The Compaction of fill dam is very important for increasing of the safty of dam. Vibration roller is used for the compaction of pervious materials such as sand and gravel. The principal objects of this study are to give a comstruction criteria of vibration roller and to find out the relationship between dry density and permeabity of pervious soil after compaction. The results in this study are summerized as follows. 1.The relationship between maximum dry density (Υdmax) and optimum moisture content(Wo) of modified compaction test is Υdmax=2. 74-0. 064w0 2.The maximum dry density decrease with increasing fine particle(n) and the relative formular is n==ae-brdmax 3.The maximum dry density is influenced more by passing rate of number 200 sieve than 4 sieve. 4.The coefficient of permeability are similar when the degrees of compaction are equal even though the spreading thickness of soil are different. 5.The coefficient of permeability(K)is greatly influence by fine particle passing number 200 sieve, and those relationship is inversely proportionate. 6.The K values of pervious soil are from 10-0 cm/sec to 10-4 cm/sec when degree of compaction by a modified method is from 90 to 95percent. 7.The coarser material is little influenced on the permeability with different density. 8.The increasing rate of permeability with decreasing degree of compaction is more influened by fine pacticle than number 200 sieve. When degree of compaction decrease from 100 percent to 90 percent the K values of SM and GM increase about 20 times but GW increase 6 times only. 9.The effect of compaction by vibration roller is greatly influenced by 6 passes and the increasing rate of the effect is decraased at 8 passes. 10. In order to get the degree of compaction of 95 percent or more, 6 to 8 passes of roller are generall required with 30 cm thickeness of soil for 4.5 ton to 6.5 ton vibration roller and 7 to 8 passes is required with 50cm thickness for 8 to 12 ton roller.

This study was attempted to get dimensionless unit hydrograph by linear model which can be used to the estimation of flood for the development of Agricultural water resources and laid emphasis on the application of dimensionless unit hydrographs for the ungaged watersheds by applying linear model. The results summarized through this study are as follows. 1.Peak discharge is found to be Qp= CAR (C =0. 895A-o.145) having high significance between peak discharge, Qp and effective rainfall, R within the range of small watershed area, 84 to 470km2. consequently, linearity was acknowledged between rainfall and runoff. Reasonability is confirmed for the derivation of dimensionless unit hydrograph by linear model. 2.Through mathematical analysis, formula for the derivation of dimensionless unit hydrograph was derived. qp--p=(tp--t)n-1[e-(n-1)](tp--t-1) 3.Moment method was used for the evaluation of storage constant, K and shape parameter, n for the derivation of dimensionless unit hydrograph. Storage constant, K is more closely related with the such watershed characteristics as length of main stream and slopes. On the other hand, the shape parameter, n was derived with such watershed characteristics as watershed area, river length, centroid distance of the basin and slopes. 4.Time to peak discharge, Tp could be expressed as Tp=1. 25 (√s/L)0.76 having a high significance. 5.Dimensionless unit hydrographs by linear model stood more closely to the observe dimensionless unit hydrographs On the contrary, dimensionless unit hydrographs by S.C. S. method has much difference in comparison with linear model at the falling limb of hydrographs. 6.Relative errors in the q/qp at the point of 0.8 and 1.2 for the dimensionles ratio by linear model and S. C. S. method showed to be 2.41, 1.57 and 4.0, 3.19 percent respectively to the q/qp of observed dimensionless unit hydrographs. 7.Derivation of dimensionless unit hydrograph by linear model can be accomplished by linking the two empirical formulars for storage constant, K, and shape parameter, n with derivation formular for dimensionless unit hydrograph for the ungaged small watersheds.

This study was carried out to investigate the relation between seepage losses and flow section area in earth canals. Totally 77 seepage measurement was gained by ponded method and the tested canals belong to the irrigation area of Farmland Improvement Association in each province, Korea. The results obtained from this study may be used as a reasonable criteria for the estimation of canal seepage losses in the design of irrigation systems. Obtained results are summarized as follows: 1. Average seepage rates in each Soils is 14cm/day in ML, 6. 3cm/day in CL and 24.9 cm/day in SM. 2. Water depth and water surface width in eath canals have little influenced on the seepage rate, while the seepage losses was increased in proportion as the water surface width lengthens. 3. A formula of S=C.An defining a relationship between seepage losses and flow section was derieved as follows. ML:S=O. 35 VA 1.20 (m$^3$/day/m) CL:S=O. 13 VA 0.84 SM:s=O.67VA-1.56 4. The average seepage loss rates per 1km of canal are as followings. Measured Time ML CL SM 0-4 hrs 2.2% 0.6% 4.5% 4-2 4hrs 1.0% 0.15% 2.0% In above table we may obtain the following results. The first row is suitable for the canal having short delivery time of irrigation, while the second row for the canal having long delivery time.

In order to investigate the characteristics of the shear strength of the unsaturated cohesive soils which has mean characters of sand and clay widely used for banking, I selected soil samples from An-sung district and, against it, performed direct shear test and unconfined compression test changing grain size, compaction energy and moisture content and also performed triaxial compression test under optimum moisture content. The results are as follows; 1.As the passing percent of the No. 200 sieve increased from 23.6% to 56.1%, cohesion increased from 0. 202kg/cm2 to 0. 398kg/cm2 under the direct shear test and from 0.38 kg/cm2 to 1. 05kg/cm2 under the tria4al compression test, internal friction angle decreased from 44. 78$^{\circ}$ to 34. 34$^{\circ}$ under the direct shear test and from 31. 88$^{\circ}$ to 13. 31$^{\circ}$ under the triaxial compression test. 2.Cohesion showed it's maximum value around OMC and internal friction angle showed a tendency to increase according to the decrease of water content but it's increasing ratio was relatively slow. 3.Decreasing ratio of cohesion and internal friction angle was relatively sensitive according to the decrease of compaction energy. 4.The smaller of the vertical stress and the coarser of the grain size of samples, changing of the volume showed a tendency to increase and as the increase of water content, the shear displacement (dh) at failure shear stress ($\tau$f) showed maximum and the $\tau$f-dh curve was gentle. 5.To synthesize the results of the direct shear test and the triaxial compression test, cohesion showed higher under the triaxial compression test and internal friction angle showed a tendeney to appear higher under the direct shear test. It seems that we can get correspondent results by removing the side friction of mold with soils and adjusting the vertical stress and shearing speed under the direct shear test.