A variety of diagenetic hematite and manganeseoxide deposits occur within well-exposed Jurassiceolian and related deposits of southeastern Utah.Hematite concretions (millimeters to tens of metersin size) and strata-bound layers occur in the perme-able Navajo, Page, and Entrada sandstones. Local-ized manganese oxide deposits without significantiron oxide occur in the overlying rocks covering theSummerville-Tidwell interval. Field, lab, and numeri-cal modeling studies indicate the diageneticdeposits are related to the Moab fault. Fluid inclu-sion studies show salinities of fault fluids range from0 to 19.7 NaCl equivalent weight percent. The δ18O(SMOW) and δ13C (PDB) values of cements andveins range from 7 to 27‰ and –12 to +5‰, respec-tively. The δ87Sr (SMOW) values of these cementsand veins range from 0.210 to 2.977‰, values sub-stantially more radiogenic than Pennsylvanian sea-water. Saline brines formed from solution ofPennsylvanian salts by meteoric water and are inter-preted to have flowed up the Moab fault and out-ward into adjacent permeable rocks. These brinesare reducing from interaction with hydrocarbon,methane, organic acids, or hydrogen sulfide, andthus remove iron, manganese, and 87Sr, and bleachthe sandstones near the fault. The isotopic evidencesuggests multiple episodes of f luid f low up theMoab fault system. When saline, reduced brinesmixed with shallow oxygenated groundwater, ironand manganese oxides were precipitated as cementsto form concretions and tabular deposits in theporous sandstones. Multiple episodes of iron oxidemineralization and concretionary geometries are
1281AAPG Bulletin, V. 84, No. 9 (September 2000), P. 1281–1310.
1Manuscript received April 16, 1999; revised manuscript received March6, 2000; final acceptance March 15, 2000.
2Department of Geology and Geophysics, 135 S. 1460 E., University ofUtah, Salt Lake City, UT 84112-0111; e-mail: [email protected]
We gratefully acknowledge the helpful input of AAPG reviewers EarleMcBride, Peter Huntoon, and M. C. Erskine. Suggestions made by Craig B.Forster and D. Kip Solomon are gratefully acknowledged. We thank J. W.Valley and M. J. Spicuzza for the oxygen isotope analyses of hematite. Thiswork was funded through a University of Utah Mineral Leasing Fund grant.Chan is grateful to Fran Barnes for sharing his ideas and for his expertise inlocating field sites.
Diagenetic Hematite and Manganese Oxides and Fault-RelatedFluid Flow in Jurassic Sandstones, Southeastern Utah1
Marjorie A. Chan, W. T. Parry, and J. R. Bowman2
evident and can be explained as the result of perme-ability heterogeneities in the host rock, presence offavorable nucleii for precipitation, a self-organiza-tion process, or the influence of microbes.
This study emphasizes the nature of the reducingf luid that mobilized iron and its relation to themovement of saline or hydrocarbon fluids along theMoab fault system to precipitate hematite and man-ganese oxides as a result of oxidation of the fluid.The preservation of diagenetic concretionary ironand manganese oxides offers an excellent insightinto permeability contrasts of sandstone units, anunderstanding of mixing fluid compositions, andsolute transport along a major fault system.
INTRODUCTION
Jurassic units of southeastern Utah contain awide variety of unusual diagenetic (postdeposition-al) concretionary iron oxide and manganese oxidedeposits. The spatial distribution of iron oxides isrelated to (1) stratigraphy, depositional structure,and permeability, (2) a source of reducing, salinefluids that were capable of mobilizing iron, and (3)oxidizing meteoric groundwater that mixed withthe saline f luids and precipitated iron and man-ganese oxides.
The localized occurrence and concentration ofthe iron and manganese oxides show a spatial rela-tionship to the Moab fault system (Figure 1). TheJurassic Navajo Sandstone adjacent to the fault hasbeen locally bleached by reduction and removalof hematite, and elsewhere cemented with bothhematite and manganese oxides and hydroxides(Figure 2). Hydrological analysis of present salini-ty in the Navajo aquifer shows that the aquifercontains a high-salinity brine characteristic ofeither oil-field brine or formation water togetherwith fresh meteoric water (Spangler et al., 1996;Howells, 1990). The hydrological history of thearea includes extensive dissolution of the salt coreof the Moab anticline. Work presented here oniron oxide and manganese oxide indicates that thebleaching and mineralization are likely a conse-quence of reducing fluids that derived salt from
Pennsylvanian salt strata and then mixed with shal-low, oxidizing, low-salinity waters.
The objectives of this study are to (1) describethe concretion-bearing rocks and the types of dia-genetic hematite and manganese oxide deposits inthe context of their geological and hydrological set-ting, (2) characterize the f luids from which thehematite precipitated by using fluid inclusion mea-surements and the isotopes of oxygen, carbon, andstrontium, and (3) geochemically model the inter-action of oxidizing and reducing fluids to accountfor the precipitation of hematite and manganeseoxide. The significance of this research is to eluci-date the characteristics and geochemical history ofa body of rocks that has been an important sourceof economic products, such as iron and manganesedeposits, petroleum, natural gas, and water. Theiron oxide concretions appear to be strongly relat-ed to the presence and movement of reducingsaline f luids through the fractures, faults, and
aquifers. In some places hydrocarbons remain astar sands and bitumen veins, but in many places thehydrocarbons are totally missing and leave onlyhints (such as bleaching) of their former presence.The thesis of our work is that the distribution ofthe iron oxide nodules and manganese oxidedeposits can be used as an index to movement ofreducing, saline fluids in the host rocks.
METHODS
Samples of hematite and manganese oxide accu-mulations, sandstone with cements and veins, andfault-related veins were collected from outcrops.Sample locations were recorded using a hand-held GPS(global positioning satellite) receiver and are shownon Figure 1. Cements, iron oxides, manganeseoxides, and clay minerals were identified by x-raydiffraction. Clay samples were prepared for x-ray
1282 Diagenetic Hematite and Manganese Oxides
Figure 1—Location map andstudy area in Grand County,near Moab, Utah. Solid blackcircles show field localitiesof this study. Map modifiedafter Doelling et al. (1987).GPS coordinates given inparentheses. LM = LaSalMountains, MF = Moab fault,LVF = Lisbon Valley fault, RR = Rainbow Rocks(38°41′25N, 109°54′ 59W),back (northwest side) ofRainbow Rocks (38°41′ 51N,109°54′ 33W), RM = RedwallMesa (38°43′10N,109°55′28W), PP = PotholePoint (38°41′38N,109°55′07W), BW = BartlettWash (38°42′50N, 109°47′6W), DP = Duma Point(38°43′10N, 109°55′28W), FF = Freckle Flat (38°44′04N,109°56′43W), DT = Determination Towers(∼ 38°41′ N, 109°45′W), CR =Courthouse Rock(38°42′35N, 109°43′44W), LV = Little Valley (38°36′04′′N,108°40′15W), CT = CastletonTower (38°40′07′′N,109°23′48′′W), AR = Arches(38°37′13′′N, 109°38′10′′W),DW = Dubinky Well (38°42′18′′N,109°53′31′′W), FIM = FlatIron Mesa (38°21′25N,109°26′55W), approximately25 km south of Moab.
Moab
Mn
Mn
MnMn
Mn
MnRR
DP
FF RMBW
DTCR
0 5 10 15 miles
0 25 kilometers5 10 15 20
Gre
e
n R
iv
er
I-70
Salt Valley
U.S.191 *
LVF
LM
FIM
MF
Detail Area
R17E R18E R19E R20E R21E R22E
T21S
T22S
T23S
T24S
T25S
T26S
Explanation
Fault, ball on hanging wall
Salt Anticline showing plunge
Collapsed salt anticline
Oil field
Manganese deposit
Sample locality
Syncline
RR
Mn
C o l ora
d o R i v er
B o o k C l i f f s
LV
CT
AR
DW Moab
Fault
PP
Utah
diffraction analysis by mild hand grinding and thenpeptizing in water using ultrasound and 1–2 mL of5% Calgon solution. The peptized suspensionswere then centrifuged to separate the desired parti-cle size. X-ray mounts were prepared using orient-ed centrifuge concentrates on glass slides. The sam-ples were glycolated for 24 hr at 60°C. Hematitemineral separates for isotopic analyses were pre-pared by grinding, sieving, magnetic separation,and final hand-picking.
Fluid inclusions that lacked a vapor bubble wereheated first to stretch the inclusion, decreasing thedensity of the contents so that a vapor bubble wouldnucleate. The inclusions were then frozen, fol-lowed by gradual heating to observe the tempera-ture of ice disappearance in the presence of liquidand vapor using the techniques described byGoldstein and Reynolds (1994). Salinity was cal-culated using the equation of Bodnar (1993).Fluid inclusions do not usually nucleate vaporbubbles if trapped at a temperature below about60°C (Goldstein and Reynolds, 1994), establishinga maximum temperature of entrapment.
The carbon and oxygen isotopic analyses of calcitewere made with standard acid dissolution, extraction,and mass spectrometry techniques (McCrea, 1950).The oxygen isotope analyses of hematite were per-formed at the University of Wisconsin Stable IsotopeLaboratory. The laser-aided BrF5 extraction technique(Sharp, 1990; Valley et al., 1995) was modified withthe addition of an airlock to the laser fluorination sys-tem that avoids partial reaction during sample pre-treatment as described by Spicuzza et al. (1998).Strontium was separated from hematite by acid disso-lution followed by cation exchange chromatography.
Strontium isotope compositions were measured bystandard techniques of thermal ionization mass spec-trometry (Nier, 1947). Isotope compositions of carbon,oxygen, and strontium are expressed in the δ notation,relative to SMOW (Craig, 1961) for oxygen and stron-tium, and relative to PDB (Craig, 1957) for carbon.
Transport and precipitation of iron as a conse-quence of groundwater mixing is modeled here usingthe computer programs SOLVEQ and CHILLER (Reed, 1982).
GEOLOGIC SETTING AND REGIONALCONTEXT
The study area (Figure 1) is located near Moab,Utah, where diagenetic iron oxide structures arelocated south of the northern splays of the Moabfault system. This region of southeastern Utah isunderlain by the Paradox basin, a late Paleozoicintracratonic basin filled with a mixture of carbon-ate, clastic, and evaporite sediments (Nuccio andCondon, 1996). The Jurassic units were reddenedby hematite formed from iron oxidation during sub-aerial exposure (Turner, 1980) and diagenesis(Walker, 1975; Walker et al., 1978, 1981); however,two prominent sandstones, the Navajo Sandstoneand the Moab Tongue, appear to have undergonebleaching from reduction of hematite and removalof the iron (Foxford et al., 1996).
STRATIGRAPHY
Marginal marine to nonmarine Jurassic units ofsoutheastern Utah are generally flat lying and well
Chan et al. 1283
Figure 2—Exposure of the Moabfault juxtaposing Jurassic Entrada–Slick Rock member (Jes)against Cretaceous Cedar Mountain Formation (Kcm). Location: Bartlett Wash (Figure1), view to the northwest. Note the bleached Slick Rocksandstones in the footwall.Arrows indicate the relativethrow in the fault; upthrownblock on the left, and downthrown block on the right.
exposed in this northern portion of the ColoradoPlateau. Six stratigraphic units (Figure 3) affectedby diagenetic iron oxide are discussed in strati-graphic ascending order from oldest to youngest,and important primary depositional characteristicsrelevant to fluid movement are presented for eachunit. Regional Jurassic paleogeography and paleo-climatology are summarized in Kocurek and Dott(1983), Blakey et al. (1988), Parrish and Peterson(1988), Chandler et al. (1992), Blakey (1994), andPeterson (1994). Ranges of permeabilities for eachJurassic unit and related underlying units areshown in Table 1.
Navajo Sandstone
The Jurassic Navajo Sandstone and its relatedequivalents form the largest eolian dune sea depositin North America (Peterson and Turner-Peterson,1989). Regional boundaries of the Navajo Sandstoneare traceable across the Colorado Plateau (e.g.,Pipiringos and O’Sullivan, 1978; Peterson andPipiringos, 1979). Previous work on this erg (e.g.,Peterson, 1988; Blakey et al., 1988; Sampson, 1992;Blakey et al., 1996; Verlander, 1995) provides a well-defined stratigraphic and sedimentologic context.
Within the study area (Figure 1), the NavajoSandstone is a well-sorted, fine-grained quartzarenitedominated by large-scale eolian cross-stratification(up to a few tens of meters). Locally there are contorted
sets from soft-sediment deformation in the upperportion (few tens of meters) of the formation. Afew thin interdune playa limestone deposits occurtoward the top of the formation and now forminverse topography where erosion at the end ofNavajo deposition left the playa limestones asresistant caps. Internally, the Navajo Sandstone iswhite colored in the study area, yet the same for-mation is orange colored just east of the Moabfault, at Moab, and along the Colorado River. Thewhite color is due to bleaching (e.g., Foxford etal., 1996) from reducing fluids. The homogeneityof the well-sorted sandstone and its preservedporosity and permeability likely allowed large vol-umes of reducing fluids to flush through the unit.The Navajo Sandstone is also an aquifer unitthroughout regions of the Colorado Plateau (Hoodand Patterson, 1984; Howells, 1990; Kimball, 1992;Spangler et al., 1996).
Page Sandstone
The Page Sandstone is locally present (generallyonly a few meters thick) and ranges from a basalchert-pebble-conglomerate lag above the J-2 uncon-formity (Pipiringos and O’Sullivan, 1975) to acoarse-grained sandstone, to sabkha and horizontalsand sheet deposits (Figure 3). The unit is thickestin the study area at the Courthouse Rock locality(Figure 1), where there is an erosional hollow
1284 Diagenetic Hematite and Manganese Oxides
EA
RLY
JU
RA
SS
IC
WINGATE SS
KAYENTA FM
NAVAJO SS
J-2E
NT
RA
DA
SS
SLICKROCK MBR
MOAB TONGUE
TIDWELL MBR
SALT WASH MBR
BRUSHY BASIN MBR
MO
RR
ISO
N F
M
~ 1
50 m
DEWEY BRIDGE MBR
PAGE SS (LOCALLY)
eolian
eolian
eolian
sabkha
coastal
fluvialtidal-flood plain
fluvial
flood plain-lacustrine
VERY HIGH
HIGH
MODERATE
HIGH
LOW
LOW
FORMATIONS PERMEABILITYENVIRONMENT
SUMMERVILLE FM }LA
TE
Interbedded mudstones +fine-grained sandstones
Large-scale, cross-beddedsandstones
Channel sandstones
Conglomerate
J-5M
IDD
LE
Figure 3—Jurassicunits in the Moab,Utah area. J-2 indicates the unconformity at thetop of the NavajoSandstone and base of the PageSandstone (Pipiringosand O’Sullivan,1978). The J-5(Pipiringos andO’Sullivan, 1978) is at the base of the Tidwell member of the Morrison Formation. Modifiedafter Peterson andTurner-Peterson(1989).
Chan et al. 1285T
able
1. R
epo
rted
Per
mea
bil
itie
s o
f M
ajo
r St
rati
grap
hic
Un
its
in t
he
Stu
dy A
rea
Rel
ativ
eFo
rmat
ion
/Mem
ber
an
d F
acie
sP
erm
eab
ility
(m
d)
Ref
eren
ceP
erm
eab
ility
*
Jura
ssic
Mo
rris
on
Fm
Bru
shy
Bas
inV
ery
low
Salt
Was
hV
ery
low
to
low
Tid
wel
lSu
mm
ervi
lleV
ery
low
Entr
ada
0.01
–140
0W
eige
l (19
86)
Ver
y lo
w t
o m
od
erat
eM
oab
To
ngu
e60
0–50
00A
nto
nel
lini a
nd
Ayd
in (
1995
)Sl
ick
Ro
ck10
–100
0A
nto
nel
lini a
nd
Ayd
in (
1995
)D
ewey
Bri
dge
Low
Pag
ein
terd
un
e/ex
tra
erg
(mea
n)
665
Ch
and
ler
et a
l. (1
989)
win
d r
ipp
le (
mea
n)
2289
Ch
and
ler
et a
l. (1
989)
grai
n f
low
(m
ean
)78
20C
han
dle
r et
al.
(198
9)N
avaj
o10
00–5
000
An
ton
ellin
i an
d A
ydin
(19
95)
Ver
y lo
w t
o m
od
erat
e4.
6–39
00W
eige
l (19
86)
Nea
r M
oab
fau
lt (
< 2
km
)60
0–50
00A
nto
nel
lini a
nd
Ayd
in (
1995
)
Kay
enta
0.3–
500
Wei
gel (
1986
)V
ery
low
to
low
Win
gate
18–5
10W
eige
l (19
86)
Ver
y lo
w t
o m
od
erat
e
Tri
assi
cC
hin
le1–
140
Wei
gel (
1986
)V
ery
low
to
low
Mo
enko
pi
1–2
Wei
gel (
1986
)V
ery
low
to
low
Per
mia
nC
utl
er5
(mea
n o
f 12
)W
oo
dw
ard
-Cly
de
(198
2)
Pen
nsy
lvan
ian
Ho
nak
er T
rail
2.5
(mea
n o
f 72
)W
oo
dw
ard
-Cly
de
(198
2)P
arad
ox
2.5
(mea
n o
f 72
)W
oo
dw
ard
-Cly
de
(198
2)
*Fro
m H
ood
and
Pat
ters
on (
1984
).
(Havholm et al., 1993). The Page Sandstone thick-ens southwestward into a separate erg (Havholm etal., 1993), and it is lithologically similar to theunderlying Navajo Sandstone. Initial permeabilityin the Page Sandstone was high (Chandler et al.,1989) (see Table 1), particularly where the unit iscoarse grained. This high permeability likely led tosome of the stratiform early iron oxide cementationdescribed in later sections. The Page Sandstone isconformable with the overlying, finer grained,Dewey Bridge Member of the Entrada Sandstone.
Entrada Sandstone
The Entrada Sandstone contains three membersin stratigraphic ascending order: the Dewey BridgeMember, the Slick Rock Member, and the MoabTongue (Figure 3). Each member exhibits distinc-tive lithologies and facies, which, in turn, affectedfluid movement and circulation.
Dewey Bridge MemberThe Dewey Bridge Member of the Entrada
Sandstone overlies the Page Sandstone and containsinterbedded sandstones, siltstones, and mudstones.The Dewey Bridge Member is a complex packageof sabkha and eolian deposits and is correlative tothe Carmel Formation, which is a marine equiva-lent to the west (Blakey et al., 1983, 1988, 1996).The red-colored, interbedded mottled sandstonesand siltstones have local bed-scale breccias in thestudy area, and largely indicate deposition in asabkha environment. The upper stratigraphic halfof the Dewey Bridge Member contains contortedand crinkled bedding with chert pieces represent-ing likely replacements for dissolved evaporites.The contorted bedding was first noted duringearly studies by Gilluly and Reeside (1927), Baker(1933, 1946), Dane (1935), and McKnight (1940).Sandstone pipes interpreted to have formed fromstrong ground motion (Alvarez et al., 1998) indi-cate probable injection of sand extending upwardthrough sabkha deposits and in places reachingclose to the base of the Slick Rock Member of theEntrada Sandstone. The overall dominance of silt inthis Dewey Bridge unit marks this interval as a low-permeability zone with only localized fluid circula-tion through sandier units (such as the pipes),when present.
Slick Rock MemberThe Entrada Sandstone is another dune-sea
deposit (Kocurek, 1981; Kocurek and Dott, 1983)that exhibits facies variations on a regional scale.The Slick Rock Member of the Entrada Sandstone is
a resistant sandstone, largely eolian in origin, withalternating white-, pink-, and salmon-colored layers.It represents a shift from the silt-dominated DeweyBridge Member to a wet-eolian dune system withhorizontal stratification alternating with meter-scale dune sets. Within the Slick Rock Member,there is a gray-colored, 9-m-thick tar-saturated sand-stone representing a sabkha deposit with abundantsoft-sediment deformation features. The gray tarsand is flanked both above and below by 5 m and11–12 m, respectively, of bleached yellowish orangeto grayish orange sandstones at the Rainbow Rockslocality (Figure 4). The tar-saturated unit commonlyexhibits concretionary iron oxide deposits. TheSlick Rock Member is interpreted as an interval ofmoderate permeability (Antonellini and Aydin,1995), which was not bleached as are the higherpermeability units (e.g., Navajo Sandstone).Permeability of the Entrada Sandstone is also het-erogeneous and ranges from 10 to 1000 md (Table1). Once oil infiltrated the lower permeabilitysabkha deposit (tar-sand unit), heterogeneities inthe crinkly bedding likely caused the oil to remaintrapped in the stratigraphic-bound unit, sand-wiched between wet-eolian dune deposits withmoderate permeabilities.
Moab TongueThe Moab Tongue is the third member of the
Entrada Sandstone. This relatively thin unit pinch-es out to the south and west. This unit is largelywhite-colored and commonly jointed, and is thusdistinguished from the underlying pink- to salmon-colored Slick Rock Member, which lacks the com-mon joints. The Moab Tongue is fine grained andcontains horizontal stratification and cross-strati-fied eolian dune sets. This unit represents transi-tional beach to eolian dune deposits (Kocurek andDott, 1983). Permeability in the Moab Tongue waslikely high and perhaps enhanced by jointing(Huntoon, 1988; Antonellini and Aydin, 1995).The Moab Tongue is also white-colored, andbleaching similar to the Navajo Sandstone isimplied.
Summerville Formation and Tidwell Memberof the Morrison Formation
The thin-bedded red sandstone and mudstoneinterval overlying the Moab Tongue in the study areaincludes the Summerville Formation (Baker et al.,1952), as well as the Tidwell member of the MorrisonFormation (Doelling, 1985). The SummervilleFormation here is only on the order of several metersthick and is composed of noncalcareous red beds(Doelling, 1993; Doelling and Morgan, 1996; H. H.
1286 Diagenetic Hematite and Manganese Oxides
Doelling, 2000, personal communication). The J-5unconformity of Pipiringos and O’Sullivan (1978)occurs at the base of the Morrison Formation
(Figure 3) and separates the Summerville For-mation from the overlying limy and cherty beds ofthe Tidwell member (Doelling and Morgan, 1996;
Chan et al. 1287
A
B
C
Figure 4—(A) Gray-colored tar sand (arrow) insabkha deposits of the Slick Rock Member ofthe Entrada Sandstone. Location: RainbowRocks (RR, Figure 1). (B) Bed-scale lamination,cross-bedding, and contorted layers in the tarsand sabkha interval (from A). (C) Weathered-outhematite concretions from beds of (B).
H. H. Doelling, 2000, personal communication).The Tidwell member in the study area is approxi-mately 10 m thick or less (H. H. Doelling, 2000,personal communication). Because the stratigraph-ic units here are so thin, we herein refer to thesered beds as the Summerville-Tidwell interval. Thisred-bed interval above the Moab Tongue representsa marine incursion of thin, interbedded mudstonesand fine-grained sandstones deposited in a coastalto tidal setting. This fine-grained interval served asa confining layer (Table 1) and hence retained itsred coloration.
STRUCTURE
The study area is located in the fold and fault beltof the Paradox basin. The evaporites of the Penn-sylvanian Paradox Formation are nearly 2 km thickand have been deformed to produce a series ofnorthwest-southeast–trending salt anticlines where
the salt is locally thickened to more than 4000 m(Cater, 1970). Four periods of structural activityrelated to salt movement were recognized byDoelling (1988). First, the period of most active saltmovement formed a series of eight salt anticlines inthe region (300–225 Ma, Table 2), including theMoab and Lisbon Valley anticlines. The Moab anti-cline is an example of a small exhumed hydrocar-bon paleoreservoir (Garden et al., 1997). Second, aperiod of localized salt movement took placewhere the salt was thickest. The resulting topo-graphic relief resulted in erosion of overlying sedi-ments (225–100 Ma, Table 2). Third, the salt struc-tures were buried by as much as 2 km ofpost-Triassic sediments. Fourth, the salt structureswere exhumed about 37 Ma (Nuccio and Condon,1996). The salt was dissolved by groundwater, andthe Moab and Lisbon Valley grabens were formed.The relative timing of these four events is shown inTable 2. The study area displays several faults influ-enced by salt tectonism and salt dissolution.
1288 Diagenetic Hematite and Manganese Oxides
Table 2. Summary of Reported Ages of Geologic Events in the Study Area*
Event Age Evidence Reference
Salt Accumulation Pennsylvanian Stratigraphy Doelling (1988)and references therein
Salt Intrusionmost active 300–225 Ma Unconformities and
thinning of postsaltsediments Doelling (1988)
localized movement 225–100 Ma Missing sediments Doelling (1988)Exhumation of Anticlines Tertiary NA Wood (1968)
37 Ma Modeling Nuccio andCondon (1996)
Salt solution Late Cenozoic Huntoon (1988)10–0 Ma Anticline collapse Doelling (1988)<5 Ma Incision of Colorado River Huntoon (1988)
Oil Generation 79 Ma Numerical model Nuccio andCondon (1996)
Oil Migration 35–40 Ma Fluid inclusions, apatite fission trackages, and time-temperature Huntoon et al. (1999)modeling
Fe reduction Late Cretaceous to Spatial association withEarly Tertiary veins and Moab fault Foxford et al. (1996)43–63 Ma Paleomagnetics R. Garden (2000)
personal communicationMoab fault 50–60 Ma K-Ar dating of illite Pevear et al. (1997)
early phase Pre-Late Jurassic NA Foxford et al. (1996)later phase Post-middle Cretaceous NA Foxford et al. (1996)
Tertiary Doelling (1988)freshwater incursion 50–60 Ma K-Ar ages and δ18O Pevear et al. (1997)
La Sal Mountains 25–28 Ma 40Ar-39Ar age Nelson et al. (1992)Manganese precipitation 21–26 Ma 40Ar-39Ar age Chan et al. (1999)Colorado Plateau uplift 0–5.5 Ma River deposits Lucchitta (1979)
Hunt (1969)
*Events are organized oldest (top) to youngest (bottom). NA = not applicable.
The salt anticlines lie above and are parallel withbasement faults that are related to salt tectonismactive from the Pennsylvanian to the present (e.g.,Atwood and Doelling, 1982; Doelling, 1985;Huntoon, 1988; Oviatt, 1988; Foxford et al., 1996).The influence of the basement faults coupled withextension along the anticlines could have producedthe incipient Moab fault, with subsequent dissolu-tion and collapse of the anticlines to create theMoab Valley (e.g, Doelling, 1985).
The Moab fault (Figures 1, 2) is a 45-km-longnorthwest-striking normal fault in the northeastpart of the Paradox basin and forms the southwestmargin of the collapsed Moab Valley. Much of thefield geology (fault geometries, diagenesis, andcementation relationships) of the Moab fault sys-tem is mapped and discussed by Foxford et al.(1996, 1998) and Garden et al. (1997, 1998). Muchof the maximum displacement of about 1 km wascaused by collapse of Pennsylvanian and youngerrocks into the dissolving core of a salt anticline(Foxford et al., 1996). Maximum throw on the faultplaces the Jurassic Morrison Formation in the hang-ing wall against Paradox Formation in the footwallnear location AR in Figure 1. Two phases of activityhave been recognized on the Moab fault by Foxfordet al. (1996) and Doelling (1988): a pre–LateJurassic and a post–middle Cretaceous phase. K-Ardating and δ18O composition of illite related to thefault by Pevear et al. (1997) shows an incursion ofmeteoric water at 50–60 Ma (Table 2).
Many localities and occurrences of iron oxidedeposits occur relatively close to the Moab fault(Figure 1) and are well exposed in the study area.Veins of bitumen, sparry calcite, pyrite, and otherminerals are present along and within 250 m of theMoab fault. Bleached Entrada and Navajo sandstonesfrom which red hematite has been removed can betraced more than 7 km from the fault (Garden et al.,1997, 1998; Foxford et al., 1996). These characteris-tics show that the fault was a conduit for migrationof aqueous solutions and hydrocarbons.
The evolution of Lisbon Valley, located southeastof the Moab anticline and fault (Figure 1), is similarto Moab Valley. The Lisbon Valley anticline is faultedalong its longitudinal axis by the 34-km-long LisbonValley normal fault. The Lisbon Valley fault is relatedto growth of a salt-cored anticline with major upliftand offset along the Lisbon Valley fault during theTertiary (Wood, 1968). At the crest of the LisbonValley anticline, the displacement is about 1200 mwith Pennsylvanian Honaker Trail Formation faultedagainst Cretaceous Dakota Sandstone (Wood, 1968;Shawe, 1970; Morrison and Parry, 1986). Dis-placement is 600 m with Chinle Formation faultedagainst Morrison Formation 13 km northwest of theanticline crest. Almost all branch faults and majorfractures are parallel or subparallel to the Lisbon
Valley fault and the long axis of the anticline.Extensive erosion of the Lisbon Valley anticlinestarted in the late Eocene, and 1500 m of sedimentswere removed. The Lisbon Valley fault has been thelocus of saline fluid movement and mineralizationthat includes manganese oxide deposits andbleaching of sandstones (Kennedy, 1961; Jacobs,1963).
The burial history of the Paradox basin has beendetermined from stratigraphic reconstructions byassuming that thicknesses of Tertiary and Cre-taceous strata removed from the Paradox basin areaby erosion are similar in thicknesses to areasaround the periphery of the basin, such as theBook Cliffs and the Henry Mountains (Nuccio andCondon, 1996; Huntoon et al., 1999). These recon-structions indicate the following sequence ofevents: (1) maximum burial in the Cretaceous–mid-dle Tertiary with oil generation in the Paradoxbasin in the Late Cretaceous (Nuccio and Condon,1996), (2) oil migration into the White Rim Sand-stone in the Tertiary (Huntoon et al., 1999), and (3)hydrocarbon movement along the Moab fault in theLate Cretaceous–early Tertiary (Foxford et al.,1996). The maximum temperature reached inJurassic sediments during burial in the Moab area ofthe Paradox basin was estimated to be 79°C(Nuccio and Condon, 1996).
HYDROGEOLOGY
Present-day groundwater in the Navajo Sand-stone includes brines with salinities of nearly20,000 mg/L dissolved solids, up to 19 mg/L dis-solved Fe, and 0.3 mg/L dissolved Mn (Kimball,1992; Spangler et al., 1996), as well as water withlow salinity. Detailed geochemical studies ofSpangler et al. (1996) indicate that the source ofhigh-salinity fluids in the Navajo Sandstone is theresult either of upward movement of water from anupper Paleozoic aquifer or from dissolution ofevaporite. Hood and Patterson (1984) also docu-mented movement of brine from below the NavajoSandstone upward in the areas east of the SanRafael swell, where relatively high artesian pres-sures are present. Water quality data and formationwater resistivities from geophysical well logs showa thick layer of moderately saline (3000–10,000mg/L) groundwater under the eastern two-thirds ofSan Juan County. Very saline (10,000–35,000 mg/L)to briny (35,000–400,000 mg/L) waters lie beneaththe moderately saline groundwater in or near areasunderlain by evaporite beds (Howells, 1990).
Circulation of groundwater has been responsiblefor four important events: (1) dissolving the saltfrom the cores of the salt anticlines and attendantcollapse of the anticlines, (2) bleaching, reduction,
Chan et al. 1289
and mobilization of iron spatially associated withthe Moab fault, (3) precipitation of vein minerals inthe fault and cements in rocks near the fault, and(4) precipitation of iron and manganese oxides.The groundwater f low is controlled by surfacetopography, rock permeabilities, and tectonicstructure.
Topographic relief allowed for the developmentof hydraulic gradients in the groundwater sys-tem. Topographic relief from uplift of the saltanticlines is indicated by consequent streamsf lowing westward from the southwest f lank ofthe Moab salt anticline in the late Tertiary–earlyQuaternary (Oviatt, 1988). The formation of the LaSal Mountains in the Oligocene resulted in 1800 mof doming that would influence groundwater flow(Hunt, 1958). The latest phase of uplift of theColorado Plateau and incision of the ColoradoRiver in the Pliocene also provided steep hydraulicgradients that enhanced removal of the salt(Huntoon, 1988).
The Jurassic eolian sandstones and associatedfluvial and sabkha deposits are aquifers separatedfrom the salt beds of the Paradox Formation byconfining aquitards of the Triassic Chinle andMoenkopi formations. Access of groundwater fromthe Lower Jurassic aquifers to the Pennsylvaniansalt was provided by erosion and structure. Erosionremoved the Moenkopi and Chinle confining lay-ers. Solution-collapse of the salt anticlines in-creased permeability of overlying layers due tofaulting, and extensional fractures developed in thecrests of the salt anticlines (Huntoon, 1988).
FIELD OCCURRENCES OF DIAGENETICHEMATITE
The Jurassic strata of the study area contain a vari-ety of very dusky red, brownish black, and black-colored (Munsell Rock Color Chart) iron oxideconcretions and cemented sandstones. In these con-cretions, hematite cement commonly composes upto 25% of the rock. Some iron deposits occur in lay-ers that are referred to as ferricretes (Wright et al.,1992). The variety of iron oxide deposits studiedhere are summarized in Table 3.
Small concretions of secondary iron oxidedeposits are exclusively hematite-cemented sand-stone. These concretions form all kinds of shapesfrom millimeters to centimeters in diameter, andcut across primary bedding structure. Some con-cretions are solid spherical balls around unknownnuclei; others are spherical rinds with plain red-stained, host rock interiors (lacking black-coloredhematite cement on the inside). Concretions alsoexhibit forms that appear to have coalesced orjoined (similar to two glued marbles), or exhibit a
variety of other odd shapes (buttons, disks, irregu-lar knobby forms to spiked-looking balls). Manyconcentric spherical concretions are locally knownas Moki marbles, a name associated with prehis-toric Indians. These small concretions are found inthe Navajo Sandstone, the Slick Rock Member ofthe Entrada Sandstone, or the Moab Tongue of theEntrada Sandstone.
Small iron oxide concretions (Figure 4) are com-mon in the tar sand unit of the Slick Rock Memberof the Entrada Sandstone. Here, the presence of thehydrocarbons implies reducing conditions.
Larger cylindrical pipes and columns that cutacross primary cross-bedding are common withinthe upper several meters to tens of meters of theNavajo Sandstone (Figures 5, 6). These range fromcentimeters to tens of centimeters in diameterand can extend several meters vertically (Figure7). Many of the pipes and columns appear to begrouped, clustered, and even aligned as though fol-lowing a fault or spring line (Figure 7B), eventhough no obvious structural control is evident onthe surface. In some places, pipes and columns arevertically slanted or inclined and hematite-cementedpipes are offset along the eolian foresets (Figure7C). These cylindrical pipes and columns are simi-lar to the small concretions in that they can betotally hematite-cemented or can be only thin (mil-limeter thick or more) outer pipes and rinds ofhematite.
On one well-exposed upper surface at the top ofthe white-bleached Navajo Sandstone are hundredsof hematite columns that have been planed off(broken) and exposed at localities at the back ofRainbow Rocks, Freckle Flat, Pothole Point, andRedwall Mesa (Figure 1). The columns typicallyhave a thick-cemented outer rind (centimetersthick) of hematite and can have varying degreesof hematite cement to a just faint red-coloredhematite staining (cement) in the core. The diame-ter of the columns is generally 10–40 cm. The out-ermost edges range from sharp to diffuse, possiblydue to groundwater movement that has streakedthe red hematite column sides/edges to create ashadow zone (showing streaked downf lowgroundwater movement and a clean, sharp upflowboundary). The result is a “comet tail” of residualhematite-stained sand that can be seen in the topcross sectional views (Figure 8). Rose diagrams ofover 150 measurements show a polymodal distribu-tion of inferred fluid flow movement (Figure 9),with average modes of 199–225° azimuths (flow tothe south-southwest) from three separate localities.This is an unusual cementation structure thatappears to preserve the direction of groundwaterflow.
A third type of iron oxide occurrence is erosion-ally resistant, large towers of very dusky red to
1290 Diagenetic Hematite and Manganese Oxides
Chan et al. 1291T
able
3. Fie
ld O
ccu
rren
ces
of
Iro
n O
xid
e D
epo
sits
Des
crip
tio
n o
fSi
zeC
emen
tP
rese
nt
in J
ura
ssic
Des
ign
atio
nD
iage
net
ic P
rod
uct
sD
imen
sio
ns
Co
mp
osi
tio
nFo
rmat
ion
s
Co
ncr
etio
ns
Dis
cret
e co
ncr
etio
nar
y fo
rms
cut
acro
ssM
illim
eter
s to
cen
tim
eter
sH
emat
ite
Nav
ajo
San
dst
on
e(M
oki
Mar
ble
s)p
rim
ary
stru
ctu
re, s
ph
eric
al b
alls
to
dis
cs,
in d
iam
eter
Slic
k R
ock
En
trad
ab
utt
on
s, s
pik
ed b
alls
, an
d o
ther
od
d s
hap
es(p
inh
ead
to
pea
siz
e,m
arb
le t
o g
olf
bal
l siz
e)
Pip
es a
nd
Cyl
ind
rica
l, cu
ts a
cro
ss p
rim
ary
Cen
tim
eter
s to
ten
s o
fH
emat
ite
Nav
ajo
San
dst
on
e to
Co
lum
ns
stru
ctu
re (
cro
ss b
edd
ing)
, flu
idce
nti
met
ers
in d
iam
eter
Pag
e Sa
nd
sto
ne
mig
rati
on
alo
ng
eolia
n f
ore
sets
Ten
s o
f ce
nti
met
ers
late
r gr
ou
nd
wat
er f
low
pro
du
cin
gve
rtic
ally
“co
met
tai
ls”
To
wer
sFo
llow
s p
rim
ary
stru
ctu
re o
fM
eter
s in
dia
met
erH
emat
ite
Dew
ey B
rid
ge t
om
ediu
m- t
o c
oar
se-g
rain
ed
up
to
10
m v
erti
cally
low
er S
lick
Ro
ck E
ntr
ada
san
dst
on
e p
ipes
Ferr
icre
teSt
rati
form
alo
ng
pri
mar
y b
edd
ing
inT
ens
of
cen
tim
eter
s th
ick
Hem
atit
eP
age
San
dst
on
eco
arse
-gra
ined
san
dst
on
eT
ens
of
met
ers
late
rally
(ch
ert
peb
ble
co
ngl
om
erat
e)
Stra
tifo
rm a
lon
g p
rim
ary
bed
din
gM
eter
s th
ick
Pyr
olu
site
,Su
mm
ervi
lle-T
idw
ell
Ten
s o
f m
eter
s la
tera
llycr
ypto
mel
ane,
calc
ite
vein
s
Lies
egan
gC
on
cen
trat
ed c
olli
form
ban
din
gT
ens
of
met
ers
late
rally
Hem
atit
e,N
avaj
o S
and
sto
ne
Ban
din
gem
anat
es f
rom
fra
ctu
res
pyr
olu
site
,cr
ypto
mel
ane
Co
llifo
rm b
and
ing
to d
iffu
seT
ens
of
met
ers
late
rally
Iro
n o
xid
esN
avaj
o S
and
sto
ne
Hu
es a
nd
Ho
mo
gen
eou
s w
hit
e to
tan
, yel
low
, red
Ho
mo
gen
eou
sIr
on
ox
ides
All
stu
die
d f
orm
atio
ns
Co
lora
tio
n
black hematite-cemented sandstone up to 10 mhigh. These appear to be fluidized sand injectedas pipelike bodies extending up through theDewey Bridge Member that nearly reach the baseof the Slick Rock Member of the EntradaSandstone. They are best exposed at Duma Point(Figure 1), where a dozen or more exhibit prefer-ential hematite cementation (Figure 10). Althoughthese sandstones are preferentially cemented byhematite, some zones contain more hematite thanothers.
A fourth occurrence of the iron oxide is a ferri-crete (Wright et al., 1992), strata-bound layer ofhematite-cemented sandstone. These ferricretesare largely localized in the Page Sandstone, where achert-pebble conglomerate and coarse-grainedsandstone occur above the Navajo Sandstone. The
ferricrete (Figure 11) is typically up to about 1 mthick and extends for tens of meters laterally, but itis clearly a local phenomenon. The hematitecement commonly encases pebble- to gravel-sizepolished and pitted chert fragments. The ferricreteis exposed in the Freckle Flat localities and theback of Rainbow Rocks (Figure 1), and typicallyoccurs in association with the pipes and columnsthat were conduits connecting f low from theunderlying Navajo Sandstone. This ferricrete canbe very dense, hard, and a metallic-looking verydusky red to black color where hematite has perva-sively cemented the sandstone. In other areas,very localized ferricretes are in fine-grained sand-stones (e.g., Determination Towers, Figure 1).
Manganese-rich zones of a totally different com-position than the hematite ferricretes (Table 4)
1292 Diagenetic Hematite and Manganese Oxides
A
B
Jp
2 m
Figure 5—(A) Weathered-outhematite columns at the top of theNavajo Sandstone. (B) Side viewsof hematite columns. Jp = PageSandstone, Jn = Navajo Sandstone.Location: back (northwest side) ofRainbow Rocks (RR, Figure 1).
occur in the Summerville-Tidwell interval andappear to be stratiform along bedding, typically infine-grained sandstone beds. Manganese oxidedeposits consisting of pyrolusite and manganite areup to 1 m or more thick and extend for tens ofmeters laterally. The manganese oxide occurs asup to 27% cement (Table 4), but is also present asvein fillings. Within a given sample, where themanganese oxide cement is present, iron is absentas indicated by whole-rock analyses (Table 4).Although other manganese oxide deposits havebeen reported in early studies (Baker et al., 1952),we focused on the manganese deposit at Flat IronMesa (Figure 1). Here, both hematite and the man-ganese oxides are associated with vein and fracturefilling along a steep fault striking 310–320°. Themineralization includes hematite, pyrolusite, andcryptomelane, cementing quartz sand grains of thehost Navajo Sandstone. The manganese oxides alsooccur as several irregular colliform bands, ema-nating from the fracture (Figure 12A). Small mil-limeter-diameter fingers of manganese oxide min-erals are contained within colliform bands (Figure12B, C). Each finger is cored by cryptomelane-hollandite with a rim of pyrolusite and an outerrim of hematite, with red sandstone occupying theinterfinger area.
Liesegang banding of all sizes, scales, and varia-tions is present in many of the iron oxide and man-ganese localities described; however, Liesegangbanding does not have any obvious correlation tothe iron oxide concentrations.
Spatial occurrence of iron and manganese oxidessoutheast of the Moab fault and mapping ofhematite reduction by Foxford et al. (1996) indi-cate that this reduction was contemporaneous withlast phases of Tertiary faulting along the Moab faultsystem. Occurrences of the diagenetic hematiteresponsible for the red coloration indicate that theywould have been present prior to bleaching andthus might have been deposited fairly soon afterJurassic deposition.
Other workers report fractures in the NavajoSandstone that are completely filled with ironoxide in the northern San Rafael swell (Hood andPatterson, 1984) north of our study area. Theseiron oxide–filled fractures could also be related toLiesegang banding or post-Jurassic faulting.
PETROGRAPHY AND MINERALOGY
Observations of more than 40 thin sections fromrepresentative field samplings indicate two possibleparagenetic sequences in the concretionary hema-tite deposits. Diagenetic minerals were identified inthin sections (Figure 13), and compared with x-raydiffraction analysis of clays. Alteration of alumino-silicate minerals in the sandstones to clay minerals iscommonly observed. The detrital feldspars, both pla-gioclase and K-feldspar, are typically altered to kaoli-nite, illite, and interstratified illite-smectite (I/S). TheI/S is R > 3 ordered and contains 90% illite layers. Inone paragenetic sequence, detrital quartz grains are
Chan et al. 1293
BA
Figure 6—Cross sectionalside views of weatheredhematite columns at the top of the NavajoSandstone. Locations: (A) Redwall Mesa and (B) back (west side) ofRainbow Rocks (RM, RR,Figure 1).
coated with iron oxide prior to early quartz over-growths (Figure 13A). In a second parageneticsequence, illite or I/S or both illite and I/S coatingssurround detrital quartz grains, and where quartz
overgrowths are present, the illite-I/S coats thequartz overgrowth (Figure 13B).
Hematite is the only iron oxide identified by x-raydiffraction. Manganese deposits contain pyrolusite,
1294 Diagenetic Hematite and Manganese Oxides
A
B
C
Figure 7—Tall clustered hematite columns (red) in thewhite (bleached) Navajo Sandstone: (A) side sectionalview, (B) oblique view, and (C) side view (column diam-eter ∼ 20 cm) showing preferred orientation of mineral-ization along eolian foresets (note small millimeter-sizeconcretions also concentrated along foresets). Location:Redwall Mesa (RM, Figure 1).
Figure 8—(A, B) Top cross sectional “comet tails” of ironoxide halos surrounding hematite columns in white(bleached) Navajo host rock. Arrows indicate inferredgroundwater flow directions. Location: back of RainbowRocks (RR, Figure 1).
B
A
manganite, and rare cryptomelane-hollandite. Weobserved no pseudomorphs of hematite afterpyrite nor other evidence that the hematite formedfrom oxidation of pyrite. Reduced (bleached) sand-stones at the bases of bleached zones are reportedto contain cubic hematite pseudomorphs of pyrite(Garden et al., 1997), and rare pyrite veins on theMoab fault are reported by Foxford et al. (1996).Occurrences of malachite and azurite (CR and RRlocalities of Figure 1) may have formed from oxida-tion of copper sulfide.
Samples of the concretionary iron oxide de-posits exhibit hematite cementation and replace-ment ranging from a stain over previous cements,to a pervasive hematite that masks cements, matrixclay, and clay coatings; thus well-rounded quartzgrains are almost f loating in a hematite cement.Samples of manganese oxide deposits exhibit earlyquartz overgrowths followed by illite coatings, thenmanganese oxide cements followed by a late-stageof calcite infilling cement.
Within the hematite concretions, quartz grainsare coated with a thick layer of hematite, and euhe-dral, hexagonal plates of hematite protrude into theunfilled pore space (Figure 13C). Hematite alsoreplaces or masks illite grain coatings. In a few con-cretion samples, porosity is filled with carbonatecements, most commonly calcite (Figure 13D), butlocally siderite is present. Some calcite pore fillingscontain disseminated hematite. Early hematite pres-ent between detrital quartz grains and the quartzovergrowth implies that hematite was present priorto cementation.
GEOCHEMISTRY
The formation of the hematite and manganeseoxide accumulations must begin with dissolutionof iron and manganese, then transport, and finallyprecipitation. Iron occurs naturally in minerals andfluids in nature in two oxidation states: Fe2+ andFe+3. The Fe+3 in hematite imparts the red colorsso common in the Jurassic Navajo and Entradasandstones. The mobility of iron in solutions isclosely related to oxidation state, pH, and com-plexing ligands such as Cl–. At ordinary tempera-tures (25°C) and geologically reasonable values of
Chan et al. 1295
N = 33 Circle = 27 %
Equal Area
N =93
Circle = 32 %
EqualArea
n = 93ave. = 199o
A
B
C
n = 33ave. = 197o
n = 38ave. = 225o
Figure 9—Rose diagrams depicting southerly directionsof inferred paleogroundwater flow from “comet tails” inthe upper Navajo Sandstone. (A) Back of Rainbow Rocks(RR, Figure 1), (B) Redwall Mesa (RM, Figure 1), and (C)Pothole Point (PP, Figure 1). Small square depicts aver-age vector for each set of data. Although some “comettails” (e.g., Figure 8B) show variable paleoflow direc-tions, the dominant mode is still toward the south-southwest.
pH (4–8), calculated iron concentrations in oxi-dizing solutions (as Fe+3) are typically less than10–6 parts per million (e.g., Stumm and Morgan,1996). Large concentrations of Fe in solution arepossible in reduced solutions, at very low pH,and at elevated salinities. Iron in red sediment isnormally present as Fe+3 (immobile) and must bereduced to Fe2+ or complexed for transport, thenoxidized to precipitate the Fe+3 minerals in con-cretionary nodules. Hematite is reduced bychemical reactions with hydrocarbons, organicacids, methane, or hydrogen sulfide (Garden etal., 1997). Typical chemical reactions for reduc-tion are (modified from Garden et al., 1997) asfollows.
For hydrocarbon:
For methane:
For organic acids:
For hydrogen sulfide:
The consequence of reduction is bleaching ofthe sandstones and removal of iron as Fe2+. Red
2H S Fe O 2H FeS 3H O Fe2 2 3 2 22+ + = + ++ +
CH COOH 4Fe O 16H 2CO 8Fe 10H O3 2 3 22
2+ + = + ++ +
CH 4Fe O 16H 10H O CO 8Fe4 2 3 2 22+ + = + ++ +
CH O 2Fe O 8H CO 5H O 4Fe2 2 3 2 22+ + = + ++ +
sandstones in the region stained red by thinhematite rims and hematite-stained clays werebleached after burial as either oil or fault-relatedreducing brines migrated through the sand-stones and reduction reactions dissolved thehematite rims and stains (Hansley, 1995; Gardenet al., 1997, 1998). The sandstones also containmanganese. An average of 289 samples of 24Paleozoic–Mesozoic sandstones f rom theColorado Plateau is 0.018% MnO (Pettijohn,1963). Laboratory leaching experiments showreducing leach solutions can typically removeiron and manganese from the red beds (Zielinskiet al., 1983).
1296 Diagenetic Hematite and Manganese Oxides
Figure 10—Hematite-cementedtowers (arrows) of resistant sandstone pipes in the JurassicDewey Bridge Member. Towersare up to 10 m tall. Location:Duma Point (DP, Figure 1).
Figure 11—Ferricrete, shown at arrow, of the JurassicPage Sandstone interval above the bleached white Nava-jo Sandstone and below the Dewey Bridge Member.Location: Freckle Flat (FF, Figure 1).
Fluid Inclusions
Salinity of f luids in fluid inclusions were mea-sured in a coarsely crystalline calcite vein from theMoab fault (sample 99-10, locality CR of Figure 1).Calcite crystals are 1–2 cm on an edge. The fluidinclusions are one-phase, liquid-filled, secondaryinclusions that occupy healed fractures in thecoarse vein calcite.
Freezing point depressions correspond to salini-ties that range from 0 wt. % NaCl to a maximum of9.9 wt. % NaCl. The most frequently observed val-ues are from 1.6 to 1.7 wt. % NaCl (Figure 14). Breitand Meunier (1990) and Morrison and Parry (1986)measured salinities of fluid inclusions trapped incarbonates related to the Lisbon Valley fault.Salinities ranged from about 0.5 to 9.5 wt. % NaCl(Breit and Meunier, 1990) and from 5 to 19.7 wt %NaCl (Morrison and Parry, 1986).
Isotopic Compositions of Carbonates andHematite
The oxygen and carbon isotope compositions ofcalcite vary widely (Table 5). The δ18O values of
calcite from fault veins within or near the Moab faultwhere the sandstone is bleached range from 7 to15.8‰. These values are significantly lower than theδ18O values of calcite cements (15–25.6‰) andveins (17.6–27.0‰) that occur in unbleached sand-stones well away (>5 km) from the Moab fault (Table5). The δ13C values of calcite cements and veinsremote from the Moab fault range from –6.0 to –1.6and –6.23 to 5.20‰, respectively. The δ13C values ofcalcites from veins within or near the Moab fault aresignificantly lower (–4.7 to –10.3‰). The oxygenisotopic compositions of four hematite separatesfrom iron oxide concretions were also measured.These δ18O values range from –4.7 to 7.0‰.
The δ87Sr values of hematite ranges from 0.210to 2.977‰ (Table 5). The δ87Sr values of two calcitesamples from the Moab fault are –0.327 and0.766‰.
Discussion of Isotope Results
The δ18O and δ13C values of calcites range wide-ly from 7.0 to 27.0 and –10.3 to +5.2‰, respective-ly. Most of this variation cannot be the result ofvariations in temperature. The temperature of
Chan et al. 1297
Table 4. Whole-Rock Analyses of Iron Oxide Samples*
Sample 97-13 97-20 97-5Unit J-Summerville-Tidwell J-Navajo J-Slick Rock EntradaLocality DP RR, Northwest Side RR (Tar Sand Unit)
*Localities are keyed to Figure 1. Note the sample from the younger Jurassic Summerville-Tidwell interval (97-13), which contains significant MnO andvirtually no Fe2O3 in contrast to the iron oxide samples of the Jurassic Navajo Sandstone and Jurassic Slick Rock Member of the Entrada Sandstone. LOI =loss on ignition.
formation of these calcite veins and cements isrestricted to the range of 30 to 80°C, based on theresults of fluid inclusion microthermometry andestimated burial and thermal history (Nuccio andCondon, 1996). Over this range of temperature, thecalculated δ18O value of calcite in exchange equi-librium with a fluid of constant δ18O compositionwould vary by a maximum of 8‰ (Table 5); there-fore, most of the observed range of δ18O values ofthe calcites must be due to variation in the isotopiccomposition of water.
The calculated δ18O and δ13C values of waters inequilibrium with the calcites at 50°C are compiledin Table 5 and shown in Figure 15. Water composi-tions were computed using the fractionation factorsof O’Neil et al. (1969) for oxygen and those ofBottinga (1968) and Mook et al. (1974) for carbon.Fluids with the lowest δ18O and δ13C values are sam-ples from calcite veins associated with bitumen onthe Moab fault (δ18O = –16.8 to –8.0‰; δ13C = –12.5to –6.9‰). The low δ13C values of these fluids indi-cate exchange with 13C-depleted hydrocarbons inthe subsurface. These f luids are likely to have areduced oxidation state. Fluids in exchange equilib-rium with calcite cements and veins away from theMoab fault generally have much higher values (veins:δ18O = –6.2 to 3.3‰, δ13C = –8.4 to 3.0‰; cements:δ18O = –8.7 to 1.9‰, δ13C = –8.2 to –3.8‰).
This large variation in δ18O and δ13C values sug-gests that these calcites were formed by mixing ofan 18O- and 13C-depleted meteoric water with amore 18O- and 13C-enriched water. The 18O- and 13C-depleted meteoric water end member would havehad δ18O values as low as –16 to –17‰. These lowvalues suggest a topographically driven hydrologicsystem characterized by high elevation recharge,deep circulation of meteoric water, and exchangewith hydrocarbons in lower stratigraphic sections.The lightest modern meteoric water in the area,represented by La Sal Mountains snow, has a compa-rable δ18O value of –16.2‰ (Table 5). Pevear et al.(1997) also proposed deep incursion of meteoricwater into the Moab fault at about 50 Ma, based onconventional K-Ar ages of 18O-depleted illite fromhanging wall Morrison Formation and shalydeformed fault zone rocks. They implied that thedated samples never reached temperatures near200°C at which illite would suffer diffusional Arloss. The association of bleached zones and themost 18O- and 13C-depleted calcites with the Moabfault suggest that 18O- and 13C-depleted, reducedmeteoric water ascended the Moab fault into theJurassic section and progressively mixed with 18O-and 13C-enriched water away from the fault zone,bleaching high-permeability zones of the Jurassicsandstone units and precipitating calcite. At 50°C,the 18O- and 13C-enriched water end member wouldhave a δ18O value of at least +1.9‰; however, at a
1298 Diagenetic Hematite and Manganese Oxides
A
B
CFigure 12—Colliform and digitate manganese mineralgrowth (A) along fractures that parallel the Moab faultstructures, (B) side view of hand sample showing cryp-tomelane cores surrounded by pyrolusite rim in ahematite-cemented matrix of Navajo Sandstone, and (C)top cross sectional view of the same. Location: Flat IronMesa (FIM, Figure 1).
more likely 25°C, the minimum required value wouldbe –0.8‰. This end member may be shallow ground-water (recharged at lower elevation) or meteoric waterthat is 18O- and 13C-enriched due to surface evapora-tion prior to recharging local shallow groundwater.
The oxygen and strontium isotopic composi-tions of f luids responsible for deposition of thehematite concretions are plotted in Figure 16.Exchange experiments suggest that oxygen isotopefractionation between ferric oxide and water issmall and is relatively insensitive to change in tem-perature (Bao and Koch, 1999). In addition, there isgenerally thought to be no significant strontium iso-tope fractionation between fluid and precipitatingsolid; therefore, in Figure 16, the oxygen and stron-tium isotopic compositions of the fluids in equilib-rium with hematite are identical values to those ofhematite cement. These δ87Sr values at 0.210 to2.977 are substantially more radiogenic than eitherPermian sea water (–3.328‰), Pennsylvanian seawater (–1.212‰), or barite from the PennsylvanianHermosa Group (–0.860‰) (Breit and Meunier,1990). Phanerozoic clastic sediments are typicallymore radiogenic, averaging 9.788‰ (McDermottand Hawkesworth, 1990). Typically, saline wateracquires 87Sr from reacting with detrital silicates inthe flow path (Hanor, 1994; Russell et al., 1988;Stueber et al., 1984; Banner et al, 1989). Fluidsfrom which the hematite precipitated likelyacquired 87Sr by reaction with the clastic sediments(detrital silicates) within the Paradox basin throughwhich they f lowed, an origin similar to that ofbrines in the Mississippi salt dome province(Russell et al., 1988), Smackover brines in southernArkansas (Stueber et al., 1984), and brines fromcentral Missouri (Banner et al., 1989).
Chan et al. 1299
A
C D
B
0
10
20
30
Breit and Meunier (1990)
Morrison and Parry (1986)
This study
Salinity in Wt. % NaCl
Num
ber
of M
easu
rem
ents
0 5 10 15 20
Figure 13—Photomicrographs ofNavajo (Jn) to Page (Jp)hematite-cementedsandstones. (A) Earlyiron oxide coating(arrow) followed byquartz overgrowth,followed by illite coats(crossed polars, sample97-8b Jn, arrow = 60mm). (B) Illite coatings(arrow) on quartz(crossed polars, sample97-8a Jn, arrow = 60mm). (C) Euhedraliron oxides (arrow)growing into pore(plane light, sample97-81 Jn, arrow = 24mm). (D) Iron oxidecoatings followed by a late stage of calcitepore-fill cement (crossedpolars, sample 97-11aJp, arrow = 150 mm).
Figure 14—Salinities of fluid inclusion fluids in carbon-ate veins associated with the Moab fault (this study) andthe Lisbon Valley fault (Breit and Meunier, 1990; Morri-son and Parry, 1986).
1300 Diagenetic Hematite and Manganese Oxides
Tab
le 5
. O
xyge
n, C
arb
on
, an
d S
tro
nti
um
Iso
top
ic C
om
po
siti
on
of
Cal
cite
, H
emat
ite,
an
d W
ater
Cal
cula
ted
Wat
ers
δ18 O
‰δ1
3 C‰
δ18 O
δ13 C
3050
8030
5080
Sam
ple
Des
crip
tio
nLo
cati
on
*Fm
(‰)
(‰)
(°C
)(°
C)
(°C
)(°
C)
(°C
)(°
C)
Cem
ents
97-2
1Sp
arry
cem
ent
BW
Nav
ajo
21.6
–2.0
–5.8
–2.1
2.2
–4.8
–4.2
–3.8
97-5
aC
emen
tR
REn
trad
a15
.0–5
.6–1
2.4
–8.7
–4.4
–8.4
–7.8
–7.4
97-1
0Sp
arry
cem
ent
FFD
ewey
Bri
dge
25.6
–1.6
–1.8
1.9
6.2
–4.4
–3.8
–3.4
rr-6
Cem
ent
top
tar
san
dR
REn
trad
a19
.2–6
.0–8
.2–4
.5–0
.2–8
.8–8
.2–7
.8rr
-7C
emen
t ta
r sa
nd
gra
y la
yer
RR
Entr
ada
18.1
–4.9
–9.3
–5.6
–1.3
–7.7
–7.1
–6.7
rr-9
Cem
ent
tar
san
d b
ott
om
RR
Entr
ada
19.4
–3.1
–8.0
–4.3
0.0
–5.9
–5.3
–4.9
Vei
ns
> 5
km
fro
m t
he
Mo
ab f
ault
97-1
2aB
asal
san
dst
on
eFF
Pag
e20
.1–3
.1–7
.3–3
.60.
7–5
.9–5
.3–4
.997
-50
Bla
ck c
alci
teD
WSu
mm
ervi
lle-T
idw
ell
17.7
–4.9
–9.7
–6.0
3–1
.7–7
.7–7
.1–6
.7rr
-4C
alci
te v
ein
320
str
ike
RR
Nav
ajo
27.0
5.2
–0.4
3.3
7.6
2.4
3.1
3.4
rr-2
7V
ein
LVW
inga
te23
.1–3
.4–4
.2–0
.63.
7–6
.2–5
.5–5
.2rr
-28
Vei
n
LVW
inga
te23
.9–4
.9–3
.50.
14.
5–7
.7–7
.1–6
.7rr
-29
Vei
n
LVW
inga
te19
.2–6
.2–8
.2–4
.6–0
.3–9
.0–8
.4–8
.0rr
-37
Vei
n
CT
Win
gate
17.6
–4.5
–9.8
–6.2
–1.9
–7.3
–6.6
–6.3
Vei
ns
in a
nd
nea
r (≤
2 km
) th
e M
oab
Fau
ltrr
-22
Mo
ab f
ault
fo
otw
all
CR
Entr
ada
7.0
–4.7
–20.
4–1
6.7
–12.
4–7
.5–6
.9–6
.5rr
-24
Mo
ab f
ault
old
Hw
y 19
1A
RM
orr
iso
n9.
3–4
.9–1
8.1
–14.
5–1
0.2
–7.6
–7.0
–6.7
rr-3
2M
oab
fau
lt h
angi
ng
wal
l, ve
inB
WC
edar
Mo
un
tain
12.3
–6.5
–15.
0–1
1.4
–7.1
–9.3
–8.7
–8.3
rr-3
4M
oab
fau
lt h
angi
ng
wal
l, ve
inA
RM
orr
iso
n8.
0–4
.7–1
9.3
–15.
7–1
1.4
–7.5
–6.9
–6.5
rr-3
6M
oab
fau
lt f
oo
twal
l bit
um
en, v
ein
AR
Cu
tler
15.8
–4.4
–11.
6–8
.0–3
.6–7
.2–6
.6–6
.399
-10-
aM
oab
fau
lt v
ein
CR
Entr
ada
8.8
–7.0
–18.
5–1
4.9
–10.
6–9
.8–9
.2–8
.899
-10-
bEn
trad
a8.
2–6
.9–1
9.1
–15.
5–1
1.2
–9.7
–9.0
–8.7
99-1
0-d
Entr
ada
8.8
–9.8
–18.
5–1
4.9
–10.
6–1
2.6
–11.
9–1
1.6
99-1
0-e
Entr
ada
8.1
–7.0
–19.
3–1
5.6
–11.
3–9
.8–9
.1–8
.899
-10-
fEn
trad
a8.
0–1
0.3
–19.
4–1
5.8
–11.
4–1
3.1
–12.
5–1
2.1
Wat
erB
ig S
pri
ng
–14.
8La
Sal
Mo
un
tain
Sn
ow
(Sp
angl
er e
t al
., 19
96)
–16.
2St
ron
tiu
mSa
mp
leM
iner
alLo
cati
on
87Sr
/86 S
rδ8
7 Sr
δ18 O
97-1
1aH
emat
ite
FF0.
7092
090.
210
0.3
97-4
0H
emat
ite
RR
0.71
0714
2.33
3–4
.7rr
-1H
emat
ite
RR
0.71
0699
2.31
2–0
.2rr
-16
Hem
atit
eD
P0.
7111
712.
977
7.0
rr-2
2C
alci
teC
R0.
7088
28–0
.327
7.0
rr-3
2C
alci
teB
W0.
7096
030.
766
12.3
*See
Fig
ure
1.
The δ87Sr values of two calcite samples from theMoab fault, –0.327 and 0.766‰, are also plotted inFigure 16. The δ87Sr values are also more radio-genic than late Paleozoic seawater. The highervalue overlaps the field of δ87Sr values for hematite.Also shown in Figure 16 are the oxygen and stron-tium isotopic compositions estimated for water
from the Lisbon Valley fault and its southeasternextension to the Dolores zone of faults (Breit andMeunier, 1990). The δ87Sr values of both theDolores and Moab fault-related fluids are also moreradiogenic than the Paradox salt beds. The isotopiccomposition of oxygen from the Lisbon Valley faultlies in the range of hematite from the Moab faultsystem, but samples from the Dolores zone of faultsare depleted in 18O.
The f luids responsible for deposition of thehematite concretions had variable, but high, δ18Oand δ87Sr values. The fluid inclusion measurementsindicate that the fluids responsible for precipitationof calcite within the Moab fault had significantsalinity (up to 9.9 wt. %). In order to transport iron,these fluids must have been reduced. Thus, thesewaters have evolved (presumably from meteoricwater) by flow and reaction with the underlyingsedimentary section. Salts were derived from evap-orites of the Paradox Formation of the HermosaGroup, and 87Sr was derived from the Jurassic redbed clastic sediments. These f luids could havebecome reduced by interaction with hydrocarbons,methane, or organic acids.
One possible mechanism to precipitate hematitefrom these geochemically evolved fluids is to mixthis evolved, reduced deep water with unevolved(relatively less saline and 18O-depleted), oxidized(shallow) meteoric water. The significant range ofoxygen and strontium isotope compositions to neg-ative δ values is broadly consistent with such mix-ing. Breit and Meunier (1990) and Morrison and
Chan et al. 1301
Figure 16—Strontiumand oxygen isotopiccomposition of concretionaryhematite cement andcalcite veins. Isotopiccompositions of minerals from theLisbon Valley faultand Dolores zone offaults (the southeastextension of the Lisbon Valley fault)are also shown.
Figure 15—Calculated δ18O and δ13C isotopic composi-tion of water in equilibrium with calcite veins andcements at 50°C.
Parry (1986) described similar fluids and mixingprocesses from the Lisbon Valley fault. They inter-pret fault fluids at 70–90°C to be evolved isotopi-cally by exchange with evaporites of the ParadoxFormation of the Hermosa Group, extraction of87Sr from red bed clastics, and exchange with 13C-depleted, carbon-rich sedimentary rocks; these flu-ids subsequently mixed with meteoric water.
Geochemical Simulations
The presence of variable saline groundwater inthe Navajo Sandstone, the history of salt dissolutionand faulting, the evidence for upward movement ofsaline water from the Paradox salt along the Moabfault, and the isotope and fluid inclusion data sug-gest a plausible mechanism for hematite precipita-tion. Saline, reduced waters contain dissolved Feand Mn, which upon mixing with shallow, oxy-genated water, would precipitate as hematite.Water compositions for mixing simulations werechosen based on the Navajo aquifer data inSpangler et al. (1996) and are shown in Table 6.The reduced, saline groundwater was adjusted toequilibrium with magnetite and hematite with atotal dissolved iron concentration of 4 mg/L. ThepH of reduced, saline brine was chosen based ondata from Hanor (1994), who observed that assalinity increases due to solution of evaporites, thepH decreases. A value of 6.5 was chosen for pHnear the maximum for salinities of 200,000 mg/L.
The oxygenated water was assumed to contain1 mg/L dissolved oxygen to provide gradual oxida-tion upon mixing with the saline, reduced water. Inthe numerical simulation (Figure 17), small fractionsof oxygenated water were added to the saline,
reduced water, and then minerals were precipitatedto reach equilibrium followed by adding the nextincrement of oxygenated water. Dissolved oxygenin the mixture increases with increasing fraction ofthe oxygenated water (Figure 17), and the oxidationpotential (Eh) increases as shown in Figure 18. At amixing fraction of oxygenated water of 0.1, 1 mg/Lof dissolved Fe precipitates as hematite. At a mixingfraction of 0.6, 90% of the dissolved Fe precipitatesas hematite, and at a mixing fraction of 1.00, essen-tially all of the Fe precipitates as hematite. Saturationof the solution with pyrolusite would be reached ata mixing fraction of approximately 0.75, but man-ganese minerals were not included in the simula-tion. If the δ18O values of the fluid end members aretaken to be the highest measured value of thehematite samples (+7‰) and of modern meteoricwater in the area (–16‰, La Sal Mountain snow),then simple oxygen isotope mass balance calcula-tions indicate that for a mixing fraction up to 0.6 (atwhich point 90% of dissolved Fe precipitates ashematite), the δ18O value of the mixed fluid wouldbe reduced to –6.8‰, a variation of about 14‰.The significant range of measured δ18O values forhematite (from +7‰ down to –4.7‰) is compatiblewith mixing fractions up to about 0.5, consistentwith these oxidation/mixing simulations.
Oxidation of ferrous iron in solution and precipi-tation as hematite generates acidity according tothe reaction:
The acidity produced can be consumed in reactionswith other minerals in the rock such as calcite or illite.
In nature, the primary products of the oxygena-tion of Fe2+ at an oxic-anoxic boundary such as wemodel here are polynuclear aggregates of Fe3+
hydroxides and ferrihydrite (Schwertmann andFischer, 1973; von Gunten and Schneider, 1991).Original precipitates are converted to goethite, lep-idocrocite [γ-FeO(OH)], akaganeite [β-FeO(OH)],and finally to hematite (Berner, 1969). Because onlyhematite was detected in x-ray diffraction scans,we modeled hematite precipitation rather than themetastable original precipitates.
Manganese mineral solubility is determined byoxidation state and pH of the solutions similar toiron mineral solubility; however, manganese miner-als are more soluble, but under much more oxidiz-ing conditions. A mechanism for separation of iron
2KAl Si O (OH) 2H 2K 3Al Si O (OH)3 3 10 2 2 2 5 4+ = ++ +
CaCO H Ca HCO32
3–+ = ++ +
2Fe 0.5O 2H O Fe O 4H22 2 2 3
+ ++ + = +
1302 Diagenetic Hematite and Manganese Oxides
Table 6. Compositions of Saline and Fresh Waters Usedin the Numerical Simulations of Fluid Mixing toPrecipitate Hematite*
*Concentrations in molality (based on Spangler et al., 1996).
and manganese is suggested by the Eh-pH diagramof Figure 18. Present-day saline groundwater in theNavajo Sandstone contains much less dissolved Mnthan Fe. The diagram is constructed assuming Mn isone-tenth of the dissolved Fe. The stability field fordissolved Mn is considerably larger than soluble Fe(Krauskopf, 1957; Maynard, 1983). The model forprecipitation of hematite proposed here involvesmixing of an oxidizing groundwater with a reduced,saline, Fe- and Mn-bearing groundwater. The oxida-tion potential of the resulting mixture increases asthe proportion of oxidizing groundwater increasesalong the reaction path (Figure 17). Hematite pre-cipitates until the groundwater mixture is very oxi-dizing; in fact, more than 90% of the Fe precipitatesas hematite before the solutions reach saturationwith pyrolusite. If the waters are advecting upwardfrom the source of saline fluids as proposed here,then manganese oxide minerals would precipitatestratigraphically above the iron oxides.
Iron and manganese removed from the Navajoand Entrada and underlying red beds during reduc-tion and bleaching constitute a plausible source ofmetals. The metals taken into solution thus precipi-tate upon mixing with oxygenated water.
Geometries
Hematite and manganese oxide deposits andconcretions occur with a variety of geometries.
The diverse geometries may be explained as theresult of permeability heterogeneities in the hostrock, presence of favorable nucleii for precipita-tion, a self-organization process, or the influence ofmicrobes.
Permeability heterogeneities in the host rocksandstones are consistent with large sandstonetowers at Duma Point encased in the fine-grainedDewey Bridge host rock, where the permeablesandstone pipes were conduits for fluid flow. Also,the coarse-grained nature of the chert pebble con-glomerate in the Page Sandstone aided cementationby hematite to form the ferricrete; however, in thesame units and sandstones Liesegang bands cross-cut primary stratification, cross-bedding, and othersedimentary structures, indicating no obvioussmall-scale heterogeneity control.
Some type of nuclei (e.g., organic matter or ironoxide originally present in the sandstone) may cat-alyze the precipitation reactions and preferentiallyprovide precipitation nucleii for cementation.Studies of carbonate concretions show complexlyzoned, late-stage cements throughout the concre-tion (e.g., Mozley, 1996). Although nucleii and suc-cessively added layers or zonation is a commonexplanation for concretions, we found no nucleiiand could see no textural evidence that the concre-tions grew through progressive addition of hema-tite to the outer edge of some nucleii.
Many concretions are simply a crystallizationgrowth with a self-organization process of naturalspacing (e.g., Nicolis and Prigogine, 1977; Orto-leva, 1984; McBride et al., 1999), although it isunclear why concentric growth can occur in onebed and not another. This model can explain smallconcretions such as the Moki marbles found in thetar sand of the Slick Rock Member of the EntradaSandstone (locality RR, Figure 4). We use a modifi-cation of the Ostwald-Liesegang cycle described byOrtoleva (1984) and Ortoleva et al. (1987a, b) toillustrate the formation of the nodules and bands(Figure 19). In this model, reducing, hydrocarbon-bearing solutions f low upward along the Moabfault, invade the red sandstones, and reduce anddissolve iron, thus bleaching the sandstone. Thereducing water is now charged with Fe2+ in solu-tion. The Fe2+-charged solution encounters oxy-genated groundwaters, and upstream diffusion ofO2 together with downstream diffusion of Fe2+ ele-vates the reaction product [(Fe2+)(O2)/(H+)] at suc-cessive times 1, 2, and 3 as shown in Figure 19. Thereaction product reaches a level of supersaturationthat promotes nucleation and precipitation ofhematite at time 3, and the reaction product falls tothe equilibrium level at time 4. Reactants are sup-plied by upstream diffusion of O2 and advection ofFe2+, and the nodule or concretion grows. Whenthe advection front of reducing, Fe2+-charged water
Chan et al. 1303
Figure 17—Grams of mineral precipitated in the numer-ical simulation of mixing of saline, reduced groundwa-ter with low-salinity oxygenated water.
proceeds sufficiently far downstream, upstream dif-fusion of O2 is no longer effective in reducing thereaction product, so a new supersaturation-nucleation level is reached downstream at time 7and growth of a new concretion or band begins attime 8 (Figure 19).
Microbial processes (Coleman, 1993) or nan-nobacteria (McBride et al., 1994) mediate mineralprecipitation and enhance kinetics (Ehrlich, 1996;Langmuir, 1997). Inorganic oxidation of iron andmanganese is very slow in natural waters at ordi-nary pH (4–8). Iron-oxidizing bacteria can increasethe rate of iron oxidation by as much as six ordersof magnitude (Fortin et al., 1997). Even thoughample time is available for precipitation, enhancedprecipitation rates at the site of microbe colonieswould concentrate precipitation there. Iron oxida-tion by iron bacteria in acid media is well known(example: Thiobacillus ferrooxidans), but un-equivocal evidence for bacterial oxidation of ironat neutral pH is lacking (with the possible example:Gallionella ferruginea) (Ehrlich, 1996). We havenot yet been able to document the microbial pro-cesses in our samples, but microbially enhancedprecipitation might explain why some of the ironforms are so clustered and exhibit colliformgrowths (e.g., Liesegang banding, as well as man-ganese pyrolusite and cryptomelane minerals) at avariety of scales.
Our data for iron oxide towers and ferricretesbest fit permeability heterogeneities, where we candetermine different heterogeneities, reflected inlithologic changes; however, the bulk of the con-cretions (Moki marbles), pipes, columns, andLiesegang banding in this study can best beexplained by a combination of self-organization andmicrobial processes where self-organization con-tributes to the physical spacing of the nodulesalong a mixing-water front with some enhance-ment of cementation from microbial processes. Wecannot rule out precipitation on nucleii becausewe cannot prove or disprove the presence ofnucleii.
DISCUSSION
Analyses of concretions (Figures 4C, 5, 11; Table4) indicate that the mineralogic composition ishematite cement within a clean, quartzarenite. Thishematite cement contains iron as Fe+3. The mobilityof iron at low temperatures depends primarily onoxidation potential and pH. Thus, iron is most easilytransported as Fe2+. These relationships imply thatprecipitation as hematite can be due to increase inpH or increase in oxidation potential or both; how-ever, the pH required for significant mobility of Fe+3
would be too low (<4, too acidic) to be geologically
1304 Diagenetic Hematite and Manganese Oxides
Figure 18—Eh-pH diagram showing the stabilities of iron andmanganese minerals with solutionand the reaction path traversedduring mixing.
reasonable, and implies that iron is mobilized asFe2+ in reducing waters, and then subsequent oxi-dation causes the hematite precipitation.
We propose that the iron is transported in theferrous state (Fe2+) in saline waters with 200,000g/L of dissolved solids, near the upper limit mea-sured in fluid inclusions, but undersaturated withhalite. The saline waters were derived from solu-tion of the salt from the Paradox Formation in thecore of the Moab-Spanish Valley anticline. Thelargest accumulations of iron are those at DumaPoint (Figure 1). A single column at Duma Pointshown in Figure 10 is 7.6 m wide and 6.1 m tall.The hematite is distributed heterogeneously in therock, which contains a maximum of 30% hematite
(Table 4). Assuming a conservative 3% hematite,the column contains as much as 2 × 107 grams ofhematite in a rock volume of 278 m3. If the watertransporting the iron contains 5 parts per millioniron (Table 6), then 2.8 × 109 kg of water arerequired for formation of a single tower. Thisapproximates the amount of water discharged in1.6 hr by the Colorado River at Phantom Ranchduring 1921–1962 before construction of GlenCanyon Dam (Kieffer, 1990).
The volume of salt water produced from dissolu-tion of a salt-cored anticline is also estimated. TheMoab anticline is 28.8 km long, and Moab Valleyformed by anticline collapse is 4.8 km wide.Assuming 1 km of salt is removed and the salt water
Chan et al. 1305
Figure 19—Schematic modelof the advection-diffusion-nucleation process of nodule and band growththrough time. (A) Reducingfluids containing Fe2+ flowin the direction of the arrow.The interface between Fe2+-bearing solution andoxygenated groundwater isshown at three differenttimes with increasing concentrations of Fe2+ and O2from time 1 to time 3. (B)The Fe2+-charged solutionencounters oxygenatedgroundwaters, and upstreamdiffusion of O2 together withdownstream diffusion ofFe2+ elevates the reactionproduct [(Fe2+)(O2)/(H+)] atsuccessive times 1, 2, and 3.The reaction product reachesa level of supersaturationthat promotes nucleationand precipitation ofhematite at time 3, and thereaction product falls to theequilibrium level at time 4.Reactants are supplied byupstream diffusion of O2and advection of Fe2+ andthe nodule or concretiongrows. When the advectionfront of reducing, Fe2+-chargedwater proceeds sufficientlyfar downstream, upstreamdiffusion of O2 is no longereffective in reducing the reaction product, so a newsupersaturation-nucleationlevel is reached downstreamat time 7 and growth of a newnodule or band begins attime 8.
is 75% of saturation with halite, then 1.5 × 1015 kgof salt water resulted from solution of the salt coreof the Moab Valley diapiric anticline. Microbialactivity in the porous sandstones that has not yetbeen fully evaluated could have enhanced iron andmanganese solution and precipitation kinetics andinfluenced the geometry of the iron oxide accumu-lations. The volumes of water required by the geo-chemical simulations could be reduced if, as sug-gested by Hansley (1995), organic acids acted asbleaching agents or other complexing ligands werepresent that would permit higher aqueous ironconcentrations.
The geochemical model proposed involves dis-solving Fe and Mn from red beds in a reducing,saline solution; moving the Fe- and Mn-chargedf luid into the Navajo, Entrada, and associatedaquifers in a dynamic flow system; and precipita-tion of hematite and Mn oxides upon encounteringan oxidizing water. A source of the salt is thePennsylvanian Paradox Formation separated fromthe aquifers by confining Triassic Chinle andMoenkopi formations. Communication of solutionsfrom the aquifers through the confining layers isprovided by the Moab fault and smaller faults andjoints related to salt intrusion and collapse of thesalt anticlines.
The large range of δ18O and δ13C values of cal-cites and of δ18O and δ87Sr values of hematites sug-gests that the deposition of these minerals was pro-duced by mixing of evolved reduced waters thatascended the Moab fault with shallow (unevolved),oxidized meteoric waters. The data, however, alsoindicate that there were at least two distinctevolved f luids ascending the Moab fault. Theevolved fluid end member responsible for calcitedeposition was 18O- and 13C-depleted and mixedwith relatively 18O- and 13C-enriched shallow mete-oric water. In contrast, the evolved fluid end mem-ber responsible for hematite deposition was 18O-enriched (and 87Sr-enriched) and appears to havemixed with a relatively 18O- and 87Sr-depleted shal-low meteoric water. Because both of these evolvedfluid end members are reduced, the precipitationof hematite concretions cannot be induced bymixing these two evolved f luids; and by exten-sion, precipitation of the vein calcite and hematiteconcretions is not contemporaneous. The isotopedata suggest the existence of two distinct hydro-logic regimes, operating at different times, inwhich chemically evolved and reduced f luidsderived from depths corresponding to the ParadoxFormation have ascended the Moab fault and subse-quently mixed with a variety of shallow, unevolvedmeteoric waters.
The hydraulic heads that could drive fluid floware produced by topography from (1) topographi-cally high crests of salt anticlines, (2) the intrusive
La Sal Mountains, or (3) topography produced byuplift of the Colorado Plateau and incision of theColorado River system. The timing of these threeevents is shown in Table 2. The Moab fault has along history of movement, but the bleaching is spa-tially associated with later Tertiary fault displace-ment. A measured age on one manganese depositon a southern extension of the Moab fault (localityFIM, Figure 1) is 21–26 Ma (Chan et al., 1999).Topography that was present then to drivegroundwater f low was the La Sal Mountains(25–28 Ma, Table 2) and the crests of the salt anti-clines. Colorado Plateau uplift and incision of theColorado River are too young (5.5 Ma, Table 2);however, because the La Sal Mountains lie to farsoutheast of the study area, that topography mightexplain northerly groundwater flow, but not thesoutherly directions observed (Figures 8, 9). It ismore likely that salt tectonism and faulting pro-duced topographically driven flow, where meteoricwater reached the salt beds and dissolved salt. Thewater was reduced by reaction with hydrocarbon,methane, organic acids, or hydrogen sulfide.Reduced water flowed into red, hematite-stainedsandstone aquifers and dissolved and transportedFe and Mn to sites where interaction with oxy-genated water to locally precipitated hematite andmanganese oxides.
Migration of saline f luids upward along faultzones has been shown to be responsible for deposi-tion of calcite, copper, and other minerals relatedto the Lisbon Valley anticline and fault by Morrisonand Parry (1986) and related to the Moab fault byGarden et al. (1997, 1998). Mixing of two solutionscaused precipitation of the minerals as a result ofchanges in oxidation state and dilution of the salinesolutions.
Stratigraphic, structural, and geochemical rela-tionships of fluid sources and mass transfer of ironand manganese have important applications. First,the presence of iron oxide cement can be used toevaluate past groundwater oxidation states andsalinities for f luids that migrated through andcemented the sandstone. Numerous studies ofgroundwater resources in the Jurassic sandstonesof southeastern Utah have demonstrated that somegroundwater is usable and some is too salty. Thesource of the salty water is probably the underlyingfluid reservoirs, including the Paradox Formation.Thus, the presence of the iron oxides can be anindicator for movement of past reduced, saline flu-ids, and can help predict regions of salt water andfresh water in groundwater planning; furthermore,field and chemical studies of fluid flow and compo-sitions in the Moab fault zone can serve as a valu-able index to tracing the pathways of hydrocarbonsand can explain the presence of economic iron andmanganese deposits.
1306 Diagenetic Hematite and Manganese Oxides
SUMMARY
In the vicinity of the Moab fault, hematite con-cretions, pipes, and strata-bound layers occur inJurassic Navajo Sandstone and adjacent strata of theColorado Plateau that overlie the Paradox basin saltdeposits. Manganese oxide deposits without signifi-cant iron occur in the overlying rocks covering theSummerville-Tidwell interval. Saline brines contain-ing significant iron and manganese in solution areknown to be present in the Navajo aquifer. We pro-pose that mixing of these saline brines with shal-low, oxygenated groundwater accounts for the pre-cipitation of iron and manganese in the porous andpermeable sandstones. The δ18O and δ13C values ofcalcites and δ18O and δ87Sr values of hematites areconsistent with products formed from mixing ofisotopically evolved, reduced waters with shallow,oxidized meteoric groundwaters. The isotope dataalso indicate that there were at least two distinctfluids ascending the Moab fault and at least two dis-tinct shallow groundwaters with which theymixed: (1) an 18O- and 13C-depleted deep waterthat mixed with a relatively 18O- and 13C-enrichedshallow meteoric water to deposit calcite and (2)an 18O- and 87Sr-enriched deep water that mixedwith an 18O- and 87Sr-depleted shallow meteoricwater to deposit hematite.
Numerical simulation of mixing together withrepresentations of mineral stabilities on an Eh-pHdiagram (Figure 18) show that iron can be quantita-tively precipitated as hematite and that iron andmanganese can be effectively separated due to thedifferences in solubility as a function of oxidationpotential.
Calculations of the quantity of water required fordissolution of Paradox basin salt in the core of theMoab Valley anticline indicate that those water vol-umes were sufficient for both the transport andprecipitation of hematite concretions. The salinebrine with hydrocarbons, methane, organic acids,or hydrogen sulfide moved up through the perme-ability created by the Moab fault and laterally intoporous and permeable sandstones. Iron was dis-solved and the sandstones were bleached. Iron wasprecipitated when shallow, oxygenated water wasencountered. Multiple episodes of hematite miner-alization and geometries are evident, showing thatthe saline brine probably moves episodically.Southerly groundwater flow directions and the esti-mated timing of mineralization (middle Tertiary)suggest a topographically driven hydraulic headfrom salt tectonic features such as the Moab Valleyanticline.
The data presented here explain the geometriesof hematite concretions. The study of the mobilityof iron is strongly related to the presence ofhydrocarbon resources and the movement of
saline fluids. The deposition of iron is stronglyrelated to the mixing of reduced fluids with shallow,oxidized meteoric waters. The understanding ofiron mobility has significant applications to tracinghydrocarbon pathways and evaluating naturalresources and contamination transport. This studyoffers new insight into understanding the importanceof fluid compositions, mixing, and solute transportalong a major fault system.
REFERENCES CITED
Alvarez, W., E. Staley, D. O’Connor, and M. A. Chan, 1998,Synsedimentary deformation in the Jurassic of southeasternUtah—a case of impact shaking?: Geology, v. 26, p. 579–582.
Antonellini, M., and A. Aydin, 1995, Effect of faulting on fluid flowin porous sandstones: geometry and spatial distribution: AAPGBulletin v. 79, p. 642–671.
Atwood, G., and H. H. Doelling, 1982, History of Paradox saltdeformation: Utah Geological and Mineral Survey Notes, v. 16,no. 2, p. 1–7.
Baker, A. A., 1933, Geology and oil possibilities of the Moabdistrict, Grand and San Juan counties, Utah: U.S. GeologicalSurvey Bulletin 841, 95 p.
Baker, A. A., 1946, Geology of the Green River Desert–CataractCanyon region, Emery, Wayne, and Garfield counties, Utah:U.S. Geological Survey Bulletin 951, 122 p.
Baker, A. A., D. C. Duncan, and C. B. Hunt, 1952, Manganesedeposits of southeastern Utah: U.S. Geological Survey Bulletin979-B, 157 p.
Banner, J. L., G. J. Wasserburg, P. F. Dobson, A. B. Carpenter, andC. H. Moore, 1989, Isotopic and trace element constraints onthe origin and evolution of saline ground waters from centralMissouri: Geochimica et Cosmochimica Acta, v. 53, p. 383–398.
Bao, H., and P. L. Koch, 1999, Oxygen isotope fractionation inferric oxide-water systems: low temperature synthesis:Geochimica et Cosmochimica Acta, v. 63, p. 599–613.
Berner, R. A., 1969, Goethite stability and the origin of red beds:Geochimica et Cosmochimica Acta, v. 33, p. 267–273.
Blakey, R. C., 1994, Paleogeographic and tectonic controls onsome Lower and Middle Jurassic erg deposits, ColoradoPlateau, in M. V. Caputo, J. A. Peterson, and K. J. Franczyk,eds., Mesozoic systems of the Rocky Mountain region, USA:Denver, Colorado, Rocky Mountain Society for SedimentaryGeology, p. 273–298.
Blakey, R. C., F. Peterson, M. V. Caputo, R. C. Geesaman, and B. J.Voorhees, 1983, Paleogeography of Middle Jurassiccontinental, shoreline, and shallow marine sedimentation,southern Utah, in M. W. Reynolds and E. D. Dolly, eds.,Mesozoic paleogeography of west-central United States:Denver, Colorado, SEPM, Rocky Mountain Section, p. 77–100.
Blakey, R. C., F. Peterson, and G. Kocurek, 1988, Synthesis of latePaleozoic and Mesozoic eolian deposits of the Western Interiorof the United States: Sedimentary Geology, v. 56, p. 3–125.
Blakey, R. C., K. G. Havholm, and L. S. Jones, 1996, Stratigraphicanalysis of eolian interactions with marine and fluvial deposits,Middle Jurassic Page Sandstone and Carmel Formation,Colorado Plateau, U.S.A.: Journal of Sedimentary Research,Section B, v. 66, p. 324–342.
Bodnar, R. J., 1993, Revised equation and table for determiningthe freezing point depression of H2O-NaCl solutions:Geochimica et Cosmochimica Acta, v. 57, p. 683–684.
Bottinga, Y., 1968, Calculation of fractionation factors for carbonand oxygen exchange in the system calcite-carbon dioxide-water: Journal Physical Chemistry, v. 72, p. 800–808.
Breit, G. N., and J-D. Meunier, 1990, Fluid inclusion and δ18O, and
Chan et al. 1307
87Sr/86Sr evidence for the origin of fault-controlled coppermineralization, Lisbon Valley, Utah, and Slick Rock district,Colorado: Economic Geology, v. 85, p. 884–891.
Cater, F. W., 1970, Geology of the salt anticline region insouthwestern Colorado: U.S. Geological Survey ProfessionalPaper 637, 80 p.
Chan, M. A., W. T. Parry, E. U. Petersen, and C. M. Hall, 1999,Iron and manganese mineralization, and timing of fault-relatedfluid flow in Jurassic sandstones, southeastern Utah (abs.):1999 Geological Society of America Abstracts with Programs,p. A-301.
Chandler, M. A., G. Kocurek, D. V. Goggin, and L. W. Lake, 1989,Effects of stratigraphic heterogeneity on permeability in eoliansandstone sequence, Page Sandstone, Northern Arizona: AAPGBulletin, v. 73, p. 658–668.
Chandler, M., D. Rind, and R. Ruedy, 1992, Pangaean climateduring the Early Jurassic: GCM simulations and thesedimentary record of paleoclimate: Geological Society ofAmerica Bulletin, v. 104, p. 543–559.
Coleman, M. L., 1993, Microbial processes: controls on the shapeand composition of carbonate concretions: Marine Geology, v. 113, p. 127–140.
Craig, H., 1957, Isotopic standards for carbon and oxygen andcorrection factors for mass-spectrometric analysis of carbondioxide: Geochimica et Cosmochimica Acta, v. 12, p. 133–149.
Craig, H., 1961, Standard for reporting concentrations of deuteriumand oxygen-18 in natural waters: Science, v. 133, p. 1833–1834.
Dane, C. H., 1935, Geology of the Salt Valley anticline andadjacent areas, Grand County, Utah: U.S. Geological SurveyBulletin 863, 184 p.
Doelling, H. H., 1985, Geological map of Arches National Parkand vicinity, Grand County, Utah, with accompanying text:Salt Lake City, Utah Geological and Mineral Survey Map 74,scale 1:50,000.
Doelling, H. H., 1988, Geology of the Salt Valley anticline andArches National Park, Grand County, Utah, in H. H. Doelling,C. G. Oviatt, and P. W. Huntoon, eds., Salt deformation in theParadox region, Utah: Utah Geological and Mineral SurveyBulletin 122, p. 1–60.
Doelling, H. H., 1993, Interim geologic map, Moab 30′ × 60′Quadrangle, Grand County, Utah, and Mesa County, Colorado:Utah Geological Survey Open File Report 287, 16 p.
Doelling, H. H., and C. D. Morgan, 1996, Interim geologic map ofthe Merrimac Butte Quadrangle, Grand County, Utah: UtahGeological Survey Open File Report 338, 89 p.
Doelling, H. H., G. C. Willis, M. E. Jensen, and F. D. Davis, 1987,Geology and Grand County: Utah Geological and MineralogicalSurvey: Utah State Department of Natural Resources, Salt LakeCity, Utah, 15 p.
Ehrlich, H. L., 1996, Geomicrobiology (3d ed.): New York,Dekker, 719 p.
Fortin, D., F. G. Ferris, and T. J. Beveridge, 1997, Surface-mediatedmineral development by bacteria, in J. F. Banfield and K. H.Nealson, eds., Geomicrobiology: interactions betweenmicrobes and minerals: Reviews in Mineralogy Volume 35,Mineralogical Society of America, p. 161–180.
Foxford, K. A., I. R. Garden, S. C. Guscott, S. D. Burley, J. J. M.Lewis, J. J. Walsh, and J. Watterson, 1996, The field geology ofthe Moab fault, in A. C. Huffman, Jr., W. R. Lund, and L. H.Godwin, eds., Geology and resources of the Paradox basin:Utah Geological Association Guidebook 25, p. 265–283.
Foxford, K. A., J. J. Walsh, J. Watterson, I. R. Garden, S. C.Guscott, and S. D. Burley, 1998, Structure and content of theMoab fault zone, Utah, USA, and its implications for fault sealprediction, in G. Jones, O. J. Flisher, and R. J. Knipe, eds.,Faulting, fault sealing and fluid flow in hydrocarbon reservoirs:Geological Society of London Special Publication 147, p. 87–103.
Garden, I. R., S. C. Guscott, K. A. Foxford, S. D. Burley, J. J. Walsh,and J. Watterson, 1997, An exhumed fill and spill hydrocarbonfairway in the Entrada sandstone of the Moab anticline, Utah,
in J. Hendry, P. Carey, J. Parnell, A. Ruffell, and R. Worden,eds., Migration and interaction in sedimentary basins andorogenic belts: Second International Conference on Fluidevolution, Belfast, Northern Ireland, p. 287–290.
Garden, I. R., K. A. Foxford, S. C. Guscott, S. D. Burley, J. J. Walsh,and J. Watterson, 1998, The spatial distribution of fault-relateddiagenesis around the Moab fault, Utah (abs.): 1998 AAPGAnnual Convention, Salt Lake City, Utah, CD-ROM.
Gilluly, J., and J. B. Reeside, Jr., 1927, Sedimentary rocks of theSan Rafael swell and some adjacent areas in eastern Utah: U.S.Geological Survey Professional Paper 150-D, p. 61–110.
Goldstein, R. H., and T. J. Reynolds, 1994, Systematics of fluidinclusions in diagenetic minerals: SEPM Short Course 31, 198 p.
Hanor, J. S., 1994, Origin of saline fluids in sedimentary basins, inJ. Parnell, ed., Geofluids: origin migration, and evolution offluids in sedimentary basins: Geological Society SpecialPublication 78, p. 151–174.
Hansley, P. L., 1995, Diagenetic and burial history of the Lower PermianWhite Rim Sandstone in the tar sand triangle, Paradox basin,southeastern Utah; U.S. Geological Survey Bulletin 2000-I, 41 p.
Havholm, K. G., R. C. Blakey, M. Capps, L. S. Jones, D. D. King,and G. Kocurek, 1993, Aeolian genetic stratigraphy: anexample from the Middle Jurassic Page Sandstone, ColoradoPlateau: International Association of Sedimentologists SpecialPublication 16, p. 87–107.
Hood, J. W., and D. J. Patterson, 1984, Bedrock aquifers in thenorthern San Rafael swell area, Utah, with special emphasis onthe Navajo Sandstone: State of Utah Department of NaturalResources, Technical Publication no. 78, 128 p.
Howells, L., 1990, Base of moderately saline ground water in SanJuan County, Utah: State of Utah Department of NaturalResources Technical Publication no. 94, 35 p.
Hunt, C. B., 1958, Structural and igneous geology of the La SalMountain, Utah: U.S. Geological Survey Professional Paper294-I, p. 305–364.
Hunt, C. B., 1969, Geological history of the Colorado River, theColorado River Region and John Wesley Powell: U.S. GeologicalSurvey Professional Paper 669-C, p. 59–130.
Huntoon, P. W., 1988, Late Cenozoic gravity tectonic deformationrelated to the Paradox salts in the Canyonlands area of Utah, inH. H. Doelling, C. G. Oviatt, and P. W. Huntoon, eds., Saltdeformation in the Paradox region, Utah: Utah Geological andMineral Survey Bulletin 122, p. 81–93.
Huntoon, J. E., P. L. Hansley, and N. D. Naeser, 1999, The searchfor a source rock for the giant tar sand triangle accumulation,southeastern Utah: AAPG Bulletin, v. 83, p. 467–495.
Jacobs, M. B., 1963, Alteration studies and uranium emplacementnear Moab, Utah: Ph.D. Thesis, Columbia University, NewYork, 227 p.
Kennedy, V. C., 1961, Geochemical studies of mineral deposits inthe Lisbon Valley area, San Juan County, Utah: Ph.D. thesis,University of Colorado, Boulder, Colorado, 155 p.
Kieffer, S. W., 1990, Hydraulics and geomorphology of theColorado River in the Grand Canyon, in S. S. Beus and M.Morales, eds., Grand Canyon geology: New York, OxfordUniversity Press, p. 333–383.
Kimball, B. A., 1992, Geochemical indicators used to determinesource of saline water in Mesozoic aquifers, MontezumaCanyon area, Utah: Selected Papers Hydrologic Sciences,1988–92: U. S. Geological Survey Water-Supply Paper 2340, p. 89–106.
Kocurek, G., 1981, Erg reconstruction: the Entrada Sandstone(Jurassic) of northern Utah and Colorado: Palaeogeography,Palaeoclimatology, Palaeoecology, v. 36, p. 125–153.
Kocurek, G., and R. H. Dott, Jr., 1983, Jurassic paleogeographyand paleoclimate of the central and southern Rocky Mountainregion, in M. W. Reynolds and E. D. Dolly, eds., Mesozoicpaleogeography of the west-central U.S.: SEPM, RockyMountain Section, Rocky Mountain PaleogeographySymposium, Denver, Colorado, p. 101–116.
Krauskopf, K. B., 1957, Separation of manganese from iron in
1308 Diagenetic Hematite and Manganese Oxides
sedimentary processes: Geochimica et Cosmochimica Acta, v. 12, p. 61–84.
Langmuir, D., 1997, Aqueous environmental geochemistry: NewJersey, Prentice Hall, 600 p.
Lucchitta, I., 1979, Late Cenozoic uplift of the southwesternColorado River region: Tectonophysics, v. 61, p. 63–95.
Maynard, J. B., 1983, Geochemistry of sedimentary ore deposits:New York, Springer-Verlag, 305 p.
McBride, E. F., M. D. Picard, and R. L. Folk, 1994, Orientedconcretions, Ionian coast, Italy: evidence of ground water flowdirection: Journal of Sedimentary Research, v. A64, p. 535–540.
McBride, E. F., A. Abdel-Wahab, and A. R. M. El-Younsy, 1999,Origin of spheroidal chert nodules Drunka Formation (lowerEocene), Egypt: Sedimentology, v. 46, p. 733–755.
McCrea, J. M., 1950, On the isotopic chemistry of carbonates anda paleotemperature scale: Journal of Chemical Physics, v. 18,p. 849–857.
McDermott, F., and C. Hawkesworth, 1990, The evolution ofstrontium isotopes in the upper continental crust: Nature, v. 344, p. 850–853.
McKnight, E. T., 1940, Geology of area between Green andColorado rivers, Grand and San Juan counties, Utah: U.S.Geological Survey Bulletin 908, 147 p.
Mook, W. G., J. C. Bommerson, and W. H. Staverman, 1974,Carbon isotope fractionation between dissolved bicarbonateand gaseous carbon dioxide: Earth and Planetary ScienceLetters, v. 22, p. 169–176.
Morrison, S. J., and W. T. Parry, 1986, Formation of carbonate-sulfate veins associated with copper ore deposits from salinebasin brines, Lisbon Valley, Utah: fluid inclusion and isotopicevidence: Economic Geology, v. 81, p. 1853–1866.
Mozley, P. S., 1996, The internal structure of carbonateconcretions in mudrocks: a critical evaluation of theconventional concentric model of concretion growth:Sedimentary Geology, v. 103, p. 85–91.
Nelson, S. T., M. T. Heizler, and J. P. Davidson, 1992, New40Ar/39Ar ages of intrusive rocks from Henry and La Salmountains, Utah: Utah Geological Survey MiscellaneousPublication 92-2, 24 p.
Nicolis, G., and I. Prigogine, 1977, Self-organization in non-equilibrium systems: New York, Wiley, p. 339–351.
Nier, A. O., 1947, A mass spectrometer for isotope and gasanalysis: Review of Scientific Instruments, v. 18, p. 398–411.
Nuccio, V. F., and S. M. Condon, 1996, Burial and thermal historyof the Paradox basin, Utah and Colorado, and petroleumpotential of the Middle Pennsylvanian Paradox Formation: U.S.Geological Survey Bulletin 2000-O, 41 p.
O’Neil, J. R., R. N. Clayton, and T. K. Mayeda, 1969, Oxygenisotope fractionation in divalent metal carbonates: Journal ofChemical Physics, v. 51, p. 5547–5558.
Ortoleva, P. T., 1984, The self organization of Liesegang bandsand other precipitate patterns, in G. Nicolis and F. Baras, eds.,Chemical instabilities: Amsterdam, Riedel, p. 289–297.
Ortoleva, P. T., E. Merino, C. Moore, and J. Chadam, 1987a,Geochemical self-organization I: reaction-transport feedbacksand modeling approach: American Journal of Science, v. 287,p. 979–1007.
Ortoleva, P. T., J. Chadam, E. Merino, and A. Sen, 1987b,Geochemical self-organization II: the reactive-infiltrationinstability: American Journal of Science, v. 287, p. 1008–1040.
Oviatt, C. G., 1988, Evidence for Quaternary deformation in the SaltValley anticline, southeastern 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 122, p. 62–76.
Parrish, J. T., and F. Peterson, 1988, Wind directions predictedfrom global circulation models and wind directions determinedfrom eolian sandstones of the western United States—acomparison: Sedimentary Geology, v. 56, p. 261–282.
Peterson, F., 1988, Pennsylvanian to Jurassic eolian transportationsystems in the western United States: Sedimentary Geology, v. 56, p. 207–260.
Peterson, F., 1994, Sand dunes, sabkhas, streams, and shallowseas: Jurassic paleogeography in the southern part of theWestern Interior basin, in M. V. Caputo, J. A. Peterson, and K. J. Franczyk, eds., Mesozoic systems of the Rocky Mountainregion, U.S.A.: Rocky Mountain Society for SedimentaryGeology, p. 233–272.
Peterson, F., and G. N. Pipiringos, 1979, Stratigraphic relations ofthe Navajo Sandstone to Middle Jurassic formations, southernUtah and northern Arizona: U.S. Geological SurveyProfessional Paper 1035-B, 43 p.
Peterson, F., and C. Turner-Peterson, 1989, Geology of theColorado Plateau: 28th International Geological Congress FieldTrip Guidebook T130, American Geophysical Union, 65 p.
Pettijohn, F. J., 1963, Chemical composition of sandstones—excluding carbonate and volcanic sands: U.S. GeologicalSurvey Professional Paper 440-S, 19 p.
Pevear, D. R., P. J. Vrolijk, and F. J. Longstaffe, 1997, Timing ofMoab fault displacement and fluid movement integrated withburial history using radiogenic and stable isotopes, inJ. Hendry, P. Carey, J. Parnell, A. Ruffell, and R. Worden, eds.,Migration and interaction in sedimentary basins and orogenicbelts: Second International Conference on Fluid Evolution,Belfast, Northern Ireland, p. 42–45.
Pipiringos, G. N., and R. G. O’Sullivan, 1975, Chert pebbleunconformity at the top of the Navajo Sandstone in southeasternUtah, in J. E. Fassett, ed., Canyonlands country: Four CornersGeological Society Guidebook, 8th Field Conference,Canyonlands, p. 149–156.
Pipiringos, G. N., and R. G. O’Sullivan, 1978, Principleunconformities in Triassic and Jurassic rocks, Western InteriorU.S.—a preliminary report: U.S. Geological Survey ProfessionalPaper 1035-A, 29 p.
Reed, M. H., 1982, Calculation of multicomponent chemicalequilibria and reaction processes in systems involvingminerals, gases, and an aqueous phase: Geochimica etCosmochimica Acta, v. 46, p. 513–528.
Russell, C. W., J. B. Cowart, and G. S. Russell, 1988, Strontiumisotopes in brines and associated rocks from Cretaceous stratain the Mississippi salt dome basin (southeastern Mississippi,U.S.A.): Chemical Geology, v. 74, p. 153–171.
Sampson, P. J., 1992, Sedimentology of the Navajo Sandstone,southern Utah, USA: Ph.D. dissertation, University of Oxford,Oxford, England, 291 p.
Schwertmann, U., and W. R. Fischer, 1973, Natural “amorphous”ferric hydroxide: Geoderma, v. 10, p. 237–247.
Sharp, Z. D., 1990, A laser-based microanalytical method for the insitu determination of oxygen isotope ratios of silicates andoxides: Geochimica et Cosmochimica Acta, v. 54, p. 1353–1357.
Shawe, D. R., 1970, Structure of the Slick Rock district andvicinity, San Miguel and Dolores counties, Colorado: U.S.Geological Survey Professional Paper 576C, 18 p.
Spangler, L. E., D. L. Naftz, and Z. E. Peterman, 1996, Hydrology,chemical quality, and characterization of salinity in the Navajoaquifer in and near the Greater Aneth oil field, San JuanCounty, Utah: U.S. Geological Survey Water-ResourcesInvestigations Report 96-4155, 90 p.
Spicuzza, M. J., J. W. Valley, V. S. McConnell, 1998, Oxygenisotope analysis of whole rock via laser fluorination: an air-lockapproach (abs.): Geological Society of America Abstracts withPrograms, v. 30, no. 7, p. 80.
Stueber, A. M., P. Pushkar, and E. A. Hetherington, 1984, Astrontium isotopic study of Smackover brines and associatedsolids, southern Arkansas: Geochimica et Cosmochimica Acta,v. 48, p. 1637–1649.
Stumm, W., and J. J. Morgan, 1996, Aquatic chemistry (3d ed.):New York, Wiley-Interscience, 1022 p.
Turner, P., 1980, Continental red beds: New York, Elsevier, 562 p.Valley, J. W., N. Kitchen, M. J. Kohn, C. R. Niendorf, and M. J.
Spicuzza, 1995, UWG-2, a garnet standard for oxygen isotoperatio—strategies for high precision and accuracy with laserheating: Geochimica et Cosmochimica Acta, v. 59, p. 5223–5231.
Chan et al. 1309
Verlander, J. E., 1995, Basin-scale stratigraphy of the NavajoSandstone: southern Utah, U.S.A.: Ph.D. thesis, University ofOxford, Oxford, England, 159 p.
von Gunten, U., and W. Schneider, 1991, Primary products ofoxygenation of iron (II) at an oxic/anoxic boundary;nucleation, agglomeration and aging: Journal of ColloidInterface Science, v. 145, p. 127–139.
Walker, T. T., 1975, Red beds in the western interior of theUnited States: U.S. Geological Survey Professional Paper 853,p. 49–56.
Walker, T. R., B. Waugh, and A. J. Crone, 1978, Diagenesis in firstcycle desert alluvium of Cenozoic age, south-western UnitedStates and north-western Mexico: Geological Society ofAmerica Bulletin, v. 89, p. 19–32.
Walker, T. R., E. E. Larson, and R. P. Hoblitt, 1981, Nature andorigin of hematite in the Moenkopi Formation (Triassic),Colorado Plateau: a contribution to the origin of magnetism inred beds: Journal of Geophysical Research, v. 86, p. 317–333.
Weigel, J. F., 1986, Selected hydrologic and physical properties of
Mesozoic formations in the upper Colorado River basin inArizona, Colorado, Utah, and Wyoming—excluding the SanJuan basin: U.S. Geological Survey Water ResourcesInvestigations Report 86-4170, 68 p.
Wood, H. B., 1968, Geology and exploitation of uranium depositsin the Lisbon Valley area, Utah, in J. D. Ridge, ed., Oredeposits of the United States, 1933–1967: American Instituteof Mining, Metallurgical, and Petroleum Engineers, p. 770–789.
Woodward-Clyde Consultants, 1982, Geologic characterizationreport for the Paradox basin study region, Utah study areas:volume II, Gibson Dome: Battelle Memorial Institute, Office ofNuclear Waste Isolation, 290 p.
Wright, W. P., R. J. Sloan, B. V. Garces, and L. A. J. Garvie, 1992,Ground water ferricretes from the Silurian of Ireland andPermian of the Spanish Pyrenees: Sedimentary Geology, v. 77,p. 37–49.
Zielinski, R. A., S. Bloch, and T. R. Walker, 1983, The mobility anddistribution of heavy metals during the formation of first cyclered beds: Economic Geology, v. 78, p. 1574–1589.
1310 Diagenetic Hematite and Manganese Oxides
Marjorie A. Chan
Marjorie A. Chan is professor of geology at theUniversity of Utah, where she joined the faculty in 1982.She received her B.S. degree from the University ofCalifornia–Davis and her Ph.D. from the University ofWisconsin–Madison. Her current research focuses on sedi-mentology and stratigraphy in Precambrian throughPleistocene rocks of the Wasatch front in Utah and inMesozoic deposits of the Colorado Plateau.
William T. Parry
William T. Parry is professor of geology and geophysicsat the University of Utah. Former positions include associ-ate professor of geosciences at Texas Tech University,Lubbock, Texas, and exploitation engineer for Shell OilCompany, Midland, Texas. He received B.S. and M.Sdegrees and a Ph.D. in geological engineering from theUniversity of Utah. His research interests are geochemistryand mineralogy related to faults and ore deposits.
John R. Bowman
John Bowman received his B.S. degree from the Collegeof William and Mary (1969), his M.S. degree from OhioState University (1971), and his Ph.D. from the Universityof Michigan (1977). He has been a faculty member at theUniversity of Utah since 1977, where he specializes inpetrology and geochemistry. His research focuses on usingstable isotopes to evaluate fluid flow and fluid-rock interac-tion in a variety of crustal hydrothermal environments,including groundwater systems.
