State Key Laboratory of Desert and Oasis Ecology, Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi 830011, China; University of Chinese Academy of Sciences, Beijing 100049, China.
State Key Laboratory of Desert and Oasis Ecology, Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi 830011, China.
Sci Total Environ. 2020 Mar 25;710:136336. doi: 10.1016/j.scitotenv.2019.136336. Epub 2019 Dec 30.
Land surface temperature (LST) is defined as an important indicator in the formation and evolution of climate. In some cases, changes in landscape patterns affect LST, even more than the contribution of greenhouse gases. Although much work has been done with respect to the correlations between urban development and thermal environment dynamics, the related questions regarding relationships between LST and landscape patterns in arid regions are not thoroughly considered. Understanding these questions is important in climate change and land planning. The objective of this study was to explore the spatiotemporal variations of LST by distribution index (DI) and Mann-Kendall mutation analysis method and to quantify the relationships between landscape patterns, climatic factors, topographic factors, and the land surface thermal environment (LSTE) by the ordinary linear regressions (OLS) model. The landscape patterns dataset, which was validated by a field trip, was extracted from the Land satellite (Landsat) TM/OLI images by the Random Forest methodology in ArcGIS software. The MODIS/LST product was validated by the "Monthly dataset of China's surface climate" and a field trip. Annual LST increased by 0.54 °C (23.15 °C in 2000 and 23.79 °C in 2015). In different landscape patterns, the percentage of areas with a high level of LST showed a significant difference. In barren land, the highest area proportion for the high LST level was larger than in other landscape patterns. Meanwhile, the area of low LST was mainly concentrated in water bodies. Considerable changes have occurred in landscape patterns, in which the most noteworthy was cultivated land encroaching on grass land (3708.44 km). The composition of landscape patterns was more important than distribution in determining the region's LST. These findings provide valuable information for land planners dealing with climate change and ecosystem conservation in arid regions.
地表温度（LST）是气候形成和演化的重要指标。在某些情况下，景观格局的变化对 LST 的影响甚至超过了温室气体的贡献。尽管已经有很多关于城市发展与热环境动态之间相关性的工作，但在干旱地区，LST 与景观格局之间的关系相关问题还没有得到充分考虑。了解这些问题对于气候变化和土地规划非常重要。本研究的目的是通过分布指数（DI）和曼恩-肯德尔突变分析方法来探索 LST 的时空变化，并通过普通线性回归（OLS）模型来量化景观格局、气候因素、地形因素与土地表面热环境（LSTE）之间的关系。通过随机森林方法在 ArcGIS 软件中从 Land 卫星（Landsat）TM/OLI 图像中提取验证过的景观格局数据集。MODIS/LST 产品通过“中国地表气候月度数据集”和实地考察进行验证。年 LST 增加了 0.54°C（2000 年为 23.15°C，2015 年为 23.79°C）。在不同的景观格局中，高 LST 水平的面积比例存在显著差异。在荒地中，高 LST 水平的最高面积比例大于其他景观格局。同时，低 LST 区域主要集中在水体中。景观格局发生了很大变化，其中最值得注意的是耕地侵占草地（3708.44km）。景观格局的组成比分布对区域 LST 的影响更重要。这些发现为处理干旱地区气候变化和生态系统保护的土地规划者提供了有价值的信息。
