A Machine Learning Strategy to Predict Peatland Revegetation Following Re-Wetting

Global warming and environmental degradation have placed increasing pressure on the conservation and restoration of vital ecosystems such as peatlands. Historical peat extraction in Ireland has led to significant habitat destruction, carbon emissions, and biodiversity loss. This study leverages satellite imagery and machine learning to predict the recovery rate of Irish peatlands, with a focus on the revegetation of decommissioned bogs. By analyzing time-series data from Sentinel-2 and applying Normalized Difference Vegetation Index (NDVI) analysis, this research quantifies vegetation regrowth. A machine learning-based regression model is developed to assess restoration success and forecast long-term ecological trends. The findings provide valuable insights into the impact of rewetting strategies and support conservation efforts aimed at improving carbon sequestration and biodiversity.