Vízügyi Közlemények, 1974 (56. évfolyam)

2. füzet - Kovács György: A felszíni lefolyás általános vizsgálata és az árvizek előrejelzése. II. rész

244 Kovács György duced into deterministic relations for estimating the liydrological parameters and time of the anticipated liydrological event. In designing problems, on the other hand, statistical methods are applied in estimating, regardless of the time of occurence, the cxpected range of discharges or stages in a flood of a particular probability, or the distribution of the parameters of floods likely to occur with different probabilities. This clear definition of concepts contributed greatly to eliminating much of the former conflicts in opinion. Considering the flood forecasts based on hydrometeorological data distincion must be made between floods originating from storm rainfalls, from snowmelt and from ice jamming. The tools for forecasting storm runoff floods are the mathematical models mentioned before and determined by systems analysis. The practical application thereof is, however, made difficult by the fact that on large catchments the processes are too complex to permit the formulation of reliable models. On small catchments, on the other hand, if the observed data of actually falling precipitation are used as the input, the advance time attainable is too short to be of any practical purpose. It is for this reason that the solution of the problem of quantitative precipitation forecasting, as mentioned before, may be delineated as the most important objective of future research, a promising approach to which involves the cyclonic classification of rainfalls, demonstrated in the paper. As a result of research concerned with the determination of the snow-water equivalent it may be concluded that there is no need for further expanding the present network of GO snow density stations, only the automation of observation (involving the placement and mounting of the snow cushions already developed) is considered desirable. The purpose of these observation stations is on the one hand to provide checks on existing data and on the results computed from them, on the other to contribute the data needed for further research. For practical purposes the volume of water stored in the snow pack is estimated in a reliable manner in all catchments where the thickness of the snow cover is observed. The further objectives of research may be stated as: determination of the rate of melting as a function of meteorological parameters and the study of effective precipitation (surface runoff) resulting from meltage water. In connection with ice-jam floods the most important advance during the period under consideration involves the development of a new observation method for quantitative ice data, the general introduction of which into general practice is believed desirable. Studies performed by the suitable arrangement and grouping of hydrological data have demonstrated the possibility of issuing warnings for the flood fighting organization once observations indicate the probability of ice-jam formation. It is believed necessary to check these results in wider practice and to improve the warning system by introducing more recent data. 6. Hydrological forecasting of the flood parameters The hydrological characteristics of a flood which has already developed over the upper reaches of a major stream can be used for preparing forecasts for the downstream reaches, since these methods afford the required advance time 011 long river sections only. The methods applied can in general be classified into three groups, namely the hgdrodynamic methods relying on the continuity — and dynamic equations alike, the methods involving the use of an empirical correlation determined from hydrological data actually observed in combination with the continuity equation (flood routing, the application of a distribution function) and finally the method using the direct correlation of observation data (the determination of gage relation curves). During the period under consideration research has been conducted and results have been attained in each of the three groups. Nevertheless, the forecasts still include a number of uncertainties, which are due in part to the complex, ramified structure of the river systems, i.e., to natural conditions, but also to the theoretically incomplete solution of the multi-variable hydrodvnamic process. Of the research work aimed at obtaining a better understanding of the theory of flood wave passage in