Pollen count forecasting.

Abstract

Pollen forecasting is becoming increasingly important to allergists as an adjunct to effective patient care. Forecasts allow patients to avoid exposure to high pollen levels and prompt them to take prophylactic medication and to plan outdoor activities for periods of low pollen levels. Investigators are making progress in developing effective models for daily and seasonal forecasts for important pollen allergens; however, current models are limited to specific geographic areas. Models for the onset of the season for spring tree pollen are based on the chilling and heat units that are required before flowering can occur. Models for pollen season severity are based on regression analysis of preseason meteorologic conditions, and models for daily forecasts are based on the normal pollen curve and responses to day-to-day meteorologic conditions. When winds are favorable, long-distance transport can introduce allergenic pollen types into a local area. The Mountain Cedar Pollen Forecasting model, which combines day-to-day release forecasts at source areas and dispersion forecasts to downwind areas, has been reasonablly successful over the past 4 years. All pollen forecasting models are dependent on accurate meteorologic forecasts, and pollen forecasting will become more accurate as meteorologic forecasts improve.