Soil mapping has made great advances in the last decades by harnessing the power of ever larger datasets of large extent and fine resolution to explain soil variation across landscapes. Many national scale soil property maps have been produced by the globalsoilmap routines and global maps using the soilgrid algorithms. Validation of maps against independent soil data typically shows that up to seventy percent of the variation in soils is explained by the maps - comparable with conventional (polygon-based) mapping approaches. Important efforts are underway to ensure adoption of the newly available datasets in global change and large-scale agricultural productivity studies.

Meanwhile, less attention is given to the other kinds of information that can be obtained from digital soil mapping studies. This can and should go beyond closer inspection of regression coefficients that quantify the correlation between covariates and soil properties. For example, the resolution at which landscape properties such as slope or curvature best match observations can help to understand at which scale soil-forming processes play a role. Assessment of the covariate feature space in relation to prediction accuracies can offer insight into soil landscape synergies, particularly in pedogenic processes. These insights can improve our understanding of processes, and the scale and complexity at which they might operate and thus aid the further development and smart implementation of process-based models in the landscape.

We hope to provide a platform for discussion about these opportunities. We invite contributions that develop or use novel methods to extract (soil-landscape) information from digital soil mapping, and contributions that show results from such methods by presenting a novel understanding of soil-landscape relations gained from digital soil mapping. We would also welcome examples of poorly performing DSM map products that may indicate a gap in our understanding of processes not currently captured by known soil-landscape paradigms.