Bending Delhi’s air pollution curve
Delhi is among the most polluted cities in the world (IQ Air 2021). In 2019 alone, air pollution caused over 17,000 premature deaths and an economic loss of $1,207 in the national Capital (Pandey et al. 2021). With regards to air quality, 2020 was an aberration. The pandemic-induced lockdown measures provided temporary respite from the year-round poor air quality. Despite the lockdown, Delhi’s annual average PM2.5 in 2020 was more than two times its permissible limit Barring a few days in April and May, Delhi experienced National Ambient Air Quality Standard (NAAQS) compliant air quality on most days during the lockdown. Yet, the annual average PM2.5 concentration in 2020 was 93 µg/m3, which is more than twice the permissible limit for PM2.5 in India. Despite low activity levels for close to eight months (March to November) in 2020, Delhi residents were exposed to NAAQS non-compliant air for more than half of the year.
Winters saw poor quality despite proactive measures by the government While the NAAQS non-compliant air quality in Delhi is not a new phenomenon, the winter of 2020 witnessed proactive measures from the State government in the wake of Covid-19 and evidence pointing at the association between high air pollution and Covid mortality (Petroni et al. 2020; Cole, Ozgen, and Strobl 2020; Wu et al. 2020). This includes the Yuddh Pradushan Ke Virudh (war against pollution) campaign and a seven-point action plan to combat air pollution in Delhi which listed measures ranging from combating dust and mitigating hotspots to a mobile application called Green Delhi for complaints and a ‘war room’ for monitoring air pollution control activities (PTI 2020). Similar to 2019, the Graded Response Action Plan (GRAP), also came into force on October 15, 2020 and the Environmental Pollution Control Authority (EPCA) oversaw its implementation until the announcement of its dissolution on October 28, 2020 (Koshy 2020; EPCA 2020).
However, despite these measures, the PM2.5 levels remained almost three times higher than the NAAQS on an average between October 2020 and January 2021. We also observe that PM2.5 levels in winter 2020 were higher than those in 2019. To explain this end, we analyse meteorological parameters, source activity levels, and contributions to establish primary drivers of pollution during different phases of the winter season. Through this brief, we intend to help the Delhi government, the Central Pollution Control Board (CPCB) and the Delhi Pollution Control Committee (DPCC) to identify priority areas of intervention for the year 2021. We summarise key highlights as follows.
Air quality in winter 2020 was worse than winter 2019. Delhi observed 92 severe and very poor air quality days in the winter of 2020 compared to 80 such days in 2019. Compared to an average PM2.5 concentration of 161 μg/m3 in 2019, between October and November 2020, this value was 172 μg/m3. It further shot up to an average level of 192 μg/m3 in the period between December 2020 to January 2021 compared to 178 μg/m3 during the same period previous year. Contributions from stubble burning and household emissions from cooking and space heating were significant fractions of the pollution pie. Modelled source contribution estimates of particulate matter (PM2.5) by UrbanEmissions suggest that relative contribution from farm fires was the highest (~30%) in the period between October 15 and November 15, 2020. We find that compared to the stubble burning period in 2019, a longer harvesting season in 2020 led to a significant increase in the number of fires. In the following months, contribution from household emissions (including domestic cooking, space heating, water heating, and lighting) primarily drove poor air quality in Delhi. It is worth highlighting that these values are modelled estimates and are subject to the sector-specific assumptions used in the model.
Calmer winds in October and November amplified the impact of farm fires on Delhi’s air quality The stubble burning phase (October 15 to November 15) in 2020 experienced 172 hours (70% higher) of calm and light winds (<5 km/h) compared to 101 hours in 2019. Winds predominantly from the north-west direction facilitated the transport of smoke emanating from farm fires and calm winds in Delhi further intensified its adverse impact on air quality. Interestingly, for brief periods in the season, even when high fire counts were reported in Punjab and Haryana, Delhi’s air quality was not affected due to favourable meteorological conditions (easterly and southerly winds) (Figure ES3).
Lowered activity levels at the start of winter due to lockdown bounced back to the previous year’s levels as the winter progressed While Delhi’s average PM2.5 concentration during the stubble-burning period (October’20 and November’20) was 172 µg/m3 , it increased to 192 µg/m3 during peak winter (December’20 and January’21). The higher PM2.5 levels in December 2020 and January 2021 were primarily caused by locally-emitted pollutants and added burden of household emissions from space heating. Activity levels were low at the start of the season, but most activities, including vehicular traffic and power generation, bounced back to the previous year’s levels (proxied by indicators such as congestion and electricity generation levels in Figures ES4 and ES5) as the season progressed.
Delhi needs a dedicated air quality forecasting cell to facilitate roll out of preventive measures. We attribute the brief periods of moderate air quality during winter 2020 largely to favourable meteorological conditions. It is evident that adverse meteorological conditions in Delhi intensified the impact of local and regional emissions on Delhi’s air quality. While meteorological conditions cannot be controlled, sustained air quality gains can be realised only by steeper emission cuts across sectors. Delhi has in place a publicly available air quality forecast system provided by UrbanEmissions for over five years. The Indian Institute of Tropical Meteorology (IITM), under the aegis of ministry of earth sciences (MoES), has also built an official air quality warning system for Delhi (PIB 2018). However, none of these forecasts were actively used to take pre-emptive measures to reduce emission loads from anthropogenic activities.
Some countries roll out emergency measures in response to air quality forecasts and not after air quality actually dips to dangerous levels. For instance, Beijing’s ministry of ecology and environment issues a red alert if the daily mean citywide air quality index (AQI) is forecasted to be greater than 200 for four days (96 hours) or more; greater than 300 for two days (48 hours) or more; or greater than 500. In contrast, the Delhi government issues orders to execute emergency measures under GRAP ex-post, that is, after air quality concentrations reach a certain threatening level. Responsive measures cannot prevent the occurrence of high pollution episodes. Further, adding relative source contributions to air quality forecasts, similar to the way UrbanEmissions issues forecasts, can help identify the primary contributors during a particular episode. Integrating such forecasts with a decision support system would enable the local regulatory agencies to implement on-demand emission control interventions targeting prominent sources during forecasted high-pollution episodes.
GRAP presents the state government with an opportunity to constitute an air quality forecasting cell that can advise the government to take necessary measures to prevent severe air quality episodes in the capital city. We recommend that going forward, the Delhi government, the CPCB, and the DPCC use the air quality forecasts not only to issue public health warnings but also for taking pre-emptive actions in the national capital. We must move from a system that enforces the GRAP as an ex-post measure to one that prevents the occurrence of high pollution episodes through pre-emptive emission control measures.
(The study has been authored by LS Kurinji and others)