The Abundance of Seafl Oor Massive Sulfi de Deposits
The Abundance of Seafl Oor Massive Sulfi de Deposits
The Abundance of Seafl Oor Massive Sulfi de Deposits
Mark Hannington1, John Jamieson1, Thomas Monecke2, Sven Petersen3, and Stace Beaulieu4
1
Department of Earth Sciences, University of Ottawa, Ottawa K1N 6N5, Canada
2
Department of Geology and Geological Engineering, Colorado School of Mines, Golden, Colorado 80401, USA
3
Leibniz Institute of Marine Sciences, IFM-Geomar, 24148 Kiel, Germany
4
Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543, USA
© 2011 Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or [email protected].
GEOLOGY,
Geology, December
December 2011
2011; v. 39; no. 12; p. 1155–1158; doi:10.1130/G32468.1; 3 figures; Data Repository item 2011342. 1155
Figure 1. Global distribution of seafloor hydrothermal systems and related mineral deposits. Version 2.0 of the InterRidge Global Database
(Beaulieu, 2010) used in this study contains information on 554 sites of seafloor hydrothermal activity (confirmed and unconfirmed) and
inactive deposits. About 300 are sites of high-temperature hydrothermal venting; 165 are confirmed sites of massive sulfide accumulation
(Table DR1 [see footnote 1]). Credits: S. Beaulieu, K. Joyce, and S.A. Soule (Woods Hole Oceanographic Institution).
ESTIMATING THE GLOBAL ABUNDANCE from heat fluxes (Baker, 2007) and plume data (Baker and German,
To estimate the abundance of seafloor massive sulfide in the neovol- 2004), the deposits are farther apart on the slow-spreading ridges (aver-
canic zones of the global oceans, we examined deposit densities in 32 age of 174 km) and closer together on the fast-spreading ridges (average
control areas of roughly equal size (5° of latitude by 5° of longitude) con- of 54 km). Deposit densities on backarc spreading centers are similar to
taining 129 occurrences (Table DR2 in the GSA Data Repository1). The those on mid-ocean ridges with similar spreading rates; along volcanic
areas were chosen to represent permissive geology throughout the oceans arcs the spacing is more closely linked to the spacing of individual vol-
and include ~7200 km of ridge and 2800 km of arc and backarc spreading canoes (de Ronde et al., 2003). The combined average spacing for all 32
centers. The map scale used for the analysis (1:2,500,000) was chosen so control areas is 107 km (Fig. 2B). Based on the cumulative strike length
that the quality of the map data was the same in all cases. The measure- of the oceanic plate boundaries (89,000 km) and an average density of one
ment of deposit densities in the control areas requires an unambiguous deposit every ~100 km, the total number of deposits expected in the global
definition of what constitutes a “deposit”; unfortunately, descriptions of neovolcanic zones is ~900. Taking into account the range of deposit densi-
deposits in the literature have included everything from a single vent or ties, the expected number of deposits is at least 500 (90th percentile) and
chimney to a large mound or cluster of vent complexes. In this study, we not more than 5000 (10th percentile).
selected a subset of 106 deposits with reported dimensions of more than This result obviously has a large uncertainty, but independent esti-
100 m2 on the seafloor as the minimum unit and clusters of mounds or vent mates of the spacing of hydrothermal vents, at least on the mid-ocean
complexes spatially separated from the next nearest cluster by more than ridges, all arrive at similar numbers. Global heat flux associated with
10 km as the maximum unit. Details of the methodology and assumptions high-temperature convection at mid-ocean ridges ranges from 1.1 to 2.5
are provided in the Data Repository. × 1012 W, with the most commonly cited values close to 1.8 ± 0.3 × 1012
An average of four deposits was included in each of the 5° × 5° W (Elderfield and Schultz, 1996; Mottl, 2003; Sinha and Evans, 2004).
control areas; 90% of the control areas have densities of two or more A large number of studies have shown that at least 90% of this axial flow
deposits per 100,000 km2, 50% have densities of six or more deposits, is diffuse flow (i.e., 350 °C fluid that has been cooled by mixing with
and 10% have densities of ten or more deposits. In this study, we consider seawater before reaching the seafloor). In order to account for the high-
only those deposits that are located within the neovolcanic zone. Thus, temperature flux (10% of 1.8 ± 0.3 × 1012 W), as many as 50,000–100,000
deposit densities can also be expressed in terms of the linear dimensions black smokers would be required, assuming heat output of 2–5 MW for
of ridge or arc segments between the occurrences (Fig. 2A). As expected a single vent, or a density of at least one black smoker for every 1 km
1
of ridge axis. Large vent fields, with up to 100 black smokers and heat
GSA Data Repository item 2011342, supplemental information on the abun-
dance of seafloor massive sulfide deposits, is available online at www.geosociety
outputs equivalent to 200–500 MW, are expected to account for most of
.org/pubs/ft2011.htm, or on request from [email protected] or Documents the high-temperature flux. A vent complex of this size would be required
Secretary, GSA, P.O. Box 9140, Boulder, CO 80301, USA. only every 50–100 km, amounting to ~1000 large fields on the global
300
250 nic zones, we can place some constraints on the total sulfide accumula-
tion by assuming a mass distribution similar to that of the known depos-
200
its (Hannington et al., 2010). Unfortunately, the sizes of most deposits
150
are incompletely known; many reported dimensions include large areas
100 of discontinuous sulfide outcrop or barren substrate between chimneys
50 and mounds, thereby overestimating the sizes of the deposits. Significant
0 deposits with explored dimensions on the order of 5000 m2 or more have
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 now been found in most of the settings considered in this study, and in sev-
Control area (n = 32) eral cases drilling information is available to judge the continuity and thick-
Figure 2. Seafloor massive sulfide deposit densities. A: Distances
ness of sulfide outcrops. Four deep drill holes that penetrated the Bent Hill
between adjacent deposits along 10,000 km of ridge, arc, and back- deposit and Ocean Drilling Program (ODP) mound at Middle Valley (ODP
arc spreading centers. Data are included only for those deposits or Legs 139 and 169) indicate a combined tonnage of 1–1.5 × 107 tonnes (Zie-
clusters of deposits more than 10 km apart (n = 106). B: Mean de- renberg et al., 1998). At the large TAG mound on the Mid-Atlantic Ridge,
posit densities (average spacing) in each of the 5° by 5° control ar- 17 holes drilled to a maximum depth of 125 m indicate 2.7 × 106 tonnes
eas listed in Table DR2 (see footnote 1); ranges are the minimum and
maximum distances between adjacent deposits in each control area. of massive sulfide in this deposit (ODP Leg 158; Hannington et al., 1998).
Closed symbols—mid-ocean ridges; open symbols—arcs and back- Data from extensive drilling of the Solwara 1 deposit in the Eastern Manus
arc spreading centers. The horizontal line indicates the combined Basin (146 holes to a maximum depth of 20 m) have been used to estimate
average value for all 32 control areas (107 km). Details are provided a resource of 2.1 × 106 tonnes in a chimney zone covering 90,000 m2 (Lip-
in the GSA Data Repository (see footnote 1).
ton, 2008). Using the drilling data as a guide, we estimated the tonnages
of the best-mapped deposits from the outcrop areas in the global database
(see Table DR1). These data are the basis for a first-order tonnage model
for seafloor massive sulfide deposits presented in Hannington et al. (2010).
Of 62 deposits used in the tonnage model, only eight have dimensions
Cumulative massive sulfides (106 tonnes)
indicating sizes larger than 2 × 106 tonnes. Three of these have been con-
80
firmed by drilling (TAG, Middle Valley, Solwara 1); the others have sulfide
70 outcrop dimensions that are consistent with similarly large tonnages (e.g.,
<2% 12% 86%
60 Sunrise, Krasnov, Semyenov, Puy des Folles, Zenith-Victory; see the Data
Repository). However, the median deposit size is only 70,000 tonnes, and
50
more than a third of the deposits are considered to be smaller than 3000
40 tonnes. The total amount of massive sulfide in the deposits used in the
tonnage model is ~5 × 107 tonnes, ~70% contained in the largest 10% of
30
deposits. Using these data, an estimate of the cumulative tonnage of seafloor
20 massive sulfide in the global neovolcanic zones can be made by populating
10 the tonnage model with the estimated numbers of deposits discussed above.
If we assume that there are ~1000 deposits with a minimum size of
0 100 tonnes, a maximum size of 1 × 107 tonnes, and the size distribution
14 12 10 8 6 4 2
Full spreading rate (cm/yr) given by Hannington et al. (2010), the total amount of massive sulfide in
the neovolcanic zones of the global oceans is on the order of 6 × 108 tonnes.
Figure 3. Expected distribution of seafloor massive sulfide deposits The largest 10% of deposits would contain ~4 × 108 tonnes. Because the
on the mid-ocean ridges as a function of spreading rate. The propor- slow-spreading centers account for ~60% of the total ridge length and host
tion of massive sulfide at different spreading rates is estimated from
the length of the ridge segments (Carbotte and Scheirer, 2004), the the largest deposits, they are expected to account for more than 85% of the
deposit density as a function of spreading rate, and the expected total tonnage of seafloor massive sulfides at ridges (Fig. 3). The contained
sizes of the deposits (see the Data Repository [see footnote 1]). The metal is more difficult to estimate, owing to the highly variable metal
total length of ridges represented in this plot is 53,000 km; the ex- grades of surface samples (Hannington et al., 2010). For a median grade
pected number of deposits (n = 537) have a total estimated tonnage
of ~3.6 × 108 tonnes. The large proportion of massive sulfide ex-
of 5 wt% Cu + Zn, which closely matches that of massive sulfide deposits
pected on the slow-spreading ridges reflects the cumulative length on land (Franklin et al., 2005), the total amount of metal (Cu + Zn) along
of the slow ridges plus the large sizes of the known deposits. the global neovolcanic zones is expected to be ~3 × 107 tonnes.