Long-term trend of ventilation coefficient over Delhi and its potential impacts on air quality

Abstract

A ventilation coefficient (VC) is an essential parameter and plays an important role in the dispersion of air pollutants and is one of the factors that determine the pollution potential over a region. A comprehensive SODAR (Sonic Detection And Ranging) data set of about 8- years is analysed to study the long term trends, variations and time series model simulation of VC. The Box–Jenkins popular ARIMA (AutoRegressive Integrated Moving Average) model was applied to simulate the VC over Delhi region. The ARIMA models have been developed for different months from January to December as most suitable for simulating and forecasting the VC over the observation site. The Stationary R-squared, R-squared, Root Mean Square Error and Normalized BIC etc. are used to test the validity and applicability of the developed ARIMA models revealing significant accuracy in the model performance. It is evident from the modeled consequences that some features of the past continue to impact on the future values of VC, and these models can be reliably used for future predictions. The long term average value of VC has been found to be 1249 ± 236 m2/s and on yearly basis VC trend increase is observed to be 70 m2/s/year.