Monday, December 11, 2023

North Korea and beyond: AI-powered satellite analysis reveals the unseen economic landscape of underdeveloped nations

THE KOREA ADVANCED INSTITUTE OF SCIENCE AND TECHNOLOGY (KAIST)

IMAGE:
IMAGE 1. NIGHTLIGHT SATELLITE IMAGES OF NORTH KOREA (TOP-LEFT: BACKGROUND PHOTO PROVIDED BY NASA'S EARTH OBSERVATORY). SOUTH KOREA APPEARS BRIGHTLY LIT COMPARED TO NORTH KOREA, WHICH IS MOSTLY DARK EXCEPT FOR PYONGYANG. IN CONTRAST, THE MODEL DEVELOPED BY THE RESEARCH TEAM USES DAYTIME SATELLITE IMAGERY TO PREDICT MORE DETAILED ECONOMIC PREDICTIONS FOR NORTH KOREA (TOP-RIGHT) AND FIVE ASIAN COUNTRIES (BOTTOM: BACKGROUND PHOTO FROM GOOGLE EARTH).
view more
CREDIT: IBS DATA SCIENCE GROUP

The United Nations reports that more than 700 million people are in extreme poverty, earning less than two dollars a day. However, an accurate assessment of poverty remains a global challenge. For example, 53 countries have not conducted agricultural surveys in the past 15 years, and 17 countries have not published a population census. To fill this data gap, new technologies are being explored to estimate poverty using alternative sources such as street views, aerial photos, and satellite images.

The paper published in Nature Communications demonstrates how artificial intelligence (AI) can help analyze economic conditions from daytime satellite imagery. This new technology can even apply to the least developed countries - such as North Korea - that do not have reliable statistical data for typical machine learning training.

The researchers used Sentinel-2 satellite images from the European Space Agency (ESA) that are publicly available. They split these images into small six-square-kilometer grids. At this zoom level, visual information such as buildings, roads, and greenery can be used to quantify economic indicators. As a result, the team obtained the first ever fine-grained economic map of regions like North Korea. The same algorithm was applied to other underdeveloped countries in Asia: North Korea, Nepal, Laos, Myanmar, Bangladesh, and Cambodia (see Image 1).

The key feature of their research model is the "human-machine collaborative approach," which lets researchers combine human input with AI predictions for areas with scarce data. In this research, ten human experts compared satellite images and judged the economic conditions in the area, with the AI learning from this human data and giving economic scores to each image. The results showed that the Human-AI collaborative approach outperformed machine-only learning algorithms.

< Image 1. Nightlight satellite images of North Korea (Top-left: Background photo provided by NASA's Earth Observatory). South Korea appears brightly lit compared to North Korea, which is mostly dark except for Pyongyang. In contrast, the model developed by the research team uses daytime satellite imagery to predict more detailed economic predictions for North Korea (top-right) and five Asian countries (Bottom: Background photo from Google Earth). >

The research was led by an interdisciplinary team of computer scientists, economists, and a geographer from KAIST & IBS (Donghyun Ahn, Meeyoung Cha, Jihee Kim), Sogang University (Hyunjoo Yang), HKUST (Sangyoon Park), and NUS (Jeasurk Yang). Dr Charles Axelsson, Associate Editor at Nature Communications, handled this paper during the peer review process at the journal.

The research team found that the scores showed a strong correlation with traditional socio-economic metrics such as population density, employment, and number of businesses. This demonstrates the wide applicability and scalability of the approach, particularly in data-scarce countries. Furthermore, the model's strength lies in its ability to detect annual changes in economic conditions at a more detailed geospatial level without using any survey data (see Image 2).

< Image 2. Differences in satellite imagery and economic scores in North Korea between 2016 and 2019. Significant development was found in the Wonsan Kalma area (top), one of the tourist development zones, but no changes were observed in the Wiwon Industrial Development Zone (bottom). (Background photo: Sentinel-2 satellite imagery provided by the European Space Agency (ESA)). >

This model would be especially valuable for rapidly monitoring the progress of Sustainable Development Goals such as reducing poverty and promoting more equitable and sustainable growth on an international scale. The model can also be adapted to measure various social and environmental indicators. For example, it can be trained to identify regions with high vulnerability to climate change and disasters to provide timely guidance on disaster relief efforts.

As an example, the researchers explored how North Korea changed before and after the United Nations sanctions against the country. By applying the model to satellite images of North Korea both in 2016 and in 2019, the researchers discovered three key trends in the country's economic development between 2016 and 2019. First, economic growth in North Korea became more concentrated in Pyongyang and major cities, exacerbating the urban-rural divide. Second, satellite imagery revealed significant changes in areas designated for tourism and economic development, such as new building construction and other meaningful alterations. Third, traditional industrial and export development zones showed relatively minor changes.

Meeyoung Cha, a data scientist in the team explained, "This is an important interdisciplinary effort to address global challenges like poverty. We plan to apply our AI algorithm to other international issues, such as monitoring carbon emissions, disaster damage detection, and the impact of climate change."

An economist on the research team, Jihee Kim, commented that this approach would enable detailed examinations of economic conditions in the developing world at a low cost, reducing data disparities between developed and developing nations. She further emphasized that this is most essential because many public policies require economic measurements to achieve their goals, whether they are for growth, equality, or sustainability.

The research team has made the source code publicly available via GitHub and plans to continue improving the technology, applying it to new satellite images updated annually. The results of this study, with Ph.D. candidate Donghyun Ahn at KAIST and Ph.D. candidate Jeasurk Yang at NUS as joint first authors, were published in Nature Communications under the title "A human-machine collaborative approach measures economic development using satellite imagery."

Image 2. Differences in satellite imagery and economic scores in North Korea between 2016 and 2019. Significant development was found in the Wonsan Kalma area (top), one of the tourist development zones, but no changes were observed in the Wiwon Industrial Development Zone (bottom). (Background photo: Sentinel-2 satellite imagery provided by the European Space Agency (ESA)).

CREDIT

IBS Data Science Group

JOURNAL

Nature Communications

DOI

10.1038/s41467-023-42122-8

METHOD OF RESEARCH

Meta-analysis

SUBJECT OF RESEARCH

Not applicable

ARTICLE TITLE

A human-machine collaborative approach measures economic development using satellite imagery

The energy footprint of architecture built by oil

After the oil boom, energy efficiency a low priority for buildings in Arab Gulf states countries

Peer-Reviewed Publication

NORWEGIAN UNIVERSITY OF SCIENCE AND TECHNOLOGY

Typical housing types in Arab Gulf countries — IMAGE:
THESE ARE SOME OF THE EXAMPLES OF BUILDING TYPES THE RESEARCHERS EXAMINED AS THEY PUT TOGETHER ENERGY AND MATERIAL USE ASSESSMENTS OF THE ARAB GULF STATES.
view more
CREDIT: GRAPHIC: SAHIN AKIN, NTNU

The six Middle East countries in the Gulf Cooperation Council are located in one of the harshest places on the planet. Temperatures on hot summer days can top 50 degrees C. Access to water in these desert lands is also challenging.

Now, as leaders from across the globe gather in Dubai for international climate talks, a new report shows how the discovery of oil in the region radically changed housing stocks, enabling a shift from traditional to energy-intensive Western architecture. As a result, four of six Gulf States now rank among the world’s worst greenhouse gas emitters, measured on a per person basis.

This high share of residential energy use is very unusual for countries with this level of income. Here we have a high-consumption economy.

“These countries are wealthy and electricity prices are subsidized by the government. So instead of investing in insulation or high quality materials, they just turn up their air-conditioning systems,” says Sahin Akin, a PhD candidate at the Norwegian University of Science and Technology's (NTNU) Industrial Ecology Programme (IndEcol).

Akin’s research was part of a larger effort at IndEcol to look more closely at regional differences in residential buildings worldwide. Residential buildings worldwide matter, because they account for fully 21 per cent of global energy use, and the materials used in their construction are responsible for 30 per cent of the global carbon footprint for overall materials production.

In this case, understanding how different kinds of housing drive energy and materials use can help policy makers adopt specific measures to help lower each country’s greenhouse gas emissions.

Building types important

Akin and his colleagues can’t look at every individual residential unit, of course, because that would be impractical. So what they did instead was to group related kinds of housing units into average types, which they called archetypes. All told, the researchers created 153 simulation models, boiled down to 6 different types, to describe typical housing in the region.

Before oil wealth changed the economic landscape in these countries, many people lived in traditional housing, Akin said.

“These were built by using traditional construction techniques and using local materials such as mud, adobe, and wood in some cases. So they were really good in terms of environmental performance,” he said, mainly because the structures were designed to cool passively, without the aid of air-conditioning, but also because they didn’t require the import of building materials.

Oil wealth, urbanization and the growth of the villa

The arrival of oil wealth resulted in more rapid urbanization in many of the countries, and the increased construction of a housing type that the researchers call a villa.

These structures can be two or three floors high, and can have separate entrances for men and women, in keeping with Muslim practice. There may also been housing for hired workers, such as housemaids and child care providers.

Population in the region went from 7.7 million in 1970 to 59.5 million in 2022, and is expected to jump to 73.4 million by 2050, according to figures from the United Nations.

The result of this trend is that while villas only accounted for 22 per cent of the total number of units in the Gulf region, they accounted for more than half of the total floor area. They tended to be very spacious, with an average of 62 m2 per person, which is nearly double the world average of 32 m2 per person.

And they can be energy intensive, Akin said.

Akin’s analysis found that among more populous countries such as Saudi Arabia and the UAE, the total amount of energy use was high, while smaller countries with higher GDP per capita, like Kuwait, Qatar, and Bahrain, had higher energy use per capita.

“This trend could be linked to the relatively high percentage of villas in the building stocks of these small countries, which use more energy than other building types,” he said.

Population growth a factor, too

Oil wealth also drove another important trend – population. Population in the region went from 7.7 million in 1970 to 59.5 million in 2022, and is expected to jump to 73.4 million by 2050, according to figures from the United Nations.

Oil wealth allowed more children to survive, and with a fertility rate of 6 births per woman, led to a quick growth in the native population. In addition, Akin says, oil wealth attracted immigrants in search of jobs. Workers were also recruited by families to work in households. While some villas might have housing for household help, many expatriates were housed in apartments or condominiums.

“We see these two distinct groups in the rich countries, expats living in condos and natives living in villas. These have differences in terms of energy use and materials, mostly by having more living space per person in villas,” said Edgar Hertwich, the senior author on the article.

Regulations can have an impact

Newly built villas with much more living space were generally the most energy intensive, but there was one surprising exemption, Akin said, which gives hope to the idea that regulations can have a positive effect.

Akin and his colleagues found that although Oman had a relatively high percentage of villas in its building stock, it had much lower energy use per capita than the other Gulf countries.

“Oman is different than the other countries because they tend to stick with the cultural habits of the past, with regulations that promote Islamic architecture,” he said.

Building regulations specify that building exteriors should be white, for example, which helps reflect the sun’s rays. That helps reduce the amount of solar heating, he said.

In contrast, Bahrain, despite being the smallest and least populous country in the region, had higher per capita energy use than the UAE, Saudi Arabia and Oman, Akin and his colleagues reported. They speculated that this was due to a number of factors such as lifestyle, climate, and building design.

Materials also important

When asked about the single most important finding of his research, Akin had a simple answer: Concrete.

Yes, concrete, or more exactly, its main component, cement, which with its production contributes nearly three times as much to global warming as air transport.

“The concrete use in all of those countries are a little bit concerning, because it dominates the entire building stock, regardless of the country,” Akin said. “Even the traditional housing styles are also constructed with concrete in recent years because it's much more affordable and it's easy to use, it doesn't require specific skills.”

Aside from the actual carbon dioxide release from making cement, concrete has another feature that makes it a climate bad guy.

“Concrete has low environmental performance compared to traditional materials,” Akin said.

For example, concrete structures in the region are rarely insulated, unless there are mandatory building regulations that require it. And with low energy prices, it can be easy and cheap to just turn up the AC, he said.

In this context, traditional houses built with local materials are best, since the materials used for their construction generally have the lowest emissions. Houses that are built using traditional methods also use less energy because they are designed to work in the hot climate.

Unusual use linked to high consumption, population

Akin reported on a study that showed that implementing energy conservation measures in Gulf States’ residential buildings could cut energy use by up to 70 per cent. That’s no small percentage, given that housing overall is responsible for fully 70 per cent of all energy use in the region.

“This high share of residential energy use is very unusual for countries with this level of income,” Hertwich says. “We find similar shares in developing countries where subsistence production is important, and residences double as places of work. This is not the case in the Gulf. Here we have a high-consumption economy.”

Another concern is demography. As the fertility declines below replacement levels, smaller and smaller families come to occupy large villas. More floor area per person naturally means higher energy use per person.

The way forward

Akin and his colleagues hope these findings can make a difference in helping individual countries find ways to cut their energy and materials use.

One clear message is that given the region’s climate — hot and sunny — there is significant potential for renewable energy, such as thermal solar, photovoltaic solar and geothermal energy, he said.

Akin and his co-authors say their results can help identify opportunities to cut emissions and develop resource-efficient construction practices and tailored policies for improving the resource efficiency of buildings in the region.

“To mitigate the climate challenges of the future, we first need to have an understanding about the current situation of what's going on in those countries. And then we can address them much more effectively by developing some targeted scenarios for particular building typologies like villas, apartments or skyscraper flats,” he said. “And the ratio among the building stocks is changing across those countries. So we believe that country-specific studies are quite important.”

A modern Oman villa, but painted white and incorporating at least some characteristics of more traditional structures.