[Prediction of atmospheric Japanese cedar pollen counts in Oita University Faculty of Medicine Complex]

Abstract

Background: We investigated the method of predicting atmospheric Japanese cedar pollen counts and the first day of pollen release at Oita University Faculty of Medicine Complex in this study.

Method: We set up a Durham sampler on the roof (30 m from the ground) of Oita University Faculty of Medicine complex and investigated atmospheric pollen counts day by day from January 1 to April 30 from 1990 through 2004.

Results: The total pollen counts per year correlated very well with averaged temperature, averaged relative humidity, and the sunshine duration during the previous July. The sunshine duration yielded the highest correlation coefficient. Multiple regression analysis showed the sunshine duration during the previous July and total pollen count of the previous year to be independent predictors of the current year's total pollen counts. Multiple correlation coefficients was 0.9518 (p<0.001). We considered the Hita area to be the source of the pollen, according to meteorological conditions. Prediction of total pollen counts was more accurate if based on factors within the previous 10 years than on factors within the previous 5 years. There was no correlation between the first day of pollen release and averaged temperature in January. Our data also showed that the first day of pollen release could not be predicted by the cumulative thermal constant method and threshold temperature for the development of male Japanese cedar flowers. Multiple regression analysis showed the minimum temperature from November 1 to the first day of pollen release, the date of minimum temperature, and the averaged temperature for the January to be independent predictors of the number of days from January 1 to the first day of pollen release.

Conclusion: We concluded that it is important to establish a method of predicting atmospheric Japanese cedar pollen counts that is based on the characteristics of each area.