Articles | Volume 11, issue 2
https://doi.org/10.5194/soil-11-849-2025
https://doi.org/10.5194/soil-11-849-2025
Original research article
 | Highlight paper
 | 
21 Oct 2025
Original research article | Highlight paper |  | 21 Oct 2025

Representing soil landscapes from digital soil mapping products – helping the map to speak for itself

David G. Rossiter and Laura Poggio

Model code and software

code associated with the paper "Representing soil landscapes from digital soil mapping products -- helping the map to speak for itself" D. G. Rossiter and L. Poggio https://doi.org/10.17027/F6VF-9R15

Download
Executive editor
The highlighted paper “Representing soil landscapes from digital soil mapping products – helping the map to speak for itself” demonstrates how operational digital soil map products can enhance decision-making. By shifting focus from traditional point-based accuracy metrics to the spatial realism and visual structure of digital soil maps, this research addresses a gap in soil mapping practice. The approach ensures that soil maps meet statistical criteria but also more faithfully represent soil landscape patterns, making them more valuable and informative for end users such as land planners and environmental managers.
Short summary
Soil maps are useful for many applications, e.g., hydrology, agriculture, ecology, and civil engineering. The dominant mapping method is Digital Soil Mapping (DSM), which uses training observations and machine-learning to predict per-pixel. Accuracy is assessed by statistical evaluation at known points, but soils occur in spatial patterns. We present methods for helping the map to "speak for itself" to reveal patterns of the soil landscape.
Share