1997 Investigations
We studied a total of five exposures (hand-dug trenches and cleaned stream cuts) along the main Gurvan Bulag thrust, and one exposure across a subsidiary fault (one of several backthrusts discovered earlier this year by A. Bayasgalan and J. Jackson north of the main thrust fault). All exposures showed stratigraphic evidence of multiple ruptures such as faulted colluvial wedges and fault terminations (Figure 5A, Figure 5B).
Figure 5A: View looking east at 2-m-high exposure of part of Gurvan Bulag fault zone in stream cut near western end of 1957 rupture. Faults dip north (left) and juxtapose well bedded alluvial units (upper left) over massive colluvial wedge units (lower right). Colluvial units formed in response to earlier fault displacement and were faulted in 1957. Little displacement occurred on these traces in 1957; most 1957 motion occurred on fault trace a few meters south (right) of photograph. (Photograph by C. Prentice, 1997)
Figure 5B: View looking west at exposure of part of Gurvan Bulag fault zone in stream cut near western end of 1957 rupture. Faults dip north (right) and juxtaposes alluvial gravel (upper right) over massive colluvial wedge units (lower left). Distance between vertical string lines is 0.5 m. Colluvial units formed in response to earlier fault displacements and were faulted in 1957. Little displacement occurred on these traces in 1957; most 1957 motion occurred on fault trace a few meters south (left) of photograph. (Photograph by C. Prentice, 1997)
We collected samples from all but one of these exposures for TL dating that we expect will provide constraints on the timing of the penultimate and pre-penultimate events, and radiocarbon samples from one site that will provide additional constraints on the timing of the penultimate event (Figure 6).
Figure 6: Eastward view of backthrust within Gurvan Bulag fault zone exposed in trench. Fault dips southward (right). Distance between nails along horizontal string line near bottom of photo is 0.5 m. Red flag marks 1957 ground surface (and upper end of fault). Dark, organic-rich horizon on hanging wall represents ground surface at time of penultimate earthquake. In 1957 this horizon was faulted over colluvial wedge (light-colored material in footwall below red flag) that formed in response to penultimate earthquake. Note that dark horizon in hanging wall is not present in footwall, indicating strike-slip displacement. On opposite wall of trench (not shown) dark horizon is only present in footwall indicating component of left-lateral slip. (Photograph by C. Prentice, 1997)
In addition, we conducted reconnaissance mapping of fluvial terraces and uplifted alluvial fan surfaces that record deformation associated with this fault zone. We collected samples for TL dating from the sediments underlying these surfaces, and measured profiles of the risers between the surfaces and across the fault scarp disrupting them.
The analyses of our TL and C14 samples will allow us to compare the timing of prehistoric earthquakes along the Bogd fault (Schwartz et al., 1996) with those on the Gurvan Bulag thrust to determine whether these two zones ruptured closely together in time during the penultimate and pre-penultimate events, as they did in 1957. In addition, one of our excavations showed clear evidence of at least a small component of left-lateral strike slip associated with the 1957 rupture of the Gurvan Bulag zone (Figure 6). Also, based on the geomorphology of the scarp at several of our sites combined with the stratigraphic evidence for multiple ruptures, we suspect that at least some measurements of 1957 scarp height may actually be measurements of composite scarps produced by multiple ruptures. This demonstrates the difficulty of determining thrust-fault scarp heights associated with single earthquakes in alluvial gravel. The whole scarp can shatter, giving the appearance of fresh rupture to the entire face of a pre-existing scarp.
In addition to our paleoseismic studies, several of us (C.P., K.K., K.B, and A.O.) were delighted to have the opportunity to hike to the summit of Ih Bogd (3957 m), the highest peak in the Gobi Altay range. Once on top, we enjoyed wandering around on the nearly flat summit plateau, throwing snowballs and observing some of the best patterned ground any of us have seen outside of textbooks. We also observed the remarkable headscarp of the huge (>130 million cubic meters) Bitüüt coseismic (1957) landslide (Kurushin et al., in press), as well as views to the north of the Hangayn Nuruu, to the south of the Gobi Tien Shan, and to the east of Orog Nuur and Baga Bogd (Figure 7).
Figure 7: View looking eastward from Ih Bogd summit plateau. Patterned ground in foreground, Baga Bogd in far distance. Kelvin Berryman for scale. (Photograph by C. Prentice, 1997)
ACKNOWLEDGMENTS
Many thanks for field support to: Altan, Bata, Chimbat, Erden, Ganboldt, and Tume, for driving, vehicle repair, trench digging, cooking and frisbee games. We wish to thank M. Ganzorig of the Informatics and Remote Sensing Center, Ulaanbaatar, Mongolia, for logistical support. Support for this project is provided by the NRC, through funding to David Schwartz at the USGS, Menlo Park.
