Groundwater recharge estimation in karst aquifers of southern Apennines (Italy) by integration of remotely sensed data

In many Italian regions, karst aquifers are main sources of drinking water and play a crucial role for socio-economic development of the territory. Hence, estimating groundwater recharge of these aquifers is a fundamental task for a proper management of water resources, also considering impacts of climate changes. In karst areas of the southern Apennines, the assessment of hydrological parameters needed for the estimation of groundwater recharge is a challenging issue, specifically for the spatial discontinuity of the rain and air temperature monitoring networks and variability of land use. In such a framework, the integration of terrestrial and remotely sensed data is a promising approach to limit these uncertainties. This research deals with the estimation of groundwater recharge for karst aquifers of southern Apennines by the application of remotely sensed data gathered by the MODIS satellite in the period 2000-2014 for assessing actual evapotranspiration (MODIS Eta). To assess uncertainties in the estimation of Eta affecting conventional methods based on empirical formulas, MODIS Eta values were compared with those calculated by Turc, Coutagne and Thornthwaite methods. In addition, annual rainfall time series of 266 rain gauge and 150 air temperature stations, recorded by regional meteorological networks in the period 2000-2014, were considered to reconstruct regional distributed models of ETa and groundwater recharge. Considering the MODIS Eta, a mean annual groundwater recharge of about 448 mm.year(-1 )was estimated for karst aquifers of southern Apennines Instead, by the Turc, Coutagne and Thornthwaite formulas, ETa mean annual values of 494, 533 and 437 mm.year(-1) were estimated respectively. The obtained results open a new perspective in the assessment of actual evapotranspiration and groundwater recharge of karst aquifer at the regional and mean annual scales allowing a reduction of uncertainties and achieving a spatial resolution greater than that of the existing meteorological networks.