Junk W.J. Wantzen K.M.
Max-Planck-Institute for Limnology, Working Group Tropical Ecology, P.O. Box
Plön, Germany E-mail: firstname.lastname@example.org
The flood pulse concept (FPC), published in 1989, was based on the scientific experience of the authors and published data worldwide. Since then, knowledge on floodplains has increased considerably, creating a large database for testing the predictions of the concept. The FPC has proved to be an integrative approach for studying highly diverse and complex ecological processes in river-floodplain systems; however, the concept has been modified, extended and restricted by several authors. Major advances have been achieved through detailed studies on the effects of hydrology and hydrochemistry, climate, paleoclimate, biogeography, biodiversity and landscape ecology and also through wetland restoration and sustainable management of floodplains in different latitudes and continents. Discussions on floodplain ecology and management are greatly influenced by data obtained on flow pulses and connectivity, the Riverine Productivity Model and the Multiple Use Concept. This paper summarizes the predictions of the FPC, evaluates their value in the light of recent data and new concepts and discusses further developments in floodplain theory.
Rivers and floodplain wetlands are among the most threatened ecosystems. For example, 77 percent of the water discharge of the 139 largest river systems in North America and Europe is affected by fragmentation of the river channels by dams and river regulation (Dynesius and Nilsson 1994). In recent reviews on wetlands, demographic trends, economic and political development, demand for hydroelectric energy and water, agriculture and animal ranching, eutrophication and pollution, fisheries, logging, recreation and eco-tourism and invasion by exotic species have been identified as the most important current determinants for the development of rivers and floodplain wetlands (Tockner and Stanford 2002; Junk 2002). The global water crisis and the threat to aquatic organisms, especially riverine biota (Dudgeon 2000; Pringle 2001), increase the necessity to develop models that serve both science and policy. The flood pulse concept (FPC) (Junk, Bayley and Sparks 1989) was primarily designed as a scientific concept, but it also outlined some strategies for use, recently specified in Junk et al. (2000). Here, the impact of advances in river ecology on this and other contemporary concepts is critically analyzed.
Limnologists classify inland waters as standing waters (lakes, ponds) and running waters (streams and rivers). Both system types receive allochthonous substances and produce autochthonous organic matter, both of which are metabolized and recycled. Standing waters, however, are closed systems that store inorganic and organic matter, circulate organic matter and dissolved inorganic substances in characteristic internal cycles in the lake basin and eventually deposit them in the sediments. These systems are characterized mainly by thermal and/or chemical stratification. Running waters are open systems that transport water and dissolved and suspended matter from the continents to the sea or to endorheic basins. This transport includes intermediate deposition and re-suspension of sediments in the river channel or in the connected floodplain, where production and degradation of organic matter also takes place.
These characteristics are reflected for lakes in the "Seentypenlehre" (Lake typology), elaborated by Thienemann and Naumann between 1915 and 1935 (e.g. Thienemann 1925; Naumann 1932) and for streams and rivers in articles by Illies (1961a) and Illies and Botosaneanu (1963) on the differentiation into different zones. These zones, described in these early studies as rhithron and potamon with epi-, metaand hypo-subzones, were mainly characterized by abiotic patterns (current, temperature) and by the occurrence of distinct animal and plant communities that depend on a given set of these abiotic patterns. For example, epirithral communities are those typical of glacier outflows and that depend on low temperatures, high oxygen concentrations and fast current. In latitudinal comparisons, Illies (1961b) found evidence for a generality of this zonation in mountain streams worldwide - high-elevation streams in the tropics have communities similar to those of low-latitude coldwater streams.
Later, Vannote et al. (1980) substituted this rather static view of river classification with the River Continuum Concept (RCC), which introduced a dynamic concept of continuously changing physical conditions and biological components along the river channel, especially regarding the allochthonous and autochthonous inputs and the processing of organic matter in the flowing water along the river continuum. The RCC predicts major changes in the load and quality of organic matter and the biota in the stream/river channel from the headwaters to the lower river courses. The high allochthonous input from riparian vegetation in the headwaters decreases with increasing channel width (increasing stream order). Autochthonous primary production in the headwaters is low because of shading by trees (P/R <1), increases in the middle reaches because of low water depth and high irradiation (P/R >1) and decreases again in the lower reaches because of high water depth and increased turbidity and turbulence (P/R <1). In contrast to the zonation concept, the RCC claims that occurring species are replaced continuously rather than in discrete stages. The percentage of shredders decreases and the number of collectors increases with increasing stream order because of decreasing input of coarse particulate organic material and an increasing amount of fine particles owing to the increasing level of processing. Headwater communities tend to optimize their use of allochthonous matter, whereas an organism living in the lower river reaches largely depend on the inefficiency of organisms living in the upper reaches to process organic material. The interplay of processing, storage and leakage is predicted to reduce the diversity of organic matter types and the maximization of energy utilization (i.e. adaptation to poorly degradable organic matter) along the river continuum.
The RCC further predicts that biodiversity of aquatic organisms is lower in the headwater regions and in the lower parts of the rivers and that highest diversity is found in the middle reaches of the streams, where the variability of temperature, riparian influence and flow are highest and allow numerous different taxa to find their thermal optima.
One of the major constraints of the RCC is that it was originally based mostly on results from northern, temperate, low-order streams with dense tree canopies and steep gradients that flowed towards more-or-less-regulated rivers in long-term-managed areas. The hydrology of small streams is strongly influenced by local rainfall and is rather erratic. Flooding of small streams occurs only for short periods and is often altered by management of rivers in intensively used areas. Therefore, flooding events and floodplains received little or no attention in the first version of the concept, but were considered later (Minshall 1985; Sedell, Richey and Swanson 1989).
Floodplains fall into the wetland category, which includes ecosystems at the interface of aquatic and terrestrial ecosystems and are therefore often called ecotones. However, large wetlands have to be considered as specific ecosystems with unique properties not adequately covered by present ecological paradigms and by limnology, estuarine ecology and terrestrial ecology (Mitsch and Gosselink 2000). Floodplains are areas that are periodically inundated by the lateral overflow of rivers or lakes and/or by rainfall or groundwater; the biota responds to the flooding by morphological, anatomical, physiological, phenological and/or ethological adaptations and characteristic community structures are formed (Junk et al. 1989).
Until the 1970s, floodplains were studied separately by different disciplines: limnologists studied floodplain lakes treating them as classical lakes, ecologists dealt with the terrestrial fauna and flora and hydrologists investigated water and sediment transport. An integrated approach was used by Welcomme (1979), who summarized data on floodplain fishery, limnology and hydrology and coupled fish production with the nutrient status of the parent rivers and the extent of flooding. The consequences of the fluctuating water level on fish have also been summarized by Lowe-McConnell (1975, 1987). Bayley (1980) pointed to limits in limnological theory with respect to fish production in river floodplains. Junk (1980) described the multiple land-water interactions of the Central Amazon River floodplain, analysed limnological concepts of rivers and lakes, pointed out a gap in limnological theory and described floodplains as specific ecosystems.
During the first Large River Symposium in Toronto in 1986, the discussion on the applicability of the RCC to large river-floodplain systems led to the formulation of the Flood Pulse Concept (FPC) (Junk et al. 1989). This concept focuses on the lateral exchange of water, nutrients and organisms between the river channel (or a lake) and the connected floodplain. It considers the importance of the hydrology and hydrochemistry of the parent river, but focuses on their impact on the organisms and the specific processes in the floodplain. Periodic inundation and drought (flood pulse) is the driving force in the river-floodplain system. The floodplain is considered as an integral part of the system that is periodically coupled and decoupled from the parent river by the aquatic/terrestrial transition zone (ATTZ). The flood pulse can be monomodal or polymodal, predictable or unpredictable and with a high or low amplitude. Predictable pulsing favors the adaptation of organisms and increases primary production and efficiency of nutrient use.
The FPC predicts that the nutrient status of the floodplain depends on the amount and quality of dissolved and suspended solids of the parent river; however, it includes the premise that internal processes of the floodplain and nutrient transfer mechanisms between the terrestrial and the aquatic phase strongly influence nutrient cycles, primary and secondary production and decomposition. At the same time, flooding is considered as a disturbance factor that leads to a regular setback of community development and maintains the system in an immature, but highly productive stage.
Another tenet of the FPC is that in the river-floodplain system, a large part of the primary and secondary production occurs in the floodplain, whereas the river is mainly the transport vehicle for water and dissolved and suspended matter. The river is also the refuge for aquatic organisms during low-water periods and serves as a route for active and passive dispersal. The "highway analogy" describing the river channel as a transport and migration corridor was used to visualize the different functions of the main river channel and its floodplain (Junk et al. 1989). The FPC was based on the personal experiences of the authors on the Amazon and Mississippi rivers but also on a vast literature about other river systems. Therefore, the concept was not restricted to large tropical rivers, as is sometimes cited (e.g. Benke et al. 2000), but was conceived as a general concept for large river-floodplain systems.
Mostly limnologists, ichthyologists and fisheries biologists study the ecology of floodplains and their organisms. They test the predictions of the FPC only for the aquatic phases of the system. However, it has to be stressed that the FPC covers the river-floodplain- system during the entire year and that its predictions are also valid for the terrestrial phases that are an integral part of the river-floodplain system.
FURTHER CONCEPTS IN RUNNING-WATER ECOLOGY
Various conceptual approaches independent of (or complementary to) the RCC and FPC were developed in the 1980s and 1990s on lotic ecosystem structure and functioning. Many early seminal papers dealt with the distribution of organisms within the lotic systems. A key aspect was the description of hyporheic zones in which important ecosystem processes occur. Early work by Schwoerbel (1961) on the distribution of benthic and stygal fauna in bed sediments was extended by the description of aquatic organisms far from the river channel area (Stanford and Ward 1988) and factors contributing to vertical distribution of the organisms (Bretschko and Leichtfried 1988) and by the understanding of organic matter dynamics in this zone (e.g. Williams 1989; Triska, Duff and Avanzino 1993). Ward (1989) included the function and occurrence of hyporheic zones in general stream theory by describing streams as four-dimensional systems.
The importance of stream hydraulics (Statzner and Higler 1986) and of disturbance (Resh et al. 1988) for the distribution for benthic organisms has shown that variation in water flow caused by climate and geomorphology can set limits to the generalizations of the RCC since flow conditions typical of upstream and downstream areas can change several times along the river course ("discontinuum", Poole 2002). Consequently, the template provided by the habitat conditions (Southwood 1977) and its alignment with species traits (Townsend and Hildrew 1994; Resh et al. 1994) might be more important for the occurrence of a species than the position of the given site along the continuum. Since geomorphology is subject to non-continuous local variations, the distribution of stream habitats appears as a mosaic (Pringle et al. 1988) of hierarchically ordinated and dynamic patches (Townsend 1989; Poole 2002). Various hierarchical concepts have been developed for riverine landscape patterns and their scale-dependent processes (e.g. Frissell et al. 1986; Townsend 1996; Petts and Amoros 1996; Poff 1997; Montgomery 1999; Ward, Malard and Tockner. 2001; Poole 2002).
Further conceptual approaches have dealt with the production and processing of organic matter, such as the nutrient spiraling concept (Elwood et al. 1983; Pinay et al. 1999), or with human impacts, e.g. the interruption of natural flow pathways by dams [the Serial Discontinuity Concept (Ward and Stanford 1983a; Ward and Stanford 1995)]. As an alternative to the RCC, Montgomery (1999) proposed a multi-scale hypothesis in which spatial variability in geomorphic processes governs temporal patterns of disturbances that influence ecosystem structure and dynamics (Concept of Process Domains). Channel networks can be divided into discrete regions in which community structure and dynamics respond to distinctly different disturbance regimes.
EXTENSIONS OF THE FPC
HYDROLOGY AND FLOW CHARACTERISTICS
The FPC was developed based upon data and long-term observations of neotropical (Amazon) and temperate zone (Mississippi) rivers. It provided a general outline and strengthened the premise that rivers and their floodplains have to be considered as one unit and therefore cannot be treated separately in ecological studies. The FPC has stimulated various studies on river-floodplain ecosystems in, e.g. Lower Rhine floodplain lakes (van den Brink et al. 1994), Missouri floodplain lakes (Knowlton and Jones 1997), the Danube (Tockner, Malard and Ward. 2000), the Murray-Darling (Humphries, King and Koehn 1999) and the Mississippi River (Sparks, Bayley and Kohlert 1990). Various studies in which the FPC has been applied and its tenets tested have lead to proposals for supplementation to the original conceptual framework.
Increasing knowledge on the hydrological characteristics of rivers has contributed much towards the understanding of their ecological processes. The hydrographs of individual rivers are influenced by a series of partly interacting factors, such as climate, gradients, landscape morphology, floodplain buffering and human impacts, which together cause very complex patterns. Several authors have identified different measures for identifying the hydrological variability of rivers and have provided tools for classifying rivers according to their hydrological signature (Richter, Baumgartener, Powell et al. 1996; Puckridge et al. 1998). Providing more detail on the type of flow variation, Puckridge et al. (1998) have stressed the general importance of water level variations even below the bankfull stage (flow pulses), which might have significant influence on the habitat size and characteristics. Irregular flood events, especially in streams (e.g. Winterbourn, Rounik and Cowie 1981) and arid zone rivers, have selected for resilient strategies of organisms to survive these events (Poff and Ward 1989; Lytle 2001) rather than to adaptations to profit from them (Junk et al. 1989). However, this view has recently been modified because in seasonal climates, the period of flash flood events can be predicted and because flash-flood events mobilize organic matter resources from stream wetlands (Wantzen and Junk 2000).
Tockner et al. (2000) extended the FPC by considering that flooding resets temperature diversity in isolated aquatic floodplain habitats. Thus, aquatic habitats within floodplains might have a much broader temperature range than the river itself, especially in rivers with wide and diversified beds, such as the alpine Fiume Tagliamento, where flow pulses occur frequently.
Unpredictable flooding and decoupling of the flood pulse from the temperature pulse leads to low temperatures during floods and high temperatures that trigger spawning of some fish species during low water level in some parts of the Murray Darling River system in Australia (Humphries et al. 1999, Low Flow Recruitment Hypothesis).
The fact that water level changes influence riverine systems four-dimensionally in space and time (Ward 1989) is important. Rising water levels not only increase the wetted surface of the channel and eventually of the floodplain, but at the same time influence the exchange between groundwater and surface water either by allowing an up-welling of groundwater or by forcing a down-welling of the surface water into the aquifer vertically and laterally. The hyporheic zone serves as an interface between groundwater and surface water (Schwoerbel 1961). Similarly, floodplains act as interfaces for the interchange between the river mainstreams and their tributaries or surface runoff from rainwater. The flow direction of the interfaces is influenced by the fourth dimension, time, such that the recent and the past hydrological situations become decisive: elevated water levels can cause blocking or even backflow of the tributaries and groundwater outflows. In the northern Pantanal wetland, frequent changes in the flow direction occur in floodplain channels that connect water bodies that receive rain and river water, depending on the respective water level (Wantzen and Da Silva, unpublished data). In the southern areas of the Pantanal, high levels of river water block the tributaries after the rainy season; therefore, a large part of the inundation occurs after cessation of the rainfall (Hamilton, Sippel and Melack 1996).
ALLOCHTHONOUS AND AUTOCHTHONOUS PRODUCTION AND NUTRIENTS
The FPC has focused on the productivity within the floodplain areas, in contrast to the RCC, which has focused on the import of more-or-less processed allochthonous matter from the upriver sections. Both concepts have been criticized by Thorp and Delong (1994) in the Riverine Productivity Model (RPM), which predicts that autochthonous production in the river channel and allochthonous inputs in the lower reaches provide a substantial portion of the organic carbon used by river animals. While analyzing the potential influence of the floodplain on the carbon budget in a channel site and a floodplain region of the Ohio River, Thorp et al. (1998) did not find significant differences in the isotopic C and N signatures, which indicated a low floodplain contribution during a short-termed, unpredictable flood event at low water temperatures in wintertime. This, however, does not necessarily contradict the predictions of the FPC.
Depending on temperature, light, nutrients and substrate conditions, river channels can show a considerable autochthonous primary production, which fuels the riverine food web as shown for Rhine River (Friedrich and Mueller 1984). Especially in those rivers where these conditions are beneficial for algal growth and where conditions for production in the floodplain are restricted by turbidity, timing of inundation during the winter and river regulation (e.g. of the Ohio River), the in-channel primary production can be substantially higher than floodplain production. River channels can support diverse and productive fish communities under these conditions (Galat and Zweimuller 2001; Dettmers et al. 2001); however, this is not the case in very large and turbid lowland rivers with a sandy, permanently moving bed load.
In those rivers that show a predictable, sufficiently long and timely inundation, such as the Mississippi River, the exploitation of floodplain resources provides a "flood pulse advantage" Gutreuter et al. (1999) for floodplain fishes compared to pure riverine species (Bayley 1991). Stable isotope ratios of many floodplain fish species of the Pantanal wetland show seasonal variation, which indicates a large variability in carbon sources and the trophic level between seasons (Wantzen et al. 2002) (Figure. 1).
Figure 1. Seasonal isotopic shifts in small omnivore floodplain fish species in the Pantanal wetland, Brazil. Filled circles: rainy season values; open triangles: dry season values. There is a general increase in d15N values from the wet to the dry season, which indicates more omnivorous feeding when the wetland is flooded and elevated carnivory during the dry season when the lake becomes confined to its basin (Modified after Wantzen et al. 2002, with permission).
In regulated rivers, where connected lakes represent remains of a floodplain, e.g. in the lower Rhine River in the Netherlands, the seepage and overflow of nutrient-rich river water determines the productivity and composition of the phytoplankton such that lakes with higher connectivity have a higher productivity (van den Brink et al. 1993). In most river-floodplain systems, primary production in the floodplain is much higher than in the channel, (e.g. Australia, review by Robertson et al. 1999). We agree with Dettmers et al. (2001) that organic matter input and production derive from the upstream sites (RCC), from the floodplains (FPC) and from the river channel (RPM). The relative contribution of these sites to the organic matter budget of a river depends on the production and transport conditions in these three units (Figure 2).
Figure 2. Schematic interplay of variable carbon sources in different segments of a river. Owing to the topographical variation in the landscape, the sequence of the segments can vary along the river course. Top: in catchment areas with strong aquatic-terrestrial interfaces, the floodplain extension is relatively small and inputs of terrestrially produced organic matter are high. Middle: in natural (mountainous, steep-bordered) or human-made (channelized) segments, the extension of the floodplains is restricted and carbon fixation occurs largely by riverine plankton and aquatic macrophytes. Below: in floodplain areas, carbon can contribute to river carbon budget via water flow from the floodplain to the mainstream or via feeding migration of fish and other aquatic animals between the floodplain and the main channel.
When considering the contribution of floodplain carbon to the entire river carbon budget, two aspects have to be stressed:
Flow conditions vary considerably between different river-floodplain systems. If geomorphology limits the exchange between river and floodplain, the contribution of the floodplain carbon to the mainstream carbon budget can be lower than expected. Lewis (1988) found that in the Orinoco and some tributaries, non-floodplain sources, including within-channel and near-channel stagnant or slow-flowing areas, accounted for 63 percent of the annual transport of phytoplankton carbon, while the floodplain accounted for only 37 percent.
Mobile organisms such as fish actively seek floodplain carbon in mass migrations as soon as flooding begins in order to feed in the floodplain (Welcomme 1985; Lowe-McConnell 1987; Winemiller 1989; Junk et al. 1997; Wantzen et al. 2002). When small floodplain fish migrate back into the main channel during the falling limb of the hydrograph, they are preyed upon in large quantities by riverine predators (Wantzen et al. 2002). Therefore, floodplain carbon can contribute significantly to river food webs without strong hydrological connectivity.
THE MULTIPLE USE CONCEPT
Floodplain management should be based on conceptional considerations in order to avoid negative side effects as much as possible (e.g. Nienhuis, Leuven and Ragas 1998). The FPC predicts exchange of nutrients and energy between the aquatic and terrestrial phases. Human use of terrestrial resources will affect aquatic resources and vice versa. These impacts have to be considered when developing management concepts.
The economic and ecological analysis of the different utilization forms led to the formulation of an integrated multiple-use concept for the central Amazon River floodplain (Junk 2000). It favours the optimization of the use of different resources instead of the maximization of the economic return of a single resource. Priority is given to the sustainable use of fishery resources because of low environmental impact, large requirement of labour force and high economic importance. Subsistence fisheries can be combined with smallholder agriculture and dairy farming on the highest levees and floodplain-adapted sustainable forest management by selective logging. Large-scale cattle and water buffalo ranching for beef production and agro-industries are considered destructive for the ecosystem because of the destruction of floodplain forests and socially unacceptable because of low labor force requirements (Junk et al. 2000). Decentralized administration of floodplain resources by local communities is considered essential to stimulate the participation of the local population in the complex management processes (Isaac et al. 1998; McGrath et al. 1999). Multiple use concepts will vary considerably for different floodplains because of the large variety of floodplain types and related socio-economic parameters.
OPEN QUESTIONS IN FLOODPLAIN RESEARCH
When river or lake water inundates the floodplain via overspill or via floodplain channels, various key processes occur simultaneously: (1) pre-flood thermal and chemical heterogeneity between main channel and floodplain water bodies temporarily resets (Sabo et al. 1999a; Tockner et al. 2000); (2) considerable inputs of mainstream (or lake) water-bound substances (dissolved and suspended, organic and inorganic) flush into the floodplain (Fisher and Parsley 1979; Lewis et al. 2000); (3) terrestrial habitats are flooded, large amounts of biomass decays and large amounts of inorganic and organic matter deposited during the terrestrial phase are mobilized by the overlaying water (Hamilton et al. 1997; Sabo et al. 1999b); (4) terrestrial organisms migrate into non-flooded habitats or show adaptations to flooding (Adis 1984; Adis, Marques and Wantzen 2001); (5) aquatic organisms are flushed or migrate into the floodplain or eclode from resting stages (Welcomme 1985; Irmler 1981) and (6) terrestrial carbon and floodplain products from the canopy of the floodplain forest, such as terrestrial invertebrates, fruits and seeds, are incorporated in the aquatic food webs (Junk et al. 1989; Wantzen et al. 2002).
When the water level falls the following key processes occur simultaneously: (1) water stored in the floodplain with any dissolved and suspended matter enters the parent river or lake (Benke et al. 2000), (2) the ATTZ falls dry and becomes colonized by terrestrial organisms (Junk and Piedade 1997; Adis and Junk 2002), (3) large amounts of water-borne organic carbon becomes stranded and incorporated in the terrestrial food webs (Junk and Weber 1996), (4) aquatic organisms move to permanent water bodies or show adaptations to periodic drought (Irmler 1981), (5) permanent water bodies become increasingly isolated from the parent river or lake and develop specific physical and chemical characteristics and specific species assemblages (Furch 1984; Tockner et al. 1999).
These changes either have a direct influence on aquatic and terrestrial flora and fauna in the floodplains and related rivers and lakes, for example through changes in the community composition and population density (e.g. Heckman 1998; De-Lamônica-Freire and Heckman 1996; Sabo et al. 1999b; de Oliveira and Calheiros 2000), or indirectly trigger various behavioral traits, such as spawning and migration of fish (Welcomme 1985; Junk et al. 1997), breeding of waterfowls (Petermann 1999; Magrath 1992) and reproduction and migration of terrestrial invertebrates (Adis and Junk 2002).
The complexity and the interdependence of these processes are yet not fully understood. Currently, questions arise about recent, past and future climatic impacts, the importance of landscape connectivity and dynamics of the flooding on biodiversity and biogeochemical cycles and how to include the results of floodplain research into sustainable management strategies.
THE IMPACT OF FLOOD PULSE ON WETLANDS IN DIFFERENT CLIMATIC ZONES
The FPC states that the flood pulse is the main driving force in river-wetland systems. This is true for the humid tropics, but in lower latitudes, there are other driving forces that also affect the biota and processes in the floodplains and that can overlap with the flood pulse. The FPC mentions these forces, but their impacts require more attention in comparative studies. In semiarid and arid regions, drought and fire affect the floodplains during the terrestrial phase, with consequences for the aquatic phase. In temperate regions, biota react to day length and/or temperature (light/temperature pulse) and this cycle is superimposed on the flood pulse (Junk 1999). Some effects (on fish fauna), because of the decoupling of the flood pulse from the temperature pulse, are discussed by Humphries et al. (1999) for the Murray-Darling River basin. In high latitudes, prolonged ice cover and low temperatures strongly affect the biota; and the biota might require as many adaptations to these events as to the flood pulse.
PALEOCLIMATOLOGICAL HISTORY OF FLOODPLAINS
The predictions of the FPC also have to be interpreted in the light of the paleo-ecological conditions that have influenced evolutionary processes and rates of speciation and extinction. For instance, the FPC states that predictable pulsing favors the development of adaptations of fauna and flora and increases species diversity. This statement holds true for some river floodplains, but not for others. The Amazon River floodplain is very rich in plant and animal species that are highly adapted to the predictable monomodal flood pulse. Approximately 1 000 flood-adapted tree species are found in the floodplains of the Amazon basin. In the floodplain of the Mamirauá Reserve near Tefé, about 800 km upstream of Manaus, which covers an area of about 11 240 km2, until today approximately 500 tree species have been recorded, about 80 percent of which are floodplain-specific (Wittmann 2002 and unpublished data). In comparison, the large majority of the about 250 tree species of the Pantanal of Mato Grosso, a wetland of approximately 140 000 km2, have broad ecological amplitude; only about 5 percent are restricted to regularly flooded areas (Nunes da Cunha and Junk 2001 and unpublished data). The number of flood-resistant tree species in bottomland hardwood forests of the USA approaches about 100 species, many of which also occur in the uplands (Clark and Benforado 1981). In northern European floodplains, only about a dozen flood-resistant tree species occur.
Many Amazonian soil arthropods are floodplain- specific and have complex survival strategies (Adis and Junk 2002). First observations indicate that flood-adapted soil arthropods in the Pantanal are less common than in Amazonia (Adis et al. 2001). Terrestrial soil invertebrates in Europe are poorly adapted to the flood pulse. Most are immigrants from the non-flooded uplands and suffer high losses during floods (Adis and Junk 2002).
Paleoclimatological history analysis of Amazonia shows that during the last ice age the temperature was probably about 5o C cooler, the precipitation about 50 percent lower (Haffer and Prance 2001), the declivity greater (Müller et al. 1995) and the floodplain area considerably smaller than today (Irion pers. comm.). However, the flood pulse continued to be monomodal and predictable with respect to dry and rainy seasons and there was sufficient floodplain area left to guarantee survival of flood-adapted plant and animal species. Despite the change in environmental conditions in the Amazon basin, basic structures and functions of the large river floodplains situated north and south of the equator were comparatively little affected and extinction rates were low. In comparison, during the ice ages, the Pantanal of Mato Grosso, about 2 500 km south of the equator, suffered from extremely dry periods that eliminated most of the flood-adapted plant and animal species.
Todays wetland conditions became reestablished in the Pantanal only about 7 000 years ago and wetland organisms of the lower Paraguay River, the surrounding Cerrado and Amazonia colonized the area (AbSaber 1988). Mobile animals, such as birds, which are very diverse in the Pantanal, were most efficient at colonization. However, pronounced annual and pluri-annual droughts in combination with frequent wildfires led to additional stress for plants and animals. A broad ecological amplitude was a better survival strategy for the organisms than adaptation to specific wetland conditions, as shown by trees that occur over a large range of habitats. The number of total species and the level of adaptation are comparatively low and endemic species are rare because the time span after the dry glacial period was too short for genetic diversification (da Silva et al. 2001). This holds true even with respect to genera that show high diversification rates, as for instance, the tree genus Inga. Most of the 300 species have developed in the last 2 million years (Richardson et al. 2001).
During the ice ages, European and North American river floodplains suffered even larger climatic changes. The temperature was lower and glaciers covered most of Northern Europe and North America. The discharge regime of the large rivers was determined by snow and ice melt. The light-temperature pulse strongly superimposed on the impact of the flood pulse. Todays wetlands of these areas began to develop about 10 000 years ago with deglaciation and there was very little time for organisms to adapt to the new conditions in the floodplains. However, during the ice ages, North American floodplain species could migrate to a certain extent to lower latitudes and later recolonize the newly formed wetlands, an option that was blocked in northern Europe by high mountains (Alps and Pyrenees), which explains the relatively small number and low level of adaptations of organisms to flooding in the European floodplains. These examples also illustrate the influence of the time scale to the degree of specialization and the development of flood-adapted communities.
CONNECTIVITY AND LENTIC-LOTIC LINKAGES
Amoros and Roux (1988) introduced the technical term "connectivity" from landscape ecology to limnology in order to describe the level of connection of the mainstream with floodplain lakes. Connectivity levels vary from permanent connection to short-term connection during extreme floods (Ward, Tockner, and Schiemer 1999; Wantzen and Junk 2000). With decreasing connectivity, the impact of the river on floodplain lakes diminishes and lakes develop their own limnological characteristics. For the Austrian Danube River floodplain near Regelsbrunn, Tockner et al. (1999) have shown that species number and community structure of many aquatic organisms change depending on the connectivity level. The quality of connectivity changes when, at very high water levels, floodplain lakes change from water storage to water transport systems, i.e. from a lentic to a lotic system (limnophase and potamophase, sensu Neiff 1990). Strong flow pulses (Puckridge et al. 1998) may lead to dramatic resetting of the limnetic succession by cleaning the lake of accumulated organic debris and profoundly modifying aquatic plant and animal communities (Marchese, Escurra de Drago and Drago 2002). On the other hand, the establishment and the cessation of flow conditions are crucial for the oxygen budget in detritus-rich floodplains. Two processes become important: 1) When terrestrially accumulated organic matter becomes flooded and decays, large amounts of oxygen can be consumed, as shown for tropical floodplains (Braum and Junk 1982; Junk, Soares and Carvalho 1983; Sabo et al. 1999b), which eventually cause fish kills (Hamilton et al. 1997) and 2) when flow ceases in floodplain water bodies, high-water-level stratification and hypoxia occur (Melack and Fisher 1983; Junk et al. 1983; Sabo et al. 1999b), which affects aquatic organisms.
Very large floodplains have complex connectivity patterns. For example, lakes can become connected to the mainstream by other lakes. In this case, migrating aquatic organisms have free access to lakes and the river; however, input of dissolved and suspended matter is concentrated in the lake near the river and is low in the remote lake, which maintains lacustrine conditions. Tributaries with different water quality can cause hydro-chemical disconnection of floodplain lakes, as shown in the Ria Lakes formed by clear water and black water tributaries in the Amazon River floodplain. These lakes can be permanently connected to the Amazon River, which transports white water. When the water level falls, the black and clear water of the tributaries advances to the lake mouth; when the water rises, the white water of the mainstream represses the river water and dominates part of the lake. The mobile frontier between river water and the water of the tributaries can become a hydro-chemical barrier for aquatic organisms despite the hydrological connection, as shown by the growth of aquatic macrophytes and the occurrence of water snails and bivalves that concentrate in the whitewater-influenced area because of better nutrient and calcium supply and higher pH values (Junk, unpubl.). Detailed studies on the impact of the hydrological and hydro-chemical connectivity level on flora and fauna in tropical river floodplain systems are still lacking.
Lentic-lotic linkages have so far been considered mainly in interconnected rivers and lakes where lentic and lotic conditions alternate along the continuum of the river course. Similar linkages exist in river-floodplains systems; however, they occur in a temporal dimension (syntopically during different hydrological periods) rather than in a spatial dimension (synchronically at different sites Figure 3). Adopting this perspective, spatially scaled processes in weakly or non-pulsing systems (e.g. regulated rivers) can be considered analogous to temporal scales in pulsing systems. For example, regulated lakes and rivers are stratified into profundal and littoral zones all year round. Processes such as open-water plankton production and shallow-site plankton filtering by benthos are spatially separated, but linked by the water movement. In river-floodplain systems, both processes can occur at the same site, but at different water levels. The moving littoral follows the rising or sinking water level in the ATTZ. In the same way, infralittoral and profundal zones migrate along the flooding gradient, provided that the depth is sufficient. When water levels recede, the functional units of the deep water disappear in floodplain systems (Figure 3).
Figure 3. Schematic comparison between hydrologically stabilized (left) and pulsing (right) aquatic ecosystems at normal water levels (0), extremely high water levels (+1) and extremely low water levels (-1). Left: in the stabilized system (regulated lake or river or natural water body without floodplain), water level 0 prevails most of the time, allowing the establishment of well defined littoral (L), pelagial (O) and profundal (P) communities that are well adapted to these environmental conditions and are optimized in using the locally occurring resources. Occasional extreme floods are catastrophic events that do not allow the use of the resources of the flooded epilittoral by flora and fauna. Right: in pulsing systems, the organisms are adapted to periodically changing water levels and profit from resources of varying origin. Flora and fauna move along the flooding gradient; therefore, the same place in the ecosystem can harbor littoral and profundal communities at different water levels. Not shown in the graph: Extreme low water levels urge the profundal and pelagial organisms either to migrate into deeper water bodies or to estivate at the sites, whereas terrestrial (epilittoral) species have developed survival strategies during flooding.
TIMING AND SHAPE OF THE FLOOD PULSE
The FPC has drawn attention to the importance of the timing of the flood pulse and the stage of the life cycle of the organisms, but for many floodplains, data are still insufficient for detailed predictions. Many floodplain organisms have a "physiological and phenological window of susceptibility" to the benefits and disturbances of the flooding. The timing decides whether an organism can profit from the flood-borne resources or apply survival strategies or not. Winter flooding does not have such deleterious effects to non-flood-adapted trees as flooding during summer when trees are physiologically fully active. Similarly, unpredicted winter flooding had no significant effect on floodplain-carbon uptake by fish in a river in the USA (Delong et al. 2001), whereas predicted timely flooding in the Pantanal did (Wantzen et al. 2002). Most fish species of the upper Paraná River have adapted their spawning to the flood pulse and are affected by the many reservoirs that in addition to interrupting connectivity between river reaches, modify timing and shape of the pulse. These changes influence spawning behavior and affect recruitment success of some species but also affect community structure, for instance by increased predator pressure (Agostinho et al. 2000; Agostinho, Gomes and Zalewski 2001).
A slowly rising water level of the Amazon River leads to interruption and/or delay of spawning migration of many migrating fish species and in extreme conditions to gonad absorption (Junk pers. obs.). Different flood patterns lead to different macrophyte assemblages, which in turn are important habitats and food sources for many fishes (Petr 2000). The effects of different flood patterns on fish populations have been summarized by Welcomme and Halls (2001).
The impact of human induced hydrological changes has been shown for seedling establishment of poplar (Populus spp.) in North American rivers (Rood and Mahoney 1990). Timing of floods for the management of grasses and herbs for ducks and geese is a major tool in polders along the Mississippi River (Fredrickson and Reid 1988; Reid et al. 1989). Comparative studies on aquatic macrophytes and water birds in the central Amazon River floodplain and the Pantanal of Mato Grosso point to the importance of the amplitude of the flood pulse for species composition and life forms. In the Amazon River flood plain, a high flood amplitude of up to 15 meters hinders the growth of submersed plants and probably also the food uptake of some wading birds. Both groups occur in large abundance and species numbers in the Pantanal of Mato Grosso, where the flood amplitude is only 1-3 meters (Junk and Petermann, unpubl. data). However, for most plant and animal species and communities such information is still missing. Considering the increasing man-made changes of river discharge, studies are required for a better understanding of the impact of the quality of the flood pulse on the biota.
EXTREME CLIMATIC EVENTS AND GLOBAL CHANGE
The effect of extreme hydrologic and climatic events on river-floodplain systems has been stressed by the FPC, but has been little studied. Long-term and deep flooding affect the ecosystem through profound levels of hydraulic energy and/or by physiological stress. Studies on streams show that 80 percent of the annual transport of particulate organic matter can occur during a single extreme flooding (Cummins et al. 1983; Hobbie and Likens 1973). Such an event reshapes the entire channel bed and the floodplain of rivers in mountainous regions, such as the Tagliamento River in the Alps (Arscott, Tockner and Ward 2000) and also strongly modifies the floodplain of lowland rivers (Sparks, Nelson and Yin 1998).
Pluriannual dry and wet periods can have long-lasting effects on community structure in floodplains. The long flood period of the Amazon River in the beginning of the 1970s led to the dieback of many floodplain trees in low-lying areas. These areas still have not yet been recolonized by trees; a pluriannual dry period is required for successful reestablishment in very low lying areas on the flood gradient (Junk 1989). The spread of Vochysia divergens, a flood-tolerant tree species, in the Pantanal of Mato Grosso during the last 30 years has been associated with a long-lasting wet period after a pluriannual dry period in the beginning of the 1960s (Nunes da Cunha and Junk unpublished). Fish catches in the central delta of the Niger River declined from 90 000 tonnes yr-1 to 45 000 tonnes yr-1 because of little rainfall in the 1980s (Lae 1994).
The study of the impacts of extreme climatic events will be crucial for wetland ecosystem management and protection strategies. The IPCC (2001) indicates that the planet Earth will suffer considerable climate changes during the next century, which will be, to a considerable extent, the result of a man-made increase in greenhouse gases, such as carbon dioxide and methane. A global average temperature increase of 1.4 to 5.8o C is predicted. Nearly all land masses, mainly those at northern high latitudes during the cold season, will warm more rapidly than the global average. Global mean sea level is projected to rise by 0.09 to 0.88 m because of temperature-related expansion of the water and melting of the glaciers of the northern polar regions and high mountains. Changes in precipitation will occur in most regions - rainfall will increase in some regions and drought will increase in others. The strongest impact will be felt in northern sub-polar regions (permafrost regions), high mountains, coastal areas, deserts and savannas, where water is already a limiting factor. In many river floodplains, man-induced changes of hydrology, pollution and wetland destruction will be more important than the effects of climate change (Vörösmarty 2002), but extreme climate change events will overlap with other human-induced modifications and aggravate the situation.
As stated by the FPC and other authors, floodplains are hot spots of species diversity (Gopal and Junk 2000). They harbour not only many wetland-specific plants and animals, but also many species from adjacent terrestrial and deep-water habitats that can have fundamental impacts on structures and functions of floodplains. For instance, terrestrial plant species substantially contribute to habitat diversity; primary production and nutrient cycles and terrestrial ungulates affect plant community structure and increase secondary production. However, inventories of floodplain species are rare and incomplete because they require interdisciplinary approaches (Gopal, Junk and Davies 2000, 2001).
One aspect of flooding is a variably strong disturbance that can modify or even reset environmental conditions in the system. Therefore, the FPC has integrated the tenets of the intermediate-disturbance hypothesis (Connell 1978; Ward and Stanford 1983b) by predicting that floodplain areas with an intermediate (and predictable) level of flooding are expected to provide the highest diversity. The two extremes for the disturbance-diversity relationship for a given floodplain habitat are, therefore, (1) frequent-to-permanent changes in the physical habitat structure caused by flooding (e.g. rainfall-driven floodplain habitats in low-order streams) and (2) low number or lack of hydrological changes with a continuous ecological succession of species, leading to a climax community (e.g. remote floodplain lakes during terrestrialization). By interrupting ecological succession in some patches, flooding causes the development of a mosaic of different successional stages at the same time on a small spatial scale. The intensity of multi-year wet and dry phases in floodplains, however, can provide additional stressors. In the Pantanal of Mato Grosso, for instance, the occurrence of numerous life forms is limited by the extreme desiccation, combined with fires during the dry phase (Nunes da Cunha and Junk 2001; da Silva et al. 2001).
In riverine floodplains, hydrological variation shapes a high diversity of physical habitat structures that might be more heterogeneous across the floodplain than along the main channel (Marchese and Ezcurra de Drago 1992; Arscott et al. 2000), thus creating the basis for a diverse flora and fauna. In Amazonia, forest diversity is related to river dynamics (Salo et al. 1986). However, flooding and drought can also reduce spatial heterogeneity by linking aquatic populations that were separated in different water bodies during the low-water period (vice versa, isolated terrestrial populations during a flooding period can mix genetically during drought). Similarly, the permanent drift of organisms from the catchment or upriver areas inoculates the riverine or near-river populations regularly and thus hinders the development of genetically distinct populations.
Connectivity between the main channel and the floodplain habitat has become a central theme in the biodiversity debate (Ward, Tockner and Schiemer 1999; Wantzen and Junk 2000; Amoros and Bornette 2002). Lateral connectivity has been suggested to determine the diversity patterns of many taxonomic groups directly (Tockner et al. 1999). Flood-pulsing systems encounter variable degrees and spatiotemporal patterns of connectivity. Therefore, the diversity of hydrological patterns is a key element for the maintenance of habitat and species diversity in river-floodplain systems.
According to the FPC, river floodplains can be considered as biogeochemical reactors that temporarily store and process organic and inorganic matter. The flood pulse exerts hydraulic forces that erode, carry and deposit these substances. Long-term storage favors in situ alteration, weathering and liberation of dissolved substances, as shown for an Amazonian Várzea lake (Weber, Furch and Junk 1996; Irion, Junk and de Mello 1997). The water level fluctuations provoke changes in water chemistry by mixing water bodies and resource input during the rising limb of the hydrograph and by increasing stratification, oxygen consumption and concentration of ions in the restricted water bodies during the falling limb. In floodplains that widely dry out periodically, like the Pantanal, a large part of the organic matter is turned over during the change of the hydrological phases.
Periodic flooding and drought of sediments leads to sequential occurrence of different redox processes. For example, organisms like cyanobacteria and legumes fix atmospheric nitrogen, but the change between anoxic and oxic conditions during the water-land transition and the availability of large amounts of organic material favor denitrification (Kern, Darwich, Furch, et al. 1996; Kern and Darwich 1997). Wassmann and Martius (1997) estimate the methane production of the Amazon River floodplain at 1-9 Tg CH4 yr-1, corresponding to 1-8 percent of the global source strength of wetlands. High primary production leads to considerable pulses in carbon dioxide uptake and release, but also to carbon storage in the sediment and carbon export to the oceans. About 1014 g of organic carbon is annually transported by the Amazon River to the Atlantic Ocean (Ritchey et al. 1980). A considerable part of it may derive from the floodplain (Junk 1985).
Junk (1980) points to an underestimate of the total wetland area in tropical South America because small wetlands are often not considered in inventories, although they might comprise about 50 percent of the total wetland area, most of them floodplains. This might also hold true to some extent for other tropical and subtropical regions. We hypothesize that these small floodplains and temporary wetlands also follow the predictions of the FPC. Mapping of these areas and inclusion of their impact on the budgets of biogeochemical cycles and the hydrological cycle and for maintenance of biodiversity are challenges for the future.
SUSTAINABLE MANAGEMENT AND RESTORATION OF RIVER FLOODPLAINS
River floodplains have provided multiple benefits since early human settlement. Predictable floods favored the management of floodplain resources and the development of ancient cultures, for example, on the Euphrates and Tigris Rivers and the Nile River several thousand years ago. Pre-Columbian human density in the floodplain of the Amazon River was several times higher than that in the adjacent upland. Rice cultivation started in China about 7000 years ago (Boulé 1994) and continues to be the nutritional basis for much of the human population worldwide.
The economic value of floodplains for buffering extreme hydrological events has been underestimated for a long time. A dramatically increasing human population during the last two centuries led to large-scale floodplain destruction and deterioration worldwide (Junk 2002, Tockner and Stanford 2002). In the past, large flood events led to heavy losses of goods and humans in Europe and brought about major flood control measures, such as the "correction" of the Rhine River by Tulla in the nineteenth century (Friedrich and Mueller 1984). The 500-year-old European tradition in river regulation (Nienhuis et al. 1998) was first transferred to North America and later applied worldwide. Only some decades ago did the negative ecological, economic and social side effects of floodplain destruction become apparent, as recently shown by the catastrophic floods along the Odra, Elbe, Rhine and Danube Rivers in 2001 and 2002 in Poland, the Czech Republic, Germany and Austria.
Management plans are required for the sustainable use of floodplain resources. The FPC provides general outlines that can be used for the development of management strategies; however, considering regional differences in the status of floodplain integrity, watershed management and demographic and economic development, there is a need for specific strategies for each floodplain and even for different stretches of large river floodplains. For instance, the importance of floodplains for protein supply by fisheries is low in most Central European rivers; however, in most tropical countries, floodplain fishery provides accessible animal protein for millions of people and is one of the most important economic activities (Welcomme 1985).
Knowledge on wetland restoration has been increasing rapidly for several decades and ambitious restoration projects are being undertaken in North America and Europe (Mitsch and Jørgensen 2003). Some restoration projects have also been started in the tropics. These projects are excellent means of validating predictions of the FPC, as shown by Heiler et al. (1995) for the Danube River. Creating and maintaining natural variation of the pulsing hydrograph and the ability of the landscape to develop a dynamic floodplain appear to be the most important elements for conservation and restoration concepts (Sparks et al. 1990, 1998; Tockner et al. 1999).
Most freshwater systems are subjected to fluctuations in water levels. All systems that are not steeply bordered by mountains, dykes, or regulating channels are fringed with floodable areas. Flooding is controlled by climate type (catchment rainfall patterns and evapotranspiration), landscape morphology (declivity and connectivity) and local effects (log jams, tributary inflows, recent local precipitation). With the knowledge of these variables, inundation-duration curves can be plotted, as for instance, for a US coastal plain river (Benke et al. 2000), for a small alpine river (Arscott et al. 2000) and for the Pantanal wetland (Hamilton et al. 1996). This general pattern makes the central tenet of the FPC - that hydrological pulsing is the driving force for the performance of organisms and for patterns of ecological processes - a unifying theme in limnological conceptualization.
Today, 17 years after its first presentation, the FPC is widely accepted and applied by most river ecologists. It provides a conceptual framework for both research and management in river-floodplain systems. Several researchers have refined its tenets. Even in upstream areas, unpredictable flood pulses can be profitable for the stream community (Wantzen and Junk 2000), but this does not seem to be the case for regulated large rivers (Thorp et al. 1998). The characteristics of the pulse shape are crucial for the establishment and survival of many aquatic organisms (Welcomme and Halls 2001; Wagner and Schmidt, unpublished manuscript). Flood pulses homogenize water quality and habitat structure of formerly isolated water bodies (Marchese and Escurra de Drago 1992; Heckman 1994; Tockner et al. 2000). It has also become clear that there is no "either/or" distribution between productivity in the catchment, the river channel and the floodplain, but rather a variable combination of these three sources for the food webs of the river-floodplain continuum (Figure 3).
Recent studies have shown that predictions of the FPC on the development of adaptations and survival strategies of organisms have to be adjusted by additional information on paleoclimatological history (Adis and Junk 2002). The interaction of the flood pulse with other environmental variables, such as the light/temperature pulse, snow melting and prolonged ice cover in high-latitude floodplain systems and rainy and dry seasons in arid regions, is not sufficiently understood (Humphries et al. 1999). Also, the impacts of short- and long-term changes of the quality of the flood pulse on life history of organisms, communities and biogeochemical processes require additional studies. The FPC also makes predictions about organisms and processes during the terrestrial phase at low water periods (Adis and Junk 2002; Parolin et al. in prep) that require additional studies. New techniques, such as stable isotope determination, remote sensing, genetic tests and techniques for wetland restoration and management provide powerful tools to test and refine the FPC further.
We thank Rüdiger Wagner (MPIL-Schlitz) for discussions on the "phenological and physiological window of susceptibility", Sabine Meier for editing the manuscript, Elke Bustorf for graphic design and Karen A. Brune for correction of the English. This paper is dedicated to the memory of Professor Jürgen Schwoerbel (1927-2002), one of the pioneers in modern running-water ecology.
AbSaber A.N. 1988. O Pantanal Mato Grossense e a teoria dos refugios. Revista Brasileira de Geographia, 50: 9-57
Adis J. 1984. Seasonal Igapo-forests of Central Amazonian blackwater rivers and their terrestrial arthropod fauna. In The Amazon. Limnology and landscape ecology of a mighty tropical river and its basin. pp. 245-268, Dordrecht, W. Junk Publishers.
Adis J. & Junk W.J. 2002. Terrestrial invertebrates inhabiting lowland river floodplains of Central Amazonia and Central Europe: A review. Freshwater Biology, 47: 711-731.
Adis J., Marques M.I., Wantzen K.M. 2001. First observations on the survival strategies of terricolous arthropods in the northern Pantanal wetland of Brazil - scientific note. Andrias, 15: 127-128.
Agostinho A.A., Gomes, L.C. & Zalewski M. 2001. The importance of floodplains for the dynamics of fish communities of the upper river Paraná. Ecohydrology & Hydrobiology, 1: 209-217.
Agosthinho A.A., Thomaz S.M., Minte-Vera C.V. & Winemiller K.O. 2000. Biodiversity in the high Paraná River floodplain. In B. Gopal, W.J. Junk & J.A. Davis eds. Wetlands Biodiversity. pp. 89-118. Leiden, The Netherlands, Backhuys Publishers.
Amoros C. & Bornette G. 2002. Connectivity and biocomplexity in waterbodies of riverine floodplains. Freshwater Biology, 47: 761-776.
Amoros C. & Roux A.L. 1988. Interaction between water bodies within the floodplains of large rivers: Function and development of connectivity. Münstersche Geographische Arbeiten, 29: 125-130.
Arscott D.B., Tockner K. & Ward J.V. 2000. Aquatic habitat diversity along the corridor of an Alpine floodplain river Fiume Tagliamento, Italy. Archiv für Hydrobiologie, 149: 679-704.
Bayley P.B. 1980. The limits of limnological theory and approaches as applied to river-floodplain systems and their fish production. Tropical Ecology & Development, 739-746.
Bayley P.B. 1991. The flood-pulse advantage and the restoration of river-floodplain systems. Regulated Rivers: Research & Management, 6: 75-86.
Benke A.C., Chaubey I., Ward G.M. & Dunn, E.L. 2000. Flood pulse dynamics of an unregulated river floodplain in the southeastern US coastal plain. Ecology, 81: 2730-2741.
Boulé M.E. 1994. An early history of wetland ecology. In Global wetlands: Old world and new. pp. 57- 74.
Braum E. & Junk W.J. 1982. Morphological adaptation of two Amazonian characoids Pisces for surviving in oxygen deficient waters. Int. Revue. Ges. Hydrobiol., 67: 869-886.
Bretschko G. & Leichtfried M. 1988. Distribution of organic matter and fauna in a second order, alpine gravel stream Ritrodat-Lunz study area, Austria. Verh. Int. Verein. Limnol., 23: 1333- 1339.
Clark J.R. & Benforado J. 1981. Wetland of bottomland hardwood forest. Dev. Agric. Managed-Forest Ecol. 11: Amsterdam, Oxford, New York, Elsevier Scientific Publishing Company.
Connell J.H. 1978. Diversity in tropical rain forests and coral reefs. Science, 199: 1302-1309.
Cummins K.W., Sedell J.R., Swanson F.J., Minshall S.G., Fisher S.G., Cushing C.E., Petersen R.C. & Vannote R.L. 1983. Organic matter budgets for stream ecosystems: Problems in their evaluation. In J.R. Barnes & G.W. Minshall eds. Stream ecology: Application and testing of general ecological theory. New York, Plenum Press. pp. 299-353.
Da Silva C.J., Wantzen K.M., Nunes da Cunha C. & Machado F.A. 2001. Biodiversity in the Pantanal Wetland, Brazil. In Biodiversity in wetlands: Assessment, function and conservation, Volume 2. pp. 187-215. Leiden, The Netherlands, Backhuys Publishers.
De Oliveira M.D. & Calheiros D.F. 2000. Flood pulse influence on phytoplankton communities of the south Pantanal floodplain, Brazil. Hydrobiologia, 427: 101-112.
De Lamônica-Freire E. & Heckman C.W. 1996. The seasonal succession of biotic communities in wetlands of the tropical wet-and-dry climatic Zone III: The algal communities in the Pantanal of Mato Grosso, Brazil, with a comprehensive list of the known species and revision of two desmid taxa. Int. Rev. Ges. Hydrobiol., 81: 253-280.
Delong M.D., Thorp J.H., Greenwood K.S. & Miller M.C. 2001. Responses of consumers and food resources to a high magnitude, unpredicted flood in the upper Mississippi River Basin. Regulated Rivers: Research & Management, 17: 217-234.
Dettmers J.M., Wahl D.H., Soluk D.A. & Gutreuter S. 2001. Life in the fast lane: Fish and foodweb structure in the main channel of large rivers. Journal of the North American Benthological Society, 20: 255-265.
Dudgeon D. 2000. The ecology of tropical Asian rivers and streams in relation to biodiversity conservation. Annual Review of Ecology and Systematics, 31: 239-263.
Dynesius M. & Nilsson C. 1994. Fragmentation and flow regulation of river systems in the northern 3rd of the world. Science, 266: 753-762.
Elwood J.W., Newbold J.D., ONeill R.V. & van Winkle W. 1983. Resource spiralling: An opearational paradigm for analyzing lotic systems. In T.D. Fontaine III & S.M. Bartell eds. Dynamics of lotic ecosystems. Ann Arbor, MI, USA, Ann Arbor Science Publishers. pp. 3-27.
Fisher T.R. & Parsley P.E. 1979. Amazon lakes: Water storage and nutrient stripping. Limnol. Oceanogr. L 24: 547-553.
Fredrickson L.H. & Reid F.A. 1988. Waterfowl use of wetland complexes. U.S. Fish Wildlife Leaflet 13.2.1. 6 pp.
Friedrich G. & Mueller D. 1984. Rhine West Germany. In Ecology of European Rivers. pp. 265-316., Palo Alto, CA, USA, Blackwell Scientific Publications.
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. Environmental Management, 10: 199-214.
Furch K. 1984. Interannuelle variation hydrochemischer parameter auf der Ilha de Marchantaria. Biogeographica, 19: 85-100.
Galat D.L. & Zweimuller I. 2001. Conserving large-river fishes: Is the highway analogy an appropriate paradigm? Journal of the North American Benthological Society, 20: 266-279.
Gopal B. & Junk W.J. 2000. Biodiversity in wetlands: An introduction. In Biodiversity in Wetlands: Assessment, function and conservation, Volume 1. pp. 1-10. Leiden, The Netherlands, Backhuys Publishers.
Gopal B., Junk W.J. & Davies J.A. 2000. Biodiversity in wetlands: Assessment, function and conservation, volume 1. 353 pp. Leiden, The Netherlands, Backhuys Publishers. Gopal B., Junk W.J. & Davies J.A. 2001. Biodiversity in Wetlands: assessment, function and conservation, volume 2. 311 pp. Leiden, The Netherlands, Backhuys Publishers.
Gutreuter S., Bartels A.D., Irons K. & Sandheinrich M.B. 1999. Evaluation of the flood-pulse concept based on statistical models of growth of selected fishes of the Upper Mississippi River system. Canadian Journal of Fisheries and Aquatic Sciences, 56: 2282-2291.
Haffer J. & Prance G.T. 2001. Climatic forcing of evolution in Amazonia during the Cenozoic: On the refuge theory of biotic differentiation. Amazoniana, XVI3/4: 579-607.
Hamilton S.K., Sippel S.J. & Melack J.M. 1996. Inundation patterns in the Pantanal wetland of South America determined from passive microwave remote sensing. Arch. Hydrobiol., 137: 1-23.
Hamilton S.K., Sippel S.J., Calheiros D.F. & Melack, J.M. 1997. An anoxic event and other biogeochemical effects of the Pantanal wetland on the Paraguay River. Limnol. Oceanogr., 42: 257- 272.
Heckman C.W. 1994. The seasonal succession of biotic communities in wetlands of the tropical wetand- dry climatic zone I: Physical and chemical causes and biological effects in the Pantanal of Mato Grosso, Brazil. Int. Revue. Ges. Hydrobiol., 79: 397-421.
Heckman C.W. 1998. The seasonal succession of biotic communities in wetlands of the tropical wetand- dry climatic zone: V aquatic invertebrate communities in the Pantanal of Mato Grosso, Brazil. Int. Revue. Ges. Hydrobiol., 83: 31-63.
Heiler G., Hein T., Schiemer F. & Bornette G. 1995. Hydrological connectivity and flood pulses as the central aspects for the integrity of a riverfloodplain system. Regulated Rivers: Research & Management, 11: 351-361.
Hobbie J.E. & Likens G.E. 1973. The output of phosphorous, dissolved organic carbon, and fine particulate carbon from Hubbard Brook watersheds. Limnol. Oceanogr., 18: 734-742.
Humphries P., King A.J. & Koehn J.D. 1999. Fish, flows and flood plains: Links between freshwater fishes and their environment in the Murray-Darling River system, Australia. Environmental Biology of Fishes, 56: 129-151.
Illies J. 1961a. Versuch einer allgemeinen biozönotischen Gliederung der Fließgewässer. Int. Revue Ges. Hydrobiol., 462: 205-213
Illies J. 1961b. Gebirgsbäche in Europa und in Südamerika - ein limnologischer Vergleich. Verh. Internat. Verein. Limnol., 14: 517-523
Illies J. & Botosaneanu L. 1963. Problemes et methodes de la classification et de la zonation ecologiques des eaux courantes, consideres surtout du point de vue faunistique. Mitt. Internat. Verein. Limnol., 12: 1-57.
IPCC 2001. Technical summary, and summary for policymakers. Third Assessment Report of Working Group I of the Intergovernmental Panel on Climatic Change. (Available at http://www. ipcc.ch)
Irion G., Junk W.J. & de Mello J.A.S.N. 1997. The large central Amazonian river floodplains near Manaus: Geological, climatological, hydrological, and geomorphological aspects. pp. 23-46. In Junk W.J. ed. The Central Amazon Floodplain - Ecology of a Pulsing System. Springer Verlag, Berlin.
Irmler U. 1981. Überlebensstrategien von Tieren im saisonal überfluteten Amazonischen Überschwemmungswald. Zool. Anz. Jena., 2061/2: 26.38.
Isaac V.J., Ruffino M.L. & McGrath D. 1998. In search of a new approach to fisheries management in the middle Amazon. In T.J. Quinn II, F. Funk, J. Heifetz, J. Ianelli, J. Power, J. Schweigert, P. Sullivan & C.I. Zhang eds. Fishery stock assessment model. pp. 889-902. AL, USA, Faibanks, Alaska Sea Grant College Program University of Alaska, As-SG-98-01.
Junk W.J. 1980. Áreas inundáveis - um desafio para limnologia. Acta Amazonica, 104: 775-795.
Junk W.J. 1985. The Amazon floodplain - a sink or source for organic carbon? Mitt. Geol. Paläont. Inst. Univ. Hamburg.
Junk W.J. 1989. Flood tolerance and tree distribution in central Amazonian floodplains. In L. B. Holm- Nielsen, I.C. Nielsen & H. Balslev eds. Tropical forests: Botanical dynamics, speciation and diversity. 47-64. London, Academic Press.
Junk W.J. 1999. The flood pulse concept of large rivers: Learning from the tropics. Arch. Hydrobiol. Suppl., 1153: 261-280.
Junk W.J. 2000. The central Amazon River floodplain: Concepts for the sustainable use of its resources. In W.J. Junk, J.J. Ohly, M.T.F. Piedade & M.G.M. Soares eds. The Central Amazon floodplain: Actual use and options for sustainable management. pp. 75-94. Leiden, The Netherlands, Backhuys Publishers.
Junk W.J. 2002. Long-term environmental trends and the future of tropical wetlands. Environmental Conservation, 294. (forthcoming)
Junk W.J. & Piedade M.T.F. 1997. Plant life in the floodplain with special reference to herbaceous plants. In W.J. Junk ed. The Central Amazon floodplain: Ecology of a pulsing system. pp. 147-186. Berlin, Springer Verlag.
Junk W.J. & Weber G.E. 1996. Amazonian floodplains: A limnological perspective. Verh. Internat. Verein. Limnol., 26: 149-157.
Junk W.J., Bayley P.B. & Sparks R.E. 1989. The flood pulse concept in river-floodplain systems. Can. Spec. Publ. Fish. Aquat. Sci., 106: 110-127.
Junk W.J., Soares G.M., Carvalho F.M. 1983. Distribution of fish species in a lake of the Amazon River floodplain near Manaus Lago Camelao with special reference to extreme oxgen conditions. Amazoniana, 7: 397-431.
Junk W.J., Soares G.M. & Saint-Paul U. 1997. The Fish. In W.J. Junk ed. The Central Amazon Floodplain - ecology of a pulsing system. pp. 385-408. Berlin, Springer Verlag.
Junk W.J., Ohly J.J., Piedade M.T.F. & Soares M.G.M. 2000. The Central Amazon floodplain: Actual use and options for sustainable management. Leiden, The Netherlands, Backhuys Publishers. 584 pp.
Kern J. & Darwich A. 1997. Nitrogen turnover in the várzea. In Junk W.J. ed. The Central Amazon floodplain - ecology of a pulsing system, 119- 136., Berlin, Springer Varlag.
Kern J., Darwich A., Furch K. & Junk W.J. 1996. Seasonal denitrification in flooded and exposed sediments from the amazon floodplain at Lago Camalão. Microbial Ecology, 32: 47-57.
Knowlton M.F. & Jones J.R. 1997. Trophic status of Missouri River floodplain lakes in relation to basin type and connectivity. Wetlands, 17: 468- 475.
Lae R. 1994. Effects of drought, dams, and fishing pressure on the fisheries of the central delta of the Niger River. International Journal of Ecology and Environmental Sciences, 20: 119-128.
Lewis W.M. Jr. 1988. Primary production in the Orinoco River Venezuela. Ecology, 69: 679-692.
Lewis W.M. Jr., Hamilton, S.K., Lasi, M.A., Rodríguez & Saunders M.A. III. 2000. Ecological determinism on the Orinoco floodplain. BioScience, 50: 681-692.
Lowe-McConnell R.H. 1975. Fish communities in tropical freshwaters: Their distribution, ecology and evolution. London, New York, Longman. 337 pp.
Lowe-McConnel R.H. 1987. Ecological studies in tropical fish communities. Cambridge, UK, Cambridge University Press.
Lytle D.A. 2001. Disturbance regimes and life-history evolution. American Naturalist, 157: 525-536.
Magrath, M.J.L. 1992. Waterbird study of the lower Lachlan and Murrumbidgee Valley wetlands in 1990/1991. A report prepared for the NSW Department of Water Resources.
Marchese M. & Ezcurra de Drago I.E. 1992. Benthos of the lotic environments in the middle Parana River system: Transverse zonation. Hydrobiologia, 237: 1-13.
Marchese M., Escurra de Drago I. & Drago E. 2002. Benthic macroinvertebrates and physical habitat relationships in the Paraná river-floodplain system. In M. McClain ed. International Association of Hydrological Sciences. Special publication No. 6. (forthcoming)
McGrath D., de Castro F., Câmara E. & Futema C. 1999. Community management of floodplain lakes and the sustainable development of Amazonian fisheries. Advances in Economic Botany, 13: 59-82.
Melack J.M. & Fisher T.R. 1983. Diel oxygen variations and their ecological implications in Amazon floodplain lakes. Arch. Hydrobiol., 98: 422- 442.
Minshall G.W. 1985. Developments in stream ecosystem theory. Journal of Fisheries and Aquatic Science, 42: 1045-1055.
Mitsch W.J. & Gosselink J.G. 2000. Wetlands. New York, USA, John Wiley & Sons.
Mitsch W.J. & Jørgensen S.E. 2003. Ecological engineering and ecosystem restoration. New York, John Wiley & Sons. (forthcoming)
Montgomery D.R. 1999. Process domains and the river continuum. Journal of the American Water Resources Association, 35: 397-410.
Müller J., Irion G., Nunes de Mello J. & Junk W.J. 1995. Hydrological changes of the Amazon during the last glacial-interglacial cycle in Central Amazonia Brazil. Naturwissenschaften, 82: 232-235.
Naumann E. 1932. Principles of regional limnology. Stuttgart, Germany, Schweizerbart Publishers. 149 pp.
Neiff J.J. 1990. Ideas for an ecological interpretation of the Paraná. Interciencia, 156: 424-441
Nienhuis P.H., Leuven, R.S.E.W & Ragas, A.M.J. 1998. New concepts for sustainable management of river basins. Leiden, The Netherlands, Backhuys Publishers. 367 pp.
Nunes da Cunha C. & Junk W.J. 2001. Distribution of woody plant communities along the flood gradient in the Pantanal of Poconé, Mato Grosso, Brazil. International Journal of Ecology and Environmental Sciences, 27: 63-70.
Parolin P., de Simone O., Haase K., Waldhoff D., Rottenberger S., Kuhn V., Kesselmeier J., Schmidt W., Piedade M.T.F. & Junk W.J. Central Amazonian floodplain forests: Tree survival in a pulsing system. (In preparation)
Petermann P. 1999. Biogeographie einer Insel-Avifauna in der Várzea des mittleren Amazonas, am Beispiel der Ilha de Marchantaria. University of Saarland. (Doctoral dissertation)
Petr T. 2000. Interactions between fish and aquatic macrophytes in inland waters: A review. FAO Fisheries Technical Paper, 396: 1-185.
Petts G.E. & Amoros C. 1996. Fluvial hydrosystems. London, Chapman & Hall. 322 pp.
Pinay G., Decamps, H. & Naiman, R.J. 1999. The spiraling concept and nitrogen cycling in large river floodplain. Archiv. Für Hydrobiologie, 143: 281-291.
Poff N.L. 1997. Landscape filters and species traits: Towards mechanistic understanding and prediction in stream ecology. Jr. N. Am. Benth. Soc., 16: 391-409.
Poff N.L. & Ward J.V. 1989. Implications of streamflow variability and predictability for lotic community structure - a regional-analysis of streamflow patterns. Canadian Journal of Fisheries and Aquatic Sciences, 46: 1805-1818.
Poole G.C. 2002. Fluvial landscape ecology: Addressing uniqueness within the river discontinuum. Freshwater Biology, 47: 641-660.
Pringle C.M. 2001. River conservation in tropical versus temperate latitudes. In Global Perspectives on River Conservation: Science, Policy and Practice. pp. 373-383. New York, John Wiley & Sons Ltd.
Pringle C.M., Naiman R.J., Bretschko G., Karr J.R., Oswood M.W. Webster, J.R., Welcomme R.L. & Winterbourn M.J. 1988. Patch dynamics in lotic systems: The stream as a mosaic. Jr. N. Am. Benth. Soc., 7: 503-524.
Puckridge J.T., Sheldon F., Walker K.F. & Boulton A.J. 1998. Flow variability and the ecology of large rivers. Marine and Freshwater Research, 49: 55-72.
Reid F.R., Kelley J.R. Jr., Taylor T.S. & Fredrickson L.H. 1989. Upper Mississippi Valley wetlands, refuges and moist-soil impoundments. In L. Smith, R. Pederson, R. Kaminski eds. Habitat management for migrating and witering waterfowl in North America. pp, 181-202. Lubbock TX, USA, Technical University Press.
Resh V.H., Hildrew A.G., Statzner B. & Townsend C.R. 1994. Theoretical habitat templets, species traits and species richness: Asynthesis of longterm ecological research on the Upper Rhone River in the context of concurrently developed ecological theory. Freshwater Biology, 31: 539-554.
Resh V.H., Brown A.V., Covich A.P., Gurtz L.H.W., Minshall G.W., Reicem S.R., Sheldon A.L., Wallace J.B. & Wissmar R.C. 1988. The role of disturbance in stream ecology. J. N. Am. Benth. Soc., 7: 443-455.
Richardson J.E., Pennington R.T., Pennington T.D. & Hollinsworth P.M. 2001. Rapid diversification of a species-rich genus of neotropical rain forest trees. Science, 293: 2242-2245.
Richey J.E., Brock J.T., Naiman R.J., Wissmar R.C. & Stallard R.F. 1980. Organic carbon: Oxidation and transport in the Amazon River. Science, 207: 1348-1351.
Richter B.D., Baumgartner J.V., Powell J. & Braun D.P. 1996. A method for assessing hydrologic alteration within ecosystems. Conservation Biology, 10: 1163-1174.
Robertson A.I, Bunn S.E., Boon P. & Walker K.F. 1999. Sources, sinks and transformations of organic carbon in Australian floodplain rivers. Marine & Freshwater Research, 50: 813-829.
Rood S.B. & Mahoney J.M. 1990. Collapse of riparian poplar forests downstream from dams in western prairies: Probable causes and prospects for mitigation. Environ. Manage., 14: 451-464.
Sabo M.J., Bryan C.F., Kelso W.E. & Rutherford A. 1999a. Hydrology and aquatic habitat characteristics of a riverine swamp: I. Influence of flow on water temperature and chemistry. Regulated Rivers: Research & Management, 15: 505-523.
Sabo M.J., Bryan C.F., Kelso W.E. & Rutherford A. 1999b. Hydrology and aquatic habitat characteristics of a riverine swamp: II. Hydrology and the occurrence of chronic hypoxia. Regulated Rivers: Research & Management, 15: 525-542.
Salo J., Kalliola R.J., Hakkinen I., Makinen Y., Niemala P., Puhaka M. & Coley P.D. 1986. River dynamics and the diversity of Amazon lowland forest. Nature, 322: 254-258.
Schwoerbel J. 1961. Über die Lebensbedingungen und die Besiedlung des hyporheischen Lebensraumes. Arch. Hydrobiol. Suppl, 25: 162-214.
Sedell J.R., Richey J.E., Swanson F.J. 1989. The river continuum concept: A basis for the expected ecosystem behavior of very large rivers? Can. Spec. Publ. Fish. Aquat. Sci., 106: 49-55.
Southwood T.R.E. 1977. Habitat: The template for ecological strategies. Journal of Animal Ecology, 46: 337-365.
Sparks R.E., Nelson J.C. & Yin Y. 1998. Naturalization of the flood regime in regulated rivers. BioScience, 48: 706-720.
Sparks R.E., Bayley P.B., Kohlert S.L., Osborne L.L. 1990. Disturbance and recovery of large floodplain rivers. Environmental Management, 14: 699-709.
Stanford A. & Ward J.V. 1988. The hyporheic habitat of river ecosystems. Nature, 335: 64-66.
Statzner, B. & Higler B. 1986. Stream hydraulics as a major determinant of benthic invertebrate zonation patterns. Freshwater Biology, 16: 127-139.
Thienemann A. 1925. Die Binnengewässer Mitteleuropas. Die Binnengewässer 1., Stuttgart, Germany, Schweizerbart Publishers.
Thorp J.H. & Delong M.D. 1994. The riverine productivity model: An heuristic view of carbon sources and organic-processing in large river ecosystems. Oikos, 70: 305-308.
Thorp J.H., Delong M.D., Greenwood K.S. & Casper A.F. 1998. Isotopic analysis of three food web theories in constricted and floodplain regions of a large river. Oecologia, 117: 551-563.
Tockner K. & Stanford J.A. 2002. Riverine flood plains: Present state and future trends. Environmental Conservation, 293: 308-330.
Tockner K., Malard F. & Ward J.V. 2000. An extension of the flood pulse concept. Hydrological Processes, 14: 2861-2883.
Tockner K., Schiemer F., Baumgartner C., Kum G., Weigand E., Zweimuller, I. & Ward, J.V. 1999. The Danube restoration project: Species diversity patterns across connectivity gradients in the floodplain system. Regulated Rivers: Research & Management, 15: 245-258.
Townsend C.R. 1989. The patch dynamics concept of stream community ecology. Jr. N. Am. Bentho. Soc., 8: 36-50.
Townsend C.R. 1996. Concepts in river ecology: Pattern and process in the catchment hierarchy. Archiv für Hydrobiologie Supplementband, 113 1-4: 3-21.
Townsend C.R. & Hildrew A.G. 1994. Species traits in relation to a habitat templet for river system. Freshwater Biology, 31: 265-275.
Triska F.J., Duff J.H. & Avanzino R. 1993. Patterns of hydrological exchange and nutrient transformation in the hyporheic zone of a gravel-bottom stream: Examining terrestrial-aquatic linkages. Freshwater Biology, 29: 259-274.
Van den Brink F.W.B., Beljaards M.J., Botts N.C.A. & van der Velde G. 1994. Macrozoobenthos abundance and community composition in 3 Lower Rhine floodplain lakes with varying inundation regimes. Regulated Rivers: Research & Management, 9: 279-293.
Van den Brink F.W.B., de Leeuw J.P.H.M., van der Velde G. & Verheggen G.M. 1993. Impact of hydrology on the chemistry and phytoplankton development in floodplain lakes along the Lower Rhine and Meuse. Biogeochemistry Dordrecht, 19: 103-128.
Vannote R.L., Minshall G.W., Cummins K.W., Sedell K.W., Cushing C.E. 1980. The river continuum concept. Can. J. Fish. Aquat. Sci., 37: 130-137.
Vörösmarty C.J. 2002. Global water assessment and potential contributions from Earth Systems Science. Aquatic Sciences, 64: 328-351.
Wantzen K.M. & Junk W.J. 2000. The importance of stream-wetland-systems for biodiversity: A tropical perspective. In B. Gopal, W.J. Junk & J.A. Davies eds. Biodiversity in Wetlands: Assessment, function and conservation, 11-34, Leiden, The Netherlands, Backhuys Publishers.
Wantzen K.M., Machado F.A., Voss M., Boriss H. & Junk W.J. 2002. Seasonal isotopic changes in fish of the Pantanal wetland, Brazil. Aquatic Sciences, 64: 239-251.
Ward J.V. 1989. The four-dimensional nature of lotic ecosystems. Jr. N. Am. Bentho. Soc., 8: 2-8.
Ward J.W. & Stanford J.A. 1983a. The serial discontinuity concept of lotic ecosystems. In T.D. Fontaine III & S.M. Bartell eds. Dynamics of lotic ecosystems, Ann Arbor, MI, USA, Ann Arbor Science Publishing. pp. 29-42.
Ward J.V. & Stanford J.V. 1983b.The intermediate-disturbance hypothesis: An explanation for biotic diversity patterns in lotic ecosystems. In T.D. Fontaine III & S.M. Bartell eds. Dynamics of lotic ecosystems, Ann Arbor, MI, USA, Ann Arbor Science Publishing. pp. 347-356.
Ward J.V. & Stanford J.A. 1995. The serial discontinuity concept: Extending the model to floodplain rivers. Regulated Rivers: Research & Management, 11: 105-119.
Ward J.V., Malard F. & Tockner K. 2001. Landscape ecology: A framework for integrating pattern and process in river corridors. Landscape Ecology, 17: 35-45.
Ward J.V., Tockner K. & Schiemer F. 1999. Biodiversity of floodplain river ecosystems: Ecotones and connectivity. Regulated Rivers: Research & Management, 15: 125-139.
Wassmann R. & Martius C. 1997. Methane emissions from the Amazon floodplain. In W.J. Junk ed. The Central Amazon Floodplain - Ecology of a Pulsing System. pp. 137-143. Berlin, Springer- Verlag.
Weber G.E., Furch K. & Junk W.J. 1996. A simple modelling approach towards hydrochemical seasonality of major cations in a Central Amazonian floodplain lake. Ecological Modelling, 91: 29-56.
Welcomme R.L. 1979. Fisheries ecology of floodplain rivers. London, Longman. 317 pp.
Welcomme R.L. 1985. River fisheries., FAO Fisheries Technical Paper 262. Rome, Food and Agricultural Organization of the United Nations. 330 pp.
Welcomme R.L. & Halls A. 2001. Some considerations of the effects of differences in flood patterns on fish populations. Ecohydrology and Hydrobiology, 13: 313-321
Williams D.D. 1989. Towards a biological and chemical definition of the hyporheic zone in two Canadian rivers. Freshwat. Biol., 22: 189-208.
Winemiller K.O. 1989. Patterns of variation in life-history among South American fishes in seasonal environments. Oecologia, 81: 225-241.
Winterbourn M.J., Rounick, J.S. & Cowie, B. 1981. Are New Zealand stream ecosystems really different? New Zealand Journal of Marine and Freshwater Research, 15, 321-328.
Wittmann F. 2002. Species distribution, community structure, and adaptations of Amazonian várzea forests depending on the annual flood-stress, using remote sensing methods. University of Mannheim. (Doctoral dissertation)