We examine a century of seismicity and earthquake-related deformation on the western margin of the Indian plate in the Pakistan region of northern Baluchistan. Several catalogs of earthquakes for this region currently exist, but early catalogs in particular suffer from errors, incompleteness, inhomogeneity, and location bias. We form a new catalog of more than 1000 earthquakes using original sources to confirm macroseismic locations and assign MS magnitudes to earthquakes since 1892. In doing so we reveal a systematic east-northeast bias in locations caused largely by the uneven global distribution of seismic stations used in their determination. An appendix provides narrative accounts and historical references to 34 significant earthquakes in the region.
The pattern of seismicity in the past century shows activity over a 700-km-long, 200-km-wide segment of the plate boundary with predominantly strike-slip faulting to the west and thrust faulting to the east. At its narrowest near 29° N, transpression of the plate boundary is partitioned into reverse and strike-slip components separated by approximately 100 km. The M 7.3 1931 Mach earthquake (slip 1.1 m on a 40° east-southeast-dipping reverse fault) released fault-normal stresses that may have “unclamped” the subsequent M 7.7 left-lateral Quetta earthquake 4 years later. The northern Chaman fault system in the past century has been largely inactive, suggesting that this time period is not representative of long-term activity in the region and that up to 4 m of potential slip is currently available to drive one or more future M > 7 earthquakes. Despite triangulation installed in the late nineteenth and early twentieth centuries, no recent geodetic data are available to permit plate boundary velocities to be measured directly.