Project description

China’s population problem has been receiving widespread attention. With urbanization process in full swing, the mass migration of people from the countryside to towns and cities is causing traditional villages “hollowing out”. Meanwhile, many towns and cities are actually losing their population and becoming so called “shrinking cities”. Netease Datablog collected and analyzed demographic data of 633 Chinese cities and discovered that nearly 40% of them are experiencing population loss.

Firstly, we calculated net migration rate of inter-provincial population movements based on the data published in the The Fifth and Sixth National Population Census, and then discovered more than half of China’s provinces had population loss, with Anhui Province experienced the highest net outflow of migrants. The concentration of population rose in the eastern region, while people in the central and western regions continued to move out.

Then we obtained urban demographics of all cities in China from the China City Statistical Yearbook 2010 and 2016. After removing cities which administrative divisions have changed, we finally got 633 cities and divided them into six groups according to their population densities. Among them, the population density of 246 cities dropped.

We also divided these cities into five categories based on urban population. The population density of four biggest cities with urban population over 10 million dropped due to Beijing’s strict population control policy.

According to research, there are five types of shrinking cities: ① The decline and shrinkage of industrial and mining cities; ② Local shrinking in megacities; ③ Shrinking in underdeveloped areas; ④ Shrinking in border cities and remote cities; ⑤ Statistical shrinkage caused by administrative division adjustment.

Although many cities are shrinking, “growth” is still the theme of China’s urbanization. The speed of urban expansion is clearly much faster than urban population growth. Even so, many Chinese cities still have crazy expansion plans and they are expecting higher population growth in the future.

The last chart shows that more than 3,000 new towns and new districts are planned all over China, many of them has already become “Ghost Town”. If we add their population growth plans up, there will be more than 3.4 billion people, which is 2.5 times than China’s current population and half of the global population.

(This peoject was originally published on 31/12/2018. This is the new version with interaction design.)

What makes this project innovative?

Shrinking cities has been a hot topic in architecture and planning area in past years. But it is too academic for general readers. We did a lot of research and made the concept much easier to understand. To demonstrate this amount of data more effectively, we created an interactive map to show population density change with color, and readers can also easily search their hometown.

What was the impact of your project? How did you measure it?

This original article was viewed more than a million times on our publishing platforms combined, and was reproduced by more than 140+ media and self-media. The founder of Beijing City Lab, China's earliest research institution on "shrinking cities" , thanked us for raising the issue to the public and pushing urban research to a good way. The story was also noticed by Chinese central government. They asked us to provide all the data and references we use to several ministries. We were also told that the government was doing research on "shrinking cities".

Source and methodology

China Urban Construction Statistical Yearbook, Urban Statistical Yearbook of China, China's New Town and New District Development Report, Academic papers

Technologies Used

Excel is used for data storage and cleaning, and Adobe Illustrator and D3 is used for data visualization. We use the Application Programming Interface of the AutoNavi Company to draw the interactive map of China. In the process, we apply different colors to the map using two layers in the levels of nation and province. The municipality directly under the central government, prefecture-level city and county-level city are considered respectively. The municipality and prefecture-level city are painted in the level of nation at first. Then the county-level cities are painted above them while other places that need not be painted are transparent. Besides, we have appended the function of information frame. When the readers search for or click on some place, we can get the latitude, longitude and name of the city according to the API as well and send back to readers with the information of the city. We put a navigation bar through the combination of jQuery and SVG in order to show the progress of reading, and anyone could focus on where they like by clicking it. We make a responsive webpage to adapt to different types of devices from personal computer to smart phone. Due to the requirement of specific animation and interaction, some targeted algorithms were designed to promote a high level user experience.

Project members

Qiuhong Zhang, Penglu Zhao, Xiaojing Guo, Xiaoqing Ma, Yingbo Li, Zhen Ye, Jinyu Cui, Zhen Wang, Yusong Wu. Xiaoyuan Sun and Junyi Li also contributed to this project.


Additional links


Click Follow to keep up with the evolution of this project:
you will receive a notification anytime the project leader updates the project page.