Coupling CFSv2 with ArcSWAT for seasonal hydrological forecasting in a Mediterranean basin

In the context of a rapidly changing world climate, where matters such as water availability and scarcity, energy demand, and adequacy of crop production grow into first-line priorities for private and government entities, the need for knowledge -in advance- and future planning with regard to water budget availability and distribution becomes more and more essential. Seasonal hydrological forecasting constitutes the key solution that informs actions regarding drinking water needs, irrigation planning, as well as management optimization of hydroelectric power schemes. The efficiency of hydrological forecasting vastly relies on the accuracy of meteorological forecasting (Sehgal et al. 2018; Ye et al. 2017). In fact, it could be stated that in relative studies the efficacy of the hydrological forecasting is determined more on the basis of the accuracy of the meteorological forecasting rather than the competence of the hydrological model -used- itself. In this study, the version 2 of the Climate Forecast System (CFSv2) is combined with the GIS version of the Soil and Water Assessment Tool (ArcSWAT) to provide mid-term hydrological predictions. More specifically, the CFSv2 is employed to forecast precipitation and temperature, two decisive meteorological parameters for hydrologic simulations. The forecasted variables are bias-corrected prior to their forcing to SWAT. The chain of the aforementioned processes is tested for a two-month period in Nestos River basin in northeastern Greece. It is a relatively large mountainous basin covering an area of approximately 840 km2 . A detailed description of Nestos River basin can be found in Kaffas et al. (2018). The results were validated with sparse discrete discharge measurements, at the outlet of the basin, and were deemed satisfactory.