Near-Field Modeling in Frenchman Flat, Nevada Test Site

Page 1

University and Community College System of Nevada Water Resources Center 600429 NEAR-FIELD MODELING IN FRENCHMAN FLAT, NEVADA TEST SITE Karl Pohlmann, Craig Shirley and Roko Andricevic Nevada Operations Office U.S. Department of Energy

Page 2

This report was prepared as an account of work sponsored by the United States Government, Neither the United States nor the United States Department of Energy, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, mark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. This report has been reproduced directly from the best available copy. Available to DOE and DOE contractors from the Office of Scientific and Technical Information, P.O. Box 62, Oak Ridge, TN 3783 T, prices available from (423) 576-8401, Available to the public from the National Technical Information Service, U.S. Department of Commerce, 5285 Port Royal Rd., Springfield, VA 22161.

Page 3

NEAR-FIELD MODELING IN FRENCHMAN FLAT, NEVADA TEST SITE Karl Pohlmann, Craig Shirley and Roko Andricevic Water Resources Center Desert Research Institute University and Community College System of Nevada Nevada Operations Office U.S. Department of Energy Las Vegas, Nevada The work upon which this report is based was supported by the U.S. Department of Energy under

Page 4

The U.S. Department of Energy (DOE) is investigating the effects of nuclear testing in underground test areas (the UGTA program) at the Nevada Test Site. The principal focus of the UGTA program is to better understand and define subsurface radionuclide migration. The study described in this report focuses on the development of tools for generating maps of hydrogeologic characteristics of subsurface Tertiary volcanic units at the Frenchman Flat Corrective Action Unit (CAU), The process includes three steps. The first step involves generation of three-dimensional maps of the geologic structure of subsurface volcanic units using geophysical logs to distinguish between two classes: densely welded tuff and nonwelded tuff. The second step generates three-dimensional maps of hydraulic conductivity utilizing the spatial distribution of the two geologic classes obtained in the first step. Each class is described by a correlation structure based on existing data on hydraulic conductivity, and conditioned on the generated spatial location of each class. The final step demonstrates the use of the maps of hydraulic conductivity for modeling groundwater flow and radionuclide transport in volcanic tuffs from an underground nuclear test at the Frenchman Flat CAU, The results indicate that the majority of groundwater flow through the volcanic section occurs through zones of densely welded tuff where connected fractures provide the transport pathway. Migration rates range between near zero to approximately four m/yr, with a mean rate of 0.68 m/yr. This report presents the results of work under the FY96 Near-Field Modeling task of the UGTA program.

Page 6

1. Generalized lithostratigraphic and hydrogeologie columns for the Frenchman Flat CAU.......................................................... ................................................. 2 2. Location of study area in Frenchman Flat, NTS. Locations of boreholes and underground nuclear tests are also shown................................................................... 5 3. Plot showing realizations AA and AB of the densely welded tuff ............................... 7 4. Plot showing realizations AC and AD of the densely welded tuff ............................ 8 5. Plot showing cells having hydraulic conductivity greater than 2.0 m/d.................. 14 6. Plot showing cells having hydraulic conductivity greater than 1.5 m/d......................... 14 7. Plot showing cells having hydraulic conductivity greater than 0.67 m/d.......... ............ 15 8. Plot showing cells having hydraulic conductivity greater than 0.1 m/d.......................... 15 9. Plot showing cells having hydraulic conductivity greater than 2.0 m/d for 10. Plot showing cells having hydraulic conductivity greater than 2.0 m/d for 11. Plot showing cells having hydraulic conductivity greater than 2.0 m/d for 12. Plot showing cells having hydraulic conductivity greater than 2.0 m/d for 13. Configuration of the physical domain for flux and transport simulations..................... 19 14. Distances traveled by the centers of mass of 30 plume realizations in 275 years. ... 21 15. Plots showing (a) tritium breakthrough at 500 m downgradient from source, (b) breakthrough curves normalized for tritium decay, and (c) ensemble breakthrough curve at mean arrival time. .......................................... 23 1. Underground nuclear tests in northern Frenchman Flat................ 4 2. Values of parameters used to generate hydraulic conductivity Fields...................... 13 3. Values of parameters used in transport simulations. ....................................................... 20

Page 7

Many years of underground nuclear testing at the Nevada Test Site (NTS) has introduced large quantities of radionuclides into the subsurface, including the saturated zone. Some of the radionuclides are readily transported in groundwater and therefore pose a risk of migrating great distances from their sources. The U.S. Department of Energy (DOE) has designated the various source areas on the NTS as Corrective Action Units (CAUs), and many of them will be the focus of detailed investigation of radionuclide transport in groundwater. Frenchman Flat is located above the regional carbonate aquifer (although the hydraulic connection is unclear) and in close proximity to an NTS boundary, thus the Frenchman Flat CAU is one of the first to be investigated. The high degree of spatial variability resulting from natural geologic heterogeneity is incorporated in the approach utilized here. The availability of hydraulic data in most areas of the NTS is limited, so geophysical data have been used as surrogates to characterize the heterogeneity of hydrogeologie units (Pohlmann and Andricevic, 1994; Shirley etal, 1996). These studies defined various hydrostratigraphic units in the volcanic tuff section and characterized their geometric configurations (thickness, horizontal and vertical continuity). Using this information and data on the spatial characteristics of selected hydraulic properties, hydraulic conductivity can be represented as a spatially-correlated random field. Contaminant transport can then be simulated using a This report presents the results of the application of newly-developed tools for generating three-dimensional maps of hydraulic conductivity of the volcanic tuffs at the Frenchman Flat CAU. Although most of the nuclear tests at Frenchman Flat occurred in the valley fill, it is generally thought that migration of radionuclides in groundwater away from the testing areas must first pass through the volcanic section before entering the underlying regional carbonate aquifer. Existence of direct pathways to the carbonates have been postulated, but information about them was insufficient at the time of this study to be incorporated in the model. The hydraulic conductivity maps are used as input for groundwater flow and transport models of radionuclide migration in groundwater from underground nuclear tests at the Frenchman Flat CAU. Modeling is conducted at the scale of one or several shots, which is termed the “near-field” scale. This report is a companion report to Shirley et al. (1996), which presents maps of hydrostratigraphic units in northern Frenchman Flat, upon which the conductivity fields presented here are based. Frenchman Flat is an alluvium-filled closed basin that is elongated along the northeast-southwest axis. The basin is bounded on the south and east by the Ranger Mountains and Mercury Ridge outcrops and on the west by the Hampill Hill-Mount Saylor High (Miller and Healy, 1986). 'Winograd and Thordarson (1975) describe the four principal hydrogeologie units that are important to groundwater flow below Frenchman Flat. They are the lower carbonate aquifer, the tuff

Page 8

(m) Geologic Units Hydrogeologie 13 ■■■■■■■■■■■ Qal: Alluvium; Quaternary to Tertiary in generally Tmr; Rainier Mesa Tuff; displays variable welding from bedded, air-fall tuffs to Tp: Paintbrush Tuff; subdivided into five formations (Dreilack et al., 1987) (0 l Tms; Tuff of Mt. Saylor; now a member of Tvu: Undifferentiated Volcanics; assumed to be a volcanic sequence similar to Yucca Flat. Not penetrated by boreholes in or near the study area. Pz: Undifferentiated carbonates; depth of contact is estimated from gravity data, displayed thickness is truncated. Figure 1. Generalized lithostratigraphic and hydrogeologie columns for the Frenchman Flat CAU, aquitard, the welded-tuff aquifer, and the valley-fill aquifer. The following summary is based primarily on Winograd and Thordarson’s work. Generalized lithostratigraphic and hydrogeologie columns for the Frenchman Flat CAU are shown in Figure 1.

Page 9

The lower carbonate aquifer is composed of Middle Cambrian through Devonian carbonate rocks, and reaches a saturated thickness of 1000 m or more. The porosity and hydraulic conductivity of the carbonate matrix are both very low, so fractures provide the means for most groundwater flow in the unit. Although the fracture frequency is widely variant, overall fracture conductivity is high enough that the carbonate aquifer is one of the most transmissive on the NTS; the IT/GeoTrans database reports a range of hydraulic conductivities from 0.09 to 731 m/d based on 17 measurements on the NTS (Rehfeldt et al, 1996). The lower carbonate aquifer, which provides for most of the interbasin flow in the regional aquifer system, appears to underlie most of Frenchman Flat. The tuff aquitard consists primarily of nonwelded to partially-welded ash-flow tuff and ash-fall tuff of Tertiary age. Other rocks included in this unit are tuff breccia, breccia flow, tuffaceous sandstone, siltstone, mudstone, freshwater limestone, and minor densely welded tuff, all of Tertiary age. The tuffs in this unit have been altered in situ to zeolites and clays and, thus, exhibit high interstitial porosity but very low interstitial hydraulic conductivity. In addition, since fractures and faults in this friable, nonwelded unit do not remain open and available for groundwater flow, overall hydraulic conductivity is very low. The tuff aquitard is as much as 400 m thick below Frenchman Flat and separates the valley-fill and welded-tuff aquifers from the underlying lower carbonate The welded-tuff aquifer is composed of late Miocene and Pliocene nonwelded to densely welded ash-flow tuff, with minor amounts of ash-fall tuff. Interstitial porosity is inversely related to the degree of welding; nonwelded tuff may exhibit interstitial porosity greater than 0.50 while the interstitial porosity of densely welded tuff is generally less than 0.05. nonwelded tuffs in the welded-tuff aquifer have very low hydraulic conductivity for the reasons previously mentioned. On the other hand, the densely welded ash-flow tuffs, which have very low interstitial porosity and hydraulic conductivity, tend to be brittle, which aids in the formation of joints and fractures. These features tend to remain open after formation, and therefore represent important pathways for groundwater flow. The UGTA database (described by Rehfeldt et al, 1996) reports the range of hydraulic conductivity values for the welded tuff aquifer to be 0.09 to 1.71 m/d based on seven measurements on the NTS. The welded-tuff aquifer is present only in the deepest parts of Frenchman The valley-fill aquifer is characterized by alluvial-fan, fluvial, fanglomerate, lakebed, and mudflow deposits in depressions created by post-Pliocence block faulting. Interstitial porosities of over 200 core samples collected from boreholes in northern Frenchman Flat are generally greater than 0.30 (Reynolds Electrical and Engineering Co., 1993a and 1993b). Hydraulic conductivity ranges from 0.30 to 41 m/d based on 14 measurements on the NTS (Rehfeldt et al, 1996). The valley-fill aquifer is present in the central, structurally deepest portion of Frenchman Flat, where the valley-fill/volcanic-tuff contact is below the water table. Water-level data in the valley are limited, but the available evidence suggests that semi-perched conditions occur between the valley-fill aquifer and the lower carbonate aquifer. Horizontal gradients in the valley fill of northern Frenchman Flat appear to be very low (approximately

Page 10

3,0x1 O'4). The evidence implies internal drainage in response to the net downward hydraulic gradient and lack of significant horizontal gradients between Frenchman Flat and adjacent valleys. Therefore, groundwater in the valley-fill aquifer is generally thought to pass through the welded-tuff aquifer and tuff aquitard before reaching the regional lower carbonate aquifer. Groundwater could also flow directly into the carbonate aquifer in locations where the volcanic section is absent. The study area is located in the northern portion of the Frenchman Flat CAU, and measures 2200 m north to south and 2600 m east to west (Figure 2). The valley fill ranges in thickness from 58 m at the northern edge to 506 m at the southeastern comer. The valley fill is saturated only in the southern half where the valley-fill/volcanic-tuff contact dips below the water table surface. Six of the seven underground nuclear tests conducted in the area of study were detonated in the valley fill, between 65 and 247 m above the valley-fill/volcanic-tuff contact (Table 1). The seventh test, U1 lb, was detonated within the volcanic tuff section, approximately 240 m below the valley-fill/ volcanic-tuff contact. All seven tests were conducted in the vadose zone, between 21.9 and 63.4 m above the water table. The volcanic section dips to the south and is estimated to range in thickness from about 500 m at the northern edge to over 700 m at the southern edge. The thicknesses of the volcanic section are based on estimated depths to the Paleozoic carbonates obtained from gravity data (Miller and Healey, 1986) because none of the boreholes in the study area are deep enough to reach the volcanic-tuff/carbonate contact. TABLE 1. UNDERGROUND NUCLEAR TESTS IN NORTHERN FRENCHMAN FLAT. Test Name Date Geologic Unit Distance above U5i Derringer 9-12-66 Valley Fill +64.5 50.0 U5k Milk Shake 3-25-68 Valley Fill +241.0 21.9 Ullb Pinstripe 4-25-66 Volcanic Tuff -238.0 63.4 Ullc New Point 12-13-66 Valley Fill + 135.0 57.0 Ulle Diana Moon 8-27-68 Valley Fill +247.0 56.7 Ullf Minute Steak 9-12-69 Valley Fill +223.0 35.6 Ullg Diagonal Line 11-24-71 Valley Fill +73.5 36.5 *Data from DRI Hydrologic Resources Management Program Geologic Data Base **Based on average water table elevation of 734 m A framework for describing subsurface heterogeneity using stochastic theory (Delhomme, 1979; Dagan, 1986; Gelhar, 1986) is the most effective way to mathematically account for the high degree of variability typically present in hydrogeologie systems. The most common approach involves generating a set of equiprobable geologic maps conditioned on the available hard data. Contaminant migration is then simulated for each map, resulting in a set of responses (e.g.,

Page 11

Zl% OOO Z1B, ooo vie. COT ZZO, OCD Figure 2. Location of study area in Frenchman Flat, NTS. Locations of boreholes and underground nuclear tests are also shown.

Page 12

contaminant migration described by breakthrough curves and/or spatial moments) that represents a distribution of equiprobable outcomes. Pohlmann and Andricevic (1994) present a methodology for generating equiprobable maps of subsurface heterogeneity using all available data while focusing on connectivity patterns of zones having high hydraulic conductivity. The methodology utilizes the sequential indicator simulation (SIS) algorithm to augment sparse hydraulic information with more widely available geophysical information. The SIS algorithm has been implemented as ISIM3D, which is a three-dimensional, multiple indicator, conditional simulation computer program developed by Gomez-Hemandez and Srivastava (1990). Selected borehole geophysical tools (e.g., resistivity, gamma) whose signals demonstrate a correlation to a particular hydrogeologie attribute (e.g., porosity, permeability) are used as conditioning data for the SIS simulations. The continuous range of the geophysical signal is divided into classes, each of which …