Digital elevation models (DEMs) are the dominant input to spatially explicit digital soil mapping (DSM) efforts due to their increasing availability and the tight coupling between topography and soil variability. This coupling is often modeled using empirical relationships between soil and terrain attributes, an approach that assumes the scale and representation of DSM inputs is appropriate for characterizing soil-landscape relationships. However, soil variability is driven by soil and geomorphic processes blending at multiple spatial scales, and a physically meaningful specification of DSM inputs remains challenging. To improve these efforts, DSM techniques should integrate robust soil conceptual models with quantitative geomorphic tools that encapsulate scalable process-response dynamics. In service of this aim we implement a DSM study that embeds conceptual models of soil distribution in a quantitative geomorphic framework using fine scale (1 m) terrestrial lidar.

Specifically we use commonplace terrain attributes to stratify a soil-mantled hillslope domain into pedogeomorphic subdomains following the catena model of soil development. Our study emulates a terrain segmentation model previously developed to extract and correlate hillslope catena positions with soil properties in a rolling post-glacial landscape. When implemented as published, the model does not translate well due to greater short-range topographic variability in our domain and high frequency spatial variability carried over from fine scale lidar data. To improve model portability and accommodate fine scale DSM inputs, we propose quantitative diagnostic tools for conditioning DSM inputs. The thresholds derived from these diagnostic tools are used to recast and condition an adaptive landscape segmentation model.

Adaptive model outputs represent a bottom up statistical aggregation of fine scale terrain attributes with morphometric breaks conditioned by quantitative diagnostics of geomorphic process-response signatures. To the degree that geomorphic processes differ in their morphometric signals, these diagnostic tools may be used to compare geomorphic processes in tectonically distinct landscape, and correlate them with the occurrence and distribution of genetic soil types. In addition, the distribution and apportionment of discrete pedogeomorphic subdomains may be also treated as empirical categorical predictors of soil and ecosystem properties.