MODELACION NUMERICA DEL TRANSPORTE DE SEDIMENTOS ...

��

�Director for Research and Education, CREAR �

Dept. of Civil and Environmental Engineering (CEE)�Dept. of Geology and Planetary Science (GPS)�

University of Pittsburgh, USA " �

Jorge D. Abad1 & Alejandro Mendoza�1Assistant Professor�

Center for Research and Education of the Amazonian Rainforest www.crearamazonia.org �

MODELACION NUMERICA DEL TRANSPORTE DE SEDIMENTOS à MORFOLOGIA�

http://www.pitt.edu/~jabad/ �

Collaborators: �@ University of Pittsburgh: �-  Center for Latin American Studies (CLAS)�-  Alejandro Mendoza (Postdoc)�-  Christian Frias, Ronald Gutierrez, Kristin Dauer (G) �-  Adrian Garcia, Brian Hone, Collin Ortals (UG) �

Programa GeneralCurso básico dehidráulica !uvial

Jiutepec, Morelos, México. 22 y 23 de octubre, 2013.

Seminario Internacional de Potamología“JOSÉ ANTONIO MAZA ÁLVAREZ”IV

TARIFAS, INSCRIPCIÓN Y FORMA DE PAGO(LOS COSTOS INCLUYEN IVA)

SeminarioProfesionales: $5,000.00Estudiantes: entrada libre a las sesiones técnicas previa inscripción al evento y presentación de credencial de estudiante vigente

Curso básico de Hidráulica FluvialProfesionales: $2,500.00 Estudiantes: $1,000.00 (previa inscripción al curso y presentación de credencial de estudiante vigente)

Para asistir al Seminario y al curso básico es requisito llenar la hoja de inscripción y cubrir la cuota correspondiente mediante depósito bancario o transferencia a favor del IMTA, a la cuenta IMTA Ingresos Propios No. 07501436317 del banco SCOTIABANK INVERLAT S.A., Número de sucursal 01 JIUTEPEC MORELOS, Plaza 075, Matriz y/o Sucursal 1717 JIUTEPEC, o bien mediante transferencia electrónica a la CLABE 044543075014363171, Dígito Referenciado 14084. El registro al seminario deberá realizarse por internet desde el micrositio del evento anunciado en la página web del IMTA: http://www.imta.gob.mx (Sección TEMAS DE INTERÉS - Foros, seminarios y eventos)

COMITÉ ORGANIZADOR

PresidenteM. I. Víctor Javier Bourguett Ortíz Tel: 777 [email protected]

Coordinador CientíficoM. A. José Raúl Saavedra Horita Tel: 777 3293600 Ext. 887, [email protected]

Comité TécnicoDr. Pedro Antonio Guido Aldana Tel: 777 3293600 Ext. [email protected]

M. en C. Alberto Güitrón de los Reyes Tel: 777 [email protected]

Lugar: Centro de Capacitación - IMTA.

Foto

: Jav

ier

de la

Maz

aRí

o U

sum

acin

ta, E

do. d

e C

hiap

as.

Σ ([1] + [2] + [3] +[4]+…) = RIVER DYNAMICS

[4]FLUID MOTION

[3]SEDIMENT TRANSPORT

[2]BED MORPHOLOGY

[1]PLANFORM DYNAMICS

RIVERS | PLANFORM| BED MORPHOLOGY AND SED. TRANSPORT| HYDRODYNAMICS| GENERALITITES

Mean and fluctuating velocities ●  Turbulence originates from the instability of laminar flows ●  The parameter that plays the most important role in this process is the Reynolds number

νHU

=Re

Re=26 Re=2000

Re=10000 Re~5x105

Wake far behind the obstacle

Re=1.54

H U

Turbulence modeling ●  We can think of turbulence as a

Ø Collection of eddies flowing Ø Largest eddies of size L and smallest eddies of size h Ø Eddies interact in a very complex manner Ø This interaction happens in a self-similar way

u Turbulent flow

Measuring mean and turbulent flow

'iii uuu +=

x z y

Free surface

ui = (u,v,w)

Measuring mean and turbulent flow

x z y

Free surface

ui = (u,v,w)

Instantaneous flow equations

Different Approaches to Modeling Turbulent Flows

Eddy size

Ener

gy d

ensi

ty

in

RANS

LES

DNS

out transfer

●  DNS: direct numerical simulation (resolves all scales)

●  LES: Large Eddy Simulation (resolves large and intermediate scales and models small scales)

●  RANS: Reynolds Averaged Navier-Stokes (resolves large scales and models intermediate and small scales)

resolved modeled RANS

modeled resolved LES

resolved DNS C

ompu

tatio

nal c

ost i

ncre

ases

Scal

e of

sim

ulat

ion

incr

ease

s

Empi

rical

mod

elin

g in

crea

ses

Comparison of Modeling Approaches

Eddy size

Eddy

ene

rgy

in

out

RANS

LES DNS

LES (Large Eddy Simulation)

15 16

WSE WSE

Abad et al. (2013), ESPL

RANS modeling: KINOSHITA CHANNEL

Modeling in river-type systems

Eddy size

Mor

phod

ynam

ic

RANS

LES

DNS

1D, 2D and 3D flow models

3D flow equations

x z y

Free surface

Channel bottom

u,v,w

h

u(x,y,z),v(x,y,z),w(x,y,z) h(x,y,z)

[1] Mass conservation

[2] x-momentum

[3] y-momentum

[4] z-momentum

x z y

U,V Free surface

Channel bottom

h

U(x,y),V(x,y) h(x,y)

2D flow equations


[2] x-momentum

[3] y-momentum

x z y

U Free surface

Channel bottom

h

U(x) h(x)

1D flow equations 1.2. FLUID GOVERNING EQUATIONS 9

(Cf = 9.81n2h�1/3). Figure 1.7 shows an example where the depth-averagedform of the governing equations were used to simulate the fluid motion arounda bend (Abad et al., 2008a). Notice only horizontal velocities are computed.

1.2.3 One-dimensional (1D) form

Similarly, by integrating the 3D governing equation in a cross section, the 1Dcross-averaged governing equations are derived as:

@A

@t+

@Q

@x= 0 (1.9)

Figure 1.8: Example of simulation using 1D governing equations. Simulationperformed using HEC-RAS model

@Q

@t+

@

@x(�U2A) + gA

@h

@x= gA(So � Sf ) (1.10)

Where Q is the water discharge or flow rate, A is the cross-section area, � isthe momentum coe�cient, So and Sf are the bed and friction slopes respectively.Figure 1.8 shows an example where the 1D form of the governing equationswere used to simulate the hydrodynamics for the Matanza-Riachuelo creek inArgentina.

1.2. FLUID GOVERNING EQUATIONS 9

(Cf = 9.81n2h�1/3). Figure 1.7 shows an example where the depth-averagedform of the governing equations were used to simulate the fluid motion arounda bend (Abad et al., 2008a). Notice only horizontal velocities are computed.

1.2.3 One-dimensional (1D) form

Similarly, by integrating the 3D governing equation in a cross section, the 1Dcross-averaged governing equations are derived as:

@A

@t+

@Q

@x= 0 (1.9)

Figure 1.8: Example of simulation using 1D governing equations. Simulationperformed using HEC-RAS model

@Q

@t+

@

@x(�U2A) + gA

@h

@x= gA(So � Sf ) (1.10)

Where Q is the water discharge or flow rate, A is the cross-section area, � isthe momentum coe�cient, So and Sf are the bed and friction slopes respectively.Figure 1.8 shows an example where the 1D form of the governing equationswere used to simulate the hydrodynamics for the Matanza-Riachuelo creek inArgentina.


[2] x-momentum

Sediment Transport Processes And Sediment Conservation Equation (Exner)

Bed load and suspended load

suspended/wash load

bedload

Schmeeckle

( )Ecvqt

)1( bsbp −+⋅∇=∂

η∂λ−

-

ybyxbxb eqeqq +=

xy

z sediment bed

( ) 0455.0~037.0,7.5q c5.1

cb =ττ−τ= ∗∗∗∗

( )( ) 05.0,17q cccb =ττ−ττ−τ= ∗∗∗∗∗∗

( )( ) 05.0,7.074.18q cccb =ττ−ττ−τ= ∗∗∗∗∗∗

∗

∗−τ+

−τ−

−

+=

π− ∫

∗

∗

b

b2)/143.0(

2)/143.0(

t

q5.431q5.43dte11

2

( ) 03.0,12.11q c

5.4

c5.1b =τ⎟

⎟⎠

⎞⎜⎜⎝

⎛

τ

τ−τ= ∗

∗

∗∗∗

Fernandez Luque & van Beek (1976)

Ashida & Michiue (1972)

Engelund & Fredsoe (1976)

Einstein (1950)

Parker (1979) fit to Einstein (1950)

A BEDLOAD TRANSPORT RELATIONS FOR UNIFORM SEDIMENT (Parker’s e-book)

)(qqor)(qq cbbbb∗∗∗∗∗∗∗ τ−τ=τ=

19

1.E-09

1.E-08

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

1.E-02

1.E-01

1.E+00

1.E+01

1.E+02

0.01 0.1 1τ*

q b*

EAMEFFLBSandP approx EFLBGrav

PLOTS OF BEDLOAD TRANSPORT RELATIONS (Parker’s e-book)

E = Einstein AM = Ashida-Michiue EF = Engelund-Fredsoe P approx E = Parker approx of Einstein FLBSand = Fernandez Luque-van

Beek, τc* = 0.038 FLBGrav = Fernandez Luque-van

Beek, τc* = 0.0455

3D simulation - Hydrodynamic

1) EXISTING CONDITION: Lidar (2011)+MBES (2012)"

DATA"-  Water slope and bed elevation calculation""

A slope of 1E-4 from gaging stations Mt. Carmel and New Harmony was calculated. Using that slope and the elevation of Mt. Carmel on February 2, 2012 a water surface was created for Maier Bend. Using the water surface and the MBES data the bed elevation surface was obtained.

FIELD MEASUREMENTS – WHAT DO WE NEED TO MEASURE?"

3D MODELING – MAIER BEND"

1)  Existing condition – Lidar (2011)

3D MODELING – MAIER BEND"1)  Existing condition - MBES (2012)


DATA"-  Water slope and bed elevation calculation""


DATA"-  MBES + Lidar + point bar (GPS)""


DATA"-  aDcp – details of the secondary flows, streamwise velocities and Umag""

CS3(011-014)

CS4(016-019)

CS5(021-024)

CS6(026-029)

CS10(044-045)


DATA"-  aDcp – details of the secondary flows, streamwise velocities and Umag""

CS4(016-019)

Umag [cm/s]


DATA"-  aDcp – details of the secondary flows, streamwise velocities and Umag"" Umag

[cm/s]

CS5(021-024)



[cm/s]

CS6(026-029)



[cm/s]

CS7(031-034)


PRE-PROCESSING FOR 3D MODEL"-  Details of Mesh generated (unstructured or structured)"

Model Domain



Background Mesh Δs=2m Δn=2m Δz=0.5m

z s

n

Mesh was generated with OpenFOAM tool snappyHexMesh. It uses a background Hexahedral mesh and an STL surface to create the final mesh.

STL surface



Mesh of hexahedral , polyhedra and pyramids elements


RESULTS"-  Flow in 3D visualization (velocities)- cavities "


RESULTS"-  Comparison between aDcp and Modeled (Umag, secondary flows, etc)"

CS6(026-029)

Umag [cm/s]



CS7(031-034)

Umag [cm/s]



Umag [cm/s]

CS8(036-039)


RESULTS"-  Prediction of bed and bank shear stresses"

Regular Mesh


RESULTS"-  Prediction of bed and bank shear stresses"

Fine Mesh

2) NO DUNES"

Gutierrez et al. (2013) –JGR-ES"-  Figure showing 1D wavelet"

2) NO DUNES"

Gutierrez et al. (2013) –JGR-ES"-  Figure showing 3D bed – not working"

2) NO DUNES"

Gutierrez et al. (2013) –JGR-ES"-  Figure showing smoothed bed condition"

2) NO DUNES"

Gutierrez et al. (2013) –JGR-ES"-  Figure showing the mesh"

Steady + unsteady bed morphology

η(t) = <η> + η’(t) <η>

η’(t)

Dunes

η(t) = <η> + η’(t)

<η>

η’(t) Bed profiles at centerline

<qb> ~ 9.68E-07 m2/s

<qb> qb’

Bed load fluxes

qb(t) = <qb> + qb’(t)

Steady (Point bars) + perturbed components (bars)

1D SEDIMENT TRANSPORT MORPHODYNAMICS with applications to

RIVERS AND TURBIDITY CURRENTS © Gary Parker November, 2004

51

TOUR OF BEDFORMS IN RIVERS: ALTERNATE BARS

Alternate bars in the Naka River, an artificially straightened river in Japan. Image courtesy S. Ikeda.

Alternate bars occur in rivers with sufficiently large (> ~ 12), but not too large width-depth ratio B/H. Alternate bars migrate downstream, and often have relatively sharp fronts. They are often precursors to meandering. Alternate bars may coexist with dunes and/or antidunes.



52

BEDFORMS IN THE LABORATORY AND FIELD: ALTERNATE BARS Alternate bars in a flume in Tsukuba University, Japan: flow turned low.

Image courtesy H. Ikeda.

Alternate bars in the Rhine River between Switzerland and Lichtenstein.

Image courtesy M. Jaeggi.



53

BEDFORMS IN THE LABORATORY AND FIELD: MULTIPLE-ROW (LINGUOID) BARS

Linguoid bars in a flume in Tsukuba University, Japan: flow turned off.

Image courtesy H. Ikeda. Linguoid bars in the Fuefuki River, Japan. Image courtesy S. Ikeda.

ρs=2566.5 Kg/m3

Grain Size Distribution

0

10

20

30

40

50

60

70

80

90

100

0.1 1 10Grain size (mm)

Perc

ent f

iner

D50=0.832 mm

Dg=0.828 mmσ

g=1.234<1.60 (well sorted)

λp=0.4 ( ) encsq *** ττφα −=

Designing the experiments: Obtaining βC and βR à Sediment transport

β

λπ 02 hk =cc k,βRR k,β

Lisle et al. (1997)

Free bars in straight channels

λ

B2

Free bars regime

Pool

Pool

Pool

Pool Free bars in HAMC

Whiting and Dietrich (1993)

Modeling of forced and free bars Alejandro Mendoza (Postdoc)

Point bars Free bars

Tools 2D 3D

•  Telemac-‐Mascaret http://www.opentelemac.org/

•  Mike 21C http://mikebydhi.com/Products/WaterResources/MIKE21C.aspx

•  Delft 3d http://www.deltaressystems.com/hydro/product/621497/delft3d-‐suite

Bed Morphology

•  Telemac-‐2D – Solves St. Venant equations – Utilizes FEM in irregular grids

•  Sisyphe – Computes sediment transport with empirical formulas

– Applies corrections for bed/bedforms effects and secondary flows when using Telemac-‐2D

FORCED BARS Hooke experiments

Hooke experiments

After Hooke (1975) (Journal of Geology)

Experiments

•  Q = [10-‐50.5] l/s •  Recirculated sediment •  diameter = 0.30 mm •  Density = 2700 kg/m3

Mesh and Boundary Conditions

HQL

QS

Inlet Outlet

Friction coefficient

Sediment roughness

Sediment transport

•  qs = 2E-‐6 m2/s (from experiments)

•  Calibration using different equations

•  Roughness of sediment particles

•  Use of corrections for bed slope and particle trajectory

Q=20 l/s

Q=50.5 l/s

Results for Q=20 l/s

Hooke (1975) (Journal of Geology)

FREE BARS Lanzoni Experiments

Lanzoni (2000a) WRR •  Development of alternate bars in ripple and/or covered beds

•  Flume 55 m long 1.5 m wide •  Sediment recirculation, •  Qs = 1.05E-‐5 m2/s (converted from Qs = 94.5 l/h pores included)

•  Sediment characteristics – d = 0.48 mm – ρ = 2650 kg/m3 – Q = [25-‐47] l/s

More experiments for free bars

•  Modeling is done with Telemac-‐2D and Sisyphe •  Equation selected was Meyer-‐Peter and Mueller. Since Telemac computes µ = c’f/cf with c’f from skin friction calculated with Nikuradse, the sediment transport in Sisyphe was calibrated with ks = 3.6D50

Sediment recirculation

Q= 30 l/s Initial perturbation: Bump in the inlet

Bed evolution

Lanzoni [2000a] (Table 2, run 1505) Wavelength = 10.0m Bar height = 7.7 cm celerity = 2.8 m/h

72 HRS SIMULATION

Since bars are advected together with the bump, a permanent perturbation is needed

Results

•  Evolution of the bed, right bank •  Profiles every 60 min. •  Wavelength 10 m aprox. (Lanzoni

measured 10m) •  Celerity 2.2 m/h (Lanzoni

measured 2.8 m/h) •  Bar height 8 cm (7.7 from

Lanzoni)

Important aspects for bed morphology modeling

•  Measurements of sediment transport from field

•  Calibrate the sediment equation in order to reproduce measurements

RVR Meander (www.rvrmeander.org)

Jorge D. Abad1, Davide Mo:a2, Eddy J. Langendoen3, Roberto Fernandez2, Nils O. Oberg2, Marcelo H. Garcia2

4River Centerline

Valley Centerline

RVR Meander User Interface

0

5

10

15

20

25

30

35

MigratedCenterlines (yrs)

2D Output

1.1 - 1.5

0.8 - 1.1

0.5 - 0.8

1.1 - 1.3

0.8 - 1.1

0.5 - 0.8Velo

city

Mag

nitu

de (m

/s)

After 5 years After 35 years

2.1 - 3.5

0.6 - 2.1

-0.9 - 0.6

1.9 - 3.2

0.6 - 1.9

-0.8 - 0.6

Wat

er D

epth

(m)

11.9 - 16.6

7.1 - 11.9

2.3 - 7.1

9.7 - 13.4

5.9 - 9.7

2.2 - 5.9

Shea

r Str

ess

Mag

nitu

de (P

a)

1D Output

Floodplain HeterogeneityBasic Input

1 Dept. of Civil and Environmental Engineering, University of Pittsburgh 2 Ven Te Chow Hydrosystems Laboratory, Dept. of Civil and Environmental Engineering, University of Illinois at Urbana-Champaign 3 USDA-ARS National Sedimentation Laboratory, Oxford, Mississippi

Meandering scales Somewhere near Detroit

Pictures courtesy Eddy Langendoen

λv Α

Meandering scales (λ) With the ongoing effort in both the United States and Europe to re-‐naturalize highly modified streams à more development of GIS Engineering Tools are needed

Proposed alignments for re-‐meandering of Trout Creek, Lake Tahoe, California.

US reach - Predesign

US reach - Design

Meandering scales (Α)

Meandering scales – free? Near Fargo, ND (Red River)

RVR MEANDER (classical approach)

Typical output of linear models

Ikeda et al. (1981)

RVR MEANDER (the classical approach) Classic migraZon-‐coefficient method for migraZon (Ikeda et al. 1981)

The coefficient E0 is usually obtained via calibraSon against historic channel centerlines.

LimitaZons of classic approach based on calibrated migraZon coefficient: q  it predicts a smooth and “conSnuous” migraSon pa:ern; q  linearity of the expression: it does not explicitly account for local, episodic mass failure mechanisms; q  the formulaSon does not account for an erosion threshold; q  it does not consider the effect of the bank geometry either; q  it does not consider the impact of the verScal heterogeneity of the bank materials, horizontal could be incorporated changing Eo

Hickin, E. J. (1974). The development of meanders in natural river channels: American Journal of Science, 274, 414-442

Complex planform pa_erns

RVR Meander: simplified two-‐dimensional (2D) hydrodynamic and migraSon model (Abad and Garcia, 2006), based on migraSon coefficient approach

RVR MEANDER (current approach) The RVR Meander pla_orm merges the funcSonaliSes of the first version of MEANDER (Garcia, Bi:ner, and Nino, 1994) and RVR Meander (Abad and Garcia, 2006) with CONCEPTS (Langendoen and Simon, 2008).

CONCEPTS (CONservaSonal Channel EvoluSon and Pollutant Transport System): one-‐dimensional (1D) hydrodynamic and morphodynamic model (Langendoen and Alonso, 2008; Langendoen and Simon, 2008; Langendoen et al., 2009)

The new plaborm q  is wri:en in C++ language; q  is composed of different libraries (preprocessing, hydrodynamics, bank erosion, migraSon, filtering, plofng, and I/O); q  stand-‐alone for Windows and Linux operaSng systems and interface in ArcGIS-‐ArcMap.

RVR MEANDER (current approach)

Typical output of linear models

Ikeda et al. (1981)

The hydraulic (fluvial) erosion rate E∗, in the horizontal direcSon, is modeled using an excess shear stress relaZon, typical of cohesive material.

Bank erosion

Jet Test for in situ measure of erosion-‐rate coefficient and criZcal shear stress.

Cohesive Strength Meter for in situ measure of criZcal shear stress.

Erosion-‐rate coefficient

CriZcal shear stress


UncertainZes on q  Stress acSng on the bank τ* (from modeling) q  CriScal shear stress τ*c (from field) q  Erosion-‐rate coefficient M* (from field)

Hydraulic erosion rate

Shear stress acZng on bank

Bank materials composed of fine-‐grained cohesive sediments. Bank erosion: combinaSon of fluvial shear erosion and gravitaZonal mass failure processes. Depending on shape of bank profile and physical properSes of the bank materials, any one of the following mass failure mechanisms might be observed

q  planar q  rotaSonal q  canSlever q  piping or sapping

Planar failure Planar failure

RotaZonal failure

CanZlever failure Piping


Pictures courtesy Eddy Langendoen

Mo:a, D., Abad, J.D, Langendoen, E.J., Garcia, M.H., 2011. A simplified 2D model for meander migra:on with physically-‐based bank evolu:on. Geomorphology

[1] RVR MEANDER: MC vs PB

Aerial pictures of the Mackinaw River in the years 1951 and 1988.

Reach of the Mackinaw River in Illinois located in Tazewell County about 15 kilometers upstream of the juncSon of the Mackinaw River with the Illinois River.

Mackinaw River study reach.

flow

flow

The proposed approach showed significant improvements over the classic approach in predicSng the observed migraSon in the period 1951-‐1988, both in terms of

q  shapes q  predicZon error

q Compound loops are captured by PB

Test for natural river

Mackinaw River study reach: historic and predicted centerlines in 1988. Flow is from right to lei (Mo_a et al., 2011).

flow

flow

flow

MC2

MC1

PB

[1] RVR MEANDER: MC vs PB

Mo:a, D., Abad, J.D, Langendoen, E.J., Garcia, M.H., 2012. The effects of floodplain heterogeneity on meander planform shapes. Water Resources Research

[2] HORIZONTAL HETEROGENEITY

Single realizaZon All realizaZons



Single realizaZon All realizaZons

Effects of length scale of floodplain heterogeneity à the larger the scale à more migrated centerline variability



High-‐frequency meander bends are starZng to appear

Single realizaZon Single realizaZon

Vertically homogeneous

Tblock = 50 daysVertically heterogeneous

Station [m]

Elevation[m]

-17 -16 -15 -14 -13

-3

-2

-1

0

1

2

2 layers with different erodibility

Layer 1

Layer 2

Station [m]

Elevation[m]

-17 -16 -15 -14 -13

-3

-2

-1

0

1

2

2 layers with same erodibility

Layer 2

Layer 1

VerZcal heterogeneity affects canZlever failure volumes and therefore migraZon Homogeneous migrates at higher rates

Area of vertical heterogeneity for erodibility

[3] VERTICAL HETEROGENEITY

wse

wse

[3] VERTICAL HETEROGENEITY Mackinaw River

Mo:a, D., Abad, J.D, Langendoen, E.J., Garcia, M.H., 2012. VerScal heterogeneity of the floodplain. Journal of Geophysical Research – Earth Surface (in review)


X - 20MOTTA ET AL.: MEANDER MIGRATION IN HORIZONTALLY AND VERTICALLY HETEROGENEOUS FLOODPLAINS

Figure 1. Examples of slump blocks in rivers characterized by di!erent spatial scales.

(a) Small scale: Trout Creek (California, USA), with small blocks but long residence

time; (b) Intermediate scale: Mackinaw River (Illinois, USA), with larger but less present

blocks; c Large scale: Pilcomayo River (Argentina), blocks are substantially absent.

D R A F T January 22, 2012, 5:31pm D R A F T


x [m]

y [m

]

0 200 400 600 800 1000 1200 1400 1600 1800

0

2006000 7000 8000 9000 10000 11000 12000 13000 14000

Cohesion [Pa]

Tblock

500100

Figure 6. Impact of slump block existence period on meander migration. Comparison

between Tblock = 100 and 500 days.



MOTTA ET AL.: MEANDER MIGRATION IN HORIZONTALLY AND VERTICALLY HETEROGENEOUS FLOODPLAINSX - 25

x [m]

y [m

]0 200 400 600 800 1000 1200 1400 1600 1800

0

2006000 7000 8000 9000 10000 11000 12000 13000 14000

Cohesion [Pa]

Cantilever + PlanarTblock = 100 days Cantilever only

Figure 7. Relative importance of cantilever and planar failure.



Figure 1. Examples of slump blocks in rivers characterized by di!erent spatial scales.

(a) Small scale: Trout Creek (California, USA), with small blocks but long residence

time; (b) Intermediate scale: Mackinaw River (Illinois, USA), with larger but less present

blocks; c Large scale: Pilcomayo River (Argentina), blocks are substantially absent.


Mo:a, D., Abad, J.D, Langendoen, E.J., Garcia, M.H., 2012. VerScal heterogeneity of the floodplain. Journal of Geophysical Research – Earth Surface (in review)

www.rvrmeander.org

SHORT COURSES q  R i v e r C o a s t a l a n d E s t u a r i n e

Morphodynamics, RCEM 2009, Santa Fe, ARGENTINA

q  NaSonal University of Engineering, NaSonal Congress 2010, Lima, Cusco, PERU

q  R i v e r C o a s t a l a n d E s t u a r i n e Morphodynamics, RCEM 2011, Tsinghua University, Beijing, CHINA

q  USDA-‐FOREST – LAKE TAHOE, Dec 2011, Nevada.

q  UNAM-‐Mexico, January 2013

www.rvrmeander.org

THANKS

LinearizaZon

Dimensionless perturbaZons of velocity in the streamwise and transverse direcZon and water depth

Important parameters:

q  Sinuosity q  Half width-‐to-‐depth raSo q  Froude number q  FricSon coefficient

Hydrodynamics and bed morphodynamics (contd.)

Dimensionless transverse velocity perturbaZon

Dimensionless depth perturbaZon

Dimensionless streamwise velocity perturbaZon

Reach-‐averaged values PerturbaZons

RVR MEANDER

The canZlever failure is associated to overhanging slumps of mass generated by preferenSal retreat of highly erodible layers or simply by the erosion of the bank below the water level with respect to its dry porSon. The occurrence of canSlever failure is determined from geometrical consideraSons, once an undercut threshold is exceeded.

Layer i

Layer i+1

Failure surface

Time

Hydraulic erosion

Bank erosion (contd.)

Sketch of canZlever failure.

RVR MEANDER

RVR Meander

GIS interface

Interface overview

Live demonstraSon

Future work

Fluvial erosion controls erosion at the toe

Mass failure occurs in upper porZon of the bank

Bank processes

Medium to long-‐term bank retreat rates are controlled by the process of fluvial erosion at the toe

Bank armoring

Increasing river scale

Bank processes

Mass failure mechanisms may impact migraSon rates and shapes through bank toe protecSon

Trout Creek, USA Mackinaw River, USA Pilcomayo, ArgenZna

River scale may affect residence Sme of slump blocks

Layering impacts mass failure volumes

Impacts bank protecSon

Impacts migraSon rates and shapes

Bank processes Bank layering

VerScal heterogeneity in the floodplain may impact migraSon rates and shapes

MODELACION NUMERICA DEL TRANSPORTE DE SEDIMENTOS ...

Documents

Transcript of MODELACION NUMERICA DEL TRANSPORTE DE SEDIMENTOS ...