[Prediction about daily scattering count of Japanese cedar pollen]

We investigated the relationship between the daily count of pollen scattered, the count of days during the scattering period and the highest temperature during the sugi (Japanese cedar) pollen scattering season from 1983 to 1997, and whether prediction of the count of pollen scattered for the following day is possible. Our studies clearly demonstrated that the count of sugi pollen scattered is small in the early period of the scattering season with some daily variations in amount; this cycle is repeated for several days in the middle period of the season then decreases again in the late period. These findings showed that both the cumulative value of the count of pollen scattered daily and the total count of pollen scattered formed an S-shaped curve when compared with the count of days after the start of the scattering season and the cumulative highest temperature for the season. Predicting the daily count of pollen scattered after the start of the scattering season against the cumulative highest temperature is possible by drawing a regression curve from the S-shaped curve. Also, a regression curve was useful in predicting the total count of pollen scattered during the season from the cumulative count of pollen scattered 10 to 20 days after the start of the scattering season. Using these data, we were able to improve the accuracy for predicting the scattering of sugi pollen.