A Method to Estimate Surface Soil Moisture and Map the Irrigated Cropland Area Using Sentinel-1 and Sentinel-2 Data

Considering variations in surface soil moisture (SSM) is essential improving crop yield and irrigation scheduling. Today, most remotely sensed products have difficulties resolving signals at the plot scale. This study aims to use Sentinel-1 radar backscatter Sentinel-2 multispectral imagery estimate SSM high spatial (10 m) temporal resolution (at least 5 days) over an agricultural domain. Three supervised machine learning algorithms, multilayer perceptron (MLP), a convolutional neural network (CNN), linear regression models, were trained changes based on variation reflectance five different crops. Results showed that CNN best algorithm as it understands relations better represents two-dimensional images. Estimated values for agreement with in-situ measurements regardless of type, RMSE=0.0292 (cm3/cm3) R2=0.92 derived RMSE=0.0317 R2=0.84 data. Moreover, time series estimated (SSM-S1), (SSM-S2), from SMAP-Sentinel1 was compared. The developed data significantly higher mean state irrigated agriculture relative rainfed cropland area during season. multiple comparisons (fisher LSD) tested found these two groups are (pvalue=0.035 95% confidence interval). Therefore, by employing maximum likelihood classification data, we managed map agriculture. overall accuracy this unsupervised 77%, kappa coefficient 65%.