The aim of the study was to characterise Artemisia pollen season types according to weather conditions in Wrocław (south-western Poland) in the years 2002–2011. Over the period analysed, the start date of the pollen season (determined by the 95 % method) ranged from 10 July 2002 to 28 July 2010. The start date of the pollen season can be determined by using Crop Heat Units (CHUs). During the period 2002–2011, the Artemisia pollen season started after the cumulative value of CHUs had reached 2,000–2,100 °C. The three distinguished types of Artemisia pollen season are best described by the frequency of weather types defined by the type of circulation, mean daily air temperature, and the occurrence of rain. The variation in these factors affected the dynamics of the pollen season. The noteworthy frequency of days with rain and high seasonal sum of precipitation totals as well as the dominance of cyclonic weather from the westerly direction had an impact on the extension of the pollen season. The meteorological factors that directly affect pollen release and transport primarily include air humidity, expressed as vapour pressure (r > 0.3, p < 0.01), temperature(r from 0.2 to 0.4, p < 0.01). The relationships between averaged meteorological data and daily pollen concentration were stronger (r > 0.5, p < 0.01). Based on the correlation analysis, the meteorological variables were selected and regression equations were established using stepwise backward regression analysis.

We have investigated the relationship between the inflow of air masses and the ragweed pollen concentration in SW Poland (Wrocław) for a 10-year period of 2005–2014. The HYSPLIT trajectory model was used to verify whether episodes of high concentrations can be related to regions outside of the main known ragweed centres in Europe, like Pannonian Plain, northern Italy and Ukraine. Furthermore, we used two different meteorological data sets (the global GDAS data set and from the WRF mesoscale model; the meteorological parameters were: U and V wind components, temperature and relative humidity) into HYSPLIT to evaluate the influence of meteorological input on calculated trajectories for high concentration ragweed episodes. The results show that the episodes of high pollen concentration (above 20 pm−3) represent a great part of total recorded ragweed pollen in Wrocław, but occur rarely and not in all years. High pollen episodes are connected with air masses coming from south and south-west Europe, which confirms the existence of expected ragweed centres but showed that other centres near Wrocław are not present. The HYSPLIT simulations with two different meteorological inputs indicated that footprint studies on ragweed benefit from a higher resolution meteorological data sets.Electronic supplementary materialThe online version of this article (doi:10.1007/s10453-017-9471-9) contains supplementary material, which is available to authorized users.

No abstract

No abstract