Pollen recovery in atmospheric samples collected with the Rotorod Sampler over multiple-day periods such as weekends

Abstract

Background: Conventional wisdom holds that multiple-day samples should not be obtained with the Rotorod Sampler because sampling efficiency declines over long sampling periods such as weekends. Objective: The purpose of this present investigation was to test this position under experimental conditions. Methods: Atmospheric pollen samples were collected over 24-hour, 48-hour, and 72-hour periods by five Rotorod Samplers operated in a compact sampling array. Pollen counts obtained from the latter two sampling periods were compared with corresponding values computed from 24-hour samples. Differences in pollen recovery were assessed using Pearson's correlation coefficient and paired comparisons t-tests. Results: Two hundred seventeen atmospheric samples were obtained between March 30 and October 8, 1998. Differences between the 48-hour and 72-hour samples and their computed counterparts were not statistically significant when the atmospheric pollen concentration was less than 100 pollen grains per cubic meter of air (p/m3). At higher pollen concentrations large quantitative and statistically significant differences were noted. Conclusions: These data suggest that particle loading on Rotorod collector rods caused sample quality to deteriorate when atmospheric pollen concentrations exceed 100 p/m3. Multiple-day sampling periods appear to be justified when the atmospheric pollen concentration is less than this threshold. Adjusting the Rotorod Sampler's duty cycle may offer a means to collect samples over periods longer than 24 hours during times of the year when pollen shed is heavy.