Back to Portfolio Research | 2026-Present

Co-Investigator

Sentinel-1 InSAR Research

This study evaluates when Sentinel-1 coherence can be trusted in freeze-thaw dominated terrain. Using Longyearbyen, Svalbard as a compact Arctic testbed, the work compares stable and highly dynamic surfaces under the same acquisition geometry to understand where C-band InSAR remains operationally reliable and where it breaks down.

Focus Season-dependent coherence reliability in Arctic terrain
Data Sentinel-1 IW SLC HH-HV stacks from 2018 to 2024
Method 12-day interferogram pairs, coherence analysis, and phase interpretation
Outcome A three-tier usability framework for risk-aware Arctic monitoring

Project Overview

The core question is simple: if coherence drops in the Arctic, is that a sign of real deformation or just seasonal surface change? Freeze-thaw transitions, snow redistribution, meltwater, and dielectric shifts can all create decorrelation before any meaningful displacement interpretation begins.

My Contribution

I work on SAR processing, interpretation logic, and framing the results in a way that is useful for operational decision-making rather than only for a one-off case study. That includes turning raw processing outputs into a clearer reliability story.

Why Longyearbyen

Longyearbyen makes it possible to compare infrastructure and exposed rock against thaw-sensitive ground, snow-dominated surfaces, and active glacier behavior without changing the broader imaging setup.

Practical Value

The result is meant to help Earth observation users read Sentinel-1 coherence more honestly in high-latitude environments, especially when climate-driven change and infrastructure risk both matter.

Full Abstract

FRINGE 2025 conference submission.

Satellite radar interferometry has become a cornerstone of operational deformation monitoring, yet its reliability depends fundamentally on the temporal stability of surface scattering mechanisms. In cryospheric and permafrost-dominated regions, surface conditions evolve rapidly due to freeze-thaw transitions, snow redistribution, liquid water infiltration, and seasonal dielectric changes. These processes alter backscatter phase consistency at scales that are often smaller than the spatial resolution of Sentinel-1, introducing decorrelation that may not correspond to actual ground displacement. As polar regions experience accelerated climate-driven change and increasing human activity, the need for robust, seasonally aware interpretation frameworks becomes critical. A systematic evaluation of coherence behavior under extreme environmental forcing therefore serves not only as a regional case study, but as a stress test for the operational limits of C-band InSAR in high-latitude monitoring applications.

Reliable interpretation of Sentinel-1 SAR coherence is critical for polar surface monitoring, but seasonal freeze-thaw processes introduce variability that may complicate data trustworthiness. High-latitude regions are characterized by seasonal snow accumulation, melt cycles, glacier motion, and permafrost dynamics, all of which can degrade interferometric coherence and phase stability. A structured assessment of these limitations is necessary to support risk-aware Earth observation services. We've selected the Longyearbyen region of Svalbard as a controlled Arctic testbed due to its combination of exposed bedrock, active glaciers, thaw-sensitive permafrost, and critical infrastructure within a compact geographic area. This diversity allows direct comparison between stable and highly dynamic surfaces under identical acquisition geometry. Sentinel-1 IW SLC HH-HV data acquired from 2018 to 2024 were processed using a single relative orbit configuration to ensure geometric consistency. Short-baseline interferograms (12-day pairs) were generated to evaluate seasonal coherence variability, temporal baseline sensitivity, and phase noise behavior across multiple freeze-thaw cycles. Preliminary analysis indicates significant coherence degradation during peak melt periods and over snow-dominated surfaces, while exposed rock and infrastructure maintain higher stability and interpretable phase trends. Based on coherence statistics and time-series consistency, a three-tier operational usability classification is proposed: Reliable, Usable with Caution, and Not Reliable. This study provides a practical framework for evaluating Sentinel-1 InSAR performance in extreme environments and supports improved decision-making for deformation monitoring in Arctic regions. By explicitly linking environmental conditions to measurement stability, the results contribute to more transparent operational use of SAR interferometry in high-latitude applications.

Project Media

Study area, acquisition geometry, and documentation.