URS LEVEE - GuidanceDocumentRev06 - Upload

DRAFT
GUIDANCE DOCUMENT
for Geotechnical Analyses
Urban Levee Geotechnical Evaluations Program

Contract 4600007418
Revision 6
March 2008
Prepared for:
DEPARTMENT OF WATER RESOURCES
Division of Flood Management
2825 Watt Avenue, Suite 100
Sacramento, CA 95821
Prepared by:
2870 Gateway Oaks Drive, Suite 150

Sacramento, CA 95833
In Association with:
TABLE OF CONTENTS
1.0 INTRODUCTION................................................................................................ 1-1

1.1 Program Overview....................................................................................... 1-1
1.2 Guidance Document Development.............................................................. 1-2
1.3 Purpose and Scope ..................................................................................... 1-2
1.4 Analyses Process ........................................................................................ 1-3
1.5 General Limitations...................................................................................... 1-3
2.0 REACH SELECTION ......................................................................................... 2-1

2.1 Reach Selection Principles .......................................................................... 2-1
2.2 Reach Selection Approach .......................................................................... 2-2
2.3 Reach Selection Process............................................................................. 2-2
Revision 06
2.3.1 Plan and Profile Preparation ........................................................ 2-2

2.3.2 Initial Reach Selection.................................................................. 2-3
2.3.3 Reach Characterization................................................................ 2-4
2.3.4 Reach Subdivision........................................................................ 2-5
2.4 Reach Refinement and Finalization............................................................. 2-5
3.0 SEEPAGE ANALYSIS USING SEEP/W ........................................................... 3-1

3.1 Objective...................................................................................................... 3-1
3.2 Software ...................................................................................................... 3-1
3.3 Water Level Conditions ............................................................................... 3-2
3.4 Criteria ......................................................................................................... 3-2
3.4.1 Seepage Gradients Defined ......................................................... 3-2
3.4.2 Seepage Gradient Criteria............................................................ 3-3
3.5 Procedure for Selecting Material Properties ................................................ 3-4
3.5.1 Anisotropy .................................................................................... 3-6
3.6 Analysis Approach....................................................................................... 3-7
3.7 Additional SEEP/W Guidelines .................................................................. 3-10
3.8 Sensitivity Analysis .................................................................................... 3-11
3.9 Past Performance Verification ................................................................... 3-11
3.10 Presentation of Results ............................................................................. 3-12
4.0 UNDERSEEPAGE ANALYSIS WITH BLANKET THEORY.............................. 4-1

4.1 Introduction.................................................................................................. 4-1
4.1.1 Blanket Theory Development and Appendix B............................. 4-1
4.1.2 Purpose of BTA ............................................................................ 4-2
4.2 Blanket Theory Assumptions ....................................................................... 4-2
4.2.1 Situations Where Blanket Theory is Invalid .................................. 4-4
4.3 BTA Guidance ............................................................................................. 4-4
4.3.1 Pervious Layer Thickness d and Permeability kf .......................... 4-5
4.3.2 Transformed Blanket Layer Thickness zb and
20080328_GuidanceDocumentRev06.doc i Issue Date: 03-2008

Revision: 6
TABLE OF CONTENTS
Permeability kb ............................................................................. 4-5

4.3.3 Landside Transformed Blanket Layer Thickness, zt, For
Seepage Gradient ........................................................................ 4-6
4.3.4 Factor c for Landside Blanket and Distance to Effective
Seepage Exit, X3 .......................................................................... 4-7
4.3.5 Factor c for Waterside Blanket and Distance to Effective
Seepage Entrance, X1 .................................................................. 4-8
4.3.6 At the Landside Levee Toe: Head Below the Blanket ho
and Average Vertical Exit Gradient io ........................................... 4-9
4.3.7 Beyond the Landside Levee Toe: Head Below the
Blanket hx and Average Vertical Exit Gradient ix .......................... 4-9
4.3.8 Waterside or Landside Ditches .................................................. 4-10
Revision 06
4.4 BTA of SEEP/W Cross Sections................................................................ 4-10

4.4.1 Boundary Conditions .................................................................. 4-11
4.5 Pore Pressures Near The Landside Toe ................................................... 4-12
4.5.1 Pore Pressures in the Landside Blanket .................................... 4-12
4.5.2 Pore Pressures Below the Levee ............................................... 4-14
5.0 STATIC SLOPE STABILITY ANALYSIS .......................................................... 5-1

5.1 Objective...................................................................................................... 5-1
5.1.1 Analysis Type ............................................................................... 5-1
5.1.2 Software ....................................................................................... 5-1
5.2 Water Level Conditions ............................................................................... 5-1
5.3 Loading Conditions...................................................................................... 5-3
5.3.1 Steady State Seepage (Landside Only) ....................................... 5-3
5.3.2 Rapid Drawdown (Waterside) ...................................................... 5-3
5.4 Minimum Factor of Safety Criteria ............................................................... 5-3
5.5 Material Properties ...................................................................................... 5-4
5.5.1 Free-Draining Versus Non-Free-Draining Materials ..................... 5-4
5.5.2 Strength Selection for Non-Free-Draining Materials..................... 5-4
5.5.3 Step-by-Step Procedure for Selecting Strength
Properties..................................................................................... 5-6
5.6 Analysis Approach..................................................................................... 5-13
5.6.1 Static Stability............................................................................. 5-13
5.6.2 Rapid Drawdown ........................................................................ 5-14
5.7 Modeling Cracks........................................................................................ 5-16
5.8 Additional Guidelines ................................................................................. 5-16
5.9 Past Performance Verification ................................................................... 5-17
20080328_GuidanceDocumentRev06.doc ii Issue Date: 03-2008

Revision: 6
TABLE OF CONTENTS
6.0 SEISMIC ANALYSIS ......................................................................................... 6-1

6.1 Seismic Loading Conditions ........................................................................ 6-1
6.2 Water Levels................................................................................................ 6-2
6.3 Liquefaction Triggering Analyses................................................................. 6-2
6.4 Seismic Slope Stability .............................................................................. 6-12
6.4.1 Seismic Shear Strengths............................................................ 6-12
6.4.2 Stability Computations................................................................ 6-13
6.5 Deformation Analyses and Vulnerability Assessment................................ 6-19
7.0 EROSION EVALUATION .................................................................................. 7-1

Revision 06
8.0 ALTERNATIVE ANALYSIS AND COST ESTIMATE ........................................ 8-1
9.0 CALCULATION PACKAGES ............................................................................ 9-1
10.0 REFERENCES................................................................................................. 10-1

10.1 Seepage Analysis References................................................................... 10-1
10.2 Blanket Theory References ....................................................................... 10-1
10.3 Stability Analyses References ................................................................... 10-1
10.4 Seismic Analyses References ................................................................... 10-2
Tables
Table 3-1. Permeability Values. ........................................................................................... 3-4

Table 4-1. Heads in Three-Stratum Blanket Layer. ............................................................ 4-13
Table 4-2. Summary Table for Blanket Theory Analysis. ................................................... 4-15
Table 5-1. Example of Strength Values Matrix................................................................... 5-11
Table 6-1. Relative Contributions of the Seismic Events...................................................... 6-2
Table 6-2. Seismic Vulnerability Classes. .......................................................................... 6-20
Table 6-3. Liquefaction Triggering Analysis Table. ............................................................ 6-21
Table 6-4. Example Seismic Strength Selection Summary Table. ..................................... 6-22
20080328_GuidanceDocumentRev06.doc iii Issue Date: 03-2008

Revision: 6
TABLE OF CONTENTS
Figures
Figure 1-1. Work Flow Chart. ............................................................................................... 1-4

Figure 1-2. Seepage and Slope Stability Analysis Process.................................................. 1-5
Figure 3-1. SEEP/W Model Geometry. ................................................................................ 3-8
Figure 3-2. SEEP/W Model Boundary Conditions. ............................................................... 3-9
Figure 3-3. Example Plot of Exit Gradient vs. Water Level. ............................................... 3-12
Figure 3-4. Example SEEP/W Model Summary. ................................................................ 3-13
Figure 3-5. Example SEEP/W Output. ............................................................................... 3-14
Revision 06
Figure 5-1. Analysis Flow Chart. .......................................................................................... 5-2

Figure 5-2. Tri-Linear Envelope Approach. .......................................................................... 5-6
Figure 5-3. Example of Cross Section Geometry Summary Plot. ........................................ 5-7
Figure 5-4. Estimates of Consolidated Undrained Strength from Triaxial CIU Test
Results. ................................................................................................... 5-12
Figure 5-5. Example Slope Stability Analysis Results. ....................................................... 5-19
Figure 5-6. Example Slope Stability Analysis Results. ....................................................... 5-20
Figure 5-7. Example Stability Plot. ..................................................................................... 5-21
Figure 5-8. Rapid Drawdown Slope Stability Analysis, 100 Year Flood Level to
Two Year Flood Level (Circular Search). ................................................ 5-23
Figure 6-1. 100-Year Return Period Mean PHA Hazard Sacramento and
Stockton Regions. ..................................................................................... 6-3
Figure 6-4. USGS 2002 Converted 100-Year Return Period Mean PHA Hazard,
Marysville Region...................................................................................... 6-6
Figure 6-7. Recommended Cyclic Stress Ratio versus Depth. .......................................... 6-11
Figure 6-8. Seed Harder Plot. ............................................................................................ 6-15
Figure 6-9. Recommended Kmax vs Input Motion PHA. ...................................................... 6-16
Figure 6-10. Strength Envelope for Post-Seismic Stability Analysis. ................................. 6-17
Figure 6-11. Newmark Displacements. .............................................................................. 6-23
20080328_GuidanceDocumentRev06.doc iv Issue Date: 03-2008

Revision: 6
TABLE OF CONTENTS
Figures
Figure 6-12. Example Stability Analysis Plot, Landside FOS (Circular Search),
Post-Earthquake Slope Stability Analysis, 500 Year Return. .................... 6-1
Figure 6-13. Example Stability Plot, Landside FOS (Non-Circular Search), Post-
Earthquake Slope Stability Analysis, 500 Year Return Period. ................. 6-2
Figure 6-14. Example Stability Analysis Plot, Waterside FOS (Circular Search),
Post-Earthquake Slope Stability Analysis, 500 Year Return
Period........................................................................................................ 6-3
Figure 6-15. Example Stability Analysis Plot, Waterside FOS (Non-Circular
Search), Post-Earthquake Slope Stability Analysis, 500 Year
Revision 06
Return Period. ........................................................................................... 6-4

Figure 6-16. Example Stability Analysis Plot, Landside FOS (Circular Search),
Pseudo-Static Slope Stability Analysis, 500 Year Event. .......................... 6-5
Figure 6-17. Example Stability Analysis Plot, Waterside FOS (Circular Search)
Pseudo-Static Slope Stability Analysis, 500 Year Event. .......................... 6-6
Figure 6-18. Example Stability Analysis Plot, Waterside FOS (Non-Circular
Search) Pseudo-Static Slope Stability Analysis, 500 Year Event.............. 6-7
Appendices
Appendix A Simplified Procedure for Shear Strength Selection of Fine-Grained Materials

Appendix B Kozeny-Carman Publication
Appendix C Shear Strength Correlations
Appendix D Lab Test Assignment Guidelines
Appendix E BTA Analysis Template
Appendix F Plan and Profile Preparation
Acronyms and Abbreviations
BTA blanket theory analysis

CIU isotropically consolidated undrained
CPT core penetration test
CRR cyclic resistance ratio
DSS direct simple shear
DWR Department of Water Resources
FEMA Federal Emergency Management Agency
20080328_GuidanceDocumentRev06.doc v Issue Date: 03-2008

Revision: 6
TABLE OF CONTENTS
Acronyms and Abbreviations
FS factor of safety
GEI GEI Consultants, Inc.
GER Geotechnical Evaluation Report
HEA horizontal equivalent acceleration
HGL hydraulic gradeline
ICB Independent Consultant Board
LiDAR light detection and ranging optical remote sensing technology
OCR over-consolidation ratio
Revision 06
PGA peak ground acceleration

P1GER Phase 1 Geotechnical Evaluation Report
QA/QC quality assurance/quality control
SEEP/W Brand name; SEEP/W is finite element software for analyzing groundwater
seepage and excess pore-water pressure dissipation problems in porous
materials like soil and rock.
SHANSEP Stress History and Normalized Soil Engineering Properties
SPT standard penetration test
URS URS Corporation
USACE Unites States Army Corps of Engineers
USCS Unified Soil Classification System
UTEXAS4 Brand name; UTEXAS4 is software for geotechnical slope stability
computations.
20080328_GuidanceDocumentRev06.doc vi Issue Date: 03-2008

Revision: 6
SECTION 1.0 INTRODUCTION
1.0 INTRODUCTION
This document provides guidance for those performing geotechnical analyses under the
state of Department of Water Resources’ (DWR’s) Urban Levee Geotechnical Evaluations
Program. The program is being executed as described in the Program Management Plan
(PMP). This Guidance Document has been prepared for the project team to facilitate
consistency in analysis approach, methodology, analysis results presentation and program
quality assurance and quality control activities.
While this Guidance Document offers a set of substantial directives on the matter of
processing data, no guidance document can identify all potential analysis variables.
Consequently, professional engineering judgment must be used in the application of any
protocols described in this document.
Revision 06
1.1 Program Overview
In May 2006, the State of California charged DWR with the task of evaluating and repairing
levee systems in California. In October 2007, DWR began a comprehensive program of
levee evaluation and upgrades. Of critical importance and highest priority to the program are
“urban levees,” the 350 miles of levee systems surrounding highly-populated urban areas
(greater than 10,000 residential), in the Central Valley.
DWR selected a URS Corporation-led team to assist them with geotechnical evaluation of
the state’s urban levees. Evaluation follows the US Army Corps of Engineers’ (USACE’s)
and program-specific protocols.
The Urban Levee Geotechnical Evaluations Program performs a logical progression of

historic data collection, field explorations, analysis, data gap assessment, supplemental data
gathering through additional field explorations and further levee analysis. This progression of
data gathering and study renders:
• An adequacy assessment for each investigated levee system addressing whether those
levees meet current DWR/USACE geotechnical criteria at selected flood levels.
• Identification of potential levee repair alternatives.
• Associated repair costs for detected geotechnical levee deficiencies.
Figure 1-1 presents a general work flow for the program. As noted above and in Figure 1-1,
the evaluation process for a given levee system begins with collecting pertinent historical
data related to:
1. Existing subsurface information

2. Levee performance and maintenance records
3. Geology and geomorphological studies
4. Field reconnaissance
Light detection and ranging optical remote sensing technology (LiDAR) topographic survey
information, subsurface exploration information including borings and cone penetration
20080328_GuidanceDocumentRev06.doc 1-1 Issue Date: 03-2008

Revision: 6
soundings typically made along the levee top and selected geophysical investigations
supplement available historical data. These supplemental data sources permit linking
historical observations with subsurface information and identifying risk factors.
Levee evaluations are being conducted in a comprehensive manner over a 2- to 3-year

period and involve a wide range of services including:
• Data review
• Exploratory drilling and testing
• Data processing
• Compilation using gINT, a graphical database program and Geographic Information
System (GIS) software
Revision 06
• Engineering analyses
• Preliminary design and construction cost estimates for recommended levee
improvements
1.2 Guidance Document Development
Since the beginning of the program, several workshops and meetings about the program’s
overall analysis approach have been facilitated by URS Corporation (URS). Input and
participation from DWR, USACE, GEI Consultants, Inc. (GEI), Fugro, GeoMatrix
Consultants, Inc. and ENGEO Inc. helped to craft these guidelines.
The program’s overall analyses approach has also been presented to and discussed with
DWR’s Independent Consultant Board (ICB). This Guidance Document summarizes analysis
protocols developed during workshops and from the ICB’s input.
This Guidance Document is updated on a regular basis as analyses protocols and

procedures used for this evaluation program are updated based on DWR, USACE and ICB
input.
1.3 Purpose and Scope
This Guidance Document’s main purpose is to outline a standard approach for completing
geotechnical analyses that evaluate current stability in DWR’s urban levees. Geotechnical
analyses discussed in this document include:
• Seepage analyses
• Stability analyses (static and seismic)
• Erosion evaluations
This document also contains guidelines on material characterization and for selecting
appropriate material parameters for analyses input.
In addition to presenting a standard approach for analyses, this document outlines a

preferred approach for selecting levee reaches. Levee reaches are the basic working blocks
for material characterization and locating study-sections.

Revision: 6
Additional Guidance Document objectives include:
• Helping to maintain a consist approach and results presentation across all teams
evaluating project levees
• Facilitating USACE and DWR guideline compliance.
• Providing an opportunity for input from the ICB, USACE and DWR.
• Ensuring the use of recent advances in material characterization and analyses
approach.
• Improving quality assurance/quality control (QA/QC) process efficiency.
1.4 Analyses Process

Revision 06
Phase 1 and supplemental analyses processes and how they relate to Phase 1 Geotechnical
Evaluation Reports (P1GERs) and final Geotechnical Evaluation Reports (GERs) are
presented in Figures 1-2A and 1-2B.
1.5 General Limitations
Recommendations in this document are solely intended to help streamline the material
characterization process and associated analyses. These recommendations are not intended
to be a final authority or to restrict users from employing engineering judgment, especially
when dealing with location-specific issues.
Readers are encouraged to perform additional analyses/evaluations on an as-needed basis

to supplement and verify analysis results discussed in this document.
It is the ultimate responsibility of individual analysis teams to be comfortable with, justify and
defend their own work.

Revision: 6
Revision 06
Figure 1-1. Work Flow Chart.

Revision: 6
Revision 06
Figure 1-2. Seepage and Slope Stability Analysis Process.

Revision: 6
Revision 06
Figure 1-2a. Seepage and Slope Stability Analysis Process (Continued).

Revision: 6
SECTION 2.0 REACH SELECTION
2.0 REACH SELECTION
This section provides general guidance for levee reach selection for analyses. These
guidelines are informational and flexible and should be applied as appropriate for site-
specific conditions and with sound professional judgment.
The entire length of urban levees must be divided into discrete lengths in order to perform
the incremental geotechnical analyses required for the overall evaluation. The descending
hierarchy for definition of lengths of levee is: levee type (urban or non-urban), levee study
area as defined by program task orders, levee reach (segments of a levee that when
combined cover the entire study area) and sub-reaches that are sub-component lengths
within a reach. A levee reach is defined for this program as a length of levee and foundation
over which evaluation results pertain consistently. Sub-reaches are shorter lengths, possibly
Revision 06
located within a reach, that are determined to be of individual analytical interest. The
rationale for creating these distinct linear sections is to perform evaluations at the lowest
level in the hierarchy that when totaled cover the entire length of the study area. Detailed
analysis and evaluation will always be performed on cross sections modeled from within
reaches or sub-reaches.
2.1 Reach Selection Principles
The principal reach selection premise is that each reach or sub-reach should be represented
adequately (in terms of geotechnical characterization and analysis) by one longitudinal soil
profile and one transverse cross section and one set of associated analysis input
parameters. The aim of reach selection is to divide the levee alignment into a minimum
number of reaches with similar geometric and geotechnical characteristics.
Reaches will be selected in consideration of the following physical factors and/or

characteristics:
• Geomorphology
• Regional geology
• Study area geology
• Reach geology including lithology
• Geotechnical interpretation
• Geotechnical properties
• Groundwater conditions
• Channel hydraulics
• Known past performance problem areas
• Levee improvements
• Levee geometry and topographic features

Revision: 6
2.2 Reach Selection Approach
The strategic approach for selecting reaches is to cover the study area with as few reaches
as is reasonably possible. The test of reasonableness should be based on a balance of the
following factors:
1. Initially select reaches of consistent physical characteristics such that the analytical
results will apply over the entire length.
2. Refine, split or create additional reaches or sub-reaches where the changes in physical
characteristics result in changes in analytical results.
3. Limit the number of reaches to that which can be reasonably evaluated given the
available project resources and budget.
Revision 06
Reaches will most commonly fall into one of the three following descriptive categories:
• Consistent conditions (either favorable or unfavorable)

• Highly variable conditions
• Anomalous or isolated specific conditions
A consistent reach is one where the factors are consistent over the entire length. Analytical
seepage and stability results will apply to the entire length of the reach.
A highly variable reach is one where the factors may not be consistent (homogenous) over
the entire length. Engineering analysis might require generalized conservative assumptions
for results to be valid over the entire length of the reach.
An anomalous reach is one that is substantially different in character than adjacent reaches.
This will most often occur as a sub-reach within a representative or random reach where
conditions dictate additional analytical attention. The anomaly could be either problematic,
such as a highly pervious bed underlying a thin blanket, or it could be non-problematic, such
as a slurry wall improved length of foundation.
2.3 Reach Selection Process
The reach selection process may be iterative, with initial reaches selected based on
stratigraphy, past performance, geometry and hydraulic loading and final selection modified
based on characterization and analyses. A minimum number of reaches should be used to
characterize the full levee alignment. However, reaches should be subdivided as needed to
achieve relative homogeneity.
2.3.1 Plan and Profile Preparation
A plan and profile drawing should be prepared to display stratigraphy and historical data. For
the P1GER analyses, this plan and profile will be primarily along the levee crest. However,
data collected on the levee toes or other areas during supplemental investigations may also
be included.

Revision: 6
The elevation profile should be for the entire length of levee under analysis. To the extent
information is available, plan and profiles should include:
• Cone penetration test (CPT) tip resistance and sleeve friction ratios.
• Geotechnical borehole logs, with standard penetration test (SPT) blow counts, Unified
Soil Classification System (USCS) material type and percent fines. The stratigraphy
should be represented for each log using standard fill patterns and standard colors
established for the Urban Levee Geotechnical Evaluations Program.
• Where space allows, historic borings should be shown stratigraphically using standard
symbols and colors.
• Levee crest, landside toe and adjacent ditch bottom profiles.
• If available, bottom profile of adjacent river or stream channel.
Revision 06
• Appropriate historical information.

• Areas of historically-observed levee performance problems (with dates and causes, if
available), including sand boils, seepage, slumping and erosion.
• Locations of any known levee penetrations or structures, like pumping stations, pipes,
cables, or similar items.
• Observed subsurface water levels, like those from piezometers, CPT dissipation tests
and or relief wells.
• Additional data deemed appropriate.
The plan should be an orthophoto plan view showing levee configuration and surrounding
surface features. Ground topographic contours can be included on the orthophoto plan view.
An orthophoto plan may also include geomorphic mapping of surface landforms or, if
practicable and useful, more complex terrain units representing both surface and underlying
landforms to some geotechnically-relevant depth (e.g., on the order of 50 feet below the
bottom of the levee). Additional guidance and graphic standards to be used in the
preparation of plans and profiles are provided in Appendix F.
2.3.2 Initial Reach Selection
The first step in the process of Phase 1 work in reach selection should be division of the
study area levees through geologic and geotechnical foundation and levee characterization
using the prepared plan and profile drawings. Contiguous lengths of levee and foundation of
similar geomorphologic, geologic, lithologic and geotechnical character and composition
should be lumped. When these factors change to the degree that analytical results will vary
significantly the reach should be divided or more properly stated a new reach should be
designated. Reach selection of this nature will be possible following the review of existing
geologic and geotechnical data, geologic characterization and geomorphologic analysis of
the study area, Phase 1 exploration and laboratory testing and the initial geotechnical study
area characterization of the levees and foundation.
The next step is to include knowledge gained from a review of past performance and levee
improvement projects. This information will identify areas of a problematic character that
could be either extensive or small features that lie within an otherwise geotechnically
consistent area. Areas of poor past performance are extremely important in this process

Revision: 6
because this knowledge can be used to quickly identify lithologies and geologic/geotechnical
combinations that could go unidentified due to the practical limitations of field investigations.
A cursory review of the surface topography and levee geometry should be performed to
identify variations in levee configuration which might warrant separate analysis. Cross
sections of the levees should be prepared as needed using ground surface elevations based
on ground survey data, LiDAR data or both. Cross sections should extend beyond levee
watersides and landsides as far as needed for seepage and stability analyses. Generally,
cross sections will be oriented perpendicular to the levee crest. In some cases (i.e., a
backfilled gravel channel passing obliquely under the levee), additional non-perpendicular
analysis cross sections may also be required.
Hydraulic loading needs to be considered prior to completing the initial reach selection
Revision 06
process. Due to channel configuration, levee geometry, topography, or other hydraulic

controls the height of water against the levee could vary over a given length. A reach should
be divided if and when the difference in hydraulic head significantly alters analytical results.
2.3.3 Reach Characterization
2.3.3.1 Substrata Properties
Reach characterization requires determining geotechnical behavioral properties for each

relevant substrata. Geotechnical characterization should be based on analysis, interpretation
and assessment of geotechnical data collected along or near the levee alignment. Necessary
analysis inputs that are part of geotechnical characterization include the following material-
property parameters:
• Soil permeabilities
• Shear strength parameters
• Unit weights
2.3.3.2 Modeled Cross Sections
Once reaches have been selected. Each reach is characterized to identify one or more:
• Critical seepage analysis sections, where the highest exit gradients are considered
likely.
• Critical slope stability analysis sections, where the lowest factors of safety are
considered likely.
To adequately represent a reach with one set of analysis material properties requires
adequate consideration of the spatial variability and the uncertainty of the reach’s properties.
Within practical constraints, each analysis-soil-profile stratum significantly affecting results
should ideally be characterized by only one (statistically) homogeneous population for each
geotechnical parameter needed. A homogeneous population for an analysis-relevant soil
property has, within the limits of available data, an approximately constant spatial average
and degree of spatial variability. Significant spatial discontinuities or systematic changes in
soil properties — in terms of either spatial average or degree of spatial variability —

Revision: 6
generally require subdivision of each reach to achieve an acceptable level of intra-reach

homogeneity for analysis.
The most problematic section or “conservative section” should be selected from within each
reach for analysis. Sections for seepage and stability analysis need not be taken at the same
location, and if necessary more than one analysis section may be selected for the reach. As
the evaluation progresses problematic areas may be identified within a reach or it may
become necessary to further sub-divide reaches or create sub-reaches. For a complete
evaluation a determination of adequacy must be made over the entire length of the levee
study area; so sub-division of reaches must continue until there is an analytical result that
corresponds to each reach or sub-reach of levee.
Since separate cross sections may be analyzed for stability and seepage, seepage analysis
Revision 06
must be performed on slope stability sections even if exit gradients are not critical at that
section. A reach may also be characterized differently for seepage analysis and stability
analysis to account for different sensitivities in analysis results to spatial variability or
uncertainty in or between reaches.
2.3.4 Reach Subdivision
Levee reaches adequately represented by one analysis cross section (separate sections
may be analyzed for stability and seepage) and set of geotechnical input parameters are
estimated and selected using geotechnical characterization data. In keeping with principal
premise of reach selection, a separate reach may be identified for any major change in
conditions potentially affecting levee performance, including:
• Improvements, like slurry walls and berms

• Cross sectional geometry or water level elevations
• Soil profiles and analysis material properties (inferred from boreholes and CPTs,
laboratory testing results, geophysical surveys and geomorphic assessment)
• Groundwater conditions
• Historic performance, including areas of past sand boils, seepage, erosion, slumping, or
over-steepened slopes
In practical cases, limited areas within a reach lacking soil homogeneity, if not excessive or
extreme, may be included in that reach. Suitably conservative properties should be
estimated for analysis.
2.4 Reach Refinement and Finalization
Reach selection may, in some cases, require a process of iterative selection,

characterization and analysis. In particular, iteration may be required to determine where an
apparent discontinuity or significant change in material properties requires a separate reach,
or where the change can be handled in an existing reach by making a more conservative
estimate of the parameter value used in analysis. Again: some reaches may have to be
defined differently for different analysis types (e.g., stability or seepage) to account for

Revision: 6
different sensitivities in the analysis results to spatial variability or uncertainty in or between

reaches.
As stated, reach selection can be an iterative process, with refinements occurring based on
analysis and evaluation results. Analysis results like seepage gradients, height of daylighting
phreatic surface, stability factors of safety and seismic yield accelerations should be
presented in both tables and graphically on elevation profiles, with standardized symbols and
colors.
Analytical results should be included on elevation profiles to facilitate direct comparison of

multi-reach subsurface interpretation data and anticipated overall levee performance.
Comparison of multi-reach analytical results also aids in the quality control process, as
anomalous results are more evident during comparison of multiple reaches. Reach selection
Revision 06
should also identify any data gaps requiring additional field or laboratory work and follow-up
analysis.
The program team anticipates it may be practicable to statistically characterize geotechnical

data (soil types and associated properties) on a landform basis, based on landforms mapped
as part of the geomorphic assessment. If so, reach-specific data could be supplemented with
applicable landform-based prior data, increasing geotechnical characterization and analysis
reliability and cost-effectiveness.
Generally, the ability to identify and characterize reaches in terms of soil profile and
associated material properties for analysis is limited by the quantity and quality of available
data. Locations where borings or CPT signatures appear to show discontinuities or
anomalies may represent data gaps that require additional borings for adequate segment
definition and property determination.
Reach refinement will occur as the investigations and evaluations progress. The first phase
of analytical results presented in the P1GER will lead to refined exploration, refined
exploration will yield a better understanding of subsurface conditions, which in turn will
enable further analysis that will more appropriately represent the reaches under
investigation. Thus the iterative exploration and evaluation process will continue from the
P1GER to the GER work phases and reach boundaries will shift until there is an analytical
result corresponding to each length of levee and foundation. The engineer can complete the
evaluation regarding the adequacy or inadequacy of the levees and foundations when
analytical results have been obtained for all of the reaches and sub-reaches that compose
the task order study area.

Revision: 6
SECTION 3.0 SEEPAGE ANALYSIS USING SEEP/W
3.0 SEEPAGE ANALYSIS USING SEEP/W
Two independent seepage analyses approaches will be used for the DWR Urban Levee
Geotechnical Evaluations Program. The first approach (described in this section) uses
SEEP/W finite element software. The second approach is based on a closed-form
formulation for underseepage called blanket theory analysis (BTA); BTA’s approach is
described in Section 4.0.
Based on a comparative study completed by URS in July 2007, for most cases both of these
approaches yield similar results. BTA has the advantage of providing a quick and relatively
easy means of estimating the excess head values along the bottom of the blanket. BTA does
not provide a way to estimate levee phreatic surface and does not consider through
seepage. By comparison, SEEP/W is a much more comprehensive and general approach for
Revision 06
analyzing through- and underseepage issues concurrently.
For DWR’s Urban Levee Geotechnical Evaluations Program, a balance between these two
analysis approaches is attempted. This overall approach takes advantage of both SEEP/W’s
and BTA’s individual strengths. In accordance with this blended approach, all sections
having a blanket meeting applicable conditions should be analyzed with the SEEP/W
program and with BTA.
In essence, BTA results help check SEEP/W results. Total head values computed from the
two analysis approaches along the bottom of the blanket will be plotted together on the
SEEP/W result figures. For cases where BTA results do not match SEEP/W results,
additional checking and justification will be needed to explain differences. Ultimately,
SEEP/W-computed phreatic surface and pore-water pressure values will be used in slope
stability analyses.
3.1 Objective
Perform two-dimensional steady-state seepage analyses using a finite element approach for
DWR’s Urban Levee Geotechnical Evaluations Program. The analyses’ primary objectives
are:
• Predict steady-state phreatic levels and pore pressures in the levee and foundation soils
for various probable future water level conditions. Pore pressure information will be used
in the slope stability analyses.
• Estimate hydraulic gradients, especially exit gradients around the levee toe area and
uplift gradients through the landside blanket, for the probable future water level
conditions.
3.2 Software
SEEP/W 2004 and 2007 versions. SEEP/W is finite element software for analyzing
groundwater seepage and excess pore-water pressure dissipation problems in porous
materials like soil and rock. SEEP/W has been developed by GEO-SLOPE International Ltd.
(http://www.geo-slope.com).

Revision: 6
3.3 Water Level Conditions
Steady-state seepage analyses will be completed for the following water level conditions:
• Top of Levee or 200-year (median) flood level plus 3 feet, whichever is lower
• 200-year (median) flood level
• 1957 design water surface for the Sacramento River basin or
• 1955 design water surface for the San Joaquin River basin
Water surface elevations for each water level condition will be determined as directed by
DWR. Water surface elevations used for each Study Area will be documented in the Phase 1
Revision 06
Geotechnical Data Report (P1GDR) and Geotechnical Data Report (GDR).
All analyses will be completed for current levee conditions without considering any potential
modifications except for planned imminent improvements which will be modeled accordingly,
e.g. RD 784, Natomas, etc.
3.4 Criteria
Criteria adopted for the seepage analyses are primarily based on the ICB’s
recommendations with input from DWR and USACE. It is the program team’s understanding
that both DWR and USACE agree about the seepage analysis criteria. It is important to note
that these criteria have no real effect on analysis methodology and calculations.
3.4.1 Seepage Gradients Defined
There are two general types of gradients, internal and external. External gradients, or
surface gradients, involve the change in total head from point to point on the ground surface.
Internal gradients involve the change in total head from one point in the ground to another
point in the ground. Following is a list of the various types of seepage gradients that are
computed and discussed in this Guidance Document:
1. Average vertical exit gradient: Average vertical exit gradient is the total head drop in
the vertical direction across the levee’s landside blanket divided by the blanket thickness.
It is also referred to as the uplift gradient. This gradient value controls a blowout type
failure through a low permeability blanket and is frequently referenced in typical levee
design/evaluation manuals.
2. Localized exit gradient: Localized exit gradient is the resultant (“xy”) gradient reported
by SEEP/W along the water exiting surface nodes of the model. Localized exit gradient
values should be investigated since these nodal values can be heavily influenced by the
model geometry and mesh density in the SEEP/W analyses. Localized exit gradient
values (which are judged realistic) should be appropriately evaluated for progressive
piping in silty and/or dispersive materials.
3. Internal gradient: Internal gradient is defined as the gradient, reported by the SEEP/W
contouring routine, in any direction across an internal soil layer. A high gradient at the
boundary between two materials having different gradations may indicate the potential

Revision: 6
for internal piping, especially when a coarse-grained layer is located downstream (down-
gradient) of a finer material that is not filter-compatible with the coarse material.
4. Average exit gradient: Average exit gradient is the exit gradient computed manually in
any direction across a reasonable distance, which may be the extents of a low
permeability layer. This gradient is especially important in the case of sandy levees that
have a relatively thin clayey berm placed on the landside face of the levee. High exit
gradients across the thin clayey berm material may indicate the potential for a blowout-
type failure. Average exit gradients are also important for reporting the exit gradients for
sections with no landside blanket.
3.4.2 Seepage Gradient Criteria
The following seepage gradient criteria are adopted for all seepage analyses.
Revision 06
For underseepage conditions, the current USACE criterion for the average vertical exit
gradients through a levee’s landside blanket is that it be less than or equal to 0.5 for the
design floodwater level condition. Please note that the USACE criterion is applicable to the
“average vertical exit gradient” and not to the “localized exit gradient” nor to the “average exit
gradient.” This statement by no means implies that the localized exit gradients and the
average exit gradients are any less important than the average vertical exit gradients. The
user should report all kinds of gradient values that are judged important for a particular
section but should only compare the average vertical exit gradients to the USACE criterion.
All cases in which the phreatic surface exits through the waterside slope of the levee need to
be identified. It is the ICB’s opinion that, for clay levees, this condition would not pose a
serious threat to levees built of clay. However for erodible materials like silts and sandy silts,
a weighted filter should be recommended as a prudent measure.
All cases with high localized exit gradients (generally greater than approximately 0.5) need to
be identified. It is the ICB’s recommendation that, at locations where high localized exit
gradients are computed, an average exit gradient value measured across a reasonable
distance should also be reported. The purpose of this exercise is to evaluate the extents and
implications of the high localized exit gradient value.
For sections with no landside blanket, average exit gradients across reasonable distances
measured along the flow path should be calculated and reported. Similarly, areas with a thin
continuous blanket should be checked for potential up-lift/blow out conditions.
When modeling sections with toe ditches, the ditches should be modeled without any water
in the ditch. If problems are found with this assumption, then the ditch should be analyzed full
of water. If this solves the problem, these areas must be flagged so that a flood preparation
plan is prepared by the local authority to assure the ditch is maintained full of water during
flood conditions.

Revision: 6
3.5 Procedure for Selecting Material Properties
Material properties required as input for steady-state SEEP/W analyses include:
• Horizontal permeability under fully saturated conditions (kh)

• Ratio between vertical and horizontal conductivities, anisotropy ratio (kh/kv)
• Hydraulic conductivity function (variation of conductivity with pressure)
Hydraulic conductivity values of various materials are selected based on:

• Laboratory gradation analysis results (sieve analysis with hydrometer test) coupled with
published Kozeny-Carman procedure (see Appendix B)
• USCS and published broad empirical relationships between soil type and permeability
Revision 06
like those presented by Terzaghi and Peck (1996). See Table 3-1.
As a first step, material is classified according to the USCS (CL, SM, SP, etc.) as
standardized in American Society for Testing and Materials (ASTM) D 2487. Depending on
soil type, either the Kozeny-Carman procedure and/or the published values shown in Table
3-1 are used to select representative permeability values for each material type. Selections
are made in accordance with the following guidelines:
• Clayey and silty materials (CL, CH, ML, MH) use values from Table 3-1.
• Sandy materials (SP, SW, SM and mixes) use Kozeny-Carman procedure.
• Gravels (GP, GW, GM) use both the Kozeny-Carman procedure and the values
presented in Table 3-1.
• SC and GC materials methods to be determined
Table 3-1. Permeability Values.

kh (kx in Resulting kv (ky in
kh/kv
SEEP/W) SEEP/W) Layer
Material Type Soil Description
(kx/ky Color
cm/s ft/day cm/s ft/day
SEEP/W)
Slurry Cutoff -6 -6
S-C-B or S-B 1x10 0.0028 1 10 0.0028 Violet
Wall
-5 -6
Compacted clay fill 1x10 0.028 4 2.5x10 0.007 Indigo
Weathered clay blanket 3 feet
-5 -3
thick with significant vertical 1x10 0.028 1/100 10 2.8 Indigo
cracks
Low to Medium Weathered clay blanket 4 to 10
-5 -4
Plasticity Clay feet thick with moderate vertical 1x10 0.028 1/10 10 0.28 Indigo
cracks
Weathered clay blanket >10 feet
-5 -6
thick, or unweathered clay 1x10 0.028 4 2.5x10 0.007 Indigo
blanket
-5 -6
Unweathered clay at depth 1x10 0.028 10 10 0.0028 Indigo
-4 -4
Silt fill (mechanically placed) 1x10 0.28 1 10 0.28 Blue
-4 -4
Silt 86 to 100% fines 2x10 0.57 1 2x10 0.57 Blue
-4 -4
Sandy silt, 50 to 70% fines 2x10 0.57 1 2x10 0.57 Blue

Revision: 6
Table 3-1. Permeability Values.

kh (kx in Resulting kv (ky in
kh/kv
SEEP/W) SEEP/W) Layer
Material Type Soil Description
(kx/ky Color
cm/s ft/day cm/s ft/day
SEEP/W)
Silty Sand Green
Sand to Silty Use permeability values calculated using
Kozeny-Carman equation and kh/kv value of 4 Yellow
Sand
Sand Yellow
-4 -4
28 to 49% fines 4x10 1.134 4 10 0.28 Green
-3 -4
18 to 27% fines 10 2.835 4 2.5x10 0.71 Green
-3 -4
13 to 17% fines 6x10 17.008 10 6x10 1.7 Yellow
Gravel -2 -3
8 to 12% fines 1.2x10 34.02 10 1.2x10 3.4 Yellow
Revision 06
-2 -3
-2 -3
-4 -4
28 to 49% fines 4x10 1.134 4 10 0.28 Green
-3 -4
18 to 27% fines 10 2.835 4 2.5x10 0.71 Green
Gravel With 13 to 17% fines 10
-2
28.35 10 10
-3
2.8 Yellow
Cobbles And -1 -2
Sand 8 to 12% fines 10 283.5 10 10 28.3 Orange
-1 -2
0 to 2% fines 2x10 566.9 0.1 2x10 56.7 Orange
-1 -2
3 to 7% fines 2x10 566.9 0.1 2x10 56.7 Orange
-0 -0
Drain Rock Gravel 10 2834.6 1 10 2834.6 Red
Notes:
1.) kh = horizontal permeability
kv = vertical permeability
cm/s = centimeters per second
ft/day = feet per day
2.) For blanket layers on the waterside, where the potential for cracking is minimal, kh/kv = 4 is recommended
When using the Kozeny-Carman procedure, the following step-by-step procedure is

recommended to select representative conductivity values for a particular layer. Note that the
permeability value estimated using the Kozeny-Carman procedure is the vertical conductivity
value (kv) of a homogeneous material (Chapuis and Aubertin, 2003). This value needs to be
multiplied by the anisotropic ratio (kh/kv) to compute the horizontal conductivity (kh).
1. Collect and summarize all available gradation (sieve and hydrometer) test results for a
particular interpreted layer from the same reach.
2. Using the Kozeny-Carman procedure, estimate conductivity values corresponding to
each set of gradation test results.
3. Plot conductivity values with elevation for a particular layer corresponding to a single
reach and compute the geometric data mean.
4. Prepare separate but similar plots of conductivity values from this layer and from the
same layer in other nearby or comparable reaches where the same material type is
present.
5. Compute the combined data set’s geometric mean and compare this value to the
geometric mean computed in Step 3.
If the two values are similar (within half an order of magnitude), select a single
approximate value to represent that particular layer in all those reaches. If the two

Revision: 6
computed values are quite different, select an approximate value close to the geometric
mean calculated in Step 3 to represent that particular layer in that particular reach.
Considering the uncertainty in estimating conductivity values, precision of the

representative value may be greater than an order of magnitude.
For situations where calibration runs are performed with the intent of matching field
observations, deviations from these guidelines are acceptable as long as the selected values
are reasonable and defendable.
Please note that, for these cases, values should not be selected merely to match field
observations without regard of the reasonableness of selected values for any particular soil
type.
Revision 06
3.5.1 Anisotropy
Clayey and silty materials’ anisotropy should be selected in accordance with the values
presented in Table 3-1. The ICB has suggested kh/kv of 3 to 10 for top clay stratum. The
project team understands this suggestion is for unweathered clay stratums that do not have
any surface cracks.
It should be noted that not all of these materials and characterizations will be used for all
levees. In particular, the higher vertical gradients for thin weathered clay blankets should
only be used for parametric evaluations, as they are considered unconservative. The higher
vertical gradients have been proposed to model the dissipation of pore pressures that could
occur through a thin cracked clay blanket; however, a thin cracked clay blanket should be
considered a problem.
For blanket layers on the levee waterside where potential for cracking is minimal a kh/kv of 4
is recommended.
Anisotropy for sandy materials is suggested as an assumed kh/kv of 4 unless there is strong
evidence for selecting a different value. Similarly, anisotropy for drain rock is suggested as
an assumed kh/kv of 1 unless there is strong evidence for selecting a different value.
As discussed above, for situations where calibration runs are performed with the intent of
matching field observations, deviations from these guidelines are acceptable as long as the
selected values are reasonable and defensible.
If a collection of two or more thin soil layers with distinctively different hydraulic conductivities
are grouped together into a single idealized layer in the SEEP/W model, then the anisotropy
of this idealized layer should be based on engineering judgment rather than on the general
material description criteria presented in Table 3-1. Likewise, engineering judgment is
required for layers that consist mainly of interfingering lenses of various materials of differing
gradation and permeability.

Revision: 6
3.6 Analysis Approach
Seepage analyses are performed using SEEP/W, a two-dimensional finite element software
program developed by Geo-Slope International. All analyses are completed for steady-state
flow conditions without considering time. Both saturated and unsaturated flow conditions are
evaluated in the same model using conductivity functions. Conductivity functions define the
variation in material conductivity with pressure (negative and positive).
The step-by-step approach for completing the seepage analyses is as follows.
1. Develop SEEP/W model (mesh) geometry

Use cross section geometry developed from survey data and boring/CPT investigations.
Investigate the presence of existing improvements like berms, slurry walls, drainage
Revision 06
ditches etc. and include them in the cross section geometry summary plot. This
information might be obtained from as-built drawings, past reports, and/or field
reconnaissance. A typical cross section geometry summary plot is shown in Figure 3-1.
The SEEP/W model geometry including internal stratigraphy should be consistent with,
and to some extent identical to, the corresponding UTEXAS4 model geometry.
UTEXAS4 is a software program for geotechnical slope stability computations.
Landside and waterside ditches or other relevant drainage features should be modeled
for partially or fully penetrating conditions through the blanket as appropriate as
discussed in Section 4.3.8.
Where a levee section is adjacent to the waterway, the thickness and lateral extent of the
waterside blanket layer may have been adversely impacted by scouring and erosion.
This may have left behind a blanket layer that is considerably thinner than (as compared
to its thickness underneath or landside of the levee) or even absent on the waterside,
and may also have severely limited the layer’s extent beyond the levee toe. For such
cases, the waterside blanket should be considered in analysis only if there is sufficient
geologic or geotechnical information (for example, from geomorphology studies,
subsurface exploration, electromagnetic or geophysical survey data, etc.) indicating its
existence, lateral extent and thickness. Where such confirmatory information is not
available, the following recommendations are made:
• Case A – Thickness of blanket layer below levee is less than or equal to 20

feet: The waterside blanket layer should be considered to be completely eroded and
should not be included in the analysis. Analysis should be performed assuming a
blanket layer truncated at the waterside toe (that is, assuming that the layer does not
exist between the waterside levee toe and the river) and replaced by the underlying
pervious layer – constituting a situation where the blanket layer has been eroded and
the area has been refilled by sands. The pervious layer would thus be directly
exposed to the river, constituting an open seepage entrance (equivalent to a
constant head at the seepage entrance equal to the river stage).
• Case B – Thickness of blanket layer below levee is greater than 20 feet: At
locations where the blanket layer thickness below the levee (as verified through
subsurface exploration) is greater than 20 feet, it seems likely that part of this layer

Revision: 6
would exist waterside of the levee, particularly if it is composed predominantly of

clayey soil (that is to say, not all of it would be eroded or scoured away). In the
absence of sufficient geologic and geotechnical information waterside of the levee,
an estimate of the lateral extent and thickness of the waterside blanket layer may be
made by using a combination of other available information, such as current surface
topographical data (from survey, LiDAR, etc.), air photos, local agency inspection
and repair reports, specific site reconnaissance and engineering judgment. Analysis
can be performed assuming this estimated waterside blanket. Analysis results should
be checked against the levee’s documented past performance.
On the levee landside, distance from the levee centerline to the model (mesh) end
should preferably be at least 2,000 feet.
On the levee waterside, the model (mesh) should preferably extend up to the middle of
Revision 06
the river. If the middle of the river is more than 1,000 feet from the levee centerline, then
the model could be terminated at 1,000 feet from the centerline.
Model (mesh) depth should be established using judgment based on subsurface

stratigraphy. Model depth below the levee toe should not be less than approximately four
times the levee’s height or 60 feet unless there is strong evidence to do otherwise.
The model should have relatively high mesh density (relatively smaller elements) in and
around the levee structure and low mesh density (relatively large-size elements) towards
the two levee boundaries. Mesh density is especially critical near the landside levee toe
and any landside ditches where the critical exit gradient will occur.
The model’s “y” coordinate system should correspond to the project elevation (North
American Vertical Datum of 1988, or NAVD 88).
Figure 3-1. SEEP/W Model Geometry.

2. Assign conductivity values to all materials
Assign conductivity values to each material using the help of conductivity functions. The
conductivity function defines the variation in material conductivity with pressure (negative
and positive). Use the appropriate standard conductivity functions available in SEEP/W
by scaling them up or down with respect to the saturated conductivity value.

Revision: 6
Use the color coding shown in Table 3-1 for presenting the conductivity values.
3. Assign external hydraulic boundary conditions (see Figure 3-2)

On the waterside, fixed total head boundary conditions, corresponding to the water level,
are applied along the top of the model (base of the river) and along the face of the levee.
The no-flow boundary condition is assigned along the vertical face of the model on the
waterside slope.
On the landside, water level is assumed to be at the ground surface unless site-specific
data is available to indicate a specific depth of water below the ground surface. Total
head boundary conditions, corresponding to the landside water level, are applied along
the vertical face of the model on the landside.
Revision 06
Bottom nodes are specified as a no-flow boundary (default condition for SEEP/W).
Review nodes are assigned along the landside to properly estimate the phreatic surface
around toe area (review nodes allow water to get out of the model boundary under zero
pressure head, without fixing the total head at the node locations) Center Line of
Levee Crest
Apply Total Head, H = Y1 Levee Crest

Water Level El = Y1 feet Review Nodes Water Level El = Y2 feet
(Model "y" Coordinate = Y1 also) (Model "y" coordinate = Y2, also)
Apply Total Head = Y2

60 ft (min)
No-Flow Boundary
Center-Line of River or 1000 feet, whichever is shorter ~ 2000 feet
Figure 3-2. SEEP/W Model Boundary Conditions.

Ditches should first be analyzed with no water in the ditch. If unacceptable seepage
gradients are found with this assumption, the ditch should be modeled full of water. For
the no water in the ditch scenario, review nodes need to be assigned in the ditch. And for
the full of water scenario, a total head boundary condition corresponding to the water
level in the ditch needs to be assigned to nodes in the ditch.
4. Assign internal hydraulic boundary conditions, if needed.

Internal hydraulic boundary conditions are assigned to simulate an internal thin drainage
layer without having to physically model the drainage layer with elements.
Internal nodes at the location of the drainage layer are assigned a pressure head
boundary condition with P=0 along the physical extent of the drainage layer.

Revision: 6
The P=0 boundary condition inherently assumes that the pressure in the drainage layer
is atmospheric and whatever quantity of water enters the drainage layer will be
transmitted by the drainage layer without developing any positive pressure head. This is
a very critical assumption that must be validated with judgment and/or hand calculations.
If this assumption can not be supported for a particular drainage layer, then that drainage
layer should be modeled as a material type (with elements) instead of using the internal
boundary conditions.
5. Solve for steady-state seepage conditions

6. Prepare summary plots of results.
Prepare all figures in PowerPoint format and use the 11 x17 title block template
developed by URS to present model results.
Revision 06
Present all model results at the same scale, with model geometry, total head and
gradients all plotted on the same figure (different plots, however). Both vertical (y)
gradients and resultant (xy) gradients need to be plotted. Flow vectors also need to be
shown on the total head plot. A scale of 1”=100’ is recommended for all plots. Use a
contour interval of 0.1 for presenting hydraulic gradients and 1.0 ft. for presenting total
head contours. Furthermore, use a consistent color scheme for presenting head and
gradient contour lines.
Any location with concentrated gradients needs to have a blowup detail of the area of
concern. On this blowup plot, appropriate contour intervals will be used for the gradient
and/or head contour lines that best illustrate the situation.
In addition to the gradient values reported by SEEP/W, average vertical exit gradients
across the landside blanket layer need to be calculated and reported on the figures.
3.7 Additional SEEP/W Guidelines
The following additional guidelines were developed based on URS’s experience with
SEEP/W analyses and suggestions made by the ICB following review of interim SEEP/W
analyses results. These guidelines are intended to further help with and improve the quality
of the SEEP/W analyses and the presentation of results.
• Set the solution convergence tolerance limit to a very low number (around 0.0001 or
lower) and the maximum number of iterations to at least 500. Do not use the SEEP/W
default values.
• To verify solution convergence, compute total model inflow and outflow values with the
help of flux sections in SEEP/W. The difference between the two values needs to be
small, typically less than 5 percent.
• Use an unexaggerated 1 horizontal to 1 vertical scale while setting up the model in
SEEP/W.
• The conductivity value input in the SEEP/W model is the horizontal conductivity value.
And the anisotropic ratio input in the SEEP/W model is the ratio of the vertical
conductivity to the horizontal conductivity (Kv/Kh) and NOT Kh/Kv, as commonly
reported.

Revision: 6
• Localized exit gradients are important and need to be evaluated and reported. The nodal
gradient values reported by SEEP/W should be examined in detail to ascertain whether
they are artifacts of mesh details or an indication of reality. Where necessary, localized
exit gradients may be calculated manually across a reasonable distance to support or
discount the nodal gradient values reported by SEEP/W.
• For sections with no landside blanket, average exit gradients should still be calculated by
estimating the total head drop across a reasonable distance along the flow path.
• In addition to calculating an average vertical exit gradient across a low permeability
stratum (blanket), a factor of safety (FS) against a blow-out type failure should also be
calculated and reported. The FS against blow-out is defined as the weight of an
overburden stratum divided by the counter acting seepage forces acting at the base of
the stratum. The calculation of FS against blow-out is especially important for situations
where a relatively thin low permeability berm is placed over the landside face of a sandy
Revision 06
levee.
• As discussed in the beginning of this section, results from the SEEP/W analyses need to
be independently checked against the BTA results. Total head values along the bottom
of the blanket computed by the two approaches should be plotted together on the
SEEP/W result figures. For cases where the BTA results do not match the SEEP/W
results, additional checking and justification should be performed to explain the
difference.
3.8 Sensitivity Analysis
Because of the inherent uncertainty involved in estimating the material conductivity values,
sensitivity analyses should be performed, on an as-needed basis, to evaluate the impact of
potential variation in conductivity values. It is recommended to consider at least one order of
magnitude change from the best-estimate conductivity values in the sensitivity analyses. It
should be noted that, if all conductivity values in a model are increased or decreased by the
same factor, the computed phreatic surface will remain unchanged. Therefore, it is important
to study the relative change in the conductivity values instead of a uniform change in values.
To assess the sensitivity of average vertical exit gradients to water levels, a plot of the
calculated average vertical ext gradients versus water surface elevations should be prepared
for each analysis cross section. An example plot is provided in Figure 3-3.
3.9 Past Performance Verification
Agreement between past performance and results from the back analyses of past conditions
is the best possible check on the reasonableness of the model input parameters. To take
advantage of this golden rule, the following two recommendations are made:
In areas where past problems, including wet spots, heavy seepage, or boils, have been
observed during previous flood events, a back analysis should be performed using the
estimated water level that existed at the time of the observed problems. If the analysis
results seem to agree with the past observations, these findings should be clearly
documented and discussed in the report. If, even with reasonable adjustments in the input
parameters, the model results do not agree with the past observations, these findings should
also be reported with appropriate explanation/recommendation.

Revision: 6
In areas where the model results, for the 100, 200-year events, seem to indicate high
gradients and/or through seepage, but the past performance data does not indicate any
historical problems, a back analysis should be performed using the estimated high water
level that that area has ever experienced in the past. If the results from this back analysis
also seem to indicate potential problems, the model geometry and/or the input parameters
should be re-evaluated and recommendations/explanations need to be provided to address
this apparent discrepancy.
3.10 Presentation of Results
For clarity of and to facilitate independent reviews SEEP/W results should be formatted and
presented as shown in Figures 3-4 and 3-5.
Revision 06
Steady State Seepage Condition

75
73
Water Level (NAVD 88)
70
Levee
Crest
68
200 yr
65
100 yr
63
60
0.4 0.5 0.5 0.6 0.6 0.7 0.7
Average Vertical Exit Gradient at Levee Toe (from SEEP/W)

Figure 3-3. Example Plot of Exit Gradient vs. Water Level.

Revision: 6
Layer USCS Soil Classification Horizontal, kh kv/kh Layer
ID (cm/sec) (ft/day) Color
1 Silt w/Sand (ML) 1.0 x 10-4 0.28 0.25
2 Silty Sand (SM) 2.0 x 10-4 0.57 0.25
3 Lean Clay w/Sand (CL) 1.0 x 10-5 0.03 0.25
4 Silty Sand (SM) 7.0 x 10-3 19.85 0.25
5 Well Graded Sand w/Silt (SW) 5.0 x 10-3 14.18 0.25
6 Silty Sand (SM) 1.2 x 10-3 3.40 0.25
7 Sandy Silt (ML) 1.0 x 10-4 0.28 0.25
8 Silty Sand (SM) 4.5 x 10-4 1.28 0.25
9 Poorly Graded Gravel w/Silt and Sand (GP) 5.0 x 10-3 14.18 0.25
10 Lean Clay w/Sand / Silt w/Sand (CL-ML) 1.0 x 10-5 0.03 0.25
a) Geometry Model
Elevation (ft)
70 70
40 40
10 10
-20 -20
-50 -50
-80 -80
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2
Distance (ft) (x 1000)
Total Head Contours Legend

b) Total Head Contours
Elevation (ft)
70 70
40 40
54
52 42
50 38
42
44
10 10
50
46
-20 -20
48
-50 -50
-80 -80
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2
36 46 56
Vertical Gradient Contours Legend

c) Vertical Gradient Contours
Elevation (ft)
70 70
40 40
10 10
-20 -20
-50 -50
-80 -80
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2

-1 -0.5 0 0.5 1
PROJECT NO. DRAWN REVISION

Gyeong-Taek Hong, PhD SEEP/W MODEL SUMMARY
Department of Water Resources URBAN LEVEE GEOTECHNICAL 100 yr Steady State Seepage 0
Division of Flood Management In association with: TASK ORDER ANALYZED
Shah Adil, PE EVALUATIONS PROGRAM Full Ditch Condition
Levee Evaluations Office FIGURE
at Station XXXX+YY
DATE CHECKED
Boring WM0003_XYZB 3-4
2/20/08 Khaled Chowdhury, PE
70 70
a) Geometry Model Elev. 54.95
60 60
50 50
40 40
30 30
20 20
10 10
0 0
-10 -10
Elevation (ft)
-20 -20
-30 -30
-40 -40
-50 -50
-60 -60
-70 -70
-80 -80
1.80 1.85 1.90 1.95 2.00 2.05 2.10 2.15 2.20 2.25 2.30 2.35 2.40
70 70
b) Total Head Contours
60 Elev. 54.95 60
50
Breakout 1.0 ft above toe 50
40 54 40
52 44 44 42
30 50 48 46 44 30
20 20
10 10
Elevation (ft)
0 0 Total Head Contours Legend

50
-10 -10
48
-20 -20
46
-30 -30
-40 -40
-50 -50 36 46 56
-60 -60
-70 -70
-80 -80
1.80 1.85 1.90 1.95 2.00 2.05 2.10 2.15 2.20 2.25 2.30 2.35 2.40
70 Average exit gradient at toe = (3.5)/(18.53)=0.19 (From BTA = 0.37) 70
c) Vertical Gradient Contours Elev. 54.95
60 Average exit gradient at bottom of ditch = (7.77)/(10.54)=0.74 60
50 Average exit gradient = (8.2)/(12.0)=0.68 50
-0.2 -0.2
40 40
0.4
-0 .
30 0.6 30
0.2
0.
0
4
6
20 20 Vertical Gradient Contours Legend
Elevation (ft)
10 10
0 0
-10 -10
-20 -20
-30 -30 -1 -0.5 0 0.5 1
0
-40 -40
-50 -50
-60 -60 Vector Scale 1’’=10 ft
-70 -70
-80 -80
1.80 1.85 1.90 1.95 2.00 2.05 2.10 2.15 2.20 2.25 2.30 2.35 2.40
Distance (ft) (X 1000)
Gyeong-Taek Hong, PhD SEEPAGE ANALYSIS RESULTS
Department of Water Resources URBAN LEVEE GEOTECHNICAL 100 yr Steady State Seepage 0
Division of Flood Management In association with: TASK ORDER ANALYZED
Shah Adil, PE EVALUATIONS PROGRAM Full Ditch Condition
Levee Evaluations Office FIGURE
TASK AREA Section at Station XXXX+YY
DATE CHECKED
Boring WM0003_XYZB 3-5
UNDERSEEPAGE ANALYSIS WITH BLANKET
SECTION 4.0 THEORY
4.0 UNDERSEEPAGE ANALYSIS WITH BLANKET THEORY
4.1 Introduction
This section discusses blanket theory analysis (BTA) of levee underseepage. For soil profiles
as described below, a BTA should be conducted on each seepage analysis cross section
analyzed using SEEP/W. BTA and SEEP/W analyses are considered complementary. BTA
results are used as a quality control check on the SEEP/W analysis results and the SEEP/W
analysis results can be used to improve the BTA and its interpretation, as discussed later in
this section.
A BTA can be applied to a soil profile that can be idealized as a “less-pervious blanket layer”
Revision 06
underlain by a “more pervious layer.” The idealization should be consistent with the blanket
theory assumptions discussed in the following subsection. Blanket theory can then be used
to estimate the head immediately below the landside blanket, the average seepage gradient
through the landside blanket and the excess pore pressures within the blanket. However,
because seepage through the levee and horizontal flow in the blanket are not considered,
blanket theory cannot be used to analyze levee through seepage or seepage along the base
of the levee.
More specifically, blanket theory provides a profile of excess pressure heads along the mid-
depth of the pervious layer. Blanket theory equates these excess heads with the excess
heads along the bottom of the landside blanket, the top of the pervious layer. The head loss
between the mid-depth and the top of the pervious layer centerline and top is thus not
considered, which is conservative.
To be clear, blanket theory deals with excess soil pore water pressure heads, symbolized as
ho or hx, and often referred to simply as “heads.” Heads are measured in feet of water and
convert to equivalent soil pore water pressure when multiplied by the unit weight of water
(62.4 pcf). Excess heads represent pore water pressures in excess of hydrostatic pore
pressures, which are proportional to depth below the water table or phreatic surface. Below
the water table, spatial differences in excess pore pressure are responsible for seepage flow.
Material properties and model geometry used in the BTA analysis should be consistent with
the ones used in the SEEP/W analysis (See Sections 3.5 and 3.6)
4.1.1 Blanket Theory Development and Appendix B
The USACE Waterways Experiment Station developed blanket theory for simplified hand
calculations of underseepage along the lower Mississippi River to ascertain whether design
of seepage berms or other forms of seepage exit control was needed (USACE TM 3-424,
October 1956). Much of the material in TM 3-424 is included in the 1961 ASCE Transactions
Paper No. 3247, “Underseepage and Its Control, A Symposium” by W.J. Turnbull and C.I.
Mansur. Blanket theory has been described more recently in EM 1110-2-1913 (USACE
2000) Appendix B, “Mathematical Analysis of Underseepage and Substratum Pressure.” EM
1110-2-1913 Appendix B is the principal reference for this section, where it is referenced
simply as “Appendix B.”

Revision: 6
SECTION 4.0 THEORY
This section complements and supplements Appendix B, which should be consulted for a
detailed description of blanket theory. It is assumed that the blanket theory analyst has an
operational understanding of the subject material in Appendix B.
4.1.2 Purpose of BTA
On this project, the purpose of the BTA for each seepage analysis cross section is to twofold.
The first purpose, as stated, is to provide a check on SEEP/W results as part of quality
control.
The second purpose is to calculate the excess hyrostatic head “ho” and associated average
vertical exit gradient gradient “io” in the blanket layer at the levee landside toe so that io can
Revision 06
then be compared with gradients empirically correlated with historic levee performance under
high-water flood conditions (ETL 1110-2-569, 1 May 2005). Operationally, io is compared
with performance criterion “ic.” The project has set a criterion of ic equal or less than 0.5,
above which levee underseepage may be of concern. While that concern increases with
increasing io, if io is less than ic=0.5, levee underseepage performance may be generally
considered acceptable.
Heads and gradients beyond the landside toe (hx and ix) are also estimated as part of a BTA.
Results are used both for quality control comparison with SEEP/W results and as part of the
analysis of levee performance under specified water levels.
A BTA is conducted for the 100-year and 200-year water levels, as well as a hypothetical
water level at the top of levee. These are the same water levels used in the SEEP/W
analyses.
In addition to using the same specified water levels, the BTA should be based on the same
soil profile, including levee dimensions and soil permeabilities, as used for the SEEP/W
analysis. The analysis cross section is developed as part of the section’s geotechnical
characterization, which is based on interpretation of available data, including CPTs, borings,
laboratory testing results, past levee performance and so on.
Determination of a BTA profile and permeabilities will generally require that the SEEP/W soil
profile and permeabilities be further idealized to make the BTA maximally consistent with the
assumptions of blanket theory that are summarized in the following section. The
characterization and analysis process requires the consistent exercise of sound engineering
judgment.
Good practice would also suggest that the engineer performing the BTA check that the
geotechnical characterization used for the SEEP/W analysis is consistent with the soil profile
and permeabilities used for the BTA. Any inconsistencies should be reconciled.
4.2 Blanket Theory Assumptions
To a large degree, the accuracy of a BTA depends on how closely the assumptions used to
derive the blanket theory equations approximate actual field conditions. This section provides

Revision: 6
SECTION 4.0 THEORY
an initial discussion of pertinent blanket theory assumptions. Please note that in blanket
theory “waterside” and “riverside” are equivalent terms; the term “waterside” is used
exclusively in this document.
Blanket theory assumes the following.
• There are two continuous horizontal strata below and adjacent to the levee: a “top
stratum,” called the “blanket layer,” that has relatively low vertical permeability and an
underlying “pervious layer” that has relatively high horizontal permeability. The blanket
layer is composed of clays or silts, or soils having vertical permeabilities equivalent to
clays or silts; the pervious layer is composed of sands or gravels, or soils having
horizontal permeabilities equivalent to sands or gravels. The blanket may comprise
several contiguous soil layers in the SEEP/W model; likewise the pervious layer may
Revision 06
comprise several contiguous layers in the SEEP/W model.

• Seepage is laminar flow, vertical through the top layer, the blanket and horizontal
through the pervious layer. Appendix B considers these assumptions reasonable if the
horizontal permeability of the pervious layer is at least 10 times the vertical permeability
of the blanket layer. There is another potentially useful and complementary way to look
at this assumption when SEEP/W results are available. That is, if the seepage in a
particular stratum is primarily horizontal (vertical total head contours), that stratum is
behaving like the pervious layer and should not be considered part of the blanket layer;
conversely, if the seepage is primarily vertical (horizontal total head contours), the
stratum is part of the blanket.
• The pervious layer is of constant thickness and permeability, or can be approximated as
such. The blanket thickness, permeability, or both, can be different on the waterside and
landside. Blankets composed of substrata of different vertical permeability can be
transformed into an equivalent blanket thickness of constant vertical permeability;
pervious layers composed of substrata of different horizontal permeability can be
transformed into an equivalent pervious layer of constant average horizontal
permeability.
• The levee (including impervious berms) and the blanket layer immediately below the
levee are considered impervious. Blanket theory therefore does not address levee
through seepage or horizontal seepage along the bottom of the levee or within the
blanket.
• Seepage may enter the pervious layer via vertical downward flow through the waterside
blanket or through a discontinuity in the waterside blanket that directly exposes the
pervious layer to river flow, thus acting as an “open seepage entrance.” An open
seepage entrance is equivalent to a constant head at the seepage entrance equal to the
river stage. Blanket theory also allows for a “seepage block” that is equivalent to a no-
flow boundary condition. A seepage block can be used for a continuous waterside
blanket where either the pervious layer has a discontinuity and is absent beyond a
certain point or is effectively blocked by an impervious feature (e.g., a thick clay plug). A
seepage block may also be used where the waterside blanket and pervious layer
extends beyond the midpoint of the river flow channel (as discussed in Section 4.3.5). A
seepage block may also be used to model a cutoff wall.
• Seepage may exit the pervious layer via vertical upward flow through the landside
blanket or through a horizontal discontinuity in the landside blanket that results in an
“open seepage exit.” An open seepage exit is equivalent to a constant head at the
seepage exit equal to the elevation of the landside toe. A seepage-block, no-flow

Revision: 6
SECTION 4.0 THEORY
boundary condition can be used for a continuous landside blanket where the pervious
layer is discontinuous or effectively blocked by an impermeable feature (as discussed in
Section 4.3.4).
The accuracy of a BTA depends on how closely the idealized soil profile and permeabilities
meet these assumptions and represent actual field behavior. With some insight and sound
judgment, a BTA can be adapted to less than ideal field conditions, including variable
analysis geometries or permeabilities. There are, however, situations where the blanket
theory assumptions are fundamentally inconsistent with field conditions, making a BTA
invalid. These invalidating conditions are discussed next.
4.2.1 Situations Where Blanket Theory is Invalid

Revision 06
There are situations where blanket theory does not apply and a BTA should not be
conducted. Most generally, blanket theory does not apply where there is no landside blanket
and the levee is founded directly on a surficial pervious layer (e.g., clean or slightly silty
sand). This restriction includes the following cases identified in Appendix B:
• Case 1 – No Top Stratum (i.e., no blanket layer)

• Case 3 – Impervious Riverside Top Stratum and No Landside Top Stratum
• Case 5 – Semi-pervious Riverside Top Stratum and No Landside Top Stratum
Also, although theoretically valid, the Appendix B cases associated with thick impervious
blankets (i.e., zero permeability blankets) are also generally not used for this project. This
includes Appendix B Cases 2 and 4.
Appendix B Case 6 and Case 7 will usually apply on this project. Case 7 is the most general
case, with a semi-pervious blanket layer on both the waterside and landside. Case 6 may be
considered a special case of Case 7 where there is no waterside blanket.
4.3 BTA Guidance

For consistence and quality assurance purposes all BTA calculations should be performed
using the Excel template developed for the Urban Levee Geotechnical Evaluations Program.
A hard copy of the template is provided in Appendix E. An electronic copy of the Excel file
can be obtained from the program website.
For each cross section, the BTA should include an estimate of the excess head “ho” beneath
the blanket at the landside levee toe and the corresponding average vertical exit gradient
through the blanket layer at the levee landside toe, “io.” The analysis should also estimate
the heads landside of the levee toe “hx” and the corresponding average gradients through
the blanket, “ix,” where subscript x represents the landside distance from the levee landside
toe. The maximum x-distance should generally be on the order of 300 to 500 feet.
• The head ho is computed using the procedures in Appendix B, as discussed here,

adapted to the site conditions represented in the cross section. Head hx decreases from
ho in a manner that depends on landside blanket properties (through factor c) and the
landside boundary condition.

Revision: 6
SECTION 4.0 THEORY
• The average vertical exit gradient io is computed as io = ho/zt, where zt is the transformed
thickness of the landside blanket layer for computation of seepage gradients or uplift.
The average gradient landside of the levee toe is computed as ix = hx/zt. Note that the
transformed thickness for gradient computation, zt, should not be confused with zbl, the
transformed thickness of the landside blanket for computation of seepage and pressure
heads as the transformations are done differently in each case.
The following paragraphs discuss technical details of blanket theory, including the input
parameters used in the analysis. For consistency and clarity, Appendix B symbols are used
wherever possible (except upper case X is used for x1 and x3).
4.3.1 Pervious Layer Thickness d and Permeability kf

Revision 06
The thickness of the pervious layer is denoted “d”; horizontal permeability is denoted “kf.”
Thickness d is the arithmetic sum of the in situ thicknesses dn of each of the individual
substrata making up the pervious layer. That is, d = ∑dn, where the sum is over all individual
substrata in the pervious layer. Generally, a substratum represents a distinct and identifiable
zone of horizontal permeability that is expected to have spatial continuity within the pervious
layer.
In some cases it may not be clear if a stratum belongs with the pervious layer or the
overlying blanket layer. This situation can occur where there is a stratum of intermediate
permeability between the blanket layer and the pervious layer. In this case, SEEP/W results
can be examined: if the seepage in the intermediate stratum is primarily horizontal, that
stratum is behaving like the pervious layer and can be considered part of the pervious layer;
conversely, if the seepage is primarily vertical, the stratum is part of the blanket.
The horizontal permeability of the pervious layer kf is the thickness-weighted average of the
horizontal permeabilities of all the substrata in the pervious layer, kfn. That is, kf = ∑dn*kfn/d,
where the sum is over all individual substrata in the pervious layer.
Blanket theory assumes that the pervious layer is continuous, with constant thickness and
horizontal permeability, from the waterside boundary condition (seepage entrance or block)
to the landside boundary condition (seepage exit or block). The blanket layer, however, may
be different on the landside and waterside, as discussed next.
4.3.2 Transformed Blanket Layer Thickness zb and Permeability kb
For computation of excess pressure heads, blanket theory requires an estimate of the
“transformed thickness” of the blanket layer, zb, and the effective vertical permeability of the
blanket layer, kb. Blanket thickness and vertical permeability can differ on the landside and
waterside of the levee.
The landside blanket thickness zbl and vertical permeability kbl are based on soil stratigraphy
and permeabilities at and near the landside levee toe. The waterside blanket thickness, zbr,
and vertical permeability, kbr, are based on soil stratigraphy and permeabilities at and near
the waterside levee toe. If the same relevant blanket layer conditions prevail for both the
waterside and landside blanket (as represented in the SEEP/W cross section), then the

Revision: 6
SECTION 4.0 THEORY
same transformed blanket thickness and associated permeability would generally be used
for both the waterside and landside blanket.
However, the original Waterways Experimentation Station levee underseepage studies found
that near surface clays/silts can be considerably less pervious on the waterside than on the
landside. The reason was attributed to the downward water pressure on the waterside
tending to plug fissures and holes, as opposed to on the landside where upward water
pressure tended to keep the fissures and holes open or worse, eroding the discontinuities or
openings, making them even larger. Another difference between waterside and landside is
that borrow areas are typically on the waterside and usually have a significant affect on
waterside blanket conditions, either thickness of the blanket, or the seepage entrance
condition.
Revision 06
In general, for either the landside or waterside, the transformed thickness of the blanket, zb
(zbl or zbr), is based on transforming the in situ thickness of each substratum in the blanket zn
having vertical permeability, kn, to an equivalent thickness of the least permeable (lowest kn)
substratum in the blanket, kn-min. The sum of the transformed thicknesses is the transformed
blanket thickness, zb. A blanket layer substratum generally represents a distinct and
identifiable zone of vertical permeability that is expected to have spatial continuity within the
blanket layer.
Mathematically, the transformed blanket thickness is zb = ∑zn*kn-min/kn, where the sum is over
all individual substrata in the blanket layer. The landside zb is equated to zbl; the waterside zb
is equated to zbr.
The permeability of the blanket layer over its transformed thickness is kn-min, the minimum
permeability used to compute the transformed blanket thickness. Permeability kn-min is
equated to kb, the “vertical permeability of the top stratum.” The landside kn-min is equated to
kbl; the waterside kn-min is equated to kbr.
Blanket theory assumes that the waterside blanket layer is continuous, with constant
thickness and vertical permeability, from the waterside levee toe to the waterside boundary
condition, either a seepage entrance or block. Similarly, blanket theory assumes that the
landside blanket layer is continuous, with constant thickness and vertical permeability, from
the landside levee toe to the landside boundary condition, a seepage exit or block. The
blanket under the levee, like the levee itself, is assumed impervious.
4.3.3 Landside Transformed Blanket Layer Thickness, zt, For Seepage Gradient
Transformed thickness zt is the actual thickness of all substrata above the base of the least
pervious substratum (having kn-min) plus the transformed thickness of the underlying more
pervious blanket layer substrata. Transformed thickness zt is used in the computation of
average gradient through the blanket layer, io = ho/zt or ix = hx/zt. Transformed thickness zt is
also used for any computation of uplift pressures.
Transformed thickness zt should not be confused with zbl, the transformed thickness of the
landside blanket for computation of seepage and pressure heads. Thicknesses zt and zbl are

Revision: 6
SECTION 4.0 THEORY
numerically equal for single-substratum blankets and for multi-substratum blankets where the
least pervious substratum (kn-min) is at the top of the blanket layer. Thicknesses zt will exceed
zbl for blanket layers where kn decreases with depth—and especially in the case where the
least pervious stratum (kn-min) is at the bottom of the blanket layer. Both zt and zbl are
assumed constant with distance x landside of the levee toe.
4.3.4 Factor c for Landside Blanket and Distance to Effective Seepage Exit, X3
The factor “c” used in blanket theory is computed from the transformed blanket thickness zb
and permeability kb along with the thickness d and permeability kf of the pervious layer;
mathematically, c = {(kb/kf )/(zb*d)}1/2. Because zb and kb can differ between the waterside
and landside, factor c can differ and should be appropriate for the existing conditions.
Revision 06
Factor c for the landside blanket is computed from the landside transformed blanket
thickness zbl and permeability kbl, along with d and kf of the pervious layer; that is, for the
landside blanket layer, c = {(kbl/kf )/(zbl*d)}1/2. Landside factor c is used to compute X3, the
distance from the landside levee toe to the effective seepage exit.
In addition to c, X3 depends on the distance L3 from the landside toe to the landside boundary
condition, either an open seepage exit or a seepage block. Examples of an open seepage
exit are a drainage ditch, or any open ditch, fully penetrating the blanket to the pervious
layer. For an open seepage exit, X3 = tanh{c*L3}/c, where c is based on the landside blanket.
For a seepage block, X3 = 1/(tanh{c*L3}*c). An effective seepage block would be represented

by a condition where drainage from the pervious layer was fully or partially blocked.
Blockage of the pervious layer may result from the natural geologic structure of the deposit,
where the layer thins and pinches out (geologic facies change) or is otherwise discontinuous.
An effective seepage block may also result from a constructed structure that penetrates
through the permeable zone—e.g., a lined channel or building basement. Surface
development that prevents free drainage at the ground surface, such as competent paved
areas, may also act as at least a partial seepage block.
In the absence of an identifiable landside open seepage exit or seepage block, L3 can be set
to infinity, in which case X3 = 1/c. A continuous blanket that extends beyond a distance 2/c
from the toe is effectively infinite, making X3 independent of the boundary condition. That is,
beyond a distance 2/c from the toe, seepage exits (or entrances) and seepage blocks give
approximately the same results for head and gradient estimates. The accuracy of the
approximation increases as the distance L3 increases beyond 2/c.
However, when the BTA is being used to check the SEEP/W analysis, the BTA boundary
conditions should match those used in the SEEP/W analysis. The DWR SEEP/W analyses
commonly use a fixed-head at 2,000 feet from the levee centerline as a default landside
boundary condition. The blanket theory equivalent is to use a seepage exit at L3 equal to
2,000 ft minus the horizontal distance from the levee centerline to the landside toe, xL3toe. For
this case, X3 is computed as follows; X3 = tanh{c*(2000 - xL3toe )}/c, where c is based on
landside blanket.

Revision: 6
SECTION 4.0 THEORY
To summarize, for a landside blanket that is continuous to distance L3 beyond the landside
toe, and using the landside c, the distance to the effective seepage exit X3 is equal to:
• tanh{c*L3}/c, for an open seepage exit at L3

• 1/(tanh{c*L3}*c), for a seepage block at L3
• tanh{c*(2000 - xL3toe )}/c, recommended for the DWR SEEP/W default landside (fixed
head) landside boundary condition
• 1/c = {zbl*kf*d / kbl}1/2, for L3 practically infinite (L3 > 2/c)
4.3.5 Factor c for Waterside Blanket and Distance to Effective Seepage

Entrance, X1
Revision 06
Factor c for the waterside blanket is computed from the waterside transformed blanket
thickness zbr and permeability kbr, along with d and kf of the pervious layer; that is, for the
waterside blanket layer, c = {(kbr/kf )/(zbr*d)}1/2. Waterside factor c is used to compute X1, the
distance from the waterside levee toe to the effective seepage entrance.
Besides c, X1 also depends on the distance L1 from the waterside toe to the waterside
boundary condition, either an open seepage entrance or a seepage block. An open seepage
entrance should be used if the waterside blanket is discontinuous, such that the river
(floodwater) can directly charge the pervious layer. For an open seepage entrance X1 =
tanh{c*L1)}/c, where L1 is the distance from the waterside levee toe to the effective end of the
waterside blanket.
If the waterside blanket is discontinuous, a BTA should use an open seepage entrance for a
discontinuous waterside blanket regardless of the SEEP/W boundary condition.
If there is no waterside blanket, L1 = 0, and Appendix B Case 6 applies. In that case,

effectively, X1 = 0.43d. More generally, X1 = 0.43d may be considered the minimum value of
X1 for L1 = 0 or L1 > 0.
If the waterside blanket is continuous, a seepage block located at the center of the river or
channel can be used to match a SEEP/W no-flow boundary condition. In this case, X1 =
1/(tanh{c*L1)}*c), where L1 is the distance from the waterside levee toe to the center of the
river or channel, as used in the SEEP/W analysis. Using a seepage block or no-flow
boundary condition effectively assumes that there is a flow divide at L1, the channel
centerline, which is theoretically accurate if the flow channel and levee conditions, including
the blanket layer, are continuous and symmetrical along the channel centerline. Otherwise,
using a seepage-block boundary condition could be unconservative, and should therefore be
checked with a more conservative analysis assuming an infinitely continuous waterside
blanket, as elaborated in the next paragraph.
To the extent L1 exceeds 2/c, X1 is effectively independent of the waterside boundary

condition, and approximately equal to the asymptotic (infinite blanket) limit X1 = 1/c.
Therefore, for L1 greater than 2/c, a seepage block (or entrance) will give the same result as
an infinite blanket.

Revision: 6
SECTION 4.0 THEORY
However, for L1 less than 2/c, a no-flow seepage-block boundary condition will underestimate
landside heads and gradients relative to using the asymptotic value X1 = 1/c, consistent with
a continuous blanket having L1 > 2/c. Therefore, for a continuous blanket, a BTA using a
waterside seepage block should also include a complementary analysis assuming an infinite
blanket.
To summarize, for a waterside blanket that is continuous to distance L1 beyond the waterside
toe, and using the waterside c, the distance to the effective seepage entrance X1 is equal to:
• tanh{c*L1}/c, for an open seepage entrance at L1

• 1/(tanh{c*L1}*c), for a seepage block at L1
• 1/(tanh{c*L1}*c), for a SEEP/W no-flow boundary condition used with a continuous
Revision 06
blanket, where L1 is the distance from the waterside toe to the center of the channel
centerline used in the SEEP/W analysis; if the blanket is discontinuous, then an open
seepage entrance should be used; if the seepage block is used to match a SEEP/W no-
flow boundary, a supplementary analysis should assume an infinite blanket (X1 = 1/c)
• 1/c = {zbr*kf*d / kbr}1/2, for L1 practically infinite (L1 > 2/c)
• X1 = 0.43d, for L1 = 0.
Recall that if there is no waterside blanket (L1 = 0) and no landside blanket (L3 = 0), a BTA is
invalid and cannot be made.
4.3.6 At the Landside Levee Toe: Head Below the Blanket ho and Average
Vertical Exit Gradient io
The blanket theory estimate of the excess pressure head at the bottom of the landside toe is
based on parameters X1 and X3, as well as the levee width from waterside toe to landside
toe, L2, and the net head on the levee, H. The net head on the levee, H, is the difference in
vertical elevation between the analysis water level (top of levee, 200-year or 100-year water
level) and the landside levee toe. If the landside toe is in a swale or ditch, the bottom of ditch
or swale should be used to estimate H. The L2 dimension should include any levee berms.
Based on these parameters, the excess pressure head beneath the blanket at the landside
levee toe, ho, is computed as ho = H*(X3/[X1 + L2 + X3]). The average vertical exit gradient
through the blanket at the landside levee toe, io, is computed as io = ho/zt, where zt is the
transformed thickness of the landside blanket layer for uplift.
4.3.7 Beyond the Landside Levee Toe: Head Below the Blanket hx and Average
Vertical Exit Gradient ix
Landside of the levee landside toe, the head below the blanket layer, hx, and the average
vertical exit gradient, ix, through the blanket layer decreases with distance “x” from the
landside toe. The magnitude and pattern of reduction depends on the landside factor c and
the landside boundary condition, as tabulated below. In each case, the ratio hx/ho is
independent of waterside conditions.
• For L1 > 0 and a landside open seepage exit: hx = ho*sinh{c(L3 – x)}/sinh{c*L3}

Revision: 6
SECTION 4.0 THEORY
• For L1 > 0 and a landside seepage block: hx = ho*cosh{c(L3 – x)}/cosh{c*L3}

• For L1 > 0 and a continuous blanket extending more than 2/c from the landside levee
toe: hx = ho*exp{-c*x)}; this is an approximation that improves with the length of the
continuous blanket, independent of actual boundary condition (seepage exit or block)
• For L1 = 0: hx = ho*{(X3 – x)}/X3} (Appendix B, Case 6).
• The average vertical exit gradient through the landside blanket is computed as ix = hx/zt
in each case.
4.3.8 Waterside or Landside Ditches

Drainage ditches or swales run parallel to many levees. Where there is a drainage ditch near
the landside toe, a separate sensitivity analysis should be conducted assuming a surrogate
Revision 06
levee toe at the bottom of the ditch nearest the de facto levee toe, with corresponding
changes to landside blanket thickness, permeability, net head H and levee width, L2.
However, if the landside ditch is fully penetrating through the landside blanket, then the ditch
represents an open seepage exit, with L3 the shortest distance between the landside toe and
the ditch. In this case, the blanket thickness and permeability are based on the blanket
conditions along this shortest distance between the landside toe and the ditch.
On the waterside, a wide ditch with reduced blanket thickness near the levee toe may act as
the major seepage source into the pervious zone. In this case, it may be appropriate to
conduct the BTA using the waterside blanket thickness and permeability below the ditch or
swale in the estimation of c and X1. Also in this case, X1 based on the ditch may be
increased by the distance from the waterside levee toe to the nearest edge of the ditch. In no
case, however, should the X1 estimate used to calculate ho and hx exceed X1 calculated by
ignoring the waterside ditch.
If the waterside ditch is fully penetrating through the waterside blanket, then the ditch
represents an open seepage entrance, with L1 the shortest distance between the waterside
toe and the ditch. In this case, the blanket thickness and permeability are based on the
blanket conditions along this shortest distance between the waterside toe and the ditch.
In all cases where there is a landside or waterside ditch (swale or excavation), the SEEP/W
analysis results should be examined to help judge the effect of the ditch on seepage heads
and gradients. This information can be used to refine the BTA.
4.4 BTA of SEEP/W Cross Sections
To reiterate, a BTA should be made for each cross section analyzed by SEEP/W as a quality
control check. The BTA should maintain as closely as possible the same levee dimensions
and soil profile geometry and permeabilities as used for the SEEP/W analysis. At the same
time, the BTA should be consistent with assumptions of blanket theory. This will generally
require that the BTA profile and permeabilities be further idealized compared with the
SEEP/W analysis.

Revision: 6
SECTION 4.0 THEORY
For example, multiple soil strata in the SEEP/W cross section that are functionally part of the
blanket will have to be transformed to an equivalent blanket thickness having the same
vertical permeability as the most impervious stratum in the blanket. Similarly, an equivalent
horizontal permeability will have to be calculated for multiple soil strata in the pervious layer.
The SEEP/W and BTA analyses should be considered complementary. In general, if

SEEP/W results show the seepage direction in a particular stratum to be primarily horizontal
(vertical head contours), that stratum is behaving like a “pervious layer.” Conversely, if the
seepage direction is primarily vertical (horizontal head contours), the stratum is behaving like
a blanket layer. For BTA, strata with predominant horizontal seepage should not be included
in blanket layers; and strata with predominant vertical seepage should not be included in
pervious layers. Examination of SEEP/W results can thus be used to help refine the BTA or
Revision 06
its interpretation, even as the BTA is used as a quality control check on the SEEP/W results.
This complementarity can be useful, for example, if there is a stratum of intermediate

permeability between the blanket layer and the pervious layer and it is not be clear if it
belongs with the pervious layer or the overlying blanket layer. If the SEEP/W results show
the seepage in the stratum as primarily vertical, that stratum is behaving like the blanket
layer and can be considered part of the blanket layer. However, if the seepage is primarily
horizontal, the stratum is part of the pervious layer and effectively is the pervious layer if the
seepage is entirely horizontal.
A continuous landside blanket layer overlying two permeable strata being charged on the
waterside but separated by a relatively low-permeability stratum is another example where
SEEP/W results could be sensibly used to help configure the BTA. If the seepage in the low-
permeability stratum is predominantly horizontal, it is part of the pervious layer, which also
includes the two permeability strata if their seepage is also horizontal. And plainly, combining
the thicknesses of all three strata as the pervious layer is a conservative assumption for a
BTA. However, if the seepage direction in the low-permeable stratum is predominantly
vertical, it would be inaccurate and it could be overly conservative to combine the three
strata, or even to combine the two permeable strata into the BTA pervious layer.
Using the BTA and SEEP/W analyses in a complementary fashion can improve the overall
quality of the seepage analyses. Examination of SEEP/W results can be used to help refine
the BTA or its interpretation where needed, even as the BTA is used as a quality-control
check on the SEEP/W results.
4.4.1 Boundary Conditions
It is important that the BTA boundary conditions are effectively consistent with the SEEP/W
boundary conditions.
• The blanket theory equivalent to a SEEP/W fixed-head boundary condition is either a

seepage entrance for the waterside or a seepage exit for the landside. SEEP/W
analyses commonly use a landside fixed-head boundary condition.
• The blanket theory equivalent to a SEEP/W no-flow boundary condition is a seepage
block, if the blanket is continuous to the boundary condition. SEEP/W analyses

Revision: 6
SECTION 4.0 THEORY
commonly use a no-flow boundary condition at the middle of the river or flow channel.
For SEEP/W analysis, a no-flow boundary condition is clearly appropriate if the blanket is
discontinuous, such that the river flow can directly charge the pervious layer between the
end of the waterside blanket and the middle of the river. For BTA, however, the correct
boundary condition would be a seepage entrance at the end of the waterside blanket,
where the discontinuity occurs.
For the special case that the waterside blanket is continuous below the entire flow channel,
such that the pervious layer can only be charged by seepage through the blanket, the BTA
could appropriately assume an infinite waterside length (i.e., L1 is infinite making X1 = 1/c).
However, case-specific judgment will be required to match the BTA boundary condition to
the boundary condition used in SEEP/W analysis, which may be a fixed-head or no-flow
boundary condition.
Revision 06
A BTA is practically independent of boundary conditions that are beyond a distance of two
times the asymptotic distance 1/c from the levee toe. That is, beyond a distance 2/c from the
toe, seepage entrances or exits and seepage blocks give approximately the same results for
head and gradient estimates; the accuracy of the approximation increases as the distance
(L1 or L3) increases beyond 2/c.
4.5 Pore Pressures Near The Landside Toe
Particularly for the purpose of slope stability, it should be noted that blanket theory can be
used to provide approximate estimates of pore pressures in the landside blanket and the
pervious layer. It can also be used to make crude estimates of pore pressures below the
levee. These estimates could be used to compare with or help check pore pressures
estimated using SEEP/W for use in slope stability analysis. Pore pressures based on blanket
theory can also be used directly in slope stability analyses if desired. However, for slope
stability, a serious drawback of blanket theory is that, because the levee and underlying
blanket are assumed impermeable, effects of levee through seepage are not considered.
4.5.1 Pore Pressures in the Landside Blanket
Heads or equivalent pore pressures within the landside blanket layer can be estimated using
blanket theory by recognizing that blanket theory assumes uniform vertical flow through the
horizontally layered blanket. The head on the underside of the blanket layer starts at hx (or ho
at the toe). The head then decreases linearly through each substratum (of uniform vertical
permeability) in the blanket, dropping to zero at the top of the blanket, the ground surface.
The linear head gradient in each stratum is inversely proportional to the substratum’s vertical
permeability, which is assumed constant within the substratum. Heads are converted to pore
pressures by multiplying by the unit weight of water. The mathematical development is
summarized in the following paragraphs.
To maintain mass continuity, the specific discharge (Darcy velocity, v = k*i) must be constant
through the blanket layer, including each substratum in the layer having constant spatially
averaged vertical permeability kn and hydraulic gradient in. This yields vn = kn*in. The
hydraulic gradient is assumed constant within each substratum because the specific
discharge and permeability are assumed constant, and in = vn/kn. Therefore, while the

Revision: 6
SECTION 4.0 THEORY
specific discharge is the same in each substratum, the gradients vary inversely with the
permeability, yielding relatively higher gradients in the lower permeability substrata and lower
gradients in the higher permeability substrata.
The specific discharge through each substratum, vn = kn*in, is equal to the specific discharge
through the entire layer, v. This yields vn = kn*in = v = k*(ho/z), where k is the “equivalent
vertical permeability” of the blanket layer and (ho/z) is the average gradient through the
blanket. The equivalent vertical permeability k should not be confused with kb, which is
based on the minimum substratum permeability; and z, the actual in situ thickness of the
blanket layer, should not be confused with the transformed thickness zb or zt.
Also, the head losses in each substratum hn is the product of the in situ substratum thickness
Revision 06
and gradient; that is, hn=zn*in, which must sum over all the substrata to ho at the levee
landside toe, or hx at distances x beyond the landside toe. This yields ho = ∑zn*in.
Combining vn = kn*in= k*(ho/z) and ho = ∑zn*in yields the equivalent vertical permeability of the
blanket k = z/∑zn/kn. It may be noted that k is the harmonic sum of the individual substratum
permeabilities. It thus follows that:
The gradient in each substratum is in = (k/kn)(ho/z)
The head loss in each substratum is hn = zn*in = ho(zn/z)(k/kn).
The heads within a blanket layer are illustrated for the case where the blanket layer is
composed of three uniform substrata. From top to bottom, the substrata are of in situ
thicknesses z1, z2 and z3 having vertical permeabilities k1, k2 and k3. Substratum 1 is at the
top of the blanket; substratum 3 is at the bottom; substratum 2 is in the middle. The head at
the bottom of the blanket, the bottom of substratum 3, is ho. The head at the top of
substratum 1, the landside ground surface, is zero.
Table 4-1 gives the heads at the top and bottom of each layer; heads between these points
vary linearly with depth. Extension to blanket layers having other numbers of substrata follow
the pattern illustrated in Table 4-1.
Table 4-1. Heads in Three-Stratum Blanket Layer.

Location in Blanket Layer Corresponding Head
Ground surface, top of substratum 1 hx{1 – (z3/z)(k/k3) – (z2/z)(k/k2) – (z1/z)(k/k1} = 0
Bottom of substratum 1, top of substratum 2 hx{1 – (z3/z)(k/k3) – (z2/z)(k/k2)}
Bottom of substratum 2, top of substratum 3 hx{1 – (z3/z)(k/k3)}
Bottom of blanket, bottom of substratum 3 hx
The pore pressures within any blanket substratum can be modeled using 1.) an “equivalent
heavy fluid” having a unit weight equal to (1+in) times the unit weight of water and 2.) an
associated piezometric surface that yields the pore pressure at the top of the substratum,
estimated using the procedure outlined in the preceding paragraphs. This approach yields a
head loss in the layer of hn = zn*in = ho(zn/z)(k/kn), where the gradient is in = (k/kn)(ho/z).

Revision: 6
SECTION 4.0 THEORY
4.5.2 Pore Pressures Below the Levee
Below the levee, heads and corresponding pore pressures in the blanket layer and pervious
layer could be approximated by the hydraulic gradeline (HGL) measured along the mid-depth
of the pervious layer. The HGL is linear and generally applies only under the levee. The HGL
has slope M = H/(X1 + L2 + X3) and is equal to ho at the landside toe. The HGL does not
apply for heads beyond the landside levee toe, where hx applies, as discussed in the
previous section.
As a point of comparison, it may be noted that a projection of the HGL beyond the landside
toe shows a steeper reduction in head with distance from the landside toe than the
reductions given by hx. It may also be noted that if adequate piezometer data collected
Revision 06
during sustained high water are available to empirically estimate the HGL, BTA can use the
empirical HGL and measured heads to estimate X1 and X3 and thus ho (and potentially hx) as
a function of net head H. An empirical HGL requires that artesian steady state flow
conditions have developed beneath the levee.
For clarity of and to facilitate independent reviews BTA cross section information and BTA
analysis results will be presented in a table format as shown in Table 4-2.

Revision: 6
SECTION 4.0 THEORY
Table 4-2. Summary Table for Blanket Theory Analysis.

Revision 06

Revision: 6
SECTION 5.0 STATIC SLOPE STABILITY ANALYSIS
5.0 STATIC SLOPE STABILITY ANALYSIS
5.1 Objective
Evaluate the waterside and landside slope stability of the levees for the DWR Urban Levee
Geotechnical Evaluations Program. Evaluation of levee landside stability will be performed in
phases as shown in Figure 5-1. An initial evaluation will be performed during the P1GER
work phase to identify levee reaches that meet program criteria. Reaches that do not meet
criteria will be re-evaluated during the GER work phase.
5.1.1 Analysis Type
Two-dimensional limit-equilibrium slope stability analyses utilizing Spencer’s method.

Revision 06
5.1.2 Software
UTEXAS4 (Shinoak Software, swright@shinoak.com)
SLOPE/W 2004/2007 (GEO-SLOPE International Ltd, info@geo-slope.com)
To maintain consistency across the teams, the UTEXAS4 computer program is to be used
for all analysis cases. The SLOPE/W computer program may be used to verify results from
key analysis cases that have already been analyzed with UTEXAS4.
5.2 Water Level Conditions
Steady-state seepage pore pressures todate (March, 2008) have been calculated by
SEEP/W for the following water level conditions:
• Top of Levee or 200-year (median) flood level plus 3 feet, whichever is lower
• 1957 design water surface for the Sacramento River basin or
• 1955 design water surface for the San Joaquin River basin
It is currently planned to change the above recurrence flood level elevations approved and
use water levels ranging from the top of the physical levee to multiple lower levels. Water
surface elevations used for each study area will be documented in the P1GDR and GDR.

Revision: 6
Revision 06
Figure 5-1. Analysis Flow Chart.

Revision: 6
5.3 Loading Conditions
5.3.1 Steady State Seepage (Landside Only)
The static stability condition represents the long-term or short-term (whichever is more
critical) stability of the levee under steady-state pore pressure conditions. For this case, the
lower of the effective stress drained or consolidated undrained shear strength is used for all
materials. For example, for normally consolidated clays where the drained strength is higher
than its undrained strength, consolidated undrained strength is used. And for over-
consolidated clays, where the drained shear strength is lower than its undrained strength,
drained strength, is used. This is accomplished using a unique strength envelope called the
“tri-linear” strength envelope which is discussed in Section 5.5.2.1 and in Appendix A.
Revision 06
5.3.2 Rapid Drawdown (Waterside)
The rapid drawdown condition represents the stability of the levee following a rapid lowering
of water level from a sustained high water level condition. The high and drawn-down water
levels need to be established based on site-specific gauging data records and/or hydrologic
and hydraulics studies. For areas where high water level information is not available, the
high water level may be assumed to be at the 100-year (median) flood level. The drawn-
down water level for these areas should be established based on reason and judgment.
A simplified drawdown (single stage) analysis should first be performed using the lower of
the drained and undrained strength envelopes. The strength envelope used in this simplified
approach, for the fine-grained materials, would essentially be the same tri-linear envelope
developed for the static stability analyses. The piezometric surface used in this simplified
analysis would correspond to the drawn-down water level on the waterside and to the high
water level condition within the levee. The ICB has suggested considering inspection of past
performance looking closely at slopes steeper than 3H:1V.
For those cases where this simplified drawdown analysis does not provide conclusive results
(good or bad), a more detailed analysis should be performed using the full three-stage
analysis procedure (Duncan, Wright and Wong, 1990) available in UTEXAS4.
5.4 Minimum Factor of Safety Criteria
The following criteria for minimum FSs are adopted for the following loading conditions:
• Static stability for landside shear surfaces: minimum FS equal to or greater than 1.40
(please note that a computed FS value of 1.36 or 1.39 may not be rounded up to a 1.40
value)
• Rapid drawdown condition: minimum FS between 1.0 and 1.20 depending upon the
duration of the high (initial) water level conditions.
The analysis will be limited to identify critical surfaces passing through the crest of the levee.
Surfaces day lighting through the side slope of the levee will be considered shallow

Revision: 6
incidents; they will not be considered slope failures but shall be given operational
consideration.
The ICB has recommended that cross section geometries with waterside slopes steeper than
2H:1V and landside slopes steeper than 2.5H:1V need to be flagged.
5.5 Material Properties
The material and strength properties of the levee and foundation soils will be developed from
the results of the various field and laboratory tests along with the historical data. It is
anticipated that not all tests, because of inherent procedural differences, would provide the
same strength parameters for any given soil layer. Therefore, engineering judgment would
be required to select the representative strength values for each soil layer. The thought
Revision 06
process behind the selection of strength values and the utilization and/or discarding of test
results should be thoroughly explained in the P1GER and GER.
5.5.1 Free-Draining Versus Non-Free-Draining Materials
As a first step, a distinction is made between free-draining materials and non-free-draining

materials. Free-draining materials are defined as coarse-grained materials with no or very
few fines, such that when sheared these materials would not generate excess pore water
pressures. Free-draining materials are assumed to remain drained and their shear strength
is characterized by effective stress drained parameters (c’ and phi’) for all loading conditions.
Non-free-draining materials are defined as coarse-grained or fine-grained materials with a
significant amount of fines, such that when sheared these materials would generate (and
sustain) excess pore water pressures. The shear strengths of saturated non-free-draining
materials are characterized as the lower of their drained or undrained shear strengths with
the tri-linear envelope.
5.5.2 Strength Selection for Non-Free-Draining Materials
While the drained shear strength of saturated non-free-draining materials (CH, CL, ML, etc.)
is relatively independent of the soil’s past history, the consolidated undrained strength of
these materials is heavily dependent upon the past history ( i.e., maximum past stress etc.).
Depending upon the over-consolidation ratio (OCR), the consolidated undrained strength of
a fine-grained material could be higher or lower than drained strength. For example, for a
normally-consolidated clay, its drained strength is usually higher than its undrained strength,
whereas for a highly over-consolidated clay, its undrained strength is generally higher than
its drained strength. For the undrained strength to be higher than the drained strength,
negative pore pressures (suction) need to be generated and sustained during shearing.
While evaluating the static stability, for short-term or long-term conditions, generation and
sustainment of negative pore pressures cannot always be depended upon. Therefore, it is
generally considered prudent to use the lower of the drained and undrained strength
envelopes to characterize the shear strength of over-consolidated (or dense) materials for
static stability analyses. On the other hand, for yield acceleration estimation, it may be
acceptable to count on negative pore pressures. This is because the duration of seismic

Revision: 6
loading is generally so short (less than a minute or so) that there is almost no potential for
the negative pore pressures to dissipate before shaking is finished.
5.5.2.1 Tri-Linear Strength Envelope
In accordance with the logic described above, a simplified procedure has been developed to
characterize the shear strength of non-free-draining, fine-grained materials for the various
loading conditions. The graphical representation of this tri-linear envelope approach is
presented in Figure 5-2. Three strength envelopes are shown on this figure:
1. Drained strength envelope;

2. Undrained strength envelope for the normally consolidated condition (OCR=1.0) and;
Revision 06
3. Undrained strength envelope for an over-consolidated condition based on the Stress

History and Normalized Soil Engineering Properties (SHANSEP) approach.
The procedure for estimating the drained and undrained strength envelopes and the
guidelines for developing the recommended strength envelopes for various loading
conditions are discussed in detail in Appendix A.
In most cases, use of the tri-linear envelope eliminates the need to subdivide a foundation
clay layer into zones (in the horizontal direction) when moving from beyond the toe to
underneath the crest of the levee. This is true for situations where a constant value of
maximum past stress can be assumed for the entire layer. For other situations where the
maximum past stress varies within the layer, the layer needs to be subdivided into zones and
a different separate tri-linear envelope needs to be developed for each zone. In either case,
the tri-linear envelope, once developed based on the maximum past stress value (before the
flood event), automatically accounts for the increase in OCR with rise in the water level
associated with a flood event. See Attachment A1 to Appendix A for further elaboration of
this concept.
Depending upon user preference, if the tri-linear envelop approach is judged too complex,
the conventional SHANSEP approach could instead be used to assign the appropriate shear
strength values to the saturated materials. For this alternative approach to be fully
compatible with the tri-linear approach, the following two basic conditions must be checked:
• The undrained strength must never be higher than the drained strength.
• The increase in the OCR value with rise in the water level (caused by a decrease in the
effective stress) should appropriately be accounted for in the SHANSEP equation, for
the (different) evaluated flood conditions.
The undrained strength (Su or Su/p’) estimated from the triaxial isotropically consolidated
undrained (CIU) tests or from the CPT is recommended to be reduced by 25 percent to
convert the triaxial compression-type strength to direct simple shear (DSS) type strength and
also to account for the rate of loading effects (see Appendix C, Figure C-3).

Revision: 6
Highly Slightly Normally

Over-Consolidated Over-Consolidated Consolidated
Shear Strength along Failure Plane
Drained Strength
NC Strength
SHANSEP Strength
Trilinear Envelope
Revision 06
Maximum Past
Pressure,
σ'B σ'v,max
Effective Stress on Failure Plane
Figure 5-2. Tri-Linear Envelope Approach.
5.5.3 Step-by-Step Procedure for Selecting Strength Properties
The step-by-step procedure to characterize and select the strength properties of the various
materials for input into the stability models is as follows:
1. Develop cross section geometry summary plot. This plot presents a summary of all
available test results that may be used for characterizing the various soil layers and
selecting the material strength values. A sample plot is shown in Figure 5-3. Data
presented on this plot includes, N1(60) values, fines content, water content, cone tip
resistance (qt) , friction ratio (FR) and water pressure profiles, if available. Interpreted
stick logs are also shown on this summary plot.
2. Develop idealized subsurface stratigraphy. Using the information presented on the
cross section geometry summary plot, identify the extents of levee and foundation
materials with similar characteristics. Additional information, if available, may also be
used to help with subsurface characterization. Plot the idealized stratigraphy lines on the
cross section geometry summary plot. While developing the subsurface stratigraphy of a
cross section, information from adjoining cross sections should also be looked at to
identify soil layers with similar characteristics.

Revision: 6
Guidance Document for Geotechnical Analyses
Rev 6
Figure 5-3.
Example Cross Section Geometry Summary Plot.
3. Estimate material strength values from each type of test. Using the correlations
presented in Appendix C, estimate drained and consolidated undrained strength values
for each layer corresponding to each type of test (SPT, CPT, Atterberg limits, etc.). The
strength values from the laboratory triaxial tests are estimated directly from the tests
results. For estimating the consolidated undrained strength from CIU triaxial test results,
use the graphical approach presented in Figure 5-4. The undrained strength (Su or
Su/p’) estimated from the triaxial CIU triaxial tests or from the CPT test is recommended
to be reduced by 25 percent to convert the triaxial compression type strength to DSS
type strength and also to account for the rate of loading effects.
Different strength values may be estimated for the same soil layer from different types of
tests. Enter these individual strength values into the matrix presented in Table 5-1 under
the appropriate type of test column. It is recognized that some of the cells in Table 5-1
Revision 06
matrix will have no data since not all test results may be available for all soil layers.
Additional guidelines recommended for selecting the strength properties are as follows:
• For clays (CH, CL), avoid assigning drained friction angles greater than 31 degrees.
• For sands (SP, SW, SM) and silts (ML), avoid assigning drained friction angles
greater than 35 degrees.
• A cohesion intercept of 100 to 200 psf may be used for cohesive materials above the
phreatic surface.
4. Estimate the representative strength values for each soil layer. The purpose of this
task is to select a single representative strength value for each soil layer based on the
information contained in the Table 5-1 matrix. Since the individual strength values
estimated for a layer from different tests may not be the same, engineering judgment
would be needed to select a single representative value for a particular layer. Criteria for
making the judgment may include, confining stresses, style of shearing, confidence or
lack of confidence in a certain test result and additional information on the soil type
outside of what is shown in Figure 5-3 and Table 5-1, including visual or historical
knowledge of the soil, if available. For situations where very limited data is available for a
certain soil layer, information from adjoining cross sections may be reviewed to see if the
same soil type is present at those adjoining locations. This cross referencing of
information from one cross section to the other is very beneficial in maintaining
consistency and/or identifying discontinuities across and between reaches. The rationale
adopted for selecting the representative value would be listed in the last column of Table
5-1.

Revision: 6
It should be noted that at this stage, it is not required to use different types of strengths along
different parts of the failure surface. It is recognized that, it may be more appropriate to use
the triaxial compression strength along the upper part of the shear surface, the direct shear
strength along the middle and the triaxial extension strength towards the toe of the shear
surface. However, considering the level of accuracy sought in P1GER or GER studies, the
amount of natural variability in material characteristics along the levee alignment and the
rather limited amount of test data available, it was decided not to unduly complicate analyses
and use direct simple shear type strength along the entire length of the failure surface. The
rationale behind this simplification was that the direct simple shear strength is a reasonable
average of the triaxial compression, direct shear and the triaxial extension strengths. And
therefore, by using this single type of strength along the entire failure surface would not
induce a significant bias in the analysis results. Having said this, it is recognized that this
simplification may not always be appropriate for all situations.
Revision 06

Revision: 6
Table 5-1. Example of Strength Values Matrix.

Revision 06

Revision: 6
Revision 06
Figure 5-4. Estimates of Consolidated Undrained Strength from Triaxial CIU Test Results.

Revision: 6
For example, if the shear surface stays entirely within the embankment and is predominantly
steeply inclined, then it would be more appropriate to select a strength value that is closer to
the triaxial compression strength rather than the DSS strength as an average value for the
entire shear surface. For these special cases, it is suggested that judgment be used in
selecting the appropriate average strength value for the entire shear surface. For these
situations where a strength value other than the DSS strength is selected as an average
strength value, the rationale behind this selection must be properly explained within the
report and on the results figure.
5.6 Analysis Approach
5.6.1 Static Stability

Revision 06
For the static stability loading condition it is assumed that:
1. Steady-state seepage has developed under the selected water level condition.
2. The levee and the underlying soils are fully consolidated under the steady-state phreatic
surface.
3. During shear, the free-draining materials will mobilize effective stress drained strength
and the non-free-draining materials will mobilize the lower of the drained or undrained
strength. In other words, reliance on negative pore pressures is not allowed.
The step-by-step approach adopted for completing the slope stability analyses for the static
stability condition is as follows:
1. Develop UTEXAS4 model geometry

• Use the cross section geometry developed from survey data and boring/CPT
investigations. A typical cross section geometry summary plot is shown in Figure 5-3.
• The UTEXAS4 model geometry, including internal stratigraphy, should be consistent
with, and to some extent identical to, the corresponding SEEP/W model geometry.
2. Import steady-state pore pressures calculated by SEEP/W into the UTEXAS4
model
• Use pressure head (in psf) values (Option 5, pressure interpolation, in UTEXAS4).
• Also, import the phreatic surface line calculated by SEEP/W into the UTEXAS4 cross
section geometry plot. This line is used to differentiate between the saturated
materials and the unsaturated materials. If needed, saturated and unsaturated
materials can be modeled with different material numbers in the UTEXAS4 input file.
If the phreatic line passes through a non-free-draining material, the portion of the
material above the phreatic line can be modeled with a different material number
than the portion below the phreatic line. This could be needed when the materials
above the phreatic line are modeled with a small amount of cohesion, or when the
material below the piezometric line is modeled with the tri-linear envelope, and the
material above with drained friction angle. No such sub-division or distinction is made
for free-draining materials, which are assigned the drained strength envelope both
above and below the phreatic line.
• The pore pressures in the UTEXAS4 file are calculated from the pressure head
values calculated by the SEEP/W model and not from the phreatic line.

Revision: 6
3. Assign shear strength envelopes to all materials

• All coarse-grained free-draining materials, both saturated and unsaturated, are
assigned effective stress drained friction angles with negligible cohesion intercepts.
Use a non-linear strength envelope (Option 6, in UTEXAS4) equivalent to the
conventional c – phi strength (Mohr-Coulomb) in the positive effective stress portion
of the envelope, and zero strength in the negative effective stress. This is done so
that so that if negative stresses are obtained, the soil strength for those slices is
computed as zero and not as a negative value.
• All fine-grained non-free draining materials that are saturated (below the phreatic
surface) are assigned a non-linear strength envelope (Option 6, in UTEXAS4
manual) developed in accordance with the procedure presented in Appendix A.
• All fine-grained non-free-draining materials that are unsaturated (above the phreatic
surface) may be assigned the effective stress drained or trilinear envelopes.
Revision 06
4. Compute the minimum FS

• Evaluate both waterside and landside shear surfaces using the automatic search
routine of UTEXAS4.
• Check both circular and non-circular (wedge) shear surfaces.
5. Perform sensitivity analyses for the potential range of water level conditions (to be
performed in the GER).
• Present results in the form of a chart (or graphs) to demonstrate the variation in FS
with different water levels. An example plot is shown in Section 5.10.
5.6.2 Rapid Drawdown
This loading condition assumes that:
1. The high water level would remain at the high level long enough to develop steady-state
seepage conditions and that the levee and the underlying soils are fully consolidated
under this condition.
2. The drop in the water level from the high water level to the drawn-down level is rapid and
the fine-grained materials will not experience any change in stress conditions due to this
drop.
3. The rapid drop in water level could induce a rapid undrained failure towards the
waterside in the saturated fine-grained materials.
As discussed earlier, it is suggested that a simplified drawdown (single stage) analysis be
first performed for all sections. The material strength envelopes used in this simplified
drawdown analysis should be the same envelopes that are used in the static stability
analysis. However, the phreatic surface used in this simplified drawdown analysis would be
different than the one used in the static analysis and would roughly correspond to the
condition shown below:

Revision: 6
For those cases where this simplified drawdown analysis does not provide conclusive results
(good or bad), a more detailed analysis should be performed using the full three-stage
analysis procedure (Duncan, Wright and Wong, 1990) available in UTEXAS4. In these
cases, the step-by-step approach adopted for completing the slope stability analyses using
the three-stage analysis procedure for the rapid drawdown condition is as follows:
1. Develop UTEXAS4 model geometry.

Use cross section geometry developed from survey data and boring/CPT investigations.
Same as that developed for steady-state seepage analysis.
Setup the UTEXAS4 data file for three-stage analysis.

Revision 06
2. Identify two water levels – high water level and drawn-down water level.
For the high water level (first stage input), import steady-state pore pressures calculated
by SEEP/W into the UTEXAS4 model.
For the drawn-down water level (second and third stage input), use a piezometric line.
No SEEP/W analysis is performed for the drawn-down water level conditions.
3. Assign drained and consolidated undrained shear strengths (Kc = Kf envelope and
the Kc = 1 envelope).
Assign effective stress drained strengths to all materials in the first-stage material
property data blocks.
Assign effective stress drained strengths to all free-draining materials in the second-
stage material property data blocks.
Assign effective stress drained strengths to all non-free-draining materials that are
situated above the phreatic line in the second-stage material property data blocks.
Assign both drained and consolidated undrained strengths to all saturated non-free-
draining materials that are expected not to drain during the rapid drawdown. These two
strength envelopes (Kc = Kf envelope and the Kc = 1 envelope) are input in the second-
stage material property data block.
4. Compute FS.
Perform a three-stage analysis (not two-stage)
Evaluate waterside shear surfaces only.
Check both circular and non-circular (wedge) shear surfaces.

Revision: 6
5.7 Modeling Cracks
Depending on the material type, cracks should be modeled in the stability analyses. Cracks
can develop in the field and therefore, the strength of the soil in these areas should not be
accounted for. It may also be necessary to model tension cracks to avoid negative normal
stresses at the interfaces between slices in the stability calculations. The following guidelines
are recommended:
• It is recommended that during site investigation notes be made regarding observed

cracks in the field. If there is evidence that deep cracks developed, the measured depth
of these cracks should be used as the minimum crack depth in the analyses.
• As a minimum, the following guidelines are recommended for water filled tension crack
depths:
Revision 06
− For CH materials 6 feet.

− For CL, MH and ML 3 feet.
− For sands and gravels no tension crack is expected.
• For stability analyses, all cracks must be filled with water to the top of the crack.
• After any analysis, it is necessary to verify that the inter-slice normal forces are not
negative (Table No. 59 in UTEXAS4 output file). If this is the case, the depth of the water
filled tension crack should be increased until no negative inter-slice normal forces are
found
5.8 Additional Guidelines
The following additional guidelines were developed based on the program team’s experience
with UTEXAS4 and suggestions made by the ICB following review of interim stability
analyses results. These guidelines are intended to further help with and improve the quality
of the stability analyses and the presentation of results.
• Use the same coordinate system in the UTEXAS4 model as the one used in the
SEEP/W model.
• Document the basis of selection for all input parameters.
• For presentation of results, at a minimum, present:
− Model geometry with critical shear surface and material zones
− Tabulated summary of material properties (for materials in which the tri-linear
envelope was used, include the three primary properties of the envelope: phi’, m
and maximum past pressure, σ’v,max)
− Both circular and non-circular critical shear surfaces plotted on the same figure
• When presenting results of stability analyses, show the shear strengths actually used in
the analyses for the various materials in a tabular form on the same figure or as strength
envelope plots on the following figure.
• When presenting results of stability analyses, instead of showing only the most critical
shear surface, show several types of potential shear surfaces that may be significant for
other reasons including compromising the crest, involving a significant portion of the
landside berm, or passing through a relatively weak foundation layer etc. Refrain from

Revision: 6
showing trivial shear surfaces that are localized artifacts of properties and idealized
model geometry. Trivial shear surfaces are typically small and shallow shear surfaces
that are not anticipated to jeopardize the performance of the levee section. This does not
mean, however, that small instabilities should be neglected if they are judged to occur in
positions and to represent situations where progressive and repeated sloughing/erosion
can evolve into a potentially dangerous situation. Engineering judgment is needed for
this determination.
• Shallow slumps/slides on the waterside due to rapid drawdown are typically largely an
operational issue as they do not typically threaten the breaching of the levees. An
exception is when these are deep-seated failures that significantly compromise the levee
crest section. Exactly what represents significant is difficult to rigorously define and
therefore engineering judgment is very much needed on a case by case basis to make
this decision.
Revision 06
5.9 Past Performance Verification
Agreement between past performance and results from the back analyses of past conditions,
is the best possible check on the reasonableness of the model input parameters. To take
advantage of past performance, the following recommendations are made:
• In areas where past slope instability problems have been observed, a back analysis
should be performed using the estimated water level that existed at the time of the
observed problems. If the analysis results seem to agree with the past observations,
these findings should be clearly documented and discussed in the report. If, even with
reasonable adjustments in the input parameters, the model results do not agree with the
past observations, these findings should also be reported with appropriate
explanation/recommendation.
• In areas where the model results, for the 100- and 200-year events indicate FS values
less than 1.0, but the past performance data does not seem to indicate any historical
problems, a back analysis should be performed using the estimated highest water level
that that area has ever experienced in the past. If the results from this back analysis also
seem to indicate potential problems (which have not actually been observed in the field),
the following set of analyses should be completed to investigate/explain this apparent
discrepancy.
− Rerun the back analysis by assuming hydrostatic pore water pressure distribution
below the calculated phreatic surface (instead of the SEEP/W calculated pore
water pressures). If this significantly increases the FS and the new FS value is
above 1.0, this means that the initial low FS value was caused by excessively
high pore water pressures and not by the use of excessively conservative
strength parameters. This important finding must be presented and discussed in
the report. In reality, the high pore water pressures may have been dissipated in
the field due to boils or other pressure relieving mechanisms, making conditions
closer to the hydrostatic scenario than to the SEEP/W scenario. Generally,
sections where such a condition occurs have high average exit gradients
calculated by SEEP/W. Therefore, these sections would require seepage
mitigation measures irrespective of the stability issues.
− If there is no significant increase in the FS with the use of hydrostatic conditions,
it means that, more than likely, either the stratigraphy of the section is not well
defined or the strength parameters used in the model do not truly represent the
field soil conditions. In either case, additional field and/or laboratory

Revision: 6
investigations are needed for further studies. For the P1GER or GER level
evaluations, the following alternative approach could be adopted for such study-
sections: Keep on increasing the shear strength of the key layers/materials until a
FS of 1.4 is achieved. If this back-calculated shear strength value is to some
extent reasonable and plausible for that section then additional investigations are
strongly recommended for that section. However, if the back-calculated shear
strength value is unreasonable for that section, then that section should be
marked as inadequate requiring remediation (irrespective of the shear strength
issue).
Slope stability results will be presented as shown in Figures 5-5 to 5-8.

Revision 06

Revision: 6
SOIL PARAMETERS
SOIL LAYER DESCRIPTION UNIT COHESION EFFECTIVE INTERSECTION OF MAXIMUM PAST Su/p’ RATIO
NO. WEIGHT (psf) FRICTION ANGLE, φ’ DRAINED AND UNDRAINED PRESSURE, σv,max’ AFTER σv,max’
(pcf) STRENGTH ENVELOPES, (psf)
σB’ (psf)
1 Levee Fill (Clay) 120 0 28 1,053 3000 0.23
2 Levee Fill (Clay 120 0 31 753 2500 0.23

w/Sand)
3 Foundation Clay 1 120 0 30 1,266 4000 0.23
4 Foundation Sand 1 125 0 40
5 Foundation Clay 2 125 0 29 3,331 10000 0.23
7 Deep Sand 130 0 40
8 Deep Clay 130 0 30 3,165 10000 0.23
Notes:
1. Analysis was limited to slip surfaces that intersect the levee crest
or waterside slope. See text for details.
2. Analysis included a 2 feet deep water filled tension crack.
3. Average exit gradient at toe of the levee is 1.09 and at bottom of
ditch is 2.98 (ditch full condition) (from SEEP/W Analysis).
4. Source of porewater pressure file: SEEP/W analysis for 100
year_Steady State.
5. Approximate landside slope 3.1(h):1(v) and waterside slope
3.6(h):1(v)
ELEVATION (NAVD 88) 100 Yr Flood Elev. 52.3
DISTANCE (feet)
Nichole Dong Slope Stability Analysis Results
Department of Water Resources
In association with: TASK ORDER ANALYZED
URBAN LEVEE GEOTECHNICAL Landside Steady State Seepage Condition 0
Levee Evaluations Office Adil Shah, PE EVALUATIONS PROGRAM 100 Year Flood Level (Circular Search)
FIGURE
Task Area WSESBP_XYZB Section at Station XXXX+YY
DATE CHECKED 5-5
SOIL PARAMETERS
SOIL LAYER DESCRIPTION UNIT COHESION EFFECTIVE INTERSECTION OF MAXIMUM PAST Su/p’ RATIO
NO. WEIGHT (psf) FRICTION ANGLE, φ’ DRAINED AND UNDRAINED PRESSURE, σv,max’ AFTER σv,max’
(pcf) STRENGTH ENVELOPES, (psf)
σB’ (psf)
1 Levee Fill (Clay) 120 0 28 1,053 3000 0.23
2 Levee Fill (Clay 120 0 31 753 2500 0.23

w/Sand)
3 Foundation Clay 1 120 0 30 1,266 4000 0.23
5 Foundation Clay 2 125 0 29 3,331 10000 0.23
7 Deep Sand 130 0 40
8 Deep Clay 130 0 30 3,165 10000 0.23
Notes:
or waterside slope. See text for details.
3. Average exit gradient at toe of the levee is 1.09 and at bottom of
ditch is 2.98 (ditch full condition) (from SEEP/W Analysis).
4. Source of porewater pressure file: SEEP/W analysis for 100
year_Steady State.
100 Yr Flood Elev. 52.3
3.6(h):1(v)
ELEVATION (NAVD 88)
DISTANCE (feet)
Nichole Dong Slope Stability Analysis Results
URBAN LEVEE GEOTECHNICAL Landside Steady State Seepage Condition 0
Levee Evaluations Office Adil Shah, PE EVALUATIONS PROGRAM 100 Year Flood Level (Non Circular Search)
FIGURE
Task Area WSESBP_XYZB Section at Station XXXX+YY
DATE
02/26/2008
CHECKED
Khaled Chowdhury, PE
5-6
Stability FOS vs. Water Level (NAVD88)

Steady State Seepage Condition
65
W/O Toe Drain
60 W/ Toe Drain
Water Level (NAVD 88)
Revision 06
Levee
Crest
55 Levee
Crest 200 yr
200 yr
100 yr 100 yr
50
45
1.0 1.1 1.2 1.3 1.4 1.5 1.6
Stability FOS (Landside)
Figure 5-7. Example Stability Plot.

Revision: 6
SOIL PARAMETERS
SOIL DESCRIPTION UNIT COHESION EFFECTIVE INTERSECTION OF DRAINED MAXIMUM PAST Su/p’ RATIO
LAYER NO. WEIGHT (psf) FRICTION ANGLE, φ’ AND UNDRAINED STRENGTH PRESSURE, σv,max’ AFTER σv,max’
(pcf) ENVELOPES, σB’ (psf) (psf)
1 Levee Fill (Silt 1) 120 0 32
2 Levee Fill (Clay Silt) 120 0 32 1,434 5000 0.23
3 Foundation Clay 1 120 0 31 1,807 6000 0.23
5 Foundation Clay 2 120 0 32 2,867 10000 0.23
7 Deep Clay 1 130 0 34 5,212 20000 0.23
8 Deep Sand 1 130 0 38
Notes:
or landside slope. The calculated factor of safety represents the
relative risk of a rapid drawdown failure impinging on the levee
crest. Shallower slip surfaces having lower factors of safety were
not considered a short-term risk to levee integrity and are not
shown. See text for details.
3. Total estimated drop = El. 70.73 – El. 53.94 = 16.79 feet
3.7(h):1(v)
2 Yr Flood Elev. 53.94
ELEVATION (NAVD 88)
DISTANCE (feet)

Nichole Dong Rapid Drawdown Slope Stability Analysis
URBAN LEVEE GEOTECHNICAL 100 Yr Flood Level to 2 Yr Flood Level 0
Levee Evaluations Office Adil Shah, PE EVALUATIONS PROGRAM (Circular Search) FIGURE
Task Area WL0001_XYZC Section at Station XXXX+YY 5-8
DATE CHECKED
SECTION 6.0 SEISMIC ANALYSIS
6.0 SEISMIC ANALYSIS
This section presents the step-by-step approach for completing the analyses to assess the
seismic vulnerability of the levees. The overall methodology for the seismic evaluations was
developed by the Seismic Evaluation and Methodology Group as described in detail in Levee
Seismic Vulnerability Assessment Methodology (Draft 2, June 5, 2007). In essence, the
seismic vulnerability assessment is conducted using six pieces of information:
1. Seismic hazard maps

2. Cyclic stress ratio charts
3. Kmax charts
4. Ky estimation
Revision 06
5. Displacement charts
6. Vulnerability classification table
This section presents the step-by-step approach of how these six pieces of information are
used to assess the seismic vulnerability of the levees. The overall approach is organized into
three sets of analyses:
• Liquefaction Triggering Analyses, to assess the potential for liquefaction using the
seismic hazard maps and the cyclic stress ratio charts.
• Seismic Slope Stability Analyses, to estimate the Ky/Kmax ratio using the Kmax charts and
results from the section-specific pseudo-static slope stability analyses.
• Deformation Analyses and Vulnerability Assessment, to estimate post-seismic
deformations using a Newmark-type approach and to assess the seismic vulnerability
using the vulnerability classification table.
Ultimately, for relative ranking purposes, the levees are classified using a four tiered
classification system. The four vulnerability classes (from least vulnerable to most
vulnerable) include:
1. Probably uncompromised
2. Possibly compromised
3. Likely compromised
4. Compromised
This analysis procedure was developed as a screening tool for the vulnerability assessment
of the levees. If more advanced methodologies are required for a site specific analysis, they
should be used.
6.1 Seismic Loading Conditions
The seismic vulnerability of the levees needs to be assessed for three seismic loading
conditions: 100-year, 200-year and 500-year return period events. The peak ground
acceleration (PGA) values associated with the three return period events are estimated from

Revision: 6
the results of the Probabilistic Seismic Hazard Assessment performed for the Delta Risk
Management Study Phase 1 Project (URS, 2007). Charts showing the contours of PGA
values for the Marysville, Sacramento and Stockton areas, taken from the Probabilistic
Seismic Hazard Assessment study are presented in Figure 6-1 through Figure 6-6. The
expected earthquake magnitudes (Mw) associated with the three return period events vary
from Mw=6.5 to Mw=8.0. The relative contributions of the three events for the Marysville,
Sacramento and Stockton areas are shown in Table 6-1.
Table 6-1. Relative Contributions of the Seismic Events.

Region 100-year event 200-year event 500-year event
Mw = 6.5 M = 7.0 M = 8.0
Marysville 0.5 0.3 0.2
Revision 06
Sacramento 0.6 0.3 0.1
Stockton 0.7 0.2 0.1
6.2 Water Levels
In accordance with the ICB’s recommendation, two water levels need to be considered in the
liquefaction triggering and seismic slope stability analyses – a typical winter water level and a
typical summer water level. For the study-sections along the rivers, the typical winter water
level is recommended to be taken as an estimated average water level for the month of
February and the typical summer water level is recommended to be taken as an estimated
average water level for the month of September. For the study-sections along the by-passes,
the first water level could be assumed to be either at the waterside levee toe elevation or at
the landside levee toe elevation, whichever is higher. The second water level could be
assumed to be at the bottom of the toe channel or ditch that typically runs parallel to the
levees. This would essentially simulate the winter and summer conditions.
6.3 Liquefaction Triggering Analyses
The purpose of the liquefaction triggering analyses is to evaluate whether or not any of the
levee and/or the underlying foundation materials could potentially liquefy during the
considered earthquake events. The results from the liquefaction triggering analyses will be
used to estimate the post-seismic strengths of the materials for use in the yield acceleration
calculations (Section 6.4) The step-by-step procedure for completing the liquefaction
triggering analyses is as follows:
1. Assess susceptibility to liquefaction using the plasticity criteria.

Based on soil type, liquid limit, plasticity index, water content and fines content, evaluate
the susceptibility to liquefaction using the procedure recommended by Seed et al. (2003).
If the material is judged susceptible to liquefaction, the remaining steps are needed to
further evaluate the liquefaction potential of the material considering the earthquake
loading. In contrast, if the plasticity of the material is high enough to preclude
liquefaction, the material may be classified as non-liquefiable, irrespective of the
earthquake loading.

Revision: 6
Sacramento
Stockton
Project No. 26815737

URS PSHA 2007 Figure
DWR Levee Seismic 100-Year Return Period Mean PHA Hazard
Vulnerability Assessment Sacramento and Stockton Regions 6-1
P:\Task Orders\TO-13 Seismic Vulnerability Assesment\Project Technical Memorandum\Figures\Fig 2-1 100Yr PGA Central South Regions.grf
Sacramento
Stockton

Sacramento
Stockton

USGS 2002 Converted Figure
Vulnerability Assessment Marysville Region 6-4
P:\Task Orders\TO-13 Seismic Vulnerability Assesment\Project Technical Memorandum\Figures\Fig 2-5 100Yr PGA North Region.grf
2. Estimate the Peak Ground Acceleration (PGA).

As discussed earlier in Section 6.2, estimate the PGA values for the three earthquake
events from the seismic hazard maps presented in Figure 6-1 through Figure 6-6.
3. Evaluate the Cyclic Resistance Ratio (CRR) using the SPT and/or CPT data.
For Coarse-Grained Materials:
For the site specific SPT blows per foot data, estimate the CRR values in accordance
with the procedure recommended by Seed et al. (2003) and Seed et al. (2004).
For the site specific CPT data, estimate the CRR values. The CPT-based correlations of
Moss et al. (2006) are compatible with the SPT-based methods from Seed at al. (2003)
Revision 06
and Seed et al. (2004). It is therefore recommended that the Moss et al. (2006)
procedure be used when evaluating liquefaction from CPT data.
For Fine-Grained Materials:
It is generally accepted that for low plasticity clays and silts (PI<7) the CRR may be
estimated within the framework of existing SPT-based (e.g. Seed et al., 2004) and CPT-
based (e.g. Moss et al., 2006) liquefaction correlations.
Boulanger and Idriss (2006, 2007) argue that the behavior of low-plasticity silts and
clays transitions from being more fundamentally like sands to more fundamentally like
clays over a range of PI between 5 and 12. Seed et al. (2003) and more recently Bray
and Sancio (2006) proposed a liquefaction susceptibility criterion which is based on the
PI and the water content to liquid limit (wc/LL) ratio. For the seismic vulnerability
assessment of the existing levees, it is recommended that the CRR for fine-grained
materials be calculated using either the Seed et al. (2003) or the Boulanger and Idriss
(2007) methodologies. If the two methods yield significantly different results, further
analyses may be required.
4. Evaluate the Cyclic Stress Ratio (CSR).

The PGA values estimated in Step 2 should then be used to estimate the CSR values
with depth using the charts shown in Figure 6-7. These charts were developed as part of
the overall levee seismic vulnerability assessment approach discussed earlier in Section
6.1. The charts are developed for three general locations along the levee cross section
– waterside free field, center of crest and landside free field. The chart that best
represents the location of the SPT boring being evaluated should be used to estimate
the CSR values for that particular boring. These CSR plots incorporate, to a certain
degree, the magnitude effects and therefore, no additional Magnitude Scaling Factor is
needed.
5. Assess FS against liquefaction (CRR/CSR).

FS against liquefaction should not be calculated for the locations that are above the
typical summer or winter water table, depending in the analysis case, or for materials
that are classified as non-liquefiable based on the plasticity criteria (Step 1).
If the FS against liquefaction at a particular depth is calculated to be less than 1.0, then
that portion of the soil layer should be classified as liquefiable.

Revision: 6
If all FS values at a particular depth are calculated to be higher than 1.2, then that soil
layer may be classified as non-liquefiable.
If the FS is calculated to be between 1.0 and 1.2, at a particular depth then engineering
judgment is needed to classify the material as liquefiable or non-liquefiable (due to the
marginal FS conditions).
6. Estimate the post-seismic shear strength for the various materials.

Soil layers classified as non-liquefiable should be assigned static shear strengths in the
post-seismic and yield acceleration analyses.
Soil layers classified as liquefiable should be assigned post-liquefaction residual

undrained shear strength using the correlation proposed by Seed and Harder (1990), as
Revision 06
shown in Figure 6-8. For estimating the residual shear strength from the Seed and
Harder plot is recommended. The value at a point lying one-third of the distance from the
lower-bound curve to the upper-bound curve. It should be noted that by using the lower
1/3 of the Seed and Harder (1990) relationship, zero post-liquefaction strength is
expected for (N1)60-cs < 6. This zero-strength must be used in such cases for post-
liquefaction analyses.
When liquefaction analyses are performed from CPT data, it is recommended that post-
liquefaction strengths be estimated by using the Seed and Harder (1990) relationship.
For the assessment of the seismic vulnerability, the following relationship is
recommended for calculating the “equivalent” (N1)60-cs from the qc,1,mod value:
qc,1,mod (tsf)/4.2 ~ (N1)60-cs
The entire liquefaction triggering analysis procedure should be programmed into a single
verified spreadsheet that is used program-wide. Spreadsheet printouts showing the input
values and the calculated results, individually for each SPT boring/CPT, should be presented
in an Appendix to the main report. Summary plots showing the extents of the liquefiable
layers and the post-liquefied strength values selected for each study-section should also be
presented.

Revision: 6
Crest Centerline
PGA of input motion 0
0.05g
0.1g
Landside Free Field
0.2g
0.3g 0
0.4g 20
0.5g
Waterside Free Field

0 20
40
20 40
60
40 60
Depth (ft)
80
Depth (ft)
Depth (ft)
60 80
100
80 100
120
100 120
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
140
CSR 0 0.2 0.4 0.6 0.8 CSR
CSR
Project No.26815737 Figure

Recommended
DWR Levee Seismic Cyclic Stress Ratio versus Depth
Vulnerability Assessment 6-7
P:\Task Orders\TO-13 Seismic Vulnerability Assesment\Project Technical Memorandum\Figures\Fig 8-3 CSR Recommendation.grf
Issues:
• The estimation of post-liquefaction residual undrained strength for fine-grained materials

(ML and CL) needs to be clarified. In the interim, it is recommended that the Seed and
Harder (1990) relationship be used also for these material.
• For gravel layers where Becker Penetration Tests were performed, recommendations for
the estimation of post-liquefaction residual undrained strength must be resolved.
• For marginal factors of safety, some assessment of strength reduction may be
appropriate.
6.4 Seismic Slope Stability

Revision 06
The purpose of the seismic slope stability analyses is to 1.) evaluate the minimum FS for the
post-earthquake condition and 2.) estimate the yield acceleration values for potential failure
surfaces. Yield acceleration (Ky) is defined as the maximum horizontal acceleration resulting
in a just-stable condition equivalent to an FS of 1.0. The yield acceleration values will be
used as an input in the deformation analyses.
6.4.1 Seismic Shear Strengths
Seismic loading is considered a rapid loading condition. Therefore, a combination of drained

and undrained shear strength evaluations is used, depending on the density and
permeability of the soil and the surrounding materials. The seismic shear strength
evaluations were developed using standard soil mechanics references, including Terzaghi,
Peck and Mesri (1996), Holtz and Kovacs (1981), the USACE Slope Stability Manual (EM-
1110-2-1902) and Duncan and Wright (2005). The following paragraphs present a brief
synopsis of the strength selection process recommended for the various materials.
6.4.1.1 Dense Coarse-Grained Materials (Dense/Compacted Sands and Gravels)
These materials, because of their high density, are not expected to liquefy. When sheared,
these materials are expected to dilate and potentially generate negative pore pressures
(when saturated). Due to the generation of negative pore pressures, the undrained shear
strength of these materials could conceivably be higher than their drained strength,
especially for rapid loading conditions. Because these materials are also highly permeable, it
is considered prudent not to rely on the negative pore pressures and to use the lower of the
drained and undrained strength for these materials. The shear strength of these materials is
generally represented by a friction angle and the pre-seismic pore pressures. Typical
material types included dense natural coarse-grained sediments and compacted fills
comprised of relatively clean gravels and sands.
6.4.1.2 Loose Coarse-Grained and/or Silty/Clayey Materials
These materials are expected to contract and generate positive pore pressures during shear.
During the rapid seismic loading, the rate of pore pressure generation typically exceeds the
rate of pore pressure dissipation resulting in an incremental increase in pore pressures in
these materials. The pore pressure increase leads to degradation in shear strength that may

Revision: 6
ultimately lead to liquefaction (or cyclic mobility). Depending upon the results of the
liquefaction triggering analyses, if these materials are classified as liquefiable, residual
undrained shear strength should be assigned to these materials. As stated earlier, residual
shear strengths are estimated using the correlation proposed by Seed and Harder (1990).
For estimating the residual shear strengths from the Seed and Harder plot, it is
recommended to use the value data point lying one-third of the distance from the lower
bound curve to the upper-bound curve. For these liquefiable materials, the residual shear
strengths are used both in the post-earthquake analyses and in the yield acceleration
analyses. The residual undrained strength should also be compared with the drained
strength, to confirm that the lower of the two is used in the analyses.
6.4.1.3 Fine-Grained Materials Not Expected to Liquefy

Revision 06
These materials, dilatant, or contractive, are not expected to liquefy based on the liquefaction
triggering analysis results. However, because of their low permeability, these materials are
likely to develop some pore pressures during seismic shearing. The rapid rate of seismic
loading could temporarily increase their shear strength while the increase in pore pressures
could degrade their shear strength. The combined effect of this increase and decrease in
strength is uncertain and depends upon the nature of the earthquake and the soil
characteristics. For this study, no increase or decrease in shear strength is recommended for
these materials. Accordingly, the static shear strength envelope, discussed earlier in Section
5.4.2, should be used for these materials.
6.4.2 Stability Computations
As with the static slope stability analyses, post-seismic slope stability and yield acceleration
analyses need to be performed using the computer program UTEXAS4. Both waterside and
landside slopes need to be evaluated using the circular and wedge-shaped failure modes
utilizing the search routines available in UTEXAS4. For the yield acceleration analyses, the
inertial force should be applied only in the horizontal direction. The step-by-step approach
recommended for the stability computations is as follows:
1. Estimate Kmax values for the potential shear surfaces.

Kmax is defined as follows: The seismic loading for a potential sliding mass can be
represented by the horizontal equivalent acceleration (HEA), which accounts for the
cumulative effect of the non-uniform horizontal acceleration distribution in the overlying
deformable material (Seed and Martin, 1966). Based on this concept, Kmax represents
the maximum seismic coefficient for the sliding mass: Kmax = max(HEA)/g
Using the Kmax vs. PGA charts shown in Figure 6-9, estimate the Kmax values for the
shallow and deep shear surfaces for the three earthquake events – 100-year, 200-year
and 500-year events. As explained in the Levee Seismic Vulnerability Assessment
Methodology document, an estimate of study-section stiffness needs to be made to
select the appropriate curve in Figure 6-9 for Kmax estimation.
2. Prepare the UTEXAS4 data files.

Revision: 6
The UTEXAS4 data files completed for the static analyses may be copied and modified
for the seismic analyses. Modifications included changing the material properties for the
liquefiable materials, and changing the piezometric line information.
Input the seismic shear strengths in accordance with the discussion presented in Section
6.4.1. To verify that the lower between the residual undrained strength and the drained
strength is used, it is recommended that in UTEXAS4, a non-linear strength envelope be
used. In the non-linear strength, the static strength (either trilinear envelope or drained
friction angle) is capped to a maximum strength equal to the post-liquefaction strength,
as shown in Figure 6-10.
Input the seismic water levels in accordance with the discussion presented in Section
6.2. For calculating the pore water pressures, piezometric surfaces, developed based on
Revision 06
judgment, could be used.
3. Perform post-earthquake analyses.

Using the seismic shear strengths, discussed earlier in Section 6.5.1, post-earthquake
analyses should be completed using UTEXAS4.
No horizontal seismic coefficient is used in the post-earthquake analyses.
If the completed post-earthquake FS, for any potential shear surface is less than or equal
to 1.0, the study-section should be classified as unstable and no further yield
acceleration analyses are needed.
4. Perform pseudo-static stability analyses with a single seismic coefficient.

Using the seismic shear strengths, discussed earlier in Section 6.4.1, pseudo-static
For each study-section, pseudo-static slope stability analyses should be completed

using a horizontal seismic coefficient, K, equal to one half of Kmax (K = 0.5•Kmax). All
waterside and landside shear surfaces should be checked with the search routines.
If the minimum completed FS, for all shear surfaces, for this acceleration level (K =
0.5•Kmax) is greater than 1.0, (which means that the Ky/Kmax will be larger than 0.5) then
minimal or negligible seismic displacements are anticipated for the study-section. Under
this condition, the study-section could be classified as Probably Uncompromised and
further refinement for the vulnerability evaluation is considered unnecessary. Computed
FS values should be rounded to one decimal place.

Revision: 6
Figure 6-8
0.5
Deep-Stiff 1:1
Deep-Medium
Deep-Soft
Shallow-Stiff
0.4
Shallow-Medium
Shallow-Soft
0.3
Kmax (g)
0.2
0.1
0
0 0.1 0.2 0.3 0.4 0.5
Input Motion PHA (g)
Project No. 26815737 Figure

Recommended
DWR Levee Kmax vs Input Motion PHA
Seismic Vulnerability Assessment 6-9
P:\Task Orders\TO-13 Seismic Vulnerability Assesment\Project Technical Memorandum\Figures\Fig 8-7 Kmax vs PGA Recommendation.grf
Revision 06
Effective Stress
Post-Liquefaction Envelope
Residual Strength
Drained Strength
su,r
Shear Strength
Figure 6-10. Strength Envelope for Post-Seismic Stability Analysis.

Revision: 6
5. Estimate the yield acceleration (Ky) values.

This step is only performed for the study-sections that have a post-earthquake FS
greater than 1.0 and a pseudo-static FS (with K=0.5•Kmax) that is less than or equal to
1.0.
Using the seismic shear strengths, discussed earlier in Section 6.4.1, pseudo-static
Yield acceleration (Ky) is defined as the minimum horizontal acceleration resulting in a

just-stable condition equivalent to a FS of 1.0.
For each qualifying study-section, pseudo-static slope stability analyses need to be

Revision 06
completed for a range of horizontal seismic coefficients (e.g., K = 0.05g, 0.10g, 0.15g,
etc.) to estimate the yield acceleration values. All waterside and landside shear surfaces
should be checked with the search routines.
The yield acceleration value, for each potential shear surface, should be estimated by
plotting the seismic coefficient (K) values versus the computed FS values and selecting
the K value that corresponds to a FS value of 1.0.
6. Estimate the minimum Ky/Kmax ratio value.

Using the Kmax value estimated in Step 1 and the Ky value estimated in Step 5, the
Ky/Kmax ratio is estimated for each potential shear surface.
The minimum value of the Ky/Kmax ratio needs to be selected for each study-section for
use in the subsequent deformation and vulnerability assessment evaluations. Note that
the minimum Ky/Kmax ratio value is not necessarily given by the minimum Ky.
6.4.2.1 Single-Stage Versus Two-Stage Stability Computations
When ICU triaxial tests are not available for the materials in the studied section, a single
stage analysis procedure should be used. In the single stage analyses, materials that are not
expected to liquefy and those materials whose liquefied undrained residual strength is larger
than the static strength will be modeled in the same way as for the static stability analyses
(i.e., drained strength or tri-linear envelope for free-draining and non-free-draining materials,
respectively). Materials that are expected to liquefy, and whose undrained residual strength
is lower than its static strength, will be modeled using the post-liquefaction residual
undrained strength (Option 1, with φ’ = 0).
When ICU triaxial tests are available for the materials in the studied section, the two-stage
procedure should be used. In these cases, two-stage stability computations will be
performed using the built-in function available in UTEXAS4. The two-stage stability
computation consists of two complete sets of stability computations. In the first one, stresses
along the shear surface are calculated. For this stage, all materials are modeled with drained
strength parameters (Option 1, c’ and φ’). For the second stage, material’s strength will be
modeled in different manners depending on its permeability:

Revision: 6
• A coarse-grained material that is not expected to liquefy and for whose liquefied residual
undrained strength is larger than its static strength will be modeled with the same
drained strength properties for in the second stage as in the first stage. A coarse-grained
material that is expected to liquefy and for whose residual undrained strength is lower
than its drained strength, will be modeled using the post-liquefaction residual undrained
strength for the second stage of the two-stage analysis (Option 1, with φ’ = 0).
• For fine-grained materials, the second set of stability computations is performed using
the undrained shear strengths to calculate the FS during an earthquake. The procedures
used to determine the undrained shear strength are based on the procedures proposed
by Duncan, Wright and Wong (1990). They propose the use of two envelopes derived
from results of ICU- triaxial compression tests with pore pressure measurements. When
these types of tests are available, the envelope proposed by Duncan, Wright and Wong
(1990) should be used in the two-stage analyses.
Revision 06
6.5 Deformation Analyses and Vulnerability Assessment
The purpose of the deformation and vulnerability evaluations is: 1.) evaluate the post-seismic
vertical displacements at the levee crest and 2.) estimate the seismic vulnerability of the
levees using a four tiered classification system. The vulnerability classification system
adopted for the project levees includes the following four tiers/classes:
2. Possibly Compromised
3. Likely Compromised
4. Compromised
Depending upon the calculated post-seismic displacements, each study-section needs to be

classified into one of the above four classes. The step-by-step approach for the deformation
and vulnerability evaluations is outlined in the following paragraphs. Further details about this
approach, including the background and rationale behind the charts and figures used in this
approach, are presented in Levee Seismic Vulnerability Assessment Methodology.
1. Estimate the Newmark-type displacements.

Using the minimum Ky/Kmax ratio, estimated for each study-section, calculate the
Newmark-type displacements from the charts shown in Figure 6-11. Displacements are
estimated for three earthquake magnitudes, Mw=6.5, 7.0 and 8.0, which represent the
range of values corresponding to the 100-year, 200-year and 500-year events.
2. Estimate Weighted Freeboard Loss.

Using a semi-probabilistic approach and the Newmark-type displacements estimated for
the three earthquake events, the weighted freeboard loss is estimated for each study-
section. The weights for the three earthquake events are presented in Table 6-1.
A rule of 0.7 foot of vertical displacement for every 1.0 foot of horizontal displacement is
recommended for estimating the freeboard loss (vertical displacements) from the
Newmark-type horizontal displacement values.

Revision: 6
• Coarse-grained materials which are not expected to liquefy and for which the liquefied
undrained residual strength is larger than the static strength will be modeled with the
same drained strength properties for the second stage as in the first stage. Coarse-
grained materials which are expected to liquefy and for which the undrained residual
strength is lower than its drained strength, will be modeled using the post-liquefaction
undrained residual strength for the second stage of the two-stage analysis (Option 1,
with φ’ = 0).
• For fine-grained materials, the second set of stability computations is performed using
the undrained shear strengths to calculate the FS during an earthquake. The procedures
used to determine the undrained shear strength are based on the procedures proposed
by Duncan, Wright and Wong (1990). They propose the use of a second envelope
derived from results of consolidated-undrained type triaxial shear tests performed on
specimens consolidated isotropically. When these types of tests are available, the
envelope proposed by Duncan, Wright and Wong (1990) should be used in the two stage
Revision 06
analyses.
6.5 Deformation Analyses and Vulnerability Assessment
The purpose of the deformation and vulnerability evaluations is: 1.) evaluate the post-seismic
vertical displacements at the levee crest and 2.) estimate the seismic vulnerability of the
levees using a four tiered classification system. The vulnerability classification system
adopted for the project levees includes the following four tiers/classes:
2. Possibly Compromised
3. Likely Compromised
4. Compromised
Depending upon the calculated post-seismic displacements, each study-section needs to be

classified into one of the above four classes. The step-by-step approach for the deformation
and vulnerability evaluations is outlined in the following paragraphs. Further details about this
approach, including the background and rationale behind the charts and figures used in this
approach, are presented in Levee Seismic Vulnerability Assessment Methodology.
1. Estimate the Newmark type displacements.

Using the minimum Ky/Kmax ratio, estimated for each study-section, calculate the
Newmark type displacements from the charts shown on Figure 6-11. Displacements are
estimated for three earthquake magnitudes, Mw=6.5, 7.0 and 8.0, which represent the
range of values corresponding to the 100-year, 200-year and 500-year events.
2. Estimate Weighted Freeboard Loss.

Using a semi-probabilistic approach and the Newmark type displacements estimated for
the three earthquake events, the weighted freeboard loss is estimated for each study-
section. The weights for the three earthquake events need to be adopted from the
information presented in Table 6-1.

Revision: 6
0.7 feet of vertical displacement for every 1.0 feet of horizontal displacement rule is
recommended for estimating the freeboard loss (vertical displacements) from the
Newmark type horizontal displacement values.
3. Estimate Vulnerability Class.

Depending upon the calculated weighted freeboard loss value, each study-section is
classified into one of four vulnerability classes shown on Table 6-2.
Table 6-2. Seismic Vulnerability Classes.

Remaining Freeboard
Significant Damage to for Post Seismic
Amount of Post Seismic Flood
Internal Structures Evaluation (2-Year
Deformation Protection Ability
(e.g. Cutoff Walls) Flood Water Surface
Revision 06
Elevation)
<1’ No >1’ Probably Uncompromised
1’ to 3’ Possibly >1’ Possibly Compromised
3’ to 10’ Likely if existing None Likely Compromised
Unlimited (flow side
Yes None Compromised
condition)
For clarity and to facilitate independent reviews, seismic analysis data and results will be
presented as shown in Tables 6-3 and 6-4 and in Figures 6-12 to 6-18.

Revision: 6
Revision 6

Task Area:
Table : Liquefaction-Triggering Analysis (Winter, 500 Year Return Period)
SPT BASED LIQUEFACTION ANALYSIS

REFERENCE: Seed, R.B., et al, 2004. Standard Penetration Test-Based Probablistic and Deterministic Assessment of Seismic Soil Liquefaction Potential. Journal of Geotechnical and Geoenvironmental Engineering ASCE, Vol. 130, No. 12, 1313-1340.
PROJECT NAME:
PROJECT NUMBER: ANALYSIS BY: Samrat Mohanty, PhD Date: 12/26/2007 Data Input By Analyist
Reach ID: CHECKED BY: Khaled Chowdhury, PE Date: 1/14/2008 Results of Analysis
Station: XXXX+YY Calculated Values
Boring/CPT ID: WL0001_XYZB
Ground Elevation (ft.)= 81.2 Earthquake Information:

Groundwater Elev. During Drilling (ft.) = 36.2 From CPT Groundwater Depth Below GS During Drilling (ft.) = 45.0 Hammer Type = Automatic Return Period (years) = 500 PGA = 0.16
Design Groundwater Elevation (ft.)= 37.2 Typical Winter Design Groundwater Depth Below GS (ft.) = 44.1 Energy Ratio = 73 Vs,40 (fps) = 800 Mw = 8.0
Lab Test Results for CL, ML Results of Analysis
Effective Fine Correction Cyclic Stress

Stress Cyclic The Below Design Liquefaction The
Depth to Unit Overburden Overburden Overburden Energy Borehole Borehole Sampler Rod Rod Length Fine Content Factor for Cyclic Ratio from Factor of Liquid Water Final Factor of
Sample USCS Drilling Sampler Reduction Stress Probability of FSL < 1 Groundwater USCS Plasticity Susceptibility Probability of
Sample Nfield Weight Pressure, sv Pressure, Correction Correction Diameter Correction Correction Length Correction N60 N1,60 Content for Fine N1,60,CS Resistance Seismic Safety, Limit, Content, Assessment Safety,
ID Classification Method Code Coefficient, Ratio, Liquefaction, (Yes/No) Level Classification Index, PI Based on Soil Liquefaction,
(ft.) (pcf) (psf) sv' Factor, CN Factor, CE (inch) Factor, CB Factor, Cs (m) Factor, CR (%) Analysis Content, Ratio, CRR Vulnerability FSL LL (w%) (Yes/No) FSL
Rd CSReq PL (Yes/No) Type (Yes/No) PL
(psf) (%) Cfines Report, CSR
S02A 5 48 SP-SM 125 625 625 1.6 1.2 M 6 1.05 S 1 3.7 0.75 46 74 5 1.0 75 0.600 1.00 0.104 0.203 0.0000 3.0 No No SP-SM Yes No - -
S04A 10 13 SP-SM 120 1225 1225 1.3 1.2 M 6 1.05 S 1 5.2 0.85 14 18 9 9 1.1 19 0.128 1.00 0.104 0.193 0.7391 0.7 Yes No SP-SM Yes No - -
S06A 15 11 SP-SM 120 1825 1825 1.0 1.2 M 6 1.05 S 1 6.7 0.91 13 13 5 1.0 14 0.080 0.99 0.103 0.184 0.9981 0.4 Yes No SP-SM Yes No - -
S08A 20 22 SP-SM 115 2400 2400 0.9 1.2 M 6 1.05 S 1 8.2 0.94 26 24 5 1.0 25 0.169 0.99 0.103 0.184 0.2021 0.9 Yes No SP-SM Yes No - -
S10A 25 22 SM 115 2975 2975 0.8 1.2 M 6 1.05 S 1 9.8 0.96 27 22 12 1.1 24 0.137 0.98 0.102 0.170 0.2948 0.8 Yes No SM Yes No - -
S13A 32.5 8 CL 120 3875 3875 0.7 1.2 M 6 1.05 S 1 12.0 0.98 10 7 35 1.4 10 0.041 0.97 0.101 0.135 0.9999 0.3 Yes No CL 29 8 23 Yes No - -
S15A 35 13 SM 120 4175 4175 0.7 1.2 M 6 1.05 S 1 12.8 0.99 16 11 52 35 1.3 15 0.055 0.96 0.100 0.127 0.9708 0.4 Yes No SM Yes No - -
S16A 40 3 CL 125 4800 4800 0.6 1.2 M 6 1.05 S 1 14.3 0.99 4 2 35 1.9 5 0.027 0.94 0.098 0.124 1.0000 0.2 Yes No CL No No - -
S18A 45 13 SM 125 5425 5425 0.6 1.2 M 6 1.05 S 1 15.8 0.99 17 10 12 1.1 11 0.046 0.92 0.096 0.125 0.9998 0.4 Yes Yes SM Yes Yes 0.37 1.00
S20A 50 9 SM 130 6075 5764 0.6 1.2 M 6 1.05 S 1 17.4 1.00 11 7 22 22 1.3 8 0.036 0.90 0.099 0.124 1.0000 0.3 Yes Yes SM Yes Yes 0.29 1.00
S28A 71 57 GW-GM 125 8775 7154 0.5 1.2 M 6 1.05 S 1 23.8 1.00 73 38 6 6 1.0 40 0.366 0.85 0.108 0.113 0.0000 3.3 No Yes GW-GM Yes No - -
S29A 75 88 SP-SM 130 9295 7424 0.5 1.2 M 6 1.05 S 1 25.0 1.00 112 58 5 1.0 60 0.600 0.84 0.109 0.109 0.0000 5.5 No Yes SP-SM Yes No - -
S30A 80 47 SP-SM 130 9945 7762 0.5 1.2 M 6 1.05 S 1 26.5 1.00 60 30 5 1.0 31 0.196 0.83 0.111 0.106 0.0000 1.9 No Yes SP-SM Yes No - -
S31A 81.5 49 CL 130 10140 7864 0.5 1.2 M 6 1.05 S 1 27.0 1.00 63 32 35 1.2 38 0.212 0.83 0.112 0.105 0.0000 2.0 No Yes CL 31 8 34 Yes No - -
S37A 100 87 SP-SM 130 12545 9114 0.5 1.2 M 6 1.05 S 1 32.6 1.00 111 52 12 12 1.1 55 0.600 0.81 0.115 0.097 0.0000 6.2 No Yes SP-SM Yes No - -
S39A 105 47 CL 130 13195 9452 0.5 1.2 M 6 1.05 S 1 34.1 1.00 60 28 35 1.2 33 0.150 0.80 0.116 0.093 0.0000 1.6 No Yes CL No No - -
S44A 115 31 CL 130 14495 10128 0.4 1.2 M 6 1.05 S 1 37.2 1.00 40 18 35 1.2 22 0.069 0.78 0.117 0.090 0.1121 0.8 Yes Yes CL No No - -
S46A 120 47 SM 130 15145 10466 0.4 1.2 M 6 1.05 S 1 38.7 1.00 60 26 12 1.1 28 0.131 0.78 0.117 0.090 0.0002 1.5 No Yes SM Yes No - -
S48A 125 39 SM 130 15795 10804 0.4 1.2 M 6 1.05 S 1 40.2 1.00 50 21 12 1.1 23 0.090 0.77 0.117 0.090 0.0558 1.0 No Yes SM Yes No - -
Table 6-3.
Example Liquefaction Triggering Analysis.
Revision 6

Task Area:
Table : Seismic Strength Selection Summary Table (500 Yr Return Period)
SUBJECT: Task Area: Seismic Strength Selection for Static Slope Stability Analysis By: Samrat Mohanty, PhD Date: 2/28/2008
PROJECT NUMBER: Chk'd By: Khaled Chowdhury, PE Date: 2/29/2008
SPT and Laboratory Test Data Strength Parameters from SPT
Effective Calculated Maximum Past Intersection of

Elevation of Depth to Elevation of Elevation of Maximum Past Overburden Pressure Using N60 Drained Friction Recommended
USCS Soil Depth to Top Layer Thickness Pocket Drained Undrained
Reference Borings Levee Crest Layer ID
Classification of Layer (ft.)
Bottom of Top of Layer Bottom of Layer
(ft.) Average Average Unit Weight Water Pressure from Stress at Mid Su, Based on N60 Based Su (75%) and Angle, φ' (psf) Maximum Past Liquefiable Residual Shear Comment
Average N Average N60 PI LL -200% OCR Penetrometer, qu Cohesion, c' Strength
(ft) Layer (ft.) (NAVD 88) (NAVD 88) N1,60 N1,60,CS (pcf) Content (%) Consolidation Layer Below (psf) Effective Overburden (Maximum φ' = Pressure, σ'v,max Layer (Yes/No) Strength (psf)
Tests, σ'v,max (psf)
(tsf)
Crest During Stress During Drilling , (psf) Envelopes, σ'B
400) (psf)
Drilling (psf) σ'v,max (psf) (psf)
Poorly Graded Sand
1 0 26 81.2 55.2 26.0 23 25 30 32 120 9 (10'), 40 (23') 2.5 (25') 1551 0 33 No
w/Silt (SP)
2 Silty Sand (SM) 26 32.5 55.2 48.7 6.5 120 16 (31') 3485 0 32 No
Sandy Lean Clay
3 32.5 35 48.7 46.2 2.5 8 10 7 10 120 23 (32.5') 8 (32.5') 29 (32.5') 4021 1200 3886 0 34 3000 782 No
(CL)
Silty Sand (SM) to
4 35 40 46.2 41.2 5.0 13 16 11 15 125 52 (35') 4906 0 36 No
Sandy Silt (ML)
40-42 feet CL and 42-45 feet
Lean Clay / Silt
5 40 45 41.2 36.2 5.0 3 4 2 5 130 38 (43') 1 (43') 26 (43') 5241 480 1157 0 27 1100 407 No ML with organic odor, trace
w/Sand (CL/ML) roots
22 (50'), 15
6 Silty Sand (SM) 45 67 36.2 14.2 22.0 15 19 11 12 130 6148 0 37 Yes 350
(55'), 12 (60')
Gravel w/Silt and
7 67 75 14.2 6.2 8.0 57 73 38 40 130 6 (71') 7154 0 40 No
Sand ( GP)
WL0001_XYZB 81.2 Poorly Graded Sand
8 w/Silt and Gravel 75 81.5 6.2 -0.3 6.5 43 54 28 29 130 7641 0 40 No
(SP)
Sandy Lean Clay /
9 81.5 87.5 -0.3 -6.3 6.0 40 50 25 29 130 34 (81.5') 8 (81.5') 31 (81.5') 8060 6000 24421 0 28 10000 3508 No
Elastic Silt (CL/ML)
10 Silty Sand (SM) 87.5 100 -6.3 -18.8 12.5 16 20 10 12 130 42 (95') 8681 0 38 Yes 350
Poorly Graded
11 100 105 -18.8 -23.8 5.0 37 47 22 24 130 12 (100') 9268 0 40 No
Gravel w/Sand (GP)
12 Lean Clay (CL) 105 118 -23.8 -36.8 13.0 45 57 26 31 130 9872 6840 27345 0 30 10000 3165 No
13 Silty Sand (SM) 118 131 -36.8 -49.8 13.0 41 52 22 24 130 10745 0 40 No
14 Lean Clay (CL) 131 140 -49.8 -58.8 9.0 62 79 33 39 130 11483 9505 39725 0 30 10000 3165 No
Levee Toe Elevation 53 (NAVD 88)

Levee Height 28.2 (feet)
Table 6-4.
Example Seismic Strength Selection Analysis.
100
M8
10 M7
M6
Newmark Displacement (ft)
0.1
0.01
0 0.2 0.4 0.6 0.8 1
Ky/Kmax

Recommended Figure
DWR Levee Newmark Displacement vs Ky/Kmax
Seismic Vulnerability Assessment 6-11
P:\Task Orders\TO-13 Seismic Vulnerability Assesment\Project Technical Memorandum\Figures\Fig 8-15 Newmark Displacement.grf
SOIL PARAMETERS Summary of Seismic Vulnerability Evaluation
SOIL LAYER DESCRIPTION UNIT COHESION EFFECTIVE INTERSECTION OF DRAINED MAXIMUM PAST Su/p’ RATIO Section: WL0001_XYZB (Circular Search) at Sta. XXXX+YY Winter Low (median) Water Level
NO. WEIGHT (psf) FRICTION ANGLE, φ’ AND UNDRAINED STRENGTH PRESSURE, σv,max’ AFTER σv,max’
(pcf) ENVELOPES, σB’ (psf) (psf) 500 Yr Return Period F.S. (Land Side) F.S. (Water Side)
1 Levee Sand 120 0 33 Scenario Deep Failure Deep Failure
2 Levee Silty Sand 120 0 32
Post-Earthquake Analysis 1.08 1.69
3 Foundation Sandy 120 0 34 782 3,000 0.23
Lean Clay Pseudo-static Analysis (K = 0.0725) 0.82 1.05
4 Foundation Silty Sand 125 0 36 PGA 0.17 0.17
5 Lean Clay/Silt w/Sand 130 0 27 444 1,100 0.23
Kmax 0.145 0.145
6 Silty Sand 130 350 0
Ky 0.02
7 Gravel/Sand 130 0 40
8 Lean Clay/Elastic Silt 130 0 28 3,508 10,000 0.23

Ky/Kmax 0.14
9 Lean Clay 130 0 30 3,165 10,000 0.23 Newmark Displacement (feet) 5.5 <1’
Seismic Vulnerability Class Likely Compromised Probably Uncompromised
ELEVATION (NAVD 88)
Winter Low (February Mean) Elev. 37.06
DISTANCE (feet)
Samrat Mohanty, PhD
Landside F.O.S. (Circular Search), Post-
URBAN LEVEE GEOTECHNICAL
Earthquake Slope Stability Analysis,
0
Division of Flood Management EVALUATIONS PROGRAM
Levee Evaluations Office Samrat Mohanty, PhD WL0001_XYZB Section at Station XXXX+YY FIGURE
TASK AREA
DATE CHECKED 500 Yr Return Period 6-12
SOIL LAYER DESCRIPTION UNIT COHESION EFFECTIVE INTERSECTION OF DRAINED MAXIMUM PAST Su/p’ RATIO Section: WL0001_XYZB (Non Circular Search) at Sta. XXXX+YY Winter Low (median) Water Level
Lean Clay Pseudo-static Analysis (K = 0.0725) 0.90
Kmax 0.145 0.145
Ky 0.055

Ky/Kmax 0.38
9 Lean Clay 130 0 30 3,165 10,000 0.23 Newmark Displacement (feet) Flow Condition 0.6
Seismic Vulnerability Class Compromised Probably Uncompromised
ELEVATION (NAVD 88)
DISTANCE (feet)
Samrat Mohanty, PhD
Landside F.O.S. (Non-circular Search), Post-
0
TASK AREA
DATE
02/28/08
CHECKED 500 Yr Return Period 6-13
Kmax 0.145 0.145
Ky 0.02

Ky/Kmax 0.14
ELEVATION (NAVD 88)
DISTANCE (feet)
Samrat Mohanty, PhD
Waterside F.O.S. (Circular Search), Post-
0
TASK AREA
Kmax 0.145 0.145
Ky 0.055

Ky/Kmax 0.38
9 Lean Clay 130 0 30 3,165 10,000 0.23 Newmark Displacement (feet) Flow Condition 0.6
Seismic Vulnerability Class Compromised Probably Uncompromised
ELEVATION (NAVD 88)
DISTANCE (feet)
Samrat Mohanty, PhD
Waterside F.O.S. (Non-circular Search), Post-
0
TASK AREA
Kmax 0.145 0.145
Ky 0.02

Ky/Kmax 0.14
1.5
0.5
0
0 0.05 0.1 0.15
ELEVATION (NAVD 88)
Horizontal Seismic Coefficient
DISTANCE (feet)
Samrat Mohanty, PhD
Landside F.O.S. (Circular Search), Pseudo-
Static Slope Stability Analysis, 500Yr Event,
0
Levee Evaluations Office Samrat Mohanty, PhD K(=Kmax/2):0.0725, WL0001_XYZB Section at FIGURE
TASK AREA
DATE CHECKED Station XXXX+YY 6-16
Kmax 0.145 0.145
Ky 0.02

Ky/Kmax 0.14
ELEVATION (NAVD 88)
DISTANCE (feet)
Samrat Mohanty, PhD
Waterside F.O.S. (Circular Search), Pseudo-
Static Slope Stability Analysis, 500Yr Event,
0
Levee Evaluations Office Samrat Mohanty, PhD K(=Kmax/2):0.0725, WL0001_XYZB Section at FIGURE
TASK AREA
DATE CHECKED Station XXXX+YY 6-17
2 Levee Silty Sand 120 0 32 Post-Earthquake Analysis 0.92 1.51
5 Lean Clay/Silt w/Sand 130 0 27 444 1,100 0.23 Kmax 0.145 0.145
Ky 0.045
Ky/Kmax 0.31
9 Lean Clay 130 0 30 3,165 10,000 0.23

Newmark Displacement (feet) Flow Condition 1.1
Seismic Vulnerability Class Compromised Possibly Compromised
1.5
0.5
0
0 0.05 0.1 0.15
Horizontal Seismic Coefficient
ELEVATION (NAVD 88)
DISTANCE (feet)
Samrat Mohanty, PhD
Waterside F.O.S. (Non-circular Search),
Pseudo-Static Slope Stability Analysis, 500Yr
0
Levee Evaluations Office Samrat Mohanty, PhD Event, K(=Kmax/2):0.0725, WL0001_XYZB FIGURE
TASK AREA
DATE
02/28/08
CHECKED Section at Station XXXX+YY 6-18
SECTION 7.0 EROSION EVALUATION
7.0 EROSION EVALUATION
Under development.
Revision 06

Revision: 6
SECTION 8.0 ALTERNATIVE ANALYSIS AND COST ESTIMATE
8.0 ALTERNATIVE ANALYSIS AND COST ESTIMATE
Under development.
Revision 06

Revision: 6
SECTION 9.0 CALCULATION PACKAGES
9.0 CALCULATION PACKAGES
Data and engineering analysis will be documented in calculation packages prepared for each
levee reach. Calculation packages will not be included in the P1GER or the GER but will be
filed in the central program file and made available to independent reviewers and third
parties as authorized by DWR. Calculation packages should be organized as per the
following Table of Contents:
1) Signed QA/QC Form

2) Data
a) Past performance
b) Past remediation
Revision 06
c) Bathymetry
d) HEM
e) Land geophysics
f) Geomorphology
g) LiDAR
h) Exploration (borings, CPTs, etc.)
i) Testing [Prefer to include the modifications of classifications based on lab tests and
SPT corrected blow counts (N60)]
j) Groundwater elevations (drilling)
k) Piezometers’ readings
3) Reach Selection
4) Working Annotated Plans, Profiles and Cross Sections
5) Selection of Strength and permeability values
6) H&H water surface elevations and profiles
7) Engineering Analysis
a) Slope stability
i) Static
ii) Rapid drawdown
b) Seismic
c) Seepage
8) Blanket theory
9) SEEPW
a) Erosion
10) Problem Areas
11) Conceptual remedial alternatives
12) Cost estimates

Revision: 6
SECTION 10.0 REFERENCES
10.0 REFERENCES
10.1 Seepage Analysis References
Krahn, J. 2004. “Seepage Modeling with SEEP/W: An Engineering Methodology,” Geo-Slope

International Ltd.
Montgomery Watson (MWH North America). As-Built Drawings, Levee Construction Contract
B, West Sacramento Project.
URS Corporation. 2007. Memorandum responding to ICB comment 4 dated June 14, 2007.
“Comparison Between Blanket Theory and SEEP/W for Seepage and Stability
Analyses.”
Revision 06
USACE. 2005. Engineering and Design – Design Guidance for Levee Underseepage. EM
1110-2-569. Proponent: CECW-EG. www.usace.army.mil/publications/eng-tech-
ltrs/etl1110-2-569/entire.pdf
USACE, 2000. Engineering and Design – Design and Construction of Levees. EM 1110-2-
1913. Proponent: CECW-EG. http://www.usace.army.mil/publications/eng-
manuals/em1110-2-1913/entire.pdf
USACE. 2004, Geotechnical Levee Practice Standard Operating Procedure (SPO EDG-03)
10.2 Blanket Theory References
USACE. 1956. Technical Memorandum 3-424.
Turnbull, W, Mansur, C. 1961. “Underseepage and Its Control, A Symposium” ASCE

Transactions Paper No. 3247.
USACE. 2000. “Mathematical Analysis of Underseepage and Substratum Pressure,”

Appendix B, EM 1110-2-1913.
10.3 Stability Analyses References
Duncan, J., Wright, S., Wong, K. 1990. “Slope Stability During Rapid Drawdown,” H. Bolton
Seed Memorial Symposium.
Robertson, P. 2006. Guide to In-Situ Testing.
Spencer, E. 1967. “A Method of Analysis of the Stability of Embankments Assuming Parallel

Inter-Slice Forces,” Geotechnique, Vol. 17, 1.
United States Air Force. 1983. Soils and Geology Procedures for Foundation Design of
Buildings and Other Structures (Except Hydraulic Structures). TM 5-818-1/AIR
FORCE AFM 88-3, Chapter 7. Distribution: Approved for Public Release;

Revision: 6
Unlimited. Proponent: CE-MP. http://www.usace.army.mil/publications/armytm/tm5-

818-1/entire.pdf
USACE. 2004. Geotechnical Levee Practice Standard Operating Procedure (SPO EDG-03)
USACE. 2003. Engineering and Design – Slope Stability. Proponent: CECW-EW. EM 1110-
2-1902. http://www.usace.army.mil/publications/eng-manuals/em1110-2-
1902/entire.pdf
USACE. 1989. Engineering and Design – Retaining and Flood Walls. EM 1110-2-2502.
Proponent: CECW-ED. http://www.usace.army.mil/publications/eng-
Revision 06
USACE. 2000. Engineering and Design – Design and Construction of Levees. EM 1110-2-
1913. Proponent: CECW-EG. http://www.usace.army.mil/publications/eng-
USACE. 2001. Engineering and Design – Geotechnical Investigations. EM 1110-1-1804.

Proponent: CECW-ET. http://www.usace.army.mil/publications/eng-
Wright, S. 1999. “A Computer Program for Slope Stability Calculations.” Technical Manuals
for UTEXAS4.
10.4 Seismic Analyses References
Bray, J., Sancio R. 2006. "Assessment of the Liquefaction Susceptibility of Fine-Grained

Soils," Journal of Geotechnical and Geoenvironmental Engineering, ASCE.
Boulanger, R., Idriss, I. 2007. "Evaluation of Cyclic Softening in Silts and Clays," Journal of
Geotechnical and Geoenvironmental Engineering.
Boulanger, R., Idriss, I. 2006. "Liquefaction Susceptibility Criteria for Silts and Clays," Journal
of Geotechnical and Geoenvironmental Engineering, ASCE.
Duncan J., Wright S. 2005. Soil Strength and Slope Stability. John Wiley and Sons, New
York.
Duncan, J., Wright, S., Wong, K. 1990. “Slope Stability During Rapid Drawdown,” H. Bolton
Seed Memorial Symposium.
Moss, R., Seed, R. Olsen R. 2006. "Normalizing the CPT for Overburden Stress," Journal of
Geotechnical and Geoenvironmental Engineering, American Society of Civil
Engineers.
Holtz, R. Kovacs, W. 1981. An Introduction to Geotechnical Engineering. Prentice-Hall,

Englewood Cliffs, NJ.

Revision: 6
Seed, R., Cetin, K., Moss, R., Kammerer, A. et al. 2003. "Recent Advances in Soil
Liquefaction Engineering: A Unified and Consistent Framework," Keynote Address,
Proceedings, 26th Annual Geotechnical Spring Seminar, Los Angeles Section of
the GeoInstitute, American Society of Civil Engineers.
Seed, R., Cetin, K., Der Kiureghian, A, et al. 2004. "Standard Penetrations Test-Based
Probabilistic and Deterministic Assessment of Seismic Soil Liquefaction Potential,"
Journal of Geotechnical and Geoenvironmental Engineering, American Society of
Civil Engineers, Vol. 130, No. 12.
Seed R., Harder L. 1990. “SPT-based Analysis of Cyclic Pore Pressure Generation and
Undrained Residual Strength”: Proceedings of H.B.Seed Memorial Symposium,
Vol. 2, BiTech Publishing, Vancouver.
Revision 06
Terzaghi, K., Peck, R., Mesri, G. 1996. Soil Mechanics in Engineering Practice, Third Edition.
John Wiley & Sons, New York.
USACE. 2003. Engineering and Design – Slope Stability. Proponent: CECW-EW. EM 1110-
2-1902. http://www.usace.army.mil/publications/eng-manuals/em1110-2-
1902/entire.pdf

Revision: 6
APPENDIX A
APPENDIX A
Simplified Procedure for Shear Strength

Selection of Fine-Grained Materials
REVISION 06
Issue Date: 03-2008

Revision: 6
APPENDIX A SIMPLIFIED PROCEDURE FOR SHEAR STRENGTH SELECTION
OF FINE-GRAINED MATERIALS
The simplified procedure presented herein allows for the development of a unique strength
envelope (tri-linear envelope) to effectively model the shear strength of non-free-draining
fine-grained materials which are below the phreatic surface line. The advantages associated
with using the tri-linear envelope are as follows:
The tri-linear envelope automatically uses the lower of the drained and undrained strength,
which is of primary significance when considering overconsolidated materials in which the
undrained strength could potentially be higher than its drained strength.
In most cases, use of the tri-linear envelope eliminates the need to subdivide a foundation
clay layer into zones (in the horizontal direction) when moving from beyond the toe to
underneath the crest of the levee. This is true for situations where a constant value of
maximum past stress could be assumed for the entire layer. For other situations where the
maximum past stress varies within the layer, the layer needs to be subdivided into zones and
a different tri-linear envelopes needs to be developed for each zone. In either case, the tri-
linear envelope, once developed based on the maximum past stress value (before the flood
event), automatically accounts for the increase in OCR with rise in the water level associated
with a flood event. See Attachment A1 for further elaboration of this concept.
A1.0 APPLICATIONS AND LIMITATIONS
The tri-linear strength envelopes obtained with this simplified procedure are proposed for use
in the slope stability analyses of the existing levees in accordance with the overall analysis
approach described in the main guidance document. None of these strength envelopes are
intended to evaluate the stability of the levee during or immediately following any
construction activity associated with any kind of remediation effort. For the case involving
new construction, the strength envelopes, including the overall analysis approach, may need
to be revised to properly account for the “unconsolidated undrained” behavior of the
materials.
A1.1 Assumptions
The following assumptions are made in the development of the procedure:
The fine-grained materials follow a normalized soil strength behavior (e.g., Stress History
and Normalized Soil Engineering Properties [SHANSEP]).
The fine-grained materials are fully consolidated with no excess pore water pressures
present.
The minimum value of the maximum past stress for a particular layer can be estimated
reasonably well from either CPT results or from Oedometer tests.
20080326_Appendix A.doc A-1 Issue Date: 03-2008

Revision: 6
A2.0 STEP-BY-STEP PROCEDURE
The simplified procedure uses the estimated maximum past stress, σ’v,max, and the estimated
effective friction angle (φ’) in conjunction with the estimated SHANSEP parameters and the
general SHANSEP equation to develop the tri-linear envelope. The recommended envelope
is comprised of portions of the drained strength envelope, the undrained strength envelope
for over-consolidated conditions and the undrained envelope for normally consolidated
conditions. A typical plot of the three envelopes is shown on Figure A1. In addition to the
procedure outlined here, it is highly recommended that engineering judgment be used along
with information from other sources to adjust the final envelopes, if needed. The final
strength envelope should be compared to measured strength values from supplemental
laboratory and field tests to corroborate the estimates of strength.

Shear Strength along Failure Plane
Drained Strength
NC Strength
SHANSEP Strength
Trilinear Envelope
Maximum Past
Pressure,
σ'B σ'v,max
Effective Stress on Failure Plane
Figure A1. Typical Strength Envelopes.
The steps in the following subsections are generally recommended for the development of
the tri-linear envelope.
A2.1 Estimate SHANSEP Parameters
The SHANSEP equation, in simple terms, provides a means to estimate the undrained
strength ratio (Su / p’) for overconsolidated conditions using the undrained strength ratio (Su /
p’)NC for normally consolidated conditions and an estimate of the OCR value.

Revision: 6
Su / p’ = m * OCRα (1)
Where,
(Su / p’) is the strength ratio value corresponding to any value of OCR;
m is equal to the strength ratio value, (Su / p’)NC, for normally consolidated
condition; and
α is a SHANSEP parameter.
The SHANSEP parameters, m and α, are typically estimated from a suite of triaxial
consolidated, undrained (CIU) tests and/or Direct Simple Shear (DSS) tests. In the absence
of this information, it is recommended to use the following values for the two parameters:
⎛ Su ⎞
⎜ ⎟ = m = 0.23 for DSS mode of shearing (m=0.3 for CIU mode of shearing),
⎜σ ' ⎟
⎝ v ,c ⎠ NC
α = 0.8
A2.2 Estimate Maximum Past Stress σ’v,max
The minimum value of the maximum past pressure, σ’v,max, for the entire layer should
preferably be estimated from oedometer test results. However, in the absence of oedometer
test results, the σ’v,max could be estimated from CPT test results in conjunction with the
SHANSEP equation. The derivation of this procedure is as follows:
α
⎛ Su ⎞ ⎛σ ' ⎞
⎜⎜ ⎟⎟ = m ⋅ OCR α = m ⋅ ⎜ v ,max ⎟ ⇒
⎜ σ' ⎟
⎝ p' ⎠ ⎝ v ,o ⎠
⎛ (S p ') ⎞
1
α
σ ' v ,max = σ ' v ,o ⋅⎜ u ⎟ (2)
⎝ m ⎠
Where,
(Su / p’) is the strength ratio value estimated from the CPT tip resistance
using published correlations to CIU triaxial strength (average value
representative of the entire layer). The estimated value of Su/p’ is reduced by
25 percent to convert the triaxial compression type strength to DSS type
strength and also to account for the rate of loading effects;
m and α are the SHANSEP parameters as discussed above; and
σ’v,o is the vertical effective stress in the middle of the layer at the time of the
CPT test.
In most cases it is reasonable to assume that the maximum past stress value will be at a
minimum close to the levee toe. However, this should be checked on a case by case basis
depending upon the quality and availability of data. In the absence of good and reliable data,

Revision: 6
it is prudent to assume that the layer is normally consolidated and the maximum past stress
for the entire layer is equal to the vertical effective stress in the middle of the layer measured
underneath the toe of the levee.
If the maximum past stress is estimated from the CPT test results, the σ’v,o , and the Su / p’
ratio should be calculated based on the best estimate of the water table at the time of testing
(i.e., “low” water table). If this information is not available, a water table at 2 feet below the
landside toe of the levee could be assumed.
Because of the inherent uncertainty involved in estimating Su / p’ and σ’v,o from CPT results,
it is recommended that layers with computed OCR < 1.5 and/or Su / p’ < 0.30 (at the time of
site investigations) be treated as normally consolidated layers with OCR = 1.0 and Su / p’ =
0.23, unless there is strong evidence (e.g., consolidation test) to justify OCR values between
1.0 and 1.5.
A2.3 Estimate Undrained Strength Envelope
Knowing the effective maximum past pressure (σ’v,max), and using the estimated SHANSEP
parameters, it is possible to estimate the undrained strength for all vertical stress levels with
the following equation:
S u = m ⋅ OCR α ⋅ σ ' v ,c = m ⋅ (σ ' v ,max ) ⋅ (σ ' v ,c )

α 1−α
(3)
(Please note that if σ ' v ,c ≥ σ ' v , max , the soil is normally consolidated, and
S u = m ⋅ σ ' v ,c )
1.1 A2.4 Estimate Drained Strength Envelope
The drained strength envelope for a non-free-draining, fine-grained materials is estimated

from the triaxial CIU test results and/or the plasticity index (PI) correlations (see Appendix C).
With limited PI data, the following guidelines could also be used:
For PI < 40 use c’ = 0 and ϕ’ = 29 degrees

For PI > 40 use c’ = 0 and ϕ’= 23 degrees
Knowing the effective stress friction angle ( ϕ ' ) it is possible to calculate the drained strength
envelope using following equation:
τ = σ ' n ⋅ tan (ϕ ')

(4)

Revision: 6
A2.5 Calculate Intersection of Drained and Undrained Strength Envelopes

σ’B
Using equations 3 and 4, the normal stress σ’B at which the drained and undrained strength
envelopes for an unconsolidated soil intersect can be calculated as follows:
α
⎛ σ ' v ,max ⎞
S u = τ ⇒ m ⋅ OCR = tan (ϕ ') ⇒ m ⋅ ⎜⎜
α ⎟ = tan (ϕ ') ⇒
⎟
⎝ σ ' v ,c ⎠
1
⎛ m ⎞ α
σ ' B = σ ' v ,c = σ ' v ,max ⋅⎜⎜ ⎟⎟ (5)
⎝ tan (ϕ ') ⎠
For stress values less than σ’B, the undrained strength is higher than the drained strength;
and for stress values between σ’B and σ’v,max , the shear strength gradually transitions
from the drained strength to the normally consolidated undrained strength.
A3.0 RECOMMENDED ENVELOPE
The recommended tri-linear envelope is shown on Figure A2. As shown on this figure, the
recommended envelope is developed by taking the lower of the drained and
overconsolidated undrained envelopes for stress values less than the maximum past stress.
For stress values greater than the maximum past stress, the recommended envelope follows
the normally consolidated strength envelope. For static analyses, the normally consolidated
undrained strength envelope may be assumed to have a slope of 0.23 unless specific test
results indicate otherwise. In other words, the SHANSEP parameter “m” for the static
analyses may be taken as 0.23.
To facilitate the input of this “non-linear” strength envelope in the UTEXAS4 data file, a
simplified procedure is developed as shown on Figure A2.

Revision: 6

Shear Strength
(σ'v,o , su )
(σ'B , σ'B ·tan φ ')

(σ'v,max , m·σ'v,max)
Effective Stress
Figure A2. Recommended Tri-linear Envelope for Input in UTEXAS4

.

Revision: 6
Attachment A1: Subdivision of Foundation Clay Layer when using the Tri-
linear Envelope
Attachment A1, provides further explanation of the following text that appears in Appendix A
and in the main document:
In most cases, use of the tri-linear envelope eliminates the need to subdivide a
foundation clay layer into zones (in the horizontal direction) when moving from
beyond the toe to underneath the crest of the levee. This is true for situations where
a constant value of maximum past stress could be assumed for the entire layer. For
other situations where the maximum past stress varies along the layer, the layer
needs to be subdivided into zones and separate tri-linear envelopes need to be
developed for each zone. In either case, the tri-linear envelope, once developed
based on the maximum past stress value(before the flood event), automatically
accounts for the increase in OCR with rise in the water level associated with a flood
event.
In estimating the stability of a levee section, for any flood event, the analysis methodology
intends to account for the following three loading stages.
Stage 1: Conditions before levee construction:
Clay Layer
• Max. past stress (σ’max) in the clay layer = σ’max_Initial

• σ’max_Initial value is generally assumed to be constant in the horizontal direction. (For
practical purposes, for the existing conditions, this value is measured around the
existing landside toe of the levee.)
Stage 2: Conditions after levee construction under normal summer water level:
This loading stage is assumed to be the initial consolidation state of the levee, prior to the
flood event.
Clay Layer

Revision: 6
Two scenarios could occur with the σ’max value in the clay layer:
• Scenario 1: If the clay layer in Stage 1 was highly overconsolidated, then the
σ’max_Initial value would still remain the maximum past stress within the clay layer
underneath the levee.
• Scenario 2: If the clay layer in Stage 1 was normally consolidated (or lightly
overconsolidated), then the maximum past stress within the clay layer would
increase from its original value at the toe (σ’max_Initial) to a higher value underneath
the levee crest equal to the new effective overburden stress.
Stage 3: Conditions following a flood event:
During this loading stage the water level rises and the effective stresses within the clay layer
decrease leading to an apparent increase in the OCR values. In accordance with the
SHANSEP approach, these new OCR values need to be used in calculating the undrained
shear strength of the clay layer.
Clay Layer
For scenario 1, the entire clay layer could be modeled as a single layer with a single tri-
linear envelope to effectively represent the change in strength with increase in OCR
associated with the rise in water level. The working of the tri-linear envelope for this scenario
is shown on Figure A4. Because the maximum past stress, for this scenario, is assumed to
be the same through out the layer, a single tri-linear envelope could effectively model the
stress/strength conditions for the entire layer.
However, for scenario 2, the clay layer would need to be subdivided into zones with a
constant σ’max value assumed within each zone. The zones, as an example, would look like
this:

Revision: 6
Please note that the subdivision of the layer into zones and the calculation of the σ’max value
for each zone needs to be based on the Stage 2 conditions and NOT on the Stage 3
conditions. Based on the σ’max value for each zone, a unique tri-linear envelope needs to be
developed for each zone. Once the tri-linear envelopes have been developed for all the
zones, the change in the stress/strength values induced by the Stage 3 conditions would
automatically be accounted for by the envelopes similar to the example shown on Figure A4.
Figure A4. Application of Tri-linear Envelope (Sample Plot).

Revision: 6
APPENDIX B
APPENDIX B
Kozeny-Carman Publication
REVISION 06
Issue Date: 03-2008

Revision: 6
APPENDIX C
APPENDIX C
Shear Strength Correlations

REVISION 06
Issue Date: 03-2008

Revision: 6
Appendix C
From CPT Data (Kulhawy and Mayne, 1990):
φ ' = 17.6 o + 11.0 ⋅ log(q c1 )

qc
where: q c1 = = normalized tip resistance
(σ 'vo σ atm )0.5
From SPT Data:
Peak Friction Angle of Sands from SPT Resistance

(adapted from Hatanakata & Uchida, 1996; Figure from FHWA NHI, 2002).
Figure C.1 – Drained Friction Angle for Cohesionless Soils
Juan F. Perri 11/27/2007 1/3

Appendix C
From Plasticity Index:
A similar correlation is given by Mitchell (1976):
sin (φ ') = 0.8 − 0.094 ⋅ ln(PI )
Where PI = Plasticity Index
Figure C.2 – Drained Friction Angle for Cohesive Soils
Juan F. Perri 11/27/2007 2/3

Appendix C
From CPT Data:

q −σv
su = t
N kt
where Nkt ≈ 15 – 20 (recommended value for calculations is 15)
From SPT Data:

Soil Consistency SPT (N) Su (psf)
Very Soft <2 < 250
Soft 2–4 250 – 500
Medium 4–8 500 – 1000
Stiff 8 – 15 1000 – 2000
Very Stiff 15 – 30 2000 – 4000
Hard > 30 > 4000
After Terzaghi and Peck (1967)
Kulhawy and Mayne (1990) approximate the results of the Terzaghi and Peck (1967)
table with the following equation:
su
≈ 0.06 ⋅ N
pa
Where pa is the atmospheric pressure.
From Plasticity Index:
(after Jamiolkowski et al., 1985)
Figure C.3 – Undrained Strength of Cohesive Soils
Juan F. Perri 11/27/2007 3/3

APPENDIX D
APPENDIX D
Lab Test Assignment Guidelines

REVISION 06
Issue Date: 03-2008

Revision: 6
GUIDELINES FOR ASSIGNING LAB TESTING FOR SEEPAGE ANALYSIS
The following general guidelines address the need for laboratory testing to assist in the
estimation of permeability for the purpose of performing seepage analysis and preparing
preliminary design of mitigation measures to improve levee performance. Guidelines for
assigning laboratory testing for the purpose of estimating soil strength characteristics are
presented in a separate section below. Many tests, such as index property tests, serve dual
purposes since they can be used to estimate both strength and permeability.
These guidelines are not meant to supercede engineering judgment. Additional index testing
may be performed to evaluate the reliability of field classification. All samples tested for
Atterberg Limits should be tested for as-received moisture content, if sample size permits.
LABORATORY TESTING WITHIN THE BLANKET LAYER

The “Blanket” is usually made up of several layers consisting of clay, sandy clay, clayey sand,
silt, sandy silt, and silty sand. In order to determine the transformed blanket layer thickness in
accordance with USACE EM 1110-2-1913 Appendix B, or to construct a seepage model that
resembles the subsurface conditions it is necessary to assign both horizontal and vertical
permeability to each layer of the blanket. Lab testing should be performed to help assess the
permeability of each layer.
Layers logged as containing about 25 percent clay or more will typically be modeled with a very
low permeability. Atterberg Limits tests should be performed in these clayey layers to confirm
the fines are clay and not silt. Sieve and hydrometer tests should be performed on samples of
clayey sand or clayey gravel containing 20 percent fines or less since Atterberg Limits testing
may not yield meaningful results. Complex mixtures of cobble, gravel, sand, silt and clay may
require both Atterberg Limits and sieve/hydrometer testing to fully characterize the sample.
Layers logged as silt or sandy silt, with slow dilatency and low plasticity, will typically be
modeled with a low permeability. Atterberg Limits tests should be performed in these silty
layers to confirm the fines are silt and not clay.
Layers logged as silt or sandy silt, with rapid dilatency and no plasticity will typically be
modeled with a moderate to low permeability based on the Kozeny-Carmen equation. Sieve and
hydrometer tests should be performed.
Layers logged as silty sand will typically be modeled with a moderate permeability based on the
Kozeny-Carmen equation. Sieve and hydrometer tests should be performed. If the estimated
fines content within a layer varies by 10 percent or more, then multiple samples should be tested.
Samples should be selected to represent the full range of estimated fines contents. This
information is needed to estimate the ratio of horizontal to vertical permeability of the layer.
Experience within the Sacramento Valley has shown that many seepage analyses are very
sensitive to the anisotropy of silty sand layers commonly encountered between the blanket and
the underlying pervious zone.
The Kozeny-Carmen equation requires an estimate of the void ratio, which is a function of
specific gravity. Some specific gravity testing may be needed to determine the site-specific
characteristics of the soil particles.
It is also important to know the unit weight of the materials making up the blanket since it
influences the allowable uplift pressure and critical gradient. Consequently, consideration
should be given to obtaining California or Modified California samples in P2 borings and
performing moisture density testing on relatively undisturbed samples.
LABORATORY TESTING WITHIN THE PERVIOUS ZONE

The “Pervious Zone” is usually made up of several layers consisting of gravel, sandy gravel,
gravelly sand, sand, and sand with silt. In order to determine the weighted average horizontal
permeability in accordance with USACE EM 1110-2-1913 Appendix B, or to construct a
seepage model that resembles the subsurface conditions it is necessary to assign both horizontal
and vertical permeability to each layer within the pervious zone. Layers logged as sand or gravel
will typically be modeled with a high to very high permeability based on the Kozeny-Carmen
equation. The presence of 5 to 10 percent silt will decrease the permeability by one to two
orders of magnitude. Sieve and hydrometer tests should be performed.
If the estimated fines content within a layer varies by 5 percent or more, then multiple samples
should be tested. Samples should be selected to represent the full range of estimated fines
contents. This information is needed to estimate the variation of horizontal permeability within
the pervious zone.
LABORATORY TESTING WITHIN THE AQUITARD

In order to perform seepage analysis and design mitigation measures such as relief wells and
cutoff walls it is necessary to know the depth to a reliable aquitard. In most cases P1A borings
(borings at intervals of 5000 feet) are being extended 10 feet into an aquitard. Other P1 borings
may encounter suitable aquitard material within the prescribed drilling depth of four times the
levee height, below the levee crown. In order to confirm the classification of these fine-grained
materials Atterberg Limits testing should be performed.
LABORATORY TESTING WITHIN THE LEVEE EMBANKMENT

Given sufficient time, high water behind a levee will eventually result in a free seepage face on
the landside of the levee. The height of the seepage face obove the landside toe depends upon
the geometry of the embankment and composition of the levee and levee foundation materials.
The susceptibility of the levee to failure due to through-seepage depends upon the erodability of
the embankment and foundation, which is usually estimated from plasticity. The height of the
free seepage face can be influenced by thin layering (on the order of inches), which results in a
horizontal permeability higher than the vertical permeability. Thicker layers of pervious
material (on the order of 1 to 2 feet) can convey water horizontally through the levee causing
seepage at or near the floodwater surface elevation. While the use of the laboratory data from
within the levee embankment is somewhat different than for the blanket and pervious zones, the
guidelines presented above can be used for selecting samples from the embankment for
laboratory testing associated with seepage analysis.
GUIDELINES FOR ASSIGNING LAB TESTING FOR STABILITY ANALYSIS

The following general guidelines address the need for laboratory testing to assist in the
estimation of strength parameters for the purpose of performing stability analysis. Guidelines for
assigning laboratory testing for the purpose of estimating permeability are presented in a separate
section above. Many tests, such as index property tests, serve dual purposes since they can be
used to estimate both strength and permeability.
These guidelines pertain to soils with strength properties dominated by their fine-grained
fractions. A typical fine-grained soil may be easily tested in situ or sampled and tested in the
laboratory. A coarse-grained soil, such as clayey gravel with sand, may be heavily influenced by
its fine-grained fraction but difficult to test either in situ or in the laboratory. In such cases,
strength parameters may need to be derived from correlations with Plasticity Index or Liquid
Limit, comparison with similar fine-grained deposit from the same formation, or strength testing
on remolded samples. Shear strength correlations for cohesionless soils are presented in
Appendix C.
These guidelines are not meant to supercede engineering judgment. Additional index testing
may be performed to evaluate the reliability of field classification. All samples tested for
Atterberg Limits should be tested for as-received moisture content, if sample size permits.
SAMPLE LOCATION AND DEPTH

Landside stability is usually a function of the strength of the embankment and the strength of the
foundation layer within about half the levee height below the levee. Thin layers of weak
material within the foundation can compromise levee performance. While some type of average
strength will likely be used for the embankment and another for the foundation the analyst
should be on the lookout for any indications in boring logs or CPT soundings that a weak layer is
present.
Samples obtained from foundation soil beneath the embankment may have been significantly
compressed as compared to samples obtained either outboard or inboard of the levee
embankment. This influence is more pronounced in normally consolidated or slightly
overconsolidated Holocene alluvium deposits and less of a factor in overconsolidated Pleistocene
deposits. Samples obtained from P2 borings are likely to provide better representation of the
minimum foundation layer strength. In situ testing, such as vane shear, should be considered
where appropriate to supplement laboratory testing.
Where the channel is close to the levee, deeper layers well below the levee toe but above the
river thalweg can influence waterside stability and should be tested.
CONSOLIDATION TESTING
The strength exhibited during shear failure is strongly influence by the over consolidation ratio
(OCR) and the maximum past vertical stress of the soil. As such, the oedometer test provides
great insite into anticipated soil behavior. Oedometer tests should be performed prior to triaxial
testing so that appropriate triaxial loading conditions may be selected.
TRIAXIAL COMPRESSION SHEAR TESTING

Isotropically consolidated undrained (ICU) triaxial compression test results can be used to
estimate the SANSEP parameters m and alpha, as well as the drained strength envelope for
dilative fine-grained materials. Consequently, the ICU test is the preferred strength test for levee
stability evaluation.
ATTERBERG LIMITS TESTING
Atterberg Limits testing should be performed in association with each ICU test in order to
facilitate correlation and extrapolation of the limited number of strength tests to the remainder of
the levee embankment and foundation.
TESTING FREQUENCY
In general, oedometer, ICU, and Atterberg Limits testing should be performed on one sample
from the levee embankment and one sample from the foundation layer for each 5000 feet of
levee. This frequency may vary depending upon levee composition, condition, and geometry.
Wherever possible, CPT correlation holes should be placed adjacent to undisturbed (Shelby
Tube) sampling locations to calibrate CPT correlation methods.
APPENDIX E
APPENDIX E
Blanket Theory Analysis Template

REVISION 06
Issue Date: 03-2008

Revision: 6
Seepage Analysis - Blanket Theory Approach SHEET 1/6
Mathematical Analysis of Underseepage and Substratum Pressure
Reference: USACE Manual EM 1110-2-1913, "Design and Construction of Levees", (Attachment

B: Mathematical Analysis of Underseepage and Substratum Pressure).
Parameters:
L1 x3
x1 L2 L3
H z kbr 1 M
kf kbl
d
s
x
L
Notes:
1. If a profile has less than 5 sub - top strata contributing to the blanket (minimum 1 layer needed), or less than 5 sub-pervious layers,
(minimum 1 layer needed) use only the necessary numbers of sub-layers
2. If a profile has more than 5 sub - top strata contributing to the blanket, or more than 5 sub-pervious layers,
this spreadsheet need to be modified.
3. For the given geometry and stratification, select an appropiate case (as shown below), and then provide the input data for that case.
4. All Elevations should be referenced to same datum.
5. Enter Value
Calculated Value
Recommended Value
6. This spreadhseet is prepared as an example. The user should understand the analytical procedures used in this spreadsheet. The
results should be verified independently.
Analysis Case:
Case 6 Semipervious Landside Top Stratum and no Riverside Top Stratum
Case 6
L
dH `
Infinite length of foundation and top stratum on landside
Case 7 Semipervious Top Strata Both Riverside and Landside
Case 7a
k h k
Infinite length of foundation and top stratum on landside
L3 = ∞
Case 7b
d k h k
Seepage block on landside
L3 is finite to a seepage block
Case 7c
k h k
Finite distance to an open seepage exit
L3 is finite to an open seepage exit
C:\DOCUME~1\LISA_D~1\LOCALS~1\Temp\notesE1EF34\BTA Spreadsheet_URS (Sacramento) for Guidance Document.xls

Transoformation of Blanket Layer SHEET 2/6
Transformed Strata for Transformed Strata for

kb Computation of SEEPAGE & Computation of Maximum
Ft =
Case Actual Strata kn SUBSTRATUM PRESSURE kb Allowable UPLIFT
z1 Layer 1 k 1 zb1 = z1 Layer 1 zb1 = z1 Layer 1

zb zb2 = z2/2 k zr
Layer 2 Layer 2 zb2 = z2/2 Layer 2
z2 2k 1/2
I z zb3 = z3/4 Layer 3 zb3 = z3/4 Layer 3
Sand Sand
z3 Layer 3 4k 1/4
Sand
zb1 = z1/4 Layer 1

z1 Layer 1 4k 1/4 zb1 = z1 Layer 1
zb2 = z2/2 Layer 2
k
z2 Layer 2 2k zb Layer 2
1/2 zb2 = z2
II z zr
zb3 = z3 Layer 3
z3 Layer 3 k 1 zb3 = z3 Layer 3

Sand
Sand Sand
zb1 = z1/2 Layer 1 Layer 1

z1 Layer 1 2k 1/2 zb1 = z1
zb zb2 = z2
Layer 2
z2 Layer 2 k zr zb2 = z2 Layer 2
1 zb3 = z3/4 Layer 3 k
III z
Sand zb3 = z3/4 Layer 3
z3 Layer 3 1/4 Sand

4k
Sand
EXAMPLE
Figure B-2 of EM 1110-2-1913
C:\DOCUME~1\LISA_D~1\LOCALS~1\Temp\notesE1EF34\BTA Spreadsheet_URS (Sacramento) for Guidance Document.xls

Seepage Analysis - Blanket Theory Approach SHEET 3/6
EXAMPLE 1 (CASE I in Transpormation of top strata and Case 6 in Underseepage flow and substratum pressure)
Reference: USACE Manual EM 1110-2-1913, "Design and Construction of Levees", (Attachment B: Mathematical Analysis of Underseepage and Substratum Pressure).
URS Project No.:

Project Name: Urban Levee Geotechnical Evaluations Program
Calculation By: Ahmed Bayoumi, PhD, Adil Shah, PE, and Gyeong Hong, PhD Date: 3/12/2008
Check By: Khaled Chowdhury, PE Date: 3/12/2008
Study Area: Study Area
Station: XYZ + 01
Reference
ABC
Boring/CPT
Case: 7
Reach: R1 Enter Value
Calculated Value
Levee and River Dimensions Recommended Value
Design flood elevation, ELdesign water = 75.1 ft.
Top of levee elevation, ELlevee crest = 81.9 ft.
Flood Level
Landside groundsurface elevation, ELlandside GS = 64 ft. Flood Stage
(NAVD 88)
Riverside groundsurface elevation, ELriverside GS = 30.9 ft. 100 yr 75.1
Levee crest width, Lcrest = 20 ft. 200 yr 78.4
Width of riverside slope of levee, L riverside slope = 102 ft. 1957 77.7
Width of landside slope of levee, Llandside slope = 50 ft. Levee Crest 81.9
Distance from River to Riverside Levee Toe, L1 = 0 ft.
Base width of levee and berm, L2, Lriverside slope+Lcrest+ Llandside slope= 172 ft.
Length of foundation and top stratum beyond landside levee toe, L3 = - ft.
Net head of levee, H = ELdesign water - ELlandside GS 11.1 ft.
L1 L2
x3
L2 L3
x1
H z kbr 1 M
kf kbl
d
s
x
L
C:\DOCUME~1\LISA_D~1\LOCALS~1\Temp\notesE1EF34\BTA Spreadsheet_URS (Sacramento) for Guidance Document.xlsExample 1 (Case I and Case 6)

Determination of Transformed Thickness for Seepage and Substratum Pressure Computations SHEET 4/6
Note: Identify appropriate case (I/II/III) based on Figure B-2 of Manual EM 1110-2-1913. Add additional layers where needed.
Determination of Transformed Thickness for Seepage and Substratum Pressure Computations

(a) Landside
Vertical
Permeability of Total
Actual Vertical Least Transformed
Layer
Least Check Transformed

Case (I/II/III) Strata Thickness, z n Permeability, K n Permeable Ft = Kb/Kn Thickness of
Permeable Case Thickness, z bl
(ft) (ft/day) Layers stratum, ztn (ft)
Layer, Kb (ft)
(ft/day)
1 Clay 5 5.0E-03 x 1.0000 5.0

Case I: Least permeable layer at top (Transformed thickness of
2 Sandy Silt 5 1.0E-02 - 0.5000 2.5 blanket = actual top layer thickness + transformed thickness of
I 3 Silty Sand 5 5.0E-02 5.0E-03 - CASE III 0.1000 0.5 11.0 all other layers)
4 Sandy Silt 5 1.0E-02 - 0.5000 2.5
5 Silty Sand 5 5.0E-02 - 0.1000 0.5
Case II: Least permeable layer at bottom (Transformed

thickness of blanket = actual total thickness of blanket)
Case III: Least permeable layer at middle (Transformed

thickness of blanket = actual thickness of least permeable
layer and all layers above it + transformed thickness of
underlying layers)
Transformed thickness of landside top stratum (blanket), z bl = 11 ft.
(a) Riverside
Vertical
Permeability of Total
Actual Vertical Least Transformed
Layer
Least Check Transformed

Case (I/II/III) Strata Thickness, z n Permeability, K n Permeable Ft = Kb/Kn Thickness of
Permeable Case Thickness, z br
(ft) (ft/day) Layers stratum, ztn (ft)
Layer, Kb (ft)
(ft/day)
1 Clay 5 5.0E-03 x 1.0000 5.0

Case I: Least permeable layer at top (Transformed thickness of
2 Sandy Silt 5 1.0E-02 - 0.5000 2.5 blanket = actual top layer thickness + transformed thickness of
I 3 Silty Sand 5 5.0E-02 5.0E-03 - CASE III 0.1000 0.5 11.0 all other layers)
4 Sandy Silt 5 1.0E-02 - 0.5000 2.5
5 Silty Sand 5 5.0E-02 - 0.1000 0.5


underlying layers)
Transformed thickness of riverside top stratum (blanket), z br = 11 ft.
For Case 7: Verify if z bl = zbr. Otherwise calculate zbr
Determination of Transformed Blanket Thickness for Uplift
Permeability of
Total
Actual Actual Vertical Least Least Transformed
Layer
Check Transformed
Case (I/II/III) Strata Thickness, z n Permeability, K n Permeable Permeable Ft = Kb/Kn Thickness, z tn
Case Thickness, z t
(ft) (ft/day) Layer, Kb Layers (ft)
(ft)
(ft/day)
1 Clay 5 5.0E-03 x 1.0000 5.0 Case I: Least permeable layer at top (Transformed thickness of
2 Sandy Silt 5 1.0E-02 0 0.5000 2.5 blanket = actual top layer thickness + transformed thickness of
I 5.0E-03 CASE III 11.0 underlying layers)
3 Silty Sand 5 5.0E-02 0 0.1000 0.5
4 Sandy Silt 5 1.0E-02 0 0.5000 2.5
5 Silty Sand 5 5.0E-02 0 0.1000 0.5

underlying layers)
Transformed thickness of landside top stratum for uplift, z t = 11 ft.

Effective Seepage Entrance and Exit SHEET 5/6
31.5 ft.
Actual thickness of top stratum on landside (blanket), Z =
Transformed thickness of landside top stratum, Z bl = 11 ft.
(from SHEET 4)
Vertical permeability of top stratum, k bl = 5.0E-03 ft/day
Thickness of pervious stratum, d = 17.5 ft.

Horizontal permeability of pervious stratum, k f = 25 ft./day
k bl
Constant for effective seepage exit calculation, C = 0.0010 1/ft. c =
k f z bl d
Distance from landside levee toe to effective seepage exit,
Landside
x3
L3 criteria for x 3 Applicable Case

L3 x3 (ft)
calculation (Yes/No) k f z bl d
1
x3 = =
Infinite Yes ∞ 981 Case 6 and 7a For L3 = ∞ c k bl Infinite when, L3>2/c
Finite distance to a 1
seepage block
No - - Case 7b For L3 = finite distance to a seepage block x3 =
c * tanh( cL3 )
Finite distance to an tanh( cL3 )

open seepage exit
No - - Case 7c For L3 = finite distance to an open seepage exit x3 =
c
For SEEP/W check, use L 3toe (distance between levee centerline to the landside
Seep/W check No - -
toe) instead of L 3, ft.
x3 = tanh( c * ( 2000 − x L 3toe ) / c
x3 for seepage and gradient calculation = 981
Distance from effective source seepage entry to riverside

Riverside
levee toe, x1
Waterside transformed blanket thickness, z br = 11 ft. Note: For Case 7 verify if z br = zbl; Otherwise calculate zbr
Vertical permeability of riverside top stratum, k br = 0.005 ft/day
k br
Constant for effective seepage entrance calculation, C = 0.0010 c =
k f z br d
L1 criteria for x1 Applicable Case

L1 x1 (ft)
calculation (Yes/No)
Open seepage tanh( cL1 )
entrance at L1
no 0 - All Case 7 x1 =
c
1
Seepage block at L1 No - - All Case 7 x1 =
c * tanh( cL1 )
Infinite No ∞ - All Case 7 For L1 = ∞ x1 =

1
c
=
k f zbr d
kbr
Infinite when, L1>2/c
No riverside blanket Yes 0 7.5 Case 6: (Restriction: not applicable with L3=0) (x1 = 0.43d)
x1 for seepage and gradient calculation = 7.5

Computation of Seepage Flow and Head below Blanket SHEET 6/6
Natural seepage per unit length of levee, Q s = (SF)kfH
Case 6: Semipervious landside top stratum (blanket) and no riverside top stratum (blanket)
Case 7: Semipervious top stratum (blanket) on both landside and riverside
SF = Shape Factor d
SF =
x1 + L 2 + x 3
kf = Horizontal permeability of pervious layer
h0 = Excess pressure head beneath the blanket at the x3

landside levee toe ho = H ( )
x1 + L 2 + x 3
Horizontal
Transformed
permeability Thickness of Distance from effective Base width of
thickness of Net Head of levee, H Distance from landside levee toe to effective Shape Factor,
of pervious pervious source seepage entry to levee and
landside blanket for (ft.) seepage exit, x3 (ft.) SF
layer, kf stratum, d (ft.) riverside levee toe, x1 (ft.) berm, L2 (ft.)
uplift, zt (ft.)
(ft/day)
11 11.07 25 17.5 981 7.5 172 0.0151
Seepage per unit length of levee, Qs Excess Hydrostatic Head beneath landside Average Exit Gradient through the blanket at
(ft3/day per LF) blanket, ho (ft.) the landside levee toe, io (=ho/zt)
4.17 9.4 0.85
Summary of Results of Blanket Theory Analysis
Seepage per unit length Excess Average Exit

Flood Level
Boring/CPT ID Station Flood Stage Net Head of levee, H (ft.) of levee, Qs (ft3/day per Hydrostatic Gradient, io Case Notes
(NAVD 88)
LF) Head, ho (ft.) (=ho/zt)
100 yr 75.1 11.07 4.17 9.4 0.85

200 yr 78.4 14.41 5.43 12.2 1.11
ABC XYZ + 01 6
1957 77.7 13.74 5.18 11.6 1.06
Levee Crest 81.9 17.92 6.76 15.1 1.38
1.5 16
14
Average Vertical Exit Gradient, I o
Excess Hydrostatic Head, ho (ft.)
12
1.0
10
6
0.5
Excess Hydrostatic Head, ho (ft.) Average Exit Gradient, io (=ho/zt) 2
0.0 0
74 75 76 77 78 79 80 81 82 83
Flood Level (NAVD 88)

APPENDIX F
APPENDIX F
Plan and Profile Presentation

REVISION 06
Issue Date: 03-2008

Revision: 6
Graphic Standards Plan and Profile Drawings
General
Plan and Profiles are on D size sheets (22 x 34 inches).
All sheets should have a 1” border on all sides
Should follow program standard template as shown in Attachments F-1 and F-2
Every sheet should include a “detail area” to locate where it is
North arrows and scale should be included for plan views. Where levee alignments bend
and additional plan view is necessary, north arrows should be included in each view.
Profiles include separate vertical and horizontal scales as detailed below.
CAD – Profile Standards

Cross section boundaries (X axis) should be defined and identified by the appropriate
station numbers.
Stationing at bottom of drawing (X axis) should be labeled “STATIONING, Feet”
Elevation on sides (Y axis) of drawing should be labeled “ELEVATION, Feet (NAVD 88)”
Historical borings/CPTs have an asterisk (*) placed at the end of the boring ID label
Scales
Horizontal Scale should be labeled, “DISTANCE, Feet” with tick marks showing 100,
200, 300, and 400 ft on the top of the scale.
Vertical Scale should be labeled “ELEVATION, Feet” with tick marks showing 10, 20,
30, and 40 ft (on the outside, or the right side of the scale. Begin “0” at the bottom of
the scale.
Labeling Lines – Label each line one time on profile view and in close proximity to each
other.
Water Level line labeling
200 year = “200-Year Water Surface”
100 year = “100-Year Water Surface”
100 year 1957 = “1957-Design Water Surface”
Crest labeling = Crest shall be labeled, “Levee Crest”
Toe labeling = Toe shall be labeled, “Landside Levee Toe” or “Waterside Levee
Toe” depending on which it is.
Lab Data on Boring Logs - % Fines = Percent fines shall be shown as a solid number,
rounded accordingly. No Decimals.
Cemented layers will be labeled (called out) as “Hard Pan”. Hard Pan will be called out on
the right, with an arrow pointing to the high “Q”. No hatching of line-work will be used. Only
upper hard pan needs to be called out.
Soft Clays should be called out when above hard pan. Soft clays may be identified by low
blow counts or on CPT signature plots, indicated by no “Q” and high FR% that goes off the
scale (soil behavior type 2). “Soft” will be shown on the left side of the CPT, with an arrow
pointing to the high (often broken) “FR%” on the graph.
GIS Plan View Standards

GIS Plan Views contain the following information:
Program levee alignment with labeled stationing
Historical explorations
Current and/or proposed explorations
Historical levee distress areas and types (seepage, boils, failures, etc.)
Graphic Standards Plan and Profile Drawings
Historic levee mile designations

Geomorphology
Known mitigation measures (berms, relief wells, slurry walls, etc.)
Cross-section locations selected for analysis
Reach designations
Plan View is at 1”=200’ scale
Reach Labels = Where practical Reach labels should be placed on the top part of bracket.
This may not always work because of bends on stationing, other things in way, etc. Can be
placed elsewhere when necessary.
Where a “reach” is not shown on a plate, add reach label, centered at the bottom of plan
view. Example, Reach A goes on for 3 plates. Identified on Plate 1 in plan view, on Plates 2
and 3, indicate by placing, “REACH A” centered at the bottom of Plan view.
Label geographic features – i.e. rivers, major roads
Cross-Section Analysis Lines – label both type of analysis and station. Example: “Slope
and Stability Analysis, STA. 1800+00.” Show stationing at top part of bracket. Label the test
performed at the bottom of the Bracket (ie., Seepage and Stability Analysis, or Seepage
Analysis).
For plan view GIS symbology standards see Attachment F-3

PLAN LEGEND
?
?
A
! Borehole Analysis Cross Section
ffffff Sheet Pile
Ç
#
( CPT
Area of Concern BBBBBB Subsidence
#

) CPT (Seismic)
# Berm D D Toe Drain
Piezometer
D D Berm Alteration (Explanation) Urban Project Levee
# Pothole
ü ü Berm Gravel
A
! Proposed Boring
R R Berm (Raised Grade Area)
Ç
#
( Proposed CPT
Ç Ç Ç
ÇÇ ÇÇ ÇÇ Boil
#

) Proposed CPT(Seismic)
# Buried Concrete Wall
Proposed Piezometer
Canal
A
! Historic Exploration
à à à Cracking
Ç Boil
ÇÇ
Ditch (Unlined)
à Cracking
Ditch (Concrete-Lined)
!
? Irrigation Well, pump, or other irrigation structure
Erosion PROFILE LEGEND
!
(2 Levee Mile Marker !
! Gate CPT - green (Fr)

/ Levee Penetration
Ground Resistivity CPT - red (Qt1)
!
? Monitoring Well
HEM Grid
2
% Proposed Exploration, USACE
Levee Penetration Grid Border
! Relief Well
Levee Stationing Levee Crest
!
* River Mile
Non- Urban Project Levee Levee Toe
!
( Sinkhole
VVVVVVVVV Overtopping Profile Interpolations
R
c Slide
Reach Wall
Test Pit
:
:
:
Riprap (Full Height) Waterside Levee Toe
Riprap (Partial Height) Water Level 100-Year
Active Page Index Box Seepage Water Level 100-Year (1957)

K:\GIS\Projects\DWR\GEOTECHNICAL\ArcMaps\PnP_Legend.mxd TNC 03.20.08 SAC
Cities !
( !
( Sinkhole Water Level 200-Year
Index Page Outline Box Rc

c Rc
RcR Slide Water Line
Levee District Boundary Slope Instability
Maintenance Area Boundary Slurry Cutoff Wall
Reclamation District Boundary Slurry Wall Gap Closure-Complete
Reclamation/Water District Boundary Slurry Wall Gap -Remaining
Plan and Profile

FT
Division of Flood Management Legend Template Updated

in association with:
3/20/08
RA
FT
Levee Evaluations Office
URBAN LEVEE
D
GEOTECHNICAL EVALUATIONS PROGRAM
RA
D
Yub
UP Intercept
k
a Rive
ee
Ja r
ck
Cr
S lou Bea
ut
gh r Riv
nc
er
Ho
Feather River
65 43 30 29
66 31 27
44 42 41 40 32
68 67 33 28 26
71 64 60 45 39
70 69 61 59 34
58 48 38
56 55 47 46 37
57 51 50 49 36 25
62 54 35
63 53 52
24
23
22
21
20
19
SURFICIAL GEOLOGY PLAN LEGEND
pt Canal
18
Surfical Geology was mapped at 1:20,000. A

! Boring
Ç
#
( CPT 17
East Interce
? Geologic contact; dashed where approximate,
?
dotted where concealed, queried where uncertain;
solid contacts within 30’ of line shown on map. A
! Historic Exploration
W 1911 Water bodies, circa 1911
# 16
Historic Piezometer
W 1937 Water bodies, circa 1937 !
(2 Levee Mile Marker
15
?
?
BP 1
Borrow pits present in 1937 Analysis Cross-Section
t Canal
14
GEOLOGIC UNITS* Reach 2
13
W 12
Canal ad 11
L
West Intercep
Levee; Culturally deposited sw
3 10
or
SP Spoils pile; Artificial fill, circa 1937 Levee Stationing th 9
Ca 8 ass
Non-Urban Project Levee 4 na u tter Byp
DT Dredge tailings; Spoils material from gold dredge operations l 7 S
Overbank deposits; sand, silt, and clay; deposited during Urban Project Levee 6
Rob high-stage water flow, overtopping channel banks 5
Crevasse splay deposits; Fine to coarse sand, with minor lenses Existing Mitigation Measure
Rcs
Tisdale Bypass
of clay deposited from breaching of natural or artifical levees
! Relief Well
Rdf Distributary fan deposits; Sand, silt and clay
Ch
Distributary channel deposits; Distributary channel deposits, Berm

er
Historical
Rdc trace gravel, sand, silt, and clay; channelized flow conducting
ok
sediment to floodplain Rip-Rap

ee
Overflow channels; Vertically stratified sand, silt and clay in

Rofc
C
channels occupiedwhen high-stage water overtops channel banks Slurry Cutoff Wall
an
Slough deposits; Sand, silt and clay, fining upward facies,

al
Rsl low-energy channel deposit

D D Toe Drain
Crevasse splay deposits; Cut off channel (chute); occurs on Historical Levee Distress
Rcu insides of meander bends within the river channel; on flood plain
– entrenchment of overflow channels into hydraulic mining debris Ç Boil
ÇÇ
Channel deposits; Sorted sands and silts; fining upward
iver
Rch (Rch 1911: channels as shown on historic topo) !
( Sinkhole
to R
Channel bar deposits; fine gravel, sand, and silt deposited
Rb in or along channel lateral margins R
c Slide
me n
Channel meander scroll deposits; Sand, silt and clay from Area of Concern
Rms lateral channel migration
ra
Ra
Alluvial deposits, undifferentiated; sand, silt, and minor lenses Erosion
Sac
of fine gravel
Bu Seepage
ttedeposited during
Overbank deposits; sand, silt, and clay;
Hob Cre Slope Instability
e
high-stage water flow, overtopping channel banks
k
Crevasse splay deposits; Fine to coarse sand, with minor
Hcs lenses of clay deposited from breaching of natural or
artifical levees PROFILE LEGEND
Hdf Distributary fan deposits; Sand, silt and clay
Levee Crest
Overflow channels; Vertically stratified sand, silt and clay in
Hofc channels occupied when high-stage water ngel Slough
Aovertops Landside Levee Toe
channel banks
Alluvial deposits undifferentiated; Sand, silt, and minor
Ha lenses of gravel; under cultivation in 1937 Waterside Levee Toe
Hch Channel deposits; Sorted sands and silts; fining upward Landside Ditch
Holocene
Channel bar deposits; Fine gravel, sand, and silt

Hb deposited in or along channel lateral margins
Waterside Ditch
But
Channel meander scroll deposits; Sand, silt and

Hms clay from lateral channel migration
100-Year Water Surface (Median)
te
Hdc 200-Year Water Surface (Median)

Slo
Distributary channel deposits, sand, silt, and clay

K:\GIS\Projects\DWR\GEOTECHNICAL\Urban\SUTTER\Arcmaps\Sutter_PnP_Overview.mxd CJK 03.11.08 SAC
ugh
Slough deposits; sand, silt and clay, fining upward facies, 1957-Design Water Surface
Hsl low-energy channel deposit
Alluvial fan deposits; Well graded gravel, sand, silt and clay; Stratigraphic Contact
Hfy volcanic lithologies
Alluvial deposits undifferentiated; Sand, silt, and minor n Weir Overflow
Qa lenses of fine gravel, under urbanization in 1937 Moulto
Basin deposits; Fine sand, silt and clay, dark yellow to dark
Qn yellowish brown, under cultivation in 1937
¯
Marsh deposits; Silt and clay, likely organic-rich; perennially
Colusa Bypass
Qs or seasonally submerged
Modesto Formation; upper member; Unconsolidated gravel,

Qmu
Pleistocene
sand, silt, and clay

Modesto Formation; lower member; Unconsolidated to 0 7,000 14,000
Qml semi-consolidated gravel, sand, silt and clay
Riverbank Formation; upper member; Semi-consolidated
Qru to consolidated gravel, sand, silt and minor clay Feet
Riverbank Formation; lower member; Consolidated gravel,
Qrl 1 inch equals 7,000 feet
sand, silt, and clay, generally associated with strong duripan
soil horizon
Pleistocene
Pliocene /
Tuff breccia; Volcanic tuff breccia (andesitic and rhyolitic) from

QTm Sutter Buttes, latest Pliocene Note: Rotary sonic boring IDs end in "S" and Department of Water Resources Plan and Profile Index and Legend
blow counts were not measured. in association with: Sutter Task Area Plate
Division of Flood Management Index
Levee Evaluations Office URBAN LEVEE GEOTECHNICAL EVALUATIONS PROGRAM MARCH 2008
Hob
RD
NION
Hob
Hob
OBA
Rob
Rob
¯
Rcs
722+26
STA. 2
Rcs CPT-97-8 Qmu

Rob L
2F-97-6 CF-88-6 0 100 200
Rob WL0001_044C
A
A A
!
2F-97-7
CPT-97-17
A
!
!
!
A
(
!#Ç WL0001_085B 2725+00
Ç
#
( Feet
2730+00
WL0001_085C
A
!
DRAFT
CPT-97-18 2720+00
A
! Seepa
ge
is
2F-97-5 Analys NOTES:
2735+00
A
! 2741+0
ÇÇ
1997 CPT-97-7 4
ÇÇ
ÇÇ
Ç
ÇÇ
Ç
2715+00
ÇÇ
Ç
CPT-97-11 1. Elevations of levee crest, landside toe, and waterside toe are
Ç
ÇÇ
CPT-97-6
Ç
ÇÇ
2F-97-11
Ç
ÇÇ
!
ÇÇ
Ç !
ÇÇ
approximate.
Ç
ÇÇ
! CPT-97-9
Ç
#
Ç
! (
Ç
!A
!! WL0001_045C
Ç
CPT-97-10 !
A A! ! 2F-97-4 2. The subsurface stratigraphy presented in this drawing is
! !
2F-97-17 ! ! ! ! WL0001_007S
!A
A
A
!!
WL0001_043C
2740+00 Ç
#
( preliminary and is based on available information. Areas
1997 ! ! WL0001_084C CPT-97-5 !A
A !A
!A!A! between explorations may differ from preliminary interpretation
Ç ! !
A
! 2710+00
ÇÇ ! !A
A
A
A
!
! ! A
A
#
!( !
Ç shown.
! CPT-97-16 3. Only explorations through the levee crest are shown on the
! Seepa 2745+00
2F-97-2 ge & profile.
CPT-97-2 Stabilit
y
CPT-97-12
CPT-97-1
Analys
W 1937is WL0001_046B 4. Historical borings on the profile are denoted by an asterisk (*)
A
! 2F-97-3 Seepa
ge CPT-97-13 WL0001_046C after the boring name. Historical boring locations and profiles
is 2750+00
Analys 04
2700+00 !
(
5
2705+00 STA. 2
705+
CPT-97-14
Ç
#
(
A
! are based on available information. Some historical explorations
are shown on the plan but not in the profile, if the log was not
1997
!
( available or was close to a new exploration.
Ra 5. USCS classification labels are not presented on the profiles for
Ra soil lenses (thickness less than 1.5 ft) and historical borings.
!
( 6. This is a color figure. Black and white reproductions should not
be relied upon as data will be lost.
Ra 7. To prevent scale distortion, this map should be printed on a “D”
Rch Rch size sheet (22x34 inches).
REACH T2
8. Surficial geology was mapped at 1:20,000
DETAIL AREA
100 100 Legend
Urban Project Levee
Non-Urban Project Levee
¯
200-YEAR WATER SURFACE
WL0001_084C WL0001_043C WL0001_044C WL0001_045C Plate Location
80 GS Elev. 69.83 ft
WL0001_007S
GS Elev. 70.13 ft
WL0001_085C WL0001_085B
100-YEAR WATER SURFACE GS Elev. 70.43 ft GS Elev. 70.37 ft 80
1957-DESIGN WATER SURFACE LEVEE CREST
GS Elev. 69.9 ft GS Elev. 69.69 ft GS Elev. 69.5 ft
Fr,% Qt1 Fr,% Qt1 Fr,% Qt1 10 Fr,% 0 Qt1 500 10 Fr,% 0
Qt1 500
10 0 500 % Fines N60(ASTM) 10 0 500 10 0 500 % Fines N60(ASTM)
GW-GM
! ! !
GW-GM
! !
! !
! ! !
68 !
!
!
! !
!
ML !
!
!
!
!
!
!
!
!
!
!
!
! ML 24
! ! ! !
!
! !
! !
ML 15
87 SILT AND CLAY
SILT AND CLAY
60 82 ML !
!
!
!
!
!
!
!
!
15 60
CL CL
20
ML LANDSIDE LEVEE TOE
!
!
!
! 11
CL !
!
!
!
!
!
!
!
! !
! !
!
! !
! !
!
! ! !
!
!
!
12
!
!
!
!
!
! ! SOFT
40 40
! !
ML
! !
! !
! !
! ! !
! !
!
CL 97 ML 87 HARD PAN HARD PAN

HARD PAN
CL 42
SILT AND CLAY
27
! !
SAND AND SILTY SAND CL !

!
!
!
SP-SM
CLAY
!
! !
!
! !
! !
!
20 20
! !
38 SM 12 28
ELEVATION, Feet (NAVD 88)
6.8 SAND AND SILTY SAND SAND AND SILTY SAND

SP-SM
SM 26
SAND AND SILTY SAND

! !
ML
! !
! !
!
!
71 !
!
11 ! !
! ! ! !
! !
30
! ! ! !
SW-SM
!
! ! ! ! ! ! ! !
! ! !
! ! ! ! !
! ! ! ! !
! ! ! !
! ! !
! ! !
! ! ! ! !
! !
21
!
! ! ! !
CLAY
! ! ! ! ! !
! ! !
! ! ! ! ! ! !
SM
!
! ! !
0 0
! ! ! !
! ! !
! !
CL
! !
SILTY SAND
! ! !
62
!
! ! !
! !
! !
!
! !
! !
57 !
!
!
!
!
SM ! ! !
ML
! ! !
SILTY SAND
!
! ! ! !
! ! !
! !
!
SANDY SILT
! ! !
!
SILT AND CLAY

! !
ML
! ! !
! ! !
! ! !
30
! ! !
! ! ! !
! !
! !
ML
! ! !
!
! ! ! !
!
! !
! !
! ! !
SP
!
! !
!
! !
SAND
!
! !
SILT AND CLAY

! !
! ! !
28
!
! ! !
! !
! !
! ! ! !!
! ! ! !
CH
! ! !
SW-SM
!
! ! ! ! ! !
! ! !
! ! !
!
! ! ! !
! ! !
!
SILT AND CLAY

! !
! ! !
! ! !
! ! !
! ! !
! ! ! ! !
!
! !
! ! ! ! ! ! !
10
!
! ! !
35
! ! ! !
! ! ! !
! ! ! !
!
! ! ! ! ! ! !
! ! !
! ! ! !
! ! ! !
! ! !
!
! ! ! ! ! !
SP-SM
!
! ! !
!
! ! ! ! ! !
! ! ! ! ! ! ! !
! ! ! ! !
! ! ! !
! !
! ! ! ! ! !
-20 -20
! ! !
! !
SM
! !
! ! ! !
10
!
SW-SM
! ! ! !
!
! ! !
SILTY SAND
! ! ! ! !
! ! ! !
44
! !! ! !
! ! ! ! !
! ! !
! ! ! !
! ! !
! ! ! ! !!
ML
!
! ! ! !
! ! ! ! ! !
! ! ! ! ! !
! ! ! !
!
! !
! ! ! ! ! ! ! !
! ! ! !
! ! ! ! ! !
! ! ! ! !
! ! !
!
! ! ! ! ! !
SM
!
SILTY SAND
! !
! ! ! !
! ! ! !
! ! !
! ! !
! ! !
! !
! ! ! !
54
! !!
! ! ! ! !
! ! !
! ! ! ! ! ! !!
! ! !
! ! !
! ! !
! !
! ! !
!
!
! ! ! ! ! ! !
! ! !
! ! ! !
! ! ! !
! ! ! ! !
! !
! ! ! ! ! !
! !
CL
! ! ! !
! ! ! !
!
! ! !
! ! ! ! ! !
! ! !
! ! ! !
! ! ! ! !
!
! ! ! ! !
! ! !!
48 SAND
! ! ! ! !
!
SM
! !
! ! ! ! ! ! !!
!
SAND AND SILTY SAND

! ! ! !
SAND
! ! ! !
! !
! ! ! !
!
! ! ! ! ! ! !
! ! !
! ! ! !
! ! ! !
! ! !
!
! ! ! ! ! !!
!
!
! ! ! !
! ! ! !
! ! !
! ! !
SAND
! ! !
!
! ! ! ! !
! !!
52
! ! ! ! !
! !
!
! ! ! ! ! ! !!
! ! ! !
! ! !
! ! ! !
! ! ! !
! ! ! !
! ! !
SW-SM
!
! ! ! !
! ! !
! !
! ! ! !
! ! ! !
! ! ! !
! !
-40 -40
! ! !
!
! ! ! ! !
! !!
! ! ! ! !
! !
!
! ! ! ! ! ! !!
8.7 ! !
! !
!
!
!
!
! !
! !
!
!
!
!
66
SAND
! ! ! ! !
! ! ! !
! ! !
SP
TD Elev. -43.03 ft
!
!
!
! !
!
! !
! ! !
GRAVEL TD Elev. -42.65 ft GRAVEL
SW-SM ! ! ! !
SM ?
! ! !
! ! !
! ! !
!
! ! ! !
! ! !
! ! !
TD Elev. -45.14 ft
! ! !
! ! !
! ! !
! !!
?
! ! ! ! !
GW
!
! !
!
! ! ! ! ! !!
K:\GIS\Projects\DWR\GEOTECHNICAL\Urban\SUTTER\Arcmaps\Sutter_PlanAndProfile_Mapbook.mxd RK 03.10.08 SAC
! ! ! !
! !
! ! ! ! !
?
! ! ! !
!
! ! ! ! ! ! !
! !
?
! ! ! ! !
12
! ! ! !
! ! !
106
! ! ! !
?
! ! !
SM
!
SM
SM 48
SILTY SAND 40
GW ?
-60 -60
! !
ML
! !
!
!
!
!
!
!
82 SILT AND CLAY

SILT AND CLAY
! !
CL
!
! !
!
! !
?
!
!
!
! !
! !
! ! ! !
! ! ! ! !
! ! !
64
! ! !
SW-SM
!
! ! !
?
! ! ! !
! ! !
! !
30
! ! !
! ! ! !
! ! ! ! ! !
! !
! ! ! !
!
! ! ! ! !
!
! !!
! ! ! ! ! !
! !
! !
! ! ! ! ! ! !! ! !
! ! ! ! ! ! !
! ! ! ! ! ! ! !
! ! ! !
ELEVATION, feet
!
! ! ! ! ! !
! ! ! ! !
! ! ! !
TD Elev. -70.10 ft TD Elev. -70.12 ft TD Elev. -70.51 ft TD Elev. -69.66 ft
20
-80 -80
10
DISTANCE, feet
0
0 100 200 300 400 500
-100 -100
2700+00 2705+00 2710+00 2715+00 2720+00 2725+00 2730+00 2735+00 2740+00 2745+00 2750+00
STATIONING, Feet
Sutter Plan & Profile Plate

in association with: Station 2700+00 to Station 2750+00 35
Levee Evaluations Office URBAN LEVEE GEOTECHNICAL EVALUATIONS PROGRAM MARCH 2008

URS LEVEE - GuidanceDocumentRev06 - Upload

Uploaded by

Copyright:

Available Formats

URS LEVEE - GuidanceDocumentRev06 - Upload

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

URS LEVEE - GuidanceDocumentRev06 - Upload

Uploaded by

Copyright:

Available Formats

DRAFT

Urban Levee Geotechnical Evaluations Program

2870 Gateway Oaks Drive, Suite 150

1.0 INTRODUCTION................................................................................................ 1-1

2.0 REACH SELECTION ......................................................................................... 2-1

2.3.1 Plan and Profile Preparation ........................................................ 2-2

3.0 SEEPAGE ANALYSIS USING SEEP/W ........................................................... 3-1

4.0 UNDERSEEPAGE ANALYSIS WITH BLANKET THEORY.............................. 4-1

20080328_GuidanceDocumentRev06.doc i Issue Date: 03-2008

Permeability kb ............................................................................. 4-5

4.4 BTA of SEEP/W Cross Sections................................................................ 4-10

5.0 STATIC SLOPE STABILITY ANALYSIS .......................................................... 5-1

20080328_GuidanceDocumentRev06.doc ii Issue Date: 03-2008

6.0 SEISMIC ANALYSIS ......................................................................................... 6-1

7.0 EROSION EVALUATION .................................................................................. 7-1

8.0 ALTERNATIVE ANALYSIS AND COST ESTIMATE ........................................ 8-1

9.0 CALCULATION PACKAGES ............................................................................ 9-1

10.0 REFERENCES................................................................................................. 10-1

Table 3-1. Permeability Values. ........................................................................................... 3-4

20080328_GuidanceDocumentRev06.doc iii Issue Date: 03-2008

Figure 1-1. Work Flow Chart. ............................................................................................... 1-4

Figure 5-1. Analysis Flow Chart. .......................................................................................... 5-2

20080328_GuidanceDocumentRev06.doc iv Issue Date: 03-2008

Return Period. ........................................................................................... 6-4

Appendix A Simplified Procedure for Shear Strength Selection of Fine-Grained Materials

Acronyms and Abbreviations

BTA blanket theory analysis

20080328_GuidanceDocumentRev06.doc v Issue Date: 03-2008

Acronyms and Abbreviations

PGA peak ground acceleration

20080328_GuidanceDocumentRev06.doc vi Issue Date: 03-2008

1.1 Program Overview

The Urban Levee Geotechnical Evaluations Program performs a logical progression of

1. Existing subsurface information

20080328_GuidanceDocumentRev06.doc 1-1 Issue Date: 03-2008

Levee evaluations are being conducted in a comprehensive manner over a 2- to 3-year

1.2 Guidance Document Development

This Guidance Document is updated on a regular basis as analyses protocols and

1.3 Purpose and Scope

In addition to presenting a standard approach for analyses, this document outlines a

20080328_GuidanceDocumentRev06.doc 1-2 Issue Date: 03-2008

Additional Guidance Document objectives include:

1.4 Analyses Process

1.5 General Limitations

Readers are encouraged to perform additional analyses/evaluations on an as-needed basis

20080328_GuidanceDocumentRev06.doc 1-3 Issue Date: 03-2008

Figure 1-1. Work Flow Chart.

20080328_GuidanceDocumentRev06.doc 1-4 Issue Date: 03-2008

Figure 1-2. Seepage and Slope Stability Analysis Process.

20080328_GuidanceDocumentRev06.doc 1-5 Issue Date: 03-2008

Figure 1-2a. Seepage and Slope Stability Analysis Process (Continued).

20080328_GuidanceDocumentRev06.doc 1-6 Issue Date: 03-2008

2.0 REACH SELECTION

2.1 Reach Selection Principles

Reaches will be selected in consideration of the following physical factors and/or

20080328_GuidanceDocumentRev06.doc 2-1 Issue Date: 03-2008

2.2 Reach Selection Approach

• Consistent conditions (either favorable or unfavorable)

2.3 Reach Selection Process