我们从中国环境总站采集实时的空气质量数据。数据API是http://106.37.208.233:20035/emcpublish/ClientBin/Env-CnemcPublish-RiaServices-EnvCnemcPublishDomainService.svc/binary/GetAQIDataPublishLives。这个API提供了一个WCF二进制数据流，使用wget命令可以下载一个二进制数据文件，然后使用python-wcfbin 对该二进制文件进行解码得到XML文本格式的数据。
这个项目，我们使用data_from_cepm.sh这个shell脚本完成整个数据获取和解码的工作。我们在Linux机器上设置了一个cron任务，每个小时定时执行两次该脚本，并将获得的XML文件存储到xml目录中，json文件存储到archives目录中。同时，我们也保存一个airnow.json的文件存储当前小时的空气质量数据。
关于数据的单位问题，通过核对官方网站可知：
- PM2.5, PM10, SO2, NO2, O3, 单位为 μg/m^3.
- CO单位是 mg/m^3.
基于此数据的中国实时空气质量地图
如果使用本方法获取数据，请引用以下论文：
Wei Lu, Tinghua Ai, Xiang Zhang and Yakun He. An Interactive Web Mapping Visualization of Urban Air Quality Monitoring Data of China[J]. Atmosphere, 2017, 8(8): 148.
我们不对使用此方法采集到的数据的真实性负责。使用此方法获取的数据出现的任何后果由使用方法者自己负责。
We collect real-time air quality data from China National Environmental Monitoring Centre. The data API is http://106.37.208.233:20035/emcpublish/ClientBin/Env-CnemcPublish-RiaServices-EnvCnemcPublishDomainService.svc/binary/GetAQIDataPublishLives. This API provides a WCF binary data stream. We can use wget to get the data, and use python-wcfbin to decode this binary data to XML text format.
In this project, we use the shell data_from_cepm.sh to do the whole process of data crawling and decoding. We set up a cron job on linux system to execute this shell twice every hour. XML files are stored in the xml directory and json files in the archives directory. We also keep a airnow.json file which contains the air quality data of current hour.
About the Units:
- For PM2.5, PM10, SO2,NO2, O3, the unit is μg/m^3.
- For CO, the unit is mg/m^3.
Base on this data, we built Air Now:Air Quality Map of China
When using the data by this method, give appropriate credit and cite the following paper:
Wei Lu, Tinghua Ai, Xiang Zhang and Yakun He. An Interactive Web Mapping Visualization of Urban Air Quality Monitoring Data of China[J]. Atmosphere, 2017, 8(8): 148.
We are not responsible for the truth of the data collected by this method. Anyone who uses the data by this method should be responsible for any consequence caused by using the data.