A spatially continuous 9-km soil moisture product based on SMAP data by using TsSMNet model

Satellite-based surface soil moisture (SSM) products often contain spatial gaps due to vegetation cover,complex surface conditions, or sensor limitations. This study presents a residual autoencoder model named TsSMNet, which combines multi-source remote sensing inputs with statistical features derived from SSM time series to reconstruct gap-free SSM estimates. The model incorporates one-dimensional convolutional layers for efficient spatial feature extraction and parameter reduction. Based on the SMAP product, TsSMNet was used to generate seamless 9-km SSM data over China for the period from January 2016 to December 2022.