Abstract
Half of the world’s land surface is covered by about 300 transboundary river basins. In many of these areas, different views on infrastructure development and management tend to result in conflicting dynamics between riparian countries, even when international water agreements are signed. Because of the natural power asymmetry between upstream and downstream countries, the socio-economic externalities of these power struggles are often sustained by the riparian communities located in the most downstream areas. Such dynamics are typically compounded by the lack of transparency on how major infrastructures are operated. In most basins, for example, there are no shared platforms providing a detailed accounting of the amount of water stored and released by large dams. An opportunity to address the problem stands in satellite observations, which provide a means to monitor the actual state of rivers and big infrastructure. Here, we turn our attention to the Mekong River Basin—the largest transboundary river basin in Southeast Asia—where the situation of power asymmetry is emblematic of the issues faced by many transboundary basins around the world. We look, in particular, at three problems. In the first one, we leverage satellite observations to infer dam storage variations and operating rules. Specifically, we use area-storage curves (derived from a digital elevation model) and time series of reservoir water surface area, which we estimate from Landsat satellite images through a novel algorithm that removes the effects of clouds and other disturbances. In the second problem, we improve the reliability of macro-scale hydrological models by making use of the inferred reservoir operations—a highly important, but often missing, element in river basin models. we couple our hydrological model with a hydraulic model used to infer discharge time series from satellite altimetry data. With the aid of Global Sensitivity Analysis, we propose an approach to avoid the pitfall occurring when co-calibrating the two models. Finally, using the output of the two first parts (reservoir operation patterns and calibrated hydrological model), we assess the impacts on the downstream areas of the impoundment of upstream reservoirs and also use a numerical framework to devise better filling strategies for those reservoirs. The data and approaches developed in this thesis will help enrich the existing tools for water resources monitoring and management in transboundary river basins.