Hidrológiai Közlöny 1958 (38. évfolyam)

5. szám - Juhász József: A beszivárgás vizsgálata

Juhász J.: A beszivárgás vizsgálata Hidrológiai Közlöny 1958. 5. sz. 3Jf5 An Investigation into Moisture Moveinents in Soils By J. Juhász The physical aspects of infiltration are extremely complex and involved. Experimentál methods and conelusions arrived at thereby are described in the first part of the present paper, while the second part is devoted to theoretical considerations. Infiltration experiments may in generál be classified into the following three groups : a) Sprinkling or flooding tests on experimentál plots, b) investigation of an isolated catehment area using natural precipitation data, c) lysimeter investigations. Actual conditions are reproduced most faithfully by the sprinkling test on experimentál plots, while lysimeter investigations yield results containing the greatest number of variables. Following the first period of accumulation, when all of the applied water was absorbed by the soil, very high rates of infiltration into water-free capillary and non-capillary soil interstices were observed during the first minutes of rainfall in all of the tests conducted. Identical infiltration curves, yet displaced relatíve to each other were found for both dry and moist soils. Infiltration rates were found to reduee rapidly and eonsiderably during subsequent periods and to approacli a constant value owing to the filling up of soil interstices, to the decreasing of the capillary force and to the air enclosed in the soil interstices. As revealed by these experiments : 1. Results obtained by sprinkling tests may bo applied to the computation of infiltration due to natural precipitation (W. W. Korner, Leonard Loyd, L. K. Xhermán). 2. According to L. K. Sherman infiltration curves obtained for the samo soil at different moisture contents are similar. 3. The rate of infiltration depends upon the season (W. W. Horner, Leonard Loyd, Edward L. Bentner, Ralph R. Oaabe, Róbert E. Horton). 4. The rate of infiltration is hardly affected by the slope of the soil if the latter is bare or covered by sparse vegetation. Experimentál results are approximated by exponential or hyperbolical relations of whieh tliose of Róbert E. Horton, M. F. Sozykshin and A. N. Befani are noteworthy. Physical aspects of infiltration have been the subject of numerous investigations. The generally accepted view is to distinguish three subsequent phases : 1. Precipitation in the initial phase falls on the surface of leaves and tends to fill local irregularities of the soil. No analytical method has as yet been developed for investigating the process occurring during the first few seconds that constitute the initial phase. 2. Infiltration into capillary ancl gravitational voids begins in the second phase. 3. The infiltration pattern is already established in the third phase and water is absorbed by two-phase percolation. Computations of percolation phenomena in phases 2. and 3. are based on the determination of the effeetive partiele size of soils composed of different fractions. A method for determining the effeetive partiele size is described. The analytical determination of infiltration phenomena may be based on the Navier-Stokes equation assuming verticai percolation. For sucli conditions a few simplifying assumptions permit the uniform treatment of several infiltration problems. Measurements of Preobrashensky sliowed a fair agreement between actual and predicted values. Air resistance to infiltration can be accounted for by the methods of N. N. Bindernann, Zunker and others. For practical application the one involving the consideration of a head reduced by a coefficient between 0,75 and 1,0 may be suggested. For rapid, approximating computations of infiltration phenomena Fig. 13 may be used, where ordinatae and abseissae represent the velocity of infiltration and the time passed since the beginning of infiltration, respectively, while the depth of infiltration and the effeetive partiele size serve as parameters. This rapid, approximating method cannot be used unless the depth of infiltration is established as a distinct front. Our investigations showed this to occur already at a depth of 4 to (i cm below the soil surface. A void ratio of 30 per cent and an originally air-dry soil have been assumed in constructing the diagram. Data of practical value can alsó be obtained by plotting the infiltration velocity, respectively the resulting infiltration depth, for various soils against the precipitation depth. Infiltration velocities into air-dry soils and into suclr having all voids fiiled with water are represented by the left, respectively right boundaries of the zones pertaining to the effeetive partiele sizes indicated in Fig. 14. The progress in time of water percolating into soils of various moisture content is represented by the zone between the two above limits assuming two-phase conditions above the front of infiltration. Infiltration veloeites of water-air mixtures, i. e., under three-phase conditions are illustrated in Fig. 15, where the coefficient of percolating velocity (abseissae) has been plotted against the amount of pores filled with water (ordinatae). As can be seen from the figure soils of different permeability change their permeability characteristies to varying degrees in case of low air contents ranging from 0 to 40 per cent. Coarse soils respond more appreciably to the air content of airwater mixtures, whereas, owing to the capillary draw, the fine ones are less affected by the presenee of air. With water contents exceeding 20 per cent no significant differences can be detected in the beliaviour of fine and coarse soils, and the same reduction factor can be applied. The method described offers a uniform basis for the treatment of all infiltration problems and for the computation of water quantities entering into the soil from natural precipitation or from irrigation supplies.