Abstract :
[en] The moisture status of peatlands is an important factor as it directly affects carbon dynamics. Therefore, it is critical to characterize and understand peatland moisture status and to monitor its spatial and temporal variations. This study aims to evaluate the potential of drone-borne ground-penetrating radar (GPR) in combination with full-wave inversion to investigate the spatial and temporal variability of peatland root-zone moisture. A secondary objective is to assess its benefits for restoration applications. This study was carried out on a 4.5 ha peatland in the Belgian Hautes Fagnes which was previously degraded by forestry activities. Ground-penetrating radar measurements were conducted every 2 to 4 weeks for 17 months, resulting in 19 peatland soil moisture maps with a 5-meter resolution. Reference soil moisture data were collected using ground-based probes to enable comparison.The temporal variability showed an overall correlation of 0.71 between the GPR and the ground-based probes, indicating that this method effectively captures overall moisture dynamics across the entire study site throughout different seasons. In contrast, the spatial comparison of GPR with the ground-based probes showed a lower correlation, namely 0.23, which is attributed to the high micro-variability of soil moisture (on centimeter to meter scales) and the spatial mismatch between the measurements and their characterization areas and depths. However, we show that the spatial data contained high information content when applying a spatial clustering analysis to produce maps of homogeneous moisture classes. These clusters aligned well with other specific site characteristics, such as peat depth and vegetation composition, and can be used to support the planning of restoration efforts. This study introduces a new approach to studying peatland root-zone moisture and shows potential to guide and monitor peatland restoration strategies.
