Analysis of Cost-Effective Rehabilitation: Principles and Tools for Reducing

Uncertainty in Design

Natasha Bankhead and Andrew Simon Cardno ENTRIX, Oxford, MS

natasha.bankhead@cardno.com

Often the Questions are Basic Ones

• Is the bank/bed stable? • How much will it erode over “x” number of years or for a given range of flows? • Can erosion be reduced/stopped using various techniques? • Can we allow for “natural” processes without threats to assets, infrastructure and human safety? • Should we do nothing, use rock, engineered wood structures, flow deflectors, vegetation? • What is the most effective? • Is it affordable?

But My Stream is Different… • The physics of erosion are the same wherever you

are…no matter what hydro-physiographic province, stream type or river style you are in…channel response is a matter of quantifying available force, and resistance of the channel boundary

• Channel adjustment is driven by an imbalance between the driving and resisting forces

• Differences in rates and magnitudes of adjustment, sediment transport rates and ultimate channel forms are a matter of defining those forces…deterministically or empirically

The Processes: Force and Resistance • Hydrologic: The principle driver (expressed as

discharge over time);

• Hydraulic: How discharge is translated to a quantifiable force acting on the channel boundary (expressed as shear stress vs. critical shear stress);

• Geotechnical: Downslope gravitational force (resisted by cohesive and frictional strengths)

The Analytic Approach Static and Dynamic Numerical Modeling

• Collect field data to define the variables that control the processes (force and resistance);

• Use that data with physically-based analytic solutions/models to test plans and designs;

• Use the best available numerical models for prediction

Hydraulic Shear Force and Resistance τo = γw R S τo = mean boundary shear stress γw = unit weight of water R = hydraulic radius = A / 2y + w S = channel gradient

τ* = τo / [ (γs – γw) d] τ* = dimensionless shear stress γs = unit weight of sediment d = characteristic particle diameter

Re = ( U* d) / υ Re = boundary Reynolds number U* = shear velocity = (g R S)0.5

υ = kinematic viscosity (function of water temperature)

= measured in the field

y and x axes of Shields diagram

Collect the Data Hydraulic Resistance (bed and banks): Non-cohesive: bulk particle size or particle count for τc Cohesive: submerged jet-test device

Equivalent to critical shear stress where rate of scour

becomes 0.0 mm/min

Geotechnical: Force vs. Resistance Resisting Forces

Driving Forces If Fs is greater than 1, bank is stable. If Fs is less than 1 bank will fail. (We usually add a safety margin: Fs>1.3 is stable.)

Resisting Forces Driving Forces (gravity)

soil shear strength bank angle matric suction weight of soil mass root reinforcement weight of water in bank confining force bank height

Factor of Safety (Fs) =

Components of Shear Strength (ie. for saturated conditions)

Where: τf = shear strength (kPa); c’ = effective cohesion (kPa); σ = normal load (kPa); µw = pore water pressure (kPa); and φ’ = effective friction angle (degrees)

τf = c’ + (σ-µw) tan φ’

= measured in the field

Collect the Data Geotechnical Resistance (banks):

• Coarse-grained, non-cohesive: particle count for φ’; c’ = 0.0 • Cohesive: borehole shear tester (BST) or laboratory analysis

for c’ and φ’

Default values by soil type are available but increase uncertainty

Differentiate Between Hydraulic and Geotechnical Protection

• Hydraulic protection reduces the available boundary hydraulic shear stress, and increases the shear resistance to particle detachment

Hydraulic Protection

Geotechnical Protection

• Geotechnical protection increases soil shear strength and decreases driving forces

Vegetation as a River Engineer • Above and below-ground biomass • Process Domains • Role of vegetation can be quantified

Process Domain

Geotechnical Hydrologic Hydraulic

Above Ground

Surcharge Interception Evapo- transpiration

Roughness Applied shear stress

Below Ground

Root reinforcement

Infiltration Matric suction

Critical shear stress

Benefits of Process-Based Approach

• Can test design scenarios for performance under a range of flow conditions before they are built; • Can modify designs and re-test; • Avoid under-design and reduce risk of project failure; • Avoid over-design and cost over-runs; • Can provide analysis of effectiveness of different designs; • Can provide analysis of cost-effectiveness in real terms of cost per amount of reduction (ie. $/m3 or $/m)

Analysis as part of the design process has been shown to be cost effective by reducing risk and uncertainty (Niezgoda and Johnson, 2007)

Technology is available and cost effective!!

Bank-Stability Model Version 5.4 • 2-D wedge- and cantilever-failures

• Tension cracks • Search routine for failures • Hydraulic toe erosion • Increased shear in meanders • Accounts for grain roughness • Complex bank geometries • Positive and negative pore-water

pressures • Confining pressure from flow • Layers of different strength • Vegetation effects: RipRoot • Inputs: γs, c’, f’, φb , h, uw, k, τc

Confining pressure

Tensiometers (pore pressure)

shear surface

12/29/97 01/05/98 01/12/98 01/19/98 01/26/98 02/02/98

FACTOR OF SAFETY

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

RIVER STAGE, IN METERS ABOVE

SEA LEVEL

80

81

82

83

84

85

BFactor of safety

Effect of confining pressure

Bank failures

Stage WAT

ER L

EVEL

, M

For Bank Erosion:

Toe Model OutputVerify the bank material and bank and bank-toe protection information entered in the "Bank Material" and "Bank Vegetation and Protection"worksheets. Once you are satisfied that you have completed all necessary inputs, hit the "Run Toe-Erosion Model" button (Center Rightof this page).

Bank Material Bank Toe MaterialLayer 1 Layer 2 Layer 3 Layer 4 Layer 5

Moderate cohesive Moderate cohesive Moderate cohesive Moderate cohesive Moderate cohesive Moderate cohesive Material

5.00 5.00 5.00 5.00 5.00 5.00 Critical shear stress(Pa)

0.045 0.045 0.045 0.045 0.045 0.045 Erodibility Coefficient(cm3/Ns)

Account for:Stream Curvature

Effective stressacting on each grain

Average applied boundary shear stress 51.060 PaMaximum Lateral Retreat 24.966 cm

Eroded Area - Bank 0.232 m2

Eroded Area - Bank Toe 0.423 m2

Eroded Area - Bed 0.000 m2

Eroded Area - Total 0.655 m2

-1.00

0.00

1.00

2.00

3.00

4.00

5.00

6.00

-1.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00STATION (M)

ELE

VA

TIO

N (M

)

Base of layer 1

Base of layer 2

Base of layer 3

Base of layer 4

Base of layer 5

Eroded Profile

Initial Profile

Water Surface

Run Toe-Erosion Model

Export New (Eroded) Profile into Model

Example: The Role of Toe Protection

Slope = 0.0035 m/m

Depth = 2.5 m

Toe material: silt

Eroded: 0.66 m2

Slope = 0.0035 m/m

Depth = 2.5 m

Toe material: wood

Eroded: 0.28 m2

Toe Model OutputVerify the bank material and bank and bank-toe protection information entered in the "Bank Material" and "Bank Vegetation and Protection"worksheets. Once you are satisfied that you have completed all necessary inputs, hit the "Run Toe-Erosion Model" button (Center Rightof this page).

Bank Material Bank Toe MaterialLayer 1 Layer 2 Layer 3 Layer 4 Layer 5

Moderate cohesive Moderate cohesive Moderate cohesive Moderate cohesive Moderate cohesive Rip Rap (D50 0.256 m) Material

5.00 5.00 5.00 5.00 5.00 204.00 Critical shear stress(Pa)

0.045 0.045 0.045 0.045 0.045 0.007 Erodibility Coefficient(cm3/Ns)

Account for:Stream Curvature

Effective stressacting on each grain

Average applied boundary shear stress 51.060 PaMaximum Lateral Retreat 24.159 cm

Eroded Area - Bank 0.232 m2

Eroded Area - Bank Toe 0.044 m2

Eroded Area - Bed 0.000 m2

Eroded Area - Total 0.276 m2

-1.00

0.00

1.00

2.00

3.00

4.00

5.00

6.00

-1.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00STATION (M)

ELE

VA

TIO

N (M

)

Base of layer 1

Base of layer 2

Base of layer 3

Base of layer 4

Base of layer 5

Eroded Profile

Initial Profile

Water Surface

Run Toe-Erosion Model

Export New (Eroded) Profile into Model

Channel-Modeling Capabilities

Process BSTEM HEC-RAS

SRH-2D

HEC+ BSTEM

SRH-2D+BSTEM

Shear in meanders

Bank-toe erosion

Mass-failures

Bed erosion

Sediment transport

Vegetation effects

‘Hard’ engineering

Channel evolution

Rapid Assessments

Alpha version of HEC-RAS/BSTEM is now being tested by CoE, Cardno

ENTRIX and USDA-ARS

0

5000

10000

15000

20000

25000

30000

12/6/1999 4/19/2001 9/1/2002 1/14/2004 5/28/2005 10/10/2006 2/22/2008

DIS

CH

AR

GE

, IN

CU

BIC

F

EE

T P

ER

SE

CO

ND

DATE

June 1, 2000 - June 30, 2008

Sandy River, OR

Flow series for calibration

Objectives: 1. Predict migration of meander bend

and timing of threats to park infrastructure;

2. Quantify reduction in retreat rates under various mitigation measures

In cooperation with Brian Vaughn, Portland METRO

50-Year Simulation Results

y

Infrastructure Type (points) Year Impacted

2011 2022 2037 2047 2062

Play Structures (1 Point)

Play Structure #1 X

Play Structure #2

Group Picnic and Campgrounds (2 Points)

Group Picnic A X

Group Picnic B

Group Picnic C X

Group Picnic D X

Group Picnic E

Group Camp #1

Group Camp #2 X

Restrooms and Showers (3 points)

2 Door Restroom #1

X

2 Door Restroom #2 X

4 Door Restroom #3 X

4 Door Restroom #4

2 Door Restroom #5 X

4-Door Showers #1

4-Door Showers #2 X

Roads (4 points)

Main Road X X X X X

Boat Ramp X X X X X

Essential Infrastructure (5 points)

Sewer System X

Total Score

Cumulative Relocation Cost Score 11 20 16 19 14

Timing of Associated Threats to

Park Infrastructure

Modeled Conditions High GW case Fs (GW 9.31 m above

bed)

Bank top retreat

(m)

Geotechnical erosion

(m3)

Existing (with stream curvature) 0.99 10.7 97.6

Grade to 1:1 (with stream curvature) 1.53 0 0

RipRap to 15% bank height (with stream curvature) 1.07 0 0

Grade + RipRap (with stream curvature) 2.18 0 0

Flow deflection (stream curvature turned off) 1.49 0 0

Flow deflection + riprap (stream curvature off) 1.50 0 0

Flow deflection + grade (stream curvature off) 1.67 0 0

Vegetation (with stream curvature) 1.00 10.7 97.6

Vegetation + grade (with stream curvature) 1.98 0 0

Vegetation + riprap (with stream curvature) 1.23 0 0

Vegetation + flow deflection (stream curvature off) 1.49 0 0

Testing Alternative Treatments

A range of alternative treatments would be successful here. Ultimate decision is a function of cost, permitting and policy

Bayou Pierre, MS Encroachment on

Gas Pipeline

Simulated Retreat: 30 Years (Calibrated with 2-D Hydraulic Modeling)

0

1

2

3

4

5

6

7

8

-30 -25 -20 -15 -10 -5 0 5 10 15

EL

EVA

TIO

N A

BO

VE

AR

BIT

RA

RY

DA

TUM

, IN

M

ET

ER

S

DISTANCE FROM BANK EDGE, IN METERS

Original Geometry5 years10 years15 years20 years25 years30 years

0

1

2

3

4

5

6

7

-250 -200 -150 -100 -50 0

EL

EVA

TIO

N A

BO

VE

AR

BIT

RA

RY

DA

TUM

, IN

M

ET

ER

S

DISTANCE FROM BANK EDGE, IN METERS

Original Geometry5 years10 years15 years20 years25 years30 years

Spoke 2-4

Spoke 2-6

Predicting Bend Migration

E = Existing pipeline; A, A/B, B and D represent alternative routes

Burnett River, Queensland, AUS

January 2013 flood of record: RI at least 140 years

Objectives: 1. Protect local assets

from bank erosion; 2. Determine cost-

effective treatments 3. Reduce fine loads to

the Great Barrier Reef

Bank erosion: 48 Mt over 3.8 years!!

Calibration: 2010-2013

Once calibrated, 10-year simulations for “no action” and alternative strategies were conducted

Erosion under Alternative Treatments

0

5

10

15

20

25

-80 -30 20 70 120

ELEV

ATIO

N, i

n m

eter

s

STATION, in meters

2013 START PROFILEEND NO ACTIONEND with vegetated bankEND with rock toeEND with rock toe and bank vegetationEND rock toe and bankEND rock toe and bank, and vegetationEND vegetation and bendway weirEND vegetation, bendway weir, rock toe and bank

0

5

10

15

20

25

-80 -30 20 70 120

ELEV

ATIO

N, i

n m

eter

s

STATION, in meters

END NO ACTIONSTART 32 degree bankSTART 2:1 bankEND 2:1 bank with vegetationEND 32 degree bank with vegetation

10-year flow series

0.001

0.010

0.100

1.000

10.000

100.000

1000.000

10000.000

100000.000

DIS

CHA

RGE,

IN

CU

BICM

ETER

S PE

R SE

CON

D

BMRG-02

“No action”

Includes 2011 and 2013 floods

Estimate Costs (similar relations for wood structures; heavy machinery;

plantings, etc.)

0

2

4

6

8

10

12

14

0 20 40 60 80 100 120 140

VERT

ICAL

HEI

GHT

OF

ROCK

, IN

M

ETER

S

TONNES OF ROCK REQUIRED

Horizontal placementVertical placement

$$ = 259.93 h 1.9528

r² = 0.9985

$$ = 187.91 h 1.4165

r² = 0.999

100

1000

10000

100000

0 5 10 15

COST

PER

MET

ER O

F RO

CK

PLAC

EMEN

T*

VERTICAL HEIGHT (h) OF ROCK, IN METERS

Bendway weirs

Rock toe protection

From D. Derrick, written comm.,

(2013)

Assuming $81.50/t (AUS)

Scenario Protection Unit Cost

Cost effectiveness Unit Cost Total Cost

Height Length By volume By retreat (m) (m) ($/m) ($/m3) ($/m) ($)

No Action - - $0 $0 $0 $0 With vegetation 13.2 530 $ 66 $81 $1,230 $35,000 Rock at toe 4.1 530 $ 1,360 $988 $15,200 $722,000 Rock toe with vegetation 4.1 530 $ 1,430 $1,020 $15,600 $757,000 Rock toe and bank face 11.1 530 $ 5,680 $3,890 $61,600 $3,012,000 Rock toe and lower bank with vegetation 6.6 530 $ 2,790 $1,930 $30,200 $1,480,000 2:1 battering with vegetation 13.2 530 $ 186 $259 $2,440 $98,600 32 degree bank with vegetation 13.2 530 $ 186 $308 $2,710 $98,600 Vegetation with bendway weirs 13.2 530 $ 1,560 $1,080 $2,700 $824,000 Vegetation with ELJs 13.2 530 $ 260 $181 $30,300 $138,000 Vegetation with bendway weir and rock toe 13.2 530 $ 2,920 $1,990 $30,300 $1,550,000 Vegetation with ELJs and rock toe 4.1 530 $ 1,620 $1,110 $16,800 $860,000

Cost Effectiveness (to hold the line or not)

Cost Effectiveness

Summary and Conclusions •Gravity and the physics of erosion and sediment transport are a constant, allowing us to quantify force and resistance mechanisms.

•We can collect all of the data we need at a site/reach to quantify driving and resisting forces operating on the channel boundary

•Using principles of hydrology, hydraulics and geotechnics we can use our field data to inform numerical tools and models.

• Analyses and models can then be used to quantify performance and effectiveness of a range of alternative strategies

• By applying a cost basis to each strategy, we can provide a range of cost-effective options to be considered.