Deformation Prediction of Landslide-Induced Highway Instability Using D-InSAR and CNN-LSTM: A Case Study in Zhouqu County, China

Geological disasters pose significant risks to highway infrastructure. This paper presents an enhanced predictive approach that integrates Differential Interferometric Synthetic Aperture Radar (D-InSAR) data with a Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) deep learning model. Using the Yahuokou landslide event in Zhouqu County, Gansu Province as a case study, this research evaluates the capability of the proposed model to forecast ground deformation affecting the G75 Expressway. Comparative analysis with traditional models such as ARIMA, SVR, and standalone LSTM demonstrates that the CNN-LSTM model achieves superior prediction accuracy, making it a reliable tool for landslide early warning systems. The research highlights the importance of combining satellite-based monitoring with intelligent predictive algorithms to enhance infrastructure resilience and disaster mitigation.