Page 20

2

Repository Performance Confirmation

The performance of geological repositories is evaluated on the basis of their ability to comply with a series of regulatory performance criteria, defined in Title 40 of the Code of Federal Regulations Part 191 (40 CFR 191; EPA, 1985). In the case of the WIPP, the time of compliance with the containment requirements formulated by the U.S. Environmental Protection Agency (EPA) is 10,000 years (see Sidebar 1.4). Of course, the issue does not end at 10,000 years; for example, the risk-controlling radionuclide in the WIPP is plutonium-239, which has a half-life of 24,000 years. The intent of a geological repository is to contain the waste for the indefinite future (e.g., > 10,000 years).

The WIPP was certified by the EPA through a comprehensive process based primarily on a detailed performance assessment (DOE, 1996; see also Sidebar 1.2). Of course, acceptance of the performance assessment (PA) is conditional on several factors that are designed to offset the many uncertainties involved. One of the EPA's requirements is that the DOE must implement a monitoring program designed to provide confidence in the assessed performance of the repository. Furthermore, every five years, the DOE must apply to the EPA for recertification of the WIPP. The recertification application must show evidence that the repository is performing as assessed.

A monitoring program that emphasizes factors contributing mostly to performance uncertainties could provide important evidence of the ability of the repository to perform its intended function. Therefore, the committee has chosen as the theme of this review to be “performance confirmation through monitoring.” The strategy of the committee is to focus on safety and monitoring activities that would best enhance confidence in the long-term performance and reduce uncertainties in the performance assessment of the WIPP.

The recommendation to implement an in-situ monitoring program was endorsed by a previous NRC committee on the WIPP in a letter report to the Hon. L. P. Duffy (NRC, 1992). Quoting from that report, “The panel emphasizes that it supports the notion of underground testing with TRU wastes, provided that the underground location does not prevent important tests from being carried out (e.g., the measurement of brine compositions in contact with real waste or progression of gas generation experiments without purging), and that the tests can be continued for sufficient time to provide useful information.”



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 20
Page 20 2 Repository Performance Confirmation The performance of geological repositories is evaluated on the basis of their ability to comply with a series of regulatory performance criteria, defined in Title 40 of the Code of Federal Regulations Part 191 (40 CFR 191; EPA, 1985). In the case of the WIPP, the time of compliance with the containment requirements formulated by the U.S. Environmental Protection Agency (EPA) is 10,000 years (see Sidebar 1.4). Of course, the issue does not end at 10,000 years; for example, the risk-controlling radionuclide in the WIPP is plutonium-239, which has a half-life of 24,000 years. The intent of a geological repository is to contain the waste for the indefinite future (e.g., > 10,000 years). The WIPP was certified by the EPA through a comprehensive process based primarily on a detailed performance assessment (DOE, 1996; see also Sidebar 1.2). Of course, acceptance of the performance assessment (PA) is conditional on several factors that are designed to offset the many uncertainties involved. One of the EPA's requirements is that the DOE must implement a monitoring program designed to provide confidence in the assessed performance of the repository. Furthermore, every five years, the DOE must apply to the EPA for recertification of the WIPP. The recertification application must show evidence that the repository is performing as assessed. A monitoring program that emphasizes factors contributing mostly to performance uncertainties could provide important evidence of the ability of the repository to perform its intended function. Therefore, the committee has chosen as the theme of this review to be “performance confirmation through monitoring.” The strategy of the committee is to focus on safety and monitoring activities that would best enhance confidence in the long-term performance and reduce uncertainties in the performance assessment of the WIPP. The recommendation to implement an in-situ monitoring program was endorsed by a previous NRC committee on the WIPP in a letter report to the Hon. L. P. Duffy (NRC, 1992). Quoting from that report, “The panel emphasizes that it supports the notion of underground testing with TRU wastes, provided that the underground location does not prevent important tests from being carried out (e.g., the measurement of brine compositions in contact with real waste or progression of gas generation experiments without purging), and that the tests can be continued for sufficient time to provide useful information.”

OCR for page 20
Page 21 The long operational phase of the WIPP repository (at least 35 years and possibly as long as 100 years) provides an unusual, and perhaps unprecedented, opportunity to implement a monitoring program and reexamine the performance assessment with information based on direct observations of the total system prior to closure of the repository. Although a time frame of 35-100 years is short compared to the 10,000-year period of compliance, the committee believes that it is long enough to develop and implement a monitoring program to observe the development of repository responses. Indeed, the rates of important processes such as salt creep, brine inflow, and microbial activity are predicted to be the highest during the first 50 to 100 years (Knowles and Economy, 2000; NRC, 1996a). If these responses confirm assumptions in the performance assessment, this will reduce uncertainty in the projections of long-term performance of the repository and could improve public confidence in the repository performance. The ongoing DOE's monitoring program required by the EPA as part of the certification, is described in the next section. The committee strongly supports such a program but believes it could be more focused and risk-informed. The difference in focus between what is planned and what the committee suggests is also discussed. REGULATORY REQUIREMENTS FOR MONITORING A monitoring plan for the WIPP was included in 40 CFR 194 under the requirements for the certification of the repository by the EPA. The purpose of the monitoring plan is to confirm that the repository is performing as expected according to the model in the Certificate of Compliance Application. The DOE proposed a monitoring plan, which was accepted by the EPA in 1998 in the certification decision, to address the requirements of the regulations in 40 CFR 194. The DOE described its monitoring program in the CCA and indicated that it would span 150 years (50 years pre-closure and 100 years post-closure). The DOE program evolved from screening 91 potentially significant parameters down to 10. The 10 parameters were divided among physical measurements in the Salado Formation, hydrological properties in the non-Salado settings, and activity levels of the waste. The four parameters to be measured in the Salado Formation relate to creep closure and stresses, extent of deformation, initiation of brittle deformation, and displacement of deformation features. The program calls for pre-closure monitoring only for the Salado parameters and pre- and post-closure for the non-Salado parameters. Waste activity is to be monitored only during pre-closure. In the DOE program, pre-closure monitoring in waste storage rooms ends with the sealing of individual panels of rooms; hence, pre-closure monitoring of emplaced waste is very limited. The parameters that the DOE is currently monitoring to comply with 40 CFR 194 are shown in Table 2.1. The committee's proposed performance confirmation monitoring plan is very similar to the DOE's monitoring program. The significant difference between the DOE monitoring program and the committee proposal is that the committee's recommended plan includes monitoring rooms and panels after sealing of the panels and extends until closure of the repository and sealing of the shafts. The committee has put greater emphasis on such issues as brine inflow, gas generation, salt rock deformation following sealing of the panels, auxiliary material inventory in the repository, and radiogenic measurements. The committee identified important issues relative to the long-term safe performance of the WIPP repository on the basis of the DOE's performance assessment (DOE, 1996), past committee reports, and numerous briefings on the WIPP. The criteria for identifying issues were related principally to the sources of uncertainty in the performance assessment and to the safety of workers and the public. Several of the issues are interrelated but are treated separately to emphasize important points. The following paragraphs describe in detail the issues of consideration in the performance confirmation monitoring program proposed by the committee. The issues have been grouped as site performance issues and site characterization issues.

OCR for page 20
Page 22 Table 2.1 Parameters Currently Monitored by the DOE to Comply with 40 CFR Part 194.42a Parameter Monitored in the WIPP Pre-closure Monitoring? Post-closure Monitoring? Salado physical parameters Creep closure and stresses YES NO Extent of deformation YES NO Initiation of brittle deformation YES NO Displacement of deformation features YES NO Non-Salado hydrological properties Culebra groundwater composition YES YES Probability of encountering a Castile brine reservoir YES YES Drilling rate YES YES Culebra change in groundwater flow YES YES Subsidence measurements YES YES Waste related parameters Waste activity YES NO a EPA (1996). SITE PERFORMANCE ISSUES The key site issues that should be monitored during the pre-closure period to confirm the performance of the WIPP repository are described below. Brine Migration and Moisture Access to the Repository The presence or absence of brine in the WIPP rooms is a key issue in the performance of the repository. Without brine there will be no radionuclide mobilization and transport or any gas generation from corrosion of the steel drums. The brine sources for the undisturbed repository are seepage from brine-filled void spaces in the undisturbed geological setting, the humidity of the repository air, and water used during mining operations. In the long term, after repository closure, additional sources of brine could include accidental fluid injections by inadvertent human intrusions (see section “ Oil, Gas, and Mineral Production” and Appendix B). Brine volumes from enhanced recovery fluid injection operations have the potential to be a source of much greater brine inflow than that expected from any other water sources in the undisturbed geological setting. A concern is the possible failure of a well casing or cement outside the casing during an injection operation and fluid leaking into overlying formations and flowing laterally along one of the several anhydrite layers in the Salado Formation. A previous NRC committee analyzed the brine accumulation issue and concluded that “the formation of an abundant mobile fluid in a repository at the WIPP site . . . is very improbable.” Nevertheless, the same committee recommended a “well conceived experimental program at WIPP to reduce remaining uncertainties” (NRC, 1988). The present committee is also in favor of a monitoring program to complement DOE's current program. It would be informative to monitor the brine flow rate into the first panel, or panels, of the WIPP facility that are filled and sealed. Monitoring from inside the face of the seal should be possible for decades after the panel is sealed and would contribute to enhancing confi-

OCR for page 20
Page 23dence in the performance of the repository. Monitoring the humidity and the accumulation of standing brine would indicate the ingress of brine, although salt mines are notoriously dry and probably no standing water will develop. The monitoring of brine inflow and of the humidity in the WIPP should continue at least until the shafts are sealed and longer if feasible. According to Knowles and Economy (2000), the brine inflow rate will be maximum within the first 50 to 100 years from the mining of the repository and will stabilize progressively after 200 years. The rate of brine inflow depends on the porosity of the medium. The mining of the salt in the repository creates alterations of the stress field of the surrounding rock and forms micro fractures in the salt around the excavation (disturbed rock zone, or DRZ). Compared to the intact salt, the DRZ will have an increased porosity because of all the micro fractures. Over time, the porosity of the DRZ decreases as salt creep continues, thereby decreasing brine inflow. Therefore, the monitoring of brine inflow is particularly important during the pre-closure phase. At closure, the panel conduits for the instrumentation would be plugged permanently to ensure the sealing of the repository. Maintaining instrumentation at the repository horizon beyond closure of the shafts could be impractical, unless new technologies allow remote monitoring of the repository avoiding instrument conduits through the seals. Recommendation: The committee recommends pre-closure monitoring of the WIPP repository to gain information on brine migration and moisture access to the repository. Observation should continue at least until the repository shafts are sealed and longer if possible. The committee recommends that the results of the on-site monitoring program be used to improve the performance assessment for recertification purposes. Gas Generation in the Repository Gas generation within the WIPP is one of the issues for consideration in the overall performance of the repository. There are two possible effects of gas in the repository. The first is a physical effect due to pressure buildup from any gas. Gas may generate sufficient pressure to eject repository materials during a human intrusion event. Gas pressure could also retard creep closure and brine inflow. Gas pressure in the repository is considered one of the main uncertainties in the PA concerning radionuclide release from the WIPP (Berglund et al., 2000; Helton et al., 2000d; Stoelzel et al., 2000; Vaughn et al., 2000). A performance assessment scenario that could cause violation of the EPA repository release limits involves ejection of waste material through a borehole. In this scenario, it is calculated that the gas pressure at the repository horizon has to be greater than approximately 8 megapascals1 to result in ejection of cuttings, cavings, and spallings that might contain radionuclides from the repository (Berglund, et al., 2000; DOE, 1996). If the gas pressure approaches the lithostatic pressure, a radionuclide release along open fractures could result. The second effect of gas generation in the repository is chemical. The main gaseous products potentially formed in the repository are carbon dioxide (CO2), hydrogen (H2), methane (CH4), nitrogen (N2 or 1The value of 8 megapascals is the pressure exerted by a column of brine-saturated drilling fluid at the depth of the repository (Stoelzel and O'Brien, 1996). This threshold in pressure was calculated in the PA on the basis of drilling technologies using mud. The public strongly criticized this assumption because it did not take into account the increasingly popular air drilling technology. However, the EPA analyzed the PA, performed supplementary calculations, and reached the conclusion that the repository would still be in compliance with release limits, even in the event of a human intrusion through air drilling (EPA, 1998).

OCR for page 20
Page 24various nitrous oxides), and hydrogen sulfide (H2S) (Lappin and Hunter, 1989). Most of concerns come from H2 and CO2. Hydrogen is a flammable gas and, in the presence of certain amounts of oxygen or water vapor, could lead to an explosion. In the case of CO2, its solubility in any brine seeping into the repository could lower the pH of the brine, which would increase actinide solubility (Novak and Moore, 1996) as shown later in this chapter. The sources of gas generation in the repository are three: radiolysis, metal corrosion, and bacterial action. The PA shows that total gas production is negligible under humid conditions for all substrates.2 The main uncertainty concerns CO2 production by microbial degradation reactions; this uncertainty was acknowledged in the PA by assigning a probability of 0.5 to the occurrence of significant microbial activity (DOE, 1996; Larson, 2000). The previous NRC committee on the WIPP (NRC, 1996a) also concluded that gas generation will be minimal, even when microbial degradation of organic material is taken into account. Although this committee concurs with the previous NRC committee and with the DOE that there is “minimal” evidence of gas generation in the WIPP, uncertainties concerning gas generation are still present. The committee is concerned that experimental data were extrapolated from laboratory experiments performed under conditions that are not indicative of the actual environment in the repository. For instance, in the case of gas generated by radiolysis of brine and organic materials, the majority of experiments were performed with high doses of radiation, which does not apply to TRU waste. Moreover, factors that significantly decrease the rate of radiolysis—matrix depletion, pressure, and inhibition by other chemical compounds—were not taken accurately into account (INEEL, 1998; Molecke, 1979b). In the case of microbial degradation of cellulosic compounds, rates of gas generation were extrapolated from laboratory experiments performed under humid conditions (70 percent humidity), which are not representative of the intrinsic dryness of salt repositories (Francis et al., 1997). In the case of metal corrosion, gas will not result without brine inflow, an event strongly affected by uncertainties. Furthermore, the nature of the interactions between gas-phase chemicals, the influence of pressure, and of corrosion rates is still not well understood (Telander and Westerman, 1996). The gas generation rate is expected to be maximum during the pre-closure period because it depends on the brine inflow rate for microbial degradation, corrosion, and radiolysis (NRC, 1996a). As shown in the previous section, the brine inflow rate is expected to be maximum at the beginning of the repository life. Therefore, it is important to monitor gas generation rates and volumes during the first 35 to possibly 100 years. Furthermore, continuous monitoring for gas could lead to the early detection of anomalous behavior of the repository. Recommendation: The committee recommends pre-closure monitoring of gas generation rates, as well as of the volume of hydrogen, carbon dioxide, and methane produced. Such monitoring could enhance confidence in the performance of the repository, especially if no gas generation is observed. Observation should continue at least until the repository shafts are sealed and longer if possible. The results of the gas generation monitoring program should be used to improve the performance assessment for recertification purposes. 2A substrate is a generic material, whether it is metal, natural fiber, or plastic, that generates gas via various mechanisms described in this section.

OCR for page 20
Page 25 Magnesium Oxide Backfill In the framework of repository performance confirmation, an issue closely related to brine access and gas generation in the WIPP is the performance of magnesium oxide (MgO) used as backfill. The backfill is introduced in the rooms to fill voids in the disposal locations, thus enhancing the healing process and facilitating the encapsulation of waste in salt. The choice of MgO as backfill is based on its chemical properties in addition to its properties as backfill. If brine is present in the repository, MgO will mix with it to form a compact material that will encapsulate the waste (Berglund et al., 1996). The water uptake of MgO from the brine will result in a volume expansion and in the precipitation of salt from the brine that will heal all pathways for later brine penetration. 3 If brine is not present in the rooms, creep encapsulation of waste would not progress as readily. The chemical role of MgO is to provide some control of the chemical environment of the waste by reacting with brine and scavenging the CO2 potentially formed in the repository. In presence of CO2, brine pH will be decreased by the formation of carbonic acid. In the acid pH range, soluble actinide carbonate complexes can then form, increasing actinide solubility (Novak and Moore, 1996). In presence of MgO scavenging all CO2, the pH will remain in the alkaline range (9.2 - 9.9), where actinides are less soluble and less likely to be released into the environment. The rationale for this expected action of MgO relies on the following assumptions: 1. There would be significant inflow of brine into the repository's rooms. 2. Microbes would be present and react with organic waste material to form CO2. 3. CO2 would dissolve and acidify the brine by forming carbonic acid. 4. MgO would react with water in the brine to precipitate brucite [Mg(OH)2]. 5. Brucite would remove carbonic acid from the solution to form magnesite (MgCO3) via intermediate products such as hydromagnesite [4MgCO3•Mg(OH)2•4H2O]. 6. These reactions with MgO would maintain the pH of brine between 9.2 and 9.9. As mentioned in the previous section, there are uncertainties concerning assumption 1 about the presence of a significant amount of brine in the rooms. Since assumption 2 relies on microbial generation of CO2 under repository conditions, it is also affected by uncertainties (see previous section). Moreover, it is unclear whether the rates of reactions in assumptions 4 to 6 are sufficiently high to be effective. The committee has several concerns. How quickly does brucite form at 25ºC (Krumhansl et al., 1999; Papenguth et al., 1999)? How quickly does brucite react with carbonate at various carbonate concentrations and brine compositions?4 When will the water trapped in hydromagnesite be returned to the fluid phase?5 There are also uncertainties in other factors related to the chemical environment, including the amount and timing of brine inflow to form the MgO-based chemical buffer and the presence and effectiveness of microbes responsible for the CO2. On the issue of brine inflow, there is evidence that the salt will creep in the rooms and fill all of the openings in 100 to 150 years (Callahan and DeVries, 1991; 3 If the rate of brine inflow is too high, it is uncertain whether MgO can form a compact material around the waste. 4 The rate to reach the compliance objective (26 mole percent MgO converted) decreases with decreasing CO2 partial pressure (Krumhansl et al., 1999). 5 The time to transform hydromagnesite to magnesite was reported to vary between 18-200 (Zhang et al., 1999) and 2.5-1,500 years (Krumhansl et al., 1999), depending on brine composition.

OCR for page 20
Page 26 Knowles et al., 2000; Stone, 1997). The absence of void spaces should provide additional protection against extensive chemical reactions between brine and the waste in a short period of time.6 The use of MgO in the repository as chemical backfill raises the additional issue of its placement. Because MgO must be in close contact with the drums to better scavenge all CO2 generated from the waste and because of the way the drums are stacked in the rooms, it is not possible to add MgO mechanically after the room is filled. MgO backfill is as a dry, granular, pelletized material packaged in bags of two different sizes: a smaller bag of about 25 pounds, called the “minisack,” and a large bag weighing approximately 4,000 pounds, known as the “supersack.” The minisacks are placed manually around and between the drums, and the supersacks are placed with a forklift on the top of each waste stack. Based on a study by the DOE, it appears that emplacing MgO around the waste adds about 0.726 person-rem per year to the collective dose caused by waste handling. Given that the expected collective dose to waste-handling personnel is 14.6 person-rem per year, this corresponds to about 5 percent of the total dose incurred from waste operations (DOE, 2000b). The committee does not consider this additional dose to be significant. However, once remote-handled (RH) waste and possibly high-specific-activity waste in CH waste such as plutonium-238 or americium-241 are introduced into the repository, the exposures to personnel placing the MgO bags will be considerably increased. Considering the uncertainties about the chemical performance of the MgO backfill, the committee questions the value of its use in the repository. The same concern was expressed by some of the peer review panels of the CCA (DOE, 1996, Chapter 9.3.2). This is especially true given the small but measurable additional radiation exposure to workers involved in MgO bags emplacement. The committee is not convinced of any major chemical advantages of the MgO backfill and, if its benefits to the long-term performance of the repository cannot be verified, the option to discontinue its use should be considered. Recommendation: The committee recommends that the net benefit of MgO used as backfill be reevaluated. The option to discontinue emplacement of MgO should be considered. Salt Healing and Disturbed Rock Zone Integrity The period between placement of waste and closure of the repository provides a window of opportunity to monitor significant deformation of the salt and self-healing of the DRZ. The DRZ is the zone around an excavation in the host rock salt where the stress field has been modified sufficiently to cause the formation of microfractures in the rock salt. Substantial deformation of the salt will occur during the operation phase, which is important in assessing the self-sealing (healing) characteristics of the repository. After an initial period of rapid deformation (a few years to decades), the rooms are expected to deform, crush, and be entirely entombed by salt within 100 to 150 years (Callahan and DeVries, 1991; Knowles et al., 2000; Stone, 1997). Since the waste drums will be immobilized in a relatively short period of time (compared to 10,000 years of compliance), the radionuclide mobility values used in the performance assessment might have been overestimated. This implies less migration of radionuclides from the repository into the environment. In addition to the general deformation and local healing of rooms and panels, an important general healing must take place in the DRZ around rooms, panels, and shafts to achieve complete closure of the disposal region. The effect of the DRZ around WIPP rooms and panels and around the shaft seal system is important in assessing the safety performance of the repository. A complete analysis of the performance of the 6A short period of time compared to the 10,000 years mentioned in the containment requirements (see Sidebar 1.4).

OCR for page 20
Page 27shaft seal system is given in Hansen and Knowles (2000). As with all the factors affecting the performance of the repository and because of the complexity of salt rock behavior, there is uncertainty in the timing and degree of self-healing of the DRZ needed to achieve the expected isolation in the mined regions. In the committee's opinion, there is also uncertainty concerning the behavior of rigid panel seals in the ductile salt surrounding them. Therefore, frequent monitoring during the pre-closure period and assessing the status of room deformation and DRZ healing are the best approaches for reducing the uncertainties associated with closure of the waste disposal area. Recommendation: The committee recommends pre-closure monitoring of the status of room deformation and DRZ healing. Seal performance should also be assessed. Observation should continue at least until the repository shafts are sealed and longer if possible. The results of the monitoring of room deformation and DRZ healing should be included in the PA and used for recertification purposes. SITE CHARACTERIZATION ISSUES The committee finds that there are a number of site characterization actions that would decrease uncertainties in the long-term performance of the repository. Among these, site characterization issues related to human activities are particularly important because they constitute the major risk of radionuclide release, according to the performance assessment (NRC, 1996a; Rechard, 2000). Site characterization issues and activities are described in the sections below. Geohydrological Characterization of the Rustler Formation The WIPP disposal panels and rooms are located in the Salado Formation, approximately 660 meters from the ground surface, as shown in Figure 1.3. The Rustler Formation, overlying the Salado Formation, consists of five sequences (members) of thin-bedded strata. The Culebra Dolomite member, also called simply Culebra, is the second member from the bottom of the formation and is the most transmissive unit in the Rustler. Thus, the Culebra is important to the groundwater flow model for the WIPP site. The geologic and hydrologic setting of the WIPP have been thoroughly described in Corbet and Swift (2000). A detailed description of radionuclide transport in the Culebra can be found in Ramsey et al. (2000). The Culebra provides pathways for the release of radionuclides into the environment in all main human intrusion scenarios (see Appendix B). These pathways can conceivably be developed when new wells are drilled through the Culebra. High-pressure fluids are used in the drilling of oil, gas, and injection wells to contain the flow from the high pressure in formations contacted during the drilling process. Formations at shallower depths, which tend to be at low pressure, are protected from the high-pressure drilling fluids by borehole casings. However, if the drilling intersects a pressurized brine reservoir before the borehole casing is placed, and if the pressure in the formation is unexpectedly higher than the pressure exerted by the drilling fluid, the high-pressure formation fluids could flow into the wellbore and cause an underground blowout into the Culebra. Drillers would use a blowout preventer to contain any immediate surface release of brine from the repository horizon. However, release to the Culebra could be synonymous with release to the accessible environment if there were high flow rates and little retardation7 of radionuclides. 7Parameter that describes the ratio of the net apparent velocity of the concentration of a particular chemical species to the velocity of a non-reactive species.

OCR for page 20
Page 28 All human intrusion assessment models in the PA require some retardation in the Culebra to meet the EPA's repository performance requirements. Similarly, the PA models require a low flow velocity in the Culebra. Unfortunately, these models are not based on sufficient hydrological characterization of the Culebra. There is uncertainty about flow directions, flow rates, retardation characteristics, and the amounts and location of groundwater recharge and discharge to and from the Culebra. This is due partly to uncertainties about the density, size, and spatial distribution of fractures and potential karstic features. These uncertainties can be reduced through a well-designed monitoring program. The monitoring program should include angled boreholes to verify assumptions about vertical fractures or karst conduits; monitoring wells to check on conditions of recharge and discharge, water levels, and chemical properties. The program should also include a series of tracer tests to determine spatial flow rates of groundwater and local tracer tests, including the use of new logging technologies. Tracer tests should include suites of conservative tracers injected in differing wells to test the complexities of the flow system over and beyond those withdrawn by the LWA. The tests should span the entire preclosure phase of the repository (35 to 100 years). New data should be implemented continually into scenario models, and PA calculations should be revised as appropriate. Recommendation: The committee recommends a monitoring program to characterize the geohydrology of the Culebra Dolomite. Tests and measurements that should be considered include angled boreholes, natural gradient tracer tests, and additional pump or injection tests. These new data should be used to confirm, or modify, the conceptual and numerical models now proposed as reasonable simulation of the actual system. Oil, Gas, and Mineral Production The oil, gas, and mineral reserves in the vicinity of WIPP are considerable. As shown in the interim report (see Appendix A1, Figure 2), there have been multiple drilling operations near the WIPP site and a future increase in production activities is expected. As indicated in the previous section, brine (or any fluid) inflow to the disposal region of the WIPP repository is a serious threat to the containment of radionuclides in the repository. Therefore, it is critical that pathways are not created by human intrusion, either intentionally or unintentionally. Such pathways would allow transport of radioactive materials from the repository to the surface or would bring water or brine in contact with the substances stored in the repository. No human intrusion should occur during the first 100 years of the repository's life because of the active institutional controls. However, drilling activity will increase progressively during the period of passive institutional controls (100 to 700 years) and will not be controlled beyond that period. Uncontrolled extraction activities would increase the probability of drilling directly into the repository. Extraction activities can be divided into drilling activities and mining activities. Drilling Activities Two scenarios related to drilling activities are of particular interest to the WIPP site: the Hartman scenario and the intersection of a pressurized brine reservoir. 1. The Hartman scenario. In 1993, while drilling in the Rhodes Yates oil field located approximately 45 miles from the WIPP site, Mr. Hartman experienced a well blowout followed by an uncontrollable flow of brine to the surface (Silva, 1996). This event has come to be known as the Hartman scenario

OCR for page 20
Page 29 (see additional details in Appendix B, Box B.1). The reason for the blowout has not been fully determined, although there is evidence that it may have been caused by a high-pressure, water-flooding operation approximately 1 mile from the well that blew out. In oil-producing regions such as southeastern New Mexico, it is common to inject high-pressure fluids into the deep rock formations of the subsurface. The purpose of these fluid injections is to stimulate secondary recovery of oil in partly depleted oil reservoirs (e.g., by water flooding) or to dispose of large volumes of brine produced simultaneously with oil. If there is a failure in the well casing or in the grout or cement outside the casing, fluid can leak into overlying formations and flow laterally along one of the many anhydrite layers in the Salado (NRC, 1996a). Mr. Hartman might have drilled into a hydraulic fracture possibly induced by such water-flooding operation, causing the well to blow out. Bredehoeft and Gerstle (Bredehoeft and Gerstle, 1997; Gerstle and Bredehoeft, 1997) studied the implication of the Hartman scenario for the safety of the WIPP. They argued that if there were an oilfield water-flooding operation in the vicinity of the WIPP, a large amount of brine could flow from a leaky injection well and induce a hydraulic fracture in the anhydrite (or marker bed) directly above or below the WIPP repository (see Appendix B, Box B.2). If, at some later time, another well were drilled through the repository and into this brine-filled fracture, the high-pressure brine in the fracture could flow through the borehole and flood the repository causing a release of radioactive materials. Bredehoeft's analysis was disputed by researchers at Sandia National Laboratories (SNL; Swift et al., 1997; Vaughn et al., 1998). The discussion focused on the size of this potential hydraulically induced fracture and on whether this fracture could reach the anhydride beds directly below or above the repository site. The committee's opinion is that there are considerable uncertainties concerning both the mechanism of the Hartman scenario and its likelihood to develop at the WIPP site. For instance, if the hypothesis of a hydraulically induced fracture were valid, and the fracture would indeed extend directly below or above the repository, a surge of brine would be expected only when the drillbit penetrates the brine-filled fracture. The volume of brine inflow would not be large enough to damage the repository because hydraulic fractures have small opening widths and high internal flow resistances. Furthermore, a leaky well could not provide sufficient energy and fluid volume to cause a brine inflow into the repository for an extended period of time; also, the energy stored in the room and in the fracture would not be enough to push the waste to the surface. In addition, the repository is partitioned into isolated rooms, which will be closed progressively by salt creep, so that radionuclides should not be mobilized by the brine inflow. Finally, based on the information gathered and on geotechnical subcommittee's discussions, it appears that the geological setting of the WIPP is different from that of the Rhodes Yates oil field. The geological configuration near the WIPP site is likely to interfere with fluid movement thereby reducing the likelihood of flow from a hydraulic fracture into the repository. Therefore, in the committee's opinion, the Hartman Scenario is not likely to cause a problem in the performance of the repository. 2. Intersection of a pressurized brine reservoir. Groundwater containing high levels of dissolved solids (brine) may occur beneath the WIPP site either as discrete pockets (brine pockets) or as a saturated continuum. The committee uses the term “brine reservoir” to refer to both of these occurrences. At present, there is a great deal of uncertainty as to the location and form (i.e., discrete pocket or saturated continuum) of brine reservoirs beneath the WIPP repository. The committee recognizes that direct drilling through the repository into underlying high-pressure brine reservoirs could result in a release of radionuclides. A survey study of brine reservoirs in the Castile Formation (Popielak et al., 1983) has suggested that the brine reservoirs in the area are not large enough to affect the safety of the WIPP site and that

OCR for page 20
Page 30there is no high-pressure brine reservoir directly underlying the repository. However, this finding is challenged by Silva et al. (1999). Using data from test well WIPP-12, Silva demonstrated that the probability of a large brine reservoir, approximately 260 meters below the repository, is rather high. The issue remains unresolved at the present time. Direct drilling (see Appendix B, Box B.1, and Box B.4) into the WIPP repository would allow circulating drill fluid to bring radioactive materials to the surface through a borehole as cuttings or spallings. In the performance assessment, SNL evaluated different possibilities of drilling into a brine reservoir (see Appendix B). In the committee's opinion, the upsurging pressure from drilling through a pressurized brine reservoir could be counteracted by the weight of drilling mud. However, the situation could be serious if the brine reservoir were large and contained a significant amount of energy. An intersection with such a reservoir, although extremely rare, could cause the well to blow out and could result in a catastrophic safety problem for the WIPP. In the committee's opinion, when the drillbit penetrates a brine reservoir below the repository, there would be an initial surge of brine flowing through the borehole into the repository, but the rate of brine inflow would decrease rapidly unless this high-pressure brine reservoir had a gas subpocket above it. Because of the low compressibility of brine, without a gas subpocket, the energy stored in the reservoir would not be sufficient to cause a large brine upsurge through the borehole into the repository. It is therefore important to determine the existence of a brine reservoir directly below the repository. This would be done using seismic techniques, which cannot measure the pressure in the reservoir but can detect its size. The committee recognizes that small brine reservoirs, including brine occurring as a saturated continuum, could not be detected by seismic surveys, or other noninvasive remote sensing techniques. Most seismic surveys are performed from the surface. However, it is possible to perform measurements at a depth, such as in wells or from within the repository. There would be advantages to performing a seismic survey at repository depth (660 meters below the surface) because the unwanted signal from near-earth formations could be eliminated. The committee is aware of the numerous geophysical surveys that have been performed on the WIPP area in the past (ETC, 1988; Popielak et al., 1983; Silva et al., 1999) and does not suggest repeating what has been already done. However, seismic interpretation technology has improved dramatically in the last decade. These improvements, including but not limited to the almost universal three-dimensional seismic techniques, have greatly enhanced resolution capability and are currently used in the oil industry. Detailed three-dimensional seismic studies results, however, are often highly proprietary because they are performed by the oil industry. The DOE could consider acquiring the results of these studies to obtain new information on possible brine reservoirs in the region. In case a brine reservoir were found beneath the WIPP site and its size were larger than what is already taken into account in the PA, then the DOE should conduct an extensive review of the impact of such a reservoir on the repository performance. A basis would then exist to take appropriate action to ensure the safety of the repository. If the reservoir is pressurized, the option of drilling a well into it to release the pressure could be considered. In case of drilling, precautionary methods, such as directional drilling, should be taken to prevent brine from entering the repository. Recommendation: The committee recommends the utilization of seismic survey techniques to detect the presence of a large brine reservoir below the repository.8 In case a brine reservoir were 8The committee recognizes that small brine reservoirs, including brine occurring as a saturated continuum, could not be detected by seismic surveys, or other noninvasive remote sensing techniques.

OCR for page 20
Page 31 found beneath the WIPP and its size were larger than what is already taken into account in the PA, then the DOE should conduct an extensive review of the impact of such a reservoir on the repository performance. A basis would then exist to take appropriate action to ensure the safety of the repository. Mining Activities A further human activity that could threaten the safety of the repository is potash mining in proximity of the WIPP site. Potash mining could impact the performance of the repository by modifying flow pathways in the overlying formations or by creating a path for brine intrusion, if methods such as flood or solution mining are employed. The potential impact of potash mining on WIPP performance is not considered significant, but it is important that the DOE monitor during the operational phase all mining activities in close proximity of the area addressed in the LWA to ensure that the WIPP repository performance is not affected. After reviewing the analyses performed for the human intrusion scenarios as a part of the performance assessment and given the reasons mentioned above, the committee finds that oil, gas, and mineral activities will not unduly threaten the integrity of the repository. However, there are uncertainties associated with these extraction operations. These uncertainties could be reduced by monitoring and documenting oil, gas, and mineral activities. The DOE could establish a database on oil, gas, and mineral activities in the WIPP area containing information such as: 1. location, depth, and type of each well surrounding the WIPP site; 2. data on accidents or unusual events reported by drilling contractors or operators; 3. data on production-enhancing activities such as water or CO2 flooding, hydraulic or cryogenic fracturing, and acidizing in surrounding wells; 4. production rates of oil, gas, and brine from nearby wells; 5. data on disposal of drill cuttings and brine from the operators; 6. data from abandoned wells, in particular those relevant to gas leakages; and 7. extent of potash mining in the vicinity of the LWA. Recommendation: The committee recommends the development of a database to collect information on drilling, production enhancement, mining operations, well abandonments, and unusual events (accidents and natural events) in the vicinity of the WIPP site. Baseline Radiogenic Analysis of Subsurface Fluids The issue of baseline values for naturally occurring radioactive material (NORM) in the vicinity of the WIPP site is important for future monitoring of any changes in radioactivity levels in and around the site. The reason for concern is that subsurface oil and gas in the vicinity of the site already contains NORM. The potential discovery of radioactive material in oil and gas could mistakenly be assumed to come from the repository and thereby cast a doubt on the performance of the nearby WIPP. One of the findings of the committee's interim report (Appendix A1) identified an absence of radiological baseline information for subsurface brines and hydrocarbons near the site, even though there has been extensive monitoring of radioactivity in the air, soils, fluvial sediments, surface water, shallow groundwater, and populace. Therefore, the committee recommended that the DOE develop and implement a plan to sample oil-field brines, petroleum, and solids associated with current hydrocarbon production to assess the magnitude and variability of naturally occurring radioactive material in the vicinity

OCR for page 20
Page 32of the WIPP site. The radionuclides of interest include those that contribute to the site's NORM background radioactivity and those present in the TRU waste inventory destined for WIPP. The NORM activity may include contributions from potassium-40, isotopes of uranium and thorium, and daughter products such as isotopes of radium. Radionuclides in TRU waste include isotopes of uranium and TRU elements and, in remote-handled TRU waste, fission and activation products. Since some TRU inventory radionuclides are not found commonly in nature, sampling to determine whether such radionuclides are present in the environment may be a good way to distinguish radioactivity due to NORM from that due to TRU waste. Further details can be found in Appendix A1. In its interim report, the committee recommended a simple but reliable analysis of the samples that do not include species depending on equilibria that can be shifted by a change in the chemical or physical parameters of the sample. In response to the interim report, the DOE stated that the New Mexico State University Carlsbad Environmental Monitoring and Research Center (CEMRC) has undertaken a project to carry out the recommended assessment, as part of CEMRC's WIPP environmental monitoring project ( Appendix A2). This project will include “completion of a database of active wells and operators, development of sample collection and handling plans, and identification of commercial sample collection services.” The CEMRC has also developed analytical methods for NORM in subsurface fluids to complement standard methods. The committee supports and encourages the pursuit of this initiative. Recommendation: The committee recommends that the DOE continue the implementation of its plan to sample oil-field brines, petroleum, and solids associated with current and future hydrocarbon production, as necessary to assess the magnitude and variability of NORM in the vicinity of the WIPP site for baselining purposes.9 9On March 12, 2001 the DOE-Carlsbad Field Office informed the committee that the efforts to collect data on NORM have received little support from oil companies and that cooperation seems unlikely. The small number of positive responses received would still not provide enough information to constitute a representative baseline of NORM in the region.