Yucca Mountain, in southern Nevada, is being investigated by the U.S. Department of Energy as a potential site for a repository for high-level radioactive waste. This report documents the results of surface-based geologic, pneumatic, hydrologic, and geochemical studies conducted during 1992 to 1996 by the U.S. Geological Survey in the vicinity of the North Ramp of the Exploratory Studies Facility (ESF) that are pertinent to understanding multiphase fluid flow within the deep unsaturated zone. Detailed stratigraphic and structural characteristics of the study area provided the hydrogeologic framework for these investigations.

Multiple lines of evidence indicate that gas flow and liquid flow within the welded tuffs of the unsaturated zone occur primarily through fractures. Fracture densities are highest in the Tiva Canyon welded (TCw) and Topopah Spring welded (TSw) hydrogeologic units. Although fracture density is much lower in the intervening nonwelded and bedded tuffs of the Paintbrush nonwelded hydrogeologic unit (PTn), pneumatic and aqueous-phase isotopic evidence indicates that substantial secondary permeability is present locally in the PTn, especially in the vicinity of faults. Borehole air-injection tests indicate that bulk air-permeability ranges from 3.5x10^-14 to 5.4x10^-11 square meters for the welded tuffs and from 1.2x10^-13 to 3.0x10^-12 square meters for the non welded and bedded tuffs of the PTn. Analyses of in-situ pneumatic-pressure data from monitored boreholes produced estimates of bulk permeability that were comparable to those determined from the air-injection tests. In many cases, both sets of estimates are two to three orders of magnitude larger than estimates based on laboratory analyses of unfractured core samples. The in-situ pneumatic-pressure records also indicate that the unsaturated-zone pneumatic system consists of four subsystems that coincide with the four major hydrogeologic units of the unsaturated zone at Yucca Mountain. In descending order, these hydrogeologic units are the Tiva Canyon welded (TCw), Paintbrush nonwelded (PTn), Topopah Spring welded (TSw ), and Calico Hills nonwelded (CHn).

Deep percolation takes place as episodic pulses of inflow that propagate rapidly to depth and apparently bypass most of the rock matrix. Field-scale and core-scale water potentials throughout much of the PTn and TSw are very high, generally greater than -0.3 megapascals, and are nearly depth invariant. Thus, the imbibition capacity of the densely welded tuffs, at least near fractures, is very small because of low matrix permeabilities and low water-potential gradients across the fracture-matrix interface. The combination of high fracture permeability, high water potentials, high matrix saturations, and low matrix permeabilities results in a percolation environment that favors deep fracture flow. The episodic pulses of inflow are evidenced in the sporadic but nevertheless commonplace occurrence of water with concentrations of radioactive isotopes indicative of origins postdating the atmospheric testing of nuclear weapons. High concentrations of tritium have been detected at many horizons within the PTn and in the top of the TSw. Much lower concentrations of tritium, indicating the mixing of a bomb-pulse component with older water, have been detected in the deeper sections of the TSw and in the CHn.

Evidence for fracture flow also is apparent in the widespread occurrence of perched water with chemical and isotopic signatures that indicate a fracture-flow origin for at least some of this water. In the North Ramp area, perched water has been detected at the base of the Topopah Spring Tuff or in the top of the underlying non welded to partially welded tuffs of the Calico Hills Formation in every dry-drilled borehole of sufficient depth to penetrate the Topopah Spring Tuff-Calico Hills Formation contact. The concentrations of the major ions of the perched water are similar to that of TSw pore water at borehole UZ-14, CHn pore water, and saturated-zone water at boreholes NRG-7 a and SD-9. The absolute chloride concentration of the perched water, however, is much lower than the chloride concentration of pore water from either the PTn or the TSw. The chemical and isotopic compositions of perched water indicate that this water was derived primarily from fracture flow, with little or no contribution from water in the matrix of the overlying rock. Carbon-14 ages of perched water range from 3,000 to 7,000 years. Strontium-87 isotope ratios indicate dissolution of surficial pedogenic calcite and calcite fracture fillings, which supports a fracture-flow origin for perched water. Moreover, carbon-13 and deuterium isotope values indicate rapid infiltration into fractures with little or no prior evaporation.

Evidence for deep fracture flow into the Calico Hills Formation at UZ-14 is indicated by carbon-14 values that are from 65 and 95 percent modem carbon, equivalent to apparent ages of about 3,500 to 500 years. Some of these ages are younger than age estimates for perched water in the overlying Topopah Spring Tuff and are much younger than any that could be derived from a matrix-flow model.

Evidence is lacking for extensive lateral flow within the PTn or for interception and diversion of this flow downward along structural pathways (faults), two key features of the original conceptual model for unsaturated flow at Yucca Mountain. Where data are available to infer lateral flow in the PTn, it is not certain that fracture flow could not have produced the same results. Pneumatic data, derived primarily from analysis of the interference effects from excavation of the North Ramp tunnel, indicate that faults within the Topopah Spring Tuff are open over substantial distances and are very permeable. Tunnel-boring-induced pneumatic disturbances have been propagated along these faults over distances that exceed 500 meters. These disturbances also have been detected in the pneumatic-pressure record of the overlying PTn in the vicinity of these faults. In spite of the apparent high permeability of faults, the existing data have neither confirmed nor refuted the hypothetical role of these faults in intercepting lateral flow from within or from above the PTn and diverting this flow downward into the deeper subsurface.

On the basis of measured temperature gradients within the TSw, deep percolation appears to be greatest beneath active channels of major drainages, diminishing toward the margins and hillslopes bordering these channels. Numerical simulations indicate that this downward percolation is accompanied by lateral spreading as the percolation front moves downward through the PTn and across the contact between the PTn and underlying TSw. Temperature data from a well-documented site in Pagany Wash indicate the presence of a significant heat-flow deficit between the PTn and underlying TSw that most likely is due to nonconductive heat-flow processes with substantial capacity to extract heat. Percolation fluxes on the order of 10 to 20 millimeters per year beneath the Pagany Wash channel and on the order of 5 millimeters per year or less beneath the hillslopes bordering this drainage accounted for the apparent heat-flow deficit. Analyses of borehole temperature gradients in Drill Hole Wash indicate similar percolation fluxes and flux distributions within that drainage. An analysis of residence times estimated from uncorrected carbon-14 activities of perched-water samples and estimates for the volume of the structurally controlled reservoir, however, showed that the perched-water reservoir intersected by borehole UZ-14 under Drill Hole Wash could be sustained by percolation fluxes through the TSw of as little as 0.001 to 0.29 millimeter per year.

The significance and implications of these findings with respect to waste isolation are discussed in the appendix of this report.