Roughness parameterization for soil moisture optimum retrieval over the Valencia anchor station

International audience; The Valencia Anchor Station (VAS) is a validation site of SMOS (Soil Moisture and Ocean Salinity) since 2010. For this purpose, different in situ measurements are performed there. This paper will focus on the parameterization of roughness, a key topic for the SMOS and SMAP (Soil Moisture Active Passive) missions. The area covered by the VAS is quite homogeneous, a surface of 50 x 50 km2 mainly composed by vineyards (65%) and other Mediterranean vegetation (30%), which is very stable along the year. The vegetative period of the vineyards lasts approximately from April to October. MELBEX-III is a station inside the area of the VAS, where among other equipments (meteorological station, thermistors, etc.), there is an ELBARA-II Lband radiometer performing multi-angular measurements over a vineyard. There are also two Delta-T ML2x soil moisture probes, one measuring between two rows of vines and the other one next to a vine stump. In the year 2013, measurements of roughness were done regularly by means of a needle board before and after all the agricultural practices, and before and after rainfall events which took place in MELBEX-III. These measurements provide the standard deviation of heights (SD) and correlation length (LC), which can be used to compute the input roughness parameters for the model L-MEB (L-band Microwave Emission of the Biosphere). L-MEB is the base of the SMOS algorithm for soil moisture (SM) retrieval. Using the ELBARA-II multi-angular brightness temperatures and testing many different parameterizations of roughness, SM has been retrieved at 6 am and 6 pm every day in 2013 and then compared to the in situ data (Theta Probe measurements). Based on this analysis, we evaluated and discussed which roughness parameterization provides the best retrieval of SM in terms of several statistical criteria: RMSE, BIAS and correlation coefficient.