Application of measured channel cross-section geometries ... · Isosceles trapezoid channel cross...

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)


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)


Lower(No.32)


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)


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)


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)


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)


Lower(No.32)


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)


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

Application of measured channel cross-section geometries ... · Isosceles trapezoid channel cross...

Documents

Transcript of Application of measured channel cross-section geometries ... · Isosceles trapezoid channel cross...