Variability between Hirst-type pollen traps is reduced by resistance-free flow adjustment

During an international campaign organized in Munich (Germany) in 2021 to test the performance of automatic pollen traps, we ran four manual Hirst-type pollen traps in parallel. All 4 Hirst-type pollen traps were set and monitored on a weekly basis for the entire campaign to 10 L/min using the same standard hand-held rotameter. Afterwards, a hand-held heat-wire anemometer (easyFlux®) was used additionally to obtain the correct flow without internal resistance. Uncorrected pollen concentrations were 26.5% (hourly data) and 21.0% (daily data) higher than those obtained after correction with the easyFlux®. After mathematical flow correction, the average coefficient of variation between the four Hirst traps was 42.6% and 16.5% (hourly and daily averages, respectively) for birch and 36.8% and 16.8% (hourly and daily averages, respectively) for grasses. When using the correct flow of each pollen trap (i.e. the resistance free anemometer measured flow), for hourly values, the median standard deviation across the traps for the eight pollen types was reduced by 28.2% (p < 0.001) compared to the uncorrected data. For daily values, a significant decrease in the median standard deviation (21.6%) between traps was observed for 7 out of 8 of the pollen types, (p < 0.05 or lower). We therefore recommend continuing to calibrate Hirst-type pollen traps with standard hand-held rotameters to avoid changing the impacting characteristics of the instruments, but simultaneously also measure with resistance-free flow meters to be able to apply flow corrections to the final pollen concentrations reported. This method improved the accuracy of the final results.