How much strain can continental crust accommodate without developing obvious through-going faults?
Published:January 01, 2007
- PDF LinkChapter PDF
B.C. Burchfiel, C. Studnicki-Gizbert, J.W. Geissman, R.W. King, Z. Chen, L. Chen, E. Wang, 2007. "How much strain can continental crust accommodate without developing obvious through-going faults?", Whence the Mountains? Inquiries into the Evolution of Orogenic Systems: A Volume in Honor of Raymond A. Price, James W. Sears, Tekla A. Harms, Carol A. Evenchick
Download citation file:
Geologic data combined with global positioning system (GPS) and paleomagnetic data from SW China indicate that continental crust can absorb tens to perhaps at least hundreds of kilometers of horizontal shear without developing either through-going faults or obvious structures capable of accommodating shear strain. The arcuate, left-lateral Xianshuihe-Xiaojiang and Dali fault systems bound crustal fragments that have rotated clockwise around the eastern Himalayan syntaxis. The two fault systems terminate to the south, but faults reappear farther south, and these continue the GPS velocity gradient. The shear must be transmitted across the Lanping-Simao fold belt without forming through-going faults. West of the Longmen Shan, a geodetically determined velocity gradient of ∼10 mm/yr at N60°E lies in an area not marked by through-going faults. If this deformation has been active for the past 8–11 m.y., it should have accumulated ∼100 km of shear across a belt ∼100 km wide. In both regions, there are no obvious structures that are capable of accommodating the shear. Paleomagnetic data from the southern Lanping-Simao belt are interpreted to indicate an unexpected zone of left-lateral shear present (Burchfiel and Wang, 2007) where rotation of crustal material is locally more than 90° across a zone unmarked by any mapped through-going faults. In these examples, the mechanism of deformation is not obvious, but we suggest it is distributed brittle deformation at a range of scales, from closely spaced faults to cataclastic deformation. In older terranes, recognition of such zones potentially adds an unknown uncertainty to field study and tectonic analyses.
Global positioning system (GPS) studies conducted over the past decade or so in several actively deforming regions have begun to reveal crustal deformation that is not obviously expressed in the geology on the ground (Burchfiel, 2004), and they have revolutionized our approach to understanding the mechanisms and time scales of crustal deformation. These short-term data can often be extrapolated back in time to show that deformation occurred over longer time intervals and produced tens to perhaps hundreds of kilometers of displacement across zones 10–100 km wide. One example of GPS velocity gradients revealing horizontal shear in the absence of through-going structures is the Eastern Mojave shear zone in California (Miller et al., 2001; McClusky et al., 2001, personal commun.  in Burchfiel, 2004). Along this shear zone, crust south of the Garlock fault, as well as the fault itself, are both being rotated in a clockwise fashion, whereas rocks both to the north and south show prominent faults related to right-lateral shear. Potentially, where the appropriate rocks are exposed in the appropriate locations, paleomagnetic studies should be able to demonstrate the rotation of the crust between the shear zones to the north and south of the Garlock fault to identify the through-going nature of the shear (Schermer et al., 1996). However, such data would still not resolve the outstanding problem: how this deformation occurred in apparently unfaulted, yet rotated crust. We submit that the Eastern Mojave shear zone is not an isolated case and additional examples are likely to come to light as geodetic techniques continue to be applied to the study of active crustal deformation.
Recent studies on deformation in the area of the eastern syntaxis of the Himalaya in southwest China have led to the recognition of several variations on the problem of differential shear passing through crustal rocks without obvious through-going structures. This observation presents a major problem for field geologic studies in this and, we suggest, numerous other areas. In several places, geodetic and paleomagnetic data have highlighted broad regions where geologic mapping has failed to identify and/or recognize considerably large magnitudes of regionally important strain.
UNRECOGNIZED SHEAR DEFORMATION IN SOUTHERN YUNNAN, CHINA
Wang et al. (1998) showed that the active left-lateral Xianshuihe-Xiaojiang fault system in the SE part of the Tibetan Plateau begins as a narrow fault zone in western Sichuan (Figs. 1 and 2). It continues to the southeast as a convex-east, arcuate fault zone, and its displacement becomes partitioned progressively onto several faults, until, ∼100 km north of the Red River, the zone is more than 100 km wide and consists of numerous subparallel faults (Fig. 2). The faults of the Xianshuihe-Xiaojiang fault system cross numerous older structures and geologic contacts within the Yangtze platform at high angles, which permit accurate measurement of their displacements. Wang et al. (1998) showed that the total displacement across the Xianshuihe-Xiaojiang fault system remained essentially constant at ∼60 km whether it was measured across a single, localized fault or summed across numerous faults where the deforming zone is broad. As the Xianshuihe-Xiaojiang fault system near the Red River, the number of mapped faults is fewer, but the displacement on individual faults also becomes less. Only three of the faults reach the Red River valley, at which point each has less than 1 km of displacement and none of them offsets the well-defined contact between the metamorphic rocks of the Ailao Shan and Cenozoic strata to the north (Schoenbohm et al., 2006). Where the fault system intersects the Ailao Shan, the Ailao Shan has a WNW strike compared to the overall NW strike of the entire Ailao Shan. The change in strike was interpreted by Wang et al. (1998) as the result of accommodation of the 60 km of left-lateral shear on the Xianshuihe-Xiaojiang fault system by bending of the Ailao Shan. They further noted that, although the displacement on individual faults within the Xianshuihe-Xiaojiang fault system becomes smaller nearer to the Ailao Shan, the total shear remains approximately constant.
South of the Ailao Shan, in the Lanping-Simao fold belt, several structures, early Cenozoic fold traces and reverse (?) faults, appear to be bent, but no through-going faults parallel to the Xianshuihe-Xiaojiang fault system are present. South of the Lanping-Simao fold belt, several prominent approximately NE-striking left-lateral faults appear and can be traced to the SW for hundreds of kilometers into Indochina (Figs. 1 and 2). Many of these faults show a curved trace convex to the SE. Wang et al. (1998) interpreted the southeastern section of most of these faults (the Dien Bien Phu fault) to be the continuation of the shear mani fested in the Xianshuihe-Xiaojiang fault system. The faults north (Xianshuihe-Xiaojiang fault system) and south (Dien Bien Phu fault and others) of the Lanping-Simao were thus interpreted as part of a boundary to a broad crustal area that has been rotating around the eastern Himalayan syntaxis for at least the past ∼2–4 m.y., and most probably longer.
Modern GPS velocities for this region calculated relative to a fixed South China show that the region west of the Xianshuihe-Xiaojiang fault system and its southern continuation is being displaced in a manner that is consistent with clockwise rotation around the eastern Himalayan syntaxis (Fig. 3; King et al., 1997; Chen et al., 2000; Zhang et al., 2004). The GPS velocities parallel the major faults, and elastic dislocation modeling indicates that the geodetically determined velocities can be explained by locking of the few, major faults at a depth of 12–17 km (Studnicki-Gizbert et al., 2004; Meade, 2007). Clearly, the Xianshuihe-Xiaojiang fault system represents the system of structures that accommodates the velocity gradient between the rotating crustal fragment to the west of the fault system and South China. What the data also show is that the velocities and the velocity gradient are continuous across the Lanping-Simao fold belt, which is consistent with the analysis from the geology as well as new block model analysis of the present-day kinematics of the area (Meade, 2007). However, there are no prominent, through-going faults in this region that can be straightforwardly related to the geodetically determined velocities.
Examination of GPS velocities also shows that the translation paths defining crustal rotation are divergent, and there is another velocity gradient with increasing westward component of velocity associated with transtension and left-lateral slip on the active Dali fault system closer to the syntaxis (Fig. 3). The Dali fault system (Fig. 4) consists of several left-lateral faults (the Zhongdian, Jianchuan, Lijiang, and Chenghai faults, among others) that generally parallel the GPS velocities (in a South China frame, see Zhang et al., 2004) and have a convex-east arcuate pattern. The faults are prominent and easily recognizable structures that can be traced in the field and in remote-sensing imagery to the south, where they intersect the eastern boundary of the Lanping-Simao fold belt, and then apparently end. They only displace the boundary locally along the easternmost fault of the Dali fault system (the Chenghai fault; Schoenbohm et al., 2006). On the south side of the Lanping-Simao fold belt, several active, convex-east, left-lateral faults are present that can be regarded as the SW continuation of the left-lateral shear gradient. Like the situation with the Xianshuihe-Xiaojiang fault system to the southeast, there are no faults that mark the shear that must cross the Lanping-Simao fold belt. Thus, like the Xianshuihe-Xiaojiang fault system, the shear gradient along the Dali fault system continues through rocks without any recognizable through-going structures that might accommodate the shear.
The total displacement across the Dali fault system remains poorly determined. Only the Zhongdian fault at the western end of Dali fault system shows a measurable offset of 15–20 km; however, much of this offset is probably early Cenozoic (Burchfiel and Wang, 2007). This fault is associated with 3 km of offset on the Jinsha River. Schoenbohm et al. (2006) has shown an offset of ∼7–9 km on the southern end of the Chenghai fault where it offsets the Red River fault. Farther to the north, the same fault offsets the Jinsha River by ∼3.5 km, but this represents a minimum value. The Lijiang, Jianchuan, and Heqing faults mostly transfer extension between actively opening basins, the subsidence histories and fill geometries of which are poorly known. The Daju fault (the active continuation of the Zhongdian fault) is associated with 2–4 km of left-slip, constrained by the offset of the pre-incision outlet to the Daju basin. An estimate of >10 km total offset on the Dali fault system is not unreasonable. The total offset across both the Xianshuihe-Xiaojiang fault system (60 km) and the Dali fault system (>10 km) is on the order of ∼70 km. Lacassin et al. (1998) have given displacements for the faults south of the Lanping-Simao belt that total 67–75+ km. Based on these observations, we suggest that the short-term GPS velocity gradient is transmitted through the Lanping-Simao belt. In addition, the long-term, total displacement of the Zhongdian-Dali continental fragment and the crust west of the Xianshuihe-Xiaojiang fault system, which rotates about the eastern Himalayan syntaxis, also appears to pass through this belt but is not marked by any through-going structure.
The nature of termination of the faults of the Xianshuihe-Xiaojiang fault system and Dali fault system on either side of the Lanping-Simao belt is different. At their north and northeast boundaries, they stop abruptly at the northern boundary of the Lanping-Simao fold belt (Fig. 2). This boundary is marked by active faults, many of which are normal faults, but there is also right-lateral displacement along part of the Red River fault (Replumaz et al., 2001; Schoenbohm et al., 2006). Those faults of the Xianshuihe-Xiaojiang fault system that reach the Red River area offset active strands of the Red River fault, indicating that both fault sets are active before they reach the northern boundary of the Ailao Shan metamorphic rocks (Schoenbohm et al., 2006). The faults of the Dali fault system end either at normal faults along the Lanping-Simao boundary or terminate within related extensional basins.
On the southwest side of the Lanping-Simao fold belt, the relationship is quite different. The southern boundary of the Lanping-Simao fold belt is marked by a continuous and sinuous (inactive) fault zone. Several of the left-lateral faults strike NE into the boundary but do not cross it, such as the Wanding and Mengliang faults (Fig. 2). Others, such as the Mengxing and NW-striking Heihe faults, cross the boundary a short way but end within the Lanping-Simao rocks. The Nantinghe fault strikes into a prominent bend in the boundary fault at the Wanling Shan (WS in Fig. 2). The Wanling Shan is cored by an anticline of Paleozoic rocks overlain by the Mesozoic strata of the Lanping-Simao fold belt. This fold has a thrust on its north side, and it refolds structures within the western part of the Lanping-Simao belt. We interpret these relations to indicate that the active Nantinghe fault dies out into a vertical-axis fold that folds the boundary fault and some structures within the Lanping-Simao fold belt. The active NW-striking Heihe fault to the south penetrates farther into the Lanping-Simao belt than any of the other faults before dying out along a belt of several faults that cuts through left-laterally sigmoidally bent folds.
The active, left-lateral Dien Bien Phu fault cuts across the entire width of the apparent SE continuation of the Lanping-Simao belt where it enters Vietnam. However, Geissman et al. (2007, personal commun.) have speculated that there may be a substantial difference between the early Cenozoic tectonic development of the SE part of the Lanping-Simao belt and that of the rocks in north Vietnam. Thus, there may be important changes in crustal structure at the location of the Dien Bien Phu fault. Nevertheless, the Dien Bien Phu fault ends before reaching the Ailao Shan metamorphic rocks.
A principal feature of the Lanping-Simao fold belt is the lack of any through-going faults that are associated with the geodetically determined velocity gradient. Seismicity indicates that this region is indeed deforming, but earthquake hypocenter locations do not serve to define any through-going fault zones. Recent block modeling by Meade (2007) also has shown that the best fit to data on active faulting and velocity gradients within the area is accommodated by shear that passes through the Lanping-Simao fold belt. All the data indicate that the shear passes through the Lanping-Simao fold belt but is not manifested by obvious through-going faults.
The key question we pose is the following: by what mechanism does shear deformation continue through the rocks of the southern Yangtze platform and Lanping-Simao belt at the southern end of the Xianshuihe-Xiaojiang fault system? The answer at present is, we do not know. Interpretation of the bedrock geology in the Lanping-Simao belt and adjacent areas is somewhat compromised because of typically poor exposure and access. For example, the area at the southern end of the Xianshuihe-Xiaojiang fault system is characterized by a thick soil cover, often terra rossa, above a well-developed karst surface, and within the Lanping-Simao, there is dense vegetation and considerable agricultural development. Nevertheless, outside of the Lanping-Simao fold belt, in areas of similarly poor exposure, Quaternary faults are well expressed geomorphically and are easily mappable both in the field and on digital elevation models (DEMs) and satellite imagery. Chinese maps of the Lanping-Simao belt show numerous small faults, some parallel to the trend of the Xianshuihe-Xiaojiang fault system, that clearly offset fold axes. Some faults have the appropriate map sense offset, but others do not, and none of these faults is continuous for very far along strike. More importantly, it is unknown if any of the mapped faults in the Lanping-Simao are active: none of these is well expressed geomorphically or is unambiguously associated with historic earthquakes. We suspect, but cannot prove, that shear strain is accommodated by brittle deformation on a broad scale by numerous small faults and gouge zones of variable orientation (thus effectively defining a system of small-scale rotating blocks) and mesoscopically ductile-mode deformation accommodated by diffuse and distributed cataclastic deformation. The problem in the context of field-based geologic observations is that most of this type of deformation would not necessarily go unnoticed, but it would not generally be mapped as parts of a large-scale system and recognized for its regional kinematic importance. However, it does seem clear that this manner of crustal deformation can absorb considerable magnitudes of strain (equivalent to tens or even hundreds of kilometers of offset on a single, localized fault) that might largely remain unrecognized. Such deformation would be increasing difficult to recognize in older more extensively deformed belts.
The Lanping-Simao belt also reveals another example of unrecognized strain. We have completed a regional paleomagnetic study of the Upper Mesozoic to Lower Cenozoic red beds within this belt (Fig. 5; Geissman, 2007, personal commun.). The uppermost crust of the Lanping-Simao belt consists largely of a Jurassic to Oligocene sequence of red beds that were folded during early Cenozoic syn- to postcollisional deformation around the eastern Himalayan syntaxis during a phase of deformation that is older than that emphasized herein but that is superposed on early Cenozoic structures. In combination with interpretations of the regional geology of the area, and our understanding of the timing and sense of displacement along major shear zones along the eastern margin of the eastern syntaxis, the paleomagnetic data are interpreted to demonstrate that during early Cenozoic time, southeastward extrusion of most of the Lanping-Simao fold belt was accompanied by some 60° to 120° of clockwise rotation. However, unexpectedly, the southeasternmost part of the Lanping-Simao belt appears to be unrotated (Fig. 5), although this observation is based on a relatively limited data set. There is a fairly well-defined boundary between rotated and unrotated rocks. Such a boundary has not been inferred to be prominent in any previous tectonic analysis of the Lanping-Simao fold belt. The boundary appears to lie parallel to the curved trend of the folds and faults within the belt. In compilations of Chinese maps we have made of this area, there are no obvious structures present that would define or accommodate such a boundary. The limits of the proposed boundary lie within a belt of arcuate structures in the fold belt and do not follow any mapped through-going fault zone. Perhaps in areas of excellent outcrop, such a boundary might be readily recognized, but in this vegetated, and relatively highly weathered area, there are no obvious mapped structures that appear to accommodate the differentially rotated rocks.
Before completion of our early GPS investigations within western Sichuan and eastern Qinghai provinces on the Tibetan Plateau, we developed a synthesis of the late Cenozoic tectonic framework of that region (Fig. 1; Burchfiel et al., 1995; Burchfiel, 2004). West of the Longmen Shan, the plateau is underlain by a thick sequence of largely Middle and Upper Triassic flysch of the Songpan Ganze basin intruded by Mesozoic plutons. The flysch is weakly metamorphosed but contains local areas of amphibolite-grade metamorphism (e.g., see Dirks et al., 1994). It is isoclinally folded and locally refolded. Folds trend NW in most of the flysch, and when traced eastward, the folds sweep into the western Longmen Shan in a sense that has been interpreted to have been caused by left-lateral shear of Mesozoic age along the eastern margin of the Songpan Ganze basin (Dirks et al., 1994; Burchfiel et al., 1995).
Our first GPS results from eastern Tibet (King et al., 1997) showed an unexpected velocity gradient west of the Longmen Shan that indicated right-lateral shear between the central Tibetan Plateau and the Longmen Shan (the eastern edge of the plateau). Newer data (VGWLS in Fig. 3; Chen et al., 2000; Zhang et al., 2004) support the existence of the NE-trending velocity gradient; however, this part of the Tibetan Plateau is not covered by an abundance of GPS stations, so the location of the velocity gradient is not well determined. The velocity gradient is ∼10 mm/yr across an ∼100-km-wide zone (not unlike the dimensions of the Xianshuihe-Xiaojiang fault system discussed previously). There are no mapped through-going structures in that area that can be related to the velocity gradient. Folds and faults in the Triassic rocks have no mapped offsets, and satellite imagery shows no obvious through-going lineaments (Burchfiel, 2004). Our field work in this area recognized an active fault zone at Zhe Da, ∼10 km west of the town of Aba (32.9°N, 101.7°E). The fault strikes N60°E, parallel with the velocity gradient. Along this fault, streams are consistently offset right-laterally a few tens of meters (Fig. 6). The mapped fault has not been followed more than 10 km to the NE, and it does not show up as a prominent linear feature on Landsat images. The location of this fault may lie too far to the west to be within the zone of the high GPS velocity gradient. Active right-lateral displacement has also been identified on the Wenchuan-Maowen fault zone within the western part of the Longmen Shan (Burchfiel et al., 1995); however, the trace of this fault appears to be too far east to be related to the GPS velocity gradient. Thus, from all the data available at present, there are no obvious through-going faults that can be unambiguously associated with the geodetically determined velocity gradient. Unfortunately, we do not know how long this velocity gradient has been in existence, but our regional analysis suggests that it may have been active since ca. 8–11 Ma (Burchfiel, 2004; Kirby et al., 2002). If the present geodetic rates can be extrapolated for this entire interval, the deformation would have resulted in ∼100 km of right-lateral shear across an ∼100-km-wide zone, similar to that of the Xianshuihe-Xiaojiang fault system.
In the region of NE Tibet, outcrop is again rather poor, but we contend that localized, through-going structures that accommodate the shear in this area are simply not present (in part, because faults elsewhere in this region are easily recognizable). These structures do not show up on either geological maps or remote-sensing images. A few short faults with NE strikes are shown in Chinese maps, but none has been investigated except the two mentioned previously, so it is not known if they are active or what their displacement is. We can only speculate that the shear is accommodated by deformation on a broad zone of spaced faults or smaller-scale brittle structures that remain unrecognized. Regardless of the actual mechanism, it is sobering to recognize that the existing geologic information would never have led us to look for such structures or infer such strains; it was only when the GPS results became available that we began to attempt to identify accommodating structures.
In the examples given here, there does not appear to be any relation between either rock type or crustal structure through which the velocity gradients pass in the absence of through-going structures. The southern part of the Xianshuihe-Xiaojiang fault system north of the Red River is within the Yangtze platform, which consists of a Neoproterozoic basement of generally low-grade metamorphic and igneous rocks with thin Paleozoic and Mesozoic cover. These rocks were deformed in both Mesozoic and Cenozoic time, and the Xianshuihe-Xiaojiang fault system cuts at a high-angle across older structures. The Lanping-Simao belt consists of at least 6 km of unmetamorphosed Mesozoic and lower Cenozoic red beds that were deformed into a fold-and-thrust belt in early Cenozoic time. The Lanping-Simao red beds were deposited in a Jurassic-Cretaceous to possibly early Cenozoic extensional basin (Yano et al., 1994). Unlike the Yangtze platform, the Lanping-Simao fold belt probably overlies a shallow, regionally extensive décollement and probably does not involve deeper (pre-Paleozoic) basement rocks. The GPS velocity gradients of the Xianshuihe-Xiaojiang and Dali fault systems pass through the structures at a high-angle. The Songpan Ganzi belt in NE Tibet consists of low-grade flysch that is tightly folded and probably also is detached within middle-crustal rocks.
In the three examples we have used here, the velocity gradient does not parallel pre-existing structures but crosses them at a high angle. The only place where older anisotropy is present is in the area of the faults along the southwest margin of the Lanping-Simao belt, and then only locally. Some of the left-lateral faults, such as the Mengxing and the Wanding faults, have early Ceno-zoic right-lateral displacement that was later reversed during late Cenozoic time (Lacassin et al., 1998). However, other left-lateral faults, such as the Nantinghe fault, appear to be new and were initiated in late Cenozoic time. The faults in this area cut through tectonic assemblages that consist of rocks distinctly different from the Lanping-Simao belt. These assemblages include the Linchang (Permian-Triassic batholith and associated metamorphic rocks), Chengling-Mengliang (blueschist metasedimentary rocks), and Baoshan (thick section of Neoproterozoic to Jurassic sedimentary rocks) units within China.
At least in these three examples, whether or not the GPS velocity gradient is marked by faults, the shear gradients of faults do not appear to be controlled by rock types or preexisting structure. The only thing they have in common is that the GPS gradient passes through a crust at a high-angle to pre-existing structures. Of course, this is also true where the well-expressed Xianshuihe-Xiaojiang fault system cuts across rocks of the Yangtze platform. Alternatively, it might be tempting to suggest that the nature of the basement rocks of the Lanping-Simao somehow controls the localization of shear strain at the surface. This explanation fails, however, to account for the right-lateral velocity gradient west of the eastern plateau margin in the Songpan Ganzi terrane, because where the western part of the Xianshuihe fault cuts through the Songpan Ganzi, it is strongly localized. The mechanism that controls the presence or absence of localized faults expressed at the surface remains unknown.
One of the more sobering consequences of the examples described herein relates to geologic field work in older, inactive settings. Without recourse to geodetic techniques, it is quite possible that regionally significant deformation, if accommodated in a diffuse manner without the development of discrete or mappable structures, may be easily overlooked. Even where bent or apparently folded geologic markers are mapped, it is not always possible to definitively conclude that the mapped geometry of some marker in fact represents deformation imposed on an originally straight marker. A problem of this nature is well represented by field relations in the area north and east of the Himalayan syntaxis. In this region, both the apparent deflection and attenuation of geologic trends and the apparent deflection and lengthening of rivers have led workers (Dewey et al., 1989; Wang and Burchfiel, 1997; Hallet and Molnar, 2001) to speculate that this area represents a broad crustal shear zone, accommodating at least part of the northward movement of the Himalayan syntaxis relative to South China. A considerable component of shear in the area may not be marked by faults. Whether or not this interpretation is correct, unambiguous geologic field evidence of some mechanism or structure that could accommodate this deformation has been difficult to find.
Short-term GPS data in the eastern part of the Tibetan Plateau and adjacent regions in Sichuan and Yunnan provinces, China, have shown that velocity gradients cross continental crust, often with pronounced regional deformation fabrics at high angles to the gradients, without the development of through-going faults. Near the Xianshuihe-Xiaojiang and Dali fault systems, the velocity gradients are marked by active structures that are several million years old. The Xianshuihe-Xiaojiang fault system has accumulated at least ∼60 km of displacement. Strain associated with this fault system and the GPS velocity gradient passes through the Lanping-Simao tectonic unit without any obvious evidence as to how the shear is accommodated. A similar relationship exists in northeast Tibet west of the Longmen Shan; however, the total amount of shear strain accommodated there is uncertain. Paleomagnetic data from the Lanping-Simao fold belt also indicate previously unrecognized crustal shear. These observations are not dissimilar to those from other areas, where contemporary deformation cannot be directly related to extant, through-going structures, such as along the Eastern California shear zone. The results indicate that continental crust can accommodate tens, and perhaps more than a hundred kilometers of shear without developing through-going faults. From one of the examples from China given in this paper (Xianshuihe-Xiaojiang fault system), the shear was recognized during regional geologic analysis and later confirmed by GPS studies. However, it remains unclear how the shear was accommodated. In a second example from NE Tibet, the shear was not recognized until after short-term GPS measurements were made. It appears from these areas that continental crust can accommodate a large amount of shear deformation without developing obvious through-going structures, although the exact conditions under which the crust will deform in such a way remain the object of speculation. We contend that such unrecognized diffuse deformation may be common in many areas of active deformation as well as older, inactive geologic terranes. This adds an unknown but potentially large uncertainty to field study and tectonic analyses of continental deformation. The potential link between this deformation and the middle and lower crust, if not the mantle part of the Indochina lithosphere, requires, if possible, further, geophysical investigation.
This work is the result of cooperative project between scientists at the Chengdu Institute of Geology and Mineral Resources and the Department of Earth, Atmospheric and Planetary Sciences at the Massachusetts Institute of Technology. The data on which the results presented in this paper were based were obtained with the support of National Science Foundation (NSF) grants EAR-0003571 and EAR-8904096 and National Aeronautics and Space Administration (NASA) grant NAGW-2155 awarded to MIT and EAR-9706300 awarded to the University of New Mexico. Support was also provided by the Chengdu Institute of Geology and Mineral Resources.