An evaporitic-source model for igneous-related

Fe oxide(-REE-Cu-Au-U) mineralization
Mark D. Barton and David A. Johnson
Center for Mineral Resources, Department of Geosciences,
University of Arizona, Tucson, Arizona 85721

Barton, M.D., and Johnson, D.A., 1996, An evaporitic-source model for igneous-related
Fe-oxide(-REE-Cu-Au-U) mineralization: Geology, v. 24, p. 259-262.

ABSTRACT
We propose that many igneous-related Fe-oxide-rich (REE-Cu-Au-U-bearing) deposits
form by hydrothermal processes involving evaporitic ligand sources, either coeval salars or
older evaporites. These deposits are abundant in both Phanerozoic and Proterozoic
extensional continental and continent-margin settings. They commonly form in global arid
zones, but they also occur where magmatism is superimposed upon older evaporites.
Magmatic compositions exert only second-order control, mainly on alteration mineralogy
and element abundances. Hot S-poor brines generated by interaction with evaporitic
materials are consistent with geologic settings and help rationalize the distinctive element
enrichments (siderophile, lithophile) and hydrothermal alteration (sodic, locally alkaline)
found in these systems. This model contrasts with immiscible oxide melt and magmatic-
hydrothermal origins commonly proposed for these deposits, although all three
mechanisms can occur.

INTRODUCTION
Evaporitic environments are widespread in the geologic record, yet they have received
scant attention as ligand sources for igneous-related mineralization, despite the importance
of saline fluids in the transport of metals and the recognized importance of brines in many
sediment-hosted mineral deposits (Kyle, 1991). Rather, the clear evidence for magma-
derived brines in many igneous-related systems has dominated ideas about the genesis of
porphyry, skarn, and related ore deposits.
In this paper, we propose that highly saline fluids of surficial or connate origin, derived
from contemporaneous or older sources, are an important source for fluids in many
igneous-related Fe oxide-rich mineral deposits (cf. Haynes et al., 1995). Such deposits
number in the thousands and are distributed globally in Proterozoic and Phanerozoic rocks,
yet they are enigmatic in character and origin. The presence of variable amounts of
accessory rare earth elements, Cu, U, Au, Ag, and Co has made these rocks of
considerable economic interest for commodities besides iron. This interest has stimulated
study of them, particularly since the discovery of the enormous Cu-Au-U-rich hematitic
breccias at Olympic Dam, Australia and the inferred association of world-class deposits of
iron (Kiruna, Sweden) and REE (Bayan Obo, China) with this group (Pratt and Sims,
1990).
Both the characteristics and origins of Fe oxide-rich systems are controversial. Fe-
oxide-rich deposits occur with many types of igneous rocks in a variety of settings (Barton
et al., 1993). Previous genetic models have invoked immiscible melt, deuteric, and
synsedimentary origins (reviewed by Hitzman et al., 1992; Hauck, 1990). The debate
about origins is largely predicated on limited data sets and interpretation of local field and
textural relations. We exclude some types, such as anorthosite-related Fe-Ti-oxide
deposits, where an immiscible melt origin fits geological and experimental data (Kolker,
1982).
 p. 2

GEOLOGICAL AND GEOCHEMICAL CHARACTERISTICS

Although origins are controversial, there is broad agreement that these deposits have
abundant Fe oxides, a characteristic suite of accessory elements, at least some hydrothermal
alteration, evidence for vigorous emplacement processes, and a genetic link to magmatism
(Hitzman et al., 1992; Table 11). These systems are typified by large masses of Ti-poor Fe
oxide (commonly >108 tonnes magnetite or hematite) with lesser amounts of phosphates
(apatite, REE phosphates), Cu-Fe sulfides, and sporadic Au, U, Ag, and Co minerals.
Proximal and deeper mineralization consists of early magnetite ± apatite rocks commonly
with abundant sodic (albite ± scapolite + hornblende) wall-rock alteration. Superimposed
or distal mineralization consists of hematite ± Cu-Fe sulfides ± REE minerals with
hydrolytic ± potassic alteration. Variants include districts with substantial sodic alteration
and Cu or U mineralization but subordinate Fe oxide. Sodic alteration occurs in nearly all
districts; it is typically voluminous, and a number of examples have local peralkaline
assemblages containing NaFe+3 -bearing amphibole or pyroxene. Calcareous host rocks
contain calcic or magnesian skarn assemblages, with hydrous assemblages (actinolite-
chlorite-carbonate) distal to and superimposed upon anhydrous assemblages (pyroxene-
garnet). In form, these deposits may consist of stratabound massive Fe-oxide in volcanic
and sedimentary rocks, although most commonly they form discordant, variably brecciated
tabular to irregular masses. The former have been interpreted both as syngenetic and
replacement bodies, whereas the latter have been interpreted as veins, dikes, and breccia
fillings.
Igneous rocks are closely linked in time and space to many of these deposits, yet there
is little correlation with igneous compositions or intrusive phases (Table 1). Crosscutting
relations indicate that mineralization proceeded concurrently with magmatism. Associated
igneous rocks range across alkalic granitoids and volcanics, calc-alkaline mafic,
intermediate, and felsic suites, continental flood basalts, and rift-related basalts. This lack
of correlation with igneous compositions and the immense amount of hydrothermal
alteration in some of the mafic systems indicates the importance of nonmagmatic controls.
In contrast, in other igneous-related deposit types, such as skarn and porphyry deposits,
metals and alteration correlate strongly with igneous compositions, and voluminous sodic
alteration is rare (e.g., Einaudi et al., 1981).
Although similar overall, mafic and felsic systems differ in aspects of their wall-rock
alteration and element enrichments. Mafic rocks have abundant early scapolitic and late
chloritic and carbonate alteration. Felsic rocks typically contain abundant early albitic ±
potassic and late siliceous and sericitic alteration, but rarely have scapolitic alteration.
Compared with mafic systems, felsic systems tend to have higher REE, U, Ba, F/Cl,
hematite/magnetite, and hydrolytic/sodic alteration volumes (Table 1; Barton et al., 1993).
Enrichments in Fe, REE, Cu, Au, and U are typically 10-100 times probable source rocks,
usually more for rare metals than for Fe or REE. These variations are consistent with
imposition of wall-rock control on otherwise similar fluids.
Ages and tectonic environments vary widely. Ages range from Early Proterozoic to
Holocene (Table 1and Fig. 1). Large deposits and districts are common in the Phanerozoic,
in contrast to the suggestion that they reflect a distinctive Proterozoic geologic environment
(Hitzman et al., 1992). Many tectonic environments are inferred — within and behind
marine and terrestrial arcs, in failed continental rifts, in active rifts, and on stable platforms
associated with hot-spot volcanism (Table 1). An anorogenic or rift-related origin is widely
postulated based on the lack of extensive deformation in some of the best known systems,

1Data Repository item 96XX, Table A, Additional Characteristics of Selected Fe Oxide (-REE-Cu-Au-U) -
bearing Regions, a supplement to Table 1, is available on request from Documents Secretary, GSA, P.O.
Box 9140, Boulder, CO 80301.
 p. 3

and the occasional association with alkaline igneous rocks (Hauck, 1990). Indeed,
extensional settings of some variety are common in both Proterozoic and Phanerozoic
districts, although in many cases they may be local closed basins (e.g., volcano-tectonic
depressions, Table 1).

EVIDENCE FOR AN EVAPORITIC COMPONENT

Rather than specific magmatic or tectonic correlations, a common thread is evidence for
an evaporitic component. Evidence for an evaporitic component includes direct association
with evaporitic materials, correlation with arid belts or evaporite-bearing basins,
voluminous sodic alteration, and geochemical data that point to evaporitic rather than
magmatic fluids (Table 1). In many Phanerozoic occurrences, older or broadly coeval
evaporites can be demonstrated. Such evidence is less common in Proterozoic regions
where evaporitic materials are only rarely observed or are inferred from distinctive
metamorphic assemblages. In some regions, evaporitic sources persisted through time and
contributed to multiple generations of iron-oxide mineralization (e.g., Iran). Holocene
generation of hydrothermal iron oxides from evaporitic sources is evident in several
settings (Table 1). In the Andes, deposits have formed in volcanic centers that are emplaced
adjacent to Cenozoic saline basins. Iron oxides are forming in the Red Sea region where
hypersaline hydrothermal systems result from basalts cutting Miocene and younger salt.
Further supporting an evaporitic connection is the striking correlation of Fe-oxide(-
REE-Cu-Au) deposits with areas having arid paleoclimates. For example, during the
Mesozoic, nearly all occurrences are restricted to arid latitudes (25 ± 10˚ N or S, Fig. 1). A
notable exception is the Permo-Triassic province of the Siberian platform, which formed at
>50˚N, but in an area where the Siberian traps were emplaced through >1 km of Cambrian
salt (Zharkov, 1984; Vakhrushev and Ryabkov, 1984). Analogous paleoclimatic relations
hold for Paleozoic deposits in Laurentia and Gondwana. For example, the scapolite-rich
multibillion-tonne iron provinces of the Turgai, Uralian, and Altai-Sayan regions are
associated with red-bed ± evaporite-bearing mid-Paleozoic arcs that formed near 30˚N
(Table 1, Zonenshain et al., 1990; Zharkov, 1984).
Enormous volumes (>100 km3) of intense sodic alteration are present in many of these
districts (Table 1). This alteration requires far more saline fluid than can plausibly be
generated from associated magmas. By comparison, corresponding volumes of intense
alteration in other types of magmatic-hydrothermal systems are considerably smaller. For
example, pervasive alkali exchange in porphyry Cu systems generally is <10 km3 except
where external fluids are involved (Dilles et al., 1995). Only seafloor hydrothermal
systems, where the amount of fluid is not limited, have greater volumes of sodic alteration.
Finally, geochemical data are consistent with an evaporitic component. Fluid inclusions
and the presence of marialitic scapolite indicate salinities > 20 wt% NaCl (e.g., Vanko and
Bishop, 1982). Available isotopic data commonly indicate nonmagmatic fluid components.
In particular, sulfur isotope ratios indicate that isotopically heavy sulfur was important in
the formation of sulfides in a number of deposits (δ34S > 5‰, Table 1). Early magnesian
silicate + magnetite to late hematite ± pyrite ± chalcopyrite assemblages indicate oxidized,
relatively sulfur-poor environments (Barton et al., 1993) as would be expected from an
evaporitic fluid source (below). This external source of volatile constituents contrasts with
magmatic sources indicated for transition metals and REE from whole-rock and radiogenic
isotope data (Barton et al., 1991; Johnson and Cross, 1991).

AN EVAPORITIC-SOURCE MODEL
The evaporitic model invokes thermal circulation of highly saline fluids originating
from coeval or older sources (Fig. 2). Magmatism provides the necessary heat to drive
circulation and generate the required high temperatures. Igneous rocks also serve as the
source and hosts for mineralization. Evaporitic sources provide the chloride necessary for
 p. 4

metal transport and the sodium observed in hydrothermal alteration, and they drive fluids
toward oxidized, relatively S-poor compositions. NaCl-CaSO4-bearing sources will be
saturated with halite and gypsum at near-surface temperatures (ca. 30 wt% NaCl, 0.5 wt%
CaSO4; Hardie et al., 1982). Reaction of these fluids with igneous and sedimentary rocks
along a circulation path would produce alteration and mineralization similar to observed
patterns (Fig. 2). These patterns will resemble those created by other igneous-related
hydrothermal systems, but they have some distinctive differences.
On the downwelling limb, sodic alteration follows from the high Na contents of low-
temperature evaporitic fluids. Voluminous sodic alteration forms independently of other
factors throughout the magmatic history; such alteration will substantially exceed the
amount of potassic alteration that can form. Alteration volumes are limited by the amount of
fluid that can be circulated (heat balance?), rather than by the S and Cl contents of causative
intrusions, as is the case in many porphyry-type systems. If temperatures are high,
scapolite-hornblende alteration will dominate, as it does in many intermediate and mafic
magmatic systems; otherwise, albite-actinolite-chlorite assemblages will form, as in most
felsic systems. Peralkaline assemblages occur in a number of Fe-oxide occurrences (Table
1); they are hard to rationalize by ordinary igneous or magmatic-hydrothermal mechanisms,
but the extremely high Na+ /H+ of an evaporitic source could drive reactions of the type
2NaClaq + Fe2O3 + 2SiO2 + H 2O = 2NaFeSi2O6 + 2HClaq. Coincident with sodic
alteration, the fluid will gain components in which it is undersaturated (Fe, K, Ca, reduced
S, REE, SiO 2), as is suggested by mass balance data on Basin and Range occurrences
(Barton et al., 1991; Dilles et al., 1995).
On the upwelling limb, cooling, wall-rock reaction, and mixing with other fluids create
distinct products depending on rock type and path (Fig. 2). Sodic alteration persists where
fluid to rock ratios are high. Hydrolytic alteration (producing K-mica + quartz in felsic
rocks, chlorite in mafic rocks) forms in response to cooling and precipitation of metals
from chloride complexes. In rocks with moderate K2O contents, potassic alteration
(secondary biotite ± K-feldspar) can form by alkali exchange (Hitzman et al., 1992). The
original sulfate-bearing fluids may be modified by partial reduction and incorporation of
sulfides, but they generally have too little sulfur to precipitate sulfide minerals at higher
temperatures. Early precipitation of magnetite-apatite is superseded by hematite ± Cu-Fe
sulfides ± other REE minerals as H-O(-Cl-S) fluids become more oxidizing and sulfidizing
with decreasing temperature (Barton et al., 1993).
Evaporitic brines have exceptionally high Cl/S; thus, metals greatly exceed the amount
of sulfur needed to precipitate them (Fig. 3). Consequently, only the least soluble
chalcophile elements (e.g., Cu) form sulfides along with the other elements that form
oxides or oxysalts (Fe, REE, U) or native elements (Au) — the suite seen in these systems.
Lacking adequate S, Pb and Zn will be removed from the system, and Fe sulfides will be
sparse. In some circumstances, reduction of these fluids might precipitate distal Ag-Co-As-
U as is observed in some systems (Hildebrand, 1986). In contrast, other ore-forming
fluids (e.g., magmatic or marine waters) have lower Cl/S leading to more abundant sulfide
precipitation, lower oxide/sulfide ratios, and commonly significant distal Zn ± Pb sulfides
(Fig. 3).

SUMMARY AND CONCLUSIONS

The involvement of fluids with evaporitic components in magmatic environments
rationalizes many features of a large group of enigmatic Fe oxide-rich mineral deposits.
Incorporation of such fluids in intrusion-driven hydrothermal systems should produce
distinctive metasomatism dominated by sodic alteration and oxide-rich ore mineral
assemblages. Hot Cl-rich, S-poor fluids can mobilize many elements while precipitating
few as sulfides. Such hydrothermal systems could form in many tectonic settings where
magmas are emplaced with contemporaneous or older evaporitic sources; thus, they will
 p. 5

primarily reflect paleoclimatic control. Igneous rocks will have only a second-order effect
on mineralization, influencing relative enrichments and alteration minerals. These
predictions are consistent with Fe oxide-dominated igneous-related hydrothermal systems.
These systems share enrichment in REE, and commonly Cu, Au, and U. Voluminous
sodic alteration, locally peralkaline, occurs in most areas along with other evidence for
evaporitic fluids. They exhibit a paleogeographic control associated with evaporitic
materials, be it in convergent, divergent, transform or intraplate settings.
Alternative models are distinguishable from the evaporitic model. Skarn and porphyry
systems associated with dioritic and alkaline intrusions have similar element enrichments
(Fe-oxide, Cu-Au), but lack the huge volumes of intense alteration and exhibit no
paleogeographic influence. Rather, they have a strong spatial and temporal correlation with
intrusive centers of restricted composition. Likewise, apatite-bearing Fe-Ti oxide-rich
bodies with accessory REE clearly form in mafic, commonly alkaline magmatic systems,
consistent with experiment (Kolker, 1982). They lack hydrothermal alteration and the
lower-temperature mineral assemblages, whereas Ti-poor systems lack clear evidence of
magmatic temperatures.

ACKNOWLEDGMENTS
Supported by National Science Foundation grant EAR91-17372. We thank M.
Marikos, J. Ruiz, V. Wall, and E. Force for discussions, and R. Bodnar, D. Crowe, M.
Einaudi, E. Force, F. Mazdab and S. Titley for helpful reviews.

REFERENCES CITED
Barton, M. D., Hassanzadeh, J., Battles, D. A., and Marikos, M. A., 1991, Magnetite-
apatite-REE mineralization in the Great Basin: Geological Society of America Abstracts
with Programs, v. 23, no. 5, p. 292.
Barton, M. D., Marikos, M. A., and Johnson, D. A., 1993, A comparison of felsic and
mafic Fe-P(-REE-Cu) deposits: Geological Society of America Abstracts with Programs,
v. 25, no. 3, p. 5.
Battles, D. A., and Barton, M. D., 1995, Arc-related sodic hydrothermal alteration in the
western United States: Geology, v. 23, p. 913-916.
Bonatti, E., Fisher, D. E., Joensuu, O., Rydell, H. S., and Beyth, M., 1972, Iron-
manganese-barium deposit from the northern Afar Rift: Economic Geology, v. 67, p.
717-730.
Denham, D. M., and Scotese, C. R. 1988, Terra Mobilus, Earth in Motion Technologies:
Austin, Texas.
Dilles, J. H., Farmer, G. L., and Field, C. W., 1995, Sodium-calcium alteration by non-
magmatic saline fluids in porphyry copper deposits: Results from Yerington, Nevada:
Mineralogical Association of Canada Short Course Series, v. 23, p. 309-338.
Einaudi, M. T., Meinert, L. D., and Newberry, R. J., 1981, Skarn Deposits: Economic
Geology, 75th Anniversary Volume, p. 317-391.
Foose, M. P., and McClelland, J. M., 1995, Proterozoic low-Ti iron oxide deposits in
New York and New Jersey: Relation to Fe-oxide (Cu-U-Au-rare earth element) deposits
and tectonic implications: Geology, v. 23, p. 665-668.
Förster, H., and Jafarzadeh, A., 1994, The Bafq Mining District in Central Iran - a highly
mineralized Infracambrian volcanic field: Economic Geology, v. 89, p. 1697-1721.
Gow, P. A., Wall, V. J., Oliver, N. H. S., and Valenta, R. K., 1994, Proterozoic iron
oxide (Cu-U-Au-REE) deposits: Further evidence of hydrothermal origins: Geology, v.
22, p. 633-636.
Haynes, D. W., Cross, K. C., Bills, R. T., and Reed, M. H., 1995, Olympic Dam ore
genesis: a fluid-mixing model: Economic Geology, v. 90, p. 281-307
 p. 6

Hardie, C. E., Eugster, H. P., and Weare, J. M., 1982, Mineral equilibria in the six-
component seawater system, Na-K-Mg-Ca-SO4-Cl-HO at 250˚C, II: Components of the
saturated solutions: Geochimica et Cosmochimica Acta, v. 46, p. 1603-1618.
Hauck, S. A., 1990, Petrogenesis and tectonic setting of Middle Proterozoic iron oxide-
rich ore deposits — an ore deposit model for Olympic Dam-type mineralization: U.S.
Geological Survey Bulletin 1932, p. 4-39.
Hildebrand, R. S., 1986, Kiruna-type deposits: Their origin and relationship to
intermediate subvolcanic plutons in the Great Bear Magmatic Zone, Northwest Canada:
Economic Geology, v. 81, p. 640-659.
Hitzman, M. W., Oreskes, N., and Einaudi, M. T., 1992, Geological characteristics and
tectonic setting of Proterozoic iron oxide (Cu-U-Au-REE) deposits: Precambrian
Research, v. 58, p. 241-287.
Johnson, J. P., and Cross, K. C., 1991, Geochronological and Sm-Nd isotopic
constraints on the genesis of the Olympic Dam Cu-U-Au-Ag deposit, South Australia, in
Pagel and Leroy, eds., Source, Transport, and Deposition of Metals, p. 395-400.
Kolker, A., 1982, Mineralogy and geochemistry of Fe-Ti oxide and apatite (nelsonite)
deposits and evaluation of the liquid immiscibility hypothesis: Economic Geology, v. 77,
p. 1146-1158.
Kyle, J. R., 1991, Evaporites, evaporitic processes and mineral resources, in Melvin, J.
L., ed., Evaporites, petroleum and mineral resources: Amsterdam, Elsevier, p. 477-533.
Lyons, J. I., 1988, Volcanogenic iron oxide deposits, Cerro de Mercado and vicinity,
Durango, Mexico: Economic Geology, v. 83, p. 1886-1906.
Ohmoto, H., Mizukami, M., Drummond, S. E., Eldridge, C. S., Pisutha-Arnond, V., and
Lenagh, T. C., 1983, Chemical processes of Kuroko formation: Economic Geology
Monograph 5, p. 570-604.
Pratt, W. P., and Sims, P. K., 1990, The midcontinent of the United States — Permissive
terrane for an Olympic Dam type deposit: U.S. Geological Survey Bulletin 1932, 81 p.
Rose, A. W., Herrick, D. C., and Deines, P., 1985, An oxygen and sulfur isotope study
of skarn-type magnetite deposits of the Cornwall type, southeastern Pennsylvania:
Economic Geology, v. 80, p. 418-443.
Smirnov, V. I., 1977, Ore deposits of the U.S.S.R., Volumes I and III: London, Pitman
Press, 491 p.
Vakhrushev, V. A., and Ryabkov, V. G., 1984, The geologic position and distinctive
mineralogical and geochemical features of the ore deposits in the Angara-Tunguska iron-
ore province: Soviet Geology and Geophysics, v. 25, p. 60-65.
Vakrushev, V. A., Ripp, G. S., and Kaviladze, M.Sh., 1981, Isotope composition of
sulfur from iron ore deposits of east Siberia: Soviet Geology and Geophysics, v. 22, no.
1, p. 59-64.
Vanko, D. A. and Bishop, F. C., 1982, Occurrence and origin of marialitic scapolite in the
Humboldt lopolith, N.W. Nevada: Contributions to Mineralogy and Petrology, v. 81, p.
277-289.
Vidal, C. E., Injoque, E. J., Sidder, G. B., and Mukasa, S. B., 1990, Amphibolitic Cu-
Fe skarn deposits in the central coast of Peru: Economic Geology, v. 85, p. 1447-1461.
Vivallo, W., Henriquez, F., and Espinoza, S., 1994, Oxygen and sulfur isotopes in
hydrothermally altered rocks and gypsum deposits at El Laco Mining District, Northern
Chile: Comunicaciones, no. 45, p. 93-100.
Zharkov, M. A., 1984, Paleozoic salt bearing formations of the world: Berlin, Springer-
Verlag, 427 p.
Zierenberg, R. A., and Shanks, W. C. I., 1983, Mineralogy and geochemistry of
epigenetic features in metalliferous sediment, Atlantis II Deep, Red Sea: Economic
Geology, v. 78, p. 57-72.
 p. 7

Zonenshain, L. P., Kuzmin, M. I., and Natapov, L. M., 1990, Geology of the U.S.S.R.:
A plate tectonic synthesis: American Geophysical Union, Geodynamics Series, v. 21,
242 p.
 TABLE 1. SELECTED CHARACTERISTICS OF SOME FE-OXIDE(-REE-Cu-Au-U)-BEARING REGIONS

Region and District Age and setting Size* and metals, sodic alteration Comments References

Red Sea region Holocene - rift basaltic >100? (>15?) FeOx + MnOx ± Zn-Cu- Modern geothermal circulation Zierenberg & Shanks
(Afar, Atlantis II) volcanism, abundant evaps Pb sulfides seds; local peralkine through evaporitic beds (1983), Bonatti et al.
(1972)
Chilean Andes (El Laco, Neogene - intermediate >1000? (>500) FeOx in 2 districts; Arid; volcanos adjacent to Vivallo et al. (1994),
Magnetita Pedernales) volcanic centers in closed minor sodic ± peralk salars containing halite Stoertz & Ericksen (1974)
basins
Mexican Altiplano (Cerro de mid-Tertiary - continental arc >300 (>100) FeOx ± REE, Cu in >20 Arid paleoclimate, sulfates in Lyons (1988), J. Ruiz,
Mercado, La Perla) felsic volcanic centers locs; modest sodic ± peralk volcanic basins, Jurassic salt 1994, personal commun.
Basin & Range, USA (Cortez Jurassic - back-arc mafic and >1000 (>75) FeOx ± REE , Cu; >40 Arid paleoclimate, evaps in Battles & Barton (1995),
Mtns., Humboldt complex) felsic volcanic complexes locs; exten. sodic ± peralk region; heavy S, saline fluids Barton et al. (1991)
Andean coastal belt, Peru & Jurassic-Cretaceous - >3000 (350) FeOx ± major Cu(Au), Arid paleoclimate, Jurassic Vidal et al. (1990)
Chile (El Romeral, Raul, extensional arc intermediate- minor Co, REE; >150 locs; evaporites; variable S, saline
Candelaria, Monterrosas) mafic volcano-plutonic extensive sodic ± peralk(?) fluids, VMS in same arc
centers
Anhui & Hankow regions, Jurassic - extensional back-arc >1000 (>300) FeOx ± Cu; >25 Red beds & Triassic evaps; Zhang (1986), Song et al.
east-central China intermediate mafic-volcanics occurrence; exten. sodic & skarn evidence for evap/SW source (1981)
Mid-Atlantic, USA Triassic-Jurassic - mafic sills >350 (>100) FeOx ± Cu, Co (± U); >50 Redbeds ± evaps; Rose et al. (1985),
(Cornwall, Grace) in rift basins locs; skarn & sodic geochemical evidence for Robinson (1988)
evap source
Siberian Platform, Russia Permo-Triassic - mafic >3000 (>650) FeOx + minor Cu, Co in Dolerites interact with Smirnov (1977), Vakrushev
(Korshunovsk, Tagar) intrusions with flood basalts >150 locs; skarn and sodic Cambrian salt, heavy S, et al. (1981), Zharkov
saline fluids (1984)
Turgai, Kazakhstan (Sarbai, Carboniferous - intermediate- >4000 (>1500) FeOx + minor Cu, Co; Arid paleoclimate, possible Smirnov (1977), Zharkov
Kachar) mafic arc volcanic centers extensive sodic & skarn evaps (1984)
Altai-Sayan Mtns., central mid-Paleozoic - mafic-felsic >2000 (>500) FeOx ± REE, Cu, U; Arid paleoclimate; evaps, Smirnov (1977), Vakrushev
Asia (Tuva, Abakan, volcano-plutonic complexes >15 locs; extensive sodic alt heavy S in stratabound et al. (1981), Zharkov
Anzas) deposits (1984)
Central Iran (Bafq, Gole Cambrian & younger - >2500 (>1000) FeOx ± Cu, REE; >25 Extensive Cambrian & younger Förster & Jafarzadeh (1994)
Gohar, Hamadan) anorogenic felsic volcanic locs; sodic ± peralk evaps; distal MnOx & Pb+Zn
centers
NW Canada (Great Bear, Proterozoic - intermediate arc >50? (>10?) FeOx ± Cu, U, Ag, Co; No evaps reported; zoned from Hildebrand (1986), Hitzman
Wernecke Mtns.) volcano-plutonic centers >25 locs; regional sodic alt FeOx to distal Ag-Co-U et al. (1992)
South Australia (Acropolis, mid-Proterozoic - anorogenic >3000 (>2000) FeOx + major Cu, High salinities, but no Reeve et al. (1990), Gow et
Olympic Dam, Emmie Bluff) felsic-mafic(?) suite REE, U, Au; >5 deposits described Proterozoic evaps al. (1994)
Northern Sweden (Kiruna, mid-Proterozoic - mafic-felsic >5000 (>2000) FeOx ± Cu, REE, Au, Evidence for rifting with coeval Hitzman et al. (1992),
Gallivare, Ekstromberg) volcanic suite, extensional(?) U; >20 deposits; exten. sodic ± evap source Hauck (1990)
peralk
Bayan Obo, Inner Mongolia, Mid-Proterozoic(?) - felsic >1000 FeOx + major REE, stratiform; Overprinted(?) by Paleozoic Chao et al. (1992), Hauck
China sediments & dolomites exten. sodic + peralk alt granitoids; controversial (1990)
Northeastern USA (Benson Mid-Proterozoic(?) - felsic- >1000 (200) FeOx + REE ± Cu, U; Evaps present; some interpret Buddington (1966),
mines, Mineville, Dover) intermdiate anorogenic suite >30 locs; exten. sodic ± peralk & FeOx to be related to Grenville Whitney & Olmsted (1988)
skarn
Bihar-Orissa area, India Mid-Proterozoic - mafic- >? (?) FeOx ± zoned REE, Cu, U; Strongly metamorphosed and Banerji (1981)
(Bihar, Singhbhum) intermediate igneous complex >15 locs; exten. sodic + peralk deformed; good zoning
SE Missouri, USA (Pilot Mid-Proterozoic - anorogenic >1000 (>300) FeOx ± Cu, REE, U, Au; No published evidence for Pratt & Sims (1990)
Knob, Pea Ridge, Iron Mtn.) felsic province >25 locs; minor sodic & skarn evaps; regional MnOx
*In millions of tonnes; first number is total district resource, numbers in parantheses are tonnages for largest deposit within the district
 p. 9

FIGURES

Figure 1. Distribution of some Fe-oxide(-REE-Cu-Au-U) districts. See Table 1. Note

correlation of most Mesozoic deposits with arid belts.
 p. 10

Figure 2. Schematic model of igneous-driven circulation of evaporitic fluids showing

alteration zoning in mafic and felsic systems. Based on occurrences in Table 1.

Figure 3. Distinction among deposit types by aqueous chlorine-sulfur contents. See text
for discussion.