Updating legacy soil data for digital soil mapping

Legacy soil point data stored in soil information systems are a valuable resource for digital soil mapping. For dynamic soil properties, however, these data may not represent the actual field conditions, which may hamper their utility for mapping exercises. Because collection of field data is a major cost component in soil mapping, updating legacy data might be an appealing alternative to collecting new data. In this paper we show how we updated the thickness of the peat layer for more than 3 000 soil profiles obtained from the Dutch soil information system with a statistical model. In addition, we illustrate how the uncertainty about the updated values can be taken into account for digital soil mapping.