2001 AND EARLIER SURFACE FAULTING IN THE EPICENTRAL AREA OF THE 26 JANUARY 2001 BHUJ, INDIA EARTHQUAKE

The 26 January 2001 Bhuj, India earthquake (M7.7) was generated by northward thrusting on an east-west fault. Reconnaissance teams from USA, Europe, and India stated that the mainshock fault plane did not rupture the surface. However, in May 2001 Dr. Mahesh Thakkar (Lalan University, Bhuj) and I mapped a >1 km long fault scarp that trends east-west, approximately where the mainshock fault plane would project to the surface. This scarp was not a lateral spread or other gravity failure, and thrust Cretaceous bedrock over Quaternary deposits. Where exposed in a stream cut, the fault dipped 13 degrees south and displaced a Holocene (?) channel fill 22 cm. Geomorphic evidence for pre-2001 displacements exists on this same fault, and could be investigated by future trenching. The largest ground deformation was a large lateral spread at Budharmora village. A French team trenched the toe thrust of this landslide, thinking it might be a primary surface rupture. I concur with other US geologists that this feature is a landslide, but the French trench also exposes a pre-2001 normal (?) fault zone in the trench that was not reactivated. We also mapped a NNW-trending dextral fault zone similar to the Manfara fault. Our unnamed dextral fault north of Bharodia village had 10-20 cm of displacement and a unique surface morphology which I term a "welt", a low, smooth linear mound with tension or dextral cracks at the crest, and a small thrust scarp on one margin. In the loose sandy soils of the epicentral region this landform will be obliterated in just a few years. Might similar structures have formed in Holocene earthquakes in the central USA, and then were obliterated? These structures form a new category of surface ruptures that may exist in stable continental regions elsewhere, and pose unrecognized hazards to infrastructure elements.