Urban heat island monitoring and analysis using a non-parametric model: A case study of Indianapolis
Urban heat island monitoring and analysis using a non-parametric model: A case study of Indianapolis
复制标题
DOI：
10.1016/j.isprsjprs.2008.05.002
复制
发表时间：
2009-01-01
影响因子：
12.7
通讯作者：
Weng, Qihao
中科院分区：
文献类型：
--
作者：
Rajasekar, Umamaheshwaran;Weng, Qihao
开发了一种监测城市热岛 (UHI) 的程序，并在美国中西部的选定地点进行了测试。选择了印第安纳州中部的 9 个县，并对其 UHI 模式进行了建模。该研究使用了 2005 年拍摄的中分辨率成像光谱仪 (MODIS) 地表温度 (LST) 图像。这些图像根据研究区域的云量进行排序。由此产生的 94 张白天和夜间图像用于建模。然后将过程卷积技术应用于图像以表征 UHI。此过程有助于将 LST 数据表征为连续表面，将 UHI 数据表征为一系列高斯函数。分析了特征图像的昼夜温度分布和 UHI 强度属性（最小值、最大值和幅度）的变化。对于白天和夜间图像，任何给定图像中的皮肤温度分别在 2-15 摄氏度和 2-8 摄氏度之间变化。白天和夜间图像中 UHI 的幅度分别为 1-5 摄氏度和 1-3 摄氏度。生成白天和夜间图像的三维 (3-D) 模型，并通过动画直观地探索模式。发现了强烈且明显的城市热岛效应，从马里恩县北部一直延伸到汉密尔顿县。这一信息与过去几年马里恩县北部的发展和扩张相吻合，而南部则相反。为了进一步探索这些结果，我们对先进星载热发射和反射辐射计 (ASTER) 2004 年土地利用土地覆盖 (LULC) 数据集的特征 UHI 进行了分析。发现具有最大热特征的区域与不透水表面具有很强的相关性。整个信息提取过程都是自动化的，以便于在全球范围内挖掘城市热岛模式。这项研究已被证明是从大量遥感图像中建模和挖掘城市热岛的有前途的方法。此外，这项研究还有助于 3D 历时分析。 (c) 2008 年国际摄影测量和遥感协会 (ISPRS)。由 Elsevier B.V. 出版。保留所有权利。
A procedure for the monitoring an urban heat island (UHI) was developed and tested over a selected location in the Midwestern United States. Nine counties in central Indiana were selected and their UHI patterns were modeled. Moderate Resolution Imaging Spectroradiometer (MODIS) land surface temperature (LST) images taken in 2005 were used for the research. The images were sorted based on cloud cover over the study area. The resulting 94 day and night images were used for the modeling. The technique of process convolution was then applied to the images in order to characterize the UHIs. This process helped to characterize the LST data into a continuous surface and the UHI data into a series of Gaussian functions. The diurnal temperature profiles and UHI intensity attributes (minimum, maximum and magnitude) of the characterized images were analyzed for variations. Skin temperatures within any given image varied between 2-15 degrees C and 2-8 degrees C for the day and night images, respectively. The magnitude of the UHI varied from 1-5 degrees C and 1-3 degrees C over the daytime and nighttime images, respectively. Three dimensional (3-D) models of the day and night images were generated and visually explored for patterns through animation. A strong and clearly evident UHI was identified extending north of Marion County well into Hamilton County. This information coincides with the development and expansion of northern Marion County during the past few years in contrast to the southern part. To further explore these results, an Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) 2004 land use land cover (LULC) dataset was analyzed with respect to the characterized UHI. The areas with maximum heat signatures were found to have a strong correlation with impervious surfaces. The entire process of information extraction was automated in order to facilitate the mining of UHI patterns at a global scale. This research has proved to be promising approach for the modeling and mining of UHIs from large amount of remote sensing images. Furthermore, this research also aids in 3-D diachronic analysis. (c) 2008 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS). Published by Elsevier B.V. All rights reserved.