Monitoring peatland dynamics in the Baltic region using EGMS products

Climate change and the increasing regulatory requirements to preserve natural carbon sinks have intensified interest in the large-scale monitoring of peatland ecosystems, which are among the most effective terrestrial environments for long-term carbon storage and hydrological regulation. In this context, Interferometric Synthetic Aperture Radar (InSAR) provides an opportunity to investigate peatland dynamics over extensive areas that are difficult to survey systematically on the ground, through repeated and cost-effective observations. This study tests the transferability of an open-source and reproducible methodology for the identification and monitoring of Peat-like Behaviour Areas (PBAs), based on Advanced Differential Interferometric Synthetic Aperture Radar (A-DInSAR) time series provided by the European Ground Motion Service (EGMS), under the demanding observational conditions of the Boreal region, where snow cover, freeze–thaw processes and coherence loss produce sparse and discontinuous displacement series. The workflow combines Principal Component Analysis (PCA), K-means clustering and Density-Based Spatial Clustering (DBSCAN) to identify coherent areas characterised by peat-like displacement patterns and to distinguish sectors affected by subsidence from those exhibiting uplift behaviour. Applied across the Baltic countries within the Boreal biogeographical region, the methodology identified 576 PBAs (424 in Estonia, 125 in Latvia and 27 in Lithuania). A comparison with the CORINE Land Cover 2018 (CLC) inventory showed that approximately 43% of the identified PBAs occurred outside existing peatland inventories, highlighting the potential of the approach to identify areas exhibiting peat-like displacement patterns. Validation based on Sentinel-2 Total Productivity and Sentinel-3 OLCI Terrestrial Chlorophyll Index products revealed a strong spatial correspondence between the detected PBAs and areas characterised by reduced vegetation productivity and chlorophyll content. By distinguishing peatland sectors that are subsiding from those that are stable or rising, the approach provides a spatially continuous indicator of ecohydrological condition that complements static inventories, enabling degradation to be located within individual peatland systems and supporting the targeting and subsequent assessment of restoration interventions across different biogeographical regions.