Technical Basis Document No. 1: Climate and Infiltration Revision 1 

1. INTRODUCTION 
1-1 
1.1 OVERVIEW 
This technical basis document is one in a series being prepared for each component of the Yucca 
Mountain repository system relevant to its design and postclosure performance. The components 
are illustrated in Figure 1-1. 
Climate controls the range of precipitation and temperature conditions at the land surface. The 
surface conditions (e.g., runoff, run-on, evapotranspiration) impact the rate of infiltration into the 
subsurface. Infiltration is defined as the flow of surface water downward across the atmosphere– 
soil or the atmosphere–bedrock interface (Flint, A.L. et al. 1996, p. 8). Net infiltration, which is 
the flow of water downward (i.e., drainage) below the root zone, controls deep percolation 
through the unsaturated zone. Percolation determines seepage that is important to the waste 
package performance, as well as groundwater recharge. Present-day and future infiltration is a 
key hydrologic parameter needed for design of the repository in the unsaturated zone at Yucca 
Mountain. The understanding of present-day infiltration is needed to provide initial conditions 
for predictions of future hydrologic changes, in association with climate predictions. 
May 2004 No. 1: Climate and Infiltration Revision 1 
To forecast the future climate conditions at Yucca Mountain, the following types of paleoclimate 
records, representing different time scales, were used: (1) regional records of stable isotope ä18O 
fluctuations from Devils Hole, Nevada, and microfossil data from Owens Lake, California; 
(2) vegetation assemblages recovered from local packrat nests (i.e., midden); and (3) global 
records of an earth-orbital clock of precession and eccentricity and Vostok (Antarctica) ice core 
isotopic compositions. The stable isotope compositions are typically reported in delta, ä, 
notation as the parts per million deviation of the ratio of the heavy to light isotopes (18O/16O) in 
the sample to that of a standard, VSMOW (Vienna standard mean ocean water, for 18O/16O) 
(USGS 2001a, p. 27). The global records show that climate is cyclic over 400,000-year periods. 
The estimation of the future climate states at Yucca Mountain for the period from the present to 
10,000 years was provided by the U.S. Geological Survey (USGS) (USGS 2001a) and by the 
Desert Research Institute (DRI) (Sharpe 2002; Sharpe 2003). 
Three potential climate states (interglacial, monsoon, and glacial transition) are forecast for the 
next 10,000 years based on the examination of past proxy climate records, including ostracode 
and diatom assemblages recovered from the Owens Lake core (USGS 2001a; Sharpe 2003). The 
interglacial climate state is comparable to the relatively warm present-day climate state. The 
monsoon climate state is characterized by hot summers with increased summer rainfall relative 
to the present-day climate. The glacial-transition (or intermediate) climate state has cooler and 
wetter summers and winters relative to the present-day climate state. 
Both USGS (2001a, p. 26) and DRI reports (Sharpe 2003, Table 6-1, p. 56) confirm the existence 
of a long-term, interglacial climate state similar to modern for at least the last 9,000 years before 
present. The DRI analysis estimated the duration of interglacial climate from 12,000 years 
before present (11,600 years before present in Sharpe 2002, Table 6-1, before rounding off) to 
1,000 years before present. Within the regulatory period of 10,000 years, no full glacial climate 
regime is expected. 
The timing of the climate intervals for the next 10,000 years is forecast on the basis of 
theoretically supported assumptions: 
• Climate conditions are (USGS 2001a, p. 62): (a) cyclic, (b) can be timed with the earth’s 
orbital clock, and (c) repeat themselves in a predictable way. 
• The time of the transition from marine isotope stages 11 and 10 (see Section 2.1) is 
selected as the analog to estimate climate for the next 10,000 years. 
• Ostracode species assemblages from the Owens Lake record for this time period and the 
sediment accumulation rate are used to determine the nature and timing for the three 
future climate states (USGS 2001a, p. 67). 
Forecasting of climatic data indicates that during the next 10,000 years at Yucca Mountain, the 
present-day climate should persist for 400 to 600 years, followed by a warmer and wetter 
monsoon climate for 900 to 1,400 years, followed by a cooler and wetter glacial-transition 
climate for the remaining 8,000 to 8,700 years (USGS 2001a, p. 76). 
May 2004 1-2 No. 1: Climate and Infiltration Revision 1 
Thompson et al. (1999, Table 4, Figures 16 and 17) reported the following for the Yucca 
Mountain region: (1) a mean annual precipitation of 125 mm and a mean annual temperature of 
13.4ºC for the present-day conditions and (2) using analog-based precipitation estimates, the 
mean annual precipitation for the last glacial maximum was approximated to be from 266 to 
321 mm/yr, which is 2 to 2.5 times the modern value, and the mean annual temperature was 
7.9ºC to 8.5ºC, which is 4.9ºC to 5.5ºC cooler than the modern value. The glacial-transition 
climate lower–bound precipitation values exceed the present-day lower-bound values by about 
150 mm (USGS 2001a, Section 6.6.2). Thus, much of the future climate (based on this analysis) 
is wetter and cooler than the present-day climate. 
Climatic conditions are key for estimating net infiltration in the unsaturated zone. Net 
infiltration is defined as water drainage below the bottom of the root system and occurs in the 
soil or the welded bedrock of the Tiva Canyon Tuff (Tpc), which is the most extensively exposed 
bedrock unit at Yucca Mountain. 
Intensive surface-based investigations at Yucca Mountain started in the early 1980s (Wang and 
Bodvarsson 2003), which have resulted in a number of conceptual models of infiltration (Flint, 
A.L. et al. 2001) and infiltration predictions for the present-day and future climates, using 
time-series data from a number of climate analog sites (Flint, A.L. et al. 2001; Hevesi et al. 
2003). To represent the large-scale (volume-averaged) Yucca Mountain infiltration processes, 
the conceptual and numerical models of infiltration involve a number of simplifications (the 
validity of these simplifications is discussed in Section 4.2), such as: 
• Describing infiltration using the water balance (bucket-type) model for one-dimensional 
piston-like flow 
• Neglecting nonlinear effects in partially saturated soils and fractured rock, lateral flow at 
the soil–rock interface, preferential flow through heterogeneous soils and fractures, flow 
under conditions where the soil moisture content is below the field capacity, and 
redistribution of water caused by fracture–matrix interaction 
• Using empirical relationships for saturated and unsaturated hydraulic conductivity and 
water retention, evapotranspiration, surface runoff, and surface run-on. 
Based on the results of numerical simulations, maps were prepared of steady-state net infiltration 
for the present-day, monsoon, and glacial-transition climate states, including lower-bound, mean, 
and upper-bound conditions for each of the states, over the Yucca Mountain region. These maps 
were then used to calculate the steady-state infiltration rates averaged over space and time for 
each of the climate scenarios (USGS 2001b). 
The net infiltration rates obtained for the present-day climate compare reasonably with those 
from corroborative studies, including geochemical and temperature measurements in the 
unsaturated zone, as well as groundwater recharge. This comparison provides a validation of net 
infiltration modeling results. 
Experimental and modeling studies show that a spatial and temporal variation in net infiltration 
at Yucca Mountain is caused by episodic storm and precipitation events (Hevesi et al. 2002), as 
May 2004 1-3 No. 1: Climate and Infiltration Revision 1 
well as the heterogeneous nature of topsoil and topography. Near-surface infiltration data 
(USGS 2001b) suggest that significant infiltration occurs only every few years. In very wet 
years, infiltration in major drainages (where flow is concentrated) can increase to hundreds of 
millimeters per year during a relatively short time. Current estimates indicate an average 
infiltration rate of about 5 mm/yr, with values ranging as high as 20 mm/yr near the ridge top of 
Yucca Mountain (USGS 2001b). 
Because long-term variations in climate occur, changes in net infiltration for future climates will 
occur as well. Predictions of spatio-temporal distribution of net infiltration for different climate 
states are likely to contain uncertainties associated with the selection of climate analog sites and 
corresponding records of precipitation and temperature (NRC 1997). The correlation analysis 
revealed that net infiltration is mostly dependent on precipitation, soil depth, bedrock 
permeability, and potential evapotranspiration (BSC 2003a). Because it is not possible to foresee 
every condition that could occur over the 10,000-year regulatory period, it is necessary to 
evaluate a range of possible scenarios to ensure that predictions of net infiltration are 
conservatively bounded within each of the climate scenarios predicted for the Yucca Mountain 
region (see Section 4.3). 
1.3 OBJECTIVES AND SCOPE 
This document summarizes the results of forecasting the climatic conditions and the evaluation 
of net infiltration at Yucca Mountain, based on the results of field, laboratory, and modeling 
studies, as well as associated uncertainties for the present-day and future monsoon and 
glacial-transition climates. 
1.2 SIGNIFICANCE OF NET INFILTRATION FOR TOTAL SYSTEM 
PERFORMANCE ASSESSMENT 
The evaluation of net infiltration is needed for assessing the nominal scenario of the repository 
system in total system performance assessment (TSPA) (DOE 2002a, Vol. II). 
The net infiltration is used as input in the prediction of deep percolation flux, the extent of 
perched-water zones, water transport time through the unsaturated zone, and potential seepage 
into waste emplacement drifts. Quantitative estimates of net infiltration under present-day and 
future climatic conditions define the upper-boundary condition of unsaturated zone flow models, 
which provide input to the TSPA (DOE 1998, Volume 3, Figure 2-1). 
The evaluation of net infiltration is also potentially important for the estimation of radiation 
doses and biosphere dose conversion factors, in the event of using contaminated groundwater for 
irrigation (Williams 2001) and in the event of the igneous groundwater contamination scenario. 
The incorporation of infiltration uncertainty analysis under different climate scenarios is required 
to support TSPA for the license application, as outlined in the Total System Performance 
Assessment-License Application Methods and Approach (BSC 2002a, Section 3.1). 
May 2004 1-4 No. 1: Climate and Infiltration Revision 1 
The scope of the document includes: 
• Analysis of present-day and future climatic conditions, including the selection of climate 
analogs (Section 2) 
• Analysis of the dominant factors and processes affecting net infiltration (Section 3) 
• Basis for the modeling approach, validation of the conceptual model, and the analysis 
from the results of simulating net infiltration for present-day and potential future 
climatic conditions (Section 4) 
• Discussion of the results of corroborative experimental studies conducted for estimating 
net infiltration (Section 5) 
• The uncertainty analysis of net infiltration (Section 6) 
• Conclusions (Section 7). 
Appendices to this document include the response to open Key Technical Issues (KTIs) imposed 
by the U.S. Nuclear Regulatory Commission (NRC), which are related to the evaluation of net 
infiltration rate (Table 1-1). 
Table 1-1. Key Technical Issues Addressed in This Report 
Appendix 
A 
B 
C 
D 
E 
Short Description 
Water balance, Richards equation 
Evapotranspiration 
Soil lateral flow 
Monte Carlo analysis of infiltration uncertainty 
Parameters of infiltration uncertainty analysis 
KTI Agreement/AIN 
TSPAI 3.18 AIN-1 
TSPAI 3.19 AIN-1 
TSPAI 3.21 AIN-1 
USFIC 3.01 AIN-1 
USFIC 3.02 AIN-1 
1.4 ENVIRONMENTAL CONDITIONS OF THE YUCCA MOUNTAIN SITE 
Physiographic Setting and Topography–Yucca Mountain is located within a transition zone 
between the northern boundary of the Mojave Desert and the southern boundary of the Great 
Basin Desert (Flint, A.L. et al. 1996, pp. 43 to 44). Physiographic subdivision of the Yucca 
Mountain area is shown in Figure 1-2. The topography of the Great Basin is characterized by 
isolated, long and narrow, roughly north-south-trending mountain ranges and broad intervening 
valleys. The topography at Yucca Mountain includes: ridge tops (about 7 percent), side slopes 
(about 47 percent, including footslopes), terraces (about 44 percent), and channels (about 
2 percent) (Flint, L.E. and Flint 1995, pp. 2 to 6; Flint, A.L. et al. 1996, pp. 38 and 39). The 
ridge tops generally are flat to gently sloping, with soils 0.5 to 2 m (1.6 to 6.6 ft) thick. Terraces 
and channels are located at lower elevations of primary washes and have thin soil cover in the 
upper washes and thick soils farther down. The surface elevations above the site of the 
repository are approximately 1,400 m. 
1-5 May 2004 No. 1: Climate and Infiltration Revision 1 
Figure 1-2. Geographic and Prominent Topographic Features of the Death Valley Region 
Source: Simmons 2004, Figure 8-2. 
May 2004 1-6 No. 1: Climate and Infiltration Revision 1 
The surface has been shaped by erosional processes on the eastward-sloping ridge of the 
mountain and along faults and fault scarps that have created a series of washes downcut to 
varying degrees into different bedrock layers. Slopes are locally steep on the west-facing 
escarpments eroded along the faults and in some of the valleys that cut into the more gentle 
eastward-facing dip slopes. Narrow valleys and ravines are cut in bedrock; wider valleys are 
covered by alluvial deposits with terraces cut by intermittent streams. Locally, small sandy fans 
extend down the lower slopes and spread out on the valley floors. East of the crest of 
Yucca Mountain, drainage is into Fortymile Wash; west of the crest, streams flow 
southwestward down fault-controlled canyons and discharge into Crater Flat (see Figure 1-2). 
The topography is different to the south and north of Drill Hole Wash (see Figures 1-3 and 1-4). 
The washes in the southern area trend eastward, are relatively short (less than 2 km), and have 
erosional channels with gently sloping sides. The washes north of Drill Hole Wash (e.g., Yucca 
Wash) trend northwest, are relatively longer (3 to 4 km) because they are controlled by fault 
features, and have steeper side slopes. The locations of Yucca Mountain watersheds are shown 
in Figure 1-3. 
Climate–In general, the present-day climate of Nevada is characterized as semiarid to arid. 
Currently, Nevada’s climate is being developed under the influence of dry winds entering 
Nevada and providing reduced precipitation. Mountain systems to the west cause a rain shadow 
effect. Moisture-laden winds traveling east from the Pacific Ocean rise, cool, and cause 
precipitation in the mountain systems west of Nevada. 
The most important climatic factors affecting water-transport processes in the unsaturated zone 
are solar radiation flux, diurnal and seasonal temperature cycles, relative humidity, and 
precipitation, in the form of either rain or snow, as well as extended periods of drought. Current 
climatic conditions for the site and the Yucca Mountain region are discussed in detail in the 
Yucca Mountain Site Description (Simmons 2004, Section 6). The Yucca Mountain Project 
environmental program collected site meteorological data using a network of nine automated 
weather stations (Simmons 2004, Section 6.2). 
Average annual precipitation over the area of the Nevada Test Site ranges from a maximum of 
370 mm (14.57 in.) in the Belted Range to a minimum of 110 mm (4.33 in.) in the Amargosa 
Desert. Annual precipitation within the repository area ranges from a minimum of about 
100 mm (3.94 in.) at low elevations along the southern boundary to a maximum in excess of 
300 mm (11.8 in.) at high elevations in the north (Simmons 2004, Figure 7-6). Based on the 
analysis of data from 114 precipitation stations in the Yucca Mountain region, which provides at 
least 8 years of complete record, a strong positive correlation between average annual 
precipitation and station elevation was found (Simmons 2004, Figure 7-7). These results 
indicate that the zones of maximum precipitation are likely to correspond to the zones of higher 
elevations of the mountain ranges. 
May 2004 1-7 No. 1: Climate and Infiltration Revision 1 
Source: USGS 2001b, Figure 6-12. 
NOTE: On this figure and other figures in this document, the inner box is the repository boundary. The outer box is 
the unsaturated zone flow and transport model simulation area. Both are shown as schematic only and are 
not to scale. 
Figure 1-3. Location of 10 Watershed Model Domains Included in the Composite Watershed Model 
Area Overlying the Area of the Unsaturated Zone Flow and Transport Model Presented in 
Simulation of Net Infiltration for Modern and Potential Future Climates (USGS 2001b) 
May 2004 1-8 No. 1: Climate and Infiltration Revision 1 
Source: USGS 2001b, Figure 6-19. 
Figure 1-4. Location of Stream-Gauging Sites and Calibration Watersheds Defined by the Gauging Sites 
May 2004 1-9 No. 1: Climate and Infiltration Revision 1 
Geology–Yucca Mountain is an uplifted, heavily block-faulted ridge of alternating layers of 
welded and nonwelded volcanic tuffs of Miocene age. The major geologic units at Yucca 
Mountain are the volcanic tuff formations of the Paintbrush (Tp) Group, the Calico Hills 
Formation (Tac), and the Crater Flat (Tc) Group. The lithostratigraphic nomenclature divides 
the Paintbrush Group into the Tiva Canyon (Tpc), Yucca Mountain (Tpy), Pah Canyon (Tpp), 
and Topopah Spring (Tpt) tuffs. The Crater Flat Group is divided into the Prow Pass (Tcp), 
Bullfrog (Tcb), and Tram (Tct) tuffs. For purposes of hydrogeologic studies, including 
infiltration, a separate stratigraphic nomenclature was developed based on the degree of welding 
and hydrologic property distributions (Simmons 2004, Tables 3-5 and 7-1). 
The major hydrogeologic units are divided into the Tiva Canyon welded (TCw), the Paintbrush 
nonwelded (PTn) (consisting primarily of the Yucca Mountain and Pah Canyon members and the 
interbedded tuffs), the Topopah Spring welded (TSw), the Calico Hills nonwelded (CHn), and 
the Crater Flat undifferentiated (CFu) units. Figure 1-5 presents a three-dimensional 
representation of the block-faulted hydrogeologic units at Yucca Mountain. 
For evaluation of net infiltration, only the TCw hydrogeologic unit of the Tiva Canyon Tuff and 
overlying alluvium are considered. The Tiva Canyon Tuff is a compositionally zoned, generally 
moderately to densely welded, tuff sequence. The sequence is a highly fractured, variably 
eroded unit with thicknesses in the repository area varying from 0 to more than 150 m. 
Figure 1-5. Three-Dimensional Geologic Presentation of Yucca Mountain 
May 2004 1-10 No. 1: Climate and Infiltration Revision 1 
Hydrogeology–Yucca Mountain is located within the Alkali Flat-Furnace Creek groundwater 
basin, which is within the larger Death Valley Regional Groundwater System. The groundwater 
flow system of the Death Valley region is very complex, involving many groundwater systems. 
In some areas, confining units allow considerable movement between aquifers; in other areas, 
confining units are sufficiently tight to support artesian conditions. Groundwater below Yucca 
Mountain and in the surrounding region flows generally south toward discharge areas in the 
Amargosa Desert and Death Valley. The area around Yucca Mountain is in the central subregion 
of the Death Valley regional groundwater system, which has three groundwater basins: Pahute 
Mesa-Oasis Valley, Ash Meadows, and Alkali Flat-Furnace Creek. Figure 1-6 depicts major 
areas of groundwater recharge and discharge, water level elevations, and flow directions within 
the Yucca Mountain area. The primary sources of groundwater recharge are infiltration on 
Pahute Mesa, Rainier Mesa, Timber Mountain, and Shoshone Mountain to the north, and the 
Grapevine and Funeral Mountains to the west and southwest (see Figure 1-2). Recharge in the 
immediate Yucca Mountain vicinity is small, consisting of water reaching Fortymile Wash, as 
well as precipitation that infiltrates into the subsurface (Hevesi et al. 2003). 
Surface Hydrology–Yucca Mountain is located in the Amargosa River drainage basin, which is 
the major tributary drainage area to Death Valley. Streamflow from Yucca Mountain can extend 
from local drainages to the Amargosa River and then to Death Valley. The Amargosa River and 
its tributaries are ephemeral streams (i.e., they are dry most of the time), with flow rarely 
occurring in direct response to precipitation. Along short distances, groundwater discharges at 
springs into the channel system. During episodic flooding, flow occurs along the Amargosa 
River, filling much of the Death Valley saltpan to depths of 1 ft (0.3 m) or more (Miller 1977, 
p. 18). During periods in which the climate has been cooler and wetter, such as 140,000 to 
175,000 years ago, Death Valley was filled with water to depths of 175 m (USGS 2001a, p. 27). 
The entire Death Valley drainage basin and several closed drainage basins are interconnected 
through the groundwater system (D’Agnese et al. 1997, Figure 9, pp. 20 and 22). About 
4,000 mi2 (10,000 km2) drain directly into Death Valley (Miller 1977, p. 18). In addition, the 
Amargosa River drains almost 3,500 mi2 (9,100 km2) north and east of Death Valley, and Salt 
Creek drains about 1,500 mi2 (3,900 km2) to the south. 
Several artificial lakes (Crystal Reservoir, Lower Crystal Marsh, Horseshoe Reservoir, and 
Peterson Reservoir) store the collective discharge from various springs located in Ash Meadows. 
Like the streams, the playas are mainly ephemeral and contain water only after heavy runoff 
periods. Throughout the Death Valley basin, perennial flow is only observed downgradient from 
spring discharges and around the margins of playas and saltpans, where the groundwater 
discharges to the land surface. Surface water flows have been monitored at five sites in the 
Yucca Mountain region (Figure 1-4). 
May 2004 1-11 No. 1: Climate and Infiltration Source: Fenelon and Moreo 2002, Figure 2. 
Figure 1-6. Hydrogeologic Map Showing Major Factors Controlling Groundwater Flow in the Yucca 
Mountain Region, Southern Nevada, and Eastern California 
No. 1: Climate and Infiltration 
Revision 1 
May 2004 1-12 Revision 1 
Soils–A soil survey of Midway Valley at the North Portal facilities and the ridges to the west 
(Resource Concepts 1989) and a more general soil survey of the entire Yucca Mountain region 
(YMP 1997) identified 17 soil series and seven map units (Table 1-2) (Resource Concepts 1989). 
Based on a wetlands assessment at the Nevada Test Site (Hansen et al. 1997), there are no hydric 
soils at the Yucca Mountain site. Yucca Mountain soils are derived from underlying volcanic 
rocks and mixed alluvium dominated by volcanic material and, in general, have low 
water-holding capacities. 
The shallow soils on ridge tops at Yucca Mountain often consist of a thin hardpan (hardened or 
cemented soil layer) on top of bedrock and range from well drained to excessively drained, 
which means that water drains readily to very rapidly. A topsoil layer is typically less than 15 
cm thick and, in some instances, has an underlying soil layer 5 to 30 cm thick. 
Soils on fan piedmonts and in steep, narrow canyons are relatively deep and well drained (water 
is drained readily but not rapidly). These soils developed from residues of volcanic parent 
material, with a component of calcareous eolian sand. Soils formed from the volcanic parent 
material generally range from moderately shallow (50 to 75 cm) to moderately deep (75 to 
100 cm) overlaying a thin hardpan on top of bedrock. The topsoil layers are generally less than 
25 cm (10 in.) thick, with an underlying soil layer of 25 to 50 cm. The mixed soils, containing 
residues from volcanic parent material and calcareous eolian sand, are often deep (100 to 
150 cm) or moderately deep, having a well-cemented hardpan. A topsoil layer is less than 15 cm 
thick, with the underlying layer of soil parent material as deep as 150 cm (60 in.). 
Soils on alluvial fans and in stream channels are very deep (greater than 150 cm) and range from 
well drained to excessively drained. A topsoil layer is generally less than 20 cm (8 in.) thick, 
with the layer of soil parent material as deep as 150 cm. Soil textures are very gravelly, fine 
sands, sandy loams, with 35 to 70 percent of rock fragments. The soils are calcareous and 
moderately alkaline. 
Alluvial deposits, consisting of fluvial sediments and debris flows, are present in the valley 
floors and washes (Rousseau et al. 1999, pp. 10 and 11). These deposits have varying degrees of 
soil development and thickness and have a gravelly texture, with rock fragments constituting 
between 20 and 80 percent of the total volume. The alluvial deposits range from 100 m thick in 
Midway Valley to less than 30 m thick in the mouths of the smaller washes. In the middle of the 
washes, most alluvial fill (soil) is less than 15 m thick. Side slopes are characterized by a very 
thin soil cover, with densely welded and highly fractured bedrock. 
The erosion of the sediments and exposed bedrock over 10,000 years is expected to be on the 
order of centimeters (Simmons 2004, Section 3.4.6), which is within the range of existing surface 
elevation irregularities and would not significantly affect the processes in the hundreds of meters 
(thousands of feet) of unsaturated zone. Therefore, the effects of soil erosion on infiltration are 
considered negligible and are reasonably excluded from the TSPA calculations (BSC 2001a, 
Section 6.4.1). 
May 2004 1-13 No. 1: Climate and Infiltration Map Unit 
Upspring-Zalda 
Gabbvally– 
Downeyville–Talus 
Upspring–Zalda– 
Longjim 
Skelon–Aymate 
Strozi variant– 
Yermo–Bullfor 
Jonnic variant– 
Strozi–Arizo 
Percent 
11 
8 
27 
22 
7 
12 
Yermo–Arizo–Pinez 13 
Source: CRWMS M&O 1999, pp. 3 and 4. 
Geographic Setting 
Mountain tops and ridges. Soils 
occur on smooth, gently sloping 
ridge tops and shoulders and on 
nearly flat mesa tops. Rhyolite 
and tuffs are parent materials for 
both soil types. 
North-facing mountain side 
slopes. Talus is stone-sized rock 
occurring randomly throughout 
unit in long, narrow, vertically 
oriented accumulations. 
Alluvial fan remnants. Soils occur 
on gently to strongly sloping 
summits and upper side slopes. 
Dissected alluvial fan remnants. 
Soils occur on fan summits, 
moderately sloping fan side 
slopes, and inset fans. They are 
formed in alluvium from mixed 
volcanic sources. 
Inset fans and low alluvial side 
slopes in mountain canyons; and 
drainages between fan remnants. 
Soils occur on moderately to 
strongly sloping inset fans near 
drainages, adjacent to lower fan 
remnants, and below foothills. 
Vegetation–At Yucca Mountain, Mojave Desert vegetation occurs mostly at lower elevations on 
bajadas (broad, continuous alluvial slopes) and in washes. Vegetation cover varies from 0 to 
90 percent (Hevesi et al. 2003, p. 49; CRWMS M&O 1996). The species most identified over 
broad areas are Coleogyne ramosissima (common name is black bush), located primarily on 
ridges and canyons. According to Yucca Mountain Biological Resources Monitoring Program 
No. 1: Climate and Infiltration 
Revision 1 
Table 1-2. Soil Mapping Units at Yucca Mountain 
Soil Characteristics 
Typically shallow (10 to 51 cm) to bedrock, or 
to thin duripan (a subsurface layer cemented 
by silica, usually containing other accessory 
cements) over bedrock. They are well to 
excessively drained, have low available 
water-holding capacity, medium to rapid 
runoff potential, and slight erosion hazard. 
Shallow (10 to 36 cm) to bedrock. 
Permeability is moderate to moderately rapid. 
They have moderate to rapid runoff potential, 
are well drained, and have low available 
water-holding capacity and moderate erosion 
hazard. 
Mountain side slopes. Soils occur Shallow (10 to 51 cm) to bedrock or to thin 
on south-, east-, and west-facing 
slopes, and on moderately sloping 
alluvial deposits below side 
slopes. 
duripan over bedrock. They are well to 
excessively drained and have moderately 
rapid to rapid permeability and runoff 
potential, very low available water-holding 
capacity, and slight erosion hazard. 
Moderately deep (51 to 102 cm) to indurated 
(hardened, as in a subsurface layer that has 
become hardened) duripan or petrocalcic (a 
subsurface layer in which calcium carbonate 
or other carbonates have accumulated to the 
extent that the layer is cemented or 
indurated) layer with low to very low available 
water-holding capacity, moderately rapid 
permeability, slow runoff potential, and slight 
erosion hazard. 
cm). They are well drained and have rapid 
permeability, very low available water-holding 
capacity, slow runoff potential, and slight 
erosion hazard. 
Alluvial fan remnants. Soils occur Moderately deep (51 to 102 cm) to deep (102 
on gently to moderately sloping 
alluvial fan remnants and stream 
terraces adjacent to large 
drainages. 
Moderately deep (36 to 43 cm) to deep (more 
than 102 cm), sometimes over strongly 
cemented duripan. They have slow or rapid 
permeability, slow or moderate runoff 
potential, very low available water-holding 
capacity, and slight erosion hazard. 
Deep (more than 102 cm), sometimes over 
indurated duripan. They are well drained and 
have very low available water-holding 
capacity, moderately slow to rapid 
permeability, slow to medium runoff potential, 
and slight erosion hazard. 
May 2004 1-14 Revision 1 
Annual Report FY89 & FY90 (EG&G 1991, Section 2.3.1), the most common vegetation 
association is Larrea-Lycium-Grayia (creosote bush-wolfberry-hopsage), which covers 
approximately 35 percent of the surface area at Yucca Mountain, Coleogyne (30 percent), and 
Lycium-Grayia (26 percent). Larrea-Lycium-Grayia predominates on the eastern bajadas of 
central Yucca Mountain, occurring at intermediate elevations in the study area ranging from 
1,000 to 1,500 m. Coleogyne occurs across the northern third of the study area from the valley 
floors at elevations of approximately 1,030 m (3,380 ft) to the flat ridge tops at roughly 1,710 m; 
it generally does not occupy the steep slopes and is present within areas with thin soils. 
Evapotranspiration is controlled by the rooting depth and density affecting the amount of water 
available to the plant (Hevesi et al. 2003). Although Mojave Desert shrubs do not achieve full 
vegetation cover (CRWMS M&O 1996), the plant systems may extensively exploit soil water in 
spaces between plants (Levitt et al. 1996). 
1.5 TYPES OF INVESTIGATIONS CONDUCTED TO FORECAST FUTURE 
CLIMATE CONDITIONS AND TO ASSESS NET INFILTRATION 
Climatic investigations involved Future Climate Analysis (USGS 2000) and other reports 
(Sharpe 2002; Sharpe 2003) to implement the following: 
• Forecasting of climatic conditions based primarily on the analysis of three types of data: 
1. Long-term regional records of stable isotope ä18O concentrations (Devils Hole, 
Nevada) and microfossil data (Owens Lake, California) 
2. Local discontinuous vegetation records from packrat nests (middens) 
3. Global records of an earth-orbital clock of precession and eccentricity and Vostok 
(Antarctica) ice core isotopic compositions. 
• Selecting present-day climate analog sites with daily precipitation and temperature 
records, which can be used for simulating net infiltration. 
Understanding of the Yucca Mountain unsaturated zone has evolved from surface-based 
investigations that began in the early 1980s into rigorous field, laboratory, and modeling studies 
(USGS 2001b; Flint, A.L. et al. 1996; Flint, A.L. et al. 2001; Wang and Bodvarsson 2003; BSC 
2003a). Studies related to the assessment of net infiltration can be grouped as follows: 
• Drilling of shallow and deep unsaturated zone boreholes 
• Mapping of faults and fractures 
• Measurements in boreholes of moisture content and water potential with time and depth 
• Determination of unsaturated hydraulic parameters using cores taken from boreholes 
• Determination of saturated hydraulic conductivity using air injection tests 
• Infiltration tests (e.g., Alcove 1 and Fran Ridge) 
May 2004 1-15 No. 1: Climate and Infiltration • Geochemical and temperature investigations in boreholes 
• Numerical modeling of net infiltration, including 
– Calibration and validation of the numerical model 
– Predictions (using 1996 and 2001 models) 
– Uncertainty analysis of net infiltration. 
The results of climate and infiltration investigations were also subjected to the expert elicitation 
projects (DeWispelare et al. 1993; CRWMS M&O 1997a). 
1.6 NOTE REGARDING THE STATUS OF SUPPORTING TECHNICAL 
INFORMATION 
This document was prepared using the most current information available at the time of its 
development. This technical basis document and its appendices providing KTI Agreement/AIN 
responses reflect the status of the Yucca Mountain Project scientific and design bases at the time 
of submittal. Information that evolves through subsequent revisions of the analysis and model 
reports and other references will be reflected in the license application as the approved analyses 
of record at the time of license application submittal. Consequently, the Yucca Mountain Project 
will not routinely update either this technical basis document or its KTI Agreement/AIN 
appendices to reflect changes in the supporting references prior to submittal of the license 
application. 
1-16 May 2004 No. 1: Climate and Infiltration 
Revision 1 Revision 1 
2. PRESENT-DAY AND FUTURE CLIMATE CONDITIONS 
2.1 TECHNICAL BASIS FOR FORECASTING CLIMATES AND SELECTING 
CLIMATE ANALOG SITES 
To forecast the climate conditions at Yucca Mountain, three types of data were used: (1) local 
records of stable isotope ä18O concentrations (Devils Hole, Nevada) and microfossil data (Owens 
Lake, California); (2) local discontinuous vegetation records from packrat nests (middens); and 
(3) records of an earth-orbital clock (precession and eccentricity) and ice core isotopic 
compositions. The technical basis for forecasting climate involves four key scientific 
assumptions (USGS 2001a, p. 19): 
• The climate is cyclical; past climates provide insight into potential future climates. 
• A relationship exists between the timing of long-term past climate change (i.e., glacial 
and interglacial cycles) and the timing of changes in certain earth-orbital parameters. 
This relation establishes a millennial-scale climate-change clock which provides a 
means to predict the timing of future climate changes. 
• A relationship exists between the characteristics of past climates and the sequence of 
those climates in the long, approximately 400,000-year, earth-orbital cycle. The 
characteristics of past glacial and interglacial climates within the long earth-orbital cycle 
differ from each other and do so in a systematic way. This climate-sequence 
relationship provides a defensible criterion for the selection of a particular past climate 
as an analog for future climate. 
• Long-term, earth-based climate forcing functions, primarily tectonics, have remained 
relatively unchanged during the last 400,000 years and will likely not change during the 
next 10,000 years. The potential and practically unpredictable impact of long-term, 
earth-based forcing functions on climate is not considered for forecasting climate change 
over the next 400,000 years. 
Because direct testing or analysis cannot be used to confirm the first three assumptions, which 
are interrelated to each other, the validity of a particular past climate as a future-climate analog 
can be confirmed only through the passage of time (within the 10,000-year period). The 
assumption that long-term, earth-based forcing functions will not change during the next 
10,000 years is consistent with the U.S. Environmental Protection Agency final rule 
40 CFR Part 197 with respect to the stability of geologic processes (40 CFR Part 197, pp. 25 
and 59). 
Earth-orbital parameters and many past climate proxy records show that climate is cyclic over 
400,000-year periods. Glacial and interglacial climates can be divided into marine isotope stages 
as shown in Figure 2-1. The marine isotope stages were first established from studies using 
marine carbonate ä18O records and are now applied to many climate proxy records. Evennumbered 
marine isotope stages represent glacial stages, whereas odd-numbered marine isotope 
stages represent interglacial stages. Marine isotope stage dates vary with the climate proxy 
record and location. The examination of time series ä18O data shown in Figure 2-1 indicate that 
May 2004 2-1 No. 1: Climate and Infiltration Revision 1 
the cycling occurs simultaneously with the overall trends of the increase in ä18O with time. 
Although the analysis of past climates shows that a strict repetition of climate characteristics is 
not expected for future climates, the general characteristics (i.e., the greatest effective moisture 
within the next 400,000 years) of future precipitation and temperature for a particular 
interglacial–glacial couplet can be inferred from corresponding interglacial–glacial couplets in 
the past. However, the magnitude or nature of climate states may be altered by positive or 
negative feedback mechanisms, which are unlikely to happen for the next 10,000 years (USGS 
2001a, Section 6.6). 
Source: USGS 2001a, p. 28. 
NOTE: Stable isotope data are reported relative to VSMOW. High Devils Hole ä18O values represent warm 
climates, and low values represent cold climates. Odd numbered marine isotope stages correspond to 
interglacial climates; even numbered marine isotope stages correspond to glacial climates. The published 
Devils Hole record stops at approximately 60,000 years before present. 
Figure 2-1. Devils Hole Stable Isotope Record Showing the Timing and Cyclical Nature of Climate 
Change 
May 2004 2-2 No. 1: Climate and Infiltration Revision 1 
The estimation of the future climate states at Yucca Mountain was provided by the USGS 
(2001a) for the period from the present to 10,000 years in the future. The USGS (2001a, p. 26) 
report confirms the existence of a long-term, present-day interglacial climate state for at least the 
last 9,000 years before present. The USGS analysis estimates that the present-day climate 
regime has lasted for about 9,000 to 10,000 years before present. The DRI analysis estimates 
that the interglacial climate state began about 12,000 years before present (Sharpe 2003, 
Table 6-1). Within the next 10,000 years no full glacial climate regime is expected. Other 
features revealed by the ice-core records and high-resolution marine sediment cores indicate that 
large and abrupt climatic changes occurred in the past. 
Although the causes of climate change between glacial and interglacial conditions in the past are 
not known with certainty (USGS 2001a, p. 21), timing relations between earth-orbital parameters 
and past climate cycle can be used to forecast the future. Any future climate analysis is 
uncertain, but perceived relations between measurable cycles provide a reasonable technical 
basis for forecasting estimated future climate conditions. For the next 10,000 years, 
three general climate conditions are expected, in order of increasing wetness (USGS 2000): 
• Present-day (interglacial) climate for 400 to 600 years 
• Monsoon climate for 900 to 1,400 years 
• Glacial-transition (intermediate) climate for the balance of the 10,000-year period. 
Table 2-1 compares the timing of the USGS and DRI forecasts of future climates. According to 
Sharpe (2003, p. 56), the difference in timing between the two reports is insignificant, differing 
by less than 1,800 years. 
Table 2-1. Comparison of U.S. Geological Survey and Desert Research Institute Future Climate 
Forecasts 
U.S. Geological Survey Desert Research Institute 
Climate 
Interglacial 
Duration 
600 years after present 
Climate 
Modern 
600 to 2,000 years after present Monsoon Monsoon 
Glacial Transition 2,000 to 30,000 years after present Intermediate 
Monsoon 
Source: Sharpe 2003, Table 6-6. 
The range in ages is likely caused by the uncertainty in evaluating sediment accumulation rates 
in Owens Lake, California, and rounding of values of Sharpe (2003). 
To address the future global climate changes in the TSPA, paleoclimatic information (from the 
records of past climate changes) was used to predict future climate changes. Forecasting of 
climate states is based on information about past patterns of climates (CRWMS M&O 2000a, 
pp. 3-38 to 3-42), which is a generally accepted approach, because climate is assumed to be 
cyclic and largely dependent on repeating patterns of the earth’s orbital parameters. 
Duration 
Not applicable. 
1,000 years before present to 500 
years after present 
500 to 18,500 years after present 
18,500 to 20,000 years after present 
20,000 to 38,000 years after present 
May 2004 
Intermediate 
2-3 No. 1: Climate and Infiltration Revision 1 
Based on Future Climate Analysis (USGS 2000), the Devils Hole, Nevada, stable isotope record 
(Landwehr et al. 1997) was used to provide the timing and sequence of climate states, and the 
Owens Lake, California, microfossil record (from sediment cores taken from drilling) is used to 
reconstruct the magnitude and nature of past climate states for the last 400,000 years. 
For the next 10,000 years, each future climate regime is represented by present-day 
meteorological stations that serve as analog sites (USGS 2000). The map showing the locations 
of meteorological stations selected as analog sites is shown in Figure 2-2. Table 2-2 summarizes 
the duration and lists the names and locations of representative present-day meteorological 
stations characterizing the three climate states. Data from these meteorological stations are, in 
turn, used to determine precipitation and temperature records, which are then used as inputs to 
the infiltration model (see Section 4). For each of the climate scenarios (present-day, monsoon, 
and glacial transition), the lower-bound, mean, and upper-bound states are considered. Table 2-3 
presents mean annual temperature and precipitation for these stations, which are discussed in 
more detail in Section 4.3 (Tables 4-4, 4-6, and 4-7). The daily records and mean annual 
precipitation and temperature values for each climate regime are then used in numerical 
simulations to quantify the net infiltration distribution in TSPA models (USGS 2001b; 
BSC 2003a). 
May 2004 2-4 No. 1: Climate and Infiltration Source: USGS 2001a, p. 70. 
Figure 2-2. Map of Meteorological Stations of Climate Analogs 
No. 1: Climate and Infiltration 
Revision 1 
May 2004 2-5 Table 2-2. Meteorological Stations Selected to Represent 10,000 Years of Future Climate States at 
Yucca Mountain, Nevada 
Modern Interglacial 400 to 600 years 
900 to 1,400 years 
8,000 to 8,700 years 
Duration Climate 
Monsoon 
Glacial Transition 
Source: USGS 2001a. 
Table 2-3. Comparison of Meteorological Characteristics of Climate Analog Sites 
Representative 
Meteorological 
Stations 
Site and regional 
meteorological stations 
Average Upper Bound: 
Nogales, Arizona 
Hobbs, New Mexico 
Average Lower Bound: 
Site and regional 
meteorological stations 
Average Upper Bound: 
Spokane, Washington 
Rosalia, Washington 
St. John, Washington 
Average Lower Bound: 
Beowawe, Nevada 
Delta, Utah 
Climate Regime 
Present-Day 
Lower Bound 
Present-Day 
Upper Bound 
Monsoon 
Lower Bound 
Monsoon 
Upper Bound 
Glacial-Transition 
Lower Bound 
Glacial-Transition 
Upper Bound 
Source: 
Nogales, Arizona 
Hobbs, New Mexico 
Delta, Utah 
Beowawe, Nevada 
Rosalia, Washington 
Spokane, Washington 
St. John, Washington 
a USGS 2001b, Tables 6-8 and 6-12. 
b CRWMS M&O 1997b, Tables 2-1 and A-10. 
c USGS 2001b, Tables 6-4, 6-5, and 6-6. 
Location 
Yucca Mountain 
region 
Yucca Mountain 
region 
Yucca Mountain 
region 
No. 1: Climate and Infiltration 
Mean Annual 
Precipitation (mm) 
185.8a 
265.6a 
188.5a 
414c 
418c 
198c 
220c 
460c 
410c 
433c 
2-6 
Revision 1 
Locations of Meteorological Stations 
Yucca Mountain region 
West Longitude North Latitude 
110° 55' 31° 21' 
103° 08' 32° 42' 
Yucca Mountain region 
West Longitude North Latitude 
117° 32' 47° 38' 
117° 22' 47° 14' 
117° 35' 47° 06' 
West Longitude North Latitude 
116° 28' 29" 40° 35' 25" 
112° 35' 45" 39° 20' 22" 
Mean Annual 
Temperature (ºC) 
Range: 15.1 to 18.2b 
Range: 15.1 to 18.2b 
Range: 15.1 to 18.2b 
15.8c 
16.8c 
10.1c 
8.8c 
8.4c 
8.9c 
9.1c 
May 2004 Revision 1 
Figure 2-3. Generalized View of Atmospheric Circulation 
2.2 PRESENT-DAY CLIMATE 
Certain semipermanent atmospheric pressure features over and near the United States play 
dominant roles in controlling the present-day synoptic weather patterns that affect the Yucca 
Mountain area. These are the Bermuda High (western north Atlantic), the Eastern Pacific High 
(eastern north Pacific), the Aleutian Low (Gulf of Alaska, winter), and the summertime thermal 
low (southwestern United States) (Ahrens 1994, pp. 288 to 289, Figure 11.3). Seasonal changes 
in position and strength of these centers of action influence winds and the movement of storms. 
A generalized view of atmospheric circulation is given in Figure 2-3. 
Source: USGS 2001a. 
Great Basin precipitation arises primarily from three airflow trajectories and moisture origins in 
the Pacific, Gulf of Mexico, and continental regions. The most important to the Yucca Mountain 
area is the Pacific trajectory. The Sierra Nevada and the transverse mountain ranges of southern 
California can either block or impede atmospheric moisture transport from the Pacific Ocean to 
Yucca Mountain, especially during winter, leading to a rain-shadow effect. In summer, moisture 
usually arrives from the south, where blockage is less effective. 
In the winter, the eastward extension of the North Pacific High steers most cyclonic storms away 
from the southwestern United States by deflecting the jet stream toward more northerly latitudes. 
Thus, the desert southwest region experiences relatively few storm passages and tends to have 
mild, dry winter weather. On occasion, the North Pacific High weakens or moves, and the jet 
May 2004 2-7 No. 1: Climate and Infiltration Revision 1 
stream shifts to the south to help bring storms, precipitation, and cooler air into the southwest 
(Mock 1996, pp. 1,111 to 1,113). Though relatively infrequent, these storms are a very 
important contributor to the annual recharge and stream flow (e.g., at Owens River) in the 
southwest. Many Pacific cyclones affecting Yucca Mountain form in the vicinity of the Aleutian 
Low and arrive from the west or northwest, often on a trajectory that curves around or over the 
southern end of the Sierra Nevada. 
In Simulation of Net Infiltration for Modern and Potential Future Climates (USGS 2001b), the 
present-day climate scenario (sometimes referred to as the modern climate scenario) was 
characterized using the results of observations conducted from 1980 to 1995, with the mean 
annual precipitation rate of 157.5 mm/yr; and results from the Nevada Test Site Station 4JA 
100-year stochastic simulation (mean annual precipitation of 150.19 mm/yr). To represent the 
lower-bound present-day climate scenario, the driest 10-year period (from 1980 to 1990) from 
the 100-year observations at the 4JA precipitation station was used. Note that the 4JA Station is 
located at the elevation of 1,044 m (within the Nevada Test Site) and has a mean annual 
precipitation of 140 mm (USGS 2001b, Table 6-3). To characterize the present-day climate 
scenario, the average data for the 1980 to 1995 observations at the Nevada Test Site Station Area 
12 Mesa (located within the northern boundary of the Nevada Test Site known as Rainier Mesa) 
were used, resulting in a mean annual precipitation of 328 mm/yr (USGS 2001b, Table 6-3). 
These data were intended to represent the wetter conditions resulting from the enhanced El Niño 
Southern Oscillation activity. 
Present-day meteorological data for the upper and lower bounds come from available stations in 
the region and include both Yucca Mountain project and nonproject data. 
2.3 FUTURE CLIMATES 
2.3.1 Monsoon Climate 
Monsoon climates in the Yucca Mountain area can be characterized by increased summer and 
annual rainfall relative to today, generated by incursions of moisture from the tropical Pacific or 
the Gulf of California. Most precipitation would likely be summer-dominated. In addition, 
monsoon climate states are most likely wetter and warmer than today, with much of the 
precipitation lost to evapotranspiration and evaporation. 
In the region in the United States that currently experiences a strong summer monsoon, two 
meteorological stations with long-term records were chosen to represent the monsoon climate 
upper bound: Hobbs, New Mexico (418 mm), and Nogales, Arizona (414 mm). The selection of 
two meteorological stations was needed to obtain averaged data in order to minimize local 
meteorological effects. The lower-bound monsoon climate scenario was defined as being 
equivalent to the average present-day climate stations in southern Nevada (USGS 2001a). 
2.3.2 Glacial-Transition Climate 
The glacial-transition climate is characterized by displacement of the polar jet from a northerly 
latitude to the Yucca Mountain area during most, but not all, winters. Continental ice sheets are 
either in the early stages of growth (interglacial moving toward glacial) or nearing the final 
stages of retreat (glacial moving toward interglacial). Mountain glaciers most likely exist in the 
May 2004 2-8 No. 1: Climate and Infiltration Revision 1 
high Sierra Nevada cirques and may expand or contract on decadal or century scales. During 
some winters, the polar front resides to the north of Yucca Mountain, but it is still close enough 
to ensure ample precipitation for regular, seasonal Sierra Nevada stream discharge. Summer 
seasons are cooler than modern. 
The glacial-transition climate occurs during the transition from glacial to interglacial or from 
interglacial to glacial climate states. 
Selecting analog meteorological stations to represent upper and lower bounds of the glacialtransition 
climate requires identifying sites with cool winter wet seasons, and warm to cool and 
dry summers (USGS 2001a, Section 6.6.2). Further, the analog sites had to lie on the east side of 
large mountain ranges and, hence, in the rain shadow of those ranges. Based on the ostracode 
assemblage in Owens Lake, the absence of cold Canadian-like climate implies that the 
upper-bound analog should lie within the contiguous United States. 
The meteorological stations representing the lower-bound glacial-transition climate should be in 
a place where mean annual temperature is higher than the upper bound, and thus these stations 
would be south of the upper bound localities (USGS 2001a, Section 6.6.2). The mean annual 
temperature, however, should be lower than that for the Owens Lake Basin at present, so that 
effective moisture is higher, consistent with a full and overflowing lake. The stations should 
have a lower mean annual precipitation than the upper-bound sites, because the record from the 
Owens Lake Basin shows episodes of either saline diatoms or ostracodes, implying less surface 
flow in the Owens River. However, the absence of abundant saline taxa implies effective 
moisture is higher than at present, reflecting cooler than the present mean annual temperature 
rather than high mean annual precipitation (USGS 2001a, p. 74). Thus, the lower-bound 
glacial-transition meteorological sites may have mean annual precipitation values similar to or 
even lower than present-day Owens Lake Basin (USGS 2001a, p. 74). As with the upper-bound 
meteorological sites, the region should be winter-precipitation dominated, north of the summer 
rain regime, and have some or all of the ostracode or diatom species found in the fossil record at 
Owens Lake. Inspection of meteorological sites that fit lower-bound glacial-transition conditions 
revealed that there were few choices available (USGS 2001a, Section 6.6.2). The sets of 
meteorological data that fit all of these criteria and also have long and complete records were 
found at Delta, Utah, and Beowawe, Nevada. 
Because a full and overflowing Owens Lake is related to seasonal and possibly annual residence 
of the polar front, upper-bound meteorological stations were selected in an area where the polar 
front currently resides through most or all of the winter season and where its average position 
prevails for most of the year (USGS 2001a, Section 6.6.2). Given all of the above qualifying 
conditions, the upper-bound glacial-transition meteorological site was selected in the 
northwestern United States, east of the Cascades, a high mountain range. The area falls within a 
rain shadow, as does Yucca Mountain. The regional mean annual precipitation is 
winter-precipitation dominated and is under the influence of the polar front during the winter, as 
well as other times throughout the year. Furthermore, unlike localities farther north in Canada, 
the region does not experience extended dominance by extremely cold, Arctic high pressure, 
typical of the cold, full-glacial periods. Based on meteorological-station data from eastern 
Washington state, three stations were selected for the upper-bound glacial-transition climate: 
Spokane, Rosalia, and St. John (see Figure 2-2). These three stations are close to each other but 
May 2004 2-9 No. 1: Climate and Infiltration Revision 1 
do not have identical records, presumably reflecting local differences in mean annual 
precipitation and mean annual temperature. As in other cases, selection of multiple 
meteorological stations was intended to minimize local effects on the climate parameters used as 
input to the infiltration model. 
2.4 COMPARISON OF CLIMATE ANALOG SITES 
A comparison of the mean annual precipitation and mean annual temperature data from the 
meteorological analog sites is shown in Table 2-3. The mean annual precipitation for the 
lower-bound glacial-transition state is only slightly higher than that for the lower-bound presentday 
and monsoon states. The mean annual precipitation for the upper-bound monsoon and 
glacial-transition climates exceed that for the present-day climate (Thompson et al. 1999, 
Figures 16 and 17, Table 4). Thus, future climate, based on the selection of meteorological 
stations having upper-bound annual mean precipitation in the low 400-mm range is wetter than 
the present-day climate. The values of mean annual temperature for the glacial-transition analog 
climate are much lower than the mean annual temperature values in the Yucca Mountain and 
Owens Lake region today (Thompson et al. 1999). 
2.5 ANALYSIS OF FUTURE CLIMATE UNCERTAINTIES 
The reasons for the climatic uncertainties can be grouped into two categories (BSC 2002b, 
Section 4). Epistemic uncertainty arises from the lack of knowledge about the processes and 
parameters, because the data are limited, or because alternative conceptual interpretations of the 
available data exist (e.g., DeWispelare et al. 1993); this type of uncertainty can be reduced 
because the state of knowledge can be improved by further testing or data collection. As a 
consequence, this type of uncertainty is also referred to as reducible uncertainty. Aleatory 
uncertainty arises from the existence of spatial and temporal variability of climatic processes and 
parameters and typically can be accounted for using geostatistical approaches (e.g., using 
appropriate probability distribution functions); as a consequence, this type of uncertainty is 
referred to as irreducible uncertainty, which cannot be reduced through further testing or data 
collection. These two categories of uncertainties are introduced to demonstrate the causes for 
uncertainties in forecasting climate states, which were identified in reports (USGS 2001a; Sharpe 
2003, pp. 40, 57, 61): the timing of future climate, the methodology of climatic forecasting, and 
the earth’s future physical processes. 
Uncertainty in the timing is caused by the following factors (Sharpe 2003, p. 40): the 
determination of age from the Devils Hole records, caused by ambiguity in selecting a precession 
value that marks the Devils Hole ä18O inflection points; and the effect of a regional climate on 
the Devils Hole signal itself, if a primary inflection point precedes or follows a global change. 
This uncertainty relates, in general, to the issue of correspondence between the orbitally tuned 
and the terrestrial-based (Devils Hole) records (Winograd et al. 1992, pp. 255, 257 to 258), as 
well as Owens Lake chronology, based on the ostracode and diatom data (USGS 2001a, 
Section 6.6). 
Mean values of the time for the onset and end of climate states, which are determined based on 
the precession methodology, are reported to the nearest 500 years (Sharpe 2003, p. 59). Each 
Devils Hole sample integrates over an average time interval of about 1,800 years (Winograd 
2-10 May 2004 No. 1: Climate and Infiltration Revision 1 
et al. 1992, p. 255). The projected duration of the monsoon climate from 900 to 1,400 years 
(USGS 2001a, p. 76) is less than the Devils Hole age resolution of 1,800 years (Sharpe 2003, 
p. 56). The Devils Hole timing resolution is likely to have little effect on the duration of the 
glacial-transition climate (8,000 to 8,700 years) (USGS 2001a, p. 76) within the 10,000-year 
TSPA regulatory period. 
Timing uncertainty is both epistemic and aleatory, because timing is based on the current 
knowledge and the best available methods of climatic investigations of different data sets. 
However, timing uncertainty is not significant for the evaluation of net infiltration rate spatial 
distribution and net infiltration weighting factors for the following reasons: 
• Net infiltration was simulated (USGS 2001b) for each of the climate states (present-day, 
monsoon, and glacial transition) using steady state, cyclic boundary conditions; 
therefore, the spatially averaged infiltration rates for each of the climate regimes are 
essentially independent of the duration of the climate state (the same results were 
obtained for the simulation of percolation in the unsaturated zone) (BSC 2003b, 
Figures 6.8-3 to 6.8-5). Also, in TSPA simulations of radionuclide transport, the actual 
transient period during which the unsaturated zone flow system responds to a climate 
change has been specified as insignificant (DOE 1998, Section 3.6.1.1, p. 3-116). 
• The net infiltration weighting factors for the whole period of 10,000 years were 
evaluated using the meteorological data for the glacial-transition climate state (higher 
precipitation and lower temperature than modern) (BSC 2003a), granting a conservative 
estimation of net infiltration needed for the NRC safety analysis (NRC 1997, p. 8). 
Uncertainty in the methodology of climate analysis arises from several reasons. It arises from 
the hypothesis of using the past cyclic climate behavior to forecast the future climate states 
(USGS 2001a, p. 19; Sharpe 2003, p. 61), which presents epistemic (reducible) uncertainty; 
however, the validity of this hypothesis about the climate cycling pattern can be confirmed only 
by the passage of time (USGS 2001a, p. 19). Uncertainty arises from the possibility for earth’s 
orbital parameters to correlate with other factors that cause climate changes (e.g., solar output or 
other mechanisms) (USGS 2001a, p. 37). The use of a conventional numerical method, 
including a correlation analysis, is restricted, because even modern science does not know with 
certainty all reasons for climate changes, so that the change in a climate system cannot be 
described numerically and long-term time series cannot be constructed (climate change is 
time-transgressive and climate proxies do not often record climate events in the same manner or 
with the same magnitude), nor is there a clear knowledge about future boundary conditions 
needed for numerical modeling; consequently, the climate analysis is based on forecasting future 
climate (USGS 2001a, p. 62). The second and the third reasons present examples of a 
combination of both epistemic and aleatory uncertainties. 
NRC (Stablein 1997, p. 2; NRC 1997) acknowledged that the paleoclimatic and paleohydrologic 
methods of analysis can be used to bound the range of past climates in the Yucca Mountain 
region and to estimate future climatic conditions, and that climatic modeling is not required 
(NRC 1997, p. 6). 
May 2004 2-11 No. 1: Climate and Infiltration Revision 1 
Uncertainty also arises from the well-known fact of the chaotic nature of the climate system 
(NRC 1997, p. 9; Sharpe 2003, p. 61). The chaotic behavior is caused by a combined effect of 
several nonlinear dynamic processes, along with positive and negative feedback mechanisms; for 
example, input of cosmic material (meteorites), changes in solar radiation, ocean salinity, land 
features (e.g., topography, vegetation, albedo), and atmospheric composition (e.g., fossil fuel 
emissions, volcanic eruptions). For such a system, precise predictions of long-term climate 
changes are limited, while the range of long-term changes can be evaluated using stochastic 
methods, with the aid of probability distributions for such meteorological parameters as 
precipitation and temperature. Therefore, a modified Monte-Carlo (Latin Hypercube sampling) 
stochastic approach was used for the evaluation of the spatial distribution of net infiltration and 
net infiltration weighting factors (BSC 2003a), using analog-site precipitation and temperature 
records, and probability distributions for daily precipitation and evapotranspiration multipliers 
(see Section 6). 
The effects of anthropogenic physical processes, such as greenhouse effects, acid rain, global 
warming, and ozone layer depletion, are still unclear and highly uncertain (NRC 1997, pp. 11, 
13). Therefore, NRC (1997, p. 13) recommended that for the purposes of evaluating repository 
performance, it is both pragmatic and conservative to postulate that such changes will be 
relatively short (several thousands years at most) and that the global cooling (cooler and wetter 
conditions) will return. Such consideration was implemented in the evaluation of the net 
infiltration weighting factors for the 10,000-year performance period, using the data for the 
glacial-transition climate (BSC 2003a). This approach is also consistent with the NRC licensing 
regulations (10 CFR 63.305(b) and 10 CFR 63.305(c)), requiring the U.S. Department of Energy 
to vary factors related to the geology, hydrology, and climate based on cautious but reasonable 
assumptions consistent with present knowledge of factors that could affect the Yucca Mountain 
disposal system over the next 10,000 years; and assume that all those factors remain constant as 
they were at the time of submission of the license application (DOE 2002b). Therefore, the 
effects of human influence on climate are excluded from the TSPA calculations (BSC 2001a, 
Section 6.5.1 through 6.5.4). 
Because various climatic physical processes affecting net infiltration cannot be simulated 
explicitly for each grid of the regional Yucca Mountain infiltration model, physically based 
empirical relationships (e.g., the relationship between the precipitation and topographic 
elevation) were used for simulations of net infiltration (USGS 2001b). To account for the 
variability of meteorological parameters caused by the effects of relatively small-scale (local) 
physical processes, such as the effect of topographic features (mountains, valleys, and other 
terrestrial factors), the Yucca Mountain regional-scale net infiltration predictions were performed 
using the averaged meteorological data (precipitation and snow rates, temperature) from two or 
three meteorological stations (USGS 2001b). 
Although reports (USGS 2001a; Sharpe 2003, p. 57) identified several types of uncertainties in 
forecasting climate states, these reports concluded that these uncertainties are adequately 
addressed by the three climate states, the estimated duration of future climate states (based on the 
precession methodology) compares favorably with that of past climate states (based on the fossil 
and isotope records), and that this forecast presents the best possible scientific assessment for 
future climates. Because there is no simple or objective way of assessing the nature of future 
climate uncertainties, only future observations could be used to validate the climate forecasting. 
May 2004 2-12 No. 1: Climate and Infiltration Revision 1 
Source: Flint, A.L. et al. 1996, Figure 3. 
3.1 EFFECT OF TOPOGRAPHY 
Topography of the land cover is a very important factor affecting both a micro- and macroscale 
surface and subsurface hydrologic processes. It has direct feedbacks to energy and water 
systems, and determinations of boundary conditions related to the exchange of momentum, 
energy and mass between the atmosphere and the subsurface. In particular, the land topography 
affects the spatial distribution of surface (including surface flow channeling, runoff, and run-on) 
and subsurface water flow, solar radiation loading, potential evaporation, and evapotranspiration. 
3. CONCEPTUAL MODEL OF PROCESSES CONTROLLING NET INFILTRATION 
This section comprises the conceptual understanding of dominant factors and processes that 
control net infiltration at Yucca Mountain. These factors and processes include: surface 
topography, precipitation, evaporation and transpiration, run-on and runoff, the redistribution of 
moisture in the shallow subsurface, infiltration, percolation, and groundwater recharge, as well as 
the types of uncertainty involved in the evaluation of flow parameters and net infiltration. The 
schematic of the hydrologic cycle involving these factors and processes is shown in Figure 3-1 
(Flint, A.L. et al. 1996, p. 9). The mathematical relationships describing the effects of these 
processes will be presented in Section 4, which is devoted to numerical modeling of net 
infiltration. 
Figure 3-1. Schematic Illustrating Field-Scale Water Flow Processes Controlling Net Infiltration 
May 2004 3-1 No. 1: Climate and Infiltration Revision 1 
Proper land surface characterization is thus crucial for the evaluation of processes controlling net 
infiltration. Topography may cause the infiltration rate to vary by several orders of magnitude 
under different geomorphic settings (Scanlon et al. 1999). 
3.3 SURFACE HYDROLOGY 
Surface hydrology is one of the main factors affecting the amount of water entering the 
subsurface. Overall, the streamflow in the desert terrain of the Amargosa River basin is 
extremely erratic and sparse. The results of monitoring show that increased streamflow in 1983, 
1992, 1993, and 1995 caused groundwater levels to rise in observation boreholes in Fortymile 
Canyon (Savard 1998, p. 27). The data analysis of streamflow indicates that: 
• The flood of February 24 to 25, 1969, was the greatest runoff event since the beginning 
of streamflow recording in the early 1960s. 
• Unit peak discharges typically are greater in small drainage basins than in larger basins. 
• Maximum peak discharges do not correlate closely with maximum unit peak discharges. 
The types of flooding at Yucca Mountain depend on the amount of precipitation. For example, 
the 1995 streamflow was dominated by high-magnitude runoff of relatively short duration in 
Beatty Wash and Fortymile Wash, probably enhanced by localized precipitation on snowpack in 
the upper altitudes of the Nevada Test Site. In 1998, sustained regional precipitation caused 
lower magnitude streamflows of longer duration in Fortymile Wash, Amargosa River, and their 
major tributaries. In both floods, much or all of Fortymile Wash and the Amargosa River flowed 
simultaneously. In March 1995, water in Fortymile Wash and Beatty Wash flowed off Nevada 
Test Site; in February 1998, water in Topopah Wash also flowed off Nevada Test Site. In 1995 
and 1998, the Amargosa River flowed from its headwaters to its terminus in Death Valley. Both 
3.2 FRACTURE ROCK CHARACTERISTICS 
At Yucca Mountain, net infiltration is mostly formed at the bottom of the root zone located 
within the first 6 m of depth, either within the soil or the fractured tuff of the TCw hydrogeologic 
unit (Flint, A.L. et al. 1996). The presence of the bedrock is considered to control net infiltration 
if the bottom of the root zone, where the net infiltration is originated, is located within the TCw 
hydrogeologic unit, which is significantly fractured. The most important characteristics of 
fractured rock that could affect net infiltration are fracture length, fracture spacing, fracture 
connectedness, fracture aperture, and fracture roughness. The degree of tuff welding determines 
the fracture characteristics, which, in turn, affect how water moves through each particular unit. 
Small fractures (less than 1 m in length) comprise about 50 percent of the total number of 
fractures, and the number of fractures less than 2 m comprises about 75 percent of the total 
number of fractures (Hinds et al. 2003). The effect of small-scale fractures, comprising a 
significant portion of the total fractures, is incorporated into the large-scale Yucca Mountain 
numerical model (with gridblock size exceeding 100 m) using calibrated unsaturated hydraulic 
parameters (Wu et al. 1999). At Yucca Mountain, geologic features such as faults may serve 
either as capillary barriers or as major conduits, focusing near-surface downward flow, 
depending on the degree of saturation and hydraulic conductivity of the infilling material. 
3-2 May 2004 No. 1: Climate and Infiltration Revision 1 
the 1995 and 1998 floods indicate, therefore, that the Amargosa River, with contributing 
streamflow from one or more among Beatty, Fortymile, and Topopah Washes, has the potential 
to transport dissolved and particulate material well beyond the boundary of the Nevada Test Site 
and Yucca Mountain areas during periods of moderate to severe streamflow. The effects of 
surface floodings on infiltration and, in particular, preferential flow in the unsaturated zone are 
discussed in Section 3.8. 
3.4 PRECIPITATION 
Precipitation at Yucca Mountain can be characterized using the precipitation time series recorded 
at a number of meteorological stations shown in Figure 3-2. As an example, Figure 3-3 (USGS 
2001b, Figure 6-18) shows the daily variations of precipitation over the period from 1980 to 
approximately 1995 characterizing the present-day climate conditions (these data were used for 
model calibration—see Section 4). Figure 3-4 shows an example of the Tule Lake, California, 
precipitation time series used for simulations of the mean glacial-transition climate (DTN: 
GS000308311221.010). 
3.5 TYPES AND DEPTHS OF SOILS 
The areal distribution of estimated soil depth in the Yucca Mountain region is shown in 
Figure 3-5. Soil textures range from gravelly to cobbled, loamy sands to sandy loams. Soils are 
calcareous (high in calcium carbonate), with lime coatings on the undersides of rocks in the 
subsoil layer. These soils are moderately to strongly alkaline, with pH ranging from 8.0 to 8.6. 
Rock fragments ranging in size from gravel to cobbles dominate 45 to 65 percent of the ground 
surface. Soils include high clay content and thin calcium-carbonate layers, and the underlying 
bedrock is moderately to densely welded and moderately to highly fractured. The presence of 
rock fragments in soil layers can impact measured hydraulic properties. Variation of soil 
hydraulic properties influences the amount, distribution, and routing of overland flow. 
3.6 DEPTH OF THE ROOT SYSTEM 
The depth of the root zone is determined by vegetation, which controls the depth distribution of 
both net infiltration and evapotranspiration (i.e., the sum of evaporation and plant transpiration). 
Vegetation cover can play a major role in hydrologic systems, contributing to recycling of 
moisture between the atmosphere and soils. Rooting depths for shrubs typically range from 0 to 
50 cm (Wallace et al. 1980). Wirth et al. (1999) compiled ranges of maximum rooting depths for 
various vegetation types under current and future (glacial-transition) climates for a study related 
to the waste disposal of transuranic waste in greater confinement disposal boreholes at the 
Nevada Test Site (Cochran et al. 2001). Wirth et al. (1999) report a range of maximum rooting 
depths of 0.35 to 17.4 m for shrubs under current and future climates, and a range of maximum 
rooting depths of 0.2 to 30 m for trees under a future climate. Under a future cooler, wetter 
climate at the Nevada Test Site, areas currently inhabited by shallow rooting shrubs may be 
replaced by deeper rooting pinon–juniper communities currently found at the upper elevations of 
the Nevada Test Site (McCurley 2003). Junipers, for example, have a rooting depth ranging 
from 6 to 60 m. 
May 2004 3-3 No. 1: Climate and Infiltration Revision 1 
Source: CRWMS M&O 2000b. 
Figure 3-2. Locations of Yucca Mountain Site Characterization Project Meteorological Monitoring Sites 
May 2004 3-4 No. 1: Climate and Infiltration Source: USGS 2001b, Figure 6-18. 
Figure 3-3. Daily Precipitation Record Characterizing Present-Day Precipitation (1980 to 
Approximately 1995) 
Source: DTN: GS000308311221.010. 
Figure 3-4. Tule Lake, California, Precipitation Time Series Used for Simulations of Net Infiltration for 
the Glacial-Transition Climate in Analysis of Infiltration Uncertainty (BSC 2003a) 
No. 1: Climate and Infiltration 3-5 
Revision 1 
May 2004 Source: USGS 2001b, Figure 6-16. 
Figure 3-5. Estimated Soil Depth in the Yucca Mountain Region 
No. 1: Climate and Infiltration 
Revision 1 
May 2004 3-6 Revision 1 
Vegetation root-zone parameters are expected to vary with changes in climate conditions. For 
example, for the monsoon climate, the root-zone parameters should be modified to reflect greater 
vegetation density and changes in vegetation type caused by wetter climate conditions 
(USGS 2001b, Section 6.9.4). 
3.7 EVAPOTRANSPIRATION 
To assess the evapotranspiration rate, field observations were conducted for several years near 
Yucca Mountain (Flint, A.L. et al. 1996, p. 59, Figure 29). The data obtained from these 
observations showed strong advective air transport effects in the arid environment at Yucca 
Mountain. The data also indicated considerable variability in evaporation, with observed 
evaporation exceeding calculated potential evaporation by about 100 percent. The data also were 
used to establish the upper limit of available energy for potential evapotranspiration modeling, 
given that modeled potential evapotranspiration should not exceed observed evaporation. 
Potential evapotranspiration rates range from a minimum of 500 mm/yr to a maximum of 
900 mm/yr, depending mostly on the land-surface slope and aspect. Minimum values of 
evapotranspiration occur at steep northerly and northeasterly facing slopes, particularly at 
locations surrounded by blocking topography. 
During winter, potential evapotranspiration is at a minimum because of shorter days, lower sun 
angle, and lower air temperatures, and root activity is either diminished or dormant. Moreover, 
after large storm or snowmelt events, water can penetrate deeper and accumulate in the root zone 
more rapidly than it can be removed by evapotranspiration. Therefore, large values of potential 
evapotranspiration do not prevent the events of episodic infiltration of precipitated or snowmelt 
water into the subsurface, causing preferential and transient flow phenomena in the shallow part 
of the unsaturated zone, as discussed in the following section. 
3.8 PREFERENTIAL AND TRANSIENT INFILTRATION PHENOMENA 
One of the most important features related to our conceptual understanding of infiltration at 
Yucca Mountain is the possibility of preferential flow. Field and modeling data provide strong 
evidence supporting the notion of fast, transient, and preferential flow through rock 
discontinuities at Yucca Mountain. Figure 3-6 shows a schematic presentation of zones of 
preferential flow (weeps). Figures 3-7 through 3-9 demonstrate evidence of fast, preferential 
flow based on the variations of the measured water content in boreholes. For example, 
Figure 3-7 demonstrates that during the period from 1984 to 1995, episodic events of preferential 
flow (shown by the increased moisture content) occurred eight times, as measured in UE-25 
UZN#1, with propagation of moisture to depths generally between 1 and 2.5 m but reaching 
11 m in 1995, following a significant storm event (see Section 3.3). Measurements from USW 
UZ-N15 demonstrate that water could propagate to a depth of about 10 m, following flooding, 
even without a noticeable increase in the moisture content within a shallow bedrock layer 
(Figure 3-8). Figure 3-9 illustrates how a zone of increased moisture content measured in UE-25 
UZN#63 (which reached at a depth of 2 m) disappeared within the following 6 months. 
3-7 May 2004 No. 1: Climate and Infiltration Revision 1 
Source: Pruess et al. 1999. 
Figure 3-6. Schematic of the Weeps Model for Significant Fracture Flow at Yucca Mountain 
Source: Flint, A.L. et al. 1996, Figure 31. 
Figure 3-7. Measured Water-Content Profiles at Borehole UE-25 UZN#1 from 1984 through 
Approximately 1995 
May 2004 3-8 No. 1: Climate and Infiltration Revision 1 
Source: USGS 2001b, Figure 6-4. 
Figure 3-8. Measured Water-Content Profiles at Borehole USW UZ-N15 for 1993 to 1995 
Source: CRWMS M&O 2000b, Figure 8.2-11. 
Figure 3-9. Measured Water-Content Profiles in Borehole UE-25 UZN#63 for 1993 through 1995 
May 2004 3-9 No. 1: Climate and Infiltration Revision 1 
The presence of preferential flow through a fracture system is also supported by the analysis of 
estimates of space-averaged net infiltration (Flint, A.L. et al. 1996) and percolation through the 
mountain. While field estimates indicate that the infiltration rate is on the order of 1 to 10 mm/yr 
(Bodvarsson and Bandurraga 1996), the calculations based on the flow through the welded tuff 
matrix show the infiltration rate on the order of one or at most several millimeters per year. 
99 
A concept of preferential flow is also supported by environmental isotope data, calcite 
deposition, and chloride data. For example, the environmental isotopes, in particular 36Cl and 
Tc, have been reported at the repository level in the Exploratory Studies Facility (ESF) by 
Fabryka-Martin et al. (1996), and elevated concentrations of tritium have been found within the 
Calico Hills unit (Yang et al. 1996). (Note that based on the results of recent investigations 
(BSC 2003c), validation studies of the 36Cl data are ongoing (Section 5.2) (Paces et al. 2003).) 
The presence of these isotopes at the repository level suggests the presence of preferential flow 
paths within the unsaturated zone. 
Calcite coating data show that calcite deposition mostly occurs within the fractures (Paces et al. 
1996). Very little evidence of calcite coatings or other similar deposits have been found within 
the matrix, indicating very low diffusion into the matrix. 
The investigations of chloride concentrations within the mountain also confirm the idea of 
preferential, fracture-dominated infiltration processes (Sonnenthal and Bodvarsson 1999). 
Chloride concentrations within the shallow Tiva Canyon unit are generally lower than 10 mg/L; 
they range from 30 to 80 mg/L in the nonwelded Paintbrush unit, and decline again to 5 to 
10 mg/L in the deep perched-water bodies found above or within the zeolitic portion of the 
Calico Hills unit (Yang et al. 1996). This suggests predominant fracture flow within the Tiva 
Canyon Tuff. Chemical concentrations suggest that the component of matrix flow to the 
perched-water bodies is rather small. 
Significant infiltration into the mountain may occur only every five years or so (Flint, A.L. et al. 
1996). In the extra wet years, infiltration in major drainages (where flow is concentrated) can 
increase to hundreds of millimeters per year during a relatively short time of perhaps not more 
than one week. Because of the fractured nature of the Tiva Canyon unit, rapid infiltration pulses 
may not dampen significantly before reaching the bottom of the Tiva Canyon unit. For example, 
A.L. Flint et al. (1996) reported a water pulse that originated during the 1995 wet winter season 
and was detected a year later. Neutron borehole data in shallow alluvium and the Tiva Canyon 
Tuff showed slower movement of the water fronts, with velocities as low as only a few 
millimeters per year (Flint, A.L. et al. 1996). However, after the water has entered permeable 
and well-connected fracture pathways, its migration velocity may be on the order of tens or even 
100 mm/yr. Flow through isolated fast-flow paths in the near-surface units might have an 
insignificant attenuation mechanism (even though isolated flow paths carry a small amount of 
water). 
Infiltration tests conducted at Fran Ridge (Nicholl and Glass 2002) also provide clear evidence of 
preferential flow at Yucca Mountain (Figure 3-10). Moreover, tests demonstrated a flow 
structure that appears remarkably similar to gravity-driven fingers observed during the 
experiments conducted by Nicholl et al. (1993). 
May 2004 3-10 No. 1: Climate and Infiltration Revision 1 
The Yucca Mountain results related to preferential flow phenomena are similar to those gathered 
at a number of sites throughout the world, indicating funneling and divergence of highly 
localized and extremely nonuniform water flow paths (Yamamoto et al. 1993; Nativ et al. 1995; 
Faybishenko et al. 2000; Nicholl and Glass 2002). 
Source: Nicholl and Glass 2002, Figure 4.10. 
NOTE: Maps were referenced to a portable 2.44 �~ 2.44 m (8�Œ �~ 8�Œ) grid subdivided at 0.305 �~ 0.305 m (1�Œ �~ 1�Œ) 
intervals. The blue tracer clearly marks both vertical and subhorizontal fractures. In the upper left-hand 
side of the image, tracer clearly extends outside of the area cleaned for mapping and disappears into the 
rubble-covered region. On the right hand side of the image, drill pipe was left standing in one of the vertical 
locator holes (blue arrow) used to position the grid. Locator holes were dry drilled following infiltration with 
the air-driven jackleg seen in the upper right-hand corner of the image. 
Figure 3-10. Photograph Showing the Distribution of Tracer and Fractures That Were Mapped Directly 
beneath the Infiltration Test Site (Scale of 1:12) at Fran Ridge 
May 2004 3-11 No. 1: Climate and Infiltration INTENTIONALLY LEFT BLANK 
3-12 No. 1: Climate and Infiltration 
Revision 1 
May 2004 Revision 1 
4.1 MODELING APPROACH 
4. SIMULATION OF NET INFILTRATION FOR PRESENT-DAY AND POTENTIAL 
FUTURE (MONSOON AND GLACIAL-TRANSITION) CLIMATIC CONDITIONS 
(Eq. 4-1) 
(Eq. 4-2) off 
May 2004 4-1 
4.1.1 Basis of the Modeling Approach 
This section describes the basis for the modeling approach used for simulating net infiltration 
with the INFIL code, including the water balance equation, and describes the concepts of 
piston-like flow, a bucket-type model, and field capacity. To perform the modeling, 
meteorological conditions were defined based on the cyclic behavior of precipitation, 
temperature, and surface water flow for each climate state. 
Water Balance Equation–Because of the complexity of processes controlling net infiltration 
(discussed in Section 3) and the significant size of the Yucca Mountain study area, the evaluation 
of net infiltration is based on the water-balance approach that requires measurements and lumped 
(distributed) estimates of basin-wide (averaged over a watershed) precipitation, snowpack depth 
and density, stream discharge, and evapotranspiration (Lichty and McKinley 1995, pp. 4 to 10; 
Flint, A.L. et al. 1996). The water-balance model is based on the principle of the conservation of 
mass for water over a particular time interval (e.g., day or year) and volume (area and depth) of 
the soil. For a watershed area with depth interval from the surface to the bottom of the root zone 
(Flint, A.L. et al. 1996, pp. 9 to 11) the water balance equation is given by (USGS 2001b): 
P+A+U+Ws+Ss+Bs+Li+Ron-Roff-I-E-T-Lo-Ex =0 
i is lateral inflow, Ron is surface run-on, Roff is surface runoff, I is net infiltration 
where P is precipitation, A is applied water (human induced), U is upward flow, Ws is the change 
in soil-water storage, Ss is the change is surface-water storage, Bs is change in biomass-water 
storage, L 
(drainage or percolation), E is evaporation, T is transpiration, Lo is lateral outflow, and Ex is 
water extraction (human induced). 
Taking into account the run-on, runoff, and evapotranspiration (ET) (which is the sum of E and 
T), and neglecting other terms of Equation 4-1, net infiltration (I) can be determined by 
I = P – SF + IRon + SM + SW – SB – ET – R 
where I is net infiltration, P is precipitation (rain and snow), SF is snowfall, SB is sublimation, 
SM is snowmelt, SW is change in water-content storage within the root zone, ET is 
evapotranspiration, IRon is infiltrated surface-water run-on, and Roff is surface-water runoff 
generated by excess precipitation, snowmelt, or run-on. 
Figure 4-1 schematically depicts the vertical profile discretization used for calculations of net 
infiltration. The left column of Figure 4-1 illustrates that for a shallow soil–bedrock interface, 
the bottom of the root zone extends into the bedrock (the propagation of the root zone into the 
bedrock was calculated using an equation from the USGS (2001b, Equation 17)), and the soil 
layer is divided into two sublayers (used for calculations of infiltration and evapotranspiration— 
see the discussion below on the bucket-type model). The middle column of Figure 4-1 indicates 
that as the depth to soil–rock interface increases, the propagation of the root zone into the 
No. 1: Climate and Infiltration Revision 1 
bedrock decreases, and the soil layer is divided into three sublayers. The right column of 
Figure 4-1 illustrates that net infiltration is formed at a depth of 6 m, within the soil unit, if the 
bedrock unit is below the 6-m depth. 
Source: Hevesi et al. 2003. 
Figure 4-1. Conceptual Model of Net Infiltration Illustrating the Layered Root-Zone Water-Balance 
Model of the Death Valley Region, Nevada and California 
Piston-Type Flow–This model assumes the one-dimensional downward progression of a sharp 
wetting front within a soil profile caused only by advection, with the evaluation of the water 
transport time for the leading edge of the plume based on the center of mass propagating through 
diffusion and dispersion. The application of a piston-type flow model is warranted for the 
evaluation of net infiltration at Yucca Mountain for areas that have root zones in unconsolidated 
soils and alluvium deposits (see the middle column of Figure 4-1); in these areas, the preferential 
flow is minimal. The application of this model is also appropriate for the areas that have root 
systems penetrating into the rock layer (see the left column of Figure 4-1) for the following 
reasons: episodic events of preferential flow usually occur within small local areas and on a short 
time scale; however, on a macroscale over sufficiently long times, pulses of preferential flow 
may average out so that they can be approximated by piston flow (Sukhija et al. 2003). 
Furthermore, parameters of the model described by Equation 4-2 were calibrated using field 
observations, as described in Section 4.2. 
May 2004 4-2 No. 1: Climate and Infiltration Revision 1 
Bucket Model and the Concept of Field Capacity–To reproduce the interaction between the 
layers within the soil–rock system, a bucket-type model was used. A “bucket” water-balance 
model used in calculations presents a series of uniform soil layers from which water is available 
for evapotranspiration or infiltration. The model assumes that downward infiltration into the 
lower layer begins when the water content in the upper layer reaches the field capacity. The field 
capacity is the water-content threshold of the near-surface soil profile (i.e., the root zone), at 
which drainage becomes negligible (several orders of magnitude less than the saturated flux rate) 
(Jury et al. 1991, p. 150). Field capacity is an old soil-physics concept intended to provide a 
characteristic index of how much water may be retained from a rainfall event after redistribution 
has ceased (Hillel 1982, pp. 243 to 248). In coarse-textured soils, such as those at Yucca 
Mountain, the drainage rate usually falls to an insignificant level within a few days, after which 
the water content changes at a slow rate. For thick soils, the saturation corresponding to the field 
capacity at a depth of 6 m would occur only at areas of localized surface-water flow, such as 
active stream channels and the base of steep side slopes. Model calibration showed that for 
upland areas with thin soils, the infiltration rate below the root zone is equal to the saturated 
hydraulic conductivity of bedrock (for filled fractures with the aperture of 250-µm fractures). 
The water potential that corresponds to the volumetric water content measured at field capacity is 
considered to be -0.1 bars (USGS 2001b, Attachment IV). While the use of the field capacity 
concept reduces the infiltration during dry periods, calculations of net infiltration during the wet 
periods using the rock saturated hydraulic conductivity overestimates the water flux during the 
periods of infiltration. As a result, the time-averaged infiltration rate closely matches the 
experimental data. 
Two competing processes affect soil moisture in the bucket model: (1) the model lacks canopy 
constraints in the release of soil water, leading to rapid evaporation of soil water, and (2) no loss 
of water from runoff occurs until the soil is saturated, and no rapid evapotranspiration occurs due 
to canopy interception of precipitation. For a bucket model, the soil water content decreases 
immediately as evaporation occurs (rapid response scheme), while for a multilayer scheme, 
water in deep soil layers is reduced through diffusion and then evaporates from the surface. 
Although the bucket model and the concept of field capacity present a simplification of the real 
soil–rock system, the bucket model could be used for regional-scale and long-term predictions to 
capture the essential features of the regional behavior of soil moisture and infiltration. 
Runoff Model–The model used to calculate the runoff is empirical. This model implicitly 
includes such effects as the two-dimensional surface flow and the differences in types of slopes 
(converging or diverging (Salvucci and Entekhabi 1995)). 
Evapotranspiration Model–Based on a thorough literature review and testing of seven types of 
models assessing evapotranspiration and bare soil evaporation under arid conditions, Levitt et al. 
(1996) concluded that the Penman-Monteith and Priestley-Taylor models appear to describe 
experimental data better than other models for both short- and long-term data sets. These 
two models require the use of the moisture content of soils. Levitt et al. (1996) also found that 
these models provide a better description of experimental data for relatively high 
evapotranspiration (i.e., greater than 1 mm/day). The modified Priestley-Taylor model used in 
INFIL is based on the empirical relationship for evapotranspiration and the climate analog site 
temperature data (USGS 2001c). 
May 2004 4-3 No. 1: Climate and Infiltration For the monsoon climate, the evaluation of evapotranspiration was based on modified root-zone 
parameters reflecting greater vegetation density and changes in vegetation type for the wetter 
climate conditions (USGS 2001b, Section 6.9.4). For the upper-bound monsoon climate, the 
root-zone weighting parameters were modified to approximate a 40 percent vegetation cover (as 
compared to 20 percent for present-day climate) and the maximum thickness of the modeled 
bedrock root-zone layer was increased from 2 to 2.5 m (6.6 to 8.2 ft). 
Alpha 
(1/Pa) 
4.1.2 Unsaturated and Saturated Hydraulic Parameters 
The unsaturated hydraulic parameters (used in calculations of the water retention and unsaturated 
hydraulic conductivity functions) were determined from rock outcrop samples (Flint, L.E. et al. 
1996; Flint, L.E. 1998) and through calibration of the INFIL model (USGS 2001b). The 
hydraulic properties determined using calibrations of the numerical model are fracture and 
matrix permeability (kf and km), and the fracture and matrix van Genuchten ƒ¿ and m parameters 
(ƒ¿f, ƒ¿m, mf, and mm). A summary of hydraulic parameters (for soils only) used in these 
simulations of net infiltration is given in Table 4-1. 
Table 4-1. Summary of Soil Properties Used as Input for INFIL V2.0 
Soil 
Unit 
1 
Saturated 
Hydraulic 
Conductivity 
(simulated, 
m/s) 
5.6 �~ 10.6 
n 
0.00052 1.24 
0.00062 1.31 1.2 �~ 10.5 2 
0.00066 1.36 1.3 �~ 10.5 3 
0.00087 1.62 3.8 �~ 10.5 4 
0.00056 1.28 6.7 �~ 10.6 5 
0.00074 1.40 2.7 �~ 10.5 6 
0.00055 1.26 5.6 �~ 10.6 7 
Porosity 
(%) 
36.6 
31.5 
32.5 
28.1 
33.0 
33.9 
37.0 
32.2 
(%) 
10.5 
11.6 
18.7 
21.9 
15.2 
11.7 
17.1 
19.1 0.00055 1.30 5.7 �~ 10.6 9 
Source: USGS 2001b, Table IV-4. 
The saturated hydraulic conductivity (used as a coefficient in the van Genuchten model of 
unsaturated hydraulic conductivity and direct evaluation of net infiltration) was determined from 
a series of air-injection tests and the calibration of the INFIL model. The results of field 
air-injection tests designed to determine the air permeability of rocks could be used to assess the 
upper limits of the saturated hydraulic conductivity values (needed for the net infiltration 
uncertainty analysis). For the Tiva Canyon welded unit, the air-permeability values are 
summarized by Patterson et al. (1996) and Rousseau et al. (1997). Figure 4-2 shows the spatial 
distribution of the saturated hydraulic conductivity of bedrocks and soils, which was determined 
from the calibration of the numerical model and was then used in the evaluation of net 
infiltration. 
Rock 
Fragments 
4-4 No. 1: Climate and Infiltration 
Revision 1 
Water 
Content at 
.60 bars 
Water 
Potential (%) 
5.4 
2.3 
1.7 
0.2 
3.5 
1.1 
4.6 
Water Content 
at .0.1 bar 
Water 
Potential 
(%) 
24.2 
17.3 
16.3 
7.3 
20.0 
15.0 
23.4 
18.9 
Bulk 
Density 
(g/cm3) 
1.60 
1.73 
1.70 
1.81 
1.69 
1.66 
1.58 
1.72 2.8 
May 2004 Source: USGS 2001b, Figure 6-15. 
Figure 4-2. Estimated Field-Scale Saturated Hydraulic Conductivity of Bedrock or Soils Underlying 
the Root Zone 
No. 1: Climate and Infiltration 
Revision 1 
May 2004 4-5 Revision 1 
In the TCw hydrogeologic unit, high values of permeability are associated with fractures and 
major faults (Kwicklis 1999; Ahlers et al. 1996, Section 3). Faults at Yucca Mountain have 
properties that cause them to function either as highly permeable conduits or as low-permeability 
barriers (Patterson et al. 1996, pp. 51 and 52), depending on the degree of saturation of infilling 
material and boundary conditions (e.g., precipitation and run-on). Presumably, the highly 
fractured breccia zones are very permeable, whereas fault gouge along the base of the faults 
seems to create low-permeability barriers to flow. Thus, permeability of fractured rock 
surrounding the monitoring boreholes incorporates the breccia zone effects as larger block 
permeability and the effects of gouge as smaller block permeability. 
The fracture permeabilities are generally in the range of 1 to 10 darcies (LeCain 1997), while 
fault permeabilities are on the order of 10 to 100 darcies (Ahlers et al. 1996, Section 3). In 
contrast, the matrix permeabilities of those units are on the order of 1 microdarcy (Flint, A.L. 
et al. 1996). For comparison, the nonwelded units at Yucca Mountain, including the PTn and 
CHn hydrogeologic units, have matrix permeabilities on the order of several hundred 
millidarcies, and these units contain much less fracturing (Flint, A.L. et al. 1996; Ahlers et al. 
1996). 
To better understand the processes governing flow in the TCw hydrogeologic unit and to 
determine the TCw hydrogeologic unit hydraulic properties, an infiltration and tracer transport 
test was performed in the ESF Alcove 1 (BSC 2001b, Section 6.8.1) near the North Portal of the 
ESF in the upper lithophysal zone of the Tiva Canyon Tuff (Tpcpul) unit (Flint, L.E. 1998, p. 3). 
The Tpcpul unit extends above the alcove to the ground surface, with the crown of the drift 
approximately 30 m below the ground surface. The infiltration test was conducted by supplying 
water at the ground surface directly over Alcove 1. For both phases, the upper boundary 
condition for water flow remained essentially the same. The seepage into the alcove and the 
tracer arrival time were recorded. The test consisted of two phases: Phase I was performed from 
March to August in 1998, and Phase II was performed from January to June in 1999. During 
Phase II, conservative bromide tracer was also introduced into the infiltrating water. To analyze 
the Alcove 1 test results, hydraulic properties for fractures and the matrix were assumed to be 
homogeneously distributed within the model domain (BSC 2001b, Section 6.8.1). Figure 4-3 
shows a comparison between observed seepage rate data for Phase I of the test and the 
simulation result from model calibration with ITOUGH2 (version 3.2) (BSC 2001b, 
Section 6.8.1). Although arrival times of three major peaks in the Phase I seepage rate data are 
matched, large differences exist between the simulated and observed seepage rate values at these 
peaks. The hydraulic parameters calculated from the Alcove 1 test were used in Analysis of 
Infiltration Uncertainty (BSC 2003a). The results of numerical modeling using the results of 
Alcove 1 observations can also be used for validation of continuum and active-fracture model 
approaches for modeling flow and transport in unsaturated fractured rock (Liu et al. 1998). 
May 2004 4-6 No. 1: Climate and Infiltration Revision 1 
Source: Liu, H-H. et al. 2003. 
4.1.3 INFIL Model 
The INFIL code represents a family of storage routing codes (that assume that gravity is the only 
driving force in water movement), with approximation of evapotranspiration as a sink. Net 
infiltration is determined in the INFIL code as water flow below the root zone. The root zone is 
defined as the zone from the ground surface to a certain depth in soil or bedrock, from which 
infiltrating water is removed by evapotranspiration (USGS 2001b, p. 20). 
The INFIL V2.0 numerical algorithm represents the solution of Equation 4-2. In this model, the 
downward infiltration rate through the root zone depends primarily on the storage capacity of the 
root zone, the field capacity, and hydraulic conductivity of the soil and bedrock. The storage 
capacity of the soil is defined as the porosity minus the residual water content, multiplied by the 
soil depth. 
The INFIL code consists of three main loops for performing a daily simulation of net infiltration 
over all model cells across a watershed model domain. Figure 4-4 (USGS 2001b, Figure 6-7) 
provides a flow chart illustration of the general model algorithm and the primary loop, which is 
driven by the daily climate input file and carries the simulation through the time domain. A 
Figure 4-3. Comparison between Observed Seepage Rate Data for Phase I of the Alcove 1 Test and 
the Simulation Result from Model Calibration with ITOUGH2 
May 2004 4-7 No. 1: Climate and Infiltration Revision 1 
grid-cell loop (nested within the primary loop) performs daily water-balance calculations within 
each layer of the root zone at each grid cell. The root zone is subdivided into layers based on the 
estimated maximum depth of bare-soil evaporation and an estimated variation in root density. 
The daily root-zone water balance consists of simulating precipitation, snowmelt, sublimation, 
evapotranspiration, water content for each root-zone layer, net infiltration, and runoff generation. 
An hourly loop (nested within the water-balance loop) is used for modeling potential 
evapotranspiration, based on the simulation of incoming solar radiation and effects on total solar 
radiation resulting from blocking ridges. After the completion of the water-balance loop, a 
surface-water flow-routing subroutine is called if runoff is generated at any grid cell. 
Surface-water flow is routed at the end of the day as a time-independent (instantaneous) total 
daily flow depth across each grid cell. The routing algorithm connects all grid cells horizontally, 
using surface-water flow-routing parameters included in the geospatial parameter input file. 
Surface-water flow is coupled to the water-balance calculation by allowing surface water to 
infiltrate into downstream grid cells according to the available root-zone storage capacity, soil 
hydraulic conductivity, and estimates for effective surface-water flow area and stream flow 
duration. Infiltrating water is added to the grid cell’s antecedent root-zone water-content term 
used in the following day’s water-balance calculation. The surface-water flow depth routed 
across the grid cell defining the outflow location of the watershed is converted to a daily mean 
discharge flow rate, which can be compared to measured stream flow for model calibration. 
Water infiltrating and percolating through the multilayered root-zone system is modeled as a 
cascading piston-flow process. Downward percolation is modeled as a “forward” cascade flow 
initiated by adding the volume of water into the topsoil root zone to the antecedent water in the 
topsoil layer. The water content is then compared with the field capacity for each grid cell. The 
volume of water exceeding the field capacity percolates downward to the underlying layer, and 
the new water content of the underlying layer is compared against the field capacity of that layer. 
If the potential percolation rate exceeds the saturated soil hydraulic conductivity (or the saturated 
bulk bedrock hydraulic conductivity) of the underlying layer, the downward percolation rate is 
set equal to the saturated hydraulic conductivity of the underlying layer, and the excess water 
volume is added to a temporary storage term for the overlying layer. The process is repeated for 
each soil and bedrock layer in the root zone until the bottom layer is reached. (For simulation of 
net infiltration, a maximum of three soil layers and one bedrock layer were used.) The volume 
of water exceeding the bedrock storage capacity is the potential net infiltration volume. 
May 2004 4-8 No. 1: Climate and Infiltration Source: USGS 2001b. 
Figure 4-4. Flow Chart of the Model Algorithm Used for Simulating Net Infiltration 
No. 1: Climate and Infiltration 
Revision 1 
May 2004 4-9 Revision 1 
For locations with the soil depth of 6 m or greater, the underlying bedrock properties are defined 
using alluvium¡Vcolluvium properties. Based on analysis of neutron moisture meter data (Flint, 
L.E. and Flint 1995), the maximum depth of infiltration in nonchannel alluvial locations is 6 m; 
therefore, bedrock properties are not taken into account. The net infiltration is calculated after 
evapotranspiration is simulated throughout the root zone and limited by the bulk saturated 
hydraulic conductivity of the underlying rock type. The potential net infiltration volume 
exceeding the bulk saturated hydraulic conductivity is added to the temporary storage term of the 
bottom root-zone layer. Starting with the bottom root-zone layer, a reverse cascade is performed 
to determine if runoff is generated. The volume of water in the temporary storage term is 
compared against the total storage capacity of each layer defined by the porosity (or effective 
fracture porosity in the case of bedrock) and layer thickness. If the volume of water in the 
temporary storage term exceeds the storage capacity, the excess water is added to the temporary 
storage term of the overlying layer. The process is repeated until the top layer is reached, 
completing the reverse cascade. The volume of water in the temporary storage term exceeding 
the storage capacity of the top layer is added to the potential runoff volume calculated for that 
grid cell. The final runoff volume is calculated following the simulation of evapotranspiration 
from the root zone. 
After the completion of the reverse cascade and the placement of excess water into temporary 
storage terms, evapotranspiration is simulated for each root-zone layer using a dynamic rootzone 
weighting function and a modified Priestley-Taylor equation. Evapotranspiration is 
simulated only for days with air temperature greater than 0¢XC. 
The vegetation characteristics most important for evaluating transpiration are distribution, 
minimum xylem water potential, and rooting depths (Flint, A.L. et al. 1996, p. 45). The 
minimum xylem water potential used to calculate the lower limit of plant-available water 
was .6,000 kPa (60 bar), which is assumed to be equivalent to the wilting point of vegetation. 
The dynamic weighting is based on calculated relative saturations for each root-zone layer and 
the relative distribution of water (based on saturation) throughout all layers. The purpose of 
dynamic weighting is to increase root activity for the wettest layer. Static root density weights 
are also incorporated into the dynamic weighting function, setting an upper limit on root activity 
within each layer. For the topsoil layer, the bare-soil evaporation term is added to the 
transpiration term. Using the calculated weighting terms, evapotranspiration is simulated by 
applying a form of the modified Priestley-Taylor equation developed by Flint, A.L. and Childs 
(1991), to each layer of the root zone (USGS 2001b, Section 6.4.6): 
(Eq. 4-3) ETk = £\¡¬ ¡P PETk 
i ƒn{wgti ¡P [ak (1.exp(bk ¡P relsati
k))]} £\¡¬ = . 
k 
i 
where ETk is total root-zone evapotranspiration for grid cell k and PETk is the adjusted clear-sky 
simulated equilibrium for grid cell k. (The equilibrium potential evapotranspiration rate is 
calculated using a £\ value of 1.0 and is used to represent the nonadvective component of the 
energy balance.) The term relsat is the relative saturation calculated for layer i within grid cell 
k; and ak and bk are the Priestley-Taylor model coefficients for grid cell k supplied as soil- and 
rock-type input parameters in the model control file. (In this analysis, the coefficients were 
identical for all soil and rock types, but varied between different climate scenarios and between 
May 2004 4-10 No. 1: Climate and Infiltration Revision 1 
soils and rocks.) After water contents for each layer are reduced according to the calculated 
evapotranspiration rates, the final runoff and net infiltration terms are calculated, and the new 
water-content for each root-zone layer are updated for the following day’s water-balance 
calculation. 
4.2 MODEL CALIBRATION AND VALIDATION 
4.2.1 Model Calibration 
The INFIL model was calibrated using comparisons of simulated streamflow with historical 
streamflow data from 31 gauging stations in the Death Valley region, and simulated 50-year 
(1950 to 1999) basinwide average net infiltration with previous estimates of basinwide average 
recharge for 42 basin areas (defined in previous studies as hydrographic areas and subareas) in 
the Death Valley region. Model calibration included adjusting the following parameters: 
bedrock saturated hydraulic conductivity, root density, average storm duration (for summer and 
winter storms), and soil saturated hydraulic conductivity and wetted area (used to represent 
stream-channel characteristics). 
The numerical model was calibrated by comparing measured volumetric water contents using 
neutron moisture meters and simulated water-content data (Flint, A.L. et al. 1996). At selected 
neutron boreholes, water-content data were summed for the soil profile and compared to the 
model simulation for the same time period by using the developed site precipitation record 
(USGS 2001b). Two examples are presented: borehole USW UZ-N50, with soil 2.7 m deep 
(Figure 4-5a), and borehole UE-25 UZN#63, with soil 1.7 m deep (Figure 4-5b). 
Changes in water-content profiles through time, measured in boreholes located in active channels 
with thick soils, were used to develop and calibrate a modified Priestley-Taylor 
evapotranspiration model (Hevesi et al. 1994). Initial model calibrations conducted using INFIL 
V1.0 in 1996, and consisting of a generalized (site-wide) calibration of the modified 
Priestley-Taylor model coefficients, were based primarily on the calculated changes in the 
measured profiles (USGS 2001b, p. 33; also discussed in Section 6.8.3). 
4.2.2 Model Validation Using Comparison of Net Infiltration and Recharge Rates 
Because of a deep unsaturated zone, net infiltration is a primary source of groundwater recharge; 
the results of net infiltration calculations using a water balance model were compared with those 
from three-dimensional numerical modeling (Wu et al. 1999) and calculations of groundwater 
recharge rates. 
May 2004 4-11 No. 1: Climate and Infiltration Revision 1 
Source: USGS 2001b, Figure 6-20. 
Figure 4-5. Simulated and Measured Water Content in Boreholes 
Using the water-balance method, Winograd and Thordarson (1975) estimated that 3 percent of 
precipitation becomes groundwater recharge; their estimate was based on discharge 
measurements from springs south of Yucca Mountain near the Nevada–California border. The 
Maxey–Eakin method (Maxey and Eakin 1950) was used in several previous water-balance 
studies of basins in the Death Valley region to estimate recharge to groundwater basins in 
4-12 May 2004 No. 1: Climate and Infiltration Nevada (Watson et al. 1976; Dettinger 1989; Avon and Durbin 1994; Harrill and Prudic 1998; 
Donovan and Katzer 2000). Given the average annual precipitation, the Maxey–Eakin method 
classifies basins into five recharge zones (Maxey and Eakin 1950): 
1. No recharge occurs for precipitation less than 203 mm/yr 
2. 3 percent recharge for precipitation from 203 to 304 mm/yr 
3. 7 percent recharge for precipitation from 305 to 380 mm/yr 
4. 15 percent recharged for precipitation from 381 to 507 mm/yr 
5. 25 percent recharge for precipitation of 508 mm/yr and greater. 
By comparing the Maxey–Eakin estimates with 40 estimates of recharge obtained from the 
Southern Great Basin using a basinwide water-budget analysis and 27 estimates of recharge 
obtained using geochemical and numerical modeling approaches, Avon and Durbin (1994) and 
Harrill and Prudic (1998) concluded that the Maxey–Eakin method provides reasonable 
estimates of recharge for basins in Nevada (Figure 4-6). Several studies have presented modified 
and updated versions of the Maxey–Eakin method, based on recent precipitation data, 
geochemical data, and basinwide water-balance data (D’Agnese et al. 1997; Donovan and Katzer 
2000; Hevesi and Flint 1998). 
Source: USGS 2001b, Figure 6-41. 
NOTE: Estimated recharge data are from Maxey and Eakin 1950, Lichty and McKinley 1995, and Winograd 1981; 
chloride mass balance data are from CRWMS M&O 2000c. 
Figure 4-6. Comparison of INFIL V2.0 Simulated Average Net Infiltration Rates with an Estimate of the 
Average Holocene Recharge Rate for the Saturated Zone at Yucca Mountain and with 
Estimates of Recharge in the Southern Great Basin Obtained Using Alternative Methods 
4-13 May 2004 No. 1: Climate and Infiltration 
Revision 1 Revision 1 
The limitation of the application of the Maxey–Eakin method is that it uses only the average 
annual precipitation, without considering other factors that can affect recharge (D’Agnese et al. 
1997). The problem is that in the Maxey–Eakin method, recharge is estimated by assuming that 
a zone-specific percentage of precipitation infiltrates to recharge the groundwater. The Maxey– 
Eakin coefficients were determined by balancing the recharge, which depends on the depth to the 
water table, with estimates of groundwater discharge for 13 valleys in east-central Nevada 
(Maxey and Eakin 1950). In the Maxey–Eakin method, the areas with annual precipitation of 
less than 200 mm are not considered to recharge the groundwater. However, at Yucca Mountain, 
recharge is known to occur within areas where annual precipitation is less than 200 mm. 
Therefore, the comparison of the calculated infiltration with that from the Maxey–Eakin 
coefficients for the annual precipitation that is less than 200 mm is invalid. Moreover, estimates 
of net infiltration for the Yucca Mountain area may not correspond directly to recharge because 
of the time lag between the net infiltration and groundwater recharge in the thick unsaturated 
zone. 
Hevesi et al. (2003) conducted numerical modeling of net infiltration for the present-day climatic 
conditions over the area of Death Valley region in Nevada and California using four types of 
models. The following parameters were varied in these models: the sublimation rate parameter 
SUBPAR1, storm duration, bedrock saturated hydraulic conductivity, the stream-channel wetted 
area, and stream-channel hydraulic conductivity for soils. Hevesi et al. (2003, pp. 96 to 99, 
Tables 22 and 23) provided a statistical analysis of the results of modeling to compare simulated 
basinwide average net infiltration and recharge rates to simulated total basinwide net infiltration 
and recharge volumes. They determined that the simulated net-infiltration volumes are in good 
agreement with the estimated recharge volumes, and the best fit of simulated and observed data 
corresponds to the net infiltration model that accounts for streamflow coupled to the root-zone 
component; this approach was used in the INFIL model for the determination of net infiltration. 
4.2.3 Model Validation Using Comparison with Simulations of Percolation in the 
Unsaturated Zone 
The three-dimensional unsaturated flow model (BSC 2003d) was used for simulations of 
percolation in the unsaturated zone for the mean, lower-bound, and upper-bound present-day, 
monsoon, and glacial-transition climates. In these simulations, the net infiltration rates for all 
nine climate state scenarios from Simulation of Net Infiltration for Modern and Potential Future 
Climates (USGS 2001b) were used to assign the upper boundary condition. 
The three-dimensional unsaturated flow model is based on using the calibrated soils and rock 
properties (BSC 2003d; BSC 2003e). The results of modeling were validated using 
mountain-scale ambient temperature measurements, gas pressure, chloride, calcite, and strontium 
data. These data were used to constrain infiltration and percolation flux. The field data (matrix 
liquid saturations, water potentials, and perched-water elevations) were collected from nine 
boreholes (USW NRG-7a, USW SD-6, USW SD-7, USW SD-9, UE-25 SD-12, USW UZ-14, 
UE-25 UZ#16, USW WT-24, and USW G-2) (BSC 2003d, Section 6.2-5). The water-potential 
data were also measured from the Enhanced Characterization of the Repository Block tunnel 
(BSC 2002c). Perched-water elevations measured in borehole USW WT-24 and pneumatic data 
measured in boreholes UE-25 SD-12 and UE-25 UZ#7a were used to validate the unsaturated 
zone model. 
May 2004 4-14 No. 1: Climate and Infiltration Revision 1 
Matching the ambient temperature distribution within the unsaturated zone from field 
measurements and modeling is important to constrain the percolation flux and infiltration rate 
(Bodvarsson et al. 2003). Temperature data measured in boreholes USW H-5, USW H-4, and 
UE-25 WT#18 were used to validate the ambient thermal model. The validation criterion was 
based on matching the observed values with less than 3°C difference (BSC 2002c, 
Attachment I-1-2-2). 
The calcite model is validated by comparing one-dimensional simulation results with the results 
of field measurements. The validation criterion is that the simulated volume fraction of calcite 
coating for each unsaturated zone model layer fall within the range of measurements for that 
layer (BSC 2002c, Attachment I-1-2-4). 
14C data from gas samples provide approximate 14C residence time for pore water in the 
unsaturated zone. The residence time can be interpreted as the tracer transport time from the 
ground surface to a depth where the gas sample was collected. 
Borehole and Enhanced Characterization of the Repository Block strontium concentrations are 
used to check the unsaturated zone model strontium modeling results. The criterion for 
validation is a qualitative agreement between the simulated strontium concentrations and the 
average of the observations at the same elevation, and agreement with the vertical trends (BSC 
2002c, Attachment I-1-2-5). 
Thus, the results of three-dimensional numerical modeling of flow and transport in the 
unsaturated zone closely agree with the types of data obtained from field measurements in the 
unsaturated zone (temperature, gas pressure, chloride and calcite concentrations, saturation, 
water potential, and perched water levels). These agreements confirm the validity of the upper 
boundary conditions used for modeling (i.e., the net infiltration rate, which was calculated using 
the water balance model). 
4.3.1 Present-Day Climate 
For infiltration simulations, using INFIL V2.0, of the present-day climate, lower- and upperbound 
mean annual precipitation were 185.8 mm and 265.6 mm, respectively (USGS 2001b, 
Table 6-8). Mean annual temperatures at Yucca Mountain range from 15.1°C to 18.2°C 
(CRWMS M&O 1997b, Tables 2-1 and A-10). 
Modeling of the present-day climate scenario was provided based on the meteorological data 
presented in Table 4-2. Results of modeling infiltration for the lower-bound, mean, and 
upper-bound present-day climate scenarios are summarized in Tables 4-3 and 4-4 (USGS 2001b, 
Section 7.1.1). For the entire flow domain of 123.7 km2 (Table 4-3), for the mean present-day 
climate scenario, average precipitation is estimated to be 188.5 mm/yr, average outflow as 
streamflow is 0.2 mm/yr (corresponding to an average stream discharge of 0.03 ft3/s), and 
average net infiltration is 3.6 mm/yr. Average net infiltration ranges from 1.2 mm/yr for the 
lower-bound present-day climate to 8.8 mm/yr for the upper-bound present-day climate. 
Although average annual precipitation for the mean present-day scenario is only 1.4 percent 
greater than precipitation for the lower-bound present-day scenario, net infiltration for the mean 
4.3 RESULTS OF MODELING FOR DIFFERENT CLIMATES 
4-15 May 2004 No. 1: Climate and Infiltration present-day scenario is 200 percent greater than the net infiltration for the lower bound. This is 
probably because of greater infiltrated surface-water run-on in channels under mean present-day 
conditions than under the lower-bound conditions. Maximum rates of infiltrated surface-water 
run-on for mean present-day conditions are greater than 100 mm/yr in the headwater sections of 
the primary watersheds such as Solitario Canyon and Yucca, Drill Hole, Pagany, and Abandoned 
Washes. Results from the upper-bound present-day scenario also indicate that net infiltration is 
significantly influenced by infiltrated surface-water run-on in channels. In all three present-day 
climate scenarios, average annual evapotranspiration is greater than 95 percent of average annual 
precipitation. 
For the area of the repository (Table 4-4), for the mean present-day climate scenario, average 
precipitation is estimated to be 196.9 mm/yr, average outflow as streamflow is 1.4 mm/yr, and 
average net infiltration is 4.7 mm/yr. For the lower-bound present-day climate, net infiltration is 
0.4 mm/yr and outflow is –0.3 mm/yr. (The estimated negative outflow is caused by 
surface-water inflow from Drill Hole Wash exceeding outflow.) For the upper-bound 
present-day climate, net infiltration is estimated to be 11.6 mm/yr. 
Table 4-2. Summary of Meteorological Data and INFIL Simulation Results Used to Develop Spatially 
Distributed Net-Infiltration Estimates for Present-Day Scenarios for the 123.7-km2 Area of the 
Net Infiltration Model Domain 
Parameter 
INFIL simulation ID used for developing 
the present-day scenario 
Simulation period (year number) 1980 to 1995 
16 Simulation time (years) 
189.3 Mean 
282.9 Maximum Average annual 
precipitation (mm/yr) 
148.0 Minimum 
182.7 Mean Average annual 
evapotranspiration 571.9 Maximum 
(mm/yr) Minimum 61.9 
6.0 Mean 
Maximum 
Average annual 
infiltrated surface 
water run-on 
Minimum (mm/yr) 
Average annual outflow (mm/yr) 
1,514.4 
0.0 
0.3 
5.1 Mean 
Maximum Average annual net 
infiltration (mm/yr) 
INFIL Simulation Results for the 123.7-km2 Area 
of the Net Infiltration Model Domain 
YM1-4ex 4JA1-4ex 4JA1-4ex 
80 to 90 0 to 100 
10 100 
182.8 187.7 
273.3 280.6 
143.0 146.8 
181.8 185.5 
689.1 652.3 
50.6 51.1 
1.6 2.7 
599.7 669.6 
0.0 0.0 
0.0 0.1 
1.3 2.2 
252.0 574.4 
0.0 0.0 
A121-4ex 
Minimum 
Source: USGS 2001b, Table 6-7. 
1,486.2 
0.0 
4-16 No. 1: Climate and Infiltration 
Revision 1 
0 to 100 
100 
342.8 
512.4 
268.1 
326.8 
788.8 
83.4 
15.1 
4,343.8 
0.0 
1.5 
14.0 
4,354.3 
0.0 
May 2004 Table 4-3. Estimation Results for Present-Day Climate Scenarios over the 123.7 km2 Area of the 
Infiltration Model Domain 
Mean 
Maximum 
Minimum 
Mean 
Maximum 
Minimum 
Mean 
Maximum 
(mm/yr) 
Minimum 
Mean 
Maximum 
Minimum 
Table 4-4. Estimation Results for Present-Day Climate Scenarios over the 4.7 km2 Area of the 1999 
Upper Bound 
265.6 
397.1 
207.8 
255.5 
700.5 
71.5 
9.7 
2,669.0 
0.0 
0.9 
8.8 
2,656.6 
0.0 
Parameter 
Average annual 
precipitation (mm/yr) 
Average annual 
evapotranspiration (mm/yr) 
Average annual infiltrated 
surface water run-on 
Average annual outflow (mm/yr) 
Average annual net 
infiltration (mm/yr) 
Source: USGS 2001b, Table 6-8. 
Design Repository Area 
Mean 
Maximum 
Minimum 
Mean 
Maximum 
Minimum 
Mean 
Maximum 
(mm/yr) Minimum 
Mean 
Maximum 
Minimum 
Parameter 
Average annual 
precipitation (mm/yr) 
Average annual 
evapotranspiration (mm/yr) 
Average annual infiltrated 
surface water run-on 
Average annual outflow (mm/yr) 
Average annual net 
infiltration (mm/yr) 
Source: USGS 2001b, Table 6-10. 
The greater net infiltration rates are within the higher elevation areas with higher precipitation 
and thinner soils (USGS 2001b, Section 6.11.1). The spatial distribution for estimated 
precipitation for the mean present-day climate is presented in Figure 4-7, while the estimated net 
infiltration is presented in Figure 4-8. Net infiltration of greater than 100 mm/yr occurs 
throughout the steep, northeast facing slope of The Prow, caused by the combined effect of high 
precipitation, reduced evapotranspiration, frequent surface-water run-on at the area of very thin 
soils, and the high permeability of bedrock. High net infiltration also occurs in the upper 
No. 1: Climate and Infiltration 
Lower Bound 
185.8 
282.2 
148.0 
184.8 
571.9 
54.7 
2.1 
474.2 
0.0 
0.2 
1.2 
252.0 
0.0 
Lower Bound 
191.6 
204.1 
178.2 
191.7 
252.9 
155.0 
1.0 
59.8 
0.0 
-0.3 
0.4 
26.6 
0.0 
4-17 
Revision 1 
Mean 
188.5 
281.8 
147.4 
184.1 
612.1 
56.5 
4.4 
994.1 
0.0 
0.2 
3.6 
958.9 
0.0 
Upper Bound 
277.5 
295.8 
258.5 
260.4 
423.0 
203.0 
8.1 
454.8 
0.0 
4.9 
11.6 
387.4 
0.0 
Mean 
196.9 
209.9 
183.4 
189.9 
273.3 
154.7 
3.4 
161.1 
0.0 
1.4 
4.7 
120.1 
0.0 
May 2004 Revision 1 
channels of Solitario Canyon and Drill Hole, Pagany, and Abandoned Washes. Maximum net 
infiltration of more than 100 mm/yr occurs within isolated areas on side slopes and in channels 
with thin soils and high-permeability bedrock. However, total net infiltration within the domain 
of the unsaturated zone flow and transport model is dominated, by the lower rates of 1 to 
20 mm/yr on side slopes and ridge tops, which make up much of the unsaturated zone flow and 
transport model domain. 
Source: USGS 2001b, Figure 6-23. 
Figure 4-7. Estimated Precipitation for the Present-Day Climate Scenario 
4-18 May 2004 No. 1: Climate and Infiltration Source: USGS 2001b, Figure 6-26. 
Figure 4-8. Estimated Net Infiltration for the Present-Day Climate Scenario 
No. 1: Climate and Infiltration 
Revision 1 
May 2004 4-19 Revision 1 
For the lower-bound present-day climate scenario within the repository area, most areas, 
including the crest, have no net infiltration (USGS 2001b, Section 6.11.1). Areas with net 
infiltration greater than 5 mm/yr are isolated to north-facing side slopes and along the 
west-facing slope of Solitario Canyon. For the upper-bound present-day climate scenario, net 
infiltration along the crest of Yucca Mountain is more than 20 mm/yr, and the relative 
contribution of net infiltration along channels to the total net infiltration is much greater than that 
for the mean present-day climate conditions. Maximum net infiltration of nearly 2,700 mm/yr 
occurs in an active channel in the northern part of Yucca Wash. Within the repository area, 
maximum net infiltration of between 100 and 500 mm/yr occurs in Drill Hole Wash and along 
the west-facing slope of Solitario Canyon. In all cases, maximum net infiltration occurs in areas 
affected by surface-water run-on, and a strong correlation exists between maximum infiltrated 
surface-water run-on in channels and maximum net infiltration for all areas and all climate 
scenarios. However, because the areas with relatively high net infiltration (greater than 
100 mm/yr) are small, most of the total net infiltration for the entire model domain occurs in 
upland areas, where net infiltration is less than 20 mm/yr. 
4.3.2 Monsoon Climate 
The two climate analog sites selected to represent the upper-bound monsoon climate scenario are 
at Nogales, Arizona, and Hobbs, New Mexico. Meteorological data for the monsoon climate 
scenario are summarized in Table 4-5. Mean annual temperatures for these two sites are 16.6°C 
and 17.5°C, respectively, which are within the range of mean annual temperatures at Yucca 
Mountain—from 15.1°C to 18.2°C. At these sites, the average annual precipitation is 
410.5 mm/yr and 414.4 mm/yr, respectively, and average annual net infiltration of 15.1 mm/yr 
and 12.1 mm/yr, respectively (USGS 2001b, Section 6.9.3, Table 6-4). 
May 2004 4-20 No. 1: Climate and Infiltration Table 4-5. Summary of Meteorological Data for Nogales and Hobbs Analog Meteorological Stations and 
INFIL Simulation Results Used to Develop Spatially Distributed Net-Infiltration Estimates for 
Average annual precipitation 
(mm/yr) 
Average annual snow fall 
(mm/yr) 
Average annual 
evapotranspiration (mm/yr) 
Average annual infiltrated 
surface water run-on (mm/yr) 
Average annual outflow (mm/yr) 
Average annual net infiltration 
(mm/yr) 
Simulation Period (begin date to end date) 
Air temperature (ºC) 
Mean annual 
Maximum daily 
Minimum daily 
Mean 
Maximum 
Minimum 
Mean 
Maximum 
Minimum 
Mean 
Maximum 
Minimum 
Mean 
Maximum 
Minimum 
Mean 
Maximum 
Minimum 
Source: USGS 2001b, Table 6-11. 
Net infiltration for the Hobbs site was less than that for the Nogales site, even though 
precipitation was greater—the Hobbs site is warmer and has a higher evapotranspiration rate 
than the Nogales site. The simulation of infiltration rates for the two ana