Landscape Soil Moisture Analysis with Machine Learning using Weather Parameters.

The correlation coefficient between soil moisture and weather variables such air temperature, precipitation, rainfall, and surface temperature is proposed by this study. The region of mid Asia was subjected to the model. The findings demonstrated a strong inverse relationship between soil moisture and surface temperature, a corresponding inverse relationship between soil moisture and air temperature, and a strong inverse relationship between soil moisture and precipitation. The soil moisture derived from NASA's LPRM_AMSR2 data was observed to have a correlation coefficient of 0.45 with gridded rainfall, and -0.59 with air temperature and ground temperature. The Aphrodite's Water Resource was employed to conduct this analysis, and the resulting meteorological factors taken into consideration were from there. Applying various weather parameters as the corresponding input to make a prediction about soil moisture and then measuring the accuracy with an LSTM model. [ABSTRACT FROM AUTHOR]