Association between multilevel landscape characteristics and rural sustainability: A case study of the water-net region in the Yangtze River Delta, China

In the Yangtze River Delta in China, known for its intricate water network, achieving harmonious development between humans and nature in rural areas is imperative. However, the identification of the water-net landscape characteristics and the relationship between rural sustainability and these landscape characteristics remain unclear. The aim of this study was to bridge this gap by proposing a novel framework for investigating the relationship between landscape characteristics and rural sustainability from a typo-morphological perspective. Specifically, through regression analysis, the influence of multilevel spatial characteristics of rural landscape on sustainability was selected as the research focus. First, multilevel metrics were introduced to delineate the rural landscape characteristics, including single and multiple landscape elements and landscape types, using deep learning methods to achieve automatic classification. Subsequently, by employing an improved entropy method, we comprehensively quantified rural sustainability indicators from the economic, social, and ecological dimensions. Finally, the ordinary least squares (OLS) model and two spatial variation coefficient models, namely, geographically weighted regression (GWR) and multiscale geographically weighted regression (MGWR), were used to quantitatively analyze the relationship between the landscape characteristics and rural sustainability. Significant regression model performances were obtained with adjusted R2 values of 0.33, 0.35, and 0.4 at each landscape characteristic level. The adjusted R2 values for the GWR and MGWR, which incorporated all of the landscape characteristics metrics, were 0.84 and 0.88, respectively. The results demonstrate that rural sustainability highly depends on the proposed multilevel characteristics and exhibits spatial heterogeneity. The findings of this study improve our understanding of the typo-morphological characteristics of the landscape and provide important planning and decision-making references for sustainable development in rural areas.