Delineating the distribution of mineral and peat soils in the northern boreal regions – Transition from discrete classification to continuous maps

To meet the sustainable development goals and enable protection of peatlands, there is a strong need to plan and align land-use management with the needs of the environment. The most critical tool to succeed in sustainable spatial planning is accurate and detailed maps. Here we present a novel approach to map mineral and peat soils based on a high-resolution digital soil moisture map. This soil moisture map was produced by combining LIDAR-derived terrain indices and machine learning to model soil moisture at 2 m spatial resolution across the Swedish landscape with high accuracy (Kappa = 0.69, MCC = 0.68). We used field data from about 20,000 sites across Sweden to train an extreme gradient boosting model to predict soil moisture. The predictor features included a suite of terrain indices generated from national LIDAR digital elevation model and other ancillary environmental features, including surficial geology, climate, land use information, allowing for adjustment of soil moisture maps to regional/local conditions. As soil moisture is an important control on peat formation, we investigated if this map can be used to improve the mapping of peatlands. In this study, we included a total of 5 479 soil pit data for organic layer thickness from the Swedish Forest Soil Inventory. Peat was defined as organic layer thickness > 50 cm. The data was split into a calibration dataset and a validation dataset using a randomized 50% split. An empirical relationship between the thickness of the organic layer and the continuous SLU soil moisture map (R² = 0.66, p < 0.001) was used to generate both a categorical map (of mineral soil and peat) and continuous map (of organic layer thickness) to demonstrate how these two mapping approaches can be useful for different management objectives. The peat coverage on the new categorical map, the quaternary deposits map and topographical map was 17.3%, 14.1% and 13.5%, respectively. Map quality measures from the evaluation dataset showed that the newly developed peat map had higher recall and MCC (80.4, 0.73) than quaternary deposits map (68.5, 0.65) and topographical map (49.8, 0.61). The continuous map of the organic layer ranged 6-95 cm with an RMSE of 4 cm.

 

Using Sweden as a test case, this study provides a guide to improved mapping of mineral and peat soils from Lidar data in other boreal forest regions for effective ecosystem management. The map of organic soils was developed to support the need for land use management optimization by incorporating landscape sensitivity and hydrological connectivity into a framework that promotes the protection of soil and water quality. The organic soil map can be used to address fundamental considerations, such as;

• guiding the restoration of drained wetlands;
• designing riparian protection zones to optimize the protection of water quality and biodiversity as the new map also include riparian peats.