Scale and Zoning Effects on Landscape Metrics for Assessing Ecological Stress from Urban Expansion

Context: Landscape metrics can effectively quantify urban landscape dynamics and their ecological stress, but incorrect metric, scale and zoning choices may lead to inaccurate result and biased land use planning. Objectives: This study aims to explore how the scale and zoning affected the magnitude and spatial pattern of ecological stress delineated by two related landscape metrics- Eco-erosion index(EEI) and percent of built-up area (PB). Methods: We quantified ecological stress from urban expansion using EEI and PB across different scales and zoning patterns, based on 30 m land use and land cover (LULC) data of Shanghai, China. Descriptive and non-parametric statistics were used to analyze the sensitivity and spatial heterogeneity of EEI and PB in response to variations in grain size, extent, and zoning. Results: Our findings indicate that EEI and PB are more sensitive to changes in extent than in grain size. EEI is better at capturing spatial heterogeneity of ecological stress in detail compared to PB at the same zoning pattern. Statistically, zoning patterns don't affect the ranking of ecological stress levels among study units. However, EEI using fine-scale zoning is more appropriate for highlighting discrepancies within the city than using broad-scale zoning at the same grain size. It is also recommended that the minimum study unit for EEI estimated with 30 m LULC grain size should be no smaller than a 5 km by 5 km grid cell. Conclusions: This study emphasizes scale and zoning effects on EEI and PB, providing guidance for using these metrics to represent ecological stress and contributing to sustainable urban planning.