El Vado Dam – Service Spillway Modification – Physical Model Study

Hydraulic Laboratory Report HL-2017-02

U.S. Department of the Interior Bureau of Reclamation Technical Service Center Hydraulic Investigations and Laboratory Services Group Denver, Colorado April 2017

Form Approved REPORT DOCUMENTATION PAGE OMB No. 0704-0188 The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing the burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. 1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 3. DATES COVERED (From - To) 04-27-2017 Hydraulics Laboratory Report January 2016-January 2017

4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER El Vado Dam – Service Spillway Modification - Physical NA Model Study 5b. GRANT NUMBER NA 5c. PROGRAM ELEMENT NUMBER NA 6. AUTHOR(S) 5d. PROJECT NUMBER NA Bryan J. Heiner 5e. TASK NUMBER Christopher C. Shupe NA 5f. WORK UNIT NUMBER NA 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION Bureau of Reclamation REPORT NUMBER PO Box 25007 HL-2017-02 Denver, CO 80225

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR'S Bureau of Reclamation ACRONYM(S) USBR 11. SPONSOR/MONITOR'S REPORT NUMBER(S) NA 12. DISTRIBUTION/AVAILABILITY STATEMENT NA 13. SUPPLEMENTARY NOTES NA 14. ABSTRACT A 1:24-scale physical hydraulic model for the feasibility design (corrective action study preferred option) of the service spillway at El Vado Dam was constructed at Reclamation’s Hydraulics Laboratory in Denver, Colorado. The model examined hydraulic performance and optimized the feasibility design for the preferred option. Modifications include relocation of the spillway entrance 20 ft farther into the right abutment and realignment of the spillway with the existing spillway alignment by means of a long radius channel bend from approximately station 2+00 to 4+00. The features included in the physical model were approximately one third of the dam crest, the spillway entrance channel, gate structure, chute, and plunge pool in the downstream river channel.

15. SUBJECT TERMS

16. SECURITY CLASSIFICATION OF: 17. LIMITATION 18. 19a. NAME OF RESPONSIBLE PERSON OF ABSTRACT NUMBER OF PAGES 65 Robert F. Einhellig a. b. a. THIS PAGE 19b. TELEPHONE NUMBER REPORT ABSTRACT 303-445-2142 Standard Form 298 (Rev. 8/98) Prescribed by ANSI Std. Z39.18

Hydraulic Laboratory Report HL-2017-02

El Vado Dam – Service Spillway Modification – Physical Model Study

Prepared: Bryan J. Heiner, P.E. Hydraulic Engineer, Hydraulic Investigations and Laboratory Services Group, 86-68560

Prepared: Christopher C. Shupe, E.I.T. Hydraulic Engineer, Hydraulic Investigations and Laboratory Services Group, 86-68560

Technical Approval: Robert F. Einhellig, P.E. Manager, Hydraulic Investigations and Laboratory Services Group, 86-68560

Peer Review: Joshua Mortensen, P.E. Date Hydraulic Engineer, Hydraulic Investigations and Laboratory Services Group, 86-68560

U.S. Department of the Interior Bureau of Reclamation Technical Service Center Hydraulic Investigations and Laboratory Services Group Denver, Colorado April 2017

Mission Statements

The mission of the Department of the Interior is to protect and provide access to our Nation's natural and cultural heritage and honor our trust responsibilities to Indian Tribes and our commitments to island communities. ______

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.

Acknowledgments The authors would like to thank Jason Black, Jimmy Hastings, Marty Poos and Dane Cheek who constructed the physical model.

Hydraulic Laboratory Reports The Hydraulic Laboratory Report series is produced by the Bureau of Reclamation’s Hydraulic Investigations and Laboratory Services Group (Mail Code 86-68560), PO Box 25007, Denver, Colorado 80225-0007. At the time of publication, this report was also made available online at https://www.usbr.gov/tsc/techreferences/hydraulics_lab/reportsdb/reportsearchall.cfm

Disclaimer The information provided in this report is believed to be appropriate and accurate for the specific purposes described herein, users bear all responsibility for exercising sound engineering judgment in its application, especially to situations different from those studied. References to commercial products do not imply endorsement by the Bureau of Reclamation and may not be used for advertising or promotional purposes.

Cover Photo: El Vado Service Spillway Model taken in May 2016, Reclamation.

CONTENTS

EXECUTIVE SUMMARY ...... 1

INTRODUCTION ...... 2 Project Background ...... 2 Prototype Operation ...... 6

MODEL OBJECTIVES ...... 7

MODEL DESCRIPTION ...... 7 Numerical Model ...... 7 Physical Model Scale ...... 9 Physical Model Features ...... 9 Physical Model Instrumentation ...... 14

MODEL TEST PLAN ...... 16

RESULTS AND DISCUSSION ...... 16 Numerical Model ...... 16 Physical Model ...... 18 Spillway Entrance Guide Walls ...... 19 Transition Section between Vertical and Sloped Spillway Walls...... 24 Final Physical Model Modifications (Configuration “J”) ...... 25 Downstream River Plunge Pool ...... 26 Approach Channel Velocities ...... 28 Spillway Entrance Water Surface Profiles ...... 31 Spillway Chute Water Surface Profiles and Velocities ...... 33 Rating Curves ...... 38 Cavitation and Air Bulking ...... 41

CONCLUSIONS...... 42

REFERENCES ...... 44

APPENDIX A: FLOW-3D NUMERICAL MODEL CONFIGURATIONS ...... 45

APPENDIX B: SPILLWAY FLOW DEPTHS AND VELOCITY DATA...... 52

TABLES

Table 1. Approach channel velocities at 60% depth for flow rate of 5,000 ft3/s. .29

Table 2. Approach channel velocities at 20% depth for flow rate of 18,600 ft3/s...... 29

Table 3. Approach channel velocities at 60% depth for flow rate of 18,600 ft3/s...... 30

Table 4. Approach channel velocities at 80% depth for flow rate of 18,600 ft3/s...... 31

Table 5. Water surface elevations through the approach channel, bridge and gate sections of the spillway...... 33

Table 6. Reservoir inflow rates and corresponding reservoir water surface elevations for various spillway entrance configurations tested in the physical model...... 39

Table 7. Spillway chute flow depths and velocities for approach channel configuration “J” in prototype dimensions...... 52

FIGURES

Figure 1. El Vado Dam and Middle Rio Grande Project general location map. ....3

Figure 2. Photograph of broken vertical weld in steel lining in spillway chute. ....4

Figure 3. Photograph of severely corroded supporting steel behind the steel plates...... 4

Figure 4. Photograph of damaged steel lining in spillway chute...... 5

Figure 5. Photograph of chute wall leaking, indicating that pressure is behind steel lining panels...... 5

Figure 6. Aerial image of the existing and proposed (overlaid in red) spillway entrance locations and partial chute alignments...... 6

Figure 7. Aerial image with approximate numerical model extents overlaid in red...... 8

Figure 8. Aerial image with approximate physical model extents overlaid in red...... 10

Figure 9. Aerial image of physical model extents...... 10

Figure 10. Cross-section of El Vado service spillway with the vertical parapet walls shown as designed and as built in the model...... 11

Figure 11. Elevated PVC spillway viewed from left of the spillway...... 11

Figure 12. Plan and section drawing of existing and proposed modified spillway alignment (in red)...... 12

Figure 13. Model viewed from upstream of the reservoir...... 13

Figure 14. Bridge, gate, and transition sections of the spillway with no gate installed...... 14

Figure 15. Bridge, gate, and transition sections of the spillway with gate installed...... 14

Figure 16. Aerial photograph with pressure port locations in the approach channel to the service spillway entrance and the reservoir...... 15

Figure 17. Schematic of approach channel alignment configurations investigated in the numerical (A-J) and physical (F, I, J) models...... 17

Figure 18. Plan view with velocity contours for visualization (configuration “F”)...... 18

Figure 19. Configuration “F” velocity contour plot looking up the spillway to visualize wave actions occurring in the spillway chute downstream...... 18

Figure 20. Entrance transition wall geometries that were included in the physical model...... 19

Figure 21. Approach channel configuration “F” with no left guide wall at flow rate of 18,600 ft3/s...... 20

Figure 22. Approach channel configuration “F” with no left guide wall at flow rate of 18,600 ft3/s...... 20

Figure 23. Approach channel configuration “F” with 30-ft circular left guide wall at flow rate of 18,600 ft3/s...... 21

Figure 24. Approach channel configuration “F” with 30-ft circular left guide wall at flow rate of 18,600 ft3/s...... 22

Figure 25. Approach channel configuration “F” with elliptical, 120-ft by 60-ft, left guide wall at flow rate of 18,600 ft3/s...... 23

Figure 26. Approach channel configuration “F” with elliptical, 120-ft by 60-ft, left guide wall at flow rate of 18,600 ft3/s...... 23

Figure 27. Approach channel configuration “F” with elliptical, 120-ft by 60-ft, left guide wall at flow rate of 18,600 ft3/s...... 24

Figure 28. Comparison of original 17-ft 3-in (bottom) and 24-ft 6-in lengthened transitions (top) of the spillway...... 25

Figure 29. Approach channel configuration “J” with elliptical left and right guide walls at flow rate 18,600 ft3/s...... 26

iii

Figure 30. Jet plunging into the plunge pool in the downstream river channel at flow rate 18,600 ft3/s...... 27

Figure 31. Jet plunging into the downstream river plunge pool at spillway flow rate of 18,600 ft3/s...... 28

Figure 32. Velocity contours in spillway approach channel at 20% depth for flow rate of 18,600 ft3/s...... 29

Figure 33. Velocity contours in the spillway approach channel at 60% depth for flow rate of 18,600 ft3/s...... 30

Figure 34. Velocity profile, computed using FLOW-3D, in spillway approach channel at 80% depth for flow rate of 18,600 ft3/s...... 31

Figure 35. Left wall water surface profile through the guide wall, bridge and gate sections of the spillway...... 32

Figure 36. Right wall water surface profile through the guide wall, bridge and gate sections of the spillway...... 32

Figure 37. Center of approach channel water surface profile. Water surface profile shown in blue...... 32

Figure 38. Center of channel water surface profile through the guide wall, bridge and gate sections of the spillway...... 32

Figure 39. Water surface profiles (blue line) on left and right spillway chute walls with station numbers...... 34

Figure 40. Spillway water surface profiles (blue line) from station 1+00 to 2+75...... 35

Figure 41. Spillway water surface profiles (blue line) from station 2+75 to 4+50...... 35

Figure 42. Spillway water surface profiles (blue line) from station 4+50 to 6+25...... 36

Figure 43. Spillway water surface profiles (blue line) from station 6+25 to 8+00...... 36

Figure 44. Spillway water surface profiles (blue line) from station 8+00 to 9+75...... 37

Figure 45. Spillway water surface profiles (blue line) from station 9+75 to 10+48.60...... 38

Figure 46. Rating curve for four entrance configurations tested in the physical model...... 40

Figure 47. Rating curve (with polynomial trend line) for service spillway at El Vado Dam with entrance configuration “J”...... 40

Figure 48. FLOW-3D approach channel configuration “A” isometric view...... 46

Figure 49. FLOW-3D approach channel configuration “B” isometric view...... 46

Figure 50. FLOW-3D approach channel configuration “C” isometric view...... 47

Figure 51. FLOW-3D approach channel configuration “D” isometric view...... 47

Figure 52. FLOW-3D approach channel configuration “E” isometric view...... 48

Figure 53. FLOW-3D approach channel configuration “F” isometric view...... 48

Figure 54. FLOW-3D approach channel configuration “G” isometric view...... 49

Figure 55. FLOW-3D approach channel configuration “H” isometric view...... 49

Figure 56. FLOW-3D approach channel configuration “I” isometric view...... 50

Figure 57. FLOW-3D approach channel configuration “J” isometric view...... 50

Figure 58. FLOW-3D approach channel configuration “J”, plan view...... 51

Page left blank

Executive Summary El Vado Dam is part of Reclamation’s Middle Rio Grande Project in north central New Mexico. The dam is located on the Rio Chama River about 160 miles north of Albuquerque. El Vado Dam is a 175-ft-high gravel fill embankment dam with a welded steel plate lining on the upstream face of the dam. There is a steel-lined service spillway at the right abutment. The 1,049-ft-long, steel-lined service spillway is currently under operating restrictions which limits discharges to 2,500 ft3/s due to the deteriorated condition of the spillway.

A corrective action study (CAS) developed three options for modifying the service and emergency spillways. The preferred option includes replacing the existing service spillway and constructing a small dike at the emergency spillway (this is referred to as option 1 in the CAS). The new service spillway would have similar dimensions and invert elevations as the existing one, and would have a new radial gate, new gate hoists, and a new bridge over the top of the new gate structure for local traffic. The new service spillway gate structure would be relocated approximately 20 ft into the right abutment and would re-align with the existing spillway alignment by means of a long radius channel bend (horizontal curve) from approximately station 2+00 to station 4+00.

The Bureau of Reclamation’s (Reclamation) Hydraulics Laboratory in Denver, Colorado constructed a 1:24-scale physical hydraulic model of the preferred option for the service spillway at the right abutment. The physical model, requested by the team leader, with approval of the Dam Safety Office, was constructed to ensure that the new spillway could pass the design maximum flow 18,600 ft3/s at the design maximum water surface elevation (WSE) 6911.0. In conjunction with the physical model, numerical modeling was used to help evaluate entrance configurations that would provide favorable hydraulic conditions from the upstream reservoir to the end of the spillway.

At least 10 configurations of the entrance channel were evaluated with numerical modeling. At least four of the configurations were tested in the physical model with differing entrance guide walls to determine if wave heights down the spillway would overtop the proposed walls. Results from the physical and numerical models compared well and were used to develop the recommended entrance approach configuration having a 1:1 sloped right bank approach with two elliptical entrance guide walls.

Spillway walls offer sufficient overtopping protection at the design maximum flow, provided the flow does not contact the radial gate or its supports when in the fully open position. Water surface profiles were collected through the spillway gate section to ensure that the new radial gate and supports, when the gate is in the fully opened position, are designed to remain above the water surface at the maximum design flow rate 18,600 ft3/s. Water surface profiles along the left and right spillway chute walls are provided for the entire length of the chute.

Velocities along the right bank of the upstream entrance channel ranged from 3.3 ft/s at station -1+75 to 20.0 ft/s at station 0+00. Velocities in the chute ranged from 27.4 ft/s at station 1+00 to 71.9 ft/s at station 10+49 at the design flow.

Cavitation was investigated in the spillway chute and cavitation potential was greatest at the highest flow rates, but was not found to be severe. The location with the minimum cavitation index value was at the bottom of the convex vertical curve where the convex curve transitions to a flatter slope near the downstream end of the chute. Software computes that an 11:1 chamfer ratio is needed to stop cavitation in the vicinity of isolated surface anomalies at the section with the lowest cavitation index. Air entrainment and bulking were investigated down the spillway, neither of which appear to negatively impact the design.

Topography downstream of the spillway was included in the model to analyze the right river bank impaction zone. During design flows the jet does not directly impact the right canyon wall (near vertical face composed of Dakota sandstone), spray may impact the right wall but will not have damaging forces. The jet trajectory will impact the left bank (directly in line with the spillway), it is likely that the left bank will receive massive impacts and will scour if the spillway operates at the maximum design flow for any length of time.

Introduction

Project Background

El Vado Dam is part of Reclamation’s Middle Rio Grande Project in north central New Mexico. The dam is located on the Rio Chama about 160 miles north of Albuquerque (Figure 1). The El Vado Project includes a reservoir, dam, emergency spillway, service spillway and outlet works. Original construction of the dam was completed by the Middle Rio Grande Conservancy District in 1934- 1935. Reclamation rehabilitated the service spillway, spillway gate, and original outlet works near the center of the dam, and constructed an emergency spillway in in 1954-57. In 1965-66, a new outlet works was constructed to accommodate additional water from the San Juan-Chama Project (Reclamation, 1981). In 1986, the original outlet works was restored to service to accommodate a Federal Energy Regulatory Commission licensed power plant owned and operated by Los Alamos County.

Figure 1. El Vado Dam and Middle Rio Grande Project general location map. El Vado Dam is 175 ft high and has a 1,326 ft long embankment of rolled gravel fill with a steel lining on the upstream face. The dam impounds a reservoir with a total capacity of 196,500 acre-ft. The service spillway is a 1,049 ft long steel membrane lined trapezoidal channel with a changing cross section the entire length of the spillway chute. When constructed, the spillway was designed to discharge 16,500 ft3/s with water at elevation 6907 and 20,000 ft3/s with water at the crest of the dam (Bunger, 1933). When the spillway excavation occurred severe weathering and over breakage led to the use of steel instead of concrete for the construction of the spillway (Seger, 1935). After Reclamation rehabilitated the service spillway in 1954-55, the spillway had a maximum design capacity of 17,800 ft3/s at elevation 6908.6 ft (Reclamation, 1981). Concerns with leakage and dilapidation of the steel membrane in the spillway chute have been identified and replacement of the steel lined chute with a concrete lined chute has been

proposed. Figure 2 through Figure 5 give examples of the type of issues identified with the steel membrane in the chute. An operating restriction has prevented the spillway from operating over 2,500 ft3/s since about 2007 due to concerns of spillway discharges getting behind the steel lining and eroding the foundation materials resulting in headcutting upstream toward the crest structure (see Figure 2 through Figure 5).

Figure 2. Photograph of broken vertical weld in steel lining in spillway chute.

Figure 3. Photograph of severely corroded supporting steel behind the steel plates.

Figure 4. Photograph of damaged steel lining in spillway chute.

Figure 5. Photograph of chute wall leaking, indicating that pressure is behind steel lining panels. Flow is from right to left. To prevent removal of the steel plates from the dam face, the new proposed spillway entrance will be moved approximately 20 ft into the right abutment (when looking at the upstream face of the dam). A long radius elbow from station 2+00 to approximately station 4+00 in the spillway chute will connect the new entrance with the existing linear downstream alignment (Figure 6). Designers were concerned that the long elbow may complicate the hydraulic conditions in

the spillway and potentially produce standing surface waves that could overtop designed spillway walls.

Existing spillway entrance

Proposed replacement spillway entrance and partial chute realignment

Long radius elbow

Figure 6. Aerial image of the existing and proposed (overlaid in red) spillway entrance locations and partial chute alignments.

Prototype Operation

The reservoir has two sources of water; snowmelt and rain from the Rio Chama, and flow from Heron Dam and Reservoir located on Willow Creek (Reclamation, 1981). For normal releases flows are passed through 5-ft by 9-ft outlet regulating gates, while the 36-ft-wide by 24-ft-high radial gate at the service spillway remains closed (at spillway invert, elevation 6879.2). For flood events, the 5-ft by 9-ft outlet regulating gates are opened to allow the discharge to equal the inflow to the reservoir. The spillway radial gate, which is not used to regulate flow but rather as a shutoff gate, remains closed until the reservoir inflow exceeds the 6,850 ft3/s capacity of the outlet gates and the WSE exceeds 6902.0 ft, at which point the gate is raised to allow a maximum discharge of 2,500 ft3/s per the current Standing Operating Procedures (SOP) for El Vado Dam. The reason for

the limited discharge is because of the deteriorated condition of the spillway. When the flood recedes and the reservoir water surface drops to the spillway invert elevation of 6879.2 ft, the spillway radial gate is closed and the 5-ft by 9-ft outlet gates are utilized to set the outlet releases equal to the reservoir inflow or until the releases equal the downstream requirement (El Vado Dam SOP, 2008).

Model Objectives The model study was proposed to make the following measurements and observations:

• Measure flow depths at the design flow rate of 18,600 ft3/s o along the center of the approach channel o along the left and right inlet (elliptical) guide walls o along the vertical left and right walls through the bridge and gate structure o through the center of the bridge and gate structure o along the left and right walls through the entire length of the chute • Measure flow velocities at the design flow rate of 18,600 ft3/s (also obtained at 5,000 ft3/s) o along the left, center, and right edges of the bottom of the approach channel (station -1+75 to station 0+00) o through the bridge and gate structure (station 0+00 to station 0+73.4) o at depths of 20%, 60% and 80% of the total depth where possible • Record observations regarding o to what extent, if at all, the discharge jet at the end of the spillway impacts the right canyon wall o horizontal and vertical extents of where the discharge jet impacts the river and the left river bank • Evaluate air entrainment and bulking potential of the spillway • Evaluate cavitation potential of the spillway

Model Description

Numerical Model

Numerical modeling was conducted at prototype (feasibility design) dimensions to avoid any possible scale effects and allow data to be analyzed directly from the output without requiring any additional conversions or computations. Numerical modeling focused on approach channel configurations, the bridge and gate

structure, and the first 200 ft of the chute (i.e., downstream to station 2+00). Figure 7 shows a representative block overlaid on an aerial image of the site with the approximate boundaries of the numerical models.

FLOW-3D, a commercially available computational fluid dynamics (CFD) software package developed by Flow Science Inc. was used for all CFD simulations. FLOW-3D utilizes the Reynolds averaged Navier-Stokes (RANS) equations to solve for fluid flow. Modifications to the standard RANS equations include algorithms to accurately track the water surface and flow around geometric objects (Hirt and Nichols, 1981; Hirt and Sicilian, 1985; Hirt, 1992). The location of the free surface and solid objects is tracked throughout a Cartesian gridded domain using the true volume-of-fluid (VOF) method presented by Hirt and Nichols (1981).

Models were run using a Cartesian grid with 1-ft cubic cells. Each model run used a constant upstream boundary supplying 18,600 ft3/s at a WSE of 6911.0 ft. Simulations ran until they reached a quasi-steady state condition where the water surface elevations, turbulent energy, kinetic energy, and velocities remained similar from one time step to the next.

Figure 7. Aerial image with approximate numerical model extents overlaid in red.

Physical Model Scale

In order for the model to accurately predict the performance of the prototype, the model and prototype must have similar geometric and kinematic ratios. The model for El Vado Dam and service spillway is an open channel flow model, where gravity is the dominant force; therefore, Froude scaling was chosen to establish similitude (Reclamation, 1980). The Froude number is defined as:

v F = r gd

Where, v = velocity, g = gravitational acceleration, and d = depth of flow. Geometric similitude is met when the ratios of all geometric parameters in the model and prototype are equal. Kinematic similitude is met when the fluid flow of model and prototype have velocities and accelerations at corresponding points that have similar directions and scaled magnitudes. For the 1:24 scale Froude model the geometric and kinematic scale ratios are as follows:

Geometric properties

Length ratio Lr = Lp/Lm = 24

2 Area ratio Ar = Lr = 576

3 Volume ratio Vr = Lr = 13,824

Kinematic properties

1/2 Time ratio tr = Lr = 4.9

1/2 Velocity ratio vr = Lr = 4.9

Acceleration ratio ar = 1.0

5/2 Discharge ratio Qr = Lr = 2,822

Physical Model Features

A 1:24 Froude-scale physical model which included approximately 410 ft of the dam crest length, the spillway approach channel, bridge and gate structure (composed of the guide wall section, bridge section, gate section, and the transition section), chute, and plunge pool in the downstream river channel was constructed. Approximate model extents and the model overview are shown in Figure 8 and Figure 9, respectively.

Figure 8. Aerial image with approximate physical model extents overlaid in red.

Head Box

Figure 9. Aerial image of physical model extents. The head box was elevated to allow the full length and dual curvature of the 1:24 scale spillway to be modeled. Figure 12 is a plan and section drawing, provided to the modelers by the design engineers, showing the entrance to the spillway offset 20 ft (prototype) toward the right bank, farther into the right abutment of the dam. The 1-horizontal:4-vertical trapezoidal spillway chute runs from station 1+00 to 10+49 and changes bottom width (32-ft at station 1+00 to 20-ft at station 10+49) and side slope height the entire length (see Figure 10). The spillway was

constructed out of polyvinyl chloride (PVC) and was supported on stepped platforms (Figure 11) that were secured to the laboratory floor and to the head and tail boxes. The inside face of the spillway walls were sanded with 300 grit sand paper to remove the water repellent wax coating and add slight roughness to more closely represent the prototype concrete lining. Spillway walls were installed at the 1h:4v side slope and were extended to the top of the vertical parapet walls as shown in Figure 10. Installing vertical parapet walls was problematic in the model due to the spillways complex geometry changes. The location where the sloped wall would have become a vertical parapet wall was precisely marked along both spillway walls in the model.

Figure 10. Cross-section of El Vado service spillway with the vertical parapet walls shown as designed and as built in the model.

Figure 11. Elevated PVC spillway viewed from left of the spillway.

Figure 12. Plan and section drawing of existing and proposed modified spillway alignment (in red).

The model included a tail box with topography that matched existing prototype conditions. The outlet works and its flows were not included in the model. Tail boards were used to control the tail water depth in the plunge pool (tail box). The head box included topography to match bathymetric data and current conditions at the site. A notch in the head box at the approximate elevation of the dam crest was included in the model to ensure that overtopping of the model’s head box would return to the laboratory recirculation system. Figure 13 shows the model as viewed from upstream of the reservoir.

Tail Box, Downstream River

Model Emergency Spillway Chute Spillway

Spillway Entrance

Figure 13. Model viewed from upstream of the reservoir.

The bridge and gate section of the spillway (station 0+00 to 0+73.4) is a 36-ft- wide (prototype) rectangular channel which was constructed in the model of High Density Overlay forming plywood with water resistant epoxy outer plies. The PVC spillway gate modeled the prototype gate curvature and was mounted to the sides of the channel with a single trunnion and corbel on each side that were located in approximately the same position as the current gate rotates. The transition pieces between the rectangular channel and the trapezoidal channel (station 0+73.4 to 0+90.7) of the spillway chute were machined using high density foam and were made to match the existing transitions. Figure 14 and Figure 15 provide views of the bridge, gate and transition sections with and without the radial gate installed. The approximate gate seal path is also shown.

Bridge Section Gate Seal Path

Gate Gate Section Attachment Location

Transition Section

Figure 14. Bridge, gate, and transition sections of the spillway with no gate installed. Flow is from top-left to bottom-right of image. View is from downstream of the reservoir.

Gate in Raised Position

Gate Corbel

Figure 15. Bridge, gate, and transition sections of the spillway with gate installed. Flow is from top to bottom of image. View is from downstream of the reservoir.

Physical Model Instrumentation

Water was supplied to the model from a 250,000 gallon sump reservoir through the laboratory pumping system by a 12-in centrifugal pump. Flow to the model was measured with the laboratory venturi meters. A 678 ft3 and 44,000 pound volumetric/weigh tank was used to calibrate the laboratory venturi meters at regular intervals to an accuracy of ± 0.25%. Water surface elevations in the

reservoir approach channel were measured using pressure taps installed along the centerline of the spillway approach at 25-ft (prototype) increments beginning at station 0+00 (tap 1) on the spillway invert to station -1+75 (tap 8). The taps were connected to a manometer board. Capillary rise in the manometer board led to uncertainty in these measurements; therefore, approach water surface elevations from the manometer board are intended to be used only to establish a relative change of water surface elevations from one tap to another. A pressure tap connected to a 12-in diameter stilling well equipped with a Lory point gauge, accurate to 0.001-ft, was utilized to measure absolute reservoir WSE. The location of the spillway approach pressure taps (Figure 16) are labeled 1 through 8 for station 0+00 through station -1+75 and the pressure tap in the reservoir is labeled 9.

8 7 6 1 5 4 3 2

Figure 16. Aerial photograph with pressure port locations in the approach channel to the service spillway entrance and the reservoir. Velocities in the spillway approach channel were measured at the flow rates of 5,000 ft3/s and at 18,600 ft3/s using a SonTek Flow Tracker handheld Acoustic Doppler Velocimeter (ADV®) to an accuracy of ±1.0% for single measurements. These velocities were then visually compared to velocity contour plots produced by the FLOW-3D numerical models. Velocities in the spillway chute were determined by measuring the water surface on the spillway walls and dividing the known flow rate by the calculated cross-sectional area to determine average flow velocity at each station marker. Water surface measurements were visually averaged and calculated velocities are intended to be used to estimate the range of velocities within the spillway.

Model Test Plan FLOW-3D numerical models were developed and run to a quasi-steady state for multiple approach channel configurations. Results of the numerical models were compared visually analyzing velocities, wave heights and wave locations from the upstream reservoir through station 2+00 of the chute. Screen shots of the simulations were collected and compared side-by-side to determine the design most likely to meet requirements. The best design as determined by the modelers and design team was tested in the physical model.

Each configuration of the physical model was tested by introducing a known flow rate into the model, allowing the model to stabilize then record the water surface elevations in the reservoir and entrance channel and take any relevant photographs and video documenting hydraulic conditions in the reservoir and spillway. Each introduced flow was subject to the same operating conditions having the radial gate fully open 24-ft (or removed). Each configuration of the physical model had unique characteristics and was tested until it was found to not meet one or more of the design criteria. Typical results for any configuration included: a rating curve, water surface profiles in the spillway bridge and gate section up to station 0+70, and video documentation of separations, eddies and or standing waves.

Results and Discussion

Numerical Model

Numerical models were developed and run, using FLOW-3D, to determine an optimum approach channel design. Figure 17 provides a plan view drawing summary of all the numerical models that were developed. For each configuration, the slope of the right bank in the approach channel was vertical unless otherwise noted. The approach channel slope was hydraulically adverse, except for configuration D (where the approach channel was flat). The approach channel slope, for all configurations, began at station 0+00 (center of spillway entrance), and remained 18-ft away from the toe of the right bank regardless of how wide the approach slab extended.

Configurations “A” through “I” were developed leaving the alignment of New Mexico State Highway (Hwy) 112 in its current location. The long and tall vertical concrete wall of the right bank approach for configurations “A” through “I” would require removal of Hwy 112 during construction to allow footings to be installed; however, vertical walls would allow the highway to be re-constructed to the original alignment. During testing, concerns were identified with the potentially cost prohibitive excavation, material, and construction costs associated

with the long vertical right bank approach wall. Design engineers compared study level costs for the long and tall right bank approach wall with the realignment of NM Hwy 112 and determined the roadway realignment to be a more cost effective solution. Configuration “J” was then modeled which requires that highway 112 be re-aligned due to laying back the right bank slope. The slope of the right bank approach was determined by design engineers to be an acceptable slope for the right bank limit of excavation and would not require additional foundation or surface armoring concrete. Figure 18 and Figure 19 provide a sample from configuration “F” of two views that were utilized for visual comparison. Images for all configurations tested are provided in appendix A.

Figure 17. Schematic of approach channel alignment configurations investigated in the numerical (A-J) and physical (F, I, J) models.

FLOW

Figure 18. Plan view with velocity contours for visualization (configuration “F”).

FLOW

Figure 19. Configuration “F” velocity contour plot looking up the spillway to visualize wave actions occurring in the spillway chute downstream.

Physical Model

The physical model was constructed with the approach channel configuration “F” to identify potential improvements to the preferred feasibility design and verify hydraulic performance. Approach configuration “F” was selected based on the

initial numerical modeling results and included a 300-ft-long vertical concrete approach wall on the right bank and a 2% sloped entrance slab. After testing of configuration “F”, the model was modified (as described in a later section) into configuration “J”.

Spillway Entrance Guide Walls The physical model was constructed such that different entrance guide walls could be quickly installed and evaluated. Figure 20 describes the different entrance guide walls that were evaluated in the physical models. The following left guide wall configurations were tested with configuration “F” in the physical model: none, a 30-ft radius curve, a 120-ft by 60-ft elliptical, and an 80-ft by 40-ft elliptical. Configuration “J” was only tested with the 60-ft x 40-ft rotated elliptical guide walls.

Figure 20. Entrance transition wall geometries that were included in the physical model (not all configurations were tested with every entrance guide wall).

Configuration “F’ with no guide walls Without a left guide wall substantial separation and eddy formation at the leading edge of the left vertical spillway wall caused a large draw down on the left side of the spillway entrance. A severely sloped water surface across the width of the entrance to the spillway existed (Figure 21). At flow rates of 16,000 ft3/s and higher, the flow through the gate section came into contact with the gate trunnion pin (Figure 22) and caused turbulent flow that moved diagonally from the left to right and overtopped the spillway on the right wall just downstream of the transition from vertical to sloped walls at approximately station 1+00.

20-foot spillway entrance offset distance Vertical concrete right approach wall

Figure 21. Approach channel configuration “F” with no left guide wall at flow rate of 18,600 ft3/s, viewed from upstream of the spillway entrance.

Figure 22. Approach channel configuration “F” with no left guide wall at flow rate of 18,600 ft3/s, viewed from downstream of the gate section. Flow contacting gate trunnion pin.

Configuration “F” with a left side 30-ft radius guide wall The sharp leading edge of the 30-ft radius guide wall (Figure 23) produced separation and eddying that created a draw down along the left side of the spillway entrance and sloped water surface across the width of the spillway entrance. A standing wave formed at the downstream edge of the left guide wall (station 0+00) which ran diagonally across the spillway gate section and rode up the right wall and nearly overtopped the spillway at station 1+00. The wave then

reflected off of the right wall and propagated downstream in the spillway chute forming a complex series of cross waves. At flow rates of 16,000 ft3/s and greater, the flowing water came into contact with the gate trunnion pin (Figure 24) which backed up the water surface through the gate section causing the flow condition under the gate to oscillate between free surface and orifice flow. With the gate removed, the spillway was able to pass the maximum design flow rate, 18,600 ft3/s while maintaining a reservoir WSE of 6910.64 ft, which is below the design maximum reservoir WSE of 6911.0 ft. The water surface was above the line representing the transition from sloped to vertical parapets at eight locations along the length of the chute and showed evidence of periodic overtopping of the left parapet wall between station 2+50 to 2+85, either by splashing or riding up the left wall.

Vertical concrete right approach wall

Figure 23. Approach channel configuration “F” with 30-ft circular left guide wall at flow rate of 18,600 ft3/s, viewed from upstream of the spillway entrance. Shown with gate removed.

Figure 24. Approach channel configuration “F” with 30-ft circular left guide wall at flow rate of 18,600 ft3/s, viewed from downstream of gate section. Flow submerging gate trunnion pin.

Configuration “F” with a left side 120-ft by 60-ft elliptical guide wall The 120-ft by 60-ft elliptical left guide wall (Figure 25) provided substantially less separation and eddy formation at the leading edge of the guide wall. The water surface across the width of the spillway entrance was nearly level with only a slight slope rising from the left to right wall. The stream lines just upstream of the entrance gently converged toward the center of the spillway as observed by dye injected at various depths. Dye also showed that the flow was generally linear in nature without much crossing through the spillway entrance. Diagonal standing waves formed at the transitions from vertical to sloped walls on both sides of the spillway. These waves crossed nearly in the middle of the spillway resulting in a pattern of diamond shaped cross waves (also called shock waves), which reflected from wall to center down the length of the spillway chute.

At flow rates of 16,000 ft3/s and greater the flow through the gate section contacted the gate trunnion pin and caused the flow condition beneath the gate to oscillate between free surface flow (Figure 26) and orifice flow (Figure 27) conditions. With the gate removed, the spillway was able to pass the maximum design flow rate of 18,600 ft3/s while maintaining a reservoir WSE of 6910.73 ft, which is below the design maximum WSE of 6911.0 ft. The water surface was approximately 1/2 inch (model) or 1 ft (prototype) above the line representing the transition from sloped to vertical parapets at both station 2+81 and station 5+54. There was no indication of overtopping the spillway parapet walls.

Vertical concrete right approach wall

Figure 25. Approach channel configuration “F” with elliptical, 120-ft by 60-ft, left guide wall at flow rate of 18,600 ft3/s, viewed from upstream of the spillway entrance. Shown with gate removed.

Figure 26. Approach channel configuration “F” with elliptical, 120-ft by 60-ft, left guide wall at flow rate of 18,600 ft3/s, viewed from downstream of gate section. Gate trunnion pin not submerged, operating in free surface flow condition.

Figure 27. Approach channel configuration “F” with elliptical, 120-ft by 60-ft, left guide wall at flow rate of 18,600 ft3/s, viewed from downstream of gate section. Gate trunnion pin submerged, operating in orifice flow condition.

Configuration “F” with a left side 80-ft by 40-ft elliptical guide wall An 80-ft by 40-ft left elliptical guide wall was installed, and the performance was observed to be similar to the larger elliptical guide wall with minimal separation and eddy formation at the leading edge of the wall and very little crossing of the streamlines through the spillway. At the maximum design flow 18,600 ft3/s the flow through the gate section contacted the gate trunnion pin causing similar open channel to orifice oscillating flow conditions. With the gate removed, the spillway was able to pass the maximum design flow rate, 18,600 ft3/s while maintaining a reservoir WSE of 6910.37 ft, which is below the maximum design WSE of 6911.0 ft.

Transition Section between Vertical and Sloped Spillway Walls The transitions in the section between the vertical gate section and sloped spillway chute walls (station 0+73.4 to station 0+90.7) were redesigned in an effort to reduce the magnitude of shock waves created by the transitions. The overall length of the transitions were increased by extending the downstream catch line to station 1+00. The lengthened transitions performed well and were successful in reducing the magnitude of the shock (diagonal standing) waves that propagated downstream the length of the spillway chute (Figure 28).

Figure 28. Comparison of original 17-ft 3-in (bottom) and 24-ft 6-in lengthened transitions (top) of the spillway.

Final Physical Model Modifications (Configuration “J”) The following modifications were implemented to improve the hydraulic performance and potentially reduce excavation, materials, and construction costs (configuration “J”). Actual estimates for the reduction of the excavation, material and construction costs go beyond the scope of this physical model study:

• Reshaping the right bank approach by changing the slope from a vertical concrete face to an unarmored 1:1 slope. • Installation of 60-ft X 40-ft elliptical guide walls on both sides of the spillway entrance. • Elliptical guide walls were attached tangent to the gate section walls with a major axis rotated 34º from the centerline of the spillway. • Approach slab slope was reduced from 2% to 1% from the invert of the spillway entrance back to the catch line of the slab and reservoir intersection. • Extend the transition walls from 17-ft 3-in to 24-ft 6-in.

Configuration “J” with 60-ft by 40-ft rotated elliptical guide walls After modification, minimal separation and eddy formation occurred at the leading edge of the elliptical guide walls (Figure 29). Injected dye showed the approach channel streamlines converging to the center of the spillway entrance without noticeable crossing. Streamlines down the spillway were almost linear.

The water surface profile at the spillway entrance was nearly symmetrical across its width with a small standing wave in the center of the entrance at approximately station 0+00 as a result of the converging approach flow. Diagonal shock waves occurred at both of the bridge and gate section transitions; from the elliptical guide walls to the vertical bridge section walls and from the vertical gate section walls to the sloped chute walls. The shock waves crossed in the center of the spillway at approximately station 0+75 and station 1+85, respectively and set up a diamond shaped cross wave pattern down the length of the spillway chute as the waves reflected back and forth from wall to center. Testing was conducted with no radial gate installed. At flow rates of 16,000 ft3/s and greater the flow through the gate section would have contacted the gate trunnion pin and caused the flow condition beneath the gate to oscillate between free surface flow and orifice flow. With the gate removed, the spillway was able to pass the maximum design flow rate of 18,600 ft3/s while maintaining a reservoir WSE of 6910.54 ft, below the design maximum reservoir WSE of 6911.0 ft.

At the maximum design flow rate the water surface exceeded the transition line between the sloped channel and vertical parapet spillway walls at numerous locations on both the right and left spillway walls. The case with the least available freeboard occurred on the left spillway wall, between station 1+00 and 1+50. The water surface exceeded the transition line by 1.96 ft and the height of the parapet wall in that location was 3.23 ft, leaving approximately 1.25 ft of freeboard.

1:1 sloped right approach wall

Figure 29. Approach channel configuration “J” with elliptical left and right guide walls at flow rate 18,600 ft3/s, viewed from upstream of the spillway entrance.

Downstream River Plunge Pool During testing, observations were made of the jet entering the plunge pool in the downstream river (Figure 30 and Figure 31). For all flow rates tested, 5,000 ft3/s-

18,600 ft3/s, there was no impact of the jet on the right canyon wall. Spray may contact the right wall but likely without damaging force. At spillway flow rates of 7,500 ft3/s and greater, the jet will impact the left bank of the river with likely damaging scouring forces. Flows entering the downstream river channel were extremely turbulent and ran up the left river bank. In order to accurately predict impact pressures and velocity patterns, the tailbox of the model needed to be much larger (more than space was available to construct) and include flows for the outlet works of the power plant. Due to the space not being available and the additional costs associated with adding the outlet works flows, modelers determined that visual inspection of impaction location was all that could be provided. Visual velocity tendencies along the right bank were not alarming to the modelers or designers. It is also believed that if the outlet flows were present in the model the velocities on the right bank would not be troublesome, but these were not measured or tested.

Figure 30. Jet plunging into the plunge pool in the downstream river channel at flow rate 18,600 ft3/s, viewed from downstream of the spillway. Note how the flow jet does not impact the right bank (canyon wall).

Figure 31. Jet plunging into the downstream river plunge pool at spillway flow rate of 18,600 ft3/s. Viewed from right bank.

Approach Channel Velocities Velocity measurements in the approach channel to the spillway entrance were made at flow rates of 5,000 and 18,600 ft3/s for approach channel configuration “J”. Velocities were measured at three locations; along the approach channel centerline and offset 12 ft (prototype) to both the left and right side, at 25 ft increments starting at station 0+00 and proceeded upstream to station -1+75. Measured model and prototype velocities are provided in Table 1 (flow rate 5,000 ft3/s at 60% flow depth) and Table 2 through Table 4 (flow rate of 18,600 ft3/s at 20%, 60% and 80% flow depths).

Physical measurements compared well with velocity contours of the same depth and flows calculated by the computational FLOW-3D model shown in Figure 32 through Figure 34. Of particular interest to design engineers were the velocities along the 1:1 unarmored sloped right bank of the approach channel. Velocities were measured at both the lower flow 5,000 ft3/s and design flow 18,600 ft3/s to identify the range of velocities present to ensure slope stability and limit scour potential. At the maximum design flow rate 18,600 ft3/s, velocity measurements in the upstream right bank approach channel show that velocities increased from 3.25 ft/s (station -1+75) to 8.5 ft/s (station -0+50) and then from 13.81 ft/s (station -0+25) to 19.98 ft/s (station 0+00) at the spillway entrance. No velocity exceeded 20 ft/s along the right bank. Velocity measurements made at 5,000 ft3/s increased to 12.54 ft/s (right bank at station 0+00) as the flow accelerated to the spillway entrance.

Table 1. Approach channel velocities at 60% depth for flow rate of 5,000 ft3/s. Velocity (ft/s) at 60% Flow Depth and 5,000 ft3/s Station Left Center Right Model Prototype Model Prototype Model Prototype -1+75 0.40 1.95 0.40 1.94 0.41 2.01 -1+50 0.51 2.50 0.49 2.42 0.49 2.39 -1+25 0.62 3.02 0.58 2.84 0.55 2.71 -1+00 0.74 3.61 0.69 3.36 0.64 3.11 -0+75 0.99 4.86 0.91 4.48 0.82 4.02 -0+50 1.45 7.09 1.31 6.43 1.17 5.75 -0+25 2.28 11.17 2.09 10.24 1.91 9.37 0+00 2.92 14.30 2.87 14.06 2.56 12.54

Table 2. Approach channel velocities at 20% depth for flow rate of 18,600 ft3/s. Velocity (ft/s) at 20% Flow Depth and 18,600 ft3/s Station Left Center Right Model Prototype Model Prototype Model Prototype -1+75 0.63 3.08 0.64 3.12 0.65 3.21 -1+50 0.74 3.64 0.76 3.74 0.75 3.69 -1+25 0.92 4.50 0.89 4.34 0.86 4.20 -1+00 1.08 5.29 1.03 5.04 0.96 4.72 -0+75 1.44 7.06 1.32 6.49 1.20 5.87 -0+50 2.15 10.52 1.86 9.11 1.56 7.65 -0+25 3.25 15.93 2.89 14.15 2.76 13.54 0+00 4.07 19.95 4.00 19.60 3.60 17.61

0+25 - St. 0+00St. St. St.

Figure 32. Velocity contours in spillway approach channel at 20% depth for flow rate of 18,600 ft3/s, computed using FLOW-3D.

Table 3. Approach channel velocities at 60% depth for flow rate of 18,600 ft3/s. Velocity (ft/s) at 60% Flow Depth and 18,600 ft3/s Station Left Center Right Model Prototype Model Prototype Model Prototype -1+75 0.62 3.05 0.65 3.17 0.65 3.19 -1+50 0.75 3.67 0.75 3.66 0.76 3.71 -1+25 0.92 4.51 0.87 4.26 0.78 3.81 -1+00 1.04 5.09 1.01 4.96 0.95 4.67 -0+75 1.44 7.05 1.28 6.29 1.21 5.94 -0+50 2.05 10.06 1.85 9.08 1.56 7.64 -0+25 3.51 17.20 3.08 15.10 2.81 13.77 0+00 NA NA 4.33 21.20 3.92 19.23

0+25 - St. 0+00St. St. St.

Figure 33. Velocity contours in the spillway approach channel at 60% depth for flow rate of 18,600 ft3/s computed using FLOW-3D.

Table 4. Approach channel velocities at 80% depth for flow rate of 18,600 ft3/s. Velocity (ft/s) at 80% Flow Depth and 18,600 ft3/s Station Left Center Right Model Prototype Model Prototype Model Prototype -1+75 0.63 3.08 0.65 3.19 0.66 3.24 -1+50 0.76 3.71 0.76 3.74 0.77 3.75 -1+25 0.88 4.31 0.88 4.31 0.86 4.20 -1+00 0.82 4.04 1.02 5.00 0.95 4.67 -0+75 1.55 7.58 1.30 6.38 1.22 5.96 -0+50 2.17 10.63 1.90 9.31 1.73 8.49 -0+25 3.41 16.72 3.18 15.60 2.82 13.81 0+00 NA NA 4.64 22.71 4.08 19.98

0+25 - St. 0+00St. St. St.

Figure 34. Velocity profile, computed using FLOW-3D, in spillway approach channel at 80% depth for flow rate of 18,600 ft3/s.

Spillway Entrance Water Surface Profiles Water surface profiles through the bridge and gate sections are provided for entrance configuration “J” in Figure 35 for the left wall, Figure 36 for the right wall and Figure 37 and Figure 38 for the center of the spillway. Table 5 provides tabulated data for Figure 35 through Figure 38 and should be used to design the gate such that it does not contact the water surface at the design flow rate of 18,600 ft3/s.

Figure 35. Left wall water surface profile through the guide wall, bridge and gate sections of the spillway. Water surface profile shown in blue.

Figure 36. Right wall water surface profile through the guide wall, bridge and gate sections of the spillway. Water surface profile shown in blue.

Figure 37. Center of approach channel water surface profile. Water surface profile shown in blue. Dashed blue line represents data points acquired using the pressure ports and manometer board.

Figure 38. Center of channel water surface profile through the guide wall, bridge and gate sections of the spillway. Water surface profile shown in blue. Dashed blue line represents data points acquired using the pressure ports and manometer board.

Table 5. Water surface elevations through the approach channel, bridge and gate sections of the spillway. Water Surface Elevations (ft) at 18,600 ft3/s (refer to Figure 16 for tap locations) Station Left Center Right -1+75 (tap 8) NA 6910.48 NA -1+50 (tap 7) NA 6910.48 NA -1+25 (tap 6) NA 6910.48 NA -1+00 (tap 5) NA 6910.36 NA

-0+75 (tap 4) NA 6910.12 NA -0+55 6908.8 NA 6910.6 -0+50 (tap 3) 6908.9 6909.88 6910.8 -0+45 6908.7 NA 6910.7 -0+40 6908.5 NA 6910.7 -0+35 6908.5 NA 6909.5

-0+30 6908.1 NA 6908.7 Channel Approach -0+25 (tap 2) 6907.8 6907.72 6908.7 -0+20 6906.4 NA 6907.8 -0+15 6905.5 NA 6906.7 -0+10 6904.0 NA 6905.3 -0+05 6903.0 NA 6903.6

0+00 (tap 1) NA 6904.84 NA 0+05 6900.9 6902.7 6901.3 0+10 6900.8 6902.0 6900.9 0+15 6900.6 6899.3 6900.4 0+20 6900.5 6898.7 6899.9 0+25 6899.8 6898.7 6899.5 Bridge Section 0+30 6899.1 6898.9 6899.0 0+35 6898.3 6898.2 6898.7

0+40 6897.9 6897.8 6898.2 0+45 6897.3 6897.3 6897.6 0+50 6896.8 6897.4 6897.1 0+55 6896.7 6897.6 6896.9 0+60 6896.5 6897.6 6896.8 Gate Section 0+65 6896.3 6897.8 6896.6 0+70 6896.2 6898.3 6896.5

Spillway Chute Water Surface Profiles and Velocities Water surface profiles at the maximum design flow of 18,600 ft3/s along the right and left walls of the spillway chute were documented for configuration “J”. Figure 39 provides an overview of the water surface along the right and left spillway chute walls with the locations and maximum water height that extends into the vertical parapet walls. Figure 40 through Figure 45 plot the water surface and freeboard along each wall at 175-ft intervals. Water surface measurements

are conservatively high, being obtained by measuring the maximum water height realized along the chute walls including any instantaneous bounce. Video footage was used to note that water surfaces consistently entered the vertical parapet walls at station 1+12 on both walls and at station 2+81 on the left wall.

Average velocities were calculated using the cross sectional area of flow at various stations along the chute. Velocities in the spillway chute ranged from 27.4 ft/s at station 1+00 (start of chute) to 71.9 ft/s at station 10+49 (end of chute). These calculated velocities visually compared well with velocity contours created using FLOW-3D of the spillway chute at 18,600 ft3/s. Appendix B contains tabulated data of both water surface heights and velocities down the spillway chute.

Figure 39. Water surface profiles (blue line) on left and right spillway chute walls with station numbers. Height of water surface above top of sloped chute walls (blue text), and height of vertical parapet walls (black text) are also shown at various locations.

Figure 40. Spillway water surface profiles (blue line) from station 1+00 to 2+75. Maximum height of the water surface above sloped spillway chute wall (blue text), and height of vertical parapet wall (black text) at various locations.

Figure 41. Spillway water surface profiles (blue line) from station 2+75 to 4+50. Maximum height of the water surface above sloped spillway chute wall (blue text), and height of vertical parapet wall (black text) at various locations.

Figure 42. Spillway water surface profiles (blue line) from station 4+50 to 6+25. Maximum height of the water surface above sloped spillway chute wall (blue text), and height of vertical parapet wall (black text) at various locations.

Figure 43. Spillway water surface profiles (blue line) from station 6+25 to 8+00. Maximum height of the water surface above sloped spillway chute wall (blue text), and height of vertical parapet wall (black text) at various locations.

Figure 44. Spillway water surface profiles (blue line) from station 8+00 to 9+75. Maximum height of the water surface above sloped spillway chute wall (blue text), and height of vertical parapet wall (black text) at various locations.

Figure 45. Spillway water surface profiles (blue line) from station 9+75 to 10+48.60. Maximum height of the water surface above sloped spillway chute wall (blue text), and height of vertical parapet wall (black text) at various locations.

Rating Curves Spillway rating curves were developed for four entrance configurations; configuration “F” with no entrance guide wall, configuration “F” with the 30-ft radius left guide wall, configuration “F” with the 120-ft by 60-ft elliptical left guide wall, and for configuration “J” with 60-ft by 40-ft elliptical left and right guide walls. Tabular rating curves are included in Table 6. The plotted rating curves for all four configurations are included in Figure 46. With the exception of configuration “F” with no entrance guide walls, all configurations could pass the design flow of 18,600 ft3/s without exceeding the design reservoir WSE of 6911.0 ft.

Design engineers selected configuration “J” as the preferred option. Figure 47 provides a plot of configuration “J” and includes a best fit polynomial trend line and equation. For configuration “J”, the inflow to the model was increased beyond the maximum design flow to determine the approximate inflow that could potentially overtop the dam and the spillway chute walls with the radial gate fully open. It is estimated that the spillway will be able to pass approximately 22,000 ft3/s before overtopping the dam. It is estimated that flows up to 22,500 ft3/s can pass through the spillway before the parapet walls are overtopped at station 2+75 at the left wall. When this occurred the reservoir water surface elevation was 6914.89 ft.

Table 6. Reservoir inflow rates and corresponding reservoir water surface elevations for various spillway entrance configurations tested in the physical model. Physical Model Rating Curves for Four Entrance Configurations Approach Channel Option Option Option Option J (preferred Configuration F F F by design team) Guide Wall Location None Left Left Left & Right 120-ft Guide Wall Size None 30-ft by 60-ft 60-ft by 40-ft Guide Wall Shape None Radius Ellipse Ellipse Spillway Flow Rate Reservoir Water Surface Elevation (ft) (ft3/s) 5000 6892.84 6892.37 6892.61 6892.46 7500 6897.04 6896.21 6896.50 6896.33 10000 6900.83 6899.81 6900.01 6899.90 12000 6903.83 6902.48 6902.65 6902.53 14000 6906.78 6905.00 6905.13 6905.06 16000 6909.81 6907.69 6907.73 6907.67 17000 6911.13 6908.81 6908.90 6908.93 18000 6912.40 6909.99 6910.04 6910.06 18600 6913.17 6910.64 6910.73 6910.54 22000 NA NA NA 6914.23 22500 NA NA NA 6914.89 (1) 23000 NA NA NA 6915.39 (2) (1) Dam overtopping (2) Spillway overtopping at left wall

6915

6910

6905

6900 Reservoir WSE (ft)

6895

6890 0 5000 10000 15000 20000 Reservoir inflow rate (ft3/s)

No Entrance Wall 30-ft radius left guide wall 60-ft by 30-ft elliptical left guide wall Option J two elliptical guide walls

Figure 46. Rating curve for four entrance configurations tested in the physical model. Configuration “F” with no guide wall, with 30-ft radius left guide wall, with 120-ft by 60-ft elliptical left guide wall, and configuration “J” with left and right 60-ft by 40-ft elliptical guide walls.

Figure 47. Rating curve (with polynomial trend line) for service spillway at El Vado Dam with entrance configuration “J”.

Cavitation and Air Bulking The potential for cavitation and aeration bulking of the flow in the spillway chute was evaluated using software developed at Reclamation in connection with Engineering Monograph No. 42, Cavitation in Chutes and Spillways (Falvey 1990). That monograph included the FORTRAN program HFWS, which has since been converted to an Excel spreadsheet. The spreadsheet computes water surface profiles and associated cavitation index values at defined sections along the length of the spillway. The cavitation index, σ, is computed as:

= 2 𝑃𝑃𝑣𝑣 −𝑃𝑃 𝜎𝜎 2 where 𝜌𝜌𝑉𝑉 ⁄

P = pressure at a flow surface; V = flow velocity; Pv = vapor pressure of water; ρ = fluid density.

Low values of the cavitation index indicate that the pressure in the flow is approaching the vapor pressure threshold at which cavitation bubbles begin to form. The dimensionless index expresses the pressure difference as a multiple of the dynamic velocity of the flow. Due to the effects of turbulent pressure fluctuations, cavitation bubbles can form before σ drops to a value of zero, especially when the flow encounters unusual surface features, such as offsets into or away from the flow. Cavitation typically starts to become a concern in spillways for values of σ < 0.5. In addition to computing cavitation index values, the software determines a damage potential factor that considers the cavitation index value, the flow velocity, and the characteristics of surface anomalies of different sizes and shapes. The software also suggests chamfer angle requirements to be specified for construction to reduce the likelihood of cavitation from occurring.

An additional output from the software is a prediction of air entrainment rates for spillway chutes. The amount of air entrained is a complex function of the chute slope and the Froude and Weber numbers of the flow. The Weber number accounts for the effects of surface tension on the air entrainment process.

Flow profiles were computed for the proposed spillway design at several different flow rates ranging from the design flow rate of 18,600 ft3/s down to 500 ft3/s. Cavitation potential is greatest at the highest flow rates, but is not severe, with a minimum cavitation index value of σ = 0.36. For the 18,600 ft3/s discharge, this occurs at the bottom of the convex vertical curve (station 9+65.71 ft) where it meets the concave curve that begins to bring the chute back to a flatter slope before the flip bucket. Damage potential values for 1-inch flow offsets are in the range of 40 at this location. Typically, significant damage has been observed in Reclamation spillways when σ < 0.2 and/or when the damage potential is 500 or

more (Falvey 1990). Thus, this spillway design does not require extraordinary aeration ramps or other features to promote air entrainment; cavitation potential can be mitigated with the use of appropriate construction tolerances. The software computes that an 11:1 chamfer ratio is needed to stop cavitation in the vicinity of isolated surface anomalies at the section with the lowest cavitation index.

Air bulking potential of the spillway chute flow was not found to be a concern. Air entrainment is not expected until the predicted thickness of the boundary layer becomes approximately equal to the flow depth in the spillway. At the design flow rate of 18,600 ft3/s the flow depth is about 15 to 17 ft in the upper reaches of the chute and 10.5 to 11 ft in the downstream reaches of the chute. In contrast, the boundary layer thickness only reaches about 6.5 ft at the downstream end of the chute. The boundary layer thickness and flow depths become approximately equal when the discharge is reduced to about 8,000 ft3/s. However, at discharges from 8,000 down to 4,000 ft3/s the software still computes no significant concentration of entrained air, since the values of the Froude number, Weber number and chute slope place the flow into a zone where air entrainment rates are low according to the mathematical models described in Engineering Monograph 42. At flow rates below 2,000 ft3/s there begins to be significant air entrainment computed in the downstream portions of the chute, but at such low discharges there is no danger for bulked flow depths to exceed the chute wall heights.

Another approach to predicting aeration in the spillway chute is based on the work of Wilhelms and Gulliver (2005) who reanalyzed laboratory data collected by others in the 1950s and 1960s to develop the following relation between the chute slope and the mean steady state air concentration at equilibrium:

= 0.656 1 . ( . ) −0 0356 𝜃𝜃−10 9 where θ is specified in degrees.𝐶𝐶��𝑒𝑒∞�� According� − 𝑒𝑒 to this relation,� at the steepest section of the spillway chute where the slope is approximately 17.9°, the value of =0.14. Since aerated flow is not expected until flow rates drop to or below 8,000 ft3/s, this air concentration still leads to bulked flow depths at these flow 𝑒𝑒∞ rates𝐶𝐶�� that are well below the clear-water flow depths associated with non-aerated flow at the design flow rate of 18,600 ft3/s.

Conclusions Numerical and physical model test results determined configuration “F” with a left elliptical entrance channel guide wall to be a viable design for the offset spillway entrance that satisfied all design criteria. The 300-ft-long approach channel, with a vertical concrete right wall, provided favorable approach flow conditions through the entrance and down the length of the spillway chute. Viewing the physical model allowed design engineers to visualize and make

estimates of excavation and construction costs associated with the 300-ft-long vertical concrete right approach wall. Comparison of those study level estimates with the cost of New Mexico State Highway 112 re-alignment led to the development of entrance configuration “J”, which included a re-aligned roadway and a 1:1 sloped unarmored right bank. Testing on configuration “F” also led to lengthening the transitions from the vertical walls in the spillway gate section to the sloped walls of the spillway chute, as well as identifying the need to raise the gate trunnion corbels in the new radial gate to positions that will not be contacted by the water surface at the maximum design flow rate 18,600 ft3/s. Should it be determined that NM State Highway 112 be reconstructed in the existing location, then configuration “F” is a viable design option that satisfies all design criteria.

The post modification results of approach channel configuration “J” with left and right elliptical guide walls demonstrate that it is also a feasible design for the spillway replacement. Design engineers preferred configuration “J” stating that it will likely represent significant cost savings when compared to the approach channel with the long and tall vertical concrete right wall of configuration “F”. No cost figures were provided to modelers. Configuration “J” was able to pass the maximum design flow rate 18,600 ft3/s without exceeding the maximum design reservoir WSE of 6911.0 ft and without overtopping the spillway walls.

At the maximum design flow rate 18,600 ft3/s, the water surface profiles along each entrance elliptical wall, through the bridge and spillway gate sections (station –0+55 to station 0+70) and down the length of the spillway chute (station 1+00 to station 10+48.60) were determined and recorded. At the design flow rate, velocities in the chute range from 27.4 ft/s at station 1+00 to 71.9 ft/s at station 10+49. The maximum velocity along the right bank in the upstream approach channel was 20 ft/s at station 0+00. Should it be determined that NM State Hwy 112 be realigned, then configuration “J” is a viable design option, that may provide cost savings, which satisfies all design criteria.

For any configuration selected, a new radial gate is to be designed and constructed that features gate supports (trunnion location and corbels) that will be located above the flow at the maximum design flow 18,600 ft3/s. Water surface profiles are provided in the body of the report to aid in the gate design.

Cavitation index value at the worst location is calculated at 0.36, which is greater than the 0.20 at which damage typically occurs. This spillway design does not require extraordinary aeration ramps or other features to promote air entrainment; cavitation potential can be mitigated with the use of appropriate construction tolerances. The software computes that an 11:1 chamfer ratio is needed to stop cavitation in the vicinity of isolated surface anomalies at the section with the lowest cavitation index.

Air entrainment and bulking were investigated down the spillway chute, neither of which appear to negatively impact the design. The flow with the greatest potential for significant bulking is a low flow when ample freeboard is available.

During testing, observations were made of the jet entering the plunge pool in the downstream river. There was no impact of the jet on the right canyon wall, spray may contact the right canyon wall but without enough force to scour the Dakota sandstone. The jet will likely impact the left river bank with potentially scouring forces when the spillway is operated at the maximum design flow rate.

References Bunger, H. P., (1933). “Metal-Faced Gravel Dam Being Built in New Mexico,” Engineering News-Record, vol. 111, p. 505.

Falvey, Henry T. (1990). “Cavitation in Chutes and Spillways”. Engineering Monograph No. 42, United States Printing Office.

Hirt, C. W., and Nichols, B. D. (1981). “Volume of fluid (VOF) method for the dynamics of free boundaries.” J. Comput. Phys., 39, 201–225.

Hirt, C. W., and Sicilian, J. M. (1985). “A porosity technique for the definition of obstacles in rectangular cell meshes.” Proc. Fourth International Conf. Ship Hydro., National Academy of Science, Washington, DC.

Hirt, C. W. (1992). “Volume-fraction techniques: Powerful tools for flow modeling.” Flow Science Rep. No. FSI-92-00-02, Flow Science, Inc., Santa Fe, N.M.

Reclamation, Bureau of (1980). Hydraulic Laboratory Techniques, U.S. Department of the Interior.

Reclamation, Bureau of (1981). Project Data, U.S. Department of the Interior.

Reclamation, Bureau of (2008). El Vado SOP (10-2008), U.S. Department of the Interior.

Reclamation, Bureau of (2010). Heron SOP (10-2010), U.S. Department of the Interior.

Seger, C. P., (1935). “Steel Used Extensively in Building El Vado Dam,” Engineering News-Record, vol. 115, p. 211.

Wilhelms, Steven C. and Gulliver, John S. (2005) Bubbles and waves description of self-aerated spillway flow, Journal of Hydraulic Research, 43:5, 522-531, DOI:10.1080/00221680509500150

Appendix A: FLOW-3D Numerical Model Configurations FLOW-3D, a commercially available computational fluid dynamics (CFD) software package developed by Flow Science Inc. was used for all CFD simulations. FLOW-3D utilizes the Reynolds averaged Navier-Stokes (RANS) equations to solve for fluid flow. Modifications to the standard RANS equations include algorithms to accurately track the water surface and flow around geometric objects (Hirt and Nichols, 1981; Hirt and Sicilian, 1985; Hirt, 1992). The location of the free surface and solid objects is tracked throughout a Cartesian gridded domain using the true volume-of-fluid (VOF) method presented by Hirt and Nichols (1981).

All numerical models were conducted at prototype dimensions to avoid any possible scale effects and allow data to be analyzed directly from the output without requiring any additional conversions or computations. Numerical models focused on the spillway entrance configurations and modeled a portion of the spillway flows down through spillway station 2+00. Models were run using a Cartesian grid with 1-ft cubic cells. Each model was run with a constant upstream boundary supplying 18,600 ft3/s at a reservoir WSE of 6911.0 ft. Simulations ran until they reached a quasi-steady state condition where the water surface elevations, turbulent energy, kinetic energy, and velocities remained similar from one time step to another.

This appendix provides screen shots of each configuration that was tested numerically and compared visually to determine which option would be modeled physically (Figure 48 to Figure 58). Velocity contours and WSE plots are not provided.

FLOW-3D approach channel configuration “A” isometric view.