Vízügyi Közlemények, 1961 (43. évfolyam)

4. füzet - IX. Képek a Föld különböző részeinek vízépítési munkáiról

(36 > НЕКОТОРЫЕ МЕТОДЫ ПРОГНОЗА ВОДНОГО РЕЖИМА РЕК К. Сестаи, кандидат техн. наук (Венгерский текст на стр. 249) Настоящая краткая монография награжденная премией АН Венгрии суммирует результаты исследовательской работы проведенной в будапештском Водохозяйственном научноисследовательском институте в прошлые годы с целью развивать службы гидрологических прогнозов страны. Глава I и глава IV дают краткий обзор о гидрологических основах организации службы прогнозов и о развитии венгерской службы. Глава II являющаяся костяком работы посвящена прогнозам отдельных событий водного режима (прогноз пика поводков, ежедневные продолжительные прогнозы, прогноз весенних максимальных уровней озер). Глава III представляет некоторые методы долгосрочного прогноза средних и екстремных величин ожидаемого водного режима. Подробное изложение главы II этой работы на русском языке встречается в рукописном докладе автора «Краткосрочные прогнозы водного режима рек в Венгрии» в библиотека ЦИП-а (Москва) и ГГИ (Ленинград). SOME METHODS OF RIVER FORECASTING By Dr. К. Szesztay For the Hungarian text see pp. 249) I. Hydrological foundations of the forecasting service The continuous observation of successive phenomena (events) related to the surface and underground runoff of waters originating from precipitation, or snowmelt, as well as the exploration of relationships existing between them provides the physical basis of forecasting river regimes (Fig 2.). The methods presented in the paper rely to a considerable extent on the graphical analysis of meteorological and hydrological observation data. Relationships involving four and more variables are preferably investigated in a form resolved to part-functions including two or three variables only [9]. The subjective errors and uncertainties of graphical correlation can be eliminated in a relatively simple and rapid manner by appropriate constructing techniques [8]. As far as the practical aim and Lhe applied methods are concerned forecasts relating to definite features of the hgdrograph (such as the peak stages of a flood wave, stages to be anticipated after a given lime "t", peak spring level of lakes, etc.) should expediently be discerned from those relating to average and extreme values characteristic for any extended period (average discharges of a year, or season, lowest limit of the anticipated stages, etc.). II. Forecasting features of the hgdrograph Forecasts relating to flood peaks should in general include two features, namely the peaking stage and the time of occurrence of the peak. In some instances also the forecasting of the shape of the flood hydrograph (the duration of the peak) may be found necessary (Fig. 18). The variables of gage relationships expressing the relation of peaks ensuing in successive sections, can be selected in a variety of ways [see Eqs. (2), (3), (4), (5) and (6) respectively Figs. 2, 4 and 5]. For reaches lying between forecasting key stations local forecasts can be given on the basis of profiles comprising the two-variable gage relationships existing between adjacent gages (Fig. 6 ). When forecasting peaks of flood waves resulting from rainfall on the basis of precipitation data (Fig. 7 ), the precipitation volumes over parts of the catchment area with different runoff conditions should be allowed for with different weights. The auxiliary variables are in this case practicably completed by the index of precedent precipitation. For small catchments the rainfall duration'and the calendar date representing evaporation conditions should also be taken into consideration. Forecasts relating to the peak of flood waves due to snowmelt rely on the magnitude of the snow cover, on the water equivalent thereof and on data concerning the daily positive degree-days (Figs. 8 and 9). In the example shown in Figs. 10 and 11 the forecast^ concerning flood runoff and peak discharge have been separated.