Both heat and sunlight are needed to drive ozone formation chemistry. Ultraviolet radiation flux/ latent heat flux plots provide information about the amount of sunlight reaching New Hampshire at the time of the ozone exceedances. The temperature plots that follow document the effects of heat on ozone levels at a few locations in New Hampshire.
The first plot shows the New Hampshire ozone season (May through September) from 1995 to 1999. The daily maximum temperature in Concord is plotted against the daily maximum one-hour ozone levels measured in the same area. A correlation between temperature and ozone levels is evident, as we expect to see higher concentrations of ozone as temperatures increase. The correlation is only so strong however; a second-degree polynomial "best-fit" line generated an R2 value of 0.3847.
The next two plots are similar and also look at the NH ozone season (May through September) from 1995 to 1999. Daily maximum temperatures were not available for Rye or Portsmouth, thus Concord daily maximum temperatures were plotted against the daily maximum one-hour ozone levels in Rye and Portsmouth. Both sites showed an even stronger correlation between temperature and ozone (Rye R2= 0.4457, Portsmouth R2= 0.4738). Take note that the polynomial line best fitted to these data sets indicates that ozone levels may also increase as temperatures decrease below 60 degrees, an anomaly that may be due to the seabreeze phenomenon, in that the colder seabreeze often transports high ozone concentrations onto the New Hampshire coast.
The last two plots were used to strengthen the merit of the explanation that the seabreeze is causing the inverse correlation between temperature and ozone. The maximum daily temperature at Kittery, Maine was plotted against the daily maximum one-hour ozone concentrations also from Kittery. Note that only the months of June through August in 1995 were used due to data availability. The polynomial best-fit R2 value was 0.4205. The last plot investigates the same three months (June-August) in 1995, but this time instead of using the maximum temperature during the day (Kittery), the actual temperature at the time of the peak (Kittery) one-hour ozone concentration was plotted against the value of the peak. This plot generated an R2 value of 0.369. A drop in the correlation occurred between these two plots. Again, this is most likely due to the cooler sea breeze responsible for advecting high levels of ozone, thus causing the second Kittery plot to be more scattered.