Importance of the watershed sediment transport system

choppedspleenΜηχανική

22 Φεβ 2014 (πριν από 3 χρόνια και 8 μήνες)

74 εμφανίσεις

Sediment Management

for River Restoration

Tom Dunne


Spring 2011

Readings for week 4


Slaymaker, O., The sediment budget as conceptual framework and
management tool,
Hydrobiologia
, v 494, pp 71
-
82.







Walling, D. E., The role of overbank floodplain sedimentation in
catchment contaminant budgets,
Hydrobiologia
, v 494, pp. 83
-
91



Brooks, A.P. and G. Brierley, Framing realistic river rehabilitation
targets in light of altered sediment supply and transport
relationships: lessons from East Gippsland, Australia,
Geomorphology, v 58, pp 107

123, 2004.



Nelson, E.J. and D. B. Booth, Sediment sources in an urbanizing,
mixed land
-
use watershed,
Journal of Hydrology
, 264, 51

68, 2002.


Managing the role of sediment supply in river restoration
is more difficult than managing the effects of flow


Flow

Sediment supply

Response of river or
ecosystem

Response of river or
ecosystem

Sediment supply from the watershed is more
difficult to predict than flow


Total amounts


Grain sizes


washload; bed
-
material load


Timing


strongly stochastic

Sediment Management


Extent and nature of sediment management depends
on


the problem to be solved


whether the project is conceived as ‘river channel restoration’
or ‘watershed restoration.’


May involve


Stabilization of sediment sources on hillslopes, etc.


Channel bank stabilization


Dredging/gravel harvest from channel


Gravel augmentation


Removal of dams and disposition of sediment (release; slurry
pipe; dredge, truck and stabilize)


Analyses ‘required’ for sediment management


Stabilization of sediment sources on hillslopes, etc.


Requires sediment budget analysis for watershed and design of erosion
controls


Channel bank stabilization


Geotechnical analysis of slope stability; hydraulic analysis of scour
velocities, and empirical analysis of channel shifting record


Dredging or gravel harvest from channel


Modeling difficult because of highly disturbed conditions, but estimation
of bed
-
material supply is necessary


Gravel augmentation (
increasing
sediment supply)


Should require transport calculation of mobile volumes and particle
sizes, but rarely done. Often determined by $$$ available for purchasing
gravel and activism of interest groups.


Removal of dams and disposition of sediment (release, slurry pipe,
dredge and truck)


Requires extensive sediment transport modeling (Matilija dam studies)

Importance of the watershed sediment
transport system


Sustainable functioning of the habitats in a river channel
-
floodplain
depends on the functioning of the sediment transport system from
the watershed through the channel reach(es) of interest


Planning and designing a restoration scheme depends on
understanding and adjusting the design to this sediment transport
system


Anticipation/prediction of responses of the sediment transport
system is one of the most sensitive and difficult components of
restoration design.


There may be harder problems, such as triggering of some
biogeochemical process
---

methylation of Hg or acidification of S/As
bearing sediments
---
, or a social process
----

failing to meet social
expectations. But sediment management is the hardest one that I know
anything about.

Importance of the watershed sediment
transport system


Arises because the morphology and functioning of the
channel/floodplain result from erosion, deposition, and transport
conditions (e.g. turbidity, bed texture and mobility).


Our prediction skill is limited.


Best evidence of future states is empirical/historical, augmented by
judicious use of theory and calculation.


Users must be made aware of the uncertainties in these predictions.


Made more difficult by the fact that the transport system may need to
be analyzed on a range of temporal and spatial scales, e.g.


channel or dike integrity on reach length/flood event


watershed
-
scale and timber
-
harvest cycles, colonization waves, etc.


watershed scale and deglaciation history


Upland zone: High sediment

supply and low storage.



Alluvial transport zone: sediment transport
rate
≈ sediment supply rate. Significant
transient sediment storage in valley floors
and tributary fans. Multi
-
threaded channels
in upper, steeper reaches; single
-
thread,
meandering channels on lower gradients.
‘Free’ alluvial landforms
.



Alluvial accumulation zone: sediment
transport capacity decreasing downstream;
floodplain aggrading.


Outlet: fans; deltas, estuaries
.

Length scale Amazon to Atascadero. Depends on plate tectonics,…. Again and always! [ESM 203]

Identification of differential transport
behavior of grain size classes

Total sediment supply

Bed material

(sediment
on channel bed/bars)

Washload

(grain sizes present in
sediment supply
---

e.g. soil


but
not in bed material)

Form channel features, aquatic
habitat, reduce flood capacity,
require dredging, etc
.

Cause turbidity, transport
contaminants, deposit
downstream

Sediment Transport Mechanisms in River Channels:


Washload and Bed material (
≈ bedload)

Bed material is exchanged with and
stored in point bars and mid
-
channel bars,

N. Fork Snoqualmie R., WA

Selective sediment transport


Particles traced by distinctive lithology, painting, radio transmitters,
magnets


Average annual transport rates:


Gravel 50

500 m/yr


Sand 100
-
10,000 m/yr


Silt
-
clay many km, or into floodplain, where it stays for a long time

Bedload

Suspendible bed
material load

Washload

Bed
-
material
load

Washload

Suspended
bed
-
material
load



Traction bedload




Suspended load

(DH
-
48)

Bedload

(Helley
-
Smith)

Transport mechanisms depend on grain size at a
particular flow rate

Observable difference of transport
state

Definition based on
measurement technologies

Transport
mechanisms

Toutle River
-
2 and Helley
-
Smith bedload
samplers

Graham Matthews and
Associates

Stochastic nature of
basin
-
scale
sediment
mobilization


The range of morphology
and behavior of a river each
depends on the power
spectrum of sediment
supply (texture and rates)
and the sediment transport
capacity of the reach
(=
f
(Q,s).


Vegetation
Climate
Fire
Storms
Topography
Soil Geotechnical
Forest Age
Slope
P
P
P
P
P
P
Climate
P
P
V
egetation
=
P
Sediment Flux
T
opography
+
+
(a)
(b)
Benda et al., Bioscience, 2004

Supply of sediment from
watershed and its storage in
channel reaches are stochastic
processes, modulated by
drainage area for a given set of
environmental conditions.


Modeling of this kind is useful for
making ‘qualitative’ interpretations of
sedimentation risk, spatial variability of
relative risk in various parts of a
catchment, but
not

for channel design.

0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0
0.2
0.4
0.6
0.8
1
Mean Depth (m)
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
0.0
0.5
1.0
1.5
0
500
1000
1500
2000
2500
3000
Year
Mean Depth (m)
0.0
0.5
1.0
1.5
0
500
1000
1500
2000
2500
3000
Year
Mean Depth (m)
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0
0.2
0.4
0.6
0.8
1
Mean Depth (m)
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0
0.2
0.4
0.6
0.8
1
Mean Depth (m)
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
Frequency
Cumulative
0.0
0.5
1.0
1.5
0
500
1000
1500
2000
2500
3000
Year
Mean Depth (m)
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0
0.2
0.4
0.6
0.8
1
Mean Depth (m)
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
180 km
2
45 km
2
25 km
2
0.0
0.5
1.0
1.5
2.0
0
500
1000
1500
2000
2500
3000
Year
Mean Depth (m)
3 km
2
TIME
FIRE SIZE (sq. km)
0
100
200
0
100
200
300
400
500
TIME
PRECIPITATION (mm/storm)
Rainstorms
Fires
0
100
200
300
400
500
100
200
(a)
(b)
(c)
(d)
(e)
Benda et al., Bioscience, 2004

Some tributary junctions are preferred sites of sediment
accumulation and spatial and temporal habitat complexity

















































































Interference
Mixing
(a)
(b)
(c)
(d)
Tributary Input
time
time
Mainstem Input
time
time
Sediment
Flux
Sediment
Flux
Sediment
Flux
Sediment
Flux
Benda et al., Bioscience, 2004

Analyses ‘required’ for sediment management


Stabilization of sediment sources on hillslopes, etc.


Requires sediment budget analysis for watershed and design of erosion
controls


Channel bank stabilization


Geotechnical analysis of slope stability; hydraulic analysis of scour
velocities, and empirical analysis of channel shifting record


Dredging/gravel harvest from channel


Modeling difficult because of highly disturbed conditions, but estimation
of bed
-
material supply is necessary


Gravel augmentation


Should require transport calculation of mobile volumes and particle
sizes, but rarely done. Often determined by $$$ available for purchasing
gravel and activism of interest groups.


Removal of dams and disposition of sediment (release, slurry pipe,
dredge and truck)


Requires extensive sediment transport modeling (Matilija dam studies)

Dam Removal


U.S. National Inventory of Dams: 76,000
>6 ft high; ~2 million smaller structures


Peak construction 1960; > 3000 built


748 removed so far; 306 since 1999


2008
---

64; 2009
---

58

Sometimes enormous payoff

Elwha Dam (30 m tall) and Glynes canyon dam upstream (70 m
tall) scheduled for removal 2012 in Washington State

Open up 110 km of prime salmon habitat

Nat. Park Service

But $308 million
---

$3 million/km; $100/fish (forever)

For the smaller Elwha Dam the first part of the removal project is the construction of a diversion channel, which will lower
the

level of Lake Aldwell by 50 feet. Once the level of the reservoir is lowered in this manner, the construction team will be ab
le
to
remove the Elwha Dam by controlled blasting.

However, the 210
-
foot Glines Canyon Dam, which is double the size of the 105
-
foot Elwha Dam, will require additional
measures to account for the larger quantities of water and sediment in Lake Mills. First, the reservoir level will gradually
be
drawn down using an outlet pipe to move water downstream. As the water level drops, demolition crews will cut and remove
7.5
-
foot sections of the dam starting from the top. These concrete blocks will be transported offsite by truck and recycled.
Finally, once the sediments behind the dam are reached, controlled blasting will be used to clear the remainder of the dam.

The removal strategies are designed carefully to minimize negative effects (such as flooding and decreased water quality from

sediment releases), but the precautions cannot completely eliminate these effects. Most of the effects are expected to be
temporary, but some infrastructure changes will need to be implemented prior to dam removal to ensure that the water supply
remains safe and steady. For example, a water filtration system will be constructed down river in order to maintain the suppl
y o
f
water to the city of Port Angeles.

Shortly after dam removal, salmon species from the Pacific Ocean will begin to recolonize gradually over 70 miles of habitat
that was not accessible to spawning salmon when the dams were in place. Much of this habitat is within the bounds of Olympic
National Park and is in excellent condition. It is estimated that within 30 years, the river will produce 390,000 salmon and
steelhead each year.

After the dams are removed, the Elwha River will regain its natural form and vast areas of land that were covered by the
reservoirs will be devoid of any vegetation. These riverbank areas will be quickly planted with native plants to begin
revegetation. However, it will take decades to restore portions of that land to the forested landscape that existed prior to
the

dams.

Some immediate effects will be negative, although they are expected to be short term. For example, the sediments that were
bound behind the dam will begin to migrate down river and eventually out to sea. It is expected this process will take place
within 5 years. When the sediment is released initially, it may kill fish and other species in the river and decrease water q
ual
ity.
Additional restoration actions downstream will help to minimize the impacts from this enormous load of sediment. However, in
the long term, the return of the natural flow of sediment will have many positive effects: improving spawning habitat, buildi
ng
up
the nearshore area, and reducing the need to build up Ediz Hook artificially.


Local example, N. Fork Matilija Creek on Ventura R.

Matilija Coalition

Most removal opportunities are small and ‘simple’

Marmot Dam, Sandy R., Oregon

Clemente Dam, Carmel R.

Marmot Dam, Sandy R., Oregon: before, during, after
.

750,000 m
3

of sand and gravel

O’Connor
et al
(2008)
Geotimes

Some are reeeeally complicated

Klamath


4 private dams Ice Harbor, Lower Snake

(agreement ‘reached’)
---

public dam, under discussion

Matilija Dam: 6 millions cu. yd. of sediment to be
managed

Matilija Dam:
physical

components of the plan

Sediment Management ( recent incarnation, but no final agreement yet)

Slurry Disposal
-

2 million cubic yards of fine sediment downstream of
Robles Diversion

Stabilize remaining 4 million cubic yards of sediment (from silt
-
clay to
large boulders) behind the dam

Desilting basin
-

construct a sediment basin along the Robles Water
Diversion Canal

Levees

Meiners Oaks
-

new levee downstream of Robles Diversion

Live Oak levee reconstruction (just upstream of Santa Ana Blvd bridge)

Bridges

Camino Cielo bridge

Santa Ana bridge

Water Supply

(Robles Diversion) Modifications
-

High flow bypass will flush sediment
past diversion

Wells
-

drill 2 new water wells at Foster Park for the City of Ventura.

Arundo (giant reed) Management



At reservoir site

http://www.matilijadam.org/documents/Slurry_Disposal_Studies/upstream%
20storage%20011410.pps

Lake Casitas flow diversion bypass structure

Flood levee construction

http://www.matilijadam.org/documents/Slurry_Disposal_Studies/upstream%
20storage%20011410.pps

Historical channel alignments, Cottonwood Creek, CA

1855
-
1999

Graham Matthews and Associates

http://www.gmahydrology.com/services.html