Predicting What We Breathe

ML & Data Science Intermediate

The City of Los Angeles is working with NASA in predicting air quality in environmental justice neighborhoods. Using convolutional networks and neural networks, the machine learning algorithms are showing 92% or better prediction capability. This project has been open sourced with other cities around the world.

The City of Los Angeles has an extensive set of policies, practices, and data related to the mitigation of particulate matter and improvement of air quality. As the City deploys new policies or undertakes place-based solutions, we are measuring the impact these solutions have. Working with NASA, the City organizes this work as Predicting What We Breathe. This work is using ground-based smart city sensor networks augmented by NASA satellite data to create a robust set of data on PM2.5, ozone, and PM10. We have developed advanced machine learning algorithms and models that link ground-based in situ and space-based remote-sensing observations of major air quality components to (a) classify patterns in urban air quality, (b) enable the deduction and forecast of air pollution events related to PM2.5 and ozone from space-based observations, and ultimately (c) identify similarities in air quality regimes between megacities around the globe for improved air pollution mitigation strategies. This work helps us understand the correlation between air pollution and health conditions all over the City of Los Angeles, and predict individuals’ health risks related to air pollution based upon air quality measurements. Using the City of Los Angeles as a test case, we will extrapolate these algorithms to other mega-cities.