Number Theory: Should Delhi’s pollution spike surprise us?

Delhi’s pollution levels, as measured by the average Air Quality Index (AQI) hit the “severe” threshold on Saturday at 468.

Delhi’s pollution levels, as measured by the average Air Quality Index (AQI) hit the “severe” threshold on November 3. According to the data at 4pm on November 3, this number was 468, which is pretty close to its maximum limit of 500. While pollution reaching very high levels around this part of the year is not surprising, one bit which is slightly counter-intuitive this year is the fact that satellite data seems to show fewer farm fires. What exactly has led to the spike in Delhi’s pollution levels? Here are four charts which try to answer this question.

HT Image

Delhi’s AQI turning severe was a question of when, not whether

• 
  Delhi’s AQI turning severe was a question of when, not whether

  Delhi’s 24-hour averages of air quality have been in the dirtier half of the “very poor” range (301-400) since October 31. It reached the “severe” range (400 and above) on November 3. However, like all averages, these hide how dirty the air was in some hours in parts of the national capital. At least four of Delhi’s 40 air quality stations were in the “severe” range of PM2.5 concentration – PM2.5 refers to particles of size 2.5 micrometres or smaller that are generally driving the AQI during this time of the year – for 75 of the 96 hours beginning 4pm on October 30. After 10 AM on November 2, at least 30 stations were in the severe range for 26 hours. The headline number breaching the 400 mark was almost a given.

What are the reasons for deteriorating air right now?

• 
  What are the reasons for deteriorating air right now?

  Air pollution levels depend on the interplay of the quantity of pollutants being pumped into the air and the quantity being pumped out. Among the different sources of air pollution in Delhi, one that is absent through most of the year – stubble burning in the neighbouring states – has become prominently active only in the last week. Active fires in Punjab and Haryana, as measured through the VIIRS instrument aboard the Suomi NPP satellite, crossed the 1,000 threshold on October 30 and have remained above that level since. In comparison, the daily average of active fires in the first week of October was 128. This number was 106 and 113 in the second and third week of October. The rising contribution of farm fires to Delhi’s pollution levels means that the contribution of other pollutants – such as from vehicles, industries, dust, etc. – must go down to keep overall pollution in check. In a way, this is exactly what the Graded Response Action Plan (Grap) seeks to achieve. However, the policy response so far suggests that is a poor reactive strategy rather than an effective proactive plan. The only other thing which can reduce pollution despite a rise in pollutants from far fires is favourable meteorological conditions such as rain, strong winds, or higher temperatures. Delhi last received rain in the middle of October, average wind speed in the past week was 35% less than in the first week of October, and compared to the first week of October, temperatures were around three degrees Celsius lower in the past week.

How does this year’s air compare with earlier years?

• 
  How does this year’s air compare with earlier years?

  There are limits to an informed comparison on this front. This is because of a change in the number of stations feeding data into the calculation of the average AQI for Delhi. For example, 35 of 40 stations went into the average for November 3. This number was 34 on November 2. CPCB does not clarify which stations did not generate the data on a particular day, and air pollution levels differ significantly across some stations. With this caveat in place, the data does show some interesting things. For the October 1 to November 2 period, Delhi’s AQI this year was 4.2% better this year than the 2018-2022 average. However, in the October 31 to November 2 period, Delhi’s AQI was 1.3% worse than the historical average. Does this mean Delhi spent most of October this year with better air quality than in the past? It appears so, but the degree of improvement is not big enough to be taken seriously given the variation in station-wise data feeding the headline number.

How can air quality remain the same this year with fires decreasing dramatically?

• 
  How can air quality remain the same this year with fires decreasing dramatically?

  Delhi’s air quality remaining roughly the same every year (unless rain or wind intervenes) is not surprising generally. However, it is noteworthy this year because the satellite count of fires has decreased dramatically: 58% compared to the 2018-2022 average for the October 1-November 1 period. What could be the reason behind this? One reason could be the measurement of fires themselves. The Suomi-NPP satellite acquires images over India, usually in two 3–4-hour periods during the day-- once around noon and once around midnight. Low-intensity fires lit outside these periods – this can be even deliberate if there is administrative pressure – are unlikely to be counted. This means it is possible the decrease in fires is not as big as the satellite numbers suggest. Moreover, the weather isn’t as favourable this year as it has been in some of the years analysed here. The average wind speed at air quality stations is 32% less than the 2018-2022 average for the October 1-November 2 period. The maximum temperature is lower by 0.73 degrees for the same period although the minimum temperature is 0.95 degrees higher. Moreover, in 2019, 2021, and 2022, October was much rainier than in 2023. These factors could have negated some of the actual decrease in farm fires.