On 24 September a shallow M7.7 earthquake rattled Pakistan. At least 300 people died and thousands of houses, most of them adobe, collapsed in Balochistan Province. The quake was felt as far away as Muscat (Oman) and New Delhi (India). Epicentral intensities reached up to IX. The earthquake appeared to be a strike slip event. Soon the media reported on an amazing effect of the quake – in roughly 400 km distance a new island appeared few hundred meters off Gwadar.
USGS reports on a maximum fault slip of ~13 m of the EQ and ~150 km rupture length, which in my opinion is a relatively large amount for this magnitude.
The new island of course caught almost all the attention. It is reported to be “18 meters high, 30 meters wide and 76 meters long” (National Post). The island is named Zalzala Jazeera and will most likely soon be washed into the sea, but it already has its own wikipedia article. Almost everyone agrees that a mud volcano was activated or created by the earthquake – since the island is a few hundred kilometers away from the epicentre, coseismic uplift can be ruled out. Witnesses also report water bubbles containing an inflammable gas and dead fish, which supports the mud volcano theory.
From the video the island appears rather solid and not very muddy. Other mud volcanos such as Lusi produce more liquid material. Even things like shear planes and cracks are visible on some photographs and people walk across it, so ‘mud’ might not be the right term.
Other things are strange, too. The location is far away from the earthquake epicentre and I guess it must be questioned if the seismic shock really triggered the rise of the island. Can liquefaction have played a role? Then I guess we would have seen liquefaction onshore in the Gwadar area, too. Plus, wouldn’t we have this phenomenon more often then? With closer but smaller earthquakes?
Mud volcanos as Earthquake Environmental Effects (EEEs)?
It is known that such mud volcanoes can be triggered by earthquakes. Similar effects have been observed after recent quakes in Pakistan in 1999 and 2010. Manga et al. (2009) provide a summary on earthquake triggering of mud volcanoes. In 1945, a M8.1 earthquake occurred at the Makran subduction zone off Pakistan and caused a tsunami that led to widespread devastation in the Gulf of Oman. It also created new islands. In a recent paper, we reviewed the impact of the 1945 Makran tsunami along the coastlines of the Arabian Sea (Northern Indian Ocean). From historic source we learned that “two islands appeared offshore, 180 miles west of Karachi as a consequence of the earthquake. Sondhi (1947) investigated the islands in January 1946. Here, a detailed account of four new islands in front of the Makran coast is given. These islands were composed of pale bluish grey clay as well as grey mudstone blocks, encrusted with marine organism. Sondhi (1947) describes the island in Gwadar west bay as 180*150 m in size with an oblong shape in a water depth of 7 m. The highest point was 7.6 m. The author refers to the second island (Hingol Island) as circular in shape, initially 180 m across and 7 m high. Gas eruptions were noticed in the vicinity of the island. Two more islands appeared in the Ormara west bay, 16 km offshore. The western island was initially 6 – 9 m high and 1.2 km long, the eastern island 18 -30 m high and 1.5 – 2.5 km long. Gas eruptions close to the islands were noticed. Sondhi(1947) further reports that a spectacular great fifi re, “thousands of feet high”, was caused by ignition of large volume of gas near Hinglaj. Pasni and Ormara were completely destroyed, Gwadar badly shaken (Sondhi 1947).” (Hoffmann et al., 2013). The 1945 islands look pretty much like our 2013 one:
So, earthquake triggering of mud volcanoes might be categorized as an earthquake environmental effect in the future, but further research is needed:
- Is there a clear distance-intensity/magnitude relationship?
- Is this a regional phenomenon only? We only know of EQ triggered mud volcanoes in areas where mud volcanoes have been known before. This might work for Azerbaijan and Pakistan, but likely not for Ohio…
- Does the size of the mud volcano correlate with earthquake magnitude, peak ground acceleration (PGA), shaking duration etc.?
- Does a sedimentological record of mud volcanoes allow to conclude on paleo-earthquakes?
Until now we have no news about other EEEs like primary surface ruptures, hydrological anomalies or other effects, but I am sure a local survey would be successful. Until then I’ll follow the news and see how the island is washed back into the sea.
- Hoffmann, G., Rupprechter, M., Al Balushi, N., Grützner, C. & Reicherter, K. 2013: The impact of the 1945 Makran tsunami along the coastlines of the Arabian Sea (Northern Indian Ocean) – a review. Zeitschrift für Geomorphologie. DOI:10.1127/0372-8854/2013/S-00134
- Manga, M., Brumm, M. & Rudolph, M.L. 2009. Earthquake triggering of mud volcanoes. Marine and Petroleum Geology 26, 1785-1798.
- Snead, R.E. 1964. Active Mud Volcanoes of Baluchistan, West Pakistan. Geographical Review 54-4, 546-560.
- Sondhi, V. P. 1947: The makran earthquake, 28th November 1945, The birth of new islands. Ind. Minerals 1: 147-154.
Victor Garcia | 2013-09-27|15:21 (UTC)
Yes! The structural pattern of the 1945 island looks very similar with the observed in the photos of the recent formed island. The presence of gas is an outstanding feature as well. I think the pattern of fractures could be indicating some kind of “intrusion” below the island…
jay patton | 2013-09-27|17:32 (UTC)
here is a great aerial view of the island. def either a mud volcano, but possibly a mud laccolith. some would argue they are the same… others would differ (the only difference being that a mud laccolith would not pierce the surface). given that the sea floor was uplifted with rocks exposed on the sea floor, i favor the laccolith interpretation.
mud volcanoes seem to be more depositional in that they emit mud to the surface, depositing this mud until it builds up some relief. this is not apparent based on the 1945 nor 2013 photos. if one looks at photos/images of what people generally agree are mud volcanoes, they successively deposit material on top of the ground/seafloor surface (building up topography in the form of an igneous volcano). this clearly did not happen in 1945 nor in 2013.
it could also be that some gas destabilization caused the uplift of the seafloor sediments. of course, that is not a mud volcano either. the anecdotal accounts in 1945 support this hypothesis.
thanks for this review christoph! this is a very exciting discussion. jay
jay patton | 2013-09-27|17:51 (UTC)
here is another update that supports the cathrate destabilization explanation for the uplift of the seafloor and discounts/nullifies the mud volcano explanation. ironically the landform is still referred to as a “volcano,” but this is likely due to the layperson audience of this web site (while being technically incorrect).
thanks again christoph, jay
jay patton | 2013-09-27|18:02 (UTC)
more here: http://earthobservatory.nasa.gov/NaturalHazards/view.php?id=82146&src=twitter-nh
Christoph | 2013-09-27|21:14 (UTC)
Thanks a lot Jay!
No, you’re right, a real mud volcano should look like this one in Balochistan: http://youtu.be/fiUyjqcKvMY
The term ‘mud volcano’ doesn’t fit the observations. But if it was a laccolithe, then we would now see the uplifted sea floor. Could this look like that?
So, we have a kind of methane-emission-related uplift of the sea floor? Is there anyone out there who has an idea on the bathymetry/the amount of uplift?
Victor Garcia | 2013-09-27|22:53 (UTC)
According Google Earth the bathymetry is between 0 to -1 meter. So, any “intrusion” in shallow marine deposits would be able to expose the sea floor…