A comparative assessment of hydrologic models of varying complexity applied to a semi-arid region (Sardinia, Italy) for climate change studies

Assessing the hydrologic impacts of climate change is of great importance in the Mediterranean basins, which are heavily sensitive to climate variability, with significant impacts on water resources and hydrologic extremes. Modeling such complex systems to manage water resources as well as to predict hydrologic extremes is a difficult task. The overall aim of the work described in this thesis is to bring a contribution in developing a modeling approach that allows evaluation of local hydrologic impacts of climate changes in two Mediterranean catchments located in Sardinia. This contribution revolves around two main themes: understanding how physical representation of hydrologic models can affect hydrologic impact assessment under climate change on a semi-arid basin of the Mediterranean region, the Rio Mannu catchment, and demonstrating how advanced hydrologic modeling can help in defining adaptation measures in a complex water system, the Flumendosa basin, under climate change. The work to achieve the general objective is elaborated into three stages. The effects of climate change are evaluated on the Rio Mannu catchment through comparison of the results from five hydrologic models, CATchment HYdrology (CATHY), Soil and Water Assessment Tool (SWAT), TIN-based Real time Integrated Basin Simulator (tRIBS), TOPographic Kinematic APproximation and Integration-eXtended (TOPKAPI-X), and Water flow and balance Simulation Model (WASIM), and using as atmospheric input outputs of four climate global (GCM) and regional (RCM) model combinations. In order to evaluate uncertainties, a recently proposed metric is used: climate and hydrologic models results are compared in terms of agreement with each other in reference and future periods using Pearson correlation values and Duveiller bias. Notwithstanding some differences, overall the five hydrologic models show good agreement, and they respond similarly to the reduced precipitation and increased temperatures predicted by the climate models, lending strong support to a future scenario of increased water shortages for this region of the Mediterranean, with negative consequences especially for the agricultural sector. Detailed comparison of the responses obtained with three physically based hydrologic models, but to varying degrees as regards physical processes and terrain features representation – CATHY, tRIBS, and TOPKAPI-X – on the same catchment is carried out, with the aim to test the transferability of parameters between the three hydrologic models, focusing in particular on the calibration and validation difficulties. While the three hydrologic models responded similarly during the calibration year, significant differences were found for the drier validation period for the CATHY model, which produced very low streamflow. To obtain satisfactory results for the CATHY model, an hypothesis of soil crusting was assumed and the first soil layer was modeled with a lower saturated hydraulic conductivity. Finally, the TOPKAPI-X model is applied on a large Sardinian basin prone to extreme flood events, the Flumendosa basin, to assess the hydrologic impact of climate change at much larger scale. The model responds with decreasing value of discharge, soil water content, and actual evapotranspiration to the reduced precipitation and increased temperature predicted by the climate models, lending strong support to a future scenario of increased water shortages also in this basin of the Mediterranean region.