Experimental evidence that rill-bed morphology is governed by emergent nonlinear spatial dynamics

Area of Science:

• Earth Science
• Hydrology
• Geomorphology

Background:

• Rill erosion involves feedback between flow hydraulics and bed roughness, forming step-pool units.
• Irregular patterns in step-pool distribution have hindered understanding of rill morphology.
• Previous research identified self-regulation of sediment transport in established rills.

Purpose of the Study:

• To test the hypothesis that step-pool succession follows nonlinear-deterministic dynamics.
• To explain the irregular fluctuations observed in experimental rill-bed profiles.
• To provide a basis for improved soil erosion modeling.

Main Methods:

• Nonlinear time series analysis was used to reconstruct state-space dynamics.
• Analysis was performed on fifteen experimental rill-bed profiles from lab and field settings.
• Echo state neural network machine learning was applied for simulation and forecasting.

Main Results:

• Results support the hypothesis of nonlinear-deterministic dynamics governing step-pool succession.
• Rill morphology is shaped endogenously by nonlinear hydrologic and soil processes.
• The study provides a benchmark for testing new theoretical framings of rill-bed roughness.

Conclusions:

• Rill morphology is primarily governed by internal nonlinear processes, not stochastic forcing.
• Understanding these dynamics is crucial for accurate soil erosion modeling.
• Machine learning can forecast rill morphological development.