2017. 07. 29.
Kangwon National University, Korea
Application of measured channel cross-section geometries data
for flow and water quality estimation
Jeongho Han, Yujin Choi, Seong Joon Kim, Bernard A. Engel, Kyoug Jae Lim, Jonggun Kim
0.0
100.0
200.0
300.0
400.0
500.0
600.0
700.0
800.0
900.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Floo
dpla
in b
otto
m w
idth
(m)
Subbasin number
0.0
20.0
40.0
60.0
80.0
100.0
120.0
140.0
160.0
180.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Bank
full
wid
th(m
)
Subbasin nunmber
Bankfull width and floodplain bottom width
-15-10
-505
10152025
Sed Org_N NH4 NO3 NO2 Org_P Min_P2013 7.8 -0.6 8.7 0.1 3.1 -1.0 19.72014 16.2 0.9 -8.4 0.0 -13.3 1.4 0.82015 12.5 -1.0 13.8 0.0 1.7 -1.7 24.02013-2015 11.2 -0.4 5.9 0.1 -2.4 -0.7 16.3
Perc
ent d
ifere
nce(
%)
-10-8-5-30358
10
Sed Org_N NH4 NO3 NO2 Org_P Min_P2013 8.1 0.0 1.2 0.1 -7.1 -0.1 4.32014 2.6 0.2 -4.2 0.1 -1.3 0.4 -3.22015 4.6 -0.4 9.9 0.0 9.2 -0.7 10.42013-2015 5.2 -0.1 2.6 0.0 -0.2 -0.2 4.2
Perc
ent d
ifere
nce(
%)
-15
-5
5
15
25
35
45
Sed Org_N NH4 NO3 NO2 Org_P Min_P2013 -2.1 0.2 -7.9 0.0 13.1 0.2 13.32014 -7.2 -0.2 13.5 -0.2 40.5 -0.3 29.62015 -4.7 -0.2 13.7 -0.2 41.4 -0.3 34.92013-2015 -4.7 -0.1 5.1 -0.1 30.2 -0.1 25.2
Perc
ent d
ifere
nce(
%)
Lower(No.32)
Middle(No.17)
Upper(No.7)
Simulation results: sediment & nutrients
Streamflow and water quality simulation models: SWAT, HSPF, APEX, AGNPS, WASP4…
Input Parameters
Model input data
DEM Landuse Soil
: curve number, soil layer depth, subbasin area, manning’s n…
Hydrological models and input parameters
Streamflow and water quality simulation models: SWAT, HSPF, APEX, AGNPS, WASP4…
Model input data
DEM Landuse Soil
Input Parameters: curve number, soil layer depth, subbasin area, manning’s n…
Right bankLeft bank Bankfull width
Floodplain
Bankfull depth
Stream-channel
Channel geometry(channel cross-section)
Cross-section area Wetted perimeter
Hydraulic radiusFlow velocity
Streamflow Water quality
Channel geometry(channel cross-section)
Channel cross-section in SWAT
Bankfull depth
Bankfull width
Main channel
Floodplain
Floodplain
y = 0.0814x0.6
0
20
40
60
80
100
120
140
0 50,000 100,000 150,000 200,000
Bank
full
wid
th(m
)
Watershed area(ha)
y = 0.0206x0.4
0.00
0.50
1.00
1.50
2.00
2.50
3.00
- 50,000 100,000 150,000 200,000
Bank
full
dept
h(m
)
Watershed area(ha)
y = 5x
0
200
400
600
800
1000
0 20 40 60 80 100 120 140 160 180
Floo
dpla
in b
otto
m w
idth
(m)
Bankfull width(m)
41
2
1
Daniel et al. (2009)
SWAT Gyeongan-river(South Korea)
Floodplain width in SWAT
Channel width
Actual floodplain width
?
Comparison SWAT and nature channel
Objectives of this study
To evaluate the effect of channel geometry on streamflow and
water quality
To improve SWAT engine to consider various channel geometry
Measure channel bankfull width and floodplain bottom width
Derive regression eqn.
Channel geometry input data
SWAT Running
Simulation result analysis
Edit SWAT source code
Study flowchart
Study area 1: regression equation
1 2
34. Banbyeon River
3. Upper Guem River
2. Yanggu River
1. Yeongpyeong River
5. Upper Nakdong River
6. Pyeongchang River
5
4
6
6. Pyeongchang River(2010-2015)
Study area 2: application edited channel data
Measure bankfull width and floodplain bottom width
13
Various aerial images were compared
• Regression equations for a small and mountainous watershed
Regression equations were derived by CurveExpert
Aerial images
Study area
0
20
40
60
80
100
120
140
160
180
0 40,000 80,000 120,000 160,000 200,000
Bank
full
wid
th(m
)
Watershed area(ha)
Measured SWAT Regression eqn.
𝒚𝒚 = 𝟎𝟎.𝟏𝟏𝟎𝟎𝟏𝟏𝟏𝟏𝟎𝟎. 𝟏𝟏𝟎𝟎𝟔𝟔
𝑹𝑹𝟐𝟐 = 𝟎𝟎.𝟔𝟔𝟗𝟗
Bankfull width regression eqn.
0
100
200
300
400
500
600
700
800
900
0 20 40 60 80 100 120 140 160 180
Floo
dpla
in b
otto
m w
idth
(m)
Bankfull width(m)
Measured SWAT Regression
𝒚𝒚 = 𝟐𝟐.𝟗𝟗𝟏𝟏𝟎𝟎𝟐𝟐𝟏𝟏0.94
𝑹𝑹𝟐𝟐 = 𝟎𝟎.𝟖𝟖𝟗𝟗
Floodplain bottom width regression eqn.
measuredSWATregression
Floodplain bottom widthBankfull width
Bankfull width and floodplain bottom width from each mothod
0.0
100.0
200.0
300.0
400.0
500.0
600.0
700.0
800.0
900.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Floo
dpla
in b
otto
m w
idth
(m)
Subbasin number
0.0
20.0
40.0
60.0
80.0
100.0
120.0
140.0
160.0
180.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Bank
full
wid
th(m
)
Subbasin nunmber
Bankfull width and floodplain bottom width from each mothod
Edit SWAT source code
Isosceles trapezoid channel cross-section various cross-section shapes can be applied
Fixed floodplain bottom width new parameters are added for user to change floodplain
as measured data or value from regression eqn.
• Improvement points
Improved channel cross-section(Red: new parameters)
(1)
(3)
(2)
(4)
Various channel cross-section
Upper(No.7)
Simulation results: streamflow
0.1
1
10
100
1000
1/1/2013 5/1/2013 9/1/2013 1/1/2014 5/1/2014 9/1/2014 1/1/2015 5/1/2015 9/1/2015 1/1/201
Stre
amflo
w(m
3/se
c)
SWAT measued channal data
7
17
34
Middle(No.17)
Simulation results: streamflow
1
10
100
1000
1/1/2013 5/1/2013 9/1/2013 1/1/2014 5/1/2014 9/1/2014 1/1/2015 5/1/2015 9/1/2015 1/1/201
Stre
amflo
w(m
3/se
c)
SWAT measued channal data
Lower(No.32)
Simulation results: streamflow
1
10
100
1000
10000
1/1/2013 5/1/2013 9/1/2013 1/1/2014 5/1/2014 9/1/2014 1/1/2015 5/1/2015 9/1/2015 1/1/201
Stre
amflo
w(m
3/se
c)
SWAT measued channal data
Upper(No.7)
Simulation results: Sediment
0
2000
4000
6000
8000
10000
12000
1/1/2013 5/1/2013 9/1/2013 1/1/2014 5/1/2014 9/1/2014 1/1/2015 5/1/2015 9/1/2015 1/1/201
Sedi
men
t(to
n/da
y)
SWAT measued channal data
1
10
100
1000
10000
100000
1/1/2013 5/1/2013 9/1/2013 1/1/2014 5/1/2014 9/1/2014 1/1/2015 5/1/2015 9/1/2015 1/1/201
Sedi
men
t(to
n/da
y)
Middle(No.17)
Simulation results: Sediment
1
10
100
1000
10000
1/1/2013 5/1/2013 9/1/2013 1/1/2014 5/1/2014 9/1/2014 1/1/2015 5/1/2015 9/1/2015 1/1/201
Sedi
men
t(to
n/da
y)
0
2000
4000
6000
8000
10000
12000
1/1/2013 5/1/2013 9/1/2013 1/1/2014 5/1/2014 9/1/2014 1/1/2015 5/1/2015 9/1/2015 1/1/201
Sedi
men
t(to
n/da
y)
SWAT measued channal data
Lower(No.32)
Simulation results: Sediment
0
10000
20000
30000
40000
50000
60000
1/1/2013 5/1/2013 9/1/2013 1/1/2014 5/1/2014 9/1/2014 1/1/2015 5/1/2015 9/1/2015 1/1/201
Sedi
men
t(to
n/da
y)
SWAT measued channal data
1
10
100
1000
10000
100000
1/1/2013 5/1/2013 9/1/2013 1/1/2014 5/1/2014 9/1/2014 1/1/2015 5/1/2015 9/1/2015 1/1/201
Sedi
men
t(to
n/da
y)
-15-10
-505
10152025
Sed Org_N NH4 NO3 NO2 Org_P Min_P2013 7.8 -0.6 8.7 0.1 3.1 -1.0 19.72014 16.2 0.9 -8.4 0.0 -13.3 1.4 0.82015 12.5 -1.0 13.8 0.0 1.7 -1.7 24.02013-2015 11.2 -0.4 5.9 0.1 -2.4 -0.7 16.3
Perc
ent d
ifere
nce(
%)
-10-8-5-30358
10
Sed Org_N NH4 NO3 NO2 Org_P Min_P2013 8.1 0.0 1.2 0.1 -7.1 -0.1 4.32014 2.6 0.2 -4.2 0.1 -1.3 0.4 -3.22015 4.6 -0.4 9.9 0.0 9.2 -0.7 10.42013-2015 5.2 -0.1 2.6 0.0 -0.2 -0.2 4.2
Perc
ent d
ifere
nce(
%)
-15
-5
5
15
25
35
45
Sed Org_N NH4 NO3 NO2 Org_P Min_P2013 -2.1 0.2 -7.9 0.0 13.1 0.2 13.32014 -7.2 -0.2 13.5 -0.2 40.5 -0.3 29.62015 -4.7 -0.2 13.7 -0.2 41.4 -0.3 34.92013-2015 -4.7 -0.1 5.1 -0.1 30.2 -0.1 25.2
Perc
ent d
ifere
nce(
%)
Lower(No.32)
Middle(No.17)
Upper(No.7)
Simulation results: sediment & nutrients
Streamflow is not almost affected by channel geometry
Water quality is influenced largely by channel geometry
except organic N, NO3 and organic P
For a small and mountainous watershed, it is required to
develop new equation for channel geometry parameters
16
ThankYou