GP43A-0838 1340h
Dynamics of Block Rotations in Strike-Slip Fault Zones: A Perspective from Distinct Element Modelling
Previous studies of the dynamics of deformation in strike-slip fault zones have approximated the deforming lithosphere as a thin sheet of viscous material governed by a power-law rheology. Such models (Sonder et al 1986, Nelson and Jones 1987, Sonder et al 1994) yield relationships between the magnitude and distribution of strike-slip offset and vertical axis rotation that match observations in some shear zones very well (e.g. Las Vegas Valley Shear Zone). However, other shear zones (e.g. North Anatolian Fault and some in the Mojave region) show no rotations or rotations that are inconsistent with that expected from thin-sheet model results using the same offset. In addition, because of their continuum nature, continuum models have limited scope in delineating the dynamics of how block size and brittle strength affect the distribution of offset and rotations in the fault zone. Therefore, we consider a complementary approach. We use a distinct element model (Saltzer 1992, Antonellini 1995) to calculate displacements and rotation of cylindrical blocks distributed in a zone between two rigid boundaries moving at a fixed rate. Forces between blocks are calculated from Hertz's (1896) and Mindlin's (1949) solutions for normal and shear interactions between elastic cylinders; the force balance is integrated over time to yield displacements. We consider the role of several factors, including block size, mechanical strength, and heterogeneity of block size and strength, in controlling the distribution of displacements and vertical-axis block rotations in shear zones.
GP43A-0839 1340h
Using Rotation Boundary Deformation to Infer Mechanics of Vertical-Axis Rotation in Southern California
Although vertical-axis rotation of crustal blocks has been recognized and studied in every type of tectonic environment, the driving mechanisms and mechanics of this process remain poorly understood. Both these aspects of rotational deformation in the upper crust can be addressed by looking at the boundaries of rotated areas where rotated crust abuts non-rotated crust. The geometry, diffusity, and kinematics of these boundary zones indicate the relative movement of crustal blocks, the relative strength of the blocks, the surficial stresses, and the nature of rotational accommodation at depth. New paleomagnetic and structural data from areas of Neogene rotation in southern California provide clues to the mechanics of vertical-axis rotation. In the western Transverse Ranges, recent data show that differential rotation occurred across dip-slip faults and folds rather than strike-slip faults. This observation suggests that: 1. rotation is accommodated at depth on discrete surfaces, and 2. rotation is driven by stresses in the upper crust rather than basal traction. Miocene rotation in the Mojave Desert was also primarily associated with dip-slip faulting and current work in the area is beginning to define and evaluate distinct rotation boundaries.
GP43A-0840 1340h
Paleomagnetic Evaluation of Crustal-Scale Block Rotations in the Mina Deflection of the Central Walker Lane
Crustal deformation resulting from relative Pacific-North America plate motion is broadly distributed on faults across the western U.S.. Geodetic observations show that some 25 percent of transform plate motion is currently accommodated by faults east of the Sierra Nevada, from the eastern California shear zone to the central Walker Lane (CWL). The northwest trending faults of the CWL are joined to the Furnace Creek - Owens Valley fault system by the Mina deflection, a belt of east-northeast striking faults that serve as a displacement transfer system linking the two fault zones. The Furnace Creek - Owens Valley fault system has a cumulative right-slip of 40 - 50 km, and this displacement is transferred in an extensional right-step to the curved high-angle faults of the Mina deflection. The high-angle fault geometry and basin depths preclude more than 10 km of right-lateral displacement accommodated along structures in the Mina deflection and thus a slip deficit of 30 to 40 km exists. We are attempting to resolve the amount of permanent Neogene strain in the Mina deflection and determine if the geodetically observed strain accumulation and local heterogeneities are consistent with the permanent Neogene strain, in particular, vertical axis rotation of blocks in the Mina deflection area. To in part quantify the magnitude and potential heterogeneities of crustal block rotation and to more clearly define block boundaries, paleomagnetic data have been obtained from numerous stratigraphic sections of upper Tertiary volcanic rocks in a 300 sqkm area centered on the Candeleria Hills, NV. Eight to ten oriented samples from 260 sites distributed across multiple tectonic blocks have been fully demagnetized with all sites yielding interpretable results (23 sites (two sections) in mid Miocene andesite flows, 102 sites (10 sections) in upper Miocene basalt flows, and 135 sites (27 sections) in lower Miocene, regionally extensive ash flow tuffs). Although some anomalous directions were observed, the stratigraphically corrected (typically less than 15° dip) data from the volcanic rocks are internally consistent, well-grouped at the site level, and discordant clockwise to late Cenozoic expected field direction (358\deg, 58\deg). For example, a Miocene basalt flow section in the central Candelaria Hills yields a group mean (D = 16.5, I = 50.0, a95 = 1.8, k = 497.3) that provides an inferred rotation of +18.5\deg +/- 6.9\deg compared to a late Cenozoic expected field. Two sites in the Candelaria Junction regional ash flow tuff located in separate structural blocks yield similar inclinations, yet statistically distinct declinations (Site1, 255.8, -48.6, 4.4; Site2, 296.4, -41.0, 4.6), providing evidence for a modest clockwise vertical axis rotation between structural blocks. Overall, we interpret most data obtained to reflect at most moderate clockwise vertical axis rotation among structural blocks in the Mina deflection region. The absence of large magnitude rotation of any individual block and, thus between blocks, places limits on the amount of cumulative slip on fault networks in the Mina deflection. The area may be accommodating strain by net northwest extension on a curved fault array with very low rates of vertical axis rotation and minor slip on individual faults. To compare the geodetic observations with long-term strain, we are developing a tectonic model, using existing fault geometries and paleomagnetic data for the CWL that links vertical axis rotations of the fragmented upper crust to the driving mechanisms associated with slip transfer across the region.
GP43A-0841 1340h
Magnetostratigraphy and Rotation of Pleistocene Sedimentary Rocks in the San Jacinto Fault zone, Western Salton Trough, CA.
Pleistocene sedimentary rocks in the San Felipe Hills (SFH) and Borrego Badlands (BB), Salton Trough, are deformed by the diffuse seismogenic San Jacinto fault zone. The Plio-Pleistocene lacustrine Borrego Fm is sharply overlain by the alluvial Ocotillo Fm in the southwest (BB and Ocotillo Badlands) and its finer-grained equivalent (Brawley Fm) in the northeast (SFH). This contact records rapid progradation of alluvium and termination of paleo Borrego Lake in response to initiation or reorganization of the San Jacinto fault. We sampled these units in measured sections at Oil Well Wash (in the SFH) and Beckman Wash (in the BB), and collected a thick composite section that includes all of the Borrego Fm in the southern BB. Sites were located at 10-50 m intervals in each section. Paleomagnetic results from 31 sites in the Oil Well Wash and Beckman Wash sections have well-defined components of magnetization in most samples. A persistent magnetic overprint, and some mineralogical alteration during thermal demagnetization, hampers analysis of magnetization vectors in some sites. Our results indicate that the age of the Ocotillo and Brawley Fms ranges from 0.5-0.6 Ma to 1.0-1.1 Ma, based on the presence of the Brunhes / Matuyama boundary and Jaramillo sub-chron in both sections, and Bishop ash in the BB. The Borrego-Ocotillo contact coincides with the base of the Jaramillo in both sections and in the nearby Ocotillo Badlands section (Brown et al., 1991). This result is consistent with evidence for a depositional hiatus at the contact in the BB section and an erosional disconformity at Oil Well Wash. Sites with well-defined directions and k$>$16 can be used to determine vertical-axis rotations. In Beckman Wash the tilt-corrected mean is D=360, I=38.2, k=48, \alpha$_{95}$=$8.1\deg$, N=8, indicating little or no rotation. In Oil Well Wash the tilt-corrected mean is D=8.5, I=61.1, k=51.0, \alpha$_{95}$=$9.5\deg$, N=6, indicating a cw rotation of $8.5\deg$\pm$$5.7\deg$. Combined, these data lead us to conclude that only minor ($8\deg$ or less) rotation has occurred in the two areas since 1.0 Ma, despite structural and stratigraphic evidence for significant fault reorganization during this time. The larger ($35\deg$) clockwise rotation documented in the Fish Creek-Vallecitos Basin (Johnson et al, 1983) may have occurred prior to 1.0 Ma, but the timing and amount of rotation is poorly known in that area and will be tested in future work.
GP43A-0842 1340h
Rotation of the Hikurangi Margin, East Coast, New Zealand: Reconciling Long-Term Deformation Patterns Indicated by Paleomagnetic and Magnetic Fabric Data With the Short-Term Velocity Field
A key question when studying continental tectonics is whether there is a practical length scale below which the continental crust is effectively rigid, allowing a 'microplate' treatment, or whether deformation is distributed at all length scales, requiring a quasi-continuous flow model. A common feature of continental deformation, the rotation of fault-bounded blocks about a vertical axis, can shed light on this question; any such blocks will have dimensions of the order of the dominant length scale of deformation. GPS and geodetic studies demonstrate that the Hikurangi Margin (East Coast, New Zealand) is actively rotating clockwise at 2-$4\deg$/ m.y.; the forearc region is acting as a linkage between regions of back-arc extension to the north and transpression to the south. Paleomagnetic studies indicate that these rotations have persisted at least since the Late Miocene, and have been used to further propose that the forearc is divided into a number of independently rotating domains. However, this partitioning is not observed in the short-term velocity field; the two datasets imply different dominant length scales of deformation. To investigate this discrepancy, we present the results of extensive sampling from 40 localities in the forearc of the Hikurangi Margin. In addition to stepwise demagnetisation to isolate characteristic remanences, the anisotropy of magnetic susceptibility was also measured. In relatively undeformed sediments the magnetic fabric is closely related to the regional stress field, and can therefore potentially be used as an independent rotation marker at sites where ancient paleomagnetic components are obscured by strong present day field overprints. The new data permit a better understanding of the manner and timing of rotations of the Hikurangi Margin. Rotations of the hitherto poorly-constrained southern part of the margin are consistent with sites further to the north, indicating that the whole margin may have rotated coherently since the Late Miocene, in agreement with the short-term velocity field. In addition, the data enable more accurate delineation of the boundary with the unrotated northern part of the margin, and allow better constraint of the structural mechanisms by which large differential rotations have been accommodated in this region.
GP43A-0843 1340h
Late Miocene block rotations in the South Chilean offshore forearc (38.5$\deg$S) linked to backarc shortening: paleomagnetic and bathymetric evidence
Oblique subduction inducing dextral shear along the southern Chilean forearc has prevailed for more than 30 Ma with rates between 15 and 5 cm/yr. At around 38°S the onshore-geology is dominated by numerous NW-striking faults parallel to the inherited Paleozoic basement structure. The swath bathymetry of cruise SONNE 161 emphasizes the continuity of these structures to the trench and shows that they segment the forearc slope and shelf into 20 to 80 km wide slices. The trench displays a pronounced saw-tooth morphology in map-view that suggests dextral rotations of these segments. However, both onshore and offshore there is no evidence for active strike-slip deformation along the NW-striking faults and the available moment-tensor solutions indicate a pre-dominance of thrust-type earthquakes within the forearc wedge. This suggests that the dextral block rotations are presently inactive or proceed at very low rates. Recently collected paleomagnetic data from 12 sites on Mocha Island 60 km east of the trench support this notion. Sites in Eocene and Miocene sediments show pre-dominantly dextral rotations of 20 to 30 $\deg$. In contrast, Pliocene rocks are not significantly rotated but only tilted to the NE in agreement with folding and thrusting over NE-dipping reverse faults. Hence, the available data imply that a major fraction of the block rotations has been accumulated between the deposition of the Miocene and the Pliocene sediments. These time constraints fit the deformation chronology of the upper-plate. At this latitude, shortening in the back arc of the Andes has been restricted to the interval between 10 and 5 Ma. Inspection of the upper-plate deformation field shows that the along-strike variations in shortening have resulted in an extension of the forearc that can account for the magnitude of the dextral rotations. In summary, forearc deformation is obviously driven by oblique subduction. However, considerable block rotations in map-view require along-strike-extension. The coincidence of the block-rotations and the Late Miocene upper-plate deformation at this latitude suggests that the required space has been provided by the deformation of the plate-boundary due to differential, southward diminishing shortening in the back-arc.
GP43A-0844 1340h
Temporal and Spatial Constraints on Multi-Phase Crustal Rotation in the Forearc of Northern Chile
The forearc of northern Chile between ~23-29$^{o}$S records some of the largest paleomagnetically detected crustal rotations reported to date in the Central Andes. In contrast to much of the rest of the Central Andes rotations appear to pre-date the main uplift and shortening of the Andean plateau between 25 Ma and the present time. We report new studies in which we have endeavoured to investigate the scale of the rotated area and timing of the rotation in the forearc area between 27-30$^{o}$S. Several authors have documented clockwise rotations in Mesozoic to Eocene units of up to 55$^{o}$ which, previously, appeared to decrease very sharply from about 30$^{o}$ of rotation at 28$^{o}$S to near zero at ~30$^{o}$S near La Serena. We present new data from over 120 sites from a range of Mesozoic to Eocene units in both the Coastal Cordillera and Precordillera. New data from two Paleocene plutons in the Tres Cruces area (29$^{o}$S) combined with existing information from contemporary plutons (66-62Ma) from as far north as Inca De Oro (26$^{o}$S) show the rotation to decrease smoothly suggesting a continuum in the deformation gradient controlling the rotations between these latitudes. These data also suggest that there was a distinct, if small ~10$^{o}$, rotation in Cretaceous times. In order to better constrain the age of the main rotation we also present new data from Triassic to Eocene units in the La Guardia area, east of the city of Copiap\'{o} (27$^{o}$S), in which we are able to demonstrate a variation in rotation during the period 60-40 Ma. In total these data strongly suggest to us that the large rotations of this region vary relatively uniformly and slowly with distance N-S and that a substantial part of this rotation pre-dates both the Andean orogeny and also the Incaic Orogeny of this part of the Central Andes. We suggest that the bulk of rotation was associated with the period of maximum obliquity of convergence between the Nazca and South American plates between 50-40 Ma. In addition, in the older rocks, of the Coastal Cordillera there is a small late Early Cretaceous component of rotation.
GP43A-0845 1340h
Two-Phase Rifting in the East Asian Continental Margin: Southwest Japan Rotated Twice
Rifting is one of the most important tectonic phenomena in continental margins. There are many rifted basins in the East Asian continental margin. Some of these basins form marginal seas with oceanic crust and the others form continental rift basins. The Japanese Islands have been recognized as continental slivers rifted from the Asian continent, associated with the opening of the Japan Sea. As a consequence of the rifting, Southwest Japan experienced clockwise tectonic rotation. The bulk of the clockwise rotation of Southwest Japan occurred during the Middle Miocene. On the one hand, Early Miocene paleomagnetic directions from Southwest Japan show declinations of about 40 to $50\deg$, showing the clockwise rotation of Southwest Japan during the opening of the Japan Sea. Late Cretaceous directions from Southwest Japan, on the other hand, show larger declination values of about $70\deg$. An unknown clockwise tectonic rotation may have superimposed on Cretaceous paleomagnetic declinations of Southwest Japan. The possible clockwise rotation may be attributed to a rifting event in East China to the west of the Korean Peninsula, because Early Tertiary rifted basins have recently been postulated in that region (e.g. Bohai Basin), and because Southwest Japan was flanked along the eastern side of the Korean Peninsula during the Cretaceous. The rift basin formation is considered to have caused the tectonic rotation of an area incorporating the Korean Peninsula and Southwest Japan. The Cretaceous paleomagnetic pole for the Korean Peninsula is rotated clockwise by $23\deg$ with respect to that for the North China Block, suggesting a rotational motion of Korea after the Cretaceous. This clockwise rotation is observed from the whole of the Korean Peninsula. The occurrence of the rotation is further confirmed by comparison of paleomagnetic poles of an older age between the two regions: Triassic data also record the consistent clockwise rotation of the Korean Peninsula ($15\deg$). Therefore the Korean Peninsula probably rotated clockwise during the Early Tertiary. A two-phase rifting occurred in the East Asian continental margin. This is characterized by a first phase (Early Tertiary) that has caused the clockwise rotation of an area incorporating the Korean Peninsula and Southwest Japan, followed by a second phase (Middle Miocene) that has resulted in the clockwise rotation of Southwest Japan.
GP43A-0846 1340h
Magnetostratigraphy of Cenozoic Sediments From the Kuche Depression, Tarim of NW China and Tectonic Significance
A combined paleomagnetic and magnetostratigraphic study is reported of Cenozoic sediments, comprising of a total of 991 red bed horizons within the upper part of the Kumugeliemu Formation and the lower part of the Kuche Formation of the Kuche Depression in the Tarim Basin, northwest China. Thermal demagnetization is used to separate the magnetic components. The characteristic remanent magnetization with both normal and reversed antipodal directions is then isolated generally between 300 and $690\deg$C and a total of 969 data were obtained for construction of a composite magnetostratigraphic sequence. The composite magnetostratigraphic sequence correlates with the C12n to the C3r, between $\sim$31.0 and 5.8 Myr, of the geomagnetic polarity timescale of Cande and Kent (1995). Results from correlation suggest that the boundaries of Kuche/Kangcun Formations, Kangcun/Jidike/ Formations, Jiedike/Suweiyi Formations, and Suweiyi/Kumugeliemu Formation are at about 5.9, 13.5, 26.0, and 29.0 Myr respectively. We argue that the Paleogene/Neogene boundary is most likely to be at the bottom of the Jidike Formation in the Kuche Depression and the dramatic transition from the marine/lacustrine to fluvial/alluvial facies in the Suweiyi Formation and the lower Jidike Formation, which may indicate the initiation of the Cenozoic thrusting in the Kuche Depression (Yin et al., 1998), is probably occurred in the late Oligocene, at about 28-25 Myr. The mean characteristic remanent magnetization directions every 5 Myr suggest that the Kuche Depression was subjected to a counterclockwise vertical-axis rotation at a rate of $\sim$1°/Myr during about 31.0-5.8 Myr, and a clockwise rotation at a rate of $\sim$2°/Myr during the past $\sim$5.8 Myr. This study thus suggests that at least two rotations of the Kuche Depression relative to the Earth rotation axis have occurred during the past 31 Myr.
GP43A-0847 1340h
Evaluating Extension in the Ross Sea vs. Vertical Axis Rotation in West Antarctica as the Cause of the Offset of West and East Antarctic ca. 100 Ma Paleomagnetic Poles
Mid-Cretaceous (ca. 100 Ma) paleomagnetic poles from eastern, central, and western Marie Byrd Land (MBL) are concordant with one another and with similar age poles from the Thurston Island (TI) and Antarctic Peninsula (AP) blocks. These West Antarctic results are, however, offset from the East Antarctic Apparent Polar Wander Path. DiVenere et al. (1994) interpreted the offset as the result of opening of the Ross Sea after 100 Ma. In contrast, Luyendyk et al. (1996) proposed vertical-axis rotations in West Antarctica, as in southern California, as an alternative explanation for the offset. This was suggested to be due to transtensional separation of New Zealand from Marie Byrd Land. The scale of the proposed West Antarctic rotational corridor (~3000 km from western MBL to the northern AP) is an order of magnitude larger than the scale observed in southern California. The method used by Luyendyk et al. (1996) (decomposing the offset between the East and West Antarctic poles into separate components of vertical-axis rotation and poleward motion) is poorly suited for the study of MBL because the close proximity of the 100 Ma paleomagnetic pole to the sample sites results in greatly differing estimates of apparent rotation vs. poleward motion depending on the proposed position for the vertical rotation axis. In fact, this method implies post-100 Ma vertical axis rotation with no statistically distinguishable horizontal displacement for the northern Ford Ranges while implying large horizontal displacement with no discernible vertical axis rotation for the Ruppert/Hobbs Coast of central MBL and the Kohler Ranges of eastern MBL. Moving eastward from MBL to Thurston Island and the Antarctic Peninsula, progressively less vertical-axis rotation is required to reconcile the paleomagnetic poles with the East Antarctic poles due to the progressive increase in the distance of the rotation pole from the paleomagnetic pole. This pattern of diminishing apparent rotations does not fit a simple "rolling ball-bearings between two plates" model. Similar amounts of vertical axis rotation would have induced progressively larger amounts of offset from the East Antarctic pole from western MBL to the northern AP. It seems unlikely that a mechanism yielding the observed post-100 Ma sequence of diminishing apparent rotations would result in such a fortuitous grouping of displaced paleomagnetic poles. The offset of the mid-Cretaceous West Antarctic poles from the East Antarctic pole is more simply explained as the result of horizontal displacement between East and West Antarctica due to extension in the Ross Sea and Ross Sea Embayment.
GP43A-0848 1340h
Pangean Reconstruction of the Yucatan Block and its Permian, Triassic, and Jurassic Geologic and Tectonic History
Paleomagnetic studies of Paleozoic sedimentary and plutonic rocks demonstrate that the Yucatan Block lay inverted off the NW coast of South America (SA) in Pangea, and rotated in a series of clockwise motions as SA and NA separated. Mid-Permian Yucatan was a fragment of the western margin of Pangea, just south of the equator. By 230 Ma, Yucatan rotated $\sim20\deg$ clockwise and moved to the equator. Clockwise rotation continued through the Jurassic: $\sim54\deg$ between 230 Ma and about Oxfordian, and another $\sim32\deg$ between $\sim$Oxfordian and Tithonian, when the approximate present orientation was achieved. Passage of the Yucatan Block into the gap created by the separation of North and South America requires left-lateral motion relative to North America, perhaps by translation along the Mojave Sonora megashear. One remanence, carried by magnetite in red bed strata from the margins of the marine Santa Rosa basin (Maya Mtns), decays exceedingly linearly to the origin of orthogonal-axes plots. Biostratigraphy indicates a Late Pennsylvanian to Middle Permian age, and the presence of dual polarities in a 110 m, four polarity, magnetostratigraphic sequence demonstrates a post-Late Paleozoic Reversed Superchron (i.e., Middle Permian) age. Plutonic rocks exhibit a second, also dual-polarity, remanence, which corresponds to a paleopole $60\deg$ clockwise of the mid-Permian pole. The spatial relationship between the Permian and pluton paleopoles is very similar to that of the North American Permian and Late Triassic poles. Moreover, exceedingly uniform K/Ar ages of 231$\pm7$ Ma characterize all Maya Mtns plutons and the southern volcanic complex, indicating a 230 Ma resetting of the K/Ar radiometric systems of plutons dated by U/Pb as Late Silurian (Steiner and Walker, 1996). Metamorphic aureoles developed in the Pennsylvanian-Permian Santa Rosa strata bordering the Silurian plutonic complexes suggest the occurrence of a post-intrusion hydrothermal event. The 230 Ma reset K/Ar systems, metamorphic aureoles in strata younger than the plutons, and a magnetization resembling a Late Triassic remanence all suggest that a 230 Ma hydrothermal event remagnetized and reset the igneous rocks, probably during the initial break-up of Pangea. The Maya Mtns pluton paleopole is statistically identical to that of the Chiapas Massif. Identical dual polarity magnetization populations characterizing both of these Yucatan plutonic complexes suggest that both were remagnetized around 230 Ma. Importantly, identical remanences in these widely separated plutonic complexes indicate that the Yucatan Block (including the Chiapas Massif) has been a structural entity since at least 230 Ma.
GP43A-0849 1340h
Interior Baja B.C. : Continuing Rotation on a Diffuse Plate Boundary
Interior Baja B.C. - the Intermontane Belt (IMB) and Yukon-Tanana (YT) terranes of northwestern North America - provide a geological record of the complex interactions between the northeastern Pacific basin plates and craton. Geophysical evidence from earthquake seismology, gravity, global positioning system and heat flow data indicate motion of the IMB terranes toward the craton today. Paleomagnetic data show the YT terrane to be parautochthonous and part of the craton's ramp onto which the IMB terranes were obducted. Conversely the IMB terranes behaved as an allochthonous reasonably-coherent microplate with its own apparent polar wander path. Relative to the craton, the path dictates that: 1) from 0-54 Ma the IMB rotated steadily on the craton's ramp at 0.29$\pm$±0.11$\deg$/Ma or 16$\pm$6$\deg$ clockwise (CW), consistent with Lithoprobe SNORCLE deep crustal seismic evidence for thin skinned tectonics; 2) from 54 to 102$\pm$14 Ma the IMB was offshore and was further rotated by 35$\pm$14$\deg$ CW and translated northward by 8.3$\pm$7.0$\deg$ (915$\pm$75 km), consistent with geological estimates for total dextral fault displacement and seafloor plate vectors; and 3) more speculatively, from Early Cretaceous to Early Jurassic, the IMB moved in concert with the craton off the western USA seaboard. This history fits with major geologic events such as extensive Eocene extension in southern British Columbia, development of the 1000 km-long Selwyn-Mackenzie orogenic arc in Yukon, YT terrane exposure on either side of the IMB, etc. Further it requires continuing crust-mantle interactions that extend some hundreds of kilometers into the craton today.