Building Affordable Air Quality Monitoring Systems for Data-Scarce Urban Areas

This World Bank study explores cost-effective air quality monitoring in low-income settings, finding that combining one regulatory-grade monitor with a network of calibrated low-cost monitors provides accurate, actionable pollution data. This hybrid model enables effective policy impact assessment and better health outcomes in resource-limited urban areas.

A recent paper by researchers Bridget Hoffmann and Sveta Milusheva, from the Development Impact Group at the World Bank, investigates the design of air quality measurement systems in regions with limited data resources, focusing on lower-income countries like Senegal. As urban populations in low- and middle-income nations are increasingly exposed to dangerous levels of air pollution, the lack of sufficient data for accurate assessment and policy intervention remains a significant challenge. While most existing air quality research focuses on high-income countries with ample monitoring data, this study centers on the city of Dakar, Senegal, to assess how different types of air quality data sources regulatory-grade monitors, satellite data, and low-cost monitors perform in measuring fine particulate matter (PM2.5) and evaluating the impact of air quality policies.

Air Quality Challenges in Low-Income Regions

Dakar, a growing urban center with substantial industrial activities, faces severe air pollution exacerbated by traffic, construction, and occasional Saharan dust storms. As in many other low-income urban settings, the high costs of deploying and maintaining regulatory-grade air quality monitors restrict their usage to just a few locations, resulting in sparse data collection that makes it difficult to obtain a comprehensive view of air quality conditions. Regulatory-grade monitors considered the “gold standard” in air quality monitoring, provide accurate and precise measurements. However, each unit requires a considerable financial investment, both in terms of installation and regular maintenance. In comparison, satellite data offers broad coverage and can be used to estimate pollution levels across large areas by measuring the density of particulate matter in the atmosphere. However, satellite data does not directly capture ground-level pollution, and its lower spatial and temporal resolution limits its effectiveness in capturing rapid pollution changes or high pollution peaks. As a more affordable alternative, low-cost monitors have been gaining popularity, but they present accuracy issues, especially when measuring pollutants like dust in urban environments where air pollution often has complex sources.

Testing a Low-Cost Network in Dakar

To better understand how these different monitoring sources work in practice, the research team established a network of 28 low-cost Purple Air monitors across Dakar, allowing for a high-density, spatially diverse data collection network. These monitors, while cost-effective, showed limitations in accurately measuring PM2.5 levels, particularly during dust storms. To enhance accuracy, the researchers employed calibration techniques, comparing data from low-cost monitors with both regulatory-grade monitors and satellite data. Specifically, they used a context-sensitive calibration approach, adjusting the low-cost monitor readings to more closely match those of regulatory-grade monitors. Through this method, the calibrated low-cost monitors produced data that better aligned with the more expensive regulatory monitors while still providing the advantage of greater spatial coverage. Satellite data, although consistent with regulatory-grade monitors at daily and weekly averages, showed misalignment at hourly levels due to its inability to capture fast, significant changes in pollution that are more evident in high-resolution ground-based monitors.

Measuring the Pandemic’s Pollution Drop

An opportunity to evaluate how these data sources could measure policy impacts arose during the COVID-19 pandemic, when Senegal’s mobility restrictions created a natural experiment to observe changes in air quality as a response to reduced traffic and industrial activities. A difference-in-differences analysis comparing air quality before and after the lockdown policies and against other periods revealed a significant drop in PM2.5 levels across Dakar following the restrictions, with all three data sources indicating similar patterns of reduction. The low-cost monitors, once calibrated, displayed results in line with the regulatory and satellite data regarding the percentage reduction in pollution, even though the absolute PM2.5 levels measured were lower than those recorded by regulatory monitors. The satellite and regulatory monitors showed a reduction of around 24-25% in PM2.5 concentrations, while the calibrated low-cost monitors closely mirrored this trend. This demonstrates that, despite differences in data sources, all three could reliably measure the impact of city-wide policies like lockdowns that affect pollution levels on a larger geographic scale.

Creating a Hybrid Monitoring Model

This research points to a viable solution for air quality monitoring in resource-limited settings: combining a single regulatory-grade monitor with a dense network of calibrated low-cost monitors. By maintaining one high-quality regulatory monitor to provide accurate calibration data, a low-cost network can offer substantial benefits, allowing for both spatial and temporal coverage without the prohibitive costs associated with regulatory-grade networks. This model could be especially beneficial in low-income cities with similar pollution sources, where dense coverage of regulatory-grade monitors is financially unfeasible. In scenarios where regulatory-grade data is unavailable, satellite-based calibration of low-cost monitors may serve as an alternative, albeit with less precision in tracking rapid pollution changes.

Empowering Policy with Air Quality Data

The findings from Dakar highlight that strategically combining data sources can yield accurate, actionable insights into urban air quality in low-resource environments. As low-cost monitors become more sophisticated, integrating them into urban air quality networks, especially with one or two regulatory monitors for calibration, offers a sustainable model for cities in low- and middle-income countries to monitor air pollution effectively. This study suggests that such hybrid systems can capture pollution trends and evaluate policy impacts even in data-scarce settings. Ultimately, as air quality monitoring data become more accessible, they could drive more targeted and impactful environmental policies, leading to improved health outcomes in rapidly urbanizing, low-income regions worldwide.