Evaluation of road traffic noise exposure based on high-resolution population distribution and grid-level noise data
Evaluation of road traffic noise exposure based on high-resolution population distribution and grid-level noise data
复制标题
基于高分辨率人口分布和网格级噪声数据的道路交通噪声暴露评估
DOI：
10.1016/j.buildenv.2018.08.037
复制
发表时间：
2019-01-01
影响因子：
7.4
通讯作者：
Wang, Haibo
中科院分区：
文献类型：
--
作者：
Cai, Ming;Lan, Ziqin;Wang, Haibo
本研究的主要目的是提出一种基于高分辨率人口和网格级噪声数据的道路交通噪声人口暴露评估方法。首先，网格化的研究区域，并设置在节点上的接收器，所有接收器的噪声值被视为线性声源的道路计算，这些值被渲染为交通噪声分布。接下来，使用兴趣点（POI）样本数据、人口样本数据和随机森林算法来训练人口分布模型。然后，将网格中兴趣点的统计量作为噪声分布输入到训练好的模型中，得到网格级的人口数据，并用于描述人口分布。然后，为了便于评估，根据网格节点的ID或位置，将网格级交通噪声分布数据和人口分布数据相结合。最后，采用综合考虑交通噪声和人口分布的评价指标，对交通噪声人群暴露进行评价。该方法适用于三种类型的区域在广州：一个住宅区，一个商业区和一个工业区。各研究区噪声污染水平均为“轻度”，但由于道路网较密集，标准较严格，居住区道路交通噪声污染最为严重，占总人口的12.00%，受影响面积的7.60%。这些结果显示居住在住宅区的市民对道路交通噪音较为敏感。
The primary objective of this study is to propose a method for assessing population exposure to road traffic noise based on high-resolution population and grid-level noise data. First, after meshing the study region and setting the receivers at the nodes, the noise values of all receivers are calculated by regarding roads as linear sound sources, and these values are rendered as the traffic noise distribution. Next, a population distribution model is trained with point of interest (POI) sample data, population sample data, and the random forests algorithm. Then, the statistics of the POIs in the grids are input to the trained model as the noise distribution; the grid-level population data are obtained and used to depict the population distribution. Then, to facilitate the assessment, the grid-level traffic noise distribution data and population distribution data are combined according to the ID or location of the nodes of grids. Finally, evaluation indexes considering the traffic noise and population distribution are used to assess the population exposure to traffic noise. The proposed method is applied to three types of regions in Guangzhou: a residential area, a commercial area and an industrial area. The noise pollution levels of all study regions are 'Mild', but the pollution of road traffic noise is most severe in the residential area with 12.00% of the population and 7.60% of the area affected due to the denser road network and stricter standards. These results indicate that people in residential areas are more sensitive to road traffic noise.