Vízügyi Közlemények, 1981 (63. évfolyam)

2. füzet - Baranyó Géza: A bevlízkárok matematikai modellezése

Belvízkárok matematikai modellezése 265 The compound losses incurred annually from 1953 to 1975 in the Körös Region have been compiled in Table II. Regarding these as sample elements and assigning influence factors to the losses, the author formulated a mathematical model. The loss occurrence-probability curve and the durations of inundation by year are illustrated in Figs. 2 and 3, respectively. For the correlation of inundation losses Xi and the corrolary influence factors two kinds of analysis are presented : — an approximate rank-correlation method and — calculation by multi-variate linear regression equations. The rank-correlation procedure was developed (Section 5) to explore the relative importance by order of magnitude of the individual factors in causing the inundation loss. The inundation loss Xi and the precipitation Хг are correlated in Fig. 5. By introducing the sensitivity index a close correlation was found to exist (R = 0.796) between the inundation loss Xi and the duration x

3 of inundation. The ratio defined as the sensitivity index is a time-variant value factor, which indicates by months the relative magnitude of damage to the main crops if inundation would occur in the particular month (Table VII). A similarly close correlation (R = 0.8068) was established between the inundation loss A'i and the water mass Xi pumped annually to the rivers. The inundation loss Xi is but loosely related to the canal density xr„ the annual average air temperature xe and the average crop yield x<i, as indicated by the low correlation coefficients R = 0.22, 0.036 and 0.1754, respectively. The inundation losses A

ri were correlated with fair closeness to the groundwater levels observed in the wells at Kondoros x

7 and at Füzesgyarmat х» (R = 0.46 and 0.74, respectively). The rank correlation analysis has revealed: a

) the influence of the individual factors on the magnitude of losses, b) the change in the magnitude of losses caused by modifying the individual factors. From estimating the inundation losses by the multi-variate linear regression equations (Section 6) the author arrived at the conclusion that for describing the loss-phenomenon the four-variate function No. 3 given by Eq. (2) is most sucessful, since this shows the lowest standard deviation (Table IX). * * * Modèles mathématiques des dégâts des eaux stagnantes par dr. Baramjó G. Dans les bassins des rivières Körösök où il y a des zones favorables á l'exploitation agricole, la sécurité hydraulique de production n'est pas suffisante á cause de retoure systématique des eaux stagnantes. Cependant ces eaux évacuées pendant la période de la stagnation, elles pourront étre utiles á la satifaction des besoins en eaux d'irrigation, c'est-à-dire leures retours seraient nécessaires. Nous avons considéré les sommes annuelles des dégâts provoqués par les eaux stagnantes dans les régions des Körösök comme les éléments des expérimentation (Tableau III), nous avons recherché les causes des dégâts, et á partir de ces données, l'auteur a formulé un modèle mathématique. La figure 2. représente les dégâts en fonction de ces probabilités d'occurance, et la figure 3. représente la série annuelle de la durée de la stagnation. L'auteur expose deux variantes d'analyse de l'interdépendance entre les dégâts des eaux stagnantes (Ai) et les causes de celles-ci: la première variante est un approche: il s'agait de l'application de la méthode de la recherche de la corrélation de rang, et