1989～2015年海口城市热环境与景观格局的时空演变及其相互关系

Spatio-temporal evolution and interrelationship between thermal environment and landscape patterns of Haikou City,1989～2015

利用1989,1999,2007和2015年的4期Landsat5TM/8OLI-TIRS遥感数据,综合应用遥感、地理信息系统、景观生态学和统计分析相结合的方法,探讨海口城市热环境与景观格局的时空演变关系.结果显示,1989～2015年间,海口城市陆地表面温度(LST)总体呈逐步上升的趋势.城市热岛面积逐步扩大,空间质心向西南方向转移,且在2007年以后转移速率明显加快;冷岛质心则向东部生态核心区转移.城市LST较高的区域往往与不透水表面分布一致,较低的LST与绿地或水体分布一致.4期绿地的平均LST比不透水表面低4.17℃,其中2015年绿地景观类型百分比(PLAND)每增加10%导致LST下降0.57℃;不透水表面每增加10%,LST则上升0.78℃.不同时期绿地和不透水表面的平均LST与PLAND、最大斑块指数(LPI)、聚集度指数(AI)均表现为一致的极显著相关关系,且相关性逐年增强,其中绿地为负相关,不透水表面为正相关;景观斑块的大小和聚集程度对城市LST有较大影响.研究结果可以拓展对城市热环境与景观格局时空演变关系的认识,为城市规划和决策者提供参考.

Landsat 5TM/8OLI-TIRS thermal infrared remote sensing data from 1989, 1999, 2007 and 2015 were used here to explore the spatio-temporal evolution and interrelationship between the thermal environment and landscape patterns in Haikou, China. Through the integration of remote sensing, geographic information systems, landscape ecology and statistical analysis methods, our results showed that the land surface temperature (LST) of Haikou City showed a gradual increase from 1989 to 2015. The area of urban heat island gradually expanded, the spatial center of mass shifted southwest, and the transfer rate accelerated significantly after 2007. In addition, the center of cold island mass was transferred to the eastern eco-centric area. The distribution of high land surface temperature in the city correlates with impervious surface, while the low LST correlated with the distribution of green land or water. The mean LST of green space was 4.17℃ lower than that of impervious surface. In 2015, 10% increase in percentage composition of landscape (PLAND) of green space caused a 0.57℃ loss in LST, whereas a 10% increase in PLAND of impervious surface caused a 0.78℃ increase in LST. The mean LST of green space and impervious surface in different years was consistently and highly significantly correlated with PLAND, largest patch index (LPI) and aggregation index (AI). In fact, the correlation grew stronger over the years. The correlation of LST with green space was negative, while it was positive for impervious surface. The size and degree of aggregation of landscape patches had great influence on urban LST. These results expand our understanding of the spatio-temporal evolution and interrelationship between urban thermal environment and landscape patterns, as well as provide a reference for urban planners and policy makers.