Delhi among 3 global cities part of Nasa's air pollution challenge

The National Aeronautics and Space Administration (Nasa) has partnered with a crowdsourcing platform to launch ‘Air-othon’, a competition aimed at utilising a combination of meteorological data, low cost-sensors and satellite imagery to develop models for air pollution in Delhi, Los Angeles and Taipei.

It has invited individuals and teams to analyse a 5x5 km grid in the three cities, and the data gathered in this manner will be shared with the US embassy in Delhi and the Delhi government. The winning team will be receiving a prize of $50,000.

Launched on Tuesday in Houston, Texas, the ‘Air-othon’ challenge is being undertaken with DrivenData, a social enterprise that uses data tools, and HeroX, a crowdsourcing platform, to generate daily estimates of surface-level NO2 and PM 2.5 across a 5km grid of the participant’s choosing.

While high NO2 levels can lead to respiratory and asthma issues and the formation of other harmful pollutants, such as ozone and particulate matter, PM2.5 (particulate matter less than 2.5 micrometers in size) is known to penetrate deep into human lungs, resulting in major health concerns in the long-term.

‘Air-othon’ officials said existing ground monitors have large gaps in coverage, which can be bridged using forecasting and real-time data models.

“This is a difficult problem where better data tools can provide timely information to millions of people to help them protect their health,” said Greg Lipstein, co-founder of DrivenData.

“The challenge will test solutions from a global community of experts, with the best approaches automatically rising to the top of the leaderboard,” he added.

Experts said models that make use of widely available, low-cost sensor data and satellite imagery also have the potential to provide daily local air quality information to thousands of people. “This effort prioritises the health of entire communities. I am eager to see the innovative ways the crowd can improve air quality data collection and enhance public health,” said Kal K Sahota, CEO, HeroX.

A Nasa official said the overarching goal of the challenge is to deliver top-performing solutions to the state department at designated embassy locations, including Delhi.

Professor SN Tripathi from IIT Kanpur, who is part of a CPCB committee that is looking at the utilisation of low-cost sensors, said satellite imagery can be used to identify NOx levels, while PM 2.5 data can be utilised from low-cost sensors and fed into a machine which analyses concentration levels at different time stamps. “Machine learning is possible and models can be developed that forecast concentration several hours in advance. Such learnings can be beneficial for India too,” he said.