• Wed. Dec 7th, 2022

Is India’s National Clean Air Programme (NCAP) working? New CSE analysis says NCAP and non-NCAP cities are showing similar trends in PM2.5 levels

A new analysis by Centre for Science and Environment (CSE), released here today on the eve of the UN International Day of Clean Air for Blue Sky, says there is barely any difference in overall PM2.5 (particulate matter 2.5) trends between cities under the National Clean Air Programme (NCAP) and those outside its ambit. The analysis says that both groups of cities reflect similar mixed trends in air quality in different climatic zones, and this means they require substantial reduction in particulate pollution levels to be able to meet the national ambient air quality standards.  

The NCAP has set a national level target of 20-30 per cent reduction in PM2.5 and PM10 concentrations by 2024 from the 2017 base year. But a latest performance assessment of NCAP cities by the Central Pollution Control Board (CPCB) for disbursement of performance-linked funds has considered only PM10 data that is largely coarse dust particles. As the monitoring of PM2.5 — the tinier particles that are much more harmful — is limited, a uniform assessment of cities based on PM2.5 reduction has not been considered for performance assessment.

CSE has carried out a national analysis of PM2.5 levels in cities for which data is available to understand the trend in both NCAP and non-NCAP cities, and the level of reduction needed in both the groups of cities to meet the national clean air standards. This has also exposed the status of air quality monitoring in terms of manual and real time monitoring, extent of PM10 and PM2.5 monitoring in cities, and challenges of data quality to construct and verify a longer term air quality trend.

“While it is encouraging that funding of clean air action is linked to performance and the cities’ ability to demonstrate improvement in air quality, dependence on only manual monitoring of PM10 evidently creates a bias in spending as it shifts focus more towards dust control and detracts attention from composite action on industry, vehicles, waste and solid fuel burning,” says Anumita Roychowdhury, executive director, research and advocacy at CSE.

“The expanding monitoring network of PM2.5 and key gases needs to be leveraged quickly to prioritise multi-pollutant action for more effective reduction of risk across all regions,” she adds.

The NCAP covers 132 cities — 82 of these have been funded by the Programme, while 50 cities have received funds from the 15th Finance Commission; Rs 6,425 crore has been released till 2021-22 and Rs 2,299 crore has been earmarked for 2022-23. Cities are required to quantify improvement starting 2020-21 – this requires 15 per cent and more reduction in the annual average PM10 concentration and a concurrent increase in good air days to more than 200. Anything less than that will be considered low and reduce the funding.

“As the system is becoming more performance oriented and real time air quality monitoring is expanding, it is necessary to develop and define robust protocols and methods for quality control of real time data and ensure adoption of standardised methods for data processing, analytics, addressing data gaps and data completeness and to construct a reliable trend to verify performance and compliance with the national air quality target,” says Avikal Somvanshi, programme manager, Urban Lab, CSE.

Methodology: This analysis has included 332 real time monitoring stations active in 2021 in India, spread across 172 cities in 27 states and Union territories. It has relied on data that is publicly available on the CPCB website. However, the minutes of the 5th meeting of the steering committee of NCAP dated July 29, 2022 have listed more stations. Accordingly, there are currently 882 manual stations in 378 cities and 361 CAAQM (continuous ambient air quality monitoring) stations in 192 cities.

This analysis is focused only on PM2.5 data, sourced from the CPCB’s real time data portal in its most granular format; the data has been further cleaned and processed, and 24-hour values calculated for each station based on the USEPA (United States Environmental Protection Agency) methodology.

The annual average has been calculated as a mean of quarterly averages as per the USEPA methodology, but given the issues of data quality the requirement of data completeness has been slightly modified to enable inclusion of a large pool of cities. The USEPA, for instance, requires each quarter to have data for a minimum 75 per cent of the monitoring days .

Key highlights

Only 10 per cent of statutory/census towns have air quality monitoring: According to the 2011 census, India has 4,041 statutory towns and 3,894 census towns. Between the national air quality monitoring programme (NAMP, manual) and the CAAQMS (real time), only 400 cities/towns have PM10 quality monitoring. Out of these, 213 (51 per cent) have only manual monitoring, 90 (21 per cent) have only real time monitoring, and 97 cities/towns (23 per cent) have both. There are 22 cities/towns (5 per cent) that have manual stations but have not reported any data since 2015 and can be considered defunct. These cities together account for 1,176 PM10 monitoring stations (804 manual stations and 372 real time stations).

PM2.5 monitoring is limited compared to PM10: Between the NAMP and CAAQMS, only 256 cities/towns are covered by PM2.5 monitoring. Out of these, 74 (29 per cent) have only manual monitoring, 131 (51 per cent) have only real time monitoring, and 51 (20 per cent) have both. These cities together account for 660 PM2.5 monitoring stations (315 manual stations and 345 real time stations). The performance assessment of NCAP cities is done only on the basis of PM10 monitoring is because the network of PM10 monitoring is more extensive than that of PM2.5. PM2.5 monitoring needs to be expanded and considered for performance monitoring for better prioritisation.

Only about half of NCAP cities have real time monitoring: In 2019, only 51 out 132 NCAP cities had real time monitoring stations. The number grew to 63 in 2021; six more NCAP cities have installed real time monitors in 2022 so far.

Almost a quarter of NCAP cities with real time monitoring do not meet minimum data completeness requirement: In 2021, 15 out of the 63 NCAP cities (24 per cent) did not meet the minimum data completeness requirement (60 days of valid 24-hour values in each quarter of the year). In 2019, the number was lower: only 16 per cent did not meet the requirement.

Performance of NCAP cities on PM2.5 levels (2019-21): Only 43 NCAP cities have adequate PM2.5 data for the period 2019-2021 – enough to create a reasonable trend for tracking progress. However, it may be noted that 2020 has been an exceptional year due to the lockdown phases and is an aberration. Nearly all cities have recorded a dip in 2020 followed by a subsequent increase in 2021. Therefore, a comparison between 2019 and 2021 shows that only 14 of the 43 cities have registered a 10 per cent or more reduction in their PM2.5 level between the two years.

Seven cities show negligible (less than 5 per cent) change: these include Delhi and Ghaziabad. There are 16 cities that have registered a significant increase (5 per cent or more) in their PM2.5 levels — Khanna, Jaipur, and Udaipur have registered the most deterioration with their 2021 annual value increasing by over 20 per cent compared to the 2019 annual value. Faridabad with 6 per cent increase is only NCR NCAP city in this pool of cities with a significant worsening of air quality. It is also the only city outside the non-attainment list.

Punjab, Rajasthan and Maharashtra cities dominate the list of cities which have registered a significant increase in PM2.5 levels between 2019 and 2021. Chennai, Varanasi and Pune show the most improvement among NCAP cities. But unlike cities with increasing pollution level which have a very clear regional pattern, there is no regional pattern seen among cities reporting significant improvement in their air quality.

Performance of non-NCAP cities: There are 46 cities that are not covered under NCAP, but have adequate real time data for both 2019 and 2021. In this group, 15 cities have registered a significant worsening of annual PM2.5 levels between 2019 and 2021. Ankleshwar in Gujarat with 34 per cent increase in annual PM2.5 value is the worst performer in the pool, followed by Satna (Madhya Pradesh), Vatva (Gujarat), Bahadurgarh (Haryana), and Bhatinda (Punjab); all of which have registered an over 20 per cent increase.

Ten cities registered negligible change (less than 5 per cent) in their annual values; 21 recorded significant improvement, with a 5 per cent or more decline between 2019 and 2021.

Cities of Haryana, Madhya Pradesh and Gujarat dominate the list of non-NCAP cities that have registered significant increase in air pollution levels between 2019 and 2020. Tirupati in Andhra Pradesh and Siliguri in West Bengal are only cities from other regions in the group. Interestingly, most of these cities are outside the NCR.

Palwal in southern Haryana with a 60 per cent improvement in its annual PM2.5 level is the best performer among non-NCAP cities. In fact, NCR cities dominate the list of most improved non-NCAP cities. Most change (positive or negative) is noted among north Indian cities. Cities within NCR show improvement, while cities outside indicate worsening levels.

Performance of NCAP vs non-NCAP cities: Data is adequate to construct PM2.5 trends for 43 NCAP cities and 46 non-NCAP cities for 2019-21. The non-NCAP cities can be treated as a control group to see how the NCAP funding has improved air quality performance of cities. This is a simplistic non-standardised assessment and should not be treated as a conclusive assessment of NCAP, as air pollution has a regional character and many non-NCAP cities are located within the influence zone of NCAP cities which are generally much bigger with asymmetrical impact on the regional air quality baseline. Says Somvanshi: “Having said that, there is hardly any difference between the performance of NCAP and non-NCAP cities between 2019 and 2021.”

There are 16 NCAP cities and 15 non-NCAP cities that registered a significant increase in their annual PM2.5 levels – with near identical numbers. Same goes for cities (20 NCAP and 21 non-NCAP) that registered a significant improvement in their annual PM2.5 levels.

NCAP vs non-NCAP cities and the impact of lockdowns: Hard and extended lockdowns in 2020 due to the ongoing pandemic had significant air quality impacts in all the monitored cities. But there is a negligible difference between the impact on NCAP and non-NCAP cities. Both groups of cities recorded nearly 12 per cent reduction in their aggregate 2020 PM2.5 level compared to the previous year.

Pollution bounced back with near similar intensity in both the groups in 2021. NCAP cities registered an aggregate 8 per cent increase in their annual PM2.5 levels in 2021 from the low of 2020. Similarly, non-NCAP cities registered a 7 per cent increase.

How have mega cities performed: All six mega cities registered drop in their annual PM2.5 levels in 2020, but pollution bounced back everywhere except in Chennai. In Delhi, pollution dropped by 13 per cent in 2020, but it rose by 13 per cent in 2021 nullifying all the gains made by the national capital.

In Kolkata, pollution dropped by 22 per cent in 2020 and rose by 16 per cent in 2021. Mumbai, with a 48 per cent increase in 2021, saw the most negative impact of unlocking of the economy after the lockdowns. Hyderabad had the least variation in its annual levels across the three years. Bengaluru air improved by 21 per cent in 2020 and with just 8 per cent increase in 2021, the southern metropolis has retained most of its gains. Chennai saw a 29 per cent drop in its PM2.5 level in 2020 and it has dropped by another 23 per cent in 2021, making it the least polluted mega city in the country.

Status of national air quality: There is an equal number (39) of NCAP and non-NCAP cities that did not meet the annual PM2.5 standard in 2021.

The CPCB classifies cities that have their annual average higher than 1.5 times the annual standard as critically polluted. In 2021, 20 NCAP cities and 24 non-NCAP cities qualified as critically polluted – this means there are more critically polluted cities that are not covered by NCAP than there under the programme.

Nine NCAP cities met the annual NAAQS in 2021. These are Chennai, Hubballi, Bengaluru, Rajamahendravaran, Mumbai, Sagar, Nashik, Dewas and Chandigarh. Cities with multiple stations like Chennai, Bengaluru, Mumbai and Chandigarh meet the standard based on average value from their trend station/s, but there are stations within the city which reported annual values higher than the standard, implying the pollution is not low throughout these city. 

Only 16 non-NCAP cities meet the standard — 13 of them are located in south India.    

Only two cities in the country out of the 103 monitored via the CAAQMS network can be classified as ‘low pollution’ (less than half the annual standard) as per the CPCB matrix. These are Kozhikode in Kerala and Madikeri in Karnataka.

Adopt a standardised method for assessing air quality trends for complying with the national ambient air quality standards: India needs to adopt a robust protocol for estimating air quality trends and their compliance with the national ambient air quality standards. It is critical that a city meets the minimum data completeness required to credibly establish its longer term air quality trend vis-à-vis the annual PM2.5 standard.

The CPCB currently uses spatial average of all stations in a city to establish its compliance with the standard. But internationally, the practice for establishing compliance with the standard has moved away from spatial averaging. The USEPA requires that all monitoring stations in a city should meet the standard for that city to be deemed compliant with the standard. However, for compliance trend analysis, it uses only a few select stations in the city.

Both NCAP and non-NCAP cities need substantial reductions to meet the national ambient air quality standards: It is acknowledged that many cities have added new stations since 2019 and many of these also meet the minimum data completeness requirement — but they have not been included in computation of annual average of the city. Nevertheless, data from these new stations is captured in the maximum and minimum information published in the parenthesis next to the annual level of the city. This is a hybrid approach that uses CPCB´s current practice of spatial averaging for cities with multiple stations, but ensures apple to apple comparison by introducing a trend station concept. It also captures the worst station data in the city allowing comparison vis-à-vis the USEPA methodology.    

There are 48 NCAP cities and 55 non-NCAP cities that meet the minimum data completeness requirement for computing annual PM2.5 levels for 2021. Only these 103 cities have been analysed by CSE.

  • Northern states: Ghaziabad is the most polluted NCAP city in the north, closely followed by Delhi. There are 16 non-NCAP cities that exceed the regional average compared to eight NCAP cities. There are as many non-NCAP cities as NCAP cities that require over a 50 per cent reduction. Most NCR cities require over 50 per cent reduction from their 2021 PM2.5 level to meet the NAAQS.
  • Western states: Jodhpur is the most polluted NCAP city in the west, followed by Kota and Jaipur. Bhiwadi — though a non-NCAP city — is still the most polluted in the region. Ankleshwar and Vatva in Gujarat are worse than most NCAP cities of the region as well. Six NCAP cities and four non-NCAP cities need reduction of over 25 per cent to meet the annual standard.
  • Eastern states: Patna is the most polluted NCAP city in the east, followed by Muzaffarpur and Durgapur. Hajipur and Siliguri are the most polluted non-NCAP cities. This is the only region where no city, NCAP or non-NCAP, meets the annual standard (see related graph in the detailed analysis). Except Talcher in Odisha and Agartala in Tripura, all cities in the region require over 20 per cent reduction to meet the annual standard.
  • Southern states: Visakhapatnam is the most polluted NCAP city in the south, followed by Hyderabad. All the other cities in the region, NCAP and non-NCAP, meet the annual standard (see related graph in the detailed analysis). Visakhapatnam needs to reduce pollution by 10 per cent and Hyderabad by 3 per cent to meet the standard.

NCAP vs non-NCAP cities pollution level: On an average at the national level, there this barely any difference in the pollution levels between NCAP and non-NCAP cities. There is some difference at the regional level, though. In west India, non-NCAP cities on an average are more polluted and these are located inland in the arid and semi-arid sub-region. In the north, non-NCAP cities on an average are marginally more polluted. In the east and south, NCAP cities are on an average significantly more polluted.

On an average, cities of east India are the most polluted in the country, while cities in south are the least polluted.

The way forward

Says Roychowdhury: “It is clear that the current practice of keeping the focus only on selected cities without considering the larger urban and regional landscape can limit the effectiveness of the NCAP programme and resource investment. The current mandate of developing state action plans has to be refined to ensure regional approach is initiated for a wider impact.”

“Moreover, the ongoing funding strategy based on performance of cities on air quality improvement requires robust air quality monitoring of all key parameters along with strong data quality control and a standardised protocol for establishing air quality trends, especially for real time data, for reporting compliance with clean air targets,” she adds

As the current focus of NCAP is to reduce particulate pollution, immediate strategy is needed to consider PM2.5 data for performance assessment of cities, points out Somvanshi. PM2.5, being smaller, is more harmful as it penetrates deep inside the lungs. The larger share of PM2.5 is emitted by combustion sources including vehicles, industry, power plants, waste and solid fuel burning. Otherwise, the expansion of PM2.5 monitoring as well as real time monitoring will be wasteful if not leveraged for performance assessment. This also needs to be supported by a roadmap to include the gases for targeted mitigation.

(A CSE Media Briefing)

Report sent by Pratyusha Mukherjee 

Leave a Reply

Your email address will not be published. Required fields are marked *