Abstract The stress indicators describing the recent (provided by active tectonics framework) and palaeo-stress (provided by micro-fault kinematics and volcanic cluster) patterns show the scale and temporal changes in stress states since the beginning of Arabian–Eurasian collision. The recent stress derived from the active fault kinematics in the Lesser Caucasus and adjacent area corresponds to a strike–slip regime with both transtension and transpression characteristics. The kinematics of active structures of various scale are conditioned by tectonic stress field with general north–south compression and east–west extension. The distribution of Neogene to Quaternary volcanic cluster geometries and micro-fault kinematic data evidence the time and orientation variability of the stress field since the beginning of the Arabian–Eurasian collision. In addition to the general north–south compression orientation, two other – NW–SE and NE–SW – secondary orientations are observed. The first one was dominant between the Palaeogene and the late Early Miocene and the second one has prevailed between the Late Miocene and the Quaternary. Since the continental collision of Arabia with Eurasia the tectonic stress regime in the Lesser Caucasus and adjacent area changed from compression (thrusting and reverse faulting) to transtension-transpression (strike–slip faulting with various vertical components).

In this century, western Mongolia and the area adjacent to it in China have been one of the most seismic intracontinental regions of the world. Four earthquakes with magnitudes (M) ≥ 8 have occurred. Average displacements of several meters along ruptures more than 100 km long characterize all of them. The dominant style of faulting for each was strike-slip: left-lateral on easterly trending planes in the 1905 Bulnay and Tsetserleg earthquakes and in the 1957 Gobi Altay earthquake and right-lateral on a north-northwesterly trending Fu-yun fault in 1931. The ruptures associated with these earthquakes, with most other, smaller earthquakes, and with one older great earthquake suggest that western Mongolia is undergoing conjugate strike-slip deformation. Equivalently, the region undergoes northeast-southwest shortening and northwest-southeast extension. The component of shortening can be seen as a manifestation of the convergence between India and Siberia. The component of extension seems to mark a transition from an area of largely crustal shortening in China to another, the Baikal Rift system, where crustal extension is dominant. The average rate of seismic deformation in western Mongolia in this century consists of 49 (± 15) mm/a of northeast-southwest shortening and 40 (± 12) mm/a of northwest-southeast extension. Such high rates imply strongly that the twentieth-century seismicity has been abnormally high. Moreover, crude estimates of average recurrence intervals for great earthquakes on the Bulnay fault and in the Gobi Altay region are about 1,000 yr. Approximate Holocene or late Quaternary average slip rates on these faults are a few millimeters per year, suggesting that Mongolia is being sheared left-laterally with respect to Siberia at about 10 mm/a (between 5 and 20 mm/a). A similarly crude estimate for right-lateral slip along the northwest-trending Mongolian Altay is also 10 mm/a. We suspect that this right-lateral shear is a manifestation of the left-lateral regional shear parallel to east-west planes and of counterclockwise rotation of both the Mongolian Altay and the strike-slip faults within the range. Correspondingly, the eastward translation of western Mongolia with respect to Siberia manifests itself as crustal extension and rifting along the Hövsgöl and Baikal rift zones. In the Hangay, the broad upland in the interior of western Mongolia, scattered minor normal faulting with no obvious preferred orientation appears to be the common style of deformation. This area seems to be underlain by the same relatively hot upper mantle that underlies the Baikal rift system. Thus, as others have suggested, the collision between India and Eurasia does not appear to be the only cause of the active tectonics of western Mongolia. The perturbations to the stress field in the crust resulting from the emplacement (and upwelling) of hot material beneath the Baikal area and the Hangay and Hövsgöl uplands also play a key role in the active tectonics. Finally, the active deformation in western Mongolia appears to be young. Some surface faulting bears no obvious relation to the present topography. The very flat summit of Ih Bogd, the highest peak in the Gobi Altay, may have risen from the surrounding lowlands since only 1 Ma. Thus, the rapid deformation in western Mongolia seems to have begun tens of millions of years after India collided with Eurasia, perhaps as recently as a few million years ago.