Hidrológiai Közlöny 1985 (65. évfolyam)

6. szám - Dr. Kovács György: Szempontok a Dunántúli Középhegység vízforgalmát szimuláló modellek felülvizsgálatához

« 348 Hidrológiai Közlöny 1985. 6. sz. Dr. Kovács Gy.: A Dunántúli Középhegység vízforgalma tosításához. A felszín és a víztartó közötti vízcserét vizsgálva ugyanis megállapítottuk, hogy a fedetlen karszton a beszivárgás nagy: nem har, yagolható azonban el a fedett területeken létrejövő táplálás sem. A függőleges megcsapolás viszont nem számottevő, mert a rések kapilláris emelőképessége jelentéktelen. A függőleges táplálás ezért a karsztos víztartókban nagyobb súllyal jelentkezik, mint más rétegekben, ezért ennek a paraméternek a meghatározására nagyobb gondot kell fordítanunk. A különböző közlések azonban éppen ebben a kérdésben mutatják a vélemények legnagyobb eltérését. Feltétlenül szükséges ezért, hogy felülvizsgáljuk az eddig alkalmazott paramétereket és részletes vízháztartási számításokkal törekedjünk a táplálás legvalószínűbb értékének és területi eloszlásának meghatározására. A hidrodinamikai modellek megbízhatóságának és karsztos áramlási rendszerek jellemzésére történő alkalmazhatóságának vizsgálata igazolta,hogy a regionális folyamatok a módszerrel kellő pontossággal leírhatók. Vitathatatlan, hogy a paraméterek és a határfeltételek meghatározását számos bizonytalanság terheli, azonban a karsztos víztartó sajátos adottságait figyelembevéve remélhető, hogy a modelltől megkívánt pontosságot elérhetjük. Ehhez módszertani segítséget a meglévő nagyszámú adat statisztikai elemzése, a gondosan elvégzett vízmérleg-vizsgálat és a modell kalibrálása (az inverz feladat megoldása) nyújthat. Some aspects for analysing the models simulating the water regime in the Transdanubiaii inountainrange Dr. Kovács, Gy. Corresponding Member of the Hungarian Academy of Sciences There has been the expression of doubts by some specialists about the applicability of the generally applied methods in the modeling of karstic aquifers. Their reasoning is the following: karstic aquifers are discretized systems by nature, because the motion of water is concentrated along the major faults and channels. The network used for the application of numerical methods cannot follow the real network of conduits so the results obtained are dubious. This reasoning is certainly valid around the outletpoints of the system where the flow is concentrated to some major channels. It can be proved, however, on the basis of continuoum approach, that the assumption of a continuous seepage, field characterized by its mean parameters is acceptable if the blocks where natural or artificail drainage takes place are excluded from the field. The basis of regional hydrodynamic models is in any case the assumption that the validity of the linear Darcy-law will bold. Hue to the fact that in the conduits of karstic systems the probability of occurrence of non-laminar flow is substantially greater than in the pores of loose sediments, non-satisfaction of this condition may also contribute to the difficulty of modeling karstic systems. Local velocity is decreasing rapidly if the distance between the point of investigation and the outlet increases. So, if the directly drained blocks are excluded from the regional field of seepage, the problems raised by the non-laminar nature of the flow are practically eliminated. To exclude directly drained blocks from a uniform regional model, it is necessary to set up methods for the calculation by which pressure losses due to local' circumstances can be determined. Meanwhile, one must not forget that the system has three dimensions and will possess the characteristics of non-steady flow and that the transport of water is concentrated. It should be taken into consideration that the variability and uncertainty of hydraulic parameters (porosity, hydraulic conductivity) are greater in karstic layers the in granular sediments. Therefore, a great number of observations is needed if the mean and standard deviation of the parameters are calculated. Observations obtained under natural circumstances should be also considered if the reliability of the quantities describing the hydraulic behaviour of karstic layers is estimated. Areal distribution of the parameters should be determined by the solution of the so-called inverse problem, in other words by the calibration of the model. By determining the boundary conditions one should secure the validity of some aspects which reflect the special nature of karstic aquifers. So, known potentials or fluxes should be considered as boundary conditions not only at the edge of the flow-field but also along the valleys cutting the mountains in parts. Valley-sections having great influence on the motion of water must be explored. Location of these sections must be embedded in the model. Here and along the edges a probable value and the probable temporal changes of this value for the boudary conditions are to be calculated. Estimated values of the above discussed boundary conditions can be applied in order to improve the accuracy of the waterbalance calculations. By investigating the exchange of water volumes between ground surface and water bearing layer it has been concluded that beside the large infiltration rates in open karsts, closed karsts display also sizable vertical water inflow. Vertical drainage, however, is not worth to mention due to available weak capillary lifting forces. The resultant vertical water motion is, therefore, in favour for an increased accumulation in the karstic layers compared to other geological formations. It is reasonable, therefore, to strive for an increased accuracy if this component is estimated. The investigation of reliability and applicability of hydrodyanimc models in karstic systems has proved that these regional processes can be well describet with required accuracy. No doubt that the determination of the parameters and boundary conditions is burdened by sevaral difficulties but it is expected that the required accuracy of the model-results can be attained with due regard on the special characteristics of a karstic aquifer. Methodological aid may be acquired for this by meticulously executed water balance investigations and by a successful calibration of the model.