ABSTRACT
Peatlands contain a large amount of the carbon stocks of the biosphere. Mapping the extent, locations and conditions of peatland at the landscape scale has implications for carbon inventories, conserva- tion and ecosystem services assessments. Digital Soil Mapping provides a (geo)statistical framework to quantitatively model relationships between soil properties measured or observed on the ground and environmental covariates. The main aim of this paper was to present some successful examples of DSM for northern peatlands and their condition, in particular mapping and modelling peat presence and condition using a range of available earth observation sensors. Remote sensing data were also used to map the potential of the peat being degraded. The two results were combined to provide a probabilistic indication of where degraded peat can be found in Scotland. The results could be validated with good outcomes that visually agree with the expert evaluation and the common knowledge of the landscape. These probabilities and uncertainties need to be taken into account to improve further modelling such as earth surface modelling or carbon accounting.
