• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.410-410
• /
• 2017

지하수 관리를 위해서는 시공간적인 변동성을 고려한 지하수 함양량의 정량적 산정이 필수적이다. 본 연구에서는 지표수-지하수 연동해석이 가능하며 토지이용 특성과 국내 토양특성을 가장 잘 표현할 수 있는 한국형 장기 유출 모형 SWAT-K를 이용하여 김천지역의 분포형 지하수 함양량을 산정하였다. 행정경계와 수자원단위지도에서 제시하는 표준단위유역을 기준으로 하여 김천시를 포함하는 유역을 SWAT-K 구동을 위한 모델영역으로 설정하여 주하도를 따라 19개의 소유역으로 구분하였다. SWAT-K를 구동하기 위해서는 기상 및 수문자료를 구축해야 하는데 강우량을 비롯하여 기온, 풍속, 일사량, 상대습도 등의 기상자료가 요구된다. 본 연구에서는 대상유역 내에 위치한 구미, 추풍령, 거창, 상주 기상관측소와 김천, 지례, 부항1, 부항2, 선산 강우관측소의 자료를 이용하여 기상 및 강우자료를 구축하였으며, 모형의 계산시간, 모형결과의 정확도 등을 판단하여 30m 공간해상도를 가지는 DEM을 300m 공간해상도로 가공하여 사용하였다. 토지이용도는 모의시 다양한 토지이용상태를 반영할 수 있도록 중분류(1:25,000) 토지이용도를 사용하였다. 토양도는 국립농업과학원에서 토양도 전산화 사업을 통해 구축된 1:25,000 축척의 정밀토양도를 사용하였다. SWAT-K를 이용하여 김천시를 포함한 전체유역에 대해 지표수-지하수 통합 물수지 분석 결과(2008년~2015년) 연평균 강수량 대비 유출률은 61.2%, 증발산률은 36.3%, 함양률은 18.0%로 나타났다. 지표수 유출과정과 지하수위 변동을 동시에 고려하여 산정한 소유역별 연간 함양량 결과를 산정하였고, 총 19개의 소유역별 연간 지하수 함양량을 제시하였다. 또한 SWAT-K 모형을 이용한 모델 영역중 김천시에 속하는 행정구역별, 표준권역별 연평균 함양량을 산출하였으며, 그 분석 결과 김천시 평균 함양률은 18.2%로 산정되었다.

• Journal of Korea Water Resources Association
• /
• v.41 no.12
• /
• pp.1231-1244
• /
• 2008

The objective of this study is to present a methodology for estimating runoff curve number(CN) using SWAT model which is capable of reflecting watershed heterogeneity such as climate condition, land use, soil type. The proposed CN estimation method is based on the asymptotic CN method and particularly, it uses surface flow data simulated by SWAT. This method has advantages to estimate spatial CN values according to subbasin division and to reflect watershed characteristics because the calibration process has been made by matching the measured and simulated streamflows. Furthermore, the method is not sensitive to rainfall-runoff data since CN estimation is on a daily basis. The SWAT based CN estimation method is applied to Chungju dam watershed. The regression equation of the estimated CN that exponentially decays with the increase of rainfall is presented.

• Journal of Korea Water Resources Association
• /
• v.45 no.9
• /
• pp.861-874
• /
• 2012

To fulfill applicability of Soil and Water Assessment Tool (SWAT) model, it is important that this model passes through a careful calibration and uncertainty analysis. In recent years, many researchers have come up with various uncertainty analysis techniques for SWAT model. To determine the differences and similarities of typical techniques, we applied three uncertainty analysis procedures to Chungju Dam watershed (6,581.1 $km^2$) of South Korea included in SWAT-Calibration Uncertainty Program (SWAT-CUP): Sequential Uncertainty FItting algorithm ver.2 (SUFI2), Generalized Likelihood Uncertainty Estimation (GLUE), Parameter Solution (ParaSol). As a result, there was no significant difference in the objective function values between SUFI2 and GLUE algorithms. However, ParaSol algorithm shows the worst objective functions, and considerable divergence was also showed in 95PPU bands with each other. The p-factor and r-factor appeared from 0.02 to 0.79 and 0.03 to 0.52 differences in streamflow respectively. In general, the ParaSol algorithm showed the lowest p-factor and r-factor, SUFI2 algorithm was the highest in the p-factor and r-factor. Therefore, in the SWAT model calibration and uncertainty analysis of the automatic methods, we suggest the calibration methods considering p-factor and r-factor. The p-factor means the percentage of observations covered by 95PPU (95 Percent Prediction Uncertainty) band, and r-factor is the average thickness of the 95PPU band.

• Journal of Korea Water Resources Association
• /
• v.42 no.1
• /
• pp.9-19
• /
• 2009

This study presents the effects of temporally varied groundwater table on hydrological components such as surface runoff, evapotranspiration, and soil water content. To this end, the SWAT-MODFLOW model in which the groundwater module of SWAT is replaced with MODFLOW model has been used with a modification to enhance the coupling between the water content in soil profile and the groundwater in shallow aquifer. The variable soil layer construction technique (VSLT) is developed in the present work to represent the direct interaction of soil water and groundwater more realistically, and then the VSLT is incorporated into SWAT-MODFLOW model. In VSLT, when the simulated groundwater table rises within the soil zone, the soil layers below the water table is regarded as a portion of the shallow aquifer, so that those layers are excluded from the initially defined soil zone and are governed by the MODFLOW. From the simulation tests for the Musim river basin, the improved SWAT-MODFLOW model with VSLT is found to correctly evaluate the spatial distributions of overland flow, soil moisture, evapotranspiration according to the groundwater table variation.

• Journal of Korea Water Resources Association
• /
• v.53 no.5
• /
• pp.347-356
• /
• 2020

The main objective of the study is to propose the most efficient SWAT model calibration method using SWAT-CUP with less computing time and high performance. In order to achieve the goal, Case1-3 (250, 500, and 1,000 simulation runs) and Case4 (1,000 simulation runs in the first iteration and then 500 simulation runs for the following iterations) were defined to compare the results. When evaluating the values of the objective function, Case2 and Case3 reached the same value after the fourth iteration, and Case1 reached the closed value of Case2-3 after the eighth iteration. However, the final estimates of the parameters had different ranges in Cases1-3, and only the results of Case3 and Case4 converged similarly. Thus, it can be considered that the parameter calibration results are highly affected by the initial number of simulation runs. On the other hand, SWAT simulation results did not show the significant difference after the first iteration, unlike the parameter ranges. From the analysis results, we can conclude that the most suitable and effective method was to repeat one or two times of iterations with a sufficient number of simulation runs, as in Case4.

• Journal of Korea Water Resources Association
• /
• v.53 no.11
• /
• pp.985-997
• /
• 2020

SWAT (Soil and Water Assessment Tool)-MODFLOW (Modular Groundwater Flow) is a coupled model that linking semi-distributed watershed hydrology with fully-distributed groundwater behavior. In this study, the groundwater simulation results of SWAT and SWAT-MODFLOW were compared for Bokhacheon watershed in Namhan river basin. The models were calibrated and validated with 9 years (2009~2017) daily streamflow (Q) data of Heungcheon (HC) water level gauge station and the daily groundwater level observation data of Yulheon (YH). For SWAT, the groundwater parameters of GW_DELAY, GWQMN, and ALPHA_BF affecting baseflow and recession phase were treated. The SWAT results showed the coefficient of determination (R²) of 0.7 and Nash-Sutcliffe model efficiencies (NES_Q, NSE_inQ) for Q and 1/Q with 0.73 and -0.1 respectively. For SWAT-MODFLOW, the spatio-temporal aquifer hydraulic conductivity (K, m/day), specific storage (S_s, 1/m), and specific yield (S_y) were applied. The SWAT-MODFLOW showed R², NSE_Q, and NSE_inQ of 0.69, 0.74, and 0.51 respectively. The SWAT-MODFLOW considerably enhanced the low flow simulation with the help of aquifer physical information. The total streamflow of SWAT and SWAT-MODFLOW were 718.6 mm and 854.9 mm occupying baseflow of 342.9 mm and 423.5 mm respectively.

• Journal of Korea Water Resources Association
• /
• v.44 no.7
• /
• pp.577-590
• /
• 2011

In this study, the watershed and water body models were linked for the simulation of the Nakding river flow. This is a pre-step study for the estimation of the effect of the flow and water quality on the climate change. For models of watershed and river flow, the SWAT and EFDC were used, respectively. The runoff discharge at each boundary points for the simulation of the river flow was provided from the drainage basin model. The calculated runoff discharge by the SWAT model was compared with the measured data of the Ministry of Environment at 13 locations along the Nakdong river and 30 locations along the tributary streams. The computed water discharge was shown to be similar with the measured data. For the model calibration and verification, % difference, NSE, and $R^2$ were computed. The computed % difference was within 15% except of a few points. The NSE and $R^2$ were also within a fair level. The Nakdong river flow of 2007 was simulated by using the EFDC model. The comparison with the measured data showed that the model reflected the actual values of low and high flow well. Also, it was confirmed that the acceleration and deceleration in the curved areas were appropriately simulated. The movement of dye injected at the upstream boundary was simulated. The result showed that the arrival time up to the estuary dam was computed to be about 65 days.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.11 no.1
• /
• pp.20-30
• /
• 2008

This study is to analyse the hydrological behavior of agricultural reservoir using SWAT model. For the upsteam watershed of Gongdo water level gauge station in Anseongcheon watershed, the streamflows at 2 reservoir (Gosam and Geumgwang) locations and Gongdo station were simulated with reservoir inclusion and exclusion. The daily water surface area and storage have been calculated considering the stage-storage curve function of the reservoir. Afterwards, the reservoir operation module in SWAT was modified from original module in SWAT for daily reservoir discharge simulated by water balance equation. Model validation results were Nash-Sutcliffe model efficiency coefficients value of 0.55, root mean square error value of 2.33 mm/day. On the other hand, the simulation results of two reservoir exclusion were Nash-Sutcliffe model efficiency coefficients value of 0.37, root mean square error value of 2.91 mm/day. The difference of Nash-Sutcliffe model efficiency coefficients between the simulation results of two reservoir inclusion and exclusion at Gongdo station was 0.18. This is caused by the storage and release operation of agricultural reservoirs for the runoff occurred at 2 reservoir watersheds.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2011.05a
• /
• pp.384-384
• /
• 2011

전 세계적으로 유역단위 수문해석 모형으로 많이 사용되는 SWAT모형은 유역 내 수문 모의시, DEM을 기반으로 유역 평균경사도를 이용하여 경사도-경사장 관계식 산정 경사장을 유역 내모든 수문학적 반응단위(HRU:Hydrologic Response Unit)에 동일하게 적용하는 문제점이 있다. 특히 SWAT모형은 미국지형 기반으로 개발되었기 때문에 유역 평균 경사도가 25% 미만일 경우 유역 평균 경사장 인자가 매우 작은 값으로 산정되어 모두 동일하게 적용된다. 이는 SWAT모형의 수문 및 지하수 함양량 산정 시 우리나라 유역 특성이 제대로 반영되지 않는 채 모의가 이루어질 수 있는 문제를 초래할 수 있다. 따라서 이러한 문제점을 해결하기 위하여 본 연구에서는 강원 발전연구원에서 전수 조사한 실측 경사장 자료가 있는 강원도 양구군 해안면 만대천 유역을 연구 대상지역으로 선정하였고 지하수 함양량을 시공간적으로 분석 할 수 있게 개발 된 SWAT HRU Mapping module에 실측 경사장 자료를 이용하여 2009년 실측 경사장(274m) 적용과 임의의 경사장(5m) 적용에 따른 유출량 및 지하수 함양량을 비교 분석 하였다. 임의의 경사장을 5m로 선정한 이유는 SWAT모형에서 소유역의 경사도가 25%이상일 경우 소유역의 경사장을 모두 0.05m로 산정하여 모두 동일하게 적용하기 때문에 실측 경사장(평균 274m)과 대조군으로 비교하기 위하여 임의의 경사장을 5m로 선정하였다. 2009년 해안면 만대천 유역의 총 강우량은 1341mm이며 실측경사장 및 임의 경사장 적용에 따른 유출특성별 유출량을 비교해보면 직접유출 결과는 두 가지 경사장 모두 큰 차이를 나타내지 않았지만 평균 경사장이 증가 할수록 중간유출은 줄어들고 기저유출이 급격하게 증가하는 것으로 분석되었다. 또한 2009년 만대천 유역의 월별지하수 함양량은 실측경사장과 임의의 경사장 적용에 따라 각각 591mm/yr(함양율 44%) 와 293mm/yr(함양율 22%) 로 나타났으며, 중간유출이 줄고 기저유출이 증가 할수록 지하수함양량 역시 증가 하는 것으로 나타났다. 따라서 SWAT모형에서 기저유출을 정확히 해석하고 지하수함양량을 제대로 산정하기 위해서는 실측 경사장에 적용에 따른 모형의 검 보정이 반드시 이루어져야 한다고 판단된다.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.34 no.2
• /
• pp.457-467
• /
• 2014

This study is to evaluate the snowmelt impact on dam inflow for the Chungju Dam watershed $6,642.0km^2$ using Terra MODIS (Moderate-Resolution Imaging Spectroradiometer) and Soil and Water Assessment Tool (SWAT). To determine the SWAT snowmelt parameter; snow cover depletion curve (SCDC) the snow depth distribution (SDD) using Terra MODIS was used, the snow depth was spatially interpolated using snowfall data of ground meteorological stations. For 10 sets (2000-2010) data during snowmelt period (November-April), the sno50cov parameter, that is, the 50% coverage at a fraction of SCDC which determines the shape of snow depletion process, showed the values of 0.4 to 0.7. The SWAT model was calibrated with average $R^2$ of 0.54 using the sno50cov of each year. The 10 years average streamflow during snowmelt period was 104.3 mm which covers 12.0% of the annual streamflow.