Group20 Report

 Air Quality In Chinese Developed Cities

OVERVIEW

PROPOSAL

POSTER

APPLICATION

REPORT

Introduction and Motivation

In the winter of 2014, Beijing suffered severe smog weather. The haze has lasted for half a year, and the number of heavy polluted days was more than one month. Air quality issues and environment issues were being discussed throughout the country, everyone was questioning himself about the consequences of developing economy by sacrificing environment. Our living conditions has become bad, and we are faced with the tracheitis, pneumoconiosis, asthma, to name just a few.

Chinese government has frequently proposed to abandon the outdated idea called “treatment after pollution” and to protect the environment while developing the economy. Therefore, after three years, has China’s air quality become better? In our project, we want to apply the visual analytics tools to visualize the changes of air quality of major cities of China. We will show the fluctuation of the historical AQI (Air Quality Index), and the pollution sources, like PM2.5, PM10, SO2, O3, CO2 and so on.

Review and critique on past work

As increasing people are attaching more importance to environment issues, there are a lot of researchers over the world studying how to improve our environment, including air quality problem, and they have contributed many substantial outcomes to this area. In China, researchers have studied air quality from various aspects based on the background of China. Some of them focused on impact of human factors on air quality, such as increasement of population and GDP, some of them focused on specific areas, like Beijing, Shanghai or eastern areas, to analyze that how the air quality is changing, and others are also interested in forecasting air quality in the future. As experts have already enriched this research area, we will focus on using visualization tools to help people better and more effectively understand the changing trend of air quality and its pollution sources. We hope that we can give some contribution raising awareness of protecting environment of people.

Data preparation

Data introduction

Our datasets were collected from various government departments containing AQI and its pollutants: AQI (Air Quality Index), PM2.5, PM10, SO2, NO2, CO, O3.We choose monthly data from 2014-01 to 2018-8 as our target dataset containing 6162 record rows. The columns are shown as below:

Variable name	Variable property
Date	Monthly data from 2014-01 to 2018-08
AQI	Air quality index, as it increases, air quality is getting poor
Range	Air quality range
Air quality level	Five levels based on range
PM2.5	One of pollutants affecting AQI, monthly data from 2014-01 to 2018-08
PM10	One of pollutants affecting AQI, monthly data from 2014-01 to 2018-09
SO2	One of pollutants affecting AQI, monthly data from 2014-01 to 2018-10
O3	One of pollutants affecting AQI, monthly data from 2014-01 to 2018-11
CO2	One of pollutants affecting AQI, monthly data from 2014-01 to 2018-12
NO2	One of pollutants affecting AQI, monthly data from 2014-01 to 2018-13
City	110 cities from China
Log	Longitude of each city gotten by R coding
Lat	Latitude of each city gotten by R coding

Data pre-processing

1) Use R to get 110 cities’ longitude and latitude.

2) Reformat the data structure to do better time series analysis.

3) Combine all the cities’ air quality data file into one csv file.

The part of final data structure is shown as below:

Visualization Approach

• Our visualization will base on R Shiny and we will do the time series analysis and forecasting by using historical data to estimate air quality index in future.

• We will conduct interactive geographic analysis to discover what factor influences the AQI such as number of cars, forest cover rate, GDP.

• In addition, we will explore whether the first-tier city only pursue high GDP and leave alone the air quality then trigger respiratory disease

Data Source

All these first-tier city AQI and responding date and some index contribute to API(PM2.5,PM10,SO2,CO,NO2,O3)(2014-2017).
GDP in those first tier city from 2014-2017.
Forest cover rate in the province of those cities.
Car number in those cities(2016-2017).
Total factory numbers in those cities.
Respiratory disease.

http://data.stats.gov.cn/easyquery.htm?cn=E0103
https://www.aqistudy.cn/historydata/
http://www.sc.stats.gov.cn/tjcbw/tjnj/2017/zk/indexch.ht
http://www.stats-hb.gov.cn/
http://www.bjstats.gov.cn/tjsj/
http://www.gzstats.gov.cn/tjsj/hgjjsjk/
http://www.gdstats.gov.cn/tjsj/zh/
http://tjj.zj.gov.cn/tjsj/ydsj/gy/
http://www.njtj.gov.cn/
http://www.cqtj.gov.cn/tjsj/sjjd/
http://stats.tj.gov.cn/
http://tj.jiangsu.gov.cn/
http://www.shaanxitj.gov.cn/
http://www.hntj.gov.cn/
http://www.ln.stats.gov.cn/
http://www.stats-sd.gov.cn/
http://www.ha.stats.gov.cn