Assessing the effects of land use spatial structure on urban heat islands using HJ-1B remote sensing imagery in Wuhan, China
Assessing the effects of land use spatial structure on urban heat islands using HJ-1B remote sensing imagery in Wuhan, China
复制标题
DOI：
10.1016/j.jag.2014.03.019
复制
发表时间：
2014-10
期刊：
影响因子：
--
通讯作者：
Hao Wu;L. Ye;W. Shi;K. Clarke
中科院分区：
文献类型：
--
作者：
Hao Wu;L. Ye;W. Shi;K. Clarke
城市热岛效应因其对生物多样性和人类生活的深刻影响而受到世界各国的关注。评估土地利用的空间结构对城市化的影响，对于更好地理解和改善城市化的生态后果至关重要。本文提出用半径分维来量化热点地区不同土地利用类型的空间分异。利用环境一号B卫星遥感影像，结合植被指数、景观指数和分形维数，综合研究了武汉市土地利用格局对城市热环境的影响。对环境一号B卫星与其他遥感卫星的植被指数和景观指数进行了比较和分析，以验证环境一号B卫星的性能。结果表明，地表温度（LST）仅与归一化植被指数（NDVI）呈负相关，而与Fv在整个数值范围内均呈负相关，表明植被分数（Fv）比植被指数（NDVI）更适合预测林区的LST。此外，平均LST是高度相关的四个类为基础的指标和三个植被为基础的指标，这表明景观组成和空间配置都影响UHIs。这表明环境一号B卫星具有与其他常用遥感卫星相当的城市热岛研究能力。分形分析结果表明，从热中心到边缘，城市建成区密度急剧减小，水域、森林和耕地密度逐渐增大。这些关系表明，水，像森林和农田，有显着的影响，在减轻UHI在武汉，由于其大的空间范围和均匀的空间分布。这些研究结果不仅证实了环境一号B卫星系统研究城市环境的适用性和有效性，而且揭示了土地利用空间结构对城市环境的影响，有助于提高城市环境的规划和管理。
Urban heat islands (UHIs) have attracted attention around the world because they profoundly affect biological diversity and human life. Assessing the effects of the spatial structure of land use on UHIs is essential to better understanding and improving the ecological consequences of urbanization. This paper presents the radius fractal dimension to quantify the spatial variation of different land use types around the hot centers. By integrating remote sensing images from the newly launched HJ-1B satellite system, vegetation indexes, landscape metrics and fractal dimension, the effects of land use patterns on the urban thermal environment in Wuhan were comprehensively explored. The vegetation indexes and landscape metrics of the HJ-1B and other remote sensing satellites were compared and analyzed to validate the performance of the HJ-1B. The results have showed that land surface temperature (LST) is negatively related to only positive normalized difference vegetation index (NDVI) but to Fv across the entire range of values, which indicates that fractional vegetation (Fv) is an appropriate predictor of LST more than NDVI in forest areas. Furthermore, the mean LST is highly correlated with four class-based metrics and three landscape-based metrics, which suggests that the landscape composition and the spatial configuration both influence UHIs. All of them demonstrate that the HJ-1B satellite has a comparable capacity for UHI studies as other commonly used remote sensing satellites. The results of the fractal analysis show that the density of built-up areas sharply decreases from the hot centers to the edges of these areas, while the densities of water, forest and cropland increase. These relationships reveal that water, like forest and cropland, has a significant effect in mitigating UHIs in Wuhan due to its large spatial extent and homogeneous spatial distribution. These findings not only confirm the applicability and effectiveness of the HJ-1B satellite system for studying UHIs but also reveal the impacts of the spatial structure of land use on UHIs, which is helpful for improving the planning and management of the urban environment.