Technology Development

   

Reclamation Dam Safety Technology Development

TITLE
REPORT NO.AUTHOR(S)LINK
ABSTRACT
 
Dynamic Direct Tensile Strength of Concrete
DSO-2020-01Veronica Madera, P.E., Catherine Lucero, P.E.DSO-2020-01 
Tensile strength of concrete is of great importance for dams located in areas subjected to seismic loading conditions. Structural analysts compare results from finite element models (FEM) of such structures to the strength of the concrete to determine whether they able to withstand those loads or require modification to maintain the infrastructure and ensure public safety. The goal of this research was to relate the number of cycles of dynamic tension that concrete can withstand prior to failure as a function of the static tensile strength and loading frequency. No such relationship was determined to exist at the load amplitudes and frequencies used for this research. Equations developed by the Comité Euro-International du Béton (CEB) in the 1990s are better suited for estimating a dynamic increase factor (DIF)rather than assuming a 50% increase in tensile strength.
 
2019 ICOLD Benchmark Workshop - Seismic Analysis of Pine Flat Concrete Dam: Formulation Package
DSO-2020-01 FormulationJerzy Salamon, Ph.D., P.E.Formulation Package 
The package includes formulation and input files used in the Seismic Analysis of Pine Flat Concrete Dam during the 15th ICOLD Benchmark Workshop in Milan on September 9, 2019. The benchmark was focused on the validation of numerical methods, the role of the foundation size, boundary conditions and mass in the model, the non-linear material response, with the ultimate goals of identifying the key uncertainties causing differences in results and developing the best practices in the dynamic analysis of concrete dams.

The summary of the benchmark results could be found in the book Numerical Analysis of Dams, Proceeding of the 15th ICOLD International Benchmark Workshop at www.springer.com/gp/book/9783030510848.
 
Evaluation of Numerical Models and Input Parameters in the Analysis of Concrete Dams - Summary Report of the 2018 USSD Benchmark Workshop
DSO-2019-13Jerzy Salamon, Ph.D., P.E.DSO-2019-13 
The report summarizes the results of the benchmark studies presented during the workshop titled, “Evaluation of Numerical Models and Input Parameters in the Analysis of Concrete Dams” during the USSD Annual Conference and Exhibition in Miami, Florida on May 3, 2018. A series of case studies, based on Pine Flat Dam, included a time history analysis for a linear concrete dam-foundation-reservoir model.
 
Evaluation of Various Approaches to Obtaining Vs30 Values
DSO-2019-01Justin B. RittgersDSO-2019-01 
The report presents an overview and evaluation of all relevant techniques currently available to Reclamation for assessing site-specific Vs30 values. The main objective of the study is to compare and evaluate all indirect Vs30 estimation techniques and all direct Vs30 measurement techniques (surfacebased and borehole seismic techniques), in order to better understand the various accuracies, advantages, and limitations of each technique. The study is based upon the quantitative and qualitative comparisons of various co-located seismic surveys and Vs30 estimates conducted at various field sites from 2016 through 201$ at several Reclamation and non-Reclamation embankment dams. Results of these various surveys and corresponding new or otherwise updated Vs30 values are also presented, where applicable. Finally, these results are utilized to inform and develop a set of practical Vs30 technique selection guidelines that can be used by project managers and engineers in order to optimize the tradeoffs between value of information and cost for future Dam Safety seismic risk analysis planning efforts.
 
Developments for the Concrete Tensile Split Test
DSO-2018-11Jerzy Salamon, Ph.D., P.E.DSO-2018-11 
This technical report examines the existing ASTM and USBR procedures for the split tension test. Specifically, a background detailing the history, overview of the current procedure, and problem with the split tension test results is presented. The results from laboratory and finite element analyses are discussed. Finally, conclusions are made on the splitting test and research completed, in addition to recommendations going forward.
 
Seismically Induced Hydrodynamic Loads on Concrete Dams and Spillway Gates
DSO-2018-10Jerzy Salamon, Ph.D., P.E.DSO-2018-10 
In the report, three methods that are commonly used in the time analysis of the dam-reservoir-foundation system during an earthquake are evaluated. These include: the “added mass”, the “acoustic fluid”, and the “fluid-like material model” methods. The theory for each of the methods is briefly presented and it is illustrated by the results for a case of two concrete dams and a large radial gate.
 
Guide for Analysis of Concrete Dam Structures using Finite Element Methods
DSO-2018-09Lan Nguyen, Ph.D., Jerzy Salamon, Ph.D.DSO-2018-09 
This report summaries our work for the Dam Safety funded research project “Guide for Concrete Dam Analysis Using Finite Element Method”. Here we discuss fundamental topics such as the mathematical formulation in term of both field partial differential equations and variational principles, boundary conditions for dam and fluid-dam interaction, the matrix formulation of the FEA equations, and solution approximation schemes for dams excited by the seismic events. Two major time integration method classes in dynamic analyses: explicit and implicit methods are covered and illustrated with algorithm steps.

We have primarily focused on topics that affect numerical stability and accuracy. These include: the effect of time step on numerical stability, sensitivity of results to mesh fineness and element-level numerical integration schemes, and degradation of accuracy due to element type and mesh distortion. These topics span a wide range from physical and mathematical modeling, passing through FE model discretization, and getting to numerical results.
 
Rock/Concrete Direct Shear Constant Normal Stiffness
DSO-2018-08Evan J. LindenbachDSO-2018-08 
In underground excavations, concrete dam foundations, drilled piers socketed into rock and rock anchors, dilation during sliding is constrained by the rock/concrete mass and the normal stress is not constant along a discontinuity. This boundary condition is better represented by a constant normal stiffness (CNS) direct shear test than the typical constant normal load (CNL) direct shear test. This is likely also the case for characterizing the strengths of concrete in large mass concrete dams where the normal stress may not be constant along the discontinuity during sliding.

This research developed the control software for a CNS boundary condition for an existing rock/concrete direct shear machine. After the programming was complete, a series of commissioning tests were performed, including a number of identical: a) 6-inch diameter concrete specimens, and b) sawtoothed hydrostone specimens to mimic rock discontinuities. Intact concrete specimens were tested for break-bond and sliding friction strengths while the hydrostone specimens were only tested for sliding friction strength. Both sets of tests were performed under both CNL and CNS conditions to investigate the effects on shear strength due to the different boundary conditions. This paper details the system upgrades and the results of the comparison testing under CNL and CNS boundary conditions, and provides examples of CNS data use in design.
 
Finite Element Model Development with LS-PrePost
DSO-2018-07Jonathan East, P.EDSO-2018-07 
Currently in the TSC, nonlinear finite element analyses (FEA) of concrete dams are completed to betterunderstand the nonlinear behavior of such structures. The process of FE model development for a nonlinear model is verylabor intensive and time consuming; some projects last multiple years and absorb Reclamation's limited personnelresources.

The quality of the finite element (FE) mesh is a key factor affecting the accuracy of FE analysis. This research investigates anew method of developing FE models using 3D AutoCAD and LSDYNA PrePost (LS-PrePost) as an alternative program touse. The tasks included in this research are to develop a new procedure for building FE models using 3D AutoCAD andLS-PrePost, develop a procedure for inputting the terrain topography into FE models using AutoCAD Civil 3D, documentthe modeling procedures of using AutoCAD and LS-PrePost in a report, evaluate the ability for local mesh refinement, anddescribe the model self-check capacity using LS-PrePost.
 
Finite Element Model Size Reduction Research and Validation Study
DSO-2018-05Hillery Venturini, Roman Koltuniuk, Jen HugginsDSO-2018-05 
In the past years, the scope of finite element models (FEM) has grown significantly. FEMs can be used to analyze various types of dams, with processing time varying based on the complexity of the model. FEMs used to inform dam safety decisions now contain over two million elements, with the potential to increase depending upon project complexity. These models are generally used to assess the likelihoods of spillway-related seismic potential failure modes (PFM) and incorporate features such as the spillway, dam, reservoir, and surrounding landscape.

Dynamic analysis run times can take as long as one day per second of ground motion time history. This means that running a simulation of a 21-second-long earthquake can require as much as three weeks of computing time. If a dam should happen to be located near a subduction zone area, in which ground motions can last as long as 300 seconds, computations may require months to run. Such complex models are time-consuming in their initial mesh generation, and as run times protract, they can extend project schedules in the latter stages as well. In an effort to deliver the greatest value to every client, this research was undertaken to investigate and test smaller FEMs and asses their quality and accuracy. Such efforts could have the potential to reduce cost to the client.
 
Finite Element Analysis for Spillway Retaining Wall Shake Table Testing Program
DSO-2018-02Mario Pereira, Roman KoltuniukDSO-2018-02 
This report presents the Finite Element (FE) analysis conducted by the Waterways and Concrete Dams Group 1. The FE analysis was performed for Shake Test Model Configuration 1 (March 30th, 2016) conducted at UCSD. Displacement and moment results from the FE analysis were compared to the measured displacement and moment output from the shake table test.
 
Concrete Shear Strength Parameters Compared to Other Laboratory Determined Properties
DSO-2017-09Evan J. Lindenbach,DSO-2017-09 
Examination of a large sample of direct shear test results on concrete specimens offers a unique insight into material properties for use in appraisal level design or a screening level evaluation of an existing structure. Several factors, such as cost, lack of suitable cores for testing, or historical data sets that do not contain direct shear test results, often make inference of direct shear strength, in both breaking and sliding friction, via correlation from other properties desirable. In order to investigate a possible correlation between the results of unconfined compressive strength, direct tensile and splitting tensile tests, and shear strength, the results of numerous tests conducted on core samples from various projects were analyzed.

Results of the study confirmed commonly used relationships between compressive and tensile strength parameters. No direct correlation between compressive/tensile strength parameters was noted. The study did note that the cohesion intercept for break-bond testing can be roughly approximated using splitting tensile or uniaxial compressive testing results, and that the Mohr-Coulomb friction angle for break-bond testing is similar to that found during sliding shear testing.
 
Evaluating Natural Pozzolans for Mitigating Temperature Rise in Mass Concrete
DSO-2017-07Catherine Lucero, MSCE, EITDSO-2017-07 
The thermal behavior of concrete has long been studied to reduce cracking, especially in massive members, where heat generation is great. Heat dissipation and restraint create volume changes that can induce stresses at an early age, when strength and modulus are low. Pozzolans or other supplementary cementitious materials can be used to reduce the temperature rise in large concrete placements. Reclamation has recently updated thermal property data sets for pure ordinary Portland cement (OPC), OPC + Class F fly ash, and OPC + slag. While Class F fly ash has traditionally been specified for mass concrete within Reclamation, sources of quality ash are becoming increasingly scarce. Class N natural pozzolans can be used as a substitute in mass concrete to decrease the temperature rise and mitigate early age thermal cracking.
 
Comparison of Thermal Properties for Concrete
DSO-2017-05Catherine LuceroDSO-2017-05 
Thermal effects from cement hydration and environmental factors can lead to thermal gradients within the concrete section and induce thermal cracking. Temperature effects are especially important in mass concrete, where the size of the placement can produce large amounts of heat with relatively little surface area to dissipate it. Even in smaller placements, concrete containing high cement contents or supplementary cementitious materials (SCMs) such as silica fume can produce high temperatures. There are several approaches for estimating temperature rise of concrete. Some of these tools could be beneficial to Reclamation’s designers when determining placement size or determining if a temperature control plan is required.
 
Temperature Rise of Mass Concrete Containing Class N Pozzolan
DSO-2017-04Catherine Lucero, MSCE, EITDSO-2017-04 
Thermal effects from hydration and the environment can lead to thermal gradients within the concrete section and induce thermal cracking. Temperature effects are especially important in mass concrete, where the size of the placement can produce large amounts of heat and relatively little surface area to dissipate it. Even in smaller placements, concrete containing high cement contents or supplementary cementitious materials (SCMs) such as silica fume can produce high temperatures. Some Reclamation Regions are experiencing a shortage in Class F fly ash which is usually the SCM used to lower the heat of hydration. This study reports the adiabatic temperature rise and heat of hydration for concrete containing Class N pozzolan from a source near several Reclamation projects. The data from this report is compared to mixtures containing Class F fly ash.
 
Colorado Front Range Streamflow, Geomorphological Hazard and Rainfall Relationship. Storm and Flood of September 11-13, 2013.
DSO-2017-03Giorgia F. deWolfe, PhD. GeologistDSO-2017-03 
A unique opportunity arose on September 11-13, 2013 for Reclamation scientists and engineers when the Northern Front Range of Colorado was impacted by one of the largest rainstorms in recorded history. Initial estimates of the total rainfall far exceed previous recorded values and the widespread flooding that resulted from the rainfall affected most of the major drainages heading in the foothills east of the continental divide and west of the major population centers in the Denver-Boulder-Fort Collins corridor.
 
Evaluation of Numerical Models for Simulating EmbankmentDam Erosion and Breach Processes
DSO-2017-02Working Group on Embankment Dam Erosion and Breach Modeling, CEATI International Dam Safety Interest GroupDSO-2017-02 
The Working Group on Embankment Dam Erosion and Breach Modeling was organized under the Dam Safety Interest Group of CEATI International to conduct research on improved methods for modeling the failure of embankment dams. From approximately 2007-2009 the group conducted anevaluation of numerical breach models using data from five large-scale laboratory tests and two real dam failures. The evaluation focused primarily ontwo models, SIMBA, developed by the USDA-Agricultural Research Service (USDA-ARS), and HR BREACH, developed at HR Wallingford. Both models performed well on 5 of the 7 test cases, with the exceptions caused by a lack of high quality data on the embankment properties, failure scenarios, and resulting floods. The evaluation showed that the ability to obtain accurate estimates of soil erodibility was crucial to achieving good model performance, with direct measurements made through lab or field-based testing being most preferable. The evaluation also showed that there is a pressing need for research that can determine whether embankment materials are likely to erode predominantly through surface erosion or headcutting mechanisms.
 
Precast Ogee Crest – Conceptual Design Study
DSO-2016-09Patrick MaierDSO-2016-09 
This research falls within the Dam Safety Program Technology Development Priority 1.4.4, Methods that investigate designs or construction techniques that would result in higher quality, in an effort to obtain a better final product. This document discusses and documents the recently completed conceptual designs for a precast concrete ogee crest. The conceptual designs included development of two general construction options for utilizing a precast concrete ogee crest as opposed to conventional cast-in-place methods of construction. Development of these designs included concrete outline drawings, reinforcement drawings, and an investigational finite element analysis. The finite element analysis was developed to understand what types of stresses a precast ogee crest would experience during fabrication, transportation and installation (including crane picks). Additionally, the team also investigated some constructability and transportation concerns regarding the precast ogee crest.
 
Characterizing Flood Seasonality in the Taylor Park Dam Watershed
DSO-2016-07Kathleen D. Holman, Ph.D.DSO-2016-07 
Taylor Park Dam is the current subject of a hydrologic hazard analysis (HHA). The HHA is motivated by results from a 2005 Comprehensive Facility Review, which indicate that Taylor Park dam overtops at approximately 54% of the rainon-snow probable maximum flood, based on a June general storm. Although theBHA will explore historical streamflow events for use in a flood-frequency analysis, an analysis of flood seasonality within the watershed, along with an explanation of potential meteorological causes is outside the scope and budget of the study. The purpose of the current project is to explore historical flood seasonality in the Taylor Park Dam watershed and identify antecedent conditions prior to major flooding events. We analyze historical observations of flow, precipitation, and snow water equivalent (SWE) in and surrounding the watershed.

Our results show that average daily flows into Taylor Park reservoir peak in June, May, and July. Average monthly precipitation, however, is lowest in June and greatest in August. SWE in the region typically peaks during March, April, and May, while the largest negative average daily changes in SWE occur during May. Annual maximum inflow events in the watershed occur most frequently during June and May, though there is large year-to-year variability in the flow magnitudes.

Composite analysis of land surface output from the Variable Infiltration Model, which is included in phase 2 of the National Land Data Assimilation System, suggests that the annual maximum inflow events in the Taylor Park Dam watershed can be described as rain-on-snow events. However, composite analysis captures "average" conditions. Consequently, additional research is needed to explore the annual maximum inflow events individually. There remains a possibility for other mechanisms (aside from rain-on-snow) to produce some of the annual maximum inflow events.
 
Underwater Performance of a Hybrid Polymeric Repair Material to Seal Cracks in Concrete
DSO-2016-04Shannon Harrell, P.E.DSO-2016-04 
This research program evaluated the crack sealing performance of a combination of hydrophilic and hydrophobic polyurethane grout, using two proportions of two different manufacturer’s grouts. The hybrid grouts were then injected underwater to determine if they could seal a crack in a concrete specimen. Successful performance of the hybrid polyurethane grout underwater consisted of sealing the crack, maintaining a 1 psi pressure, keeping the water clear during the injection phase, and having minimal floatation of the cured product. None of the hybrid grouts maintained the 1 psi pressure. However, the concrete specimens were qualitatively evaluated after the testing to determine how the grouts performed in the categories of penetration, number of gaps, adhesion, and shrinkage. The post evaluation also evaluated the final reacted product in the crack, characterizing the hybrid grouts as either flexible or rigid foams. Two of the four grout combinations were recommended based on the “good” rating they received in the post evaluation. Further study is recommended to determine if a hybrid polyurethane grout will fully seal cracks and sustain pressure underwater.
 
Aggregate and SCM Alkali Release
DSO-2016-03Catherine Lucero, MSCE, EITDSO-2016-03 
This research aims to find a better understanding of how alkalis are released from concrete aggregates and SCMs (mainly fly ash) into the concrete pore solution. This can be accomplished using existing structures information and performing alkali release testing utilizing multiple methods, in particular an electron microprobe analyzer equipped with wavelength dispersive spectroscopy. This testing can provide analytical geochemistry and quantitative microscopy to enhance understanding of how various minerals in concrete aggregates and SCMs release alkalis into concrete pore solution. For existing Reclamation structures, records have been kept on the materials and proportions used. In this scoping study, a data search has been performed to find relevant structures and information about the initial alkalis present at the time of construction.
 
Concrete Tensile Strength Specimen Size Effects
DSO-2016-02Concrete, Geotechnical and Structural Laboratory StaffDSO-2016-02 
Three concrete mixtures with different nominal maximum size aggregates (NMSA) were developed and 3-foot cubical blocks were cast from which cores of various diameter were extracted for testing. Specimens were cut from the cores and tested in compression, splitting tension, static direct tension, and cyclic dynamic direct tension. Due to the complexity and cost associated with obtaining and testing the specimens, only a very limited number of tests were conducted. The study was not intended to present a statistically representative sample size, but to see if any trends could be observed for each combination of NMSA and core diameter. The Principle Investigator for the project left Reclamation prior to completing a draft report. Subsequently, staff from the Concrete, Geotechnical and Structural Laboratory assembled the figures and data to present in this report. By documenting this program and the test results, future studies can be performed that will add to the available data and perhaps lead to more robust findings.
 
Finite Element Research Compilation
DSO-2016-01Ron Kurzdorfer, P.E., Nate SpragueDSO-2016-01 
The purpose of this Technical Memorandum is to compile all the work that has been performed in the past to validate the finite element analyses that have been performed by the Waterways and Concrete Dams Groups 1 and 2. The TM was written in conjunction with the research project "Dam Performance During Seismic Loading Research Program"
 
Updating Thermal Data Sets to Better Evaluate thermal Effects of Concrete
DSO-2015-02Katie Bartojay, P.E., Catherine Lucero, M.S., E.I.T.DSO-2015-02 
The thermal behavior of concrete has long been studied to reduce cracking, especially in massive members, where heat generation is great. Heat dissipation and restraint create volume changes that can induce stresses at an early age, when strength and modulus are low. Many designers and contractors use Figure 31 of Reclamation Engineering Monograph (EM) No. 34 “Control of Cracking in Mass Concrete Structures” or Figure 4.1 of ACI 207.2R “Thermal and Volume Change Effects on Cracking of Mass Concrete” when preparing their temperature control plan for mass concrete. These figures were generated from pre 1960’s data and can be traced back to the 1936 Boulder Canyon studies. Due to changes in physical and chemical properties of cement over the years, these figures no longer provide accurate guidance to designers. In an effort to provide a reliable set of thermal properties, nine (9) concrete mixtures were tested for adiabatic temperature rise, diffusivity, conductivity and thermal expansion. This study includes mixtures containing various amounts of OPC, Class F fly ash, and Grade 100 Slag.
 
Spillway Wall Structural Capacity
DSO-2015-01Roman M. Koltuniuk, P.E.DSO-2015-01 
This report compares how well finite element analyses of spillway counterfored walls match lab tests of a model of a typical counterforted wall. This is part of an ongoing validation study to gain confidence in using finite element models to predict the seismic performance of dams.
 
Evaluation of Nonlinear Material Models in Concrete Dam Finite Element Analysis
DSO-2014-08Jerzy Salamon, Ph.D., P.E., David W. Harris, Ph.D., P.E.DSO-2014-08 
This report presents a study performed to validate a new concrete constitutive material model implemented in LS-DYNA Finite Element program by ADAPTEC in 2007 for concrete dam applications. The new Continues Surface Cap Model (CSCM) was successfully used by Federal Highway Administration for the highway related projects. The main objective of this study are to evaluate the effectiveness of the CSCM in the nonlinear analysis of concrete dams and to calibrate the input parameters of CSCM against the data obtained from laboratory test.
 
Scanning Sonar Technology Development – Dam Safety Technology Development Program
DSO-2014-06Tracy B. VermeyenDSO-2014-06 
The problem with performing underwater inspections of hydraulic structures using divers or remotely operated vehicles is that visibility is often poor, which increases the time the inspection takes and reduces the inspection quality. Commercially available scanning sonar systems can be used to overcome this limitation because sonar works in highly turbid water to produce detailed images of underwater infrastructure. Scanning sonar can also collect survey grade bathymetry and three-dimensional (3-D) point clouds of underwater features. Point cloud data can be used to develop 3-D computer models or to compute estimates of lengths, areas, or volumes. Typical scanning sonar applications include inspections of scour, undercutting, concrete abrasion, exposed rebar, and accumulation of sediment around or debris on intakes. A scanning sonar system is currently being used for a variety of projects, which is critical for developing experience with sonar operation and interpretation of sonar images. This experience includes developing deployment techniques for inspecting features unique to large hydraulic structures such as embankment and concrete dams. The first few applications revealed the need to upgrade the sonar system with accessories to improve data quality, data collection efficiency, and the quality of geographic position data. This report describes the system enhancements and deployment improvements that have been achieved.
 
Large-Scale Filter Performance Tests
DSO-2014-05Ted Howard, Caleb Rudkin, Peter IreyDSO-2014-05 
The Bureau of Reclamation, in partnership with the U.S. Army Corps of Engineers, has been conducting large-scale filter testing for several years. These tests have been performed to gain a better understanding of the performance of filters subject to cracking. In particular, they explore how well a crack can form within a filter, how well a cracked filter subject to a concentrated leak can heal, and how effectively a filter can stop or control a concentrated leak. This paper discusses the construction of the testing apparatus, filter materials tested, test procedures used, results for 14 tests that have been performed, and lessons learned pertinent to the design engineer.
 
Gridded Snow Water Equivalent Data Set Development - Processing Methods Using the Geospatial Data Abstraction Library
DSO-2014-04Jeffrey P. Niehaus, M.S., P.E.DSO-2014-04 
The purpose of this report is to document the data processing steps developed to convert the SNODAS SWE flat binary grids to a suitable format for input to stand-alone hydrology models and as a data set for climate change or extreme flood analysis in relation to Dam Safety Technology.
 
Seminoe Dam - Assessment of Concrete by Quantitative Methods – The Petrographic Damage Rating Index
DSO-2014-03Douglas R. HurcombDSO-2014-03 
The Damage Rating Index (DRI) method was conducted on mass concrete cores to determine the value of the method for overall evaluation of damage to Reclamation’s concrete structures. This study compares the current Seminoe Dam compressive strength trends with DRI results and contemplates how the results could be used to evaluate concrete from structures suffering from ASR in Reclamation’s aging concrete infrastructure. The DRI can be performed on a limited number of samples and provides a semi-quantitative value related to alkali-silica reaction (ASR) aggravated concrete damage. The DRI value can be used to evaluate the extent of damage caused by ASR and appears to correlate well with the traditional petrographic concrete quality approach. The approach may be a useful way to estimate the extent of damage in a concrete structure.
 
Verification of FLAC Mohr-Coulomb Model for Granular Materials under Monotonic Loading
DSO-2014-02Kevin Zeh-Zon Lee, Ph.D., P.E.DSO-2014-02 
The computer program FLAC has been used extensively by Bureau of Reclamation (Reclamation) for determining embankment dam deformations under seismic loads for Dam Safety projects. However, there are substantial uncertainties regarding the predictive capability of FLAC. Verification and validation (V&V) process is needed in order to evaluate the predictive capability of FLAC for embankment dam applications. One of the components in the V&V process for numerical model simulation is to examine material behavior at the element level. At Reclamation, geologic materials are modeled almost exclusively using the FLAC Mohr-Coulomb model for its simplicity, as it only requires four shear strength parameters and two elastic moduli. Without the verification assessment, it is uncertain if FLAC Mohr-Coulomb model could generate stress-strain responses comparable to laboratory element tests. This research study focuses on the verification of FLAC Mohr-Coulomb model subjected to monotonic loading.
 
Construction Flood Case Histories
DSO-2014-01Daniel D. Mares, David P. Keeney, Victoria Sankovich BahlsDSO-2014-01 
The purpose of this paper is to review case histories of significant floods that have occurred during construction of dam modifications, the meteorological and hydrologic conditions that occurred, and the response of the cofferdams or design features to the loading. The construction flood case histories include Jackson Lake Dam in 1986, Roosevelt Dam in 1993, and Glendo Dam in 2010 and 2011. Another case history is added (appendix A), which documents the Auburn cofferdam breach and the large flood event that caused the breach. The case histories are useful in reviewing the design floods selected, the cofferdam type and its performance, and in reinforcing the importance of evaluating risks during construction as presented in the Bureau of Reclamation’s Best Practices. The meteorological data (i.e., rain gauge observations, synoptic maps, National Center for Atmospheric Research/National Centers for Environmental Prediction reanalysis, and radar) were reviewed, and a summary of the meteorological conditions is presented.
 
Soil-Structure Interaction - Phase 3 - Feasibility Level Design - Full-Scale Retaining Wall Shake Table Testing
DSO-2013-03Ahmed ElgamalDSO-2013-03 
This ongoing research will help answer the questions that continue to arise as Reclamation is faced with detailed analysis and modifications of spillways and other earth-retaining structures subjected to significant seismic loading. The objective of this research is a better analytical tool to predict the seismic lateral earth pressures for configurations that include groundwater, cohesion, non-horizontal zone, and compaction/in-place density effects under various ground accelerations. This report represents the conclusion to the third phase (Feasibility Level Design Phase 3) of this research project. The ultimate project purpose is to complete a full scale shake table test of a concrete cantilever retaining wall.
 
Extreme Floods in a Changing Climate
DSO-2013-02Godaire, J., J. Caldwell, N. Novembre, T. Harden, V. SankovichDSO-2013-02 
This study explored the potential connection between extreme floods, streamflow and climate change. The analyses concluded that relationships do exist between records of fluvial deposition and shifts in climate in that less fluvial deposition (or lower streamflow) is occurring during periods of drier climate and more fluvial deposition (or higher streamflow) is occurring during periods of wetter climate in the Sierra Nevada of central California. This relationship can be defined on a broad scale for about the last 1,000 years. Paleofloods that have been documented along rivers in the Sierra Nevada appear to fall within both dry and wet periods in the paleoclimate record. Paleofloods that have been documented in the southwestern U.S. in the Colorado River Basin appear to occur within specific wet or dry periods, depending on the geographic location.
 
Capping Study for Large Diameter Cylinders
DSO-2013-01J. Bret RobertsonNot Available 
The Materials Engineering and Research Laboratory (MERL) conducted testing for four proposed methods of capping large diameter cylinders. These methods include, lapping, unbonded/unconfined pad caps, unbonded/confined (pad) caps, and sulfur. To accurately test these four methods, 30 cylinders for each capping method will be made, 120 specimens total. All the cylinders need to be made at same time with the same mixture proportions. Testing needs to take place within a reasonable time frame at a late stage of hydration (around 90-180 days of age).
 
Green Mountain Dam Climate Change
DSO-12-03Victoria Sankovich, R. Jason Caldwell, & Kelly MahoneyDSO-12-03 
The Weather Research and Forecasting (WRF; Skamarock et al., 2007) model was investigated to understand its capabilities concerning its ability to model historic, large precipitation events; allow model users to maximize storm parameters; and allow model users to perturb storm parameters for climate change scenarios. Extreme storms in the vicinity of the Green Mountain Dam watershed were examined due to the orographic nature of the region and as a result of ongoing questions concerning high-elevation precipitation estimates originating from the Hydrologic Hazard Study for Green Mountain Dam. Two historical storm events that occurred in the vicinity of the Green Mountain watershed were modeled using the WRF model. The 4-6 Sept. 1970 storm event was deemed sufficient to serve as the basis for experiments for moisture maximization and climate perturbation simulations. This proof-of-concept study demonstrates that the WRF model is capable of simulating extreme storms in orographic locations within the vicinity of Green Mountain Dam. This modeling approach offers potential utility for use in model-based storm-maximization and climate-change assessment studies. The results presented in this study are considered preliminary and exploratory in nature. Running a high resolution model requires significant resources and, for the results to be robust, further improvements to the research methodology and additional applications are required. Based on the preliminary results, the WRF model is beneficial to the extreme storm analysis process needed for Dam Safety Hydrologic Hazard Analyses.
 
Thermal Properties of Reinforced Structural Mass Concrete
DSO-12-02Katie BartojayDSO-12-02 
The in-situ temperature rise of structural reinforced mass concrete (RSMC) differs significantly from traditional unreinforced mass concrete, such as dams. The compressive strength of RSMC structures is often higher and required at earlier ages. This leads to higher cementitious materials contents and subsequent higher potential temperature rise. Two RSMC placements were monitored for temperature rise and potential harmful temperature gradients, one with a 4,000 psi design strength and the other with a design strength of 7,000 psi. The temperature rise was monitored at the center, near the surface, and at mid-point locations for up to 90 days. One mixture was also tested in the laboratory for compressive strength, elastic properties, and thermal properties under standard curing conditions and under a simulated in-situ temperature regime. This report will highlight the in-situ properties of RSMC and compare methods taken to reduce the temperature rise.
 
Development of a Test to Determine Cementation Potential of Embankment Dam Granular Filter Material – Results of Phase III Research
DSO-12-01Robert V. RinehartDSO-12-01 
Granular filters are used in embankment dams to protect against internal erosion either through the embankment or the foundation. For proper performance, it is required that the filter material not be able to sustain a crack during embankment settlement or seismic loading. Historically, the mechanism used to limit cracking potential is a restriction of no more than 5% nonplastic fines, and this requirement is easily measured with the standard soil tests for grain size distribution (gradation) and plasticity (Atterberg Limits). Unfortunately, material meeting this requirement has been observed to cement, and hence has cracking potential. More robust test procedures are needed in order to better quantify the desired material properties (cementation potential), and to this end, modification of the Vaughan and Soares Sand Castle Test has been undertaken jointly by the Bureau of Reclamation and U.S. Army Corps of Engineers. This report presents the third phase of the research, describing specimen preparation and test procedures and results from 16 source materials from around the United States. All materials met ASTM C-33 fine aggregate gradation requirements as well as the additional requirement of less than 2% fines. Additionally, unconfined compression tests were performed on each material to help quantify the strength gain from cementation. The Sand Equivalency Value, which further quantifies the amount and characteristics of the clayey portion of a material, was also determined for all sources to see how well it correlated with the Modified Sand Castle Test results. The Modified Sand Castle Test is shown to be a good indicator of cementation potential and correlates well with unconfined compressive strength, but to a lesser degree with sand equivalency value. Petrographic analysis aided in understanding the cementation mechanisms. A discussion is presented describing how the tests results could be carried forward into practice.
 
Extreme Storm Data Catalog Development
DSO-11-07Victoria Sankovich & R. Jason CaldwellDSO-11-07 
Meteorological inputs are necessary for dam safety hydrologic hazard studies: storm-based precipitation is the dominant forcing variable for the generation of extreme floods in rainfall-runoff models. Extreme storm probability estimates are needed to better understand extreme flood processes and are also required for watershed models. However, the majority of historical storm data exists only in paper format and is not stored in a logical manner or in a central archive. There are no current procedures to update historical storm data sets. Furthermore, methods for estimating extreme storm probabilities, up to and including the Probable Maximum Precipitation, are currently lacking. In response, extreme storm data research was completed in three main areas: 1. a comprehensive electronic database of historical extreme storm events in GIS was developed; 2. methodologies and datasets that could be used to process new storms based on Multisensor Precipitation Estimates were investigated and documented; and 3. in-house capabilities and programs to calculate regional precipitation frequencies up to the PMP in a format suitable for flood runoff models were developed. This research will save time and resources for ongoing and future dam safety hydrologic hazard studies, provide a data repository for future storm analyses, and demonstrate the capabilities of Reclamation-developed programs and procedures for meteorological needs.
 
Seismic Induced Loads on Spillway Gates - Phase I – Literature Review
DSO-11-06Jerzy W. SalamonDSO-11-06 
This report includes a literature review and compilation of technical references regarding seismic induced loads on spillway gates. The literature review included widely-accepted analytical methods as well as studies focused on the dam-reservoir and spillway gates-reservoir interaction. One of the primary objectives of the literature review was to identify the factors and limitations that are important in determination of hydrodynamic loads on the spillway gates during an earthquake.
 
Binding Agents in Embankment Dam Protective Filters
DSO-11-04Robert V. Rinehart & Mark W. PabstDSO-11-04 
Granular filters are used in embankment dams to protect against soil that may erode through cracks in the embankment core. It is required that the filter material itself not be able to sustain a crack. Design recommendations exist to limit cohesive behavior of filters. However, based on the observation of cemented behavior of filter materials in the field, it became clear that requirements beyond the existing ones related to grain size and plasticity were needed. The goal of the present research is to evaluate a new index test method to quantify the cementation potential for candidate filter materials for use in embankment dams. This report describes a modified sand castle test method and apparatus, test results for 12 different materials, conclusions from the research, as well as recommendations for future study.
 
Evaluation of Paleoflood Peak Discharge Estimates in Hydrologic Hazard Studies
DSO-11-03Travis Bauer & Ralph KlingerDSO-11-03 
The findings of this study are a compilation of two different research projects. The initial project considered the differences in peak discharge estimates made using a single cross-section slope-conveyance calculation with the Manning equation (Method 1) to discharge estimates made using a more sophisticated one-dimensional hydraulic model (i.e., HEC-RAS; Method 2). Initial results were documented in a draft report with the same title as this report, but dated December 2007. The findings of that report have been updated and are included in their entirety in this report, which supersedes the findings of the draft report.

The second study included a similar evaluation as the previously mentioned study, but in this case comparing one-dimensional model results (Method 2) to two-dimensional hydraulic model (SRH2D) results (Method 3). At the time of the first study, the two-dimensional hydraulic model SRH2D was not fully developed, and the detailed topographic data needed as model input were not readily available. Over several years the development of SRH2D improved to the point that it can easily be applied to a paleoflood study as long as the topographic data are sufficiently detailed (Lai, 2009). After using SRH2D to estimate peak discharge on several projects (Los Banos IE, Klinger and Bauer, 2004;

Whiskeytown CAS, Klinger and Bauer, 2009; Anderson Ranch IE, Klinger and Bauer, 2010), enough data were available with which to compare one and two-dimensional model results.
 
Soil-Structure Interaction - Planning Phase 2
DSO-11-02Steve Dominic & Larry NussDSO-11-02 
This ongoing research will help answer the questions that continue to arise as Reclamation is faced with detailed analysis and modifications of spillways and other earth retaining structures subjected to significant seismic loading. The objective of this research is a better analytical tool to predict the seismic lateral earth pressures for configurations that include groundwater, cohesion and compaction/in-place density effects under various ground accelerations. This report represents the completion of the first year of the second phase (Planning Phase 2) of this research project. The ultimate project purpose completing a full scale shake table test of a concrete cantilever retaining wall.
 
Soil Structure Interaction on Spillway Walls Adjacent to Embankments
DSO-11-01Roman M. KoltuniukDSO-11-01 
An extremely important aspect of the analyses of an embankment dam with a gated concrete spillway is the interaction of soil with the spillway walls during a seismic event. If the soil load becomes too great, the concrete spillway retaining wall may fail, opening up a large seepage path between the failed wall and the soil behind it, leading to erosion of the soil and, eventually, the entire dam. A more rigorous approach is required to evaluate seismic loads on concrete spillways other than traditional pseudostatic methods. This research will propose the use of the LSDYNA finite element code to model spillway walls with soil behind them in order to achieve this and will validate this code against results from centrifuge experiments at Berkeley .
 
Soil-Structure Interaction Scoping Phase 1
DSO-10-01Steve Dominic and Liu MingDSO-10-01 
This ongoing research will help answer the questions that continue to arise as Reclamation is faced with detailed analysis and modifications of spillways and other earth-retaining structures subjected to significant seismic loading. The objective of this research is a better analytical tool to predict the seismic lateral earth pressures for configurations that include groundwater, cohesion, nonhorizontal zone, and compaction/in-place density effects under various ground accelerations. This report represents the conclusion to the first phase (Scoping Phase 1) of this research project. The ultimate project purpose is to complete a full scale shake table test of a concrete cantilever retaining wall.
 
Probabilistic Slope Stability Analysis Using the Random Finite Element Method (RFEM)
DSO-09-02Giorgia Fulcheri deWolfeDSO-09-02 
This research developed a computer program based on the random finite element method (RFEM), able to estimate the probability of failure of slopes while fully accounting for spatial variability controlled through an input parameter called the spatial correlation length. The elasto-plastic finite element slope stability method makes no a priori assumptions about the shape or location of the critical failure mechanism and therefore, offers very significant benefits over traditional limit equilibrium methods in the analysis of highly variable soils. Loadings of particular interest in this project include seismic events and potential instability caused during construction or modification of a dam. The probability of failure can be used to prioritize dam safety activities within a risk context. The risk reduction benefits in terms of money spent per unit reduction in failure probability of risk can provide a useful measure for comparing dam safety modification alternatives.
 
Physical Properties of Plastic Pipe Used in Reclamation Toe Drains
DSO-09-01Jay Swihart and Mark PabstDSO-09-01 
A significant number of plastic pipes have either collapsed or suffered some other type of poor performance in toe drains of embankment dams. To better understand the comparative strength relationships between different plastic pipe types, a laboratory study was undertaken with an emphasis on perforated pipe. A variety of material types, pipe sizes, and perforated versus non-perforated pipes were load tested. This report presents those results as well as recommendations on preferred plastic pipe for use in toe drain applications.
 
Determining Erosion Indices of Cohesive Soils with the Hole Erosion Test and Jet Erosion Test
DSO-08-05Tony L. WahlDSO-08-05 
Two methods of soil erodibility testing, the hole erosion test (HET) and submerged jet erosion test (JET), were investigated to determine the correlation between their results and to evaluate and improve them for potential application to the modeling of embankment dam erosion and breach processes. Basic assumptions regarding the behavior of the friction factor for flow through the predrilled hole in the HET were investigated, and it was found that the friction factor was best correlated with the hole diameter rather than the test time as previous investigators had assumed. This finding and others were used to develop improved HET testing and data analysis procedures including a method that does not require measurement of the final eroded hole diameter. The HET and JET methods were compared to one another by using them to determine erodibility parameters of identically prepared remolded soil specimens. The JET method indicated much greater erodibility in a direct comparison of quantitative results, which indicates that results of each test should be interpreted using criteria adapted to each particular test. Differences in erodibility were one or more orders of magnitude in erosion rate and two or more orders of magnitude in critical shear stress. The JET method seemed to be more sensitive to variations in soil fabric, and specimens with a coarse and nonuniform soil structure seemed to produce the greatest differences between HET and JET results. Differences between HET and JET results are also thought to include simplified stress descriptions in each test environment and fundamental differences in the mechanisms of erosion exploited by each test. The JET proved to be a more easily applied test method, with a higher ratio of successful tests and a greater ability to successfully test soils of widely varying erodibility. The JET also has the advantage of being suitable for in situ field testing wherever a soil surface of interest can be exposed. Ultimately, selection of a test for any particular purpose should be made primarily based on the application and the erosion mechanisms of importance.
 
Non-Linear Analysis Research
DSO-08-02Barbara Mills-BriaDSO-08-02 
To study the effects of hourglassing parameters particular to LS-DYNA, a finite element model was generated and analyzed for static loading conditions. A number of internal variables within the program were determined to be likely key players in the hourglassing behavior of the model and were varied to determine their effects on the model as a whole, as well as the individual materials of which the model was composed. Results indicate that the application of traction loads as seismic input more closely matches the target surface motions. A comparison of maximum and minimum values of velocity and acceleration as well as frequency content leads to this conclusion. The comparison of ASI values indicates that any structure modeled on the surface of the foundation would be subject to excess energy during a seismic analysis.
 
Development of a Regional Paleoflood Approach
DSO-08-01Jeanne GodiareNot Available 
A regional paleoflood investigation was undertaken during Fiscal Year 2005 and 2006 as part of Dam Safety Office Research to develop regional paleoflood data in Northeastern Utah. The objective of this study is essentially to determine whether regional paleoflood data can be correlated using variables such as basin area, paleoflood peak discharge estimates, and paleoflood age estimates and to explore the variability of paleoflood data in basins within a particular region. Soil descriptions, radiocarbon analysis and hydraulic modeling were used to develop paleoflood information at sites along the Wasatch Front, Strawberry River Basin, High Uintas, and Price River Basin. In the Wasatch Front, previous detailed studies and reconnaissance level studies were reviewed and analyzed. Six new sites in the Strawberry River Basin, three new sites in the High Uintas and six new sites in the Price River Basin were developed during the course of this study. The hydrologic record of stream gaging stations with drainage areas less than 100 mi2 in the Wasatch Front and High Uintas show that the majority of annual peak flows are during the spring and early summer and are largely related to snowmelt runoff. A smaller number of annual peak flows happen during the late summer and are related to thunderstorms at lower elevations in the area. Variability in the hydrologic records is greater for the Strawberry River Basin and Price River Basin when compared to the Wasatch Front and High Uintas, presumably because these basins have a greater mix of thunderstorm-generated floods and snowmelt floods.
 
Underwater Crack Repair
DSO-07-10Tim DolenDSO-07-10 
Leakage through cracks in concrete dams and seepage through voids in embankments have both operational and dam safety concerns. Most corrective actions require draining the reservoir to inject various chemical and/or cement base grouts on a pattern of drill holes or placement of protective membranes. This is expensive due to the loss of water and the time to perform repair operations. Three attempts have been made to eliminate the leakage through several thermal cracks at Upper Stillwater Dam, Utah. The cracks were caused by thermal gradients in mass roller-compacted concrete (RCC) during the winter months after placement in 1986 and 1987. The RCC dam was designed without formed contraction joints, allowing transverse cracking instead with the intent to grout cracks after they formed. However, filling the cracks with a durable material has proved to be very difficult in practice. Two attempts to grout the cracks with chemical grouts were temporarily effective. But the chemical grouts deteriorated over time due to continued thermal movement essentially grinding the seal. A third repair involving a combination of chemical grouts and embedded steel barriers is under way at an estimated final cost of about $6 million (Reclamation, 2002).
 
Evaluation of In Situ Methods for Liquefaction Investigation of Dams
DSO-07-09David Gillette and William EngemoenDSO-07-09 
This study uses the literature on liquefaction in dams to compare and contrast the various in situ methods of determining embankment dams' susceptibility to liquefaction. The study concludes taht three soil penetration tests usually used to measure soil density, the principle factor in liquefaction, predict liquefaction fairly reliably when used in combination with other tests and with good data and sound engineering judgement. The study lists advantages and disadvantages of each method and points out conditions where a method may be better (or less) suited to particular site conditions.
 
High Velociy Uplift
DSO-07-07Warren FrizzelDSO-07-07 
This report describes laboratory tests that were completed to extend the available data on uplift pressures generated by high velocity flows over offset joints. In addition, magnitudes of flows into the joints or cracks were measured for various configurations. Lab data were supplemented with numerical modeling using the computer program, Flow-3D. Uplift pressure and flow rate data from this study will be used to reduce the uncertainty in analyses used during risk assessments to provide improved estimates of the level of effort required to bring a spillway or outlet works into a safe operating condition for a variety of geometries.
 
Static and dynamic strength of shear keys in concrete dams
DSO-07-05Larry NussDSO-07-05 
The objectives of this study are to determine how much strength and resistance the shear keys develop relative to the typically implemented finite element (FE) models of concrete arch dams. This study will give risk teams a more defensible basis when judging if an arch dam maintains its arch action during seismic events that move the dam upstream and open contraction joints. This study will also help risk assessment teams better estimate probabilities of failure for arch dams with and without shear keys. This report proposes a method that assesses the behavior in the joint to provide further evaluation on the stability of arch dams with and without shear keys. Generically, the proposed method compares the behavior of the typically implemented contraction joint model with known joint behavior and geometry. The known joint behavior is based on past research on shear key failure (Kaneko et al., 1993a, 1993b), sliding in rough, undulating rock surfaces (Hoek and Bray, 1981; Ladanyi and Archambault, 1970), and the behavior of an FE model of typical shear keys that was developed as part of this study. The proposed method provides a means of assessing the effects of contraction joints without changing the current methods utilized or increasing the complexity of the current methods utilized for structural evaluation of concrete arch dams.
 
Use and Mis-use of Rayleigh Damping
DSO-07-03Larry NussDSO-07-03 
Rayleigh damping is commonly used to provide a source of energy dissipation in analyses of structures responding to dynamic loads such as earthquake ground motions. In a finite element model, the Rayleigh damping matrix consists of a mass-proportional part and a stiffness-proportional part; the latter typically uses the initial linear stiffness matrix of the structure. Under certain conditions, for example, a nonlinear analysis with softening nonlinearity, the damping forces generated by such a matrix can become unrealistically large compared to the restoring forces, resulting in an analysis being unconservative. Potential problems are demonstrated in this paper through a series of examples. A remedy to these problems is proposed in which bounds are imposed on the damping forces.
 
Perfectly Matched Layers
DSO-07-02Larry NussDSO-07-02 
The modelling of linear wave propagation on unbounded domains is of interest in various fields of both science and engineering. It is especially of interest in the earthquake analysis of dams because the foundation rock and impounded water may be modelled as unbounded domains undergoing wave motion generated by the motion of the dam. One approach to the numerical solution of a wave equation on an unbounded domain uses a bounded domain surrounded by an absorbing boundary or layer that absorbs waves propagating outward from the bounded domain. A perfectly matched layer (PML) is an absorbing layer model that absorbs, almost perfectly, all waves incident upon it. This report develops the concept of a PML for elastic and acoustic waves using some of the insights obtained in the context of electromagnetics and presents PMLs for (1) a rod on elastic foundation, (2) acoustic waves in two and three dimensions, and (3) elastic waves in two and three dimensions. Furthermore, this report develops displacement-based finite-element implementations for the PMLs, both in the frequency-domain and the time-domain. In particular, an efficient finite-element implementation suitable for explicit integration is presented for the three-dimensional elastic PML, thus allowing the solution of realistic three-dimensional problems without the overhead of solving a large system of equations at each time step.
 
Fault-Based Probababilistic Seismic Hazard Analysis Software
DSO-07-01Roland LaForgeDSO-07-01 
Probabilistic Seismic Hazard Analysis (PSHA) is the preferred technique for translating seismic hazard into ground shaking characterizations that engineers can use for testing structures and incorporating into risk assessments. The results are presented in terms of the annual probability of exceedance (APE) of a ground motion parameter such as peak horizontal acceleration, damped peak spectral acceleration at some oscillator frequency, or peak horizontal velocity. A uniform hazard spectrum for a specified annual probability of exceedance or return period is often used to estimate the probability of damage determined from a dynamic analysis of the structure. Software developed at Reclamation for performing PSHA is described in this manual. The main categories are programs for generating APE data for fault and areal seismic sources, postprocessing these results in a logic tree format, and programs for other, more general seismic hazard analysis.
 
Investigation of the Failure Modes of Concrete Dams - Physical Model Test
DSO-06-03Dave HarrisDSO-06-03 
This investigation focuses on two case histories to develop failure modes under loadings similar to earthquakes. The models are intended primarily to produce cases which can be compared to computer analyses. The first is a 1/50th scale model of the Koyna Gravity dam section, which cracked during an earthquake. This is a simpler model using a 2-dimensional cross section and no water in the simulation. The second model is a simulation of an arch dam in a wide canyon with a reservoir in flight. Conclusions are for trends observed are: (1) The Koyna Model gave results similar to previous studies and to what actually happened in the field. The 3-D arch dam model compares with previous models and linear measurements such as response frequencies made in the field. (2) All models show the onset of sudden cracking and pronounced structural nonlinearity following cracking. This nonlinearity is characterized by the bottom of the dam being able to slip back and forth beneath the top of the dam. (3) The arch dam model demonstrates a critical acceleration of 0.70 g's for first cracking of this specific model independent of joints in the models. (4) The crack pattern in the models is dominated by the joint patterns. (5) The time to final full failure when converted to full scale times exceeds the duration of any recorded earthquake. (6) Final failure is a push through of the dam into downstream. This failure mechanism appears to require abrasion in the joint before it can be established. (7) Water does pass through cracks in the model in approximately1/10th of a second.
 
LS-DYNA vs. DYNA3D Benchmark and Validation Testing
DSO-06-02Kevin O'SheaDSO-06-02 
This report summarizes the basic results comparing program performance and stability between DYNA3D and LS-DYNA as well as the validation and benchmark testing of LS-DYNA. The finite element models (FEM) for Parker Dam and Black Canyon Diversion Dam were selected to be used as the test cases in comparing DYNA3D and LS-DYNA due to their size and contact surface complexity. The load application in this model is as follows: (1) Gravity and forces are applied from 0 to 3 seconds. (2) The model stabilizes from 3 to 6 seconds. (3) The ground motion begins at 6 seconds and runs until 20 seconds. Both DYNA3D and LS-DYNA are highly complex programs that produce valuable results for structural analysis issues. Several problem areas have been examined and compared using both codes.
 
Computer Material Models for Soil, Rock, and Concrete using FLAC and DYNA
DSO-06-01Dave HarrisDSO-06-01 
This report summarizes material models available for the analysis of soils in the computer codes, FLAC and DYNA. The input constants needed for these models are emphasized. Typical values or ranges of values are given, and methods to calculate the parameters from laboratory data are discussed. In addition, dynamic damping in computations is discussed. Two components, mass (structural inertia) and stiffness (material hysteresis), are discussed. Field-measured values are presented and methods to calculate material damping in the lab shown. The report concludes that nonlinear analyses can be used to better understand complex site response of dams; models are available for these analyses. It is recommended that measured values from the structure in question be used for analyses.
 
Effect of Spatially Varying Earthquakes on Concrete Gravity, Arch, and Slab-and-Buttress Dams
DSO-05-06Barbara Mills-BriaDSO-05-06 
The objective of this research is to study the effects of spatially varying ground motions and, if necessary, incorporate the capability of varying ground motion input into the analysis procedures used by Reclamation's Structural Analysis Group.
 
Materials Properties Model of Aging Concrete
DSO-05-05Tim DolenDSO-05-05 
A database model of aging concrete was developed to identify the changes in materials properties over time for Bureau of Reclamation (Reclamation) mass concrete dams. Materials properties data on mass concrete were input to the Reclamation Aging Concrete Information System (ACIS). The data were analyzed for trends in the deterioration of concretes subject to aging, including alkali aggregate reaction (AAR) and general aging of early twentieth century concrete with high water-to-cement ratios. The aging concretes were compared to dams of similar age, but not suffering from aging processes. The aging concretes were also compared to known good quality concretes that were manufactured after about 1948 to specifically resist deterioration from AAR, freezing and thawing (FT), and sulfate attack. Trends were established for comparing the compressive strength, splitting and direct tensile strength, and elastic properties of aging and non-aging dams. The strength and elastic properties of aging mass concrete differed significantly from those of comparable non-aging concretes. Both spatial variations within a structure and long-term changes in strength and elastic properties were identified. The ACIS database can be used to track the long-term materials properties behavior of dams through comprehensive concrete coring and testing programs. Laboratory core test data are included for dams ranging from about 10 to more then 83 years old. These data provide the necessary supporting documentation for the Dam Safety Office Comprehensive Facilities Review evaluation process and for dams in need of corrective action.
 
Guidance on Sampling, Transportation, and Analysis of Materials
DSO-05-04Doug HurcombDSO-05-04 
This report is a practical guide on how to sample, transport, and analyze sediments samples from the drains of concrete and embankment dams and from other hydraulic structures.
 
Seepage Chemistry Manual
DSO-05-03Doug CraftDSO-05-03 
Gypsum, anhydrite, calcite, dolomite, and halite are soluble minerals that are common in the western United States where the Bureau of Reclamation has constructed many dams. Dams sited on foundations and abutments containing soluble minerals have the potential to develop seepage problems that require monitoring by water resource managers. When mineral dissolution is suspected at a dam, seepage samples may be collected, analyzed and compared to reservoir water to help determine whether soluble minerals pose a structural safety problem. Seepage chemistry investigations are interdisciplinary and require collaboration among chemists, geologists, engineers, and geophysicists. This peer reviewed technical report summarizes the basic chemistry associated with mineral dissolution, weathering, biotic processes and mixing, all of which may contribute to changes in seepage chemistry during structural transit. This report also provides references and good practice procedures to guide seepage chemistry investigations, and includes examples from successful seepage investigations performed over the past 20 years by Bureau of Reclamation Dam Safety Program professionals.
 
Spillway, Outletworks, and Water Conveyance Structure Needs - Survey Results
DSO-05-02Kathy FrizellDSO-05-02 
This report is the result of the research project entitled "Evaluation and Innovative Solutions to Reclamation Spillway and Water Conveyance Structure Needs." The goal of the research project was to define common problems with Reclamation's hydraulic structures and determine potentially innovative and cost-effective solutions to the identified needs. This report summarizes the results of the survey that was conducted under the research project. In addition, the report summarizes some of the current repairs under way on specific projects and provides ideas for future research beyond the scope and funding level of this project.
 
Improved Flood Frequency Extrapolation Procedures
DSO-05-01John EnglandDSO-05-01 
This research focuses on new methods to estimate extreme floods for dam safety and hydrologic engineering. Physically-based, distributed watershed models are used as an avenue to estimate extreme floods, and as a basis to extrapolate frequency curves. The main elements of this research include improving and using a physically-based rainfall-runoff model to estimate extreme floods and probabilities for dam safety on a large watershed. The test watershed is the Arkansas River above Pueblo, Colorado.
 
Resolution of Crosshole Shear-Wave Testing
DSO-04-10Lisa BlockDSO-04-10 
This project investigates the ability of crosshole shear-wave testing to resolve low-velocity layers of varying thickness. The resolution of crosshole shear-wave testing, as currently implemented by Reclamation, is investigated by generating and analyzing synthetic waveform data for seven shear-wave velocity models. The models include velocity layers of varying thickness and velocity contrast. Three experienced geophysicists from the Bureau of Reclamation, U. S. Geological Survey, and U. S. Army Corps of Engineers independently estimated the shear-wave arrival times on the synthetic waveforms. Velocity-depth profiles were computed from the arrival times using Reclamation's standard crosshole processing procedures. The thin-bed resolution limit of the crosshole shear-wave method is controlled mainly by the frequency content of the propagated seismic energy. Tests conducted during this investigation indicate that incorporation of ray bending into the data processing procedures has little effect on the resolution of velocities within layers less than two feet thick. The most effective way to improve the resolution of Reclamation's crosshole shear-wave surveys would be to use a higher-frequency borehole shear-wave source. These tests also demonstrate that computed crosshole velocities are smeared at layer interfaces, with velocities being overestimated in the lower-velocity layer near an interface and velocities underestimated in the adjacent higher-velocity layer. The smearing at layer interfaces is caused by refraction of seismic energy and the finite frequency range of the propagated seismic energy. Because of this smearing near layer interfaces, the most accurate crosshole velocities are those computed near the centers of velocity layers. The exact depths of layer interfaces are better determined from geologic drill logs and geophysical borehole logs rather than from crosshole velocity profiles.
 
Guidance on Sampling, Transportation, and Analysis of Materials in Drains
DSO-04-09Doug HurcombDSO-04-09 
The topic of monitoring seepage sediments in dams is a reoccurring one. There is no guidance on how to sample, transport, and analyze sediment samples from drains in any current Reclamation document. Examples of sediments that may be present in a drain are soils, clay to sand size sediments, biological growths and films, and precipitates. Investigators often need guidance to estimate costs and specify sampling procedures for dam safety monitoring programs. Without guidance, investigators may sample critical material and handle it inappropriately, which requires re-sampling and loss of time and resources. This brief guidance document provides advice on how to effectively use the Technical Service Center (TSC) services to help you determine what is fouling a drain. This report includes a discussion of drains, inspections, materials, sampling, transportation, testing, costs, and TSC contacts. This document should be used as a practical guide as the title suggests. It should not be considered complete or a definitive dissertation on microbiology and sampling.
 
Hydraulic Hazard Curve Estimating Procedures
DSO-04-08Robert Swain, John England, Kenneth Bullard, David RaffDSO-04-08 
The objective of this project is to establish a prescriptive procedure for generating hydrologic hazard curves for use in dam safety evaluations. These curves can then be used for risk analysis and prioritization of further work at Bureau of Reclamation (Reclamation) dams and other U.S. Department of the Interior facilities. Hydrologic hazard curves are defined as graphs of peak flow and volume (for specified durations) versus Annual Exceedance Probability (AEP). The range of AEPs that are displayed on these graphs is from 0.99 to 0.00000001 (1 x 10^8 or 100 million years). Reclamation has developed an approach toward developing hydrologic hazard curves for use in evaluating dam safety issues. The procedure relies on extracting information from existing studies to the fullest extent possible. The procedures and analysis techniques defined in this report allow for the possibility, and even plausibility, that peak discharge and volume estimates may exceed the probable maximum flood (PMF). This is a function of the uncertainty and inconsistency among and between analysis techniques. Therefore, in these cases, the PMF is believed to represent the upper limit to hydrologic risk. This report recommends that the approach for developing hydrologic hazard curves consider the dam safety decision criteria, potential dam failure mode and dam characteristics, available hydrologic data, possible analysis techniques, resources available for analysis, and tolerable level of uncertainty.
 
Investigation of the Ability of FIlters to Stop Erosion Through Cracks in Dams
DSO-04-06Youngjin Park, Tom Brandon, Mike Duncan DSO-04-06 
Since Terzaghi (1922) developed grain size criteria for granular soils in dam filters, many researchers have studied embankment dam filters. The major function of the filter is to prevent erosion and piping. In order to have this ability, filters must restrain the particles of the protected soil (the base soil) and allow water to pass freely out of the base soil. Sherard et al. (1984) modified these criteria for cohesive soils, and developed the concept of "critical filters," that can prevent erosion even under the severe condition where the base soil is cracked, and where concentrated flow occurs through the crack. In addition to grain size criteria that ensure restraint of the base soil while allowing free passage of water, a filter must also be graded so that the filter itself will not crack. To ensure that filters will not support cracks, most current filter gradation criteria require that no more than 5% of the filter material should be finer than the #200 sieve, and that the fines within the filter should be non-plastic. However, it is not clear that this criterion is sufficient. At Ochoco Dam shown in Figure 1.1, a sinkhole developed in a filter that was designed to have a maximum of 3% passing the #200 sieve. This incident at Ochoco Dam gave rise to renewed interest in filter criteria, and resulted in sponsorship of the research described in this dissertation. This research was designed to investigate the crack-preventing and crack-stopping abilities of filters, and to develop criteria that can be relied upon to ensure that a filter will perform its essential function even when subjected to deformations that cause cracks in the adjacent core.
 
Case Histories of Dam Failures and Accidents Caused by Internal Erosion
DSO-04-05David MiedemaDSO-04-05 
The vast majority of embankment dams have exhibited good behavior. However, about 50 percent of large embankment dam failures have been attributed to internal erosion. Therefore, a study of case histories of incidents (both accidents and failures) can be instructive. The purpose of this report is to aid in the risk analyses, in comprehensive facility reviews, and in decisions about modifying existing Bureau of Reclamation (Reclamation) embankment dams. The goal has been to collect some helpful case histories of dam failures and accidents caused by internal erosion. Also, a few case histories of Reclamation's response to piping incidents at their dams and at one Bureau of Indian Affairs dam have been included. A careful review of a few case histories of dams that are similar to the one under study can result in better assessments of possible failure mechanisms and insights into factors that can contribute to satisfactory or poor performance of embankment dams.
 
Small-Scale Permeameter Test to Determine Compatibility of Pipe Wall Perforations, Geotextile Socks, and Sand/Gravel Envelopes
DSO-04-04Jay SwihartDSO-04-04 
Previous studies using a full-scale pipe box test have shown that a geotextile sock can significantly improve performance of perforated drain pipe by simultaneously increasing inflow rates and decreasing soil loss (table ES-1). Unfortunately, these full-scale tests are timeconsuming and expensive. This study looks at a small-scale permeameter test (Modified Gradient Ratio) as a possible replacement for the full-scale pipe box test.
 
Evaluation of Protective Filter Erosion Boundaries
DSO-04-03Chris SlavenDSO-04-03 
The objectives of the experiments described in this paper are to provide insight into the selfhealing capabilities of protective filters used in embankment dams and to verify the results and reproducibility of previous research. Existing dams may have filters that do not meet Sherard's No Erosion Filter (NEF) criteria; however, the filters may still perform adequately after "some" erosion of the base material into the filter. The research documented in this paper will focus on the soil retention aspect of protective filters.
 
Program RSVR2 User's Manual
DSO-04-02Author UnknownDSO-04-02 
RSVR2 is a modification of the RSVR program which originated from Kuo (J. S.-H. Kuo, Fluid- Structure Interactions: Added Mass Computations for Incompressible Fluid, Report No. UBC/EERC-82/09, Earthquake Engineering Research Center, University of California, Berkeley, California, August 1982).
 
US Bureau of Reclamation Digital Photogrammetry Research Report
DSO-04-01Peter Shaffner, Lisa Krosley, Joseph KottenstetteDSO-04-01 
Based on this completed phase of the research program, the Bureau of Reclamation (Reclamation) now has the capability to use digital photographs to map geologic discontinuities at dam sites. Several computer software programs and digital cameras were evaluated. Software designed specifically for mapping geologic features was purchased from the Commonwealth Scientific Industrial and Research Organization (CSIRO) in Australia. The SiroVision software consists of two integrated components: Siro3D, which allows for the creation of a three-dimensional image; and SiroJoint, which allows users to measure mapped parameters defining discontinuities using 3D images. The selection of this particular software over several more robust programs available on the market was based on several variables, including cost, ability to map geologic features directly within the program, and ease of use. There are several software products on the market that may produce higher accuracy and may be more useful for other applications, but SiroVision was a logical first choice for geologic mapping because the product is much faster to learn and does not require additional software to produce stereonets to display discontinuity orientations. Coupled with a Nikon D100 digital camera, SiroVision was used to map the discontinuity orientations at three locations, including East Canyon Dam. Initial problems with accuracy and processing were reduced significantly following separate field measurements at test sites. This testing allowed corrections to the data acquisition process as more was learned about the sensitivity of the software to various camera parameters, such as tilt and convergence, and more was learned to improve computer processing. The research testing allowed many iterations of the data processing during this learning period, with accuracy improvements each time.
 
Effects of Concrete Deterioration on Safety of Dams
DSO-03-05Tim Dolen, Gregg Scott, Kurt von Fay, Bob HamiltonDSO-03-05 
The objective of this project was to determine the influences of concrete deterioration on the safety of the Bureau of Reclamation (Reclamation) structures and how deterioration can impact Reclamation's risk evaluation process. In addition, tasks needed to accomplish the objective of evaluating the risks as a function of time for deteriorating concrete structures are identified. A major component of this report is an evaluation of models to predict the influences of various deterioration mechanisms on the strength and elastic properties of concrete. Although the focus of this report relates to risk assessment, we envision that models to predict concrete properties resulting from ongoing deterioration would be very useful for operations and maintenance planning and budgeting. The results of this report can serve as a basis for evaluating future research proposals to develop specific tools for use in evaluating risks associated with concrete deterioration.
 
Stochastic Modeling Methods
DSO-03-04John EnglandDSO-03-04 
The purpose of this project is to perform a detailed analysis and review of stochastic rainfall/runoff models, including those currently used by Reclamation; compile and review other stochastic models available and evaluate their applicability to the BOR Dam Safety program objectives. The MGS stochastic model is one tool currently used by DSO to determine hydrologic risk. This study will allow the Probabilistic Flood Hazard Cadre to identify the strengths, weaknesses, and limitations of the MGS model and other stochastic models that will enable DSO managers to effectively and correctly apply the information. It is important that paleoflood data, one of several sets of data used by DSO to evaluate hydrologic risk, can be incorporated into these stochastic models. The study tasks are to: undertake an extensive search, review, compilation, and evaluation of other stochastic rainfall/runoff models to better understand the current state-of-knowledge. In addition, a detailed analysis and review of the MGS stochastic model shall be performed. The detailed analysis of the MGS model will include review by independent experts. Important aspects of the review shall address issues of uncertainty, sensitivity analysis, and plausibility of incorporating paleoflood data into the stochastic model.
 
Probabilistic Extreme Flood Hydrographs That Use PaleoFlood Data for Dam Safety Applications
DSO-03-03John EnglandDSO-03-03 
Extreme flood hydrographs are needed to evaluate dam safety issues for situations where the reservoir inflow peak discharge is greater than the maximum spillway capacity, the reservoir has large surcharge storage, and/or the reservoir has dedicated flood control space. Flood runoff hydrographs integrate the drainage basin and channel response to precipitation, given some initial, variable state of moisture throughout the watershed. To conduct risk analyses and dam safety evaluations, probability estimates for extreme floods are required. Probabilistic extreme flood hydrographs are developed to assess the adequacy of the spillway and reservoir flood/surcharge space to temporarily store a portion of the flood volume, and to attenuate or pass the hydrograph peak without overtopping the dam. The hydrographs and probability estimates are used in risk analyses for dam safety. These hydrographs can also be used to establish reservoir operating rules and determine diversions needed for construction. This report documents statistical and rainfall-runoff techniques used to develop probabilistic extreme flood hydrographs for dam safety risk analyses. A probabilistic extreme flood hydrograph is defined in this report as one that preserves a peak discharge exceedance probability and dependence between volume and peak for a fixed duration. An extreme flood is considered to have an Annual Exceedance Probability (AEP) of 0.005 or less. There are many methods of estimating extreme flood runoff hydrographs, such as unit hydrograph approaches (e.g., Chow et al., 1988), continuous rainfall-runoff modeling (e.g., Bradley and Potter, 1992), and statistical techniques (e.g., USACE, 1975a). These and other extreme rainfall-runoff methods are discussed in-part by NRC (1988), Pilgrim and Cordery (1993), and Nathan and Weinmann (1999). Other than a method presented by Nathan and Weinmann (1999), these references do not describe methods that can be used by practitioners to estimate both extreme floods (peak, volume, duration) and associated probabilities for dam safety. The methods and examples presented in this report are an attempt to bridge part of that gap.
 
Stochastic Event Flood Model Improvements and Extreme Storm Analyses for A.R. Bowman Watershed
DSO-03-02John EnglandDSO-03-02 
The research on improving SEFM and analyzing extreme storms was motivated by three developments: (1) results from the 1997 Reclamation-sponsored workshop in Logan, Utah (USBR, 1999); (2) the successful work and modeling by MGS using SEFM for Reclamation (Schaefer and Barker, 1997; 1998); and (3) the creation of the Reclamation Flood Cadre in 1999 to coordinate and develop extreme flood methods. The Logan workshop highlighted the continued need for development and research in extreme floods. The focus was on ultimately developing practical hydrological modeling tools that the output would be used in risk analysis.
 
Causey Dam, Utah, 1,000-Year Rainfall-Runoff Report
DSO-03-01Monica NorvalDSO-03-01 
A research proposal to develop the 1,000-year thunderstorm (local) hydrograph for Causey Dam was submitted and approved by Dam Safety for fiscal year 2002. The study purpose was to use rainfall-runoff modeling procedures to compute both peak discharge and volume amounts. These results were then compared with previous frequency peak estimates at Causey Dam that used peak streamflow discharge estimates and paleoflood information from the Ogden River drainage basin. The study used the National Weather Service (NWS) draft precipitation values for the southwest region that include the 1,000-year return period. Rainfall-runoff modeling was done using these precipitation values to develop the 1,000-year thunderstorm hydrograph for Causey Dam. The thunderstorm event was chosen to eliminate snowfall as a variable in the computation.
 
Seismic Tomography of concrete structures Phases II & III
DSO-02-03Lisa BlockDSO-02-03 
Between 1997 and 2001, the Seismotectonics and Geophysics Group at the Bureau of Reclamation conducted a research project on the application of the seismic tomography method to investigation of large concrete structures. The hardware and software needed to efficiently acquire high-quality seismic tomography data on large concrete structures were developed. This hardware and software, as well as the effectiveness of the seismic tomography method as applied to concrete structures, was evaluated by acquiring and processing seismic tomography data from a concrete arch dam undergoing alkali-aggregate reaction. A total of approximately 13,700 seismic waveforms were acquired across three upstream-downstream cross sections during 9 days of data acquisition. Seismic ray paths ranging from horizontal to approximately 70 degrees from horizontal were obtained through most of each cross section, with excellent signal-to-noise ratio on most seismic traces. The P-wave velocity tomograms computed from these data show velocities ranging from 16,000 ft/s to less than 7,000 ft/s. The areas with the lowest P-wave velocities are interpreted as regions with the most severely deteriorated concrete. The pattern of P-wave velocity variations is consistent with information obtained from core analyses, fracture mapping, and deformation studies, and provides valuable additional information about the spatial distribution of concrete deterioration.
 
Investigation of the Failure Modes of Concrete Dams - Physical Model Tests
DSO-02-02Dave HarrisDSO-02-02 
This report describes the design and testing of physical models of concrete arch dams. Two series of models are included. In the first series, a planar, two-dimensional model which represents a 1/50 scale model of the Koyna Dam in India is tested. The model is tested without water behind the dam to utilize conditions as simple as possible for comparison with numerical modeling. Cracks formed in the test are similar to cracks which resulted on the Koyna Dam following an actual earthquake event. The second series of models are approximately 1/150 scale of a typical wide canyon dam. Material properties are adjusted to the model scale; a reservoir and foundation are included. The models utilize different joint patterns of a monolith, single horizontal and vertical joints, and multiple vertical and horizontal joints. Models are tested to collapse of the structure. Conclusions are that initial cracking of the structure is not influenced by the joints. Final collapse is a push through of sections of the dam downstream and is controlled by the joints both with the pattern of the failure and with the acceleration required to fail the structure. Water is observed as passing through the model during the loading.
 
Shake Table Study to Investigate Failure Modes of Arch Dams
DSO-02-01Terry PagneDSO-02-01 
Determining the stability of concrete dams using linear elastic finite element analyses can be very subjective when the results indicate nonlinear response. To reduce uncertainty associated with failure of concrete arch dams during an earthquake, large laboratory shake table tests of a representative arch dam and reservoir were performed at the U.S. Bureau of Reclamation's Material Engineering and Research Laboratory. These models were analyzed using the ABAQUS finite element computer code to determine how accurately the response could be modeled and the failure predicted. Failure modes predicted by the computer analyses corresponded well with the laboratory tests.
 
Spillway Gate Failure of Misoperation Case Histories
DSO-01-01Wayne GrahamDSO-01-01 
The failure or misoperation of spillway gates has the potential to cause injury and loss of life. This report discusses the various mechanisms that may lead to the sudden and unexpected failure or opening of spillway gates. Downstream flooding is possible from any of the following: spillway gates cannot be opened; spillway gates open accidentally; spillway gates are opened during a flood; spillway gates fail structurally; debris impeded spillway flow; spillway gates are operated incorrectly. Case studies are used to describe what went wrong, what happened as a result, and what could have been done to prevent the failure or misoperation. This report also contains guidance on estimating the consequences of spillway gate failure or misoperation and discusses preventative measures and design options for existing dams.
 
Nonlinear Response of Steel Beams
DSO-00-01Terry PagneDSO-00-01 
This paper presents the results preliminary research completed toward the development of plastic failure criteria for components of large fixed wheel spillway gates subjected to seismic loads. Nonlinear finite element analyses were used to model the local buckling mode of failure in plastic hinge regions of beams subjected to cyclic loading. Material testing data from a Stanford University Research Project was used for comparisons to analysis results.
 
Full-Scale Laboratory testing of a Toe Drain with Geotextile Sock
DSO-99-14Jay SwihartDSO-99-14 
This report describes the full-scale laboratory testing (pipe box testing) of a 15-inch-diameter, corrugated, polyethylene toe drain with a knitted geotextile sock, backfilled with a sand envelope material. The test results are compared with previous small-scale and full-scale tests using perforated pipe with 1/8-inch and 1/4-inch perforations, but no geotextile. Use of the geotextile optimized toe drain performance both with respect to flow and with respect to loss of the sand envelope. The long-term flow rate was 21 gpm per linear foot of pipe, which was significantly higher (by a factor of 3 to 12) than the earlier tests without geotextile. The total loss of sand envelope was only 50 grams per linear foot of pipe, which was significantly lower (by a factor of 4 to 17) than the earlier tests without geotextile. The test with geotextile was run for 31 days at a constant head of 2½ feet above the pipe invert with no indication of clogging. Based on these results, use of geotextile sock in conjunction with a sand envelope is recommended for all future toe drain installations in areas with fine native soils. Use of the geotextile sock will improve toe drain performance by increasing flow rates and decreasing loss of envelope material. The experience gained with the use of geotextiles in toe drains will also improve Reclamation's ability to use geotextiles in other applications.
 
A Procedure for Estimating Loss of Life Caused by Dam Failure
DSO-99-06Wayne GrahamDSO-99-06 
Risk assessments and other dam safety studies often require that an estimate be made of the number of fatalities that would result from dam failure. To assist in this effort, an extensive evaluation of dam failures and the factors that contributed to loss of life was conducted. Every U.S. dam failure that resulted in more than 50 fatalities and every dam failure that occurred after 1960 resulting in any fatalities was investigated with regard to warning, population at risk (PAR) and number of fatalities. These dam failure data are used to provide a historical perspective of the risk associated with the U.S. dam inventory. Loss of life resulting from dam failure is highly influenced by three factors: 1)The number of people occupying the dam failure flood plain, 2)The amount of warning that is provided to the people exposed to dangerous flooding and 3)The severity of the flooding. The procedure for estimating loss of life due to dam failure relies heavily on data obtained from U.S. dam failures. The procedure is composed of 7 steps: 1) Determine dam failure scenarios to evaluate. 2) Determine time categories for which loss of life estimates are needed. 3) Determine when dam failure warnings would be initiated. 4) Determine area flooded for each dam failure scenario. 5) Estimate the number of people at risk for each dam failure scenario and time category. 6) Apply empirically-based equations or methods for estimating the number of fatalities. 7) Evaluate uncertainty.
 
Full-Scale Toe Drain Test, Many Farms Dam
DSO-99-05Jay SwihartDSO-99-05 
Full-scale laboratory testing was conducted on the proposed toe drain design for Many Farms Dam. The toe drain design calls for 10-inch single-wall corrugated high-density polyethylene (HDPE) drain pipe with 1/16-inch wide slotted perforations, covered with a knitted geotextile sock, and backfilled with a fine sand envelope. The geotextile sock has an apparent opening size (AOS) of #30 sieve (0.6 mm), and is placed on the pipe at the factory. The 1/16-inch slotted perforations are intended to clog with sand particles if the geotextile sock should tear. However, the slot width could potentially double if the drain pipe elongates significantly during installation. The test was conducted for 13 days, and the outflow quickly stabilized at 7.3 gallons per minute (gpm) per foot (see figure 1). The envelope loss was stabilizing at about 1,000 grams per linear foot, which was higher than expected, but believed acceptable. This loss equates to a calculated thickness loss of 0.087 inches around the pipe circumference. After testing, the pipe was carefully exhumed, and a coarser natural soil filter had formed adjacent to the geotextile sock, measuring ½ to 1 inch thick. Because of the loss of 1,000 grams of sand envelope per foot of toe drain, a coarser more broadly graded envelope material has been proposed. Traditional two-stage filters use a gravel envelope surrounded by a sand filter, and require extensive bench-cut (open-cut) excavation. Use of the geotextile sock in lieu of the gravel envelope allows construction with trenching equipment, which can offer significant cost savings. The knitted geotextile sock used in this study is the only geotextile that is preinstalled on the pipe at the factory. The knitted sock is only available with AOS of #30, which limits design options. Other geotextile products (such as monofilament woven) would have to be attached to the pipe in the field, but are available in a wide range of AOS.
 
Earthquake Response Records
DSO-99-04Structural Analysis GroupDSO-99-04 
This study was completed to examine typical acceleration responses at the top of concrete dams subjected to earthquake loadings. Comparisons were made between the input accelerations applied to finite element models and the accelerations generated at the dam crests during the dynamic analyses. This information will be instrumental in determining the number of cycles which would impact a spillway gate at the dam crest level during and immediately following an earthquake. The finite element analyses were continued for a long enough period of time after the ground motions had ceased for the response amplitude at the dam crest to decay to half its maximum response amplitude. Four arch and two gravity type dams were included in the study. In all cases, the input accelerations were significantly amplified at the dam crest. For the dams included in this study, the acceleration amplification ranged from 0.002 g per foot of dam height at Hoover Dam, to 0.043 g per foot of dam height at Deadwood Dam. Greater amplification occurred in the three thin arch dams than for either the gravity dams or the thick arch dam (Hoover). The crest and input acceleration curves generated for the gravity and thick arch dams tended to remain in phase, which was not the case for the thin arch dams. The crest accelerations generated for the two gravity dams and the thick arch dam also tended to decay very rapidly after the input motion ended. The crest accelerations generated for these dams decreased by half within 2 to 3 cycles after the input motion ended. The crest accelerations generated for the three thin arch dams decayed more slowly, requiring 5 to 25 cycles to reach half amplitude after the input motion ended.
 
CME Automatic Hammer Operations Bulletin
DSO-99-03Jeff Farrar, Doug ChitwoodDSO-99-03 
This report is a summary of experience with the Central Mine Equipment (CME) automatic hammer operation during performance in the Standard Penetration Test (SPT). The CME automatic hammer is run by a hydraulic chain-cam lifting system, and the drop height of the hammer depends on the speed of the chain cam. This report addresses energy transmission characteristics of the hammer system. Historical energy transmission data are reviewed. Manufacturer's operating instructions are reviewed. In a field study, the rate of the hammer was changed, and the drop height and energy measurements were taken. The data show that when the hammer is run at a slow rate of 30 blows per minute, energy loss was almost 15 percent. The data show that it is important to observe the hammer rate while testing and to operate the hammer at the design speed of 50 blows per minute
 
Effects of Sinkholes on Earth Embankments
DSO-99-02University of ColoradoDSO-99-02 
Centrifuge experiments were conducted on small models of earth embankments to study the effects of sinkhole formations on the stability of dams. A special container and various trap door assemblies were designed to simulate formation of sinkholes in a displacement controlled fashion. Significant efforts were invested in the initial design and subsequent modifications of the trap door assemblies in order to measure correct soil loads in its undisturbed state, i.e. before the trap door is lowered. Calibration tests on level ground sand and compacted Bonnie silt models were conducted to investigate this aspect of the experimental developments. The results from level ground model tests were found to be repeatable, reliable, and consistent. Finite element simulations of these experiments using PLAXIS compared well with the experimental results. Centrifuge tests conducted on embankment models revealed mechanisms involved in the cavity formation inside embankments, as a result of sinkhole development. The two models of compacted Bonnie silt were very similar except the presence of water reservoir in the second model. In the test with no water reservoir, the surface of the embankment did not suffer major deformations; although a large cavity was formed above the sinkhole. In the test with water reservoir, the cavity reached the dam surface which created a 12 ft (3.6 m) deep and 35 ft (11 m) diameter depression in the dam surface in terms of prototype scale. Results from centrifuge tests on level ground and embankment model tests as well as numerical simulations of the level ground tests are presented in this report. Possibilities for continuing research are also discussed.
 
Measurement of In-Place Relative Density in Coarse Grained Alluvium for Comparison to Penetration Tests
DSO-99-01Jeff FarrarDSO-99-01 
This paper summarizes results of investigations into coarse grained alluvial soils at several Bureau of Reclamation dams. In-place density tests have been performed in coarse grained deposits to measure relative density. Relative density data can be correlated to liquefaction resistance and to penetration testing. Penetration tests are the primary method of determining liquefaction resistance.
 
Standard Penetration Test, Driller's / Operation's Guide
DSO-98-17Jeff FarrarDSO-98-17 
In the 1980's, the benefits of this technology for hydrodemolition of concrete for repair projects became recognized. Deteriorated concrete could be removed without damaging the remaining sound concrete and without damaging the reinforcement. The resulting surface was ideal for bonding to the new repair concrete. The use of water jetting to clean concrete dam foundation drains was also developing in the 1980's. This technology seemed to have the potential for solving problems with existing methods---the economical methods were often not effective, and the effective methods were very expensive. The purpose of this report is to examine possible increased use of water jetting technology for cleaning foundation drains in concrete dams maintained by the Bureau of Reclamation. A Reclamation Drain Cleaning Seminar was held January 20-21, 1998, in Denver, and a summary report of the seminar lists drain cleaning needs and goals, as well as concerns about the water jetting method. By comparing the seminar information with current literature on water jetting for drain cleaning, this report attempts to assess the potential for increased use of water jetting for Reclamation drain cleaning work.
 
An Evaluation of Water Jetting For Bureau of Reclamation Use in Cleaning Concrete Dam Foundation Drains
DSO-98-16Todd Rutenbeck, Gregg DayDSO-98-16 
High pressure water jetting with pump pressures of 10,000 to 25,000 pounds/square inch, and ultra-high pressure water jetting with pump pressures up to about 35,000 (and sometimes as high as 60,000) pounds/square inch, have been used for many industrial and construction applications in recent years. The technology has developed from the Russian invention of a water cannon that could fracture hard rock in 1937, to hand held water blasting equipment for cleaning surfaces in the 1950's, to the use of abrasives and improved pump pressures for concrete removal in the 1960's, to later improvements including even higher pressures, remotely operated equipment, hole boring, rotating heads, waste water collection systems, and increased removal rates and thus lower costs.
 
Dynamic Properties of Mass Concrete Obtained From Dam Cores
DSO-98-15Caroline Mohorovic, Dave Harris, Tim DolenDSO-98-15 
The dynamic properties of concrete are an important consideration in the analysis and review of the safety of structures such as concrete dams. Concrete tests can be designed to predict the behavior of a structure under various static and dynamic loading conditions. In laboratory tests, different dynamic conditions are modeled by varying the strain rate at which the test is performed. Thus, the strain rate of the tests is key to the interpretation of results. The U.S. Bureau of Reclamation is particularly concerned with the performance of its dams when subject to earthquake loads. For approximately fifteen years, Reclamation's laboratory core test programs typically include dynamic tests performed at strain rates corresponding to seismic loads. This paper summarizes the results of a Reclamation research project designed to provide a broad database of the behavior of mass concrete from existing dams under dynamic loading conditions that simulate earthquake loadings. Laboratory tests performed on cores at both traditional, static loading conditions (strain rates of 10^-6 to 10^-4) and dynamic loading conditions (strain rates of 10^-3) are compared. Dynamic and static measurements of compressive strength, modulus of elasticity, compressive failure strain, Poisson's ratio and splitting tensile strength are summarized. In some cases, the core size or moisture condition among similar samples was varied to determine if either of these parameters affected test results. Linear, elastic assumptions that are typically used for finite element analyses of structural deformation and structural failure are reviewed for these data.
 
Full Scale Laboratory Testing of a Toe Drain with a Geotextile Sock
DSO-98-14Jay SwihartDSO-98-14 
This report describes the full-scale laboratory testing (pipe box testing) of a 15-inch-diameter, corrugated, polyethylene toe drain with a knitted geotextile sock, backfilled with a sand envelope material. The test results are compared with previous small-scale and full-scale tests using perforated pipe with 1/8-inch and l/4-inch perforations, but no geotextile. Use of the geotextile optimized toe drain performance both with respect to flow and with respect to loss of the sand envelope. The longterm flow rate was 21 gpm per linear foot of pipe, which was significantly higher (by a factor of 3 to 12) than the earlier tests without geotextile. The total loss of sand envelope was only 50 grams per linear foot of pipe, which was significantly lower (by a factor of 4 to 17) than the earlier tests without geotextile. The test with geotextile was run for 31 days at a constant head of 2.5 feet above the pipe invert with no indication of clogging.
 
Shake Table 2-D Models o a Concrete Gravity Dam For Computer Code Validation
DSO-98-13Dave Harris, Nate Snorteland, Tim Dolen, Fred TraversDSO-98-13 
One of the most famous and studied cases of dams subjected to earthquake loading is the Koyna Dam in India. In this study, a 2-dimensional model of Koyna Dam at 1/50 scale was used on a shake table to simulate effects and serve as data for nonlinear computer model calibration. A new concrete mix was designed for the nonlinear similitude modeling. This new mix provided the correct kinematic failure of concrete at scale. Two models were tested to failure: one with an initial shrinkage crack and one monolith. Reservoir effects were not modeled. The results of both models are discussed and compared. The ability to model nonlinear effects is discussed.
 
A Laboratory Simulation of In-Situ Grouting Using Standard Portland Cement and Polyurethane Materials with Standard Methods
DSO-98-12Dave Harris, Kurt von Fay, Nate SnortelandDSO-98-12 
Grouting features below ground is a common construction activity. The main purposes for placing grout have been previously identified (Shannon and Wilson, 1987), and are: 1) lower permeability, 2) reduce hydrostatic pressures under structures, 3) reduce water losses, 4) increase strength and bearing capacity, 5) stabilize rocks and subgrades, 6) backfill annular openings, and 7) fill cavities. Various requirements on the need for additional research have been identified in the review of the state-of-the-practice in grouting (Shannon and Wilson, 1987). One suggestion was the use of laboratory models as an interim step between theoretical mathematical models and full scale field tests. The objectives of this research study were to develop procedures and to perform tests to satisfy some of these previously suggested research needs. In-situ conditions were simulated in the laboratory, portland cement grout and polyurethane grout were placed under field-like conditions, and the grouted features were recovered for forensic testing and analysis of performance. Grout properties were measured and grouted modules were examined.
 
Shake Table Model of Concrete Gravity Dam for Computer Code Validation in Monolithic Models
DSO-98-11Terry PagneDSO-98-11 
Safety of dams under earthquake loading has always been a major concern in seismically active regions because of the high potential hazard that uncontrolled release of the reservoir poses to the population downstream of the dam. The current practice for the evaluation of the dynamic response of concrete dams at the U.S. Bureau of Reclamation is to initially perform a linear elastic finite element analysis of the structure under static loads plus the maximum credible earthquake predicted to occur at the site. If the response of the structure indicates that significant cracking damage would occur based on high tensile stress results, then non-linear analyses would be completed. In recent years a number of finite element programs have become available to structure engineers which provide the capability of modeling vertical contraction joint opening and concrete cracking which may occur in concrete arch dams during a seismic event. Reclamation has recently purchased the ABAQUS finite element program which is capable of performing nonlinear cracking analyses. The ABAQUS program includes a brittle cracking material model for concrete. This is a fairly complex analytical tool and engineers in Reclamation's structure analysis group need to develop expertise in performing and understanding non-linear modeling and analysis techniques. A set of calibration models would provide a valuable data set for those performing the numerical studies to assure that known results can be obtained. Structure engineers and material engineers need to work closely to develop new standards for material testing so that the appropriate material properties are available for these analyses. Therefore a shaking table test of a 1/50 scale cantilever monolith of Koyna Dam [in India] was built at Reclamation's Material Testing Laboratory for use in calibrating linear-elastic and non-linear material properties, as well as to provide a means of evaluating the effect of various analysis parameters available in the ABAQUS program. Two models were tested, a model with a natural but preexisting crack and a monolithic model. The ultimate goal of the shaking table test program was to develop a correlation between the material testing program and material property requirements for the non-linear numerical concrete cracking model. But before this goal can be accomplished, the basic linear elastic material properties needed to be calibrated and the effects of various analysis parameters needed to be understood. These requirements were accomplished through comparisons of analytical results with laboratory measurements. This report is concerned primarily with these preliminary studies, although some non-linear analysis results are also presented.
 
Verification of Block Kinematic Responses as Calculated by ABAQUS
DSO-98-09Barbara Mills-BriaDSO-98-09 
The study of rigid body motion is important in the consideration of the behavior of the dam under earthquake motion. Discrete blocks can be formed during the earthquake and the response of these blocks to subsequent earthquake motion is an important consideration. Research on the response of rigid bodies has been done since the late 1890's, although tools to predict the response of rigid bodies are scarce. The nonlinear capabilities of ABAQUS are studied here to understand how closely it can model the rigid body motion of a block compared to the theoretical solutions posed thus far.
 
An Investigative Study and Evaluation of the Aggregate Database
DSO-98-06Jeff Morris, Dave HarrisDSO-98-06 
The United States Bureau of Reclamation has developed the first electronic library of aggregate data entitled the "Aggregate Database". The International Center for Aggregates Research (ICAR) at the University of Texas at Austin entered into a cooperative agreement with Reclamation and the Federal Highway Administration to further develop and maintain the Aggregate Database. The Aggregate Database is a relational database that consists of aggregate data including location of the aggregate source, physical properties of the aggregate, concrete data such as freezing and thawing and alkali-aggregate reactivity results, and petrographic results. The Aggregate Database is in its infancy and there are limitations and drawbacks. Therefore, ICAR distributed a survey to determine the needs of professionals in the aggregates industry in terms of a database consisting of aggregate data, the effectiveness of the Aggregate Database in fulfilling those needs, and improvements that need to be made to the database. According to the results, users need a database of aggregate data to find suitable aggregate, have a listing of aggregate sources, and perform comprehensive analyses on the data. The results also show that the Aggregate Database is effective at meeting the needs of the user. However, there are many improvements that need to be made to the database to make it more effective and efficient. Therefore, a set of recommendations has been developed to provide ICAR and Reclamation with a comprehensive list of recommended improvements.
 
Concrete Dams Case Histories of Failures and Nonfailures with Back Calculations
DSO-98-05Chuck Anderson, Caroline Mohorovic, Larry Mogck, Bitsy Cohen, Gregg ScottDSO-98-05 
This compilation of case history summaries is intended to assist risk analysis teams in estimating probabilities related to concrete dams. Much can be learned examining the successes and failures of other projects. What makes this compilation unique and hopefully more useful is the back calculations of strength and response which are provided. If a case history can be found in this compilation that is similar to a dam being considered in a risk analysis, then the information may be useful in making reasonable estimates. Case histories can provide valuable insight for identifying failure modes and for breaking them down into sequences of events. Failures and incidents that did not lead to failure are equally valuable, and provide information about what happened at other dams. This information provides the means for conceptualizing and specifying the occurrences, conditions, and interventions that could be pertinent to the dam under consideration. In addition, back analysis of these situations provides valuable insights when interpreting the results of analyses for the dam under consideration.
 
Prediction of Embankment Dam Breach Parameters - A Literature Review and Needs Assessment
DSO-98-04Tony WahlDSO-98-04 
This peer-reviewed report examines the role, importance, and methods for predicting embankment dam breach parameters needed for analysis of potential dam-failure floods. Special emphasis is given to dam breach analysis within the context of the risk assessment process used by the Bureau of Reclamation (Reclamation). Current methods for predicting embankment dam breach parameters and numerically modeling dam breach events are reviewed, and the needs and opportunities for developing improved technologies are discussed. Recent technical advances that could contribute to improvements in dam breach simulation are identified.
 
Study of the Properties of Aged Concretes Containing Various Cements, Parapet Wall, Green Mountain Dam
DSO-98-03Caroline Mohorovic, Dave HarrisDSO-98-03 
This project resumed the study of the durability of various concrete mix designs after 50 years of environmental exposure. After completion of a visual inspection and nondestructive testing of all panels, 14 cements were selected for coring and laboratory testing. Cored samples were tested for static compressive strength, dynamic compressive strength, splitting tensile strength, cyclic (nonlinear) compressive strength, modulus of elasticity, sonic modulus, Poisson's ratio, and unit weight. This report emphasizes the results of material properties tests conducted in Reclamation's Materials Engineering and Research Laboratory. Additional papers concerning other aspects of this study were written by University of Colorado at Denver contributors and were submitted to the American Society of Civil Engineers Journals. Osama Mohamed, a graduate student of civil engineering at the University of Colorado at Denver, is preparing a thesis based on this study. All data and publications resulting from this project reflect the combined intellectual, material, and financial resources of both parties.
 
Seismic Tomography of Concrete Structures Phase 1
DSO-98-02Lisa BlockDSO-98-02 
This report contains the results of Phase I of the seismic tomography research project currently being performed by the Seismotectonics and Geophysics Group, with funding from the Dam Safety Office. The ultimate goal of this research project is to develop Reclamation's capabilities for efficiently acquiring high-quality seismic tomography data on concrete structures for the purpose of imaging variations in concrete quality within the structure. As part of this project, the usefulness of the seismic tomography method as applied to concrete structures will be evaluated by acquiring and processing seismic data from one of Reclamation's concrete dams. The project is divided into 4 major phases: gathering background information from other researchers and testing various types of equipment (Phase I); preparing for the field test by purchasing and constructing equipment, modifying data processing software, and performing forward modeling (Phase II); data acquisition and preliminary on-site data processing (Phase III); and final data analysis and report preparation (Phase IV). The primary purpose of Phase I of this research project is to determine the types of sources, receivers, and coupling methods that are likely to produce the highest quality seismic data on concrete structures. Other important factors that are considered when evaluating various field equipment are equipment costs and difficulty of installation. A secondary purpose of Phase I is to determine the frequencies of the seismic signals that are likely to be acquired through concrete dams over the distances and angles necessary for application of the seismic tomography method.
 
Comparison of Failure Modes From Risk Assessment and Historical Data from Bureau of Reclamation Dams
DSO-98-01Joe Tatalovich, Dave HarrisDSO-98-01 
To determine which dams need to be renovated because of their risk of failure or accident, the Bureau of Reclamation (Reclamation) has developed a process of calculating that risk. It takes into account the failure possibilities and the number of people who would be affected by certain occurrences. Uncertainties that are inherent in the process make this study of the risk assessment process all the more important. The purpose of this project was to present findings from various risk assessments of Reclamation dams. The failure modes of these dams were compared to each other as well as to historical failures (Von Thun, 1985). To compare the risk assessment and historical data, it was necessary to put together a comprehensive list of all Reclamation dams classified by their age, height, and type. It was hoped that, through the process of examining the risk assessments and the historical data, Reclamation would learn more about both their dams and how the risk of failure is analyzed. By doing this, the prioritizing of dam servicing in the future may be improved.
 
Comparison of Failure Modes From Risk Assessment and Historical Data from Bureau of Reclamation Dams
DSO-97-01Joe Tatalovich, Dave HarrisDSO-97-01 
To determine which dams need to be renovated because of their risk of failure or accident, the Bureau of Reclamation (Reclamation) has developed a process of calculating that risk. It takes into account the failure possibilities and the number of people who would be affected by certain occurrences. Uncertainties that are inherent in the process make this study of the risk assessment process all the more important. The purpose of this project was to present findings from various risk assessments of Reclamation dams. The failure modes of these dams were compared to each other as well as to historical failures (Von Thun, 1985). To compare the risk assessment and historical data, it was necessary to put together a comprehensive list of all Reclamation dams classified by their age, height, and type. It was hoped that, through the process of examining the risk assessments and the historical data, Reclamation would learn more about both their dams and how the risk of failure is analyzed. By doing this, the prioritizing of dam servicing in the future may be improved.
 

 

Last Updated: 2/14/22
For site maintenance contact: Webmaster