The authors describe a tectonic scenario for the Ross Sea region for the last 100 Myrs. They propose three phases: 1) An extensional phase 105-80 Ma during which the four major depocenters (basins) were formed., 2) an extensional phase 50-30 Ma. with localized basin subsidence, magmatism, and the formation of the Transantarctic Mtns. (TAM), and 3) a transtensional phase 30 Ma - present with right lateral strike-slip motion along NW-SE faults that extend from the central Ross Sea to the TAM.
Evidence for the last phase is based largely on marine seismic sections. Faults that are active in the last phase are determined by whether they pierce the RSU6 unconformity, which is poorly estimated to be 30-26 Ma and is widespread across the Ross Sea. The main evidence for net offset is the 25 km displacement of the gravity-inferred Moho high, by the Lanterman-Avaitor faults. Extrapolating this estimate to all the major NE-SW faults gives 200 km of strike slip displacement. Vertical displacements appear to not exceed 1-2 km.
The authors also infer the current rheologic state of the lithosphere by the apparent intensity of phase 3 faulting. The assumption is that dispersed smaller faults indicate a ductile lithosphere while focused in larger faults are indicative of a more brittle lithosphere. Under this assumption, the lithosphere under the central and eastern Ross Sea is ductile, while central to western Ross Sea grades to increasingly more brittle.
Discussion: The displacement the faults in phase 3 seem to be quite small, making this a relatively minor component of the overall tectonics. The sedimentary thickness and style of faulting (focused in a few larger faults) does not appear to agree with the Adare Trough section which has thin sediments and almost no apparent faulting except for the Trough itself. This may indicate significant activity on strike-slip faults on the northern edge of the region covered by this paper.
The paper presents kinematic observations of brittle faults (strike, dip, offset, and striation direction) along the boundary between the Transantarctic Mountains and the Ross Sea. The study area covers ~50 km along the shoreline and more westerly outcrops between Granite Harbour and Marble Point on the W side of McMurdo Sound. The faults are exposed mainly in Proterozoic to early Paleozoic basement rocks. The author distinguishes three fault arrays: 1) NNE faults (~020), usually moderately to steeply E-dipping with normal to slightly oblique displacement sometimes overprinted by a shallower rake displacement; 2) NE faults (~040), dipping moderately SE to moderately NW and often subvertical,with variable rakes of striae, again with shallower striae on vertical planes as the later event ; 3) ENE faults, seen mostly at Granite Harbour, with moderate N or S dips.
The timing of motion on these faults is constrained as follows: the fission track isochrons are offset by the NNE and NE striking faults. It is assumed that the fission track isochrons record uplift of the mountain range. Thus, these faults represent structural events that post-date the ca 55 to 60 Ma start of uplift of the mountain range. However it appears that the NNE set of faults formed during continued Cenozoic uplift; the NE faults are interpreted as reactivated Paleozoic faults which are parallel to Pz dike trends. The ENE faults are interpreted to have first formed in the Jurassic but to have been reactivated during Cenozoic time as the TAM uplifted.
The main conclusion is that none of these fault sets are parallel to the NNW trend of the edge of the TransAntarctic mountains here; thus the author concludes that the TAM front, which is a huge topographic escarpment, is not likely to be controlled by a large normal fault formed in orthogonal extension. Many previous authors had described the TAM as a rift-flank uplift formed by orthogonal normal faulting. Wilson proposes that in fact the TAM flank was formed by oblique (dextral transtensional) rifting. By comparison with laboratory models of structures formed during oblique rifting, she proposes the following history to fit the observations: an early period of rifting with the displacement direction 30 to 45 degrees dextrally oblique to the margin , followed by a later phase of rifting with the displacement direction 15 degrees dextrally oblique from the rift trend.
The author further concludes that, if the crust in the Ross Sea indeed was thinned due to orthogonal extension perpendicular to the TAM front, this event must predate the transtension she documents, and therefore must be pre-Cenozoic. The author relates the orthogonal extension to major block motions during the late Cretaceous breakup of this sector of Gondwana. Thus she concludes that the Cenozoic faults that she studied cannot have produced much extension in this region in the Cenozoic; and therefore that plate motion models requiring such extension really don't work.
Discussion: 1) We have observations of 150 to 200 km of Tertiary extension
in the Adare trough; how does that fit into this picture? It seems incompatible.
2) These data are all from a 50-km stretch of the 1000-km-long transantarctic
mountain front. If the front is structurally segmented (into zones of opposing
polarity of normal faults, as is known from the Gulf of California, for
example) then these results might all be from one segment and may not actually
be representative of the entire mountain front. Perhaps studies elsewhere
along strike would reveal large normal faults parallel to the TAM front
in other sectors of the rift system. 3) There was considerable discussion
of Figure 7 and what it actually meant- we went back to look at the figures
in Withjack and Jamison's paper to clarify the diagrams. 4) Salvini et
al propose a 3 stage history of rifting, with one phase going from 50 Ma
to 30 Ma. Wilson refers to uplift from 60 Ma to 55 Ma. Which of Salvini
et al's episodes does this correspond to? Probably the 50 Ma to 30 Ma one.
We concluded that Wilson's numbers are based on the fission track ages
and Salvini et al.'s numbers are based on the seismic stratigraphy, which
lacks exact age control. In future meetings we will read papers about both
the seismic stratigraphic details and the fission track results.
In the first paper, a faulted elastic plate model is used to explain three NW striking features in Victoria Land: the Wilkes Basin (WB), the Transantarctic Mtns (TAM), and the Victoria Land Basin (VLB). The TAM is modeled as a tilted uplift due to extensional forces acting on a 62 degree normal fault (Vening rift model), thermal uplift forces in the athenoshpere, and erosional unloading on th eastern flank of the TAM. The WB is an elastic depression that is a balance of the ice load and bending moment of the elastic plate. Its distance from the uplift constrains the effective elastic thickness of the plate to be 115 km in East Antarctica. The Victoria Land Basin is the corresponding downwarp with the gravity high in the middle of the basin modelled as the flexural bulge. The distance to the bulge constrains the elastic plate thickness to be 19 km in West Antarctica. The two differing thicknesses give rise to thermal ages of 600 Ma for East Antarctica and 25 Ma for West Antarctica.
Two alternative model are discussed. A 30 km slab of eclogite at the crust-mantle boundary is dismissed as being as unlikely coincidence. Low-angle/lithospheric thinning (a la Basin and Range) dose not appear to produce enough uplift for fit the TAM.
The second paper expands on some of the points and calculations made in the first paper. The Wilkes Basin gravity anomaly is now observed to be < -50 mgals in some places. A refined estimate of the thermal uplift is determined based on thermal conduction between the thinner (hotter) West Antarctica and the thicker (cooler) East Antarctica lithospheres.
Discussion: Models appears to explain most of the observations. However,
the rejection of the two alternate models is not entirely convincing. The
eclogite slab is one of the leading candidate models for the support of
the southern Sierras. The crustal thinning model does have the advantage
of naturally introducing a localized heat source into the model. The restriction
on being only able to support 1 km of uplift does not seem to be well documented.
The evolution of the stratigraphic environment in the Ross Sea consists of:
between basement and unconformity RSU6. approximately 2 km thick along the coast. May have extended further west than presently seen. Missing part probably eroded during the uplift of the Transantarctic Mountains; structurally deformed in places; source rivers or glaciers depositing materials in isolated basins. sampled in CIROS-1 drill site. Eocene to early Oligocene (55? to 30 m.y.)__________________________________________________________________
UPPER BOUNDARY: RSU6 - prominent feature characterized by a strong that can be traced nearly continuously; may correlate with a hiatus between middle Eocene and Early Oligocene or a major sea level low 30.5 Ma years ago.
more continuous reflectors than RSS-1. Sometimes directly over basement. sampled at two sites (30 to 22 m.y.); deposited in open marine conditions regional subsidence through early Miocene (18.m.y.) and western Ross Sea was submerged. coastline shifted eastward, following uplift of the TAM; sub-horizantal layers__________________________________________________________________
UPPER BOUNDARY RSU5 -bounds the top of this unit - identified with sea level low at 21 m.y. (late Oligocene)
widespread across the Western Ross Sea stratified and non-reflective units, more continuous & regular stratification than RSS-2. sampled in the eastern Ross Sea (DSDP Site 272) (silty claystone with clasts) age: (between 21 and 18.5 m.y. - late Oligocene to early Miocene) western side has channels indicating that glaciers advanced and retreated from the TAM , while east of the coast uniformly stratified units suggest that the sequence was deposited in an open marine environment.__________________________________________________________________
UPPER BOUNDARY -RSU4a - 18.5 m.y. no sea level change - some tectonic event ???
widespread, uniformly and continuously stratified; sampled at DSDP site 273 and 272. diatomaceous silt/clay with rare ice-rafted debris, shell fragments and some bioturbation. (top at 18.2 my.y., base 19.2 and 18.34m.y.)) upper age boundary suggested to be 16.5 m.y. because of a known major sea-level drop. deposited in an open water environment.__________________________________________________________________
UPPER BOUNDARY - RSU4 - sampled at Site 273 - marked by fine sand, possibly resulting from a change in ocean circulation or to a relative sea-level low. (15.5-16.5 m.y.)
basin filling, aggrading sequence, in DSDP SITE 273, semi-lithified pebbly, silty clay that has diatoms and is sparsely bedded; glacial marine, and deposited in open waters.__________________________________________________________________
UPPER BOUNDARY - RSU3 (10.5 m.y. low sea level)
- shelf-edge deposit (10.5 to 4 m.y.), not a lot of stratigraphic control 10.5 to 5.5 m.y. period of low sea level, 5.5 m.y. to 4.m.y. sea level rise__________________________________________________________________
UPPER BOUNDARY - RSU2 (4.m.y.), abrupt change in erosional patterns, development of a full scale ice sheet grounded in the Ross Sea.
- morphology similar to what is now on the sea floor, wide and deep erosional troughs and depositional banks.__________________________________________________________________
UPPER BOUNDARY - RSU1 - not well defined
- thin, highly eroded, and regionally isolated set of sequences in the western Ross Sea. Only the uppermost part has been sampled in many piston and gravity cores. Oldest stratigraphic units are younger than the major sea level low at 2.5 Ma. Probably formed as the polar ice sheet was advancing and retreating (due to sea level fluctuations) across the shelf many times, eroding sediment from the older sequences on the inner parts of the __________________________________________________________________DISCUSSION: The paper ties the seismic stratigraphy with existing geological information from nearby drill hole data. However, the paper does not supply any information about the evidence for the unconformities in the drill holes,(although we assume that it is based on fossil assemblages). It is very difficult to correlate the seismic units with the units in the drill holes, as well as with other seismic sections, because: 1) the stratigraphic units are time-transgressive, and not laterally continuous; 2) the stratigraphic units vary in thickness in different parts of the ROSS sea. In spite of these difficulties in correlation, there does not appear to be any evidence in these data that conflicts with the idea that tectonic activity in the Ross sea appears to have changed from extensional to transtessional around 30 m.y. ago (~RSU6 time).
This paper presents a fission track study of the Transantarctic Mountains in the Granite Harbour and Wilson Piedmont Glacier areas of southern Victoria Land. All of the Apatite ages sampled were less than 175 million years old, suggesting that there was a thermal event during the Jurassic accompanying the emplacement of the Ferrar Dolerite which "erased" fission tracks that existed before that time. Apatite age profiles from this study show a sharp break in slope that indicates a rapid uplift at approximately 55 Ma, and age dates from several sites and traverses delineate a series of N-S trending steeply dipping normal faults, with displacements of 40-1000 m down to the east throughout the area of study. The Jurassic Ferrar Dolerite sills in basement and cover rocks also proved as a useful tool in mapping out these normal faults. In several places, the westward dip of the dolerite is steeper as one moves east which indicates minor rotation of the fault blocks and suggests that the fault planes are listric at depth. These north trending normal faults lie at an acute angle to the trend of the axis of maximum uplift, indicating a dextral component of motion in this predominately EW extensional regime. There is evidence for several large scale dextral strike slip faults which may offset the northward trending normal faults. The axis of maximum uplift of the Transantarctic Mountains lies at Mount Termination, about 30 km west of the McMurdo Sound Coast, and there has been ~6 km of uplift since the early Cenozoic and 4.5-5 km of erosion along this axis. The amount of uplift decreases to the west at the same rate of the decrease in dip of the Kukri Peneplain.
Discussion:
Figure 16 addresses the question of rapid uplift of the TM during the past 10 million years. Evidence for this recent rapid uplift came from studies of diatoms which were found in rocks within the TM; however, it is now believed that these diatoms were transported to their present locations by ice or wind thus refuting the hypothesis that there was rapid uplift within the past 10 million years. It is also worth noting that the uplift rates determined in this study are strongly dependent on which value of the geothermal gradient is used in the calculation. It appears that the geothermal gradient in the region of study is not very well known, so the uplift rates may not be very accurate.
The authors studied lower crustal xenoliths that were erupted from various Cenozoic volcanoes within ~100 km of McMurdo sound, including some sites in the Transantarctic Mountains and some sites in the Ross Embayment. These inclusions were 2-pyroxene granulites, whose bulk chemistry suggests a tholeiitic or calc-alkaline, rather than alkalic, protolith; in other words they are unrelated genetically to the younger alkalic basalts in which they were erupted.
The 2-pyroxene granulites were further grouped into the following compositions: olivine granulite, spinel granulite, garnet granulite, and garnet clinopyroxenite (ranging in origin from lower pressure to higher pressure, respectively).
1. Pressure determinations. The whole-rock major element chemistry of the samples is similar to some glasses for which experimental petrological studies had been done. This provides bounds on the range of pressure stability for each of the granulite compositions. The authors found that the inclusions beneath the Transantarctic Mountains originated at depths from 15-45 km and those beneath the Ross Embayment originated at depths less than 25-30 km. One method of barometry used involves the Al content in opx based on reactions and thermodynamics, corrected for Fe and Na content. The equations are shown in the paper. The authors did these thermodynamic calculations for the entire suite of rocks. There was no other method available for the lower pressure compositions (spinel granulite and olivine granulite). For the garnet bearing compositions, 3 other geobarometers were also used.
2. Temperature determinations. The major element composition of the garnet, spinel, olivine, plagioclase, opx, and cpx was determined. Temperatures were based on the 2-pyroxene thermometer of Wells (1977). In a few samples they were able to use 2-feldspar geothermometry or albite geothermometry with similar results. One instance of the magnetite-ilmenite thermometer also gave consistent results. The temperatures were all high (827 to 977 C).
3. The uncertainties in pressure are estimated to be +/- 4 kbar, and in temperature to be +/- 75 C.
4. The authors conclude that a single geotherm can fit all of their observations. This geotherm is very hot compared to a typical shield geotherm. It is 50-100 C/km in the upper crust, with the top 15 km having an average gradient of 50 C/km or more.
It is inferred to be "contemporaneous" with the eruptions because there is no evidence for compositional zonation in the minerals studied - suggesting they came to the surface rapidly with no time to re-equilibrate to shallower T & P. High T phases are preserved (high sanidine, uninverted pigeonite). The geotherm is close to the basaltic solidus in the lower crustal part, perhaps due to the presence of basaltic melt throughout the lithospheric column being sampled. The upper crustal part of the geotherm is close to the dry granite solidus. The authors think that it's unlikely that granitic melts are actually being produced, though, because there aren't any silicic lavas being erupted.
However, all of the data for the shallow part of this geotherm come from the Ross Embayment, so the shallow geotherm for the TAM may be different - their data don't constrain it.
5. The authors state that this geotherm cannot represent conductive heat transport; there must be significant advection going on, to keep the T so high at shallow depths. However they only consider conduction from below; presumably if there is conduction from the Ross Sea side to the TAM side then maybe it could reproduce their observations for the TAM.
6. All of the inclusions are relatively mafic (< 66% SiO2). The authors infer a sharp compositional gradient at about 15-20 km depth, with silicic rocks overlying more mafic rocks, because it is thought that basic rocks are absent in the TAM within 15 km of the surface.
7. Since all of the inclusions studied are from the lower crust (not the upper mantle) the results would require the crust to be 27 km thick in the Ross Embayment; this disagrees with some geophysical estimates of the crustal thickness here as 20-23 km. However, the base of the crust here may consist of ultramafic cumulates whose seismic velocities make them look like "mantle" to the geophysicists (the old problem of the petrological vs. the geophysical Moho). The thickness of the crust beneath the TAM is inferred to be 40-43 km and this appears to have been in reasonable agreement with the geophysical data available at the time.
8. The high geothermal gradient means extensive heating of the lithosphere. The authors suggest that heating and simple thermal expansion of the lithosphere could have been an important mechanism in the uplift of the TAM.
9. We would have liked to have seen some discussion of the upper crustal xenoliths in both study areas, to see if there were a different shallow geothermal gradients in these two regions. Also it would have been useful to see if there were any P-T constraints on the mantle xenoliths, to aid in constraining the depth to the Moho in the two regions.
This paper outlines evidence that the Transantarctic Mountains have been rising since about 60 Ma, with the most rapid uplift occurring since mid-Pliocene time. The authors suggest that the Late Pliocene and Pleistocene uplift created a mountain barrier that resulted in climatic cooling, promoted ice-sheet thickening by damming outward ice flow, and changed the character of the ice sheet from temperate to polar. This correlates well with a cold period that began 2.5 Ma ago.
The Cenozoic West Antarctic rift system (Transantarctic Mountains) has a maximum relief of 5 km in the Ross embayment and 7 km in the Ellsworth Mountains-Byrd subglacial basin area. This rift scarp is interpreted to be the expression of a major transtensional fault zone. The main cause of uplift of this scarp is interpreted to be the result of a Cenozoic tectonic event, e.g. a flexed continental lithospheric plate (Stern and ten Brink, 1989). Variation of topography along the Transantarctic Mountains is partly caused by differential uplift on transverse faults.
Fitzgerald(1989) using fission-track dates from the Dry Valleys, interpreted 5-6 km of uplift in the Transantarctic mountains of southern Victoria Land. This uplift began approximately 60 Ma ago, and has continued at an average calculated rate of ~100 m/my. since that time. Because of the presence of angular unconformities seen in seismic-reflection data, the authors conclude that this uplift must have been more episodic, and therefore faster than the average rate (i.e. ~1 km/Ma since early of middle Pliocene time.
This conclusion is based on the following lines of qualitative evidence:
Wilch, T.I., Lux, D.R., Denton, G. and McIntosh, W.C., 1993, Minimal Pliocene-Pleistocene uplift of the dry valleys sector of the Transantarctic Mountains: A key parameter in ice-sheet reconstructions, Geology, v. 21, p. 841-844, September 1993
- 1) The relief (5-7 km) and the elevations (4-5 km) are the greatest of the rift mountains of the earth, therefore the uplift must be relatively recent.
- 2) in marine seismic reflection profiles, Holocene fault scarps with ~10m displacement are observed.
- 3) Normal faults displace late Pliocene moraines, by as much as 300 vertical meters.
- 4) The presence of unconformities in the Victoria Land Basin, in conjunction with an average uplift rate of 100 m /my. since 60 Ma requires episodic uplift. Constant uplift rates over 60 my. are not generally observed in active rift zones.
- 5) Glacial striae are found on peaks in the Transantarctic Mountains 600-1000m above present ice levels. Because models of maximum and minimum ice-sheet elevations indicate that the maximums were much lower than the highest striae, their presence at these elevations are most likely not caused by higher ice-sheet elevations in the past.
- 6) The presence of Pliocene marine microfossils (diatoms) in the Sirius Group (2.5 Ma) have been glacially transported from the Wilkes basin by the East Antarctic ice sheet to elevations between 1750 and 2500 m in the Transantarctic Mountains. The locations of these fossils can be more easily explained if the mountains were lower 3 Ma ago. (suggested uplift > 1000m since deposition)
- 7) Similar uplift rates in other rifts (2 and 8 km/Ma)
The authors agree that the Transantarctic Mountains consist of different tectonic blocks that were uplifted at different rates. They disagree that an episode of rapid wide-spread surface uplift occurred in Pliocene-Pleistocene time. They feel that their data outlining minimum uplift rates is more accurate than the previous studies, and less controversial than biostratigraphic evidence. This is because their data is from the rift shoulder where maximum uplift should have occurred, versus in down-dropped fault blocks.
They also argue that Behrendt and Cooper do not have much chronologic,
outside of fossil evidence in Sirius outcrops used elsewhere in Antarctica
to provide evidence for significant Pliocene-Pleistocene uplift, control
to support their findings. The argue that if the Sirius outcrops in the
Dry Valleys are not used, evidence from such outcrops in other areas should
also not be used. The assumption is that the Pliocene age diatoms present
in high-elevation glacial deposits were derived from glacial erosion from
the East Antarctic marine basins. Recent studies in _The Case for a Stable
East Antarctic Ice Sheet_, present arguments that that polar-desert conditions
extended back to middle Miocene time, and that the Pliocene marine diatoms
in the Sirius Group outcrops in the dry valleys block were not emplaced
by the East Antarctic ice sheet.
DISCUSSION:
We generally agreed that the Behrendt, J.C. and Cooper, A., (1991)
paper was short on actual data to support their claims of rapid uplift
in the Pleistocene. They neglected to mention that glaciers could have
a similar effect on producing the young looking topography and high relief
associated with the West Antarctic rift that was used to support their
claim. The Wilch, T.I., et al, (1993) paper had good data backup to support
their claim that there was not significant uplift of the Dry Valleys since
the Pleistocene. However, the question remains how the Dry Valleys fit
into the overall tectonic history of the West Antarctic Rift, and wether
they represent an anomaly along the rift, or are an indicator of the tectonic
history elsewhere.
The authors map the extent of a 28-30 Ma plume event in Marie Byrd Land (MBL), West Antarctica by examining the uplift of an erosional surface. This surface had its origins around the time of the breakup off of New Zealand from MBL. The two vestiges of it (one in West Antarctica and the other being the Waipounamu erosion surface in New Zealand) cuts through rocks that may have been initially in contact or perhaps on opposite sides one a shallow separation sea. The formation of the surface took about 10-15 Myrs. and occurred after the break up.
The marker of the erosional surface in W. Ant. is taken to be flat mountain tops that cut a variety of Cretaceous and older rocks, an uncomformities that separate Cretaceous rock from overlying late-Cenozoic volcanic rocks, and the U6 unconformity in the Ross Sea. The surface shows very little erosional cutting (100-200m). In New Zealand, the authors go to some length to say that the surface is not a just a "peneplain" (subarial erosional surface) but also involves marine transgression and wave planation. To emphasize this point and the importance of their paper, they rename the surface the "Waipounam erosional surface". The marker for the surface is taken to be the base of the marine transgressive sequence. The topography of the surface is deeply cut by fluvial processes, but generally shows the large uplift associated with the plate collision and the associated 20-km scale folding due to the transpresion of the Alpine Fault. The is no evidence of a plume uplift feature in the topography of the surface.
In MBL, the plume dome manifests itself as a broad arch of uplift
measured at the high points on the grabens (the elevation of the bottom of
the Grabens as not observed as yet). It is centered on Mount Petras, at the
edge of MBL, and has a width of about 500 km and an uplift of about 3 km.
The authors place the age of the dome at 28-30 to be in conjunction with a
magmatic activity in MBL. The authors make no attempt to model the uplift.
DISCUSSION:
I did not see the point of including the New Zealand erosional
surface in the paper since it was formed separately and after the New Zealand-
MBL breakup. I don't see how it can be used to prove or disprove the
presence of the plume associated with the break up process. An alternative
hypothesis to the plume might be crustal thinning and dynamic uplift due to
the inflow of mantle material.