Identifying the Transition Zone between Peat and Mineral Soils Using Airborne Radiometric Data: a national scale case study from Ireland

Abstract

National and international climate change mitigation plans require a knowledge of peat soil extent across large geographic areas. Peat soils, which play a vital role in carbon storage and climate regulation, have a physical margin where soils change from high to low organic content. Accurate delineation of both national extent of peat soils and peat to mineral soil transition is required for assessing land use and planning effective conservation and carbon loss mitigation strategies. This abstract presents a novel approach for defining both peat soil extent nationally and transition zones between peat and mineral soils at field scale. At a national scale, peat soil maps are created using optical satellite remote sensing or legacy soil/quaternary maps or a combination of both. However, optical remote sensing cannot detect peatlands under landcover such as forest or grassland and legacy maps are often created from sparse in-situ auger data making the accurate delineation of the boundary between peat and mineral soils difficult and cost prohibitive. Airborne radiometric data, which measures natural environmental radiation, has been shown to differentiate between peat and mineral soils due to high attenuation of gamma rays in organic soils. Radiometric data is considered a direct measurement of the subsurface and so is minimally affected by landcover. Additionally, as airborne radiometric data can be acquired in a spatially consistent manner, it has the potential to identify areas of peat soil across the landscape and highlight areas of transition between high and low organic soils. In Ireland, the Tellus survey, acquired by Geological Survey Ireland (GSI) aims to acquire airborne data (including radiometric data), consistently across the country (flight line spacing of 200m) at a resolution of 50 x 50 m. Utilising this national radiometric dataset, a machine learning classification methodology is presented. Data are classified as peat (> 30 % organic material) or non-peat, with 85 % accuracy, is validated using a national soils sampling survey. A confidence value is extracted, once data are classified, which results in the identification peat soils. Several field sites across the midlands of Ireland, which are located at verified transition zones, are then used to show the effectiveness of the classification at identifying transition zones at the field scale. The methodology is robust and can be applied in all areas where these data exist. The results highlight that inclusion of an airborne radiometric dataset in a national climate plan can be used to update national and international carbon inventories of peatlands areas and inform European policy. Understanding the location of these peat to mineral soil transitions is paramount when considering the impact on climate change mitigation strategies such as potential impact of rewetting of peat soils.