cycles of pollen emission for different days, so that
averaging the data sets would be meaningless. In order to
compare the different days, pollen concentrations were
normalized by the maximum concentration at canopy height
for the same day, cmax. A subsequent comparison between
diurnal pollen concentration distributions at z/h = 1 and 2 is
presented in Fig. 6 where days exhibiting similar diurnal
pollen concentrations are grouped. To emphasize trends,
average values are depicted as black squares connected by a
thin line. It can be seen clearly that in each group variations of
c(ts)/cmax are small compared to inter-group variations,
especially during the mornings. The large inter-group variation
is mainly due to the fact that initial morning pollen
emission starts at different times for different days. A detailed
analysis of the effect of meteorological conditions on the
diurnal cycle of pollen emission is presented in Sections 4 and
5.