China Night Lightmap

Observe Changes Before and After the COVID-19 Pandemic Using Satellite Images

 

Introduction

Chinese authorities officially suspended travel and closed business activities in late January 2020 in response to the rapid surge of the novel coronavirus. Satellite observations since then have revealed apparent changes in terms of city-level activities. Starting on December 12, 2019, the first cluster of patients was detected in Wuhan, Hubei province. Besides health issues, people's day-to-day lives have been affected in a myriad of ways. For instance, the lockdown of cities in China reduced the number of collective gatherings in public spaces. It also caused an instant reduction in tourism and transportation flow. The restrictions also curtailed economic production because workers could not return to their workplaces. Some manufacturers were shut down involuntarily. The changes varied depending on the outbreak timing and the restrictions put in place by cities and provinces. The GIS question I am interested in investigating is the difference between cities' nighttime brightness levels before January 2020 and afterward. The question that this GIS project tried to answer is the quantitative human activities through time. The period started before the pandemic to the present. The analysis of nighttime satellite images would be the primary tool and result to answer the question of depicting the changes. Nighttime lighting is a reliable and robust data source to observe broader patterns of a city's capacity to adapt to new living conditions. Still, it also provides fascinating insights into people's activities in adapting to COVID-related restrictions.

 

Methods & Analysis

 

Data Collection

Project Criteria

The primary data for this project comes from the Earth Observation Group, a monthly cloud-free DNB composite available on its website. The vcmsl config for the tile in 75N060E (Northeast Hemisphere) is the data set to cover the location of China.

Data Sources

Earth Observation Group

Global country boundaries

China - Subnational Administrative Boundaries

I used night light observations to track variations in cities' energy use based on whether light pollution increased or decreased, which highways were shut down, and which cities kept a status quo. I also included other economic and census indicators to help me better contextualize such spatial outcomes.

First, analyzing the timeline for all the restrictions and lockdowns helped me find a set of comparable timestamps. Using the monthly Day/Night Band (DNB) composite, I could quantify the average amount of light being projected. The monthly composite is relatively meaningful for undertaking year-to-year comparisons. To choose which month to analyze, I divided the timeline into three periods:

 

Methodology

  1. Removed the outliers and classified the brightness from 0 to 50 for the country's scale. Used the single-band color palette to show the map and generated maps for February 2019, February 2020, and February 2021.

  2. Selected cities for analysis (Wuhan, Shanghai, Shenzhen, Baotou)

  3. Clipped the tiled image for four cities and for months from January to April during 2019 to 2022.

  4. Classified the brightness from 0 to 10 for the city's scale for a better visual representation.

  5. Analyzed the mean brightness for four cities by zonal statistics.

  6. Compared the mean and created bar charts for four cities.

  7. Researched cities' GDP growth rate and census data to testify the bar charts.

  8. Compared the brightness outside of cities to notice the impact of transportation.

  9. Researched the events during the pandemic in Wuhan since Wuhan had an opposite outcome to other cities.

  10. Calculated the percentage changes in February by subtracting the mean from 2020 to 2019 and graduated by the same color theme to make the thematic map.

  11. Listed the top five increasing and decreasing cities.

  12. Repeated steps 10 and 11 for the changes from 2020 to 2021.

 

City Selection

In order to compare fluctuations in terms of city-level activities and their changes throughout the pandemic timeline, I concentrated this analysis on distinct cities that include Wuhan, Shanghai, Shenzhen, and Baotou. Wuhan was the city where the first Covid case was found. With a longer lockdown and more restricted rules for public activities, we should expect to observe dynamic changes in the city. Shanghai is the most populated city in China, meaning we expect to observe the most dramatic changes in human activities. Besides Shanghai, Shenzhen is the largest city for manufacturing, with a high foreign population and tons of manufacturers located in the Pearl River Delta. We expect to observe the brightest set of colors from the satellite imagery. The last city is Baotou, located in northern China. It belongs to Inner Mongolia, which has the largest amount of rare-earth metals globally and is also an excellent comparator to observe the changes brought by the pandemic.

 

Brightness at the Country’s Scale

Results

Comparing the brightness on an annual basis, we can observe this brightness indicator increasing from 2019 to 2022 both at a national level and at a city level. Human activities bounced back to the highest level in 2022. Even by looking at the result for only four months, we can conclude that it corresponds to the GDP growth rate.

On the other hand, there are some exciting outcomes if we compare the results monthly. Cities' nighttime brightness before January 2020 and after varies depending on which city was being observed. The key results and related observations are as follows:

 

Brightness at Cities’ Scale

Results

  • The brightness in Shanghai shows a stable trend in human activities from 2019 to 2022. The drop in February 2020 is not significant.

  • The brightness level sharply drops in April 2022, corresponding to the lockdown timeline since this is the first and the most strict lockdown since 2020.

 
 
 

Results

  • The brightness increased after the lockdown on January 23 and decreased in March. The reasons can be various. First, the floating population is low compared to tier-1 cities, ex. Shanghai and Shenzhen. Therefore, not many people traveled out before the lockdown. Instead, people might migrate back to Wuhan before the Chinese New Year.

  • Second, the increase in February might impact the existing population, plus the medical workers, suppliers, and supporters. According to the National Health Commission of the People's Republic of China, more than 38,000 medical workers were sent to Wuhan before March 2020.

  • Without access to additional information, it is unclear why there was a drop in brightness levels between January 2021 to April 2021. However, we notice the same pattern at the country level. The observed rebound is higher in 2022.

  • By looking at the reference map, we found the answer for highlighting roads connecting to Wuhan downtown. One of the locations was the temporary mobile cabin hospital for Covid-19 patients.

  • Interestingly, there was a lighting show on April 30 to celebrate lifting the ban. The aerial photograph shows the lighting from the city, which has the "Wuhan Fighting" slogan in February.

 
 

Results

  • During the four months in 2020, we notice a significant drop in March, corresponding to the restriction that started in March. Many people could not return to work because of the strict enterprise resumption report system. According to the website www.gov.cn, the return to work rate was only 68.7%.

  • After March 2020, the trend resumed, and it bounced back to the highest level in April 2022. This phenomenon is also related to the high number of floating populations.

  • Transportation near the Pearl River Delta decreased in 2020.

  • The 2019 to 2022 trend shows the centralized human activities in cities in 2019, 2021, and 2022. The purple color displays the dispersed distribution.

 
 
 

Results

  • With fewer permanent residents, Baotou has less lighting compared to other cities.

  • March 2020 has the lowest brightness of the analysis months. Compared to the increasing rate of the country's level, the brightness level in 2020 was supposed to be higher. However, the pandemic impact was negatively significant in the city. The brightness changes from 2019 to 2021 corresponded to its GDP growth rate.

  • We can notice the same outcome apparently from the Bayan Obo mining district.

  • We can observe further disconnections in terms of transportation/ commuting flows between cities and towns from neighboring areas during 2020.

 

Brightness at Provinces’ Scale

 
 

Takeaways

 

From the country’s scale, we see the brightness increasing yearly.

Overall the decreasing provinces are in the north, where the heavy industry is located. The impact of the pandemic based on the industrial structures.

 

During the lockdown, the city did not shut down its function. More people separately staying at home means more lighting consumption in houses instead of roads.

Cities in the south have more tolerance for upheaval and are more flexible.

Wuhan has a different pattern from other cities. The brightness increased in Feb. 2020 shows the ambulance were buzy in the city.

 

Reference

 

Meghan Bartels, MB. (2020). New satellite views show impact of coronavirus on emissions, China's night lights. Space.com.

https://www.space.com/coronavirus-


The World Bank. (January 26, 2015). East Asia's Changing Urban Landscape: Measuring a Decade of Spatial Growth. 

https://www.worldbank.org/en/topic/urbandevelopment/publication/east-asias-changing-urban-landscape-measuring-a-decade-of-spatial-growth


AP News. (January 22, 2021) Timeline: China's COVID-19 outbreak and lockdown of Wuhan

https://apnews.com/article/pandemics-wuhan-china-coronavirus-pandemic-e6147ec0ff88affb99c811149424239d


COVID-19 lockdown in China. (n.d.). Wikipedia. Retrieved Month Day, Year, from

https://en.wikipedia.org/wiki/COVID-19_lockdown_in_China#2020_lockdowns


Nighttime Lights. (2021, May 18). Earth Data NASA.

https://earthdata.nasa.gov/learn/backgrounders/nighttime-lights


Nighttime Images Capture Change in China. (n.d.). NASA Earth Observatory. Retrieved Month Day, Year, from

https://earthobservatory.nasa.gov/images/146481/nighttime-images-capture-change-in-china


Emily Cassidy, EC. (2020, April 3) Nighttime Images Show Changes in Human Activity

https://earthdata.nasa.gov/learn/articles/feature-articles/nighttime-images-wuhan


Bayan Obo Rare Earth Mine. (n.d.).  NS Energy. Retrieved Month Day, Year, from

https://www.nsenergybusiness.com/projects/bayan-obo-rare-earth-mine/


包头市第七次全国人口普查公报 2020 [Baotou Seventh National Census communique 2020].

红黑统计公报库.

https://tjgb.hongheiku.com/13665.html 


武汉市第七次全国人口普查公报 2020  [Wuhan Seventh National Census communique 2020]. 武汉市人口统计局.

http://tjj.wuhan.gov.cn/ztzl_49/pczl/202109/t20210916_1779157.shtml 


深圳市第七次全国人口普查公报[1](第一号)——全市常住人口情况 2020. [communique of the Seventh National Population Census of Shenzhen [1] (No. 1) - the city's permanent population 2020]. 深圳市统计局.

http://tjj.sz.gov.cn/ztzl/zt/szsdqcqgrkpc/szrp/content/post_8772112.html


深圳市第七次全国人口普查公报[1](第六号)——流动人口[2]情况 2020. [communique of the Seventh National Population Census of Shenzhen [1] (No. 6) - Floating Population 2020]. 深圳市统计局.

http://tjj.sz.gov.cn/ztzl/zt/szsdqcqgrkpc/szrp/content/post_8772123.html


“始终把人民群众生命安全和身体健康放在第一位”——中国抗疫彰显“生命至上、人民至上”理念. ["Always put people's life safety and health first" - China's anti-epidemic highlights the concept of "life first, people first"]. National Health Commission of the People's Republic of China.

http://www.nhc.gov.cn/xcs/fkdt/202005/c9eb389e93e44c02a8c1b978ab361abe.shtml


Brian, M., KIM, B., TREVOR, M., RAN G., BENJAMIN, S., WALKER, KB., & DAYNAN, C. (2021, January 21). Light Every Night – New nighttime light data set and tools for development.

https://blogs.worldbank.org/opendata/light-every-night-new-nighttime-light-data-set-and-tools-development


Over 38,000 medical workers sent to Wuhan to fight virus. (2020, March 24). National Health Commission of the People's Republic of China

http://en.nhc.gov.cn/2020-03/24/c_80407.htm


范梓萌. 深圳规上工业企业开工率超97% 复工复产进入“快车道”  [The operating rate of industrial enterprises above designated size in Shenzhen exceeds 97%, and the resumption of work and production has entered the "fast lane"]. (2020, March 14). The State Council The People’s Republic Of China.

http://www.gov.cn/xinwen/2020-03/14/content_5491266.htm