Multi-point analysis of airborne Japanese cedar (Cryptomeria japonica D. Don) pollen by Pollen Robo and the relationship between pollen count and the severity of symptoms

Abstract

An important theme related to aerobiology is the spatial distribution of airborne pollen, which is a final goal of our research efforts. In relation to this, pursuing the severity of pollinosis symptoms and problems with airborne pollen and allergens is required. In the past, we simulated the distribution of airborne Japanese cedar (Cryptomeria japonica D. Don) pollen (Kawashima and Takahashi in Grana 34:142–150, 1995, Grana 38:316–324, 1999). The simulation results of recent years will be published in the near future. At the time, it was not possible to compare the results of actual airborne pollen data observed from multiple points at the same time. However, now we can achieve this objective if using C. japonica pollen characteristics. The pollen is dispersed in large quantities over a long distance, remains in the air for a long time, and most airborne pollen at this time is C. japonica pollen. Recently, several automatic C. japonica pollen monitoring systems have been developed and deployed nationwide. Pollen Robo is one and is the most widely installed nationwide. In this study, we addressed Pollen Robo. First, we examined whether Pollen Robo can accurately count C. japonica pollen and found that Pollen Robo was useful during the major pollen season. However, it sometimes counts non-pollen particles and other pollen rather than C. japonica pollen. Therefore, it is not useful when a small amount of pollen is dispersed such as the start day of the pollen season. Next, we examined the daily dispersion of C. japonica pollen and it was shown that it is related to the distance between sampling locations. However, factors other than distance may be involved. Another matter of concern is the relationship between the number of airborne pollen and pollinosis symptoms. Symptoms began to appear from approximately the start day of the pollen season in moderate and light cases. The most serious symptoms occurred when the pollen count was the highest. Meanwhile, in severe cases, symptoms were observed at the time the pollen was scarcely detected and plateaued when the pollen count reached 100 counts/m3 on average over 6 days. Thus, Pollen Robo was not useful when symptoms began in severe cases. This problem was solved using pollen antigen measurement. The time for the symptoms to begin in severe cases is between 2 and 4 weeks before the start day of the pollen season. The amount of Cry j 1 on the start day during some years was considerably larger than that calculated from the number of pollen. The results obtained in this study will be the basic data used for achieving our final goal.