Mapping Active Peatland on Upland Mires Using Remote Sensed Data on Dartmoor, UK: A Methodological Analysis

Investigation of suitable techniques for mapping of peatlands on Dartmoor using remote sensed data has highlighted the variation between and within peatlands which makes their classification challenging. This study found the benefits of using both Compact Airborne Spectrographic Imager (CASI) and Light Detection and Ranging (LIDAR) data to identify peatlands, in particular the active sections. LIDAR data provided topographic analysis to pinpoint possible areas of peatland at a landscape scale for further small scale investigation using CASI data. Different techniques were found to be most suitable dependant on whether the site is a valley mire or blanket bog. Soft classification techniques were not used in this study due to their complex nature and the fine spatial resolution of the CASI data limiting the possible number of mixed pixels. Visual analysis was most effective with band ratios for the valley mire while an unsupervised classification of the Near Infra-Red waveband was most effective for blanket bog. This research illustrates the possible methods which could be utilised in a conceptual model for mapping the extent of peatlands at a landscape scale