Dicussion of 'An integrated statistical approach for evaluating the exceedance of criteria pollutants in the ambient air of megacity Delhi' Atmospheric Environment

摘要

The introduction to this article is admirably well-written, clear and understandable. However, the article's methods of estimating exceedance probabilities over multiple time periods are erroneous.First, on page 7, the paper says the theory "assumes the concentration levels measured in successive non-overlapping periods are independent of one another". This is an incorrect assumption. Concentration levels have high serial correlations, as a calculation of the autocorrelation coefficient for any empirical pollutant concentration data series will show. Any realistic physical model of concentrations will conclude that the concentration level in one period is more likely to be close to the concentration level in the immediate previous period than to a random selection from all periods. This erroneous assumption is apparently not unique or original in the present paper, but may be part of a current through the literature.