Nevenka O`ani}
Josip Rubini}
ANALYSIS OF DISCHARGE FLUCTUATION TRENDS IN THE MOUNTAIN PART OF THE KUPA RIVER CATCHMENT AREA
SUMMARY
The upper part of the Kupa river catchment area is situated in the mountain massif of Gorski Kotar, with the highest peaks reaching about 1500 m. It is in this mountain area that the Kupa catchment area, and hence the catchment area of the Black Sea, is only some 15 km away from the Adriatic Sea, and therefore directly exposed to its climatic influence. Therefore, this area is the most abundant in rainfall in Croatia, with rainfall reaching even 4000 mm per annum. With that respect, water resources of the upper part of the Kupa catchment area, in addition to high enviromental value, have an extremely important water supply and hydropower potential which is at present utilized only to a minor extent. However, in accordance with the growing water and power demand, larger use of water resources from this part of the catchment area is planned.
Such plans are being intensified by the fact that, opposite to the global growing trend of water demand, the contrary trends are also present - the falling trends of mean annual discharges in most watercourses throughout the wider region. The situation is similar in the upper part of the Kupa catchmant area, where the falling trend of mean annual rainfall has been particularly noticed in the past ten-year period. Therefore, the approach to planning of the area development and finding of optimum ways to meet the expected water demands should take into account also the falling trend of discharges in watercourses and sources in the area.
Key words:the Kupa river catchment area, discharge fluctuation trends, changes of water regime
1. INTRODUCTION
The analyses of annual discharge fluctuations belong to the basic hydrological data in assessing the capacity of water resources in a given area. The analyses carried out in the upper part of Kupa river catchment area have shown a considerable falling trend of mean annual discharges during the past three decades. The presence of this trend has been somewhat neglected by the public due to the fact that during the last ten years, when the largest decrease of annual discharges was noticed, there were at the same time some very high, even catastrophic floods in some areas. The obvious examples are the flood of the ^abranka river on September 13-14, 1988, and of the Kupica on September 13-14, 1993. However, such short extreme flood events cannot have essential influence on the general trend of discharge fluctuations in the analysed area.
The paper shows the results of analyses carried out at several typical rain and hydrological gauging stations on the Kupa river and its tributaries in the mountain part of the catchment area. The analysis covers the joint 30-year period of stations operation, i.e. 1965-1994. Fig. 1 shows the position of the rain gauges and the hydrological gauging stations.
2. COURSE OF ANNUAL RAINFALL AND MEAN ANNUAL DISCHARGES
The upper part of the Kupa catchment area, due to its geographic position and orographic properties, is comparatively heterogeneous regarding the annual rainfall which ranges from 1500 to 4000 mm. To illustrate the oscillations of annual rainfall in this area, the analysis included two stations - station Mrzla Vodica (751 ma.s.l.) situated on the periphery of the catchment area, with rainfall properties typical for the most upstream part of the Kupa basin, and the station Skrad (675 m a.s.l.) representing the more downstream part of the analysed area. Table 1 shows their typical values, while the course of modular values of annual rainfall and their trends are shown in Fig. 2.
Fig. 2Course of modular values of annual rainfall (1965-1994)
The observed trends of annual rainfalls also influenced the regime of discharges in the area. The hydrological stations selected for the analysis were: Hrvatsko on the Kupa river (370 sq.km. of catchment area), Brod na Kupi on the tributary Kupica (291 sq.km.), and Luke on the downstream tributary Gornja Dobra (175 sq.km.). Their typical hydrological parameters are also shown in Table 1. Fig. 3 shows the course of modular values of the mean annual discharges.
Table 1 Basic data of the analysed gauging stations (1965-1994)
RAIN GAUGES / HYDROLOGICAL GAUGING STATIONSstation / Mrzla Vodica (mm) / Skrad (mm) / Hrvatsko-Kupa (cu.m./sec) / Brod na Kupi-Kupica
(cu.m./sec) / Hrvatsko - Kupa
(cu.m./sec)
Mean annual values / 2688,4 / 1690,3 / 20,00 / 13,467 / 7,010
/ 378,8 / 249,5 / 3,95 / 2,564 / 1,47
Cv / 0,142 / 0,148 / 0,197 / 0,160 / 0,210
Annual max. / 3589,9 / 2110,0 / 419,0 / 18,50 / 9,70
Annual min. / 2054,9 / 1169,0 / 2,11 / 8,40 / 4,60
Fig. 3Modular values of mean annual discharges in typical hydrological gauging stations in the upper part of the Kupa catchment area (1965-1994)
Presentation of typical modular annual values of rainfalls and discharges show a falling trend of annual values in all stations. In rain gauge station Mrzle Vodice this trend is 0.40 percent per annum, and in Skrad 0.57 percent. It may be noticed that in the same period of analysis the calculated falling trends of mean annual discharges are even more evident. In the station Hrvatsko-Kupa the falling trend of mean annual discharges is 1.39 percent per annum, while in the stations Brod na Kupi-Kupica and Luke-Dobra the trends are the same, i.e. 1.34 percent per annum.
The analysis of correlation between modular values of mean annual discharges from the considered stations was also carried out, resulting in a very high degree of coincidence. The coefficient of linear correlation of interrelations of modular values of mean annual discharges in the stations Hrvatsko and Brod na Kupi is k=0.93, for stations Hrvatsko and Luke it is k=0.77, and Brod na Kupi and Luke k=0.71. A comparatively high degree of correlation has also been observed in the analysis of modular values of annual rainfalls and mean annual discharges - for the relation between the Kupa discharge at Hrvatsko and rainfall in Mrzle Vodice k=0.77, for the relation between the Kupica discharge and rainfall in Skrad k=0.81, and for the relation between the Gornja Dobra discharge in Luke and rainfall in Skrad k=0.62.
It may be seen that dry periods occur in series of several years in succession. In order to determine whether the hydrological system of analysed watercourses “remembers” wet or dry intervals in time periods longer than one season, i.e. over one year, autocorrelation analysis of the time series has also been caried out. The analysis was carried out in steps up to four years, and its results are shown in Table 2.
Table 2Results of autocorrelations analysis of annual rainfalls and mean annual discharges in the upper part of the Kupa catchment area (1965-1994)
station / AUTOCORRELATION STEPparameter / 1 year / 2 year / 3 year / 4 year
Mrzla Vodica (rainfall) / -0.08 / 0.36 / 0.14 / -0.017
Skrad
(rainfall) / -0.11 / 0.27 / 0.18 / 0.27
Hrvatsko - Kupa (discharges) / 0.17 / 0.41 / 0.28 / 0.23
Brod na Kupi - Kupica (discharges) / 0.17 / 0.45 / 0.26 / 0.25
Luke - Dobra (discharges) / 0.14 / 0.39 / 0.12 / 0.07
It may be seen from Table 2 that no significant autocorrelation has been determined for any of the stations subject to the analysis. Still, although statisticaly important autocorrelations of time series have not been proved, it has been determined that within the analysed time step up to 4 years the calculated autocorrelation coefficients retain positive values. This, in part shows that their time series have a tendency of slow changes of values, or a certain inertia in changes.
3. ANALYSIS OF CHANGES IN THE HYDROLOGICAL REGIME
In order to obtain a more detailed picture of the occurrence of falling trends of mean annual discharges, the analysis of time series by RAPS method (Rescaled Adjusted Partial Sums) has been carried out on the example of hydrological stations Hrvatsko - Kupa, Brod na Kupi - Kupica, and Luke - Dobra. The results of this analysis are shown in Fig. 4 where prevailingly wet periods “W”, prevailingly dry “D”, and neutral “N” periods are indicated. Noticeably high coincidence is evident, indicating their mutual interconnections.
Fig. 4Transformed series of mean annual discharges in typical hydrological stations (1965-1994)
It may be sean in Fig. 4 that the last decade was characterized by a long lasting dry period. In addition to reduced annual rainfall and thus also reduced discharges, the reasons of this phenomenon should also be sought in the changes of the intra-annual water regime. Therefore, a comparison has been made between mean monthly rainfalls and discharges for the entire period of analysis (1965-1994) and for the last decade (1985-1994), which is shown in Fig. 5.
Fig. 5Mean monthly rainfalls in the rain gauging station Mrzle Vodice and an monthly discharges in the hydrological station Hrvatsko-Kupa
It may be seen that during the last decade the changes occured resulting, in almost all months except October and November, in rainfalls and discharges lower than the average. Regarding the absolute discharge balance, the runoff deficit is particulary marked during the winter months (December-March). However, from the standpoint of the use of water resources, even more concern is raised by the deficit occuring during the dry summer period (July-September).
4. CONCLUSIONS
The results of the hydrological analyses, although modest with regard to the number of stations involved, have indicated the presence of clear falling trends of both the annual rainfalls and, even more, the mean annual discharges in watercourses from the mountain part of the Kupa catchment area. Also, it has been noticed that the last decade was characterized by an expressedly dry period.
Such trends and phenomena make it necessary for future water mnagement development plans to be based, instead of the usual individual hydrological parameters, on more complex hydrological data which include also the assumptions of possible further occurrence of such unfavourable hydrological conditions. With respect to the exceptionally abundant water resources of the upper part of Kupa catchment area, as well as to the high water quality in areas not exposed to human influence, the Kupa water resources will have, in future, to meet water requirements of the wider region. Therefore, it becomes imperative to preserve and to increase the possibility of regulation of the water regime in the upper part of the catchment area, through construction of storage reservoirs, as well as by taking care not to disturb the biological balance with the existing vegetation contributing to alleviation of the surface runoff regime.
5. LITERATURE
1. Bonacci,O., Bonacci,T.:Drought Periods Identification at the Osjek Climatologic Station (Croatia), XVIII. Conference of the Danube Countries, Graz, 1996.
2. O`ani},N., Rubini},J.:Problemi pra}enja stanja i gospodarenja vodnim resursima na primjeru sliva Rje~ine (Problems of Situation Monitoring and Water Resources Management on the Example of the Rje~ina Catchment Area), Hrvatska vodoprivreda 50, Zagreb, 1996.
3. Rubini},J., O`ani},N.: Prirodne hidrolo{ke zna~ajke povr{inskih vodnih pojava (Natural Hydrological Properties of Surface Water Phenomena), basic data for County physical plan (unpublished), Rijeka, 1997.
kontakt adrese autora:
dr.sc. Nevenka O`ani}, dipl.in`.gra|.
HRVATSKE VODE, VGO Rijeka, Ciottina 17b, 51000 RIJEKA
Josip Rubini}, dipl.in`.gra|.
HRVATSKE VODE, VGI Labin, Zelenice 18, 52220 LABIN