Spatial heterogeneity patterns along the human footprint gradient and their ecological implications: A case study in South Korea

Assessing spatial heterogeneity patterns along human footprint (HF) gradients is important for understanding biodiversity conservation in human-modified landscapes. This study investigates how human footprint intensity shapes spatial heterogeneity patterns and their ecological implications in South...

Full description

Saved in:
Bibliographic Details
Main Authors: Kyoung-Ho Kim, Minkyu Park, Taeho Park, Donghae Baek, Gyobeom Kim, Namjoo Heo, Dong Jun Chun
Format: Article
Language:English
Published: Elsevier 2025-08-01
Series:Ecological Indicators
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1470160X25007010
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Assessing spatial heterogeneity patterns along human footprint (HF) gradients is important for understanding biodiversity conservation in human-modified landscapes. This study investigates how human footprint intensity shapes spatial heterogeneity patterns and their ecological implications in South Korea by integrating multiple heterogeneity metrics and machine learning approaches. We quantified HF heterogeneity through local indicators of spatial association and gray-level co-occurrence matrix indices and identified three distinct zones through Self-Organizing Map clustering: urbanized (low heterogeneity), transitional (high heterogeneity), and natural (moderate heterogeneity) areas. The relationships between heterogeneity metrics and ecological indicators—forest patch number, mammal species richness, and protected area coverage—exhibited complex nonlinear patterns. While forest fragmentation increased with HF intensity, mammal species richness showed local increases in transitional zones despite declining protected area coverage, supporting the heterogeneity-diversity relationship. Machine learning models incorporating multiple heterogeneity metrics significantly improved predictive performance compared to HF intensity alone, with Random Forest achieving the highest accuracy. These findings indicate that moderate levels of human activity create complex landscape mosaics that can sustain biodiversity, whereas intensive urbanization leads to homogenization. Our results suggest that landscape planning should consider spatial heterogeneity patterns, particularly in transitional zones, to maintain ecological connectivity. This study provides a quantitative framework for assessing human impacts on ecosystems through spatial heterogeneity analysis, contributing to more effective conservation strategies in human-modified landscapes.
ISSN:1470-160X