Litcius/Paper detail

Sentinel-1 Backscatter Assimilation Using Support Vector Regression or the Water Cloud Model at European Soil Moisture Sites

Dominik Rains, Hans Lievens, Gabriëlle De Lannoy, Matthew F. McCabe, Richard de Jeu, Diego G. Miralles

2021IEEE Geoscience and Remote Sensing Letters27 citationsDOIOpen Access PDF

Abstract

Sentinel-1 backscatter observations were assimilated into the Global Land Evaporation Amsterdam Model (GLEAM) using an ensemble Kalman filter. As a forward operator, which is required to simulate backscatter from soil moisture and leaf area index (LAI), we evaluated both the traditional water cloud model (WCM) and the support vector regression (SVR). With SVR, a closer fit between backscatter observations and simulations was achieved. The impact on the correlation between modeled and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in situ</i> soil moisture measurements was similar when assimilating the Sentinel data using WCM ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\Delta R = +0.037$ </tex-math></inline-formula> ) or SVR ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\Delta R = +0.025$ </tex-math></inline-formula> ).

Topics & Concepts

Backscatter (email)Environmental scienceCloud computingRemote sensingSupport vector machineData assimilationMoistureRegression analysisAssimilation (phonology)Water contentMeteorologyGeologyComputer scienceGeographyMachine learningGeotechnical engineeringPhilosophyOperating systemWirelessTelecommunicationsLinguisticsSoil Moisture and Remote SensingSoil and Unsaturated FlowPrecipitation Measurement and Analysis