Pakistan earthquake and its new island

A view of the new island off the coast of Pakistan
from NBC 

A magnitude 7.7-7.8 earthquake hit Pakistan today with an epicenter at about 15 km depth. It was well inside the borders of Pakistan and far from the ocean. Although only a few dozen (46) casualties have been reported as I write this (10:00 p.m. Central Time), the location of the quake is remote and it is likely that casualties will rise, with some estimates being in the thousands  (WAPMERR, The World Agency of Planetary Monitoring and Earthquake Risk Reduction). As this unfolds, an intriguing observation with speculations has hit the news--a mysterious island has risen from the ocean. It is reported to be 20-40 feet high and about 100 feet wide and lies about 350 feet out in the sea from the Gwadar coast (these numbers are very tentative--one report says that it's a mile out to sea, which seems more consistent with the picture that I've included above). It appears to be several hundred miles from the epicenter.
     Residents of Gwadar have reported that an earthquake in 1968 produced an island that remained for about a year before vanishing (but it's reported also to have occurred in the 1940's so I don't know if this is an inconsistency in historical reporting or if it has happened more than once). What could have caused this mysterious phenomenon? Speculation at the moment is our old culprit, liquefaction--the sudden transformation of wet, but solid, sediments to a weak mush upon shaking, just like jiggling quicksand turns it to a mush (for more on liquefaction and the general phenomena discussed here, see my book "The Dynamics of Disaster" to be released by Norton Press on October 21!!). When this happens, structures such as buildings can sink into the ground because it can no longer support their weight.  The liquified material can also squirt up through cracks in the ground forming sand- or mud-volcanoes, and the speculation here is that the island is a big mud volcano.
     This area of the Arabian Sea is known for its mud volcanoes.* Here the oceanic crust of the Arabian Sea is being subducted under Eurasia at a rate of about 4 cm/year.  A sediment pile has been building up on top of this subduction region to a thickness exceeding 6 km, a pile of wet gooey muck. The coastal region is known for mud volcanoes on land, and for the episodic formation of islands of them in the shallow waters off the coast. These are typically destroyed within months. Here's a seismic reflection image of a buried mud volcano from the reference * listed below:

Sometimes these mud islands appear in places that others have appeared before. One, called Malan island, appeared in 1999, unaccompanied by any noticeable earthquake, reoccupying a site of one that had been formed in 1945. It appears to have been driven by methane of bacterial origin.

One of the largest reported mud volcanoes is a complex 100-m high. It has been hanging around since at least 1840 and appears not to have changed much in the intervening decades. Methane of bacterial origin is persistently discharged into its crater mud lake. Eruptions of these gases often show periodicities of several hours that may be related to ocean tides. I'm not sure how the dating is done, but the article * below says that the presence of mud volcano activity in this region can be demonstrated for the last 460,000 years.

*G. Delisle, "The mud volcanoes of Pakistan," Environmental Geology, 46: 1024-1029, 2004.

Links between earthquakes and other geologic activity

Nature Geoscience (August volume 6(8), pp. 585-672) has a fairly long section ("a Web Focus") and a number of papers on geologic activity associated with or triggered by earthquakes. The introductory editorial reflects that in 1835 Charles Darwin voyaging on the Beagle experienced a large earthquake near Concepcion, Chile, and noted that within the hour a train of volcanoes in the Andes spouted out a dark column of smoke (though it would take a journey into Darwin's notes to determine whether he thought this was volcanic gas or perhaps debris from landslides. The implication in the editorial is that it was the former).
   

Illustration of the elastic rebound part of volcanic arc
subsidence after a megathrust earthquake

The first paper in this section (by Sigurjon Jonsson) summarizes the deflation of volcanic areas in response to the 2011 Tohoku and 2010 Maule (Chile) earthquakes. Both settings are at subduction zones (see figure), and the volcanoes that subsided were on the overriding plate. Prior to the earthquake, strain accumulates and compresses the overriding plate. During and after the earthquake, the overriding plate extends and subsides. However, subsidence beyond that which can be explained by this process is observed.
     In the case of the Tohoku earthquake, Takada and Fukushima documented 5-15 cm of subsidence at a distance of 150-200 km from the rupture earthquake, but no volcanic eruptions. They suggest that subsidence is caused by sinking of magma reservoirs and their warm host rocks through the colder surrounding crust. Prichard and colleagues noted that two earthquakes (1906, 1960) were followed by eruptions in the Andes within a year, but that no eruptions have been clearly associated with the 2010 earthquake. They were, however, able to document the 15 cm of subsidence, and suggest that hydrothermal fluids were released from hydrothermal systems surrounding the volcanoes in Chile during the 2010 quake, and that the escape of these fluids caused the volcanic areas to deflate.
     A second example of a proposed connection between earthquakes and geologic activity is more controversial: the Lusi mud volcano eruption. In 2006, mud erupted through and around a drill hole, flooding towns and displacing thousands of people.  Paul Davis summarizes a paper by Lupi et al. that proposes that the 2006 Lusi mud eruption in Indonesia (still continuing) was triggered by a M6.3 earthquake two days prior to the eruption and 275 km away.  Lupi et al. argue that strains, which are unarguably small at such a distance in homogeneous media,  were amplified by a downward concave layer of shale that acted as a parabolic reflector. Their simulations suggest that the stresses could have been about 100 kPa, five times higher than original estimates of 21 kPa. Such pressures, the assert, could have liquified the mud that resides at depth, resulting in the eruption of mud through the drill hole. This conclusion remains controversial (see discussion by R.J. Davies, et al., Earth and Planetary Science Letters, 272, 627-638, 2008).
     For a third example, Fischer et al. examine subduction zone earthquakes as triggers of submarine hydrocarbon seepage.  Offshore of Pakistan, the Arabian Plate subducts beneath the Eurasian plate. This is a region of intense seismicity, in particular a major earthquake (M8.1) occurred there in 1945. It occurred in an area where gas hydrates (methane clathrates) are present, and leakage of hydrocarbon gas is known to occur here. Methane and sulfates both occur in the ocean with sulfate being stable above about 5 mbsf, and methane at greater depths. The concentration of both goes to nearly zero at a depth known as the sulfate-methane transition (SMT). In a complicated chemical reaction, sulphate is consumed through anaerobic oxidation of methane (CH4 + SO24􏰀 ! HCO􏰀3 + HS􏰀 + H2O). Barium, being present in sea water, is precipitated at the SMT in so-called "barite fronts" and the abundance of barite can be used to reconstruct changes in upward methane flux.  The authors calculated that it would take approximately 38-91 years to produce the observed barite enrichments. This leads them to conclude that the barite production could have been initiated by the 1945 earthquake and an accompanying increase in methane flux due to release from the hydrates. If confirmed, submarine gas release triggered by earthquakes needs to be added to the list of processes that can add methane to the hydrosphere, and possibly to the atmosphere, in the carbon budget.




References: Takada, Y., and Fukushima, Y., Nature Geoscience, 6, 637-641, 2013.
Pritchard, M.E., Jay, J.A., Aron, F., Henderson, S.T., and Lara, L.E., Subsidence at southern Andes volcanoes induced by the 2010 Maule, Chile earthquake, Nature Geoscience, 6, 632-626, 2013.
Lupi, M., Saenger, E.H., Fuchs, F., and Miller, S.A., Lusi mud eruption triggered by geometric focusing of seismic waves, Nature Geoscience, 6, 642-646, 2013.
Fischer, D., et al., Subduction zone earthquake as potential trigger of submarine hydrocarbon seepage, Nature Geoscience, 6, 647-651, 2013.