Distinguishing pristine and drained peatland sites: How effective is open source Sentinel-2 imagery?

Peatlands are among the most carbon-dense terrestrial ecosystems and play a crucial role in climate regulation and landscape hydrology, while supporting a unique and highly specialized biodiversity. In response, several European countries have developed national peatland strategies to support conservation, restoration, and sustainable management. A major limitation in implementing these strategies is the lack of consistent, up-to-date, and spatially explicit information on peatland health. Many existing peatland inventories are affected by missing data, outdated classifications, or heterogeneous data quality, which restricts their applicability for monitoring and management. In a previous study (Gross-Schmoelders et al., 2025), we demonstrated that high-resolution PlanetScope satellite imagery provides reliable information on peatland health, showing strong agreement with biogeochemical soil properties and enabling a clear distinction between pristine and drained peatland areas. Building on these findings, the present study evaluates whether freely available Sentinel-2 data are equally effective in distinguishing pristine from drained peatlands. Sentinel-2 offers moderate spatial resolution (10 m), a revisit time of approximately five days, and global open access, making it a highly attractive data source for large-scale peatland monitoring. Our analysis covers 13 peatland sites across Europe, representing a wide range of climatic conditions, peatland types, and management histories. Both, established reference sites (Gross-Schmoelders et al., 2025) and newly introduced test sites are included to enhance the robustness and transferability of the results. We analyze a suite of vegetation, moisture, and surface indices commonly applied in peatland remote sensing, including NDVI, GI, gNDVI, EVI, FAPAR, SAVI, MSAVI2, and albedo. In addition, ground motion metrics are incorporated to capture surface dynamics related to drainage and peatland health. The analysis focuses on the early growing season, when differences in vegetation structure, productivity, and moisture conditions between pristine and drained peatlands are expected to be most pronounced, consistent with findings from our previous work. Preliminary results show that both Sentinel-2 and PlanetScope data reliably differentiate between pristine and drained peatland conditions across all sites, particularly when using NDVI, GI, and EVI. Preliminary results show that FAPAR, SAVI, and MSAVI2  also exhibit consistent differences between pristine and drained conditions. Overall, these results demonstrate that Sentinel-2 represents a robust, cost-effective, and scalable data source for peatland health assessment. This has direct relevance for remote sensing–based peatland monitoring and supports the development of consistent, comparable, and transparent peatland inventories. The findings highlight the strong potential of open-access satellite data to support national peatland strategies, large-area monitoring frameworks, and evidence-based ecosystem management.

Reference:

Gross-Schmölders, M., Gupta, S., Grady, M., Wania, A., Bengtsson, F., and Alewell, C., (2025) Building a Framework to Differentiate between Pristine and Drained Peatlands in Europe by comparing Molecular and Spectral Data (submitted).