J.01_Fabens_Geotech_Report_Supplemental_20120820.pdf

SUPPLEMENTAL GEOTECHNICAL ANALYSIS FOR FINAL DESIGN

RECTIFICATION PROJECT LEVEE FROM RIVERSIDE DAM TO FABENS

FOR

U.S. SECTION OF THE INTERNATIONAL BOUNDARY AND WATER COMMISSION

(USIBWC)

Prepared for

URS GROUP, INC.

Austin, Texas

Prepared by

RABA-KISTNER CONSULTANTS, INC.

San Antonio, Texas

PROJECT NO. AEA10-028-00

June 22, 2012 andrea.glover Typewritten Text Amended August 20, 2012

Project No. AEA10-028-00

TABLE OF CONTENTS

i

INTRODUCTION

LIMITATIONS

PROJECT DESCRIPTION

FINAL LEVEE GEOMETRY AND DESIGN VALUES

DESIGN PERMEABILITY

DESIGN SHEAR STRENGTH

End of Construction Condition Steady State Condition Rapid Drawdown Condition Selection of Values

EARTHEN LEVEE SYSTEM

DESIGN CONSIDERATIONS

PROBABLE FAILURE MODES

Embankment Seepage Under Seepage Slope Erosion Slope Stability Instability of the Riverside Canal Extension and Drain Liquefaction

ITERATIVE DESIGN PROCESS/RECOMMENDATIONS

LEVEE SEEPAGE ANALYSIS

OTHER APPROACHES

LEVEE SLOPE STABILITY ANALYSIS

MINIMUM FACTOR OF SAFETY

LOADING CONDITIONS

External Loads Steady State Seepage from Design Flood Stage Sudden Drawdown from Design Flood Stage Seismic Coefficients

RESULTS OF ANALYSES

LIQUEFACTION

WASTEWAY CULVERT SYSTEM

DESIGN CONSIDERATIONS

TABLE OF CONTENTS

ii

PROBABLE FAILURE MODES

Under Seepage Sliding Stability Liquefaction

LOADING CONDITIONS

Steady State Seepage from Canal Full of Water Seismic Coefficients

ITERATIVE DESIGN PROCESS/RECOMMENDATIONS

WASTEWAY SEEPAGE ANALYSIS

WASTEWAY SLIDING STABILITY ANALYSIS

MINIMUM FACTOR OF SAFETY

RESULTS OF ANALYSES

LIQUEFACTION

CANAL CONSIDERATIONS

LEVEE TRAFFICABILITY

CONCLUSIONS

RECOMMENDATIONS

CONSTRUCTION RECOMMENDATIONS

GENERAL CONSTRUCTION CONSIDERATIONS

CLAY FILL LEVEE MATERIALS

VEGETATIVE COVER FOR CLAY LEVEES

CRUSHED ROCK BASE

ATTACHMENTS

Field Exploration ................................................................................................................ Appendix E Laboratory Testing ............................................................................................................. Appendix F Liquefaction Potential During Earthquakes....................................................................... Appendix G Levee Slope Stability and Seepage Analysis and Results ............................................... Appendix G Structure Seepage and Sliding Analysis and Results ...................................................... Appendix G

INTRODUCTION

Raba-Kistner Consultants, Inc., (R-K), has completed its review of the analysis prepared for our original Geotechnical Analysis and Engineering Evaluation Report (R-K Report No. AEA10- 028-00, dated November 19, 2010) for the levee system for the Rio Grande Flood Control project from Riverside Dam to Fabens within El Paso County, Texas based on the updated survey and levee geometry data provided by URS. On the basis of our review of the new data, it was determined that the levee geometry and survey conditions warranted additional analysis to evaluate seepage and slope stability conditions along the levee alignment. Additionally, R-K was asked to perform analysis and provide input on the existing structures. This report should not be utilized separately from the referenced, original Geotechnical Report. These services were performed in accordance with the scope of work pursuant to the executed Subcontract for Consultant Services between URS Group, Inc. (CLIENT) and R-K for the above-referenced indefinite delivery - indefinite quantity contract (IDIQ) with the United States International Boundary and Water Commission (USIBWC).

As referenced above, new slope stability and seepage analyses were performed based on the updated survey and levee geometry. Additionally, several segments of the levee are being widened from their current configuration, and two structures were analyzed that were not included in the original report. Consequently, the supplemental analyses performed for this study are presented in Appendix G which is included as an attachment to this report.

LIMITATIONS

This geotechnical addendum has been prepared in accordance with accepted Geotechnical Engineering practices in the region of West Texas for the use of URS Group, Inc. (CLIENT) and its representatives for design purposes. This addendum may not contain sufficient information for purposes of other parties or other uses.

The recommendations submitted in this supplemental report are based on the data obtained in completion of the original Geotechnical Report, and should not be utilized separately from the report. All other recommendations and limitations remain as stated in that report.

PROJECT DESCRIPTION

The segment of the Rectification Project under study consists of approximately 15.8 miles of existing levees, all of which run parallel to the Rio Grande in El Paso, Texas. The seepage and stability analyses are based on survey data provided by URS Group, Inc. Our analysis included a review of the Condition Assessment of U.S. International Boundary & Water Commission, Texas and New Mexico Levees, dated June 2004 prepared by the U.S. Army Corps of Engineers.

We have not been provided with any information regarding specific levee deficiencies in the study section, or any plans for modification. We understand the levee system has been designed with a minimum free-board of 3 feet with respect to the design flood stage. All stability and seepage calculations are based on this 3-foot assumption.

SUPPLEMENTAL BORINGS AND LABORATORY TESTS

Between September 1 and September 19, 2011, 16 supplemental borings were drilled at the locations shown on the Boring Location Map, Figure E-1. Ground surface elevations were provided by URS and are noted on the boring logs. The borings were drilled to depths ranging from 30 to 50 ft below the existing ground surface using a truck-mounted drilling rig. During drilling operations, 190 split-spoon samples (with Standard Penetration Test) were collected.

Each sample was visually classified in the laboratory by a member of our geotechnical engineering staff. The geotechnical engineering properties of the strata were evaluated by the following tests:

Type of Test Number Conducted

Natural Moisture Content 172

Atterberg Limits 42

Percent Passing a No. 200 Sieve 72

Grain Size Analysis (with Hydrometer) 5

The results of all laboratory tests are presented in graphical or numerical form on the boring logs illustrated on Figures E-2 through E-17. A key to classification terms and symbols used on the logs is presented on Figure E-18. The results of the laboratory and field testing are also tabulated on Figure E-19 for ease of reference.

Standard penetration test results are noted as “blows per ft” on the boring logs and Figure E-19, where “blows per ft” refers to the number of blows by a falling hammer required for 1 ft of penetration into the soil/weak rock.

Following completion of six of the borings, the boreholes were converted to permanent piezometers to permit long-term observation of groundwater levels. Details regarding construction of the piezometers are presented on Figures E-20 through E-25.

FINAL LEVEE GEOMETRY AND DESIGN VALUES

LEVEE GEOMETRY

A portion of the project alignment runs parallel to the Riverside Canal Extension Alignment, which is an unlined channel (except for a small segment adjacent to the wasteway structures) sharing the landside slope of the project levee. This portion is in Subreach 2. The following graphical representation is presented to communicate the general cross-section observed along the Subreach 2 of the project which extends from Station No. 933+30 to 1319+23.

The borings performed for this study were drilled through the top of the U.S. Levee. The analysis presented herein also refers to the U.S. Levee and Riverside Canal Extension Alignment.

The levee sections in Subreach 1 and 3 are adjacent to drainage structures that run along their landside toe. The following is a graphical representation of the general cross-section observed along Subreach 1 and 3 of the project. Subreach 1 extends from Station No. 805+37 to 933+30, while Subreach 3 extends from 1319+23 to 1633+50.

LEVEE WIDENING

After issuance of our original referenced report, it was determined that several segments of the levees had to be widened. In general, this widening is towards the riverside of the existing levee. The segments requiring widening are presented below:

Starting Station Ending Station Length (ft) Existing Levee Width (min., ft)

916+90 919+60 270 15.7

919+70 937+90 1820 11.1

939+85 947+80 795 13.4

950+15 953+50 335 14.9

963+45 966+40 295 14.9

967+20 968+50 130 15.2

969+85 972+40 255 15.2

972+40 976+30 390 14.2

976+60 979+40 280 15.0

1035+25 1043+05 780 13.6

1052+55 1053+90 135 15.0

1056+50 1058+60 210 14.2

1076+80 1086+80 1000 14.6

1090+30 1093+30 300 12.0

1170+15 1182+20 1205 14.1

1184+25 1192+50 825 14.3

1194+25 1197+40 315 12.1

1198+90 1203+95 505 14.2

1207+30 1216+65 935 12.6

1227+05 1228+60 155 14.3

1242+85 1251+45 860 14.5

1252+40 1257+85 545 14.8

Starting Station Ending Station Length (ft) Existing Levee Width (min., ft)

1270+70 1272+90 220 15.2

1283+65 1290+00 635 13.2

1308+30 1313+90 560 13.9

1318+10 1342+75 2465 11.8

1345+00 1348+35 335 15.2

1351+25 1352+40 115 15.2

1473+30 1485+90 1260 12.8

1491+40 1499+70 830 13.7

1512+75 1527+70 1495 13.0

1536+80 1538+75 195 15.3

1539+40 1549+50 1010 15.0

1571+65 1575+55 390 14.7

1577+90 1587+35 945 14.5

1589+05 1592+20 315 15.2

1592+90 1596+70 380 15.8

1599+35 1606+60 725 14.3

1610+80 1617+15 635 15.4

1619+55 1622+70 315 15.5

1633+70 1636+40 270 14.6

1640+85 1643+60 275 15.1

As a part of our updated analysis, R-K analyzed a widened levee section corresponding to Station 1603+62 based upon cross sections provided by URS. For this section, the widening is towards the riverside, an approximate 2.5 ft of levee crest will be replaced and benches will be cut into the existing riverside slope, the final riverside and landside slopes will be 3:1 (horizontal:vertical) and the levee crest will be 16 ft. wide.

DESIGN PERMEABILITY

Based on the USACE EM 1110-2-1901 Seepage Analysis and Control for Dams, indirect methods for determining hydraulic conductivity were employed for this study. For uniform, loose clean sands classified as SP (poorly graded sand) using the Unified Soil Classification System, the Hazen1 equation is suggested for use. We have also used the Hazen equation to estimate poorly graded sands with some silt (SP-SM) as well as the limited number of gravel samples (GP, GP- GM, GW and GW-GM.) The following graph presents the results calculated using the Hazen equation plotted against the sample fraction passing the No. 200 sieve.

1 Hazen, Allen, “Discussion of ‘Dams on Sand Foundations’ by A.C. Koenig”, Trans. American Society of Civil Engineers, Vol. 73, p. 199, 1911.

Since the silty sands (SM) at this site contained significant quantities of silt, we chose a design horizontal permeability of 10-4 centimeters per second for these soils based on the silt fraction as well as prior area experience with similar soils. Similarly, we chose a design horizontal permeability of 1·10-6 cm/sec for the lean clays. The following table summarizes the assumed horizontal permeability values used in our analyses.

Material

Horizontal Saturated Design Permeability, cm/sec

Gravels (GP, GP-GM, GW, GW-GM) 1.00E-01*

Poorly Graded Sands (SP) 1.00E-02*

Poorly Graded Silty Sands (SM) 1.00E-04*

Lean Clays (CL) 1.00E-06*

Gravel (Drainage Blanket) 1.00E+00**

Sand (Drainage Blanket) 1.00E-02**

Concrete 1.00E-08*

Cement Treated Base 1.00E-02

Lean Clay FILL (CL) 1.00E-06

Relief Well (Gravel Filter) 1.00E-01** *Vertical permeability was estimated to be 20% of horizontal permeability

**Vertical permeability was estimated to be 100% of horizontal permeability The selection of the design permeabilities was based on the discussions presented above, as well as the relative density of the materials based on SPT values and the plasticity of the materials.

The boring logs were reviewed and the classification of the various strata into one of the above

Permeability Values Calculated Using Hazen's Equation materials was based predominately on the gradation and percentage of fine grained materials encountered in our borings.

DESIGN SHEAR STRENGTH

The selection of the design shear strength values was based on several parameters including but not limited to the soil classification testing (PI and gradation), the moisture contents, the results of the SPT testing, and our experience with similar soils.

For granular soils, the following table summarizes the range in angle of internal friction values based on the results of our field and laboratory testing.

Angle of Internal Friction, degrees

Material SPT Blow

Count Range Inferred Range

Gravels (GP, GP-GM, GW, GW-GM) 14 - 50+ 35-43

Poorly Graded Sands (SP), Silty Clayey Sands and with Gravel (SC-SM), Silty Clayey Gravel and Sand (GC-GM), Poorly Graded Sand with Silt (SP-SM)

2 – 50+

30 - 43

Poorly Graded Silty Sands (SM), Sandy Silt (ML), Silty Gravel (GM)

2 - 20 30 - 35

Clayey Sand with Gravel (SC) 2 – 37 30 - 37

For cohesive soils, the following table summarizes the range of undrained shear strength values based on the results of our field and laboratory testing.

Undrained Shear Strength, psf

Material SPT Blow

Count Range Inferred Range

Lean Clays (CL) 2 - 9 250 – 1,125

Fat Clays (CH) 37 4,625

The boring logs were reviewed and the classification of the various strata into one of the above materials was based predominately on the soil classification and the SPT test results. The design shear strength parameters for this study were selected based on the resistance to driving while performing the SPT test, prior area experience, and professional judgment. The following tables summarize the design values used in our analyses.

End of Construction Condition

Total Stress

Total Unit Weight

Cohesion, psf

Angle of Internal Friction degrees

Gravels (GP, GP-GM, GW, GW-GM) 135 0 35

Poorly Graded Sands (SP) 117 0 34

Poorly Graded Silty Sands (SM) 120 0 32

Poorly Graded Silty Sands (SM) FILL 120 0 32

Lean Clays (CL) 114 300 0

Lean Clays (CL) with debris 114 300 0

Gravel (Drainage Blanket) 130 0 35

Sand (Drainage Blanket) 120 0 34

Cement Treated Base 125 7000 0

Concrete 150 1500 0

Lean Clay FILL (CL) 125 400 12

Relief Well (Gravel Filter) 120 0 32

Silt (ML) 117 200 14

Steady State Condition

Effective Stress

Weight

Cohesion, psf

Angle of Internal Friction degrees

Gravels (GP, GP-GM, GW, GW-GM) 135 0 35

Poorly Graded Sands (SP) 117 0 34

Poorly Graded Silty Sands (SM) 120 0 32

Poorly Graded Silty Sands (SM) FILL 120 0 32

Lean Clays (CL) with debris 114 0 24

Lean Clays (CL) 114 200 24

Gravel (Drainage Blanket) 130 0 35

Sand (Drainage Blanket) 120 0 34

Cement Treated Base 125 7000 0

Effective Stress

Weight

Cohesion, psf

Angle of Internal Friction degrees

Concrete 150 1500 0

Lean Clay FILL (CL) 125 200 24

Relief Well (Gravel Filter) 120 0 32

Silt (ML) 117 100 28

Rapid Drawdown Condition

For rapid drawdown analyses, both total stress (End-of-Construction) and effective stress (Steady State) parameters were used to calculate the resistance to sliding for each slice within the analysis, with the weaker of the total stress analysis and effective stress parameters used to calculate the sliding resistance.

Selection of Values

A review of the above indicates that for our analyses, the lower range of strength values for a given soil type was typically selected based on the above parameters in lieu of average values based on statistical analyses. The use of the lower range of strength values for design purposes was considered appropriate as levees must be designed for local, not average, soil conditions.

The exception was the lower range of blow counts recorded for some of the granular strata.

A number of SPT blow counts at depth for this material were less than 5, suggesting these materials are weaker than we have assumed for design. However, published correlations2 between SPT and relative density would imply that for an angle of internal friction of 32 degrees, the soil’s relative density would be less than 30 percent. Based on the observed condition of the overlying soils during field exploration and sampling, we do not believe that in situ relative densities lower than 30 percent are likely, and consider the minimum likely angle of internal friction in granular soils to be 32 degrees at this site.

EARTHEN LEVEE SYSTEM

DESIGN CONSIDERATIONS

Subreach No.’s 1, 2 and 3 will consist of an earthen levee system. The existing levees should meet three important criteria: they should be resistant to the forces of erosion, should exhibit a slope stability design allowable factor of safety of at least 1.3 for end of construction, 1.4 for steady state and 1.0 for rapid drawdown, and the calculated exit gradients should not be greater than 0.5 and have a factor of safety of at least 1.2 against heave at the toe where a “blanket” condition exists. The levee structure must meet these criteria so that the calculated risk of failure is consistent with criteria established by the U.S. Army Corps of Engineers and USIBWC design guidelines.

2 “Penetration Tests and Bearing Capacity of Cohesionless Soils”, Meyerhof, G.G., JSMFD, ASCE, Vol. 82, SM 1,

pp. 1-19.

PROBABLE FAILURE MODES

Our review of the site and expected conditions indicates that the following major modes of failure could affect the levee system during a flood event:

Embankment (“Through”) Seepage Under seepage Slope erosion Slope stability Instability of the Riverside Canal Extension alignment Liquefaction induced settlements and lateral deformations

The following sections address each of these failure modes. Please note that while these failure modes have been separated for ease of discussion, it is possible that two or more of the failure modes may occur simultaneously.

Embankment Seepage

Embankment seepage consists of water flowing through the embankment as a result of the rise in river level during the design flood event. The principal failure mechanism related to embankment seepage occurs when the seepage gradients in the soils are sufficient to transport fine soil particles through the soil mass and create a “pipe.” As some of the existing levee embankment materials commonly consist of granular materials, embankment seepage was evaluated as a risk for the current constructed conditions. However, based on the results of our analyses, the phreatic surface never exits above the levee toe indicating embankment seepage is not a significant risk factor for design flood conditions.

Under Seepage

We generally consider under seepage to be the major risk for the existing and proposed levee embankments as a result of the location of the canal and drains. Under seepage consists of water flowing beneath the embankment as a result of a rise in river level during the design flood event. The principal failure mechanisms related to under seepage occur when: 1) the seepage force equals or exceeds the buoyant weight of the soil in the toe area; and/or 2) when the seepage gradients in the soils are sufficient to transport fine particles through the foundation soils and create a pipe. The first process, often referred to as a “quick condition”, represents a potentially serious condition for overall levee stability. The second process is referred to as piping and is also potentially dangerous to the overall stability of the levee system. When the upward force of water equals or exceeds the buoyant weight of the soil, the soil particles can become dislodged and carried away by the water. The soil particles are lost where the flow exits the ground surface; this “loss of ground” can migrate toward the river as more soil is lost, forming a “pipe” or open channel. If a pipe becomes large enough, and collapses during the flood event, the levee can be breached in a sudden, catastrophic event. As discussed in a subsequent section, all drains and canals should be maintained full of water during a flood event to reduce the seepage exit gradients in the soils beneath the canal and reduce the risk of piping beneath and adjacent to the canal.

Slope Erosion

The existing embankments and the underlying soils are generally composed of silty sands, fine to medium sands, clayey silts and lean clays. It appears that the existing embankments were constructed using the soils available within the meander band of the Rio Grande. These materials

– particularly the sandy silts and silts – are generally considered to be poor materials to use when constructing levees, dams and slopes. Our field exploration activities also indicate that minimal effort was made to compact or treat the supporting soils before constructing the levees in certain reaches. As a result, the soils directly below the levees are frequently in a loose condition in some areas.

The existing slopes are considered to be highly susceptible to erosion by water and, to a lesser extent, wind. Common methods to improve resistance to erosion include plating the top few feet of the entire embankment with clay, treating the upper two feet of embankment with a blend of Portland cement and fly ash, and/or armoring the embankment surface with a layer of crushed rock or concrete riprap. Although plating of the levee is not recommended at this time, R-K recommends inspection of the riverside face of the levee following flooding events to determine if isolated remedial activities should be performed.

Slope Stability

Based on our review of available data and our visual observations during drilling, the existing levee embankments along the Rectification Project alignment appeared to exhibit slopes ranging from about 1.6:1 to 3.3:1 (H:V) on the landside and 3:1 (H:V) or flatter on the riverside. It is our understanding that a slightly steeper riverside slope exists in two isolated areas, but that in all cases, the riverside slope is 2.5:1 (H:V) or flatter. Due to the relatively high factors of safety for the riverside slopes in all analyses performed and our understanding that these may be flattened to meet the 3:1 (H:V) criteria, independent analyses of these isolated slopes was not considered necessary.

In general, slopes 3:1 (H:V) or flatter on the riverside would be expected to exhibit the required factors of safety for a normal (flood) seepage conditions. In general slopes 2.5:1 (H:V) or flatter on the landside would be expected to exhibit the required factors of safety for normal flood seepage conditions. As some of the landside slopes were steeper, a detailed slope stability analyses was performed for critical sections, as discussed in a subsequent section of this report.

As discussed below, slope stability of the slopes in the canal and drain are also considered a risk which would impact the stability of the levee due to their proximity to the levee toe.

Instability of the Riverside Canal Extension and Drain

As referenced in a previous section of this report, the Canal alignment is located immediately adjacent to and parallel to the toe of the levee alignment, in Subreach 2 and a drain is located immediately adjacent to the canal in Subreaches 1 and 3. For this condition, the results of our analyses indicate that under seepage will result in pressures that will destabilize the bottom of the canal and erode the soils near the landside of the levee which could result in significant levee failures. Methods to address this and other under seepage issues are addressed in the LEVEE SEEPAGE ANALYSIS section of this report and include filling the canal and drains to capacity.

Additionally, due the location of the canal and the side slopes of approximately 1.5:1 (H:V) in some areas, slope stability failures within the canal/drain slopes are a significant concern during flood conditions, which is also addressed in the LEVEE SLOPE STABILITY section of this report.

Liquefaction

Soil liquefaction results from loss of strength during cyclic loading, such as imposed by earthquakes. Soils most susceptible to liquefaction are clean, loose, saturated, uniformly graded, and fine-grained sands. Empirical evidence indicates that loose silty sands are also potentially liquefiable. When seismic ground shaking occurs, the soil is subjected to cyclic shear stresses that can cause excess hydrostatic pressures to develop. If excess hydrostatic pressures reach the effective confining stress from the overlying soil, the sand may undergo deformations. If the sand undergoes virtually unlimited deformation without developing significant resistance, it is said to have liquefied, and if the sand consolidates or vents to the surface during and following liquefaction, ground settlement may occur.

When groundwater is present, the entire length of levee has a significant potential for liquefaction during earthquake events, particularly when the earthquake magnitude exceeds 6.5.

Following a discussion with the USIBWC staff regarding the risk of damage to the two wasteway structures as a result of liquefaction and the measures that would be required to limit the damage, the USIBWC indicated that mitigation measures would likely be more expensive than structure replacement. As a result, this report does not contain recommendations for measures to limit liquefaction-induced damage to the levees or to the two wasteway structures.

CRITICAL SECTIONS

R-K worked with URS Group, Inc. to analyze the existing cross sections to determine the critical sections for analyses. On the basis of that review and multiple iterations, the following sections were selected for complete analysis.

Subreach Number 1: Station 863+00 Subreach Number 2: Station 1089+55 Subreach Number 3: Station 1529+00 Subreach Number 3: Station 1607+00

The selection of the critical sections was based on multiple factors including levee height, levee slopes, levee width, subsurface soil conditions, head differential and the proximity to and slopes of the adjacent canal and drains.

ITERATIVE DESIGN PROCESS/RECOMMENDATIONS

It should be noted that the analyses of the levee system, as well as potential remedial activities, is an iterative design process. Furthermore, remedial activities performed to correct one deficiency will impact the results of the remaining analyses for the various failure modes. As referenced above, several failure modes were identified for the levee portions within this reach.

During the design process for the various levee subreaches in this system, a final recommended design solution was determined, which consist of no recommended changes to the existing levee geometry but maintaining all drains and canals at full capacity throughout a flood event.

Additionally, our recommendations include flattening the slopes of the drain to 2.5:1 (H:V) or flatter from Station No. 1565+00 to 1655+00.

The analyses based on these improvements, as well as discussions on other options considered, is provided in the following sections of this report.

LEVEE SEEPAGE ANALYSIS

We performed a steady-state seepage analysis of the slope geometries for the critical sections using SEEP/W, a commercially available engineering program produced by Geo-Slope International which utilizes finite element analyses to evaluate ground water seepage. Our analyses included the surveyed slope geometry for each of the critical sections. The lateral boundaries of the model on the landside were located at a distance from the levee profile following the guidelines established in USACE DIVR 1110-1-4003. Potential seepage face conditions were assumed at the landside ground elevation. The bottom boundary of our model was assumed to range from approximately 50 to 100 ft depending upon the subsurface profile within the given subreaches. No flow conditions were assumed at the bottom boundary. The analyses were performed assuming that the hydraulic properties of the soils exhibited a horizontal permeability that was five times greater than the vertical permeability.

For the case of riverside and/or landside blanket, blanket permeabilities were selected based on material type, blanket thickness and general guidance provided in the USACE DIVR 1110-1-400 but modified for conditions in the Rio Grande Valley based on local experience. Assumptions made for these seepage analyses included extending the “blanket” materials discovered in the levee crest borings hundreds of feet towards both the riverside and landside.

The full hydrostatic pressure associated with the design 100 year storm was assumed to act on the boundary on the flood side model boundary, typically corresponding to the centerline of the river. Our analyses were performed assuming that the far field landside groundwater table corresponds with the ground surface elevation which is the typical design assumption recommended by the USACE.

Conditions exceeding the maximum allowable seepage exit gradient of 0.5 were observed for the critical sections located Subreaches 1, 2 and 3, as presented in the revised Appendix G. When evaluating the seepage conditions based on filling the canal with water, in order to efficiently determine where the flow changes direction and determine the location of the phreatic surface on the canal face, nodal exit gradients were selected and are presented in Attachment C. For Subreaches 1, 2 and 3, the critical exit gradient was generally encountered at the bottom of the canal/drain adjacent to the levee. These analyses indicate the need to lengthen the seepage flow paths, provide pressure relief, intercept the seepage with one or more drainage layers and/or fill the drains and canal with water for Subreaches 1, 2 and 3.

3 USACE DIVR 1110-1-400, “Landside Seepage Berms for Mississippi River”, December 12, 1998.

Maintaining the drains and canal at capacity during a flooding event was selected as the preferable alternative for Subreaches 1, 2 and 3. For operational purposes, emergency operations including filling the drains should begin when the Rio Grande reaches 60 percent of its flow capacity.

OTHER APPROACHES

Other seepage cutoffs could be used to improve the overall stability of the levees, such as cement or chemical grout curtains and soil-cement jet-grouted walls. In general, we consider these options to be less desirable due to their probable costs and relatively high risk of failure

LEVEE SLOPE STABILITY ANALYSIS

We performed a steady-state slope stability analysis of the critical slope geometries using SLOPE/W, a commercially available engineering program produced by Geo-Slope International which evaluates slope stability. Our analyses were based on the Spencer method, and included existing ground surface geometries.

MINIMUM FACTOR OF SAFETY

For a given slope configuration, the forces that “drive” slope failure (including gravity and external loading conditions) are compared to the slope’s resistance to failure, which is a function of dewatering controls and internal shear strength (cohesion and internal angle of friction) of both the foundation soils and the fill soils utilized for construction of the levees.

For levee projects, the USACE has specified minimum safety factors against slope failure with respect to loading conditions. The minimum acceptable factors of safety for levees, provided in Table 6-1b on Page 6-5 of EM 1110-2-1913, are listed in the following table.

Condition Required Factor of Safety

End of Construction 1.3

Sudden (“Rapid”) Drawdown 1.0 to 1.2*

Long Term (Steady Seepage) 1.4

* Sudden drawdown analysis. FS = 1.0 applies to pool levels prior to drawdown for conditions where these water levels are unlikely to persist for lengthy periods preceding drawdown. FS = 1.2 applies to pool level, likely to persist for long periods prior to drawdown. For this project, 1.0 is considered applicable.

We consider a significant slope failure to involve a volume of slope material that is large enough to substantially impair the serviceability or operation of the levee or that could imperil human life.

Shallow, sloughing slope failures that involve a relatively small volume of material or that can be repaired locally without substantially impacting the levee operations are considered to be minor slope failures and do not control the conclusions of our stability analyses.

LOADING CONDITIONS

For satisfactory performance, an earth embankment should have an acceptable factor of safety during construction and throughout its projected service lifetime. Stability analyses should include variations in stress conditions brought on by construction practices and sequencing, external loadings, and any anticipated changes in hydraulic conditions. The following paragraphs discuss each stability condition analyzed in our study.

External Loads

External loads for the roadways along the levee crest have also been modeled. A loading of an equivalent uniform vertical surcharge of 300 psf was applied to the crest of the levee for all analyses.

Steady State Seepage from Design Flood Stage

The long term, steady-state seepage loading condition was also analyzed. This loading condition models the design flood stage for such a duration that the levee soils are fully saturated and a condition of steady state seepage occurs through the embankment. For this loading condition, effective stress soil parameters were used in the analysis.

Sudden Drawdown from Design Flood Stage

This condition represents the situation when the flood water recedes at such a rapid rate that the saturated levee soils do not have time to drain. Sudden drawdown was considered for the drains and canal from a completely full to a completely empty state and for the riverside slope of the canal when flood water is expected to act on the slope face (Subreach 2 and 3). In the case of the widened section, sudden drawdown analyses were only performed for the riverside slope of the levee.

End of Construction

This condition represents the situation immediately following the widening of the levee sections and is considered for failures towards the landside and riverside. This analysis was only performed for the widened section at Station 1603+62 or where the canal slopes were flattened.

Seismic Coefficients

The Rectification segment lies within an active seismic zone. As such, a liquefaction analysis has been performed and is presented later in this Geotechnical Memorandum. In addition, the anticipated deformation of an earth embankment due to earthquake is also presented.

RESULTS OF ANALYSES

The results of the analyses indicate that the riverside slopes of the levee are not of primary concern. However, the landside slopes are of concern and do not meet minimum factors of safety without the canals and drains being maintained full of water during a flood event for Subreaches 1, 2 and 3.

The following table contains a summary of the results from our slope stability analyses for loading conditions and slope configuration discussed above. The values presented below assume that the canal and drains are full. For the steady state condition at the flood design stage for Critical Sections 863+00 and 1607+00 where the drain is located some distance away from the landside toe, a slope stability analyses for the drain slope was also performed and is reported below. For Sections 1089+55 and 1529+00 where the canal/drain slope is located at the toe of the levee slope, a single analysis was performed that incorporated both slopes. A graphical presentation of the most critical failure surfaces from our SLOPE/W analyses for all three subreaches can be found at the end of this Geotechnical Memorandum in Appendix G.

Computed Factors of Safety

Subreach 1 – Station 863+00

Steady State at Flood Stage – Riverside

Steady State at Flood Stage – Landside

3.1 2.3

Subreach 2 – Station 1089+55

Steady State at Flood Stage – Riverside

Steady State at Flood Stage – Landside

3.5 1.9

Subreach 3 – Station 1529+00

Steady State at Flood Stage – Riverside

Steady State at Flood Stage – Landside

Rapid Drawdown Following Flood Stage – Riverside

2.4 1.4 2.2

Subreach 4 – Station 1607+00

Steady State at Flood Stage – Riverside

Steady State at Flood Stage – Landside

Rapid Drawdown Following Flood Stage – Riverside

2.9 1.5 2.1

As indicated in the footnote of the table for Section 1607+00, the values reported for the drain slopes are based on the requirement to flatten the slope in the drain to 2.5:1 (H:V). For the existing slope conditions in the drain, the calculated factor of safety for steady state and sudden drawdown conditions were 0.7 and 0.6, respectively. These analyses are also presented in Appendix G. Consequently, to achieve the required factor of safety for the drain slopes, the slopes should be flattened to 2.5:1 (H:V) from Station 1565+00 to 1655+00. However, it should be noted that due to the distance of the drain from the levee toe, failure of the drain slopes is not expected to impact the stability of the levee.

Based on the results of our analyses, the factors of safety computed for riverside and landside slopes meet the USACE requirements for minimum safety factors immediately after widening of the levee sections and for the assumed flood event based the drains and being maintained full of water during a flood event.

LIQUEFACTION

Due to the presence of saturated, very loose to loose cohesionless soils near the ground surface (within the top 10 ft), we evaluated the potential for liquefaction.4 This procedure examines the relationship between the stress reduction factor, earthquake magnitude scaling factor for cyclic stress ration, overburden correction factor for cyclic stress ratios, and the overburden normalization factor for penetration resistances (SPT N-values).

The procedure relies on both field (SPT N-values) and laboratory data (percent passing a No.

200 sieve). Often, due to a lack of information, it is acceptable to extrapolate N-values and passing 200 values from a limited number of samples to represent a larger stratum or strata of soil. In this case, the field and testing program was directed with this type of analysis in mind.

Therefore, a large number of samples having both N-values and passing 200 data were available for analysis.

The results of the Idriss and Boulanger’s semi-empirical procedure are presented on Figures 1 through Figure 3, of Appendix G. It should be noted that the curves presented on these figures represent the cyclic stress ratio required to cause liquefaction. That is, data plotting above the curve has a significant potential for liquefaction during an earthquake, while data plotting below the curve does not.

In addition to the available field and laboratory data discussed above, our analysis was based on published ground accelerations available in Section 1615 Earthquake Loads – Site Ground Motion of the 2006 International Building Code and were presented in earlier the Seismic Coefficients section of this Geotechnical Memorandum. Our analyses are based on a Site Class B designation. With this site class designation, a significant portion of the soils would be expected to experience liquefaction.

The 2006 International Building Code recommends that the seismic coefficients be increased based on the actual site classification. In our opinion, these site classifications were intended for the design of buildings, not low-lying earthen structures. Therefore, based on our

4 Idriss, I.M. and Boulanger, R.W., “Semi-Empirical Procedures for Evaluating Liquefaction Potential During Earthquakes”, Proceedings 11th ICSDEE, pp. 32-56, Berkeley, CA, 2004.

experience, the entire length of the Rectification Project has been designated Site Class B (Fa and FV multipliers of 1.0).

Based on these results, it is our opinion that the entire length of levee has a significant potential for liquefaction during earthquake events, as discussed below. USACE design guidelines do not require the design of earthen levees to include a liquefaction analysis, but do require that the risk of liquefaction be considered for structures such as wasteways and floodwalls.

SETTLEMENTS

A preliminary settlement analysis was performed for our original study, but references that a more detailed settlement analysis should be performed once it had been determined what additional fill materials would be placed to raise the levee height. However, it is our understanding that no new embankment materials will be placed resulting in heightening of the levee. As the existing levees are in excess of 50 years old and no additional height or weight will be added, new theoretical settlements would be zero. Per FEMA requirements, a calculation reflecting this condition is presented in Appendix G.

For the widened levee segment, a settlement analyses was performed for Station 1603+62.

Based on our experience with soils similar to those encountered in the underlying native soils, the SPT blow counts recorded in the field, settlements on the order of 3-3/4 in. are estimated, with approximately ¼ in. of the settlements being elastic settlements occurring at the time of construction.

It is our understanding that the levee height will be constructed with a free board of 3 ft., and that a six inch overbuild will be included to account for potential settlements. Consequently, based on our analyses, settlements are not considered to be a major risk for the levees.

WASTEWAY CULVERT SYSTEM

DESIGN CONSIDERATIONS

Two existing wasteway structures on the project are used to divert flows from the canal to the Rio Grande River. Each of these will be replaced with new structures. The design case for these two structures is full hydrostatic head in the canal and normal water level in the Rio Grande River.

Both steady state seepage and stability calculations were performed.

It should be noted that due to the high permeability of the underlying soils and the relatively shallow nature of the groundwater, the use of dewatering systems should be anticipated during construction. Given the high permeability of the soils, high capacity dewatering systems may be required.

DESIGN VALUES

R-K recommends the following design values be utilized for the design of the culvert system.

Design Parameter

Recommended Design Value

Active Earth Pressure Coefficient 0.42

Passive Earth Pressure Coefficient 1.19

At Rest Earth Pressure Coefficient 0.6

Coefficient of Sliding 0.62

Unit Weight of Soil 125 pcf

Modulus of Subgrade Reaction WW1 14 pci

Modulus of Subgrade Reaction WW2 17 pci

Allowable Net Bearing Pressure 4,000 psf

The above design values for backfill materials assume that clay will be utilized for all backfill materials.

PROBABLE FAILURE MODES

Our review of the site and expected conditions indicates that the following major modes of failure could affect the wasteway culvert structures:

Under seepage Sliding stability Liquefaction induced settlement and lateral deformation

The following sections address each of these failure modes. Please note that while these failure modes have been separated for ease of discussion, it is possible that two or more of the failure modes may occur simultaneously.

Under Seepage

We generally consider under seepage to be the major risk for the proposed wasteway structures.

Under seepage consists of water flowing beneath the structures as a result of water in the canal.

The principal failure mechanisms related to under seepage occur when: 1) the seepage force equals or exceeds the buoyant weight of the soil in the outfall area; and/or 2) when the seepage gradients in the soils are sufficient to transport fine particles through the foundation soils and create a pipe. The first process, often referred to as a “quick condition”, represents a potentially serious condition for stability of the wasteway structure. The second process is referred to as piping and is also potentially dangerous to the overall stability of the structure. When the force of water exceeds the effective stresses in the soil, the soil particles can become dislodged and carried away by the water. The soil particles are lost where the flow exits the ground surface; this “loss of ground” can migrate toward the outfall area where soil is lost, forming a “pipe” or open

August 20, 2012

LIQUEFACTION

Due to the presence of saturated, very loose to loose cohesionless soils near the ground surface (within the top 10 ft), we evaluated the potential for liquefaction.5 This procedure examines the relationship between the stress reduction factor, earthquake magnitude scaling factor for cyclic stress ration, overburden correction factor for cyclic stress ratios, and the overburden normalization factor for penetration resistances (SPT N-values).

The procedure relies on both field (SPT N-values) and laboratory data (percent passing a No.

200 sieve). Often, due to a lack of information, it is acceptable to extrapolate N-values and passing 200 values from a limited number of samples to represent a larger stratum or strata of soil. In this case, the field and testing program was directed with this type of analysis in mind.

Therefore, a large number of samples having both N-values and passing 200 data were available for analysis.

The results of the Idriss and Boulanger’s semi-empirical procedure are presented on Figures 1 through Figure 3, of Appendix G. It should be noted that the curves presented on these figures represent the cyclic stress ratio required to cause liquefaction. That is, data plotting above the curve has a significant potential for liquefaction during an earthquake, while data plotting below the curve does not.

In addition to the available field and laboratory data discussed above, our analysis was based on published ground accelerations available in Section 1615 Earthquake Loads – Site Ground Motion of the 2006 International Building Code and were presented in earlier the Seismic Coefficients section of this Geotechnical Memorandum. Our analyses are based on a Site Class B designation. With this site class designation, a significant portion of the soils would be expected to experience liquefaction.

The 2006 International Building Code recommends that the seismic coefficients be increased based on the actual site classification. In our opinion, these site classifications were intended for the design of buildings, not low-lying earthen structures. Therefore, based on our experience, the entire length of the Rectification Project has been designated Site Class B (Fa and FV multipliers of 1.0).

Based on these results, it is our opinion that the entire length of levee has a significant potential for liquefaction during earthquake events, as discussed below.

CANAL CONSIDERATIONS

Based on the results of the steady state seepage analyses, we recommend that the canal and the drains be filled with water to capacity (estimated to be 10 ft deep). This level must be achieved within 6 hours of the beginning of a storm event or the existing exposed slopes of the canal and the levee toe may be subjected to significant damage. If filling the drains and canal to a depth of 10 ft is not possible, R-K should be contacted immediately to review and revise these recommendations.

5 Idriss, I.M. and Boulanger, R.W., “Semi-Empirical Procedures for Evaluating Liquefaction Potential During Earthquakes”, Proceedings 11th ICSDEE, pp. 32-56, Berkeley, CA, 2004.

The operation of the canal system could have a deleterious impact on the performance of the levee. We recommend that the USIBWC modify the operation agreement with the canal operator to ensure that any drawdown in canal level be accomplished in a relatively slow manner, with the water level within the canal drawn down at a steady rate over a period of one week or more. Inclusion of this requirement in the operation of the canal will reduce the likelihood that a rapid drawdown failure could occur in the canal slopes, which could adversely impact the levee itself. We understand that inclusion of this requirement into the canal operator’s operations and maintenance agreement, and as a result the performance of a rapid drawdown analysis was not required for this study.

LEVEE TRAFFICABILITY

Based on the information provided, the top of the levee will generally be utilized for light traffic to facilitate access, but may be subject to infrequent heavy equipment loads during flood emergencies and maintenance. Consequently, traffic loading was taken into account during the performance of our slope stability analyses by applying a 300 psf surcharge load on the levee crown. The existing subgrade soils at the top of the levee are sufficient to hold light vehicular traffic, although they may be subject to pumping under a combination of inundated conditions and heavy loads.

Prior to placing the gravel surfacing along the crest, the subgrade should be scarified and compacted to 95% of the maximum density as determined by ASTM D1557 at a moisture content ranging between 3 percent below optimum moisture content to optimum moisture content. We recommend that the gravel layer consist of a minimum of 6 inches of aggregate base selected in accordance with Texas Department of Transportation, 2004 Standard Specifications for Construction and Maintenance of Highways, Streets, and Bridges” for Item 247 Flexible Base, Type A or C, Grade 3.

CONCLUSIONS

The existing levee embankments were constructed using soils that are highly susceptible to erosion and through seepage, over subgrade soils that are susceptible to under seepage.

Seepage through and/or beneath the existing embankments represents a serious risk to the levee system during a design flood event. Additionally, the Riverside Canal Extension alignment is located immediately at the landside toe of the levee, and runs parallel to the levee over Subreach 2, while a drain is located adjacent to the levee over a majority of Subreaches 1 and 3. The Riverside Canal Extension alignment and drains in each reach have relatively steep side slopes of 1.5:1 (horizontal:vertical). The results of our seepage analyses for steady state flow indicate that a significant risk of damage and failure of the levee and side slopes of the adjacent Riverside Canal Extension alignment and drains if the canal and drains are dry. The slope stability analyses indicated similar results.

As a result of the analyses, that the Riverside Canal Extension alignment and drains should be maintained full of water during flood events.