Anna Sieczko - Interactions between dissolved organic matter and bacteria in a river floodplain system

(Ph.D. thesis, supervisor: Peter Peduzzi)

Within my study I focus on the relationship between quantity and quality of DOM, originating from diverse sources, and the involved bacterial responses.

The dissolved organic matter (DOM) is probably the most dominant form of organic material in aquatic ecosystems. Bacterial utilization of this material can be highly variable depending on its quality and may act as an important control mechanism for bacterioplankton growth. The sources and properties of DOM are considered as keys that control microbial activity and processing. There is also evidence that the hydrological connectivity influences bacterial biomass and production.

The interactions outlined above are potentially of great importance for the global carbon budget, but unfortunately poorly understood, especially in floodplain areas. The high heterogeneity of such systems generates many difficulties in investigating the relationships between the reactivity and origin of DOM and bacteria. Naturally occurring river-floodplain systems can serve as sinks, sources or transformers of organic material, inorganic nutrients, and biota. Although the autochthonous, more labile carbon is preferentially consumed by bacterioplankton, terrigenous DOM, considered to be mostly recalcitrant, has been recently recognized to be more bioavailable for the uptake by heterotrophic prokaryotes than previously thought.

In order to better understand the issues outlined above, I am going to test the following hypotheses:

1. Retention zones in the main channel exhibit conditions of higher diversity of DOM due to mixing of long distance transported DOC, autochthonously produced organic matter and local input of terrestrial DOM. The heterogeneity of sources of carbon significantly promotes bacterial production and creates zones of enhanced microbial activity.

2. Freshly introduced terrigenous DOM during seasonal floods, although more refractory also may stimulate microbial processes, due to changed physical and biological settings of backwaters. In “active” river-floodplain systems there is opportunity for significant utilization also of allochthonous carbon introduced from flood pulses.

3. Retention zones provide additional opportunities to augment the metabolic performance of fluvial ecosystems. Retention time and the degree of connectivity and water exchange is an important driving factor for C-remineralization.

This study is expected to provide information on the influence of hydrology on the quantity and quality of DOM in a river floodplain system and on the significance of variable bacterial response in relation to changing DOM quality. I am especially interested in the importance and bioavailability of allochthonous, recalcitrant DOM for bacterial utilization.

Charlotte Wöber - Species delineation and autecology of Spirogyra (Link 1820)

(Ph.D. thesis, supervisor: Michael Schagerl)

The object of my investigation is the green algae genus Spirogyra (Link 1820). Although this algae is very common in fresh water habitats and often found in large quantities, little is known about it’s autecology. Species determination is especially difficult, because all stages of the life-cycle are necessary for correct determination – especially the hypnozygotes (which are products of the sexual reproduction). Species delineation is also difficult because the traditional species concepts don’t fit this genus (see project description).

As a first step, I collected algae samples from 137 field sites together with my colleagues from the Spirogyra team. By analyzing the chemical, physical and environmental parameters of the sites we want to get insight in the autecology and distribution of Spirogyra in Middle Europe. The collected strains were isolated to unialgal cultures, photographed and kept for further investigations.
My investigation also involves the induction of sexual reproduction, so called conjugation experiments. With these experiments we want to find a “key” to induce conjugation and make species found in vegetative form determinable.
The third big part is the investigation of the genetics – especially the primary and secondary structure of the ITS. We want to test the species concepts commonly used and get further information on the relatedness of the Spirogyra species.

Christina Fasching - Dissolved organic carbon (DOC) biogeochemistry in fluvial networks

(Ph.D. thesis, supervisor: Tom Battin, mentor: Gabriel Singer)

Inland waters are active components of the global carbon cycle for they do not only transport, but mineralize and store a large amount of carbon, yet carbon fluxes in these systems have not received much attention in the global carbon budget. My PhD thesis focuses on the biogeochemistry of dissolved organic carbon (DOC). I will investigate the chemical diversity and bioreactivity of DOC, relate it to environmental and microbial parameters in order to better understand how fluvial networks in alpine regions function. The thesis is funded by the START-project ARCANET (“Architecture of carbon fluxes in fluvial networks”).

Karoline Wagner - Stream biofilms: a prime site for priming

(Ph.D. thesis, supervisor: Tom Battin, mentor: Mia Bengtsson)

Mechanisms of priming and its implications for global carbon cycling is increasingly studied and understood in soils but not in aquatic ecosystems. Priming describes the increased mineralization of terrestrial recalcitrant organic carbon in the presence of labile organic carbon. Biofilms are supposed to be hotspots for priming because of the close proximity between algae and bacteria. The objective of my PhD thesis is to study the mechanism of priming in benthic biofilms in streams. A combined genomic, transcriptomic and proteomic approach will lead to the identification of key players and possible mechanisms involved, including shifts in community structure and enzymatic activities. The thesis is funded by the FWF “Stream biofilms: a prime site for priming”.

Linda Wilhelm - Linkage of microbial diversity and function in fluvial networks

(Ph.D. thesis, supervisor: Tom Battin, mentor: Katharina Besemer)

Fluvial networks, including streams and rivers, transport solutes and microorganisms from land to the ocean. On this way, carbon is transformed and metabolized by microbial communities, primarily living in biofilms or suspended aggregates.  My PhD thesis focuses on the questions, whether microbial communities and their metabolic capacities acclimate to their geophysical environment, and on how microbial community structure and diversity change in fluvial networks. The thesis is funded by the START-project ARCANET “Architecture of carbon fluxes in fluvial networks”.

Mary Kaggwa – Ecophysiological studies on Arthrospira fusiformis and Anabaenopsis abijatae

(Ph.D. thesis, supervisor: Michael Schagerl)

I am working on some interesting cyanobacteria namely Arthrospira fusiformis and Anabaenopsis abijatae. These cyanobacteria are filamentous and grow extensively in some of the alkaline-saline waters, such as the Rift valley lakes, having a high pH of up to 11 and often form thick algal mats. These lakes are inhabited by huge flocks of the Lesser flamingos, Phoenicanaias minor which feed mainly on this cyanobacteria. Previously, A. fusiformis has been regarded to be the predominant cyanobacteria species in these lakes. However, changing dominances have been observed between A. fusiformis and A. abijatae and the factors that trigger these changes are not yet known. Additionally, in the past massive deaths of the Lesser flamingos have been observed to occur of upto about 20,000 in a short periods. Some explanations have been put forward in this regard though still not yet conclusive. A. fusiformis is commonly regarded as non-toxic though some recent investigations have indicated a possible toxicity of some of its strains hence being a presumed threat to the flamingos. A. abijatae too, has been cited as another possible source of toxins though no cyanobacterial toxins have been detected so far from this taxon. My research work therefore, involves carrying out an in-depth study using culture experiments as well as in-situ analyses, of the ecophysiology of both A. fusiformis and A.abijatae strains which will be obtained from various saline Rift valley lakes in Kenya, Tanzania and Ethiopia. It is expected that the outcome of this research will expound on the scientific knowledge of the overall prevalence of these two taxa in these soda lakes. It will also provide some useful information that will contribute towards explaining the cyanobacteria occurrences in relation to the flamingo population dynamics.

Martin Gruber – Factors controlling abundance of Arthrospira fusiformis

(Ph.D. thesis, supervisor: Michael Schagerl)

 
The Great Rift Valley offers a number of endorheic saline-alkaline lakes representing some of the most alkaline naturally occurring environments on earth. They feature an enormous phytoplankton biomass dominated by the cyanobacterium Arthrospira fusiformis, which is the basis of life in these lakes. However, phytoplankton standing crop has been observed to be extremely variable, with no simple explanation available. I will focus on this fundamental question by investigating two types of Kenyan lakes: the very shallow Lake Nakuru, which is occupied by the planktivorous fish Oreochromis and mixes daily and Lake Bogoria, which is fish-free and stratified over longer periods.

Nadja Straubinger – Fish-cage farming in Kenyan reserviors – interactions of the pelagic food web

(Ph.D. thesis, supervisor: Michael Schagerl)

My investigations are running within the framework of BOMOSA cage fish-farming project (http://bomosa.oeaw.ac.at) in Kenya. There are 9 sites in 3 different regions (Machakos, Mount Kenya and Lake basin region). Sampling started in May 2007 to get the reference status before the fish cages were employed. One target is to detect the effects of the fish farming for the phytoplankton community and bacterial numbers. A second target is to find possible risks based on facultative toxic cyanobacteria. Therefore I analyse samples for the toxin producing Genecluster (PCR) and Microcystins using HPLC.

Simon Vitecek - Phylogenetic analysis of the Drusinae subfamily (Insecta, Trichoptera)

 (Ph.D. thesis, supervisor: Johann Waringer)

This PhD project will be conducted in the framework of a FWF project (P23687-B17) on taxonomy, phylogeny and sensitivity to climate change of the Limnephilidae subfamily Drusinae granted to Univ. Prof. Dr. Waringer (as principal investigator). Within the framework of this project, phylogeny and ecological metrics of Drusinae will be assessed in an international cooperation. As part of the FWF project, the proposed PhD project will deal with taxonomy and phylogeny of the Drusinae subfamily.

In a first step, in-depth morphological analysis of adult and larval specimens will be conducted in order to a) perform phylogenetic analysis based on morphological characters; b) construct a reliable identification key. Additionally, a molecular phylogeny of the taxon will be constructed, and combined with the morphological data matrix. The expected outcome will provide a robust basis for further analysis of the evolutionary history of this subfamily.

Within the framework of the FWF project, the proposed PhD project will focus on the following research aims: 1) clarify the phylogeny of recent Drusinae with emphasis on certain evolutionary traits (e.g. feeding ecology); 2) provide a comprehensive key for all Drusinae species, in order to enhance ecological analyses of water quality using macro-invertebrates.

Specimens will be provided either from collections or sampled in the framework of the FWF project. Adults will be identified to species level based on Malicky (2004). Unknown larvae collected at the same sites will be affiliated to adult species primarily by mitochondrial cytochrome oxidase I (mtCOI) and large subunit rDNA D2 (LSU rDNA D2) sequence analyses. Analysis of morphological characters of larvae and adults will be performed at the University of Vienna under supervsion of Ao Univ. Prof.J. Waringer. To construct a molecular phylogeny, analysis of five genes (mtCOI, mtLSU, nuclear wingless (nuWG), elongation factor 1a (EF-1a) and LSU rDNA D2) will be performed at Biodiversity and Climate Research Centre (BiK-F), Frankfurt am Main, Germany, under supervision of Dr. Steffen Pauls. The publication effort will comprise a minimum of three papers: 1) aspects of the morphological character matrix, trait evolution and cladistics; 2) a comprehensive key for all Drusinae species; and 3) a synthesis paper on Drusinae phylogeny combining morphological and molecular genetic data.

Anita Hutmann -  Verteilungsmuster und ökologische Ansprüche von Großkrebsen im Einzugsgebiet des Kamp (Niederösterreich)

(M.Sc. thesis, supervisor: Johann Waringer, mentor: Manfred Pöckl)

Im Rahmen einer Kartierung im gesamten Einzugsgebiet des Kamp sollen die Flusskrebsvorkommen erfasst und die ökologischen Schlüsselfaktoren für ein Erreichen hoher Abundanzen erfasst werden. Insbesondere soll das Vorkommen der autochthonen Arten im Vergleich zu Ausbreitungstendenzen eingebürgerter Arten dokumentiert werden.

Bernhard Zens - Linking fish behaviour and hydraulics: rheoreaction and movement patterns in fish larvae

(M.Sc. thesis, supervisor: Hubert Keckeis)

Movement patterns of different developmental stages of fish larvae (Chondrostoma nasus, Cyprinidae) were studied using an artificial racetrack flume. In order to resemble conditions of the nase-carp`s natural habitat, our experimental mesocosm offered characteristic features of river-bed morphology as well as a velocity gradient. Experiments were carried out at three different velocity scenarios (representing sub-, near- and supercritical flow conditions with respect to swimming abilities) and were conducted by daylight and during night. Based on video recordings, fish larvae`s travel paths were analysed by means of their exact [±5 cm] position, their swimming speed as well as their orientation in the water column. Information on hydraulic parameters (i.e. flow velocity in different layers) at any position was obtained from a fine-scaled 3D numerical model. Focusing on fish movement in relation to flow velocity and the orientation of the fish against the current vector enables the differentiation between active, active-passive and passive modes of downstream migration (drift). Rates of downstream movement differed significantly between flow scenarios and developmental stages. These results provide basic information for a mark-recapture study aiming to model the dispersion of individual fish larvae in the River Danube.

Carina Zittra - Ökologie der Culicidae (Insecta: Diptera) im Nationalpark Donauauen

(M.Sc. thesis, supervisor: Johann Waringer)

Die Diplomarbeit untersucht das Arteninventar und die Phänologie der Culiciden-Arten im Nationalpark Donauauen. Weitere Aspekte der Arbeit umfassen die Charakterisierung der Standortsbedingungen der einzelnen Arten, insbesondere die Brutgewässerbeschaffenheit, die Dynamik der Wasserführung und den Einfluß des natürlichen Prädatorenvorkommens auf den Bruterfolg.

David Ramler - Seasonal course of larval drift of selected native and invasive benthic fish species along two different shore types in the main channel of a large river (Danube, Austria).

(M.Sc. thesis, supervisors: Hubert Keckeis & Harald Ahnelt)

The aim of my study is to compare the larval drift of an artificial and a restored shoreline (Witzelsdorf, Donau-Auen National Park), by using three selected fish families. The Percidae and Cottidae which can be found in the Danube are native and include several endemic and endangered species, while the Gobiidae are represented by one native and three invasive species. Drift samples, collected in short intervals (daily – weekly) during May and June in 2011, will be analysed. At each shore type, samples were taken at different stations along the site, and at each station in different distances from the shore (in-, mid, and offshore) from dusk until midnight, when the highest drift intensities occur on a diurnal basis. In a first step the fish larvae are identified to genus and (if possible) to species levels. I then assess family and species specific seasonal aspects, like first occurrence, period, pattern (permanent, irregular), and peak occurence (unimodal, multimodal). The spatial distribution of fish species within each shore will also be analyzed in both transversal and longitudinal directions, and drift densities of the two shorelines are compared and discussed with regard to structural and hydraulic conditions of the shores.

Franz Loisl - Comparison of the fish community of three Danube segments on the basis of different fish sampling methods

(M.Sc. thesis, supervisor: Hubert Keckeis)

Fish assemblages, as well as single fish species are important indicators for the assessment of type and status of large water bodies. Regarding the growing number of endangered, invasive and exotic species, it is important to gain as much information as possible concerning the community composition, species diversity as well as the spatial and temporal distribution of single species. The free-flowing main channels of large rivers are important habitats for many riverine species; many of those are classified endangered. It is a difficult task to get data reflecting the ”true picture” of communities in large rivers. By combining different sampling methods, it should be possible to approach the natural picture of the fish assemblage.

This kind of information is very important for conservation and restoration projects, like the “River Engineering Project” within the National Park “Donauauen”, to investigate the effects of restructuring measures on the fish communities. Increasing quality, availability and structural diversity of habitats is an essential aspect for nature conservation and restoration, not only for fish, but for the whole aquatic environment.

The data I am referring to has been collected within a period of three months (from June to August) in 2007. Three different fish sampling methods have been applied to quantify the fish assemblage of three reaches of the Danube main channel east of Vienna.

The questions of my study:

(i) Do the three sample sites differ according to their species number, fish species composition and size structure?

(ii) Are there differences between mesohabitats irrespective of the river segment?

(iii) Does the fish species composition change during the sampling period within and between the segments?

(iv) Is there a possibility to integrate results from different fish sampling methods?

Julia Wukovits - Carbon assimilation and carbonic anhydrase activity in desmids

(M.Sc. thesis, supervisor: Michael Schagerl)

The availability of the different carbon sources in water is highly related to pH. Thus, at pH values 0-6, free CO2 is the predominant carbon species, at pH 6-10 bicarbonate is the main carbon source and beyond pH 10 CO3²- ions prevail.
As CO2 is the only DIC species, which reacts with RuBisCO for carbon fixation, eukaryotic algae have developed different strategies to use bicarbonate dominating in many surface waters. One mechanism is to produce external CA which catalyses the dehydration of HCO3- to CO2. Another one is to excrete acid (H+) across the cell membranes (Graham & Wilcox 2000). Desmids occurring at acid conditions are expected to produce no external CA. The same holds true for Oocardium stratum, which is found in CO2 super saturated tuff springs.
My methods include pH-drift experiments (Allen & Spence 1980) for determining the affinity to CO2 and HCO3- of the algae and a potentiometric method (Wilbur & Anderson 1948) as well as a spectrophotometric method (Pocker & Stone 1967) to determine CA activity. I want to show if the species studied produce CA and if they are able to use HCO3- at low levels of CO2.

Katharina Stefan - Migration corridors and spawning activity of Amphibia at the Exelberg ponds

(M.Sc. thesis, supervisor: Andrea &Johann Waringer)

This thesis investigates the phenology, breeding success, larval development and migration corridors at 14 ponds in the Exelberg area in the Wienerwald. Based on the results of biometrical data, we expect insights into breeding biology of Bufo bufo, Rana dalmatina, Rana temporaria, Hyla arborea, Pelophylax ridibundus, Triturus carnifex, Ichthyosaura alpestris and Lissotriton vulgaris.

Lukas Kräftner - Amphibian colonisation dynamics of newly-created ponds in the Ybbs catchment

(M.Sc. thesis, supervisors: Andrea & Johann Waringer)

 In the framework of revitalization measurements at the Ybbs River several ponds were created. In this investigation the acceptance and colonization by Bombina variegata and other amphibians will be studied. The interaction with colonization patterns of invertebrate and vertebrate predators will be a further focus.

Martina Adlboller - Hydrological stochasticity and biodiversity of algae in biofilms.

(M.Sc. thesis, supervisor: Tom Battin, mentor: Iris Hödl)

Worldwide fluvial ecosystems effect climatic processes,  landscape, water support, …, and all organisms. Biofilms, algae and bacteria growing on several surfaces, play a major role in self-purification processes and primary production, which responds quantitatively to changes in light and further to variation in community composition and structure of benthic algae. Rivers in urban regions are regulated and so water bodies (groundwater, ...) are separated and influences on ecosystem services are rarely known. The ambition of this experiment was to receive an impression of what is influencing algae growth and composition in an artificial system.  In ecological terms the biodiversity should lead to a clearer understanding whether different light conditions or frequent discharge changes have a greater impact on algae. To determine this effect of variation the diversity within one experimental flume and between flumes will be measured.

Martin Soneira - Auswirkungen unterschiedlicher Beleuchtungstypen auf den Anflug von Köcherfliegen

(M.Sc. thesis, supervisor: Johann Waringer)

In der oberen Lobau soll die Auswirkung unterschiedlicher Beleuchtungstypen des öffentlichen Raumes (z.B. Quecksilber- und Natriumdampflampen, UV-Lampen etc.) auf den Anflug adulter Köcherfliegen getestet werden. Zusätzlich sollen Phänologie, Artenzusammensetzung und Ökologie der dominanten Arten untersucht werden.

Michaela Leuchtenmüller - Comparison of the aquatic macrophyte assemblages in two major Danube River branches: Maly Dunaj and Mosoni Duna

(M.Sc. thesis, supervisor: Georg Janauer)

My investigations are focused on the aquatic macrophyte assemblages of Maly Duna and Mosoni Duna, which are the side branches of the Danube River fringing the two islands of Zitny ostrov (SK) in the North and Szigetköz (HU) in the South. Three hypotheses form the background of this study: (i) What changes in macrophyte composition are detected on a temporal basis for each of these water bodies? (ii) How do the two river channel differ in species and abundance with respect to the aquatic environment? (iii) How are the two water bodies characterised in conservational context, especially regarding rare species and the danger of extinction under changed habitat properties?

Philipp Abel - Vergleich der Fischlarvendrift des Hauptarmes der Donau und der Schwechat

(M.Sc. thesis, supervisor: Hubert Keckeis, mentor: Aaron Lechner)

Der Hauptarm der Donau und die Schwechat unterscheiden sich hinsichtlich ihrer gewässermorphologischen und –typologischen Variablen (unterschiedliche Flussordnungen, Fließgeschwindigkeit, Wassertemperatur, etc.). Es wird die saisonale Fischlarvendrift an vier verschieden Probenpunkten, eine an der Schwechat, eine am linken Donauufer und zwei am rechten Donauufer simultan mit standardisierten Driftnetzen untersucht. Die Proben werden in Labor ausgewertet.  Es wird die Driftdichte gemessen, die Larven auf Art- bzw. mindestens Familienniveau, deren Entwicklungsstadium bestimmt und die Länge gemessen (Notochordlänge, Totallänge). Folgende Fragestellungen sollen untersucht werden:

Gibt es saisonale und quantitative Unterschiede bezüglich der Driftdichte an den verschiedenen Probenpunkten? Gibt es Fischfamilien bzw. –arten die bevorzugt in einem der Gewässer vorkommen? Unterscheidet sich die Artenzusammensetzung an der Schwechat von der Donau im saisonalen Verlauf und simultan zwischen den Probenpunkten? Gibt es Unterschiede im Entwicklungsstadium der Fischlarven und unterscheiden sie sich in der Länge?

Reinhard Krusch - Development and evaluation of daily increments in otoliths of Chondrostoma nasus during early ontogeny

(M.Sc. thesis, supervisor: Hubert Keckeis)

This study focuses on development and validation of daily increments in otoliths of embryonal and larval life stages of nase carp. Larvae used for the study were reared in through-flow tanks at two different temperatures (warm: avg. 16,7°C ±1,1; cold: avg. 11,4°C ±0,9). Samples were taken in daily intervals starting at fertilization of the eggs until 47 days post hatch for cold reared and 90 days for warm reared larvae, respectively. Individual size is measured and Sagittae and Lapilli otoliths are extracted, fixed in artificial resin on microscope slides and used for measurement of otolith size (diameter) and age determination (counts of rings) under a light microscope.

Main points of interest of this study are:

(i)   At which age and developmental stage are otoliths formed?

(ii)  Are the light-dark patterns in otoliths produced in a circadian rhythm?

(iii) How accurate is ring counting for age determination?

(iv) Are there specific structures, like broader or darker rings, identifiable in otoliths and are they related to specific developmental events or external influences?

(v)  Is there a relationship between otolith size and individual size?

(vi) To which extend is otolith size and ring formation temperature dependent?

With the results of this study I aim to produce a template, usable in the FWF project P22631-B17 “Modelling dispersal patterns of fish larvae in a large river”, for determining age of wild caught larvae, their range of spawning time as well as back-calculating growth of released and recaptured larvae.

Sabrina Hengsberger - Relating CO2-outgassing to DOM properties in a fluvial network

(M.Sc. thesis, supervisor: Tom Battin, mentor: Gabriel Singer)

Fluvial networks link landscapes, soils and groundwater with the atmosphere and the oceans. Statistics show a worldwide CO2 outgassing from streams of 1,2 Gt per year. This is controlled by microbial processes and influences global carbon fluxes. My thesis focuses on metabolism, CO2 outgassing and gas-exchange in a fluvial network of the upper river Ybbs in Lower Austria. The network is sampled at 120 sites ranging from first order streams to the fifth order river Ybbs at Göstling. I will investigate networke-wide patterns of partial CO2-pressure, dissolved organic carbon (DOC), optical parameters of dissolved organic matter (DOM) and bioavailability of carbon (BDOC).

Stefan Singer - Intercalibration of two sonar systems to increase the applicability of these systems for investigations of deep-pools in the Danube

(M.Sc. thesis, supervisor: Hubert Keckeis, mentor: Georg Rakowitz)

The ecology and the ecological function of deep pools for the fish-zoenosis of large rivers are fairly unknown. There is a great lack of knowledge especially in quantitative and qualitative information about the fish-community in these habitats. A reason for this deficit could be the fact that conventional fish-sampling methods are limited in their applicability to investigate these habitats. The use of hydroacoustic methods seems to be an appropriate method to investigate the fish-community of deep pools (Rakowitz et al., 2009; Rakowitz, 2009).

The aim of my diploma-thesis is to compare recordings from two different sonar-systems, a split beam echosounder (Simrad EK60) and a mulibeam echosounder (DIDSON), regarding their appropriateness for investigating the fish-community of deep pools. The dataset contains seasonal 24-hour recordings of two pools which differ in their hydrological conditions (overflow-pool, groyne-head-pool). For the analysis of the hydroacoustic data I use the post-processing software SONAR5-Pro developed by Balk & Lindem (http://www.fys.uio.no/~hbalk/).

I want to investigate the following scientific questions:

1) Are there differences in the pattern of the seasonal and diel abundance-, biomass- and size-distribution between the two different hydroacoustic systems (splitbeam, multibeam)?

2) What are the causes of possible differences in the estimates of the two different hydroacustic systems?

3)What are the possibilities and limits of the two systems with regard to investigate the in-stream habitat fish-community of large rivers?