8550-2020-06_Heart_Butte_Dam_Lab_Testing_Report_0002.pdf

United States Department of the Interior

BUREAU OF RECLAMATION

PO Box 25007

Denver, CO 80225-0007

86-68550 2.2.4.23

April 9th, 2020

MEMORANDUM

To: Geotechnical Engineering, Group 2 Geotechnical Services Division, Technical Service Center, Denver, CO Attention: 86-68312 (Jie Yu, P.E., Civil Engineer)

From: Gergo Arany, Civil Engineer (Geotechnical) Geotechnical Laboratory and Field Support (86-68550), Technical Service Center, Denver, CO

Subject: Transmittal of TM No. 8550-2020-06, “Geotechnical Laboratory Testing Results for Heart

Butte Dam Conduit Floor, Foundation, Borrow, and Rock Material,” Bureau of Reclamation, Missouri Basin Region, ND

Attached is the Technical Memorandum (TM) documenting the geotechnical laboratory testing program to evaluate samples obtained from beneath the conduit floor, embankment and foundation materials, and borrow materials obtained during field investigations at Heart Butte Dam that were submitted by Bureau of Reclamation’s Missouri Basin (MB) Region personnel.

If you have any questions or comments regarding this report, please contact Gergo Arany at 303-445-2389 or via email at garany@usbr.gov.

Attachment:

Electronic copy CC recipients (w/att):

kgwalker@usbr.gov – Kent Walker, Dam Safety Program Manager (84-44000) jyu@usbr.gov – Jie Yu, TSC Civil Engineer (86-68312) jwhall@usbr.gov – Justin Hall, TSC Supervisory Civil Engineer (86-68312) rlung@usbr.gov – Robert Lung, TSC Geologist (86-68320) dscanson@usbr.gov – David Scanson, MB Region Civil Engineer (GP-2400) cclark@usbr.gov – Cody Clark, MB Regional Geologist (MB-2300) rrinehart@usbr.gov – Robert Rinehart, TSC Supervisory Civil Engineer (86-68550) elindenbach@usbr.gov – Evan Lindenbach, TSC Civil Engineer (86-68550) barmstrong@usbr.gov – Blake Armstrong, TSC Civil Engineer (86-68550)

U.S. Department of the Interior Bureau of Reclamation Technical Service Center Denver, CO April 2020

Geotechnical Laboratory Testing Results for Heart Butte Dam Conduit Floor, Foundation, Borrow, and Rock Material Pick-Sloan Missouri Basin Program, Heart Butte Dam

Interior Region 5: Missouri Basin Region, North Dakota

Geotechnical Laboratory and Field Support, 86-68550, Technical Memorandum No. 8550-2020-06 ii

Mission Statements The U.S. Department of the Interior protects America’s natural resources and heritage, honors our cultures and tribal communities, and supplies the energy to power our future.

The mission of the Bureau of Reclamation is to manage, develop, and protect water and related resources in an environmentally and economically sound manner in the interest of the American public.

iii

Geotechnical Laboratory and Field Support, 86-68550

Technical Memorandum No. 8550-2020-06

Geotechnical Laboratory Testing Results for Heart Butte Dam Conduit Floor, Foundation, Borrow, and Rock Material

Prepared by: Gergo Arany, M.Sc.

Civil Engineer, Geotechnical Laboratory and Field Support, 86-68550

Checked by: Blake Armstrong, M.Sc.

Technical Approval: Robert Rinehart, Ph.D., P.E.

Manager, Geotechnical Laboratory and Field Support, 86-68550

Peer Review: Evan Lindenbach, P.G., P.E.

iv

Contents Introduction Conclusions

Conduit Floor Downstream Embankment Materials Downstream Alluvium Materials Sandstone Foundation Materials Borrow Soils Bedrock

Laboratory Testing Methods & Results Physical Properties Testing

Conduit Floor Materials Emb Materials Qal Materials Sandstone Materials Test Pit Materials

Jet Erosion Testing (JET) Horizontal Flexible Wall Permeability Testing Direct Shear Testing Petrographic Examination

Conduit Floor Materials Sandstone Materials Shale Materials

Rock Uniaxial Compressive Strength Testing Rock-type UCS Soil-type UCS

References Physical Property Test Reports ......................................................................A-1 Test and Microscope Photos ...........................................................................B-1 Petrography Reports .......................................................................................C-1 Rock Uniaxial Compressive Strength Test Reports .......................................D-1

Figures Figure 1: Spillway Conduit Schematic and Drill Hole Locations within Conduit (Wagner, 2018) Figure 2: Test Pit Locations (Wagner, 2018) Figure 3: Drill Hole Locations (Wagner, 2018) Figure 4: JET Erodibility Chart. The second specimen from 10U-291 (2), DH-18-2, from approximate elevations of 1991.6 to 1989.1 feet ceased to erode during testing and was promptly ended. The arrow drawn through points 10U-291 (1) and 10U-292 (2) indicate the likely behavior of the material had testing been performed to a natural ending. Testing and report was performed in accordance with guidelines outlined in Hanson and Simon (2001), Hanson and Cook (2004), Hanson and Hunt (2006), and Wahl (2014) v

Figure 5: Static Direct Shear Test Results

Tables Table 1. Test Request and Lab Indices for Heart Butte Soil Samples Table 2: Test Request and Lab Indices for Heart Butte Soil Samples (cont.)

Table 3: Test Request and Lab Indices for Heart Butte Rock Samples Table 4: Summary of Physical Properties Test Results Table 5: Summary of Physical Properties Test Results (cont.)

Table 6: Summary of Jet Erosion Test Results Table 7: Summary of Horizontal Flexible Wall Permeability Test Results Table 8: GLFS XRD Analysis Results Table 9: XRD Analysis Results Table 10: Rock Uniaxial Compressive Strength and Soil-Type Physical Properties Results

Introduction Heart Butte Dam is a 142-ft high earthen embankment that is 1,850 ft long at the crest and is located approximately 73 miles west of Bismarck, ND. The dam impounds Lake Tschida, with a capacity of about 221,800 acre-feet at the maximum reservoir water surface. Geology at the dam and abutments is predominantly comprised of fine-grained, light gray sandstone of the Ft. Union Formation, containing hard and concretionary strata near the abutment crests and a series of interbedded shales in the exposed section beneath the dam.

Due to risk of internal erosion failure associated with cracking of the outlet works conduit and seepage into the structure, the Dam Safety Office has recommended the completion of final designs of the preferred alternative to the outlet works conduit based on a 2016 Corrective Action Study (CAS). This lab testing program was performed in support of the final design and had multiple purposes, including to characterize: 1) in-place materials from beneath the existing conduit floor, 2) embankment and foundation materials from the periphery of the existing outlet works conduit, and 3) material from an on-site borrow area to potentially be used in modifications of the existing conduit and embankment features. The testing program was developed in cooperation between the Bureau of Reclamation’s Technical Service Center’s (TSC) Geotechnical Laboratory and Field Support group (GLFS) and Geotechnical Services Division personnel. The sample inventory and test request summary is shown in Tables 1 through 3.

Samples from nine drill holes and three on-site test pit sources were delivered to the GLFS group in 2018; drill hole and test pit locations are shown in Figures 1 through 3. Each sample was assigned a GLFS index number as shown in Tables 1 through 3. Material was tested to determine physical properties, permeability, compressive strength, shear strength, erosion potential, and minerology. Laboratory tests reported were performed by GLFS and private sector contract labs selected by GLFS (Blue Rock Labs, Denver, CO; Golder and Associates, Denver, CO; and CAMET Research, Inc, Goleta, CA). All tests were performed in general accordance with ASTM and USBR standards as noted in the text.

Conclusions Samples received by the GLFS group were collected from the downstream side of the dam, in an effort to identify any risks of internal erosion failure associated with cracking of the conduit works and any subsequent seepage into this structure. The samples consisted of embankment (Emb) and foundation materials, material sampled from beneath the outlet works conduit floor, and from test pits located in an on-site test borrow area. Foundation materials from below the Emb can further be categorized as quaternary alluvium (Qal) and sandstone, as outlined in final and draft drill hole logs provided by TSC and Great Plains (GP) region geologists. For identification of subsurface trends, elevation – not depth – were used in Emb, Qal, and sandstone summaries; approximate elevations were obtained from Google Earth using coordinates of proposed drill hole locations obtained from the 2019 Field Exploration Request (FER).

Conduit Floor

Samples obtained from below the conduit floor of the outlet works consisted primarily of non-plastic silty sands. Small to trace amounts of gravel were encountered, not exceeding more than 4% gravel content in any sample from this area. Neither of the collected samples from these locations exhibited plastic behavior. Testing for carbonate presence, loose sands in all samples exhibited no reaction to hydrochloric acid (HCl). Where present, gravels (usually intact sandstone fragments) generally exhibited a strong reaction to HCl. In-situ dry unit weights from tube densities are generally low for silty sands compared to typical values, ranging from 95.4 to 97.9-pound force per cubic feet. Water contents from these measurements ranged from 20.9 to 29.0%, and given the sample locations, it is likely that the in-place material is saturated.

Consolidated drained direct shear tests on conduit floor materials (HBC-18-3) found total stress cohesion (c’) to be 292-pound force per square feet, with an effective stress peak friction angle (’) equal to 39.4°.

Downstream Embankment Materials

Embankment material (Emb) sampled downstream of the crest of the dam in proximity of the toe drain of the dam (DH-18-7 and DH-18-9) consists primarily of silty materials with interbedded sands and clays of low plasticity. Typically, as depth increases, gravel content increases.

Moving downstream from the toe to the off-toe area, significant increases in gravel content are observed, ranging from no gravel at the toe and up to 36% gravel content off-toe. LLs and PIs of the collected samples remained relatively constant, ranging from 24 to 27 and 3 to 7, respectively.

Downstream Alluvium Materials

Quaternary alluvium material (Qal) sampled downstream of the crest of the dam from the toe and off-toe areas of the dam (DH-18-7 and DH-18-9, respectively) consists primarily of sandy materials with interbedded silts and clays of low plasticity. It is observed that gravel content increases with increasing depth at both drill holes. Much like Emb materials, an observed increase in gravel content is seen when moving downstream, increasing from a maximum of 13% gravel content at the toe up to a maximum of 22% gravel content off-toe. LLs and PIs of the collected samples remained relatively constant, ranging from 21 to 27 and 3 to 7, respectively.

Sandstone Foundation Materials

Moving downstream from the crest to off-toe drill holes, sandstone foundation samples consist primarily of silty sands and clayey sands of low-to-moderate plasticity. Gravel contents within the collected samples generally increase with increasing depth and downstream direction, ranging from small amounts (0-2%) of gravel content at the crest and mid-slope (DH-18-2 and DH-18-5, respectively) to a maximum of 38% gravel content off-toe (DH-18-9). However, no gravel was encountered at an off-toe drill hole near the outlet works (DH-18-10). Generally, LLs and PIs increased moving downstream from the crest to the off-toe areas, with some pockets of non-plastic materials; LLs and PIs increased from 25 to 60 and 4 to 40, respectively.

In-situ dry unit weights of sandstone materials from tube densities at the crest (DH-18-2) were variable with depth, while dry unit weights at the toe remain relatively constant with depth (DH- 18-7 and DH-18-9). In-situ dry unit weight ranged from 102.5 to 126.8-pounds per cubic feet at DH-18-2, from 101.2 to 107.3-pounds per cubic feet at DH-18-7, and from 102.8 to 110.5-pounds per cubic feet at DH-18-9. In-situ water contents from tube densities at the crest show an increase in water content with depth, while water contents from toe tube samples were variable with increasing depth. Water contents ranged from 10.3 to 17.4% at DH-18-2, from 21.5 to 25.8% at DH-18-7, and from 20.9 to 24.3% at DH-18-9.

Jet erosion testing (JET) results on sandstone materials from a crest drill hole (DH-18-2) indicates soils ranging from “ extremely erodible” to “moderately resistant.” The erodibility of the material varied due to the relative intactness of the interbedded materials within the specimens. Detachment rate coefficients (kd) ranged from 0.299 to 422.7 feet per hour per pound-force per square feet (ft/hr/psf) at critical shear stresses (τc) ranging from 0.999 to 0.045-pound force per square feet. Flex wall permeability measurements on materials from similar depths in the same drill hole exhibited a horizontal hydraulic conductivity (kh) of 7.8 x 10-4 centimeters per second at an effective consolidation stress (σc’) 12,960-pound force per square feet.

Borrow Soils

Borrow area materials generally consisted of non-plastic silty sand materials, with the exception of a pocket of silty sand with low plasticity (TP-18-3). Two of three borrow area materials do not exhibit plastic behavior; the pocket of silty sand material exhibits plastic behavior with a LL of 22 and a PI of 3. Results from standard Proctor testing on minus 3/8-inch, silty sand material (TP-18-1) indicates a maximum dry unit weight of 130.5-pounds per cubic feet and an optimum water content of 11.2%; corrected for oversize particles, the maximum dry unit weight is 135.6-pounds per cubic feet and the optimum water content is 9.2%.

Bedrock

Uniaxial compressive strength (UCS) testing was performed on six wax-sealed rock core samples (two sandstone samples and four shale samples) to determine the compressive strength of the material at the approximate in-situ water content. Due to the sample quality, one specimen was tested following typical rock test procedures, and the remaining five specimens were tested following typical soil testing procedures. The compressive strength of the cemented sandstone (tested as a rock-type) was determined to be 1,870-pound-force per square inch; the compressive strength of the remaining uncemented sandstone sample (tested as a soil-type) was determined to be 4.74-pound-force per square inch. All shale samples were tested as a soil-type; the compressive strength of the shale samples ranged from 62.3 to 185-pound-force per square inch.

For soil-type UCS specimens, the water content at-test ranged from 22.1% to 26.8%. The minus No. 40 U.S. sieve size materials for shale samples exhibited plastic behavior, with LLs between 43 and 98 and PIs between 22 and 78; the uncemented sandstone sample did not exhibit plastic behavior.

Laboratory Testing Methods & Results

Physical Properties Testing

Gradation analyses were performed in general accordance with USBR 5325-89 and 5330-89. To complete the classification of the samples, Atterberg limits tests were performed on air-dry, minus No. 40 U.S. sieve size material in general accordance to ASTM D4318-17; shale materials were balled-milled to pass the No. 40 U.S. sieve. Standard Proctor testing was also performed in general accordance with ASTM D698-12, Method B. In-situ density measurements were performed in general accordance with ASTM D7263-09. The presence of carbonate cementation was tested for based on reactivity with dilute HCl. Results are summarized in Table 4, and full reports are available in Appendix A.

Conduit Floor Materials

HBC-18-3

Two 3-inch, thin wall Shelby tubes obtained from HBC-18-3 from the depths of 3.2 to 6.9 feet below the conduit floor classified as SM - silty sand, per the Unified Soil Classification system (USCS). The samples consisted of approximately 7 to 8% clays, 10% silts, 81 to 83% sands, and 0 to 1% gravels. The minus No. 40 U.S. sieve size material did not exhibit plastic behavior. The average specific gravity of minus No. 4 U.S. sieve size material was found to be 2.67. The sands of the two samples exhibited no reaction to HCl; gravels found in these samples exhibited a “strong” reaction to HCl. In-situ dry unit weights and water contents from tube density measurements ranged from 95.4 to 97.9-pound force per cubic feet and 26.2 to 26.7%, respectively.

HBC-18-3A

Two grab samples of material obtained from HBC-18-3A from the depths of 3.2 to 4.0 feet below the conduit floor classified as SM - silty sand to SP-SM - poorly graded sand with silt, per USCS. The samples consisted of approximately 3 to 11% clays, 4 to 12% silts, 74 to 93% sands, and 0 to 4% gravels. The minus No. 40 U.S. sieve size material did not exhibit plastic behavior.

The specific gravity of minus No. 4 U.S. sieve size material was found to be 2.70. The sands of the two samples exhibited no reaction to HCl; gravels found in these samples exhibited a “slight” and “strong” reactions to HCl. In-situ water contents from tube density measurements ranged from 20.9 to 24.5%.

HBC-18-4

Two grab samples obtained from HBC-18-4 from the depths of 3.1 to 8.0 feet below the conduit floor classified as SM - silty sand, per USCS. The samples consisted of approximately 11 to 15% clays, 13 to 20% silts, 63 to 76% sands, and 0 to 2% gravels. The minus No. 40 U.S. sieve size material did not exhibit plastic behavior. The specific gravity of minus No. 4 U.S. sieve size material was found to be 2.66. The sands of the two samples exhibited no reaction to HCl;

gravels found in these samples exhibited a “strong” reaction to HCl. In-situ water contents from tube density measurements ranged from 23.1 to 25.6%.

HBC-18-4A

Two grab samples obtained from HBC-18-4A from the depths of 3.1 to 8.0 feet below the conduit floor classified as SM - silty sand to SP-SM - poorly graded sand with silt, per USCS.

The samples consisted of approximately 6 to 15% clays, 6 to 20% silts, 65 to 89% sands, and no gravel. The minus No. 40 U.S. sieve size material did not exhibit plastic behavior. The specific gravity of minus No. 4 U.S. sieve size material ranged from 2.67 to 2.69. The sands of the two samples exhibited no reaction to HCl; no gravel was encountered. In-situ water contents from tube density measurements ranged from 28.3 to 29.0%. The results of the physical properties testing from all samples are summarized in Table 4. Full physical property reports are available in Appendix A and select test and microscope photos are available in Appendix B.

Emb Materials

DH-18-7

Two grab samples from DH-18-7 from the approximate elevations of 2020.4 to 2014.9 feet classified as SC-SM - silty, clayey sand, per USCS. The samples consisted of approximately 7 to 11% clays, 15% silts, 74 to 78% sand, and no gravel. The minus No. 40 U.S. sieve size material exhibited behavior with a range of LLs between 24 and 27 and PIs between 4 and 7.

DH-18-9

Two grab samples from DH-18-9 from the approximate elevations of 2013.0 to 2006.0 feet classified as (SM)g - silty sand with gravel and (SC-SM)g - silty, clayey sand with gravel, per USCS. The samples consisted of approximately 4 to 7% clays, 13 to 14% silts, 46 to 51% sand, and 29 to 36% gravels. The minus No. 40 U.S. sieve size, (SM)g and (SC-SM)g material exhibited plastic behavior with a LL of 23 and 24 and PI of 3 and 5, respectively. The results of the physical properties testing from all samples are summarized in Table 5. Full physical property reports are available in Appendix A.

Qal Materials

DH-18-7

Four grab samples from DH-18-7 from the approximate elevations of 2010.4 to 1997.4 feet classified as SM - silty sand and SC-SM - silty, clayey sand, per USCS. The samples consisted of approximately 7 to 9% clays, 11 to 21% silts, 60 to 82% sand, and 0 to 13% gravels. The minus No. 40 U.S. sieve size, SM and SC-SM material exhibited plastic behavior with a LL of 21 and 25 and PI of 3 and 6, respectively; two SM samples did not show plastic behavior.

DH-18-9

Two grab samples from DH-18-9 from the approximate elevations of 2002.0 to 1998.4 feet classified as (SC-SM)g - silty, clayey sand with gravel, per USCS. The samples consisted of approximately 6% clays, 11 to 12% silts, 61 to 64% sand, and 18 to 22% gravels. The minus No.

40 U.S. sieve size material exhibited plastic behavior with LLs between 24 and 27 and PIs between 4 and 7. The results of the physical properties testing from all samples are summarized in Table 5. Full physical property reports are available in Appendix A.

Sandstone Materials

DH-18-2

One grab samples from DH-18-2 from the approximate elevations of 2045.0 to 2044.4 feet classified as SM- silty sand, per USCS. The samples consisted of approximately 9% clays, 11% silts, 80% sands, and no gravel. The minus No. 40 U.S. sieve size material did not exhibit plastic behavior.

Five grab samples from DH-18-2 from the approximate elevations of 2016.8 to 1996.6 feet classified as SP-SM – poorly graded silty sands, SM – silty sand, and SC-SM – silty, clayey sand, per USCS. The samples consisted of approximately 2 to 4% clays, 6 to 12% silts, 84 to 92% sands, and 0 to 2% gravels. The minus No. 40 U.S. sieve size, SC-SM material exhibited plastic behavior with an average LL of 26 and average PI of 4; SM and SP-SM material did not exhibit plastic behavior.

Two, 4-inch Pitcher tubes from DH-18-2 from the approximate elevations of 1996.6 to 1994.1 feet and 1991.6 to 1989.1 feet classified as SC- clayey sand, per USCS; two additional Pitcher tubes from the approximate elevations of 1994.1 to 1991.6 feet and 1989.1 to 1987.1 feet were also received, but not classified. The two classified samples consisted of approximately 7 to 11% clays, 14 to 28% silts, 61 to 79% sands, and no gravels. The minus No. 40 U.S. sieve size material of the classified samples exhibited plastic behavior with an average LL of 29 and average PI of 13. All Pitcher tubes were tested for tube densities, with in-situ dry unit weights and water contents ranging from 102.5 to 126.8-pound force per cubic feet and 10.3 to 17.4%, respectively.

DH-18-5

One grab sample from DH-18-5 from the approximate elevations of 2043.3 to 2042.8 feet classified as SM – silty sand, per USCS. The sample consisted of approximately 7% clays, 7% silts, 84% sands, and 2% gravels. The minus No. 40 U.S. sieve size material exhibited plastic behavior with a LL of 32 and PI of 8.

Four grab samples from DH-18-5 from the depths of 2036.0 to 2018.0 feet classified as SP-SM – poorly graded silty sands, per USCS. The samples consisted of approximately 3 to 4% clays, 6 to 8% silts, 84 to 91% sands, and no gravels. The minus No. 40 U.S. sieve size material did not exhibit plastic behavior.

DH-18-7

Two grab samples from DH-18-7 from the approximate elevations of 1996.4 to 1994.9 feet classified as SM - silty sand and SC – clayey sand, per USCS. The samples consisted of approximately 4 to 6% clays, 9 to 12% silts, 74 to 83% sand, and 1 to 11% gravels. The minus No. 40 U.S. sieve size, SC material exhibited plastic behavior with a LL of 37 and PI of 21; the SM sample did not exhibit plastic behavior.

Additionally, five, 5-inch thin wall Shelby tubes from the approximate elevations of 1994.9 to

1984.4 feet were tested for tube densities following collection of samples. In-situ dry unit weights and water contents from tube density measurements ranged from 101.2 to 107.3-pound force per cubic feet and 21.5 to 25.8%, respectively.

DH-18-9

Two grab samples from DH-18-9 from the approximate elevations of 1998.4 to 1996.5 feet classified as (SM)g - silty sand with gravel and (SP-SM)g – poorly graded sand with silt and gravel, per USCS. The samples consisted of approximately 11 to 15% clays and silts, 51 to 60% sand, and 25 to 38% gravels. The minus No. 40 U.S. sieve size material did not exhibit plastic behavior.

Additionally, four, 4-in Pitcher tubes from the approximate elevations of 1995.5 to 1986.5 feet were tested for tube densities following collection of samples. In-situ dry unit weights and water contents from tube density measurements ranged from 102.8 to 110.5-pound force per cubic feet and 20.9 to 24.3%, respectively.

DH-18-10

One grab sample from DH-18-10 from the approximate elevations of 2004.8 to 2002.8 feet classified as s(CL) - sandy lean clay and SC - clayey sand, per USCS; stick logs indicate that interbedded shale fragments in this interval. The samples consisted of approximately 7 to 12% clays, 11 to 42% silts, 43 to 73% sands, and no gravel. The minus No. 40 U.S. sieve size, s(CL) material exhibited plastic behavior with a LL of 32 and a PI of 10, while the SC material showed a LL of 60 and a PI of 40. The results of the physical properties testing from all samples are summarized in Tables 4 and 5. Full physical property reports are available in Appendix A.

Test Pit Materials

TP-18-1

One, 5-gallon bucket of soil obtained from TP-18-1 from a depth of approximately 4.5 feet classified as SM - silty sand, per USCS. The sample consisted of approximately 11% clays, 23% silts, 64% sands, and 2% gravels. The minus No. 40 U.S. sieve size material did not exhibit plastic behavior. The specific gravity of minus No. 4 U.S. sieve size material was found to be 2.69.

An additional, 5-gallon bucket of soil obtained from TP-18-1 from a depth of approximate 7.0 feet was used for specific gravity and standard Proctor testing; USCS classification was not performed on this sample. The specific gravity of minus No. 4 U.S. sieve size material was found to be 2.82. Standard Proctor compaction tests on minus 3/8-inch material determined the maximum dry unit weight to be 130.5-pound force per cubic feet with an optimum water content of 11.2%. Corrected for oversized particles, the representative maximum dry unit weight is 135.6-pound force per cubic feet with an optimum water content of 9.2%.

TP-18-2

One, 5-gallon bucket of soil obtained from TP-18-2 from a depth of approximately 3.0 feet classified as SM - silty sand, per USCS. The sample consisted of approximately 10% clays, 25% silts, 64% sands, and 2% gravels. The minus No. 40 U.S. sieve size material did not exhibit plastic behavior. The specific gravity of minus No. 4 U.S. sieve size material was found to be approximately 2.67.

TP-18-3

One, 5-gallon bucket of soil obtained from TP-18-3 from a depth of approximately 5.0 feet classified as SM - silty sand, per USCS. The sample consisted of approximately 8% clays, 11% silts, 81% sands, and 1% gravels. The minus No. 40 U.S. sieve size material exhibited plastic behavior with a Liquid Limit (LL) of 22 and a Plasticity Index (PI) of 3. The specific gravity of minus No. 4 U.S. sieve size material was found to be approximately 2.87. The results of the physical properties testing from all samples are summarized in Table 4. Full physical property reports are available in Appendix A.

Jet Erosion Testing (JET)

One JET on sandstone, clayey sand (SC) materials from DH-18-2 from the approximate elevations of 1996.6 to 1994.1 feet showed a kd 422.7 feet per hour per pound-force per square feet (ft/hr/psf) at a τc of 0.045-pound force per square feet. JET erodibility classification was determined to be “extremely erodible.”

Two jet erosion tests (JET) on sandstone, clayey sand (SC) materials from DH-18-2 from the approximate elevations of 1991.6 to 1989.1 feet showed a kd of 0.454 to 0.299 feet per hour per pound-force per square feet (ft/hr/psf) at a τc of 0.527 to 0.999-pound force per square feet. JET erodibility classifications were determined to be “erodible” to “moderately resistant.”

JET measurements on the samples obtained from approximate elevations of 1991.6 to 1989.1 feet contained interbedded layers of variably erodible material (i.e., erodible sandstone and moderately resistant siltstone and shale) (see Figure 4). The sandstone specimen was tested under 48-inches of head until erosion ceased. The head was subsequently raised to 72-inches, and a test was repeated with a new specimen (siltstone and shale) and was tested until erosion ceased. The test on the sandstone indicated "erodible", while the test on the siltstone and shale indicated "moderately resistant" erodibility.

Hard, cemented sandstone underlies the material that was used for JET measurements.

According to final drill logs, drilling was met with refusal due to this harder sandstone around the approximate elevation of 1987.1 feet. It is possible that this cemented sandstone is interbedded within the tested samples from the approximate elevations from 1991.6 to 1989.1 feet, which could contribute to the low kd and high τc values. The results of JET measurements from all samples are summarized in Figure 4 and Table 6.

The submerged JET test was developed at the Agricultural Research Service Hydraulic Engineering Research Unit, Stillwater, Oklahoma (Hanson and Cook, 2004). This test can be performed in-situ, or in the laboratory using tube samples or remolded samples in compaction molds (Hanson and Hunt, 2006). Testing has been successfully carried out on specimens as small as 3 inches in diameter, but a minimum specimen size has not been firmly established.

There is currently no testing standard, but the test apparatus is detailed in the now withdrawn

ASTM D5852.

The JET apparatus is designed to apply shear stress at the soil surface with an impinging submerged jet, which is produced by a 0.25-inch diameter nozzle initially positioned between 6 and 30 nozzle diameters from the soil surface. The starting nozzle position and test head may be adjusted to vary the stress applied to the soil sample, although once a test head is selected it is usually held constant for the duration of a test. Scour of the soil surface beneath the jet is measured over time (typically a few minutes up to 2 hours) using a point gauge aligned with the axis of the jet. Data from the JET have typically been analyzed using a volumetric form of the same erosion model used to analyze HET data (as discussed below):

𝜀ሶ ൌ 𝑘ௗሺ𝜏 െ 𝜏ሻ where, 𝜀ሶ = the volume of material removed per unit surface area per unit time (ft/s/ft², or ft/s) kd = the detachment rate coefficient (ft/hr/psf) τ = the applied shear stress (psf) τc = critical shear stress, or the threshold shear stress for soil detachment (psf)

Reclamation’s JET apparatus is installed in the hydraulics laboratory in Denver, Colorado. Data is collected manually during a test using the procedures described by Hanson and Cook (2004).

Hanson and Simon (2001) proposed a qualitative classification of the erodibility of soils. Their classification scheme identifies five erodibility groupings ranging from very resistant to very erodible based on the kd and τc value of the soil. In connection with JET testing of sediment samples from Paonia Reservoir, Wahl (2014) suggested a sixth category for extremely erodible soils.

The solution method utilizes an Excel spreadsheet and the Solver utility, a goal seeking optimization tool. Two parameters are calculated with the Solver to determine the τc and kd. The solution procedure begins by fitting the measured scour and time data to an asymptotic function that predicts the equilibrium depth of scour that would occur at t = ∞. Once this distance is determined, the corresponding stress that would be produced by the jet at this scour distance is defined as the τc. Note that the stress on the soil surface decreases as scour occurs, so the test is progressing from a condition of high stress to one of low stress. The τc condition is not actually reached in most tests. In these cases, the τ exceeds the τc, and erosion never ceases. The second part of the analysis determines the value of the kd that produces a best fit of the scour vs. time data (Hanson and Cook, 2004).

Horizontal Flexible Wall Permeability Testing

Three flex wall permeability tests were performed on clayey sand (SC) materials from DH-18-2 from an approximate elevation of 1995.1 feet. Trimming was attempted on samples from DH- 18-2 from the approximate elevations from 1991.6 to 1989.1 feet, but shale lenses prohibited trimming an intact specimen. A total of three specimens were tested, with the first two used to refine the methods used during testing; only the third test data is considered valid and reported herein. The third test showed a kh of 7.8 x 10-4 centimeters per second at a σc’ of 12,960-pound force per square feet. The results of flex wall permeameter testing for this sample is summarized in Table 7.

Falling head rising tailwater permeability testing was performed on soft, sandstone materials in a flexible wall permeameter in accordance with ASTM D5084, Method C to determine the kh of intact, hand-trimmed soil specimens. The specimens were prepared by carefully removing the material from the sample tube without extruding it (Ladd and DeGroot, 2003). The material was then placed on its side to trim horizontally. The specimens were hand trimmed according to USBR 5215 using a 2-inch inner diameter cutting shoe affixed to a hand operated press. Once trimming was complete, specimen ends were squared using trimming tools and a mold.

The specimen was placed in a latex membrane and set in the permeation chamber. After the chamber was filled and all lines were attached to the permeability panel, the saturation began.

The specimen was backpressure saturated to a minimum B-value of 0.95. After saturation, the specimen was isotropically consolidated by increasing the cell pressure to the desired effective stress and left until primary consolidation was complete and a well-defined state of secondary consolidation was established.

A maximum hydraulic gradient of five was chosen for this soil based on guidance from ASTM D5084 and the soil classification. The gradient was applied via head differential across the top of the specimen and the bottom of the specimen. Inflow and outflow readings were taken every minute, and the test concluded prior to the gradient decreasing to 75% of the initial gradient.

After kh was recorded, the initial gradient was reset, and the test resumed until at least four values of kh fell within ± 25% of the mean kh value. This testing was performed using laboratory tap water.

Direct Shear Testing

Consolidated-drained direct shear testing on silty sand (SM) material from HBC-18-3 from a depth of 5.5 to 6.9 feet below the conduit floor showed a ’ equal to 39.4° with a c’ of 292-pound-force per square foot. Well-defined, peak drained shear stresses for this sample were fully developed between 3.4% and 7.5% relative horizontal strain. Specimens exhibited contractive behavior during shear, with all three specimens contracting less than 0.6% axial strain at failure.

The results of direct shear testing for these samples are summarized in Figure 5.

Consolidated-drained, soil direct shear strength testing was performed according to ASTM D3080. A nominal 2x2x1.25-inch (length by width by height), two-part shear box was employed for testing samples. Specimens were prepared by hand-trimming intact samples from 3-in Shelby tube samples into the shear box apparatus; the trimmed material was classified as minus No. 4 material. Once trimmed, the shear box was mounted into the testing machine and inundated in a water bath prior to consolidation under a seating load provided by the shear box top plates and the weight of the loading yoke.

One test suite, consisting of testing three separate specimens consolidated at three normal stresses, was performed. The separate specimens were consolidated at 2.5 lbf/in2, 5.0 lbf/in2, and 10 lbf/in2 (nominal) before shearing the samples. The specimen was allowed to consolidate until it was in a well-defined state of secondary consolidation before any shear deformations were applied.

Normal force was applied to the specimen by means of a loading yoke and hanger mass. The normal force was manually controlled and kept constant throughout each consolidation phase.

The normal force was calculated based on the desired normal stress for consolidation and shear stages, adding the necessary weights on the hanger mass to apply the force.

The shearing portion of the test was strain controlled via a programmable gear box and performed in a drained manner – i.e., the applied strain rate was sufficiently slow to ensure that excess pore water pressures did not develop in the specimen. For the testing performed here, a strain rate of 0.0005 in/min (equal to 1.5% relative horizontal displacement/hour) was used.

Tests were run overnight and until a well-defined post-peak shear strength was established. The normal and horizontal displacements were monitored with linear variable differential transformers (LVDTs), and the shear load was monitored with a strain-gage based load cell. All sensors were connected to PC-based data acquisition.

While the state of stress in the sheared zone is not fully known, the analysis is carried out assuming that the failure plane is horizontal, i.e., the applied normal stress is the normal stress on the failure plane and that the measured horizontal stress is the shear stress on the failure plane.

The shear stress was corrected for the changing shear surface area as straining progresses. The stress required to shear each specimen was then plotted against the relative horizontal strain.

The peak shear stresses for each testing suite are plotted against the associated normal stresses, and a best-fit line is drawn through the data. This Mohr-Coulomb linear failure envelope defines the peak, drained shear strength parameters for the soil: the slope of the line is the ’, and the y-intercept is the c’. The best-fit line is then fit through the origin and ’ is determined from the slope.

Note, that drained strength parameters from direct shear tests are typically higher than those obtained from triaxial shear strength testing due to the failure plane being restrained horizontally (Lambe and Whitman, 1969).

Petrographic Examination

Conduit Floor Materials

X-ray diffraction (XRD) on oven-dried, air-dried, and natural water content materials from HBC- 18-3A found significant variability in non-clay and clay composition. Oven-dried material contained significantly more (>10%) non-clay minerals as compared to air-dried and natural water content material.

XRD on air-dried material from HBC-18-4 also indicates significant variability in non-clay and clay percentages compared to air-dried and natural water content samples from HBC-18-3A.

Compared to the other conduit floor materials from HBC-18-3A, the air-dried material from HBC-18-4 contains significantly less non-clay minerals. No discernable trends are observed across the two sampling locations. The results of XRD and specific minerology for conduit floor materials can be found in Table 8.

Sandstone Materials

XRD on sandstone materials from DH-18-7 at the approximate elevation of 1995.4 feet and from DH-18-10 from the approximate elevations of 1996.1 feet and 1991.8 contained similar non-clay and clay mineral percentages. Most notably, the materials from DH-18-10 from the approximate elevation of 1991.8 contained slightly more (<10%) quartz non-clay minerals compared to the previously mentioned samples.

XRD on the remaining sandstone materials from DH-18-10 at the approximate elevations of

2005.8 and 2002.1 indicate that these samples contain significantly less non-clay minerals and significantly more clay minerals as compared to the other sandstone samples. It is observed, that there is a decrease in the proportion of clay minerals in the samples from DH-19-10 with depth.

Additionally, the sample from DH-18-7 and the sample from DH-18-9 at a similar depth contain similar mineral contents. The results of XRD and specific minerology for sandstone materials can be found in Table 9.

Shale Materials

XRD on shale materials from DH-18-2 at the approximate elevation of 1950.5 and DH-18-9 at the approximate elevation of 1968.4 indicates that the sample from DH-18-2 contained significantly more non-clay minerals than the sample from DH-18-9. No discernable trends are observed across the two sampling locations. The results of XRD and specific minerology for shale materials can be found in Table 9.

Petrographic examinations of four conduit floor materials, five sandstone samples, and two shale samples were performed by GLFS and CAMET contract lab personnel using XRD to understand the mineralogical composition of the received materials. Detailed in-house and contract lab petrography test procedures and results are available in Appendix C.

Rock Uniaxial Compressive Strength Testing

The UCS specimens were delivered to the TSC as waxed core specimens, and were tested at the approximate water content at the time of waxing. Due to the fragile nature of the samples, specimens were not checked for tolerances after cutting. All specimens were tested with spherical platens.

Rock-type UCS

Rock-type UCS was performed on one of the six specimens. The specimen was tested on the servo-hydraulic stiff testing equipment which was controlled by lateral-circumferential strain as measured during testing. Clip-on gauges measured axial and lateral strain during the tests while an internal load cell measured the applied load.

The uniaxial test followed a procedure to settle the strain gages and reduce the gage hysteresis effects at small strains (Hakala and Heikkilä, 1997):

1. Zero load cell and strain gages

2. Start data acquisition

3. Seat specimen at 0.1 kips axial load

4. Start test in load control to 0.8 kips axial load

5. Switch to circumferential strain control with testing performed at rate determined by operator such that failure occurred with 15 minutes

6. Test to failure as determined by the operator, stop data acquisition

Rock-type testing on the sandstone specimen resulted in a UCS of 1.87E+03-pound-force per square inch. The average elastic (Young’s) modulus (the average Young’s modulus is defined as the Young’s modulus evaluated from 40% to 70% strain at failure) was determined to be 2.14E+05-pound-force per square inch, and the tangent (evaluated from 20% to 50% strain at failure) and secant (evaluated from 0% to 50% strain at failure) Young’s moduli were determined to be 1.56E+05 and 1.36E+05-pound-force per square inch, respectively. The average Poisson’s ratio (Poisson’s ratio evaluated from 40% to 70% strain at failure) was determined to be 0.41, and the tangent (evaluated from 20% to 50% strain at failure) and secant (evaluated from 0% to 50% strain at failure) Poisson’s ratios were determined to be 0.39 and 0.33, respectively.

Soil-type UCS

The remaining specimens (four shale samples and one sandstone sample) could not hold the applied seating load in the rock testing machine and could not be tested for compressive strength;

these specimens were subsequently tested as a soil-type UCS on another machine. The results from soil-type UCS testing were considered more reliable if the peak axial load exceeded seating load from the rock-type UCS procedures, and are likely within an order of magnitude of the true compressive strength for the material. Elastic modulus and Poisson’s ratio measurements also could not be obtained on the specimens that were tested as a soil, since these measurement apparatuses were not available in the soil UCS machine.

The soil-type UCS specimens were tested in general accordance with ASTM D2166. For all soil-type UCS specimens, the at-test water content and Atterberg limits were determined, in lieu of elastic modulus determination and detailed strain measurements. The approximate, at-test water contents were also determined on soil-type UCS specimens in general accordance with ASTM D2216. Atterberg limits testing for sandstone was performed in general accordance with ASTM D4318, but sample preparation of shale samples were performed in general accordance with guidelines outlined in Stark et al. (2005).

According to Stark et al. (2005), heavily-overconsolidated clays, mudstones, claystones, and shales show variability in the bonds between clay mineral particles and various ionic compounds.

The degree to which these bonds are disaggregated can influence the measured index properties.

The researcher determined that traditional grinding methods – such as the mortar and pestle – cannot adequately disaggregate the bond within these materials. Therefore, ball mill grinding was used as this technique was found to sufficiently disaggregate clay mineral bonds, so that the behavior of the sample mimicked the broken down (disaggregated) particles in the aforementioned materials during shear testing. The minus No. 200 U.S. sieve size particles from ball milled samples resulted in a more accurate estimate of the actual LL of the material, since the elimination of the ionic bonding allowed for an increased susceptibility to hydration due to the free area created by effective disaggregation of the material.

Atterberg limit specimens from the shale samples were ball milled for 10 minutes and allowed to hydrate for at least 6 days prior to testing. Since the tested shale samples contained some amount interbedded sample, minus No. 40 U.S. sieve size material was used, instead of the recommended minus No. 200 U.S. sieve size material as outlined by Stark et al. (2005). The fine sand particles are likely to affect the plasticity of the material, and were thus included to obtain accurate estimates of LL and PI. However, specimen preparation for the sandstone sample was prepared in general accordance with the standardized method outlined in ASTM D4318.

Soil-type testing on the four shale samples resulted in a range of UCS from 6.23E+01-pound-force per square inch to 1.85E+02-pound-force per square inch, with an average of 1.28E+02-pound-force per square inch and a standard deviation of 5.31E+01-pound-force per square inch.

The specimens experienced peak axial strains ranging from 9.90E-02 to 2.45E-01, with an average axial strain of 1.91E-01 and a standard deviation of 6.36E-02. Soil testing on the remaining sandstone sample resulted in a UCS of 4.74E+00-pound-force per square inch, experiencing a peak axial strain of 2.03E-01. The results from the shale sample are considered relatively reliable, since the peak axial loads exceeded the rock-type UCS seating load; however, one shale sample was slightly below the recommended two-to-one length-to-diameter ratio (1.90). The results from the uncemented sandstone sample are not considered reliable, since the peak axial load did not exceed the rock-type UCS seating load. Due to the limitations of the soil testing equipment, average elastic modulus and Poisson’s ratio were not measured for any of these UCS specimens. UCS test results for the shale samples were relatively variable; the strength of the strongest specimen is approximately three times higher than the weakest specimen, with the standard deviation being about 40% of the mean value.

The testing water content was determined for all specimens tested as a soil-type UCS. At the time of testing, the approximate water content of the four shale samples ranged from 22.1% to 26.8%, with an average of 24.2% and a standard deviation of 2.3%. The approximate water content of the sandstone sample was 22.4%. Minimal differences were observed in the testing water contents across the sandstone and shale samples.

Atterberg limits testing was also performed on all specimens tested as a soil-type UCS using the top and bottom cut-ends of the samples, which are typically used for rock-type physical properties testing. The minus No. 40 U.S. sieve size material from the shale samples exhibited plastic behavior, with LLs between 43 and 98 and PIs between 22 and 78, with averages of 75 and 56 and standard deviations of 24 and 25, respectively. The minus No. 40 U.S. sieve size material from the sandstone sample did not exhibit plastic behavior. The results of rock- and soil-type UCS and soil-type physical properties testing are summarized in Table 10, with full reports in Appendix D.

References ASTM D698, (2012). Standard Test Methods for Laboratory Compaction Characteristics of Soil

Using Standard Effort (12 400 ft-lbf/ft3 (600 kN-m/m3)), ASTM International, West Conshohocken, PA.

ASTM D854, (2014). Standard Test Methods for Specific Gravity of Soil Solids by Water Pycnometer, ASTM International, West Conshohocken, PA.

ASTM D2166, (2016). Standard Test Method for Unconfined Compressive Strength of Cohesive Soils, ASTM International, West Conshohocken, PA.

ASTM D3080, (2011). Standard Test Methods for Direct Shear Test of Soils Under Consolidation Drained Conditions, ASTM International, West Conshohocken, PA.

ASTM D4318, (2017). Standard Test Methods for Liquid Limit, Plastic Limit, and Plasticity Index of Soils, ASTM International, West Conshohocken, PA.

ASTM D5084, (2016). Standard Test Methods for Measurement of Hydraulic Conductivity of Saturated Porous Materials Using a Flexible Wall Permeameter, ASTM International, West Conshohocken, PA.

ASTM D5852, (2007). Standard Test Methods for Erodibility of Soil in the Field or in the Laboratory by the Jet Index Method, ASTM International, West Conshohocken, PA.

ASTM D7263, (2009). Standard Test Methods for Laboratory Determination of Density (Unit Weight) of Soil Specimens, ASTM International, West Conshohocken, PA.

Germaine, J. T. and Germaine, A. V., (2009). “Geotechnical Laboratory Measurements for Engineers,” John Wiley and Sons, Inc., Hoboken, NJ.

Hanson, G.J. and Simon. A., (2001). “Erodibility of cohesive streambeds in the loess area of the midwestern USA,” Hydrological Processes, Vol. 15, pp. 23-38.

Hanson, G.J. and Cook, K.R., (2004). “Apparatus, test procedures, and analytical methods to measure soil erodibility in situ,” Applied Engineering in Agriculture, Vol. 20, No. 4, pp. 455- 462.

Hanson, G.J. and Hunt, S.L., (2006). “Lessons learned using laboratory jet method to measure soil erodibility of compacted soils,” Applied Engineering in Agriculture, Vol. 23, No. 3, pp.

305-312.

Ladd, C. C. and DeGroot, D. J.