Comprehensive Monitoring of Air Pollution Hotspots Using Sparse Sensor Networks

What is it about?

High air pollution is a major public health risk, especially in densely populated cities like New Delhi, India. However, city governments often rely on a small number of expensive sensors that miss many dangerous "hotspots", areas with dangerously high levels of pollution. In this study, we added 28 low-cost sensors to the city’s existing network and collected data for over two years. We found nearly 200 hidden pollution hotspots that were not picked up by government sensors. We then used advanced mathematical techniques to fill in the gaps between sensors, allowing us to predict where pollution is highest even if sensors are missing. We also built a physics-based model to understand how pollution spreads and which sources, like traffic, brick kilns, or power plants, are most responsible. Our findings can help cities take smarter, more targeted actions to reduce pollution and protect public health, even when resources are limited.

Featured Image

Photo by abhijeet gourav on Unsplash

Photo by abhijeet gourav on Unsplash

Why is it important?

This study shows that cities can uncover critical pollution hotspots using a scalable and low-cost approach. We conducted a longitudinal deployment of low-cost air quality sensors across New Delhi for more than two years, generating a rich dataset that revealed nearly 200 pollution hotspots that were previously missed by the government network. What makes this work particularly timely is that governments, including New Delhi, are already adopting hotspot-based strategies for air pollution mitigation. However, the current hotspot identification efforts are constrained by sparse public sensor networks, which miss highly localized but impactful sources of pollution. Our methodology directly addresses this gap. We combine statistical interpolation and physics-based dispersion modeling to uncover hidden hotspots and trace them back to likely sources like traffic, domestic emissions, or industrial activity. Crucially, our approach works with minimal new infrastructure, our entire deployment cost was under $8,000, a fraction of the cost of traditional monitoring stations. By offering an affordable and extensible toolkit for hotspot detection, our work empowers local governments to make more informed, targeted interventions in communities facing the worst pollution. Cities already using hotspot frameworks can plug in our methods to generate fine-grained, actionable insights that were previously out of reach. This enables faster, data-driven responses to one of the most pressing public health issues in the developing world.

Perspectives