The graph on the left in red shows the trend of AOT data and the
graph on the right in blue shows the trend of PM10 data selected for the same date and time of acquiring
by the satellite and the measurement ground station. It is observed that, the trend of the two parameters, in
this case, go together in similar direction, therefore we can have good correlation coefficient and linear
regression for this year. Similar way was applied to the dataset of matching pairs for year 2009 and 2010
and for the months of February, March and April of each year, and only February for 2011. The total
matching pairs used in this step are 140 pairs as shown in Figure 1(b). In this Figure, the total number of
matching pairs available within year 2007-2011were used to calculate the correlation coefficient and
generated linear regression equation. Here the correlation is 0.32 which is not high, due to scattered points
that are unusual observations, according to the statistical analysis which need to be investigated. Other
possible reasons related to meteorological and environmental conditions could be also the factors that
cause unusual observations.