Department Surface Waters - Research and Management

The current boom of dam construction at low latitudes endangers the integrity and function of major tropical river systems. A deeper understanding of the physical and chemical functioning of tropical reservoirs is essential to mitigate dam-related impacts. However, the development of predictive tools is hampered by a lack of consistent data on physical mixing and biogeochemistry of tropical reservoirs. In this study, we focus on Lake Kariba (Southern Africa), the largest artificial lake in the world by volume. Kariba Dam forms a transboundary reservoir between Zambia and Zimbabwe, and therefore its management represents a socio-politically sensitive issue because the Kariba Dam operation completely changed the downstream hydrological regime. Although Lake Kariba represents a unique and scientifically interesting case study, there is no consistent dataset documenting its physical and chemical behaviour over time. This limits the scope for quantitative studies of this reservoir and its downstream impacts. To address this research gap, we aggregated a consistent database of in situ measurements of temperature and oxygen depth profiles for the entire 60 years of Lake Kariba’s lifetime and performed a detailed statistical analysis of the thermal and oxygen regime of the artificial lake to classify the different behaviours of the lake’s sub-basins. We demonstrate that the seasonal stratification strongly depends on the depth of the water column and on the distance from the lake inflow. Satellite data confirm these spatiotemporal variations in surface temperature, and reveal a consistent longitudinal warming trend of the lake surface water temperature of about 1.5°C from the inflow to the dam. Finally, our results suggest that the stratification dynamics of the lacustrine sub-basins have the potential to alter the downstream Zambezi water quality. Future research should focus on assessing such alterations and developing strategies to mitigate them.

This study deals with the impact of spatio-temporal heterogeneities on the assessment of lake ecological status according to the European water framework directive (WFD). A method, based on three-dimensional coupled hydrodynamic and ecological modeling, is presented to assess the variability of lake ecological status, and to locate the most representative sampling station of Lake Geneva (France/Switzerland). Five variables used in the lake ecological status evaluation were simulated by using the free software Delft3D. The numerical simulation results showed that the simulated ecological status based on chlorophyll a and total phosphorus concentrations measured at the regulatory monitoring station depend on the choice of the sampling date. Results also indicated a strong spatial heterogeneity in ecological status that varies from "poor" to "good" along an East-West gradient. Finally, the numerical simulation results showed that the most representative point of a mean theoretical ecological quality for Lake Geneva would be located in the center of the upper basin, close to the historical sampling station.

Cette étude traite de l'impact des hétérogénéités spatio-temporelles sur l'évaluation de l'état écologique des lacs conformément à la directive-cadre européenne sur l'eau (DCE). Une méthode, basée sur la modélisation tridimensionnelle couplée hydrodynamique et écologique, est présentée pour évaluer la variabilité de l'état écologique et pour localiser la station d'échantillonnage la plus représentative du lac Léman. Cinq variables utilisées dans l'évaluation de l'état écologique ont été simulées en utilisant le logiciel libre Delft3D. Les résultats des simulations numériques ont montré que l'état écologique simulé basé sur les concentrations en chlorophylle a et en phosphore total mesurées à la station de suivi réglementaire dépend du choix de la date d'échantillonnage. Les résultats ont aussi indiqué des fortes hétérogénéités spatiales de l'état écologique allant de «médiocre» à «bon» le long d'un gradient est-ouest. Finalement, les résultats de simulations numériques ont montré que le point le plus représentatif d'un état écologique théorique moyen du lac Léman serait situé au centre du grand bassin, proche de la station d'échantillonnage historique.