 Rapid urban development in China has caused land subsidence, which could pose a potential threat to sustainable urban development. Monitoring and predicting land subsidence over large areas is important to address these issues. We selected Hunan Province as our study area and applied small baseline subset interferometric synthetic aperture radar, technology to obtain land deformation information from November 2019 to February 2022 with 364 multi-track Sentinel-1A satellite images. Traditional time series deep learning models have limitations when it comes to extracting a sequence of information that is too long or extracting the feature information between the influence factor and the land subsidence. Therefore, we developed a long short-term memory temporal convolutional network, LSTMTCN, deep learning model to predict land subsidence and explore the influence of environmental factors such as the volumetric soil water layer and monthly precipitation on land subsidence in this study. Leveling data was used to verify the effectiveness of SPS in SAR in land subsidence monitoring. The result of this article was authored by Hongliangua, Yonghao Yuan, Jin Yongwang, and others. We are article.tv. Links in the description below.