This study was conducted to derive optimal design floods by Gerneralized Extreme Value(GEV) and Weibull-3 distributions for the annual maximum series at ten watersheds along Han,Nagdong, Geum, Yeongsan and Seomjin reiver systems. Adequency for the analysis of flood data used in this study was established by the tests of Independence. Homogeneity , detection of Outlines, L-moments. Design flood sobtaine dby /methods of Moments and L-Moments using different methods for plotting positions in BEV and Weibull-3 distributions were compared by the Relative Mean Errors(RME) and Root MEan Square Errors(RMSE). The result wa found that design floods derived by the L-moments using the other formulas for plotting positions from the viewpoint of Relative Mean Errors and Root Mean Square Errors.

Field experimnet of constructed wetland for rural wastewater treatment was performed from July 1998 to April 1999 including winter to examine the seasonal effect on the wetland performance. The system worked without freezing even under $-10^{\circ}C$ of air temperature as long as watewater was flowing. BOD removal rates varied in similar pattern as the air temperature, and winter performance was relatively lower than that in the growing season. However, removing performance during winter was still significant, and BOD removal rates were almost the the same as in the growing season. SS removal rate was relativelyless affected by temmperature, but lower decay rate during the winter can result in accumulation of the SS in the system, which releases constituents in the next spring and can affect whole system performance. The winter removal rates of nutrients like T-N and T-P were decreased about half compared to the growing season and low temperature. To maintain stabilized wetland performanced including winter time, supplying minimum heating for plants could be an alternative in field application. Experimental data was compared with NADB(North Americal Wetlands for Water Quality treatment database), and general performance of the system was within the reasonable range. The pollutant loading and effluent concentration of the experimented system were in high margin. Base on the experiment and databases, the required effluent water quality could be achieved if loading rate adjusted as ilulstrated in the database.

The concentration and load of T-N and T-P in paddy field areas in Chongwon, Chungbuk , Korea , were investigated during the irrigation period. The concentration of T-N and T-P in ponded water in paddies is higher than that in the irrigation, drainage and percolation waters. For T-N, the average concentration indrainage water was higher than that in irrigation water until end of Jung, But lower thereafter. The average T-N concentration in irrigation water was 2.3 ∼3.2mg/l and therefore the agricultural water quality standard at on-farm level for T-N should be established realistically . The expontial L(load)-Q(discharge) equations have higher coefficients of determination than the linear L-Q equations . Especially, the exponential L-Q equation of T_N showed a coefficient of determinatino of over 0.9 for irrigation water.

A modified gas tracer method was used to obtain reaeration coefficient from an artificial channel and a reach of Bokha stream, Ichon city. Propane was used as the tracer gas and Rhodamine-B dye as a dispersion and dulution tracer. Concentrations of propane in water sample were measured using a gas chromatograph and concentrationsof dye using UV-Spectrophotometer. To compare measured values with predicted values,commonly used 14 equations were selected . Results of this study suggested that the modified gas tracer method is a potentially useful procedure for th edetermination of reaeration cofficients. However, estimated reaeration coefficients from predictive equations were significantly different from that of this study. Therefore, when using predictive equations, careful selection of equation with consideration for hydraulic characteristics such as flow depth and average velocity, or use of newly derived predictive equation which is adequate for questioned stream would be needed.

In this study , the optimum structural components for the rural house design using the light gauge cold-formed steel frame is proposed. The proposed components were optimzed by the developed model composed with the analysis model and LGC database. The analysis model adapts FEA(finite element analysis). LGC database and calculation of element force adapt the design criteria of KISC. The structure of house is divided into header, bearing wall and foof truss. The variable of the each structure of house are defined component which designed by the case of load, aize and space. The designed weight were used for optimization procedure of the divided components.

In this study, a computer program considering initial imperfection of axisymmetric reinforced concrete shell which plastic deformation by large external loading was developed . Initial imperfection of dome was assumed as 'dimple type' which can be expressed as Wi=(Wo/h)(1-x$^2$)$^3$. The developed model applied to the analysis of dynamic response of axisymmetric reinforced concrete shell when it has initial imperfection. The initial imperfection of 0.0, -5.0, and 5cm and steel and steel layer ratio 0,3, and 5% were tested for numerical examples . The results can be summarized as follows ; 1. Dynmaic response of vertical deflection at dome crown showed slow increased if it has not inital imperfection . But the response showed relatively high amplitude when initial imperfection was inner directed (opposite direction to loading). Similar trends also appeared for different steel layer ratios. 2. Dynamic responses of radial displacement at the junction of dome and wall showed the highest amplitude when initial imperfection was inward directed (opposite direction to loading). The lowest amplitude occurred when initial imperfection was outward directed (same direction to loading). Vibration period also delayed for inward directed initial imperfection . These trends were obvious as steel layer ratio increasing. 3. The effects of imperfection for the dynamic response of radial displacement a the center of wall scarely appeared. The effects of initial imperfection of dome on the dynmaic response of the wall can be neglected. 4. Effect of steel on the dynmic response of axisymmetric shell structure was great when initial imperfection did not exist. And the effect of direction of initial imperfection (inward or outward) did not show big difference.

This study is performed to evaluate the physical and meanical properties of oyster shell concrete. The result shows that the unit weights of concrete with oyster shell are less by 15 ~2% than that of the normla cement concrete. The highest strengths are achieved by 2.5% oyster shell concrete , with increased compressive strength by 4% , tensile strength by 6% and bending strength by 7% than that of the normal cement concrete, respectively. The static modulusof elasticity is in the range of 290$\times$10$^3$~314 $\times$10$^3$kgf/㎤ for 2.5~7.0% oyster shell concrete,which has showed about the same compared to that of the normal cement concrete. The Poisson's number of oyster shell concrete is less than that of the normal cement concrete. Accordingly, oyster shell concrete will improve the properties of concrete.

Recently structural damage has been frequently observed in reinforced concrete brdiges due to repeated loads such as vehicular traffic an due to continual overloads by heavy duty trucks. Therefore, the purpose of this experimental stduy is to investigate the damage mechanism due to fatigue behavior of high-strength reinforced concrete beams under repeated loads. From the test results, the relation of cycle loading to deflection is on the mid-span , the crack growth and the modes of failure according to cycle number, fatigue life and S-N curve were observed through the fatigue test. Based on the fatigue test results , high-strength reinforced concrete beams failed to 57 ∼66 percent of the static ultimate strength . Fatigue strength aobut two million cycles from S-N curves was certified by 60 percent of static ultimate strength.

Tensile behavior in concrete has been neglected until recently. However, the effect of tensile stresses in concrete must be considered where the member primarily carries tensile forces or when ultimate strength is affected by the cracking history. In this paper, a series of experiments were performed with a reinforced rectangular beams of 15 specimens in order to investigate the effect of tension stiffening into the nonlinear analysis and cracking behavior. The experimental results were analyzed in terms of load-deflection curves and strain fracture energy with respect to the main experimental variables such as types of specimen, strength of concrete and steel ration. The results from experiments and finite element analysis were compared in terms of load-deflection relationship and cracking pattern. The results are as follows ; The tension stffening effects of reinforced concrete beams were observedc up to yielding of members after cracking showing strain energy difference of 35 % at the beam of 0.57% steel ratio compared with that of beam ignoring the tension stiffening effect. The tension stiffening of concrete strength 400kgf/$\textrm{cm}^2$ and 600kgf/$\textrm{cm}^2$ increased by 8% and 13%, respectively, compared with that of concrete strength 200kgf/$\textrm{cm}^2$. The tension stiffening effects were greater at a ductile member rather than a brittle one. The load-deflection results of finite element analysis showed very similar results from experiment. The crack growth and pattern might be predicted from the nonlinear finite element analysis considering concrete stiffening.