Delineating Ground Subsidence and Flood Risk in Southeast Texas (SETX) Using PS-InSAR and Deep Learning

Authors

Arip Syaripudin Nur
Jinwoo An, The University of Texas Rio Grande ValleyFollow
Yong Je Kim

Document Type

Conference Proceeding

Publication Date

8-2024

Abstract

This study investigates ground subsidence and its impact on flood risk in Southeast Texas (SETX), a region particularly susceptible to inundation due to rising sea levels and subsidence. Understanding subsidence patterns is crucial for geotechnical risk management and ground-use planning. Employing Persistent Scatterer Interferometry (PS-InSAR) with Sentinel-1 SAR data (2020-2023), the study delineates the spatiotemporal distribution of subsidence and its influence on flooding. PS-InSAR, wellsuited for urban areas, captures stable scatterers detected by SAR satellites, enabling detailed subsidence analysis. Time series analysis reveals subsidence evolution and its implications for flood risk management. Validation utilizes strategically placed GPS station data. Furthermore, susceptibility mapping is conducted using Geographic Information Systems (GIS) and deep learning models. InSAR measurements serve as the foundation for generating ground subsidence susceptibility maps. Contributing factors are systematically analyzed for their correlation with subsidence occurrences using methods like the frequency ratio and multicollinearity tests. A deep learning algorithm, particularly Convolutional Neural Network (CNN), was innovatively incorporated to generate susceptibility results. Accuracy assessments are ensured through robust metrics like Root Mean Square Error (RMSE) and area under the receiver operating characteristic (ROC) curve. This comprehensive study offers valuable insights for mitigating subsidence, strengthening flood warning systems, and implementing preventive measures in SETX. The findings contribute significantly to the field of geotechnical engineering, safeguarding coastal communities and infrastructure from escalating threats of sea level rise and subsidence.

Comments

Copyright the Authors.

Recommended Citation

Nur, Arip Syaripudin, Jinwoo An, and Yong Je Kim. 2024. “Delineating Ground Subsidence and Flood Risk in Southeast Texas (SETX) Using PS-InSAR and Deep Learning.” In The 2024 World Congress on Advances in Civil, Environmental, & Materials Research (ACEM24) Proceedings. Seoul, Korea.

Publication Title

The 2024 World Congress on Advances in Civil, Environmental, & Materials Research (ACEM24) Proceedings