[HGM03-P01] Stochasticity controls and the central role of “internal variability” in soil erosion system
キーワード：Soil erosion, Geomorphic Internal Variability, Geomorphic External Variability, Stochasticity index, Soil erosion variability
Accurate prediction of soil loss rates remains a problem because erosion exhibits a non-unique behavior given the same rainfall/runoff forcing. The effects and causes of uncertainties in soil surface erodibility resulting in such a behavior have not been fully addressed from a mechanistic perspective in previous research. We use a large database of empirical data on soil loss and a comprehensive physical model of runoff – overland flow – erosion – transport processes that dynamically updates the mass and composition of soil substrate at the hydrologic-event scale to address reasons of unpredictability in soil erosion. We explain the role of micro-scale erodibility (referred to here as ‘geomorphic internal variability’) on geomorphic response, which acts as an intermediary between larger-scale forcings and soil loss response. Accounting for a possible range of internal variability illustrates the high sensitivity of erosion response to initial conditions of soil bed, resulting in extremely large uncertainties in short-term predictions. Furthermore, the reduction of geomorphic response variability at larger temporal scales is primarily attributed to a ‘compensation effect’: temporal alternation of events that exhibit either ‘source-limited’ or ‘transport-limited’ regimes. We relate this reduction to a novel stochasticity index that reflects the degree of variability of intra- and inter-event hydrometeorologic conditions. A higher stochasticity index implies a larger reduction of soil loss variability (higher predictability) at the aggregated temporal scales with respect to the mean hydrologic forcing.
Acknowledgement:
This study was supported by the Basic Science Research Program of the National Research Foundation of Korea funded by the Ministry of Education (2016R1D1A1B03931886).
Acknowledgement:
This study was supported by the Basic Science Research Program of the National Research Foundation of Korea funded by the Ministry of Education (2016R1D1A1B03931886).