Juniperus ashei (JA) pollen is responsible for a severe form of allergic rhinitis known as "cedar fever" in the Texas-Oklahoma area. Pollination of JA occurs during December and January and long distance transport of JA pollen has been well documented in previous studies. Starting in Dec 1998 pollen forecasting has been undertaken to warn downwind populations of the potential threat of high pollen concentrations in the atmosphere. Forecasts are posted on the internet daily during the pollination season, (http://pollen.utulsa.edu ). Over the past four seasons a total of 628 forecasts were issued. Forecasting requires a thorough knowledge of the ecology, phenology, and aerobiology of JA along with knowledge of local and regional weather. The daily forecast includes both a release forecast and a downwind forecast. The release forecast is based on meteorological conditions and the phenology of the plants. Once pollen cones are mature, the meteorological conditions necessary for pollen release include air temperatures above 45^oF and relative humidity below 50%. Sunshine and the absence of rainfall during the previous 24 hours also enhance pollen release; however, extended periods without rain resulted in reduced pollen production. Assessment of the forecasts has been based on air sampling data from our Burkard spore traps at two sites within the JA population on the Edwards Plateau of central Texas. These data show a strong relationship between recorded pollen levels and predicted conditions for release. The downwind forecast is based on trajectories generated using the HY-SPLIT model from NOAA Air Resources Laboratory (http://www.arl.noaa.gov/ready/hysplit4.html) and meteorological conditions along the path. Air sampling data from our three samplers in the Tulsa area were used to test the accuracy of the forecast trajectories and predicted threat to downwind communities. Analysis of the 98/99, 99/00, and 00/01 seasons revealed only two occurrences of "high" or "very high" pollen concentrations in Tulsa that were not directly linked to a wind trajectory from a source area. Results indicate that our pollen forecasting was successful in predicting the influx of JA pollen concentrations dispersed over long distances. Aerobiology, plant phenology, and meteorology can be successfully used to warn individuals about the exposure risk due to high atmospheric pollen levels.