Network optimization for enhanced resilience of urban heat island measurements
Network optimization for enhanced resilience of urban heat island measurements
复制标题
DOI：
10.1016/j.scs.2015.02.004
复制
发表时间：
2015-12-01
影响因子：
11.7
通讯作者：
Grimmond, C. S. B.
中科院分区：
文献类型：
--
作者：
Honjo, Tsuyoshi;Yamato, Hiroaki;Grimmond, C. S. B.
城市热岛效应是一个众所周知的现象，影响着各种各样的城市运营。随着廉价气象传感器的增加，有可能以更高的分辨率测量城市大气特征的空间模式。为了开发鲁棒性和弹性网络，识别传感器可能发生故障，重要的是要知道测量点何时提供额外的信息，以及为特定应用提供空间信息所需的最小传感器数量。在这里，我们考虑的温度数据的例子，和城市热岛，通过在东京都区（扩展METROS）的传感器网络的分析。考虑了现有气象测量网减少观测点的影响，采用随机抽样和聚类抽样。结果表明，分层聚类的采样可以产生类似的温度模式，在东京的测量点减少了30%。所提出的方法在评估现有城市温度网络的鲁棒性和弹性以及如何通过新的移动的和开放的数据源增强网络方面具有更广泛的实用性。(C)2015爱思唯尔有限公司版权所有。
The urban heat island is a well-known phenomenon that impacts a wide variety of city operations. With greater availability of cheap meteorological sensors, it is possible to measure the spatial patterns of urban atmospheric characteristics with greater resolution. To develop robust and resilient networks, recognizing sensors may malfunction, it is important to know when measurement points are providing additional information and also the minimum number of sensors needed to provide spatial information for particular applications. Here we consider the example of temperature data, and the urban heat island, through analysis of a network of sensors in the Tokyo metropolitan area (Extended METROS). The effect of reducing observation points from an existing meteorological measurement network is considered, using random sampling and sampling with clustering. The results indicated the sampling with hierarchical clustering can yield similar temperature patterns with up to a 30% reduction in measurement sites in Tokyo. The methods presented have broader utility in evaluating the robustness and resilience of existing urban temperature networks and in how networks can be enhanced by new mobile and open data sources. (C) 2015 Elsevier Ltd. All rights reserved.