Despite successful pilot, MPCB ditches low-cost air quality monitors in city

Between November 2020 and May 2021, as many as 40 SAAQMs produced by four Indian companies were installed alongside existing CAAQMS. This was part of the pilot project called ‘Technical Assessment of Low-Cost Sensor based PM2.5 and PM10 Monitoring Network in Maharashtra.

Mumbai: Despite once calling its pilot project of using low-cost air quality monitors across the city a success, the Maharashtra Pollution Control Board (MPCB) has decided to ditch the idea of using them to support its limited network of costlier continuous ambient air quality monitoring stations (CAAQMS) across the Mumbai Metropolitan Region (MMR).

To give you an idea of how SAAQMs and CAAQMS function, the former only measures concentrations of particulate matter (PM2.5 and PM10), which is on most days the primary pollutant across India and the first visible sign of poor air. When compared with nearby CAAQMS, sensors from three start-ups saw less than 25% error for uncalibrated values. (HT PHOTO)

Experts said that the decision to not install the low-cost devices, also known as Sensor Ambient Air Quality Monitors (SAAQMs) in Mumbai is a major setback for the city where winter pollution levels in 2022-23 were at their highest in four years.

It remains unclear why the MPCB has abandoned a project it once declared a success.

Between November 2020 and May 2021, as many as 40 SAAQMs produced by four Indian companies were installed alongside existing CAAQMS. This was part of the pilot project called ‘Technical Assessment of Low-Cost Sensor based PM2.5 and PM10 Monitoring Network in Maharashtra’.

The pilot project was carried out in partnership with IIT-Kanpur, but it has not been scaled up despite availability of funds under the National Clean Air Program (NCAP) and the 15th Finance Commission. VM Motghare, joint director (air), refused to comment when contacted by Hindustan Times.

The pilot had been spearheaded by SN Tripathi, principal investigator at IIT-K’s ‘Advanced Technologies for Monitoring Air-quality iNdicators’ (ATMAN) centre, who also serves on the NCAP’s steering and monitoring committees.

To give you an idea of how SAAQMs and CAAQMS function, the former only measures concentrations of particulate matter (PM2.5 and PM10), which is on most days the primary pollutant across India and the first visible sign of poor air. When compared with nearby CAAQMS, sensors from three start-ups saw less than 25% error for uncalibrated values. After calibrating, or “adjusting” the values based on the main monitor, the error was reduced to less than 15% for three types of sensors. The devices’ performance could be improved over time.

“I am still hopeful for a revival of the Mumbai project, which was a clear success. Just last week, at a meeting with the Brihanmumbai Municipal Corporation and the MPCB, I demonstrated how a mix of CAAQMS, sensors, and satellite imagery can provide air quality data at a resolution of at least 500 sq meters, if not 100 sq meters. For Greater Mumbai alone, which has an area of around 900 sq km, low-cost sensor devices can help provide at least 1200-1400 air quality data points. Emission plumes and individual sources of pollution can be identified. Such a detailed picture is unprecedented for Indian cities,” Tripathi said to HT.

Pravin Darade, Maharashtra’s environment secretary and member secretary, MPCB, did not respond to requests for comment.

Three sub-regional officers, meanwhile, claimed the project is a “failure”, while an independent scientist who was closely involved said it had been a success but was “put on ice for no good reason”.

The low-cost sensors can be purchased for about ₹50,000 with a standard 10% annual maintenance cost, but a regulatory-grade CAAQMS -- which also measures gaseous pollutants -- can cost around ₹20 lakh, plus ₹2 lakh annual maintenance, as per a recent paper-- ‘Plugging the ambient air monitoring gaps in India’s national clean air programme (NCAP) airsheds’ -- published by the Elsevier journal Atmospheric Environment.

The paper also points out that Mumbai has a total of 21 CAAQMS, which falls short of the 67 that are recommended by NCAP criteria for the city’s larger ‘airshed’. This includes not just Greater Mumbai but also Badlapur, Navi Mumbai, Thane, Ulhasnagar, Vasai-Virar, Kalyan and Karjat. It would cost ₹268,000,000 to operate 67 full-fledged monitoring stations across this region over 10 years, but a hybrid network (of 20 CAAQMS and 40 sensors) would cost less than half of that, at ₹129,350,000.

“To meet the... scientific assessment required by the NCAP, significant costs associated with augmenting monitoring infrastructure can be supplemented by using a combination of reference- and sensor-grade monitors as a calibrated hyperlocal hybrid network,” the researchers point out.

In 2021, Sudhir Srivastava, former chairman, MPCB, had said to HT, “CAAQMS are prohibitively expensive. As a result, our existing network is patchy. They only give a broad picture of air pollution while low-cost devices can be used for a more hyper-local picture. This study was needed to ensure that the information they provide is reliable.”

Sunil Dahiya, an independent analyst with the Centre for Research on Energy and Clean Air (CREA), emphasised on the need to establish monitors at sub-ward levels, so that granular pollution hotspots can be identified and hopefully acted on. “Low-cost sensors cannot replace CAAQMS, but a hybrid network can tell us reasonably well how air pollution changes every 2-3 sq km using PM2.5 or PM10 data only. Authorities and at-risk groups, like pregnant women or people with asthma, can make valuable decisions with that kind of information. They can’t do this by looking at the city’s overall air quality index (AQI), which is just an average number that masks regional variations,” said Dahiya.

While hybrid networks have been established across the USA, and in Beijing and London, with the US’ Environmental Protection Agency utilising data from 10,000 low-cost sensor-based devices on the official US government air quality portal, their development in India is limited. Under the ATMAN project, 1,400 low-cost sensors are being deployed in Bihar and Uttar Pradesh under the ‘Ambient air quality Monitoring over Rural areas using Indigenous Technology (AMRIT)’ scheme, conceived by the Principal Scientific Advisor to the Govt. of India. Under a different scheme, the Municipal Corporation of Surat has established a network of 20 sensor-grade monitors (in partnership with Respirer Living Sciences), but the city has just one CAAQMS.

Respirer Living Sciences, an independent technology company and partner on the AMRIT scheme, was also involved in the MPCB’s Mumbai study. Founder Ronak Sutaria confirmed that the pilot project, though successful in his view, has not moved forward since concluding in early 2022.

Months later, in December 2022, the Central Pollution Control Board instructed states that no central funds can be spent on buying new CAAQMS equipment, which is almost always imported. “States will have to buy Indian-made devices that will be certified by the National Physical Laboratory. They have been working on the certification criteria since 2019. But in UP, Bihar and Surat, other institutes have stepped in to do the capacity building because they are seeing value in such hybrid networks. States can still use central funds to develop these,” said Sutaria.