Summary

The occurrences of smog in the National Capital Region of India has gone up, the concentration of PM 2.5 is through the roof. It has been reported as one of the worst levels of air quality in Delhi since 1999 and probably one of the worst air quality in the world. Low visibility has resulted in accidents across the city, notably a 24 vehicle pile-up on the Yamuna Expressway. Deteriorating air quality has far reaching effects on health such as multiple sclerosis and lung cancer It also led to cancellation and delay of public transport, primarily trains and flights, causing much hindrance to the people. The primary sources of smoke are stubble burning, lit garbage, road dust, power plants, factories, vehicles and a lack of vegetation across the city. It has been observed for the past five years that surface concentration levels of PM2.5 dramatically increase during the winter months in the national capital region of India. Major reasons for this trend include stubble burning during the harvest months in winter. This project aims to understand the increase in PM2.5 concentration using climatic variables using statistical learning techniques.

On the eve of new year 2018, the air quality was found to be the worst when compared to the last 2 years (2016 and 2017). The Air Quality Index during the period from 25 Dec’17 to 3rd Jan’18 remained roughly 50 notch points higher as compared to the last 2 years. The lead pollutant was identified as PM2.5. Thus, it is very essential to understand the reasons behind the poor air quality index in Delhi and predict the air quality index using statistical learning techniques.

Our study has statistically established the association of environmental conditions such as rain, fog, temperature, relative humidity etc, with concentration of PM2.5. The inclusion of such predictors paves the way for implementing the resulting model in other regions of India as well.