new features for hec-ras 5 - cdn.ymaws.com · integrating hec-ras sediment transport capabilities...

New Features for HEC-RAS 5.0

Martin Teal, P.E., P.H., D.WRE

August 22, 2017

New Sediment Features in HEC-RAS 5.0

(and 5.0.1/2/3/x)

Thanks to Stanford Gibson, HEC

Topics

1. Unsteady Sediment Transport

2. USDA-ARS Bank Stability and ToeErosion Model (BSTEM)

3. Other New Sediment Features

Develop and Calibrate a Good Hydraulic Model First

-Variable n-values-Ineffective Flow Areas

HEC-RAS Sediment Data InputStarts here:

New Sediment Features in HEC-RAS 5.0.3




Quasi- vs Unsteady Sediment Files

Q

t

Quasi-Unsteady Flow

QQ

tt

Quasi-Unsteady Flow Unsteady Flow

Case Study:Spencer Dam

Spencer Dam Flush

Gibson, S. and Boyd, P. (2016) “Monitoring, Measuring, and Modeling a Reservoir Flush on the Niobrara River in the Sandhills of Nebraska,” Proceedings, River Flow 2016, ed Constantinescu et al., 1448-1455.

Gibson, S. and Boyd, P. (2016) “Designing Reservoir Sediment Management Alternatives with Automated Concentration Constraints in a 1D Sediment Model,” Proceedings International Symposium on River Sedimentation.

.

Spencer HEC-RAS Model

400 600 800 1000 1200 1400 1600

1485

1490

1495

1500

1505

1510

Station (ft)

Elev

atio

n (ft

)

32

D/S Concentration Threshold

HEC-RAS Model Concentration

Mississippi River Unsteady HEC-RAS Sediment Model (HEC)

Conowingo Dam Studies

Lower Susquehanna River Reservoirs Dynamic equilibrium Potential impacts on

Chesapeake Bay water quality

20

Figure 1-2 from Lower Susquehanna River Watershed Assessment (2015)

Introduction and Background• Safe Harbor Dam Constructed 1931 75 feet tall, impounds Lake Clarke (7,360 acres)

• Holtwood Dam Constructed 1910 55 feet tall, impounds Lake Aldred (2,400 acres)

21

Safe Harbor Dam, October 24, 2010 (U.S.C.G.Aux5DNR on Flickr) Holtwood Dam, December 2015 (http://suereno.blogspot.com)

Study Goals• Develop unsteady, 1D sediment transport model Marietta, PA to Holtwood Dam (20.4 miles)

• Aid in parameterization of Chesapeake Bay Watershed Model

• Input for HDR model of Conowingo Pond

22

Study Goals – Model Improvements• Improve upon past studies using HEC-RAS 5.0 Model (unsteady) Solves the unsteady flow equation, routing flow and

explicitly accounting for storage and travel time Conserves volume, important in reservoir systems Cohesive and non-cohesive sediments

23

Model Boundary Conditions and SetupGeometric DataFlow &

Temperature DataSediment Data

Bed sediment (12 size classes)

Inflow loads

Transport FunctionSediment

ParametersDam Operations

Rating curves

24

Safe Harbor Dam

Holtwood Dam

• Sediment Volume Change 2008-2013

(calibration) 2013-2015

(verification)

Model Calibration and VerificationHydraulic Calibration at Marietta, PA

25

20 31 10 20 28 10 20 31Jan2010 Feb2010 Mar2010

230

235

240

245

250

0

100000

200000

300000

400000

500000

Plan: Gate2-T River: SUSQ Reach: LSUS RS: 187225.3

Time

Stag

e (ft

)

Flow

(cfs)

Legend

Stage

Obs Stage

Flow

Missing Data

28 10 20 31 10 20 30 10 20Mar2011 Apr2011 May2011

230

235

240

245

250

0

100000

200000

300000

400000

500000

Plan: Gate2-T River: SUSQ Reach: LSUS RS: 187225.3

Time

Stag

e (ft

)

Flow

(cfs)

Legend

Stage

Obs Stage

Flow

Missing Data

Model Calibration and Verification (2)Sediment

Transport Calibration January 2008 –

August 2013 Several large

events, including Tropical Storm Lee

Sediment Transport Verification August 2013 -

October 2015 Unusually low flows

26

Observed vs. Modeled Cumulative Volume Change: Lake Clarke Sub-Areas 1-5, January 2008 – October 2015

Observed vs. Modeled Cumulative Volume Change: Lake Aldred Sub-Areas 3-6, January 2008 – October 2015

Summary of Results

Additional Runs Historical flow events applied to end of simulation

Results

27

Increased transport on rising limb

Decreased transport on falling limb

Lateral Structures and Diversions

Sedi

men

t Dive

rted

(tons

)

Lateral Structures and Diversions

Mixed Flow Regime Sediment Model

Collaborators: Eddy Langendoen, PHD – USDA-ARSAndrew Simon, PhD – Cardno-Entrix

USDA-ARS Bank Stability and Toe Erosion ModelHEC-RAS/BSTEM

Three Interacting Processes

1. Vertical Bed Adjustment

2. Geotechnical Bank Failure

3. Toe Scour

∆z ~ Qs - Gs

∆x ~ M(τ-τC)

FS ~ FD - FR

-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)

ELEV

ATIO

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

Toe Erosion Model

Bank StabilityModel

-1.00

0.00

1.00

2.00

3.00

4.00

5.00

6.00

-10.00 -5.00 0.00 5.00 10.00 15.00

STATION (M)

ELEV

ATIO

N (M

)

bank profile

base of layer 1

base of layer 2

base of layer 3

base of layer 4

base of layer 5

failure plane

water surface

water table

USDA-ARS Bank Stability and Toe Erosion Model (BSTEM)

USDA-ARS BSTEM Interface

Three Methods:1. Pre-Defined Default Parameters2 .Single Set of User Defined Material3. Layers of Unique Material at a Bank

Defining Bank Material

Clark and Wynn (2007)

Soil Properties

(a) Incision

(b) ToeScour

(c) BankFailure

Interaction of all three process

http://www.eng.buffalo.edu/glp/events/summer2008/week1/full/8-GoodwinCreekDesign.pdf

Images by Andrew Simon (CardnoEntrix)

1996 2006

GoodwinCreek

Goodwin Creek HEC-RAS/BSTEM Model

Goodwin Creek HEC-RAS/BSTEM Model

Gibson, S., Simon, A., Langendoen, E., Bankhead, N., and Shelley, J. (2015) “A Physically-Based Channel-Modeling Framework Integrating HEC-RAS Sediment Transport Capabilities and the USDA-ARS Bank-Stability and Toe-Erosion Model (BSTEM),” Federal Interagency Sediment Conference, SedHyd Proceedings.

Missouri River HEC-RAS BSTEM Model

Modeled and Target Longitudinal Cumulative Mass Change over 1960 to 2011 Calibration Period.

Calibration Reach

RM 773.76

Bank Protrusion

2007 2015

RM 773.76

RM 794.25

RM 766.13

Reservoir Deposition Method

Initial Bed

Veneer Method

Initial Bed

ReservoirMethod

Reservoir Deposition Method

950 1000 1050 1100

1070

1080

1090

Station (ft)

Ele

vatio

n (ft

)

25000 30000 35000

1040

1050

1060

1070

1080

1090

Main Channel Distance (ft)

Ch

Inve

rt E

l (ft)

HEC-RAS Results (profile over time)

HEC-RAS XS Results (Reservoir Sediment Erosion)

Springville Dam Removal ModelBy Waleska Echevarria(ERDC-CHL)

Simplified Channel

Evolution Model

Multiple Movable Bank Limits

Another Veneer Method Departure:Floodplain Deposition

Floodplain Deposits on the Missouri River from the

2011 Flood

T. Branß, A. Dittrich, and F. Núñez-González(Technische Universität Braunschweig)

Reproducing natural levee formation in an experimental flume River Flow 2016

0.7 m0.6 m0 m 2 m

z

y0.0 cm

10.4 cm

1:1D

hm

ydymax

Floodplain Deposition Options in 5.1

Limerinos Brownlie

Bed Roughness Predictors

Bed Roughness Predictors

Bed Roughness Predictors:Reach Average

Computing an n-value at each XS based on gradational and hydraulic parameters can be noisy and prone to negative feedbacks. Therefore, cross sections can be grouped together into reaches to compute a single n-value from reach weighted parameters.

Navigation Dredging

Substrate Mining

Copeland Mixing Method (Exner 7)

Major differences between Exner 5 and Exner 7

Thomas (Exner 5) Copeland (Exner 7)Active Layer (La) computation based on Equilibrium Depth (Deq).

Active Layer (La) computation based on 15% of Water Depth or 3d90.

Linear Armor Ratio. Polynomial Armor Ratio.

Armors more rapidly, ↓ erosion. Armors more gradually, ↑ erosion.

“The arbitrary maximum cover layer thickness of 2 ft may hinder deposition during low energy conditions. Mixing of fine material will probably result in underestimation of scour during high flows. Erosion of fine material may be too severely constrained by (cover formation assumptions). “ (HEC 1996 HEC 6 Manual, p 28)

Questions?

Holtwood Dam, September 10, 2011 (TheGates1210 on YouTube)

new features for hec-ras 5 - cdn.ymaws.com · integrating hec-ras sediment transport capabilities...

Documents