Saturday, December 1, 2007

Earthquakes in South Maharashtra

The Amateur Seismic Centre, Pune website keeps a track of earthquake activity in south Asia. From their list of recent earthquakes these four:

24th November Chandoli region Mag: 4.3 10.57 a.m
24th November Chandoli region Mag : 3.8 11.35 a.m
24th November Chandoli region Mag: 3.8 12.35 p.m
24th November Chandoli region Mag: 2.8 3.53 p.m

Chandoli region lies in southern Maharashtra, which shows elevated levels of seismic activity as seen in figure below.

Source: Amateur Seismic Centre, Pune;

Two distinct clusters are seen in south Maharashtra, one centered along the western ghats in Satara and Kolhapur districts and another to the east around Latur. Both these areas have experienced devastating earthquakes in recent times. In 1967, the Koyna earthquake caused few hundred deaths and structural damage and then in 1993 the Killari earthquake in Latur district caused thousands of deaths and enormous structural damage.

Earthquakes as we learn in school are caused mostly in and around plate boundaries. There rock masses along faults are under frictional stress, i.e. friction between the rock masses on either side of the fault inhibits slip or motion. Plate movements and pore pressures keep adding shear stresses to this rock mass to the extent that the shear stress exceeds a critical value that overcomes the frictional stress. When that happens the rock masses suddenly lurch past each other releasing energy and causing an earthquake. But southern Maharashtra is far away from plate boundaries. What are causing these intra-plate earthquakes? Figure below (source: Sheth 2006) shows zones of structural weakness in south Maharashtra.

The Dharwar trend refers to very ancient zone of crustal weakness, one that formed during orogenic activity and associated rupturing and faulting of the crust in Precambrian times maybe around 2 billion years ago. This zone of weakness is buried in south Maharashtra below the Deccan volcanic pile (brown region) which is around 65 million years old. When plate motions caused India to split from Africa first, then Madagascar and then Seychelles, a process that occurred episodically starting around 170 million years ago and continuing to about 60 million years ago, the tensional forces involved created a series of faults and zones of weakness along the western margin of the Indian continent. Preexisting zones of weakness were more likely to fail during crustal stresses, hence the coincidence of the Dharwar trend with later zones of weakness such as the Koyna fault or rift zone. The Kurduvadi rift also probably coincides with a deep zone of crustal weakness, currently buried under the Deccan traps. The Killari earthquake does not coincide with the Kurduvadi "rift" but may represent a similar zone of crustal weakness further to the east of the Kurduvadi zone. The enhanced seismic activity resulting in the clusters of earthquakes in south Maharashtra occur in the vicinity of these zones of structural weaknesses and need to be understood within the context of the tectonic forces acting presently within these zones.

Beginning around 55 million year ago, the northward moving Indian plate collided with the Asian plate. The northward movement was met by resistance from the Asian continent and so the Indian continent has been under compressional stresses since. At first, the thought that a continental collision in northern India can have an effect on earthquakes in south Maharashtra about 1500 kms away seems absurd, but two types of evidence are available. The first is the nature of the earthquakes themselves. The mechanism of earthquakes are investigated by deriving a fault plane solution, which is an analysis of particle motions in the earth during the passage of the P or longitudinal waves which radiate in all direction from the hypocentre. The hypocentre is the point of origin of the earthquake and coincides with the point in the fault where the first slip of the rock masses occur. So, a fault plane solution tells us the direction of motion of the rocks along the fault. Such as analysis on the Koyna earthquake (Rao et al 1974) and the Killari earthquake (Mandal et al 1997, Gahalaut et al 2002) shows that the slip in case of Koyna is a strike-slip motion (rocks move past each other with very little vertical motion) while in the case of Killari it is a reverse thrust motion (fault block on the hanging wall move up relative to the footwall). Both these types of motions indicate compressive stresses.

The second type of evidence is from the Neogene deformation (23 million years to recent) of the Indian peninsular region around the Deccan traps. In the vicinity of Mahabaleshwar , the Deccan traps are folded into a south plunging regional anticline (Widdowson & Cox 1996). To the north in the Satpura, the high topography of the Pachmari hills has also been interpreted as being due to recent crustal warping and uplift (Venkatakrishnan 1984, 1987). And in the south, crustal arching along the Mangalore -Chennai axis (Subrahmanya, 1994) has resulting in the NE and SE flow of rivers on either side of the arch (see figure below; source Sheth 2006).

All this deformation has been interpreted to be of Neogene age on geomorphological evidence such as youthful steep topography and antecedent drainage (evidence summarized in Sheth 2006) and indicates a compressional stress regime in the Indian peninsula.

The fundamental zones of crustal weakness along which south Maharashtra earthquakes occur originally developed in the Precambrian and are 2-3 billion years old. These zones have been periodically re-activated first as tensional fractures and faults when India broke up from Madagascar and Seychelles and later as strike-slip and reverse faults as manifestations of the compressional stresses periodically building up in the Indian peninsula. There is a very ancient geological control on the Koyna and Killari earthquakes.

Finally a few thoughts on that nagging question: Did the Koyna dam cause the Koyna earthquake in 1967? The popular notion at that time was that the impounding of water in the reservoir imposed stresses on the crust caused the earthquake. But calculations of rock mechanics show that the increase in pore pressures at depth due to loading of even a few hundred metres of water is very small, orders of magnitudes less than required to cause rock failure by itself. However, if the tectonically created stress along faults are already at a critical level, then a small increase in pore pressure due to impounding of water may act as a tipping point causing slip along the fault. This seems to be the case at Koyna. The deep crustal faults appear to be at a critical level of stress and the pattern of seismicity correlates with the loading and unloading cycles of the reservoir (Pandey and Chadha 2003). Under the right circumstances of crustal stresses, dams may induce earthquakes but are not a primary cause. This is of significance in understanding the risk imposed by the various proposed dams in the Himalayas. There faults and tectonic instability is unavoidable. Understanding the stress regime along individual faults in the vicinity of the proposed dams is of critical importance.


Gahalaut V.K, Kalpna and P. S. Raju 2003; Rupture mechanism of the 1993 Killari earthquake, India: constraints from aftershocks and static stress change; Tectonophysics, Volume 369, Issues 1-2, 3 July 2003, Pages 71-78

Mandal P, Manglik A and Singh R.N., 1997; Intraplate stress distribution induced by topography and crustal density heterogeneities beneath the Killari, India, region; JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 102, NO. B6, PAGES 11,719–11,730

Pandey Ajeet P and Chadha R. K. Surface loading and triggered earthquakes in the Koyna–Warna region, western India; Physics of the Earth and Planetary Interiors, Volume 139, Number 3, 31 October 2003 , pp. 207-223(17)

B. S. R. Rao, T. K. S. Prakasa Rao and V. S. Rao 1974; Focal mechanism study of an aftershock in the Koyna region of Maharastra State; Journal Pure and Applied Geophysics

Subrahmanya, K. R., 1994, Post-Gondwana tectonics of peninsular India: Current Science, v. 67, p. 527-530

Venkatakrishnan, R., 1984, Parallel scarp retreat and drainage evolution, Pachmarhi area, Madhya Pradesh, central India: Journal of Geological Society of India, v. 25, p. 401-413

Venkatakrishnan, R., 1987, Correlation of cave levels and planation surfaces in the Pachmarhi area, Madhya Pradesh: A case for base level control: Journal of Geological Society of India, v. 29, p. 240-249

Widdowson, M., and Cox, K. G., 1996, Uplift and erosional history of the Deccan Traps, India: Evidence from laterites and drainage patterns of the Western Ghats and Konkan coast: Earth and Planetary Science Letters, v. 137, p. 57-69

1 comment:

  1. very interesting and since I've just visited the region, found your blog to give a very interesting insight into its geological history.