Exploring the spatial and temporal driving mechanisms of landscape patterns on habitat quality in a city undergoing rapid urbanization based on GTWR and MGWR: The case of Nanjing, China

Abstract

Landscape pattern significantly impacts habitat quality, especially in cities undergoing rapid urbanization, where landscape patterns are changing dramatically. However, the spatial and temporal driving mechanisms of landscape pattern on habitat quality are still unclear, and the proposed methods of Geographically and Temporally Weighted Regression (GTWR) and Multiscale Geographic Weighted Regression (MGWR) provide possibilities for the exploration of these mechanisms. This study was conducted in Nanjing from 2001 to 2020. Landscape pattern indices indicating aggregation, connectivity, diversity and compactness were calculated using Fragstats from 2001 to 2020. The habitat quality was computed using the Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model. By combining two new spatial measurement models, GTWR and MGWR, the spatial and temporal driving mechanisms of landscape patterns on habitat quality were explored. The results show that (1) as Nanjing’s land under construction has expanded, habitat quality has decreased significantly, and the overall landscape pattern has fluctuated drastically. (2) GTWR and MGWR are well-suited to such analysis and provide important insights. (3) Overall, aggregation and compactness were negatively associated with habitat quality in areas of low-quality habitat. Increased connectivity on low habitat substrates had a positive effect on habitat. The increase of diversity in proximity had a positive effect on habitat, while the opposite was true in high habitat zones. (4) As the urbanization level increases, the negative effects of aggregation expand, as do the positive effects of connectivity and diversity. (5) The extent of influence of landscape pattern effects are ranked from largest to smallest: compactness, diversity, connectivity, and aggregation, while the intensity of effects is reversed. Based on these findings, a reference point for urban planners is provided to plan urban landscape patterns in a sustainable and rational manner. It also provides a new means of integrating GTWR and MGWR into the study of landscape ecology.