ABSTRACT: Once a nuclear waste repository is licensed and operational, it is prudent to dispose of the largest volume of waste allowed by regulation. In addition to the contact handled transuranic waste that will be placed in boxes and drums in the excavations at the Waste Isolation Pilot Plant, remote handled waste will be placed in large-diameter boreholes drilled into the walls of the openings. Field testing has been done on the original borehole layout design. Because space is so valuable in a nuclear waste repository, it may be desirable to alter the borehole layout (spacing and depth) to increase the number of waste canisters that can be stored in the facility. To that end, a series of three-dimensional finite difference numerical models were developed to determine if any or all of the alternative layouts are viable. The models investigate various combinations of borehole spacing and length. It was particularly important that all of the models be consistent with field observations, earlier modeling efforts, and internally with one another. The results show that reducing the spacing between boreholes and lengthening the boreholes will increase ground control level of effort and prematurely age the openings. The model results give project personnel the tools needed to determine the optimal trade off between maintenance costs and increased storage capacity.
INTRODUCTION The Waste Isolation Pilot Plant (WIPP) is a U.S. Department of Energy facility located in southeastern New Mexico near the city of Carlsbad. WIPP?s mission is to receive, handle, and permanently dispose of transuranic waste generated by defense activities. About 97% of the waste is designated contact handled or CH waste. CH waste will typically be disposed of in steel barrels and boxes which are placed on the floor of disposal rooms. The remainder of the waste is designated remote handled or RH waste. RH waste emits more radiation than CH waste and must be handled and transported in shielded canisters. The canisters will be emplaced in 76 cm (30 in) diameter horizontal boreholes drilled into the walls of the disposal rooms. The current RH borehole design calls for the boreholes to be drilled about 2.44 m apart and 5.18 m long. Reducing the borehole spacing or increasing the length of the boreholes would increase the ultimate RH waste capacity of WIPP. However, most previous studies only looked at the geomechanical implications of the original borehole layout [1, 2]. Some alternative layouts have been investigated, but not in detail and not in the stratigraphic location that future disposal rooms will be mined in.
To provide technical bases for alternative layouts of RH waste boreholes at WIPP, a series of threedimensional finite difference models were developed. Borehole spacings of 1.83 m (6 ft) and 2.44 m (8 ft) were modeled. Borehole lengths were 5.18 m (17 ft) long, 10.36 m (34 ft) long or alternated between the two lengths. The disposal rooms are located 640 m below the surface in a thick evaporate formation primarily composed of halite with thin interbeds of clay and anhydrite.
ABSTRACT: Seismic tomography in the ground is typically limited to panels between pairs of boreholes. It usually offers reliable structural sections along the panels. However the resolution between panels is poor and deteriorates with distance between the panels. The authors successfully employed the Three-dimensional Reflector Tracing (TRT?) technique to enhance resolution of the ground images between neighboring panels. The technique uses a source and a set of receivers in the same borehole (Single- Hole TRT?) for detecting seismic waves reflected from anomalies in the ground due to changes in acoustic impedance (velocity * density) along the wave path. The Single-Hole TRT? uses volumetric velocity model derived from the velocity tomography to measure the distance from a single borehole to surrounding anomalies, but in general it is insufficient to define the direction to these anomalies. This approach may suffice for anomalies located in known direction from the borehole. However, a truly threedimensional imaging of anomalies in the ground requires triangulation using Single-Hole TRT? images from at least three boreholes. Presented Case Studies demonstrate a broad range of applications for this technique, from defining natural structural features in the ground, to a number of an old infrastructure imaging efforts (piles, sewers, foundations).
INTRODUCTION It has been a long-standing goal of being able to convert geophysical data into three-dimensional digital image of underground structural features that could be dissected and measured at any direction, angle and depth. Using a combination of transmitted and reflected seismic waves the authors were able to produce images allowing to identify piles and their depth, boreholes, lamination, old sewer walls, and even general shape of solution mining caverns.
ABSTRACT: This paper extends the 1980 Kulhawy and Goodman paper that proposed a simplified bearing capacity calculation model for foundations on discontinuous rock masses that are dominated by vertical and horizontal discontinuities. The theoretical models and the empirical correction factor are refined. A finite element model also is used to verify the refined models. Furthermore, Monte Carlo simulations are performed to evaluate the results of the refined models relative to an empirical bearing capacity approach. In addition, the results are compared to field test data.
INTRODUCTION It is desirable that foundations be constructed on or in the surface or near surface rock masses. The first step in predicting the behavior of the foundations is geological characterization. Rock masses in general are inhomogeneous, discontinuous media composed of the rock material and naturally occurring discontinuities, and they are among the most variable of all engineering materials. This characterization process can be found elsewhere [e.g., 1, 2].
Analytical predictions of foundation capacity on rock masses require adoption of a model for the behavior of the ideal material, as well as the selection of the basic model input parameters. Invariably, the choice of the ideal material depends on the type of prediction to be made and on how much information is available. Adoption of a highly sophisticated model for rock masses would be pointless if the necessary geological information and test data were not available to support its use. Research in this area has advanced significantly. At present, however, the advances find major use in the research environment and rarely find application in design practice. Simplified methods still are used widely in current design practice.
In the design of foundations on rock masses, displacement criteria and the ultimate capacity must be addressed. The capacity evaluation consists of two factors. First, the foundation element itself must be adequate to carry the applied loading. Second, the rock mass must provide sufficient resistance to the loading. In both cases, a margin of safety is needed to ensure that no adverse performance will occur. This paper focuses on the second factor.
The bearing capacity of foundations on rock masses is complex because it is usually a function of both the intact rock material and the rock discontinuities. The simplified typical failure modes relevant to this paper are shown in Fig. 1.
In 1980, Kulhawy and Goodman proposed a simplified calculation model for foundations on discontinuous rock masses that are dominated by vertical and horizontal discontinuities, focusing on general wedge behavior. This paper examines the problem further. Theoretical treatments and an update of the empirical correction factor are presented, followed by numerical verification using PLAXIS . Monte Carlo simulations of the problem are discussed to examine the effect of rock mass property variability. The results then are compared with those of field tests.
ABSTRACT: A large rockslide from an existing rock cut on U.S Highway 6 near Denver not only pushed two trucks off the road and closing the road with debris, but also left an unstable rock face with overhangs perched above the road, creating a hazard for highway users. The highway is a major thoroughfare linking Denver with several mountain communities, and it was closed to through traffic for 83 days to reconfigure and stabilize the cut. This paper reviews the facts of the incident and the resulting project, including the original rockslide, the ground conditions, the engineered stabilization, and construction. Ground conditions consisted of foliated gneiss with two prominent joint sets and pegmatite intrusions. A prominent planar pegmatite intrusion was a prime factor in creating the geologic conditions that led to the slide. This same pegmatite plane, along with natural jointing, was also used as the basis for the cut reconfiguration and stabilization measures implemented. Work began with removal of the overhangs for safety considerations. Stabilization involved reconfiguring the cut to a stable slope and installation of rock reinforcement followed by draping the face with rockfall mesh. Rock excavation was performed by blasting benches from the top down with concurrent installation of rock reinforcement.
SETTING Clear Creek Canyon is situated on the western side of the Denver metropolitan area. Highway 6, which runs through the canyon, is a major thoroughfare providing access to the high mountains and the gaming towns of Black Hawk and Central City, handling 12,000 vehicles per day. The section of Highway 6 through the canyon from Golden to its intersection with Interstate 70 is approximately 14.5 miles long. At approximately 11.5 miles up canyon, Highway 119 diverges to Black Hawk and Central City. Although Interstate 70 is an alternative route to Black Hawk/Central City, the route is inconvenient. Figure 1 shows the location of the site.
The canyon is steep and rugged, and is heavily used for recreation including hiking, fishing, boating, climbing, and gold panning. Historically the canyon was exploited for gold primarily from placer mining, and at one time a railroad snaked up the canyon. Much of the land, including the site of the slide, is owned by Jefferson County as part of their open space program.
At the site of the event, Highway 6 is on the north side of the canyon, north of Clear Creek, with the rock cut south-facing, the road takes a sharp 80 degree bend, as shown in Figure 2.
The original cut was over 120 ft high and 500 ft long at an inclination of approximately 0.5H to 1V, and wrapped around with the road curve. The location of the slide is milepost 261.65.
ABSTRACT: This paper develops a series of room closure and porosity surface calculations, which are used to assess performance of the Waste Isolation Pilot Plant. The concept of a porosity surface comprises calculation of room closure as salt creep is resisted by back stress created by the waste packages and by hypothetical gas generation within the rooms. The physical and mechanical characteristics of some of the waste packaging are appreciably different from the assumed waste upon which the original compliance was based and approved. These analyses provide insight into the structural response of a room full of various wastes, including the influence of the waste in the absence of gas generation, as well as the lack of influence on room closure when gas generation is modeled. All of the underlying assumptions pertaining to the original compliance certification including the same finite element code are implemented; only the material parameters describing the more robust waste packages are changed from the certified baseline. As modeled, more rigid waste tends to hold open the rooms and create relatively more void space in the underground than identical calculations run on the standard waste packages, which underpin the compliance certification. Several porosity surfaces were developed to cover a range of possible packaging.
INTRODUCTION 1.1. Objective In 1996, the U.S. Department of Energy (DOE) completed a performance assessment (PA) for the Waste Isolation Pilot Plant (WIPP). The PA was part of the Compliance Certification Application (CCA)  submitted to the U.S. Environmental Protection Agency (EPA) to demonstrate compliance with the long-term disposal regulations in 40 Code of Federal Regulations (CFR) 191 (Subparts B and C) and the compliance criteria in 40 CFR 194. In 1997, EPA required a verification of the calculations performed for the CCA, termed the Performance Assessment Verification Test (PAVT). On the basis of these submittals, WIPP was certified for operations. Since March 1999 the DOE has disposed of radioactive waste at WIPP in accordance with provisions of compliance certification.
One provision of the certification itself is a requirement for recertification on a five-year interval. This requirement recognizes that repository operations would likely change from the baseline conditions assumed in the original certification. The compliance recertification application (CRA), which was prepared and submitted in 2004, includes analyses of conditions that depart from the bases underlying the original certification. PA is charged with the responsibility of evaluating the consequences of these changes. This document examines actual and potential structural/mechanical changes in waste packaging that are substantially different from the compliance basis.
ABSTRACT: An underground limestone mine with thin and rigid limestone roof and high horizontal stresses create a burstprone condition and present a major ground control and safety problem at the mine: rock-burst caused severe roof falls. To help the mine address this problem, an Acoustic Emission/Microseismic system was installed at the mine by the National Institute for Occupational Safety and Health (NIOSH). However, the efficiency of this system was severely hampered by high background noise and inaccurate event location. The focus of this research is therefore to resolve these problems. The most beneficial outcome of this research is the demonstration that the source location accuracy at the mine can be significantly improved through a comprehensive and highly efficient source location approach. This includes digital filtering, phase association, sensor array optimization, absolute value based optimization method, advanced Simplex location algorithm, and reliability analysis. The development of this approach provides a unique and reliable solution to the problem.
INTRODUCTION It is generally accepted that most solids emit lowlevel seismic signals when they are stressed or deformed. In the geotechnical field this phenomenon is generally referred to as Acoustic Emission/Microseismic (AE/MS) activities . When rock fractures, it will produce microseismic signal which will transmit through rock as elastic waves. The application of AE/MS techniques, which monitors self-generated acoustic signals occurring within the ground, has now rapidly increased for stability monitoring of underground structures such as mines, tunnels, natural gas and petroleum storage caverns, as well as surface structures such as foundations, rock and soil sloped.
An AE/MS system was installed by U.S. National Institute for Occupational Safety and Health (NIOSH) at an underground limestone mine in southwestern Pennsylvania, where roof fall was a big problem to the production and safety. The traditional room-and-pillar and a new stress control layout were in use at this mine site. In order to monitor the stability of the roof, minimize the hazardous ground conditions and provide safer working conditions for miners, NIOSH installed a microseismic system, which had 24-channel uniaxial geophones connected with DOS-based data acquisition system. The system started to monitor microseismic activity on Feb 9, 2000 (Figure 1).
ABSTRACT: The current conceptual design for the proposed nuclear waste repository at Yucca Mountain Nevada includes placing most of the facility in lithophysal Tuff. It is difficult (if not impossible) to conduct standard confined triaxial laboratory tests on lithophysal tuff. It was decided to use UDEC to conduct numerical triaxial testing of simulated lithophysal tuff samples to supplement existing uniaxial data. The results of these numerical experiments are intended to provide guidance on the variability of physical properties as a function of lithophysal porosity. These data then can be used as a basis for large scale modeling of the behavior of the repository drifts. The results of the numerical testing were encouraging and show consistent trends as a function of lithophysal porosity.
INTRODUCTION The repository horizon for the proposed nuclear spent-fuel waste repository at Yucca Mountain Nevada is located in both lithophysal and nonlithophysal rock units in the Topopah Spring Tuff. The nonlithophysal rocks are characterized as hard, strong, jointed. The lithophysal rocks typically contain about 20 to 30 percent cavities, are more deformable, and have a lower unconfined compressive strength. The lithophysae vary in size from a few centimeters to more than 1 meter in diameter. For this paper, lithophysal porosity is defined to be the ratio of volume of lithophysal voids to total sample volume.
Normal testing procedures recommend that the test sample dimension should be at least 10 times the size of any mineral grains or other inclusions. The void size would dictate that very large samples would need to be tested. The representative elementary volume of lithophysal rock is of the order of cubic meters to cubic decameters depending on the size of lithophysae. In order to develop an adequate correlation between lithophysal porosity and mechanical properties of lithophysal rock, sufficient numbers of laboratory tests on large-size rock samples were desired. However, as a consequence of the size of rock samples required, the lack of high-capacity equipment needed to test such large samples and the cost and time that would be required to produce an adequate statistical database, a suitable laboratory testing effort was impractical to carry out. The presence of voids intersecting the sample surface also makes standard triaxial testing difficult or impossible. It is apparent that the uniqueness of lithophysal rock poses formidable challenges to obtaining data directly by the process of testing larger rock specimens. Sampling logistics are difficult to manage, and applicable testing standards are inadequate when dealing with lithophysal rock. The challenge, therefore, is how to characterize the range in variability of the properties of this material.
To overcome the inability to conduct adequate physical testing, a numerical approach was used to supplement the existing intact rock property database and to estimate mechanical properties of lithophysal rock. This paper presents a systematic method of creating a numerical model of the material, calibrating that model against existing data, and then conducting numerical triaxial tests to supplement existing data. The supplemental data can be used to predict the larger-scale behavior of the repository.
ABSTRACT: Synthetic Aperture Radar (SAR) interferogram was used to detect the ground surface movement due to oil extraction in this study. Interferometric SAR (InSAR) technique makes use of the phase shifts of radar signals between repeated SAR observations of the same area to extract information of vertical elevation changes on the ground surface. A detailed procedure for generating interferogram was described. The procedure was applied in detecting subsidence in an oilfield area (100 km by 100 km, centered approximately at Latitude 70° 21? and Longitude 149° 7?) in the North Slope of Alaska. The results reveal that this method could produce promising results of ground surface movement during a certain period of time. The precision of the measurement is within the range of centimeters or less. The ability to map surface deformation from remote sensing data permits subsidence measurements in areas where access is difficult or expensive, such as the Arctic Regions.
INTRODUCTION Land subsidence, loss of surface elevation due to removal of surface support, at rates that are of practical significance to man-made structures and natural environment, affects most of the United States . Subsidence is caused by a diverse set of human activities and natural processes, including underground mining, extraction of fluids from subsurface, melting of permafrost, liquefaction, natural sediment compaction, earthquakes, and volcanic deformation. Subsidence due to natural processes, such as sediment compaction, is relatively slow and rarely causes problems on human timescales. Subsidence attribute to human activities, such as the extraction of fluids from underground and underground mining, are relatively rapid. Subsidence due to underground mining has long been known to cause damages to surface structures, alter subsurface hydrological conditions and disturb the environment surrounding the mining operations. In most cases, oil field subsidence may not cause severe impact to surface structures as underground mining does. However, the long-term effect of oilfield subsidence to the local ecological environment, the subsurface hydrological system and the surface drainage pattern can be significant . Furthermore, in the environmentally sensitive high latitude regions, such as Alaska, subsidence may cause disturbance to permafrost and potentially induce loss of thermal equilibrium in the permafrost areas.
The traditional methods of monitoring subsidence by setting up surveying lines across the fields and employing common surveying instruments for measurements are extremely time-consuming and too expensive to be applied in vast and remote areas of oilfields in Arctic Regions. More efficient subsidence monitoring techniques are, therefore, needed for oilfield subsidence measurements. In this paper, we present a procedure for detecting oil extraction induced ground movement using differential interferometric SAR technology.
ABSTRACT: Different methods for increasing the fracture gradient of underground formations have been proposed over the years; they are commonly referred to as wellbore strengthening techniques. Perhaps, one of the most successful is the stress cage concept. In this method, the tangential stress around the wellbore is increased by inducing and propping open a controlled fracture at the borehole wall. This technique, albeit very efficient in permeable formations, has proven very ineffective when applied to low permeability formations. This paper presents a new procedure for creating a stress cage in low-permeability formations (e.g. shales). In this novel method, changes in temperature are induced in the formation to be treated, before setting the stress cage. The drilling fluid is used to cool down the formation in order to reduce the tangential stress at the borehole wall. The magnitude of this temperature change is determined by the required increment in fracture resistance, which also establishes the opening of the fractures in the stress cage. Subsequently, the stress cage is set up following normal procedures. Fracture growth is expected to be arrested once the fracture leaves the low-temperature zone around the borehole (i.e. once the fracture encounters the higher original closure stress). After the stress cage is set up, the fractures are locked open by allowing the formation to return to its original temperature. The entire procedure is performed without significant variations of hydrostatic pressure; thus, reducing greatly the probability of dislodging the proppant particles within the fractures. The stability as well as the hydraulic seal of the stress cage may both be further increased by adding graphite and polymer together with the proppant particles.
INTRODUCTION The current world?s need for energy is forcing oil and gas producers to drill in environments of ever increasing complexity. In some instances, low fracture gradient formations (condition probably caused by earlier depletion) must be drilled in order to reach hydrocarbon reserves located at greater depths. Operating companies that desisted of pursuing such accumulations in the past are hastily trying to produce them as the current oil and gas prices conditions make such resources economically attractive.
During the drilling stage, the presence of low fracture gradient formations translates into narrow, and even non-existent, mud weight ranges; thus leading to either borehole collapse or fluid losses due to hydraulic fracturing of the rock. Thus, the drilling engineer is left with no alternative other than setting additional strings of casing to isolate the problematic formations and reach target. This solution entails large additional cost not only due to the need for supplementary materials but also due the extra rig time involved. In some cases, this added cost is in the order of millions of dollars per well, as the average daily price tag for an offshore rig is more than $400K. Thus, the economical benefits of creating technologies for increasing a formation fracture gradient are not only evident but critical.
ABSTRACT: The effects of both limited frequency bandwidth and presence of noise in recordings of mine seismicity are investigated using Brune?s source model and Gaussian white noise. The results indicate that not accounting for these effects in the evaluation of spectral levels leads to underestimated or overestimated values for large and small magnitudes, respectively, with biases ranging between approximately 50 and 300% for signal-to-noise ratios of 40-60 dB. Corner frequency estimates exhibit opposite biases, with typical bias of 50-100%. Even when employing approximate and full frequency range corrections, the results show that a noise level increase by 20 dB alone leads to a bias of 25-50% in spectral parameter estimates from velocity data for smaller magnitude events, and several times larger as derived from acceleration data. Limited frequency range and noise corrections are subsequently applied to source parameter catalogs recorded at three hard rock underground mines: Kidd, Creighton (both in Ontario, Canada), and Ridgeway (New South Wales, Australia). These catalogs contain several thousands to tens of thousands of seismic events recorded between 2001-2005 with magnitudes between approximately -2 and 1.5. With corrections applied, log energy vs. log moment slope decreases from 1.6-1.9 to 1.1-1.4 and source estimates are brought much closer to theoretical expectations for a constant stress drop model. The application of limited frequency range and noise corrections also reduces the b-value of the frequency-magnitude distribution, with critical implications on the assessment of seismic hazard.
INTRODUCTION Seismic energy and moment are source parameters that can be independently estimated from recorded waveforms. A large number of source estimates on earthquakes over a wide range of magnitudes indicate that the relationship between these two parameters is approximately constant, independent of magnitude (e.g., Choy and Boatwright, 1995). This observation supports the idea that the seismic source behaves similarly at both large and small scales, and could be described by a constant stress drop model. Other studies though (e.g., Kanamori et al., 1993; Abercrombie, 1995) appear to indicate that the ratio of seismic energy to seismic moment decreases with decreasing magnitude, at least for limited magnitude ranges. The break in energy scaling indicates the non-similarity of the seismic sources. By including observations from mineinduced seismicity, the magnitude range of these studies spans over 12 units, from -4 to 8. The implications of similarity or non-similarity are multiple, from fundamentals regarding the process of seismic failure to seismic hazard assessment. It was long argued (Di Bona and Rovelli, 1988) that the apparent break in energy scaling could be an artifact caused by uncertainties involved in energy and moment estimates from frequency-limited seismic recordings. Ide and Beroza (2001) demonstrated that accounting properly for uncertainties in source spectrum estimates significantly lowers the perceived decrease in apparent stress for small earthquakes. Even while accepting as a working hypothesis the existence of a decreasing trend in scaled energy with seismic moment, Abercrombie and Rice (2005) outline that model assumptions can lead to large uncertainties and earthquake scale independence is possible within the resolution