From the short-term view, understanding the effect of air pollution on urban activity is also crucial in terms of reducing and assessing the exposure risk. The theoretical base is that a rational individual may cancel or postpone discretionary activities when heavy pollution happens, resulting in an overestimate of their exposure risk. In contrast, the exposure risk of those who have more indispensable activities may be comparatively underestimated. Recently, some studies have pioneered the use of cellphone data (Nyhan et al., 2016) or simulated individuals’ mobility patterns (Burke, Zufall, & Ozkaynak, 2001) to provide a better exposure evaluation by taking into consideration daily mobility patterns.
Air pollution in China has been causing severe health consequences. Research demonstrates that air pollution in China may have caused health-related economic losses of 1.63% to 2.32% of the GDP (Li, Lei, Pan, Chen, & Si, 2016), and is calculated to contribute to 1.6 million deaths per year — roughly 17% of yearly deaths in China (Rohde & Muller, 2015). In North China, the most affected area (Li & Sun, 2018), long-term exposure to total suspended particulates may have reduced life expectancies by about 5.5 years (Chen, Ebenstein, Greenstone, & Li, 2013). In fact, China has been one of the countries with the highest particulate matter levels in the world (Chen et al., 2013). Air pollution in most Chinese cities exceeds 6 to 20 times the values suggested by the World Health Organization Air Quality Guidelines (Chan & Yao, 2008; Long, Wang, Wu, & Zhang, 2014). Meanwhile, both the recent industrialization and urbanization of China are aggravating the problem (Sheng & Tang, 2016; Zheng & Kahn, 2013).
Additionally, studies related to environmental injustice also shed light on the potential association between air pollution and urban activity. In Western cases, researchers (Boone, Fragkias, Buckley, & Grove, 2014; Jerrett et al., 2001) showed that people with lower socioeconomic status are exposed to higher levels of air pollution. And studies suggest that there are serious environmental inequalities associated with income level (Bevc, Marshall, & Picou, 2007; Krieg & Faber, 2004), with the poor being exposed to environmental pollution more than the middle class. Apparently, understanding how urban activity responds to air pollution may contribute significantly to a better environmental injustice evaluation.
The supporting policies for environmental injustices in China consist of national policies, laws and regulations, air quality standards. Among all the three segments, air quality standards were mostly newly introduced in the past five years as air was getting significantly unhealthy. The protection from central government were expanding from simple petrol-powered vehicle usage regulations to serious punishment of pollutants mis-management or allowance of electronic automotive purchasing.
Percentage of secondary industry over GDP
In the past decade (2010-2018), secondary industry, as a main source of pollution among all the three industries, kept decreasing, which directly plummet from 4.64 billion yuan in 2010 to 4.05 billion yuan in 2017. The plummet of industries was supported by the Chinese government strategy on industry conversion.
Source: China National Bureau of Statistics The past decade was the only period in the past forty years when percengt of secondary industry plummet significantly. It happened in the random years but had never been a trend.
China’s 13th Five-Year (2016–2020) Plan for Economic and Social Development contains strong commitments to improve air quality and control emissions. The Plan describes the country’s clean-air action plan as follows:
“We will formulate a plan for ensuring air quality standards in cities are met, strictly enforce obligatory targets, see that cities at and above the prefectural level achieve a 25% reduction in the number of days of heavy air pollution, and channel greater effort into reducing fine particulate matter emissions in key regions. We will establish a monitoring system to ensure that environmental protection standards for vehicles, watercraft, and fuel oil are achieved. We will work to increase the proportion of natural gas users in cities. We will strengthen monitoring of windblown dust from unpaved roads and construction sites and prohibit open straw burning.”
The Five-Year Plan commits to reduce emissions, ensure compliance with emissions standards, and promote the use of clean energy:
“We will ensure that all industrial polluters meet emissions standards. We will improve emissions standards, strengthen supervisory monitoring of industrial pollution sources, publish a blacklist of enterprises that fail to meet emissions standards, and require such enterprises to make corrections within a stipulated time frame. All heavily polluting enterprises located within urban districts will either be relocated, upgraded, or, in accordance with the law, shut down. We will conduct the second national survey of pollution sources. We will reform the total emissions control system for major pollutants so that more pollutants are covered. . . . We will promote the use of alternative clean energy in urban “villages” and [outskirts], and replace small and medium coal-fired facilities. . . . Chief industries will be transformed to achieve clean production.”
The supporting policies showed both wider social awareness and governmental determination on solving the environmental problem in China. In an attempt to solve the problem from most populated regions, heavy industries (mostly referred to industry outputting high pollution, such as steel, chemical, Metallurgical Industry) had been gradually transferred from populated cities to rural areas, a process known as “industry shift policy” in China. Focusing on service, high-tech, and environmentally-friendly industries, those developed areas benefited from both economic growth and health protection awareness. However, for regions where manufacturing was transferred to, citizens may suffer from both deteriorating air quality and the lack of avoidance behaviors. If this kind of regional balanced-development policy is not combined with public health policy such as spreading the health effects of air pollution and adjusting work schedules in heavy pollution days, then the vision of “economic growth – health problem – protection awareness” should be archived quickly to avoid accumulated health problems. The worst scenario would be that an economic slowdown occurs together with the absence of public health policy and air pollution in the Middle and broader South China increases. Obviously, this problem could be a global lesson for all developing countries with unbalanced regional development.
All data manipulation, organization and cleaning was done in R with RStudio. The code and PDF of everything computed is available for reproducing our results and is included in the following github repository:
All of the negative / NA values were removed to prevent skewing of distributions. Negative values do not make sense for air quality measurement units, and most negative values were -999, which usually indicates missing data points.
The distributions were analyzed as single points on scatterplots, as well as aggregated weekly & monthly into line graphs using average PM2.5 measurements.
Variance was calculated after aggregation, but was not significant enough to drastically effectinterpretation
Variance formonthly aggregation was much worse, so we decided not to use it. Instead, daily aggregation was plotted on an axis refitted to show months for better analysisand visual interpretation.
Standard Deviations of aggregated AQI’s are shown to the left, and the variance is extremely high, which means these measurements are not reliable. Additionally, the mean aggregated values are too even, which shows a very even distribution throughout the year which is not accurate. The measurements vary greatly from 0 all the way up to almost 800.
Pollution data was visualized all together and also organized by city to view marginal distributions.
When all five cities were graphed together, the trends through 2016 seem to follow the same patterns by visual inspection. Since historical smog data was only released for these five cities, this shows that the combined data set can be used as an adequate proxy for the pollution patterns of the entire country.
DEMOGRAPHICS
We were only able to find mean income data and population data for each city by year, which was not ideal for our analysis. However, we did graph their distributions next to that of the cities and did not find any obvious correlations. We were not able to draw any conclusions based on these data points. If in the future we are able to find better proxies for economic data possibly on poverty levels or land use in each region, we could further speculate if those have any relationships with pollution.
FACE MASK SALES
This data has been organized into a monthly distribution. Additionally, the pollution data was aggregated and subsetted to plot on a six month timeline so it could be compared directly with the mask trends.
PROVINCES
The shapefile did not provide names, so over 50 provinces have been hand-labeled and coordinated with the table.
Indicator values have been set to the 5 provinces relevant to our region-specific pollution data.
We obtained a Google API key to access use in a mapping package (ggmap) made for R.
The province data was then mapped onto Carto to analyze relationships between mobile activity and location.
An essential and foremost thing we were thinking about was which was indicator we could use for representing urban activities. In the research paper done collectively by MIT and Tongji University (Yan, Duarte, Wang, Zheng & Ratti, 2018), researchers used the social media check-in data from Sina Weibo micro-blogging microform, the Chinese version of Twitter as the indicator of people’s urban activity data. As mobile devices are ubiquitous (Chinese person surveyed owns at least a basic mobile phone (98%)). We thus used the data from Kaggle. The Data is collected from TalkingData SDK integrated within mobile apps. TalkingData serves under the service term between TalkingData and mobile app developers. Kaggle data was collected based on full recognition and consent from individual user of those apps have been obtained, and appropriate anonymization have been performed to protect privacy.
Kaggle Data Sample
We started by tabling observations by device ID to retrieve frequency of entries for each unique person in the dataset and removed users with fewer than 20 entries to exclude the users not living in that specific region. In order to identify the residential city of each user, we cleaned the data by extracting the people with more than 20 entries in that specific region.
Observations from the first (April 30) and last day (May 8) were removed because they only have 800 and 2 points, respectively. This compared to the rest of the days containing upwards of 400,000 entries is very insignificant and will not provide much insight.
The points have been tabled by device ID to retrieve frequency of entries from each unique person in the dataset.
Users with fewer than 20 entries were removed to extrapolate where each user lives.
K-MEANS CLUSTERING MAP
K-means clustering, an unsupervised machine learning method was applied to all mobile activity observations to see where most of the data are gathered together. The centers of each cluster were then plotted onto the province map. The map shows that a lot of the biggest clusters are near our 5 cities which gives stronger evidence that the mobile check-in data can be used as an indicator for activity.
frequency of 2016 measurements in unhealthy and hazardous ranges
The measurement table above highlights how severe the air quality is in Beijing. 42% of all unhealthy measurements recorded were in Beijing and 94% of all hazardous level measurements. The maximum PM2.5 value was 782 in Beijing while on the other hand, Chengdu, Guangzhou, and Shanghai all had zero measurements in the hazardous range through all of 2016.
FACE MASK SALES
The shape of the pollution trend loosely follows that of the mask sales trend. Even though it is more granular because we only have mask data from six months, we can see that they both start at their highest and drop in the second month, and rise back up again. Both have two maximums at roughly the same times. This corresponds to our hypothesis: people tend to buy more masks when the air quality goes worse.
However, as has been said, more conclusion could be made if we could get access to the provincial sales data.
INCOME & POPULATION
According to the bar charts, Shanghai & Beijing have both the highest income and population values, while Shenyang and Chengdu are significantly lower.
Beijing has the highest proportion of hazardous measurements at 50.9% and also has the highest income and population values. However, Chengdu has the second highest proportion at 26.6% but is at the lowest in terms of both income and population. Additionally, the city with second highest mean income and population, Shanghai, takes up only 6.3% of all dangerous values reported in 2016.
From these observations, we were not able to conclude any intuitive relationships between these two demographic measures and air quality. The potential limitations could be the listed factors:
Seasons and climate also play an important role in pollution.
Dataset was found within a too short period to summarize a intuitive trend.
Although income could be a proper variable for measuring economic development, detailed factors like industries would also have influence on the trends.
MOBILE ACTIVITY
mobile activity check-ins by day
Mobile activity kind of shows some similar trends when compared marginally by city. Activity in Beijing and Shanghai seemed to have risen as air quality got better, which is what we expected. When air quality got worse, activity in Guangzhou dropped. Not all five of the cities followed the trend of our hypothesis, but the two biggest cities most affected by smog (Beijing and Shanghai) did, which highlights their greater sensitivity to pollution than the smaller regions.
SUMMARY
Our study showed air pollution does have a significant impact on urban activities, in terms of the frequency of local residents’ entry of certain locations. The majority of local people still travel or commute as usual without much potential to choose at-house working. Secondly, people in more developed cities (such as Beijing and Shanghai) do have greater sensitivity to pollution than people in less developed regions. Due to the limitation of the data, it was harder for us to see whether people in different cities share similar purchasing behavior in smog days. But we were able to see the similar trend shared among pollution fluctuations and people’s purchasing intentions.
The available analysis together with the policies indirectly reveals new insights about environmental injustice in China. New injustice may arise in underdeveloped areas where manufacture industry is transferred to but people barely take avoidance behavior. China had worked really hard to rise the environemntal-related standards. But there is still a long way for the whole Chinese society to defend themselves from pollution and environmental injustice.
Source: US Department of State Air Quality Monitoring Program (http://stateair.net/)
PM 2.5 concentration levels which indicate air quality for five different cities in China
Each dataset spans one year and concentration measurements are recorded hourly every day
We selected 2016 for our analysis because our mobile activity and mask sales data is also within that year, so our research is narrowed into this time frame.
BACKGROUND
PM2.5 refers to atmospheric particulate matter (PM) that have a diameter of less than 2.5 micrometers, which is about 3% the diameter of a human hair.
Since they are so small and light, fine particles tend to stay longer in the air than heavier particles. This increases the chances of humans and animals inhaling them into the bodies. Owing to their minute size, particles smaller than 2.5 micrometers are able to bypass the nose and throat and penetrate deep into the lungs and some may even enter the circulatory system. Studies have found a close link between exposure to fine particles and premature death from heart and lung disease. Fine particles are also known to trigger or worsen chronic disease such as asthma, heart attack, bronchitis and other respiratory problems.
Due to its detrimental effect on people’s health, we chose the concentration of PM2.5 as the indicator of air quality. This hourly-recorded concentration data was collected from the U.S Department of state, focusing on the five cities we worked on.
Recent years China saw a growing demand for filtration masks, pollution monitors, air purifiers and other anti-pollution gadgets. From December 16 to December 20, the five-day stretch when China’s air pollution was at its worst in 2016, domestic consumers bought 110,000 air purifiers through its online marketplaces, up 210 percent year-on-year, according to data from JD.com Inc.
BEIJING, CHINA – NOVEMBER 15: Chinese women wear masks as haze from smog caused by air pollution hangs over the Forbidden City Photo by Kevin Frayer/Getty Images
Knowing this, we decided to use the sales volume of anti-pollution masks as an indicator of people’s reaction to smog. We gathered this monthly data from Hui Dian Shang–a trade analysis website dedicated to Taobao (china’s biggest e-commerce platform) and made a line chart.
Unfortunately, the sales data doesn’t give any information about which city those masks were sold to. Also, it only shows the sales volume of all types of masks, including anti-smog ones and regular ones. However, Taobao is China’s biggest online-shopping platform and is significant enough to analyze patterns from, although it is not completely representative on region-specific areas.
Because we also wanted to see if there’s any connection between each city’s economic status and its air quality, we chose the disposable income as an indicator. The data was collected from each city’s bureau of statistics.
source: national bureau of statistics of china
Additionally, we wanted to see if there was any relationship between population density and air quality. We collected population values for each region in 2016.
This shapefile was found for mapping purposes. We were able to map the boundaries of each province and highlight the 5 regions we are focusing our research on.
The province areas are important to be able to map to show where patterns of mobile activity occur as well as using to calculate how much activity was in each region.
The dataset is collected from TalkingData, a third-party mobile data platform in China that works with various mobile apps to log information every time a user opens one of the many integrated apps.
3 million data points (~1000 MB)
device ID’s (distinguishes unique users)
timestamps
coordinates
An essential and foremost thing we were thinking about was which was indicator we could use for representing urban activities. In the research paper done collectively by MIT and Tongji University (Yan, Duarte, Wang, Zheng & Ratti, 2018), researchers used the social media check-in data from Sina Weibo micro-blogging microform, the Chinese version of Twitter as the indicator of people’s urban activity data. As mobile devices are ubiquitous (Chinese person surveyed owns at least a basic mobile phone (98%)). We thus used the data from Kaggle. The Data is collected from TalkingData SDK integrated within mobile apps. TalkingData serves under the service term between TalkingData and mobile app developers. Kaggle data was collected based on full recognition and consent from individual user of those apps have been obtained, and appropriate anonymization have been performed to protect privacy.
Check-in data can be used as a proxy indicator for activity within a region. This is meant to be analyzed in relation to air quality over the same time period as the check-in dataset.
Smog in China 