Assessment of Ground-Subsidence Risk in Coastal Megacities Using Multi-Source Satellite InSAR and GNSS Data Fusion

Project Abstract

Coastal megacities face escalating ground subsidence risks driven by rapid urbanization, groundwater extraction, sediment compaction, and sea-level rise, necessitating a robust, integrated monitoring framework. This study presents a comprehensive assessment of ground-subsidence risk by fusing multi-source satellite InSAR (Interferometric Synthetic Aperture Radar) data with GNSS (Global Navigation Satellite System) observations to achieve high-resolution, long-term spatiotemporal insights across dynamic coastal environments. We develop an end-to-end methodology that harmonizes SAR time series from Sentinel-1 and TerraSAR-X with high-precision GNSS networks, leveraging advanced time-series decomposition, atmospheric disturbance mitigation, and multi-sensor data fusion through Bayesian inference and machine learning techniques. The core objective is to quantify vertical ground movements at sub-centimeter precision, identify pervasive subsidence corridors, and attribute them to anthropogenic and natural drivers through integrated socio-hydrological analyses. The research innovates by addressing critical challenges in coastal subsidence monitoring (i) heterogeneous data quality and acquisition geometries across satellites; (ii) sparse GNSS coverage in rapidly urbanizing littoral zones; (iii) separating vertical displacement from surface deformation induced by atmospheric, hydrological, and thermal effects; and (iv) establishing a near-real-time risk assessment framework that can inform adaptation planning. We implement a multi-layered data fusion pipeline that first harmonizes InSAR line-of-sight measurements into reliable vertical deformation estimates using a probabilistic unwrap and tie-point network approach, followed by GNSS-based constraint assimilation to reduce ambiguities and enhance absolute height accuracy. Temporal downscaling and long-term trend extraction are performed with decomposed time-series models to distinguish episodic events (e.g., aquifer drawdown, rapid urban consolidation) from gradual subsidence trends. We integrate auxiliary datasets such as groundwater extraction records, land-use change, soil type maps, sediment compaction indices, and tide-gauge records to disentangle driving mechanisms and quantify contribution shares. The expected outcomes include (i) high-resolution, multi-decadal subsidence maps with quantifiable uncertainties for selected coastal megacities; (ii) a robust, transferable methodology for multi-source InSAR-GNSS fusion tailored to heterogeneous data environments; (iii) a catalog of subsidence hotspots and their dominant drivers, enabling targeted mitigation measures such as managed aquifer recharge, infrastructure redesign, and land-use regulation; and (iv) a decision-support prototype that computes risk metrics under various sea-level rise and urban growth scenarios. Validation will be performed through cross-comparison with GNSS time series, leveling data, and independent tide-gauge records, as well as through sensitivity tests to data sparsity and atmospheric errors. The study aims to advance the precision and reliability of coastal subsidence assessment, supporting proactive urban resilience planning and sustainable development in vulnerable coastal megacities.

Project Overview

What This Project Is About

---

A plain-language overview of how ground movement in coastal cities can be measured using satellite images and ground sensors, and how combining these methods helps predict subsidence risks and inform planning. The project looks at how land slowly sinks due to natural and human factors and how coastal areas are especially vulnerable to flooding and infrastructure damage.

---

---

The Problem It Addresses

---

Subsidence reduces the height of land relative to sea level, increasing flood risk, damaging buildings, and raising maintenance costs. Traditional methods cover small areas or rely on sparse data; this project explores using multiple satellite viewpoints and on-site measurements together to get a clearer, more accurate picture of where and why subsidence happens in big coastal cities.

---

---

Objectives of the Project

---

1. Identify key factors that drive ground subsidence in coastal megacities.
2. Compile and compare data from different satellite sources and ground stations.
3. Develop a simple method to map subsidence patterns over time.
4. Assess which areas are most at risk and why.
5. Suggest practical measures for monitoring and mitigation for planners.

---

---

What You Will Do Step by Step

---

1. Review basic concepts: what subsidence is and how it is measured in simple terms.
2. Gather public satellite data and any available local ground measurements.
3. Apply easy-to-understand methods to identify sinking areas on a map.
4. Cross-check satellite signals with ground data to improve reliability.
5. Analyze how subsidence correlates with land use and population density.
6. Summarize findings into a clear risk map and short report.

---

---

Expected Outcome

A straightforward risk assessment highlighting where subsidence is happening, why it occurs, and recommended steps for monitoring and policy makers to reduce future damage.