The aim of this study is to optimize an optical and radar data synergy for a regional mapping of soil water content, through experimental campaigns over agricultural fields in the Kairouan plain, in the central of Tunisia, during two agricultural seasons (2015-2016 and 2016-2017). Firstly, a radiative transfer model, Water Cloud Model is calibrated using NDVI index acquired from Sentinel-2 images to eliminate the vegetation effects on radar signal. The second research axe is to propose a semiempirical inversion method, using an inversion of the calibrated Water Cloud Model, and applied over bare soils and wheat fields (Irrigated and non irrigated fields). In this context, a mapping of surface moisture is proposed at 20 m spatial resolution with a six day repeat frequency for the entire studied site. This study reveals the high potential of Sentinel-1 data, when combined in synergy with optical images (Sentinel-2), for the recovery of moisture and vegetation characteristics. In this context, the proposed approach is validated ground truth measurements during the period (2015-2017). The maps produced from radar acquisitions are found to be reasonably correlated with the field measurements.