Spatial heterogeneous relationship between environmental quality and human disturbances: a case study in Ili Valley, China

Increased human disturbance poses a profound threat to ecosystem sustainability worldwide. The spatial heterogeneity of the relationship between environmental quality and human disturbance increases the complexity of this issue. Different regions experience varying degrees of human disturbances and environmental conditions, resulting in diverse ecological responses. However, the heterogeneous relationship between the mechanisms of human activity and environmental quality has not been fully investigated. Therefore, based on multi-source data, remote sensing ecological and human footprint indices were used to assess the environmental quality and the intensity of human disturbance in the Ili Valley during 2009-2021. After determining that environmental quality has strong spatial autocorrelation through Moran's I, the Getis-Ord Gi* was used to identify spatially heterogeneous units of environmental quality distribution. Furthermore, LISA maps were plotted to observe the spatial aggregation relationship between environmental quality and human disturbances. Finally, the Spatial Error Model (SEM) was identified as the most appropriate model for measuring the spatial dependence between environmental quality and human disturbance, and it was used to explain this dependence across heterogeneous units. The results showed that: (1) the environmental quality in the valley was good, whereas the slopes on both sides of the valley had poor environmental quality. Hills at middle altitudes had good environmental quality, whereas mountains at high altitudes had poor environmental quality; (2) The human disturbance intensity was higher in the valley and lower at higher elevation areas, and high-intensity areas also showed a distribution along traffic roads; (3) four different patterns of local correlations between the two variables at each location were visualized; (4) SEM was more appropriate for assessing spatial dependence of environmental quality on human disturbance among heterogeneous units because it considered the effects of spatial autocorrelation; and (5) SEM results indicated that the effects of spatial dependence between environmental quality and human disturbance were different across heterogeneous units. These findings highlight the complex relationship between environmental quality and human activity, and provide valuable insights into the spatial dependence effects of environmental quality on human disturbances and potential guidance for coordinating environmental and human activities.