Abstract
Designed ecology aims to proactively create sustainable ecosystems, yet quantifying the ecological effects of design parameters remains challenging due to the complexity of mathematical models and the high cost of long-term field observations. This study presents a scaled physical model system simulating nonpoint source pollution processes in sloping landscapes, enabling rapid and controlled analysis of key design parameters in designed ecology. The results indicate that (1) through 24 laboratory experiments (slope length 0.25-1 m), runoff pollutant concentrations consistently follow a power-law decay over time, with 17 groups achieving a goodness-of-fit above 0.90. (2) Orthogonal analysis identified slope material and soil thickness as the most influential factors, while slope length and width had limited effects. (3) Field validation in navel orange orchards (slope length 4-16 m) in the Three Gorges Reservoir area demonstrated that the physical model reliably reproduced the trends of runoff and pollutant loss observed in real slopes. These findings demonstrate that scaled physical models can effectively bridge laboratory and field studies, providing actionable guidance for parameter selection in landscape design, and supporting the implementation of nature-based solutions for water ecosystem restoration.