Abstract :
[en] Lake Edward is one of the East African Rift lakes (surface 2325 km2, maximum depth 112 m), is located on the border between the Democratic Republic of Congo (DRC) and Uganda. It is connected to the smaller lake George (surface 250 km2, maximum depth 3m) through Kazinga Channel, and is part of the Nile watershed. On both sides of the border, Lake Edward is bordered by two major natural reserves, the Virunga National Park, Queen Elisabeth National Park. A collapse of the fisheries of Lake Edward was observed from the 1960’s to 1980’s that matched the reduction of the numbers of hippos observed from the 1960s and 1970s. The reason of this collapse is unknown, nor if fisheries continued to decline until present. The aim of HIPE was to test the causal relationship between the recent environmental changes and the drastic reduction of fisheries productivity. Our working hypothesis was that several environmental pressures in the watershed of Lake Edward have disrupted the biogeochemical, structural and functional links between the terrestrial and aquatic ecosystems, leading to a collapse of the main ecosystem service provided by Lake Edward. The fishing effort has dramatically increased since the 1960s. This in combination with a relaxed enforcement, resulted in an overexploitation of the fish stock and a spread of the use of damaging fishing techniques around the lake. Other potential causes are changes in the hydrological cycle, which could have also modified the transfer of nutrients from the watershed to the lake, and climate variability affecting nutrient cycling and availability in the Great Lakes. Shallow sediment cores from Lake Edward show a shift in stable carbon isotope composition in the upper 10 cm of the core (approximately last 40 yrs) that could be explained by (i) an increase in the delivery of C3-derived organic matter to the lake, and/or (ii) a decrease in phytoplankton productivity, which would increase isotope fractionation in phytoplankton due to alleviation of diffusive CO2 limitation. Museum-archived fish specimens confirm a shift in the baseline δ13C values in the lake, corresponding to a lower productivity or a change in the composition of terrestrial organic matter input to L. Edward over the past century. A range of bivalve shells analyzed for C and O isotope ratios along their growth lines indicate these are excellent tools to differentiate habitats, but this compromises their use as archives for paleo-environmental reconstruction in the area since exact provenance of museum collection specimens is typically not sufficiently well documented. The data from tooth enamel from Hippopotamus canines show a long-term trend towards decreasing δ13C values of the Hippopotamus diet in the region, although showing a high inter-individual variability in δ13C values, most pronounced in more recent specimens, indicating that dietary differences exist on small spatial scales. As Hippopotamus are thought to be rather indiscriminate grazers, this should reflect local differences in relative C3-C4 cover. The oxygen stable isotope data of Hippopotamus were shown to be likely good proxies for their lacustrine versus riverine habitat use. The phytoplankton community in Lake Edward was largely dominated cyanobacteria (60% of the phytoplankton biomass), followed by diatoms (25% of the phytoplankton biomass), and by green algae, chrysophytes and cryptophytes. 248 taxa phytoplankton were identified with clear prevalence of cyanobacteria (104 taxa), from the morphological groups of coccal and filamentous species (non-heterocytous and heterocytous). Compared to historical data from the 1930’s it seems a shift in algal diversity (number of species) from diatoms to cyanobacteria. Lack of historical data does not allow to determine if a change in total phytoplankton biomass occurred. The primary production measured in Lake Edward during the three cruises was approximately 4 times lower than the single historical value (June 1960), but values seemed comparable in Lake George compared to those from the 1960’s. This is consistent with an overall decrease of present phosphorus loading compared to historical data (1990’s). Primary production rates were high, and in excess of community respiration, indicating a net autotrophic status leading to low values of dissolved CO2. Consequently, Lake George and the part of Lake Edward influenced by the outflow from Lake George acted as sinks for atmospheric CO2. This challenges the paradigm that inland waters are systematically sources of atmospheric CO2 that was derived from data mostly collected in boreal and temperate lakes. The sampled sites were in some cases very strong CH4 emitters to the atmosphere, although a very strong spatial variability was observed due to a combination of several factors such as the occurrence of anoxia in bottom waters and productivity, both related to depth (the shallowest systems being the most productive and devoid of bottom anoxia). Most systems were sinks of atmospheric N2O due to denitrification. Data on stable C, N and H isotope composition indicate that aquatic primary production is the dominant energy source at the basis of the foodweb, but the wide variability in consumer stable isotope signatures also indicates that a wider variety of sources is used, with contributions of terrestrial and possibly CH4-derived carbon. The basin of Lake Edward has a fish fauna that is typical for East Africa and consists of two components: a species-rich assemblage of Haplochromis species and a relatively species-poor assemblage of non-Haplochromis species. A total of 34 non-Haplochromis species belonging to 10 families and 21 genera are recorded from the system. These include six species of major importance to the fisheries. Three non-Haplochromis species are endemic to the system and two others have been introduced in the region. Six species are new records for the Lake Edward system. The Haplochromis species diversity was seriously underreported and went from 27 species known to 56, of which 13 new ones have been or are being described. We have currently no indication that species have disappeared from the system in the last few decades, though one catfish species known from historical records, H. longifillis, could not be retrieved during the recent expeditions. Substantial dietary niche overlaps were found between several commercially important fish species, while niche overlaps between Haplochromis species were small, which suggests that they display a high trophic differentiation in accordance with their specialised morphologies. We did not find indications that different stocks exist for the commercial species. However, we found that the populations in the lakes were separated from those inhabiting the rivers. We did find intraspecific differences in morphology and ecology between populations from Lake Edward and Lake George (including the Kazinga Channel) of some Haplochromis species. No signs of large-scale hybridization were found between the two important tilapias. Recent reports on the state of the fisheries of the Lake Edward system remain scarce. Lakes Edward and George are almost completely within protected areas. However, occupied land area, the number of inhabitants and the number of boats has risen substantially during the last decades. Total annual yields have been increasing in the last ten years, but Catch Per Unit Effort (CPUE) appears to be continuously decreasing. Tilapias, which were traditionally dominant in the catches, seem to have crashed in the eighties and their share in the catches keeps decreasing. Recent estimates showed poor stock status for most commercially important species with most stocks defined as either collapsed, recruitment impaired or overfished. Higher catches could be obtained under sustainable management. The immediate target of management should be rebuilding biomass to the biomass at maximum sustainable yield (Bmsy). A survey was conducted at the different landing sites along lakes Edward and George. This revealed that fishers were aware of the changes in the state of the fishery. Fishers attributed declining catches to excessive effort, fishing malpractices, destruction of breeding grounds, and pollution. Communities reported large-scale non-compliance with fisheries regulation, especially at Lake George.
