Volume XXXVII-XXXVIII (37-38)
Michał Kasina, Joanna-Pociask-Karteczka, Zenon Nieckarz
Tendencje występowania wysokich przepływów w dorzeczu Dunajca w II połowie XX wieku.
(Trends in high river flows in the Dunajec River drainage basin in the second half of the 20th century)
Kasina M., Joanna-Pociask-Karteczka , Nieckarz Z., 2006-2007, Tendencje występowania wysokich przepływów w dorzeczu Dunajca w II połowie XX wieku., Folia Geographica ser. Geographica-Physica, vol. XXXVII-XXXVIII (37-38): 5-35 (in polish).
Keywords: floods, trends, Carpathians, parametric tests, nonparametric tests
Summary: Detection of temporal changes in the extremes of hydrological elements is a very important task and contribution in the context of climate changes studies. Studies were based on the normalized daily discharge data from 20 catchments with area differentiated from 58.4 to 5316 sq. km in the Dunajec River basin. The Dunajec River (L = 247 km) is one of the longest tributaries of the Vistula River. Its basin (A = 6 798 sq. km) includes the Tatra Mts. - the highest part of the Carpathian mountain chain with Gerlach (2655 m a.s.l.). The area spreads through southern Poland and northern Slovakia. The location of the Carpathians is noticeable and distinctive: they are forming a border between two areas with different influences of Atlantic and continental air masses. Investigated catchments represent typical for mountain rivers flow regimes with two maxima: in spring and summer. Snowmelts are the most consistent factor influencing spring peak (March, April). The summer rain peak occur in July and August. Summer rainfall and floods is more variable in occurrence, amount and intensity.
The main goal of this paper is detecting changes of high river discharges in the Dunajec River Basin. The high river discharge taken into consideration was defined as discharge bigger than the percentile 5% of daily river discharges in particular years. Three different methods were used to estimate whether there is a significance positive or negative tendency in the high annual, winter and summer discharges: parametric linear regression, and non-parametric tests (Mann-Kendall test and Sperman's rank correlation test). Test statistic for the linear regression is r Pearson's correlation coefficient of the time variable and the observations values. Also the slope (change per year) of the best-fitted regression line was calculated by the method of least squares. Using linear regression needs to satisfy the following conditions: the series values must be normally distributed and independent. When the observations don't have the normal distribution, a non-parametric and distribution-free test should be used. Mann-Kendall's test is a non-parametric test based on t correlation coefficient. The test statistic is Kendall's sum divided by the square root of its variance. Sperman's rank correlation can be understand as a Pearson correlation coefficient, but calculated not from original data, but from ranks. Independence of observations is the main assumption of those non-parametric tests.
It is worth to emphasize that all test used in trend detection are based on strong assumption on the time series: independence (all parametric and non-parametric tests) and normal distriburtion (the parametric tests) (a common mistake is to apply any statistical test without checking its assumptions). The W Shapiro-Wilk's test was used for checking normality in observation series, and three for checking the data independence were used - the series test, median crossing test devised by Fisz, and Kendall's turning point test. Distributions of many series weren't normal and elements of many series weren't independence. It means that theoretical formulae for estimation of trend significance are unsuitable for this study. Instead, significance levels were estimated using bootstrapping method.
For each station, tables 2-7 shows the results of trend detection for high annual, winter and summer discharges in three periods: 1971-1990, 1984-2003, 1951-2003. The results of trend detection, estimated using Mann-Kendall's test, are also shown on figures 2-7. Only one series shows statistically significant changes (winter high discharges in Lepietnica for 1971-1990). For the great part of series (90%), trends are not significant even at very high values of a significance level. Higher values of critical significance level were found in 1971-1990 period than in 1984-2003. The lowest critical significance level were found in high winter discharges. Low values of the critical significance level means that the hypothesis of a lack of a change is rejected in the light of evidence. If the critical significance level is 0,01 (1%), there is a high chance that a change exist in the hydrological time series. Usually the critical significance level was higher than 0,02 (2%), so changes weren't detected at the 0,05 (5%) significance level, and even at the 0,10 (10% ) significance level. Links between time and observation series were very small - all correlation coefficient used in this paper (r Pearsona, R Spearmana, t Manna-Kenadalla) were very slight (from -0,3 to 0,3).
Finally it may be stated, that statistically significant increase or decrease of high river discharges can not be found (except Lepietnica). In means, that climate change expressing in warming and decreasing precipitation in Poland do not lead to changes in high river discharges in the Dunajec River basin.
Download full version (0.92 MB)
Ladislav Holko, Hubert Kostka
Snow cover in northern Slovakia - past, present and future
Holko L., Kostka H., 2006-2007, Snow cover in northern Slovakia - past, present and future, Folia Geographica ser. Geographica-Physica, vol. XXXVII-XXXVIII (37-38): 37-51.
SummaryDownload full version (0.97 MB)
Pavol Miklanek, Pavla Pekarova
Influence of forest on snowmelt runoff in small highland basins in Slovakia.
Miklanek P., Pekarova P., 2006-2007, Influence of forest on snowmelt runoff in small highland basins in Slovakia., Folia Geographica ser. Geographica-Physica, vol. XXXVII-XXXVIII (37-38): 53-64.
SummaryDownload full version (0.64 MB)
Zjawiska lodowe na Morskim Oku w Tatrach
(Ice phenomena on the Morskie Oko Lake in the Tatra Mts.)
Choiński A., 2006-2007, Zjawiska lodowe na Morskim Oku w Tatrach, Folia Geographica ser. Geographica-Physica, vol. XXXVII-XXXVIII (37-38): 65-77 (in polish).
SummaryDownload full version (0.48 MB)
Użytkowanie terenu a zmiany sieci dróg polnych na Pogórzu Cięzkowickim w XX wieku.
(The correlation between the land use and changes in the unmetalled road network in Pogórze Ciężkowickie in the 20th century.)
Kroczak R., 2006-2007, Użytkowanie terenu a zmiany sieci dróg polnych na Pogórzu Cięzkowickim w XX wieku., Folia Geographica ser. Geographica-Physica, vol. XXXVII-XXXVIII (37-38): 79-94 (in polish).
SummaryDownload full version (1.26 MB)
Relacje między temperaturą wody w rzece a temperaturą powietrza (na przykładzie rzeki Ropy).
(Relationship between water temperature in the river and air temperature (on the Ropa River as an example).)
Wiejaczka Ł., 2006-2007, Relacje między temperaturą wody w rzece a temperaturą powietrza (na przykładzie rzeki Ropy)., Folia Geographica ser. Geographica-Physica, vol. XXXVII-XXXVIII (37-38): 95-105 (in polish).
SummaryDownload full version (0.64 MB)