• 농업과학연구
• /
• 제46권4호
• /
• pp.843-856
• /
• 2019

Real-time flood prediction has an important role in significantly reducing potential damage caused by floods for urban residential areas located downstream of river basins. This paper presents an effective approach for flood forecasting based on the construction of a deep neural network (DNN) model. In addition, this research depends closely on the open-source software library, TensorFlow, which was developed by Google for machine and deep learning applications and research. The proposed model was applied to forecast the flowrate one, two, and three days in advance at the Son Tay hydrological station on the Red River, Vietnam. The input data of the model was a series of discharge data observed at five gauge stations on the Red River system, without requiring rainfall data, water levels and topographic characteristics. The research results indicate that the DNN model achieved a high performance for flood forecasting even though only a modest amount of data is required. When forecasting one and two days in advance, the Nash-Sutcliffe Efficiency (NSE) reached 0.993 and 0.938, respectively. The findings of this study suggest that the DNN model can be used to construct a real-time flood warning system on the Red River and for other river basins in Vietnam.

• 한국환경과학회지
• /
• 제22권8호
• /
• pp.965-977
• /
• 2013

Reliable long-term streamflow forecasting is invaluable for water resource planning and management which allocates water supply according to the demand of water users. It is necessary to get probabilistic forecasts to establish risk-based reservoir operation policies. Probabilistic forecasts may be useful for the users who assess and manage risks according to decision-making responding forecasting results. Probabilistic forecasting of seasonal inflow to Andong dam is performed and assessed using selected predictors from sea surface temperature and 500 hPa geopotential height data. Categorical probability forecast by Piechota's method and logistic regression analysis, and probability forecast by conditional probability density function are used to forecast seasonal inflow. Kernel density function is used in categorical probability forecast by Piechota's method and probability forecast by conditional probability density function. The results of categorical probability forecasts are assessed by Brier skill score. The assessment reveals that the categorical probability forecasts are better than the reference forecasts. The results of forecasts using conditional probability density function are assessed by qualitative approach and transformed categorical probability forecasts. The assessment of the forecasts which are transformed to categorical probability forecasts shows that the results of the forecasts by conditional probability density function are much better than those of the forecasts by Piechota's method and logistic regression analysis except for winter season data.

• 한국환경과학회지
• /
• 제22권8호
• /
• pp.953-963
• /
• 2013

Reliable long-term streamflow forecasting is invaluable for water resource planning and management which allocates water supply according to the demand of water users. Forecasting of seasonal inflow to Andong dam is performed and assessed using statistical methods based on hydrometeorological data. Predictors which is used to forecast seasonal inflow to Andong dam are selected from southern oscillation index, sea surface temperature, and 500 hPa geopotential height data in northern hemisphere. Predictors are selected by the following procedure. Primary predictors sets are obtained, and then final predictors are determined from the sets. The primary predictor sets for each season are identified using cross correlation and mutual information. The final predictors are identified using partial cross correlation and partial mutual information. In each season, there are three selected predictors. The values are determined using bootstrapping technique considering a specific significance level for predictor selection. Seasonal inflow forecasting is performed by multiple linear regression analysis using the selected predictors for each season, and the results of forecast using cross validation are assessed. Multiple linear regression analysis is performed using SAS. The results of multiple linear regression analysis are assessed by mean squared error and mean absolute error. And contingency table is established and assessed by Heidke skill score. The assessment reveals that the forecasts by multiple linear regression analysis are better than the reference forecasts.

• 한국농공학회논문집
• /
• 제50권6호
• /
• pp.3-12
• /
• 2008

It is important to estimate accurate effective rainfall to analyse flood flow and long-term runoff for the rational planning, design, and management of water resource. The initial abstraction is also important to estimate effective rainfall. The Soil Conservation Service (SCS) has developed a procedure and it has been most commonly applied to estimate effective rainfall. But the SCS method still has weak points, because of unnatural assumptions such as antecedent moisture conditions and initial abstraction. The coefficient of initial abstraction(K) is depending on the soil moisture condition and antecedent rainfall. The maximum storage capacity of Umax which is calibrated by stream flow data in the proposed watershed was derived from the DAWAST(DAily WAtershed STreamflow) model. The values of K obtained from 69 storm events at the five watersheds are ranging from 0.133 to 0.365 and its mean value is 0.207. Effective rainfall could be estimated more reasonably by introducing new concept of initial abstraction. The equation of $K=0.076Sa^{0.255}$ was recommended instead of 0.2 and it could be applicable to the small-medium rural watersheds.

• 환경영향평가
• /
• 제9권3호
• /
• pp.239-248
• /
• 2000

The aggravating water quality from the expansion of industrialization along with increasing population lead to develop more intensive physical measures to secure better drinking water quality. This study was mainly initiated to establish a water-pollutant buffering zone for the upper stream basin of Paldang--the major source area of drinking water for the metropolitan Seoul and suburban areas with a population more than 13 million. Two different criteria were considered in determining the buffering distance from the edge of the streamflow : 1km-width buffer zone for the special protection area which has been strictly controlled by the conventional laws for the protection of drinking water supply, and 500m-width buffer zone for the rest of the area. To delineate the exact boundaries of the water-pollutant buffering zone, GIS database was created integrating topography, hydrography, cadastral, and other related layers. The newly designated water-pollutant buffering zone would contribute to improve the water quality in a long term along with the conservation of the wet land. More study, however, should be made within the water-pollutant buffering zone such as the detailed survey of the pollutants, vegetation, and ecosystem for more effective management of the buffering zone.

• 한국농공학회논문집
• /
• 제48권1호
• /
• pp.27-38
• /
• 2006

In this study, we developed a program to estimate discharge efficiently considering major hydraulic characteristic including water level, river bed, water slope and roughness coefficient in a natural river. Stream discharge was measured at Gongju gauge station located in the down stream of the Daechung Dam during normal and dry seasons from 2003 to 2004. The developed model was compared with the results from the existing rating curve at T/M gage stations, and was used for runoff analyses. Evaluating the developed river discharge estimation program, it was applied during 1983-2004 that base flow separation method and RRFS (Rainfall Runoff Forecasting System) which is based on SSARR (Streamflow Synthesis And Resevoir Regulation). The result presents the stage-discharge curve creator range at the Gong-ju is overestimated by approximately $10-20\%$, especially at the low stage. It is attributed to the hydraulic characteristics at the study. The discharge simulated by the RRFS and base flow separation, which is calibrated using the measurement at the early spring and late fall season during relatively d]v season, shows the least errors. The coefficient of roughness at Gongju station varied with the high and low water level.

• 한국환경과학회지
• /
• 제20권11호
• /
• pp.1457-1464
• /
• 2011

In this study, the method of estimating hydrologic information (water depth, submerged period etc.) on the proper selection of construction point and scale as well as vegetation type suggested for the design of natural riparian rehabilitation structure. Long-term comprehensive watershed model SWAT-K(Korea) was applied to this purpose. Flow duration analysis was conducted to analyze the hydrologic characteristics of Pyungchang watershed at which the 'bangtul' construction method was tested. For this purpose 20 years (1989-2008) rainfall runoff analysis was carried out. Based on the simulated daily streamflow data, flow duration curve was made to analyze the flow characteristics, and the water depth hydrograph was made to analyze the water depth distribution at the cross section. Finally, the information for the selection of proper vegetation according to the submerged period is suggested.

• 농촌계획
• /
• 제11권1호
• /
• pp.17-24
• /
• 2005

The purpose of this study is to evaluate the hydrological impact due to temporal land cover change by gradual urbanization of Anseong-cheon Gongdo watershed ($371.8km^2$). Land covers of 1981, 1990, 2000 Landsat TM images were classified by maximum likelihood method. The watershed showed a trend that forest & paddy areas decreased about $33km^2$ and $27km^2$, respectively and urban area increased about $11km^2$ during the periods. To identify the impact of streamflow due to urbanization, WMS HEC-1 was used. According to apply Huffs quartile storm events by changing land cover data, peak runoff discharge of each frequency rainfall (50, 100, 500 years) increased about 56, 36, $192m^3/sec$, respectively.

• 한국산림과학회지
• /
• 제85권4호
• /
• pp.634-646
• /
• 1996

본 연구는 산지유역을 대상으로 하여 임지의 수질정화능을 구명하고자 실시했다. 대한리 및 박달리의 2개 유역에서 임외우 및 수관통과우, 수간류, 토양수, 계류수의 산지 물순환 각 과정에 대한 수량과 산도, 전기전도도, 용존산소 등을 측정하여 수질변화를 분석한 결과는 다음과 같다. 1. 임외우의 평균 pH는 박달리 7.6, 대한리 6.4로 나타났다. 2. 적송과 리기다소나무의 수간유하수와 수관통과우의 수질을 비교한 결과, 형균 pH는 대한리에서 각각 4.32와 4.22, 박달리에서 각각 3.34와 4.81로 나타났으며, 평균 전기전도도는 박달리에서 각각 $230.0{\mu}S/cm$, $82.0{\mu}S/cm$이었고, 대한리에서 각각 $119.7{\mu}S/cm$, $96.8{\mu}S/cm$로 나타났다. 3. 계류수의 pH는 유출량의 증가에 따라 감소하였으나, 전기전도도 및 용존산소량은 값이 높아지는 것으로 나타났다. 4. 적송의 임내우(수간류와 수관통과우)가 상수리나무의 임내우보다 pH는 낮게 나타났으며, 전기전도도는 높게 나타났다. 5. 두 시험지의 산지 물순환 과정 중에서 산림토양의 수질정화능력이 가장 높은 것으로 나타났다. 6. 물순환 과정에 따라 수질 변화를 분석한 결과, 습성 강하물에 의한 오염물질의 부하보다는 건성강하물에 의한 부하와 수목의 수피에 의한 영향이 더 큰 것으로 판단되었다. 그러나 산성 임내우도 토양을 거치면서 모암에 의해 중화되고 토양입자에 의한 침착이 일어나면서 계류로 유출되는 유출수의 수질은 양호하게 나타났다.

• 한국수자원학회논문집
• /
• 제51권6호
• /
• pp.503-514
• /
• 2018

최근 기후변화 및 유역개발로 인하여 메콩강 유역의 수문환경이 급격히 변화하고 있으며, 메콩강을 공유하는 국가의 수재해 예방 및 지속가능한 수자원개발을 위해서는 메콩강 주요지점에서의 유량 정보의 분석 및 예측이 요구된다. 본 연구에서는 물리적 기반의 수문모형인 SWAT과 데이터기반 딥러닝 알고리즘인 LSTM을 이용하여 메콩강 하류 Kratie 지점의 유출모의를 수행하고, 유출모의 정확도 및 두 가지 방법론의 장 단점을 비교 분석한다. SWAT 모형의 구축을 위해 범용 입력자료(지형: HydroSHED, 토지이용: GLCF-MODIS, 토양: FAO-Soil map, 강우: APHRODITE 등)을 이용하였으며 warming-up 및 매개변수 보정 후 2003~2007년 일유량 모의를 수행하였다. LSTM을 이용한 유출모의의 경우, 딥러닝 오픈소스 라이브러리인 TensorFlow를 활용하여 Kratie 지점기준 메콩강 상류 10개 수위관측소의 두 기간(2000~2002, 2008~2014) 일수위 정보만을 이용하여 심층신경망을 학습하고, SWAT 모형과 마찬가지로 2003~2007년을 대상으로 Kratie 지점에 대한 일수위 모의 후 수위-유량관계곡선식을 이용하여 유출량으로 환산하였다. 두 모형의 모의성능 비교 검토를 위하여 모의기간에 대해 NSE (Nash-Sutcliffe Efficiency)을 산정한 결과, SWAT은 0.9, LSTM은 보다 높은 0.99의 정확도를 나타내는 것으로 분석되었다. 메콩강과 같은 대유역의 특정 지점에 대한 수문시계열 자료의 모의를 위해서는 다양한 입력자료를 요구하는 물리적 수문모형 대신 선행 시계열자료의 변동성을 기억 학습하여 이를 예측에 반영하는 LSTM 기법 등 데이터기반의 심층신경망 모형의 적용이 가능할 것으로 판단된다.