Nitrogen Dioxide (NO2) is a direct byproduct of combustion and its atmospheric concentration has been used as a measure of economic activity. Its short atmospheric lifetime — less than one day — means that observed densities remain closely correlated with their emission sources, making it a particularly localized and timely indicator. Global concentration of NO2 is available through satellite remote sensing datasets making it a valuable resource for many data-poor countries.
On the mobility data side, Waze collects real-time traffic information from its users and produces a jams dataset with records of traffic slowdowns at the road segment level, updated at a two-minute frequency. The World Bank processes this data to produce the Traffic Congestion Intensity Index (TCI), which summarizes both the spatial extent and duration of traffic jams within any polygon of interest.
Our hypothesis is that NO2 values can be used to predict changes in mobility — as captured by the TCI — and that this relationship can be exploited in countries where mobility data are unavailable. The research goal is twofold: first, to establish the statistical and geospatial relationship between NO2 and TCI in countries where both datasets are available; and second, to use the resulting model to extrapolate mobility estimates for countries where no direct mobility data exist.
The usage of nitrogen dioxide data as a proxy for economic trends is relatively new compared to other indicators such as nighttime lights which have been extensively used to track GDP. NO2 data is easily accessible and available globally through satellite imagery.
Economic activity and people’s mobility are highly intertwined. Research has also shown that NO2 and mobility are highly correlated. This relationship between pollution, transportation and the economy was explored especially in economic trend monitoring studies after COVID-19 .
This direction of research was chosen to build on last year's project. Your colleagues explored the relationships between NO2 and economic activity, as well as between NO2 and mobility more broadly. The current phase concentrates on refining and deepening that analysis, with a focus on the spatial prediction of mobility patterns across diverse geographies.
TCI is available for the following cities. If you pick cities outside of this, you are free to obtain mobility proxies from any other source where data is available.
- Kathmandu, Nepal
- Buenos Aires, Argentina
- Cairo, Egypt
- Jakarta, Indonesia
- Lima, Peru
- Manila, Philippines
- Bogota, Colombia
- New York, USA
- Mumbai, India
- Paris, France
- Qualitative analysis for Myanmar
The students from the previous batch of MSc Transport and Data Science worked on this research in the past. You can find the resources here:
The template maintains a Code of Conduct to ensure an inclusive and respectful environment for everyone. Please adhere to it in all interactions within our community.
The template is licensed under the MIT License together with the World Bank IGO Rider. The Rider is purely procedural: it reserves all privileges and immunities enjoyed by the World Bank, without adding restrictions to the MIT permissions. Please review both files before using, distributing or contributing.