Soil moisture is one of the biosphere's most essential climatic variables. However, its periodical monitoring at regional scales using in-situ measurements is complex. In this context, L-band microwaves observations from passive remote sensors appear as an important tool for the periodical estimation of soil moisture at regional scales. In this work, a semi-empirical model to estimate soil moisture values from L-band observation was calibrated and evaluated over several land cover classes in a heterogeneous study area in Chile. For this, 3 years (2010, 2011 and 2012) of brightness temperature data from SMOS, soil temperature and volumetric water content from ERA-Interim, and NDVI from MODIS were used. Results showed an increase in the average r(2) when a vegetation index was used in the calibration of the approach. These increases ranged from a 3% for Crops, to a 49% for Closed Shrublands. The ubRMSD showed a decrease in its value of up to 1% m(3)/m(3) for Woodlands, Open Shrublands and Woody Shrublands and of up to 2% m(3)/m(3) for Closed Shrublands.