A satellite-based approach to estimating spatially distributed groundwater recharge rates in a tropical wet sedimentary region despite cloudy conditions

Groundwater recharge (GWR) is one of the most challenging water fluxes to estimate, as it relies on observed data that are often limited in many developing countries. This study developed an innovative water budget method using satellite products for estimating the spatially distributed GWR at monthly and annual scales in tropical wet sedimentary regions despite cloudy conditions. The distinctive features proposed in this study include the capacity to address 1) evapotranspiration estimations in tropical wet regions frequently overlaid by substantial cloud cover; and 2) seasonal root-zone water storage estimations in sedimentary regions prone to monthly variations. The method also utilises satellite-based information of the precipitation and surface runoff. The GWR was estimated and validated for the hydrologically contrasting years 2016 and 2017 over a tropical wet sedimentary region located in North-eastern Brazil, which has substantial potential for groundwater abstraction. This study showed that applying a cloud-cleaning procedure based on monthly compositions of biophysical data enables the production of a reasonable proxy for evapotranspiration able to estimate groundwater by the water budget method. The resulting GWR rates were 219 (2016) and 302 (2017) mm yr(-1), showing good correlations (CC = 0.68 to 0.83) and slight underestimations (PBIAS =-13 to-9%) when compared with the referenced estimates obtained by the water table fluctuation method for 23 monitoring wells. Sensitivity analysis shows that water storage changes account for +19% to-22% of our monthly evaluation. The satellite-based approach consistently demonstrated that the consideration of cloud-cleaned evapotranspiration and root-zone soil water storage changes are essential for a proper estimation of spatially distributed GWR in tropical wet sedimentary regions because of their weather seasonality and cloudy conditions.