• Journal of Korean Society of Rural Planning
• /
• v.28 no.4
• /
• pp.105-117
• /
• 2022

This study tested SVM(support vector machine), RF(random forest), and ANN(artificial neural network) machine-learning models that can predict net irrigation water requirements in paddy fields. For the Jeonju and Jeongeup meteorological stations, the net irrigation water requirement was calculated using K-HAS from 1981 to 2021 and set as the label. For each algorithm, twelve models were constructed based on cumulative precipitation, precipitation, crop evapotranspiration, and month. Compared to the CE model, the R² of the CEP model was higher, and MAE, RMSE, and MSE were lower. Comprehensively considering learning performance and learning time, it is judged that the RF algorithm has the best usability and predictive power of five-days is better than three-days. The results of this study are expected to provide the scientific information necessary for the decision-making of on-site water managers is expected to be possible through the connection with weather forecast data. In the future, if the actual amount of irrigation and supply are measured, it is necessary to develop a learning model that reflects this.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.56 no.5
• /
• pp.1-9
• /
• 2014

The DIROM (Daily Irrigation Reservoir Operation Model) was modified to estimate amounts of water release from reservoirs, considering customary irrigation water management practices, such as water supply for puddling and transplanting paddy rice from seeding beds and mid-season drainage. The applicability of the modified model was investigated by simulating amounts of water release from three study reservoirs: Hwamae, Ogi, and Doya Reservoirs. In terms of annual amounts of water release, the relative errors between the observed and simulated values in 2012 and 2013 ranged -26.20 % to 10.28 % and 4.90 % to 30.06 %, respectively; in case of reservoir water levels, the RMSE values ranged 0.45 m to 1.34 m and 0.40 m to 1.27 m, respectively. Also, it was revealed that the model provided better simulation results for monthly water releases than the original model. In addition, the model presented better performance in simulating 10-day amounts of water release from April to June. However, the model had still significant errors in the simulation results from July to September because the reservoirs were practically operated to adapt to water management circumstances. Finally, it is concluded that the modified DIROM can estimate the amounts of water release from reservoirs, reflecting irrigation water management customs in paddy-field districts. To achieve higher prediction accuracy of the model, it is necessary to incorporate practical reservoir operation rules into the model.

• Korean Journal of Environmental Agriculture
• /
• v.8 no.2
• /
• pp.93-102
• /
• 1989

Water pollution status of the irrigation water, inorganic nutrient contents of rice plant, relationship between water quality of irrigation water and inorganic nutrient contents of rice plant, and nitrogen supply by irrigation water at six sites of pumping stations in Kimhae plain were investigated. The results were as follows : 1. Average values of water components analyzed at all pumping stations were in the range of 6.5-7.0 pH, 0.6-7.7ppm DO, 4.4-63.5ppm BOD, 7.3-73.9ppm COD, 1.84-16.22ppm $NH_4$-N, 1.22-15.49ppm $NO_3$-N and 0.07-1.35ppm $PO_4$-P respectively. 2. Comparatively heavy polluted sites were Sikman, Bongrim and Noksan, and less polluted sites were Daejeo, Myeongje and Jangyou, judging from BOD, COD and $NH_4$-N of the irrigation water. 3. Nitrogen supply by irrigated water for a year were 6.82Kg/10a, 5.98Kg/10a, 6.64Kg/10a, 2.31kg/10a, 6.22Kg/10a and 2.54Kg/10a in Daejeo, Sikman, Bongrim, Myeongji and Noksan area, respectively. 4. Positive correlation was observed between ammonium nitrogen of the water and total nitrogen contents of rice plant. Total nitrogen contents of rice plant were higher in Sikman, Borgim and Noksan area than in the other areas.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.42 no.6
• /
• pp.706-711
• /
• 1997

Low supply of water is generally the most serious factor limiting rice production. The experiment was conducted to identify the reasonable method for minimum irrigation under ridge direct seeding on dry paddy, at National Honam Agricultural Experiment Station, RDA in 1996. The results showed that the reduction ratio of irrigation water was high in order of furrow irrigation at 15-day>furrow irrigation at 10-day>flooding irrigation at 10-day>, and furrow irrigation at 5-day intervals. However, milled rice yield was high in the furrow irrigation at 5-day intervals and in flooding irrigation at 10-day intervals due to high ripened grain as compared with other treatments indicating two treatments were the most reasonable irrigation methods in terms of saving the labor cost and water supply as well as the admittable yield performance.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.44 no.2
• /
• pp.75-80
• /
• 2002

The severity of drought in (the) irrigation reservoir could be evaluated by the accumulative rainfall method, soil moisture content method, storage ratio method, and water supply restricted intensity method, etc. The storage ratio method would be the most reliable one for irrigation reservoir. The pattern of drought might be forecast with the most similar pattern of accumulative rainfall and/or storage ratio out of the file of past operation history.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.65 no.2
• /
• pp.21-33
• /
• 2023

While the main purpose of irrigation reservoirs is to supply agricultural water, the needs of environmental flow and flood control has been expanded. The agricultural reservoirs have been operated in the form of carry-over system until now. Therefore, the supply of agricultural water is difficult when the storage rate is not sufficiently secured after large volume of irrigation. In addition, there are regulation of the upper storage rate for some large reservoirs during the flood season, but lower storage rate is not regulated. Accordingly, this study aims to evaluate the capacity of agricultural water and environmental flow supply by setting the management lower storage rate of reservoir. The changes in the supply of agricultural and environmental flow was simulated according to the three different regulating lower storage rate scenarios. As a result, it was judged effective in terms of water supply managing the lower storage rate up to 30% when the initial storage rate of farming period is above annual average for the Naju reservoir considering existing water management practice. If the lower storage rate would have been controlled above 30%, the supply of agricultural water might be increased and non-effective discharge amount would be decreased compared to other scenarios during dry period of 2016-2018.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.65 no.6
• /
• pp.23-35
• /
• 2023

In this study, we developed a method to determine the flood-limited water levels of agricultural reservoirs, considering both their irrigation and flood control functions. Irrigation safety and flood safety indices were defined to be applied to various reservoirs, allowing for a comprehensive assessment of the irrigation and flood control properties. Seasonal flood-limited water level scenarios were established to represent the temporal characteristics of rainfall and agricultural water supply and the safety indices were analyzed according to these scenarios. The optimal scenarios were derived using a schematic solution based on Pareto front analysis. The method was applied to Obong, Yedang, and Myogok reservoirs, and the results showed that the characteristics of each reservoir were well represented in the safety indices. The irrigation safety of Obong reservoir was found to be significantly influenced by the late-stage flood-limited water level, while those of Yedang and Myogok reservoir were primarily affected by the early and mid-stage flood-limited water levels. The values of irrigation safety and flood safety indices for each scenario were plotted as points on the coordinate plane, and the optimal flood-limited water levels were selected from the Pareto front. The storage ratio of the optimal flood-limited water levels for the early, mid, and late stages were 65-70%, 70%, and 75% for Obong reservoir, 75%, 70-75%, and 65-70% for Yedang reservoir, and 75-80%, 70%, and 50% for Myogok reservoir. We expect that the method developed in this study will facilitate efficient reservoir operations.

• KCID journal
• /
• v.12 no.1
• /
• pp.4-14
• /
• 2005

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.29 no.3
• /
• pp.125-137
• /
• 1987

The purpose of this paper is to examine some difficulties in water management of agricultural reservoirs in Korea, for there are approximately more than 15,000 reservoirs which are now being utilized for the purpose of irrigation, along with the much amount of expenses and labors to be invested against droughts and floods periodically occurred. Recently, the effective use of water resources in the agricultural reservoirs with a single purpose, is becomming multiple according to the alterable environment of water use. Therefore, the task to allocate agricultural water rationally and economically must be solved for the multiple use of agricultural reservoirs. On the basis of the above statement, this study aims at suggesting the rational method of water management by introducing an optimal technique to allocate the water in an existing agricultural reservoir rationally, for the sake of maximizing the economic effect. To achieve this objective, a reservoir, called "0-Bongje" as a sample of the case study, is selected for an agricultural water development proiect of medium scale. As a model for the optimum allocation of water in the multi-purpose use of reservoirs a linear programming model is developed and analyzed. As a result, findings of the study are as follows : First, a linear programing model is developed for the optimum allocation of water in the multi-purpose use of agricultural reservoirs. By adopting the model in the case of reservoir called "O-Bongje," the optimum solution for such various objects as irrigation area, the amount of domestic water supply, the size of power generation, and the size of reservoir storage, etc., can be obtained. Second, by comparing the net benefits in each object under the changing condition of inflow into the reservoir, the factors which can most affect the yearly total net benefit can be drawn, and they are in the order of the amount of domestic water supply, irrigation area, and power generation. Third, the sensitivity analysis for the decision variable of irrigation which may have a first priority among the objects indicate that the effective method of water management can be rapidly suggested in accordance with a condition under the decreasing area of irrigation. Fourth, in the case of decision making on the water allocation policy in an existing multi-purpose reservoir, the rapid comparison of numerous alternatives can be possible by adopting the linear programming model. Besides, as the resources can be analyed in connection with various activities, it can be concluded that the linear programing model developed in this study is more quantitative than the traditional methods of analysis. Fifth, all the possible constraint equations, in using a linear programming model for adopting a water allocation problem in the agricultural reservoirs, are presented, and the method of analysis is also suggested in this study. Finally, as the linear programming model in this study is found comprehensive, the model can be adopted in any different kind of conditions of agricultural reservoirs for the purpose of analyzing optimum water allocation, if the economic and technical coefficients are known, and the decision variable is changed in accordance with the changing condition of irrigation area.

• Korean Journal of Agricultural Science
• /
• v.50 no.4
• /
• pp.903-916
• /
• 2023

This study aims to establish basic data for efficient management of rural water by analyzing regional irrigation facilities and benefitted areas in the statistical yearbook of land and water development for agriculture at the watershed level. For 511 domestic rural water use areas, water storage facilities (reservoirs, pumping & drainage stations, intake weirs, infiltration galleries, and tube wells) are spatially distributed, and the benefitted areas provided at the city/county level are divided by water use area to provide agricultural water supply facilities. The characteristics of rural water district areas such as benefitted area, were analyzed by basin. The average area of Korea's 511 rural water districts is 19,638 ha. The average benefitted area by rural water district is 1,270 ha, with the Geum River basin at 2,220 ha and the Yeongsan River basin at 1,868 ha, which is larger than the overall average. The Han River basin at 807 ha, the Nakdong River basin at 1,121 ha, and the Seomjing River basin at 938 ha are smaller than the overall average. The results of this basic analysis are expected to be used to set the direction of various supply and demand management projects that take into account the rational and scientific use and distribution of rural water and the characteristics of water use areas by presenting a quantitative definition of Korea's agricultural water districts.