Regional-Scale Thermal Landscapes and Their Connectivity and Spatiotemporal Evolution: Prediction Using Artificial Neural Network and Cellular Automaton-Markov Modeling

Thermal landscapes are crucial for the study of the urban heat island phenomena because they allow for the analysis of the spatial and temporal patterns of urban thermal environments using methods and theories from landscape ecology. Recent research has primarily examined the effects of land surface composition and landscape pattern on heat islands, neglecting to examine the thermal landscape's connectivity from a holistic standpoint. We have opted to examine the spatial and temporal development of thermal landscape connectivity on a regional scale in this article. In view of this, the spatial pattern of heat islands in 2030 was predicted using artificial neural network and Cellular Automaton and Markov (ANN-CA-Markov) model based on the heat island evolution characteristics from 2014 to 2022 in Deqing County, Zhejiang Province, China. The structural characteristics of heat islands were analyzed utilizing the Morphological Spatial Pattern Analysis model, and thermal landscape connectivity was quantified using Conefor 2.6 software, and its spatial and temporal evolution patterns were analyzed. The analysis showed that, first, the ANN-CA-Markov model simulated a KAPPA coefficient of 0.78, which has a high degree of accuracy and can simulate the distribution of thermal landscapes during various time periods. Second, analysis of heat island connection trends showed that connectivity expanded quicker between 2014 and 2022 than between 2022 and 2030. Although the largest patches were also the most connected, areas with a concentrated distribution of small patches also produced patches with extremely high connectivity metrics and are likely to develop into larger, more threatening patches. This study can quantify the importance of heat islands based on thermal landscape connectivity metrics and identify clusters of important patches, which are important for mitigating and preventing the worsening of the heat island effect.