Economic and Exploration Significance

of Red Sea Metalliferous Brine Deposits

W. C. Shanks, Ill*
Department of Geology and Geophysics,
University of Wisconsin,
Madison, WI

ABSTRACT bathymetry and brine-inte$ace reflectors using

Seventeen deeps, which contain hot or cold a high-resolution narrow-beam echo sounder
has been the most ejJective exploration tool in
brines andlor nletalliferous sediments, have been
discovered in the axial rift zone of the Red Sea. the Red Sea. Followup coring has discovered
Metalliferous activity results from the unique additional metalliferous deposits in basins with-
geological setting of this incipient ocean basin: out present-day brines, and potential exists for
oceanic crust intruding in a central rift zone additional discoveries. Geochemical explora-
flanked by thick evaporite sequences. Brine pools tion using heavy metals values in sediments (iron,
tend to occur in structural depressions where- manganese, zinc, copper, and mercury) and sul-
fur isotope values of sulfides are considered to
transform faults intersect the spreading axis.
The most significant metalliferous occurrence be the most promising techniquesfor discovery
is in the Atlantis II Deep, which contains ap- of currently inactive deposits. Positive sulfur
proximately 227 Mt (dry weight) of metallijer- isotope values in the range I to 17 are an in-
dication of hydrothermal sulfides in the Red Sea,
ous sediment which averages 1.04% Zn and
and there is considerable contrast from back-
0.28% Cu. Higher potentially economic grades
(3-6% Zn, 1% Cu, 50 glt Ag) occur in the ground bacreriogenic sulfides ( - 10 to - 45).
Southwest Basin of the Atlantis II Deep, com-
prising total metal tonnages of 1.7 Mt Zn, 0.4
Mt Cu, and 5000 t Ag. The Pre-Pilot Mining INTRODUCTION
Test conducted in 1979 raised 15 000 m3 of mud Metalliferous sediments andlor saline brines
by a slurry process and demonstrated the fea- have now been discovered in numerous seafloor
sibility of mining, concentration by froth flo- deeps along 900 km of the axial rift zone of the
tation, and tailings disposal at sea. A Pilot Mining Red Sea (Fig. 1). The Atlantis I1 Deep is the
Test is scheduled for 1983-1 984. most interesting of the bnne deeps both in terms
The metalliferous discoveries in the Red Sea of intensity of submarine hydrothermal pro-
provide a model example of deep marine min- cesses and in terms of potential economic im-
eral exploration. Detailed mapping of bottom portance. This deposit is unconventional mainly
by virtue of occurrence in an inaccessible phys-
* Present address. US Geolog~calSurvey, Reston, VA ical setting. However, successful completion of
 UNCONVENTIONAL MINERAL DEPOSITS

-
B r ~ n ePool w ~ t hM e t a l l ~ f e r o u sSediments
A Metall~ferousSed~ments(no brine at present)

-
----
Transform Faults from bathymetr~cdata
Continental F r a c t u r e Lines
L a t e r a l extens~onof on-shore faults

FIG. 1. Location map of

Red Sea Deeps w ~ t hbrines
and/or metalliferous sedl-
ments (after Blgnell, 1975).
Axial nft zone defined by
the 1000-m depth contour.
Transform faults and con-
tinental fracture line loca-
tions from Bignell (1975)
with fault identification (A
to 0 ) according to Garson
and Krs (1 978).

a prepilot mining test in 1979 indicates that the vide significant insight into exploration for sea-
Atlantis 11 Deep may become the first deep-sea floor b~e-relatedore deposits and for analogous
base-metal sulfide mine (Mustafa and Amann, ancient deposits. In this paper I will concentrate
1980). Under the most optimistic scenario, the on the economic potential of Red Sea deposits
Atlantis LI mine might be established four to six and seafloor exploration for brine related de-
years after completion of a year-long pilot min- posits, but some ramifications of relevance to
ing test in 1983-1984. If established, mining ancient deposits are extant.
would produce about 100,000 tpd of wet mud
with about 200 tpd contained zinc, and propor-
tional amounts of comer and silver. Exwcted GENERAL STATEMENT
mine life would be about 15-20 years (Anon., The Red Sea is an embryonic ocean basin
1981). which has formed by a two-stage rifting process
In addition to direct economic importance, the (Girdler and Styles, 1974). The initial nfting
Red Sea brines and metalliferous deposits pro- stage occurred in Late Eocene to Oligocene.
 RED SEA METALLIFEROUS BRINE DEPOSITS 159

Much of the Miocene was characterized by ac- simple derivation from connate brines of evap-
cumulation of thick sequences of evaporites re- orative origin.
lated to subsidence along bounding normal faults. Subsurface circulation and the evaporite dis-
Renewed rifting began about 5 milllon years solution processes, which convert Red Sea waters
(m.y.) ago and continues today. This sequence to brine, are not precisely understood. However,
of events, believed to be fairly typical of con- Bignell(1975) has proposed that northeast strik-
tinental rifting in arid regions, has produced the ing transform faults (Fig. l), which cut obliquely
unique geologic setting which is of pivotal im- across the rift zone and through flanking eva-
portance to the formation of submarine brines porite~,may provide pathways for fluid circu-
and associated deposits. This setting includes lation. Manheim (1974) has proposed that
thick subsurface evaporite sequences (>4 km) seawater penetrates along the flanks of the Red
of mainly halite and anhydrite in the southern Sea and circulates in Middle Miocene clasticd
Red Sea. These flank the narrow axial rift zone beneath the capping, impermeable evaporite se-
(10-30 km, Fig. 1) which is a result of the quence. Dissolution of basal evaporites then
present stage of spreading. Upper Miocene eva- produces dense brine, thus inducing a hydraulic
porite~are overlain by a relatively thin veneer gradient toward the rift zone.
of Pliocene to Recent marine marls. Brine buoyancy due to heating in the rift zone
Oceanic basalts outcrop throughout the axial has not been specifically considered, but is clearly
rift zone, and transform and graben faulting pro- important as shown by the detailed hydrography
duce numerous structural depressions su~table of Atlantis I1 brines (Schoell, 1976). In partic-
for brine pools. In particular, significant brine ular, Schoell (1976) has shown that new brine
and metalliferous activity occur where major vents through the metalliferous sediments of the
transverse structures intersect the axis of the Southwest Basin (Fig. 2), rises buoyantly to the
main Red Sea trough (Fig. 1). Moreover, Gar- density interface between the lower (62°C) and
son and Krs (1976) have suggested that the N60°E upper (50°C) brine layers, and then spreads lat-
transform structures in the Red Sea developed erally to other portions of the Deep.
along Precambrian fracture lines which cut both
Nubian and Arabian shield rocks. However, ECONOMIC POTENTIAL OF THE
transforms and associated continental fractures
are not continuous across the Red Sea (Fig. 1, ATLANTIS II DEPOSIT
Garson and Krs, 1976) due to the left lateral The Atlantis I1 Deep is a local tectonic depres-
movement of 75-80 km during opening of the sion withln the actively spreading Red Sea rift,
Gulf of Aden in Late Cretaceous to Eocene. and is filled with stratified hot saline brine to a
Seafloor brine pools occur in Kebrit, Nereus, minimum depth of about 2000 m (Fig. 2). Me-
Valdiv~a,Albatross, Shagara, and Suakin Deeps talliferous sediments have been precip~tatedin
(Fig. 1) and in the hot brine area which includes this Deep over the last 25,000 yr, with intense
Atlantis 11, Chain and Discovery Deeps (Fig. and continuous hydrothermal activity for the last
2). Many of the brines (Table 1) reach salinities 10,000 yr, forming a complexly layered deposit
similar to the Atlantis I1 brine, about 25% total with a thickness of 5-20 m over an area of 65
dissolved salts. but chemical compositions are km2. Lithostratigraphic relations are systematic
quite variable (Backer and Schoell, 1972). The over much of the Deep (Fig. 3. Backer and
Atlantis I1 brine has uniquely high heavy metal Richter, 1973), but are disrupted and chaotic in
contents (iron, manganese, copper, lead. zinc), the Southwest Basin where new brine currently
is depleted in magnesium and SO,, and has the vents through earlier deposited metalliferous
highest in situ temperature (62-65"C), (Luck muds. The presently depositing AM zone, the
and Nawab, 1981). SU zones, and all the zones in Southwest Basin
Brine formation is clearly related to the flank- (Fig. 3) are of special economic importance.
ing evaporite sequences. Oxygen and hydrogen Zinc and copper occur as sphalerite and chal-
isotope studies (Craig, 1969; Schoell and Faber, copyrite, which are intimately intermixed or finely
1978) indicate that the brines are a result of interlayered with pyrite, iron oxides, iron-rich
evaporite leaching by circulated Red Sea paleo- clay minerals, and anhydrite.
waters. Invariably high ClIBr ratios in Red Sea Preussag AG, Hannover, Germany, has played
brines (Table 1, Schoell et al., 1974) preclude the leading role in geological, geochemical, and
 UNCONVENTIONAL MINERAL DEPOSITS

FIG. 2. Bathymetric map

for Atlantis I1 Deep area
with core locations and sites
(1-4) of Pre-Pilot Mining
Test. Shaded areas are b m e
filled. Bathymetry from
BEker and Richter (1973)

economic evaluation of the Atlantis I1 deposit Basin of the Deep, and mining operations would
over the last ten years. Since 1976 Preussag has be expected to concentrate on that area. Average
been the main contractor of the Saudi-Sudanese grades (dry salt-free basis) of mined material
Red Sea Joint Commission. Approximately 650 would be approximately 3-6% Zn, 1% Cu, and
core samples, including some long cores (>8 50 g/t Ag (Anon., 1981).
m) and some large volume cores, have been Preussag and various subcontractors mined and
taken from the Atlantis I1 Deep. Recently pub- concentrated approximately 15 000 m3 of mud
lished estimates of total metal contents in the from four mine sites in the Atlantis I1 Deep (Fig.
muds are: 1.7 Mt Zn, 0.4 Mt Cu, and 5000 t 2) over a three-month period in 1979 (Luck and
Ag (Anon., 1981). An early estimate of overall Nawab, 1981). This prepilot mining test (Table
average grade for dry, salt-free sediment is 1.04% 2) demonstrated effective mining at 2200-111depth
Zn and 0.28% Cu (Backer, 1975). However, using a conventional 5-in. drill string and a suc-
grades are significantly higher in the Southwest tion head with conical vibrating screen (Fig. 4),
 RED SEA METALLIFEROUS BRINE DEPOSITS

Table 1. Chemical Composition of Red Sea Brines (g/Kg)

T, "C Na K Mg Ca so4 C1 Ref.
Suak~n 24 6 50 6 I I 14 22 32 84 8 1
Port Sudan 36 2 78 1.5 15 11 40 125 2
Erba 27 9 50 I .O 15 10 4.1 86 5 2
Shagara - 62 20 25 48 12 113 2
Albatross 24 4 86 20 10 40 11 143 2
Vald~vla 29 7 95 19 19 0.8 68 140 2
Thet~s 22 6 II 05 13 0.9 34 23 2
Nereus 30 2 - - 15 79 09 130 2
Kebnt 23 3 95 15 24 7 1 22 153 2
Atlant~sI1 61.5 93 19 08 52 08 156 3
Red Sea Water 22.2 12 0.4 14 05 30 21 4
References ( I ) .
Baumann. et al 1973
(2) Backer and Schoell. 1972
(3) Brewer and Spencer. 1969
(4) Crag, 1969

concentration at sea by froth flotation, and suc- Concentration by froth flotation was carried
cessful disposal of tailings at 400-m depth (Mus- out in the mud mixing tanks of the SEDCO 445.
tafa and Amann, 1980). Minimal environmental The flotation process was fairly standard and
impact or contamination of the mine site oc- similar to that used for bulk flotation of complex
curred dunng t h ~ disposal
s test. Further tests are sulfide ores (Mustafa and Amann, 1980; Anon.,
planned, but monitoring of the disposal plume 1981), except that seawater was used to dilute
using the Valdivia indicated that the discharge the slurry to about 30 g/L. Concentrates were
penetrates t o 1200-m depth within 20 min, with produced continuously with 30-35% Zn, 3.5-
high solid concentrations at about 800 m. Solids 4.5% Cu, 500-700 g/t Ag, and a few ppm Au
are dispersed to only twice normal Red Sea lev- at 60-70% recovery rates (Table 2, Weber et
els within a few days (Anon., 1981). al., 1981).
The prepilot test was carried out using the A number of possibilities are currently being
dnllship SEDCO 445, wlth suction head loca- tested for metallurgical processing of the con-
tion and movement controlled to within l m by centrates (Weber et al., 1981). Pyrometallurg-
dynamic positioning of the ship relative to two ical processes have been eliminated from
bottom-tethered acoustic beacons. A submers- consideration, largely because of the 7-12% salts
ible radial pump located on the drill string just in the filter cake. The most promising methods.
above the brine interface produced 100 m3/hr being tested in four separate laboratories and
of slurry with solid concentration of 60- 100 gl differing in detail, involve chloride leaching or
L. Vertical positioning of the head was con- oxygen pressure leaching in the presence of
trolled by an acoustic sediment thickness meter chloride followed by solvent extraction and
and the pump location relative to the suction electrowinning. Recoveries in excess of 99% for
head allowed a maximum penetration of 18 m zinc and 95% for copper and silver have been
below the mud-bnne interface. demonstrated (Anon., 1981).

Table 2. Results of the 1979 Pre-Pilot Mining Test

Zn content, %
Quantity
Site Time, hr mined, m3 Heads Tails Concentrate
 RED SEA METALLIFEROUS BRINE DEPOSITS

FIG. 4. Schematic draw-

ing of the metalllferous
mud mining system. After
Mustafa and Amann
(1980).
A

Suction head
positioning beacons

It is interesting to note that the "unconven- Finally, deep-sea mining requires no expensive
tionality" of the Atlantis I1 deposit significantly mine development and demonstrably can be car-
increases the economic potential of this occur- ried on continuously even under heavy sea con-
rence. If the deposit was lithified and emplaced ditions. A factor which favors the prospect is
within a continental land mass, it would be a the Saudi-Sudanese agreement (1974) to hold
sheetlike body with dimensions of approxi- joint equal and exclusive sovereign rights to eco-
mately 6 to 14 km. Expulsion of the 50-90% nomic development of the so-called common
interstitial brine would leave a bulk deposit zone of the central Red Sea below water depth
thickness of less than 5 m. Mining, crushing, of 1000 m.
and processing of a lithified equivalent of the Future prospects for mining the Atlantis I1
Atlantis I1 deposit would be economic only if deposit will depend critically on the results of
metamorphism and deformation produced con- additional metallurgical testing, the year-long
siderable thickening along fold hinges. How- pilot mining test scheduled for 1983-84. and
ever, it is also germane to note that the bulk of on base and precious metal prices. However,
Atlantis deposit has developed over only 10-12 present results are moderately encouraging.
thousands years whereas many ancient massive
sulfides required millions of years to develop.
Potential advantages of marine mining of the EXPLORATION SIGNIFICANCE
Atlantis 11 deposit include abundant inexpensive Because metalliferous sediments have never
transport and supply by sea, and processing of been commercially mined from the seafloor,
the unconsolidated muds without crushing and economic exploration for such deposits outside
grinding. Tailings disposal can apparently be of the Red Sea has not been systematically at-
canied out on site without deletenous effects. tempted. The attractiveness of the Atlantis 11
 UNCONVENTIONAL MINERAL DEPOSITS

deposit hinges on its unconsolidated nature and proposed. (Rona, 1978; Cronan. 1980), but in
occurrence on the surface of the seafloor. Buried practice detalled bathymetric profiling has been
Red Sea type deposits would be uneconomic ~f most effective in delineating favorable deeps
removal of significant overburden were neces- and in detecting above-bottom reflectors related
sitated. Thus, seafloor exploration targets are to suspended matter at brine-seawater interfaces
limited to actively forming or recently formed (Backer, 1975). In addition, high-resolution
metall~ferousmuds. narrow-beam echo sounding, combined with si-
The unique geological setting of the Red Sea multaneous gravity and magnetic profiling. has
is of critical importance. The Atlantis I1 deposit provided a wealth of detailed geological and
contains abundant base metal sulfides because tectonic information on the Red Sea rift zone.
of the ability of hot chloride-rich brines to trans- Follow-up sediment sampling in brine-con-
port heavy metals (Shanks and Bischoff, 1977). taining deeps and in basins without bnne has
The depos~tis an unconsol~datedmix of sul- led to the discovery of additional recently formed
fides, sulfates, silicates, and oxides because dense metalliferous deposits (Fig. 1; Backer and
brines pool tn bathymetric depressions. T h ~ an-s Schoell, 1972) and to an improved understand-
oxic hot brine pool provides a quiescent, sterile ing of the distribution of metalliferous phases
environment In which finely layered metallifer- in Recent Red Sea sediments. Recent detrital
ous precipitates settle out and are preserved from sedimentation in the Red Sea is dominated by
bacterial degradation, oxidation. and bioturba- biogenic carbonates (mainly pteropods and for-
tion. Therefore, the fonnat~onof an Atlantis I1 aminifera). Lesser amounts of detrital quartz.
type deposit requires evaporites to produce brine feldspars, and clays also occur. and metallifer-
and igneous activity to heat the brine for effec- ous components are admixed or interlayered in
tive metal transport. In addition, a subsurface some areas.
circulation pathway which ends at the seafloor The intensity of metallization of sediments in
is required. Red Sea deeps is nicely illustrated by bulk
Global exploration for seafloor brine-related chemical proportions of detrital (aluminum, cal-
base metal sulfide deposits can thus be targeted cium, magnesium), oxide (manganese, iron),
to general geological situations like the Red Sea. and sulfide (zinc, copper) components (Fig. 5,
At present the Red Sea is unique. The Gulf of Schoell et a]., 1974). The sediments in many
California is tectonically similar, but significant of the Red Sea deeps are dominated by detntal
evaporites have not formed there. The Atlantlc materials with thin layers enriched in metalll-
Ocean basin was probably very similar to the ferous components. However, significant me-
Red Sea during initial opening. Evaporites did talliferous sediments occur in Nereus, Thetis,
form extens~vely,but any sulfide deposits which Gypsum, and Suakin Deeps and in areas flank-
formed would now be deeply buned beneath ing the Atlant~sI1 Deep (Discovery, Cham,
thick clastic wedges related to the inactive basin Wando, and Atlantis Terrace). The Atlantis I1
margins. In fact, metalliferous sediments con- Deep contains the only sediments sampled thus
taining iron and manganese oxides and some far which are dominated by the sulfide com-
base metal sulfides have been drilled off eastern ponent and which attain average grades and ton-
North America in Cretaceous sedimentary rocks nages of potential economic levels (Fig. 5).
by DSDP hole 105 (Hollister et a]., 1972). An- The careful exploration of the Red Sea rift
other potential area for Red Sea type deposit zone by Preussag (Backer, 1975) has most likely
exploration is in the Gulf of Mexico, where discovered any currently active brine pools.
evaporites and subsurface brines occur exten- However, discovery of metalliferous deposits
sively, but igneous activity is generally not spa- not related to present brine pools is dependent
tially related. Indeed, one hypersaline bnne pool on direct sediment sampling and, considering
has been reported (Shokes et al., 1977) but me- the enormous area of the rift zone, important
tallization is insignificant. deposits may remain undiscovered. In addition,
Exploration within the Red Sea, by Preussag, metalliferous deposits which became inactive a
has concentrated on the axial rift zone and pro- few tens of thousands of years ago may still
vides a model example upon which similar ex- have economic sign~ficanceeven though buried
ploration programs might be developed. A under a few meters of recent sediments and ef-
number of possible exploration tools have been fectively beyond the reach of gravity corers of
 RED SEA METALLIFEROUS BRINE DEPOSITS

FIG.5. Relative weight proportions of detrital (aluminum, calclum. magne-

sium), oxide (iron, manganese), and sulfide (zinc, copper) components in sed-
iments from the Red Sea nft zone. Based on averaged bulk chemical data. From
Schoell et al. (1974). Abbreviations: VA-Valdivla Deep; KD-Kebrit Deep;
SDP-Suakin Deep Sea plam; MA-Commission Plain, Marsamlnt; VED-
Vema Deep; ABT-Atlantis Terrace; WT-Wando Terrace; VEDW-West
of Vema Deep; NDE-Nereus Deep, East Basm; NDW-Nereus Deep. West
Basin; SD-Suakin Deep; AL-Albatross Deep; PSD-Port Sudan Deep;
TDWE-Thetis Deep (West, East Basin); DETD-Depression east of Thetis
Deep; GYP-Gypsum Deep; CH-Cham Deep; DIS-D~scovery Deep; A-
'4tlantis I1 Deep.

grab samplers. Finally, the recent spectacular provided excellent general indications of poten-
discoveries of base-metal sulfide deposits on the tial hydrothermal vent areas (Craig et al., 1980).
East Pacific Rise (Hekinian et a]., 1980; Speiss Similarly, Lupton et al. (1977) have measured
et al., 1980) and the Juan de Fuca Rise (Delaney a 3000-fold 3He enrichment in the Atlantis I1
et a1., 1982) suggest the possibil~tyof additional brine and Holmes and Tooms (1972) have dem-
Red Sea deposits formed by normal-sallnity sea- onstrated enrichment of iron, manganese, zinc
water. Recently discovered dispersed metalli- and copper by as much as five orders in the
ferous occurrences on the Commission Plain seawater column above the Atlantis II Deep.
(Figs. 1 and 5, Backer, 1979) may have formed Helium isotope and trace element studies of Red
from buoyant seawater hydrothermal flulds and Sea Deep Water should be sensitive methods
there may be high grade base-metal deposits capable of locating buoyant hydrothermal sea-
related to the vent areas for these fluids. water vent areas. Follow-up investigations uti-
What are the tools which might be useful in lizing bottom photography, deeply towed
exploration for inactive brine related deposits instrumental surveys, and manned submersible
and for possible seawater related deposits? studies could then be used to pinpoint sulfide
Discovery of massive sulfide deposits in the deposits.
Pacific has required very detailed explorat~on Continued search for inactive sulfide deposits
uslng submersibles, and a substantial element formed by venting of dense brines has, perhaps,
of luck. However, studies of manganese and a greater potential for success in the Red Sea.
helium isotope values m the water column have Both geophysical and geochemical explorat~on
 UNCONVENTIONAL MINERAL DEPOSITS

have potential applicability. Rona ( 19781 has Deeps and the area around the Atlantis I1 Deep
suggested that a large residual magnetic low (Table 3). Specific core locations with respect
spatially related to.the Atlantis I1 Deep is a result to brine pools are indicated in Figs. 2 and 6,
of hydrothermal alteration of underlying oceanic and sulfur isotope values are plotted against car-
crust by venting hot brine. Similar magnetic bonate carbon content in Fig. 7. Carbonate car-
anomolies might locate inactive deposits, but bon content is useful as an indicator of the relative
very detailed local surveys would be required. proportion of detrital matenal in a given sample.
In addition, dense brines may travel significant Normal detrital samples in the Red Sea rift zone
distances along bottom topography before pool- have about 6% carbonate carbon. and are thus
ing In suitable deeps. Thus, hydrothermally al- about 50% carbonate minerals with the balance
tered rocks in vent areas may not be directly consisting of clays and other silicates. Iron-ox-
related to sulfide deposits. ide-containing samples contain less of the de-
Bignell et al. (1976) have demonstrated the trital component due to the dilution effect.
usefulness of geochemical exploration in the Red Hydrothermal samples from the Atlantis I1 Deep
Sea using surficial sediment samples. Disper- have very low carbonate carbon contents (Fig.
sion halos of manganese, copper, zinc, and mer- 7), due to very rapid accumulation of hydro-
cury can be recognized for up to 10 km away thermal precipitates (50- 150 cm/103 yr accord-
from Atlantis I1 and Nereus Deeps. These dis- ing to Garson and Krs, 1976; Shanks and
persion halos significantly enlarge the explo- Bischoff, 1980). Most samples with high car-
ration target and are potentially very useful in bonate carbon contents have typical marine bac-
exploration for inactive as well as active de- teriogenic sulfur isotope values, but it is
posits. However, the geochemical anomolies are encouraging that a number of samples from the
significantly greater in the area around the Ner- DOP zone (Fig. 3) of the Atlantis I1 sediments
eus Deep even though Nereus Deep contains and from other areas have high carbonate con-
much less intensely metallized sediments, far tent and still retain hydrothermal sulfur isotope
below potentially economic grades and vol- signatures.
umes. Thus, numerous geochemical halos may The potential usefulness of sulfur isotope stud-
exist in the Red Sea trough which are unrelated ies in determining proximity to hydrothermal
to significant mineralization. Once such an- activity in the Red Sea is dependent on two
omolies are located. additional geochemical factors: dispersion of hydrothermal sulfides and
techniques which might distinguish significant the meaning of positive sulfur isotope values in
and insignificant signals would be extremely terms of significant metallization. Examination
useful. of samples from the Atlantis I1 area (Fig. 2,
Sulfur isotope studies of sulfide phases in sed- Table 3) is instructive. Sulfides with positive
iments have considerable potential for explo- sulfur isotope values have been analyzed in sam-
ration in the Red Sea, and may be more definitive ples from cores 8K and 10K from the Wando
than other geochemical indicators. Shanks and Basin (or Terrace), from core 118K (10 km south
Bischoff (1980) and Kaplan et al. (1969) have of the Atlantis I1 Deep), and from core 634K
shown that hydrothermal sulfides in the Atlantis from the Shagara Deep. Wando samples are typ-
I1 Deep have sulfur isotope values which cluster ical of sulfide and silicate zone Atlantis 11 sed-
+
about 6, and range from 1 to 14. Marine bac- iments, 118K samples are iron-oxide rich, and
teriogenic sulfides in the Red Sea and elsewhere Shagara samples cannot be macroscopically dis-
give negative and erratic sulfur isotope values, tinguished from normal detrital sediments. Thus,
ranging from about - 10 to - 45 (Table 3). there is a large range in the metalliferous com-
Thus there is significant contrast between hy- ponent, but all of these samples have positive
drothermal and bacterial values, and sulfur iso- sulfur isotope values. Dispersion of hydrother-
tope studies on metalliferous layers in otherwise mal sulfides occurs to at least 18 km from the
normal detrital sediments might be extremely Atlantis 11 Deep, due either to brine overflow
useful in determining proximity to hydrothermal from the main pool or to peripheral venting sites
centers. related to the same brine plumbing system. This
Systematic sulfur isotope and geochemical assertion is supported by the presence of brine
studies have been carried out on specially se- in the Albatross Deep which is chemically iden-
lected metalliferous samples from the Red Sea . tical to the Atlantis I1 lower brine (Table I).
 RED SEA METALLIFEROUS BRINE DEPOSITS

Table 3. Sulfur Isotope Values of Red Sea Sulfides

Mineralogy*
Depth, C l4
Deep Core cm Age SUS* Major Minor
Kebnt
Gypsum

Vema
Nereus

Hadarba-Hat~ba
Vald~v~a
Wando

A11 vlclnlty
(from Kaplan
.
et a1 1969)

Albatross

Shagara

(from Gant, 1980)

*Brackets lndlcate duplicate analyses of separate sp11u of the sample

tSulfates
$Abbrevlatlons Ar-aragonite. Bs-bassan~te. Ct--calc~te. D-dolomite. Gc-goeth~te. Gy-gypsum. Hm-hernat~te. I-lll~te, K-
liaollnltr, M-montmor~llon~te. Mt-magnetite. Py-pynte. Q-quartz. Sp-sphalente
168 UNCONVENTIONAL MINERAL DEPOSITS

0
- KM 5
THETIS DEEP

FIG. 6. Bathymetric maps of Thetis, Nereus, and Suakin Deeps showing locations of cores utilized in this
study. From Bkker and Schoell (1972). Stippled areas are brine filled.

Hydrothermal sulfides were not detected in core +

average 6. Thus positive sulfur isotope values
388K from Albatross, probably due to poor core in sediments of the Red Sea rift zone indicate
placement with respect to the brine pool. high temperature brine activity, as opposed to
Shanks and Bischoff (1977, 1980) have con- low temperature brines derived from paleowa-
cluded that the sulfides in the Atlantis I1 Deep ters which have not interacted with hot nft zone
sediments have formed due to a sulfide reduc- intrusives. Indeed, outside of the Atlantis II area,
tion process in heated subsurface brine. Sub- hydrothermal sulfides are found only in the The-
surface brine temperatures in excess of 200°C tis and Suakin Deeps (Table 3, Figs. 6 and 7).
are probably required to produce the character- Thetis has no brine at present, and Suakin brine
istic "hydrothermal" sulfur isotope values which has oxygen and hydrogen isotope values which
 RED SEA METALLIFEROUS BRINE DEPOSITS

8.0 t ATLANT1S.n
THETIS
A WANC)O
. SHAGARA
SUAKIN
x OTHERS

NORMAL
DETRlTAL\ x

OXIDE
TAl N l NG

FIG. 7. Plot of carbonate carbon content (welght percent) vs. sulfur Isotope value of sulfide minerals (CDT
scale). Carbonate carbon content is a general indicator of the proportion of detntal material In a given Red
Sea sample. Hydrothermal sulfides have posltlve lsotope values and contain sulfide derlved from hot brlne
venting. Detrital samples have negative sulfur Isotope values typlcal of marine bactenogenic sulfides.

dlffer from the paleowater-derived brines and values (Table 3). The Thet~sDeep area certainly
are not amenable to unambiguous interpretation should be carefully evaluated for more extensive
(Schoell and Faber, 1978). deposits.
The presence of sulfides with positive isotope
values in the Red Sea does not ensure ore-grade SUMMARY AND CONCLUSIONS
accumulation of metalliferous muds. Positive The metalliferous sediments of the Atlantis I1
isotope values occur in some metalliferous lay- Deep represent the only known deposlt in the
ers of the Atlantis I1 DOP zone (Fig. 3) and the Red Sea with reasonable potential for devel-
Suakin sediments are very slmilar to the DOP opment as a seafloor base metal sulfide ore de-
zone (Baumann et a]., 1973; Cant, 1980). Thin posit. Atlantis I1 muds are estimated to contain
nontronitic and oxide layers in these sediments 1.7 Mt Zn, 0.4 Mt Cu, 5000 t Ag, and a small
indicate only intermittent hydrothermal activity. gold byproduct. A pre-pilot mining test con-
However, positive sulfur isotope values appear ducted in 1979 has demonstrated effective min-
to be a prerequisite for significant hydrothermal ing of the muds by slurry methods, concentration
deposition and are thus a very positive geo- by froth flotation at sea, and tailings disposal
chemical indicator. The most significant dis- without apparent dire consequences.
covery outside of the Atlantis Ll area is the Thetis The Atlantis I1 deposit is the most attractive
Deep, which contains sediments comprised al- thus far discovered in the Red Sea, but the over-
most entirely of magnetite with minor but sig- all grade is low and the economies of mining
nificant base metal sulfides with positive isotope remain very questionable. Discovery of addi-
 UNCONVENTIONAL MINERAL DEPOSITS

tional deposits in the Red Sea remains a pos- Part of the Red Sea," Geol R~tndschau.Vol 62, pp
sibility, especially inactive deposits without 684-697.
B~gnell,R D , 1975, "T~ming,D~stnbut~on and O n g ~ nof
associated brine pools. Submanne Mlnerallzation In the Red Sea," Transacr~ons.
Exploration for active or inactive seafloor brine Institution of Mlnlng & Metallurgy. Vol 84, pp BI-
deposits in the Red Sea and elsewhere should B6
proceed at approximately three levels of refine- .
Blgnell. R D., Cronan, D S , and Tooms. J S I91b. "Metal
ment. Level 1 targets exploration in general areas, Dlspers~onIn the Red Sea as an Aid to Manne Geo-
chem~calExplorat~on." Transact~ons.lnst~tutlonof Mln-
such as the Red Sea rift, known to contain the ing and Metallurgy, Sec. B. Vol 85, pp. B274-B278
critical geological factors: evaporites, igneous Brewer, P G and Spencer, D W , 1969, "A Note on the
activity, and fluid circulation pathways. Level Chermcal Compos~t~on of the Red Sea Bnnes," Hor Brrr~es
2 involves detailed bathymetric, gravity, and and Recenr Hear? Meral Depos~rsIn the Red Sea. E T
Degens and D.A. Ross. eds , Spnnger-Verlag. New York,
magnetic profiling to establish the geologic and pp 174-179
bathymetric framework. This step should dis- Cra~g.H., 1969, "Geochemlstry and O n g ~ nof Red Sea
cover active brine pools by above-bottom re- Bnnes," Hor Brrnes and Recent Heayv Metal Deposlrs
flections. Level 3 involves sediment and water In the Red Sea, E.T Degens and D A Ross, eds , Spnn-
sampling, geochemical analysis for metallic ele- ger-Verlag, New York, pp 208-242
Cralg, H , et a].. 1980. "Geochemical Stud~esof the 21°N
ment dispersion, and sulfur isotope analysis of EPR Hydrothermal Flulds, EOS," Transactrons. Amer-
sulfides. Level 3 should be directed at discovery ]can Geophys~calUn~on,Vol 61. p 992.
of inactive deposits In favorable bathymetric Cronan. D S.. 1980. Underwater Mrnerals. Acadermc Press.
areas, especially if magnetic lows or geochem- New York. 361 pp
Delaney, J.R., et a]., 1982, "Mass~veZlnc and Iron-Rich
ical anomolies are detected. Sulfur isotope stud- Sulfide Deposlts Associated w ~ t hHot Spnngs, Juan de
ies of sulfide minerals provide a geochemical Fuca Rldge." Abstracts, Amencan Assoc~at~on of Petro-
exploration tool which may explicitly indicate leum Geologists, Annual Convent~on,p 43.
proximity to hydrothermal sulfide deposits. Gant. J , 1980, "M~neralog~caland Geochem~calStud~es
of Bnne-Related Metall~ferousSedlments In the Suakln
Deep, Red Sea." M.S. Thesls. Unlverslty ot hr.,c.,bln-
Madlson. 178 pp.
ACKNOWLEDGMENTS Garson, M.S , and Krs, M , 1976, "Geophys~caland Geo-
I gratefully acknowledge the cooperation of logical Evldence of the Relat~onsh~p of Red Sea Trans-
H. Backer, Preussag AG, in providing samples verse Tectomcs to Anc~entFractures," Bullenn. Geological
Soc~etyof Amenca. Vol 87. pp 169-181.
and published information on the current status Glrdler, R W., and Styles, P , 1974, "Two Stage Red Sea
of economic investigations. Floor Spreading," Nature, Physical Sc~ence,Vol 247,
This study was supported by National Science pp 7- 1 1
Foundahon Grants OCE780548 1 and Hekln~an.R , et al , 1980, "Sulfide Depos~tsfrom the East
Pac~ficRlse Near 2I0N: A Mlneralog~caland Geochern-
OCE8018644. lcal Study," Science. Vol 207. pp. 1433-1444
Holllster, C D., et a1 , 1972, "S~te105-lower Contlnen-
tal R ~ s Hills,"
e Initial Reports. Deep Sea Dnlllng Project,
Vol 11. pp. 219-312
REFERENCES .
Holmes. R., and Tooms. J.S 1972, "Dispersion from a
Anon , 1981, "Two Thousand Meters Under the Sea," Submarine Exhalatlve Ore Body," Proceedmgs. 4th In-
M ~ n ~ nMagaz~ne,
g pp 114- 122. ternational Geochem~calExplorat~onSympos~um,Instl-
Backer, H , 1975, "Explorat~on of the Red Sea and Gulf tutlon of Mlnlng & Metallurgy. London. pp 193-201.
of Aden dunng the M S VALDlVIA Cru~ses'Erzsch- Kaplan, I R., Sweeney, R E , and N~ssenbaum,A., 1969,
lamme A' and 'Erzschlamme B'," Geol Jarbuch, D 13, "Sulfur Isotope Studles on Red Sea Geothermal Bnnes
pp. 3-78. and Sedlments," Hor Brtnes and Recenr Heavy Metal
Backer, H , and R~chter,H., 1973, "Die rezente hydro- Depos~tsof the Red Sea. E.T Degens and D.A Ross,
thermal-sed~mentarelagerstatte Atlant~sI1 t ~ e fIm Roten eds , Spnnger-Verlag. New York, pp. 474-498.
Meer," Geol. Rundschau, Vol 62. pp. 697-741. Luck, K., and Nawab, Z , 1981. "Metalllferous Sediments
Backer, H., and Schoell M , 1972, "New Deeps w~thBrines In the Atlant~sI1 Deep-Recovery and Selected Speclal
and Metall~ferousSediments In the Red Sea," Nature. Problems, Conference Report, lnternatlonal Conference
Physical Science, Vol. 240, pp. 153- 158. for Ocean Mmlng. Dusseldorf. pp. 103-1 15.
Backer. H., 1979, "Metall~ferous Sed~mentsof the Red Manhelm, E.T., 1974, "Red Sea Geochemlstry," I n ~ t ~ a l
Sea, Occurrence, Exploration and Resource Assess- Reports. Deep Sea Dnlllng Project. Vol 23, pp. 975-
ment," Proceedmgs, Internat~onalSeminar on Offshore 998.
Mlneral Resources, Orleans, France, BRGM Document Mustafa, Z., and Amann, H., 1980. "The Red Sea Prepllot
NO. 7, pp. 317-338 Minlng Test 1979," Proceedmgs, 12th Annual Offshore
Baumann. A., R~chter.H.. and Schoell. M , 1973. "Suak~n Technology Conference, Society of Petroleum Engineers,
Deep. Brines and Hydrothermal Sedlments In the Deepest Dallas, Paper No 3874
 RED SEA METALI-IFEROUS BRINE DEPOSITS

Nawab, Z , and Luck, K , 1979, "Test Mlnlng of Metal- Shanks, W.C , 111. and Blschoff. J.L , 1977, "Ore Trans-
llferous Mud from the Red Sea Bottom," Meerestechnrk. port and Depos~tlonIn the Red Sea Geothermal System.
Vol 10, No 6. pp. 181-187 A Geochem~calModel," Geoch~m~ca et Cosmochimrca
Rona, P.A., 1978, "Cntena for Recogn~t~on of Hydro- Acta. Vol 41. p 1507-1519
thermal Mineral Deposits In Oceanlc Crust," Economrc
Geology, Vol 73, pp. 135-160
.
Shanks, W C 111, and Bischoff, J.L , 1980, "Geochem-
~ s qSulfur
, Isotope Compos~t~on.
and Accumulat~onRates
Schoell. M , Backer, H., and Baumann. A., 1974, "The
of Red Sea Geothermal Deposlts," Economrc Geology,
Red Sea Geothermal Systems, New Aspects on Thew
Bnnes and Assoc~atedSediments," lntemat~onalAsso- Vol. 75, pp. 445-459
clatlon on the Genes~sof Ore Deposlts (IAGOD). Vama Shokes, R.F , et al.. 1979. "Anox~c,Hypenallne Basln In
Meetlng. the Northem Gulf of Mexlco." Science, Vol 196. pp '
Schoell, M , 1976, "Heatlng and Convect~onwlthln the 1443-1446
Atlantls 11 Deep Geothermal System of the Red Sea," Spiess, F N , et al. (RISE Project Group), 1980, "East
Proceedrngs. 2nd Unlted Natlons Symposium on the De- Paclfic R~se.Hot Spnngs and Geophys~calExpenments,':.
velopment and Use of Geothermal Resources. Vol. 1. Sc~ence.Vol 207. pp. 1421-1432
pp 583-590
Schoell, M.. and Faber, E , 1978, "New lsotoplc Ev~dence . .
Weber, H et a1 1981, "Metallurg~cal Treatment of Red
for the Ongln of Red Sea Bnnes." Nature, Vol. 275. Sea Concentrates," Corijerence Reports, International
pp 436-438 Conference for Ocean Mlnlng. Dusseldorf, pp 1 16- 122