Inhomogeneous anisotropic analysis of the available water content of the upper soil layer according to ground-based and Remote Sensing on the territory of Russia
The Hydrometeorological Center of Russia receives 2 agrometeorological information from about 950 stations one time 3 per ten days and the remote sensing Advanced Scatterometer 4 (ASCAT) data from three Meteorological Operational (MetOp) 5 satellites. We suggest a combined objective analysis (OA) of the 6 available water content based on the available water content 7 measurements at agrometeorological stations and on remote 8 sensing data. The new version of OA is constructed using two 9 neural networks and the backpropagation of error to learn it 10 simultaneously. The first neural network is used to convert the 11 ASCAT data into the available water content values, and the 12 second network is used to estimate the inhomogeneities of soil 13 moisture fields. We use the optimal interpolation (OI) method 14 for assimilation of the ground-based data. In the new version, 15 we evaluate the correlation functions (CFs) of inhomogeneous 16 non-Gaussian fields, not from sample statistics but from machine 17 learning methods. The method takes into account the combin18 ing of various datasets: ASCAT data, Food and Agriculture 19 Organization (FAO) soil types, European Space Agency (ESA) 20 GlobCover, and National Center for Atmospheric Research 21 (NCAR) climate data.