1. PURPOSE This Scientific Analysis report describes the methods used to develop numerical grids of the unsaturated hydrogeologic system beneath Yucca Mountain. Numerical grid generation is an integral part of the development of the Unsaturated Zone Flow and Transport Model (UZ Model), a complex, three-dimensional (3-D) model of Yucca Mountain. This revision incorporates changes made to both the geologic framework model and the proposed repository layout. The resulting numerical grids, developed using current geologic, hydrogeologic, and mineralogic data, provide the necessary framework to: (1) develop calibrated hydrogeologic property sets and flow fields, (2) test conceptual hypotheses of flow and transport, and (3) predict flow and transport behavior under a variety of climatic and thermal-loading conditions. The technical scope, content, and management of this Scientific Analysis report was initially controlled by the planning document, Technical Work Plan (TWP) for: Unsaturated Zone Sections of License Application Chapters 8 and 12 (BSC 2002 [159051], Section 1.6.4). This TWP was later superseded by Technical Work Plan for: Performance Assessment Unsaturated Zone (BSC 2002 [160819]), which contains the Data Qualification Plan used to qualify the DTN: MO0212GWLSSPAX.000 [161271] (See Attachment IV). Grids generated and documented in this report supersede those documented in previous versions of this report (BSC 2001 [159356]). The constraints, assumptions, and limitations associated with this report are discussed in the appropriate sections that follow. There were no deviations from the TWP scope of work in this report. Two software packages not listed in Table IV-2 of the TWP (BSC 2002 [159051]), ARCINFO V7.2.1 (CRWMS M&O 2000 [157019]; USGS 2000 [148304]) and 2kgrid8.for V1.0 (LBNL 2002 [154787]), were utilized in the development of the numerical grids; the use of additional software is accounted for in the TWP (BSC 2002 [159051], Section 13). The use of these software packages is discussed in Sections 3 and 6.1.1. The steps involved in numerical grid development include: (1) defining the location of important calibration features, (2) determining model grid layers and fault geometry based on the Geologic Framework Model (GFM), the Integrated Site Model (ISM), and definition of HGUs, (3) analyzing and extracting GFM and ISM data pertaining to layer contacts and property distributions, (4) discretizing and refining the two-dimensional (2-D), plan-view numerical grid, (5) generating the 3-D grid, with finer resolution at the proposed repository horizon and within the Paintbrush nonwelded (PTn) and ch1 (Uppermost Calico Hills Formation (Table 11)) hydrogeologic units, and (6) formulating the dual-permeability mesh. The products of grid development include a set of one-dimensional (1-D) vertical columns of gridblocks for hydrogeologic-property-set inversions, a 2-D UZ Model vertical cross-sectional grid for fault hydrogeologic-property calibrations, and a 3-D UZ Model grid for additional model calibrations and generating flow fields for Performance Assessment (PA). Note that the repository layout utilized in constructing the numerical grids (BSC 2002 [159527]) has been superseded by a revised repository design (BSC 2003 [161726]; BSC 2003 [161727]) that does not include the lower block area. Because the repository layout used for grid construction includes all of the area covered by the most recent repository design, the use of the older repository design for grid construction will not impact License Application (LA) model calculations that utilize these grids. Numerical grid generation is an iterative process that must achieve a proper balance between desired numerical accuracy in terms of gridblock size and computational time controlled by the total number of gridblocks. Gridblock size should reflect the scale of the process to be modeled. For example, to capture flow and transport phenomena along individual waste emplacement drifts, gridblock thickness and width should not exceed the drift diameter or the drift spacing. For large models, such as the site-scale UZ Model of Yucca Mountain, flow and transport phenomena occurring on scales of less than a few meters cannot be captured. Rather, the model is intended to provide an overview of key UZ characteristics and processes potentially affecting repository performance. Grids must also be adapted to the particular needs of the processes to be modeled because sharp gradients may occur in different domains for different flow processes. At Yucca Mountain, the heterogeneous, variably fractured layers are better represented by a dual-continuum (matrix and fracture) model, rather than a single-continuum approach. Once developed, the UZ Model numerical grids are evaluated for appropriate resolution, representation of important features, and proper gridblock connections. The following list of Features, Events, and Processes (FEPs) was taken from the LA FEP List (DTN: MO0301SEPFEPS1.000 [161496]). The LA FEP List is a revision to the previous project FEP list (Freeze et al. 2001 [154365]) used to develop the list of included FEPs in the Technical Work Plan for: Performance Assessment Unsaturated Zone (BSC 2002 [160819], Table 2-6). The selected FEPs are those taken from the LA FEP List that are associated with the subject matter of this report, regardless of the anticipated status for exclusion or inclusion in Total System Performance Assessment for License Application (TSPA-LA) as represented in BSC (2002 [160819]). The results of this analysis are part of the basis for the treatment of FEPs as discussed in the Total System Performance Assessment-License Application Methods and Approach (BSC 2002 [160146], Section 3.2.2). The cross-reference for each FEP to the relevant sections of this report is also given below. • Stratigraphy (LA FEP Number 1.2.02.02.0A). Stratigraphic information has been extracted from GFM2000 (DTN: MO0012MWDGFM02.002 [153777]), modified using the hydrogeologic unit (HGU) definitions of Flint 1998 [100033], and explicitly included in this report. See Sections 5.2, 6.3, 6.4 and 6.6.3 for discussion of stratigraphy. • Faults (LA FEP Number 2.2.03.01.0A). Faults, which can focus flow, are included as discrete features in the unsaturated zone (UZ) Flow Model and the numerical grids generated in this report. See Sections 5.2, 6.2, 6.3, and 6.6.1 for discussion of faults. 2. QUALITY ASSURANCE Development of this Scientific Analysis report and the supporting analyses have been determined to be subject to the Yucca Mountain Project’s quality assurance program as documented in Technical Work Plan for: Unsaturated Zone Sections of Licence Application Chapters 8 and 12 (BSC 2002 [159051], Attachment I, Work Package P421224UP2) and Technical Work Plan for: Performance Assessment Unsaturated Zone (BSC 2002 [160819], Section 8.2 Work Package AUZM06). Approved quality assurance (QA) procedures identified in the technical work plan (BSC 2002 [159051], Attachment IV; BSC 2002 [160819], Attachment II) have been used to conduct and document the activities described in this Scientific Analysis report. The technical work plan also identifies the methods used to control the electronic management of data (BSC 2002 [159051], Attachment II, Work Package P421224UP2; BSC 2002 [160819], Section 8.4) during the analysis and documentation activities. The procedure AP-SIII.2Q, Qualification of Unqualified Data and the Documentation of Rationale for Accepted Data, was utilized to qualify an input data file (DTN: MO0212GWLSSPAX.000 [161271]) used to delineate the water table. This file was derived from the unqualified DTN: MO0110MWDGFM26.002 [160565]. The derivative file was reviewed and qualified using the Data Qualification Plan found in the Technical Work Plan for: Unsaturated Zone Performance Assessment (BSC 2002 [160819], Attachment III). The data reviews for DTN: MO0212GWLSSPAX.000 [161271] are presented in Attachment IV. This Scientific Analysis reports on hydrogeologic units (HGUs) that are identified as natural barriers that are included in the Q-List (YMP 2001 [154817]) as “Quality Level – 1” items important to waste isolation. The report contributes to the analysis and modeling data used to support performance assessment; the conclusions do not directly impact engineered features important to safety as defined in AP-2.22Q, Classification Criteria and Maintenance of the Monitored Geologic Repository Q-List. INTENTIONALLY LEFT BLANK 3. USE OF SOFTWARE The software used in this study, listed in Table 1, was obtained from Software Configuration Management (SCM), was appropriate for the intended application, and was used only within the range of validation in accordance with applicable software procedures. The qualification and baseline status of each of these codes is given in the Document Input Reference System (DIRS). Table 1. Qualified Software Used in Numerical Grid Development Software Name Version Software Tracking Number (STN) DIRS Reference Number EARTHVISION 5.1 10174-5.1-00 152614 ARCINFO 7.2.1 10033-7.2.1-00 157019 ARCINFO 7.2.1 10033-7.2.1-01 148304 WINGRIDDER 2.0 10024-2.0-00 154785 2kgrid8.for 1.0 10503-1.0-00 154787 TOUGH2 1.4 10007-1.4-01 146496 The use of the codes identified in Table 1 is documented in Section 6 and in the supporting scientific notebooks identified in Section 6. EARTHVISION V5.1 (Dynamic Graphics 1998 [152614]) is used to evaluate and extract data from the Geologic Framework Model (GFM2000) and Integrated Site Model (ISM3.1) files listed in Attachment I, and to create grids utilizing the hydrogeologic units (HGUs) of Flint (1998 [100033], pp. 21–32). ARCINFO V7.2.1 (CRWMS M&O 2000 [157019]; USGS 2000 [148304]) was used to convert potentiometric contours into ACSII format for use in qualification of water table data used in this report (see Attachment IV). The WINGRIDDER V2.0 (LBNL 2002 [154785]) software program is used to generate 1-, 2-, and 3-D gridblock element and connection information in a TOUGH2 format (the primary mesh is an “effective-continuum model,” or “ECM,” mesh) (Pruess 1991 [100413]). Data extracted from the HGU grids generated by EARTHVISION V5.1 are used as input to WINGRIDDER V2.0 to construct the TOUGH2 grid files. WINGRIDDER V2.0 contains new functionality that allows for a proposed repository with multiple subregions. The software program 2kgrid8.for V1.0 (LBNL 2002 [154787]) generates a dual-permeability mesh from a primary ECM mesh for modeling applications, using the TOUGH2 family of codes. TOUGH2 V1.4 (LBNL 2000 [146496]) was used to perform a test simulation to check the 3-D grid, as described in Attachment III. EARTHVISION V5.1, WINGRIDDER V2.0, ARCINFO V7.2.1 and 2kgrid8.for V1.0 are qualified under AP-SI.1Q, Software Management. Microsoft Excel (97 SR-2) and Adobe Illustrator V8.0 were used to plot data and illustrate information generated in the gridding process. Several computations were performed using this commercial off-the-shelf software and are exempt from AP-SI.1Q. All information needed to reproduce the work, including the input, computation, and output, is included in this Scientific Analysis report and the references specified. A fault slope analysis was conducted in Section 6.3. The Slope Grid Calculation utility in EARTHVISION V5.1 was used to determine the slope (rise/run) of each fault within the UZ: this input is listed in the second column of Table 13. Excel97 (SR-2) was used to make the following conversions: (1) arctangent of slope = fault dip in radians, and (2) radians to degrees. The output of these conversion calculations is given in columns 3–5 of Table 13. The specific details of these calculations can be found in Wang 2002 ([159673], SN-LBNL-SCI-213-V1, pp. 73–74). The relative proximity of all boreholes within the UZ Model grid area was examined to determine whether or not neighboring boreholes should be grouped as composite locations. Boreholes that were closer than 80 m to another borehole were paired with the neighboring borehole, and an average borehole location was determined for use in grid construction. All borehole coordinates were converted from Nevada State Plane (NSP) (feet) to NSP (meter) coordinates for the use in the UZ Model grid construction, as discussed in Section 6.2. These calculations are performed using Excel97 (SR-2) in the file “borehole loc.xls” (Output- DTN: LB0208HYDSTRAT.001). The input (NSP feet) coordinates for the boreholes are listed in columns A and B and rows 1 and 2 of the worksheet “All Boreholes” in the Excel file “borehole loc.xls” (Output-DTN: LB0208HYDSTRAT.001). For each borehole combination, where x1, y1 are the coordinates of borehole 1, and x2, y2 are the coordinates of borehole 2, the distance between the boreholes was calculated as the square root of the sum of the squares of the differences in x and y coordinates, given as the equation: 22 (x -x2 ) +(y - y ) (Eq. 1) 1 12 This calculated distance was then converted from feet to meters using the conversion factor 1 ft = 0.3048 m, and the output values are listed in the worksheet “All Boreholes” in the Excel file “borehole loc.xls” (Output-DTN: LB0208HYDSTRAT.001). Boreholes that are within 80 m of one another were then paired together in the worksheet “Selected Boreholes (ft)” in the Excel file “borehole loc.xls” (Output-DTN: LB0208HYDSTRAT.001). Average x, y coordinates (NSP ft values from worksheet “All Boreholes” in the Excel file “borehole loc.xls” (Output-DTN: LB0208HYDSTRAT.001)) were calculated as (x1 + x2)/2 and (y1 + y2)/2. All of the borehole coordinates were then converted to meters using the conversion factor 1 ft = 0.3048 m. The output for this calculation is in the worksheet “Selected Boreholes (m)” in the Excel file “borehole loc.xls” and the file “boreholes_Rick_updated.hol” (Output-DTN: LB0208HYDSTRAT.001), and is also given in Table 10 and Figure 1b. The specific details of these calculations can be found in Wang 2003 ([162380], SN-LBNL-SCI-213-V1, p. 71). Contact elevations from the input file “contacts00el.dat” (see GFM2000 files in Attachment I) were converted from feet to meters using the conversion factor 1 ft = 0.3048 m, and the resulting values are listed in Table II-1. These calculations were performed using Excel97 (SR-2). As mentioned in Section 6.4.1, some of the GFM2000 isochore files were combined or subdivided using the EARTHVISION V5.1 Formula Processor to generate the UZ Model HGU isochores. For validation purposes (see Attachment II), the output UZ Model HGU contact elevations for boreholes in the file “Boreholes.mck” from Output-DTN: LB02081DKMGRID.001 were compared to layer contact elevations in the file “contacts00el.dat” from DTN: MO0012MWDGFM02.002 [153777]. The GFM2000 borehole elevations from “contacts00el.dat” were first converted to feet to meters using the conversion factor of 1 ft = 0.3048 m. The unit contact elevations were then adjusted in the same manner as described in Section 6.4.1 to make the GFM2000 stratigraphic units correspond to the UZ Model HGUs. These calculations were performed using Excel97 (SR-2). The output data for these calculations are recorded in Table II-1 under the columns labeled GFM2000. There are actually two different "foot" units. One of these, the U.S. Survey foot, used for geodetic survey coordinates, is defined as 1,200 m = 3,937 ft, while the standard foot is equal to 0.3048 m (IEEE/ASTM SI 10-1997 [151762], pp. 18, 25). By using the standard foot-to-meter conversion factor (instead of the more appropriate U.S. Survey foot conversion), a small error is introduced into the model. For example, the NSP coordinates for the borehole G-1 (given as 561,000 E, 770,502 N in NSP ft in "contacts00el.dat") convert to 170,993.1 E, 234,849.0 N in NSP m using the conversion factor of 0.3048 m/ft, and to 170,993.4 E, 234,849.5 N using the more appropriate U.S. Survey feet conversion factor. The model grid is not sensitive to the magnitude of the maximum difference (0.5 m) resulting from the use of the 0.3048 m/ft conversion factor. Models used in the development of the UZ Model numerical grids include the Geologic Framework Model (GFM2000) (DTN: MO0012MWDGFM02.002 [153777]), and the Rock Properties Model of the Integrated Site Model, Version 3.1 (RPM3.1) (DTN: MO9910MWDISMRP.002 [145731]). Data from the Mineralogic Model of the Integrated Site Model, Version 3.1 (MM3.1) (DTN: MO9910MWDISMMM.003 [119199]) and the Rock Properties Model 2000 (RPM2000) (DTN: SN0112T0501399.004 [159524]) were evaluated during grid development, but were not directly incorporated into the UZ Model grids (discussed in Section 6.6.3). INTENTIONALLY LEFT BLANK 4. INPUTS The initial stage of grid development begins with the definition of lateral domain and proposed repository boundaries, along with the location of important calibration features (e.g., boreholes). In order to generate a 3-D grid, WINGRIDDER V2.0 (LBNL 2002 [154785]) requires specification of three reference horizons: an upper and lower model boundary (usually the bedrock surface and water table, respectively) and a structural reference horizon that defines layer displacement along fault traces and sets the elevation of the remaining layer interfaces. These reference horizon files consist of regularly spaced x, y, and elevation data. Isochore (borehole unit thickness) maps, consisting of regularly spaced x, y, and thickness data for each model layer, are then stacked above or below the structural reference horizon to build the vertical component of the UZ Model. 4.1 DATA AND PARAMETERS The input data used directly in numerical grid development are summarized in Table 2. The Q- status of each of these Data Tracking Numbers (DTNs) can be determined by referring to the Document Input Reference System (DIRS). Uncertainty in the input data and parameters is discussed in Section 7.1. Table 2. Summary of Direct Input Data Used in Numerical Grid Development Description DTN Data Use3 Geologic Framework Model (GFM2000) MO0012MWDGFM02.002 [153777] 6.1, 6.2, 6.4, Attachments I, II, III Water Table Elevations MO0106RIB00038.001 [155631] GS010608312332.001 [155307] MO0212GWLSSPAX.000 [161271]1 6.3, 6.4.2, Attachment IV Fracture Data for Hydrogeologic Units LB0205REVUZPRP.001 [159525]2 LB0207REVUZPRP.001 [159526]2 6.7 NOTE: 1 2 See Attachment IV for qualification of DTN MO0212GWLSSPAX.000 [161271] Use of these Qualified-Verification Level 2 DTNs as direct input is permissible because this report does not directly support any principal factor. 3 Sections where the data used are described in detail. The primary data feed for UZ Model grids is the Geologic Framework Model (GFM2000) (DTN: MO0012MWDGFM02.002 [153777]). The GFM2000 is a representation of lithostratigraphic layering and major fault geometry in the Yucca Mountain area that was created using geologic mapping and borehole data as primary input data (BSC 2002 [159124], Section 4.1). The model contains information about layer thickness and layer contact elevation, and defines major fault orientation and displacement. The data for each layer and each fault within GFM2000 are available on a regular horizontal grid spacing of 61 × 61 m over the model’s domain (methodology described in BSC 2002 [159124], Section 6.4; data files in DTN: MO0012MWDGFM02.002 [153777]). A total of 50 geologic units and 44 faults are represented in GFM2000. As listed in Attachment I, 42 of these units and 19 faults (those that lie within the UZ Model domain) are incorporated into the 3-D UZ Model grids. Alternate geologic models are not available for use in the UZ Model, nor were they developed. The conceptual model used in the development of GFM2000 is founded on the observation that Yucca Mountain is composed of volcanic rocks originating from several calderas or vent sources (BSC 2002 [159124], Section 6.4.1). The resulting geologic interpretation it represents is the ORD’s geologic model to be used in site-scale process models. GFM2000 files used in UZ Model grid development are listed in Attachment I. As discussed in Sections 5.1 and 6.4.2, the lower UZ Model boundary is based on water table elevations given in DTN: MO0106RIB00038.001 [155631] and the contoured potentiometric surface (DTN: MO0212GWLSSPAX.000 [161271]). Table 3 includes water-level data from DTN: MO0106RIB00038.001 [155631], along with the water-level elevation for USW WT-24 as reported in DTN: GS010608312332.001 [155307]. Table 3. Water Levels in Selected Boreholes Borehole ID1 Water Level Elevation (masl) USW G-2 2 1020.2 USW G-3 2 730.5 USW H-1 2 730.8 USW H-3 2 731.5 USW H-4 2 730.4 USW H-5 2 775.5 USW H-6 2 776.0 USW WT-1 2 730.4 USW WT-2 2 730.6 UE-25 WT#3 729.6 UE-25 WT#4 2 730.8 UE-25 WT#6 2 1034.6 USW WT-7 2 775.8 USW WT-10 776.0 USW WT-11 730.7 Table 3. Water Levels in Selected Boreholes (cont.) Borehole ID1 Water Level Elevation (masl) UE-25 WT# 12 729.5 UE-25 WT#13 729.1 UE-25 WT#14 729.7 UE-25 WT#15 729.2 UE-25 WT#16 2 738.3 UE-25 WT#17 729.7 UE-25 WT#18 2 730.8 UE-25 WT-24 2 840.13 UE-25 J#13 728.4 UE-25 b#1 2 730.6 UE-25 c#2 730.2 UE-25 c#3 730.2 Source: From DTN: MO0106RIB00038.001 [155631] (except data for WT-24). WT-24 datum from DTN: GS010608312332.001 [155307]. NOTES: 1 For simplicity, the borehole names used throughout the remainder of this document drop the USW and UE-25 prefixes. 2 These boreholes lie within or along UZ Model boundaries. 3 Elevation as reported in DTN: GS010608312332.001 [155307]). Fracture hydrogeologic properties (DTNs: LB0205REVUZPRP.001 [159525] and LB0207REVUZPRP.001 [159526]) describing UZ Model layers are used to formulate the dual- permeability (dual-k) meshes for 1-D hydrogeologic-property-set inversions, for 2-D fault property calibration, and for 3-D UZ Model calibration and flow fields for PA. Fracture hydrogeologic properties used for dual-k grid generation are listed in Table 4. Table 4. Fracture Hydrogeologic Properties Model Layer Fracture Porosity (m3/m3) Fracture Aperture (m) Fracture Frequency (m-1) Fracture Interface Area (m2/m3) tcw11 2.4E-02 7.3E-04 9.2E-01 1.6E+00 tcw12 1.7E-02 3.2E-04 1.9E+00 1.3E+01 tcw13 1.3E-02 2.7E-04 2.8E+00 3.8E+00 ptn21 9.2E-03 3.9E-04 6.7E-01 1.0E+00 ptn22 1.0E-02 2.0E-04 4.6E-01 1.4E+00 ptn23 2.1E-03 1.8E-04 5.7E-01 1.8E+00 ptn24 1.0E-02 4.3E-04 4.6E-01 3.4E-01 ptn25 5.5E-03 1.6E-04 5.2E-01 1.1E+00 ptn26 3.1E-03 1.4E-04 9.7E-01 3.6E+00 tsw31 5.0E-03 1.6E-04 2.2E+00 3.9E+00 tsw32 8.3E-03 2.0E-04 1.1E+00 3.2E+00 tsw33 5.8E-03 2.3E-04 8.1E-01 4.4E+00 tsw34 8.5E-03 9.7E-05 4.3E+00 1.4E+01 tsw35 9.6E-03 1.5E-04 3.2E+00 9.7E+00 tsw36 1.3E-02 1.6E-04 4.0E+00 1.2E+01 tsw37 1.3E-02 1.6E-04 4.0E+00 1.2E+01 tsw38 1.1E-02 1.3E-04 4.4E+00 1.3E+01 tsw39 4.3E-03 2.2E-04 9.6E-01 3.0E+00 ch1VI 6.1E-04 3.0E-04 1.0E-01 3.0E-01 ch2VI 7.7E-04 2.7E-04 1.4E-01 4.3E-01 ch3VI 7.7E-04 2.7E-04 1.4E-01 4.3E-01 ch4VI 7.7E-04 2.7E-04 1.4E-01 4.3E-01 ch5VI 7.7E-04 2.7E-04 1.4E-01 4.3E-01 ch6VI 7.7E-04 2.7E-04 1.4E-01 4.3E-01 ch1Ze 1.6E-04 2.0E-04 4.0E-02 1.1E-01 ch2Ze 3.7E-04 1.3E-04 1.4E-01 4.3E-01 ch3Ze 3.7E-04 1.3E-04 1.4E-01 4.3E-01 ch4Ze 3.7E-04 1.3E-04 1.4E-01 4.3E-01 ch5Ze 3.7E-04 1.3E-04 1.4E-01 4.3E-01 ch6Ze 1.6E-04 2.0E-04 4.0E-02 1.1E-01 pp4 3.7E-04 1.3E-04 1.4E-01 4.3E-01 pp3 9.7E-03 2.4E-04 2.0E-01 6.1E-01 pp2 9.7E-03 2.4E-04 2.0E-01 6.1E-01 pp1 3.7E-04 1.3E-04 1.4E-01 4.3E-01 bf3 9.7E-04 2.4E-04 2.0E-01 6.1E-01 bf2 3.7E-04 1.3E-04 1.4E-01 4.3E-01 tr3 9.7E-04 2.4E-04 2.0E-01 6.1E-01 tr2 3.7E-04 1.3E-04 1.4E-01 4.3E-01 tcwf1 2.9E-02 5.5E-04 1.9E+00 1.3E+01 ptnf1 1.1E-02 4.1E-04 5.4E-01 1.3E+00 tswf1 2.5E-02 4.6E-04 1.7E+00 8.7E+00 chnf1 1.0E-03 3.3E-04 1.3E-01 4.6E-01 Source: DTN: LB0205REVUZPRP.001 [159525] and LB0207REVUZPRP.001 [159526] NOTES: 1Values for fault fracture properties within the Tiva Canyon welded (tcwf), Paintbrush nonwelded (ptnf), Topopah Spring welded (tswf), and Calico Hills nonwelded (chnf) units. VI = Vitric Subunit, Ze = Zeolitic Subunit ANL-NBS-HS-000015 REV01 22 April 2003 4.1.1 Other Inputs The inputs in Table 5 are associated with scientific analyses and interpretations of assumptions listed and discussed in Section 5. The first five rows of inputs are collectively used to assign hydrogeologic nomenclature to layers in the numerical grids. The last two rows of inputs in Table 5 are used to represent the location of the proposed repository and to interpret hydrologic features away from the repository area. Table 5. Summary of Other Inputs Used in Numerical Grid Development Description Reference Use Hydrogeologic Unit Definitions Flint 1998 [100033]1 Assumption 3 Rock Properties Model (RPM3.1) of Integrated Site Model (ISM3.1) MO9910MWDISMRP.002 [145731] Assumptions 4, 5 Mineralogic Model (MM3.1) of Integrated Site Model (ISM3.1) MO9910MWDISMMM.003 [119199] Assumption 5 Rock Properties Model (RPM2000) SN0112T0501399.004 [159524] Assumptions 4, 5 Rock Property Data (saturation, porosity, hydraulic conductivity) Boreholes SD-6, SD-7, SD-9, SD-12, UZ-14, UZ-16, NRG-7a, and WT-24: LB0207REVUZPRP.002 [159672] All other boreholes: MO0109HYMXPROP.001 [155989] SD-6: GS980808312242.014 [106748], GS980908312242.038 [107154] SD-7: GS951108312231.009 [108984] SD-12: GS960808312231.004 [108985] Assumptions 4, 5 Repository Layout Configuration2 BSC 2002 [159527] Assumption 8 Perched-Water Elevations GS010608312332.001 [155307] MO0106RIB00038.001 [155631] Assumption 2 NOTE: 1 Hydrogeologic unit definitions (Flint 1998 [100033]) used qualitatively; individual sample data not used. 2 The latest version of the repository layout (BSC 2003 [161726]; BSC 2003 [161727]) does not include the lower block area. Geologic data alone cannot adequately capture all important features that affect flow and transport in the UZ at Yucca Mountain. Hydrogeologic rock-property data have also been considered, as discussed in Section 5.2 (Assumption 3). Based on analyses of several thousand rock samples performed by the U.S. Geological Survey (USGS), 30 hydrogeologic units (HGUs) have been identified, based on “limited ranges where a discrete volume of rock contains similar hydrogeologic properties” (Flint 1998 [100033], p. 1, Table 1). Since the hydrogeologic property sets to be calculated with the UZ Model grid use, to a large extent, the matrix-property data collected and analyzed by Flint (1998 [100033]), layering within the numerical grid was chosen to correspond as closely as possible to HGUs to facilitate data usage. The boundaries of HGUs are not defined by regularly spaced data, but are more qualitative in nature. The qualitative descriptions (but not any sample or other data) given in Flint (1998 [100033], pp. 21– 32), when correlated with GFM2000 data (DTN: MO0012MWDGFM02.002 [153777]), are used to develop a set of hydrogeologic layers whose thickness and elevation are described by regularly spaced data for the UZ Model. Because of the importance of mineral (especially zeolitic) alteration for flow and transport calculations, boundaries between vitric and zeolitic areas are defined within certain UZ Model grid layers below the proposed repository horizon. Alteration to zeolites has been shown to greatly reduce permeability (Flint 1998 [100033], p. 32; Loeven 1993 [101258], pp. 18–19, 22) and may increase the rock’s ability to adsorb some radionuclides. As discussed in Section 5.2 (Assumptions 4 and 5), the data considered in the numerical grid development for defining low- permeability, zeolitic volumes of rock are obtained from the Rock Properties Model of the Integrated Site Model, Version 3.1 (RPM3.1) (DTN: MO9910MWDISMRP.002 [145731]), along with supporting information from the Rock Properties Model (RPM2000) (DTN: SN0112T0501399.004 [159524]) and the Mineralogic Model of the Integrated Site Model, Version 3.1 (MM3.1) (DTN: MO9910MWDISMMM.003 [119199]). The specific Integrated Site Model (ISM3.1) files used in UZ Model grid development are listed in Attachment I. As discussed in Section 5.2 (Assumption 8), an assumed proposed repository layout configuration, based on Data Sheets 2 and 3 from Repository Design, Repository/PA IED Subsurface Facilities Plan Sht. 1 of 5, Sht. 2 of 5, Sht. 3 of 5, Sht. 4 of 5, and Sht. 5 of 5 (BSC 2002 [159527]), is used during numerical grid generation to locate areas of fine spatial resolution. The repository layout used in the formulation of the numerical grids consists of an extended upper repository area (consisting of two parts) that covers much of the footprint of the previous repository as presented in REV 00 ICN 01 (BSC 2001 [159356], Figure 1), and an additional lower repository area that is situated just east of the upper repository area. The areal boundary coordinates for, and elevations of the proposed repository (in meters above sea level, [masl]) are summarized in Table 6 and the proposed repository outline is shown in Figure 1a (Section 6.2). As noted in Section 5.2 (Assumption 8), the proposed repository layout may be subject to future design modifications. The most recent version of the repository layout (BSC 2003 [161726]; BSC 2003 [161727]), created after the formulation of the numerical grids described in this report, does not include the lower block area designated in Table 6. Table 6. Proposed Repository Boundary Coordinates Used for UZ Model Grids NSP Easting (m) NSP Northing (m) Elevation (masl) Upper Block, Part A 170861.2 236062.5 1037.7 170656.9 235911.0 1038.8 170479.8 235768.2 1040.0 170140.8 235061.9 1048.2 170072.5 234017.7 1062.2 170116.0 233520.8 1069.3 170126.2 233438.9 1070.5 Table 6. Proposed Repository Boundary Coordinates Used for UZ Model Grids (cont.) NSP Easting (m) NSP Northing (m) Elevation (masl) 170158.7 233364.4 1071.6 170217.3 233298.2 1072.8 171227.6 233626.5 1072.8 171259.4 234233.0 1064.6 171742.0 234475.0 1063.4 171858.7 235023.9 1056.4 172309.4 235936.8 1045.9 172230.8 236081.7 1043.5 172140.2 236137.4 1038.8 171577.8 236039.8 1041.2 171619.7 236138.6 1040.0 171665.6 236323.9 1037.7 Upper Block, Part B 171214.0 233366.5 1076.3 170482.2 233128.8 1076.3 170482.2 233128.8 1076.3 170468.6 232357.8 1086.8 170490.0 231513.1 1098.6 170487.4 231341.9 1100.9 170496.4 231089.3 1104.4 170523.6 231013.0 1105.6 170572.7 230943.8 1106.8 171018.4 231088.6 1106.8 171062.8 231188.2 1105.6 171098.4 231284.9 1104.4 Lower Block* 172628.5 235345.4 1001.3 172613.7 235425.7 1001.9 172084.8 235253.9 1001.9 172012.0 235145.1 1001.3 171890.6 234850.1 999.3 171870.2 234588.0 997.4 171836.4 234491.8 996.7 171339.5 234245.2 996. 1 171225.9 232079.1 979.8 171431.4 232060.7 979.2 Table 6. Proposed Repository Boundary Coordinates Used for UZ Model Grids (cont.) NSP Easting (m) NSP Northing (m) Elevation (masl) 171816.0 232100.5 978.5 171907.2 232045.0 977.9 172369.7 232195.2 977.9 172437.3 233239.2 985.7 172655.8 233906.4 990.2 172414.2 234594.4 996.1 172435.7 234771.7 997.4 172444.1 235030.0 999.3 172542.7 235232.3 1000.6 Source: Repository layout obtained from BSC (2002 [159527]). The coordinates in this table were chosen to represent the basic outline of the proposed repository footprint (as depicted in Figures 1a and 1b). The complete IED coordinate set was utilized to construct the numerical grids. NOTE: * The latest version of the repository layout (BSC 2003 [161726]; BSC 2003 [161727]) does not include the lower block area which is being used in AMRs supporting License Application (LA) documents. 4.2 CRITERIA Technical requirements to be satisfied by performance assessment (PA) are based on 10 CFR 63.114 [156605] (Requirements for Performance Assessment) and identified in the Yucca Mountain Project Requirements Document (Curry and Loros 2002 [157916]). The acceptance criteria that will be used by the Nuclear Regulatory Commission (NRC) to determine whether the technical requirements have been met are identified in the Yucca Mountain Review Plan, Information Only (YMRP; NRC 2003 [162418]). For this Scientific Analysis report, the pertinent requirement is PRD-002/T-015 from Curry and Loros (2002 [157916]), which is linked to 10 CFR 63.114(a-c) [156605]. The acceptance criterion for Flow Paths in the Unsaturated Zone, identified in Section 2.2.1.3.6.3 of the YMRP (NRC 2003 [162418]) is Criterion 1, System Description and Model Integration are Adequate. As applied to this Scientific Analysis report, the criterion is: The aspects of geology and hydrology that may affect flow paths in the unsaturated zone are adequately considered. Conditions and assumptions in the abstraction of flow paths in the unsaturated zone are readily identified and consistent with the body of data presented in the description. 4.3 CODES AND STANDARDS No specific formally established codes or standards have been identified as applying to this Scientific Analysis activity. INTENTIONALLY LEFT BLANK 5. ASSUMPTIONS The assumptions presented below are necessary to develop the UZ Model numerical grids. This section presents the rationale and supporting data for the assumptions, and references the section of this report in which each assumption is used. The assumptions presented in this section are similar to those presented in REV00 ICN01 (BSC 2001 [159356], Section 5), and are based on interpretation and synthesis of a variety of geologic and hydrologic inputs. The enhancement in justification of the original assumptions (primarily associated with the inputs in Table 5 presented in Section 4.1.1) is discussed in this section to substantiate interpretations used for the basis of development of numerical grids representing the Yucca Mountain site. The basis of grid development is presented in this section and serves as an overview of detailed scientific analyses presented in Section 6. Assumptions used in developing the numerical grids are of two kinds: assumptions made about the physical world, and assumptions made about the effects of certain features of the grid upon the results of model calculations. None of the assumptions listed below requires confirmation. No hydrologic and rock property values are assigned, justified, or qualified for gridblocks in this Scientific Analysis report. Certain features of the grid are simplifications known to be different from the physical prototype. These simplifications are necessary for calculations to be done with existing computers and qualified software. Assumptions about the effects of such simplifications upon the results of calculations can be verified through sensitivity analyses; that is, by running simulations with the assumptions as stated and with alternative assumptions. The effects of numerical grid resolution on flow and transport model simulation results are discussed through the utilization of previous studies. 5.1 ASSUMPTIONS REGARDING PHYSICAL CONDITIONS EXTERIOR TO THE MODELING PROCESS The following two assumptions pertain to the elevation of the water table, which defines the lower UZ Model boundary. 1. The lower boundary for the UZ Model was established using the regional water table as represented by the potentiometric surface presented in DTN: MO0212GWLSSPAX.000 [161271]. This surface is consistent with borehole water level measurements (DTNs: MO0106RIB0038.001 [155631] and GS010608312332.001 [155307]), but does not represent a unique interpretation of the data (see Sections 6.4.2 and 7.1.1 and Attachment IV for discussion). The water table rises to the west and north, with a base elevation of approximately 730 m above sea level (masl) in the vicinity of the North and South Ramps. The potentiometric surface elevations are based upon reported water table elevations for boreholes in the Yucca Mountain area (see Table 3). The contoured water table elevations, derived from EARTHVISION V5.1 (Dynamic Graphics 1998 [152614]) gridding of the borehole water table elevation and digitized potentiometric map contours taken from DTN: MO0212GWLSSPAX.000 [161271], are presented in Section 6.2 (Figure 1b) and discussed in Sections 6.4.2 and 7.1.1. 2. It is assumed that the observed water levels in boreholes WT#6 and G-2 (at 1,034 and 1,020 masl, respectively) can be interpreted to be perched water (Section 6.2). Observed water levels in these two boreholes from northern Yucca Mountain (located east of the Solitario Canyon fault) are much higher than 840 masl, the elevation of the water level encountered in the nearby USW WT-24 borehole, which is interpreted to represent the regional water table. In boreholes WT#6 and USW G-2, water levels measure about 1,034 masl and 1,020 masl, respectively (Table 3). The UZ Model simulates and calibrates to perched-water data under selected portions of northern Yucca Mountain. These two assumptions are supported by a variety of studies on the water table at Yucca Mountain (e.g., BSC 2001 [155950], Figure 12.3.1.2–2; Ervin et al. 1994 [100633], p. 15; Czarnecki et al. 1994 [142594]; Czarnecki et al. 1995 [103371]), as discussed in more detail in Section 6.2. 5.2 ASSUMPTIONS REGARDING NUMERICAL GRID CONSTRUCTION The geologic data provided in Geologic Framework Model (GFM2000) (DTN: MO0012MWDGFM02.002 [153777]) cannot, by themselves, adequately capture all important features that affect flow and transport in the UZ at Yucca Mountain. Hydrogeologic rock- property data must also be considered. 3. It is assumed that the 30 hydrogeologic units (HGUs) identified by the USGS (Flint 1998 [100033], p. 1, Table 1) based on similarities in rock hydrogeologic properties are adequate to define the layering scheme used for the UZ Model grids (Section 6.3). Since the hydrogeologic property sets to be utilized in UZ flow and transport modeling use, to a large extent, the matrix properties data collected and analyzed by Flint (1998 [100033]), layering within the numerical grid was chosen to correspond as closely as possible to HGUs to facilitate data usage. The boundaries of HGUs are defined by irregularly spaced data and thus additional borehole data could lead to future adjustments to HGU contact locations. The qualitative descriptions given in Flint (1998 [100033], pp. 21–32), when correlated with GFM2000 data (DTN: MO0012MWDGFM02.002 [153777]), are used to develop a set of hydrogeologic layers (whose thickness and elevation are described by regularly spaced data) for the UZ Model grids. The detailed analysis of hydrogeologic properties and definition of HGUs by Flint (1998 [100033]) provides justification for the use of these units in development of the UZ Model grids. The distribution of low-permeability zeolites within the Topopah Spring welded (specifically, tsw39) and Calico Hills nonwelded (CHn) HGUs impacts flowpaths and groundwater travel times from the proposed repository horizon to the water table and is, therefore, an important feature to capture in the UZ Model grids. The data considered in numerical grid development for defining low-permeability, zeolitic volumes of rock come from the Rock Properties Model of the Integrated Site Model, Version 3.1 (RPM3.1) (DTN: MO9910MWDISMRP.002 [145731]), from measurements of borehole rock matrix hydrologic properties (DTNs: LB0207REVUZPRP.002 [159672], MO0109HYMXPROP.001 [155989], GS980808312242.014 [106748], GS980908312242.038 [107154], GS951108312231.009 [108984], GS960808312231.004 [108985]), and from corroborative evidence using data from Rock Properties Model (RPM2000) (DTN: SN0112T0501399.004 [159524]) and the Mineralogic Model of the Integrated Site Model, Version 3.1 (MM3.1) (DTN: MO9910MWDISMMM.003 [119199] [see Assumptions 4 and 5]). The following three assumptions pertain to the definition of low-permeability, zeolitic regions within UZ Model layers corresponding to portions of the TSw and CHn. Within UZ Model layers tsw39, ch1, ch2, ch3, ch4, ch5, and ch6, the tuff has been altered from vitric to zeolitic in some areas and remains unaltered in other areas. For the purposes of flow and transport modeling, the principal differences between these two types of tuff are the adsorptive properties and the saturated hydraulic conductivity. Each gridblock within these UZ Model layers is assigned to either the vitric or zeolitic material. A combination of geologic data is used to define vitric-zeolitic boundaries, including saturated hydraulic conductivity values, matrix saturation measurements, the difference between oven-dried and relative-humidity porosities, and the relative structural position of these layers within the UZ Model area. The assumptions associated with these data are described below. 4. It is assumed that saturated hydraulic conductivity (Ksat) data from the RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) can be used as a surrogate for assigning gridblocks either vitric or zeolitic material names (and thus, separate hydrogeologic properties) within certain layers of the Topopah Spring welded (TSw) and Calico Hills nonwelded (CHn) hydrogeologic units. Vitric rock properties are assigned for areas within UZ Model layers tsw39, ch1, ch2, ch3, ch4, ch5, and ch6 where Ksat is greater than 10-10 m/s, whereas zeolitic properties are used where Ksat is less than 10-10 m/s (Section 6.6.3). There are two main reasons why Ksat data are used as a surrogate to assign gridblocks either vitric or zeolitic material names. First, existing data show that the Ksat of vitric tuff is orders of magnitude greater than that of zeolitic tuff (Flint 1998 [100033], Table 7). Also, there are much more available data on Ksat values than on mineralogic alteration (e.g., percentage of zeolite). Results from analyses by Flint (1998 [100033], Table 7) indicate that vitric Ksat values are on the order of 10-7 m/s, while zeolitic Ksat values are on the order of 10-10 to 10-11 m/s. No definitive Ksat cutoff value exists by which to distinguish vitric from zeolitic material, because this transition occurs over about three orders of magnitude. The Ksat-value cutoff of 10-10 m/s is somewhat arbitrarily chosen; however, the sensitivity of the 10-10 m/s cutoff is not expected to be significant compared to using a 10-9 m/s or 10-8 m/s cutoff, since these contours are closely spaced in the proposed repository footprint within the RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) (see Figure 3). Based on these observations, no additional confirmation of this assumption is required. 5. It is assumed that, in UZ Model layers tsw39, ch1, ch2, ch3, ch4, ch5, and ch6, tuff is vitric where matrix saturations are relatively low (< ~90%) and the difference between oven-dried (105°C) and relative-humidity porosities are less than 5% (Section 6.6.3). Results from analyses by Flint (1998 [100033], p. 29) indicate that altered (i.e., zeolitic) nonwelded tuffs have oven-dried porosities that are typically more than 5% higher than relative- humidity porosities. The loss of water from hydrous secondary minerals (such as zeolites and clays) from oven-dried altered tuffs results in higher estimates of the matrix porosity (relative to those obtained using the relative-humidity method) for these samples. Boreholes where oven- dried porosities exceed relative-humidity porosities by more than 5% for each of the UZ Model layers in question (tsw39, ch1-ch6) generally coincide with zeolite-rich zones, as predicted by MM3.1 (DTN: MO9910MWDISMMM.003 [119199]). Based on these observations, no additional confirmation of this assumption is required. 6. It is assumed that major faults with significant vertical displacement may serve as lateral boundaries for vitric (unaltered) areas within UZ Model layers tsw39, ch1, ch2, ch3, ch4, ch5, and ch6 (Section 6.6.3). Vitric portions of the CHn and Tsw may be laterally continuous within fault blocks that have a higher structural position above the water table compared to adjacent structural blocks. For example, the Solitario Canyon fault offsets the CHn by more than 300 m in the southern part of the UZ Model domain. CHn layers west of the Solitario Canyon fault lie near or below the water table in this area, and thus these tuffs likely have abundant zeolitic alteration. The correlative CHn layers on the east side of the fault may be over 300 m above the water table and are much less likely to have undergone zeolitization, owing to limited water-rock interaction. Because major faults (i.e., Solitario Canyon and Dune Wash faults) determine the proximity of the CHn layers to the water table, they are used as boundaries between vitric and zeolitic areas, where appropriate. The observed structural offsets provide sufficient justification for this assumption. The next assumption pertains to the representation of faults within the UZ Model grids. 7. It is assumed that the simplification of (a) representing steeply dipping faults as vertical in the UZ Model grids and (b) representing related, near-parallel faults as a single feature that incorporates the cumulative offset (e.g., the Solitario Canyon and Solitario Canyon (west) faults) will not significantly affect model calculations (Sections 6.3, 6.6.1, Attachment III). The use of a single fault to represent the offset observed for the Solitario Canyon and the Solitario Canyon (west) faults is in part required by the use of wide vertical columns to model dipping faults. If the projection of near-parallel dipping faults overlap over the depth interval of the UZ Model, then separate faults are difficult to portray in the UZ Model grids without the use of very fine gridding. By accomodating the cumulative offset along a single structural feature, the overall structural and stratigraphic integrity of the UZ geology (as represented by GFM2000 in DTN: MO0012MWDGFM02.002 [153777]) is preserved, albeit in a somewhat simplified manner. The representation of structural offset by the UZ Model grids is evaluated in Attachment III, and the results of the grid verification studies indicate that this assumption is justified. The configuration of the proposed repository layout constitutes the final assumption. 8. It is assumed that the proposed repository layout configuration presented on Data Sheets 2 of 5 and 3 of 5 from Repository Design, Repository/PA IED Subsurface Facilities Plan Sht. 1 of 5, Sht. 2 of 5, Sht. 3 of 5, Sht. 4 of 5, and Sht. 5 of 5 (BSC 2002 [159527]) is appropriate to define those areas within the numerical grid that require enhanced numerical resolution (Section 6.6.2). The proposed repository design used was the most recent representation of the repository layout at the time the numerical grids presented in this document were generated and was considered to be the best source for this information. This design consists of an upper (primary) block located west and north of the ESF, and a lower elevation region located east of the primary repository block and areally overlapping part of the ESF. It is recognized that the proposed repository design is still undergoing change, and that future adjustments to the grid resolution may be necessary, depending on final design decisions. As noted in Section 4.1.1, a revised version of the proposed repository layout was created after the formulation of the numerical grids described in this report. The new layout does not include the lower block area delineated in Table 6 and Figures 1a and b. As discussed in Section 3.3.4.8.1 of FY 01 Supplemental Science and Performance Analyses, Volume 1: Scientific Bases and Analyses (BSC 2001 [155950]), the use of more refined gridding in the area of the proposed repository layout (see Section 6.6.2, Figures III-3 and III-4) provides needed resolution for flow models. Based on these observations, no additional confirmation of this assumption is required. INTENTIONALLY LEFT BLANK 6. SCIENTIFIC ANALYSIS DISCUSSION 6.1 NUMERICAL GRID DEVELOPMENT—OVERVIEW & APPROACH Numerical grids of the UZ beneath Yucca Mountain are used to develop calibrated hydrogeologic property sets and flow fields, to test conceptual hypotheses of flow and transport, and to predict flow and transport behavior under a variety of climatic and thermal-loading conditions. This report describes the development of three different sets of grids. The purpose and general characteristics of each grid set are summarized in Table 7. A description of the steps involved in the generation of these grids is provided in the following sections and in scientific notebooks. Key scientific notebooks used for numerical grid generation activities described in this Scientific Analysis report, along with relevant page numbers and accession numbers, are listed in Table 8. Table 7. Summary of Grids Developed for FY02 UZ Modeling Activities Output DTN (filename) Purpose Grid Description LB02081DKMGRID.001 (Boreholes.mesh)1 (Mesh_1d.dkm)2 (Boreholes_NF.mesh)3 (Boreholes.mck) 1-D hydrogeologic property set inversions and calibrations Consists of 1-D columns centered at borehole locations. Uses borehole contact elevation picks based on the GFM2000 file "contacts00el.dat" (DTN: MO0012MWDGFM02.002 [153777]) and HGU boundaries defined by Flint (1998 [100033]) and Assumption 3. Hydrogeologic data and fault locations used to define the vitric-zeolitic boundary (Assumptions 4-6). Borehole locations used in the 1-D meshes include: b#1, G-1, G-2, G-4, H-1, H-3, H-4, H-5, H-6, SD-6, SD-7, SD-9, SD-12, NRG#4, NRG#5, NRG-6, NRG-7a, N-11, N-15/16, N-17, N-27, N-31/32, N-33, N-36, N-37, N-38, N-53-54, N-55, N-57/58, N-59/61, N-62, N#63, N-64, UZ#4/5, UZ-6, UZ-7a, UZ-1/14, UZ#16, WT-1, WT-2, WT#4, WT#6, WT-7, WT#18 and WT-24. Uses fracture hydrogeologic data in Table 4 to generate the dual-permeability meshes. See Attachment II for additional details. (EWUZ7a.mesh)1 (Mesh_2d.dkm)2 (EWUZ7a_NF.mesh) 3 (EWUZ7a.mck) 2-D fault hydrogeologic property calibration East-west, cross-sectional grid through borehole UZ-7a. Grid columns are generated using GFM2000 isochore and elevation data provided on a regular grid spacing of 61 × 61 m (DTN: MO0012MWDGFM02.002 [153777]). Uses fracture hydrogeologic data in Table 4 to generate the dual-permeability meshes. See Section 6.5 and Attachment III for additional details. LB03023DKMGRID.001 (Grid_LA_3D.mesh)1 (Mesh_3dn.dkm)2 (Grid_LA_3D_NF.mesh) 3 (Grid2002_3D.mck) 3-D UZ Site Scale Modeling Three-dimensional site-scale model with enhanced discretization along major faults and proposed repository drifts. The 3-D grids are generated using GFM2000 isochore and elevation data provided on a regular grid spacing of 61 × 61 m. Uses fracture hydrogeologic data in Table 4 to generate the dual-permeability meshes. See Sections 6.6 and 6.7 and Attachment III for additional details. NOTES: 1 The primary mesh represents matrix blocks only; also referred to as an effective-continuum model (ECM) grid. 2 Dual-permeability model (DKM) mesh generated with fracture properties from Table 4 and a 1-D fracture continuum (Type #1 fractures: See Section 6.7 for details). 3 The “*_NF.mesh” files were used to generate the DKM mesh files, and are not considered output files. ANL-NBS-HS-000015 REV01 35 April 2003 Table 8. YMP Scientific Notebooks Used for FY02 Numerical Grid Development and Grid Verification Analyses LBNL Scientific Notebook ID M&O Scientific Notebook ID Relevant Pages Citation YMP-LBNL-YSW-JH-2 SN-LBNL-SCI-143-V1 137–140 Hinds 2001 [155955] YMP-LBNL-YSW-WZ-1 SN-LBNL-SCI-115-V1 52–56, 66–72 Zhang 2000 [159531] YMP-LBNL-YSW-JH-3 SN-LBNL-SCI-213-V1 7–34, 63–99 Wang 2002 [159673] 67, 71, 100-134 Wang 2003 [162380] YMP-LBNL-GSB-LP-2 SN-LBNL-SCI-103-V1 111–115, 122, 134–141, 145–151 Wang 2002 [159673] 139 Wang 2003 [162380] YMP-LBNL-YSW-3 SN-LBNL-SCI-199-V1 82–92 Wang 2002 [159673] 86, 88-89, 237–238 Wang 2003 [162380] YMP-LBNL-GSB-LP-2.1 SN-LBNL-SCI-103-V2 17–28 Wang 2003 [162380] Data extracted from Geologic Framework Model (GFM2000) (DTN: MO0012MWDGFM02.002 [153777]) and Rock Properties Model of the Integrated Site Model, Version 3.1 (RPM3.1) (DTN: MO9910MWDISMRP.002 [145731]) form the basis for numerical grid development. With these data, an initial 2-D (plan-view) grid is developed (see Section 6.5) defining borehole, fault, and repository column locations, where appropriate. Using the 2-D grid as the basis for column locations, a 3-D effective-continuum model (ECM) grid is constructed (see Section 6.6) using layer reference and bounding horizons, along with thickness data from GFM2000 (DTN: MO0012MWDGFM02.002 [153777]). Initial grid generation is followed by an iterative process of grid evaluation and modification to achieve appropriate spatial resolution and representation of important features, such as the proposed repository, faults, and calibration boreholes, and to ensure proper connections between the various elements of the grid. Revisions are made accordingly until these criteria are met. Next, the 3-D ECM grid is modified to allow for modeling dual-continuum processes (matrix and fracture flow) using a dual-permeability (dual-k) mesh maker, 2kgrid8.for V1.0 (LBNL 2002 [154787]). The 2kgrid8.for V1.0 software program incorporates information (i.e., fracture porosity, spacing, aperture, and fracture-matrix interaction area) from fracture data analyses (see Table 4) into the grids. The computer code WINGRIDDER V2.0 (LBNL 2002 [154785]) is used to generate 1-, 2- and 3-D integral finite difference (IFD) grids for the UZ Model domain. The type of grid generated by WINGRIDDER V2.0 is consistent with the computational requirements for V1.4 and later versions of the TOUGH2 numerical code simulator (Pruess 1991 [100413], pp. 27–30, 41–42). TOUGH2 and the inverse modeling code iTOUGH2 (Finsterle 1999 [104367]) use cells, or gridblocks, and connections between those gridblocks to represent the flow system without requiring the global location of each gridblock or connection. This approach provides great flexibility in describing complex flow geometry and relationships between individual objects within the system. Unlike other gridding software, WINGRIDDER V2.0 has the capability of designing complex, irregular grids with large numbers of cells and connections, and it can incorporate nonvertical faults and other embedded refinements, such as waste emplacement drifts within the proposed repository area at Yucca Mountain. WINGRIDDER V2.0 can generate a grid that includes a repository with multiple subregions and drifts. A bilinear interpolation between points of known elevation of a regular grid is used in WINGRIDDER V2.0 to determine the thickness or elevation at intermediate points, thus helping to conserve layer discontinuity resulting from faulting. The grids produced by this work are integral finite difference (IFD) grids. Alternative gridding methods include finite difference and finite element methods, but IFD was chosen for compatibility with the TOUGH2 family of codes employed by downstream users. Compatibility with TOUGH2 for flow and transport simulations is a requirement of this work. Described in this report are the methods used to develop numerical grids for hydrogeologic- property-set inversions, for model calibration, and for calculation of 3-D UZ flow fields for PA. The stages of grid development include the following: 1. Establish domain boundaries and location of important calibration features such as boreholes (Section 6.2). 2. Determine UZ Model layers and fault geometries based on GFM2000 (DTN: MO0012MWDGFM02.002 [153777]), RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]), and correlation with Flint’s (1998 [100033]) HGUs (Section 6.3). 3. Extract and format GFM2000 (DTN: MO0012MWDGFM02.002 [153777]) and RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) data for incorporation into 3-D grids (Section 6.4). 4. Generate a 2-D grid, incorporating information from Steps 1 and 2, and refine as needed to capture spatial variability (Section 6.5). 5. Generate a 3-D (ECM) grid, based on the column locations established in the 2-D grid and data from Step 3 (Section 6.6). 6. Combine the results of fracture analyses with the ECM grid from Step 5 to generate a dual-permeability mesh (Section 6.7). The process of verifying that appropriate gridblock material names, gridblock volumes and locations, connection lengths and directions, and interface areas between gridblocks are used in the UZ Model numerical grids is documented in Section 6.8 and in Attachments II and III. Section 6.8 also summarizes results from corroborative studies that support the use of fairly coarse numerical grids to model flow and transport processes. 6.1.1 Summary of Changes to the UZ Model Grids Some of the input data, software, and assumptions used in this revision differed from those used in REV00 ICN01 (BSC 2001 [159356]), thus resulting in changes to the UZ Model grids generated. These include the following items: • Use of GFM2000 (DTN: MO0012MWDGFM02.002 [153777]) instead of GFM3.1 (DTN: MO9901MWDGFM31.000 [103769]) for construction of UZ Model grids • Change in UZ Model areal boundaries • Change in position of water table (forms base of UZ) • Change in proposed repository layout • Enhanced resolution in vertical gridding • Change in software • Test simulation to verify connections in 3-D grid • Addition of vitric and zeolitic subunits to tsw39 and ch6, and modification of vitric/zeolitic boundaries for layers ch1-ch5 • Addition of Bow Ridge/Toe fault and simplification of Solitario Canyon fault system • Use of additional boreholes for 1-D properties calibration • ESF and ECRB not incorporated into formulation of 3-D grid • Changes in fracture hydrogeologic properties. The primary change in the development of numerical grids for UZ flow and transport modeling is the change of geologic input data. Revision 00, ICN01 of this report (BSC 2001 [159356]) utilized GFM3.1 (DTN: MO9901MWDGFM31.000 [103769]), while the grids generated in this report are based on GFM2000 (DTN: MO0012MWDGFM02.002 [153777]) (see Section 4.1). As noted in the Geologic Framework Model (GFM2000) (BSC 2002 [159124], Section 6.1), the changes between GFM3.1 and GFM2000 relating to elevation changes in geologic layers are relatively small in magnitude, rarely as large as 7.6 m (25 feet), and are primarily near the margins of the GFM boundaries. Thus, changes in HGU thicknesses and contact elevations should also be relatively minor. The current UZ Model area has also undergone modifications, with an extension made to the east to allow for an expanded proposed repository footprint (Section 6.2). The proposed repository layout used for numerical grid generation consists of two different layers (Section 4.1.1) and includes regions north and east of the repository layout used in prior versions of the UZ Model grids. Note that the eastern extension to the proposed repository (the lower block) has been removed from the most recent revision of the repository layout (BSC 2003 [161726]; BSC 2003 [161727]), and is not used in any LA calculations presented in this report. The maximum thickness of any gridblock was reduced from 60 to 20 m (5 m for ch1, ptn22, ptn24, ptn25, and ptn26, and 2 m for ptn21 and ptn23 HGUs), and the minimum grid thickness was reduced from 1.5 m to 1.0 m. Another modification consists of using the contoured regional water table instead of using two fixed elevations to define the water table, and thus the base of the UZ (Section 5.1, Assumptions 1 through 3 of BSC 2001 [159356]; Section 5.1, (Assumptions 1 and 2) and sections 6.2 and 6.4.2 of this Scientific Analysis report). This modification results in little change to the portion of the model near the ESF and to the SE, but it does impact the region to the north, where elevated water levels in boreholes were previously all interpreted as representing perched water. The current UZ Model grids will have a thinner UZ interval in the northern part of the model area as a result of this change. The current UZ Model was constructed using updated software versions (Sections 3 and 6.1). EARTHVISION V5.1 (Dynamic Graphics 1998 [152614]) was used instead of V4.0 (Dynamic Graphics 1997 [134134]), and the updated version of WINGRIDDER V2.0 was utilized to construct the UZ Model grids. The macro DKMgenerator V1.0 (LBNL 1999 [140702]) was replaced with the qualified software program 2kgrid8.for V1.0. The use of updated software provides some additional functionality, but does not result in significant modifications to the generated grids. Some additional software packages were utilized in the water table data qualification and review process (see Attachment IV). ARCINFO V7.2.1 (CRWMS M&O 2000 [157019]; USGS 2000 [148304]) was utilized to convert potentiometric contours from the ARCINFO file “pot_contours_e00” (DTN: GS010608312332.001 [155307]) into an ASCII format that could be used to qualify data from DTN: MO0212GWLSSPAX.000 [161271]. To facilitate comparison with the USGS contour data, coordinate data from the new DTN (MO0212GWLSSPAX.000 [161271]) were converted from Nevada State Plane coordinates (NAD27 datum) to UTM coordinates to using ARCINFO V7.2.1. This updated report also includes a test simulation using TOUGH2 V1.4 (LBNL 2000 [146496]) to verify the connections of the ECM 3-D grid (Attachment III). Several modifications were made to the delineation of vitric and zeolitic zones within the UZ Model grid (Section 5.2, Assumptions 4 through 6; Section 6.6.3). The primary data source used to delineate the vitric and zeolitic zones, RPM3.0 (DTN: MO9901MWDISMRP.000 [103771]), was superseded by RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]). In addition to the information obtained from RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]), the location of major faults, rock-property data from boreholes, and corroborative information from MM3.1 (DTN: MO9910MWDISMMM.003 [119199]) and RPM2000 (DTN: SN0112T0501399.004 [159524]) were used to modify the vitric/zeolitic boundaries for the layers ch1-ch5. New vitric and zeolitic subunits were created for the layers tsw39 and ch6. The main impact of these changes is the modification of rock properties for layers tsw39 and ch6, and an overall enlargement of the vitric subunit for layers ch2-ch5. Two changes were made to the faults within the UZ Model grids. One new fault (a combination of the Bow Ridge and Toe faults) was added to the model; this change resulted from the eastward extension of the UZ Model boundary. This fault is located away from the current proposed repository, and its addition to the UZ Model grids should not significantly impact subsequent users. The second change relates to the simplification of the Solitario Canyon fault system (Solitario Canyon, Splay “N”, Splay “G”, Splay “S”, and Solitario Canyon [west]) into a single structural feature (Section 5.2, Assumption 7). The splay faults are relatively small features with minor offsets, but the Solitario Canyon (west) fault has significant (up to more than 300 m) vertical offset. However, this feature is very close to the Solitario Canyon fault, and given the coarsely spaced gridding used in the SW portion of the UZ Model area where the Solitario Canyon (west) fault is located, a single feature containing the cumulative offset of the two faults was used for the UZ Model. This simplification does not retain all of the structural details contained within GFM2000 (DTN: MO0012MWDGFM02.002 [153777]), but it preserves the general stratigraphic and structural relations, and is also located away from the current proposed repository area. These changes in modeling the Solitario Canyon fault system should not impact downstream users, provided that the proposed repository area is not subsequently shifted into the areas where the Solitario Canyon (west) and Splay faults are situated. Over 40 boreholes were used to calibrate matrix properties (Section 6.2), more than twice the number used in REV00 ICN01 of this report (BSC 2001 [159356]). Because of the increased number of boreholes used for calibration, the ESF, ECRB, and associated niches and alcoves were not used in the calibration process, and therefore it was not necessary to incorporate these features into the 3-D grid. Updated fracture properties (Section 4.1) were utilized in the construction of the dual-permeability 2-D and 3-D grids. 6.2 BOUNDARIES AND CALIBRATION FEATURES The areal domain of the UZ Model encompasses approximately 40 km2 of the Yucca Mountain area. Yucca Wash lies near the northern model boundary, while the approximate latitude of borehole G-3 defines the southern boundary. The eastern model boundary lies just to the east of the Bow Ridge fault, and the western boundary lies approximately 1 km west of the Solitario Canyon fault. These boundaries encompass many of the existing hydrology wells for which extensive moisture saturation and water potential data are used as calibration points for determining layer properties. One important objective of selecting these boundaries was to minimize potential boundary effects on numerical simulation results within the proposed repository footprint. Figures 1a and 1b show map views of the model domain, including the proposed repository boundary, the paths of the ESF and ECRB, major faults defined in GFM2000 (DTN: MO0012MWDGFM02.002 [153777]), calibration boreholes, and the contoured regional water table. Table 9 lists the NSP coordinates for the domain boundary. Development of Numerical Grids for UZ Flow and Transport Modeling U0000 5ALAH9=ID.=K.J -NE.A0E.. .=K.J 5.. EJ=HE. +=.O.. .=K.J /D.IJ,=. ?A 9A IJ .=K.J /D.IJ,=.?A 9AIJ .=K.J5..EJ=HE. +=.O.. .=K.J 9AIJ ,K.A9=ID.=K.J ,K.A 9AIJ /D.IJ,=.?A.=K.J 1.>HE?=JA.=K.J 6.A .=K.J 5F.=O 5 ,K.A : *.M4E@CA.=K.J 2=C=.O9=ID.=K.J 5....=J .=K.J 5....=J .=K.J,HE.. 0..A9=ID.=K.J 5K.@=.?A .=K.J 5K.@=.?A.=K.J5F.=O C 5F.=O C 5F.= O . 5F.=O . NSP Northing (m) NSP Easting (m) .-/-., 7. ..@A. *.K.@=HO 4AF.IEJ.HO *.K.@=HO 4AF.IEJ.HO ..MAH *..?. -5. =.@ -+4* /.. .=K.JI Source: (Proposed Repository Design) BSC 2002 [159527]; (GFM2000 Faults) DTN: MO0012MWDGFM02.002 [153777] NOTE: 2002 Repository Lower Block will not be used in any LA calculations. Figure 1a. Plan-View Schematic Showing the UZ Model Boundary, the Proposed Repository Outline, Major Faults from GFM2000, the ESF, and the ECRB 860 820 820 780 780 780 760 740 740 730 730 1000 980 940 900 860 760 96.% .$ .#'.$ . .#%.#&. 0.! .$" 7..$ 5,.% 96. 7..%= 0."5,.$ 5,. 0.$ . . % . #. $. / 96. " .!$ . % .!! /. 0. 96 $ 0.# 5,.' 96. .! .! . /." > .4/.$ .4/ # . $! 7. ".#. 96 & 96 " .!& .!% .4/.%= .4/ " .#!.#". .## 7. $ 7. . ". NSP Northing (m) NSP Easting (m) .-/-., 7. *.HAD..AI7. $ 900 ..JA. . ..@A. *.K.@=HO4AF.IEJ.HO *.K.@=HO4AF.IEJ.HO ..MAH *..?. -5. =.@ -+4* 9=JAH 6=>.A -.AL=JE.. +..J.KHI ,A..JAI IE.C.A ..?=JE.. KIA@ B.H =@.=?A.J >.HAD..AI. Source: (Proposed Repository Design) BSC 2002 [159527]; Output-DTNs: LB0208HYDSTRAT.001 (UZ2002 Model Boundary, Modified Borehole Locations); LB02092DGRDVER.001 (Water Table). NOTE: 2002 Repository Lower Block will not be used in any LA calculations. Figure 1b. Plan-View Schematic Showing Boreholes, the Contoured Water Table (Elevations in m), the UZ Model Boundary, the Proposed Repository Outline, the ESF, and the ECRB Table 9. UZ Model Areal Boundary Coordinates NSP Easting NSP Northing (m) (m) 168100 229500 169600 238900 171400 238550 173910 236320 172820 229500 Output-DTN: LB0208HYDSTRAT.001 The upper boundary of the UZ Model is the bedrock surface (topography minus alluvium), which is defined by the GFM2000 file “s00bedrockRWC.2grd” (see Attachment I, GFM2000 files). The lower boundary is the water table, or potentiometric surface, derived from water- level-elevation data (Assumption 1). Borehole water-level elevations beneath northern Yucca Mountain suggest a large hydraulic gradient, as seen in the contoured potentiometric surface (BSC 2001 [155950], Figure 12.3.1.2–2) and the water-level data contained in DTN: MO0106RIB00038.001 [155631], with water levels increasing northward from about 730 masl at the south end of the proposed repository area to 840 m (USW WT-24; see DTN: GS010608312332.001 [155307]) less than a kilometer north of the proposed repository area. Two boreholes north of WT-24, G-2 and WT#6, have significantly higher water levels (>1000 masl). One explanation for the fairly abrupt water-level difference between WT-24 and the G-2 and WT#6 boreholes is the occurrence of perched or semi-perched water under portions of northern Yucca Mountain (USGS 2001 [157611]; Ervin et al. 1994 [100633], p. 15; Czarnecki et al. 1994 [142594]; Czarnecki et al. 1995 [103371]). For the purpose of developing UZ Model grids, water table elevations beneath portions of northern Yucca Mountain are assumed to represent perched water, as stated in Section 5.1, Assumption 2. The contoured regional water table elevations (Figure 1b) are represented by the surface defined in the file “gwl_sspac_60.96.2grd” (see Attachment III, Table III-1). Details on how this surface was generated are presented in Section 6.4.2 and Attachment IV. UZ Model borehole calibration features, represented as column centers in the 1-D inversion and 3-D calibration grids, are listed in Table 10. For simplicity, the borehole names used throughout the remainder of this document drop the USW and UE-25 prefixes. Where boreholes are closer than 80 m to one another, the boreholes (as indicated on Table 10 by an asterisk) are jointly represented by an intermediate location calculated by averaging the coordinates of the two boreholes. Because borehole UZ-7a is located adjacent to a fault, the position of the column center associated with the borehole was shifted slightly to accommodate the fault geometry. Table 10. Borehole Locations Used in the UZ Inversion and Calibration Models NSP Easting(m) NSP Northing(m) Feature 170993 234849 G-1 170842 237386 G-2 171627 233418 G-4 171416 234774 H-1 170216 230594 H-3 171880 232149 H-4 170355 233670 H-5 168882 232654 H-6 172767 233806 NRG#4 172142 234053 NRG#5 171964 233698 NRG-6 171598 234355 NRG-7a 171178 232245 SD-12 171066 231328 SD-7 170264 232386 SD-6 171242 234086 SD-9 170744 235090 UZ-1/14* 172168 231811 UZ#16 172559 234286 UZ#4/5* 170178 231566 UZ-6 171363 231866 UZ-7a1 171398 236739 WT-24 171828 229802 WT-1 171274 231850 WT-2 173138 234243 WT#4 172067 237920 WT#6 168826 230298 WT-7 172168 235052 WT#18 172644 233246 b#1 170390 237919 N11 170563 237171 N15/16* 170687 237203 N17 170344 235175 N27 171534 232951 N31/32* 171051 234717 N33 171780 235885 N36 171820 233934 N37 171707 233924 N38 171983 231704 N53/54* 171983 231801 N55 170946 230186 N57/58* 170960 230230 N59/61* 170171 230772 N62 172568 234342 N#63 170516 233394 N64 Source: Output-DTN: LB0208HYDSTRAT.001 (file “boreholes_Rick_updated.hol”). NOTES: *Single location used for boreholes in close proximity to one another, as explained in text. Original northing and easting values (in feet) converted to meters by multiplying by 0.3048. See discussion of metric conversion in Section 3. 1 Location of UZ-7a shifted to accommodate fault grid geometry. Borehole locations used for 1-D column construction. See file “Boreholes.mck” in Output-DTN: LB02081DKMGRID.001. A subset of the listed boreholes was used for property inversion and calibration. For the earlier versions of this report, many fewer boreholes were used for the calibration, but these were supplemented by the ESF, ECRB, and associated alcoves and niches. Because more boreholes were used for the UZ 2002 grid model calibration, the ESF and ECRB features were not needed for the present model calibrations, and thus these features were not discretized in the UZ 2002 Model grids. The GFM2000 file “contacts00el.dat” (see Attachment I, GFM2000 files) is used to define the location of most of the boreholes that serve as column centers within the various UZ Model grids. Since the coordinates contained within this file are listed in feet, rather than meters (which is the desired unit of measure in the UZ Model), a simple unit conversion was performed (1 ft = 0.3048 m; see metric conversion discussion in Section 3). The locations of the N-series boreholes not listed in this file (N15/16, N17, N27, N36, N57/58, N59/61, N#63, and N64) that were used for model calibration were obtained from the Yucca Mountain Site Characterization Project Site Atlas 1995 (DOE 1995 [102884], vol. 1, p. 9.14). Where boreholes were located within 80 m of one another, the boreholes were listed as a pair, and the average location of the two boreholes was used for property calibration. The spatial relationship between boreholes and faults (determination of fault locations in the 2-D grid is described in Sections 6.3, and 6.5, and 6.6.1) is such that these features may intersect or lie within 30 m of each other (which is typically less than the desired lateral resolution of the grid). As a result, the location of certain features (e.g., column centers) is prioritized. In general, the location of column centers at boreholes was given highest priority, followed by the proposed repository layout, followed by faults. 6.3 UZ MODEL LAYERS AND FAULT GEOMETRIES As discussed previously in Section 4, layering within the UZ Model grid is chosen to correspond as closely as possible to HGUs, to facilitate usage of rock-property data. Table 11 provides a correlation between major HGUs, GFM2000 lithostratigraphic units (BSC 2002 [159124], Table 4), UZ Model layers, and Flint’s (1998 [100033]) HGUs. In many cases, HGUs correlate 1-to-1 with, or are simple combinations of, GFM2000 layers. In a few instances, multiple HGUs can be present within one GFM2000 layer, such as within the Yucca Mountain Tuff (Tpy), the lower nonlithophysal zone of the Topopah Spring Tuff (Tptpln), or the Calico Hills Formation (Tac). Using Table 11 as a basis for UZ Model layering, GFM2000 layer-thickness (isochore) grid files (DTN: MO0012MWDGFM02.002 [153777]) are combined or subdivided, as appropriate (see Section 6.4.1), to correspond to Flint’s (1998 [100033]) HGUs. Table 11. GFM2000 Lithostratigraphy, UZ Model Layer, and Hydrogeologic Unit Correlation Major Unit (Modified from Montazer and Wilson 1984 [100161]) GFM2000 Lithostratigraphic Nomenclature* FY02 UZ Model Layer Hydrogeologic Unit (Flint 1998 [100033], Table 1) Tiva Canyon welded (TCw) Tpcr tcw11 CCR, CUC Tpcp tcw12 CUL, CW TpcLD Tpcpv3 tcw13 CMW Tpcpv2 Paintbrush nonwelded Tpcpv1 ptn21 CNW (PTn) Tpbt4 ptn22 BT4 Tpy (Yucca) ptn23 TPY ptn24 BT3 Tpbt3 Tpp (Pah) ptn25 TPP Tpbt2 ptn26 BT2 Tptrv3 Tptrv2 Topopah Spring welded Tptrv1 tsw31 TC (TSw) Tptrn tsw32 TR Tptrl, Tptf tsw33 TUL Tptpul, RHHtop Tptpmn tsw34 TMN Tptpll tsw35 TLL Tptpln tsw36 TM2 (upper 2/3 of Tptpln) tsw37 TM1 (lower 1/3 of Tptpln) Tptpv3 tsw38 PV3 Tptpv2 tsw39 (vit, zeo) PV2 Table 11. GFM2000 Lithostratigraphy, UZ Model Layer, and Hydrogeologic Unit Correlation (cont.) Major Unit (Modified from Montazer and Wilson 1984 [100161]) GFM2000 Lithostratigraphic Nomenclature* FY02 UZ Model Layer Hydrogeologic Unit (Flint 1998 [100033], Table 1) Calico Hills nonwelded Tptpv1 ch1 (vit, zeo) BT1 or BT1a (altered)(CHn) Tpbt1 Tac (Calico) ch2 (vit, zeo) CHV (vitric) or CHZ (zeolitic) ch3 (vit, zeo) ch4 (vit, zeo) ch5 (vit, zeo) Tacbt (Calicobt) ch6 (vit, zeo) BT Tcpuv (Prowuv) pp4 PP4 (zeolitic) Tcpuc (Prowuc) pp3 PP3 (devitrified) Tcpmd (Prowmd) pp2 PP2 (devitrified) Tcplc (Prowlc) Tcplv (Prowlv) pp1 PP1 (zeolitic) Tcpbt (Prowbt) Tcbuv (Bullfroguv) Crater Flat undifferentiated Tcbuc (Bullfroguc) bf3 BF3 (welded) (CFu) Tcbmd (Bullfrogmd) Tcblc (Bullfroglc) Tcblv (Bullfroglv) bf2 BF2 (nonwelded) Tcbbt (Bullfrogbt) Tctuv (Tramuv) Tctuc (Tramuc) tr3 Not Available Tctmd (Trammd) Tctlc (Tramlc) Tctlv (Tramlv) tr2 Not Available Tctbt (Trambt) and below NOTE: * Buesch et al. (1996 [100106]) define the units in the Paintbrush Group (layers beginning with “Tp”). Moyer et al. (1995 [103777]) describe the Tac and Tacbt. Buesch and Spengler (1999 [107905]) describe the symbols for the Crater Flat Tuffs. GFM2000 nomenclature (BSC 2002 [159124], Table 4) uses the symbols that are included parenthetically below layer Tpbt1. Additional details on how the GFM2000 units were combined or subdivided to obtain the UZ Model units are found in Wang 2002 ([159673], SN- LBNL-SCI-213-V1, pp. 11–15). Faults are important features to include in the UZ Model grids, because they may provide fast pathways for flow or serve as barriers to flow. A fault can be a surface with arbitrary shape in the 3-D UZ Model domain and is represented as a surface (defined by a set of x, y, z data on a regular grid spacing) in GFM2000 (BSC 2002 [159124], Sections 6.4.2 and 6.4.4). In UZ Model grids, fault surfaces are represented by a series of connected columns of gridblocks. Faults can be represented in the grid as either vertical or nonvertical features. Many of the faults at Yucca Mountain are steeply dipping, particularly within the UZ. For UZ flow and transport modeling studies of Yucca Mountain, it is believed that flow through faults is much more sensitive to the rock properties assigned to fault zones than to slight variations in fault dip. Since large numbers of gridblocks are needed to discretize nonvertical fault zones (which adds significantly to the computational time of model calibration and forward simulations), certain criteria have been developed under Assumption 7 (Section 5) to reduce the total number of gridblocks along faults in order to simplify the UZ Model grids. Faults are modeled as vertical if they meet any of the following criteria: (1) their average dip exceeds 85 degrees, (2) their average dip exceeds 80 degrees and they lie sufficiently far (>1 km) from the proposed repository layout area so as not to significantly affect flow and transport calculations, (3) they lie west of the Solitario Canyon fault, (4) they are adjacent to UZ Model boundaries, or (5) they pass through or abut the proposed repository (see Figure 1a). Fine-resolution gridding of the repository is deemed to be more important than incorporating dipping faults, which require larger gridblocks (see Section 6.6.1). Table 12 lists the GFM2000 faults that lie within or along UZ Model boundaries. Table 12. Faults Within the UZ Model Domain Fault Name GFM2000 File Name Solitario Canyon f00sol.2grd Solitario Canyon (west) f00solwest.2grd Unnamed joins Solitario Canyon f00soljfat.2grd & Fatigue Wash faults Splay “G” f00splayg.2grd Splay “N” (north) f00splayn.2grd Splay “S” (south) f00splays.2grd Sundance f00sundance.2grd “Toe” f00toe.2grd Sever Wash f00sever.2grd Pagany Wash f00pagany.2grd Drill Hole Wash f00drill.2grd Ghost Dance f00ghost.2grd Ghost Dance (west) f00ghostw.2grd Dune Wash f00dune.2grd Dune Wash “X” f00dunex.2grd Dune Wash (west 1) f00dunew1.2grd “Imbricate” f00imb.2grd Bow Ridge f00bow.2grd Exile Hill (or Bow Ridge east) f00exile.2grd Source: DTN: MO0012MWDGFM02.002 [153777] The average slope of each fault was evaluated to determine which faults can be reasonably approximated by vertical columns of gridblocks in UZ Model grids. This task involves the calculation of slope along each fault (as it transects the UZ) using the Slope Grid Calculation utility in EARTHVISION V5.1 (Dynamic Graphics 1998 [152614]). Refer to the scientific notebook by Wang (2002 [159673], SN-LBNL-SCI-213-V1, pp. 73–74) for details regarding this calculation. The results are summarized in Table 13. Table 13. Results of GFM2000 Fault-Slope Analysis Fault Name Slope Range (average) Minimum Dip(degrees) Maximum Dip(degrees) Average Dip (degrees) Solitario Canyon 1.0–6.8 (2.4) 44.7 81.7 67.5 Solitario Canyon (west) 5.3–10.5 (6.4) 79.3 84.6 81.1 “Soljfat” 3.2–4.5 (3.8) 72.9 77.4 75.1 Splay “G” 1.6–2.9 (2.2) 58.7 70.8 65.4 Splay “N” 1.3–4.1 (2.0) 53.0 76.4 63.2 Splay “S” 1.3–2.7 (2.0) 52.1 69.7 63.8 Sundance 7.1–12.3 (11.9) 82.0 85.4 85.2 "Toe" 3.6–5.2 (4.2) 74.3 79.1 76.6 Sever Wash 5.6–8.4 (7.0) 79.9 83.2 81.8 Pagany Wash 8.8–13.8 (11.5) 83.5 85.8 85.1 Drill Hole Wash 10.7–14.0 (11.9) 84.7 85.9 85.2 Ghost Dance 8.4–14.5 (11.6) 83.2 86.1 85.1 Ghost Dance (west) 10.0–13.4 (11.7) 84.3 85.7 85.1 Dune Wash 1.4–3.0 (1.9) 55.0 71.3 62.5 Dune Wash “X” 3.7–5.0 (4.5) 75.0 78.6 77.5 Dune Wash (west1) 3.1–4.4 (3.7) 72.2 77.1 74.7 "Imbricate" 9.2–15.8 (12.2) 83.8 86.4 85.3 Bow Ridge 0.4–36.9 (3.8) 23.4 88.4 75.1 Exile Hill 0.02–6.5 (4.7) 1.0 81.3 78.0 Output-DTN: LB0208HYDSTRAT.001 In accordance with Assumption 7, the following faults are represented by vertical columns of gridblocks (i.e., are assumed to be vertical) in the UZ Model grids: “Soljfat,” Sundance, “Toe,” Sever Wash, Pagany Wash, Drill Hole Wash, Ghost Dance, Ghost Dance (west), and “Imbricate” faults. The “Toe” and Bow Ridge faults are represented by a single structural feature, which, due to its proximity to the eastern boundary of the UZ Model area, is considered as a vertical fault. The remaining faults (Solitario Canyon, Dune Wash, Dune Wash “X,” and Dune Wash [west1]) are represented by nonvertical columns of gridblocks in the 3-D grids. Splay faults “N”, “S”, and “G” lie close to the Solitario Canyon fault and intersect it at a relatively shallow depth. This presents complications when generating the 3-D grids because of the preferred numerical grid resolution and fault representation method (described in Section 6.6.1). Thus, these three splay faults are considered part of the Solitario Canyon fault zone and are not explicitly defined. However, after grid generation, fault properties can be assigned to the gridblocks closest to the location of these faults, as needed. As mentioned in Assumption 7, the Solitario Canyon (west) fault was not depicted as a distinct feature in the UZ Model grids. The relatively coarse gridding used in the SW portion of the UZ Model area (resulting from its location away from the proposed repository area) precludes the individual portrayal of these closely spaced west-dipping normal faults. However, the cumulative vertical offset observed in the GFM2000 (DTN: MO0012MWDGFM02.002 [153777]) for the Solitario Canyon and Solitario Canyon (west) faults is captured by the single nonvertical fault (Solitario Canyon) and the adjacent columns used in the UZ Model grids, thus preserving the general stratigraphic and structural relations of GFM2000. Preparation of GFM2000 fault data (DTN: MO0012MWDGFM02.002 [153777]) for incorporation into UZ Model grids first involves a simple unit conversion from feet to meters. The spatial position of the faults is then determined by intersecting each fault surface (*.2grd, listed in Table 12) with one or more horizontal planes, producing data files describing fault-trace locations at prescribed elevations. Faults represented as vertical features in the UZ grids use fault-trace information at an arbitrary elevation of 1,100 masl, chosen because it is just above the proposed repository and near the middle of the UZ. During grid generation, vertical columns of gridblocks are assigned along each fault trace. Faults represented as nonvertical features (i.e., by nonvertical columns of gridblocks) use fault- trace information at three elevations (one near the land surface, one near the water table, and one located approximately midway between the other two) to capture variations in dip. The UZ Model gridding process interpolates the location of each nonvertical fault using data points at the three prescribed elevations. With this approach, the dip of a fault within a given fault column is uniform in the upper interval between the highest and middle elevations, and is again uniform in the lower interval between the middle and lowest elevations. This allows the dip in the upper interval to be different from the dip in the lower interval (which may occur if the fault surface is curved, rather than planar). Furthermore, dip angles within the same vertical interval can be different in different columns (i.e., laterally along a fault). Thus, even a fault with variable dip along its trace can be represented with this method. In some cases, the upper and lower portions at dipping faults have been adjusted to a vertical orientation to ensure appropriate grid resolution and comply with the requirement that gridblock columns adjacent to fault columns be at least as wide as the fault columns (see Dune Wash fault in Figure III-4). For specific details regarding manipulation of fault data, refer to the scientific notebook by Wang (2002 [159673], SN-LBNL- SCI-213-V1, p. 19) and Hinds 2001 ([155955], pp. 137–140). 6.4 EXTRACTION OF GFM2000 AND ISM3.1 DATA 6.4.1 Isochores Geologic layers are correlated with Flint (1998 [100033]) HGUs in Table 11, and UZ Model layers are determined based on this correlation (Assumption 3). Because of its large thickness beneath northern Yucca Mountain, layer Tac is vertically subdivided equally into four layers throughout the UZ Model domain. Based on the relationships provided in Table 11, certain GFM2000 (DTN: MO0012MWDGFM02.002 [153777]) layers (represented by isochore grids) are combined, while others were subdivided, to create hydrogeologic model layers for the UZ grids. GFM2000 isochore grids (DTN: MO0012MWDGFM02.002 [153777]) used in FY02 UZ grid development include those lying between the upper Tpcpv3 contact and the lower Trambt contact. Layers are combined if (1) they have similar hydraulic properties based on analyses by Flint (1998 [100033]), (2) they are very thin across Yucca Mountain, or (3) property data are very limited for the rock units. GFM2000 isochores (DTN: MO0012MWDGFM02.002 [153777]) are subdivided if rock-property data exist that suggest two or more distinct hydrogeologic layers within a geologic unit. For specific details describing the manipulation and formatting of GFM2000 isochore files (DTN: MO0012MWDGFM02.002 [153777]), refer to the scientific notebook by Wang (2002 [159673], SN-LBNL-SCI-213-V1, pp. 11–15). Below is a brief summary of the steps taken. GFM2000 isochore files (DTN: MO0012MWDGFM02.002 [153777]) that are not combined or subdivided include: • ia00cpv1RWC.2grd • ia00tppRWC.2grd • ia00tpmnRWC.2grd • ia00tpllRWC.2grd • ia00tpv3RWC.2grd • ia00tpv2RWC.2grd • ia00tacbtRWC.2grd • ia00prowuvRWC.2grd • ia00prowucRWC.2grd These grids, which contain regularly spaced (61 × 61 m) data, require no manipulation other than simple formatting for incorporation into the UZ grids. EARTHVISION V5. (Dynamic Graphics 1998 [152614]) is used to export the regularly spaced data and to convert the units (x, y, and thickness) from feet to meters. Since GFM2000 data coverage (BSC 2002 [159124], Figure 1) extends well beyond the UZ Model boundaries, each data file is reduced to the approximate UZ Model domain, using the EARTHVISION V5.1 Graphic Editor to remove data points lying south of N 228,820 m and east of E 174,860 m. GFM2000 isochore files (DTN: MO0012MWDGFM02.002 [153777]) that are combined include: • ia00cpv3RWC.2grd + ia00cpv2RWC.2grd • ia00bt4RWC.2grd + part of ia00tpyRWC.2grd (see discussion of Tpy below) • ia00bt3RWC.2grd + part of ia00tpyRWC.2grd (see discussion of Tpy below) • ia00bt2RWC.2grd + ia00trv3RWC.2grd + ia00trv2RWC.2grd • ia00trv1RWC.2grd + part of ia00trnRWC.2grd (see discussion of Tptrv1 and Tptrn below) • ia00trltfRWC.2grd + ia00tpulRWC.2grd • ia00tpv1RWC.2grd + ia00bt1RWC.2grd • ia00prowmdRWC.2grd + ia00prowlcRWC.2grd • ia00prowlvRWC.2grd + ia00prowbtRWC.2grd + ia00bulluvRWC.2grd • ia00bullucRWC.2grd + ia00bullmdRWC.2grd + ia00bulllcRWC.2grd • ia00bulllvRWC.2grd + ia00bullbtRWC.2grd + ia00tramuvRWC.2grd • ia00tramucRWC.2grd + ia00trammdRWC.2grd + ia00tramlcRWC.2grd The EARTHVISION V5.1 (Dynamic Graphics 1998 [152614]) Formula Processor is used to add the *.2grd files as shown above. The resulting files are then formatted as previously described for uncombined isochores. Subdivided GFM2000 isochore files (DTN: MO0012MWDGFM02.002 [153777]) are described below and include: • ia00tpyRWC.2grd • ia00trv1RWC.2grd + ia00trnRWC.2grd • ia00tplnRWC.2grd • ia00tacRWC.2grd GFM2000 layer Tpy (Yucca Mountain Tuff)—Based on the HGUs defined by Flint (1998 [100033]), GFM2000 layer Tpy is subdivided vertically into three layers (see Table 11). The upper portion is typically nonwelded and has properties similar to Tpbt4 (BT4); therefore, it is combined with layer Tpbt4 (GFM2000 isochore file “ia00bt4RWC.2grd” (see Attachment I, GFM 2000 files)) to make UZ02 Model layer “ptn22.” The middle portion can become moderately welded to the north (porosity <30%), where layer Tpy is generally thicker. This middle portion corresponds to HGU “TPY” and is designated “ptn23” in the UZ02 grid. The lower portion is typically nonwelded and has properties similar to Tpbt4 and Tpbt3, and is therefore combined with layer Tpbt3 (GFM2000 isochore file “ia00bt3RWC.2grd” (see Attachment I, GFM2000 files)) to make UZ02 Model layer “ptn24.” Because the presence of the hydrologically distinct middle portion of layer Tpy depends on the overall thickness of the unit, the isochore for layer Tpy is subdivided as follows: • Where Tpy is <6 m thick, the total Tpy thickness is combined with layer Tpbt4 to create UZ02 Model layer “ptn22” (corresponding to HGU “BT4”). • Where Tpy thickness is between 6 and 9 m, the thickness is split in half: the upper half is combined with Tpbt4 to make UZ Model layer “ptn22,” while the lower half is combined with Tpbt3 to make UZ02 Model layer “ptn24” (corresponding to HGU “BT3”). • Where Tpy thickness is between 9 and 12 m, 2 m is assigned to UZ02 Model layer “ptn23” (corresponding to HGU “TPY”); the remainder is split in half, and these equal portions are combined with Tpbt4 to make UZ02 layer “ptn22” and TPbt3 to make layer “ptn24.” • Where Tpy thickness is between 12 and 15 m, 3 m is assigned to UZ02 Model layer “ptn23” (corresponding to HGU “TPY”); the remainder is split in half, and these equal portions are combined with Tpbt4 to make UZ02 layer “ptn22” and TPbt3 to make layer “ptn24.” • Where Tpy thickness is greater than 15 m, the unit is divided in thirds, with one third assigned (in combination with Tpbt4) to “ptn22,” one third to “ptn23,” and the remaining third is combined with Tpbt3 to make “ptn24.” GFM2000 layers Tptrv1 and Tptrn (upper Topopah Spring Tuff)—The densely welded Tptrv1 is relatively thin (0–2 m thick, typically <0.5 m) across Yucca Mountain (Flint 1998 [100033], p. 27). Given a minimum vertical resolution of 1.0 m for the UZ Model grids, this layer would be missing from UZ simulations across most of Yucca Mountain. To capture this potentially important flow unit at the PTn/TSw interface (see Table 11), GFM2000 isochores (DTN: MO0012MWDGFM02.002 [153777]) for Tptrv1 and Tptrn were combined, and then the upper 2 m of this combined unit were assigned a distinct model layer name corresponding to Flint’s “TC” HGU. The remaining thickness of the combined unit (Tptrv1 + Tptrn - 2 m) corresponds to Flint’s “TR” HGU. Where the combined thickness of Tptrv1 and Tptrn is less than 0.5 m, the isochore for the “TC” HGU is assigned zero thickness. GFM2000 layer Tptpln (Topopah Spring, lower nonlithophysal)—Tptpln is characterized by HGUs “TM2” and “TM1” (see Table 11). According to the proportions given in Flint (1998 [100033], p. 3), GFM2000 layer Tptpln is vertically subdivided into an upper portion (with 2/3 the total thickness of Tptpln) and a lower portion (with 1/3 the total thickness of Tptpln) for incorporation into the UZ Model. GFM2000 layer Tac (Calico Hills Formation)—The Tac is subdivided vertically into four equal layers because of its large thickness beneath northern Yucca Mountain (see Table 11). After the isochores have been subdivided according to the specified criteria/proportions, they are formatted using the same steps that were used to format the uncombined isochores. 6.4.2 Reference Horizons, and Top and Bottom UZ Model Boundaries WINGRIDDER V2.0 (LBNL 2002 [154785]) generates a numerical grid based on the elevations of three major horizons: (1) a top boundary (e.g., the topographic or bedrock surface), (2) a structural reference horizon, which identifies faults and their associated offsets, and (3) a bottom boundary (i.e., the water table). The reference horizon is a surface from which elevations of all hydrogeologic-unit interfaces are calculated by stacking layer thicknesses above or below it, based on their stratigraphic position. All offsets resulting from faulting are described by the reference horizon data. Any portions of HGUs lying above the top boundary or below the bottom boundary after stacking are removed (clipped). GFM2000 horizons used (see Attachment I, GFM2000 files): • s00bedrockRWC.2grd (bedrock/present-day erosional surface; UZ Model top boundary) • s00TpcpEXuncut.2grd (top of Tpcp; surface used in the absence of Tpcp isochore) • s00Tptpv3EXuncut.2grd (top of Tptpv3; primary structural reference horizon for UZ grids). The top of layer Tpcp (the contact between the crystal-rich and crystal-poor tuffs of the Tiva Canyon, defined as a surface in GFM2000) (DTN: MO0012MWDGFM02.002 [153777]) is used to separate UZ Model layers “tcw11” and “tcw12” (see Table 11), since no GFM2000 isochore grids (DTN: MO0012MWDGFM02.002 [153777]) exist for these layers. As with the isochore grids, the horizon grids, which also contain regularly spaced (61 × 61 m) data, require no manipulation other than simple formatting for incorporation into the UZ Model. EARTHVISION V5.1 is used to export the regularly spaced data and to convert the units (x, y, and elevation) from feet to meters. The complete details for formatting these GFM2000 horizon grids (DTN: MO0012MWDGFM02.002 [153777]) are documented in the scientific notebook by Wang (2002 [159673], SN-LBNL-SCI-213-V1, pp. 20–22). The lower boundary of the UZ Model (the water table) was discussed previously in Sections 5.1 and Section 6.2. The input data set (gwl_sspac2.asc) used to define the water table at the base of the UZ was obtained from DTN: MO0212GWLSSPAX.000) [161271]. These input data consist of borehole water-level elevations (consistent with qualified data in DTNs: MO0106RIB00038.001 [155631] and GS010608312332.001 [155307]) along with interpreted potentiometric surface contour lines. This surface was constructed under the assumption that the water levels in G-2 and WT-6 represent perched water, and the level in WT-24 represents the regional groundwater surface (USGS 2001 [157611]; also see Assumptions 1 and 2 in Section 5.1). The data were derived from the Vulcan GFM2000 layer “GWL_SSPAC” (DTN: MO0110MWDGFM26.002 [160565]). The review and qualification process for this data set is documented in Attachment IV. The file containing the water table data was then edited to make it compatible with EARTHVISION V5.1. The resulting data were gridded using the 2-D minimum tension gridding function in EARTHVISION V5.1 to produce a surface defined by a regularly spaced (182.88 by 182.88 m) data set. The data defining this surface were then exported using the 2-D and 3-D grid export function in EARTHVISION V5.1, and subsequently regridded using the 2-D minimum tension gridding function to produce a surface defined by a regularly spaced (60.96 by 60.96 m) data set (“gwl_sspac_60.96.2grd” in output-DTN LB02092DGRDVER.001). The 2-D and 3-D grid export function was then utilized again to produce a file with the 60.96 by 60.96 m regularly spaced data set required as input for grid generation using WINGRIDDER V2.0. The file was edited to ensure that a minimum elevation of 730 m was used, thus revising lower elevations that resulted from the minimum tension gridding process (Wang 2003 [162380], SN-LBNL- SCI-213-V1, p. 117). This file was then edited (by cropping the data, removing xy coordinates, and modifying the header) to create a reference horizon file (“REF_wt_sspac.dat” in output DTN: LB0208HYDSTRAT.001) suitable as input for WINGRIDDER V2.0. The details of these steps can be found in the scientific notebook by Wang (2002 [159673], SN-LBNL-SCI- 213-V1, pp. 21–22). The gridding procedure used to define the water table in EARTHVISION V5.1 was conducted using a two-step process (irregularly spaced data to a coarsely spaced grid, followed by a finely spaced grid) to avoid generating large deviations from the contoured potentiometric surface as represented by the contours from DTN: MO0212GWLSSPAX.000 [161271]. However, this gridding process, which is required to produce the data input needed for numerical grid generation using WINGRIDDER V2.0 does result in small deviations in the water table relative to the surface initially defined by DTN: MO0110MWDGFM26.002 [160565]. The deviations in water table elevation are typically < 5 m in the area of the proposed repository footprint. Further minor modification to this surface occurs when the reference horizon file “REF_wt_sspac.dat” (Output-DTN: LB0208HYDSTRAT.001) is used to constrain the lower bounds for each column of the numerical grids produced by WINGRIDDER V2.0. However, there are larger (up to 60 m) observed discrepancies in the original (USGS 2001 [157611], Figure 6-1) and output (Figure 1b) water table elevations that may result from errors associated with contour digitization prior to generation of the Vulcan water table representation (See Attachment IV). Further discussion of the uncertainties associated with the definition of the water table is presented in Section 7.1.1 and Attachment IV. 6.5 2-D GRID GENERATION Used by WINGRIDDER V2.0 (LBNL 2002 [154785]) to organize grid information, the 2-D (map-view) grid (Figure III-3) defines the structure of columns and segments that provide the basis for projecting the 3-D grid. Each column is represented by a node in map-view indicating the column's position in the x-y plane. Additionally, the shape of each column is a polygon in the x-y plane whose boundaries consist of segments defined prior to 3-D grid generation. Grid development begins with the assignment of nodes in map view for each object (e.g., domain nodes, fault nodes, repository nodes) with specified orientation and density. Based on the location of these nodes, a primary 2-D grid is generated using Voronoi tessellation techniques (e.g., Aurenhammer 1991 [160333]) embedded in the WINGRIDDER V2.0 numerical code. The 2-D grid is then improved systematically and interactively by deleting physically incorrect or unnecessary connections. A few iterations of these steps, including adding, moving, and deleting certain nodes, are necessary to create a final 2-D grid, or column scheme, that serves as the basis for generating the vertical component of the grid. Two-dimensional grid generation for the UZ Model incorporates the location of domain and proposed repository boundaries, borehole locations, and map-view traces of major faults. As mentioned in Section 6.3, the fault trace information taken from an elevation of 1,100 masl (DTN: MO0012MWDGFM02.002 [153777]), from GFM 2000 was used to define the map-view traces for the 2-D grid. Various subsets of these features are included in the different UZ Model grids, depending on their intended use. Because the 1-D hydrogeologic-property-set inversions only consider rock-property data from vertical boreholes, only borehole locations are relevant when generating this particular grid. The 3-D UZ Model grid assigns nodes in 2-D to the location of all data sources (i.e., boreholes), as well as within domain and proposed repository boundaries and along faults. Another issue considered in 2-D grid generation is spatial resolution. Grid resolution (node spacing) is a compromise between computational efficiency and a need to capture spatial variability in rock properties and boundary conditions (such as infiltration rate). As discussed in Section 6.6, additional grid resolution was added to the PTn units and the repository, two features that previous Yucca Mountain flow model studies identified as needing enhanced numerical resolution to capture the effects of spatial variability on flow (BSC 2001 [155950], Section 3.3.4.8.1). The 3-D grid captures the needed spatial variability in the infiltration rate at the bedrock surface for calibration purposes, while containing sufficient numerical resolution within the proposed repository boundary, the area most important to PA studies. 6.6 3-D GRID GENERATION Once UZ Model grid nodes are assigned in plan view and polygons are generated representing the lateral extent of each grid column, model layer contact elevations are determined for each vertical column within the UZ Model grid, using a bilinear interpolation method to determine values between the regularly spaced (61 x 61 m) nodes of the GFM2000 grid (DTN: MO0012MWDGFM02.002 [153777]). The estimated maximum error in layer contact elevations at UZ Model column centers associated with this interpolation method is about 5 m, except in areas affected by faulting (see Attachments II and III for grid verification), assuming that the hydrogeologic layers dip 10 degrees. Dips are generally less than 10 degrees (BSC 2002 [159124], Section 6.4), and thus a value of 10 degrees was used to calculate the maximum error value. This amount of potential error is considered insignificant to grid development and subsequent site-scale UZ Model simulation activities because lateral column dimensions almost always exceed 61 × 61 m (except along faults), thus encompassing the nearest GFM2000 data point. The 3-D grid describes the location, rock material name, and connection information for each 3-D gridblock in the UZ Model domain. All 3-D gridblocks are generated column by column with WINGRIDDER V2.0 (LBNL 2002 [154785]), based on the 2-D (plan-view) grid design, to ensure that each vertical connection occurs between adjacent gridblocks and that each gridblock has at least one vertical connection. Lateral connections are then generated segment by segment within a model layer, with each segment joining two neighboring columns. This ensures that only gridblocks in two adjacent columns have lateral connections and that no connections between two adjacent columns are missing. For a given column, 3-D gridblocks are built for each HGU, first above the Tptpv3 structural reference surface until reaching the bedrock surface, and then below this reference surface down to the water table. The interfaces of the generated gridblocks are located exactly at the interfaces of the corresponding hydrogeologic layers. Vertical connections within the column are generated after each gridblock is built. A dummy gridblock is added to the top and bottom of each column to enable assignment of model boundary conditions. When building lateral connections, each pair of two adjacent columns are searched top-to- bottom. If gridblocks in the adjacent columns belong to the same layer, a lateral connection is built for them. The lateral interface area is determined by the length of the shared side multiplied by the height of the shorter of the two gridblocks that are connected. If the layer is missing in one of the two neighboring columns (resulting from a layer pinching out), the gridblock representing the last occurrence of the pinch-out layer is laterally connected to the adjacent gridblock, now occupied by the next hydrogeologic layer. The height of that interface at the pinch-out margin is reduced to 0.10 m (10% of the minimum gridblock height). This value was chosen assuming that the pinch-out layers are not just layer discontinuities, and that permeable connections are preserved. If one of the two adjacent columns is a fault, the lateral connections are built based on elevations only. The maximum thickness of any cell within the UZ grids is 20 m (Wang 2002 [159673], SN- LBNL-SCI-103-V1, pp. 135–136). If the thickness of a model layer within a column exceeds 20 m, the layer is subdivided equally into two layers. Minimum vertical grid resolution is 1.0 m; thus, if the thickness of a hydrogeologic layer is less than 1.0 m within a column, the layer is considered absent, and no gridblock is generated for the layer at this location. To conserve the total thickness of the UZ, layer thicknesses below this cutoff are added to the overlying layer if they lie above the structural reference horizon (i.e., top of Tptpv3), or are added to the underlying layer if they lie below the reference horizon. Still, this may lead to a significant discontinuity if many thin, adjacent layers exist. Within UZ Model boundaries, however, no more than two adjacent hydrogeologic layers, each with a thickness less than 1.0 m, occur in any vertical column, except for a few locations near the land surface where erosion has removed most of the crystal-poor Tiva Canyon Tuff (Tpcp), and the underlying Tpcpv units (model layers tcw13 and ptn21) are also less than 1.0 m thick. In this rare case, the small layer thicknesses are added to the underlying layer, ptn22. Further vertical grid resolution is added within the Ptn units ptn22, ptn24, ptn25, and ptn26, as well as the unit ch1 and the proposed repository horizon, where a maximum cell thickness of 5 m is used (Wang 2002 [159673], SN-LBNL-SCI-103-V1, pp. 135–136). Sensitivity studies examining the effects of grid refinement on flow and transport models indicate that a vertically refined grid is needed to capture lateral flow caused by capillary barriers formed by the layers ptn21 and ptn23 (BSC 2001 [155950], Sections 3.3.3.4.2 and 3.3.4.8.1; Wu et al. 2002 [161058], pp. 7–8, 11, Fig. 7), and thus enhanced grid refinement (maximum cell thickness of 2 m) was assigned to ptn21 and ptn23. Detailed grid resolution to the repository footprint (Assumption 8) allows flow models to better capture spatial variability (BSC 2001 [155950], Section 3.3.4.8.1). The proposed repository itself is represented by five grid layers, each 5 m thick. Material properties are assigned to gridblocks depending on the hydrogeologic layer to which the gridblock corresponds. For layers with multiple properties, such as the vitric and zeolitic zones within the lowermost Topopah Spring and Calico Hills units, polygons defining the areal extent of these zones are created (see Section 6.6.3). Assignment of material properties (i.e., vitric or zeolitic) to model gridblocks is then confined to the appropriate polygon. 6.6.1 Faults Although faults may occur as displacement surfaces only or as deformation zones of variable width, each fault within the current UZ Model domain is represented by columns of gridblocks having an arbitrary width of 30 m. Nevertheless, adjustments can be made within a grid to assign appropriate rock properties to each fault zone to handle various fault configurations. Conceptually, there are three important features of a fault that are conserved in the numerical grid. First, a fault is a separator that causes discontinuity of geological layers and may serve as a structural barrier to lateral flow. Second, a fault zone is continuous and may serve as a fast path for vertical flow depending on its hydraulic properties. Third, a fault may or may not be vertical, and its angle of inclination may vary spatially. To implement these features in the UZ grids, three parallel rows of fault-related columns are built for each fault. Each section of a fault in map view consists of three connected columns, with the fault column located in the middle (Figure 2). Each fault column is connected to two side columns and two neighboring fault columns only. Columns on opposite sides of a fault are always separated by a fault column. Figure 2. Schematic Illustration of Fault-Related Gridblocks in Map View and in Cross Section The three fault-related columns (the fault column and its two side columns) are processed together to generate 3-D gridblocks representing the fault and layer offset. From the bedrock surface to the water table, the x, y location of fault gridblocks may shift according to the elevation and dip of the fault. Similarly, the volumes and the center (nodal point) location of the corresponding side cells are adjusted accordingly. As a result, the inclination of the fault is described by a series of connected gridblocks whose x, y locations vary with elevation. The fault-related gridblocks are connected vertically, if they belong to the same column, regardless of the fault angle. Columns of side cells are connected in a similar fashion regardless of the horizontal shifting of position and change in volume. To look at it from another perspective, each set of three fault-related columns (i.e., the fault column plus its two side columns) can be viewed collectively as one vertical column that is subdivided into three nonvertical columns to capture the angle of inclination along a fault. One limitation of this method is that intersecting faults cannot be represented. This method of representing the three-dimensionality of faults requires that all fault gridblocks have the same elevation and thickness as the laterally adjacent gridblock to facilitate vertical displacement of geologic layers. Because Yucca Mountain is comprised of hydrogeologic layers with variable thickness, simply reassigning material properties from one row of gridblocks to another to establish offset along faults is insufficient for representing the true layer configurations. This approach removes certain layers from columns adjacent to fault columns and often misrepresents layer thicknesses. To avoid such error, additional vertical resolution is added to fault-related gridblocks based on the elevation of hydrogeologic layer contacts on both sides of the fault. Therefore, vertical grid discretization in each set of three fault-related columns is identical, and all interfaces between hydrogeologic units in both side columns correspond to the interfaces between gridblocks. The layer and rock properties of fault gridblocks are then assigned according to the stratigraphy of the fault column. The assignment of lateral connections that involve fault-related gridblocks is different from the way lateral connections are assigned to normal (non-fault-related) gridblocks. Fault-related lateral connections are of two types, fault-fault gridblock connections and fault-side cell connections. In these two cases, lateral connections occur between gridblocks that share the same interface. The interface area is precisely determined by the contact area between the two gridblocks. As mentioned in Section 6.3, some simplification of the GFM2000 faults was made in creating the UZ Model grids, including the representation of the Solitario Canyon and Solitario Canyon (west) faults as a single fault. During the evaluation of the 3-D grid described in Attachment III, it was discovered that some matrix columns adjacent to fault columns exhibited fault-related stratigraphic offset with their neighboring columns. To ensure proper flow behavior in the grid, these columns were classified as "faults” while building the 3-D grid so that lateral connections between gridblocks in these columns and those in the adjacent columns were made with the closest lateral neighbor, and not with the same stratigraphic interval. A total of 18 columns, all adjacent to faults, were treated in this manner. 6.6.2 Repository For numerical gridding purposes, the proposed repository is defined as a 3-D object that is subdivided into a regular mesh of gridblocks. The repository design used in the constuction of the numerical grids (BSC 2002 [159527]) calls for two sets of waste emplacement drifts to be constructed, with the primary proposed repository area located west and/or north of the ESF Main Drift, and the lower elevation block located east of the primary block (Assumption 8). Note that the lower elevation block has been removed from the most recent revision of the proposed repository layout (BSC 2003 [161726]; BSC 2003 [161727]). All repository columns are aligned along the direction of the emplacement drifts, as currently designed, and each column of gridblocks (except those corresponding to borehole locations) has four sides to facilitate the representation of a drift with a series of connected 3-D gridblocks. Local refinement is added vertically at the proposed repository horizon in the UZ Model grids for PA. For each repository column, a repository thickness of 25 m is assigned at the appropriate elevation. This thickness is then divided vertically into five layers, each 5 m thick. For the interfaces between repository gridblocks, lateral connections are established if two adjacent gridblocks belong to the same layer within the five-layer grid structure of the proposed repository horizon. For interfaces between a repository gridblock and a nonrepository gridblock, the connection is built based on their hydrogeologic-layer similarity. The assignment of rock properties to repository gridblocks is determined by the elevation of the gridblock and the corresponding hydrogeologic layer present at that elevation. 6.6.3 Vitric/Zeolitic Boundaries The ISM3.1 Rock Properties Model (Assumptions 4 and 5) is used together with measured rock- property measurements from boreholes and corroborative data from the RPM2000 and Mineralogical Model 3.1 (MM3.1) to add resolution to UZ Model grids within the lowermost Topopah Spring tuffs (TSw) and Calico Hills nonwelded unit (CHn). Of great importance to UZ flow and transport modeling is the distribution of low-permeability zeolites, because of their potential to significantly alter flowpaths and travel times and to retard radionuclides migrating from the proposed repository horizon to the water table. At high matrix saturations, groundwater flow within the TSw and CHn should be diverted around zeolitic volumes of rock and preferentially flow through the less-altered, higher- permeability vitric matrix. Consequently, only a low percentage of the total percolation flux is expected to travel through significantly zeolitized tuffs. This suggests that sorption within the slightly altered (mostly vitric) tuffs is of far greater importance. As such, high- and low- permeability regions are defined within certain UZ Model layers corresponding to the tuffs of the lowermost TSw and upper CHn (above lithostratigraphic unit Tcpuv). Lateral boundaries between high- and low-permeability tuffs within the lowermost TSw and upper CHn were determined using results from the geostatistical RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]), as well as information found in RPM2000 (DTN: SN0112T0501399.004 [159524]) and MM3.1 (DTN: MO9910MWDISMMM.003 [119199]). The details and results of this exercise and a comparison between RPM3.1, RPM2000, and MM3.1 are provided below. The net result is the subdivision of the lithostratigraphic unit Tac (see Table 11) vertically into four grid layers (ch2, ch3, ch4, ch5), and laterally into vitric and zeolitic regions for which separate hydrogeologic and sorptive properties are assigned. The UZ Model layers tsw 39 (corresponding to the Tptpv2), ch1 (corresponding to the combined lithostratigraphic units Tptpv1 and Tpbt1), and ch6 (corresponding to the Tacbt) are also laterally subdivided into vitric and zeolitic regions. Note that the horizontal and vertical resolution of the UZ Model grids is too coarse to capture meter-scale heterogeneity within the CHn. Small-scale heterogeneity is, however, observed within the CHn and may have an impact on flow and transport calculations. Data from the Rock Properties Models 3.1 (DTN: MO9910MWDISMRP.002 [145731]) and 2000 (DTN: SN0112T0501399.004 [159524]) and Mineralogic Models 3.1 (DTN: MO9910MWDISMMM.003 [119199]) are analyzed in EARTHVISION V5.1 (Dynamic Graphics 1998 [152614]) by generating map-view figures of percent-zeolite distribution (from MM3.1), interpreted saturated hydraulic conductivity (Ksat) data (from RPM3.1), and a contoured region with <0.5 hydrous-phase alteration (from RPM2000). Results from the RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) were used as the primary means to define vitric and zeolitic boundaries. Because RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) does not include more recent rock-property data from SD-6, saturation, porosity, and hydraulic conductivity data from this borehole (DTNs: GS980908312242.038 [107154] and GS980808312242.014 [106748]) are used to modify zeolitic and vitric boundaries where appropriate. In general, vitric material is characterized by relatively low saturation (<~90%), relatively high Ksat (>~10-10 m/s), and oven-dried porosity that is less than 5% higher than relative-humidity porosity. The MM3.1 (DTN: MO9910MWDISMMM.003 [119199]) and RPM2000 (DTN: SN0112T0501399.004 [159524]) data are used as corroborative evidence for the presence of vitric and zeolitic tuffs. Major faults are assumed to represent appropriate lateral boundaries for unaltered areas (Assumption 6). It is reasonable to assume that vitric portions of the CHn may be laterally continuous within fault blocks that have a higher structural position above the water table compared to adjacent structural blocks. For example, the Solitario Canyon fault system offsets the CHn by more than 300 m in the southern part of the UZ Model domain. CHn layers west of the Solitario Canyon fault lie near or below the water table in this area; consequently they are most likely altered to zeolites. In contrast, CHn layers east of the Solitario Canyon fault may be up to 300 m above the water table and are less likely to have undergone alteration because of limited rock/water interaction. The vertical offset along the Dune Wash fault suggests that this is another possible boundary for vitric and zeolitic subunits within the CHn. As a result, major faults are considered as potential boundaries between vitric and zeolitic areas when interpreting data from RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) and corroborative evidence from RPM2000 (DTN: SN0112T0501399.004 [159524]) and MM3.1 (DTN: MO9910MWDISMMM.003 [119199]). RPM3.1 uses porosity (data that are relatively abundant at Yucca Mountain) as a surrogate to predict Ksat values. The limitations of this correlation are discussed in Rautman and McKenna (1997 [100643], pp. 13–14). In the RPM3.1 (DTN: MO9910MWDISMRP.002 [145371]), the CHn consists of the volume of rock lying between the upper Tptpv1 contact and the lower Tacbt contact (in other words, geologic layers Tptpv1, Tpbt1, Tac, and Tacbt, shown in Table 11, equivalent to the UZ Model HGUs ch1–ch6). Ksat distributions within the CHn (represented by 24 grid layers in the Rock Properties Model) are plotted in EARTHVISION V5.1 by contouring (2-D minimum tension gridding) the regularly spaced (200 × 200 m) Ksat data for each of the 24 rock-property grid layers. The 24 rock-property grid layers are not stratabound; rather, they are equally thick at any given x, y coordinate. An equivalent GFM2000 isochore file was created by combining the thicknesses of the layers mentioned above. Using the midpoint surface positions for each of the UZ Model layers, Ksat isosurfaces were then back-interpolated from the RPM3.1 file “ChnZksStrat.3grd” (see Attachment I, ISM 3.1 files). The plots show Ksat data that range from approximately 10-5 to 10-12 m/s; note that Ksat values >10-10 m/s are assumed to represent vitric tuffs (Assumption 4). Figure 3 shows an example of one of these Ksat plots for the upper Tac (UZ Model layer ch2) lithostratigraphic unit. Details explaining the extraction of relevant ISM3.1 rock-property data used to define vitric boundaries within UZ Model grid layers are documented in the scientific notebook by Wang (2002 [159673], SN-LBNL-SCI-213-V1, p. 24). A similar approach was used to evaluate vitric and altered tuffs using data from RPM2000 (DTN: SN0112T0501399.004 [159524]). This version of the RPM contains data from boreholes not included in RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]), but is not qualified, and thus can be used only for corroborative purposes. The RPM2000 file “CHn_hmap_etype.out” (see Attachment I, RPM2000 files) is an “E-type” model of hydrous- phase mineral alteration in the form of a probability distribution, with values close to 1 indicating a strong probability of mineral alteration to phases such as zeolites and clays. For more discussion on E-type models, see Rock Properties Model Analysis Model Report (BSC 2002 [159530], Sections 6.1 and 6.4.8.3). Using the mid-point elevation of UZ Model layers ch1, ch2, ch3, ch4, ch5, and ch6, faces files were created for each unit, where the 0.5 probability contour is interpreted to represent the vitric-zeolitic boundary, and where altered (zeolitic) tuffs lie on the >0.5 probability side of the contour line. Figure 4 shows an example of one of these alteration- probability contour plots for the upper Tac (UZ Model layer ch2) lithostratigraphic unit. Details explaining the extraction of RPM2000 rock-property data used to define vitric boundaries within UZ Model grid layers are documented in the scientific notebook by Wang (2002 [159673], SN- LBNL-SCI-213-V1, pp. 24–25). Percent-zeolite plots were also made from MM3.1 data (DTN: MO9910MWDISMMM.003 [119199]) in EARTHVISION V5.1 by contouring (2-D minimum tension gridding) the regularly spaced (61 × 61 m) percent-zeolite data for the CHn contained in the ISM3.1 file “mineralsM.pdat” (see Attachment I, ISM 3.1 files). The plots essentially represent the exact results of the Mineralogic Model. The plots show a general trend of increased zeolitic alteration to the north and east across the model area. Figure 5 is an example of one of these plots for the upper one-fourth of the Tac lithostratigraphic unit. This representation of zeolite distribution is not appropriate as the primary means of defining vitric-zeolitic boundaries in the numerical grids discussed in this Scientific Analysis report because of the lack of mineralogic sample data and the interpolation technique used in the development of the Mineralogic Model. However, these data can be used for corroborative purposes. The interpreted extent of the vitric-zeolitic boundaries from the above analysis are shown in Figures 6a and 6b. These boundaries are used in WINGRIDDER V2.0 to assign material names to gridblocks (i.e., “vitric” or “zeolitic,” for which associated rock properties will be assigned) within UZ Model layers tsw39, ch1, ch2, ch3, ch4, ch5, and ch6. These boundaries were selected using the results of the RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) Ksat plots (Assumption 4), measured rock-property data for boreholes within the UZ Model area (Assumption 5), and the location of faults with significant vertical offset (Assumption 6). A summary of how vitric/zeolitic boundaries were defined for each UZ Model layer is presented Development of Numerical Grids for UZ Flow and Transport Modeling U0000 ANL-NBS-HS-000015 REV01 63 April 2003 below; additional details can be found in the scientific notebook by Wang (2002 [159673], SN- LBNL-SCI-213-V1, pp. 25–34, 63–66; 2003 [162380], SN-LBNL-SCI-V1, p. 67). 101010101010-5-6-7-8-9-10Property Color KeyDisplay: Tac4-mid.facesUnits: Ksat (m/s) NSP Easting (m) NSP Northing (m) DTN: MO9910MWDISMRP.002 (Rock Properties Model 3.1) [145731] NOTE: Values less than 10-10 m/s given by white. Figure 3. Distribution of Ksat from ISM3.1 Rock Properties Model, Upper 1/4 of Layer Tac (UZ Model layer “ch2”). Ksat Contour Units are m/s. Development of Numerical Grids for UZ Flow and Transport Modeling U0000 ANL-NBS-HS-000015 REV01 64 April 2003 NSP Easting (m) NSP Northing (m) DTN: SN0112T0501399.004 (Rock Properties Model 2000 (Non-Q)) [159524] NOTE: Hachured area within UZ Model boundary indicates vitric tuff Figure 4. Alteration Probability Contour (0.5) Plot from RPM2000, Upper 1/4 of Layer Tac (UZ Model layer “ch2”) 239000 238000 Property Color Key Display: Tac431.pdat NSP Northing (m) Units: % zeolites 237000 72 68 64 60 56 52 48 44 236000 235000 40 36 32 28 24 234000 20 16 12 8.0 233000 4.0 0.0 232000 231000 230000 229000 168000 169000 170000 171000 172000 173000 174000 G-2 WT-24 UZ#4 H-6 SD-6 SD-12 UZ#13 SD-7 H-3 G-3 H-1 G-4 NRG#5a#4 NRG-7A SD-9 H-4 UZ-14 G-1 H-5 NRG-6 NRG#4 NSP Easting (m) DTN: MO9910MWDISMMM.003 (Mineralogic Model 3.1) [119199] NOTE: Vitric region denoted by purple Figure 5. Percent Zeolite Distribution from ISM3.1 Mineralogic Model, Upper 1/4 of Layer Tac (UZ Model layer “ch2”) Development of Numerical Grids for UZ Flow and Transport Modeling U0000 ANL-NBS-HS-000015 REV01 66 April 2003 ch1ch2 NSP Northing (m) NSP Northing (m) NSP Easting (m)NSP Easting (m) tsw39NSP Easting (m) NSP Northing (m) LEGEND Boreholes Vitric Region UZ Model Boundary FaultUZ#16NOTE:Non-hachured areas withinUZ Model boundary indicate "zeolitic" material Output-DTN: LB0208HYDSTRAT.001 Figure 6a. Extent of Vitric Region in FY02 UZ Model Layers tsw39, ch1 and ch2 Development of Numerical Grids for UZ Flow and Transport Modeling U0000 ANL-NBS-HS-000015 REV01 67 April 2003 ch5ch6ch4ch3NSP Easting (m) NSP Northing (m) NSP Easting (m) NSP Northing (m) NSP Easting (m) NSP Northing (m) NSP Easting (m) NSP Northing (m) Output-DTN: LB0208HYDSTRAT.001 Figure 6b. Extent of Vitric Region in FY02 UZ Model Layers ch3, ch4, ch5 and ch6 Tsw39 (Tptpv2) Because RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) cannot be easily used to evaluate Ksat values for the unit Tptpv2, rock-property data from boreholes SD-6, SD-7, SD-9, SD-12, UZ-14, UZ-16, NRG-7a, and WT-24 (DTN: LB0207REVUZPRP.002 [159672]; see Wang 2002 [159673], SN-LBNL-SCI-213-V1, pp. 63–66, 93 for details) were the primary input used to define the vitric and zeolitic regions for layer tsw39. Tuffs were characterized as vitric when the following properties were observed: relatively low saturation (<~90%), relatively high Ksat (>~10-10 m/s), and a difference between oven-dried and relative-humidity porosities of less than 5% (Assumptions 4 and 5). An evaluation of these rock properties within this unit for the boreholes listed above suggests that the boreholes SD-6, SD-7, SD-9 and SD-12 contain vitric tuffs, UZ-14, UZ-16, and WT-24 contain zeolitic tuffs, and that NRG-7a has samples with both vitric and zeolitic properties. However, to reconcile the presence of perched water above this unit in boreholes SD-9 and NRG-7a (Rousseau et al. 1999 [102097], pp. 170–171), these boreholes were assigned to lie near the boundary, but within the zeolitic region. In general, the vitriczeolitic boundary for this unit is similar in shape to that determined for the underlying ch1 unit. The Dune Wash fault system was used to bound a portion of the eastern margin of the vitric zone. Ch1 (Tptpv1 + Tpbt1) The vitric region is initially defined by RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) Ksat data and data from boreholes SD-6, SD-7, SD-12, G-3, H-3, H-5, H-6, and WT-2. Rock- property data for SD-6 (DTN: GS980808312242.014 [106748]) within this unit (corresponding to a depth interval of 463.3–475.8 m) report low saturations (29–51%), and differences in oven- dried and relative-humidity porosities less than 5%, indicating that the ch1 interval in this borehole is vitric. Two of the three hydraulic conductivity values reported for this borehole (DTN: GS98090831224.038 [107154]) are greater than 10-10 m/s, consistent with the vitric interpretation. The Dune Wash fault system was used to bound a portion of the eastern margin of the vitric zone. Ch2 (upper ¼ of Tac) The vitric region is initially defined by RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) Ksat data and data from boreholes SD-6, SD-7, SD-12, G-3, H-3, H-5, H-6, and WT-2. Rock- property data for SD-6 (DTN: GS980808312242.014 [106748]) within this unit (corresponding to a depth interval of 475.8–483.6 m) report low saturations (<70%, avg. 35%), and differences in oven-dried and relative-humidity porosities less than 5%, indicating that the ch2 interval in this borehole is vitric. The three hydraulic conductivity values reported for this borehole (DTN: GS98090831224.038 [107154]) are greater than 10-10 m/s, consistent with the vitric interpretation. The Dune Wash fault system was used to bound a portion of the eastern margin of the vitric zone, and the Solitario Canyon fault, which downdrops the region to the west by over 200 m (Figure III-4), was assumed to form the western boundary of the vitric zone for this unit, thus resulting in assigning the H-6 borehole as zeolitic (consistent with the results of RPM3.1 (DTN: MO9910MWDISMRP.002 [145731])). Ch3 (mid-upper ¼ of Tac) The vitric region is initially defined by RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) Ksat data and data from boreholes SD-6, SD-7, SD-12, G-3, H-3, H-5, H-6, and WT-2. Rock- property data for SD-6 (DTN: GS980808312242.014 [106748]) within this unit (corresponding to a depth interval of 483.6–491.5 m) report low saturations (25–30%), and differences in oven- dried and relative-humidity porosities of less than 5%, indicating that the ch3 interval in this borehole is vitric. The Dune Wash fault system was used to bound a portion of the eastern margin of the vitric zone, and the Solitario Canyon fault, which downdrops the region to the west by over 200 m (Figure III-4), was assumed to form the western boundary of the vitric zone for this unit, thus resulting in assigning the H-6 borehole as zeolitic (consistent with the results of RPM3.1 (DTN: MO9910MWDISMRP.002 [145731])). Ch4 (mid-lower ¼ of Tac) The vitric region is initially defined by RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) Ksat data and data from boreholes SD-6, SD-7, G-3, H-3, H-5, H-6, and WT-2. Rock-property data for SD-6 (DTN: GS980808312242.014 [106748]) within this unit (corresponding to a depth interval of 491.5–499.3 m) report low saturations (25–43%), and differences in oven-dried and relative-humidity porosities less than 5%, indicating that the ch4 interval in this borehole is vitric. The hydraulic conductivity value (2.31 × 10-5 m/s) reported for this borehole (DTN: GS98090831224.038 [107154]) is greater than 10-10 m/s, consistent with the vitric interpretation. While SD-12 lies within the vitric region as defined by RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]), rock-property data (DTN: GS960808312231.004 [108985]) for samples from this borehole within the ch4 interval (depths of 458.9–473.2 m) indicate elevated saturation values (92–100%), suggesting that this borehole lies within the zeolitic zone. The Dune Wash fault system was used to bound a portion of the eastern margin of the vitric zone, and the Solitario Canyon fault, which downdrops the region to the west by over 200 m (Figure III-4), was assumed to form the western boundary of the vitric zone for this unit, thus resulting in assigning the H-6 borehole as zeolitic (consistent with the results of RPM3.1 (DTN: MO9910MWDISMRP.002 [145731])). Ch5 (lower ¼ of Tac) The vitric region is initially defined by RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) Ksat data and data from boreholes SD-6, SD-12, G-3, H-3, and H-5. No rock-property data are available for SD-6 (DTN: GS980808312242.014 [106748]) within this unit (corresponding to a depth interval of 499.3–507.2 m). While SD-7 lies within the vitric region as defined by RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]), rock-property data (DTN: GS951108312231.009 [108984]) for samples from this borehole within the ch5 interval (depths of 465.4–477.7 m) indicate elevated saturation values (87–100%, avg. 97%) and differences in oven-dried and relative-humidity porosities typically greater than 5%, suggesting that this borehole lies within the zeolitic zone. The Dune Wash fault system was used to bound a portion of the eastern margin of the vitric zone, and the Solitario Canyon fault was assumed to form the western boundary of the vitric zone for this unit. Ch6 (Tacbt) The vitric region is defined by data from boreholes SD-6 and G-3 (no vitric region is indicated by RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]). Rock-property data for SD-6 (DTN: GS980808312242.014 [106748]) within this unit show low saturations (54–67%), and differences in oven-dried and relative-humidity porosities less than 5% in one of two samples, indicating that the ch6 interval in this borehole is vitric. The hydraulic conductivity value (1.2 × 10-9 m/s) reported for this borehole (DTN: GS98090831224.038 [107154]) is greater than 10-10 m/s, consistent with the vitric interpretation. The Solitario Canyon fault was assumed to form the western boundary of the vitric zone for this unit. 6.7 DUAL-PERMEABILITY GRID GENERATION The software program 2kgrid8.for V1.0 (LBNL 2002 [154787]) generates dual-k numerical grids for heterogeneous, fractured rocks. The 2kgrid8.for V1.0 generates a dual-k grid using (a) a primary single-continuum mesh (ECM grid) with 8-character element names, and (b) fracture properties for multiple hydrogeological units. The program is adapted from the software macro DKMgenerator V1.0 (LBNL 1999 [140702]). The 2kgrid8.for V1.0 software is designed to handle three types of fractured media: 1. A set of parallel, infinite fractures (Type #1, 1-D fracture continuum) with uniform spacing within each hydrogeological unit 2. Two sets of parallel, infinite, orthogonal fractures (Type #2, 2-D fracture continuum) with the same spacing within each hydrogeological unit 3. Three sets of parallel, infinite, orthogonal fractures (Type #3, 3-D fracture continuum) with the same spacing within each hydrogeological unit. Volumes of fracture and matrix elements are computed with 2kgrid8.for V1.0 using the following formulas: Vf =F f V (Eq. 2) n and V =(1-F )V (Eq. 3) m fn where Vf and Vm are volumes of fracture and matrix elements, respectively, for the dual-k grid, Vn is the volume of element n of the primary mesh from which a dual-k grid is being generated, and f f is the fracture porosity or fractional volume of fractures within the bulk rock. The connection information in the dual-permeability grid is determined as follows: • Global fracture-fracture and matrix-matrix connection data are kept the same as the connections in the primary mesh for the corresponding gridblocks. This implies that permeabilities used for both fracture and matrix systems are the “continuum” values for both, relative to the bulk-connecting areas. • Inner-connection distances between fractures and matrix within a primary gridblock are calculated as: Df = 0 (Eq. 4) D D= for Type #1 fractures (Eq. 5) m 6 D D= for Type #2 fractures (Eq. 6) m 8 D D= for Type #3 fractures (Eq. 7) m 10 and 1 D = (Eq. 8) F where Df is the distance from the fracture center to the surface of a matrix block; Dm is the calculated distance for flow crossing fracture/matrix interfaces, based on the quasi-steady state assumption (Warren and Root 1963 [100611], p. 247; Pruess 1983 [100605], Table 1); D is the fracture spacing; and F is the fracture frequency within the unit. The interface area (A) between fractures and matrix blocks is estimated by: A =AfmVn (Eq. 9) where Afm is a volume-area factor, which represents the total fracture-matrix interface area per unit volume of rock, determined from site fracture characterization studies. Fracture properties incorporated in the UZ Model are listed in Table 4. Only Type #1 fractures were used in the generation of dual-k numerical grids. The program 2kgrid8.for V1.0 must first be compiled using a FORTRAN compiler to create the executable file for the operating platform. Three input files are required to run 2kgrid8.for V1.0. These files are called “2kgrid.dat,” “connec.dat,” and “framtr.dat,” and contain the following information: 1. The “2kgrid.dat” file contains the two parts of ELEME and CONNE data blocks from the primary single-continuum mesh using the same formats. 2. The “connec.dat” file contains connection indexes from the primary single-continuum mesh using the same formats. 3. The “framtr.dat” file contains fracture properties (DTN: LB0205REVUZPRP.001 [159525] and DTN: LB0207REVUZPRP.001 [159526]) with the following format and data: Format (A5,5X,4(E10.3)) urock(i), volf(i), xxx, dspac(i), afm_v(i) urock(i) rock type name as rock(n) volf(i) porosity or volume fraction of fractures within bulk rock xxx aperture, not used dspac(i) fracture frequency afm_v(i) a volume-area factor, representing the total fracture-matrix area per unit volume rock, as determined from site fracture characterization studies. Execution of “2kgridv1” creates three output files: 1. The “2kgrid.out” file contains information from the primary mesh and new dual-k meshes for grid verification purposes. 2. The “eleme.dat” file contains “ELEME” data blocks for the new dual-k grid. 3. The “conne.dat” contains “CONNE” data blocks for the new dual-k grid. 6.8 GRID VERIFICATION The Scientific Analysis report presents the grids to represent the geological framework model, refined from borehole data for the unique representation of Yucca Mountain. Since there are no alternative geologic models developed (Section 4.1), no alternative grids are presented in this report. The grids are intended for use by the UZ Model for site-scale flow and transport processes. Numerical grids are fixed objects, or frameworks, that alone do not capture physical processes or phenomena occurring at Yucca Mountain. As such, the process of “model validation,” in the usual sense, does not apply. However, the process of grid “verification”—an evaluation of how accurately the numerical grid represents the geologic and hydrogeologic input—does apply, and is discussed in this section. The parameters generated for each numerical grid include gridblock material names, gridblock volumes and locations, connection lengths and interface areas between gridblocks, and direction of absolute permeability for each connection. Because of the number and size of the numerical grids developed for UZ Model activities, it is not practical to verify each parameter for each gridblock generated. Consequently, a subset of gridblocks from each mesh is taken, and the associated parameters are verified to ensure the accuracy and representativeness of the mesh. The criteria by which the numerical grids are evaluated are not as rigorous as, for example, those specified for engineering design. This is because of the simplified approximation and large uncertainty inherent in modeling studies, where variations in modeling results up to an order of magnitude may be considered acceptable. For the 1-D numerical grids (Output-DTN: LB02081DKMGRID.001), which consist of columns of gridblocks at borehole locations only, gridblock material names and elevations are verified through comparison with stratigraphic information from GFM2000 (see Attachment II for details). For the 2-D cross-sectional grids through borehole UZ-7a (Output-DTN: LB02081DKMGRID.001), gridblock material names and elevations are verified through visual comparison with stratigraphic and structural information from GFM2000 exported surface horizons (see Attachment III for details). For the 3-D UZ Model grids (Output-DTN: LB03023DKMGRID.001) for calibration and calculation of flow fields, gridblock material names and elevations are verified through comparisons at borehole locations with the GFM2000 file “contacts00el.dat” (see Attachment I, GFM2000 files) and through visual comparison with stratigraphic and structural information from GFM2000 exported surface horizons (see Attachment III for details). A spot check involving hand calculation of gridblock volumes, connection lengths, and interface areas between gridblocks showed consistency with calculated results for all UZ Model grids generated. A spot check of the direction of connectivity confirmed vertical connections for all connections within gridblock columns (except for columns associated with nonvertical faults, where the x-y locations of grid nodes can vary with depth). These spot checks are documented in the scientific notebook by Wang (2002 [159673], SN-LBNL-SCI-213-V1, p. 93). An additional test of the 3-D grid was performed through the use of a TOUGH2 V1.4 (LBNL 2000 [146496]) simulation. For this fully saturated isothermal (25°C) simulation, all gridblocks were assigned the same rock properties and an initial fluid pressure of 500 bars. Several large volume gridblocks at the base of the grid were assigned constant pressures, and the remainder of the grid was allowed to attain equilibrium pressure conditions over time. Thus, for an ideally configured grid, there should be a linear relation between gridblock elevations and steady-state pressures. Small deviations from this relation were observed for the gridblocks in inclined fault columns, where vertical connections between gridblocks deviate from 90 degrees. The shift in pressure for a given elevation for these gridblocks is a function of the relative deviation from vertical for the fault columns, with more inclined faults exhibiting a greater deviation from the predicted pressure. A more detailed discussion of this simulation is presented in Attachment III. Corroborative Studies Sensitivity studies that examine the effect of grid resolution (i.e., gridblock size) on flow and transport simulation results were documented in FY97 (Haukwa and Wu 1997 [107934]; Haukwa et al. 1997 [101243]) and FY98 (Zhang 2000 [159531], pp. 52–56, 66–72) UZ Models, and are summarized below as corroborative material for this Scientific Analysis report. FY97 UZ Model Sensitivity Study—Both coarse and refined 2-D, cross-sectional grids of the UZ at Yucca Mountain were developed by Haukwa and Wu (1997 [107934], pp. 4-12–4-13) to address concerns over the use of appropriate numerical grid resolution in UZ moisture flow modeling. The cross sections were developed along a north-south (N-S) transect through the proposed repository area, extending from borehole G-2 in the north to borehole G-3 in the south. The coarse grid used an average horizontal spacing of 50 m within the proposed repository area and 100 m outside the proposed repository area. The fine grid used a horizontal spacing as small as 6 m within the proposed repository area and as high as 50 m outside the proposed repository area. The coarse grid was comprised of 23 vertical layers; the refined grid had 61 layers (Haukwa et al. 1997 [101243], pp. 12-2–12-3). Identical layer-averaged rock properties were used in both grids. From comparison of flow simulation results using the coarse and refined grids, it was concluded by Haukwa et al. (1997 [101243], p. 12-16) that the 100 m lateral grid resolution within the proposed repository area, used in the 3-D UZ Model, was sufficient for ambient site-scale flow modeling purposes. Results indicated that moisture flow is predominantly vertical (Haukwa et al. 1997 [101243], p. 12-4), except where zeolites are present, suggesting that modeling results are less sensitive to lateral gridblock dimensions than to vertical changes in grid resolution, unless a sudden change in rock hydrogeologic properties occurs at a layer contact, resulting in significant lateral diversion. Below the proposed repository horizon, lateral diversion is most likely to occur above zeolites in the CHn. Calculated saturation and percolation flux distribution could be adequately resolved by adding a few grid layers at the PTn-TSw interface and at the vitric-zeolitic interfaces within the CHn, since these are transitional areas where rock properties change rapidly over short distances. The current (FY02) 3-D UZ Model is vertically resolved with about 57 layers in the proposed repository footprint; about 26 of these layers are above the proposed repository horizon, 5 layers are within the proposed repository horizon, and about 26 layers lie between the proposed repository horizon and the water table). The transitional areas at the PTn-TSw and vitric-zeolitic interfaces are generally captured by several thin layers. FY98 UZ Model Sensitivity Study—In this study, the influence of gridblock size on flow and transport simulation results was examined along an east-west (E-W) cross section through borehole SD-9. Four meshes, each with a different nominal gridblock size, were developed along the E-W transect (for details, refer to Zhang 2000 ([159531], pp. 52–56, 66–72)). Three simulation scenarios were considered in this study. In the first simulation scenario (Scenario #1), no modifications are made to the calibrated FY98 hydrogeologic property sets to represent perched water. In the second simulation scenario (Scenario #2), FY98 calibrated perched-water hydrogeologic properties are used. In the third simulation scenario (Scenario #3), perched-water properties are used, but fracture flow is ignored in zeolitic units (except in fault zones). Both conservative and reactive tracers are considered in the transport simulations for each of the three scenarios. Under the conditions prescribed in Scenario #1 (no perched water), the effect of gridblock size is minimal. Results from the coarsest of the four cross-sectional grids (which has a nominal horizontal spacing of 112 m and a maximum layer thickness of 60 m) compared with the results from the finest of the four cross-sectional grids (which has a nominal horizontal spacing of 28 m and a maximum layer thickness of 15 m) show an approximate 20% difference in the time at which half of the tracer mass (both conservative and reactive) reaches the water table. Under the conditions prescribed in Scenario #2 (perched water), model results for the coarsest mesh and finest mesh show differences of about 10% in the time at which half of the tracer mass reaches the water table for conservative tracers. For reactive tracers, results for the coarsest mesh differ from those for the finest mesh by a factor of two. Under the conditions prescribed in Scenario #3 (perched water, no fractures in zeolitic units), the effect of gridblock size is once again minimal. Results from the coarsest of the four cross- sectional grids compared with the results from the finest of the four cross-sectional grids show an approximate 20% difference in the time at which half of the conservative tracer mass reaches the water table, as well as an approximate 15% difference in the time at which half of the reactive tracer mass reaches the water table. The results of this FY98 modeling study suggest that the numerical grid resolution used in the FY02 site-scale UZ Model grids, at least within the proposed repository area, is appropriate for capturing important flow and transport phenomena. INTENTIONALLY LEFT BLANK 7. CONCLUSIONS Data from the GFM2000 geological model (DTN: MO0012MWDGFM02.002 [153777]) were integrated with hydrogeologic units defined by Flint (1998 [100033]) and adjusted using rock- property data contained in ISM3.1 to create integral finite-difference numerical grids for the UZ at Yucca Mountain. The layer subdivision and assignment of material properties resulted in numerical grids that are appropriate for UZ flow and transport modeling. These grids were verified for accuracy by inspection of gridblock material names, volumes, location, interface areas, and connection length and direction. The grids were also verified against known stratigraphy in reference boreholes and the GFM2000 (DTN: MO0012MWDGFM02.002 [153777]). The results show that the resulting 1-D, 2-D, and 3-D grids accurately reflect the stratigraphy and structural features of GFM2000, with contact elevations and unit thicknesses usually within 5 m of those of GFM2000. Larger deviations may occur in the vicinity of faults with large vertical offsets or with nonvertical fault slopes. Corroborative sensitivity studies show that the grids developed are valid and appropriate for UZ flow and transport modeling. The FY02 UZ Model grids incorporate closer spacing of layers (maximum of 5 m) for the PTn units (where lateral flow may occur), the proposed repository, and the unit ch1, thus allowing for adequate resolution of flow and transport phenomena within the UZ. Results from the development of numerical grids to simulate the UZ at Yucca Mountain include: • One primary mesh and one dual-k mesh consisting of 1-D columns at borehole locations (Output-DTN: LB02081DKMGRID.001) used for developing calibrated hydrogeologic property sets for the UZ at Yucca Mountain. • One primary mesh and one dual-k mesh comprising a 2-D cross section through borehole UZ-7a (Output-DTN: LB02081DKMGRID.001) used to calibrate fault hydrogeologic properties in the UZ at Yucca Mountain. • One primary mesh and one dual-k mesh (Output-DTN: LB03023DKMGRID.001) used for 3-D UZ Model calibration and to generate 3-D UZ flow fields for Performance Assessment. 7.1 LIMITATIONS AND UNCERTAINTIES The numerical grids developed in this report are intended for use in mountain-scale flow and transport modeling of the Yucca Mountain UZ system. A model of a complex system such as Yucca Mountain must be used with recognition of its limitations. For the site-scale UZ Model, a key limitation is imposed by numerical grid resolution. Since computational time rapidly increases with grid size (i.e., number of gridblocks and connections), the use of large refined grids is currently limited by both simulation time and CPU requirements. Refining an entire 3-D model with gridblocks having dimensions roughly equivalent to the expected drift spacing in the proposed repository and using comparably refined vertical resolution would increase current grid sizes by more than an order of magnitude. Thus, it is not feasible at the mountain scale to characterize flow behavior on horizontal scales less than a few tens of meters. Current lateral resolution (up to 300 m in areas outside the proposed repository boundary) can sometimes lead to high aspect ratios within very thin layers. This may lead to inaccuracies when trying to calculate lateral flow components; however, fracture spacing and orientation data suggest that groundwater flow is primarily downward, except within the altered tuffs. Previous modeling studies at Yucca Mountain have established that sufficient vertical grid resolution is critical to capturing important flow and transport processes, such as lateral flow (Wu et al. 2002 [161058]; BSC 2001 [155950], Sections 3.3.3.4.2 and 3.3.4.8.1). Wu et al. (2002 [161058]) evaluated the effect of grid refinement on percolation fluxes and noted that simulations using a vertical grid spacing of 10 m within the PTn were unable to resolve the effects of lateral flow. In contrast, the use of a more refined grid with a maximum vertical spacing of 2 m within the PTn could capture the capillary barrier effects of ptn21 and ptn23, resulting in significant lateral flow. The results of this sensitivity study were used to design the current numerical grids by employing a variable maximum vertical grid spacing with enhanced grid resolution within the PTn (See Section 6.6 for details). The impact of utilizing nonorthogonal grids on TH modeling at Yucca Mountain was evaluated by Haukwa et al. (2003 [161647]). With a nonorthogonal grid, cross-term contributions in the numerical discretization are neglected because of vertical separation of laterally connected nodes. A comparison of simulations conducted using orthogonal and nonorthogonal grids for the Yucca Mountain UZ system (where represented layers typically have dips less than 10 degrees; see Section 6.6) indicated little impact on both steady-state and transient solutions, because the cross-term connections contribute less than 6% to the total flux. As mentioned in Section 6.8 and Attachment III, the use of non-vertical columns for inclined faults does lead to some deviations in the flow behavior for the affected grid blocks. The accuracy of UZ Model grids depends largely on the accuracy of the GFM2000 (DTN: MO0012MWDGFM02.002 [153777]) and RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]) input data. Both of these models, which are assumed to provide a representative picture of subsurface geology and rock properties, are constructed with limited data resources. GFM2000 includes assumptions about the lateral continuity and thickness trends of layers at Yucca Mountain based on limited borehole data. The UZ Model numerical grids attempt to match this layered approach as closely as possible to constrain UZ flow and transport processes. While the degree of lateral continuity of layers represented in GFM2000 is a valid interpretation, the impact of more lateral discontinuity resulting from the inclusion of small faults on flow could be significant, especially in areas where little or no information has been collected. However, these areas typically lie too far from the proposed repository area to have any significant impact on repository performance. The GFM2000 bedrock surface ("s00bedrockRWC.2grd"; listed in GFM2000 files in Attachment I) was used to define the upper boundary of the UZ Model grids (see Section 6.4.2). The use of the bedrock surface thus results in the exclusion of alluvial cover from the model. In the area of the proposed repository, bedrock is typically exposed at the surface, with alluvium confined to washes and other topographic lows (BSC 2002 [159124]), Figure 13). Because alluvial cover is mostly absent above the proposed repository, any insulating effects of this material are likely to be minimal. Sensitivity studies to test the effect of alluvial cover on thermal modeling are not within the scope of this Scientific Analysis report. Within RPM3.1 (DTN: MO9910MWDISMRP.002 [145731]), the interpretation of saturated hydraulic conductivity (Ksat) distribution and mineral alteration at Yucca Mountain is also based on limited data and assumed correlations (e.g., using porosity as a surrogate for predicting Ksat). The spatial heterogeneity of low-permeability alteration products such as zeolites has a profound impact on UZ flow and transport modeling, yet the nature of their distribution is not fully understood. Though currently represented per hydrogeologic layer (i.e., UZ Model layers tsw39, ch1, ch2, ch3, ch4, ch5, and ch6), true mineral alteration and rock-property variation may not strictly follow a layered model. While a variety of geologic and rock property data were used to define vitric-zeolitic boundaries (see Sections 5.2 and 6.6.3), the location of vitric to zeolitic transitions are not consisely resolved. Grid verification exercises show that UZ Model layer thicknesses and elevations are reasonable representations of the hydrogeologic input data. Using visual cross-sectional comparisons with GFM2000 (DTN: MO0012MWDGFM02.002 [153777]), UZ Model layer contact elevations are shown to have some large (up to 50 m) differences in areas immediately adjacent to inclined fault zones, reflecting the coarse lateral grid resolution used as well as certain limitations of the gridding software. Given the large uncertainties associated with fault zone hydrogeologic characteristics, the effect of these differences along faults on modeling results has yet to be determined, but is likely limited in extent to the area immediately surrounding the fault zones. Additional hydrogeologic property data and analyses within fault zones would reduce uncertainty in this area. There are some limitations relating to the modeling of faults in the UZ Model grids. As noted earlier (Section 6.6.1), faults cannot be modeled as intersecting features. To simplify the model, subsidiary faults related to the Solitario Canyon fault (“Splay N,” “Splay G,” “Splay S,” and the Solitario [west] faults) were omitted from the UZ Model grids because of their proximity to the dipping Solitario Canyon fault (making them difficult to incorporate as separate features to the model). Faults observed within the ESF and ECRB that are not part of the GFM2000 (owing to either insufficient length or offset) are also not incorporated in the UZ Model grid. As mentioned in Section 5.2, the proposed repository design used for the UZ Model numerical grid generation (BSC 2002 [159527]) was the most recent representation of the repository layout at the time the grids were generated. It is recognized that the proposed repository layout may be subject to future design modifications, and that the most recent revision of the proposed repository layout (BSC 2003 [161726]; BSC 2003 [161727]) does not include the lower block area (See Section 4.1.1). If additional design changes are made, the numerical grids should be evaluated to ensure that sufficient grid resolution in the area of the proposed repository exists, and if this is no longer the case, new numerical grids should be generated utilizing the revised repository design. 7.1.1 Water Table Uncertainty The water table by definition forms the base of the UZ (Assumption 1). The potentiometric- surface map as defined by USGS (2001 [157611], Figure 6-1) was constrained by borehole water levels in the Yucca Mountain area (USGS 2001 [157611], Table I-2). Contours for this map were hand-drawn to conform to the borehole water levels, assuming that the measured water level in WT-24 represents the regional water table, whereas the water levels in boreholes G-2 and WT-6 represent perched conditions. The water table is well constrained in the area near the ESF where abundant borehole data exist, but is poorly constrained to the north and west, where there are very few control points and the potentiometric surface has a higher gradient. Thus, any definition of the water table elevations will inevitably include some uncertainty, especially in the areas where few borehole constraints are available. The water table is defined in the qualified DTN: GS010608312332.001 [155307] through the use of borehole locations and their associated water table elevations and potentiometric map contours. The DTN: GS010608312332.001 [155307] from USGS 2001 [157611] contains the ARCINFO files (“pot_contours.e00” and “wells.e00”). The layer “GWL_SSPAC” in the Vulcan GFM2000 Representation database (DTN: MO0110MWDGFM26.002 [160565]) was derived by digitizing the contours depicted on the potentiometric surface map included in FY 01 Supplemental Science and Performance Analyses, Volume 1: Scientific Bases and Analyses (BSC 2001 [155950], Figure 12.3.1.2-2), which appears to be identical to that presented in USGS (2001 [157611], Figure 6-1). The data defining this layer (contours and borehole coordinates) were then extracted and the resulting data set (gwl_sspac2.asc) was submitted to the TDMS as DTN: MO0212GWLSSPAX.000 [161271] (see Attachment IV for details). This representation of the water table was qualified using the procedure AP-SIII.2Q, Qualification of Unqualified Data and the Documentation of Rationale for Accepted Data and the Data Qualification Plan found in the Technical Work Plan for: Performance Assessment Unsaturated Zone (BSC 2002 [160819], Attachment III). The data qualification reviews for DTN: MO0212GWLSSPAX.000 [161271] are presented in Attachment IV. The file gwl_sspac2.asc was used as input for the generation of the UZ water table reference horizon (see Section 6.4.2 for details). Data files in DTNs: GS010608312332.001 [155307] and MO0212GWLSSPAX.000 [161271] both contain water table contours, but these contours do not uniquely define a surface from which regularly spaced water table data could be obtained. Thus, the digitized potentiometric contour data and borehole water-level data must be used to create a numerical surface to facilitate production of a regularly spaced set of water table elevations. The x, y data contained in the ARCINFO files (DTN: GS010608312332.001 [155307]) are given in Universal Transverse Mercator (UTM) coordinates, whereas those from the gwl_sspac2.asc file (DTN: MO0212GWLSSPAX.000 [161271]) are in Nevada State Plane (NSP) meters. The ARCINFO file “pot_contours.e00” was modified using a text editor so that it could be read as a “.dat” file in EARTHVISION V5.1 (Dynamic Graphics 1998 [152614]). The coordinate transformation utility of EARTHVISION V5.1 was used to convert the UTM coordinates to NSP coordinates. Elevation values were then assigned to each point on the basis of visual comparison to the potentiometric map in the report, Water-Level Data Analysis for the Saturated Zone Site- Scale Flow and Transport Model (USGS 2001 [157611], Figure 6-1). The contour line locations Development of Numerical Grids for UZ Flow and Transport Modeling U0000 ANL-NBS-HS-000015 REV01 81 April 2003 were then compared with those extracted from the Vulcan data base, and significant deviations in water table elevations between the data sets were observed in areas north (up to 60 m) and northwest (up to 30 m) of the ESF within the UZ Model grid area. These variations may result from errors associated with the digitization of the contour lines. The data from these two sources were imported into EARTHVISION V5.1 to construct gridded surfaces to permit more rigorous comparison of the data (Wang 2003 [162380], SN-LBNL-SCI- 213-V1, pp. 110–115). Borehole water table data included in the input file converted from Vulcan were appended to the modified ARCINFO file (which contained only water table contour data), with the only modification being that SD-7 was removed from the input data set, as it was considered to be unreliable in the report, Water-Level Data Analysis for the Saturated Zone Site- Scale Flow and Transport Model (USGS 2001 ([157611], Table I-2). Using the 2-D minimum tension gridding utility in EARTHVISION V5.1, the data were coarsely gridded and then finely gridded, using the same steps as outlined in Section 6.4.2 that were employed to create the water table utilized for numerical grid generation. The contoured water table surfaces created in EARTHVISION V5.1 using the two data sets are displayed in Figure 7. NSP Easting (m) NSP Northing (m) NSP Easting (m) LEGEND UZ2002 Model Boundary 2002 Repository Boundary 2002 Repository Lower Block Water Table Elevation Contours (m) Boreholes with Water Table Elevations (m) 727.6900(a)(b) DTNs: (a) GS010608312332.001 [155307] (b) MO0212GWLSSPAX.000 [161271] NOTE: 2002 Repository Lower Block will not be used in any LA calculations. Figure 7. Comparison of EARTHVISION V5.1 Gridded Potentiometric Surfaces Because neither data set uniquely defines a water table that could be used to extract a set of regularly spaced data needed for creating the numerical grids, it was necessary to create a numerically defined surface in EARTHVISION V5.1 through gridding of the borehole water table elevations and potentiometric surface contour lines. The two-step process used to create the fine-spaced grid (see Section 6.4.2) does result in small changes in the appearance of the contoured surface, such as the creation of a small ridge in the potentiometric surface where the water table is around 730 m elevation. Such features are artifacts of the irregularly spaced input data and the use of 2-D minimum tension gridding, and result in minor shifts in the water table. As mentioned in Section 6.4.2, the resulting regularly spaced data set was edited to have a minimum water table elevation of 730 m, and thus the final numerical grids generated by WINGRIDDER V2.0 (LBNL 2002 [154785]) have a lower boundary no lower than 730 m. Both contoured surfaces are consistent with the measured borehole water table elevations; however, significant differences exist between the locations of the contour lines used to define the potentiometric surface. These differences translate into significant deviations (up to 60 m) in the water table elevations in the areas to the north and west, where there is a pronounced gradient to the water table and few borehole constraints. In general, the water table elevations as indicated by DTN: MO0212GWLSSPAX.000 [161271] are higher than the corresponding elevations from DTN: GS010608312332.001 [155307], resulting in a shorter distance for radionuclide transport through the UZ. It is recommended that for future revisions of this document, a single water table be generated using the best available data and conceptual models. Such a surface could be applied to the full range (e.g., UZ modeling, SZ modeling, and repository design) of Yucca Mountain studies, and would reduce potential inconsistencies and differences between these products. 7.2 RESTRICTIONS FOR SUBSEQUENT USE The UZ Model numerical grids developed herein shall be used only for development of UZ hydrogeologic property sets, for UZ Model calibration, and for development of UZ flow fields for Performance Assessment. These activities will involve the use of software from the TOUGH2 family of codes. 7.3 TECHNICAL PRODUCT OUTPUT The technical product output files for this Scientific Analysis report have been submitted to the TDMS and are included in the following Output-DTNs: LB02081DKMGRID.001 LB0208HYDSTRAT.001 LB02092DGRDVER.001 LB03023DKMGRID.001 8. INPUTS AND REFERENCES The following is a list of the references cited in this document. Column 1 represents the unique six digit numerical identifier (the Document Input Reference System [DIRS] number), which is placed in the text following the reference callout (e.g., BSC 2002 [155950]). The purpose of these numbers is to assist the reader in locating a specific reference. Within the reference list, multiple sources by the same author (e.g., BSC 2002) are sorted alphabetically by title. 8.1 DOCUMENTS CITED 160333 Aurenhammer, F. 1991. “Voronoi Diagrams–A Survey of a Fundamental Geometric Data Structure.” ACM Computing Surveys, 23, (3), 345-405. [New York, New York]: Association for Computing Machinery. TIC: 240932. 159356 BSC (Bechtel SAIC Company) 2001. Development of Numerical Grids for UZ Flow and Transport Modeling. ANL-NBS-HS-000015 REV 00 ICN 01. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20020211.0002. 155950 BSC (Bechtel SAIC Company) 2001. FY 01 Supplemental Science and Performance Analyses, Volume 1: Scientific Bases and Analyses. TDR-MGR-MD-000007 REV 00 ICN 01. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20010801.0404; MOL.20010712.0062; MOL.20010815.0001. 159124 BSC (Bechtel SAIC Company) 2002. Geologic Framework Model (GFM2000). MDL-NBS-GS-000002 REV 01. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20020530.0078. 159527 BSC (Bechtel SAIC Company) 2002. Repository Design, Repository/PA IED Subsurface Facilities Plan Sht. 1 of 5, Sht. 2 of 5, Sht. 3 of 5, Sht. 4 of 5, and Sht. 5 of 5. DWG-MGR-MD-000003 REV A. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20020601.0194. 159530 BSC (Bechtel SAIC Company) 2002. Rock Properties Model Analysis Model Report. MDL-NBS-GS-000004 REV 00 ICN 03. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20020429.0086. 160819 BSC (Bechtel SAIC Company) 2002. Technical Work Plan for: Performance Assessment Unsaturated Zone. TWP-NBS-HS-000003 REV 02. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20030102.0108. 159051 BSC (Bechtel SAIC Company) 2002. Technical Work Plan for: Unsaturated Zone Sections of License Application Chapters 8 and 12. TWP-NBS-HS-000003 REV 01. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20020613.0192. 160146 BSC (Bechtel SAIC Company) 2002. Total System Performance Assessment-License Application Methods and Approach. TDR-WIS-PA-000006 REV 00. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20020923.0175. 161726 BSC (Bechtel SAIC Company) 2003. Repository Design, Repository/PA IED Subsurface Facilities. 800-IED-EBS0-00401-000-00B. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20030109.0145. 161727 BSC (Bechtel SAIC Company) 2003. Repository Design, Repository/PA IED Subsurface Facilities. 800-IED-EBS0-00402-000-00B. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20030109.0146. 161731 BSC (Bechtel SAIC Company) 2003. Repository Design, Repository/PA IED Subsurface Facilities. 800-IED-EBS0-00403-000-00B. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20030109.0147. 107905 Buesch, D.C. and Spengler, R.W. 1999. “Correlations of Lithostratigraphic Features with Hydrogeologic Properties, a Facies-Based Approach to Model Development in Volcanic Rocks at Yucca Mountain, Nevada.” Proceedings of Conference on Status of Geologic Research and Mapping in Death Valley National Park, Las Vegas, Nevada, April 9-11, 1999. Slate, J.L., ed. Open-File Report 99-153. Pages 62-64. Denver, Colorado: U.S. Geological Survey. TIC: 245245. 100106 Buesch, D.C.; Spengler, R.W.; Moyer, T.C.; and Geslin, J.K. 1996. Proposed Stratigraphic Nomenclature and Macroscopic Identification of Lithostratigraphic Units of the Paintbrush Group Exposed at Yucca Mountain, Nevada. Open-File Report 94-469. Denver, Colorado: U.S. Geological Survey. ACC: MOL.19970205.0061. 157916 Curry, P.M. and Loros, E.F. 2002. Project Requirements Document. TER-MGR- MD-000001 REV 00. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20020806.0027. 103371 Czarnecki, J.B.; Nelson, P.H.; O’Brien, G.M.; Sass, J.H.; Thapa, B.; Matsumoto, Y.; and Murakami, O. 1995. “Testing in Borehole USW G-2 at Yucca Mountain: The Saga Continues.” Eos, 76, (46), 191-192. Washington, D.C.: American Geophysical Union. TIC: 240933. 142594 Czarnecki, J.B.; O’Brien, G.M.; Nelson, P.H.; Sass, J.H.; Bullard, J.W.; and Flint, A.L. 1994. “Is There Perched Water Under Yucca Mountain in Borehole USW G2?” Eos (Supplement), 75, (44), 249-250. Washington, D.C.: American Geophysical Union. TIC: 226992. 102884 DOE (U.S. Department of Energy) 1995. Yucca Mountain Site Characterization Project Site Atlas 1995. Two volumes. Washington, D.C.: U.S. Department of Energy. ACC: MOL.19960311.0262. 100633 Ervin, E.M.; Luckey, R.R.; and Burkhardt, D.J. 1994. Revised Potentiometric- Surface Map, Yucca Mountain and Vicinity, Nevada. Water-Resources Investigations Report 93-4000. Denver, Colorado: U.S. Geological Survey. ACC: NNA.19930212.0018. 104367 Finsterle, S. 1999. ITOUGH2 User’s Guide. LBNL-40040. Berkeley, California: Lawrence Berkeley National Laboratory. TIC: 243018. 100033 Flint, L.E. 1998. Characterization of Hydrogeologic Units Using Matrix Properties, Yucca Mountain, Nevada. Water-Resources Investigations Report 97-4243. Denver, Colorado: U.S. Geological Survey. ACC: MOL.19980429.0512. 154365 Freeze, G.A.; Brodsky, N.S.; and Swift, P.N. 2001. The Development of Information Catalogued in REV00 of the YMP FEP Database. TDR-WIS-MD-000003 REV 00 ICN 01. Las Vegas, Nevada: Bechtel SAIC Company. ACC: MOL.20010301.0237. 107934 Haukwa, C. and Wu, Y.S. 1997. “Grid Generation and Analysis.” Chapter 4 of The Site-Scale Unsaturated-Zone Model of Yucca Mountain, Nevada, for the Viability Assessment. Bodvarsson, G.S., Bandurraga, T.M., and Wu, Y.S., eds. LBNL-40376. Berkeley, California: Lawrence Berkeley National Laboratory. ACC: MOL.19971014.0232. 101243 Haukwa, C.; Wu, Y.; and Bodvarsson, G.S. 1997. “Modeling Study of Moisture Flow Using a Refined Grid Model.” Chapter 12 of The Site-Scale Unsaturated Zone Model of Yucca Mountain, Nevada, for the Viability Assessment. Bodvarsson, G.S.; Bandurraga, T.M.; and Wu, Y.S., eds. LBNL-40376. Berkeley, California: Lawrence Berkeley National Laboratory. ACC: MOL.19971014.0232. 161647 Haukwa, C.B.; Wu, Y-S.; and Bodvarsson, G.S. 2003. “Modeling Thermal- Hydrological Response of the Unsaturated Zone at Yucca Mountain, Nevada, to Thermal Load at a Potential Repository, (In Press), Journal of Contaminant Hydrology, Vols. 62-63, April-May 2003.” Memorandum from C.B. Haukwa (LBNL), Y-S. Wu (LBNL), and G.S. Bodvarsson (LBNL), to Records Processing Center (RPC), February 5, 2003, with attachments. ACC: MOL.20030210.0259. 155955 Hinds, J. 2001. Unsaturated Zone Modeling & Synthesis. Scientific Notebook YMP-LBNL-YSW-JH-2. ACC: MOL.20010725.0216. 151762 IEEE/ASTM SI 10-1997. Standard for Use of the International System of Units (SI): The Modern Metric System. New York, New York: Institute of Electrical and Electronics Engineers. TIC: 240989. 101258 Loeven, C. 1993. A Summary and Discussion of Hydrologic Data from the Calico Hills Nonwelded Hydrogeologic Unit at Yucca Mountain, Nevada. LA-12376-MS. Los Alamos, New Mexico: Los Alamos National Laboratory. ACC: NNA.19921116.0001. 100161 Montazer, P. and Wilson, W.E. 1984. Conceptual Hydrologic Model of Flow in the Unsaturated Zone, Yucca Mountain, Nevada. Water-Resources Investigations Report 84-4345. Lakewood, Colorado: U.S. Geological Survey. ACC: NNA.19890327.0051. 103777 Moyer, T.C.; Geslin, J.K.; and Buesch, D.C. 1995. Summary of Lithologic Logging of New and Existing Boreholes at Yucca Mountain, Nevada, July 1994 to November 1994. Open-File Report 95-102. Denver, Colorado: U.S. Geological Survey. TIC: 224224. 162418 NRC (U.S. Nuclear Regulatory Commission) 2003. Yucca Mountain Review Plan, Information Only. NUREG-1804, Draft Final Revision 2. Washington, D.C.: U.S. Nuclear Regulatory Commission, Office of Nuclear Material Safety and Safeguards. TIC: 254002. 100605 Pruess, K. 1983. GMINC - A Mesh Generator for Flow Simulations in Fractured Reservoirs. LBL-15227. Berkeley, California: Lawrence Berkeley Laboratory. ACC: NNA.19910307.0134. 100413 Pruess, K. 1991. TOUGH2—A General-Purpose Numerical Simulator for Multiphase Fluid and Heat Flow. LBL-29400. Berkeley, California: Lawrence Berkeley Laboratory. ACC: NNA.19940202.0088. 100643 Rautman, C.A. and McKenna, S.A. 1997. Three-Dimensional Hydrological and Thermal Property Models of Yucca Mountain, Nevada. SAND97-1730. Albuquerque, New Mexico: Sandia National Laboratories. ACC: MOL.19980311.0317. 102097 Rousseau, J.P.; Kwicklis, E.M.; and Gillies, D.C., eds. 1999. Hydrogeology of the Unsaturated Zone, North Ramp Area of the Exploratory Studies Facility, Yucca Mountain, Nevada. Water-Resources Investigations Report 98-4050. Denver, Colorado: U.S. Geological Survey. ACC: MOL.19990419.0335. 154625 USGS (U.S. Geological Survey) 2001. Water-Level Data Analysis for the Saturated Zone Site-Scale Flow and Transport Model. ANL-NBS-HS-000034 REV 00 ICN 01. Denver, Colorado: U.S. Geological Survey. ACC: MOL.20010405.0211. 157611 USGS (U.S. Geological Survey) 2001. Water-Level Data Analysis for the Saturated Zone Site-Scale Flow and Transport Model. ANL-NBS-HS-000034 REV 01. Denver, Colorado: U.S. Geological Survey. ACC: MOL.20020209.0058. 159673 Wang, J.S. 2002. “Scientific Notebooks Referenced in AMR U0000, Development of Numerical Grids for UZ Flow and Transport Modeling, ANL-NBS-HS-000015 REV01.” Memorandum from J.S. Wang (BSC) to File, September 13, 2002, with attachments. ACC: MOL.20020917.0174. 162380 Wang, J.S. 2003. “Scientific Notebooks Referenced in Scientific Analysis Report U0000, Development of Numerical Grids for UZ Flow and Transport Modeling, ANL-NBS-HS-000015 REV01.” Memorandum from J.S. Wang (BSC) to File, March 04, 2003, with attachments. ACC: MOL.20030306.0526. 100611 Warren, J.E. and Root, P.J. 1963. “The Behavior of Naturally Fractured Reservoirs.” Society of Petroleum Engineers Journal, 3, (3), 245-255. Dallas, Texas: Society of Petroleum Engineers. TIC: 233671. 161058 Wu, Y-S.; Zhang, W.; Pan, L.; Hinds, J.; and Bodvarsson, G.S. 2002. “Modeling Capillary Barriers in Unsaturated Fractured Rock.” Water Resources Research, 38, (11), 35-1 through 35-12. [Washington, D.C.]: American Geophysical Union. TIC: 253854. 154817 YMP (Yucca Mountain Site Characterization Project) 2001. Q-List. YMP/90-55Q, Rev. 7. Las Vegas, Nevada: Yucca Mountain Site Characterization Office. ACC: MOL.20010409.0366. 159531 Zhang, W. 2000. UZ Modeling and Synthesis. Scientific Notebook YMP-LBNL- YSW-WZ-1 (SN-LBNL-SCI-115-V1). ACC: MOL.20001025.0166. SOFTWARE CITED 155383 BSC (Bechtel SAIC Company) 2001. Software Code: VULCAN. V3.5NT. PC. 10044-3.5NT-00. 157019 CRWMS M&O (Civilian Radioactive Waste Management System Management and Operating Contractor) 2000. Software Code: ARCINFO. V7.2.1. SGI Irix. 10033- 7.2.1-00. 134134 Dynamic Graphics. 1997. Software Code: Earthvision. V4.0. Silicon Graphics Indigo R4000. 30035-1 V4.0. 152614 Dynamic Graphics. 1998. Software Code: EARTHVISION. 5.1. Silicon Graphics Indigo R4000. 10174-5.1-00. 140702 LBNL (Lawrence Berkeley National Laboratory) 1999. DKMGenerator. V1.0. Unix. MOL.19990909.0315. 146496 LBNL (Lawrence Berkeley National Laboratory) 2000. Software Code: TOUGH2. V1.4. Sun Workstation and DEC/ALPHA. 10007-1.4-01. 154785 LBNL (Lawrence Berkeley National Laboratory) 2002. Software Code: WINGRIDDER. V2.0. PC. 10024-2.0-00. 154787 LBNL (Lawrence Berkeley National Laboratory) 2002. Software Routine: 2kgrid8.for. V1.0. DEC-Alpha, PC. 10503-1.0-00. 148304 USGS (U.S. Geological Survey) 2000. Software Code: ARCINFO. V7.2.1. 10033- 7.2.1-01. 8.2 CODES, STANDARDS, REGULATIONS, AND PROCEDURES 156605 10 CFR 63. Energy: Disposal of High-Level Radioactive Wastes in a Geologic Repository at Yucca Mountain, Nevada. Readily available. AP-2.21Q, Rev. 1, ICN 0, BSCN 001. Quality Determinations and Planning for Scientific, Engineering, and Regulatory Compliance Activities. Washington, D.C.: U.S. Department of Energy, Office of Civilian Radioactive Waste Management. ACC: MOL.20010212.0018. AP-2.22Q, Rev. 0, ICN 0. Classification Criteria and Maintenance of the Monitored Geologic Repository Q-List. Washington, D.C.: U.S. Department of Energy, Office of Civilian Radioactive Waste Management. ACC: MOL.20020314.0046. AP-2.27Q, Rev. 0, ICN 0. Planning for Science Activities. Washington, D.C.: U.S. Department of Energy, Office of Civilian Radioactive Waste Management. ACC: MOL.20020701.0184. AP-SI.1Q, Rev. 4, ICN 0. Software Management. Washington, D.C.: U.S. Department of Energy, Office of Civilian Radioactive Waste Management. ACC: MOL.20030113.0149. AP-SIII.2Q, Rev. 1, ICN 0. Qualification of Unqualified Data and the Documentation of Rationale for Accepted Data. Washington, D.C.: U.S. Department of Energy, Office of Civilian Radioactive Waste Management. ACC: MOL.20021105.0164. AP-SIII.9Q, Rev. 0, ICN 1. Scientific Analyses. Washington, D.C.: U.S. Department of Energy, Office of Civilian Radioactive Waste Management. ACC: MOL.20020404.0083. AP-SV.1Q, Rev. 0, ICN 3. Control of the Electronic Management of Information. Washington, D.C.: U.S. Department of Energy, Office of Civilian Radioactive Waste Management. ACC: MOL.20020917.0133. YMP-LBNL-QIP-SV.0 Rev. 2, Mod. 1. Management of YMP-LBNL Electronic Data. Berkeley, California: Lawrence Berkeley National Laboratory. ACC: MOL.20020717.0323. 8.3 SOURCE DATA, LISTED BY DATA TRACKING NUMBER 149947 GS000508312332.001. Water-Level Data Analysis for the Saturated Zone Site-Scale Flow and Transport Model. Submittal date: 06/01/2000. 155307 GS010608312332.001. Potentiometric-Surface Map, Assuming Perched Conditions North of Yucca Mountain, in the Saturated Site-Scale Model. Submittal date: 06/19/2001. 108984 GS951108312231.009. Physical Properties, Water Content, and Water Potential for Borehole USW SD-7. Submittal date: 09/26/1995. 108985 GS960808312231.004. Physical Properties, Water Content and Water Potential for Samples from Lower Depths in Boreholes USW SD- 7 and USW SD-12. Submittal date: 08/30/1996. 106748 GS980808312242.014. Physical Properties of Borehole Core Samples and Water Potential Measurements Using the Filter Paper Technique for Borehole Samples from USW SD-6. Submittal date: 08/11/1998. 107154 GS980908312242.038. Physical Properties and Saturated Hydraulic Conductivity Measurements of Lexan-Sealed Samples from USW SD-6. Submittal date: 09/22/1998. 159525 LB0205REVUZPRP.001. Fracture Properties for UZ Model Layers Developed from Field Data. Submittal date: 05/14/2002. 159526 LB0207REVUZPRP.001. Revised UZ Fault Zone Fracture Properties. Submittal date: 07/03/2002. 159672 LB0207REVUZPRP.002. Matrix Properties for UZ Model Layers Developed from Field and Laboratory Data. Submittal date: 07/15/2002. 153777 MO0012MWDGFM02.002. Geologic Framework Model (GFM2000). Submittal date: 12/18/2000. 155631 MO0106RIB00038.001. Water-Level Data and the Potentiometric Surface. Submittal date: 06/22/2001. 155989 MO0109HYMXPROP.001. Matrix Hydrologic Properties Data. Submittal date: 09/17/2001. 160565 MO0110MWDGFM26.002. Vulcan GFM2000 Representation. Submittal date: 10/18/2001. 161271 MO0212GWLSSPAX.000. ASCII File, Extracted from DTN: MO0110MWDGFM26.002, Which Includes 1) Contours Digitized from DTN: GS010608312332.001 and 2) Water Levels from DTNS: MO0106RIB00038.001 and GS010608312332.001. Submittal date: 12/23/2002. 161496 MO0301SEPFEPS1.000. LA FEP List. Submittal date: 01/21/2003. 103769 MO9901MWDGFM31.000. Geologic Framework Model. Submittal date: 01/06/1999. 103771 MO9901MWDISMRP.000. ISM3.0 Rock Properties Models. Submittal date: 01/22/1999. 119199 MO9910MWDISMMM.003. ISM3.1 Mineralogic Models. Submittal date: 10/01/1999. 145731 MO9910MWDISMRP.002. ISM3.1 Rock Properties Models. Submittal date: 10/06/1999. 159524 SN0112T0501399.004. Three-Dimensional Rock Property Models (RPM2000). Submittal date: 12/04/2001. 8.4 OUTPUT DATA, LISTED BY DATA TRACKING NUMBER LB02081DKMGRID.001. 2002 UZ 1-D and 2-D Calibration Grids. Submittal date: 08/26/2002. LB03023DKMGRID.001. UZ 3-D Site Scale Model Grids. Submittal date: 02/26/2003. LB0208HYDSTRAT.001. 2002 UZ Model Grid Components. Submittal date: 08/26/2002. LB02092DGRDVER.001. Files for 2D Grid Verification. Submittal date: 09/30/2002. 9. ATTACHMENTS Attachment I—Electronic GFM2000, ISM3.1, RPM2000, and Rock- and Fracture-Property Data Files Used to Develop UZ Model Numerical Grids Attachment II—Development of Numerical Grids for 1-D Hydrogeologic-Property-Set Inversions Attachment III—Grid Verification Attachment IV—Qualification of Water Table Contour Data INTENTIONALLY LEFT BLANK ATTACHMENT I ELECTRONIC GFM2000, ISM3.1, RPM2000, AND ROCK- AND FRACTUREPROPERTY DATA FILES USED TO DEVELOP UZ MODEL NUMERICAL GRIDS GFM2000 Files: (DTN: MO0012MWDGFM02.002 [153777]) Isochores: Faults: ia00cLDRWC.2grd f00bowex.dat ia00cpv3RWC.2grd f00solEX.dat ia00cpv2RWC.2grd f00solwestEX.dat ia00cpv1RWC.2grd f00soljfatEX.dat ia00bt4RWC.2grd f00splaygEX.dat ia00tpyRWC.2grd f00splaynEX.dat ia00bt3RWC.2grd f00splaysEX.dat ia00tppRWC.2grd f00sundanceEX.dat ia00bt2RWC.2grd f00toeex.dat ia00trv3RWC.2grd f00severEX.dat ia00trv2RWC.2grd f00paganyEX.dat ia00trv1RWC.2grd f00drillEX.dat ia00trnRWC.2grd f00ghostEX.dat ia00trltfRWC.2grd f00ghostwEX.dat ia00tpulRWC.2grd f00duneEX.dat ia00tpmnRWC.2grd f00dunexEX.dat ia00tpllRWC.2grd f00dunew1EX.dat ia00tplnRWC.2grd f00imbex.dat ia00tpv3RWC.2grd f00exileEX.dat ia00tpv2RWC.2grd ia00tpv1RWC.2grd ia00bt1RWC.2grd ia00tacRWC.2grd ia00tacbtRWC.2grd ia00prowuvRWC.2grd ia00prowucRWC.2grd ia00prowmdRWC.2grd ia00prowlcRWC.2grd ia00prowlvRWC.2grd ia00prowbtRWC.2grd ia00bulluvRWC.2grd ia00bullucRWC.2grd ia00bullmdRWC.2grd ia00bulllcRWC.2grd ia00bulllvRWC.2grd Surface Horizons: ia00bullbtRWC.2grd s00bedrockRWC.2grd ia00tramuvRWC.2grd s00TpcpEXuncut.2grd ia00tramucRWC.2grd s00Tptpv3EXuncut.2grd ia00trammdRWC.2grd ia00tramlcRWC.2grd ia00tramlvRWC.2grd Other: ia00trambtRWC.2grd boreholepaths.dat contacts00el.dat ANL-NBS-HS-000015 REV01 Attachment I-1 April 2003 ISM3.1 Files: mineralsM.pdat* (DTN: MO9910MWDISMMM.003 [119199]) CHnKsatEtype.out (DTN: MO9910MWDISMRP.002 [145731]) CHnZksStrat.3grd (DTN: MO9910MWDISMRP.002 [145731]) ISM31.seq (DTN: MO9910MWDISMRP.002 [145731]) *Data considered for corroborative purposes. RPM2000 Files: CHn-hmap_etype.out* (DTN: SN0112T0501399.004 [159524]) Rock and Fracture Property Data General borehole rock property data (DTN: LB0207REVUZPRP.002 [159672]) Rock fracture property data (DTN: LB0205REVUZPRP.001 [159525]) Fault fracture property data (DTN: LB0207REVUZPRP.001 [159526]) Specific Borehole Rock Property Data Borehole DTN and Q-status Description SD-6 GS980808312242.014 [106748] qualified saturation, porosity SD-6 GS980908312242.038 [107154] qualified hydraulic conductivity SD-7 GS951108312231.009 [108984] qualified saturation, porosity SD-12 GS960808312231.004 [108985] qualified saturation, porosity ATTACHMENT II DEVELOPMENT OF NUMERICAL GRIDS FOR 1-D HYDROGEOLOGIC- PROPERTY-SET INVERSIONS UZ Model numerical grids developed for the FY02 1-D hydrogeologic-property-set inversions are comprised of numerous 1-D columns centered at borehole coordinates, or in the case of boreholes closer than 80 m to each other, the midpoint location between the two boreholes (Wang 2003 [162380], SN-LBNL-SCI-213-V1, p. 71). Layer subdivision within these 1-D columns is based on a combination of borehole stratigraphic picks identified in the GFM2000 file “contacts00el.dat” (DTN: MO0012MWDGFM02.002 [153777]) and HGU boundaries defined by Flint (1998 [100033]). The mesh files identified by Output-DTN: LB02081DKMGRID.001 and created for use in 1-D hydrogeologic property set inversions and calibration for the UZ Model include: • The primary effective-continuum model (ECM) mesh, “Boreholes.mesh” • The ECM mesh “Boreholes_NF.mesh” with rock (rather than fault) matrix properties used for fault grid nodes, in turn used for generation of the dual-k mesh • The dual-k mesh “mesh_1d.dkm” for transient (pneumatic) and steady-state simulations based on the “Boreholes_NF.mesh” file and the fracture values given in Table 4. The detailed steps describing the generation of these files are documented in scientific notebooks (Wang 2002 [159673], SN-LBNL-SCI-103-V1, pp. 134–140, 145–151; SN-LBNL-SCI-199-V1, pp. 85–91; Wang 2003 [162380], SN-LBNL-SCI-103-V1, p. 139; SN-LBNL-SCI-199-V1, pp. 86, 88–89). Table II-1 summarizes the layer contact elevation input to the 1-D inversion grids based on the GFM2000 (DTN: MO0012MWDGFM02.002 [153777]) file “contacts00el.dat.” Note that the GFM2000 borehole elevations, which have been converted from feet to meters, are also adjusted in the same manner as described in Section 6.4.1 of this Scientific Analysis report to correspond with Flint’s (1998 [100033]) HGUs. The corresponding elevations for each of these HGU contacts as determined from the UZ Model grid file “Boreholes.mck”, is also given to provide a means of verifying the accuracy of the UZ Model results (Wang 2002 [159673], SN-LBNL-SCI- 213-V1, pp. 67–69; Wang 2003 [162380], SN-LBNL-SCI-103-V1, p. 67). A total of 45 borehole locations were cross-checked. Note that in most cases, the differences in contact elevations are less than 5 m. There are several cases where deviations exceed this amount. A number of boreholes (e.g., UZ-7a, H-6, NRG#7, UZ#4/5) had greater than 5 m discrepancies for the elevation of the uppermost unit present. These differences (primarily at the bedrock surface) arise from channel erosion that produces surfaces with large local variations in slope and elevation. Although the nearest GFM2000 data point may be only meters to a few tens ANL-NBS-HS-000015 REV01 Attachment II-1 April 2003 of meters away, the highly variable surface elevations may result in the observed mismatches in the upper contact surfaces. These differences are restricted to the upper unit only, and thus should not have a significant impact on UZ Model flow and transport modeling results. Two boreholes (b#1 and N11) exhibit poor matches for most of the contact elevations, with an abrupt shift in elevations occurring below a given unit contact. Both of these boreholes are near faults, and differences in how faults were modeled in GFM2000 and the UZ Model grids may explain these discrepancies. In the case of N11, where there is a difference of over 50 m in most of the contact elevations, the borehole lies on the west side of the Solitario Canyon fault in the GFM2000 representation, but is situated on the east side of this fault in the UZ Model grid. The difference in contact elevations is similar to the observed vertical offset on the fault. The N11 borehole is located ~2 km north of the proposed repository footprint (Figure 1b), and thus this discrepancy should have little impact on UZ flow and transport models for the proposed repository area. Because of the observed differences between GFM2000 (DTN: MO0012MWDGFM02.002 [153777]) and UZ Model grid contact elevations, the b#1 and N11 boreholes were not used for 1-D rock property calibration calculations. Table II-1. Comparison of Borehole Layer Contact Elevations from GFM2000 and UZ Model Grid UZ Model GFM2000 HGU UE25#b1/a#1 USW G-1 USW G-2 USW G-4 USW H-1 Unit Unit GFM20001 UZGrid GFM2000 UZGrid GFM2000 UZGrid GFM2000 UZGrid GFM2000 UZGrid tcw11 Tpcr CCR, CUC 1553.87 1553.712 tcw12 Tpcp CUL, CW 1,153.4 1,156.1 1512.6 1512.659 1260.958 1262.655 1293.632 1293.934 tcw13 Tpcpv3,2 CMW 1,146.0 absent 1485.29 absent 1234.135 1234.133 1284.732 1284.968 ptn21 Tpcpv1 CNW 1,145.4 1134.212 1484.376 1483.961 1230.478 1230.418 1281.989 1282.185 ptn22 Tpbt4+upper Tpy BT4 1,143.3 absent 1309.2 1309.45 1482.242 1481.804 1227.125 1227.101 1275.893 1276.083 ptn23 mid Tpy TPY absent absent 1304.299 1303.468 1469.39 1468.958 absent absent 1267.663 1267.865 ptn24 lower Tpy+Tpbt3 BT3 1,142.4 1131.843 1301.299 1300.468 1459.586 1459.17 1224.747 1224.72 1260.958 1261.173 ptn25 Tpp TPP 1,138.7 1127.739 1286.34 1285.567 1403.238 1402.876 1218.834 1218.847 1245.413 1245.609 ptn26 Tpbt2+Tptrv3,2 BT2 1,126.8 1114.494 1255.86 1255.032 1331.123 1330.732 1209.477 1209.484 1218.286 1218.412 tsw31 Tptrv1 TC 1,117.1 absent 1245.192 1244.388 1320.15 1319.759 1197.254 1197.259 1202.741 1202.952 tsw32 Tptrn TR 1,115.1 1102.955 1243.192 1242.388 1318.15 1317.759 1195.254 1195.259 1200.741 1200.952 tsw33 Tptrl+Tpul TUL 1,075.0 1061.772 1193.986 1193.215 1276.777 1276.417 1148.06 1148.069 1149.401 1149.597 tsw34 Tptpmn TMN 993.5 981.825 1110.044 1109.192 1174.09 1174.063 1064.666 1064.668 1063.142 1065.354 tsw35 Tptpll TLL 967.7 955.920 1079.137 1078.217 1163.726 1163.586 1034.186 1034.197 1029.919 1030.21 tsw36 upper Tptpln TM2 856.5 844.713 961.9718 961.1646 1064.971 1064.956 926.3177 926.425 899.7696 900.3949 tsw37 lower Tptpln TM1 825.4 816.428 944.1309 943.1175 1058.916 1058.846 887.9942 888.0687 882.2944 882.9135 tsw38 Tptpv3 PV3 809.9 802.286 935.2104 934.094 1055.888 1055.791 868.8324 868.8906 873.5568 874.1729 tsw39 Tptpv2 PV2 793.7 785.864 918.3245 917.0564 1044.854 1044.759 860.0237 860.0588 855.4212 856.0834 ch1 Tptpv1+Tpbt1 BT1, BT1a 788.8 780.979 912.8076 911.5274 1040.435 1040.345 857.5243 857.5441 850.331 850.9791 ch2 upper 1/4 Tac CH 778.8 771.669 892.9956 892.0233 1018.337 1018.324 840.5165 840.5243 844.6008 845.1284 ch3 mid 1/4 Tac CH 743.7 735.789 869.305 868.4537 955.8757 955.8505 817.9613 817.9547 821.9694 822.432 ch4 mid 1/4 Tac CH 845.6144 844.8841 795.4061 795.3852 799.338 799.7356 ch5 lower 1/4 Tac CH 821.9239 821.3145 772.8509 772.8157 776.7066 777.0392 ch6 Tacbt BT 798.2333 797.7448 750.2957 750.2462 754.0752 754.3427 pp4 Prowuv PP4 779.1528 778.724 732.8306 732.8007 736.092 736.3386 pp3 Prowuc PP3 759.798 759.4007 pp2 Prowmd+Prow PP2 pp1 Prowlv+Prowbt+Bulluv PP1 bf3 Bulluc+Bullmd+Bulllc BF3 bf2 Bulllv+Bullbt+Tramuv BF2 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777]; Output-DTN: LB02081DKMGRID.001 NOTE: 1 GFM2000 data for b#1. A subset of these boreholes was used in 1-D Property Set Inversions. Depths given in meters. Table II-1. Comparison of Borehole Layer Contact Elevations from GFM2000 and UZ Model Grid (cont.) UZ Model GFM2000 HGU USW H-3 USW H-4 USW H-5 USW H-6 UE#25 NRG#4 Unit Unit GFM2000 UZGrid GFM2000 UZGrid GFM2000 UZGrid GFM2000 UZGrid GFM2000 UZGrid tcw11 Tpcr CCR, CUC 1483.467 1482.424 1478.89 1477.951 1292.962 1303.709 1249.988 1249.887 tcw12 Tpcp CUL, CW 1466.207 1465.925 1248.766 1248.831 1445.3 1444.681 1244.194 1244.537 1246.108 1245.239 tcw13 Tpcpv3,2 CMW 1370.747 1371.286 1195.761 1196.322 1355.75 1355.991 1241.146 1241.459 1153.058 1152.775 ptn21 Tpcpv1 CNW 1365.199 1365.668 1192.378 1192.874 1350.874 1351.079 1222.858 1223.221 1151.534 1151.223 ptn22 Tpbt4+upper Tpy BT4 1361.542 1361.983 1189.939 1190.416 1345.54 1345.763 absent absent 1146.962 1146.662 ptn23 mid Tpy TPY absent absent absent absent absent absent 1218.286 absent absent absent ptn24 lower Tpy+Tpbt3 BT3 1360.353 1360.77 1188.415 1188.841 1342.492 1340.3 1217.371 1217.639 1142.086 1141.823 ptn25 Tpp TPP 1356.36 absent 1182.929 1183.442 1335.329 1335.52 1213.714 1214.006 1135.685 1135.431 ptn26 Tpbt2+Tptrv3,2 BT2 1356.36 1356.737 1180.49 1181.028 1323.442 1323.615 1201.522 1201.762 1110.386 1110.143 tsw31 Tptrv1 TC 1347.826 1348.126 1172.261 1172.831 1307.592 1307.783 1199.522 1199.762 1102.157 1101.87 tsw32 Tptrn TR 1345.826 1346.126 1170.261 1170.831 1305.592 1305.783 1177.442 1177.645 1100.157 1099.87 tsw33 Tptrl+Tpul TUL 1322.862 1323.116 1134.161 1134.673 1265.53 1265.736 1103.071 1103.359 1048.664 1048.393 tsw34 Tptpmn TMN 1276.167 1276.471 1073.201 1073.696 1177.747 1178.021 1059.79 1060.018 tsw35 Tptpll TLL 1224.961 1225.293 1034.491 1035.049 1147.267 1147.494 967.74 968.0029 tsw36 upper Tptpln TM2 1163.452 1163.756 947.928 948.4726 1036.93 1037.192 944.1688 944.4087 tsw37 lower Tptpln TM1 1134.171 1134.473 907.6944 908.251 1010.107 1010.321 932.3832 932.6117 tsw38 Tptpv3 PV3 1119.53 1119.832 887.5776 888.1402 996.696 996.8859 902.8176 903.0982 tsw39 Tptpv2 PV2 1084.783 1085.145 880.2624 880.734 973.2264 973.4256 899.16 899.4235 ch1 Tptpv1+Tpbt1 BT1, BT1a 1074.725 1075.087 868.68 869.2362 969.264 969.4739 888.7968 889.0499 ch2 upper 1/4 Tac CH 1056.742 1057.069 847.344 848.0614 959.2056 959.3875 881.0244 881.2959 ch3 mid 1/4 Tac CH 1053.922 1054.238 827.913 828.6188 945.8782 946.072 873.252 873.542 ch4 mid 1/4 Tac CH 1051.103 1051.407 808.482 809.1762 932.5508 932.7565 865.4796 865.788 ch5 lower 1/4 Tac CH 1048.283 1048.575 789.051 789.7336 919.2235 919.441 857.7072 858.0341 ch6 Tacbt BT 1045.464 1045.744 769.62 770.2909 905.8961 906.1255 842.4672 842.7937 pp4 Prowuv PP4 1027.786 1028.084 752.8865 753.6395 886.0841 886.3395 828.1416 828.495 pp3 Prowuc PP3 1020.775 1021.054 742.188 743.0525 879.348 879.5576 813.816 814.126 pp2 Prowmd+Prow PP2 983.5896 983.8958 843.3816 843.6558 788.5176 788.8992 pp1 Prowlv+Prowbt+Bulluv PP1 964.692 964.9361 829.6656 829.9347 bf3 Bulluc+Bullmd+Bulllc BF3 897.636 897.8414 bf2 Bulllv+Bullbt+Tramuv BF2 752.856 753.1556 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777]; Output-DTN: LB02081DKMGRID.001 NOTE: A subset of these boreholes was used in 1-D Property Set Inversions. Depths given in meters. Table II-1. Comparison of Borehole Layer Contact Elevations from GFM2000 and UZ Model Grid (cont.) UZ Model GFM2000 HGU UE#25 NRG#5 UE#25 NRG-6 UE#25 NRG-7a USW SD-6 USW SD-7 Unit Unit GFM2000 UZGrid GFM2000 UZGrid GFM2000 UZGrid GFM2000 UZGrid GFM2000 UZGrid tcw11 Tpcr CCR, CUC 1495.349 1495.412 tcw12 Tpcp CUL, CW 1241.007 1239.184 1277.722 1283.19 1277.569 1291.754 1472.299 1472.224 1363.98 1362.601 tcw13 Tpcpv3,2 CMW 1206.307 1206.628 1261.659 1261.554 1239.347 1234.845 1368.979 1369.086 1271.016 1271.179 ptn21 Tpcpv1 CNW 1201.278 1201.715 1258.763 1258.603 1233.221 1229.901 1364.59 1364.668 1267.663 1267.791 ptn22 Tpbt4+upper Tpy BT4 1199.205 1199.598 1251.814 1251.722 1229.045 1225.738 1360.505 1360.602 1264.676 1264.795 ptn23 mid Tpy TPY absent absent 1245.433 1244.768 absent absent absent absent absent absent ptn24 lower Tpy+Tpbt3 BT3 1197.925 1198.265 1240.394 1240.869 1227.247 1224.024 1356.451 1356.551 1263.213 1263.303 ptn25 Tpp TPP 1194.237 1194.494 1230.478 1230.335 1223.802 absent 1349.045 1349.14 1259.434 1259.514 ptn26 Tpbt2+Tptrv3,2 BT2 1180.247 1180.591 1204.021 1203.916 1222.675 1220.493 1346.363 1346.453 1255.471 1255.592 tsw31 Tptrv1 TC 1168.359 1168.713 1192.621 1192.461 1213.531 absent 1335.115 1335.18 1246.236 1246.327 tsw32 Tptrn TR 1166.359 1166.713 1190.621 1190.461 1211.531 1209.552 1333.115 1333.18 1244.236 1244.327 tsw33 Tptrl+Tpul TUL 1116.787 1117.237 1137.148 1136.969 1174.151 1172.198 1302.715 1302.787 1217.676 1217.75 tsw34 Tptpmn TMN 1030.224 1030.781 1057.351 1057.121 1235.354 1235.444 1155.954 1156.091 tsw35 Tptpll TLL 1000.658 1002.85 1015.411 1015.292 1192.073 1192.177 1119.134 1119.219 tsw36 upper Tptpln TM2 904.0368 903.9169 1097.585 1097.692 1053.084 1053.035 tsw37 lower Tptpln TM1 869.127 869.2099 1066.902 1067.005 1020.166 1020.158 tsw38 Tptpv3 PV3 851.6722 851.8564 1051.56 1051.661 1003.706 1003.719 tsw39 Tptpv2 PV2 838.8096 838.919 1037.234 1037.334 972.312 972.4217 ch1 Tptpv1+Tpbt1 BT1, BT1a 833.4451 833.5818 1032.053 1032.144 965.3016 965.3709 ch2 upper 1/4 Tac CH 826.3128 826.387 1019.556 1019.646 935.5531 935.7576 ch3 mid 1/4 Tac CH 1011.707 1011.792 923.2392 923.4592 ch4 mid 1/4 Tac CH 1003.859 1003.938 910.9253 911.1608 ch5 lower 1/4 Tac CH 996.0102 996.0834 898.6114 898.8624 ch6 Tacbt BT 988.1616 988.2292 886.2974 886.564 pp4 Prowuv PP4 972.6168 972.6871 869.7468 869.9841 pp3 Prowuc PP3 965.0273 965.0948 862.1268 862.45 pp2 Prowmd+Prow PP2 924.7632 924.841 826.008 826.324 pp1 Prowlv+Prowbt+Bulluv PP1 913.7904 913.8426 793.3944 793.756 bf3 Bulluc+Bullmd+Bulllc BF3 848.4413 848.5061 bf2 Bulllv+Bullbt+Tramuv BF2 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777]; Output-DTN: LB02081DKMGRID.001 NOTE: A subset of these boreholes was used in 1-D Property Set Inversions. Depths given in meters. Table II-1. Comparison of Borehole Layer Contact Elevations from GFM2000 and UZ Model Grid (cont.) UZ Model GFM2000 HGU USW SD-9 USW SD-12 UE#25 UZ#4/5 UE#25 UZ-6 USW UZ-1/14 Unit Unit GFM2000 UZGrid GFM2000 UZGrid GFM20002 UZGrid GFM2000 UZGrid GFM20003 UZGrid tcw11 Tpcr CCR, CUC 1501.446 1500.024 tcw12 Tpcp CUL, CW 1303.02 1301.601 1323.746 1319.003 1189.33 1194.513 1480.806 1480.864 tcw13 Tpcpv3,2 CMW 1285.585 1285.393 1250.747 1250.691 1179.454 1178.984 1384.706 1382.479 ptn21 Tpcpv1 CNW 1279.703 1279.489 1245.718 1245.633 1177.442 1176.162 1372.819 1372.966 ptn22 Tpbt4+upper Tpy BT4 1275.131 1274.913 1243.371 absent 1171.042 1170.298 1369.619 1369.766 1339.6 1338.763 ptn23 mid Tpy TPY 1268.447 1268.24 absent absent 1163.474 1163.726 absent absent 1334.733 1332.634 ptn24 lower Tpy+Tpbt3 BT3 1265.447 1265.24 1242.67 1242.548 1160.474 1160.726 1368.186 1368.322 1332.733 1330.592 ptn25 Tpp TPP 1255.624 1255.428 1238.921 1238.791 1148.212 1148.619 1364.254 1364.416 1320.58 1317.966 ptn26 Tpbt2+Tptrv3,2 BT2 1233.952 1233.767 1234.989 1234.878 1108.253 1108.838 1362.608 1362.807 1278.427 1276.032 tsw31 Tptrv1 TC 1221.181 1221 1224.839 1224.731 1096.061 1096.891 1352.398 1352.604 1265.595 1263.237 tsw32 Tptrn TR 1219.181 1219 1222.839 1222.731 1094.061 1094.891 1350.398 1350.604 1263.595 1261.237 tsw33 Tptrl+Tpul TUL 1165.86 1165.689 1190.732 1190.622 1326.185 1326.38 1220.637 1217.538 tsw34 Tptpmn TMN 1080.516 1080.378 1121.451 1121.398 1264.31 1264.554 1133.769 1131.36 tsw35 Tptpll TLL 1045.22 1045.081 1083.899 1083.817 1221.943 1222.135 1099.326 1096.919 tsw36 upper Tptpln TM2 942.7464 942.5479 998.982 998.8207 1138.733 1139.025 1004.838 1001.555 tsw37 lower Tptpln TM1 906.9832 906.7829 955.7817 955.6774 1109.675 1109.94 976.1666 973.3775 tsw38 Tptpv3 PV3 889.1016 888.9004 934.1815 934.1058 1095.146 1095.398 961.8309 959.2885 tsw39 Tptpv2 PV2 870.6917 870.5053 925.068 925.0046 1081.126 1081.264 937.7822 935.5172 ch1 Tptpv1+Tpbt1 BT1, BT1a 868.4666 868.2757 916.0764 915.9899 1068.019 1068.212 930.1622 927.9533 ch2 upper 1/4 Tac CH 851.9465 851.7716 893.5212 893.4719 1056.437 1056.566 918.8236 916.2651 ch3 mid 1/4 Tac CH 830.2676 830.1 879.1956 879.1375 1049.792 1049.948 897.96 895.2167 ch4 mid 1/4 Tac CH 808.5887 808.4284 864.87 864.8032 1043.148 1043.329 877.0965 874.1683 ch5 lower 1/4 Tac CH 786.9098 786.7568 850.5444 850.4688 1036.503 1036.711 856.2329 853.1199 ch6 Tacbt BT 765.2309 765.0852 836.2188 836.1344 1029.858 1030.093 835.3694 832.0715 pp4 Prowuv PP4 748.0706 747.9223 821.3141 821.1733 1016.203 1016.409 818.2396 814.8519 pp3 Prowuc PP3 733.4402 733.2986 812.5968 812.5241 1009.498 1009.693 798.4581 795.0506 pp2 Prowmd+Prow PP2 779.0688 778.9805 968.0448 968.2735 787.8206 784.2032 pp1 Prowlv+Prowbt+Bulluv PP1 755.2944 755.2224 943.9656 944.2452 bf3 Bulluc+Bullmd+Bulllc BF3 bf2 Bulllv+Bullbt+Tramuv BF2 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777]; Output-DTN: LB02081DKMGRID.001 NOTE: A subset of these boreholes was used in 1-D Property Set Inversions. Depths given in meters 2 GFM2000 data for UZ#4 3 GFM2000 data for UZ-14 Table II-1. Comparison of Borehole Layer Contact Elevations from GFM2000 and UZ Model Grid (cont.) UZ Model GFM2000 HGU UE-25 UZ#16 USW UZ-N11 USW UZ-N31/32 USW UZ-N33 USW UZ-N37 Unit Unit GFM2000 UZGrid GFM2000 UZGrid GFM20004 UZGrid GFM2000 UZGrid GFM2000 UZGrid tcw11 Tpcr CCR, CUC 1591.754 1590.4 tcw12 Tpcp CUL, CW 1207.709 1206.519 1589.294 1590.4 1267.358 1268.672 1316.096 1317.546 1245.931 1249.623 tcw13 Tpcpv3,2 CMW 1176.894 1176.393 1584.594 1527.931 1238.098 1238.968 1316.096 1315.646 1223.65 1224.214 ptn21 Tpcpv1 CNW 1173.175 1172.661 1583.223 1526.528 1234.592 1234.821 1313.2 1312.708 1220.084 1220.49 ptn22 Tpbt4+upper Tpy BT4 1170.828 1170.255 1578.132 1524.853 1232.886 1233.371 1306.617 1306.123 1218.072 1218.335 ptn23 mid Tpy TPY absent absent 1573.804 1510.692 absent absent 1305.672 1299.612 ptn24 lower Tpy+Tpbt3 BT3 1166.957 1166.424 1231.392 1231.866 ptn25 Tpp TPP absent absent 1227.734 1228.838 ptn26 Tpbt2+Tptrv3,2 BT2 1162.263 1161.761 1219.048 1220.769 tsw31 Tptrv1 TC 1149.888 1149.557 1206.581 1209.008 tsw32 Tptrn TR 1147.888 1147.557 1205.667 1207.008 tsw33 Tptrl+Tpul TUL 1110.752 1110.504 tsw34 Tptpmn TMN 1053.694 1053.462 tsw35 Tptpll TLL 1015.898 1015.648 tsw36 upper Tptpln TM2 934.8216 934.4863 tsw37 lower Tptpln TM1 899.7696 899.3915 tsw38 Tptpv3 PV3 882.2436 881.8441 tsw39 Tptpv2 PV2 864.6566 864.3239 ch1 Tptpv1+Tpbt1 BT1, BT1a 860.7552 860.3443 ch2 upper 1/4 Tac CH 854.964 854.3603 ch3 mid 1/4 Tac CH 835.2739 834.6529 ch4 mid 1/4 Tac CH 815.5838 814.9455 ch5 lower 1/4 Tac CH 795.8938 795.2381 ch6 Tacbt BT 776.2037 775.5308 pp4 Prowuv PP4 767.1816 766.3932 pp3 Prowuc PP3 763.3106 762.4906 pp2 Prowmd+Prow PP2 740.9688 740.0902 pp1 Prowlv+Prowbt+Bulluv PP1 bf3 Bulluc+Bullmd+Bulllc BF3 bf2 Bulllv+Bullbt+Tramuv BF2 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777]; Output-DTN: LB02081DKMGRID.001 NOTE: A subset of these boreholes was used in 1-D Property Set Inversions. Depths given in meters 4 GFM2000 data for N32 Table II-1. Comparison of Borehole Layer Contact Elevations from GFM2000 and UZ Model Grid (cont.) UZ Model GFM2000 HGU USW UZ-N53/54 USW UZ-N55 USW WT-1 USW WT-2 USW WT-7 Unit Unit GFM20005 UZGrid GFM2000 UZGrid GFM2000 UZGrid GFM2000 UZGrid GFM2000 UZGrid tcw11 Tpcr CCR, CUC tcw12 Tpcp CUL, CW 1227.43 1229.417 1241.45 1240.629 1192.073 1181.052 1282.903 1297.756 1184.758 1195.65 tcw13 Tpcpv3,2 CMW 1188.872 1189.809 1187.501 1188.169 1080.821 1081.838 1242.365 1242.33 1092.098 1096.113 ptn21 Tpcpv1 CNW 1184.819 1186.203 1183.538 1184.155 1074.115 1075.147 1235.659 1235.841 1088.746 1092.733 ptn22 Tpbt4+upper Tpy BT4 1182.106 1183.028 1179.302 1180.216 1069.848 1070.891 1232.002 1232.137 1084.326 1088.3 ptn23 mid Tpy TPY absent absent absent absent absent absent absent absent absent absent ptn24 lower Tpy+Tpbt3 BT3 1179.728 1180.797 1176.254 1177.255 1068.629 1069.64 1231.087 1231.072 1082.802 1086.783 ptn25 Tpp TPP absent absent absent absent absent absent absent absent absent absent ptn26 Tpbt2+Tptrv3,2 BT2 1174.882 1175.823 1173.907 1174.418 1065.276 1066.249 1225.906 1226.111 1077.773 1081.723 tsw31 Tptrv1 TC 1162.324 1163.564 1167.079 1166.659 1053.694 1054.672 1215.847 1216.156 1065.276 1069.231 tsw32 Tptrn TR 1166.287 1164.659 1051.694 1052.672 1213.847 1214.156 1063.276 1067.231 tsw33 Tptrl+Tpul TUL 1025.957 1026.927 1185.367 1184.726 1039.978 1043.926 tsw34 Tptpmn TMN 977.7984 978.7964 1121.359 1120.757 981.7608 985.726 tsw35 Tptpll TLL 930.5544 931.6222 1079.602 1079.452 904.6464 908.6525 tsw36 upper Tptpln TM2 839.4192 840.6009 992.124 992.2496 864.4128 868.3616 tsw37 lower Tptpln TM1 816.6608 817.8055 958.596 958.5879 824.5856 828.5462 tsw38 Tptpv3 PV3 805.2816 806.4078 941.832 941.7571 804.672 808.6385 tsw39 Tptpv2 PV2 793.6992 794.7491 928.4208 928.2658 785.1648 789.0901 ch1 Tptpv1+Tpbt1 BT1, BT1a 784.2504 785.3049 915.924 915.8546 782.4216 786.3244 ch2 upper 1/4 Tac CH 779.3736 780.3893 899.16 898.901 ch3 mid 1/4 Tac CH 765.6576 766.6763 883.7676 883.5388 ch4 mid 1/4 Tac CH 751.9416 752.9632 868.3752 868.1765 ch5 lower 1/4 Tac CH 738.2256 739.2502 852.9828 852.8143 ch6 Tacbt BT 837.5904 837.452 pp4 Prowuv PP4 absent absent pp3 Prowuc PP3 815.34 815.3688 pp2 Prowmd+Prow PP2 781.2024 781.2358 pp1 Prowlv+Prowbt+Bulluv PP1 754.38 754.3675 bf3 Bulluc+Bullmd+Bulllc BF3 bf2 Bulllv+Bullbt+Tramuv BF2 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777]; Output-DTN: LB02081DKMGRID.001 NOTE: A subset of these boreholes was used in 1-D Property Set Inversions. Depths given in meters 5 GFM2000 data for N54 Table II-1. Comparison of Borehole Layer Contact Elevations from GFM2000 and UZ Model Grid (cont.) UZ Model GFM2000 HGU UE-25 WT#18 USW WT-24 Unit Unit GFM2000 UZGrid GFM2000 UZGrid tcw11 Tpcr CCR, CUC 1336.246 1335.836 1493.518 1492.268 tcw12 Tpcp CUL, CW 1310.106 1309.962 1469.098 1468.345 tcw13 Tpcpv3,2 CMW 1240.536 1240.952 1427.988 1428.093 ptn21 Tpcpv1 CNW 1235.05 1235.45 1415.796 1415.992 ptn22 Tpbt4+upper Tpy BT4 1232.611 1232.983 1408.603 1408.822 ptn23 mid Tpy TPY 1221.537 1221.926 1399.184 1399.395 ptn24 lower Tpy+Tpbt3 BT3 1214.425 1214.821 1390.802 1391.022 ptn25 Tpp TPP 1184.758 1185.218 1349.045 1349.198 ptn26 Tpbt2+Tptrv3,2 BT2 1137.818 1138.332 1292.565 1292.599 tsw31 Tptrv1 TC 1122.578 1123.119 1281.074 1281.107 tsw32 Tptrn TR 1120.578 1121.119 1279.074 1279.107 tsw33 Tptrl+Tpul TUL 1068.324 1068.839 1231.087 1231.096 tsw34 Tptpmn TMN 1007.669 1007.687 1142.482 1142.494 tsw35 Tptpll TLL 979.6272 980.1595 1108.954 1109.128 tsw36 upper Tptpln TM2 absent absent 998.22 998.5361 tsw37 lower Tptpln TM1 absent absent 987.044 987.4208 tsw38 Tptpv3 PV3 878.7384 879.184 981.456 981.8631 tsw39 Tptpv2 PV2 859.536 859.9577 969.0202 969.4164 ch1 Tptpv1+Tpbt1 BT1, BT1a 851.0016 851.4074 966.155 966.5284 ch2 upper 1/4 Tac CH 842.4672 842.8865 954.3898 954.7154 ch3 mid 1/4 Tac CH ch4 mid 1/4 Tac CH ch5 lower 1/4 Tac CH ch6 Tacbt BT pp4 Prowuv PP4 pp3 Prowuc PP3 pp2 Prowmd+Prow PP2 pp1 Prowlv+Prowbt+Bulluv PP1 bf3 Bulluc+Bullmd+Bulllc BF3 bf2 Bulllv+Bullbt+Tramuv BF2 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777]; Output-DTN: LB02081DKMGRID.001 NOTE: A subset of these boreholes was used in 1-D Property Set Inversions. Depths given in meters ATTACHMENT III — GRID VERIFICATION This attachment describes the verification activities associated with the 1-D, 2-D, and 3-D UZ Model Grids. III.1 GRIDBLOCK ATTRIBUTE VERIFICATION Because the total number of gridblocks within the 3-D UZ Model grids is quite large, a subset of gridblocks from the model is evaluated to ensure the accuracy of the calculated gridblock volumes, connection lengths, and interface areas. These verification activities are described in scientific notebooks by Wang (2002 [159673], SN-LBNL-SCI-213-V1, p. 93). Spot checks of the 1-D and 2-D mesh files were conducted to verify that the proper gridblock connections were created in mesh generation. For all 1-D and 2-D grid columns examined, gridblocks had the correct gridblock volumes and vertical connections with the adjoining gridblocks within the column (BETAX = -1). The lateral connections between gridblocks in adjoining columns for the 2-D mesh file were also spot-checked. These checks revealed that the examined gridblocks were laterally connected to neighboring blocks (in adjoining columns) and had the same assigned rock properties, with two exceptions. These exceptions were: (1) the neighboring column, or the column under investigation, was a fault block (fault blocks have different properties assigned to them), and (2) the rock type might be absent in the adjacent column, in which case the lateral connection was made with the stratigraphically closest rock type. Note that connections between gridblocks within columns associated with nonvertical (inclined) faults may be nonvertical, because the x, y locations of grid nodes within these columns can vary with depth. III.2 CONTACT ELEVATION VERIFICATION Model layer contact elevations for 45 grid columns were compared against the observed stratigraphic contact elevations contained in the GFM2000 (DTN: MO0012MWDGFM02.002 [153777]) file “contacts00el.dat.” Given an estimated maximum error in layer contact elevations at column centers of about 5 m (see first paragraph of Section 6.6), a grid validation criterion of plus-or-minus 5 m for layer contact elevations in grid columns corresponding to borehole locations was established. Differences in layer contact elevations (values from UZ Model calibration grid subtracted from values from “contacts00el.dat”) are plotted in Figures III-1 and III-2. Line discontinuities indicate missing, or pinched out, layers for that particular location. Output-DTN: LB02081DKMGRID.001 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777] NOTE: A negative value means the UZ Model layer contact elevation is higher than the stratigraphic pick. Figure III-1. Upper Contact Elevation Differences at Select Borehole Locations (GFM2000 file “contacts00el.dat” Minus UZ Model Grid) Output-DTN: LB02081DKMGRID.001 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777] NOTE: A negative value means the UZ Model layer contact elevation is higher than the stratigraphic pick. Figure III-2. Upper Contact Elevation Differences at All Borehole Locations (GFM2000 file “contacts00el.dat” Minus UZ Model Grid) Note that in most cases, the differences in contact elevations are less than 5 m. There are several cases where deviations exceed this amount. A number of boreholes (e.g., UZ-7a, H-6, NRG#7, UZ#4/5) had greater than 5 m discrepancies for the elevation of the uppermost unit present. These differences (primarily at the bedrock surface) arise from channel erosion that produces surfaces with large local variations in slope and elevation. Although the nearest GFM2000 data point may be only meters to a few tens of meters away, the highly variable surface elevations may result in the observed mismatches in the upper contact surfaces. These differences are restricted to the upper unit only, and thus should not have a significant impact on UZ Model flow and transport modeling results. Two boreholes (b#1 and N11) exhibit poor matches for most of the contact elevations, with an abrupt shift in elevations occurring below a given unit contact. Both of these boreholes are near faults, and differences in how faults were modeled in GFM2000 and the UZ Model grids may explain these discrepancies. In the case of N11, where there is a difference of over 50 m in most of the contact elevations, the borehole lies on the west side of the Solitario Canyon fault in the GFM2000 representation, but is situated on the east side of this fault in the UZ Model grid. The difference in contact elevations is similar to the observed vertical offset on the fault. The N11 borehole is located ~2 km north of the proposed repository footprint (Figure 1b), and thus this discrepancy should have little impact on UZ flow and transport models for the proposed repository area. Because of the observed differences between GFM2000 (DTN: MO0012MWDGFM02.002 [153777]) and UZ Model grid contact elevations, the b#1 and N11 boreholes were not used for 1-D rock property calibration calculations. III.3 2-D CROSS SECTION VERIFICATION To verify the accuracy of the 2-D E-W cross section (Figures III-3 and III-4), ten selected adjacent pairs of grid columns were compared to a series of GFM2000 cross sections constructed using the location of each pair of grid column nodes as ends of the cross sections (Wang 2002 [159673], SN-LBNL-SCI-213-V1, pp. 94–99). The apparent vertical offset between adjacent columns as seen in Figure III-3 is an artifact of the visualization generated by WINGRIDDER V2.0 (LBNL 2002 [154785]), and does not reflect how the layers are connected in the numerical grids (see Section 6.6 for more details). Cross sections constructed using EARTHVISION V5.1 (Dynamic Graphics 1998 [152614]) and the following GFM surfaces (see Table III-1) were compared with the correlative UZ Model grid columns (Figures III-5 to III-7). Table III-1. UZ Model Layers and GFM2000 Surfaces File Name Corresponding UZ Model Layers REF00bedrock.m.2grd tcw11 REF00tpcp.m.2grd tcw12, tcw13 s00Tpcpv1EX.m.2grd ptn21, ptn22, ptn23, ptn24 s00PahEX.m.2grd ptn25, ptn26 s00Tptrv1EX.m.2grd tsw31, tsw32 s00TptrlEX.m.2grd tsw33 s00TptpmnEX.m.2grd tsw34 Table III-1. UZ Model Layers and GFM2000 Surfaces (continued) File Name Corresponding UZ Model Layers s00TptpllEX.m.2grd tsw35 s00TptplnEX.m.2grd tsw36, tsw37 s00Tptpv3EX.m.2grd tsw38 s00Tptpv2EX.m.2grd tsw39, ch1 s00CalicoEX.m.2grd ch2, ch3, ch4, ch5, ch6 s00ProwuvEX.m.2grd pp4, pp3, pp2 s00ProwlvEX.m.2grd pp1 s00BullfrogucEX.m.2grd bf3 s00BullfroglvEX.m.2grd bf2 s00TramucEX.m.2grd tr3 gwl_sspac_60.96.2grd base of UZ The corresponding pairs of column coordinates used for each of the traverses are listed in Table III-2. Table III-2. Cross Section Traverse Columns Traverse # ID of W Column W Column Easting W Column Northing ID of E Column E Column Easting E Column Northing 1 q40 168882.0938 232653.5 a63 169150 232650 2 e64 170094.6094 232454.7344 q44 170263.7031 232385.9062 3 i24 170539.8438 232295.8125 i29 170564.875 232218.7812 4 i40 170614.9375 232064.7031 i41 170769.0156 232114.7656 5 i47 170948.1094 232087.7812 i52 170973.1406 232010.75 6 i60 171023.2031 231856.6875 i61 171177.2812 231906.7344 71 p 3 171338.0469 231868.3125 p 2 171388.5781 231860.8594 8 q62 171982.7969 231801.2969 q51 172168.4062 231811.0938 92 o 2 172299.2812 231776.0312 o 1 172358.8906 231769.2031 103 a48 172750 231750 q19 173079.3906 231774.7656 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777]; Output-DTN: LB02081DKMGRID.001 NOTE: 1- Columns separated by Ghost Dance Fault 2- Columns separated by Imbricate Fault 3- Column q19 adjacent to Toe Fault Using these traverse endpoints and the stacked GFM2000 surfaces listed in Table III-2, ten 2-D cross sections were created. The results of this comparison are shown below. Figures III-3 and III-4 depict the 2-D plan-view grid design and an East-West cross section from the UZ Model grid (file “EWUZ7a_profile.eps” from Output-DTN: LB02081DKMGRID.001) and illustrate the location of each of the column pairs used to construct the 10 traverses. Figures III-5 to III-7 depict each of the GFM2000 traverse cross sections, sandwiched between the corresponding UZ Model columns. Output-DTN: LB02081DKMGRID.001 NOTE: Line A-A’ indicates location of cross section shown in Figure III-4. Figure III-3. 2-D (plan-view) UZ Model Grid Design Output-DTN: LB02081DKMGRID.001 NOTE: UZ Model layer ptn23 does not occur within this traverse. Numbered column pairs were used to construct the comparison plots between the UZ Model grid and GFM2000. Figure III-4. Two-dimensional Cross Section from the UZ Model Grid Output-DTN: LB02081DKMGRID.001 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777] Figure III-5. Traverses 1–4 of 2-D Cross Section, Comparing Results of UZ Model and GFM2000 Grids Output-DTN: LB02081DKMGRID.001 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777] Figure III-6. Traverses 5–7 of 2-D Cross Section, Comparing Results of UZ Model and GFM2000 Grids Output-DTN: LB02081DKMGRID.001 Source DTN: MO0012MWDGFM02.002 (GFM2000) [153777] Figure III-7. Traverses 8–10 of 2-D Cross Section, Comparing Results of UZ Model and GFM2000 Grids For Traverses 1–6 and 8, the matches between the unit contacts for the GFM2000 cross sections and the UZ Model columns are extremely good, with minimal offset of units observed. These intervals are not intersected by faults, and thus a good match is expected. Discrepancies between unit contacts are observed for traverses (7, 9, 10) that cross or are immediately adjacent to faults. Most of the GFM2000 unit thicknesses in Traverse 7 (where the Ghost Dance fault passes) correlate with their counterparts for the two UZ Model columns; however, there are some differences in the location of the contact elevations. More significant differences are observed in Traverses 9 and 10. Substantial (~50 m) vertical offset is observed along the Imbricate fault, which cuts through Traverse 9, and discrepancies of up to 10–20 m are observed between the UZ model column and GFM2000 contacts. Even larger discrepancies are observed between the GFM2000 cross section in Traverse 10 and the eastern UZ Model column. This difference may result from the nearby presence of the Toe fault, which is modeled as a vertical feature by the UZ model, but as a dipping fault in GFM2000 (DTN: MO0012MWDGFM02.002 [153777]). The comparisons made using column centers around faults in the UZ Model are affected by the closely spaced nature of the column nodes (50–60 m), similar to the data resolution (61 × 61 m) of the GFM2000 grid (BSC 2002 [159124], Section 6.4.2). The localization of contact elevation discrepancies between the GFM2000 and UZ Model grids near faults results in part from the differences in the way that faults are represented in the two systems. The simplification of faults as required by the use of vertical columns in the UZ Model grids (Assumption 7) results in localized discrepancies between the two grids. However, as demonstrated by good matches observed in Traverses 1–6 and 8, the UZ Model grids accurately portray the stratigraphic representation of geologic units within structural blocks. III.4 3-D MESH VERIFICATION To verify the accuracy of the 3-D mesh and its connections, test simulations using isothermal, saturated conditions were conducted on the ECM mesh using TOUGH2 V1.4 (LBNL 2000 [146496]). The goal of these simulations was to look for improperly connected gridblocks that would be identified by anomalous points on a pressure-elevation plot. Under steady-state conditions, the observed fluid pressures should vary linearly as a function of gridblock elevation. A description of the simulations and their results are given in Wang 2003 ([162380], SN-LBNLSCI- 213-V1, pp. 125–131; SN-LBNL-SCI-103-V2, pp. 17–28; SN-LBNL-SCI-199-V1, pp. 237– 238). Initial conditions of 25°C, 500 bars water pressure, and a single suite of rock properties were assigned to all of the gridblocks. Large volume gridblocks located at the base of the grid served as a constant pressure boundary and the remaining gridblocks in the mesh were allowed to come to pressure equilibrium with this boundary condition. The simulations were run for 0.316 × 1018 s to ensure that a steady-state solution would be obtained (Wang 2003 ([162380], SN-LBNLSCI- 199-V1, p. 238). Several modifications to some of the lateral connections in the 3-D mesh were made as a result of the simulation results. First, improper lateral connections between adjoining fault and repository columns were corrected (Wang 2003 [162380], SN-LBNL-SCI-103-V2, pp. 17–23). During further evaluation of the 3-D grid, it was discovered that anomalous pressures were associated with some matrix columns adjacent to fault columns (Wang 2003 [162380], SN-LBNL-SCI-213-V1, pp. 125–131; SN-LBNL-SCI-103-V2, pp. 23–26). As mentioned in Section 6.3, some simplification of the GFM2000 faults was made in creating the UZ Model grids, including the representation of the Solitario Canyon and Solitario Canyon (west) faults as a single fault. The gridblocks with the anomalous pressure-elevation relations exhibited faultrelated stratigraphic offset with their neighboring columns. To ensure proper flow behavior in the grid, the columns with apparent fault-related offset were classified as "faults” while reconstructing the 3-D grid so that lateral connections between gridblocks in these columns and those in the adjacent columns were made with the closest lateral neighbor, and not with the same stratigraphic interval. A total of 18 columns, all adjacent to faults, were adjusted in this manner. The pressure-elevation relation results from the test simulation conducted using the final 3-D mesh exhibited very little deviation from linearity (Figure III-8). A few small deviations were observed in this simulation were attributed to the presence of nonvertical connections associated with inclined fault columns. Larger pressure shifts were observed for gridblocks associated with faults with dips that had the largest deviation from vertical. This feature is a result of the non-orthogonal configuration of the 3-D grid (Wang 2003 [162380], SN-LBNL-SCI-213-V1, pp. 130–131). The changes in the 3-D mesh resulting from these test simulations were captured in the output DTN: LB03023DKMGRID.001. This DTN supersedes DTN: LB02083DKMGRID.001. These changes do not impact the 1-and 2-D grids of DTN: LB02081DKMGRID.001. Output-DTN: LB03023DKMGRID.001 Figure III-8. Pressure-Evaluation Relations of 3-D Mesh (124,795 elements) after TOUGH2 Test Simulation