The influence of flood cycle and fish predation on invertebrate production on a restored California floodplain

Although floodplains are known to be tightly controlled by the flood cycle, we know comparatively little about how flooding influences predators and their consumption of secondary production, particularly in highly seasonal floodplains typical of Mediterranean climates. In this study, we investigate how the seasonal dynamics of a central California floodplain influence the timing and magnitude of fish predation and the abundance and composition of invertebrates. For 3 years (2000–2002), we compared changes in abundances and size distributions of invertebrates through the flood season (January–June) with seasonal changes in the abundance of larval and juvenile fishes. Using diet analysis of fishes and manipulative feeding experiments with fishes in field enclosures, we link specific changes in invertebrate populations directly to feeding preferences of seasonally abundant fish. Early in the flood season prior to March, we found little influence of fish predation, consistent with the near absence of larval and juvenile fishes during this period. Coinciding with the midseason increase in the abundance of larval and juvenile fishes in April, we found significant declines in zooplankton abundance as well as declines in the size of zooplankton consistent with fish feeding preferences. Our results were consistent with results from feeding enclosure experiments that showed that fish rapidly depressed populations of larger cladocerans with much less effect on smaller cladocerans and calanoid copepods. At the end of the flood season, zooplankton abundances rapidly increased, consistent with a switch in the feeding of juvenile fish to aquatic insects and subsequent fish mortality. We also found that zooplankton biomass on the floodplain reached a maximum 2–3 weeks after disconnection with the river. We suggest that floodplain restoration in this region should consider management strategies that would ensure repeated flooding every 2–3 weeks during periods that would best match the peaks in abundance of native fishes. Electronic supplementary material Electronic supplementary material is available for this article at and accessible for authorised users.     

Towards a model of a floodplain fish population and its fishery

Synopsis A model is developed which describes the way in which the fish populations of African rivers and their fisheries are influenced by the different types of flood Regime. Ichthyomass and fish catch are dependent on both the extent of flooding during high water and the amount of water remaining in the system during the dry season. The relative number and individual weight of fish are determined by the intensity of flooding, whereas the total number surviving to the next year depends more on the low water regime. Catch per unit effort falls with increasing difference between areas flooded at high and low water. A negative log-log relationship exists between catch and the ratio of maximum area flooded to minimum area of water remaining in the system. This relationship of catch to flood ratio may form the basis for a general index for the evaluation of both year-to-year variations within a floodplain, and differences between floodplains. Lines of equal catch are also derived for various combinations of high and low water areas; these might be used as guidelines for the hydrological management of tropical flood-plains.     

Cyanobacterial blooms in a temperate river-floodplain ecosystem: the importance of hydrological extremes

In the past decade, extreme hydrological events were expressed with extreme droughts and floods in temperate regions. The aim of this paper is to explain how such changes in hydrology can influence cyanobacterial populations in floodplain ecosystems. We therefore analyzed a 6-year (2003–2008) study of the phytoplankton in the Kopački Rit floodplain, one of the largest natural floodplains in the middle section of the Danube River (Europe). During the studied period, the shallow floodplain lake shifted between a state of turbid water, characterized by high phytoplankton biomass and regular appearance of cyanobacteria blooms, to a state of clear water with very low phytoplankton biomass and absence of cyanobacteria, and back to the turbid state. Apparently, the major forces driving the cyclic shift were closely related to extremely high and long-lasting flood events. Significant increase in water level, low hydraulic residence time of water, decrease in transparency and low-light climate, together with mass developed aquatic macrophyte vegetation in the whole inundated floodplain were unfavorable conditions for growth and proliferation of cyanobacteria. With the establishment of the flood regime characterized by long-lasting periods without flooding, in-lake processes prevailed leading to cyanobacterial bloom. The most successful were filamentous non-N-fixing cyanobacteria tolerant to mixed and low-light conditions (Planktothrix and Limnothrix) and invasive species Cylindrospermopsis raciborskii. Their massive development led to the establishment of a phytoplankton steady state. All our results demonstrate that the altered intensity and frequency of flood events will have pronounced effects on the appearance of cyanobacterial blooms and generally on alternative stable states in the floodplain. Relating to this, management objectives should be focused on qualifications of changes in hydrology and projecting those effects for potential floodplain restoration.     

Linking Flow Regime,Floodplain Lake Connectivity and Fish Catch in a Large River-Floodplain System,the Volga–Akhtuba Floodplain (Russian Federation)

River-floodplain systems are amongst the most productive—but often severely impacted—aquatic systems worldwide. We explored the ecological response of fish to flow regime in a large river-floodplain system by studying the relationships between (1) discharge and inundated floodplain area, with a focus on spatial and temporal patterns in floodplain lake connectivity, and (2) flood volume and fisheries catch. Our results demonstrate a non-linear relationship between discharge and floodplain inundation with considerable hysteresis due to differences in inundation and drainage rate. Inundation extent was mostly determined by flood volume, not peak discharge. We found that the more isolated lakes (that is, lakes with a shorter connection duration to the river) are located at higher local elevation and at larger hydrological distance from the main rivers: geographical distance to the river appears a poor predictor of lake isolation. Although year-to-year fish catches in the floodplain were significantly larger with larger flood volumes in the floodplain, they were not in the main river, suggesting that mechanisms that increase catch, such as increased floodplain access or increased somatic growth, are stimulated by flooding in the floodplain, but not in the river. Fish species that profit from flooding belong to different feeding guilds, suggesting that all trophic levels may benefit from flooding. We found indications that the ecological functioning of floodplains is not limited to its temporary availability as habitat. Refugia can be present within the floodplain itself, which should be considered in the management of large rivers and their floodplain.     

Structure of aquatic assemblages of Coleoptera and Heteroptera in relation to habitat type and flood dynamic structure

The abundance and taxonomic composition of the aquatic insect fauna were investigated, with focus on adult water bugs, water beetles and water scavenger beetles (Heteroptera: Nepomorpha and Gerromorpha, and Coleoptera: Hydradephaga and Hydrophilidae) in two different freshwater habitats: (1) a periodically flooded area of the Special Zoological Reserve in Kopa?ki rit Nature Park, Croatia; and (2) melioration canals in the wider area of the Nature Park during 2005. Aquatic insects are generally abundant in various water systems, including floodplains that are exposed to water level fluctuations. Our aims were (1) to determine abundance and species richness in relation to habitat type; (2) to determine the influence of high flood peaks and oscillations during high water levels on the diversity of aquatic Coleoptera and Heteroptera in the flooded area; and (3) to identify characteristic species associated with each habitat type. We collected 71 species; 41 were captured at canals and 64 at the flooded area. Diversity of the two habitat types varied depending on the months but there were remarkable differences in species pool and their abundance. Both high and low water levels as well as oscillations during high water levels had major influence on species assemblages at the flooded area. Diversity of aquatic Coleoptera and Heteroptera was higher when the water level decreased after high water level peaks. A total of 11 species fulfilled the criteria for specificity and were sufficiently abundant to be suitable species characteristic for these two habitats. A higher proportion of significant characteristic species was present in the flooded area than in the canals. The presence of two Red List species of water beetles (Graphoderus bilineatus De Geer, 1774 and Berosus geminus Reiche et Saulcy, 1856) and nine significant characteristic species at the flooded area clearly indicates that the contribution of floodplains in maintaining freshwater biodiversity is not only important regionally but also at the international level. Also, our results suggest that the power of high water levels is an important factor that can be used in analyses on aquatic Coleoptera and Heteroptera assemblages, showing the uniqueness of large floodplain areas.     

Thermal Heterogeneity in River Floodplains

River floodplains are composed of a shifting mosaic of aquatic and terrestrial habitats. Each habitat type exhibits distinct environmental and ecological properties. Temperature is a key property driving ecological processes and controlling the composition and distribution of biota. However, given the size and complexity of floodplains, ground surveys based on point measurements are spatially limited. In this study, we applied thermal infrared (IR) imagery to quantify surface temperature patterns at 12–15 min intervals over 24 h cycles in two near-natural Alpine river floodplains (Roseg, Tagliamento). Furthermore, vertical temperature distribution was measured at 3–5 min intervals in unsaturated gravel sediment deposits (at 1 cm distances; 0–29 cm depth). Each habitat type exhibited a distinct thermal signature creating a complex thermal mosaic. The diel temperature pulse and maximum daily temperature were the main thermal components that differentiated habitat types. In both floodplains, exposed gravel sediments exhibited the highest diel pulse (up to 23°C), whereas in aquatic habitats the pulse was as low as 11°C (main channel in the Roseg floodplain). In the unsaturated gravel sediment deposits, the maximum diel kinetic temperature pulse ranged from 40.4°C (sediment surface) to 2.7°C (29 cm sediment depth). Vertically, the spatiotemporal variation of temperature was about as high as horizontally across the entire floodplain surface. This study emphasized that remotely sensed thermal IR imagery provides a powerful non-invasive method to quantitatively assess thermal heterogeneity of complex aquatic and terrestrial ecosystems at a resolution required to understand ecosystem processes and the distribution of biota.     

The Significance of Meander Restoration for the Hydrogeomorphology and Recovery of Wetland Organisms in the Kushiro River,a Lowland River in Japan

The meanders and floodplains of the Kushiro River were restored in March 2011. A 1.6‐km stretch of the straightened main channel was remeandered by reconnecting the cutoff former channel and backfilling the straightened reach, and a 2.4‐km meander channel was restored. Additionally, flood levees were removed to promote river–floodplain interactions. There were four objectives of this restoration project: to restore the in‐stream habitat for native fish and invertebrates; to restore floodplain vegetation by increasing flooding frequency and raising the groundwater table; to reduce sediment and nutrient loads in the core wetland areas; to restore a river–floodplain landscape typical to naturally meandering rivers. In this project, not only the natural landscape of a meandering river but also its function was successfully restored. The monitoring results indicated that these goals were likely achieved in the short term after the restoration. The abundance and species richness of fish and invertebrate species increased, most likely because the lentic species that formerly inhabited the cutoff channel remained in the backwater and deep pools created in the restored reach. In addition, lotic species immigrated from neighboring reaches. The removal of flood levees and backfilling of the formerly straightened reach were very effective in increasing the frequency of flooding over the floodplains and raising the water table. The wetland vegetation recovered rapidly 1 year after the completion of the meander restoration. Sediment‐laden floodwater spread over the floodplain, and approximately 80–90% of the fine sediment carried by the water was filtered out by the wetland vegetation.     

Spatial and temporal variations in densities of small fishes across different temporary floodplain types of the lower Okavango Delta,Botswana

Small-sized and juvenile fishes as well as physicochemical water parameters in various primary, secondary and rarely flooded temporary floodplains of the Okavango Delta were assessed during the different hydrological phases of the 2009–2010 flooding season. Small fishes were sampled in the marginal zone of the floodplains using a throw-trap net and a suite of physicochemical properties were measured. Both physicochemical water quality parameters and densities of small fishes showed spatiotemporal variations across the temporary floodplain types. Turbidity, dissolved oxygen, conductivity, woody debris and chlorophyll a were all highest in rarely flooded floodplains compared to both primary and secondary floodplains (ANOVA, p < 0.05). Fish from 38 fish taxa belonging to 11 families were identified and classified during the study. The fish families Cichlidae (dominated by tilapias), Poeciliidae (Aplocheilichthys spp.) and Cyprinidae (Barbus spp.), were the most abundant across the study floodplains. The poeciliids were most abundant in frequently flooded primary and secondary floodplains, whereas juvenile cichlids dominated in rarely flooded floodplains. During high floods the rarely flooded portions of the delta function as important nursery habitats for juvenile cichlids, implying that a significant reduction in flooding may have negative effects on the delta's tilapia stocks due to reduced recruitment success.     

Impact of Juvenile Stocking Size on the Major Carp Production in a Minor Reservoir,Bibinagar, India

The present study deals with the impact of juvenile stocking sizes of the major carp on production in a minor reservoir, Bibinagar, Nalgonda, India. Reservoir surface area equaled over 23.8 ha and water from the reservoir was used for irrigation and fisheries. Four experiments were conducted for 4 years from 2000–2001 to 2003–2004. The experiments were planned in such a way that every year juvenile stocking size was held constant and subsequent fish production analyzed. During the first year’s experiment (2000–2001) a stocking size of 25–30 mm (fry) was maintained. Similarly 50–55 (advanced fry), 75–80 mm (fingerling) and 100–105 mm (advanced fingerling) were stocked during 2001–2002, 2002–2003 and 2003–2004 respectively. Uniform yearly stocking densities (2000/ha) were established in each experiment in the month of July and fish were harvested in June of the subsequent year. Major carp production was enhanced with larger stocking size. Yearly productions equaled 144.00, 231.48, 632.91 and 1005.03 kg/ha/year with the above stocking sizes. Variation in stocking sizes had significant (P<0.05) effects on fish production. The number of fish per kg production decreased gradually with increases in stocking size with an average value of 3.97. Reservoir production of catla was the most abundant species, followed by rohu, common carp, mrigal and grass carp. These results show that stocking size has a great impact on fish production and the stocking of advanced fingerlings will provide maximum carp production in minor reservoirs in India.     

Fish recruitment in a large,temperate floodplain: the importance of annual flooding,temperature and habitat complexity

1. Large river floodplains are considered key nursery habitats for many species of riverine fish. The lower Volga River floodplains (Russian Federation) are still relatively undisturbed, serving as a suitable model for studying the influence of flooding and temperature on fish recruitment in floodplain rivers. 2. We examined the interannual variability in recruitment success of young‐of‐the‐year (YOY) fish in the lower Volga floodplain in relation to flood pulse characteristics and rising water temperatures in the spring. We sampled four areas with different flooding regimes, in three consecutive years (2006–2008). 3. Extensive areas with a long duration of flooding accommodated high densities of young fish. This suggests that extended inundation improves the recruitment success of river fish. In areas with extensive flooding, the biomass of YOY of most fish species was about three times higher in 2006 and 2007 than in 2008. We hypothesise that low spring temperatures in 2008 may have caused this reduced recruitment and that a flood synchronised with rising temperature enhances recruitment success. 4. Extensive flooding was particularly favourable for species characterised by large body size, delayed maturation, high fecundity and low parental investment, such as pike Esox lucius, roach Rutilus rutilus and ide Leuciscus idus. Gibel carp Carassius gibelio, a species tolerant of high temperature and hypoxia, did particularly well in small waterbodies in the driest parts of the floodplain. 5. Structural characteristics of floodplain waterbodies explained much of YOY fish density. These species–environment associations varied from year to year, but some species such as common bream Abramis brama, roach and gibel carp showed consistent relationships with structural habitat characteristics in all years, despite large interannual fluctuations in flood pulse and spring temperature.     

Nitrate loss from a restored floodplain in the Lower Cosumnes River, California

Floodplain restoration has been advocated as a means to restore several ecological services associated with floodplains: water quality improvement, fish rearing habitat, wildlife habitat, flood control, and groundwater recharge. A history of agricultural encroachment on the lower Cosumnes River has resulted in extensive channelization and levee construction. In fall 1998, an experimental floodplain was established by breaching a levee in order to restore the connection between the main channel and its historic floodplain. In this study, we examined how effective this newly restored floodplain was at processing nitrate (NO ₃ ⁻ ) before reentering the main channel downstream. Two methods were used to examine nitrate loss. In December 2001, we determined denitrification potentials of the floodplain soils before seasonal flooding inundated the floodplain. Next, we conducted a series of field soil column (mesocosm) experiments from March to June 2002 to study NO ₃ ⁻ -N loss from the overlying water in both sandy and clayey soils and at three levels of NO ₃ ⁻ -N (ambient, +1 mg N l⁻¹, +5 mg N l⁻¹). In addition, we examined NO ₃ ⁻ -N loss from mesocosms with water only to determine if loss was related primarily to soil or water column processes. Denitrification potentials were highly variable ranging from 1.6 to 769 ng N₂O–N cm⁻³ h⁻¹. In addition, the denitrification potential was highly correlated with the amount of bioavailable carbon indicating that carbon was a limiting factor for denitrification. Nitrate-N loss rates from mesocosms ranged from 2.9 to 21.0 μg N l⁻¹ h⁻¹ over all treatments and all 3 months examined. Significant loss of NO ₃ ⁻ -N (60–93%) from the water only treatment only occurred in June when warmer temperatures and solar radiation most likely increased NO ₃ ⁻ -N uptake by phytoplankton. When scaled up, potential NO ₃ ⁻ -N loss from the restored floodplain represented 0.6–4.4% of the annual N load from the Lower Cosumnes River during a typical wet year and ~24% during a dry year. During dry water years, the effectiveness of the floodplain for reducing nitrate is limited by the amount of N supplied to the floodplain. Results from this study suggest that restored floodplains can be an effective NO ₃ ⁻ sink.     

Managing floodplain inundation for native fish: production dynamics of age-0 splittail (Pogonichthys macrolepidotus) in California’s Yolo Bypass

We used data gathered across seven hydrologically diverse years (1998–2004) from Yolo Bypass, the primary floodplain of the Sacramento River, California (U.S.A.), to examine how physical and hydrological characteristics of floodplain habitat influence spawning and patterns of occurrence and production of age-0 splittail (Pogonichthys macrolepidotus). We estimated that spawning in Yolo Bypass occurred from January to May but typically peaked in March in close association with the vernal equinox. Production of age-0 splittail varied significantly among years. Using an information-theoretic approach to evaluate several factors, the amount of inundated floodplain habitat available during the primary spawning and rearing period of January–June was the most important factor we examined in explaining annual production (81% of variance explained in a linear regression model). Peak emigration of age-0 splittail from Yolo Bypass varied temporally among years but always occurred when fish were 30–40 mm in length, suggesting an ontogenetic influence on floodplain emigration. Annual system-wide production of age-0 splittail derived from a separate long-term monitoring program exhibited an apparent positive relationship with annual production in Yolo Bypass, suggesting that site-specific floodplain-derived production within Yolo Bypass may be important at a regional scale. Our results support the flood pulse concept for cyprinids in regulated temperate river–floodplain systems and demonstrate floodplain inundation in regulated systems can be managed to benefit native fish.     

Short-term response of nutrients,carbon and planktonic microbial communities to floodplain wetland inundation

Environmental flows were released to the Macquarie Marshes (~210,000 ha) in north-west NSW of Australia between October and December 2005, inundating an estimated 24,600 ha of floodplain area. According to the flood pulse concept, the marsh floodplains would have stored large amounts of nutrients and carbon during dry antecedent conditions, which would be released into the overlaying flood water. Field studies were conducted in mid-December 2005 at two sites, one on open floodplain woodland with a sparse canopy of River Red Gum and ground cover dominated by saltbushes and the other on open floodplain with black roly-poly. At each site, nutrients, dissolved organic carbon (DOC), planktonic bacteria and phytoplankton were monitored daily for a 6-day period from the overlaying water of a floodplain inundated by the environmental water release. Those in mesocosms deployed in situ, containing marsh floodplain sediments that had been inundated artificially, were also monitored. The mesocosm results from both the sites showed that release of nitrogen was rapid, attaining mean concentrations of total nitrogen of 3.7–14.8 mg l⁻¹, followed by more gradual increases in total phosphorus (mean concentrations 0.6–0.8 mg l⁻¹) and DOC (26.1–50.2 mg l⁻¹) within the 6-day experiment; planktonic microbial communities developed concomitantly with the increasing concentrations of nutrients and DOC, attaining mean densities of (6.0–6.9) × 10⁶ cells ml⁻¹ of planktonic bacteria and (80.7–81.4) × 10³ cells ml⁻¹ of phytoplankton; and for each site the overall measured condition of the mesocosm tended to approach that of the Marshes, over the course of the 6-day experiment. The present study (both observational and experimental) demonstrates that the floodplain sediments in the Marshes, which have been exposed to dry antecedent conditions, release nutrients and carbon to the overlaying flood water following inundation. These resources are thought to have been stored during the dry antecedent phase in accord with the flood pulse concept. Based on the mesocosm experiment, the released nutrients and carbon are in turn most likely to be used by microbial components, such as bacteria and algae, which develop within days of inundation of the floodplain sediments. Thus, environmental flow release provides potential for floodplains to attain a series of ecological responses including initial release of inorganic nutrients and dissolved organic matter and increase in planktonic bacteria and phytoplankton.     

Winter fish community structures across floodplain backwaters in a drought year

We examined fish community structures in floodplain backwaters in the Kiso River, Central Japan, in an uncommon drought year. Nine floodplain backwaters with varying levels of flood inundation frequency were surveyed for winter fish communities, and measured for local environmental variables in summer and winter periods. The winter fish community was better represented, with less diverse and physiologically tolerant species in backwaters with a decreasing flood pulse frequency. The temporal frequency of hypoxic conditions (dissolved oxygen of <2 mg l^?1) and chlorophyll a in summer explained the heterogeneity of the fish community along the gradient of inundation frequency, while no variables measured in winter had an explanatory power. These results suggest that drought-induced, infrequent flood pulses in the autumn–winter period limited the dispersal opportunity of fishes, and thus maintained community structures selected by summer limiting factors until the survey took place in winter. Future comparisons of our data with those in non-drought years would provide important implications to the sustainable management of floodplains for wintering fishes against the increasing probability of future flow regime changes.     

Significance of a floodplain in the formation of fish communities in small rivers of Ryazan oblast

In total, 37 fish species from 30 genera and 10 families are registered in small rivers of the Ryazan oblast. In addition to other significant factors, the floodplain width is important for the formation of species structure of fish communities. In the rivers with small floodplains (less than 0.2 km in width), mainly small and short-lived fishes are represented in the catches. In the rivers or in their parts with wide floodplains (0.2–0.5 km), the dominant complex (in addition to short-lived fish species) includes roach Rutilus rutilus and/or perch Perca fluviatilis. In the rivers with wider floodplains (more than 0.5 km), long-lived phytophils represented by mainly the limnophilous group prevail.     

The dependence of fish community structure and dynamics on floodplain and riparian ecotone zone in Parana River,Brazil

Flood intensity was a main factor determining access to the riparian/floodplain ecotonal resources of the upper Parana River, Brazil, and in consequence determining yield of the dominant trophic groups, which are fish feeding on flooded grasslands and on riparian fruits and leaves. Prey and predator density both declined in years of low floods, and predators did not recover until the next year of high flood, more slowly than in the case prey fish, most probably due to differences in life cycle length. The life cycle of one of the most important species, Curimba (Prochilodus lineatus — Characidae) depends on access to the floodplain lagoons and return to the reservoir after maturation for intensive growth. The riparian/floodplain habitat complexity and connectivity has great significance for fish community dynamics and fisheries yield, and may serve a reason to create a national park in the last floodplain section of the upper Parana River.     

Spatial and temporal diversity of small shallow waters in river Lužnice floodplain

To propose a concept of their mutual diversity, twenty-nine permanent shallow floodplain pools and oxbows in the river Lužnice floodplain were analysed for area, depth, shape, flooding, and shading by terrestrial vegetation, and sampled in all seasons for their water chemistry, phytoplankton composition and biomass, and zooplankton composition. The sites are regularly flooded, eutrophic, and often shaded by surrounding vegetation. Cryptophyceae, Chrysophyceae and Euglenophyceae dominated the phytoplankton, while Cyanophytes were rare. Within the rich zooplankton assemblage (63 species), cladocerans and rotifers dominated. Correlation matrices and multivariate analyses indicated that shaded and relatively deeper sites had lower oxygen saturation and higher concentrations of PO₄–P and NH₄–N. Shade and relative depth correlated negatively with phytoplankton biomass and number of phytoplankton taxa, and positively with Cryptophytes and large cladocerans—thus indicating poor mixing, poor light availability and low fish pressure on herbivores. Decomposition of leaf litter increased oxygen consumption, while shade from terrestrial vegetation restricted photosynthesis and decreased oxygen production. Larger sites were more species-rich in phytoplankton and supported Euglenophyceae, green algae and rotifers.     

Stream–floodplain connectivity and fish assemblage diversity in the Champlain Valley, Vermont, U.S.A.

To evaluate the influence of main channel–floodplain connectivity on fish assemblage diversity in floodplains associated with streams and small rivers, fish assemblages and habitat characteristics were surveyed at 24 stream reaches in the Champlain Valley of Vermont, U.S.A. Fish assemblages differed markedly between the main channel and the floodplain. Fish assemblage diversity was greatest at reaches that exhibited high floodplain connectivity. Whereas certain species inhabited only main channels or floodplains, others utilized both main channel and floodplain habitats. Both floodplain fish α-diversity and γ-diversity of the entire stream corridor were positively correlated with connectivity between the main channel and its floodplain. Consistent with these results, species turnover (as measured by β-diversity) was negatively correlated with floodplain connectivity. Floodplains with waterbodies characterized by a wide range of water depths and turbidity levels exhibited high fish diversity. The results suggest that by separating rivers from their floodplains, incision and subsequent channel widening will have detrimental effects on multiple aspects of fish assemblage diversity across the stream–floodplain ecosystem.     

Fish community characteristics of the lower Gambia River floodplains: a study in the last major undisturbed West African river

1. The Gambia River is the last major West African river that has not been impounded. However, a hydroelectric dam is being constructed and substantial changes to the hydrology and ecology of the system are expected. 2. Little information is available on the impact of water impoundments in semi‐arid regions on downstream floodplain fish communities, due to the scarcity of pre‐intervention data. Because profound impacts on physical habitat, salinity and nutrient transport can occur downstream of such impoundments, a knowledge of the species‐habitat associations of biota such as fishes is necessary for understanding likely changes and how to limit them. 3. Fish were sampled using cast and hand nets along two transects on the floodplain, and with fyke nets in two ‘bolongs’ (creeks) from May to November 2005 and 2006 in the lower reaches of the Gambia River, close to the salt water front where ecological changes due to the construction of the dam are likely to be pronounced. 4. Greatest fish species richness was associated with low conductivity, low pH and deep water. Bolongs held greater species richness compared with other floodplain habitats, probably because they acted as conduits for fish moving on and off the floodplain. Species richness and catch biomass increased rapidly following the first rains and then declined. 5. Using a multivariate analysis, three main species groups were identified on the floodplain; one associated with deeper water, one with less brackish water and one with shallow, open water. Tilapia guineensis was the commonest species on the floodplains. 6. The floodplains provide nursery habitats as many fish captured were immature, particularly for species where adults are mainly encountered in the main channel. Several small‐sized floodplain specialists were also represented by a high proportion of mature individuals. 7. Impoundment is expected to reduce seasonal flooding of the floodplain in the lower Gambia River, downstream of the impoundment, resulting in reduced occurrence of aquatic habitats, especially bolongs, together with lower dissolved oxygen and increased salinity, leading to alteration of the floodplain fish communities, benefiting salt‐tolerant species, reducing overall species richness and probably reducing floodplain fish production.