关键词: 日平均气温, 地理加权回归克里格, 海南岛, 线性回归

Abstract: We analyzed the temporal and spatial patterns of the land use types in Nanjing using TM/ETM remote sensing image data from 1988, 2000 and 2013. Then, using the morphological spatial pattern analysis (MSPA) method, we obtained the landscape types of the green infrastructure (GI) for each of the three periods, and quantitatively analyzed its spatial and temporal patterns of changes. Finally, changes in the landscape connectivity of the core area were quantitatively evaluated using landscape connectivity index and landscape graph theory. The results show that, from 1988 to 2013, the total area of GI in the study area had a slower rate of increase, showing that the implementation of a series of “green Nanjing” strategic project had obtained certain achievements. From 1988 to 2013, core area and edge area increased, but the bridge zone, branch continued to decrease, loop zone overall decreased, indicating the increase in the degree of urban landscape fragmentation and the decrease of landscape connectivity. The analysis of GI network based on the landscape connectivity index showed that both of the integral index of connectivity and the probability of connectivity decreased at first and then increased; the level of GI network connectivity which was based on graph theory increased at first and then decreased. Generally speaking, the level of connectivity of GI in the study area was still relatively low. The case analysis of Nanjing GI network pattern results showed that the research which combined MSPA with landscape connectivity and graphic theory provided a new framework for the green infrastructure pattern change analysis. These results can serve as a reference for green infrastructure network planning and optimization in the study area, as well as important guidance for other regions.

Key words: daily mean temperature, geographically weighted regression-Kriging, Hainan Island, linear regression.