Evaluating the robustness of conceptual rainfall-runoff models under climate variability in northern Tunisia
Résumé
Climate change is likely to have significant impacts on runoff in the Southern rim of the Mediterranean basin that already suffers from scarcity of water resources. To investigate how hydro-climatic conditions could evolve in this region, not only future projections of climate are necessary but also robust rainfall-runoff models that are able to be fairly reliable under changing climate conditions. This study thus aims at assessing the robustness of two conceptual rainfall-runoff models (GR4j and IHACRES) under long-term climate variability on eight basins covering the main hydrographic characteristics in Northern Tunisia (High Medjerda, Zouaraâ, Ichkeul and Cap bon). The catchment areas are between 81 km² and 418 km². The streamflow regime of the basins can be considered as natural since these basins are located upstream from storage-dams and withdrawals. A 30-year common period (1970-2000) was considered to capture a large spread of hydrological conditions. The models were calibrated and validated using various goodness-of-fit criterions: NSE, KGE and Bias. Their efficiency was evaluated according to a 10-day time step using three differential split sample tests that aimed at assessing the models under various precipitation and PE conditions. Results showed that the two models were able to correctly reproduce runoff from the basins with a good agreement of the hydrograph shapes, of the high and low flows as well as the water volumes. However, the models prove a limited transferability of their parameters under contrasted climate conditions in almost all basins. Consequently, many parameter sets will have to be considered to represent the hydrological uncertainties associated in the assessment of possible climate change impact on runoff in the basins.