Assessing the results of scenarios of climate and land use changes on the hydrology of an Italian catchment: modelling study

Hydrological models are recognized as valid scientific tools to study water quantity and quality and provide support for the integrated management and planning of water resources at different scales. In common with many catchments in the Mediterranean, the study catchment has many problems such as the increasing gap between water demand and supply, water quality deterioration, scarcity of available data, lack of measurements and specific information. The application of hydrological models to investigate hydrological processes in this type of catchments is of particular relevance for water planning strategies to address the possible impact of climate and land use changes on water resources. The distributed catchment scale model (DiCaSM) was selected to study the impact of climate and land use changes on the hydrological cycle and the water balance components in the Apulia region, southern Italy, specifically in the Candelaro catchment (1780 km2). The results obtained from this investigation proved the ability of DiCaSM to quantify the different components of the catchment water balance and to successfully simulate the stream flows. In addition, the model was run with the climate change scenarios for southern Italy, i.e. reduced winter rainfall by 5-10%, reduced summer rainfall by 15-20%, winter temperature rise by 1·25-1·5°C and summer temperature rise by 1·5-1·75°C. The results indicated that by 2050, groundwater recharge in the Candelaro catchment would decrease by 21-31% and stream flows by 16-23%. The model results also showed that the projected durum wheat yield up to 2050 is likely to decrease between 2.2% and 10·4% due to the future reduction in rainfall and increase in temperature. In the current study, the reliability of the DiCaSM was assessed when applied to the Candelaro catchment; those parameters that may cause uncertainty in model output were investigated using a generalized likelihood uncertainty estimation (GLUE) methodology. The results showed that DiCaSM provided a small level of uncertainty and subsequently, a higher confidence level.