Zmiany intensywności opadów w Hornsundzie (Spitsbergen) w okresie 1978-2008

Artykuł porusza problematykę zmian intensywności opadów atmosferycznych w Hornsundzie (Spitsbergen) w okresie 1979-2008. Szczegółową analizę opadów dobowych przeprowadzono w przedziałach/ klasach kwantyli równych sum opadów, które wyznaczono metodą zaproponowaną przez Karla i in. (1995) oraz Karla i Knighta (1998), a zmodyfikowaną przez Osborna i in. (2000), Brunettiego i in. (2001) oraz Brunettiego (2004). Wydzielone klasy kwantyli sum opadów zawierają w sobie łączną informację o sumie i liczbie dni z opadem, dzięki czemu zostały uznane za uproszczony wskaźnik intensywności opadów. Dla każdego z wydzielonych wskaź-ników intensywności, obliczono tendencje udziału opadów w całkowitej sumie rocznej oraz w sumach półroczy i sezonów. W Hornsundzie, w okresie 1979-2008 istotne trendy wzrostowe intensywności opadów zaobserwowano jedynie w okresie rocznym, w półroczu chłodnym oraz jesienią. W pozostałych porach roku oraz w półroczu ciepłym tendencje były nieistotne statystycznie, chociaż na wiosnę zaznaczył się spadek intensywności opadów.

This paper deals with problems of temporal changes in distribution of precipitation intensity in Hornsund (Spitsbergen) in the period 1979-2008. Trends in precipitation totals, number of days with precipitation and average daily precipitation amount were calculated in order to assess general changes in precipitation intensity. Detailed analysis of changes in distribution of daily precipitation amount was conducted by categorizing all daily precipitation into 10 classes of precipitation intensity and then computing the proportion of each seasonal total which was provided by events of each class. This class intervals, which can be thought of as 10 equal amount quantiles, were calculated after Osborn et al. (2000). In the next step trends in contribution of each precipitation intensity class-interval to annual and seasonal (warm and cool half-year, spring, summer, autumn, winter) precipi-tation total were calculated. Linear regression fitted by the least square method was used to assess trends magnitude and direction. Trends were expressed in percentage of average indices values for the 1979-2008 period (relative trends) which enabled the comparison of the trends magnitude calcu-lated for indices expressed with different units and for class-intervals characterized by various range of daily precipitation amounts. Statistical significance of trends was checked by Mann-Kendall test. Trends with significance level p between 0.2 and 0.1 were recognized as slightly significant whereas trends with p equal to or less than 0.1 were recognized as significant. The 1979-2008 period was the base for trend analysis. Moreover, several arbitrary selected long-term periods (1979-00, 1979-01, 1979-02, …, 1979-08) were also considered in order to assess the trends stability. In Hornsund in the period 1979-2008 significant changes in daily precipitation intensity appeared on annual basis as well as in cool half year and in autumn. There was a significant increase in annual precipitation total and no sign of any changes in the number of days with precipitation. These were a result of an increase in contribution of extreme precipitation (K10) to total annual precipitation and a decrease in contribution of daily precipitation of low intensity. Similar pattern of precipitation intensity changes was noted in cool half year. In this case the changes are a consequence of decreasing trends in contribution of lower intensity class-intervals (K1, K2 I K4) to total precipitation. In autumn statistically significant were increasing trends in precipitation totals, number of days with precipitation and average daily precipitation amounts – at the same time the increases in precipitation totals and average precipitation amounts were much more higher than in the number of days with precipitation. These changes were a result of decreasing trends in contribution of K1, K2 and K3 intensity classes and increasing trends in contribution of K7 and K8 intensity classes in total autumn precipitation. In spring statistically significant drop in precipitation totals accompanied by unchanged number of days with precipitation indicates the decrease in precipitation intensity. However, detailed analysis of trends in contribution of class-intervals to total winter precipitation has not proven any significant changes of precipitation intensity in this season. It is worth to notice that there were negative trends in contribution of majority of precipitation intensity class-intervals to total spring precipitation which probably resulted in significant decrease in intensity of total precipitation. It also should be mentioned that there are opposite trend directions in spring and autumn precipitation totals. In Hornsund the relationships between precipitation and atmospheric circulation depend on daily precipitation intensity. In majority of seasons daily precipitation of the least intensity (class-interval K1) are the most frequent during an inflow of air masses form east under influence of low pressure system. Precipitation form K4 and higher intensity classes were usually noted in SWc type. It must be said that the results of this study are probably influence by well-known problem of precipitation measu-rement errors in polar regions especially in relation to snow and cool period of year.