Iahr2015 net transport of fine sediments in a narrow, converging estuary, winterwerp, deltares, 29062015

IAHR, June 29, 2015

net transport of fine sediments in a narrow, converging estuary – the Sea Scheldt

Han Winterwerp, Miguel de Lucas Pardo, Bas van Maren, Julia Vroom & Zheng Bing Wang Deltares and Delft University of Technology

the hyper-turid Ems River

Emden

ebb Qriv < 30 m3/s

flood

salinity

suspended sediment

salinity

suspended sediment

Talke et al., 2009 - data Aug 2, 2006

no relation between SPM & salinity

tidal excursion

Herbrum (weir)

historical evolution of the tidal range

summary tidal evolution

0

2

4

6

1875 1900 1925 1950 1975 2000 2025

year

tidal

rang

e [m

]

AntwerpBremenHamburgNantesPapenburg

tidal amplification in Loire (Nantes) and Ems (Papenburg) can be explained for 50% by deepening and narrowing and for 50% by sediment-induced reduction in hydraulic drag

our research question – the stability diagram

00 0.2 0.4 0.6 0.8 1

volumetric concentration

flux

Ric

hard

son

num

ber R

i f

Ricr

super-saturated

sub-saturated

U2

U1

U2 < U1

present Scheldt

conditions

present Ems & Loire hyper-turbid

conditions

pote

ntia

l ene

rgy

/ kin

etic

ene

rgy

how do we get to hyper-turbid state?

our research question

reduced river flushing

too much deepening

tidal amplification

reduction in hydraulic drag

increase tidal asymmetry

pumping of mud

this takes time (decades?)

too big ships

what triggers the increase in the net up-estuary transport of mud, e.g. the “pumping of mud”

net sediment transport processes

net sediment balance SPM over water column

water-bed exchange

∆SPM/∆t = import by estuarine circulation 2DV or 3D not relevant

import (or export) by tidal asymmetry see next slide necessary

export by river flushing depth-averaged not relevant

source by bed/bank erosion depth-averaged -

sink by sedimentation depth-averaged -

transport by tidal advection = gross effect

2DV or 3D: vertical SPM gradients required for net transport vertical gradients are affected by sediment-induced stratification

net transport by tidal asymmetry

we start here, thus we look for balance between: • estuarine circulation • river-induced flushing • tidal asymmetry in vertical mixing and slack water

net transport by tidal asymmetry SPM over water column

water-bed exchange

alluvial bed

(hyper-turbid estuary)

peak flow asymmetry depth-averaged necessary

ebb/flood asymmetry

(settling lag)

depth-averaged necessary

starved bed

(“normal” estuary)

internal asymmetry

(vertical mixing)

2DV or 3D necessary

slack water asymmetry

(scour lag)

depth-averaged necessary

4T U∝

( )d dT

u t

2z Uε ∝

( ) 0d d

uu t

=

asymmetry for starved bed conditions

for starved-bed systems: slack-water asymmetry & asymmetry in mixing

-2

-1

0

1

2

0 5 10 15 20

vert

ical

mix

ing ε z

[m2 /s

]

time t [hrs]

mixing ~U^2

22 -φ4 = ...o

floodebb

22 -φ4 = ...o

flood = 1.26×ebb-2

-1

0

1

2

0 5 10 15 20

velo

city

U [m

/s]

time t [hrs]

U4 = 0.2 U2

22 -φ4 = ...o

ucrit

flood

ebb

HWSLWS

our approach

• idealized, converging estuary with constant depth • dimensions, tidal forcing and river flow from Sea Scheldt river • no intertidal area, hence flood-dominant • implemented in Delft3D • switch on/off various processes one by one (or combinations) • vary boundary conditions (Qriv and external M4 amplitude & phase) • vary internal processes (equation of state, erosion/deposition)

ETM

Bath

Ghent

idealized model of Sea Scheldt (+ small part Western Scheldt)

tide

b.c.

Qriv

3 km

104 km

first results – development of asymmetry

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

0 5 10 15 20 25 30 35 40 45

velo

city

U [m

/s]

time t [hrs]

flow velocities in idealized estuary

sea boundary (x = 0)max sal. intrusion (x = ~18 km)x = 40 km

LWSHWS

mouth x = 0 maximum salinity intrusion x = 40 km

LWS HWS LWS HWS LWS HWS

45 min 24 min 33 min 24 min 24 min 21 min

• profound development of asymmetry in peak velocity, hence vertical mixing: induces up-estuary transport • tLWS < tHWS, hence down-estuary transport, but reducing in up-estuary direction

first results – effect estuarine circulation

simulating 3 months sediment input from sea-side: • no water-bed exchange • no sediment-induced buoyancy destruction

salinity sediment (SPM)

first results – effect buoyancy destruction

with buoyancy destruction without buoyancy destruction

sediment-induced buoyancy destruction = effect sediment-induced stratification on vertical turbulent mixing

maximal up-estuary ETM after 6 months simulation time

cumulative sediment flux through kmr 5

without buoyancy destruction

with buoyancy destruction

discussion and conclusions

• we used Delft3D for studying net transport of fine sediments in schematized estuary, resembling Sea Scheldt,

• this approach seems promising, but many more analyses are required,

• the first results indicate: • ETM can be produced by estuarine circulation alone, • sediment-induced buoyancy destruction keeps fines closer to

the bed, reducing horizontal transports, • the realistic effect of sediment-induced buoyancy destruction is

large, hence cannot be omitted in sediment transport studies, • tidal asymmetry without water-bed exchange does not induce

net transport of fine sediment, • these numerical results are in line with theory, but have never

been quantified.