Balancing sediment starvation and hydropower development: the case study of the Vjosa river

The functioning of a fluvial ecosystem is controlled by abiotic and biotic factors and their interactions in space and time. Specifically, sediment transport and connectivity are key factors for the functioning of fluvial eco-systems. Disturbances to these factors, like the construction of dams for hydropower production, are hence bound to cause changes in the river ecosystems. These trade-offs should be taken into consideration in multi-objective analysis. Modelling dam impacts on sediment connectivity requires a whole-network perspective, which is challenging to provide with common sediment transport models. CASCADE (CAtchment Sediment Connectivity And DElivery) (Schmitt et al., 2016) is a modelling framework that has been developed to tackle this problem. CASCADE pro- vides disaggregated information about the provenance and destination of sediment derived from a specific source such as to quantitatively describe sediment connectivity in river networks. In this work, we propose a new version of the CASCADE model, where 1) sediment supply from each source is described not by a single sediment size, but as a distribution of sediment classes and 2) calculation of transport capacity is based on the empirical Wilcock and Crowe equation. This new framework is applied to the Vjosa basin, a gravel bed river in south Albania. The Vjosa is one of the last untouched braided rivers in Europe with a notable ecological valuable to be protected. In the basin currently, there are plans for hydropower development that pose a serious threat to this valuable ecosystem. We applied CASCADE on the Vjosa to assess the impact on sediment transport due to alternative scenarios of hydropower development. We defined indicators for sediment connectivity alterations and hydropower production and adopted them in a multi-objective analysis to find optimal dam siting. Furthermore, given the data scarce context, we performed a sensitivity analysis about initial sediment distribution, which is unknown, and evaluated to robustness of the ranking about the planning scenarios in function of this model uncertainty. The results show that there are alternatives that guarantee around 50% of the maximum possible hydroelectric pro- duction without drastically altering the sediment transport or damaging the braided sections of the river. Usually, those portfolios rely on the construction of hydropower dams on the tributaries, that have lower impact on the over- all sediment connectivity. The sensitivity analysis shows that, while changes on the initial conditions significantly affect the estimation of sediment fluxes across the basin, the ranking of alternatives dam siting, and the shape and composition of the Pareto front do not change significantly. Planning solutions are robust in terms of uncertainty related to sediment connectivity assessment.