High-Resolution Insights from Photogrammetry: Tracking Soil Surface Changes Across Scales with Time-Lapse Data over 3.5 Years for Enhanced Soil Erosion Modelling

Current process-based soil erosion models face significant challenges stemming from data availability, parameter uncertainty, and the dynamic nature of evolving environmental conditions. These limitations restrict the precision and applicability of existing models in capturing the complexities of soil erosion processes. To address some of these challenges, this study introduces an innovative approach that leverages nested, continuous, high-resolution spatio-temporal data obtained through Structure from Motion (SfM) photogrammetry. The technique enables detailed monitoring of soil surface changes caused by precipitation events across multiple spatial scales, ranging from plot-scale observations to slope-scale and micro-catchment-scale analyses.

The study's methodology incorporates an extensive and unprecedented dataset, blending time-lapse photogrammetry, comprehensive field measurements, and data collected via uncrewed aerial vehicles (UAVs) over an extensive period of 3.5 years. This robust dataset allows for a detailed monitoring of soil erosion dynamics, including flow velocity, soil consolidation and compaction process measurements. Moreover, it provides a critical foundation for the calibration and evaluation of soil erosion models, demonstrated by its application in refining the RillGROW model.

To further advance the field, the study offers an open-access dataset to the scientific community, intended for model parameterisation, calibration, and testing. Researchers are invited to build on this work by employing similar methods to collect complementary soil erosion data, thereby contributing to an expanded, high-resolution dataset. This collective effort aims to foster the development of more accurate and reliable soil erosion models, ultimately improving our ability to predict and mitigate soil degradation in diverse environmental contexts.