HIGHLIGHTS:

• Earth observation data is increasingly being used as an alternative data source. 
• The initiative analyzed whether there were correlations between proxy indicators using EO data and economic activities. 
• Data acquired from this initiative, including the ships and containers in ports, can be used for further economic analyses.

__________

 

FAST FACTS:

Country: Georgia, Philippines, Republic of Korea 

Year of implementation: 2021

Technology: Remote sensing, GIS, Big data, artificial intelligence   

ADB Department: Economic Research and Regional Cooperation Department

ADB Partners: 

• European Space Agency
• Earthlab AI Systems—technology service provider 

In line with ADB’s Operational Priorities:

• Fostering regional cooperation and integration 
• Global and regional trade and investment opportunities expanded 
• Regional public goods increased and diversified
• Strengthening governance and institutional capacity
• Strengthened public management and financial stability

 

__________

 

BACKGROUND  

Earth observation (EO) uses remote sensing technologies to gather imaging data on land, marine resources, and the atmosphere.1 While EO is not new, various developments have occurred in the past years, including additional investments in satellite capabilities, advancements in algorithms and data processing tools, as well as increasing data accessibility.2 The COVID-19 pandemic further accelerated its use as more governments and organizations are exploring alternative data sources with the pandemic affecting the collection and sharing of other data.  

 

Earthlab AI Systems, which was the team selected in the “Earth Observation Data Challenge”, explored whether it was possible to use different sources of data collected by satellite missions for correlating and predicting economic activities.3 

 

SOLUTION 

The original intent was to have an integrated dashboard that can be used to monitor multiple economic indicators across different countries. However, the team decided during implementation to focus instead on investigating whether there were correlations between proxy indicators using EO data and economic activities in three countries: Georgia, the Philippines, and Republic of Korea. 

 

TECHNOLOGY 

Besides remote sensing and GIS mapping, the team used artificial intelligence (AI), specifically machine learning, and big data to make sense of the earth observation images taken by satellites. The amount of earth observation data collected by satellites is huge. Making sense of the information can quickly become overwhelming, particularly given the extent and capacity of humans to interpret data.4 Machine learning can be used to detect patterns and similarities in large amounts of data. It can also be used to automate the performance of repetitive and menial tasks.  

  

RESULTS 

1. Ships in ports. Some ports showed significant positive or negative correlations. Areas that have shown highly significant positive correlations between the number of ships in their ports and economic indicators included the Port of Davao in the Philippines and the Port of Gwangyang in Republic of Korea. The Port of Busan, meanwhile, showed a negative correlation between the average monthly number of ships in the port and export as an economic indicator. Other factors may affect the presence of ships in ports. For instance, the movement restrictions in the first half of 2020 saw an increase in the number of ships in the Port of Manila because of quarantine restrictions.  

 

2. Nitrogen dioxide emissions. While all calculated correlation coefficients were positive and some were statistically significant, results were inconclusive on whether there was a significant correlation between nitrogen dioxide measurements and economic indicators. This could be due to various reasons, including the complexity of the relationship between nitrogen dioxide and the different economic indicators (including the impact of other variables) and meteorological conditions (such as precipitation and wind). It is also possible that the time series of three years was too short. There was also a clear seasonal effect on the nitrogen dioxide concentrations across all three countries.  

 

3. Containers in ports. The small sample size may have contributed to inconclusive results. There was a strong but not significant negative correlation between the number of containers and the import and export activity. This may have been affected by the port’s space utilization. For instance, containers may have piled up when trade slowed down. As with the ships, different ports have different dynamics and contexts, so generalization may not be possible to explain the number of containers vis-à-vis trade volume.  

 

There were some challenges with the manual annotations. The current resolution of 3 meters per pixel resolution was not sufficient to effectively determine what objects were shown in some of the images. Even with the high resolutions of the images, some sections were either very bright or had shadows, making interpretation difficult. 

 

4. Nighttime light. The results were inconclusive. Further research may show the relationship between economic activities and nighttime light. It is also possible that the available instruments at the time the satellite images were collected were not sensitive enough to capture the subtle differences in nighttime light. Additional calibration, cleaning, and processing of the data, which was not covered by the initiative, may be needed as the available images did not seem to show significant differences between the average pixel radiance per month. Another possible explanation is that there were other variables that influenced the relationship between economic activity and nighttime light.