Integrating remote sensing and mechanistic model for spatial evaluation of shelterbelt porosity and windbreak effectiveness in agricultural landscapes

Shelterbelts serve critical functions in protecting agricultural ecosystems through soil erosion mitigation, wind damage reduction, and enhancement of farming system resilience. However, traditional evaluations of windbreak effectiveness have predominantly focused on localized zones close to shelterbelts, overlooking spatial heterogeneity and impeding landscape-level assessments. To fill this gap, this study first explored the relationship between shelterbelt structural parameters and remote sensing pixel, and further improved the method for extracting shelterbelt width by pixel decomposition. Then, we examined the influence of key shelterbelt parameters on porosity, and developed a mechanistic model to quantify shelterbelt porosity. We calculated the friction coefficient of shelterbelts and constructed a windbreak speed attenuation model within spatial computation domain based on porosity. Finally, by integrating regional prevailing wind direction and farmland distribution, we proposed the Windbreak Effectiveness Index (WEI) to assess shelterbelt protection in agricultural landscapes. The principal findings are summarized as follows: (1) The dimidiate pixel model reliably estimated shelterbelt fractional coverage across different ages (R² = 0.764, RMSE = 0.151), while the improved width extraction method demonstrated strong alignment with field measurements (R² = 0.758, RMSE = 2.12 m, MAE = 1.78 m) with minimal directional bias. (2) The developed mechanistic porosity model accurately characterized structural complexity of shelterbelt (R² = 0.775, RMSE = 0.066). Remote sensing data effectively captured porosity spatial variations (R² = 0.759, RMSE = 0.071) to extract shelterbelt horizontal structural parameters. (3) The friction coefficient effectively quantified wind speed attenuation near shelterbelts (R² = 0.628, RMSE = 0.080). The proposed WEI demonstrated to be a robust index for spatially revealing windbreak performance, with optimal windbreak effectiveness reaching 53.13 % in the study area. This research proposes a novel spatial framework for evaluating windbreak effectiveness of shelterbelt, offering actionable methods for optimizing shelterbelt design and management across agricultural landscapes.