Automated UAV-based DInSAR for Real-Time Forest Deformation Monitoring and Erosion Risk Assessment in Coastal Geographies

Project Abstract

This study presents an integrated framework that leverages automated Unmanned Aerial Vehicle (UAV) synthetic aperture radar (inSAR/DInSAR) data acquisition, processing, and analysis to enable real-time monitoring of forest deformation and erosion risk in coastal geographies. Building on advances in low-power, high-resolution SAR sensors and on-board processing, the methodology automates flight planning, data collection, calibration, and change detection to produce timely deformation maps and erosion indicators. The work addresses the critical need for near-real-time spatial intelligence in fragile coastal forests subjected to hurricane impacts, storm surges, sea-level rise, and anthropogenic pressure, where rapid information can inform emergency response, forest management, and coastal resilience planning. The research design integrates UAV-mounted DInSAR workflows with cloud-enabled processing pipelines to extract millimeter- to centimeter-scale vertical displacements across heterogeneous terrains. A multi-temporal SAR dataset is generated through repeated UAV flights over selected coastal forest stands and dune systems, ensuring consistent orbital geometry and radiometric calibration. Advanced interferometric phase unwrapping, atmospheric artifact mitigation, and coherence optimization are employed to enhance displacement sensitivity in treetop canopies, understory vegetation, and soils. To address rapid environmental changes, the framework incorporates real-time telemetry, automated quality control, and adaptive sampling strategies that prioritize regions of high deformation potential and erosion risk, such as slope transitions, dune toe areas, and drainage networks. Key contributions include the development of a hybrid processing chain that fuses SAR-derived deformation signals with high-resolution multi-spectral imagery, LiDAR-derived canopy metrics, and topographic data to discriminate between vibrational noise, seasonal growth, and tectonically or hydrodynamically induced movements. The study introduces a probabilistic erosion risk index that combines deformation rates with soil moisture proxies, vegetation health indicators, and wind/wave exposure models to forecast potential sediment loss and dune retreat. Validation is conducted against ground-based survey data, terrestrial LiDAR, and independent GNSS benchmarks, demonstrating robust performance under varying atmospheric conditions and vegetation densities. The outcomes offer actionable geospatial products including deformation velocity maps, cumulative displacement catalogs, erosion susceptibility hotspots, and scenario-based risk forecasts under different climate and storm events. The automated UAV-DInSAR system reduces manpower and turnaround time while enhancing the spatial resolution of deformation and erosion assessments, enabling proactive management of coastal forests, buffer zones, and protective dune systems. The research also discusses scalability, data governance, and transferability to other vulnerable coastal regions, along with limitations such as retrieval sensitivity in dense canopies and the need for standardized calibration protocols.

Project Overview

What This Project Is About

---

This project explores how small unmanned aircraft (drones) equipped with radar-like sensing can track tiny changes in forests and nearby coastlines over time. It uses a method called Differential Interferometric Synthetic Aperture Radar (DInSAR) with UAVs to detect movement and deformation, helping assess erosion risk and forest health in real time.

---

---

The Problem It Addresses

---

Forests near coastlines are affected by erosion, flooding, and landslides. Traditional ground surveys are slow and expensive, and satellite data may not capture rapid changes. The project combines drone-based sensing with simple data analysis to provide timely information that can guide conservation and risk management.

---

---

Objectives of the Project

---

1. Explain how UAV-based DInSAR can detect small ground movements.
2. Assess erosion risk by linking deformation data with coastal features.
3. Develop an easy-to-use workflow for data collection and basic analysis.
4. Test the method in a coastal forest environment to evaluate practicality.
5. Provide clear visuals and recommendations for stakeholders.

---

---

What You Will Do Step by Step

---

1. Study the basic concepts of UAVs, radar imaging, and deformation monitoring.
2. Collect UAV data over a chosen coastal forest area at multiple times.
3. Process data to generate simple deformation indicators (maps and graphs).
4. Analyze how deformation relates to erosion and forest conditions.
5. Validate findings with available ground information or simple field checks.
6. Prepare a user-friendly guide for future researchers or local managers.

---

---

Expected Outcome

Anticipated results include a practical workflow, deformation maps, and risk assessments that can support coastal forest management and early warning efforts, with recommendations for when to conduct follow-up surveys.