Abstract
Coupled magmatic and tectonic activity plays an important role in high-temperature hydrothermal circulation at mid-ocean ridges. The circulation patterns for such systems have been elucidated by microearthquakes and geochemical data over a broad spectrum of spreading rates, but such data have not been generally available for ultra-slow spreading ridges. Here we report new geophysical and fluid geochemical data for high-temperature active hydrothermal venting at Dragon Horn area (49.7°E) on the Southwest Indian Ridge. Twin detachment faults penetrating to the depth of 13 ± 2 km below the seafloor were identified based on the microearthquakes. The geochemical composition of the hydrothermal fluids suggests a long reaction path involving both mafic and ultramafic lithologies. Combined with numerical simulations, our results demonstrate that these hydrothermal fluids could circulate ~ 6 km deeper than the Moho boundary and to much greater depths than those at Trans-Atlantic Geotraverse and Logachev-1 hydrothermal fields on the Mid-Atlantic Ridge.
Similar content being viewed by others
Introduction
Slow (1.2– 5.5 cm year−1) and ultra-slow (<1.2 cm year−1 full spreading rate) spreading ridges are typically characterized by low melt production1. Conductive cooling limits melt production and the mantle rocks are often tectonically exposed onto the seafloors1,2,3,4. This distinctive geological setting allows faulting to be particularly extensive and to penetrate deep into the crust and upper mantle, thereby potentially forming pathways for enhanced hydrothermal fluid circulation5. The thermal structure of a spreading center is controlled in part by this hydrothermal circulation that can result in the formation of large polymetallic sulfide deposits6. The depth of the hydrothermal circulation is important in the efficiency of lithospheric cooling7, the style of accretion of the lower oceanic crust8, and the potential to leach ore-forming elements from the rock6. The depths of hydrothermal circulation on detachment faults at oceanic core complexes (OCC) of slow spreading ridges, such as Trans-Atlantic Geotraverse (TAG) and Logatchev-1 hydrothermal fields, can be traced up to ~7 and ~6 km below the seafloor (bsf), respectively9,10. The chlorine-excess in melt inclusions hosted by basaltic rocks in the South Mid-Atlantic Ridge (MAR) and Gakkel Ridge offers further geochemical clues that hydrothermal alteration reaches lower crustal depths11. Despite these observations, however, the origin and depth of the hydrothermal fluids in active high-temperature vent fields at the ultra-slow spreading ridge has not, until now, be investigated in any detail.
The Longqi-1 hydrothermal vent field (~49.7°E) in the Dragon Horn region (Fig. 1) of the ultra-slow spreading Southwest Indian Ridge (SWIR) exhibits high-temperature hydrothermal vents associated with a major detachment fault system and has been the subject of recent intensive studies12, providing an opportunity to examine this problem. Here, we show the comprehensive geophysical and geochemical investigations on this hydrothermal field, which are based on the ocean bottom seismometers (OBS), the Jiaolong human-occupied vehicle, and Qianlong II autonomous underwater vehicle. A concerted effort was also undertaken to track the pathway of hydrothermal circulation using a two-dimensional (2D) numerical model of circulation in a NaCl-H2O fluid system to verify whether the hydrothermal fluids derived from the depth could circulate up to the seafloor and vent at the observed high temperatures. Our results show that the hydrothermal circulation below the Longqi-1 field is associated with a detachment system penetrating to the depth of 13 ± 2 km below the seafloor, and the hydrothermal fluids could circulate ~6 km deeper than the Moho boundary that is much deeper than those at TAG and Logachev-1 hydrothermal fields on the MAR.
Results
Geological setting
The Dragon Horn area is located on the south flank of the SWIR segment 28 (~49.7˚E), following the nomenclature described previously13. The corrugated surface of the detachment fault of the OCC exposed on the southern ridge flank covers an area of approximately 5 km × 5 km (Fig. 1b). The basalt-hosted Longqi-1 active vent field is located at the edge of this OCC at water depths of 2700–2900 m on the southwest wall of the axial rift valley and comprises two sulfide-bearing vent zones named S zone and M zone, respectively (Fig. 1b, d)12. The highest measured fluid temperature of 379 °C is associated with the DFF6 (Dragon Flag Field: English name of the Longqi-1 Field) vent site of the M zone (Fig. 1d, Supplementary Table 1). The active Longqi-1 vent field together with the inactive Longqi-3 hydrothermal field and a hydrothermal plume anomaly site form a possible linear mineralized zone along terminus 2 of the detachment fault 2 (DF2) (Fig. 1b, c), where serpentinized peridotite outcrops have been observed and sampled. Additionally, consolidated carbonate sediments representing low-temperature hydrothermal processes were also found on the east side of the OCC (Fig. 1b)14.
Seismic observation
We located 512 microearthquakes with local magnitudes of −0.5 < ML < 3 by a double-difference relocation method during a series of discontinuous OBS monitoring experiments. The relocated hypocenters were divided into two sections near the non-transform discontinuities (NTDs) and OCC, respectively (Supplementary Fig. 1). Seismic activity on the west side of Longqi-1 is attributed to a network of brittle faults and fractures in the NTD zone. Therefore, only hypocenters in the OCC zone, which are associated with detachment faulting beneath the southern flank of the axial valley (within white rectangle in Fig. 1b), are considered in this study. Most events are located along the axial volcanic ridge at a depth of 13 ± 2 km bsf and shoale to ~3 km bsf to the south. The cross-sectional view suggests that most of the hypocenters are focused along two seismically active structures within 1 km offset (inside of dotted curve in Fig. 1c), with some of the earthquakes occurring outside the two seismic zones.
We infer that the earthquakes beneath Longqi-1 field are more possible induced by tectonic activities rather than by magmatic intrusions, based on the following observations. Firstly, active-source wide-angle seismic data conducted along the SWIR from 49.3˚E to 50.8˚E15,16 clearly shows that there is no crustal magma chamber beneath Longqi-1 hydrothermal field and that no volcanic tremors were recorded by the OBS network. Secondly, dike-induced earthquakes associated with magma intrusions typically show hypocenters migrating with time along a narrow path17; however, the distributions of the earthquakes in our case show no migration behavior with time. In addition, the similar spatial distribution of earthquakes has been observed at Logatchev-1, TAG, and Irinovskoe hydrothermal fields at the MAR, which were explained as the response to the extension of the detachment faults9,10,18. The occurrence of scattered microearthquakes outside the regions where a majority of hypocenters are located has also been observed in previous studies at TAG9 and 13˚20′N on the MAR18. These scattered earthquakes are typically inferred to be due to the volume expansion accompanying serpentinization19.
Overall, we define two detachment faults (DF1 and DF2 in Fig. 1c) in the Dragon Horn area based on the morphology, geological sampling, and seismic data. First, we have documented that the smooth seafloor is a detachment fault by sampling of exhumed mantle-derived peridotite (Fig. 1b). Second, most of recorded microearthquake epicenters project along the trace of two detachment fault zones (Fig. 1c). Finally, a three-dimensional P-wave seismic velocity model shows that the footwall of the detachment faults beneath the Dragon Horn area is characterized by shallow, high crustal velocities and a strong vertical velocity gradient (Fig. 1c, Supplementary Fig. 2).
Fluid chemistry
The volatile gas and element concentrations and O–H isotopic compositions of the Longqi-1 venting hydrothermal fluids are shown in the Supplementary Tables 1 and 2. The boron (B), potassium (K), chlorine (Cl), silicon (Si), and lithium (Li) concentrations are compared with those of venting fluids from other vent sites along the global mid-ocean ridges (Fig. 2). Compared with vent fields hosted by mafic rocks, the Longqi-1 vent fluids show obvious depletion in dissolved boron (Fig. 2a) and lower K/Cl ratios (Fig. 2b), while their Si and Li concentrations are within the typical range of global mafic rock-hosted vent fluids (Fig. 2c, d). The concentrations of H2, CH4, and H2S in the high-temperature vent fluids are considerably higher than those observed at TAG (Supplementary Table 1), a detachment-related hydrothermal system at 26°N on the MAR. The δ18O and δD (δ18O (D) [‰] = (Rsample/Rreference−Rreference) × 1000, where R is the ratio of 18O/16O or D/H, reference is the Vienna Standard Mean Ocean Water (VSMOW)) of the vent fluids vary from 0.18‰ to 1.21‰, and from 0.8‰ to 3.9‰, respectively, both of which are considerably higher than that of local bottom seawater (Supplementary Table 2).
Discussion
The distribution of microearthquakes along the DF2 is similar to that observed at TAG on the MAR; however, our results demonstrate that DF2 penetrates ~6 km deeper than the detachment fault at TAG as inferred from seismic data9 (~13 ± 2 vs. ~7 ± 1.1 km) (Fig. 1c). Although some microearthquakes at ultra-slow ridges have been detected deeper than at the Dragon Horn area, e.g., 16 km bsf for 85°E/85°N Gakkel Ridge20 and 17 km bsf for 12.5°~14.5°E SWIR21,22, no high-temperature hydrothermal activities were found along these ridges. The maximum depth of microearthquakes in these regions reflect the variation in crustal thickness along-axis because of poor and discontinuous melt supply at these ultra-slow spreading centers22. The steep and deep DF2 with associated serpentinized peridotites exposed at the seafloor may be at an initial stage in the rolling-hinge model, in which the oceanic detachment faults initiate at high dips and rotate to low-angle geometries as displacement increases23. In contrast, DF1 with scattered epicenters has a much lower angle, as indicated by flexural exhumation of the OCC (dashed line delineated area in Fig. 1b) and the dome-shaped surface. Following the rolling-hinge model and based on the lower angle of the fault, it is believed that DF1 is more mature than DF2 (ref. 23). These two detachment faults jointly compose a twin detachment faulting system (Fig. 1c). The detachment faults appear to provide the main circulation path for the hydrothermal fluids9,10,15,24. An important consequence of this geometry is that deep detachment faults would allow the seawater to penetrate to a great depth along it and a long fluid circulation/reaction path.
The relatively long circulation/reaction path for Longqi-1 hydrothermal fluids is supported by the geochemical data. The depletion in dissolved boron in the vent fluid from Longqi-1 hydrothermal field (Fig. 2a) suggests seawater reaction with ultramafic rocks, where experimental data have documented the tendency of boron to partition into serpentine and/or chlorite from solution at a range of temperatures and pressures (200–300 °C at pressure around 500 bars)25,26,27. As with TAG vent fluids (Fig. 2a), boron removal could occur during recharge or during more deeply seated reaction where increasingly higher temperatures and lower fluid/rock mass ratios would prevail. Seawater (usually with Cl of 550 mmol kg−1, and B of 420 μmol kg−1 (ref. 28)) passage through this region of the crust would not only lead to boron removal from the fluid, but the continued hydrolysis and reaction of olivine and plagioclase could contribute to the observed elevated dissolved chloride concentrations27,29. Chloride concentrations in vent fluids elevated relative to seawater is a noteworthy feature of many detachment-related hydrothermal systems at mid-ocean ridges30,31,32. Although moderately low and high dissolved chloride concentrations in vent fluids are often accounted for by phase separation effects in the NaCl–H2O system, especially in basalt-hosted hydrothermal systems with well-defined shallow magma chambers (e.g., East Pacific Rise 9°N), the same is less likely here owing to the high hydrostatic pressures that prevail at the inferred depths of the SWIR vent system. The preponderance of olivine and plagioclase bearing plutonic rocks, needed for hydrolysis reactions, is also suggested by the usually low K/Cl, high Ca/Cl, and moderately high dissolved methane of Longqi-1 vent fluids (Fig. 2b, Supplementary Table 1). Entrapment of mantle-derived CO2 and its reduction to methane in fluid inclusions in plutonic rocks of the ocean crust at the SWIR has been long recognized33,34,35,36 and subsequent leaching by hydrothermal alteration37 could account for the methane dissolved in the vent fluids reported here.
Consistent with hydrothermal vent fluids from the slow spreading MAR, hydrogen and oxygen isotope data for Longqi-1 vent fluids (δ18O from 0.18‰ to 1.21‰, δD from 0.8‰ to 3.4‰, Supplementary Table 2) are distinctly elevated and indicate long reaction paths accompanied by intensive water–rock interaction as isotopic fractionation increases with decreasing water/rock ratios, even at moderately high temperatures32,38,39,40,41,42. The moderately high dissolved Si and Li concentrations (Fig. 2c, d; Supplementary Table 1), however, are not consistent with serpentinization, although these data could be explained by subsequent passage of the evolved seawater through basaltic rocks, as is likely the case at TAG, where near seafloor tectonic processes inherent to detachment faulting systems make this possible36,37. Cs and Rb concentrations are also similar between the TAG and Longqi-1 vent fluids (Supplementary Table 1). These data, together with the low dissolved H2 (Supplementary Table 1), suggest phase equilibria involving plagioclase, chlorite, and perhaps epidote solid solutions ± quartz31. These water–rock reactions would help to account for the relatively low measured pH (~3.15–3.6) and high Fe and Ca, especially considering the relatively high dissolved chloride concentrations (Supplementary Table 1, Fig. 2d), which can enhance the solubility of these elements35,43. In effect, a reasonable model for the chemical composition of the high-temperature Longqi-1 vent fluids is somewhat analogous to hybrid models (both mafic and ultramafic rocks are involved in the water–rock interaction zone) advanced earlier to account for vent fluid chemistry associated with regions of the Kairei44 and the Nibelungen29 hydrothermal fields. This is consistent with constraints imposed by the existence of deeply penetrating and high permeability faults that focus fluid flow and allow reaction with heterogeneous crustal components inherent to detachment fault systems. As shown by the velocity structure (Fig. 1c), the thickness of the mafic oceanic crust beneath the Longqi-1 hydrothermal field would be 3–6 km (7.3 km isovelocity contour16 in the Supplementary Fig. 2). This means that the fluid-chemical signatures of the ultramafic rock could not be from shallow depth. Thus, the reaction path of fluids with the hybrid mafic and ultramafic rocks should be long and derive from the depth.
For a hydrothermal system associated with steep and deeply penetrating detachment faults and a considerably long reaction path, the location of the underlying heat source is of particular interest. The highest measured temperature of the venting fluid of the Longq-1 hydrothermal fluids is as high as 379 °C, which, combined with an estimated heat output of ~250 ± 100 MW45, requires a magmatic heat source46. The seismic velocity structures and the distribution of the microearthquakes preclude the existence of a crustal magma chamber or gabbro intrusion beneath the Longqi-1 field shallower than 13 ± 2 km bsf (Fig. 1c). Therefore, the heat source would most likely be located below the depth of 13 ± 2 km and presumably under the axial volcanic ridge. To investigate the dynamics of the hydrothermal circulation and whether the discharge of high-temperature fluids at Longqi-1 can be derived from a very deep magmatic heat source through such a deep and steeply dipping fault zone, we conducted numerical simulations using a hydrothermal model solving for porous flow of seawater24,47 (see Methods for the details and Supplementary Fig. 3 for the setting of initial conditions). This numerical modeling allows us to investigate the temperature of the venting fluids at the seafloor for a range of given initial conditions including the width of the fault zone and the permeability contrast between the fault zone and the background (here the background is the oceanic lithosphere without fault zone) (kdf/kb). In the modeling, we set the heat source depth at the bottom of the lower limit of the microearthquakes (~13 km bsf), and the temperature to 650 °C, which is consistent with the isotherm deduced from the brittle–ductile transition of the lithospheric mantle inferred from the maximum depth of the microearthquakes22,48,49. The results show several important clues to the Longqi-1 hydrothermal circulation. As shown in Fig. 3c, for a fault zone with a width around 400 m, the temperature of the venting fluids varies from ~380 to ~410 °C when the permeability of the fault zone varies from 10 to 60 times the background. For a fault zone with relatively low permeability (kdf/kb < 30), the temperature of the venting fluid is higher than 350 °C but is not so sensitive to the width of the fault zone. For a wide range of fault zone width (200–1000 m) and a wide range of kdf/kb (10–100), the predicted temperature of the venting fluids is greater than 300 °C (Fig. 3c, Supplementary Fig. 4). When the kdf/kb is set as 60, which sets the permeability of the fault zone close to the estimated permeability of the discharge zone (3 × 10−14 m2), based on the heat output (see Methods for details), a fault zone width of ~200–400 m will allow the venting fluids to maintain a temperature consistent with the observed value (379 °C) (Fig. 3a). This response of vent temperature to permeability contrast and fault width is similar to that reported by ref. 24, while the simulated discharge mass flux (10−4–5 × 10−4 kg m−2 s−1) is also in the same range as reported by refs. 24,47. Pressure–temperature paths of numerical fluid tracers (Fig. 3b) further suggest that venting fluids had reached depths of up to 13 km bsf and that the fluid always remained in the single phase region of seawater. Both findings are very different from modeling results for the fast-spreading East Pacific Rise47, where fluids circulate no deeper than ~7 km bsf and pressure–temperature paths do intersect with seawater phase boundaries. This is also consistent with inferences based on the fluid geochemistry as discussed above. We also conducted a set of numerical models to test the response of vent temperature to the permeability contrast of DF1 and DF2 (kdf2/kdf1) (Supplementary Fig. 5). The results show that if the permeability of DF1 is higher than the background permeability, DF1 becomes the pathway for recharge flow toward the foot of DF2, but hot rising fluids never flow through DF1. This is in agreement with the absence of high-temperature hydrothermal activity near the terminus of DF1. The observations that there is no obvious signature of hydrothermal circulation near the terminus of DF1, and that there are considerably fewer microearthquakes along DF1 compared with DF2 (Fig. 1c), supports the interpretation that DF1 has much lower permeability than DF2 (high kdf2/kdf1). Thus, the simulation results support the idea that the high-temperature hydrothermal fluids at Longqi-1 originate from as deep as ~13 ± 2 km bsf, given that the temperature of the heat source could be up to 650 °C.
It has been widely accepted that a heat source of magmatic origin is required for high temperature and high heat output hydrothermal activity46. The existence of such a melt zone in the deep lithospheric mantle could be possible, if the focused melt delivery from the neighboring cold areas to the mantle beneath the Dragon Horn area, as revealed by the recent 2D seismic observations, are considered16. With the existence of this focused melt, the temperature of the deep front of the DF2 could easily reach 650 °C to drive the hydrothermal circulation. Overall, we suggest that the heat source for the Longqi-1 high-temperature hydrothermal circulation would most likely be a melt zone in the lithospheric mantle at a depth of ~13 ± 2 km bsf, which is associated with the deeply seated large-scale multi-stage detachment faulting system (Fig. 4).
Combining insights from the microearthquake activity in the brittle lithosphere, the long fluids-rock reaction path revealed by the complex fluid chemistry and fault geometry, and the numerical modeling of hydrothermal circulation system, we suggest that the hydrothermal circulation along the ultra-slow spreading ridge may reach as deep as ~13 ± 2 km bsf. This is considerably greater than that of the TAG and Logatchev-1 hydrothermal vent fields at the MAR9,10, and approximately 6 km deeper than the seismic Moho boundary in this region (~7 km bsf16). This means that hydrothermal circulation would penetrate into the upper lithospheric mantle. These inferred deeper and longer hydrothermal circulation paths facilitate significant interaction between the seawater and the oceanic lithosphere, thereby affecting the thermal architecture and the composition of the oceanic lithosphere22,50 and the ore-forming capacity of the hydrothermal activity in ultra-slow spreading environments6.
Methods
Microearthquakes location
More than 9 months of passive seismic data were recorded using two kinds of free fall OBS with hydrophone: German Geopro Sedis IV (60 s–50 Hz) and Chinese 1–4C long-period OBS (30 s–50 Hz). These were deployed during the Chinese Cruises DY34th, DY40th, and DY43th. We manually picked 3885 P-phases on the hydrophone channel and 3232 S-phases on the horizontal channels with average picking uncertainties of 0.05 and 0.15 s, respectively, as the input data for 967 events. Hypocenters were located by the least-squares HYPOSAT routine51 on the basis of a one-dimensional P-wave velocity profile derived from the three-dimensional velocity structure of the crust and upper mantle from wide-angle reflection data15. Events with azimuthal station gaps less than 270°, a root-mean square (RMS) travel time residual less than 0.4 s, with at least three P-phases and one S-phase and a free depth inverted solution, were located as the initial hypocenters prior to relocation. The mean horizontal error semi-major and minor axis of the 95% confidence error ellipsoid are 2.19 and 1.79 km, respectively, with a depth error of 2.21 km. Finally, 512 well located events were relocated by the double-difference algorithm. In double-difference theory, the nearest-neighbor events over distance at common stations are much smaller than the length scale of the assumed velocity variation sampled by the seismic waves, and thereby model errors originating from outside the source area are reduced52. Compared with the events located initially by HYPOSAT, the average root-mean square travel-time residual for the relocated events decreased from 0.2 to 0.018 s.
Sampling and analysis of hydrothermal vent fluids
Hydrothermal fluid samples from the sulfide structures at Longqi-1 vent area were collected using pressurized, piston-driven sampling devices, constructed entirely of titanium. These devices make use of compressed nitrogen to maintain each sample at seafloor pressure before and during sampling, and are similar in design and concept to IGT samplers developed in ref. 53. Immediately after arrival of the samplers on board, aliquots were taken for pH, and measurement of the dissolved gases (H2, H2S, CH4) were undertaken. For a detailed description of on-board analytical methods see refs. 28,53. Additional sub-samples were preserved for shore-based analysis of major dissolved anions/cations and trace metals. The trace metal aliquot was immediately acidified with analytical grade HCl (Optima). A fraction of this sub-sample was then diluted 50-fold and utilized for analysis of aqueous silica. Both fractions were stored in pre-weighed and acid cleaned high-density polyethylene bottles. Precipitates that formed in the titanium samplers were collected on a 0.45-μm nylon filter and subsequently re-dissolved in HCl/HNO3 (Ultrex). Based on the total volume of the particular sampler used, the amount of metals measured in the precipitate was recombined with metals that remained in solution to obtain a complete metal inventory for the vent fluid samples. Essentially, this process only affected Fe, Cu, and Zn. For vent fluid samples above 300 °C, the precipitates account for 90% of the reported concentrations of Cu and Zn, but less than 10% of the Fe28. The shore-based analysis of all major dissolved cations/anions was conducted using ICP-OES and/or ion chromatography, while trace metals were determined by ICP-MS. The analytical uncertainties (2σ) is ±2% for Cl (Ion chromatography), ±10% for H2, CH4 (gas chromatography–TCD) and H2S (CuCl2-precipitation/H2O2 reduction). Other elements are measured with ICP-MS, and 2σ uncertainties are ±2% for major species and ±5% for minor species27. Detection limits by ICP-MS are approximately an order of magnitude below reported concentrations. The hydrogen and oxygen isotopes of hydrothermal vent fluids were measured at the University of Arizona using a Thermo Finnigan Delta XP isotope ratio mass spectrometer. All samples were measured at least in duplicate and results are principally given in the standard delta notation in per mil (‰) vs. VSMOW according to δ18O (D) [‰] = (Rsample/Rreference−Rreference) × 1000 fluid samples54. The endmember composition of hydrothermal fluids was calculated using a least-squares regression of the individual components versus Mg with extrapolation to Mg zero and including the respective seawater values. Vent fluid chemistry is summarized in Supplementary Tables 1 and 2.
2D numerical model
Based on the vent temperature and seismic data discussed above, we constructed a 2D numerical model for hydrothermal circulation along a profile across Longqi-1 field and the ridge axis (shown as AA′ in Fig. 1b). The governing equations for convection of seawater and thermodynamic properties calculated from equations in ref. 55, and are solved on an unstructured triangular mesh (see Supplementary Fig. 3) using a hydrothermal model solving for porous flow of water(see ref. 47 for details). The heat source is simulated by a fixed temperature boundary condition at the base. The detailed boundary conditions and mesh structure are shown in Supplementary Fig. 3.
No high-temperature vents and fewer discrete earthquakes are associated with DF1 and thereby, we inferred that DF1 might be inactive. So, here we only consider DF2 as the permeable zone in the 2D numerical model. The precise width of DF2 cannot be inferred from the seismic data, so we conducted a series of calculations with width (d) of DF2 varying from 100 to 1000 m. DF2 is considered as the main pathway of hydrothermal upwelling. There are no data to constrain the precise permeability distribution within DF2; however, the permeability of the discharge zone (kd) can be estimated from the heat output (H), vent temperature (Td = 379 °C), and area (Ad = 8 × 104 m2) of vent field based on the single pass model56,
where cht ≈ 5 × 103 J kg−1 °C−1, αd ≈ 10−3 °C−1, and νd ~ 10−7 m2 s−1 are the specific heat, thermal expansion coefficient, kinematic viscosity of the high-temperature venting fluid, respectively. \(\rho _{f_0}\) = 1000 kg m−3 is the density of fluid at 0 °C; g = 9.8 m s−2 is the gravitational acceleration. The total heat output of high-temperature focused flow is H = 250 ± 100 MW and is estimated by the heat output of hydrothermal plumes detected by the autonomous underwater vehicle data45. Substituting these parameters in Eqs. (1) and (2), we can obtain a mass flow rate of Q = 132 ± 52 kg s−1 and the permeability of the discharge zone kd = 3 × 10−14 ∼ 6 × 10−14 m2, which is also used as the permeability of detachment fault kdf. The background permeability kb = 5 × 10−16 m2 is chosen such that a homogeneous model without detachment faults predicts high-temperature venting24.
Data availability
The geochemical data for the fluids used in this study are reported in the Supplementary Tables, and the seismic data may be requested from the corresponding author Chunhui Tao (taochunhuimail@163.com).
Code availability
The readers can access the open source code on the website as follows:
1. HYPOSAT for seismic location (ftp://ftp.norsar.no/pub/outgoing/johannes/hyposat), for some details about program package see ref. 51
2. Numerical modeling codes are available upon request from the corresponding author Hasenclever et al. (2014).
References
Dick, H. J., Lin, J. & Schouten, H. An ultraslow-spreading class of ocean ridge. Nature 426, 405–412 (2003).
Sauter, D. & Cannat, M. The ultraslow spreading southwest Indian ridge diversity of hydrothermal systems on slow spreading ocean sridges. Geophys. Mon. Ser. 188, 153–173 (2010).
Cannat, M. Emplacement of mantle rocks in the seafloor at mid‐ocean ridges. J. Geophys. Res. Solid Earth 98, 4163–4172 (1993).
Michael, P. et al. Magmatic and amagmatic seafloor generation at the ultraslow-spreading Gakkel ridge, Arctic Ocean. Nature 423, 956 (2003).
Cannat, M., Mével, C. & Stakes, D. Stretching of the deep crust at the slow-spreading Southwest Indian Ridge. Tectonophysics 190, 73–94 (1991).
German, C. R., Petersen, S. & Hannington, M. D. Hydrothermal exploration of mid-ocean ridges: where might the largest sulfide deposits be forming? Chem. Geol. 420, 114–126 (2016).
Stein, C. & Stein, S. A model for the global variation in oceanic depth and heat flow with lithospheric age. Nature 359, 123–129 (1992).
Coogan, L. A., Kasemann, S. A. & Chakraborty, S. Rates of hydrothermal cooling of new oceanic upper crust derived from lithium-geospeedometry. Earth Planet. Sci. Lett. 240, 415–424 (2005).
DeMartin, B. J., Sohn, R. A., Pablo Canales, J. & Humphris, S. E. Kinematics and geometry of active detachment faulting beneath the Trans-Atlantic Geotraverse (TAG) hydrothermal field on the Mid-Atlantic Ridge. Geology 35, 711–714 (2007).
Grevemeyer, I., Reston, T. J. & Moeller, S. Microseismicity of the Mid-Atlantic Ridge at 7°S–8°15′S and at the Logatchev Massif oceanic core complex at 14°40′N–14°50′N. Geochem. Geophy. Geosyst. 14, 3532–3554 (2013).
van der Zwan, F. M. et al. Lower crustal hydrothermal circulation at slow-spreading ridges: evidence from chlorine in Arctic and South Atlantic basalt glasses and melt inclusions. Contrib. Mineral. Petr. 172, 97 (2017).
Tao, C. et al. First active hydrothermal vents on an ultraslow-spreading center: Southwest Indian Ridge. Geology 40, 47–50 (2012).
Cannat, M., Rommevaux Jestin, C., Sauter, D., Deplus, C. & Mendel, V. Formation of the axial relief at the very slow spreading Southwest Indian Ridge (49°to 69°E). J. Geophys. Res. 104, 22825–22843 (1999).
Liao, S. et al. Surface sediment geochemistry and hydrothermal activity indicators in the Dragon Horn area on the Southwest Indian Ridge. Mar. Geol. 398, 22–34 (2018).
Zhao, M. et al. Three-dimensional seismic structure of the Dragon Flag oceanic core complex at the ultraslow spreading Southwest Indian Ridge (49°39′E). Geochem. Geophys. Geosy. 14, 4544–4563 (2013).
Li, J. B. et al. Seismic observation of an extremely magmatic accretion at the ultraslow spreading Southwest Indian Ridge. Geophys. Res. Lett. 42, 2656–2663 (2015).
Weekly, R. T. et al. Termination of a 6 year ridge-spreading event observed using a seafloor seismic network on the Endeavour Segment, Juan de Fuca Ridge. Geochem. Geophys. Geosyst. 14, 1375–1398 (2013).
Parnell-Turner, R. et al. Oceanic detachment faults generate compression in extension. Geology 45, 923–926 (2017).
Horning, G., Sohn, R. A., Canales, J. P. & Dunn, R. A. Local seismicity of the rainbow Massif on the Mid-Atlantic Ridge. J. Geophys. Res. Solid Earth 123, 1615–1630 (2018).
Korger, E. I. M. & Schlindwein, V. Seismicity and structure of the 85°E volcanic complex at the ultraslow spreading Gakkel Ridge from local earthquake tomography. Geophys. J. Int. 196, 539–551 (2013).
Grevemeyer, I. et al. Constraining the maximum depth of brittle deformation at slow- and ultraslow-spreading ridges using microseismicity. Geology 47, 1069–1073 (2019).
Schlindwein, V. & Schmid, F. Mid-ocean-ridge seismicity reveals extreme types of ocean lithosphere. Nature 535, 276–279 (2016).
Buck, W. R. Flexural rotation of normal faults. Tectonics 7, 959–973 (1988).
Andersen, C., Rüpke, L., Hasenclever, J., Grevemeyer, I. & Petersen, S. Fault geometry and permeability contrast control vent temperatures at the Logatchev 1 hydrothermal field, Mid-Atlantic Ridge. Geology 43, 51–54 (2015).
Janecky, D. R. & Jr, W. E. S. Hydrothermal serpentinization of peridotite within the oceanic crust: experimental investigations of mineralogy and major element chemistry. Geochim. Cosmochim. Acta 50, 1357–1378 (1986).
Seyfried, W. E. Jr & Dibble, W. E. Seawater-peridotite interaction at 300 °C and 500 bars: implications for the origin of oceanic serpentinites. Geochim. Cosmochim. Acta 44, 309–321 (1980).
Seyfried, W. E. Jr., Foustoukos, D. I. & Fu, Q. Redox evolution and mass transfer during serpentinization: an experimental and theoretical study at 200 degrees C, 500 bar with implications for ultramafic-hosted hydrothermal systems at mid-ocean ridges. Geochim. Cosmochim. Acta 71, 3872–3886 (2007).
Seyfried, W. E., Seewald, J. S., Berndt, M. E., Ding, K. & Foustoukos, D. I. Chemistry of hydrothermal vent fluids from the Main Endeavour Field, northern Juan de Fuca Ridge: geochemical controls in the aftermath of June 1999 seismic events. J. Geophys. Res. Solid Earth 108, 2429–2452 (2003).
Schmidt, K. et al. Fluid elemental and stable isotope composition of the Nibelungen hydrothermal field (8°18′S, Mid-Atlantic Ridge): constraints on fluid–rock interaction in heterogeneous lithosphere. Chem. Geol. 280, 1–18 (2011).
Pester, N. J. et al. Subseafloor phase equilibria in high-temperature hydrothermal fluids of the Lucky Strike Seamount (Mid-Atlantic Ridge, 37°17′N). Geochim. Cosmochim. Acta 90, 303–322 (2012).
Seyfried, W. E. J. & Ding, K. Phase euilibria in subseafloor hydrothermal systems: a review of role of redox, temperature, pH and dissolved Cl on the chemistry of hot spring fluids at mid-ocean ridges, seafloor hydrothermal systems: physical, chemical, biological, and geological interactions. Am. Geophys. Union 91, 248–272 (1995).
Bach, W. & Humphris, S. E. Relationship between the Sr and O isotope compositions of hydrothermal fluids and the spreading and magma-supply rates at oceanic spreading centers. Geology 27, 1067–1070 (1999).
Kelley, D. S. Methane-rich fluids in the oceanic crust. J. Geophys. Res. Solid Earth 101, 2943–2962 (1996).
Kelley, D. S. Fluid evolution in slow-spreading environments. Proc. Ocean Drill. Program Sci. Results 153, 399–415 (1997).
Kelley, D. S. & Früh-Green, G. L. Abiogenic methane in deep‐seated mid‐ocean ridge environments: insights from stable isotope analyses. J. Geophys. Res. Solid Earth 104, 10439–10460 (1999).
Proskurowski, G. et al. Abiogenic hydrocarbon production at lost city hydrothermal field. Science 319, 604–607 (2008).
Mcdermott, J. M., Seewald, J. S., German, C. R. & Sylva, S. P. Pathways for abiotic organic synthesis at submarine hydrothermal fields. Proc. Natl Acad. Sci. USA 112, 7668–7672 (2015).
Schmidt, K., Koschinsky, A., Garbe-Schönberg, D., Carvalho, L. M. D. & Seifert, R. Geochemistry of hydrothermal fluids from the ultramafic-hosted Logatchev hydrothermal field, 15°N on the Mid-Atlantic Ridge: temporal and spatial investigation. Chem. Geol. 242, 1–21 (2007).
Shanks, W. C., Böhlke, J. K. & Seal, R. R. Stable isotopes in mid-ocean ridge hydrothermal systems: interactions between fluids, minerals, and organisms. Am. Geophys. Union 91, 194–221 (1995).
Shanks, W. C. Stable Isotopes in seafloor hydrothermal systems: vent fluids, hydrothermal deposits, hydrothermal alteration, and microbial processes. Rev. Miner. Geochem. 43, 469–525 (2001).
Mccaig, A. M., Delacour, A., Fallick, A. E., Castelain, T. & Frühgreen, G. L. Detachment fault control on hydrothermal circulation systems: interpreting the subsurface beneath the TAG hydrothermal field using the isotopic and geological evolution of oceanic core complexes in the Atlantic. Geophys. Monogr. Ser. 188, 207–239 (2010).
Mccaig, A. M., Cliff, R. A., Escartin, J., Fallick, A. E. & Macleod, C. J. Oceanic detachment faults focus very large volumes of black smoker fluids. Geology 35, 935–938 (2007).
Edmonds, H. N. et al. Continuation of the hydrothermal fluid chemistry time series at TAG, and the effects of ODP drilling. Geophys. Res. Lett. 23, 3487–3489 (1996).
Kumagai, H. et al. Geological background of the Kairei and Edmond hydrothermal fields along the Central Indian Ridge: implications of their vent fluids’ distinct chemistry. Geofluids 8, 239–251 (2008).
Liu, W. Y. Research of Characteristics of Plume and Heat Flux Output of Hydrothermal Field at the Ultraslow Spreading Southwest Indian Ridge. Master Dissertation (in Chinese) (2010).
Lowell, R. P. Hydrothermal circulation at slow spreading ridges: analysis of heat sources and heat transfer processes. Diversity of hydrothermal systems on slow spreading ocean ridges. Geoophys. Monogr. Ser. 188, 11–26 (2010).
Hasenclever, J. et al. Hybrid shallow on-axis and deep off-axis hydrothermal circulation at fast-spreading ridges. Nature 508, 508–512 (2014).
McKenzie, D. K., Jackson, J. & Priestley, K. Thermal structure of oceanic and continental lithosphere. Earth Planet. Sci. Lett. 233, 337–349 (2005).
Schlindwein, V., Demuth, A., Geissler, W. H. & Jokat, W. Seismic gap beneath Logachev Seamount: indicator for melt focusing at an ultraslow mid-ocean ridge? Geophys. Res. Lett. 40, 1703–1707 (2013).
Staudigel, H., Plank, T., White, B. & Schmincke, H. U. Geochemical fluxes during seafloor alteration of the basaltic upper oceanic crust: DSDP Sites 417 and 418. Subduction: Top to Bottom Vol. 96, 19–38 (American Geophysical Union, 1996).
Schweitzer, J. HYPOSAT—an enhanced routine to locate seismic events. Pure Appl. Geophys. 158, 277–289 (2001).
Waldhauser, F. & Ellsworth, W. L. A double-difference earthquake location algorithm: method and application to the northern Hayward fault, California. Bull. Seismol. Soc. Am. 90, 1353–1368 (2000).
Seewald, J. S., Doherty, K. W., Hammar, T. R. & Liberatore, S. P. A new gas-tight isobaric sampler for hydrothermal fluids. Deep Sea Res. I 49, 189–196 (2002).
Seyfried, W. E. Jr. & Mottl, M. J. Hydrothermal alteration of basalt by seawater under seawater-dominated conditions. Geochim. Cosmochim. Acta 46, 985–1002 (1982).
Driesner, T. The system H2O–NaCl. Part II: Correlations for molar volume, enthalpy, and isobaric heat capacity from 0 to 1000°C, 1 to 5000bar, and 0 to 1X NaCl. Geochim. Cosmochim. Acta 71, 4902–4919 (2007).
Lowell, R. P., Farough, A., Hoover, J. & Cummings, K. Characteristics of magma-driven hydrothermal systems at oceanic spreading centers. Geochem. Geophy. Geosyst. 14, 1756–1770 (2013).
Von Damm, K. L. et al. Chemistry of submarine hydrothermal solutions at 21°N, East Pacific Rise. Geochim. Cosmochim. Acta 49, 2197–2220 (1985).
Spivack, A. J. & Edmond, J. M. Boron isotope exchange between seawater and the oceanic crust. Geochim. Cosmochim. Acta 51, 1033–1043 (1987).
Von Damm, K. L. Chemistry of hydrothermal vent fluids from 9°–10°N, East Pacific Rise: Time zero the immediate posteruptive period. J. Geophys. Res. Solid Earth 105, 11203–11222 (2000).
Von Damm, K. L. & Lilley, M. D. Diffuse flow hydrothermal fluids from 9°50’N East Pacific Rise: origin, evolution and biogeochemical controls. Geophys. Monogr. Ser. 144 (2004).
Pester, N. J., Rough, M., Ding, K. & Seyfried, W. E. J. A new Fe/Mn geothermometer for hydrothermal systems: Implications for high-salinity fluids at 13°N on the East Pacific Rise. Geochim. Cosmochim. Acta 75, 7881–7892 (2011).
Butterfield, D. A., Massoth, G. J., Mcduff, R. E., Lupton, J. E. & Lilley, M. D. Geochemistry of hydrothermal fluids from axial Seamount hydrothermal emissions study vent field, Juan de Fuca Ridge: subseafloor boiling and subsequent fluid‐rock interaction. J. Geophys. Res. 95, 12895–12921 (1990).
Von Damm, K. L. & Bischoff, J. L. Chemistry of hydrothermal solutions from the southern Juan de Fuca Ridge. J. Geophys. Res. 92, 11334–11346 (1987).
Butterfield, D. A. & Massoth, G. J. Geochemistry of north cleft segment vent fluids: temporal changes in chlorinity and their possible relation to recent volcanism. J. Geophys. Res. Atmos. 99, 4951–4968 (1994).
Edmond, J. M., Campbell, A. C., Palmer, M. R. & German, C. R. Geochemistry of hydrothermal fluids from the Mid-Atlantic Ridge: TAG and MARK 1990. Am. Geophys. Union 71, 1650–1651 (1990).
Charlou, J. L., Donval, J. P., Jean-Baptiste, P., Dapoigny, A. & Rona, P. A. Gases and helium isotopes in high temperature solutions sampled before and after ODP Leg 158 drilling at TAG Hydrothermal Field (26°N, MAR). Geophys. Res. Lett. 23, 3491–3494 (1996).
Douville, E. et al. The rainbow vent fluids (36°14′N, MAR): the influence of ultramafic rocks and phase separation on trace metal content in Mid-Atlantic Ridge hydrothermal fluids. Chem. Geol. 184, 37–48 (2002).
Gallant, R. M. & Damm, K. L. V. Geochemical controls on hydrothermal fluids from the Kairei and Edmond Vent Fields, 23°–25°S, Central Indian Ridge. Geochem. Geophys. Geosyst. 7, 1–4 (2006).
Acknowledgements
We thank Science Party, captains and crew of COMRA Cruise DY35th, DY40th, and DY43th, the submersible Jiaolong group and the 4500 m Qianlong-II Autonomous Underwater Vehicle team. We thank Lars H Rüpke and Jörg Hasenclever for the numerical modeling code and Fernando J.A.S. Barriga, Maurice A. Tivey, and Honglei Shen for polishing English. We thank F. van der Zwan for her fruitful comments. This work was supported by National Key R&D Program of China under contract no. 2018YFC0309901, 2017YFC0306603, 2017YFC0306803, and 2017YFC0306203, COMRA Major Project under contract No. DY135-S1-01-01 and No. DY135-S1-01-06. All the co-authors give their deep appreciation to R.P. Lowell, who passed away during the revision of this work. His contributions to this paper, however, were seminal, in the absence of which key findings and implications of the study would not have been possible.
Author information
Authors and Affiliations
Contributions
C.T. designed and led the whole project. W.E.S. and K.D. finished analyses and explanation of vent fluid chemistry. R.P.L. and Z.G. finished the numerical simulation of hydrothermal circulation. Y.L., L.Q., and H.W. hosted OBS experiments and processed the seismic data. J. Liang. and W.C. participated in the sampling of the vent fluid at Jiaolong human-occupied vehicle. H. Zhang, I.E., and G.Z. outlined structure line from the bathymetry data. S.L. and H.L. participated in sampling using TV-grab and geology interpretation. M.Z. provided the three-dimensional velocity model acquired by seismic refraction experiments. J. Liu and J. Lin provided the advice of the geological interpretation. J.Z., X.D., H. Zhou, and W.L. participated in data collection on board. All authors contributed the interpretation of the results and participated in the review of this manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Peer review information Nature Communications thanks the anonymous reviewers for their contribution to the peer review report of this work. Peer review reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Tao, C., Seyfried, W.E., Lowell, R.P. et al. Deep high-temperature hydrothermal circulation in a detachment faulting system on the ultra-slow spreading ridge. Nat Commun 11, 1300 (2020). https://doi.org/10.1038/s41467-020-15062-w
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-020-15062-w
This article is cited by
-
Microearthquake reveals the lithospheric structure at mid-ocean ridges and oceanic transform faults
Journal of Oceanology and Limnology (2024)
-
Microseismicity and lithosphere thickness at a nearly-amagmatic oceanic detachment fault system
Nature Communications (2023)
-
High-precision double-spike Cd isotopic measurements of seawater by MC-ICP-MS and its application to seawater affected by hydrothermal vent fluids
Acta Geochimica (2023)
-
Record of hydrothermal activity in the Yuhuang hydrothermal field and its implications for the Southwest Indian Ridge: evidence from sulfide chronology
Acta Oceanologica Sinica (2023)
-
Mineralogy and trace element geochemistry of hydrothermal sulfides from the Ari vent field, Central Indian Ridge
Mineralium Deposita (2023)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.