Post on 15-Jul-2020
The Tasmantid Seamounts: A window into structural inheritance of ocean floor fabricFred Richards1, 2*, Lara Kalnins3, Tony Watts1, Benjamin Cohen4 & Robin Beaman5
1Department of Earth Sciences, University of Oxford, Oxford, UK 2 Bullard Laboratories, Department of Earth Sciences, University of Cambridge, Cambridge, UK 3Department of Earth Sciences, Durham University, UK 4Scottish Universities Environmental Research Centre, East Kilbride, UK 5College of Science, Technology and Engineering, James Cook University, Cairns, Australia *fdr22@cam.ac.uk
V21A-3034
Introduction The Tasmantid Seamounts extend for >2000 km off the east coast of Australia, constituting one of three contemporaneous, sub-parallel Cenozoic hotspot tracks that traverse the region, locally separated by ≤500 km (Cohen et al., 2013). The sea-mounts young north to south, spanning ~50 - 6 Ma (McDougall & Duncan 1988; Kalnins et al., 2015).
At multiple locations, the Tasmantid chain intersects the extinct slow-spreading Tasman Sea ridge system, which was active between 84 - 53 Ma (Muller et al., 2008). Despite the >20 Ma hiatus between spreading cessation and seamount emplacement, palaeo-ridge structure appears to exert significant control on sea-mount morphology and structure.
Geophysical datasets acquired on voyage ss2012_v07 have been analysed to answer the following:
1. What mechanism can account for the morphologi-cal diversity of the seamounts?
2. What is the relationship between pre-existing tec-tonic fabric, intraplate magmatism and the strength of the Tasman Sea lithosphere?
3. How has the strength of the Tasmantid melting anomaly evolved through time?
152
152
15
15
1 0
1 0
−2 −2
−2 −2
−24 −24
−22 −22
−20 −20
−1 −1
−1 −1
− 000 −5000 −4000 −3000 −2000 −1000 0 1000Bathymetry (m)
MAINSTUDYREGION
NovaBank
ArgoBank
Tweed
Brisbane Seamounts
Moreton & MooloolabaSeamounts
Recorder Guyots
FraserGuyots
CatoReef
WreckReefs
Britannia Guyots
QueenslandGuyot
Stradbroke Seamount
TASM
AN
TID
SEA
MO
UN
T CH
AIN
LORD
HO
WE
SEAM
OU
NT
CHA
IN
CORAL SEA
MainRange
Flinders
Glasshouse - Maleny
Noosa
Fraser
EAST
AU
STRA
LIAN
VO
LCAN
IC C
HAI
N
LORDHOWE
RISE
Gi�ord Guyot
Capel Bank
KelsoBank
KennReefs
FrederickReef
Mellish Plateau
MarionPlateau
Dampier Ridge
Kenn Plateau
Middleton Basin
Cato Trough
TASMAN SEA
BRISBANE
Chester�eldReefs
AUSTRALIA
BellonaReefs
MellishReef
Bunya
Toowoomba
Figure 1. Map of the study area. Dotted red line = shiptrack of voyage ss2012_v07. Star = position of Cenozoic seamounts and volcanoes. Solid red line = extinct ridge axis, solid black line = transform fault, dashed black = fracture zone. Bathymetry derived from GBR100 dataset (Beaman 2010).
1 Volcanic Architecture• Tasmantid morphologies fall
into four distinct categories: a) conical seamounts char-acterised by steep, smooth flanks; b) elongate and stel-late terraced seamounts with subaerially eroded peaks; c) rugged seamounts con-structed via repeated fissure eruptions along crosscutting volcanic rift zones (VRZs) and d) shield seamounts with shallow slopes and dispersed cinder cones.
• The chain exhibits low mass wasting rates and fluctuating edifice volume, with dramat-ic morphological variation between seamounts separated by ≤10 km.
• There is a clear link between overall morphology and sea-mount position relative to the extinct spreading centre.
154 154 9 155 0 155 1 155 2
−25 4
−25 3
−25 2
−25 1
−25 0
−24 9
−24
Perspective view
Recorder F
Z
−4500 −3500 −2500 −1500 −500Bathymetry (m)
3 identifiable summit benches
Hummocky debris deposits with pelagic
sediment drape
Rugged flanks with narrow embayments
Slumped debris deposits coat terraces
−25 4
155 7 155 155 9 15 0 15 1 15 2
−29 1
−29 0
Perspective view
2 main volcanic centres
Steep gullied flanks
ENE -WSW trending volcanic ridge
Pillow basalts and talus deposits
Flank crater−3000–4500 –1500
Bathymetry (m)
155 3 155 4
−24 2
−24 1
−24 0
Perspective view
Carbonate-capped, flat summit
Flank cone
Hummocky debris deposit
Shallow headscars above narrow, steep embayments
−4000−3000−2000−1000Bathymetry (m)
Steep, fluted flanks
155 0 155 1 155 2 155 3 155 4 155 5 155 155 7 155
−23 5
−23 4
−23 3
−23 2
−23 1
−23 0
West
Bellona FZ
Perspective view
Cato
FZ
KENN PLATEAU
CATOTROUGH
Volcanic cone field
Hummocky debris field with blocks up to 800m across
Fissure-fed volcanic ridges
Bathymetry (m)−3000 −2000 −1000 0
Gentle slope gradients on E side toward Lord
Howe Rise
Small satellite guyot
Steeper slopes on W side into
Cato Trough
2ba
c
d
Figure 2. (a) N Fraser: conical seamount emplaced off-axis, adjacent to the ocean-con-tinent boundary (OCB). (b) N & S Recorder: both seamounts have terraced upper slopes. N Recorder is off-axis and stellate; S Recorder is emplaced on a fracture zone and is elongated. (c) Stradbroke: rugged seamount straddling an inside corner. (d) Cato: shield seamount located on the Lord Howe Rise microcontinent. Vertical exaggeration is x4.5. N.B. Bathymetry colour scales are different for each seamount.
Figure 3. Average slope gradient by sector. (a) South Moreton, a con-ical seamount. (b) Cato, a shield seamount. (c) Britannia, a terraced seamount: (i) basal and (ii) upper, subaerially-eroded slopes. (d) Strad-broke, a rugged seamount.
S 5
EW
N
S 5
EW
N
S 5
EW
N
S 5
EW
N
S 5
EW
N
S
EW
N
Slope Analysis• Tectonic setting exerts major control on slope characteristics: conical seamounts with elevated slope gradients but lower inter-sector variance occur off-axis and
at outside corners; rugged seamounts with low slope gradient but high inter-sector variability occur at inside corners. • Terraced and shield seamounts have notably variable, low gradient upper slopes suggesting elevated rates of mass wasting due to sub-aerial exposure.• The relatively low slope variance and elevated backscatter on the lower sections of the edifices indicate that large mass-wasting events are rare. This is consistent
with minimal shallow deformation and may reflect modest eruption rate with a high intrusive-to-extrusive magmatic budget (Ramalho et al., 2013).
Figure 4. Detecting mass wasting. (a) SE-ward view of S Moreton, red dashes = mass wasting deposit with reduced mean slope gradient (b) and low backscatter (e). Slope gradient is fairly constant (c) so bathymetric curvature (e) is ~ 0. The run-out distance:headscar aspect ratio suggests a debris avalanche mechanism (Mitchell et al., 2002).
3
b ba a
cc e
dd
i
ii
Structural Orientations4
Figure 7. Effect of cross-transform mechanical coupling on principal stress orientations (modified after Behn et al., 2002). Focal mechanisms show optimal, calculated fault plane solutions at 4 km depth for differing degrees of mechanical coupling. χ = mechanical coupling, OC = inside cor-ner, IC = outside corner, red = ridge axis, black = transform fault. τdev/σdev+lith is the ratio of shear to normal stress on the fault plane incorporating both deviatoric and lithostatic components. (a) χ = 0, (b) χ=0.05, (c) χ = 0.1 & (d) χ = 0.15. Horizontal displacement = 100m (10 ka at half-spreading rate of 1 cm a-1).
Figure 5. Linear feature orientation vs. tectonic setting. (a) Stradbroke: (i) trends oblique-to-spreading; (ii) inside-corner setting. (b) S & N Recorder: (i) S Recorder: volcanic trends parallel to Recorder Fracture Zone and (iii) fracture zone setting (ii) N Re-corder has ridge-parallel features and (iii) off-axis setting. (c) N Fraser: (i) trends ridge-parallel or subparallel to fracture zones; (ii) ocean-continent boundary setting. Red = major VRZs, blue = minor VRZs and black = faults. IC = inside corner, OC = outside corner.
• Alignments suggest deep faulting of the oceanic lithosphere allowing channelisation of magma along pre-existing structures despite emplacement post-dating active extension by >20 Ma. • Dominance of the tectonic signal points to low melt production, implying that the Tasmantid “plume” was a relatively weak melting anomaly.• Seamounts located at or adjacent to ridge-transform intersections have dominant trends oblique to spreading, implying strong mechanical coupling across the transforms.• The high degree of mechanical coupling is consistent with slow Tasman Sea spreading rates and low rates of melt production.
0.15
0
0.1
0.05
dev /dev+litho
125
100
75
50
25
Y (k
m)
OC IC
50 100X (km)
50 100X (km)
50 100X (km)
50 100X (km)
INCREASING CROSS-TRANSFORM MECHANICAL COUPLING
500 km100 km
Ridge Axis & FZs
25 km
−
25 km
Fracture Zone
25 km
Inside Corner
25 km
Outside Corner
25 km
Continental
S
EW
N
S
EW
N
S
EW
N
S
EW
N
S
EW
N
S
EW
N
10 km
10 km 10 km
10 kmS
EW
N
S
EW
N
S
EW
N
S
EW
N
CATO FZ
RECORDER FZ
IC
ICOC
IC
OC
OC
IC
IC
ICOC
OC
10 km 10 km 10 km
OCIC
b
ba
a c
c d
iii
ii iiii
d f
Figure 6. Linear feature orientations of all seamounts. (a) Axial ridges (red) and fracture zones (black). (b) Off-axis sea-mounts. (c) Fracture zone seamounts. (d) Inside corner seamounts. (e) Outside corner seamounts. (f) Continental seamounts.
ba c
e
iii
Deep Structure • Bouguer anomalies are calculated by removing the combined gravity effect of the water-sediment and
and sediment-seamount interfaces to investigate gravity signals related to deeper density interfaces.• Calculations assume that seamount topography is uncompensated, i.e. the lithosphere is infinitely rigid.• The reduction density of a seamount refers to the input seamount density that minimises spectral
coherence between calculated Bouguer anomalies and the bathymetry.• 20-50 mGal Bouguer highs over many seamounts, coupled with reduction densities up to ~3100 kg m-3,
suggest extensive intra-basement intrusion of primary magmas (cf. Contreras-Reyes et al., 2010).• Low reduction densities over Cato and Wreck are consistent with both seamounts being emplaced on
continental basement and significant fractionation of parent magmas (Hammer et al. 1990).
5
−22 2
720000 750000 7 0000
−3000 −2000 −1000 0
i ii
7530000
75 0000
720000 750000 7 0000
−200 −100 0 100 200
i ii
155 1 155 4 155 7
−22 2
−200 −100 0 100 200−
i ii
155 1 155 4 155 7
7530000
75 0000
052522002571
i ii
−300
−200
−100
0
100
200
300
400
500
700000 725000 750000 775000
−5000
−4000
−3000
−2000
−1000
0
1000
i ii 0 0
0 1
0 2
0 3
0 4
0 5
0
0 7
0
0 9
1 0
102 103 104 105
ρ
2400 −3
2 00 −3
2 00 −3
3000 −3
3200 −3
230025002700290031003300
−3
−3
−3
−3
−3
−3 2500 −3
−
Figure 8. Gravity reduction over Wreck seamount. (a) Bathymetry. (b) Combined basement-seafloor inter-face topographic anoma-ly (sediment density (ρsed) = 2000 kg m-3; seamount density (ρsea) = 2500 kg m-3). (c) Free-air anomaly. (d) Bouguer anomaly. (e) Gravity anomaly and to-pography comparison. (f) Coherence between Bouguer anomaly and bathymetry for range of ρsea (2300 - 3300 kg m-3).
Seamount Reduction Density (kg m-3)
Mean of Error(kg m-3)
Stradbroke 2760 12.5
Queensland& Britannia 2715 15.0
North & South Brisbane 3050 35.0
South Moreton 3135 30.0
North Moreton 2800 7.5
North & SouthRecorder 2805 37.5
South Fraser 2860 15.0
North Fraser 2825 60.0
Cato 2510 37.5
Wreck 2500 42.5
Table 1. Reduction density (i.e. seamount density, ρsea, that minimises weighted average coherence between gravity and bathymetry) for each of the Tasmantids with associated mean of error. Coherence and associat-ed error are weighted using the power spectrum of the bathymetry.
a b
c d
e f
Gravity Modelling to Determine Lithospheric Strength• Models assume a finite-strength lithosphere and that the seamounts constitute loads, potentially associated with crustal roots and flexural moats. • Inclusion of large sediment loads and extensive intra-basement intrusion of mafic magmas is required to achieve acceptable fits to observed gravity anomalies.• There is no observable elastic thickness vs. age relationship along the chain.• Seamounts emplaced at inside corners, fracture zones and on highly-extended continental basement have the lowest elastic thicknesses suggesting that structural
inheritance, not age, is the dominant control on lithospheric strength.
−500 −400 −300 −200 −100 0 100 200 300 400 500−40
−35
−30
−25
−20
−15
−10
−5
0
Distance (km)
Dep
th (k
m)
Water DepthBasementTTSBackstrip MohoFlexed Moho
Seamount InfillDense Core
2900
2700
3000
2900 2700
2000
2000
2000
1030
2500 2500
3300
−200
−100
0
100
200
Gra
vity
Ano
mal
y (m
Gal
)
Satellite FAAModelled FAAShipboard FAA
Figure 9. (a) Te vs. age of crust at time of loading with predicted isotherms for half-space plate cooling model. Data are coloured by tectonic context. (b) Best-fitting crustal model for Stradbrokeseamount from process-ori-ented gravity and flexure modelling. Values indicate densities (kg m-3)). TTS = Total Tectonic Subsidence.
0
10
20
30
Ela
stic T
hic
kness (
km
)
0 10 20 30 40 50 60
Age Differential (tseafloor − tload (Ma))
0
10
20
30
0 10 20 30 40 50 60
150°
300°
450°
600°
900°
Off−Axis
Fracture Zone
Inside Corner
Ocean−Continent
Continental
6
a b
7 Conclusions1. The modest rates of mass wasting revealed by slope analysis, combined with the prevalence of dense cores indicated by gravity signatures and lithospheric
modelling, suggest that subsurface intrusion, rather than sub-aqueous eruption, was the dominant magmatic growth mechanism.
2. Low overall Te and the >20 Ma time separation between seamount emplacement and spreading cessation suggest deep intra-lithospheric faulting must have accompanied spreading in order to allow Tasmantid magmas to exploit and align with pre-existing structural weaknesses.
3. Slow rate of magma supply, as indicated by the dominance of tectonic controls, high intrusive:extrusive ratios and scarcity of large mass-wasting deposits, points to a relatively weak Tasmantid melting anomaly.
4. Structural inheritance dominates the magmatic evolution of the Tasmantids as demonstrated by: i) dependence of morphology on tectonic setting; ii) lack of a Te-age relationship and iii) strong alignment of volcanic features with principal stress directions predicted for the Tasman Sea ridge system.
5. The strong dependence of intraplate magmatic fabric on extinct ridge structure demonstrates the importance of understanding tectonic inheritance and its influence on magmatic systems in both continental and oceanic settings.
References: Beaman, 2010, MTSRF Project 2.5i.1a Final Report, pp. 13 & Appendix 1; Behn et al., 2002, Geophys. Res. Lett., 29(24), 2207; Cohen et al., 2013, Tectonics, 32, pp.1371–1383. Contreras-Reyes et al., 2010, EPSL, 289(3-4), pp. 323-333. Hammer et al., 1990, Geophysics, 56(1), pp. 68–79. Kalnins et al., 2015, DI41A-2591, AGU Fall Meeting 2015. McDougall & Duncan (1988), Earth Plan. Sci. Lett., 89, pp. 207-220; Mitchell et al., 2002, Journ. Volc. & Geo. Res., 155(1-2), pp. 82–107; Müller et al., 2008, G3,9, Q04006; Ramalho et al., 2013, Earth Sci. Rev., 127, pp.140–170; Watts & Zhong, 2000, Geophys. J. Int., 142, pp. 855-875.