Geochemistry of CO2
sequestration in the Jurassic
Navajo Sandstone, Colorado
Plateau, Utah
W. T. Parry, Craig B. Forster, J. P. Evans,
Brenda Beitler Bowen, and Marjorie A. Chan
ABSTRACT
The Jurassic Navajo Sandstone on the Colorado Plateau of Utah
may be considered for sequestration of CO2, because it is thick,
widely distributed, has a high porosity and permeability, is typically horizontal to gently folded, is favorably located with respect
to seal strata, and underlies many large point sources of CO2.
However, faulting common on the Colorado Plateau may provide
pathways for leakage of the CO2 similar to present-day geysers and
CO2-charged springs. Natural groundwater present in the Navajo
Sandstone includes a range of low-salinity, moderate-salinity with
high bicarbonate, and high-salinity waters. Higher salinity waters
may have moved from deeper strata under artesian pressures or
originated from solution of evaporite in pre-Jurassic rocks. These
characteristics make the Navajo Sandstone an excellent analog
for examining the geochemistry of CO2 injection into deep saline
aquifers. The storativity of CO2 in solution is a function of the
solubility of CO2 in these waters, which is dependent on salinity,
temperature, and pressure. Geochemical modeling shows that the
coolest, least saline water can contain the most CO2 in solution.
Dissolving CO2 in the water lowers the pH, so that no minerals
precipitate, and the Navajo Sandstone contains only small amounts
of mineral that may consume the H + . The reaction of the acidic
water produced by dissolving CO2 with K-feldspar and minor clays
and calcite in the sandstone throughout 500 yr consumes little H +
and produces only small amounts of product minerals. The Navajo
Sandstone likely would not store significant CO2 as mineral precipitate, and thus, stored volumes of CO2 would be limited by its solubility in the in-situ water and storage as free CO2 in pore space.
AUTHORS
W. T. Parry Department of Geology and
Geophysics, University of Utah, Salt Lake City,
Utah, 84112-0111; [email protected]
W. T. Parry is professor emeritus of geology
and geophysics at the University of Utah. His
former positions include associate professor
of geoscience at Texas Tech University, Lubbock, Texas, and exploitation engineer for Shell
Oil Company, Midland, Texas. He received
his B.S. and M.S. degrees and his Ph.D. in geological engineering from the University of
Utah, where he taught geology and engineering. His research interests are geochemistry and
mineralogy related to ore deposits and faults.
Craig B. Forster Department of Geology
and Geophysics, University of Utah, Salt Lake
City, Utah, 84112-0111; [email protected]
Craig Forster holds degrees in geology (his B.Sc.
degree and his Ph.D. from the University of
British Columbia and his M.Sc. degree from
the University of Waterloo). Throughout the past
30 years, he has been using laboratory, field,
and numerical simulation methods to assess
fluid-solute-heat flow in fractured geologic systems. He is a research associate professor at the
University of Utah and Utah State University.
J. P. Evans Department of Geology, Utah
State University, Logan, Utah 84322-4505;
[email protected]
J. P. Evans is a professor of geology at Utah
State University. He was educated at the University of Michigan (B.S. degree in geology and
B.S. degree in engineering), and Texas A&M
University (M.S. degree and Ph.D. in geology).
Brenda Beitler Bowen Department of
Geology and Geophysics, University of Utah,
Salt Lake City, Utah, 84112-0111; present
address: Department of Earth and Atmospheric Science, Purdue University, 550 Stadium
Mall Drive, West Lafayette, Indiana 47907;
[email protected]
Brenda Beitler Bowen is a postdoctoral researcher at Central Michigan University studying fluid-sediment interactions in acid saline
systems. She will begin as assistant professor
of earth sciences at Purdue University in Fall
2007. She earned her B.S. and M.S. degrees
Copyright #2007. The American Association of Petroleum Geologists/Division of Environmental
Geosciences. All rights reserved.
DOI:10.1306/eg.07120606004
Environmental Geosciences, v. 14, no. 2 (June 2007), pp. 91 –109
91
at the University of California, Santa Cruz, and
her Ph.D. at the University of Utah studying
the diagenesis of the Navajo Sandstone.
Marjorie A. Chan Department of Geology and Geophysics, University of Utah, Salt
Lake City, Utah, 84112-0111;
[email protected]
Marjorie A. Chan is a professor of geology and
department chair at the University of Utah.
She received her B.S. degree from the University of California, Davis, and her Ph.D. from
the University of Wisconsin, Madison. Her recent and current research focuses on Mesozoic
sedimentology and stratigraphy on the Colorado Plateau, with applications to eolian reservoirs and terrestrial iron oxide concretion
analogs to Mars.
ACKNOWLEDGEMENTS
Funding was provided by the U.S. Department of Energy grant DE-FG03-00ER15043.
We gratefully acknowledge Matthias Grobe
and Brian McPherson for their review of this
manuscript.
92
INTRODUCTION
Large quantities of anthropogenic CO2 generated by burning fossil fuels are contributing to a rise in CO2 concentrations in the
Earth’s atmosphere. Average releases at United States power plants
are 2.7 million t of CO2 each year, with large power plants releasing more than 18 million t (Hovorka et al., 2001). This has led to
concerns that growth in concentrations of CO2 in the atmosphere
is contributing to a greenhouse effect that is affecting global climate. Despite uncertainty in estimating the impact of CO2 emissions on future climate, the general scientific consensus is that
CO2 emissions should be minimized (Intergovernmental Panel
on Climate Change, 2001). One element of a multifaceted strategy
for limiting CO2 emissions is to sequester CO2 in geologic formations. Carbon dioxide can be stored in the subsurface as (1) a dissolved constituent in naturally occurring groundwater; (2) mineral
precipitates produced by chemical reactions; and (3) free CO2 as a
supercritical fluid in the pore spaces of reservoir rock (Gunter and
Perkins, 1993; Bachu et al., 1994; Law and Bachu, 1996; Perkins
and Gunter, 1996; van der Meer, 1996; Pruess et al., 2001; Bachu
and Adams, 2003).
Regional-scale sedimentary aquifers in subsurface geological
structures offer a large storage capacity for CO2 (Hendriks and Blok,
1993; Law and Bachu, 1996; McPherson and Cole, 2000; Lackner,
2003). Deeper aquifers are more attractive than shallow aquifers
(van der Meer, 1993; Bergman and Winter, 1995). The aquifer
should be capped by a regional aquitard, the depth should be more
than 800 m (2600 ft) deep, and the aquifer should possess adequate porosity and permeability (Bachu et al., 1994). The residence
or traveltime in regional aquifers is of the order of thousands to
millions of years (Bachu et al., 1994). Deep brines absorb less CO2
than low-salinity waters, but additional storage capacity may be
provided because of water-rock reactions; in fact, large amounts
of carbonate minerals can precipitate in some cases (Gunter and
Perkins, 1993). The acidity produced by dissolving CO2 may be
consumed by dissolution of calcite and by reaction with aluminosilicate minerals such as feldspar if they are sufficiently abundant
(Gunter and Perkins, 1993). However, some numerical modeling
has shown that only small amounts of the carbonates are precipitated because they are quite soluble under the reservoir conditions of high CO2 pressure and resulting low pH. The ferrous
carbonate siderite can trap CO2 if organic matter, petroleum, or
some other reductant can chemically reduce the iron. Long-term
CO2 storage may lead to a decrease in porosity because of mineral
precipitates and alteration. Alteration product hydrous aluminosilicates generally occupy larger volumes than the precursor minerals
(Pruess et al., 2001).
The process of collecting the CO2 at a point source (e.g., a
power plant or gas field production well), transporting it to a well
site, and injecting it into a subsurface reservoir is commercially
proven, and commercial geologic sequestration has begun in natural
Geochemistry of CO2 Sequestration in the Jurassic Navajo Sandstone
CO2 reservoirs, enhanced oil-recovery operations, and
experimental sites (Bowker and Shuler, 1991; Holloway, 2001; Haszeldine et al., 2005; Kharaka et al.,
2006).
Assessing the viability of a CO2 sequestration reservoir requires studying the basic geology of a sequestration site, evaluating potential pathways for leakage,
and performing geochemical analyses that help to determine the likely behavior of CO2 introduced to the
reservoir-seal system (Law and Bachu, 1996). Numerical modeling of the fluid flow and geochemical processes
is necessary to understand the long-term sequestration
of CO2 because chemical reactions with aluminosilicates and fluid flow in the porous reservoirs are slow
and cannot be duplicated in laboratory experiments
(Pruess et al., 2001).
Injection of CO2 into petroleum reservoirs for secondary recovery may also provide an indication of the
nature of fluid-rock interaction. For example, CO2 has
been injected into the Pennsylvanian–Permian Weber
Sandstone, an eolian deposit (Fryberger, 1979) with
mineralogy and sedimentary petrography similar to
the Navajo Sandstone (Bowker and Shuler, 1991). The
Weber Sandstone is the major petroleum-producing
reservoir at the Rangely field, Colorado. Much of the
Weber Sandstone is cemented with ferroan calcite,
ferroan dolomite, and illite, and interstratified illitesmectite cements comprise approximately 5% of the
sandstone. The pH of the brine produced from wells
decreased from 7.5 to 4.5 from initiation to end of injection. The concentrations of Fe, Ca, and Mg in produced brine increased as a function of solution of carbonate, clays, and feldspars. The pH increased from
3.5 at injection wells to 4.5 at production wells. No net
change in permeability was observed as a consequence
of CO2 injection (Bowker and Shuler, 1991).
Kharaka et al. (2006) investigated the potential for
CO2 storage in a saline aquifer. They injected 1600 t
of CO2 into the Frio Formation, a regional oil reservoir
in the United States Gulf Coast. Injection of CO2 resulted in a sharp drop in pH to 5.9 and increases in alkalinity and dissolved Fe. The solution of carbonate
and iron oxyhydroxides buffered the pH, which geochemical models show would otherwise have dropped
to 3.
This study focuses on the geochemical aspects of
CO2 sequestration in faulted and folded eolian Navajo
Sandstone located in the Colorado Plateau region of
Utah (Figure 1). Issues addressed include (1) the chemical characteristics of water naturally present in the
these sandstones; (2) the chemical characteristics of
water that may flow into the sandstones from aquifers
with contrasting water types; (3) the solubility of CO2
in these waters; (4) the geochemical consequences of
solution of CO2 in the diverse water types; (5) mixing of
waters derived from aquifers beneath the Navajo; and
(6) the reaction of CO2-charged waters with reactive
minerals in the sandstones and overlying seal rocks. The
likely effects of mineral dissolution and precipitation
are also evaluated.
The Colorado Plateau province is an elevated block
of mildly deformed rocks and thickened crust encompassing an area of 3.6 106 km2 (1.4 106 mi2) centered around the four corners of Utah, Colorado, Arizona, and New Mexico. The upper geologic section of
the Colorado Plateau comprises a thick Mesozoic sedimentary sequence that includes permeable Jurassic eolian sandstones (prospective sequestration reservoirs)
beneath a Late Jurassic to Cretaceous sedimentary sequence that includes low-permeability mudstone, shale,
and carbonate beds (prospective reservoir seals).
Several factors make the Colorado Plateau region of
Utah an attractive target for CO2 sequestration (Haszeldine et al., 2005). Natural CO2 reservoirs already exist
to show sequestration feasibility and provide natural
reservoir analogs. Allis et al. (2001) identified 12 major CO2 accumulations with more than 30 billion m3
(1059 billion ft3) of gas in place. Some reservoirs have
already been partially depleted through drilling and production. Other CO2 reservoirs are releasing CO2 to the
atmosphere through fault-related breaches in seal rocks
and geysering drill holes (Shipton et al., 2004).
Mantle degassing, metamorphism of carbonate sediments, and oxidation of hydrocarbons are among possible sources of this CO2 (Shipton et al., 2004). Helium
isotopes suggest that only a minor mantle component
is present, and carbon isotopes suggest that the probable source is thermal decarbonation of carbonate sediments within the contact aureoles of igneous intrusions
(Shipton et al., 2004).
Large, coal-fired power plants operating in the region produce CO2, and new power plants are planned.
The CO2 produced may be sequestered in favorable
sedimentary strata (Allis et al., 2001). Wastewater from
coal-bed gas production is presently injected into the
Navajo Sandstone (Conway et al., 1997). The eolian Navajo Sandstone forms a well-studied and extensive candidate sequestration reservoir with high porosity and
permeability, relatively simple chemical and mineralogical composition, and good reservoir seals, although
faults may permit the seals to leak (Shipton et al., 2004;
Heath et al., in press).
Parry et al.
93
Figure 1. Index map of the Colorado Plateau in Utah showing
uplifts and basins with structural
contours (in thousands of feet)
on the Permian Kaibab Limestone and locations described in
the text. The upper left inset
shows the distribution of the
Navajo Sandstone in outcrop
and in the subsurface on the Colorado Plateau of Utah. The
Navajo Sandstone has been removed by erosion from the
San Rafael swell (S), from the
Circle Cliffs uplift (C), from
the Monument uplift (M),
and from the Kaibab uplift (K).
Fault systems shown are Hurricane fault (HF), Sevier fault (SF),
Paunsaugunt fault (P), Moab
fault (MF), and Little Grand Wash
fault (G). Localities where CO2
is present are Farnham Dome (F)
and Woodside Geyser (W). Tertiary igneous intrusions are
the La Sal Mountains (L), Abajo
Mountains (A), and Henry
Mountains (H).
METHODS
The geochemistry of CO2 sequestration in the Navajo
Sandstone is evaluated using petrographic examination,
whole-rock chemistry, clay mineralogy, calculated modal
mineralogy, and geochemical modeling. Samples of Navajo Sandstone were collected in outcrop on the Colorado Plateau. Sampling areas and detailed sample localities are given in Beitler et al. (2005) and Bowen
(2005). Samples of host rock (n = 74) representing diagenetic types were submitted to Actlabs for whole-rock
analyses by inductively coupled plasma–mass spectroscopy for major, minor, and trace elements. Clay min94
erals were extracted from 10-g samples of host rock by
mild hand grinding in a mortar and then peptizing in
water. Centrifuging separated the less than 2-mm-size
fraction. Centrifuged slurry was smeared on glass slides
for x-ray diffraction analyses. Thin sections (n = 66)
were stained for K-feldspar and examined in the petrographic microscope. Polished sections were examined
under reflected light. The modal mineralogical composition of Navajo Sandstone samples was calculated
using a linear, least-squares fit of whole-rock chemistry, mineral composition, and mineral mass abundance described by Parry et al. (1980). Petrography, bedding, thickness, permeability, porosity, and mineralogy
Geochemistry of CO2 Sequestration in the Jurassic Navajo Sandstone
of subsurface samples studied by Lindquist (1983) from
the Nugget Formation, an equivalent to the Navajo Sandstone, in southwestern Wyoming are similar to petrography and mineralogy of outcrop samples in this study,
suggesting that outcrop samples adequately represent
the Navajo Sandstone.
Geochemical reaction paths for mixing of different
waters, for the reaction of CO2-charged waters with
Navajo Sandstone, and for leakage of CO2-charged
waters through faults were calculated using The Geochemists Workbench (Bethke, 1998). Model calculations
were done for a temperature of 60jC and 100 bars
(1450 psi), well above the critical temperature for CO2
of 31jC and 74 bars (1073 psi). Water chemistry typical of the Navajo Sandstone used in the model calculations was from Hood and Patterson (1984), Kimball
(1992), and Spangler et al. (1996).
STRATIGRAPHY
The sequence of Mesozoic sedimentary rocks on the
Colorado Plateau of southeastern Utah includes three
permeable eolian sandstone units of the Glen Canyon
Group that are exposed on the flanks of the regional
Laramide uplifts (Figure 2). The eolian Wingate Formation, the Navajo Sandstone, and the Entrada Sandstone are targets of this work and are the site of deepwater injection (Conway et al., 1997). The Kayenta
Formation that separates the Wingate Formation from
the overlying Navajo Sandstone is fluvial, and the Chinle
(fluvial) and Moenkopi (marine) formations beneath
the Wingate Formation are fine-grained aquitards. The
Carmel Formation that forms a seal unit above the Navajo Sandstone represents a complex sabkha sequence
of sandstone, siltstone, mudstone, limestone, anhydrite,
and gypsum.
THE NAVAJO SANDSTONE
The Early Jurassic Navajo Sandstone and its related
equivalents, the Nugget Sandstone of northern Utah
and Wyoming and the Aztec Sandstone of Nevada,
form the largest eolian dune deposit in North America
(Blakey et al., 1988; Peterson and Turner-Peterson, 1989;
Blakey, 1994). The Navajo Sandstone is the thickest,
most laterally extensive (Figure 1), and most permeable unit in the Mesozoic section of the Colorado Pla-
teau. In the eastern half of the Colorado Plateau, the
Navajo Sandstone and its equivalents cover an area of
more than 3.5 105 km2 (1.35 105 mi2), with deposits more than 670 m (1213 ft) thick, and a volume
of up to 1.4 105 km3 (3.35 104 mi3) (Blakey, 1988).
The thickest parts of the Navajo Sandstone (up to 700 m
[2300 ft]) occur in southwestern Utah, which represents the main accumulation of eolian sand. The eolian
sandstone is characterized by high-angle, large-scale
cross-stratification and striking red to white variations
associated with primary depositional patterns and extensive bleaching caused by fluid migration (Chan
et al., 2000; Parry et al., 2004). Wingate Sandstone,
Navajo Sandstone, and Kayenta Formation are extensively bleached along steeply dipping eastern limbs,
along the flanks, and along the remaining crests of the
San Rafael swell, Circle Cliffs uplift, Kaibab uplift,
and Monument uplift (Figure 1) (Beitler et al., 2003).
Bleached sandstone contains less calcite and dolomite
cement, less hematite, and more K-feldspar alteration
to clay (Beitler et al., 2005). The Navajo Sandstone is
a fine-grained quartz arenite that is texturally mature,
and fine to medium grained. Cordova (1978) divided the
Navajo Sandstone into three informal units. Although
there are some internal stratigraphic differences, Hood
and Danielson (1979), Hurlow (1998), and others, however, suggest that the entire formation can be represented
as one relatively homogeneous unit particularly for injection and chemical-reaction modeling purposes. The
Navajo Sandstone is unconformably overlain by the
J-2 unconformity (Figure 2). The Navajo Sandstone is
known as an important groundwater aquifer throughout much of the Colorado Plateau (Cordova, 1978; Hood
and Patterson, 1984; Howells, 1990; Kimball, 1992;
Spangler et al., 1996; Hurlow, 1998).
In various areas of Utah, the Navajo Sandstone typically contains different facies units such as cross-bedded
dune sandstone; massive, uniform sandstone (sometimes with contorted cross-strata); horizontal stratified
interdune sandstone; and some carbonate playa and
interdune deposits (e.g., Verlander, 1995; Thomas et al.,
2000). In addition, internal stratification with some
thin horizons of very high permeability (grain flow laminae of Hunter, 1977) occur along dune foresets, and
very low-permeability horizons up to 30 cm (12 in.) thick
lie at the base of some dune sequences (wind ripple
laminae of Hunter, 1977, with finer grained sand, silt,
and mud). Dune sets typically range in thickness from
4 to 20 m (13 to 66 ft) thick in the northern San Rafael
swell (S in Figure 1), but can also be several tens of
meters thick elsewhere.
Parry et al.
95
Figure 2. Stratigraphy of Jurassic and Triassic formations in the Colorado Plateau study area showing the position of the Jurassic
Navajo Sandstone and the Page Sandstone with the overlying Carmel Formation seal unit. The underlying Kayenta and similar
Wingate Sandstone units are also potential reservoirs that are a part of the Glen Canyon Group. The Triassic Chinle Formation and
Moenkopi Formation are aquitards that inhibit fluid flow upward into the Navajo Sandstone. General depositional environments of
the formations are indicated right of the column. Major unconformities (e.g., Tr-1, 3, and J-0 to J-5) are from Pipiringos and
O’Sullivan (1978).
Beitler et al. (2005) interpret bleaching patterns of
the normally red Navajo Sandstone to indicate the pathways of paleofluid flow. Bleaching on a regional scale
occurs on the crests of uplifts and commonly occurs at
structural inflections. On an outcrop scale, bleaching is
96
restricted by low-permeability deformation bands, interdune limestone-chert layers, and fine-grained to muddy
wind ripple laminations (Beitler et al., 2005). These
relationships suggest that flow and storativity of CO2
are determined by regional-scale characteristics.
Geochemistry of CO2 Sequestration in the Jurassic Navajo Sandstone
Locally, another eolian unit, the Page Sandstone,
overlies the Navajo Sandstone (Figure 2) (Blakey, 1994).
Because the Page Sandstone is similar in lithology and
facies and occurs only locally, it can be treated as part
of the Navajo Sandstone for modeling purposes.
THE CARMEL FORMATION
The Carmel Formation overlies the Navajo Sandstone
and/or the Page Sandstone where present throughout most of the Colorado Plateau (Wright and Dickey,
1963). The Carmel Formation is a heterogeneous sabkha
sequence of interbedded sandstone, siltstone, mudstone,
limestone, anhydrite, and gypsum that overlies the Navajo Sandstone and forms the primary seal that would
limit upward migration of CO2 from the Navajo Sandstone reservoir. The Carmel Formation reaches a maximum thickness of 400 m (1312 ft) and averages more
than 100 m (330 ft) across much of the Colorado Plateau
region. Anhydrite and alabaster gypsum beds in the
upper Carmel Formation comprise 50% of the unit.
Individual gypsum beds are more than 2 m (6.6 ft) thick
(Trimble and Doelling, 1978). Calcite cement is common in the clastic strata, but silica, gypsum, and hematite cements are also present (Wright and Dickey,
1963; Doelling, 1975; Trimble and Doelling, 1978;
Doelling and Davis, 1989; Doelling et al., 2000). Mercury injection porosimetry reported by White et al.
(2005) shows that anhydrite with displacement pressures of 30–60 bars (435–870 psi) forms the most
effective seal. The Mancos Shale is the next most effective seal. Thus, the gypsum beds (anhydrite in the subsurface) in the Carmel Formation may form a good seal
stratum.
STRUCTURE
Monoclinal folds, anticlines, domes, basins, and faults
deform the generally flat-lying Navajo Sandstone and
overlying Carmel Formation (Hunt, 1956; Davis, 1999).
Early Tertiary compression formed a series of monoclines, with the Navajo Sandstone dipping at angles in
excess of 60j to the east (e.g., the Kaibab, Circle Cliffs,
Monument, and San Rafael monoclines) (Figure 1).
The Navajo Sandstone has been removed from much
of the crest area of these broad, asymmetrical monoclinal folds, which are bounded by steeply dipping, blind,
reverse faults (Davis, 1999; Bump and Davis, 2003).
Deformation bands are common in the steep parts of
these structures. The deformation bands significantly
reduce the porosity and permeability of fault-affected
rocks (Antonellini and Aydin, 1994; Shipton et al.,
2002). Shipton et al. (2002) and Sigda and Wilson
(2003) show that hydraulic conductivity is three orders
of magnitude less in cataclastic deformation bands
than in undeformed sandstone. Smaller basins occur
between the monoclinal uplifts, such as the Henry
Mountain Basin between the Circle Cliffs and Monument uplifts (Figure 1).
The Navajo Sandstone has also been folded into
broad anticlines and domes that provide adequate structural traps for storage of buoyant CO2 fluids, including
the Farnham Dome field (Morgan and Chidsey, 1991).
Northwest-trending anticlines and associated faults occur in east-central Utah underlain by thick evaporite
beds. The evaporite beds of the Pennsylvanian Paradox Formation have been deformed to produce eight,
northwest-southeast–trending salt anticlines, where the
salt is locally thickened to more than 4 km (2.5 mi)
(Cater, 1970). Salt intrusion structures were covered
by 2 km (1.2 mi) of post-Triassic sedimentary rocks
(Doelling, 1988), and exhumed about 37 Ma (Nuccio
and Condon, 1996). Salt was dissolved by groundwater,
forming the Moab and Lisbon Valley graben. The salt
anticlines lie above and parallel to basement faults.
Northward-trending fault-bounded blocks of the
geographical High Plateaus are bounded by Basin and
Range normal faults that include, from west to east,
the Hurricane, Sevier, and Paunsaugunt faults. Finally,
a series of domes and folds are related to Tertiary laccolithic intrusions in the Henry, La Sal, and Abajo
Mountains.
Abundant, closely spaced faults displace Jurassic
and Cretaceous formation along the northern part of
the Kaibab monocline, the Escalante anticline, and
the San Rafael swell. The Teasdale fault zone forms the
northern margin of the Miners Mountain uplift. Trace
lengths of some of the faults are many kilometers
(Davis, 1999). Faults in the Navajo Sandstone are related to regional trends, such as the northwest-striking
set of faults that cut the San Rafael swell and the central part of the Colorado Plateau (Witkind, 1991; Bump
and Davis, 2003) or the north- and north-northeast–
striking trend observed near the transition with the Basin and Range (Witkind, 1991). Small faults in the Navajo Sandstone are typically deformation-band faults
where grain crushing along narrow zones results in
low-permeability surfaces or networks (Antonellini
Parry et al.
97
and Aydin, 1994; Shipton et al., 2002). Faulting and
fracturing may have an important effect on the variability of permeability in the Navajo Sandstone and the
overlying Carmel Formation.
PETROGRAPHY OF THE NAVAJO SANDSTONE
The Navajo Sandstone averages 89% quartz. Alteration
of detrital feldspars to illite, illite-smectite, and kaolinite is common. Interstratified illite-smectite is R > 3 ordered (each smectite layer is surrounded by more than
three illite layers) and contains 90% illite layers. Detrital quartz grains are sometimes coated with iron
oxide, illite, and/or illite-smectite. Where quartz overgrowths are present, these grain coatings postdate the
quartz overgrowths. Most of the sand grains are quartz,
but K-feldspar grains are scattered throughout. The
sand grains are mostly 0.1–0.25 mm (0.004–0.009 in.)
in size and are commonly coated with 2–4-mm-thick
illite and hematite. Illite also occasionally fills pore
space. Kaolinite commonly occurs as pore fillings, and
calcite cement is sporadic.
NAVAJO SANDSTONE MODAL MINERALOGY
Mineral abundances of the phases that affect CO2 sequestration were calculated by linear least-squares fit
of mineral abundance, mineral composition, and rock
composition. Mg was calculated as dolomite; the remaining Ca was calculated as calcite; K, Al, and Si were
fit to kaolinite, K-feldspar, and K-mica; and the remaining Si was calculated as quartz. The results of modal
mineralogical calculations for 74 samples are shown
in Figure 3. In decreasing weight percent abundance,
K-feldspar averages 4.6%, illite averages 4%, kaolinite
averages 1.8%, and calcite averages 0.81%.
PERMEABILITY AND POROSITY
The most common controls on sandstone matrix porosity and permeability are thermal maturity, composition, sorting, and grain size. Porosity and permeability
of the Navajo Sandstone are also related to mineral
cements and deformation. Porosity of undeformed sandstone ranges from 10 to 30% (Cooley et al., 1969; Dunn
et al., 1973; Cordova, 1978; Hood and Patterson, 1984;
Freethey and Cordy, 1991; Shipton et al., 2002). Shipton
et al. (2002) note that deformation bands that develop
98
Figure 3. Histograms of mineralogical composition of 74 samples of the Navajo Sandstone. The mineralogy has been computed from whole-rock chemistry using a linear, least-squares fit
of rock chemistry, mineral chemistry, and mineral abundance.
in response to faulting can reduce the original sandstone porosity from an initial 19 down to 12%.
Laboratory-scale permeability measurements of undeformed Navajo Sandstone yield values ranging from
50 to 5000 md ( Weigel, 1986; Antonellini and Aydin,
1994; Shipton et al., 2002). Locally, permeability values
of up to 9000 md are observed (Shipton et al., 2002).
In-situ well tests yield permeability estimates that range
from 80 to 13,000 md (Heilweil et al., 2000). The
2.5 order of magnitude variation in permeability of Navajo Sandstone reflects both facies variations and the
consequences of faulting and fracturing.
When considering CO2 injection into a sandstone
reservoir, relative permeability relationships for the sandstone with respect to water and CO2 must be identified
Geochemistry of CO2 Sequestration in the Jurassic Navajo Sandstone
Table 1. Chemical Composition of Groundwater from the Navajo Sandstone and Related Strata on the Colorado Plateau*
Sample
Farnham
Paradox
Mississippian
Low salinity
Ca
Mg
Na + K
HCO3
SO4
Cl
Sum
874
76,200
10,120
56
61
9480
1410
52
422
58,300
86,830
14, 6.1
3070
919
366
305
586
49
300
100
172
249,000
155,000
14
3607 (0.36 wt.%)
394,000 (39 wt.%)
254,500 (25 wt.%)
402
*Concentrations are from Hood and Patterson (1984) in milligrams per liter.
because the relative proportion of water and liquid or
supercritical CO2 in the pore space will vary as a function of time and space in the reservoir. Unfortunately,
few data are available for estimating these relationships for CO2-water systems. Consequently, Pasala et al.
(2003) and Ennis-King et al. (2004) used models developed by van Genuchten (1980) and Corey (1986)
to compute plausible relative permeability curves for
sandstone.
Typical waters from below the Navajo include late Paleozoic Paradox, Kaibab, Coconino, and Little Grand
Wash fault waters shown in the trilinear diagram in
Figure 4. Waters from the Navajo Sandstone that contain substantial bicarbonate include Woodside, Farnham,
and Crystal Geyser locations (Figures 1, 4).
GEOCHEMICAL MODELING
GROUNDWATER CHARACTERISTICS
In several locations, the Navajo Sandstone is a natural
reservoir for CO2-charged waters (e.g., Morgan and
Chidsey, 1991). Natural water in the Navajo Sandstone
in the northern San Rafael swell varies from fresh to
moderately saline (Table 1) (see Hood and Patterson,
1984; Kimball, 1992; Spangler et al., 1996; Heath et al.,
in press). Groundwater in the Navajo Sandstone includes
brines with salinities of nearly 20,000 mg/L dissolved
solids and low-salinity water (Kimball, 1992; Spangler
et al., 1996). Water with salinities in the range of 3000–
10,000 mg/L occurs in the southeasternmost part of Utah,
and very saline to briny water with salinities of 10,000–
35,000 and 35,000 – 400,000 mg/L ( Table 1) underlies
the moderately saline groundwater near areas underlain by evaporite beds in southeastern Utah (Howells,
1990). The lowest salinity water is of the calcium, magnesium bicarbonate type. Higher salinity water contains
significant sulfate and chloride that results from interformational leakage from below the Navajo Sandstone
caused by relatively high artesian pressures and likely
conduction pathways along faults (Hood and Patterson,
1984). Very saline waters in aquifers below the Navajo
Sandstone are in the Pennsylvanian Paradox Formation, which has dissolved evaporite salts, and Mississippian limestones. The saline waters may leak through
the Triassic Moenkopi and Chinle formations into the
Navajo Sandstone along prominent faults in the region.
Injection of CO2 into water-saturated, porous rock forms
regions of free CO2 and CO2-saturated water. During
injection, the dissolution of CO2 in water lowers the
pH as aqueous CO2 dissociates into H + and HCO
3.
The ionizable hydrogen is capable of driving chemical reactions with aluminosilicate and carbonate minerals in the rock. Significant amounts of water remain
in the pore space after maximum saturation with free
supercritical CO2. The original details of this waterrock reaction depend on the chemical composition of
the groundwater. We illustrate this interaction by first
calculating the solubility of CO2 in water of varying
salinities, modeling the consequences of adding CO2
to two waters of contrasting composition, mixing contrasting waters, and modeling reaction of two contrasting waters with minerals in the Navajo Sandstone.
Simulation of rock-water reactions uses a representative volume of rock with a porosity of 20%. The volume
of rock required to contain 1 kg of solution is 5000 cm3
(305 in.3), a cube of rock 17 cm (6.7 in.) on an edge. The
beginning mineralogical composition of the rock derived from chemical analysis and calculated modal mineralogy of Navajo Sandstone rock samples is shown in
Table 2.
In the simulations, water compositions from Table 1
were used, dissolved SiO2 was set at quartz saturation,
Al3 + was set at kaolinite saturation, and K + was set at
muscovite saturation because of the common occurrence
of these minerals. The four stages in the geochemical
simulation are (1) the equilibrium state of the fluid was
Parry et al.
99
Figure 4. Piper diagram showing the compositions of some typical groundwater in the Navajo Sandstone and from other stratigraphic
units on the Colorado Plateau. Water compositions are taken from Hood and Patterson (1984).
calculated, (2) equilibrium fluids could then be mixed,
(3) CO2 was reacted with the equilibrium fluid, and
(4) the equilibrium CO2-charged fluid was reacted with
K-feldspar.
CO 2 SOLUBILITY
The solubility of CO2 in natural water in sandstone
aquifers is a function of temperature, pressure of CO2,
and salinity of the water. The likely temperature range
for sequestration is from a surface temperature of about
20jC to a temperature of approximately 80jC at a depth
100
of a few kilometers. Maximum CO2 pressure would
be between hydrostatic and lithostatic pressure at the
aquifer depth. Salinity in water from our study area
ranges from fresh to 20 wt.% dissolved salts. Stored CO2
would likely be a dense (between 0.6 and 0.8 g/cm3),
supercritical fluid at 60jC and 100 bars (1450 psi) pressure considered here. CO2 would occur as a dense, supercritical fluid phase, as a dissolved constituent in groundwater, and as a reactant with minerals in the reservoir
rock. The solubility of CO2 was calculated in water of
various salinities using the equations and coefficients of
Duan and Sun (2003) and shown in Figure 5. The solubility of CO2 decreases systematically as salinity increases
Geochemistry of CO2 Sequestration in the Jurassic Navajo Sandstone
Table 2. Average Mineralogical Composition of 74 Samples of Navajo Sandstone*
Mineral
Quartz
K-feldspar
Illite
Kaolinite
Calcite
Porosity
Sums
Weight %
Volume %
Cubic Centimeters in
5000 cm3 (305 in.3) of Rock
88.8
4.6
4.0
1.8
0.8
–
100.0
71.0
3.8
3.0
1.5
0.6
20.0
99.9
3555
191
150
74
31
1000
5000
Grams in 5000 cm3
(305 in.3) of Rock
9413
488
424
191
85
–
10,601
*In terms of weight %, volume %, and volumes of minerals in 5000 cm3 (305 in.3) of rock.
(Figure 5). For example, with fresh water at a CO2
pressure of 50 bars (725 psi), the solubility is 29.5 g/kg
of water at 60jC, decreasing to 14 g/kg in a 20 wt.%
(4.28 molal) salt solution. In a 2 wt.% (0.35 molal) salt
solution, the solubility of CO2 is 61.34 g/kg of water
at 20jC and CO2 pressure of 100 bars (1450 psi) that
decreases to less than 35.7 g at 80jC. Solubility increases
with increasing pressure of CO2. For example, increasing
CO2 from 10 to 100 bars (145 to 1450 psi) in a 10 wt.%
(1.90 molal) salt solution at 60jC increases the solubility from 5.04 to 31.27 g/kg of water.
CO 2 REACTION WITH WATER
Adding CO2 to water that contains constituents other
than Na and Cl may result in saturation of the water
with carbonate minerals. Water from Mississippian aquifers contains significant calcium and magnesium. Adding CO2 to this water could result in the precipitation
of dolomite and calcite. If the pH is buffered at 6.0 by
reaction with minerals in the rock, then the addition of
approximately 0.5 mol of CO2/kg of water would result
in precipitation of 10.4 g of dolomite and 1.9 g of calcite. These mineral precipitates would occupy 4.5 cm3
(0.27 in.3) of pore space. Adding CO2 to water that is
discharged from a CO2 geyser (Crystal Geyser) could
result in the initial precipitation of 2.26 g of dolomite
at a pH of 5.5. The dolomite would dissolve as additional CO2 is added, and the pH falls to 4.5. However,
the Navajo Sandstone is notably lacking in appropriate mineral buffers. Little calcite is present (average =
0.81 wt.%), and reactions with the principal aluminosilicate mineral K-feldspar are very slow.
Adding 2.37 mol of CO2 to water typical of the
Navajo Sandstone in Farnham Dome (0.36 wt.% salinity)
raises P CO2 to 138 bars (2001 psi) and lowers the pH to
3.8 (Figure 6). No minerals reach saturation at this low
pH value. If the Navajo Sandstone contains calcite at
these conditions, then pH is buffered by CO2-H2Ocalcite at a higher value of 4.6, but no minerals are precipitated (Figure 6). The buffer involves solution of
0.13 cm3 (0.007 in.3) of calcite in 1 kg of fluid.
Adding 0.85 mol of CO2 to water typical of Mississippian aquifers (25 wt.% salinity) raises P CO2 to 138 bars
(2001 psi) and lowers the pH to 3.2 (Figure 6). Despite
the higher Ca levels in this water, no minerals reach
saturation. If the pH is buffered by calcite at 4.7, then
4.5 cm3 (0.27 in.3) of calcite dissolves. Note that the
average Navajo Sandstone contains 31 cm3 (1.9 in.3)
of calcite in 5000 cm3 (305 in.3) of rock with 1000 cm3
(61 in.3) of pore space (20% porosity).
WATER MIXING
Mixing of waters of diverse composition may result in
mineral precipitation with attendant changes in porosity and permeability. The effect is most pronounced
Figure 5. Solubility of CO2 in waters of varying salinity in
terms of weight percent NaCl. Solubilities are calculated at 60jC
using the equations of Duan and Sun (2003).
Parry et al.
101
reactions. Reaction products are a function of the proportion of reacting CO2 compared with rock mineral
reactants. The reaction of K-feldspar with fluid is slow
compared to the rate at which CO2 dissolves. Therefore, the reaction simulations proceeded as follows.
First, the equilibrium state of the fluid was calculated;
then CO2 was reacted with the equilibrium fluid; and
finally, the fluid was reacted with K-feldspar.
The reaction rate law for K-feldspar reacting to
product clay minerals used in the simulations is (Bethke,
1998)
þ
rate ¼ As k anHþ ð1 Q=KÞ in mol=cm2 =s
Figure 6. Plot of pH change as CO2 is added to various waters
at 60jC. (a) Plot of pH change as CO2 is added to Farnham water
at 60jC. The pH is buffered by calcite, CO2, and water. (b) Plot
of pH change as CO2 is added to Farnham water at 60jC. No
mineral buffers are included in the calculation. (c) Plot of pH
change at 60jC as CO2 is added to water from Mississippian –
age strata. No mineral buffers are included in the calculation.
when water that contains substantial bicarbonate from
shallow, low-salinity aquifers is mixed with calciummagnesium brine from deeper aquifers. Mixing bicarbonate water from Farnham Dome with saline brine
from a Mississippian aquifer results in the initial precipitation of dolomite and dawsonite (NaAlCO3(OH)2) as
shown in Figure 7. The dolomite begins to dissolve with
increasing mixing fraction of saline water because of
dilution of the bicarbonate content of Farnham Dome
water. Dawsonite was also formed in the CO2 injection
model experiments of White et al. (2005) and in the
CO2 reaction models of Kharaka et al.(2006). Moore
et al. (2005) observed dawsonite in the SpringervilleSt. Johns field, Arizona and New Mexico, where CO2
leaks to the surface 5.
ð1Þ
where A s is the surface area; k + is the rate constant;aaHH+þ
is the activity of hydrogen ion; Q is the reaction quotient;
and K is the equilibrium constant. The overall rate of
reaction of K-feldspar in the rock is rate multiplied by
surface area in square centimeters per gram times the
number of grams of K-feldspar in the rock. The rate constant k + at 25jC is 1.168 10 14 mol/cm2/s, and the
activation energy is 51.7 kJ/mol (Blum and Stillings,
1995). The rate constant at an elevated temperature
may be calculated using the Arrhenius equation
þ
kt ¼ AeE=RT
ð2Þ
where A is the pre-exponential factor 1.347 10 5,
E is the activation energy, R is the gas constant, and
T is the absolute temperature. The rate constant calculated for 60jC is 1.047 10 13 mol/cm2/s.
REACTION OF WATER AND CO 2 WITH ROCK
The consequences of reacting aquifer waters with reactive minerals in the rock as CO2 is added were evaluated by treating CO2 as a reactant along with feldspar
in the rock. Calcite and clay were not treated as reactants
because the solutions were saturated or nearly saturated
with these minerals, and addition as reactant simply results in the precipitation of the amount of reactant added.
Clays and carbonates instead became products in the
102
Figure 7. Plot of minerals precipitated and dissolved as Mississippian water is added to Farnham water at 60jC.
Geochemistry of CO2 Sequestration in the Jurassic Navajo Sandstone
The mean grain size of K-feldspar in the Navajo
Sandstone from measurements of Hood and Patterson
(1984) and from our own measurements is 0.125 mm
(0.0049 in.). The specific surface area of K-feldspar is
374 cm2/g (57.9 in.2/g), calculated using the equation
of Brantley and Mellot (2000). The reaction of lowsalinity Farnham Dome water saturated with CO2 at
138 bars (2001 psi) CO2 pressure is shown in Figure 8.
No significant mineralogical changes occur within the
first few years. Kaolinite precipitates and then dissolves
as muscovite begins to precipitate. The maximum kaolinite reaches 4 g or 1.5 cm3/kg of water (0.091 in.3/kg
of water). A tiny quantity, 0.02 g, of alunite also precipitates then dissolves. Muscovite continues to precipitate over the 500-yr time span of the simulation,
reaching a maximum of 51 g or 18 cm3 (1.09 in.3). The
solution reaches dolomite saturation at about 225 yr,
and 0.08 g (0.03 cm3; 0.0018 in.3) of dolomite has
precipitated in 500 yr. K-feldspar available for reaction was 488 g or 191 cm3 (11.6 in.3). After 500 yr,
42 cm3 (2.5 in.3) had reacted to 22% of the total available K-feldspar.
Reaction of high-salinity Mississippian formation
water saturated with CO2 at 138 bars (2001 psi) produces no aluminosilicate clays. The solution reaches
saturation with dawsonite (NaAl(CO3)(OH)2) within
the first year; and after 500 yr, 95 g (39 cm3; 2.37 in.3) of
dawsonite precipitated along with 8 g (3 cm3; 0.18 in.3)
of dolomite (Figure 9a). After 500 yr, 72 cm3 (4.4 in.3) of
K-feldspar had reacted (38% of the total). Suppression
of dawsonite results in the precipitation and then dis-
Figure 9. (a) Plot of minerals precipitated as Mississippian
water saturated with CO2 at 138 bars (2001 psi) pressure is
reacted with K-feldspar for 500 yr. (b) Similar to (a), but with
dawsonite suppressed.
solution of 5.8 g (2.2 cm3; 0.13 in.3) of kaolinite within
the first 10 cm3 (0.61 in.3) of K-feldspar reaction. At
500 yr of reaction time, 80.5 g (28.4 cm3; 1.73 in.3) of
muscovite, 9.5 g (3.3 cm3; 0.20 in.3) of dolomite, and 6.1 g
(2.3 cm3; 0.14 in.3) of calcite precipitate (Figure 9b).
LEAKAGE OF CO 2 ALONG FAULTS
Figure 8. Plot of minerals precipitated and dissolved as Farnham water (saturated with CO2 at 138 bars [2001 psi] pressure)
is reacted with K-feldspar for 500 yr.
Oil and gas fields in the Navajo Sandstone and its equivalent Nugget Sandstone include the Anschutz Ranch,
Anschutz Ranch East, Farnham Dome, Lodgepole,
Parry et al.
103
Pineview, and the most recently discovered Covenant
fields (Hill and Bereskin, 1996; Brown, 2005). The presence of oil, natural gas, and CO2 reservoirs, together
with modeling studies, shows that reservoirs and seals
exist that can contain gas for long periods of time
( White et al., 2002; Allis et al., 2003). However, some
natural CO2 reservoirs have failed to contain the gas
in the past and are leaking today. The Carmel Formation has sustained the same sequence of structural deformation as that in the Navajo Sandstone and is therefore broken in places by faults and fracture zones. Some
of these faults are known to transmit CO2 through the
reservoir seal. Two fault systems have provided conduits for discharge of CO2 from the subsurface. The
Crystal Geyser and surrounding springs along the Little
Grand Wash fault system discharge CO2-charged waters and deposit travertine, both modern and ancient.
The Little Grand Wash fault brings CO2-charged water from the Navajo Sandstone to a borehole where the
CO2 and water are regularly discharged in geyser activity. Ten Mile Geyser and Torrey’s Spring also discharge CO2-charged waters from the Salt Wash fault
system with the attendant deposition of travertine. The
CO2 discharge appears in the form of slow leaks instead of catastrophic discharges. The presence of travertine terraces, tufa, and calcite-filled vein structures,
CO2 geysers, and CO2-charged springs at the Little
Grand Wash and Salt Wash fault zones suggests that
CO2-charged water has been discharging from the
fault throughout recent geologic time (Dockrill et al.,
2004; Shipton et al., 2004; Heath et al., in press). Travertine has formed from the leakage of CO2 adjacent to
fault zones from the Springerville-St. Johns field over a
long period of time (Allis et al., 2002; Moore et al.,
2005). Thus, the fault zones that breach the reservoir
seals do not appear to have been sealed by mineral precipitation from CO2-charged water in the fault zone.
The geochemical modeling of the interaction of
CO2-charged water with minerals in the Navajo Sandstone described above results in water that is saturated
with carbonate minerals calcite and dolomite because
of the solution of calcite cement in the sandstone and
the consumption of hydrogen ion by reaction with detrital feldspar in the sandstone. Leakage of this water
to the surface or into shallower aquifers would result
in decreased pressure, CO2 loss, and a rise in pH. Carbonate minerals would reach supersaturation and precipitate along the flow path and as travertine at the
surface. Calculating the consequence of CO2 loss indicates that less than 1 cm3 (0.061 in.3) of carbonate mineral precipitates per kilogram of water as CO2 pressure
104
is lowered to 1 bar (14.5 psi). Negligible quantities of
dolomite and clay minerals are precipitated. These calculations are consistent with observed CO2 leakage
from faults that results in precipitation of travertine, but
does not result in sealing the fault conduit as seepage
continues today (Shipton et al., 2004).
Modeling studies of White et al. (2005) suggest that
neither the Wingate nor the Navajo Sandstones can contain CO2 for long periods of time along the transect that
was modeled and are not suitable for storage because
of leakage through seals.
CONSEQUENCES OF ROCK-FLUID INTERACTION
A small amount of clay mineral (modeled as muscovite)
precipitated in the Farnham Dome simulation (51 g.,
18 cm3 [1.09 in.3]). The total pore space containing
1 kg of water was 1000 cm3 (61 in.3), so the pore volume occupied by clay is very small (1.8% of the total
pore space), decreasing the porosity from the assumed
20% to 19.6%. The decrease in porosity probably would
have a negligible effect on permeability.
The formation of clay minerals as CO2 reacts with
K-feldspar in the rock affects permeability in two ways.
The clay minerals fill pore space and reduce porosity,
and more significantly, the kaolinite platelets and illite
fibers can break loose during fluid flow and partially
block pore throats, the main cause of permeability reduction (Vinopal and Taylor, 1999). Several empirical
relations have been developed to predict the effect of
clay abundance on permeability. The relationship developed by Baker et al. (2000) for sandstones in the Carnarvon Basin in Western Australia is
log k ¼ ð0:137f 0:821 Mgs 0:0476 Cv 1:81 Ro
þ 2:18Þ
ð3Þ
where k = permeability in millidarcys; f = porosity in %;
Mgs = mean grain size in f units; Cv = total clay volume
%; and Ro = vitrinite reflectance in %. This equation
accounts for 87% of the permeability variation, with
most predicted values within one order of magnitude of
measured permeability.
Bloch (1991) formulated a second empirical relationship between permeability and clay content. According to Bloch (1991), the most important factors in
sandstone permeability are detrital composition, grain
Geochemistry of CO2 Sequestration in the Jurassic Navajo Sandstone
size, sorting, and temperature-pressure history. The
empirical equation is
log k ¼ ð1:34 Mgs þ 4:08ð1=SoÞ
þ3:42 ðrigid grain content=100Þ 4:67Þ
ð4Þ
where k = permeability in millidarcys; Mgs = grain size
in millimeters; and So = Trask coefficient = (Q 1/Q 3)0.5,
where Q 1 = particle size in millimeters having 75% of
the sample smaller in size; and Q 3 = particle size in
millimeters having 25% of the sample smaller in size
(Trask, 1932, p. 71; Krumbein and Pettijohn, 1938,
p. 230–231). This equation is valid for sandstones with
less than 5% pore-filling cement.
The contribution of varying clay content on permeability for the empirical equations of Bloch (1991)
and Baker et al. (2000) is plotted in Figure 10, together
with the measured permeability and clay content of
samples of Navajo Sandstone from the northern part of
the San Rafael swell. The two equations are in reason-
Figure 10. Calculated permeability as a function of clay
content of sandstone using the equation of Bloch (1991) and
Baker et al. (2000). Superimposed are measured permeabilities
and clay contents of Navajo Sandstone samples.
able agreement, but neither equation predicts the Navajo Sandstone permeability. Both equations predict a
negligible change in permeability with a change in clay
content of a few volume percent.
In the simulations of CO2 injection into the Navajo
Sandstone, a small amount of calcite is dissolved, increasing the porosity by a small amount. Later, dolomite and
dawsonite are precipitated in pore space, decreasing
the porosity. The maximum porosity change is precipitation of 39 cm3 (2.37 in.3) of dawsonite, which amounts
to a decrease in porosity from 20 to 19.2%. These changes
in porosity probably are insignificant in terms of permeability. It takes, on average, a 4% porosity increase to
cause a one-order-of-magnitude change in permeability (Archie, 1950).
DISCUSSION
To understand and use a potential CO2 sequestration
site, the influence of rock mineralogy and groundwater chemistry should be examined. The temperature,
pressure, and salinity of groundwater into which the
CO2 is to be injected determine the CO2 solubility.
As CO2 is injected, the pH is substantially reduced, unless mineral buffers are present. This decrease in pH
inhibits the precipitation of carbonate minerals and sequestration of CO2 in solid phases. The CO2 thus remains either in solution in the natural groundwater or
as a separate CO2 phase.
The quantity of CO2 that can be stored in solution
in groundwater is determined by the CO2 solubility
and by the quantity of groundwater into which the CO2
can be dissolved. Injection of CO2 into a dynamic groundwater system in which the regional groundwater flow
field continually supplies unsaturated water to the injection site and sweeps away saturated groundwater
down the flow path would result in greater storativity
than injection into a static or low-gradient groundwater
system. The San Rafael swell region of the Colorado Plateau (Figure 1) is bounded on the west by high plateau
recharge regions, and the regional groundwater flow
is from west to east at a nominal velocity of about
2 cm/day (0.08 in./day) ( White et al., 2005). Such
groundwater flow in an injection well would continually present unsaturated water at the site of injection
and remove CO2-saturated water.
The Navajo Sandstone provides the porosity, permeability, structure, and seal for an effective CO2 sequestration reservoir. However, only small amounts of
reactive minerals are present to consume the acidity
Parry et al.
105
produced by dissolving CO2 in the water naturally present in the Navajo Sandstone. The primary reactant
is K-feldspar, with an average abundance of 4.6 wt.%,
but the reaction rate is slow. The maximum K-feldspar
abundance is about 8.5 wt.%. Consequently, little CO2
can be stored as carbonate minerals. Storage of CO2 as
precipitate would require the selection of alternative
strata such as the Permian Kaibab Limestone, which
is also favorably positioned with regard to seals and
would also exhibit a favorable structure.
We show that CO2 would be stored as a dissolved
phase in the water and as free CO2, and thus, the total
volume that can be stored is a function of the volume
of water in the system, its chemistry, the rock porosity,
density of CO2 at storage depth conditions, and the
rates of groundwater movement in the system. These
results have important implications for any sequestration site because clean eolian sandstones have many of
the attributes required for candidate sites.
The fact that the porosity and permeability of Navajo Sandstone are little affected by CO2 injection suggests that coupled multiphase flow and geochemical
reaction simulators that also represent temporal changes
in permeability are not required in evaluating prospective CO2 sequestration strategies in the undeformed
Navajo Sandstone. Because the mineral composition of
faulted Navajo Sandstone differs little from that of the
undeformed sandstone, this conclusion also applies to
deformed Navajo Sandstone. Depending on the mineral
composition of faulted Carmel Formation seal rocks,
however, a fully coupled simulator should be used to
evaluate the potential for fault-controlled leakage from
the sandstone reservoir.
The CO2-charged waters injected into the Navajo
Sandstone reach or nearly reach saturation with the carbonate minerals calcite and dolomite. Upward leakage
through faults in the Carmel seal would result in systematic decreases in CO2 pressure as the fluids travel
toward the surface. The decrease in CO2 pressure leads
to supersaturation in the carbonate minerals and precipitation of carbonate within the fault fractures and
as travertine at the surface as has been observed in the
natural, fault-controlled leakage on the Little Grand
Wash fault and Salt Wash fault systems (Shipton et al.,
2004). This precipitation has not completely sealed
the leakage along the faults as seepage continues today.
The effectiveness of sealing with carbonate precipitate
is related to the coupling between precipitation rate
and fluid flux rate through the seal leakage, but in the
models described above, carbonate precipitation is an
ineffective seal.
106
CONCLUSIONS
More CO2 can be stored in solution in aquifers at lower
temperature and lower salinity because CO2 solubility
in water decreases as salinity and temperature increase.
Addition of CO2 to groundwater reservoirs with dissolved calcium and magnesium may store CO2 as mineral
precipitate, providing that pH is buffered by reaction
with rock, but the mineral precipitate plugs the pore
volume and permeability. The volume of mineral precipitate is related to the proportion of reactant CO2
compared to the proportion of reactant mineral in the
rock. The Navajo Sandstone contains little reactant mineral that would consume H + at a sufficient rate to provide effective buffering. Therefore, little CO2 can be
stored as mineral precipitate. Careful studies should
be done to evaluate the potential for fault-controlled
leakage from the sandstone reservoir through the overlying Carmel Formation seal rocks.
The potential for sequestering CO2 in the Jurassic
Navajo Sandstone is promising because of its favorable
hydrologic properties, geographic distribution, thickness,
geologic structure, seal strata, and mineralogy. The presence of petroleum, natural gas, and CO2 reservoirs in the
Jurassic Navajo Sandstone suggests that long-term storage is possible. Present-day leakage from natural CO2
reservoirs along faults is indicative of a leakage risk.
REFERENCES CITED
Allis, R., S. White, T. Chidsey, W. Gwynn, C. Morgan, M. Adams,
and J. N. Moore, 2001, Natural CO2 reservoirs on the Colorado Plateau and southern Rocky Mountains: Candidates for
CO2 sequestration: Proceedings of the First National Conference on Carbon Sequestration, Washington D.C., May 2001,
U.S. Department of Energy – National Energy Technology Laboratory Report 2001/114: http://www.netl.doe.gov/publications
/proceedings/01/carbon_seq/6a2.pdf (accessed February 1, 2007).
Allis, R. G., J. H. Moore, T. Chidsey, C. Morgan, W. Gwynn, H. H.
Doelling, M. Adams, S. Rauzi, and S. P. White, 2002, CO2
geysers, springs and massive travertine deposits on the Colorado Plateau; examples of natural leakage of fluids saturated in
CO2 (abs.): Geological Society of America Abstracts with Programs, v. 34, no. 6, p. 392.
Allis, R. G., T. Chidsey, C. Morgan, J. N. Moore, and S. P. White,
2003, Geological storage of CO2 beneath the southern Rocky
Mountains –Colorado Plateau region (abs.): AAPG Annual Meeting Expanded Abstracts, v. 12, p. 5.
Antonellini, M. A., and A. Aydin, 1994, Effect of faulting on fluid
flow in porous sandstones: Geometry and spatial distribution:
AAPG Bulletin, v. 79, p. 642 – 671.
Archie, G. E., 1950, Introduction to petrophysics of reservoir rocks:
AAPG Bulletin, v. 34, p. 943 – 961.
Bachu, S., and J. J. Adams, 2003, Sequestration of CO2 in geological
media in response to climate change: Capacity of deep saline
Geochemistry of CO2 Sequestration in the Jurassic Navajo Sandstone
aquifers to sequester CO2 in solution: Energy Conversion and
Management, v. 44, p. 3151 – 3175.
Bachu, S., W. D. Gunter, and E. H. Perkins, 1994, Aquifer disposal
of CO2: Hydrodynamic and mineral trapping: Energy Conversion and Management, v. 35, p. 269 – 279.
Baker, J. C., P. J. Havord, K. R. Maftin, and K. A. R. Ghori, 2000,
Diagenesis and petrophysics of the Early Permian Moogoolo
sandstone, southern Carnarvon Basin, Western Australia: AAPG
Bulletin, v. 84, p. 250 – 265.
Beitler, B., M. A. Chan, and W. T. Parry, 2003, Bleaching of Jurassic
Navajo Sandstone on Colorado Plateau Laramide highs: Evidence of exhumed hydrocarbon supergiants: Geology, v. 31,
p. 1041 – 1044.
Beitler, B., W. T. Parry, and M. A. Chan, 2005, Fingerprints of fluid
flow: Chemical diagenetic history of the Jurassic Navajo Sandstone, southern Utah. U.S.A.: Journal of Sedimentary Research,
v. 75, p. 547 – 561.
Bergman, P. D., and E. M. Winter, 1995, Disposal of carbon dioxide
in aquifers in the U.S.: Energy Conversion and Management,
v. 36, p. 523 – 526.
Bethke, C. M., 1998, The geochemist’s workbench: University of
Illinois, Urbana, Illinois.
Blakey, R. C., 1988, Late Paleozoic and Mesozoic eolian deposits of
the Western Interior of the United States: Sedimentary Geology, v. 56, p. 127 – 151.
Blakey, R. C., 1994, Paleogeographic and tectonic controls on some
Lower and Middle Jurassic erg deposits, Colorado Plateau,
in M. V. Caputo, J. A. Peterson, and K. J. Franczyk, eds., Mesozoic systems of the Rocky Mountain region, U.S.A.: Denver,
Colorado, Rocky Mountain Society for Sedimentary Geology,
p. 273 – 298.
Blakey, R. C., F. Peterson, and G. Kocurek, 1988, Synthesis of late
Paleozoic and Mesozoic eolian deposits of the Western Interior
of the United States: Sedimentary Geology, v. 56, p. 3 – 125.
Bloch, S., 1991, Empirical prediction of porosity and permeability
in sandstones: AAPG Bulletin, v. 75, p. 1145 – 1160.
Blum, A. E., and L. L. Stillings, 1995, Feldspar dissolution kinetics, in
A. F. White and S. L. Brantley, eds., Chemical weathering rates
of silicate minerals: Reviews in Mineralogy, v. 31, p. 291 – 351.
Bowen, B. B., 2005, Sandstone bleaching and iron concretions: An
index to fluid pathways and diagenetic history of Jurassic Navajo Sandstone, southern Utah: Ph.D. thesis, University of Utah,
Salt Lake City, Utah, 221 p.
Bowker, K. A., and P. J. Shuler, 1991, Carbon dioxide injection and
resultant alteration of the Weber Sandstone, Rangely field, Colorado: AAPG Bulletin, v. 75, p. 1489 – 1499.
Brantley, S. L., and N. P. Mellott, 2000, Surface area and porosity
of primary silicate minerals: American Mineralogist, v. 85,
p. 1767 – 1783.
Brown, D., 2005, Covenant field keeping promises; Utah play makes
lots of headlines: AAPG Explorer, v. 26, p. 4, 8.
Bump, A. P., and G. H. Davis, 2003, Late Cretaceous – early Tertiary Laramide deformation of the northern Colorado Plateau,
Utah and Colorado: Journal of Structural Geology, v. 25, p. 421 –
440.
Cater, F. W., 1970, Geology of the salt anticline region in southwestern Colorado: U.S. Geological Survey Professional Paper 637, 80 p.
Chan, M. A., W. T. Parry, and J. R. Bowman, 2000, Diagenetic
hematite and manganese oxides and fault-related fluid flow in
Jurassic sandstones, southeastern Utah: AAPG Bulletin, v. 84,
p. 1281 – 1310.
Conway, M. W., R. D. Barree, J. Hollingshead, C. Willis, and M.
Farrens, 1997, Characterization and performance of injection
wells in the Wingate and Navajo sandstones: Proceedings of
the International Coalbed Methane Symposium, Tuscaloosa,
May 1997: Tuscaloosa, Alabama, University of Alabama, p. 467 –
480.
Cooley, M. E., J. W. Harshbarger, J. P. Akers, and W. F. Hardt,
1969, Regional hydrogeology of the Navajo and Hopi Indian
reservations, Arizona, New Mexico, and Utah, with a section
on vegetation by O. N. Hicks: U.S. Geological Survey Professional Paper 400-B, 61 p.
Cordova, R. M., 1978, Ground-water conditions in the Navajo Sandstone in the Central Virgin River Basin, Utah: Utah Department of Natural Resources Technical Publication no. 40, 64 p.
Corey, A. T., 1986, Mechanics of immiscible fluids in porous media:
Littleton, Colorado, Water Resources Publications, 230 p.
Davis, G. H., 1999, Structural geology of the Colorado Plateau
region of southern Utah with special emphasis on deformation bands: Geological Society of America Special Paper 342,
157 p.
Dockrill, B., Z. K. Shipton, D. L. Kirschner, and J. P. Evans, 2004,
Understanding leakage from a fault-sealed CO2 reservoir in eastcentral Utah; a natural analogue applicable to CO2 sequestration (abs.): Geological Society of America Abstracts with Programs, v. 36, no. 5, p. 541.
Doelling, H. H., 1975, Geology and mineral resources of Garfield
County, Utah: Utah Geological and Mineral Survey Bulletin,
v. 107, 175 p.
Doelling, H. H., 1988, Geology of the Salt Valley anticline and
Arches National Park, Grand County, Utah, in H. H. Doelling,
C. G. Oviatt, and P. W. Huntoon, eds., Salt deformation in the
Paradox region, Utah: Utah Geological and Mineral Survey Bulletin, v. 122, p. 1 – 60.
Doelling, H. H., and F. D. Davis, 1989, The geology of Kane
County, Utah: Utah Geological and Mineral Survey Bulletin,
v. 124, 192 p.
Doelling, H. H., R. E. Blackett, A. H. Hamblin, J. D. Powell, and
G. L. Pollock, 2000, Geology of Grand Staircase-Escalante National Monument Utah, in D. A. Sprinkle, T. C. Chidsey, Jr.,
and P. B. Anderson, eds., Geology of Utah’s parks and monuments: Utah Geological Association Publication 28, p. 189 –
231.
Duan, A., and R. Sun, 2003, An improved model calculating CO2
solubility in pure water and aqueous NaCl solutions from 273
to 533 K and from 0 to 2000 bar: Chemical Geology, v. 193,
p. 257 – 271.
Dunn, D. E., L. J. La Fountain, and R. E. Jackson, 1973, Porosity
dependence and mechanisms of brittle fracture in sandstones:
Journal of Geophysical Research, v. 78, p. 2403 – 2417.
Ennis-King, J., C. M. Gibson-Poole, S. C. Lang, and L. Paterson, 2004,
Long-term numerical simulation of geological storage of CO2
in the Petrel sub-basin, north west Australia: GEODISC, 2004:
http://www.apcrc.com.au/GEODISC (accessed February 1, 2007).
Freethey, G. W., and G. E. Cordy, 1991, Geohydrology of Mesozoic
rocks in the upper Colorado River Basin – excluding the San
Juan Basin – in Arizona, Colorado, New Mexico, Utah and Wyoming: U.S. Geologic Survey Professional Paper 1411-C, 118 p.
Fryberger, S. G., 1979, Eolian-fluviatile (continental) origin of ancient stratigraphic trap for petroleum in Weber Sandstone,
Rangely oil field, Colorado: The Mountain Geologist, v. 16,
p. 10 – 36.
Gunter, W. D., and E. H. Perkins, 1993, Aquifer disposal of CO2rich gases: Reaction design for added capacity: Energy Conversion and Management, v. 9 – 11, p. 941 – 948.
Haszeldine, R. S., O. Quinn, G. England, M. Wilkinson, Z. K. Shipton,
J. P. Evans, J. Heath, L. Crossey, C. J. Ballentine, and C. M.
Graham, 2005, Natural geochemical analogues for carbon dioxide storage in deep geological porous reservoirs, a United
Kingdom perspective: Oil and Gas Science and Technology:
Revue de l’Institut Français du Pétrole, v. 60, p. 33 – 49.
Parry et al.
107
Heath, J., T. E. Lachmar, J. P. Evans, P. T. Kolesar, and A. P. Williams,
in press, Hydrogeochemical characterization of leaking, carbondioxide charged fault zones in east-central Utah with implications
for geologic carbon storage, in B. J. McPherson and E. Sundquist,
eds., Science and technology of carbon sequestration: American
Geophysical Union Monographs.
Heilweil, V. M., G. W. Freethey, B. J. Stolp, C. D. Wilkowske, and
D. E. Wilberg, 2000, Geohydrology and numerical simulation
of ground-water flow in the central Virgin River Basin of Iron
and Washington counties, Utah: State of Utah Department of
Natural Resources Technical Publication no. 116, 178 p.
Hendriks, C. A., and K. Blok, 1993, Underground storage of carbon
dioxide: Energy Conversion and Management, v. 34, p. 949 –
957.
Hill, B. G., and S. R. Bereskin, eds., 1996, Oil and gas fields of Utah:
Utah Geological Association Publication 22, 300 p.
Holloway, S., 2001, Storage of fossil fuel derived carbon dioxide
beneath the surface of the Earth: Annual Reviews of Energy
and the Environment, v. 26, p. 145.
Hood, J. W., and T. W. Danielson, 1979, Navajo Sandstone, a source
of ground water for future energy related development: U.S.
Geological Survey Professional Paper 1150, p. 128.
Hood, J. W., and D. J. Patterson, 1984, Bedrock aquifers in the
northern San Rafael swell area, Utah, with special emphasis on
the Navajo Sandstone: State of Utah Department of Natural
Resources Technical Publication no. 78, 128 p.
Hovorka, S. D., C. Doughty, P. R. Knox, C. T. Green, K. Preuss,
and S. M. Benson, 2001, Evaluation of brine-bearing sands of
the Frio Formation, upper Texas Gulf Coast for geological sequestration of CO2, in First National Conference on Carbon
Sequestration, 2001: U.S. Department of Energy – National
Energy Technology Laboratory Report 2001/114: http://www
.netl.doe.gov/publications/proceedings/01/carbon_seq/4a2%20
.pdf (accessed February 1, 2007).
Howells, L., 1990, Base of moderately saline ground water in San
Juan County, Utah: State of Utah Department of Natural Resources Technical Publication no. 94, 35 p.
Hunt, C. B., 1956, Cenozoic geology of the Colorado Plateau: U.S.
Geological Survey Professional Paper 279, 99 p.
Hunter, R. E., 1977, Basic types of stratification in small eolian
dunes: Sedimentology, v. 24, p. 361 – 387.
Hurlow, H. A., 1998, The geology of the Central Virgin River Basin,
southwestern Utah, and its relation to ground-water conditions:
Utah Geological Survey Water-Resources Bulletin, v. 26, 53 p.
Intergovernmental Panel on Climate Change (IPCC), 2001, Third
assessment report: Climate change: http://www.ipcc.ch/ (accessed
February 1, 2007).
Kharaka, Y. K., D. R. Cole, S. D. Hovorka, W. D. Gunter, K. G.
Knauss, and B. M. Freifeld, 2006, Gas-water-rock interactions
in Frio Formation following CO2 injection: Implications for
the storage of greenhouse gases in sedimentary basins: Geology,
v. 34, p. 577– 580.
Kimball, B. A., 1992, Geochemical indicators used to determine
source of saline water in Mesozoic aquifers, Montezuma Canyon area, Utah: Selected papers hydrologic sciences, 1988 –
1992: U.S. Geological Survey Water-Supply Paper 2340, p. 89 –
106.
Krumbein, W. C., and F. J. Pettijohn, 1938, Manual of sedimentary
petrography: New York, Appleton-Century Co., 549 p.
Lackner, K. S., 2003, A guide to CO2 sequestration: Science, v. 300,
p. 1677 – 1678.
Law, D. H.-S., and S. Bachu, 1996, Hydrogeological and numerical
analysis of CO2 disposal in deep aquifers in the Alberta sedimentary basin: Energy Conversion and Management, v. 37,
p. 1167 – 1174.
Lindquist, S. J., 1983, Nugget Formation reservoir characteristics
108
affecting production in the overthrust belt of southwestern
Wyoming: Journal of Petroleum Technology, v. 35, p. 1355 –
1365.
McPherson, B. J. O. L., and B. S. Cole, 2000, Multiphase CO2 flow,
transport and sequestration in the Powder River Basin, Wyoming, U.S.A.: Journal of Geochemical Exploration, v. 69 – 70,
p. 65 – 69.
Moore, J., M. Adams, R. Allis, S. Lutz, and S. Rauzi, 2005,
Mineralogical and geochemical consequences of the long-term
presence of CO2 in natural reservoirs; an example from the
Springerville-St. Johns field, Arizona, and New Mexico, U.S.A.:
Chemical Geology, v. 217, p. 365 – 385.
Morgan, T. C., and T. Chidsey, 1991, Gordon Creek, Farnham Dome
and Woodside fields carbon, Carbon and Emery counties, Utah,
in T. C. Chidsey, Jr., ed., Geology of east-central Utah: Utah
Geological Association Publication 19, p. 301 – 309.
Nuccio, V. F., and S. M. Condon, 1996, Burial and thermal history
of the Paradox Basin, Utah and Colorado, and petroleum potential of the middle Pennsylvanian Paradox Formation: U.S.
Geological Survey Bulletin, v. 2000-O, 41 p.
Parry, W. T., J. M. Ballantyne, and N. L. Bryant, 1980, Hydrothermal alteration enthalpy and heat flow in the Roosevelt Hot
Springs thermal area, Utah: Journal of Geophysical Research,
v. 85, p. 2559 – 2566.
Parry, W. T., M. A. Chan, and B. Beitler, 2004, Chemical bleaching
indicates episodes of fluid flow in deformation bands in sandstone: AAPG Bulletin, v. 88, p. 175 – 191.
Pasala, S. M., C. B. Forster, M. D. Deo, and S. J. Lim, 2003, Simulating the impact of faults on CO2 sequestration in sandstone
aquifers (abs.): AAPG Annual Meeting Expanded Abstracts,
v. 12, p. 133.
Perkins, E. H., and W. D. Gunter, 1996, Mineral traps for carbon
dioxide, in B. Hitchon, ed., Aquifer disposal of carbon dioxide.
Hydrodynamics and mineral trapping — Proof of concept: Alberta, Canada, Geoscience Publishing Ltd., p. 93 – 113.
Peterson, F., and C. Turner-Peterson, 1989, Geology of the Colorado Plateau, 28th International Geological Congress Field Trip
Guidebook T130: American Geophysical Union, 65 p.
Pipiringos, G. N., and R. G. O’Sullivan, 1978, Principal unconformities in Triassic and Jurassic rocks, Western Interior U.S. —
A preliminary report: U.S. Geological Survey Professional Paper 1035-A, 29 p.
Pruess, K., T. Xu, J. Apps, and J. Garcia, 2001, Numerical modeling
of aquifer disposal of CO2: Society of Petroleum Engineers,
SPE/EPA/DOE Exploration and Production Environmental Conference, San Antonio Texas, February 2001, SPE Paper 66537,
16 p.
Shipton, Z. K., J. P. Evans, K. R. Robeson, C. B. Forster, and S. H.
Snelgrove, 2002, Structural heterogeneity and permeability in
faulted eolian sandstone: Implications for subsurface modeling
of faults: AAPG Bulletin, v. 86, p. 863 – 884.
Shipton, Z. K., J. P. Evans, D. Kirchner, P. T. Kolesar, A. P.
Williams, and J. Heath, 2004, Analysis of CO2 leakage through
‘‘low-permeability’’ faults from natural reservoirs in the Colorado Plateau, east-central Utah, in S. J. Baines and R. H. Worden,
eds., Geological storage of carbon dioxide: Geological Society
(London) Special Publication 233, p. 43-58.
Sigda, J. M., and J. L. Wilson, 2003, Are faults preferential flow
paths through semi-arid and arid vadose zones?: Water Resources Research, v. 39, no. 8, p. 1225.
Spangler, L. E., D. L. Naftz, and Z. E. Peterman, 1996, Hydrology,
chemical quality, and characterization of salinity in the Navajo
aquifer in and near the Greater Aneth oil field, San Juan County,
Utah: U.S. Geological Survey Water-Resources Investigations
Report 96-4155, 90 p.
Thomas, K., Z. K. Shipton, and J. P. Evans, 2000, Stratigraphic and
Geochemistry of CO2 Sequestration in the Jurassic Navajo Sandstone
structural heterogeneity in faulted eolian sandstone from borehole geophysics and cores, San Rafael swell, central Utah: Geological Society of America Abstracts with Programs, v. 32, no. 7,
p. A411.
Trask, P. D., 1932, Origin and environment of source sediments of
petroleum: Houston, Texas, Gulf Publishing Co., 323 p.
Trimble, L. M., and H. H. Doelling, 1978, Geology and uraniumvanadium deposits of the San Rafael River mining area, Emery
County, Utah: Utah Geological and Mineral Survey Bulletin,
v. 113, 122 p.
van der Meer, L. G. H., 1993, The conditions limiting CO2 storage
in aquifers: Energy Conversion and Management, v. 9 – 11,
p. 959 – 966.
van der Meer, L. G. H., 1996, Computer modeling of underground
CO2 storage: Energy Conversion and Management, v. 37,
p. 1155 – 1160.
van Genuchten, M. Th., 1980, A closed-form equation for predicting the hydraulic conductivity of unsaturated soils: Soil Science
America Journal, v. 44, p. 892 – 898.
Verlander, J. E., 1995, Basin-scale stratigraphy of the Navajo Sandstone: Southern Utah, U.S.A.: Ph.D. thesis, University of Oxford, Wellington Square, Oxford, United Kingdom, 159 p.
Vinopal, R. J., and L. Taylor, 1999, Revisiting the rocks — Upper
Almond Formation: Table Rock field, Sweet Water County,
Wyoming, U.S.A. (abs.): AAPG Bulletin, v. 83, p. 1190.
Weigel, J. F., 1986, Selected hydrologic and physical properties of
Mesozoic formations in the upper Colorado River Basin in
Arizona, Colorado, Utah, and Wyoming excluding the San Juan
Basin: U.S. Geological Survey Water-Resources Investigations
Report 86-4170, 68 p.
White, S. P., R. G. Allis, T. Chidsey, J. N. Moore, and C. Morgan,
2002, Simulation of CO2 injection near the Hunter power
station, central Utah (abs.): Geological Society of America Abstracts with Programs, v. 34, no. 6, p. 390 – 391.
White, S. P., R. G. Allis, J. N. Moore, T. Chidsey, C. Morgan,
W. Gwynn, and M. Adams, 2005, Simulation of reactive transport of injected CO2 on the Colorado Plateau, Utah, U.S.A.:
Chemical Geology, v. 217, p. 387 – 405.
Witkind, I. J., 1991, Implications of distinctive fault sets in the San
Rafael swell and adjacent areas, east central Utah, in T. C.
Chidsey, ed., Geology of east central Utah : Utah Geological
Association Publication 19, p. 141 – 148.
Wright, J. C., and D. D. Dickey, 1963, Relations of the Navajo and
Carmel formations in southwest Utah and adjoining Arizona:
U. S. Geological Survey Professional Paper 450-E, Short Papers in Geology, Hydrology and Topography, p. E63 – E66.
Parry et al.
109