POLLEN SEASONS OF SELECTED TREE AND SHRUB TAXA IN KRAKÓW AND ITS NEIGHBOURHOOD

The aim of the study was to compare the dynamics of pollen seasons of selected tree and shrub taxa among measurement sites in Kraków and its neighbourhood. The study was performed in Kraków and Piotrkowice Małe in 2002, as well as in Kraków and Giebułtów in 2006. During the study the volumetric method was applied and pollen grains were counted along four horizontal lines. The lowest percentage of Corylus pollen and the highest percentage of Betula pollen were found in the analysed sites. The differences among start dates in various measurement sites in a given year were inconsiderable. Statistically signifi cant differences of SPI values for the majority of taxa were found between measurement sites and between seasons for Kraków. The pollen season dynamics showed one (Betula, Pinaceae) or more maximum values (Corylus, Populus, Fraxinus, Salix). The occurrence of many peaks could be explained by the appearance of several species within one genus in the studied area or by various weather conditions. In 2002 maximum pollen concentrations were recorded earlier than in 2006. The differences in these dates could be explained better by cumulative temperature >5C than >0C.


INTRODUCTION
Information about pollen concentration measurements in urban and suburban areas performed in short time series are known in literature (G o t t a r d i n i and C r i s t o f o l i n i , 1997; K a s p r z y k , 1999).The results of these studies indicate disparities in annual pollen concentrations in various sites, and slight differences in the course of pollen seasons (start dates, duration).
Piotrkowice Małe (PM) (50 o 12'N, 20 o 14'E) a small village, is located about 40 km north of Kraków.Forests occupy about 2% of its area.Mixed forests, subcontinental forests growing on dry ground, pine-oak forests and communities of elm-ash and willow-poplar marshy meadows dominate in this area (T o w p a s z et al. 1998).
Giebułtów (GB), (50 o 08'N, 19 o 53'E) is located about 14 km north west of Kraków.Poplar, willow, alder and hazel prevail.There is a low contribution of pine and beech (T o w p a s z , 1996).
In Kraków a heat island occurs (air temperature increased by 1-2 o C) which is responsible for change of the thermal season duration in the year.In the center of the city, summer lasts longer by 25 days and winter lasts shorter by 23 days than in suburban area.In Kraków the number of hot days is higher by 10-11 days and also accumulated rainfall is higher than in suburban area (L e w i ń s k a , 2000).Mean temperature for January 2002 was -0.5 o C, although in the second half of this month temperature increased up to over 10 o C. February 2002 was the warmest since 1991 (mean temperature was 4.3 o C).The beginning of 2006 was one of the coolest since 1991 (mean temperature for January was -7.0 o C, for February -2.2 o C, for March 1.0 o C) and snow cover was long lasting up to the 22 nd of March (communication of the Department of Climatology, Jagiellonian University).
The study was performed using the volumetric method.In Kraków and Giebułtów the traps were installed at a height of 20 m and in Piotrkowice Małe at a height of 15 m above ground level.The duration of the study in Kraków was from the 15 th of January to the 2 nd of December in 2002 and from the 1 st of March to the 31 st of December in 2006; in Piotrkowice Małe from the 6 th of February to the 11 th of November in 2002 and in Giebułtów from the 8 th of March to the 15 th of August in 2006.
The saw like scatter plot of data describing day to day variation in pollen concentrations were smoothed by means of Generalized Additive Model (H a s t i e et al. 2001).In this modelling we assumed that the number of pollen grains collected in successive days undergoes the Poisson distribution.The pair wise comparison of SPI values calculated for given sites and pollen seasons was done using a version of the Student t-test adopted for counts (S o k a l and R h a l f , 1998).

RESULTS
Table 1 presents the taxa pollen percentage in annual sums of tree pollen, and table 2 shows pollen season parameters.The differences among start dates of various measurement sites in a given year were inconsiderable and amounted to several days except at Fraxinus (12 days) and Quercus (33 days).In 2002 the pollen seasons of all the studied taxa started earlier than in 2006.The highest differences in start dates between particular years were found for Alnus (55 days) and Corylus (53 days).
In the studied sites and years, clearly signifi cant differences in SPI values were recorded, except at KR vs PM in 2002 for Corylus and Pinaceae (borderline signifi cant; p=0.0496) and at KR02 vs KR06 for Salix (p=0.9569).In 2002 SPI values were signifi cantly higher in Kraków for Betula, Fraxinus, Quercus and Salix, and in 2006 for Quercus, Corylus, Salix, Alnus and Pinaceae (Fig. 2, Tab.2).
Figure 3 presents the dynamics of particular taxa pollen seasons.For Betula, Quercus and Pinaceae, one pollen peak concentration was observed.In case of Corylus, Populus, Fraxinus, Salix, more than one peak concentrations were noted.Generally speaking, the highest pollen concentrations were observed earlier in 2002 than in 2006.The delay in 2006 was smaller when it referred to the taxa whose pollen occurred later in the course of the season (Fig. 4).
Different dates of the highest pollen concentrations in 2002 and 2006 could be explained in a qualitative way by using cumulative temperature >0 o C and >5 o C (Fig. 5).Cumulative temperature (from the 1 st of January to the 30 th of March 2006) of 5 o C and over 5 o C was 290 o C on the 90 th day of a year.The same temperature in 2002 was noted on the 75 th day of a year.The ninetieth day in 2006 was also the day of the peak concentration of Corylus pollen.In 2002 the fi rst Corylus pollen peak concentration was almost two months earlier.
Even though the second peak occurrence is taken into consideration, the difference between the studied years was about 25 days.Figure 5 shows that the differences between cumulative temperature curves disappear as the season proceeds.The delay of the season on the 130 th day in 2006 was 8 days in comparison with 2002.The same delay occurred for the fl owering of the latest taxa (Quercus, Pinaceae).PM -Piotrkowice Małe; GB -Giebułtów.
1 The consecutive day from the 1 st of January.

DISCUSSION
The lowest pollen percentage was recorded for Corylus in Kraków and its neighbourhood (Tab.1).The only species Corylus avellana L. occurs commonly, especially on the outskirts of the city on small scattered sites.The highest percentage and SPI value were noted for Betula pollen at all the sites.In 2006 these values were considerably higher than in 2002.In 2006 these values were higher for Giebułtów than for Kraków.In Kraków and in the suburban area, Betula pendula Roth.dominates.G o t t a r d i n i and C r i s t o f o l i n i (1997), performing a similar study in two cities in northern Italy, found a higher pollen concentration of Betula in the bigger city (Trento) where Betula was planted as an ornamental tree.
Our study shows that the SPI value depends on both place and pollen season (Tab. 2 and Fig. 2).Variability in the SPI value in the different sites is associated with plant physiology (alternative cycles of high and low concentrations) and meteorological parameters (S p i e k s m a et al. 1995).E l -G h a z a l y et al. (1993), performing observations of pollen concentrations in Stockholm and Huddinge (15 km away from Stockholm), found signifi cant differences between pollen annual totals.The highest percentage and annual total were observed for Pinus in Huddinge, while for Betula and Quercus in Stockholm.E l -G h a z a l y et al.
(1993)  and R o d r i g u e z -R a j o et al.(2004)  insist on the fact that pollen annual totals are different between close monitoring sites.However, this relationship could be disturbed because of unstable weather conditions.Different start and end dates of pollen seasons depend rather on meteorological factors than on monitoring sites (F r e i , 1998; E m b e r l i n etal.2002).G o t t a r d i n i and C r i s t o f o l i n i (1997) did not observe signifi cant differences between start dates and pollen season duration in two sites in Italy, whereas K a s p r z y k (2003) noted clear differences in start pollen emission in some arboreal taxa in Southwestern Spain.Grana, 38: 306-310.R o d r i g u e z -R a j o F. J., D o p a z o A., J a t o V., 2004.Environmental factors affecting the start of pollen season and concentrations of airborne Alnus pollen in two localities of Galicia (NW Spain).Ann.Agric.Environ.Med.11: 35-44.S o k a l R. R., R h a l f F. J.,1998.Biometry, the principles and practice of statistics in biological research.NY, USA, W. H. Freeman and Company.S p i e k s m a F.TH.M., E m b e r l i n J., H j e l m r o o s M., J ä g e r S., L e u s c h n e r R. M., 1995.Atmospheric birch (Betula) pollen in Europe: Trends and fl uctuations in annual quantities and the starting dates of the seasons.Grana, 34: 51-57.To w p a s z K., 1996.Dolina Kluczwody.Soc.Polonica, 3. Kluczwody Valley.To w p a s z K., K o t a ń s k a M., Tr z c i ń s k a -Ta c i k H., 1998.Notatki fl orystyczne z Płaskowyżu Proszowickiego (Wyżyna Małopolska).Fragm.Flor.Geobot.Ser.Polonica 5: 31-39.Floristic notes from Płaskowyż Proszowicki (Małopolska Upland).Tu r z a ń s k i K. P., P a u l a -W i l g a J., 2002.Raport o stanie środowiska naturalnego miasta Krakowa za lata 1999--2001 z analizą porównawczą pięciolecia 1994-1998./ The report of the Cracow natural environment in the years 1999-2001 with regard to the comparative study in the period 1994-1998.2002.Praca zbiorowa.Biblioteka Monitoringu Środowiska, Kraków.