Measuring spatio-temporal heterogeneity and interior characteristics of green spaces in urban neighborhoods: A new approach using gray level co-occurrence matrix
Chenghan Xie, Jingxia Wang, Dagmar Haase, Thilo Wellmann, Angela Lausch
Abstract
Urban green space (UGS) is a complex and highly dynamic interface between people and nature. The existing methods of quantifying and evaluating UGS are mainly implemented on the surface features at a landscape scale, and most of them are insufficient to thoroughly reflect the spatial-temporal relationships, especially the internal characteristics changes at a small scale and the neighborhood spatial relationship of UGS. This paper thus proposes a method to evaluate the internal dynamics and neighborhood heterogeneity of different types of UGS in Leipzig using the gray level co-occurrence matrix (GLCM) index. We choose GLCM variance, contrast, and entropy to analyze five main types of UGS through a holistic description of their vegetation growth, spatial heterogeneity, and internal orderliness. The results show that different types of UGS have distinct characteristics due to the changes of surrounding buildings and the distance to the built-up area. Within a one-year period, seasonal changes in UGS far away from built-up areas are more obvious. As for the larger and dense urban forests, they have the lowest spatial heterogeneity and internal order. On the contrary, the garden areas present the highest heterogeneity. In this study, the GLCM index depicts the seasonal alternation of UGS on the temporal scale and shows the spatial form of each UGS, being in line with local urban planning contexts. The correlation analysis of indices also proves that each type of UGS has its distinct temporal and spatial characteristics. The GLCM is valid in assessing the internal characteristics and relationships of various UGS at the neighborhood scales, and using the methodology developed in our study, more studies and field experiments could be fulfilled to investigate the assessment accuracy of our GLCM index approach and to further enhance the scientific understanding on the internal features and ecological functions of UGS.