Dynamic Fuel Mapping in the South Pennines using an Earth Observation Multisensor and Multitemporal Approach

This research presents an Earth Observation (EO) multisensor (optical and radar) and multitemporal approach to understand the dynamics of fuel types in the South Pennines. Fuel properties of peatland vegetation vary spatially due to differences in land management activities, wildfire occurrence and phenology [1]. The 10m Land Cover Map from the Centre for Ecology and Hydrology is used to implement a stratified sampling technique for extracting Synthetic Aperture Radar (SAR) intensity, Interferometric SAR (InSAR) coherence and optical Vegetation Indices (VI) for bog, heather, heather grassland and acid grassland. Areas unburnt are sampled to act as a control and mode of comparison to known burned areas supplied by the Moors for the Future Partnership.

We examined the detectability, structural dynamics and recovery of different land covers from 2017 – 2023 using Sentinel-1 SAR intensity and InSAR coherence [2]. Sentinel-2 optical VIs such as Soil Adjusted Vegetation Index (SAVI) [3], Normalised Difference Moisture Index (NDMI) [4] and Plant Senescence Reflectance Index (PSRI) [5] are analysed to characterise the phenology of fuel properties and to delineate periods of highest fuel flammability. An analysis of pre-/post-fire signals enables better understanding of wildfire-induced changes. Results are compared with field data from known wildfire areas and from a fieldwork campaign of wildfire sites (July 2023). The outcomes of this research provide a detailed understanding of the South Pennines wildfire regime for the last seven years. The dynamic fuel maps produced will help to inform future wildfire resilience work for this heterogenous peatland environment at a landscape scale.