Modelling the hydrologic behavior of the upper Tiber river basin under climate change scenarios

Abstract

Quantification of the various components of hydrological processes in a watershed remains a challenging topic as the hydrological system is altered by internal and external drivers. This study investigates hydrological behavior of the Upper Tiber River Basin (UTRB) in central Italy under climate change scenarios. It addresses the response of the watershed by evaluating the relative changes in magnitudes of various hydrological processes using downscaled climate data in a watershed model. The observed data from regional concerned offices of the basin and the readily available global and regional climate model data were collected for analysis. First, the reliability of the precipitation and temperature data through two statistical downscaling methods are evaluated. The Statistical Downscaling Model (SDSM) and the Long Ashton Research Station Weather Generator (LARSWG) are used to downscale the HadCM3 GCM predictions of the A2 and B2 scenarios for the Chiascio sub-basin in the UTRB. The results show that the downscaling methods used have different performance to reproduce the precipitation patter but they agree on both minimum and maximum temperature. However it is difficult to choose which method is of downscaling as both have their own limitations and associated uncertainties Second, a physically-based watershed simulation model called Soil Water Assessment Tool (SWAT) is used to understand the hydrologic behavior of the basin. The model is successfully calibrated for the period of 1963-1970 and validated for the period of 1971-1978 using observed weather and flow data. A total of eighteen hydrologic parameters are evaluated and the model showed high relative sensitivity to groundwater flow driven parameters than the surface flow driven parameters. The objective function for model evaluation statistics showed coefficient of determination (R2) from 0.68 to 0.81 and Nash Sutcliffe Efficiency (ENS) between 0.51 and 0.8 for the validation period. Based on the calibrated parameters the model is proved to be capable of predicting impact of climate changes in water resources planning and management. Third, the calibrated watershed model is used to evaluate the response of the basin under different climate change scenarios. Bias correction to three regional climate model (RCM) from the PRUDENCE including RegCM, RCAO, and PROMES outputs are applied. The current (1961-1990) and future (2071-2100) time periods were considered for the analysis. The result indicated that there will be significant decrease in the hydrologic components and water yields of the basin which resembles some previous findings in the basin. The result of the present study is different from the others in that most of them used indices based on observation. However, future work on the uncertainty issues related to projected climate variables is highly recommended. Despite some common limitations, this study is relevant to assist the development of climate change adaptation in the study.