Analysis of soil moisture retrieval and response to meteorological factors using AMSR-2

Soil moisture is an important component of the hydrologic cycle in terrestrial ecosystems and it is critical for predicting and understanding various hydrological processes, including changes in weather conditions, precipitation patterns, runoff generation and irrigation scheduling. Soil moisture is a function of the total effect of environmental factors. The Mongolia Plateau is an ideal area for studying the interaction between soil moisture and environmental factors, because of its arid and semi-arid location and its high ecological fragility and sensitivity to global climate change. Therefore, it was necessary to study the response of soil moisture to environmental factors, which was favorable to monitor and predict droughts, adjust agricultural production structures and improve regional eco-environment in the Mongolia Plateau. A soil moisture retrieval model for the Mongolia Plateau was built using microwave radiance transfer function and Qp model based on AMSR-2 brightness temperature and SPOT normalized difference vegetation index (NDVI) data. Soil moisture was retrieved, and the retrieval precision was verified during vegetation growth period from April to October 2013 in the Mongolia Plateau. Combination with TRMM 3B43 precipitation and air temperature data acquired by meteorological stations, the study explored response characteristics between soil moisture, meteorological factors and vegetation. The results showed that 1) the coefficient of determination (r) between retrieved and ground-based soil moisture was 0.680 6, with a root-mean square error (RMSE) of 0.031 6 cm³.cm^-3. The retrieval result was much better than soil moisture product data of JAXA (RMSE = 0.044 1 cm³.cm^-3). 2) The developed model had a high accuracy and was applicable in surface soil moisture estimation. The regression coefficient of the linear fit of the TRMM 3B43 precipitation measure (rainfall) was 0.859 8 and with a slope line of 0.941 5, which suggested that TRMM 3B43 data were applicable in the Mongolia Plateau. 3) Total precipitation, mean NDVI and soil moisture during the growing season decreased gradually from north to south and from northeast to southwest. In the arid region of the study area, soil moisture was significantly and positively correlated with temperature, followed by precipitation and vegetation. In the semi-arid region of the study area, vegetation was the key factor driving soil moisture, and the effects of temperature and precipitation on soil moisture showed seasonal variations. The response of soil moisture to the three factors was in the order of vegetation > precipitation > temperature in the semi-humid region of the study area. In conclusion, the response of soil moisture to both environmental factors and vegetation could provide scientific basis for constructing healthy regional eco-environments with reducing disasters risk.