Sediment yield in a small mountain basin during extreme events

Giovanna Grossi1 & Paolo Caronna1

FRIEND project - MED groupUNESCO IHP-VII (2008-13)

4th International Workshop on Hydrological Extremes From prediction to prevention of hydrological risk in Mediterranean countries

University of Calabria, 15-17 September 2011

1Department of .Civil Engineering, Architecture, Land and Environment,University of Brescia, Italy, via Branze 43

Objectives, methodologies, case study-------------------------------------------------------

Hec‐Ras hydraulic model

Application of a ‘lumped type’ sedimenttransport flow equation

GIS implementation of the RUSLE equation

2)   Sediment yield in the river network

3)   Sediment transport

1) Water erosion on the hillslope

Area (A) 30.9 km2

Contour (P) 33.3 kmMain stream length (L) 10.6 km

Minimum elevation (Hmin) 195 m s.l.m.Maximum elevation (Hmax) 1131 m s.l.m.

Mean elevation (Hmedia) 643 m s.l.m.Max elevation range (∆Hmax) 1136 m

Mean elevation range(∆Hmedio) 448 m

Guerna watershed (BG – North Italy)-------------------------------------------------------

N

Lithology & Land use-------------------------------------------------------

Permeable rocks (karst)

Gravity sediments

Fields

Woods

Lithology and land use – maps were used for the GIS application

‐ Revised version of USLE: Universal Soil Loss Equation (Wischmeier e Smith, 1978)

PCSLKRA ⋅⋅⋅⋅⋅=

Erosion: Revised Universal Soil Loss Equation (RUSLE) (Renard et al.1991)-------------------------------------------------------

Erosion estimate: R and K factors-------------------------------------------------------

yearPR ⋅+= 48.346.38 (“Actual Erosion in the Alpine Space” , European Soil Bureau)

mmPyear 1260=yhha

mmMJ⋅⋅

⋅ 4423.26=R

K

Van Der Knijff et al. (2000)

t ha hha MJ mm

⋅ ⋅⋅ ⋅

.

0.035=K

Erosion estimate: L and S factors-------------------------------------------------------

β the slope

( )03.0sin8.1013.22

+= ⎟⎠⎞

⎜⎝⎛ β

λ m

LS

( )05.0sin8.1613.22

−= ⎟⎠⎞

⎜⎝⎛ β

λ m

LS

ff

m+

=1

( )56.031

0869.0sin

sin 8.0 +⋅=

ββf

if tanβ < 0.09

if tanβ ≥ 0.09

Being:

Renard et al. (1991) write: 

Topographic factors L and Swere computed on the basis of a20 m resolution digital elevationmodel.

Erosion estimate: C and P factors-------------------------------------------------------

The coltural factor C is computed as the ratio between the soil loss under actualconditions to losses experienced uder the reference conditions.

Tables depending on crops and crops rotation

Description C factorUrban area 0.003

Unproductive soil 0.36Old and new forest 0.003

Woodland 0.451Lawn 0.04

Sown ground 0.4Wild vegetation 0.003

The P factor is defined as the ratio between the soil loss with contouring and/orstripcropping to that with straight row farming up‐and‐down slope.

C

P=1 ;  P=0.34 (woodland)

Erosion estimate: obtained results-------------------------------------------------------

Average soil loss map for the Guerna watershed:

Category [t/y] Area [Km2]

0 ‐ 1 7.66

1 – 3 8.40

3 – 5 3.82

5 ‐ 10 3.83

10 – 20 2.62

20 ‐ 40 2.17

> 40 2.30Parcel soil loss, averaged over the basin area: 13 t/y 

,

Total soil loss: 225 000 t/y(Vol. 86 790 m3/y)Specific loss: 2.8 mm/y.

Sediment transport in the river network: HEC-RAS-------------------------------------------------------

Requested data for the definition of the hydraulic model of the creek:

A) Reaches and cross sections geometryB) Grain classesC)   Quasi‐unsteady flow computational scheme (definition of the upstream inflow event)

Hydrologic Engineering Centers River Analysis System (HEC‐ RAS)

Main stream legnth: 10.6 KmUpstream reach: 6.2 Km

Downstream reach: 4.4 Km (i = 0.006)

A1) Reaches

Sediment flow equation  Meyer‐Peter‐MullerFall velocity Ruby

Basic data requirements (I)-------------------------------------------------------

Upstream reach: 10 rectangular  cross sections 

width [m] height [m]2 0.4

Downstream reach: topographic survey

A2) cross section geometry

B) Roughness of the banks and of the stream bed

Manning coefficient (0.02  0.07 s/m1/3)

C) Grain classes

Basic data requirements (II)-------------------------------------------------------

C) Quasi‐unsteady flow computational scheme (precipitation event)

The peak discharge for a fixed return period was computed on the basis of the results ofa regionalisation study by Bacchi et al. (1999)

)(, cTTc QmXQ ⋅=73.024.3)( AQm c ⋅=

072.0033.1)))/)1ln((ln((0521.0(exp(53.01 −−−−⋅

⋅+=TTX T

2km 401 ≤≤ Awith:

Basic data requirements (III)-------------------------------------------------------

Section A [Km2] Tc [h] Qc,100 [m3/s]66 2.74 0.55 1752 2.47 0.53 1732.4 5.7 0.91 310 30.9 2.25 109

Q0 - (Q66 + Q52 + Q32.4) = 44 m3/s

HLATC ∆

+=

8.05.14

(Giandotti)

02468

1012141618

00:00

:0000

:02:00

00:04

:0000

:06:00

00:08

:0000

:10:00

00:12

:0000

:14:00

00:16

:0000

:18:00

00:20

:0000

:22:00

00:24

:0000

:26:00

00:28

:0000

:30:00

00:32

:0000

:34:00

00:36

:0000

:38:00

00:40

:0000

:42:00

00:44

:0000

:46:00

00:48

:0000

:50:00

00:52

:0000

:54:00

00:56

:0000

:58:00

01:00

:0001

:02:00

durata [ore]

portata [m3/s]

02468

1012141618

00:0

0:00

00:0

3:00

00:0

6:00

00:0

9:00

00:1

2:00

00:1

5:00

00:1

8:00

00:2

1:00

00:2

4:00

00:2

7:00

00:3

0:00

00:3

3:00

00:3

6:00

00:3

9:00

00:4

2:00

00:4

5:00

00:4

8:00

00:5

1:00

00:5

4:00

00:5

7:00

01:0

0:00

01:0

3:00

durata [ore]

portata [m3/s]

0

5

10

15

20

25

30

35

00:00

:0000

:05:00

00:10

:0000

:15:00

00:20

:0000

:25:00

00:30

:0000

:35:00

00:40

:0000

:45:00

00:50

:0000

:55:00

01:00

:0001

:05:00

01:10

:0001

:15:00

01:20

:0001

:25:00

01:30

:0001

:35:00

01:40

:0001

:45:00

durata [ore]

portata[m3/s]

To fill the 44 m3/s gap, a constant intake (0.47 m3/s) for 

each section during the simulation time has been 

considered

Sediment yield: modeling results (I)-------------------------------------------------------

Water levelsAnd profiles

Sediment yield: modeling results (II)-------------------------------------------------------

Obtained results: Sediment yield at the end of the simulation period

At the outlet in the Oglio river (section 0) sediment yield is 1192 t (γs=2600 kg/m3

Vs~460 m3)

Estimate of sediment transport: Meyer-Peter Müller equation-------------------------------------------------------

The sediment transport rate qs is the sediment volume discharge per section widthunit.

(Meyer‐Peter Müller, 1948)

The index Φ, is the non dimensional form for qs :

∆⋅⋅=−⋅=Φ

gd

qsc 3

50

5.1)(3.13 φφ

6/190

23

50

26' ; '1

; 047.0 ; −⋅=⎟⎟⎠

⎞⎜⎜⎝

⎛⋅

⋅⎟⎟⎠

⎞⎜⎜⎝

⎛−

⋅==

−=∆ dk

kk

d

iRS

S

S

SC

S

ρρ

φφρρρ

ρ : water unit mass

ρs : sediment unit mass

φC : Shields number

R : hydraulic radius

i : bed slope

ks : Strickler roughness due to bed grain and shapek’s : Strickler roughness due only tograinsd50 : 50% of sediment weight is lower insized90: 90% of sediment weight is lower insize

Thank you for your attention

d50[m]

d90[m]

Slopei [%]

Water dischargeQ [m3/s]

Meyer‐Peter Müller

QS [m3/s]I reach 0.8 1.5 9.1 48 0.298II reach  0.4 0.8 2.8 80 0.008III reach 0.1 0.3 0.6 109 0.051

Sediment transport: Meyer-Peter equation-------------------------------------------------------

1°

2°

3°On the basis of Meyer‐Peter Müller equation alone, sedimenttransport rates would be very low and their distribution wouldbe much different from those obtained through the detailedhydraulic model.

Conclusions-------------------------------------------------------

Sediment Transport analysis was performed using Hec‐Ras modeling framework,providing the sediment yield for a 100‐ year return period event

The GIS implementation of RUSLE has provided an estimate of the annual sediment yieldon the hillslopes of the watershed

Results are in agreement with literature values

Future research :

• Geometry of the upstream reach

• Improvement of the sediment gradation through additional field and laboratoryanalysis

• Link of the Gis‐ RUSLE and Hec‐RAS frameworks

225000 t/y (~ 86790 m3/y )

1192 t (~ 460 m3)

Thank you for your attention