Upcoming Changes for EM 1110-2-1913 Di C i dE l i fDesign ...€¦ · Di C i dE l i fDesign,...

transcript

Upcoming Changes for EM 1110-2-1913 D i C i d E l i fDesign, Construction, and Evaluation of LeveesScott E. Shewbridge, PhD, PE, GENational Technical Specialist – Risk & Geotechnical Engineeringp g gRisk Management Center, DenverMichael Navin, PhD, PE US Army Corps of Engineers, St. Louis DistrictNeil Schwanz, PE ,US Army Corps of Engineers, St. Paul DistrictNoah Vroman, PE US Army Corps of Engineers, MVD Dam and Levee Safety Production Center, Vicksburgy g

USSD Fall 2015 Levee Workshop

November 3, 2015

ObjectiveObjective

To provide an overview of the updatesTo provide an overview of the updates being proposed for EM 1110-2-1913 Design Construction and Evaluation ofDesign, Construction, and Evaluation of Levees

ContentsContents

Introduction and overview of someIntroduction and overview of some regional design differences Review of Chapters Review of Chapters Questions and Discussion

Complaint About Old EMComplaint About Old EM

“It’s great if you are from the MississippiIt s great if you are from the Mississippi Rivers and Tributaries Project, but th ’ thi b t th t f lthere’s nothing about the types of levee designs we do here in _____________.”

Primary ObjectivesPrimary ObjectivesTo update the USACE EM for Design,

Construction and Evaluation of Levees►Compile information on regional design►Compile information on regional design

approaches and performance expectations including operations, maintenance, flood-fighting efforts and associated documentation

►Use risk-based potential failure modes to establish design objectives and approaches

►Assess adequacy of traditional design with risk l i th danalysis methods

EM Revision TeamEM Revision TeamEM Revision PDT

1 Scott Shewbridge RMC2 Noah Vroman MVD - Mississippi Valley Division MVK - Vicksburg3 Ken Klaus Emeritus ERDC3 Ken Klaus - Emeritus ERDC4 Mike Navin MVD - Mississippi Valley Division MVS - St. Louis5 Neil Schwanz MVD - Mississippi Valley Division MVP - St. Paul6 Tom Brandon – Adjunct to ERDC Virginia Tech

Regional Representatives “EM Collaborators”Regional Levee Design PFMA Session / ATR

EM Collaborators / Regional Levee Design Contributors1 Pat Conroy Geotech MVD - Mississippi Valley Division - St. Louis2 Doug Chitwood, PE GE Geotech SPD - South Pacific Division - Los Angeles3 Glen M Bellew, PE Geotech NWD - Northwestern Division - Kansas City4 Derek Morely PE, Geotech SPD - South Pacific Division - Sacramento4 Derek Morely PE, Geotech SPD South Pacific Division Sacramento5 Christina Neutz Geotech LRD - Great Lakes and Ohio River Division - Louisville6 Tom Mack Chief Geotech MVD - Mississippi Valley Division - Rock Island7 Mark Woodward Geotech MVD - Mississippi Valley Division - New Orleans8 James Snyder Geotech NAD - North Atlantic Division - Baltimore

Example Case HistoriesMississippi Valley Division – New Orleans

Marchand Levee Failure 1983

Flow Slides in SandFlow Slides in Sand

ACM provides the best level of protection along the rivers in the ACM provides the best level of protection along the rivers in the New Orleans District’s jurisdiction due to the depth of water and New Orleans District’s jurisdiction due to the depth of water and construction issues with using other materials in those depths. construction issues with using other materials in those depths.

Hurricane Protection – StabilityWith Reinforcement FabricWith Reinforcement Fabric

Geotextile Installation

Example Case HistoriesMississippi Valley Division – Rock Island

Design of Sand LeveesDesign of Sand LeveesDesign Philosophy

• The through seepage control must prevent the slope failure and excessive erosion of the landside slope.

• The under seepage design criteria must control the hydraulic gradients landward of the levee to prevent piping and excessive uplift pressures on the landsidepiping and excessive uplift pressures on the landside impervious stratum.

• Stability usually not an issueStability usually not an issue

Drury and Iowa River/Flint Creek Test S ti 1962 d 1964Sections 1962 and 1964

1616/81

Results of Test Section: Typical Sand Levee Cross Section

4:1 Riverside slope 10 foot crown 5:1 Land side slope

► Steeper slopes resulted in erosion due to through-seepageseepage.

10 h base width (The width at the base of the levee is 10 time the height of the levee.)► On heights greater than 10 feet a berm is required

• 3 feet thick usually 20 feet wide. • Design based on seepage and stability analysis (EM 1913)g p g y y ( 9 3)

Example Case HistoriesNorthwestern Division – Kansas City

2011Missouri River Performance History

Example Case HistoriesNorth Atlantic Division – Baltimore

Typical Levee Section( i h MSE W ll)(with MSE Wall)

NAB – Scranton22/81

Example Case HistoriesSouth Pacific Division – Los Angeles

Site GeologySite Geology

Levee Embankment – Soil CementLevee Embankment Soil Cement

Soil Cement

Original Levee

ChannelTypical Soil Cement Grade

ypEmbankment Section

Soil Cement Original Levee

ChannelGrade

Landside Slope

Example Case HistoriesSouth Pacific Division – Sacramento

Design Concept: Deep CutoffDesign Concept: Deep Cutoff Degrade ½ levee heightg g Extend cutoff wall down

into low perm layer Regrade with clay corelevee

blanket

high perm layer

low permlow perm layer

Example Case HistoriespGood Coverage for Entire Country

EM Chapters and AppendicesEM Chapters and Appendices Chapter 1. Introduction

Ch t 2 L F il M d d Ri k B d Appendix A. References Appendix B Typical Regional Designs and Typical Chapter 2. Levee Failure Modes and Risk-Based

Decisions During Design and Construction Chapter 3. Field Investigations for Levees Chapter 4. Laboratory Testing for Levees

Ch t 5 B A

Appendix B. Typical Regional Designs and Typical Design Documentation Report Contents

Appendix C. An Overview of Probabilistic Analysis for Geotechnical Engineering Problems –

Methods to Assess of Levee ReliabilityA di D S b f I t t ti G hi Chapter 5. Borrow Areas

Chapter 6. Subsurface Interpretation Chapter 7. Seepage Evaluation and Control Chapter 8. Slope Design

Appendix D. Subsurface Interpretation Graphics Appendix E. Levee Seepage Analysis using Blanket

Theory and Finite Element Methods Appendix F. Analysis and Design of Seepage

Berms Chapter 9. Settlement Chapter 10. Erosion Control Chapter 11. Design and Bid Packages Chapter 12. Levee Construction

Appendix G. Toe Drain Analysis and Design Appendix H. Finite Element Method Seepage

Analysis for Relief Wells Appendix I. Soil Behavior for Slope Stability Appendix J. Geotextile Reinforced Embankment on

Chapter 13. Special Features Chapter 14. Development of Levee Operations

and Maintenance Manual

Appendix J. Geotextile Reinforced Embankment on Soft Foundation

Appendix K. Use of soil cement for levee protection Appendix L. Emergency flood protection Appendix M. Notation

Ch t 5 B A

Significant Updates30/81

Ch t 5 B A

Design Water Surface ElevationDesign Water Surface Elevation The top of barrier associated with the Design Water Surface

Elevation plus superiority plus over-wash height (or wave runupoverbuild) is termed the project (final) levee grade. Where superiority and over-wash height are zero, the final levee grade equals the DWSE.

All standard geotechnical analyses (such as seepage, stability, and erosion) shall use the DWSE with the standard design requirements (such as effective stress/vertical gradient factor of safety, slope t bilit f t f f t t )stability factor of safety, etc.).

StandardsStandards

“in this engineering manual (EM)in this engineering manual (EM), traditional design standards remain largely unchanged from previous editions but aunchanged from previous editions, but a risk-informed process for evaluating required levee reliability for requestingrequired levee reliability for requesting variances from the standards when reliability is either too high (too expensive)reliability is either too high (too expensive) or too low (too much risk).”

Old Levee, Updated Standards?Old Levee, Updated Standards?

For existing levees, all future evaluations will be based g ,on the same above risk-informed reliability evaluation process regardless of the method for how it was designed in the past If the reliability of the existing leveedesigned in the past. If the reliability of the existing levee is too low for the current estimated consequences, then the levee becomes a candidate for reevaluation within

S C fthe USACE levee safety program routine processes, leading to further risk assessments and potential follow-on feasibility studies, congressional funding y , g gauthorizations, and improvement design and construction.

Ch t 5 B A

ASTM Standard TestsASTM Standard Tests

Ch t 5 B A

Geomorphology and GeophysicsGeomorphology and Geophysics

TM 3-424 Mississippi River

Martis Creek Sierra Glacial

Texas Coast Littoral

Standardized Material Graphics and ColorsSta da d ed ate a G ap cs a d Co o s

Guidance On Interpretation Process d P t l f S ti D tand Portrayal of Supporting Data

Interpreted and Analyzed SectionsInterpreted and Analyzed Sections

Ch t 5 B A

Analysis MethodsAnalysis Methods

Focus on modern FEM computer methodsFocus on modern FEM computer methods while still considering “Blanket Theory”

Effective Stress Heave/Uplift Design Criteriaect e St ess ea e/Up t es g C te a

Seepage Berm DesignSeepage Berm DesignFS = 1.6DWSE FS = 1.0

Berm Crown

BermTop Stratum

LeveeRiverChannel

Pervious Substratum

Levee Toe

Berm Width

Berm Toe

Use existing factor of safety criteria for critical vertical gradient / effective stress factor of safety for uplift at end of berms.f f S 1 6 f If end of berm FoS is less than 1.6, then also assess likelihood of

progression of backwards erosion

Backwards Erosion / Piping FailureC R ti d/ C iti l H i t l G di tCreep Ratios and/or Critical Horizontal Gradients

Example With Vertical and H i t l G di t C id ti

Landside BermMeets USACE Vertical Gradient/

Seepage Path Length No BermCreep Ratio Too Low

Horizontal Gradient ConsiderationsMeets USACE Vertical Gradient/Effective Stress Design Criteria

& Increases Seepage Path Length

Creep Ratio Too Low

Seepage Path Length With BermCreep Ratio Sufficient (?)

Clay Blanket Thin Blanket Does Not MeetClay Blanket Thin Blanket Does Not MeetUSACE Vertical Gradient/

Effective Stress Design CriteriaVertical Boils Expected Here

Sand Aquifer

Failure Modes & Event TreesFailure Modes & Event TreesFlood Fighting Impacts

Traditional Creep Ratio or Modern Horizontal Gradient Computations

49/81Traditional Vertical Gradient / Effective Stress Computations

Filtered Toe Drain (aka Trench) DesignFiltered Toe Drain (aka Trench) Design

Relief WellsRelief Wells

Ch t 5 B A

“Critical State” Soil Behavior Framework

Summary of Types of Clays, Relative Strengths, Loading Conditions Conventional Names and Conditions ControlledConditions, Conventional Names and Conditions Controlled

Selection of StrengthsSelection of StrengthsTime for 99% DrainageTime for 99% Drainage

dense“strength”

NC “ t th”

Undrained Strengths Drained Strengths

dense and NC “strength”NC “strength”

aka “Rapid Flood Loading” of Soft Materials57/81

For more information see 2013 ASDSO C f P di2013 ASDSO Conference Proceedings

Levee Seismic RiskP t S i i W i / R Ti S iPost Seismic Warning / Response Time Scenarios

Earthquake

WSELeveed Area

Elevation

Earthquake

TiLong None

(evacuations and repairs)

Ch t 5 B A

Analytical FrameworkAnalytical Framework

Typical Analysis ParametersTypical Analysis Parameters

Design of Overtopping - Resiliency Measureses g o O e topp g es e cy easu es

Transition Area ArmoringTransition Area Armoring

Ch t 2 L F il M d d Ri k Appendix A. References Appendix B Typical Regional Designs and Typical Chapter 2. Levee Failure Modes and Risk-

Based Decisions During Design and Construction

Chapter 3. Field Investigations for Levees Chapter 4 Laboratory Testing for Levees

Methods to Assess of Levee ReliabilityA di D S b f I t t ti G hi Chapter 4. Laboratory Testing for Levees

Chapter 5. Borrow Areas Chapter 6. Subsurface Interpretation Chapter 7. Seepage Evaluation and Control

Ch t 8 Sl D i

Berms Chapter 8. Slope Design Chapter 9. Settlement Chapter 10. Erosion Control Chapter 11. Design and Bid Packages

Chapter 12. Levee Construction Chapter 13. Special Features Chapter 14. Development of Levee Operations

USBR / USACE “Best Practices”USBR / USACE Best Practices

Failure Modes & Event TreesFailure Modes & Event TreesFlood Fighting Impacts

Traditional Creep Ratio or Modern Horizontal Gradient Computations

Traditional Vertical Gradient / Effective Stress Computations68/81

Risk-Informed Design ProgressionRisk Informed Design Progression1

TRRL0.1

0.01 Existing ConditionsFailu

Existing Conditions

Low Population Levee High Population Levee

Existing Conditions

Minimum Berm

Existing Conditions

Minimum Berm

Mi i B1E-4

Higher Standard Berm

Higher Standard BermFl d Fi hti

Minimum Berm + Flood Fighting

u + Flood FightingHigher Standard Berm+ Flood Fighting+ Improved Emergency Evacuation

Life Loss10.10.01 10 100 1,000 1E4 1E5

How “Reliable” Does the Levee Need to Be For Prior-to-O t i F il M d Wh O t i C t l ?Overtopping Failure Modes When Overtopping Controls?

Existing ConditionsFailu

Existing Conditions

Low Population Levee High Population Levee0 0

Existing Conditions

Minimum Berm

Existing Conditions

Minimum Berm

Mi i B

Overtopping1/300 years

Higher Standard Berm

Higher Standard BermFl d Fi hti

Minimum Berm + Flood Fighting

How Far?

u + Flood FightingHigher Standard Berm+ Flood Fighting+ Improved Emergency Evacuation

Life Loss10.10.01 10 100 1,000 1E4 1E5

Taylor Series Approximate Solution – First Order Second MomentCan be used with virtually any limit state analysis.y y y

Only Models Aleatory•Vary each factor by +/1 sigma,L k t i ti i•Look at variation in response

function,•Use that as estimate of variance of response function

•Compare distance of “Limit State” to “Expected” value and based on variance estimated above, calculate percentage of area under lognormal curve for values greater

lognormal curve for values greater than the limit value.

Epistemic Uncertainty“U k U k ”“Unknown-Unknowns”

For geotechnical engineering, epistemic uncertainty g g g, p y(knowledge uncertainty) in many situations is more important than aleatory (natural variability), because it often cannot be estimated directly and can haveoften cannot be estimated directly and can have dramatic impact.

For geotechnical failure modes, both for design and reliability assessments, the epistemic uncertainty “unknown-unknown” challenge is common and well-known to the profession and has been addressedknown to the profession and has been addressed through a classic inductive-reasoning approach referred to as the Terzaghi and Peck “Observational Method” (P k 1969)(Peck 1969).

Epistemic Uncertainty Risk Mitigation and L S f t P N R ti A ti itiLevee Safety Program Non-Routine Activities

Recognizing that it is not possible to eliminate all “unknown-unknowns” on levee systems, per current and past USACE practice, performance monitoring and flood fighting will remain important risk reduction measures to reduce epistemic uncertainty over time. All f il d t t l ill i l d t d fAll failure mode event tree analyses will include event nodes for Unsuccessful Detection and Unsuccessful Intervention.

As a part of the proposed levee safety program (Draft EC 1110-2-6072, USACE 2014) if f f l i d d t dUSACE, 2014), if performance of a levee is deemed too poor and flood fighting activities too demanding and unreliable, a risk-informed process will be initiated to evaluate the needs for additional actions If risk is high enough a variety of structural and nonactions. If risk is high enough, a variety of structural and non-structural actions to increase levee reliability (i.e., change the fragility curves) and/or decrease potential consequences will be consideredconsidered.

Failure Modes & Event TreesFailure Modes & Event Trees

Ch t 5 B A

Explicit Description of Expected “Flood Fighting”p p p g g

Example Flood Fighting Evaluation“More than expected and, but for flood fighting, levee would have failed”p g g

Ensley Berm, Memphis 2011

Levee did not fail, but internal erosion pipes projecting towards the river found in 2012.

Example Flood Fighting Evaluation “Flood fighting occurred but levee failed”g g

L-575 Breach, NW Atchison County Levee District, Hamburg Iowa 2011

Possibly due to defects in riverside cap - fourth pipe formed and breached on June 13, 2011.

Ch t 5 B A

ScheduleSchedule

Independent External Peer Review (IEPR)Independent External Peer Review (IEPR)►Contract in Negotiation►Anticipated Completion December 2015►Anticipated Completion December 2015

More Review? Publication FY2016 Q3 or Q4

Questions / DiscussionQuestions / Discussion

Upcoming Changes for EM 1110-2-1913 Di C i dE l i fDesign ...€¦ · Di C i dE l i fDesign,...

Documents