Towards improved spatio-temporal resolution soil moisture retrievals from the synergy of SMOS and MSG SEVIRI spaceborne observations

Abstract

Earth Observation (EO) technology is today at a maturity level that allows deriving operational estimates of Surface Soil Moisture (SSM) from a variety of sensors; yet, such products are provided at present at a coarse spatial and/or temporal resolution, which restricts their use in local or regional scale studies and practical applications. Herein, a methodology to derive SSM estimates from space at previously unattained spatio-temporal resolutions is proposed. The method is based on a variant of the >triangle> inversion technique leveraging on the strengths and synergies of SMOS (Soil Moisture and Ocean Salinity mission) microwave observations and geostationary optical/infrared data. The SSM retrieval technique allows for: i) enhancing the spatial resolution of SMOS SSM product estimates to 3 km spatial resolution, and, ii) providing a temporal average daytime SM product from the instantaneous fine-scale SSM estimates acquired every 15 min; the latter is allowing higher coverage in presence of clouds and representativeness (up to 96 estimates per day) in comparison to the instantaneous estimate at the time of satellite overpass.The proposed technique has been implemented to SMOS and MSG (Meteosat Second Generation) SEVIRI (Spinning Enhanced Visible and Infrared Imager) observations acquired over the Iberian Peninsula and Southern France during year 2011. SSM instantaneous estimates at the time of SMOS overpass and daytime-averaged SSM estimates have been obtained and evaluated separately against collocated in-situ measurements acquired from a total of 40 stations belonging to the REMEDHUS, VAS and SMOSMANIA permanent soil moisture measurement networks. Statistical agreement between compared datasets has been evaluated both at individual stations and considering the network average on the basis of several statistical terms computed including correlation, bias, root-mean-squared errors, slope and intercept of linear regression. Results showed that the proposed method not only preserves the quality of SMOS SSM at finer spatial scales, but also allows achieving higher temporal coverage and representativeness in daytime averages. The synergy of SMOS and SEVIRI provides a pathway to enhance water cycle EO capabilities taking full advantage of the new observational records of SSM and operational geostationary information.