Evaluation of the Effect of Channel Geometry on Streamflow and Water

water

Article Evaluation of the Effect of Channel Geometry on Streamflow and Water Quality Modeling and Modification of Channel Geometry Module in SWAT: A Case Study of the Andong Dam Watershed

Jeongho Han 1 , Dongjun Lee 1, Seoro Lee 1, Se-Woong Chung 2 , Seong Joon Kim 3, Minji Park 4, Kyoung Jae Lim 1 and Jonggun Kim 5,* 1 Department of Regional Infrastructure Engineering, Kangwon National University, Gangwon-do, Chuncheon-si 200-701, Korea; [email protected] (J.H.); [email protected] (D.L.); [email protected] (S.L.); [email protected] (K.J.L.) 2 Department of Environmental Engineering, Chungbuk National University, Chungdae-ro 1, Chungbuk-do, Cheongju-si 28644, Korea; [email protected] 3 Department of Civil and Environmental engineering, Konkuk University, 1 Hwayang-dong, Gwangjin-gu, Seoul 143-701, Korea; [email protected] 4 Han River Environment Research Center, 42, Dumulmeori-gil, 68 beon-gil, Yangseo-myeon, Yangpyeong-gun, Gyeonggi-do 12585, Korea; [email protected] 5 Agriculture and Life Sciences Research Institute, Kangwon National University, Gangwon-do, Chuncheon-si 200-701, Korea * Correspondence: [email protected]; Tel.: +82-33-250-6468

Received: 25 March 2019; Accepted: 4 April 2019; Published: 6 April 2019

Abstract: The impact of the channel geometry on water quantity and quality simulation of the Soil and Water Assessment Tool (SWAT) was evaluated for the Andong Dam watershed. The new equations to determine the bankfull width of the channels and the bottom width of the floodplains were developed using aerial photographs, and its performance was compared with the current equations of SWAT. The new equations were more exact than the current equations since the current equations tended to overestimate the widths of the channel and floodplain. When compared with the observed data, the streamflow of the scenario 2 (S2, applying the new equations) showed lower deviation and higher accuracy than scenario 1 (S1, applying the current equations) because the peak flow of S2 captured the observed data better due to the impact of the change geometry. Moreover, the water quality results of S2 outperformed S1 regarding suspended solid, total nitrogen, and dissolved oxygen. This is attributed to the variables, such as flow travel time, which is directly related to the channel geometry. Additionally, SWAT was modified to consider the various channel cross-sectional shapes. The results of this study suggest that the channel geometry information for the water quantity and quality estimation should be carefully applied, which could improve the model performance regarding streamflow and water quality simulations.

Keywords: channel geometry; SWAT; hydrological modeling; regression equation; water quality modeling; Andong Dam Watershed; aerial photographs

1. Introduction Various studies have been conducted to predict the hydrology and water quality for river and watershed management in ungauged watersheds. As a result of these eﬀorts, various watershed-scale hydrological models have been developed, such as the Precipitation Runoﬀ Modeling System (PRMS), Hydrologic Simulation Program—Fortran (HSPF), Better Assessment Science Integrating Point and

Water 2019, 11, 718; doi:10.3390/w11040718 www.mdpi.com/journal/water Water 2019, 11, 718 2 of 18

Nonpoint Sources (BASINS), and Soil and Water Assessment Tool (SWAT) [1]. These hydrological models generally simulate the streamflow and water quality by calculating various parameters based on input data such as topographic and meteorological data [2,3]. In addition to the parameters calculated using the topographic and meteorological data (e.g., curve numbers, elevation, slope, and soil properties), the hydraulic characteristics of the channels in watersheds also play a crucial role in the water quantity and quality simulation as well as habitat evaluation since they are important influential factors [4,5]. For this reason, in most hydrological models, the channel shape parameters (e.g., channel width, depth, slope, length, and floodplain width) are required as input parameters [6]. Among these parameters, channel width and the water depth of channels are used to apply Manning’s equation for flow routing in channels in watersheds [7]. Due to the valid influence of channel width and depth, they are essential input parameters in hydrological modeling. The channel length and slope can be accurately estimated by digital elevation models (DEMs) and channel extraction algorithms; however, the channel width and depth are difficult to estimate accurately [6]. Thus, in most hydrological models, the channel width and depth are estimated using the regression equations for the upstream watershed area [8–11]. The current and typically used regression equations were developed only from large-scale studies over multiple regions in the United States. Ames [6] and Staley et al. [12] suggested that those equations may not be applicable to small-scale watershed studies. Thus, it is essential to evaluate the applicability of the current regression equations of SWAT in small or middle-scale watersheds with different characteristics. Moreover, although natural channels have various cross-sectional shapes, most hydrological models assume channel cross-sections as simplified forms such as rectangles, trapezoids, and triangles [13–15]. Similar to other commonly used hydrological models (e.g., PRMS, HSPF, and BASINS), SWAT, one of the most widely used models worldwide [16–20], also considers channel geometry through simplification with a predefined geometric shape, i.e., a compound channel composed of two trapezoids [21]. Thus, the current channel geometry estimation module in SWAT needs to be modified as it is limited in terms of taking into account multiple channel cross-section shapes. Therefore, the objectives of this study were to evaluate the current regression equations of SWAT for channel geometry estimation by comparing with the new regression equations for a middle-scale watershed in Korea, to modify SWAT for application to various channels with different cross-section shapes, and to investigate the impact of channel geometry on the model simulation results and water quantity and quality.

2. Materials and Methods

2.1. Overview of SWAT SWAT is a physically-based and long-term continuous time model. SWAT is used to predict the impact of management practices, land use and land cover change, and climate change on hydrology and water quality at a watershed scale [22]. Weather, soil properties, topography, and land management practices are the main inputs for simulating hydrologic and water quality processes in a watershed [23]. The model divides a watershed into a number of sub-watersheds according to the threshold values; the sub-watersheds are further divided into main channels and hydrologic response units (HRUs), a unique combination of land use, soil, and slope classes within each sub watershed. A HRU is the smallest computational unit in SWAT. Surface runoff, soil water content, baseflow, nutrient cycles, and erosion are simulated for each HRU, and then the HRUs are combined and calculated for a sub-watershed by a weighted value [7]. SWAT regards channels as compound channels that have a double trapezoidal shape as shown in Figure1[ 6–8]. The lower part of the trapezoid shape of the channels is the main channel, and the upper part is the floodplain. The floodplain dimension is considered when the volume of water in the channel exceeds the maximum amount of water that can be stored by the channel. Water 20192018,, 1110,, 718x FOR PEER REVIEW 33 of of 18

Figure 1. Channel dimensions in SWAT. Figure 1. Channel dimensions in SWAT. SWAT currently uses Equations (1) and (2) for estimating the width and depth of the main channel SWAT currently uses Equations (1) and (2) for estimating the width and depth of the main at the bankfull stage, which means that the channel is ﬁlled up to the top of the bank. Equation (3) channel at the bankfull stage, which means that the channel is filled up to the top of the bank. is used to estimate the bottom width of the ﬂoodplain. These equations were developed using the Equation (3) is used to estimate the bottom width of the floodplain. These equations were developed methodology based on Muttiah et al. [8], Allen et al. [9], and Leopold and Maddock [10]. using the methodology based on Muttiah et al. [8], Allen et al. [9], and Leopold and Maddock [10].

0.6. W𝑊bnk f ull ==1.29∙1.29 A 𝐴,, (1) ·

0.4. D𝐷bnk f ull ==0.13∙0.13 A 𝐴,, (2) · Wbtm, f ld = 5 Wbnk f ul, (3) 𝑊, =5∙𝑊· , (3) where Wbnk f ull and Dbnk f ull are the bankfull stage width and depth, respectively, of the main channel (m); where 𝑊 and 𝐷 are 2the bankfull stage width and depth, respectively, of the main A is the upstream drainage area (km ); and Wbtm, f ld is the bottom width of the ﬂoodplain (m). Moreover, channel (m); 𝐴 is the upstream drainage area (km2); and 𝑊 is the bottom width of the SWAT assumes that the slopes of the main channel and ﬂoodplain, have a 2:1 and 4:1 run to the rise, floodplain (m). Moreover, SWAT assumes that the slopes of the main channel and floodplain have a respectively. Based on these assumptions, SWAT computes the bottom width of the main channel 2:1 and 4:1 run to the rise, respectively. Based on these assumptions, SWAT computes the bottom using the bankfull width and depth using Equation (4). width of the main channel using the bankfull width and depth using Equation (4).

𝑊Wbtm ==𝑊Wbnk f ull −2∙𝑧2 zch Dbnk∙𝐷 f ull, (4) − · · , (4) 𝑊 𝑧 where Wbtmis is the the bottom bottom width width of of the the main main channel channel (m) (m) and andzch is the is inverse the inverse of the of channel the channel side slope. side Then,slope. SWATThen, calculatesSWAT calculates other variables other variables including including the cross-section the cross-section area of flow, area the of wettedflow, the perimeter, wetted andperimeter, the hydraulic and the radius hydraulic using radius channel using geometry channel data geometry to calculate data theto calculate flow velocity the flow and velocity simulate and the flowsimulate and the water flow quality and water over timequality (Equations over time (5)–(7)). (Equations (5)–(7)). 𝑃 =𝑊 +2∙𝑑𝑒𝑝𝑡ℎ∙q1+𝑧, (5) 2 Pch = Wbtm + 2 depth 1 + z , (5) · 𝐴· ch 𝑅 = , (6) 𝑃 /Ach / Rch 𝑅= ∙𝑠𝑙𝑝, (6) 𝑣 = P , (7) ch𝑛 2/3 1/2 where 𝑃 is the wetted perimeter of flow (m),R 𝑊slp is the bottom width of the channel (m), 𝑧 is v = ch · ch , (7) the inverse of the channel slope, depth is thec depth ofn water in the channel (m), 𝑅 is the hydraulic 2 radius (m), 𝐴 is the cross-sectional area of flow (m ), 𝑃 is the wetted perimeter of flow (m), 𝑣 where Pch is the wetted perimeter of flow (m), Wbtm is the bottom width of the channel (m), zch is is the flow velocity (m/s), 𝑠𝑙𝑝 is the slope along the channel length (m/m), and n is Manning’s the inverse of the channel slope, depth is the depth of water in the channel (m), Rch is the hydraulic coefficient for the channel. 2 radius (m), Ach is the cross-sectional area of flow (m ), Pch is the wetted perimeter of flow (m), vc is the SWAT uses the five different stream power equations for sediment routing in stream channels; flow velocity (m/s), slpch is the slope along the channel length (m/m), and n is Manning’s coefficient for simplified Bagnold equation, simplified Bagnold model, Kodatie model, Molinas and Wu model, and the channel. Yang sand and gravel model [7]. All these models predict the maximum concentration of bed load SWAT uses the five different stream power equations for sediment routing in stream channels; using a non-linear function of peak velocity, where the channel geometry-related variables such as simplified Bagnold equation, simplified Bagnold model, Kodatie model, Molinas and Wu model, flow velocity, bottom width of the channel, and flow depth are used. and Yang sand and gravel model [7]. All these models predict the maximum concentration of bed load The outflow of water quality related to nitrogen and phosphorus from a channel, such as organic using a non-linear function of peak velocity, where the channel geometry-related variables such as nitrogen, ammonia, nitrite, nitrate, and organic P, is estimated by subtracting the amount of change flow velocity, bottom width of the channel, and flow depth are used. in each item’s concentration from the total amount in the channel and adjusting travel time in the The outflow of water quality related to nitrogen and phosphorus from a channel, such as organic reach segment that is decided by the flow rate and the volume of the channel. Similar to the nitrogen nitrogen, ammonia, nitrite, nitrate, and organic P, is estimated by subtracting the amount of change in and phosphorus simulation processes in the channel, the dissolved oxygen (DO) is estimated by each item’s concentration from the total amount in the channel and adjusting travel time in the reach subtracting the variation of DO from the total amount of DO in the channel considering the flow travel time.

Water 2019, 11, 718 4 of 18 segment that is decided by the flow rate and the volume of the channel. Similar to the nitrogen and phosphorusWater 2018, 10, simulationx FOR PEER REVIEW processes in the channel, the dissolved oxygen (DO) is estimated by subtracting4 of 18 the variation of DO from the total amount of DO in the channel considering the flow travel time. 2.2. Development of New Regression Equations for Estimating Channel Geometries 2.2. Development of New Regression Equations for Estimating Channel Geometries 2.2.1. Study Area 2.2.1. Study Area This study was carried out for the Andong Dam watershed located upstream of the Nakdong This study was carried out for the Andong Dam watershed located upstream of the Nakdong River, which is one of the four major rivers in South Korea. Figure 2 shows the location and other River, which is one of the four major rivers in South Korea. Figure2 shows the location and other information for the study area. The dam was built in 1976, and it has a total watershed area of 1584 information for the study area. The dam was built in 1976, and it has a total watershed area of 1584 km2, km2, which corresponds to a middle-scale watershed according to the standard of watershed which corresponds to a middle-scale watershed according to the standard of watershed classification classification by size [24]. The recent 10-year average temperature is 10.25 °C, and average annual by size [24]. The recent 10-year average temperature is 10.25 C, and average annual precipitation is precipitation is 1258 mm. Most of the watershed consists of◦ forest areas, showing that the well- 1258 mm. Most of the watershed consists of forest areas, showing that the well-formed, narrow river formed, narrow river valley is naturally distributed in both the main channels and tributaries from valley is naturally distributed in both the main channels and tributaries from upstream to downstream. upstream to downstream.

Figure 2. Study area: Location of the Andong Dam watershed in South Korea. 2.2.2. Development of New Regression Equations to Reflect Real Channel Geometry Information 2.2.2. Development of New Regression Equations to Reflect Real Channel Geometry Information The regression equations to determine the bankfull width of the main channels and floodplain bottomThe width regression in middle-scale equations watersheds to determine were the developed bankfull width as follows. of the First, main the channels sub-watersheds and floodplain of the studybottom area width were in middle-scale delineated, and watersheds the areas were of each develo sub-watershedped as follows. including First, th thee sub-watersheds upstream area of were the calculated.study area were Then, delineated, using aerial and photographs the areas of providedeach sub-watershed by Google including Earth Pro the and upstream the sub-watershed area were shapecalculated. file formed Then, using by SWAT, aerial the ph bankfullotographs stage provided width by of theGoogle main Earth channels Pro and and the the sub-watershed bottom width ofshape the file floodplain formed by were SWAT, measured the bankfull at a constant stage widt distanceh of the (i.e., main approximately channels and 500the bottom m intervals) width for of eachthe floodplain sub-watershed. were measured However, at the a constant distance wasdistance flexibly (i.e., changed approximately considering 500 m the intervals) condition for of each the sub-watershedssub-watershed. suchHowever, as the the sub-watershed distance was area, flexibly hydraulic changed structures, considering and topographic the condition characteristics. of the sub- watershedsAccording such to as Cinotto the sub-watershed [25], a bankfull area, stage hydraulic occurs structures, within a and range topographic of 1~2 years, characteristics. and there are primaryAccording indicators to Cinotto identifying [25], aa bankfull bankfull width stage stage:occurs topography within a range (a change of 1~2 inyears, the slope and there from theare channelprimary bankindicators to a flatidentifying valley bottom), a bankfull vegetation width stage: (a change topography in vegetation (a change species in the or slope its presence), from the andchannel sediment bank to texture a flat (thevalley size bottom), distribution vegetation of surface (a change sediments). in vegetation However, spec theseies or indicatorsits presence), are and not fullysediment applicable texture when (the size using distribu aerialtion photographs of surface without sediments). a field Howeve investigation.r, these Thus,indicators as an are alternative, not fully theapplicable distance when with using the widest aerial waterphotographs surface without through a various field investigation. aerial photographs Thus, as was an assumedalternative, as the distance with the widest water surface through various aerial photographs was assumed as the bankfull stage because the bankfull stage means that the channel is filled with water up to the top of the bank. To determine the widest water surface, the aerial photographs provided by Google Earth Pro at different shooting times from 2012 to 2017 were compared for the entire study area because the

Water 2019, 11, 718 5 of 18 bankfull stage because the bankfull stage means that the channel is filled with water up to the top of the bank. To determine the widest water surface, the aerial photographs provided by Google Earth Pro atWater different 2018, 10 shooting, x FOR PEER times REVIEW from 2012 to 2017 were compared for the entire study area because5 of 18 the shooting time was different depending on the region. To guarantee the reliability of the data, it was confirmedshootingthat time there was haddifferent been depending no extreme on flood the thatregion. could To causeguarantee changes the reliabil in the morphologicality of the data, it changes was ofconfirmed the study areathat duringthere had this been period. no extreme Figure3 floodshows that an could example cause of thischanges process. in the In morphological this process, no measurementschanges of the were study made area in during sub-watersheds this period. where Figure hydraulic 3 shows structures an example aff ectingof this theprocess. surface In areathis of theprocess, water, suchno measurements as weirs, were were located. made This in sub-wate is becausersheds the new where regression hydraulic equations structures were affecting only for the the channelsurface widths area of of the natural water, rivers, such as not weirs, those were regulated located. or This affected is because by human the new activities. regression equations were only for the channel widths of natural rivers, not those regulated or affected by human activities.

FigureFigure 3. 3.An An example example of of a comparisona comparison of of the the top top width widt ofh of the the main main channel channel according according to shootingto shooting time (fromtime Google (from Google Earth Pro). Earth Pro).

TheThe average average widths widths were were calculated calculated from from the measuredthe measured bankfull bankfull stage stage widths widths of the of main the channels main andchannels the measured and the bottommeasured width bottom of thewidth floodplain of the floo fordplain each for sub-watershed, each sub-watershed, and these and werethese usedwere as dataused for as developing data for developing the new the regression new regression equations. equations. Table 1Table summarizes 1 summarizes the channel the channel geometry geometry data measureddata measured by Google by Google Earth ProEarth in Pro the in sub-watersheds. the sub-watersheds. In this study, the new regression equation for estimating the bankfull width of the main channels was developedTable as a 1.powerMeasured function channel of the and upstream floodplain drainage geometry area data (the for same the study as the area. current equation in SWAT). To estimate the bottom width of the floodplain, the current SWAT model exploits the No. Area (km2) Ch_w (m) Fld_w (m) No. Area (km2) Ch_w (m) Fld_w (m) regression equation expressed as a function of the bankfull width of the main channels, while this 1 54.1 8.1 20.0 18 545.2 34.2 68.4 study developed a new power regression equation as a function of the upstream drainage area. 2 41.5 8.3 17.9 19 104.4 18.3 37.1 3The optimal 134.8 coefficients 14.4 for each regression 30.2 equation20 were 141.0 derived by 22.0 using CurveExpert 37.4 Professional4 (v.2.2.0). 59.2 CurveExpert 7.8 is a cross-platfo 18.8 rm22 program 661.4 that provides the 43.8 curve fitting 66.9 results and5 data analysis 201.2 at the same 15.7 time, providing 39.9 various25 linear 169.9 regression models 17.7 and nonlinear 32.6 regression6 models. 49.3 CurveExpert 8.5 basically 21.1provides mo26re than 702.060 regression models 45.6 and also 75.2 has the advantage8 of 42.9allowing users to 8.2 enter their 18.8desired regression27 models 754.7 [26]. 39.2 82.6 9 260.8 21.5 43.6 28 90.8 9.3 24.6 11 50.4 12.4 19.8 30 197.6 19.4 35.7 Table 1. Measured channel and floodplain geometry data for the study area. 12 28.6 9.3 21.4 33 41.0 6.1 12.8 13No. Area 362.1 (km2) Ch_w 23.8 (m) Fld_w 54.1 (m) 34No. 85.2Area (km2) Ch_w 15.4 (m) Fld_w 31.2 (m) 141 35.954.1 7.18.1 21.820.0 3518 39.3545.2 9.634.2 23.368.4 15 404.7 26.4 56.2 40 1187.5 62.5 89.0 2 41.5 8.3 17.9 19 104.4 18.3 37.1 16 57.0 13.2 27.6 3 134.8 14.4 30.2 20 141.0 22.0 37.4 Note: No.: sub-watershed number. Ch_w: bankfull width of the main channel. Fld_w: bottom width of Floodplain. 4 59.2 7.8 18.8 22 661.4 43.8 66.9 5 201.2 15.7 39.9 25 169.9 17.7 32.6 In this study, the new regression equation for estimating the bankfull width of the main channels 6 49.3 8.5 21.1 26 702.0 45.6 75.2 was developed as a power function of the upstream drainage area (the same as the current equation in 8 42.9 8.2 18.8 27 754.7 39.2 82.6 SWAT). To estimate the bottom width of the floodplain, the current SWAT model exploits the regression 9 260.8 21.5 43.6 28 90.8 9.3 24.6 equation expressed as a function of the bankfull width of the main channels, while this study developed 11 50.4 12.4 19.8 30 197.6 19.4 35.7 a new power regression equation as a function of the upstream drainage area. 12 28.6 9.3 21.4 33 41.0 6.1 12.8 The optimal coefficients for each regression equation were derived by using CurveExpert 13 362.1 23.8 54.1 34 85.2 15.4 31.2 Professional14 (v.2.2.0).35.9 CurveExpert 7.1 is a cross-platform21.8 program35 that39.3 provides the9.6 curve fitting23.3 results and data15 analysis404.7 at the same time,26.4 providing56.2 various linear40 regression1187.5 models and62.5 nonlinear regression89.0 16 57.0 13.2 27.6 Note: No.: sub-watershed number. Ch_w: bankfull width of the main channel. Fld_w: bottom width of Floodplain.

Water 2019, 11, 718 6 of 18 models. CurveExpert basically provides more than 60 regression models and also has the advantage of Water 2018, 10, x FOR PEER REVIEW 6 of 18 allowing users to enter their desired regression models [26]. 2.3. Modification of Channel Geometry Estimation Module in SWAT 2.3. Modification of Channel Geometry Estimation Module in SWAT Due to meteorological and geographical characteristics, natural rivers have various cross-section Due to meteorological and geographical characteristics, natural rivers have various cross-section shapes as shown in Figure 4. For example, as shown in Figures 4a–d, there are many different shapes shapes as shown in Figure4. For example, as shown in Figure4a–d, there are many di fferent shapes of cross-sections in natural rivers with a symmetric floodplain, including a floodplain biased to one of cross-sections in natural rivers with a symmetric floodplain, including a floodplain biased to one side as well as no floodplain. Additionally, the slope of both sides of the stream is commonly not side as well as no floodplain. Additionally, the slope of both sides of the stream is commonly not symmetrical. Therefore, this study adjusted the SWAT 2012 code (v.664) written in FORTRAN symmetrical. Therefore, this study adjusted the SWAT 2012 code (v.664) written in FORTRAN language language to allow users to apply various cross-section shapes of rivers. to allow users to apply various cross-section shapes of rivers.

(a) (b) (c) (d)

FigureFigure 4. Various 4. Various channel channel cross-section cross-section shapes shapes (a–d )in in natural natural rivers. rivers.

2.4.2.4. Input Input Data Data for for SWAT SWAT TheThe SWAT SWAT input input data data were were constructed constructed to to simulate simulate the the Andong Andong Dam Dam watershed watershed to to analyze analyze the the effeffectect of of channel channel geometry geometry data data on on the the hydrological hydrological and and water water quality quality simulation. simulation. The modelThe model was run was usingrun using two scenarios. two scenarios. Scenario Scenario 1 (S1) used 1 (S1) current used regressions current regressions in SWAT for in channelSWAT for geometry channel estimation, geometry andestimation, Scenario and 2 (S2) Scenario used the 2 regression(S2) used equationsthe regressi developedon equations in this developed study. These in this two study. scenarios These used two thescenarios same input used data the same except input for the data channel except geometryfor the channel data, i.e.,geometry bankfull data, width i.e., ofbankfull the main width channel, of the andmain the channel, bottom widthand the of bottom the floodplain. width of the floodplain. TheThe elevation elevation data data were were rasterized rasterized as as a a 30-m 30-m resolution resolution digital digital elevation elevation model model (DEM) (DEM) from from 1:50001:5000 vector vector maps maps obtained obtained from from the the KoreaKorea NationalNational GeographyGeography Institute. The The soil soil data data pertinent pertinent to totexture, texture, depth, depth, and and chemical chemical attributes, attributes, whic whichh were suppliedsupplied byby thethe KoreaKorea Rural Rural Development Development Administration,Administration, were were rasterized rasterized fromfrom 1:25,000 vector vector maps. maps. The The land land use use and and land land cover cover (LULC) (LULC) map mapfor 24 for classes 24 classes for 2010 for 2010was obtained was obtained from fromthe Korea the Korea Ministry Ministry of Environment. of Environment. The dominant The dominant land use landof the use Andong of the AndongDam watershed Dam watershed is forest is(81%), forest followed (81%), followed by cultivated by cultivated croplands croplands (6.5%), including (6.5%), includingboth rice bothpaddy rice fields paddy and fields upland and crops. upland Eight crops. years Eight (2008–2015) years (2008–2015) of daily weather of daily weatherdata were data collected were collectedfrom three from weather three weather stations stations of the ofKorea the Korea Meteor Meteorologicalological Administration Administration (Andong, (Andong, Bonghwa, Bonghwa, and andTaebaek) Taebaek) containing containing precipitation, precipitation, maximum maximum and and minimum minimum temperature, relativerelative humidity,humidity, wind wind speed,speed, and and solar solar radiation radiation (Figure (Figure2). 2). 2.5. Streamflow and Water Quality Simulation Using SWAT 2.5. Streamflow and Water Quality Simulation Using SWAT The Andong Dam watershed was divided into 34 sub-watersheds and 1799 HRUs based on the The Andong Dam watershed was divided into 34 sub-watersheds and 1799 HRUs based on the DEM, LULC map, and soil map. Among the two water-routing options (variable storage routing DEM, LULC map, and soil map. Among the two water-routing options (variable storage routing method and Muskingum routing method), the Muskingum routing method was applied in this study. method and Muskingum routing method), the Muskingum routing method was applied in this The first 3 years (2008–2010) over the whole period were set as the warm-up period. The calibration for study. The first 3 years (2008–2010) over the whole period were set as the warm-up period. The streamflow and water quality was carried out from 2011 to 2013 for 3 years, and the validation was calibration for streamflow and water quality was carried out from 2011 to 2013 for 3 years, and the performed from 2014 to 2015. This study used the daily dam inflow data of Andong Dam, monitored by validation was performed from 2014 to 2015. This study used the daily dam inflow data of Andong the Korea Water Resources Corporation (lower yellow point in Figure2). Additionally, the streamflow Dam, monitored by the Korea Water Resources Corporation (lower yellow point in Figure 2). data of Dosan Station were used to improve the accuracy of the model simulation results; Dosan Station Additionally, the streamflow data of Dosan Station were used to improve the accuracy of the model is at the outlet of the 25th sub-watershed as shown in Figure2. However, the calibration and validation simulation results; Dosan Station is at the outlet of the 25th sub-watershed as shown in Figure 2. for water quality was carried out for only one observation station, Nakbon B Station which is located However, the calibration and validation for water quality was carried out for only one observation nearby Dosan Station, due to the absence of available observed water quality data for Andong Dam station, Nakbon B Station which is located nearby Dosan Station, due to the absence of available Station. This study analyzed sediment, total nitrogen (TN), and DO among various water quality items. observed water quality data for Andong Dam Station. This study analyzed sediment, total nitrogen (TN), and DO among various water quality items.

Water 2019, 11, 718 7 of 18

2.6. Model Calibration and its Evaluation To guarantee the objectivity of the calibration, the SWAT Calibration and Uncertainty Program (SWAT-CUP) [27] was used instead of manual calibration. SWAT-CUP is one of the most popular auto-calibration programs as it can produce parameter estimation by uncertainty analysis and simultaneously reduce labor on the part of the user [23,27,28]. SWAT-CUP provides five calibration procedures for the SWAT calibration: Sequential Uncertainty Fitting algorithm version 2 (SUFI-2), Generalized Likelihood Uncertainty Estimation (GLUE), Parameter Solution (ParaSol), Markov chain Monte Carlo (MCMC), and Particle Swarm Optimization (PSO). In this study, calibration was performed using SUFI-2, which has been used to optimize the SWAT parameters in many studies [23,29,30]. For the streamflow and water quality calibration, the sensitivity analysis for the selected parameters based on the previous studies (16 parameters for the streamflow calibration and 22 parameters for the water quality calibration) was initially carried out [30]. Through the sensitivity analysis, 12 parameters for streamflow and 15 parameters for water quality were selected (Tables2 and3), and these selected parameters were identically applied to S1 and S2; the same ranges were set for each parameter’s value. In the calibration process, the channel-related parameters were excluded to prevent the interpretation of the effect of channel geometry information on the model simulations.

Table 2. Calibrated parameters for streamﬂow.

Fitted Value Variation Parameter Definition Method Dosan Andong Dam S1 S2 S1 S2 ALPHA_BF Baseflow alpha factor Replace 0.61 0.61 0.83 0.83 CH_K2 Effective hydraulic conductivity in main channel alluvium Replace 13.34 13.47 0.77 0.78 CN2 SCS runoff curve number Multiply 0.82 0.83 0.75 0.76 ESCO Soil evaporation compensation factor Replace 0.45 0.46 0.09 0.09 GW_DELAY Groundwater delay time Replace 113 114 269 271 Threshold depth of water in the shallow aquifer required for GWQMN Replace 2451 2476 822 830 return flow to occur LAT_TTIME Lateral flow travel time Replace 3.16 3.19 1.45 1.47 Calibration coefficient used to control impact of the storage MSK_CO1 Replace 4.03 4.03 4.03 4.03 time constant for normal flow Calibration coefficient used to control impact of the storage MSK_CO2 Replace 3.94 3.94 3.94 3.94 time constant from low flow Weighting factor controlling relative importance of inflow MSK_X Replace 0.01 0.01 0.01 0.01 rate and outflow rate in determining water storage in a reach SOL_AWC Available water capacity of the soil layer Multiply 0.83 0.84 1.24 1.25 SURLAG Surface runoff lag time Replace 6.65 6.72 6.65 6.72

Table 3. Calibrated parameters for water quality of Nakbon B Station.

Variation VALUES Parameter Definition Method S1 S2 CDN Denitrification exponential rate coefficient Replace 2.63 2.51 N_UPDIS Nitrogen uptake distribution parameter Replace 16.70 38.10 PRF_BSN Peak rate adjustment factor for sediment routing in the main channel Replace 0.46 0.21 ADJ_PKR Peak rate adjustment factor for sediment routing in tributary Replace 1.13 0.62 RSDCO Residue decomposition coefficient Replace 0.06 0.07 SOL_NO3 Initial NO3 concentration in the soil layer Replace 7.90 27.50 LAT_ORGN Organic N in the baseflow Replace 0.05 0.45 HLIFE_NGW Half-life of nitrate in the shallow aquifer Replace 56.66 77.80 SLSUBBSN Average slope length Multiply 1.23 0.86 HRU_SLP Average slope steepness Multiply 1.03 1.24 USLE_P USLE equation support practice Replace 0.35 0.84 BIOMIX Biological mixing efficiency Replace 0.14 0.16 USLE_C Min value of USLE C factor applicable to the land cover/plant Multiply 1.19 1.03 USLE_K USLE equation soil erodibility (K) factor Multiply 1.45 1.30 AI1 Fraction of algal biomass that is nitrogen Replace 0.07 0.08 Water 2019, 11, 718 8 of 18

To evaluate the model performance, this study used a graphical technique. According to Legates and McCabe [31], the graphical techniques are essential to appropriate the model evaluation because they provide a visual comparison of the simulated and measured constituent data and a first overview of model performance. This study conducted the model evaluation using a hydrograph. A hydrograph is beneficial to identify the model bias and differences in timing, the magnitude of peak flows, as well as the shape of the recession curves [32,33]. In addition to the graphical model evaluation, various statistical techniques are concurrently used, and the following statistic criteria were selected: coefficient of determination (R2), Nash-Sutcliffe efficiency (NSE), root mean square error (RMSE)-observations standard deviation ratio (RSR), and percent bias (PBIAS). R2 is the square of the coefficient of correlation between the simulated and observed values and it ranges from 0 to 1. When R2 is 1, it means that the simulated data are perfect. NSE indicates the fit of the data to a linear 1:1 measured versus the simulated best-fit line [34], and it ranges from minus infinity (poor model) to 1.0 (perfect model). A limitation of NSE is the fact that larger values in a time series are significantly overestimated while lower ones are neglected because the differences between the observed and simulated values are computed as squared values [31]. The RMSE summarizes the average error between the observed and predicted variables using the same units as those variables, and RSR is estimated as the ratio of the RMSE and standard deviation of the measured data. The optimal value of RSR is 0, which signifies the perfect model simulation [33]. PBIAS measures the average tendency of the simulated data to be larger or smaller than their corresponding measured values. PBIAS is the deviation of values being evaluated and is expressed as a percentage. The optimal value of PBIAS is 0, and positive values mean model underestimation bias whereas negative values indicate model overestimation [35]. Table4 is the classification of the model efficiencies suggested by Moriasi et al. [33]. However, this study applied the standards of TN to DO due to the absence of reference for DO.

Table 4. The SWAT performance evaluation classiﬁcation.

PBIAS Performance Rating R2, NSE RSR Streamﬂow Sediment N, P Very good 0.75–1.00 0.00–0.50 – 10 – 15 – 25 ± ± ± Good 0.50–0.75 0.50–0.60 10 – 15 15 – 30 25 – 40 ± ± ± ± ± ± Satisfactory 0.25–0.50 0.60–0.70 15 – 25 30 – 55 40 – 70 ± ± ± ± ± ± Unsatisfactory 0.00–0.25 0.70– 25 – 55 – 70 – ± ± ±

3. Results and Discussion

3.1. Comparison between Current and New Regression Equations for Applicability Assessment of the Middle-Scale Watershed The average channel bankfull width and the average ﬂoodplain bottom width data for each sub-watershed, which were calculated from the channel geometry data measured by Google Earth Pro, were applied to the CurveExpert program to develop the regression equations. The new regression equations were expressed as a function of the upstream watershed area. Equations (8) and (9) are the new regression equations for the bankfull stage width of the main channel, and for the ﬂood width for the Andong Dam basin, respectively.

W = 1.0768 A0.5519 (8) bnk f ull · W = 3.4986 A0.4611 (9) btm, f ld · where Wbnk f ull is the bankfull stage width of the main channel (m); A is the upstream drainage area 2 (km ); and Wbtm, f ld is the bottom width of the ﬂoodplain (m). Figure5 shows the comparisons of the average measured bankfull width of the main channels and the ﬂoodplain bottom width of each Water 2018, 10, x FOR PEER REVIEW 9 of 18

WateraverageWater2019 2018 ,measured,11 10,, 718 x FOR PEER bankfull REVIEW width of the main channels and the floodplain bottom width of each9 9 ofsub- of 18 18 watershed constructed through Google Earth Pro, the regression equations of the current SWAT model,average and measured the new bankfull regression width equations. of the main Figure channe 6 showsls and the the bar floodplain graphs bottomthat indicate width the of eachresults sub- of sub-watershed constructed through Google Earth Pro, the regression equations of the current SWAT thewatershed calculated constructed main channel through width Google and floodplaiEarth Pro,n bottom the regression width for equations each sub-watershed of the current using SWAT the model,currentmodel, andSWATand thethe regression new regression equations equations. equations. and the Figure Figurenew regressi 66 shows showson the theequations. bar bar graphs graphs It can that that be indicate indicateseen in the Figure the results results 5 that of ofSWATthe the calculated calculatedtended mainto mainslightly channel channel overestimate width width and and thefloodplai floodplainbankfulln bottom width bottom widthof widththe formain foreach channels. each sub-watershed sub-watershed In particular, using using thethe thebottomcurrent current widthSWAT SWAT ofregression the regression floodplain equations equations was and significantly and the thenew new regressi overestimated regressionon equations. equations. when It comparedcan It be can seen be to seenin the Figure in measured Figure 5 that5 thatdata.SWAT SWAT On tended the tended contrary, to slightly to slightlythe new overestimate overestimateregression the equations thebankfull bankfull showed width width aof high the of degree themain main channels. of accuracy channels. In forparticular, In both particular, results the thecomparedbottom bottom width with width of the ofthe current the floodplain floodplain regression was was significantly equations significantly in overestimated SWAT. overestimated According when when to compared comparedthe analysis to to ofthe the the measured measured channel data.geometrydata. OnOn thethe estimation contrary, results thethe newnew of regression S1 (current equationsequations regression showed equations) a high an degree d S2 (new of accuracy regression for for both bothequations), results results comparedthecompared channel withwith bankfull thethe currentcurrent width and regressionregression the floodplain equations bottom in SWAT. width AccordingAccording of S1 were to estimated the analysis to be of 35% the andchannel 73% geometrylarger,geometry respectively, estimationestimation than resultsresults those ofof S1S1 of (current(current S2, for regressionall of the sub-watersheds. equations) an andd S2 These (new resultsregression regression reflect equations), equations), that the thecurrentthe channelchannel regression bankfullbankfull equations widthwidth andand in thethe SWAT floodplainfloodplain are inapprop bottombottom riate width for of middle-scaleS1 were estimated watersheds to be be 35% 35% such and and as 73% 73%the larger,Andonglarger, respectively, respectively, Dam because than than of those overestimation those of S2,of S2, for allfor as of allAmes the of sub-watersheds. theet al. sub-watersheds. [6] stated. TheseSpecifically These results ,results the reflect current reflect that theregression that current the regressionequationcurrent regression of equations SWAT forequations in estimating SWAT arein SWAT inappropriatefloodplain are inappropwidths for middle-scale wasriate found for middle-scale to watersheds be considerably suchwatersheds as inaccurate. the Andong such as Dam the becauseAndong of Dam overestimation because of overestimation as Ames et al. as [6 Ames] stated. et al. Specifically, [6] stated. theSpecifically current, regressionthe current equation regression of SWATequation for of estimating SWAT for floodplain estimating widths floodplain was foundwidths to was be considerablyfound to be considerably inaccurate. inaccurate.

(a) (b)

Figure 5. Comparisons between(a) measured data and the calculated results from the(b current) equations and the new equations. (a) Bankfull stage width of the main channel. (b) Bottom width of the Figurefloodplain.Figure 5.5. ComparisonsComparisons betweenbetween measured data and the calculatedcalculated results from the current equations andand the the new new equations. equations. (a) Bankfull(a) Bankfull stage stage width width of the of main the channel. main channel. (b) Bottom (b) widthBottom of thewidth ﬂoodplain. of the floodplain.

(a)

Figure(a 6.) Cont.

Water 2018, 10, x FOR PEER REVIEW 10 of 18

Water 2019, 11, 718 10 of 18 Water 2018, 10, x FOR PEER REVIEW (b) 10 of 18

Figure 6. Results of channel geometry estimation for each sub-watershed. (a) Bankfull width of the main channel for each sub-watershed in the Andong Dam watershed. (b) Floodplain bottom width for each sub-watershed in the Andong Dam watershed.

3.2. Model Performance Evaluation and Analysis of the Effect of Channel Geometry Data on Model Simulation

3.2.1. Comparison of Flow Velocity Estimation Results from Current and New Regression Equations. Figure 7 indicates the results of the flow velocity calculations in the #25 (Dosan Station) and #34 (Andong Dam Station) sub-watersheds where the streamflow and water quality calibration were performed. As shown in Figures 7a,b, the flow velocity of S1 (current regression equations) was smaller than that of S2 (new regression equations)( bat) both stations. When peak flow occurred, the gap of theFigure flow 6.6. velocity Results ofbetween channelchannel S1 geometrygeometry and S2 tended estimationestimation to increase. for eacheach sub-watershed.sub-watershed.The average percentage ( a) Bankfull of width the difference of thethe in themain flow channelchannel velocity forfor eachofeach S1 sub-watershed andsub-watershe S2 for 5 yearsd in in the the was Andong Andong 48% Dam and Dam watershed.50%, watershed. respectively. (b )(b Floodplain) Floodplain bottom bottom width width for eachforThis each sub-watershed result sub-watershed suggests in the thatin Andongthe the Andong impact Dam Dam watershed.of channel watershed. geometry caused such a difference in the flow velocity. According to Neitsch et al. [7], the flow velocity is proportional to the hydraulic radius in 3.2.SWAT Model (Equation Performance (7)). Evaluation The hydraulic and Analysis radius ofis thethe Evalueffect of of Channel the cross-sectional Geometry Data area on Modeldivided Simulation by the 3.2.wetted Model perimeter Performance (Equation Evaluation (6)), andwhich Analysis is proportion of the Effectal to of the Channel channel Geometry width (EquationData on Model (5)). Due to 3.2.1. Comparison of Flow Velocity Estimation Results from Current and New Regression Equations. Simulationthese relationships between the channel width, wetted perimeter, hydraulic radius, and flow velocity, the Figureflow velocity7 indicates decreases the results when ofthe the channel flow velocitywidth increases calculations under in the the cond #25ition (Dosan of an Station) identical and #343.2.1.cross-sectional (Andong Comparison Dam area of Station) ofFlow flow Velocity sub-watersheds in the channel, Estimation wherewhic Resultsh themeans streamflow from that Current the amount and and water Newof water quality Regression flowing calibration into werethe performed.Equations.channel is Asthe shown samein (combining Figure7a,b, surface the flow runoff, velocity lateral of S1 (currentflow, an regressiond baseflow). equations) Therefore, was smallerwhen thanconsidering that of S2 the (new flow regression velocity esti equations)mation procedure, at both stations. the results When where peak S1 flow showed occurred, a lower the value gap of for the Figure 7 indicates the results of the flow velocity calculations in the #25 (Dosan Station) and #34 flowflow velocity velocity between than S2 were S1 and due S2 to tended the fact to that increase. the channel The average width of percentage S1 was more of extensive the difference than inthat the (Andong Dam Station) sub-watersheds where the streamflow and water quality calibration were flowof S2. velocity of S1 and S2 for 5 years was 48% and 50%, respectively. performed. As shown in Figures 7a,b, the flow velocity of S1 (current regression equations) was smaller than that of S2 (new regression equations) at both stations. When peak flow occurred, the gap of the flow velocity between S1 and S2 tended to increase. The average percentage of the difference in the flow velocity of S1 and S2 for 5 years was 48% and 50%, respectively. This result suggests that the impact of channel geometry caused such a difference in the flow velocity. According to Neitsch et al. [7], the flow velocity is proportional to the hydraulic radius in SWAT (Equation (7)). The hydraulic radius is the value of the cross-sectional area divided by the wetted perimeter (Equation (6)), which is proportional to the channel width (Equation (5)). Due to these relationships between the channel width, wetted perimeter, hydraulic radius, and flow velocity,

the flow velocity decreases when the channel width increases under the condition of an identical (a) (b) cross-sectional area of flow in the channel, which means that the amount of water flowing into the channelFigure is 7.theFlow same velocity (combining estimation surface results forrunoff, Dosan lateral Station flow, and Andongand baseflow). Dam Station. Therefore, (a) Dosan when considering Station (No. the 25flow sub-watershed). velocity estimation (b) Andong procedure, Dam Station the (No.results 34 sub-watershed).where S1 showed a lower value for flow velocity than S2 were due to the fact that the channel width of S1 was more extensive than that of S2.This result suggests that the impact of channel geometry caused such a difference in the flow velocity. According to Neitsch et al. [7], the flow velocity is proportional to the hydraulic radius in SWAT (Equation (7)). The hydraulic radius is the value of the cross-sectional area divided by the wetted perimeter (Equation (6)), which is proportional to the channel width (Equation (5)). Due to these relationships between the channel width, wetted perimeter, hydraulic radius, and flow velocity, the flow velocity decreases when the channel width increases under the condition of an identical cross-sectional area of flow in the channel, which means that the amount of water flowing into the channel is the same (combining surface runoff, lateral flow, and baseflow). Therefore, when considering the flow velocity estimation procedure, the results where S1 showed a lower value for flow velocity than S2 were due to the fact that the channel width of S1 was more extensive than that of S2.

(a) (b)

Water 2019, 11, 718 11 of 18

3.2.2. Comparison of Simulated Streamflow for Scenarios 1 and 2 The calibration and validation were carried out during the periods from 2011 to 2013 and from 2014 to 2015, respectively. The optimal values of the parameters for Dosan and Andong Dam Stations are presented in Table2, and the calibrated values for S1 and S2 were similar. Figure8 shows the streamflow calibration and validation results for Dosan and Andong Dam Stations. The simulated streamflow adequately captured the observed data during the peak flow and recession limb, while the low flow or baseflow was underestimated at both stations. This is because the streamflow calibration was carried out focusing on the high flow (peak flow) periods due to the fact that channel geometry information primarily had an impact on the high flow. In Figure8, spring and winter were omitted because there were few rainfall events. In these hydrographs, one point of interest was that the simulated peak flow and recession limb tended to be relatively higher in S2 (red lines) than in S1 (blue lines) for both stations. This was due to the channel-routing method, the Muskingum routing method. The Muskingum routing method is based on an assumed linear relationship between a channel’s storage and inflow and outflow (Equation (10)) [36,37].

V = K (X q + (1 X) qout), (10) stored · · in − · 3 K = 1000 Lch , (11) · ·5vc where Vstored is the storage volume, K is the storage time constant, X is the weighting factor, qin is the inflow, qout is the flow out, Lch is the channel length (km), and vc is the flow velocity. Here, K is inversely proportional to the flow velocity and is proportional to the channel length (Equation (11)) [7]. Therefore, if the channel width is relatively larger like in S1, the flow velocity will be estimated to be smaller, and thus, K will be calculated to be larger so that the amount of water stored in the channel will increase. This means that the streamflow or outflow from the channel decreases because more water is stored by the channel when the channel width increases. This trend, where the peak flow as well as the falling limb of S1 showed that the higher values were smaller than those of S2 due to the difference between the channel widths, can be easily seen in all hydrographs in Figure8. Table5 shows the values of R 2, NSE, RSR, and PBIAS for the calibration and validation periods. According to the classification of model efficiencies (Table4), the values of R 2 and NSE during the calibration periods at both stations were ‘very good’ and those during the validation periods were ‘good’. The values of the RSR of S1 and S2 during the calibration period at Dosan Station were ‘good’ and ‘very good’, respectively, and those values were ‘satisfactory’ during validation. At Andong Dam Station, all values of RSR of S1 and S2 during the calibration and validation were ‘good’. Overall, although there was a slight difference in R2, NSE, and RSR between S1 and S2 at both stations, no specific differences were indicated. This is because the amount of water flowing into the channel was calculated as the sum of the surface runoff, lateral flow, and baseflow at the watershed scales, which was estimated using meteorological and topographical data before flowing into the channel. In other words, the parameters related to the cross-sectional shape of the channel had little effect on the flow rate calculation for the channel.

Table 5. Statistical criteria for the streamﬂow calibration and validation results.

Dosan Station Andong Dam Station Criteria Calibration Validation Calibration Validation S1 S2 S1 S2 S1 S2 S1 S2 R2 0.79 0.78 0.61 0.62 0.78 0.77 0.69 0.70 NSE 0.77 0.77 0.52 0.54 0.74 0.74 0.67 0.69 RSR 0.51 0.48 0.68 0.69 0.51 0.51 0.57 0.56 PBIAS 29.7 16.3 36.0 31.2 24.9 14.8 35.0 20.7 Water 2019, 11, 718 12 of 18

Unlike the previous three criteria, however, the values of PBIAS showed distinct diﬀerences between S1 and S2 during the calibration and validation periods at both stations. The values of S2 for the calibration and validation results were much greater than those of S1 at both stations. This means that the channel geometry of S2 estimated by the new regression equations reduced the deviation and improvedWater 2018, the10, x accuracyFOR PEER ofREVIEW the streamﬂow simulation. This can be seen in Figure8. 12 of 18

(a) (b)

(e) (f)

FigureFigure 8. 8. Hydrographs during during summer summer and and fall fall seasons seasons of ofeach each year. year. (a), ( ca),,c ,(ee)) DosanDosan Station Station and and (b(b,d),,f ()d Andong), (f) Andong Dam Dam Station. Station.

3.2.3. Comparison of Simulated Water Quality for Scenarios 1 and 2 ComparedTable to the 5. analysisStatistical results criteria offor streamflow the streamflow simulation, calibration theand distinct validation di ffresults.erences appeared in the water quality simulationDosan results Station for S1 and S2. The adjusted parameters Andong Dam for S1Station and S2 showed distinctlyCriteria diff erent valuesCalibration for all parametersValidation as shown in TableCalibration3. For example, the calibratedValidation values of N_UPDIS (rangesS1 from 0 to S2 100) wereS1 16.70 andS2 38.10 for S1S1 and S2, respectively, S2 S1 and SOL_NO3S2 (rangesR from2 0 to 100) 0.79 were 7.90 0.78 and 27.50,0.61 respectively. 0.62 0.78 0.77 0.69 0.70 TheNSE results of 0.77 model performance 0.77 0.52 evaluations 0.54 also reveal0.74 that the 0.74 channel 0.67 geometry-related 0.69 2 parametersRSR affected 0.51 water quality 0.48 simulation.0.68 According0.69 to0.51 the performance 0.51 ratings0.57 in Table0.564,R , NSE,PBIAS RSR, and PBIAS 29.7 of S2 for 16.3 the calibration36.0 results31.2 improved24.9 more compared 14.8 35.0 with ‘satisfactory’20.7 for Sediment, TN, and DO as shown in Table6. On the other hand, NSE, RSR, and PBIAS of S1 for3.2.3. TN Comparison during the calibrationof Simulated period Water were Quality within for Scenarios the ‘unsatisfactory’ 1 and 2 range. During the validation periods, the simulation results of S1 and S2 improved more than the ‘satisfactory’ ranges apart from Compared to the analysis results of streamflow simulation, the distinct differences appeared in DO. Although the DO simulation results indicated unsatisfactory, S2 showed the higher values for the water quality simulation results for S1 and S2. The adjusted parameters for S1 and S2 showed all model evaluation indicators compared to S1. This suggests that the water quality simulation was distinctly different values for all parameters as shown in Table 3. For example, the calibrated values influenced by the channel geometry information when considering the same parameters and ranges of N_UPDIS (ranges from 0 to 100) were 16.70 and 38.10 for S1 and S2, respectively, and SOL_NO3 were identically applied to S1 and S2 for calibration. (ranges from 0 to 100) were 7.90 and 27.50, respectively. The results of model performance evaluations also reveal that the channel geometry-related parameters affected water quality simulation. According to the performance ratings in Table 4, R2, NSE, RSR, and PBIAS of S2 for the calibration results improved more compared with ‘satisfactory’ for Sediment, TN, and DO as shown in Table 6. On the other hand, NSE, RSR, and PBIAS of S1 for TN during the calibration period were within the ‘unsatisfactory’ range. During the validation periods, the simulation results of S1 and S2 improved more than the ‘satisfactory’ ranges apart from DO. Although the DO simulation results indicated unsatisfactory, S2 showed the higher values for

Water 2019, 11, 718 13 of 18

Table 6. Statistical criteria for the water quality calibration and validation results.

Sediment TN DO Criteria Calibration Validation Calibration Validation Calibration Validation S1 S2 S1 S2 S1 S2 S1 S2 S1 S2 S1 S2 R2 0.65 0.87 0.84 0.84 0.65 0.76 0.71 0.72 0.79 0.79 0.21 0.32 NSE 0.65 0.87 0.78 0.80 0.41 0.65 0.32 0.41 0.59 0.62 0.49 0.21 − RSR 0.59 0.36 0.46 0.44 0.77 0.59 0.46 0.31 0.64 0.62 1.03 1.00 PBIAS 7.7 9.9 37.6 22.6 75.3 57.4 18.8 14.4 62.3 49.9 95.8 66.6 − −

The same results were also demonstrated by the graphical comparisons between the observed and simulated water quality for S1 and S2 during the calibration and validation periods. As shown in Figure9, even though some peak and low water quality values were underestimated in both scenarios, S2 tended to estimate water quality to be higher than S1 (closer to observation) concerning all water quality items, i.e., sediment, TN and DO. When comparing the total amount of sediment, TN and DO for ﬁve years between S1 and S2, the 5-year sum of S2 for sediment, TN and DO were higher with 20.8%,Water 2018 81.1%, 10, x and FOR28.6% PEER REVIEW respectively (Table7). 14 of 18

(a) (b)

(e) (f)

FigureFigure 9. 9. ComparisonsComparisons ofof water quality simulation results. results. ( (aa)) Sediment Sediment during during the the calibration calibration period. period. (b(b)) Sediment Sediment during during thethe validationvalidation period.period. (c) TN during the calibration period. period. (d (d) )TN TN during during the the validationvalidation period. period. ((ee)) DODO duringduring thethe calibrationcalibration period. ( (ff)) DO DO during during the the validation validation period. period.

From the analysis above, it seems that inaccurate channel geometry data could increase uncertainty in the water quality simulation. This misinterpretation, caused by the uncertainty of the simulation results, could eventually impede the proper establishment of watershed management. Figure 10 gives an example pertaining to this situation. Figure 10 is the load duration curve (LDC) for the sediment. The LDC is a graphical analytical tool that describes the relationships between streamflow and water quality. The LDC can assist in decision making regarding this relationship and provides basic information for the total maximum daily load (TMDL) [38], one of the most used watershed management approaches. Although the LDCs in Figures 10a,b were made for the same station, each LDC shows different tendencies for all flow duration intervals. In the case of Figure 10a, S1 exceeded the standard load under all flow duration intervals except the moist condition. S2, on the other hand, did not exceed the standard load under dry conditions and low flows, which means that the water quality for S2 met the target water quality under these duration intervals. Likewise, in Figure 10b, S1 and S2 showed a difference where the exceedance frequency of S1 was much higher under all conditions except for low flows. Consequently, S1 and S2 could result in different judgments on whether the current water quality of the target river was acceptable. Based on the example of Figure 10, this suggests that if the LDC is developed by erroneously simulated results derived from inaccurate channel geometry information, it could lead to misjudgments in determining the water quality status of a target river.

Water 2019, 11, 718 14 of 18

Table 7. Total amount of the simulated water quality for 5 years.

Sediment TN DO Values S1 S2 S1 S2 S1 S2 5-year sum (tons) 256,000 309,000 2427 4396 11,142 14,334 Percent diﬀerence 20.8% 81.1% 28.6%

Figure 10a,b shows the sediment simulation results. The simulated sediment from S2 was higher than in S1, especially when peak flow occurred. The difference between the two scenarios was due to the way that SWAT simulates the instream transportation for sediment. SWAT estimates the maximum concentration of sediment that can be transported by water in each sub-watershed depending on the flow velocity. Then, based on the maximum concentration of sediment, the deposition and degradation are determined. Accordingly, as the maximum concentration of sediment of S2 with a faster flow velocity was estimated to be larger, the amount of sediment at the sub-watershed outlet would be larger. Consequently, if the incorrect channel cross-section information is applied, the uncertainty in sedimentWater 2018, simulation 10, x FOR PEER could REVIEW be increased. 15 of 18

(a) (b)

FigureFigure 10. 10.Load Load durationduration curvecurve ofof sediment forfor the simulation periods periods (2011–2015). (2011–2015). (a (a) )Dosan Dosan Station. Station. (b(b)) Andong Andong Dam Dam Station. Station.

The results of TN were akin to the sediment results (in Figure9c,d); TN of S2 was higher than that of3.3. S1 Modification during peak of flow the SWAT periods 2012 and Code this tendencyto Apply Various was also Shapes caused of Channel by the e ffCross-Sectionsect of the channel geometry. The amount of TN at an outlet includes the values of organic nitrogen, ammonia, nitrite, as well as The channel geometry, as a symmetrical trapezoid, of the current SWAT model was improved nitrate, and each nitrogen-related item is estimated by subtracting the change in nitrate concentration. so that various cross-sectional shapes of a channel could be considered (Figure 11). In Figure 11, the The amount of change in the nitrate concentration is a function of nitrite, algae, biomass concentration, red letters are newly added input variables in order to take into account the various channel cross- and flow travel time. Here, the travel time is inversely proportional to the flow rate and is proportional sectional shapes. The inclination of each side of both the channel and floodplain were divided into to the volume of the channel. That is, the larger the cross-sectional area, the longer the flow travel time. independent input variables to consider different slopes. Additionally, in the present SWAT model, For this reason, the TN for S1 with a large cross-sectional area was smaller than that of S2. the floodplain width is automatically calculated by the model when the model is running, so it cannot The DO simulation results showed a similar trend with the TN results; S2 predicted higher DO be adjusted by the user. However, in the improved module, the floodplain bottom width is added thanand S1divided for the into calibration two sides and (left validation and right) periods. as new Thisinput is variables, due to the which same reasonhelps the as theuser TN to adjust estimation the process.floodplain Here, bottom the DO width. was computed by an equation related to atmospheric reaeration, algal biomass concentration, CBOD, channel depth, ammonium concentration, nitrite concentration, and flow travel In figure 11, Wfld,left and Wfld,right are the left and right side width of the floodplain; slpfld,left and time. As a result, because of the effect of travel time on the DO outflow, S1 with a large amount of slpfld,right are the side slope of the floodplain; and slpch,left and slpch,right are the slope of the main channel. changeThe cross-sectional of DO in the shape channel input discharged table should a smaller use the amount new improved of DO at thechannel watershed geometry outlet. module. If the userFrom has theactual analysis channel above, data it seemson the that cross inaccurate section, channelthe channel geometry shape data can could be accurately increase uncertainty reflected inthrough the water the quality newly simulation.added variables. This misinterpretation,Furthermore, the convenience, caused by the when uncertainty exploiting of thethe simulation modified results,module, could was improved eventually by impede adjusting the the proper SWAT establishment so that the user of could watershed select the management. current module Figure or an 10 givesimproved an example one. In pertainingaddition, the to SWAT this situation. 2012 code Figure was modified 10 is the to load display duration the sub-watershed curve (LDC) forname the sediment.where the The river LDC cross-section is a graphical data analyticalthat users generated tool that describes were applie thed relationships when using the between modified streamflow SWAT, andwhich water helps quality. the user The to LDC see if can cross-section assist in decision information making is applied regarding to each this sub-watershed. relationship and provides

Figure 11. Modified channel cross-section.

4. Conclusions Using multiple aerial photographs, the new regression equations were developed for estimating the bankfull stage width of the main channels and the bottom width of the floodplain in the middle- scale watershed, the Andong Dam watershed. Through a comparison with the new regression equations, the applicability of the current regression equations in SWAT were evaluated for the middle-scale watershed. Then, it was assessed how the channel geometry information affected the flow and water quality simulation. Finally, the channel geometry estimation module in the current SWAT model was improved by modifying SWAT to consider various cross-sectional shapes of channels.

Water 2018, 10, x FOR PEER REVIEW 15 of 18

Water 2019, 11, 718 15 of 18 basic information for the(a total) maximum daily load (TMDL) [38], one of(b the) most used watershed management approaches. Figure 10. Load duration curve of sediment for the simulation periods (2011–2015). (a) Dosan Station. Although(b) Andong the Dam LDCs Station. in Figure 10a,b were made for the same station, each LDC shows different tendencies for all flow duration intervals. In the case of Figure 10a, S1 exceeded the standard load under all flow duration intervals except the moist condition. S2, on the other hand, did not exceed the standard3.3. Modification load under of the dry SWAT conditions 2012 Code and to lowApply flows, Various which Shapes means of Channel that theCross-Sections water quality for S2 met the target water quality under these duration intervals. Likewise, in Figure 10b, S1 and S2 showed The channel geometry, as a symmetrical trapezoid, of the current SWAT model was improved a difference where the exceedance frequency of S1 was much higher under all conditions except for so that various cross-sectional shapes of a channel could be considered (Figure 11). In Figure 11, the low flows. Consequently, S1 and S2 could result in different judgments on whether the current water red letters are newly added input variables in order to take into account the various channel cross- qualitysectional of theshapes. target The river inclination was acceptable. of each side Based of both on thethe examplechannel and of Figure floodplain 10, this were suggests divided that into if theindependent LDC is developed input variables by erroneously to consider simulated different results slopes. derived Additionally, from inaccuratein the present channel SWAT geometry model, information,the floodplain it couldwidth leadis automatically to misjudgments calculated in determining by the model the when water the quality model status is running, of a target so it cannot river. be adjusted by the user. However, in the improved module, the floodplain bottom width is added 3.3. Modification of the SWAT 2012 Code to Apply Various Shapes of Channel Cross-Sections and divided into two sides (left and right) as new input variables, which helps the user to adjust the floodplainThe channel bottom geometry, width. as a symmetrical trapezoid, of the current SWAT model was improved so that variousIn figure cross-sectional 11, Wfld,left and shapesWfld,right ofare a channel the left could and right be considered side width (Figure of the 11 floodplain;). In Figure sl 11pfld,left, the and red lettersslpfld,right areare newly the side added slope input of the variables floodplain; in order and to slp takech,left into and account slpch,right theare variousthe slope channel of the main cross-sectional channel. shapes.The cross-sectional The inclination shape of each input side table of should both the use channel the new and improved floodplain channel were dividedgeometry into module. independent If the inputuser variableshas actual to channel consider data diff erenton the slopes. cross Additionally,section, the channel in the present shape SWATcan be model, accurately the floodplainreflected widththrough is automatically the newly added calculated variables. bythe Furthermore, model when the the convenience, model is running, when exploiting so it cannot the be modified adjusted bymodule, the user. was However, improved in by the adjusting improved the module, SWAT so the that floodplain the user could bottom select width the iscurrent added module and divided or an intoimproved two sides one. (left In addition, and right) the as SWAT new input 2012 variables,code was whichmodified helps to display the user the to sub-watershed adjust the floodplain name bottomwhere width.the river cross-section data that users generated were applied when using the modified SWAT,

FigureFigure 11.11. ModiﬁedModified channel cross-section.

4. ConclusionsIn Figure 11 , Wfld,left and Wfld,right are the left and right side width of the floodplain; slpfld,left and slp are the side slope of the floodplain; and slp and slp are the slope of the main channel. fld,rightUsing multiple aerial photographs, the new regrch,leftession equationsch,right were developed for estimating The cross-sectional shape input table should use the new improved channel geometry module. If the the bankfull stage width of the main channels and the bottom width of the floodplain in the middle- user has actual channel data on the cross section, the channel shape can be accurately reflected through scale watershed, the Andong Dam watershed. Through a comparison with the new regression the newly added variables. Furthermore, the convenience, when exploiting the modified module, was equations, the applicability of the current regression equations in SWAT were evaluated for the improved by adjusting the SWAT so that the user could select the current module or an improved one. middle-scale watershed. Then, it was assessed how the channel geometry information affected the In addition, the SWAT 2012 code was modified to display the sub-watershed name where the river flow and water quality simulation. Finally, the channel geometry estimation module in the current cross-section data that users generated were applied when using the modified SWAT, which helps the SWAT model was improved by modifying SWAT to consider various cross-sectional shapes of userchannels. to see if cross-section information is applied to each sub-watershed. 4. Conclusions Using multiple aerial photographs, the new regression equations were developed for estimating the bankfull stage width of the main channels and the bottom width of the floodplain in the middle-scale watershed, the Andong Dam watershed. Through a comparison with the new regression equations, the applicability of the current regression equations in SWAT were evaluated for the middle-scale watershed. Then, it was assessed how the channel geometry information affected the flow and water quality simulation. Finally, the channel geometry estimation module in the current SWAT model was improved by modifying SWAT to consider various cross-sectional shapes of channels. Water 2019, 11, 718 16 of 18

The new regression equations developed using aerial photographs showed higher accuracy than the current regression equations in the SWAT modeling. The current regression equations are likely to overestimate both the main channel width and the floodplain bottom width. Consequently, the current regression equations in SWAT showed poor applicability to the middle-scale watershed as mentioned by Ames et al. [6]. The flow velocity, streamflow, and water quality results estimated in two different scenarios, S1 (using current equations) and S2 (using new equations), were compared to assess the impact of the channel geometry information. While the flow velocity of S1 with the larger channel width showed smaller values than S2, the streamflow simulation results showed no significant differences in general. However, the peak flow rate and falling limb of S2 were estimated to be somewhat higher (closer to observations). Such results occurred because the variation in the channel geometry had an effect on the storage time constant and then caused the outflow from the channel to be diminished by increasing the amount of water stored in the channel. The water quality simulation results (e.g., sediment, TN, and DO) showed the obvious improvement of model performances when the new regression equations were applied; all statistic criteria (R2, NSE, RSR, and PBIAS) indicated greater values than those applying the current equations. The simulation results of S2 replicated the observed water quality data better by predicting peak values to be higher than those of S1. Moreover, the analysis of LDCs developed by the results of S1 and S2 suggested that the application of faulty channel geometry information to the channel not only leads to the uncertainty of the water quality simulation but also to the erroneous analysis of water quality condition. Therefore, it was confirmed that the accurate channel cross-section data are as critical as meteorological and topographical input data. However, this study has a limitation as the new regression equations were developed targeting only one region, the Andong Dam watershed; it is premature to conclude whether the results from this study can be applicable worldwide. Thus, further research applying this study’s approach should be conducted with respect to other middle- and small-scale watersheds with different characteristics. Nonetheless, this study demonstrated that channel geometry could increase the accuracy of modeling water quality simulations by changing the hydraulic characteristics of the channel, although it does not substantially affect streamflow simulation. The demonstration derived in this study is also critical in other fields, as the hydraulic properties of a channel are one of the significant factors determining the aquatic ecosystem or hydrological component.

Author Contributions: Conceptualization, J.H.; Data curation, D.L.; Formal analysis, S.L.; Investigation, D.L.; Methodology, J.H., M.P. and J.K.; Project administration, S.J.K. and J.K.; Supervision, K.J.L.; Writing – original draft, J.H.; Writing – review & editing, S.-W.C., S.J.K., K.J.L. and J.K. Acknowledgments: This work is supported by the Korea Environmental Industry & Technology Institute (KEITI) grant funded bythe Ministry of Environment (RE201901083) and by a grant from the National Institute of Environment Research (NIER), funded by the Han Gang Watershed Management Committee (HGWMC) of the Republic of Korea (NIER-2017-05-01-018, Integrated monitoring and management plan for non-point source pollution). Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Borah, D.K.; Bera, M. Watershed-scale hydrologic and nonpoint-source pollution models: Review of mathematical bases. Trans. ASAE 2003, 46, 1553. [CrossRef] 2. Bárdossy, A. Calibration of hydrological model parameters for ungauged catchments. Hydrol. Earth Syst. Sci. 2007, 11, 703–710. [CrossRef] 3. Borah, D.K.; Bera, M. SWAT model background and application reviews. In Proceedings of the 2003 ASAE Annual Meeting, Lass Vegas, NV, USA, 27–30 July 2003. 4. Choi, B.; Kang, H.; Lee, W. Baseﬂow Contribution to Streamﬂow and Aquatic Habitats Using Physical Habitat Simulations. Water 2018, 10, 1304. [CrossRef] 5. Stewardson, M. Hydraulic geometry of stream reaches. J. Hydrol. 2005, 306, 97–111. [CrossRef] Water 2019, 11, 718 17 of 18

6. Ames, D.P.; Rafn, E.B.; Van Kirk, R.; Crosby, B. Estimation of stream channel geometry in Idaho using GIS-derived watershed characteristics. Environ. Model. Softw. 2009, 24, 444–448. [CrossRef] 7. Neitsch, S.; Arnold, J.; Kiniry, J.; Williams, J. Soil & Water Assessment Tool: Theoretical Documentation Version-version 2009; Texas Water Resources Institute: College Station, TX, USA, 2011; pp. 1–647. 8. Muttiah, R.S.; Srinivasan, R.; Allen, P.M.Prediction of two-year peak stream discharges using neural networks. J. Am. Water Resour. Assoc. 1997, 33, 625–630. [CrossRef] 9. Allen, P.M.; Arnold, J.C.; Byars, B.W. Downstream Channel Geometry For Use In Planning-Level Models. JAWRA J. Am. Water Resour. Assoc. 1994, 30, 663–671. [CrossRef] 10. Leopold, L.B.; Maddock, T.J. The Hydraulic Geometry of Stream Channels and Some Physiographic Implications; USGS Professional Paper: Pennsylvania, MD, USA, 1953; pp. 1–57. 11. Bieger, K.; Rathjens, H.; Allen, P.M.; Arnold, J.G. Development and Evaluation of Bankfull Hydraulic Geometry Relationships for the Physiographic Regions of the United States. J. Am. Water Resour. Assoc. 2015. [CrossRef] 12. Staley, N.; Bright, T.; Zeckoski, R.W.; Benham, B.L.; Brannan, K.M. Comparison of Hspf Outputs Using Ftables Generated with Field Survey and Digital Data. JAWRA J. Am. Water Resour. Assoc. 2006, 42, 1153–1162. [CrossRef] 13. Chan Min, S. Improving HSFP model’s hydraulic accuracy with FTABLES based on surveyed cross sections. J. Korean Soc. Water Environ. 2016, 32, 582–588. 14. Richards, K.S. Channel and flow geometry: A geomorphological perspective. Prog. Phys. Geogr. 1977, 1, 65–102. [CrossRef] 15. Ficklin, D.L.; Luo, Y.; Luedeling, E.; Zhang, M. Climate change sensitivity assessment of a highly agricultural watershed using SWAT. J. Hydrol. 2009, 1, 65–102. [CrossRef] 16. Cohen Liechti, T.; Matos, J.P.; Ferràs Segura, D.; Boillat, J.L.; Schleiss, A.J. Hydrological modelling of the Zambezi River Basin taking into account floodplain behaviour by a modified reservoir approach. Int. J. River Basin Manag. 2014, 12, 29–41. [CrossRef] 17. Wang, W.; Xie, Y.; Bi, M.; Wang, X.; Lu, Y.; Fan, Z. Effects of best management practices on nitrogen load reduction in tea fields with different slope gradients using the SWAT model. Appl. Geogr. 2018, 90, 200–213. [CrossRef] 18. Woldesenbet, T.A.; Elagib, N.A.; Ribbe, L.; Heinrich, J. Catchment response to climate and land use changes in the Upper Blue Nile sub-basins, Ethiopia. Sci. Total Environ. 2018, 644, 193–206. [CrossRef] 19. Lee, S.; Sadeghi, A.M.; McCarty, G.W.; Baffaut, C.; Lohani, S.; Duriancik, L.F.; Thompson, A.; Yeo, I.Y.; Wallace, C. Assessing the suitability of the Soil Vulnerability Index (SVI) on identifying croplands vulnerable to nitrogen loss using the SWAT model. CATENA 2018, 167, 1–12. [CrossRef] 20. Zhang, Y.; Xia, J.; Shao, Q.; Zhai, X. Water quantity and quality simulation by improved SWAT in highly regulated Huai River Basin of China. Stoch. Environ. Res. Risk Assess. 2013, 27, 11–27. [CrossRef] 21. Her, Y.; Jeong, J.; Bieger, K.; Rathjens, H.; Arnold, J.; Srinivasan, R. Implications of Conceptual Channel Representation on SWAT Streamflow and Sediment Modeling. J. Am. Water Resour. Assoc. 2017, 53, 725–747. [CrossRef] 22. Arnold, J.G.; Srinivasan, R.; Muttiah, R.S.; Williams, J.R. Large area hydrologic modeling and assessment part I: Model development. J. Am. Water Resour. Assoc. 1998, 34, 73–89. [CrossRef] 23. Jang, W.S.; Engel, B.; Ryu, J. Efficient flow calibration method for accurate estimation of baseflow using a watershed scale hydrological model (SWAT). Ecol. Eng. 2018, 125, 50–67. [CrossRef] 24. Singh, V.P. Elementary Hydrology; Prentice Hall: Englewood Cliffs, NJ, USA, 1992. 25. Cinotto, P.J. Development of Regional Curves of Bankfull-Channel Geometry and Discharge for Streams in the Non-Urban; US Department of the Interior, US Geological Survey: Pennsylvania, MD, USA, 2003. 26. Hyams, D.G. CurveExpert Professional. 2018, pp. 1–213. Available online: https://www.curveexpert.net/ products/curveexpert-professional/ (accessed on 6 April 2019). 27. Abbaspour, K.C. SWAT Calibration and Uncertainty Programs—A User Manual; Eawag: Zurich, Switzerland, 2008. 28. Abbaspour, K.C.; Vejdani, M.; Haghighat, S. SWAT-CUP Calibration and Uncertainty Programs for SWAT. In Proceedings of the Fourth International SWAT Conference, Delft, The Netherlands, 2–6 July 2007. 29. Arnold, J.G.; Moriasi, D.N.; Gassman, P.W.; Abbaspour, K.C.; White, M.J.; Griensven, V.; Liew, V. SWAT: Model use, calibration, and validation. Trans. ASABE 2012, 55, 1491–1508. [CrossRef] Water 2019, 11, 718 18 of 18

30. Abbaspour, K.C.; Rouholahnejad, E.; Vaghefi, S.; Srinivasan, R.; Yang, H.; Kløve, B. A continental-scale hydrology and water quality model for Europe: Calibration and uncertainty of a high-resolution large-scale SWAT model. J. Hydrol. 2015, 524, 733–752. [CrossRef] 31. Legates, D.R.; McCabe, G.J. Evaluating the use of “goodness-of-fit” measures in hydrologic and hydroclimatic model validation. Water Resour. Res. 1999, 35, 233–241. [CrossRef] 32. ASCE Task Committee on Definition of Criteria for Evaluation of Watershed Models of the Watershed Management Committee, I. and D. D. Criteria for evaluation of watershed models. J. Irrig. Drain. Eng. 1993, 119, 429–442. 33. Moriasi, D.N.; Arnold, J.G.; van Liew, M.W.; Bingner, R.L.; Harmel, R.D.; Veith, T.L. Model Evaluation Guidelines for Systematic Quantification of Accuracy in Watershed Simulations. Trans. ASABE 2007, 50, 885–900. [CrossRef] 34. Nash, J.E.; Sutcliffe, J.V. River flow forecasting through conceptual models part I—A discussion of principles. J. Hydrol. 1970, 10, 282–290. [CrossRef] 35. Gupta, H.V.; Sorooshian, S.; Yapo, P.O. Status of Automatic Calibration for Hydrologic Models: Comparison with Multilevel Expert Calibration. J. Hydrol. Eng. 1999, 4, 135–143. [CrossRef] 36. Tung, Y.K.; Asce, A.M. River flood routing by nonlinear Muskingum method. J. Hydraul. Eng. 1985, 111, 1447–1460. [CrossRef] 37. Nguyen, V.; Dietrich, J.; Uniyal, B.; Tran, D. Verification and Correction of the Hydrologic Routing in the Soil and Water Assessment Tool. Water 2018, 10, 1419. [CrossRef] 38. U.S. EPA Office of Wetland, Oceans and Watersheds. An Approach for Using Load Duration Curves in the Development of TMDLs; U.S. Environmental Protection Agentcy: Washington, DC, USA, 2007; EPA 841-B-07-006.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).