 6. DISCUSIÓN GENERAL

Los resultados de esta Tesis Doctoral, muestran que Castilla-La Mancha es una región rica en macroalgas. En general, los filos Cyanophyta (=Cyanobacteria) (algas verdeazules), Rhodophyta (algas rojas) y Chlorophyta (algas verdes) están ampliamente distribuidos en el área de estudio y cuentan con numerosos representantes. Muchos de ellos están mundialmente extendidos, mientras que algunas especies han sido documentadas de forma ocasional en Europa. Estas especies raras han sido: Nostochopsis lobata Wood ex Bornet et Flahaut, Batrachospermum atrum (Hudson) Harvey, Chroothece richteriana Hansgirg, Oocardium stratum Nägeli y Tetrasporidium javanicum Möbius. Sin embargo, con la revisión en el futuro de todo el material recolectado a nivel específico, no se descarta la aparición de más especies de interés regional, nacional o europeo. En concreto N. lobata únicamente ha sido detectada en cuatro localidades europeas, localizadas en cuatro países (Italia, Francia, Portugal y España), y en sólo una única ocasión por localidad. Sin embargo, esta especie es muy abundante en la zona tropical, donde desarrolla colonias de mayor tamaño que las descritas en Europa. Tal y como se refleja en el Artículo 5.3., las citas de esta especie tropical en la zona templada han aumentado en los últimos años. A pesar de no ser una especie apta para la bioindicación de presiones antropogénicas debido a su escasez en nuestro continente, s podría ser útil como indicadora de calentamiento global. De

la misma manera, la íinestabilidad de poblaciones de Hydrurus foetidus (Vilars) Trevisan, en algunas zonas montañosas, también puede ser un efecto de cambios en el clima, ya que es una especie estenoterma, mundialmente extendida, que se desarrolla en aguas frías dentro de un rango muy reducido de temperatura (entre 2 y 12°C).

En cuanto al resto de taxones, la mayoría presentan una distribución geográfica amplia, tal y como se espera en especies indicadoras. Sin embargo, sólo unos pocos de ellos han sido incluidos en algún índice de macrófitos. En la comparación de métricas realizada por Demars et al. (2012), se observó que menos de la mitad de los índices incluían macroalgas y, de hacerlo, se consideraban preferentemente a nivel de género, a pesar de que la diversidad ecológica intragenérica puede ser muy

105 elevada. Todo ello indica que mejorar la precisión de las métricas basadas en macrófitos sería posible si se incluyeran macroalgas indicadoras a nivel de especie.

Otro de los requisitos que debe cumplir una especie indicadora es, obviamente, que sea capaz de aportar información sobre los impactos que afectan al medio en el que vive, por lo que el estudio de estas presiones es un punto clave en la búsqueda de especies indicadoras. Una de las contribuciones más relevantes de este trabajo tiene que ver con el análisis del impacto de los usos de suelo en la calidad del agua y, más concretamente, en cómo analizar este factor de la manera más adecuada. En Castilla- La Mancha las principales presiones sobre los ríos y arroyos son la agricultura (principal fuente de contaminación difusa) y los vertidos de aguas residuales urbanas (contaminación puntual). Según los resultados obtenidos en el Artículo 5.4., la agricultura aparece como la causa principal de eutrofización en los sistemas fluviales, en concordancia con otros estudios realizados previamente en la zona (Moreno et al., 2006; Lassaletta et al., 2009). Comparando las diferentes técnicas agrícolas, los resultados indicaron que el regadío es la que más impacto causa en el sistema fluvial castellano-manchego, a pesar de ocupar una superficie mucho menor que el secano. Además, se detectó un umbral de impacto que sugiere que al superar un 50% de suelo dedicado al cultivo en regadío en zonas muy cercanas a los cauces (1km), se incrementa considerablemente el riesgo de eutrofización. Este hecho apoya la idea de que la intensidad del uso agrícola puede ser un mejor indicador del impacto que el porcentaje de suelo que ocupa (Harding et al., 1999).

En el caso del regadío, la superficie que ocupa es mucho menor que la del secano por lo que empleando escalas espaciales amplias (p. ej. área total de drenaje)

el secano enmascara, los resultados y no es posible detectar el efecto del regadío,. Para evitar este efecto, se testaron diversas escalas, y los resultados mostraron que la más adecuada fue la zona de influencia de 1km de radio aguas arriba, ya que con ella se obtiene el gradiente de datos más completo para todos los usos de suelo y se evita el problema de la autocorrelación espacial (Figura 4). Estos resultados subrayan que la escala de estudio afecta a los resultados y que, tal y como sugiere Chang (2008) es necesario considerar los enfoques ‘multiescala’ para poder aclarar la dinámica de la calidad del agua en el tiempo y el espacio. La contribución más novedosa del Artículo

106 5.4. es la propuesta de una metodología a seguir para la elección de la escala más adecuada en este tipo de estudios.

Figura 4. Elección de la escala espacial adecuada al caso de estudio.

Una vez estudiada la diversidad de macroalgas e identificadas las presiones se procedió a analizar cómo afectan las variables ambientales a la comunidad de macroalgas (Artículo 5.5.). En este caso, los usos de suelo se contabilizaron a través de áreas de influencia de 1 km de radio aguas arriba, tal y como se estableció en el trabajo anterior. El grupo de algas elegido como objeto de estudio fue el de las cianobacterias, ya que estaban ampliamente representadas y fueron detectadas en una gran variedad de ambientes (zonas con impacto, aguas puras, etc.). Según los resultados obtenidos, la conductividad fue la variable más determinante en la composición de especies de la comunidad de cianobacterias (Figura 5). Una vez descartado su efecto, se pudo observar el efecto de los nutrientes, fosfato y nitrato, por orden de importancia. Tal y como apunta Biggs (1990), esto puede deberse a que la conductividad es un factor que integra otros factores de gran envergadura ecológica que actúan a gran escala, tales como litología, clima, etc. Aunque no se encontraron grandes diferencias entre cianobacterias con heterocitos y sin ellos, las primeras fueron más frecuentes en los sitios con un nivel de nitrato bajo, tal y como

107 se esperaba y como se ha descrito en otras zonas geográficas (p. ej. Stancheva et al., 2013).

Figura 5. Factores determinantes y especies típicas en las comunidades de cianobacterias.

Observando los rangos ecológicos de las especies y dados los resultados de análisis de la composición de las comunidades según las condiciones ambientales, se pudieron identificar cinco especies adecuadas para la bioindicación de las presiones humanas. Nostoc verrucosum Vaucher ex Bornet et Flahault parece indicar de ausencia de presiones, de acuerdo con descripciones previas de la ecología de la especie (Ennis, 1978; Mollenhauer et al., 1999). Phormidium autumnale Gomont se encontró en sitios impactados tanto por vertidos urbanos como por agricultura, en concordancia con los estudios que relacionan esta especie con aguas contaminadas (Barinova et al., 2010; Loza et al., 2013). Tanto Plectonema tomasinianum Bornet ex Gomont como Tolypothrix distorta Kützing ex Bornet et Flahault aparecen

108 relacionados en la literatura con aguas oligotróficas (p. ej Schneider and Melzer, 2003 y Kastovsky et al., 2010, respectivamente). Sin embargo, en el área de estudio P. tomasinianum parece indicar impacto urbano y T. distorta impacto agrícola, lo que apoya la teoría sugerida por Rott y Schneider (2014) de que antes de establecer el potencial indicador de una especie es necesario estudiar los rangos ecológicos de la misma a nivel regional. Por último, Rivularia haematites C. Agardh ex Bornet et Flahault mostró predilección por los sitios con alto nivel de nitrato, de acuerdo con trabajos previos que relacionan esta especie con ratios elevados de N:P (Stancheva et al., 2013; Aboal et al., 2014). Sin embargo, no puede considerarse un buen indicador de impacto agrícola ya que también se desarrolla en lugares pobres en nitrato. Por otro lado, su tolerancia al fosfato fue muy reducida, por lo que sí puede considerarse un buen indicador de ausencia de impacto urbano.

Tanto la metodología establecida para el estudio de las presiones como el enfoque aplicado en la evaluación del potencial indicador de las cianobacterias, pueden servir de base para identificar más especies de algas indicadoras en ésta y otras áreas geográficas. Ampliar el abanico de bioindicadores disponibles podrá contribuir, en un futuro, al perfeccionamiento de los índices biológicos de calidad del agua empleados en aplicación de la Directiva Marco De Agua (2000/60/CE).

109 110 CONCLUSIONES Río Júcar en Las Mariquillas (Albacete) Fotografía: José Luis Moreno Alcaraz

111 112 7. CONCLUSIONES

De los resultados obtenidos en los trabajos comprendidos en esta Tesis Doctoral, se desprenden las siguientes conclusiones, agrupadas según las líneas de trabajo planteadas en la misma:

7.1. Diversidad y distribución de las macroalgas bentónicas de Castilla-La Mancha:

. Castilla-La Mancha es una región con una gran diversidad de macroalgas fluviales. Los filos Cyanophyta (=Cyanobacteria) y Chlorophyta son los más frecuentes y más ampliamente distribuidos. . Hasta el momento, la región cuenta con la presencia de cinco especies consideradas ‘raras’ a nivel europeo: Nostochopsis lobata Wood ex Bornet et Flahaut, Batrachospermum atrum (Hudson) Harvey, Chroothece richteriana Hansgirg, Oocardium stratum Nägeli y Tetrasporidium javanicum Möbius. . El hallazgo de Nostochopsis lobata en el Arroyo de Nava del Rey (Ciudad Real), supone la primera cita nacional publicada de esta especie y la cuarta europea. . Los ejemplares de Hydrurus foetidus (Villars) Trevisan encontrados en el Parque Natural del Hayedo de Tejera Negra (Guadalajara), suponen la primera cita de esta especie en la región.

7.2. Presiones antropogénicas en el área de estudio:

. El regadío es la principal causa de eutrofización en el sistema fluvial castellano-manchego. Lo que indica que tanto la intensidad de la práctica agrícola como su proximidad a los cauces son factores importantes a tener en cuenta. . Sobrepasar el umbral de presión del 50% del suelo dedicado al regadío en un radio de influencia de 1km en ambas márgenes de los ríos, puede ocasionar un aumento rápido de la concentración de nitrato y, por tanto, eutrofización de las aguas.

113 . La elección de la escala espacial puede afectar a los resultados en estudios que relacionan los usos de suelo con la calidad del agua y, por tanto, puede influir también en la interpretación ecológica de los mismos. . En el área de estudio, la escala espacial más adecuada para analizar el impacto de la agricultura fue el radio de influencia de 1km aguas arriba de las estaciones de muestreo. . La escala más adecuada puede variar según el objeto de estudio y, en ocasiones, puede haber incluso más de una escala espacial que proporcione resultados reales y fiables. . Como contribución novedosa, la siguiente metodología puede servir de base en la búsqueda de la escala más apropiada para el estudio de los usos de suelo como causantes de la eutrofización en ríos y arroyos: a) Es importante analizar qué escala(s) proporciona(n) el rango o gradiente más amplio de datos para cada clase de uso de suelo que se estudia. Cuanto más amplio sea este rango, más fiables serán los resultados obtenidos. b) Es necesario comprobar que no exista autocorrelación espacial de los datos, algo frecuente en distancias geográficas reducidas. c) La escala más adecuada será aquella que nos permita trabajar con la muestra más completa, sin generar autocorrelación.

7.3. Relación entre la comunidad de cianobacterias y la calidad del agua:

. En el área de estudio, la conductividad es la variable más determinante en la composición de especies de la comunidad de cianobacterias. Comparando los nutrientes estudiados, el fosfato influye más en la comunidad que el nitrato. . Para evaluar el potencial indicador de una determinada especie es conveniente analizar previamente sus rangos de tolerancia y sus óptimos ecológicos a nivel regional. . El potencial indicador de una determinada especie puede variar de una zona geográfica a otra. . Nostoc verrucosum Vaucher ex Bornet et Flahault, Phormidium autumnale Gomont, Plectonema tomasinianum Bornet ex Gomont, Rivularia haematites C.

114 Agardh ex Bornet et Flahault y Tolypothrix distorta Kützing ex Bornet et Flahault han mostrado ser especies sensibles a los niveles de nutrientes y tener potencial como bioindicadores de las presiones antropogénicas. Por tanto, podrían tenerse en cuenta en el desarrollo de índices biológicos.

115

116

BIBLIOGRAFÍA Tapetes de Phormidium sp.

Fotografía: José Luis Moreno Alcaraz

117

118 8. BIBLIOGRAFÍA

Aboal, M., 1996. Comunidades fitobentónicas del Parque Natural de las Tablas de Daimiel., en: Las Tablas de Daimiel. Ecología Acuática Y Sociedad. Álvarez, M. & Cirujano, S. (eds.), Madrid, pp. 117–128. Aboal, M., 1989a. Contribución al conocimiento de las algas epicontinentales del SE de España, II: los rodófitos (Rhodophyceae). Lazaroa 11, 115–122. Aboal, M., 1989b. Aportación al conocimiento de las algas epicontinentales del SE de España. V. Xantofíceas (Xanthophyceae P. Allorge ex Fritsch, 1935). Bol. Soc. Broteriana 62, 239–248. Aboal, M., 1989c. Epilithic algal communities from River Segura basin, South-eastern Spain. Arch. Für Hydrobiol. 116, 113–124. Aboal, M., 1988a. Aportación al conocimiento de las algas epicontinentales del Sudeste de España. III: Cianofíceas (Cyanophyceae Schaffner 1909). An. Jardín Botánico Madr. 45, 3–46. Aboal, M., 1988b. Aportación al conocimiento de las algas epicontinentales del SE de España.VII. Clorofíceas (Chlorophyceae Wille in Warming 1884). Candollea 43, 521–548. Aboal, M., 1988c. Zygnemataceae (Conjugales, Chlorophyceae) of the river of the Segura basin, southeastern Spain. Nova Hedwig. 47, 389–402. Aboal, M., García-Fernández, M.E., Roldán, M., Whitton, B.A., 2014. Ecology, morphology and physiology of Chroothece richteriana (Rhodophyta, Stylonematophyceae) in the highly calcareous Río Chícamo, south-east Spain. Eur. J. Phycol. 49, 83–96. doi:10.1080/09670262.2014.893018 Aboal, M., Llimona, X., 1985. Aportación al conocimiento de las algas del SE de España. I. Caráceas (Characeae). An. Biol. 6, 7–17. Álvarez Cobelas, M. A., 2007. Ecología acuática y sociedad de las lagunas de Ruidera. Consejo Superior de Investigaciones Científicas, Madrid. Alvarez-Cobelas, M. A., Sánchez-Andrés, R., Sánchez-Carrillo, S., Angeler, D.G., 2010. Nutrient contents and export from streams in semiarid catchments of central Spain. J. Arid Environ. 74, 933–945. doi:10.1016/j.jaridenv.2010.01.002 Barinova, S., Tavassi, M., Glassman, H., Nevo, E., 2010. Algal indication of pollution in the Lower Jordan River, Israel. Appl. Ecol. Environ. Res. 8, 19–38. Bellinger, E.G., Sigee, D.C., 2010. Freshwater algae: identification and use as bioindicators. Wiley- Hoboken, NJ. Berzas, J.J., García, L.F., Martín-Álvarez, P.J., Rodríguez Martín-Doimea, R.C., 2004. Blackwell, Chichester, West Sussex, UK ; Quality Assessment and Chemometric Evaluation of a Fluvio-Lacustrine System: Ruidera Pools Natural Park (Spain). Water. Air. Soil Pollut. 155, 269– 289. doi:10.1023/B:WATE.0000026532.72630.10 Biggs, B.J.F., 1990. Periphyton communities and their environments in New Zealand rivers. N. Z. J. Mar. Freshw. Res. 24, 367–386. doi:10.1080/00288330.1990.9516431

119 Carpenter, S.R., Caraco, N.F., Correll, D.L., Howarth, R.W., Sharpley, A.N., Smith, V.H., 1998. Nonpoint pollution of surface waters with phosphorous and nitrogen. Ecol. Appl. 8, 559–568. doi:10.1890/1051- 0761(1998)008[0559:NPOSWW]2.0.CO;2 Cavero, J., Beltrán, A., Aragués, R., 2003. Nitrate Exported in Drainage Waters of Two Sprinkler-Irrigated Watersheds. J. Environ. Qual. 32, 916–926. Chang, H., 2008. Spatial analysis of water quality trends in the Han River basin, South Korea. Water Res. 42, 3285–3304. doi:10.1016/j.watres.2008.04.006 Cirujano Bracamonte, S., Medina Domingo, L., 2002. Plantas acuáticas de las lagunas y humedales de Castilla-La Mancha. Real Jardín Botánico Junta de Comunidades de Castilla-La Mancha, Madrid. Cirujano, S., Meco, A., Cezón, K., 2011. Flora acuática: macrófitos, TAXAGUA. Consejo Superior de Investigaciones Científicas, Madrid. Demars, B.O.L., Potts, J.M., TréMolièRes, M., ThiéBaut, G., Gougelin, N., Nordmann, V., 2012. River macrophyte indices: not the Holy Grail!: River macrophyte indices: an enticing mirage. Freshw. Biol. 57, 1745–1759. doi:10.1111/j.1365-2427.2012.02834.x D.O.C.E. 2000. Directiva 2000/60/CE del Parlamento Europeo y del Consejo de 23 de octubre de 2000 por la que se establece un marco comunitario de actuación en el ámbito de la política de aguas. D.O.C.E. L 327 de 22.12.00. 69 pp. Dodkins, I., Aguiar, F., Rivaes, R., Albuquerque, A., Rodríguez-González, P., Ferreira, M.T., 2012. Measuring ecological change of aquatic macrophytes in Mediterranean rivers. Limnol. - Ecol. Manag. Inland Waters 42, 95–107. doi:10.1016/j.limno.2011.09.001 Ennis, G.L., 1978. Algal indicators of water quality in the Flathead River Basin. J PhycolUnited States 14. Ferreira, M.T., Rodríguez-González, P.M., Aguiar, F.C., Albuquerque, A., 2005. Assessing biotic integrity in Iberian rivers: Development of a multimetric plant index. Ecol. Indic. 5, 137–149. doi:10.1016/j.ecolind.2005.01.001. Flor-Arnau, N., M. Real, G. González, J. Cambra, J. L. Moreno, C. Solà y A. Munné., 2015. Índice de Macrófitos Fluviales (IMF), una nueva herramienta para evaluar el estado ecológico de los ríos mediterráneos. Limnetica 34 (1): 95-114. Frissell, C.A., Liss, W.J., Warren, C.E., Hurley, M.D., 1986. A hierarchical framework for stream habitat classification: Viewing streams in a watershed context. Environ. Manage. 10, 199–214. doi:10.1007/BF01867358 Harding, J.S., Benfield, E.F., Bolstad, P.V., Helfman, G.S., Jones, E.B.D., 1998. Stream biodiversity: The ghost of land use past. Proc. Natl. Acad. Sci. U. S. A. 95, 14843 –14847. Harding, J.S., Young, R.G., Hayes, J.W., Shearer, K.A., Stark, J.D., 1999. Changes in agricultural intensity and river health along a river continuum. Freshw. Biol. 42, 345–357. doi:10.1046/j.1365-2427.1999.444470.x Jafari, N.G., Gunale, V.R., 2006. Hydrobiological study of algae of an urban freshwater river. J. Appl. Sci. Environ. Manag. 10, 153–158.

120 Johansson, C., 1982. Attached algal vegetation in running waters of Jämtland, Sweden. Johnson, L., Richards, C., Host, G., Arthur, J., 1997. Landscape influences on water chemistry in Midwestern stream ecosystems. Freshw. Biol. 37, 193–208. doi:10.1046/j.1365-2427.1997.d01-539.x Kastovsky, J., Hauer, T., Komarek, J., Skacelova, O., 2010. The list of cyanobacterial species of the Czech Republic to the end of 2009. Fottea 245–249. doi:10.5507/fot.2010.015 Larkum, A.W.D., Kennedy, I.R., Muller, W.J., 1988. Nitrogen fixation on a coral reef. Mar. Biol. 98, 143–155. doi:10.1007/BF00392669 Lassaletta, L., García-Gómez, H., Gimeno, B.S., Rovira, J.V., 2009. Agriculture-induced increase in nitrate concentrations in stream waters of a large Mediterranean catchment over 25 years (1981–2005). Sci. Total Environ. 407, 6034–6043. doi:10.1016/j.scitotenv.2009.08.002 Lee, R.E., 2008. Phycology, 4th ed. ed. Cambridge University Press, Cambridge, (England); New York. Loza, V., Berrendero, E., Perona, E., Mateo, P., 2013. Polyphasic characterization of benthic cyanobacterial diversity from biofilms of the Guadarrama river (Spain): morphological, molecular, and ecological approaches 1. J. Phycol. 49, 282–297. doi:10.1111/jpy.12036 Loza, V., Perona, E., Mateo, P., 2014. Specific responses to nitrogen and phosphorus enrichment in cyanobacteria: Factors influencing changes in species dominance along eutrophic gradients. Water Res. 48, 622–631. doi:10.1016/j.watres.2013.10.014 Lund, J.W., 1967. Eutrophication. Nature 214, 557–558. doi:10.1038/214557a0 Marquardt, J., Palinska, K.A., 2007. Genotypic and phenotypic diversity of cyanobacteria assigned to the genus Phormidium (Oscillatoriales) from different habitats and geographical sites. Arch. Microbiol. 187, 397–413. doi:10.1007/s00203-006-0204-7 Mateo, P., Berrendero, E., Perona, E., Loza, V., Whitton, B.A., 2010. Phosphatase activities of cyanobacteria as indicators of nutrient status in a Pyrenees river. Hydrobiologia 652, 255–268. doi:10.1007/s10750-010-0338-0 Ministerio de Agricultura, Alimentación y Medio Ambiente, 2014. Anuario de Estadística 2013. Madrid. Mollenhauer, D., Bengtsson, R., Lindstrøm, E.-A., 1999. Macroscopic cyanobacteria of the genus Nostoc: a neglected and endangered constituent of European inland aquatic biodiversity. Eur. J. Phycol. 34, 349–360. doi:10.1080/09670269910001736412 Moreno, J., Navarro, C., De las Heras, J. 2006. Abiotic ecotypes in south-central Spanish rivers: Reference conditions and pollution. Environ. Pollut. 143, 388– 396. doi:10.1016/j.envpol.2005.12.012. Moreno, J. L., C. Navarro & J. De Las Heras. 2006. Propuesta de un índice de vegetación acuática (IVAM) para la evaluación del estado trófico de los ríos

121 de Castilla-La Mancha: Comparación con otros índices bióticos. Limnetica, 25 (3): 821-838. Moreno, J.L., J. De Las Heras, N. Prat & M. Rieradevall. 2008. Evaluación del estado trófico de tres cuencas interiores de Cataluña (Foix, Besos y Llobregat) mediante la vegetación acuática: aplicación de un índice trófico (IVAM-FBL). Limnetica, 27 (1): 107-118. Munné, A., C. Solà, M. Rieradevall & N. Prat. 1998. Índex QBR. Mètode Per A L’avaluació De La Qualitat Dels Ecosistemes De Ribera. Estudis De La Qualitat Ecológica Deis Rius. Diputació De Barcelona, Área De Medi Ambient. Barcelona. Nerbonne, B.A., Vondracek, B., 2001. Effects of Local Land Use on Physical Habitat, Benthic Macroinvertebrates, and Fish in the Whitewater River, Minnesota, USA. Environ. Manage. 28, 87–99. doi:10.1007/s002670010209 Omernik, J.M., Abernathy, A.R., Male, L.M., 1981. Stream nutrient levels and proximity of agricultural and forest land to streams: Some relationships. J. Soil Water Conserv. 36, 227 –231. Osborne, L., Wiley, M., 1988. Empirical relationships between land use/cover and stream water quality in an agricultural watershed. J. Environ. Manage. 26, 9– 27. Parikh, A., Shah, V., Madamwar, D., 2006. Cyanobacterial Flora from Polluted Industrial Effluents. Environ. Monit. Assess. 116, 91–102. doi:10.1007/s10661-006-7229-x Potts, M., 1979. Nitrogen fixation (acetylene reduction) associated with communities of heterocystous and non-heterocystous blue-green algae in mangrove forests of Sinai. Oecologia 39, 359–373. doi:10.1007/BF00345445 Quinn, J.M., 1991. Guidelines for the Control of Undesirable Biological Growths in Water (Consultancy Report No. 6213/2). Water Quality Centre, Hamilton, New Zealand. Richards, C., Host, G., 1994. Examining Land Use Influences on Stream Habitats and Macroinvertebrates: A GIS Approach. J. Am. Water Resour. Assoc. 30, 729– 738. doi:10.1111/j.1752-1688.1994.tb03325.x Robson, B.J. 2000. Role of residual biofilm in the recolonization of rocky intermittent streams by benthic algae. Mar Freshw Res 51, 725-732. Robson, B.J., Matthews, T.G., Lind P.R., Thomas, N.A. 2008. Pathways for algal recolonization in seasonally-flowing streams. Freshwat. Biol. 53:2385-2401. Romani, A.M., Sabater, S. 1997. Metabolism recovery of a stromatolitic biofilm after drought in a Mediterranean stream. Arch. Hydrobiol. 140, 261-271.Roth, N.E., Allan, J.D., Erikson, D.L., 1996. Landscape influences on stream biotic integrity assessed at multiple spatial scales. Landsc. Ecol. 11, 141–156. Rott, E., Schneider, S.C., 2014. A comparison of ecological optima of soft-bodied benthic algae in Norwegian and Austrian rivers and consequences for river monitoring in Europe. Sci. Total Environ. 475, 180–186. doi:10.1016/j.scitotenv.2013.08.050

122 Sabater, S., Aboal, M., Cambra, J., 1989. New Rhodophyceae records for the NE and SE spanish continenatl waters. Limnetica 5, 93–100. Schiller, D., D. Von, Martí, E., Riera, J.L., Ribot, M., Marks, J.C., Sabater, F., 2008. Influence of land use on stream ecosystem function in a Mediterranean catchment. Freshw. Biol. 53, 2600–2612. doi:10.1111/j.1365- 2427.2008.02059.x Schneider, S., Melzer, A., 2003. The Trophic Index of Macrophytes (TIM) – a New Tool for Indicating the Trophic State of Running Waters. Int. Rev. Hydrobiol. 88, 49–67. doi:10.1002/iroh.200390005 Smith, V.H., 2003. Eutrophication of freshwater and coastal marine ecosystems a global problem. Environ. Sci. Pollut. Res. 10, 126–139. doi:10.1065/espr2002.12.142 Smith, V.H., Tilman, G.D., Nekola, J.C., 1999. Eutrophication: impacts of excess nutrient inputs on freshwater, marine, and terrestrial ecosystems. Environ. Pollut. 100, 179–196. doi:doi: DOI: 10.1016/S0269-7491(99)00091-3 Stancheva, R., Sheath, R.G., Read, B.A., McArthur, K.D., Schroepfer, C., Kociolek, J.P., Fetscher, A.E., 2013. Nitrogen-fixing cyanobacteria (free-living and diatom endosymbionts): their use in southern California stream bioassessment. Hydrobiologia 720, 111–127. doi:10.1007/s10750-013-1630-6 Suárez, M. L., A. Mellado, M. M. Sánchez-Montoya y M. R. Vidal-Abarca., 2005. Propuesta de un índice de macrófitos (IM) para evaluar la calidad ecológica de los ríos de la cuenca del Segura. Limnetica, 24 (3-4): 305-318. Thacker, R.W., Paul, V.J., 2001. Are benthic cyanobacteria indicators of nutrient enrichment? Relationships between cyanobacterial abundance and environmental factors on the reef flats of guam. Bull. Mar. Sci. 69, 497–508. Townsend, C., Arbuckle, C., Crowl, T., Scarsbrook, M., 1997. The relationship between land use and physicochemistry, food resources and macroinvertebrate communities in tributaries of the Taieri River, New Zealand: a hierarchically scaled approach. Freshw. Biol. 37, 177–191. doi:10.1046/j.1365- 2427.1997.00151.x Tran, C.P., Bode, R.W., Smith, A.J., Kleppel, G.S., 2010. Land-use proximity as a basis for assessing stream water quality in New York State (USA). Ecol. Indic. 10, 727–733. doi:10.1016/j.ecolind.2009.12.002

123

124 AGRADECIMIENTOS

Durante todos estos años, ha sido mucha la gente que me ha tendido la mano y ha contribuido, de una forma u otra, a que hoy esté escribiendo las últimas líneas de esta Tesis. Os quiero dar las gracias a todos por haber formado parte de esta etapa de mi vida, espero no dejarme a nadie en el tintero.

La primera persona que me viene a la mente eres tú, José Luis. Muchas gracias por todo esto que hoy queda materializado en papel, por haber hecho siempre todo lo posible y más para ayudarme, por contagiarme de ilusión y esperanza siempre que lo he necesitado, por tu comprensión… pero sobre todo, por tu generosidad. Has compartido conmigo tu tiempo, conversaciones geniales, recursos de todo tipo, sabiduría y ¡hasta alguna lección de conducir!. Me quedo con muchas cosas de ti que me acompañaran siempre. Que conste que esto no es una despedida, ¡aún te debo unos cuántos mapas!.

Gracias a mis padres. Aunque se me encoja un poco el corazón pensando que no leeréis estas líneas, me consuela pensar que de alguna forma os llega el mensaje. Gracias por absolutamente todo. Me enseñasteis lo más importante en esta vida: que un hogar no tiene nada que ver con cemento ni ladrillos, que tiene que ver con el amor, y el amor es eterno. Y precisamente por eso, gracias también a mis hermanos, Alberto y Ricardo, porque siempre y en cualquier parte del mundo, estando con vosotros, estaré en casa.

Jose, gracias por recordarme que la vida es muy sencilla siempre que se me olvida. Por compartir conmigo miles de pequeños momentos que me llenan el corazón y por hacerme reír. Me alegro de no haber hecho caso a aquello de ‘no te eches un novio murciano que son muy mentirosos y muy negociantes’ jejeje. Y, cómo no, gracias al pequeño Mojo que me regala el sol, el aire y la lluvia sacándome a pasear todos los días, y que es capaz de hacer des parecer un mal día haciéndome sentir la persona más importante del mundo. Todoa es mejor con vosotros dos.

Gran parte del impulso necesario para llegar hasta aquí me lo ha dado mi familia pricense al completo, que es un tesoro. Sobretodo tú Tita, que eres experta en

125 querer y cuidar a los que te rodean. Me alegro mucho de poder decirte que esto ya está.

Gracias a mis amigos, que son geniales… En especial a Myriam, Lourdes y Milla, porque sois personas excepcionales (literalmente, dudo mucho que abunde la gente como vosotros). Sois la mejor compañía que una podría tener en esta vida y es un privilegio poder contar con vosotros siempre. Cuanto más os conozco, más os admiro y os quiero. Y a Estela, qué buenas esas tardes de cañas y de buen cine, el mejor aderezo para estos años de trabajo.

Al grupo de los miércoles, gracias por haber compartido tanto con mi madre y con nosotros. Habéis llenado su memoria y la nuestra de muy buenos recuerdos, de risas, viajes y un sinfín de anécdotas. Gracias por haber sido nuestros ángeles de la guarda, sin vosotras todo hubiera sido mucho más difícil.

Gracias a todas las personas que formaron parte del Área de Limnología Aplicada e Hidrobiología antes de mi llegada y cuyo trabajo y esfuerzo forma parte de esta memoria.

Al grupo de Ecología Acuática de la Universidad de Murcia, mil gracias por acogerme como una más en diversas ocasiones académico-sociales, espero poder veros en muchas otras… me suena que algo se cuece en Villalgordo.

A Pedro Tomás, gracias por invitarme a compartir con vosotros aquellos días de muestreo en Huesca y ¡enhorabuena por tu Tesis!.

Por último, gracias a David Sánchez y Daniel Moya por su amabilidad revisando este trabajo.

126