Lava flows come in all sizes. The Reykjanes eruption produced flows of several kilometers length. Other volcanoes can do ten times that length, although we do not have a lot of recent experience with such long flows. Flood basalts can do hundreds of kilometers, and volcanoes on Venus manage thousands of kilometers – helped by the fact that the lava barely cools under its hothouse conditions. On the other hand, some flows only manage a few hundred meters. And one lava flow is known for being the shortest in living memory: it had a length of only 8 meters. There is a story here.
It all happened in a mountainous region of Pakistan. The country is a fractured one. It lies on one of the boundaries between the Eurasian and the Indian plate, and carries collateral damage from the collision of the two. Pakistan borders India, China, Afghanistan and Iran. (Not all the borders are well defined.) The eastern parts of the country are the fertile delta of the Indus river, among the major rivers of the world, and the origin of the name ‘India’. East of the Indus the land is dry and bare: this is the Thar desert which extends into India. In the far north is the final flourish of the Himalayas, where the gorge of the Indus defines its westernmost extend. The Himalayas took the brunt of the India-Asia collision: it is the crumple zone of a continent. The western half of Pakistan is very different, with long curved mountain chains along the edge of the Indian plate.
Pakistan’s mountains are impressive, and that is on a world-class scale. There are over 100 peaks above 7 km, including the famous but nameless – and deadly – K2. (If you wonder why the second highest mountain in the world has no name: it is in effect invisible from any inhabited location, so lacks a local name. The term ‘K2’ means it is the second peak in the Karakorum range, from a survey done from a great distance at a time when the true height was not known. For lack of an alternative, the name stuck.) (If you wish to climb it, be aware that the death rate among its climbers is a staggering 25%. Nanga Parbat is almost as bad, at 22%.) All of these 7-k’s are in the north, in the Himalayas, the Karakoram (along the northern border of Pakistan, where K2 is located) and the Hindu Kush (which is mostly in Afghanistan but extends into northwestern Pakistan).
Many more, but lower, mountain ranges occupy the west of the country. The map below shows the names and location of the various ranges. Heights vary from 1 to 5 kilometers. The curved chains of the Sulaiman range give spectacular satellite images. They form a fold belt at the edge of the Indus delta; the peculiar shape is related to some rotation of the Indian plate in its collision with Asia. Everything here is India’s fault.
Among all the natural wonders, there is one notable lack. Sadly, Pakistan does not have volcanoes. This is not entirely true: there are some mud volcanoes, if that is your thing. There is an extinct, pleistocene volcano in the south, near Iran: Koh-e-Sultan. It is located in the Chagai arc and was active perhaps 2 million years ago. Near it, Neza-e-Sultan is an old magma pipe which sticks out of the ground. It takes a long time to erode volcanoes down to their conduit. This clearly is old. Various reports claim that eruptions here have been dated to as recent as 90,000 years ago. But that claim comes from an unpublished PhD thesis and we can consider it as ‘unconfirmed’ – or we may assume that it is unpublished for a reason. This volcano looks millions of years old. And ever since it went extinct, Pakistan has had to live with only mud volcanoes.
But this volcanic deficiency suddenly came to an end. It happened in January 2010. Pakistan lacks volcanoes no longer. Or does it?
Tor Zawar is a steep mountain side some 40 km northeast of the city of Quetta. A major earthquake (M6) had hit the region two years earlier, in October 2008. A small earthquake was reported in the region on 27 January 2010. Following this, reports began to emerge of smoke emerging from the ground and black, molten rock reaching the surface. This was surprising, as the region had never been seen as volcanically active. The black melt was highly viscous and it remained within meters of the eruption site. The volume that was erupted amounted to no more than 10m3. But it cannot be denied that molten rock coming up to the surface, however minor, is by definition volcanic. It beat the mud volcanoes hands down. Scientists began the arduous journey to the dry, mountainous region. They arrived a few days after the event. The district governor got there first, the day after the event: he confirmed that the eruption had begun with ‘chemical gasses’ (a bit more specific would have been nice -another report mentions methane) followed by the ejection of a molten rock, 1 meter wide.
This was a unique event. There had never been an eruption in the area, and in fact there was no volcano within hundreds of kilometers. No one would have predicted an eruption here. But there is in fact a volcanic legacy in the region, albeit a very ancient one.
Once, this was within the Tethys ocean. India’s venture into Asia was part of the closure of the Tethys, some 50 million years ago. (It was not the first, nor the last, terrain to make the jump. A series of microcontinents and volcanic islands had done the same, arriving ahead of India.) In this particular region, the rocks date from the Jurassic to the Eocene (the time of India’s arrival). They are mainly sedimentary, sand and debris deposited in the Tethys ocean. But in the late Cretaceous, 70-74 million years ago and ahead of India’s arrival at the shores of Asia, there was a period of basaltic volcanism with submarine pillow lavas and tuff. The remains of these volcanic rocks, interspersed with sediment (sandstone and mudstone) are now found on land, in the region of Tor Zawar. At the surface there is a thin band of volcanic rock stretching east to west over a length of 10km and a width of 1-2 km. Satellite images show the band of dark rock well. It is called the Bibai volcanics and lies between two thrust faults, the Bibai and Gogai thrusts. Tor Zawar is at the southern rim of this band, and this where the eruption took place. Further north, the mountains show sills, attributed to the same ancient volcanic event. They consist of dolerite, typical for solidified basaltic dikes and sills. The Bibai volcanics may have been a local intrusion but it occurred over a wider region than just the thin band now at the surface. Was this related to the eruption? It would be a record long lead time!
The molten rock
The locals reported that emission of black, molten material started along with tremors occurred on the night of 27 January 2010. Steam was continuously emitted from six fissures, and rock fragments were too hot to handle with bare hands. Heat was still rising from the site during the 3 days of observation.
The eruption had caused a series of fissures. But in keeping with the size of the eruption, these were only meter size! Six fissures were reported by the locals. The longest was measured at 1.33 meters and the shortest was a mere 30 centimeters. The width varied from 15 to 50 cm. The lava came from the two largest fissures, and had crept along the surface along a distance of 8 meters, in a flow that was 2 meters wide and 15 cm thick before it solidified. Presumably this followed the orientation of the fissures. In the centre it had formed small concentric circles. Here, a small scoria cone had formed: small meaning 60 cm tall! This was truly a non-Jesperian event: a miniature volcano. But in this country void of volcanoes, no one minded. You have to start somewhere, and for the local newspapers this was plenty.
It was plenty for the locals as well. People came to this isolated region, including the local governor and the scientists. Stories were told. Not all were equally believable: the one about fire and flames on top of the mountain overlooked the fact that this did not happen on the top of the mountain but near the bottom. People wanted a souvenir, and thus much of the rock quickly disappeared in the hands of the local people.
A trench was dug to find the origin of the lava, a few days after the eruption. The vent was found to be a tube only 5 cm wide, going down 1 meter. At this depth it connected to a cone-shaped cavity some 60 cm wide. a shiny, black, fine coating on the walls made it look like an oven. Even at this time the sides of the cavity were still too hot to touch. Bushes caught fire when placed close to the vent: temperatures were estimated to be up to 300 C.
Nothing remains. Whilst the rock was carried off by the locals, the cone was removed by the scientists. It now appears to be in a museum, an unusual place for an entire volcanic cone. I tried to find the exact location of the eruption on satellite images but even with a photo of the location at hand no remnant could be recognized, except possible a whitish spot. Pakistan had its volcano only for a few days.
Two rocks from the minute scoria cone made it to the laboratory in Cardiff, and were analyzed there. The composition of one was andesitic-basaltic and the other was trachy-basalt with a bit lower SiO. Based on other elements both rocks were deemed to derive from an alkaline basalt, in spite of the andesitic affinity. The composition was similar but not identical to those of the some of the sills and the Bibai volcanics. But it seemed unlikely that a magma chamber would have lain undisturbed for 70 million years!
The eruption site was close to an electricity pylon, and the cable reportedly had snapped, apparently in the earthquake. ‘Close’ meant a distance of some 7 meters. But this could not have caused the melt. The cable had been overhead, not buried, and any melt from its short-circuit would have been on the surface. In fact, there are cases known where a broken electricity cable melted surface rocks. But at Tor Zawar the molten rock had definitely come from below. Also, the basaltic composition of the analyzed stone would have required a very high melt fraction of the surface rocks, much higher than expected from a short-circuit. So that suggestion did not work.
The problem remained that the two analyzed rocks had different compositions. The suggestion was made that the more evolved rock could have had some sand melted into the rock. That pointed at crustal contamination. Subtracting that, the compositions could be fitted with an origin 60 km deep, with a very low melt fraction of 1%. The Bibai thrust fault could have allowed the melt to come to the surface, perhaps after the 2008 earthquake. But in this collision zone, 60 km deep is still within the lithosphere, and there is no indication of excess heat. But if magma had indeed managed to travel up from so deep, then further small eruptions could well be possible.
That was a prediction made in August 2010, when this first analysis was written up. It turned out to be well-timed. In January 2011, a second eruption was found, 300 meters north of the previous one. This was even smaller (and so immediately took over the record for the smallest volcanic eruption even recorded). But it was so small (3 m3, less than half of the 2010 eruption) that it had not in fact been recorded. The exact date of this eruption is therefore not known: the ejecta were discovered in January but could have been made months earlier. It was extraordinary to think that after 70 million years of dormancy, the region would re-activate with two such miniature eruptions within one year!
This was in fact so extraordinary as to be considered utterly implausible. There had to be another explanation. As far as we know, volcanic eruptions arise from melt from geothermal heat at depth. The magma then rises because molten rock is less dense than solid rock. How could this be different in Pakistan?
Problems with the original interpretation were spotted. The lithosphere here is the Indian craton; the thrust faults are near the surface, but do not go anywhere near the depth of 60 km that had been suggested. The deep origin of the magma seemed unlikely.
The second eruption in 2011 provided the all important clue. For it too was next to an electricity pylon, in fact one on the same line. But the main electricity cable was not damaged – in fact it was now discovered that reports that the line had broken on the first eruption were wrong. Whatever had happened had involved the pylons but not the electricity cable.
In the 2011 eruption, basaltic melt was found in two places: at the base of the pylon, and at the base of one of the support cables of the pylon. Going back to the 2010 event, it was found that photographs showed that the hole that had been dug around the conduit also had a supporting cable going into it. The authors of the original paper had not known this because they had not been at the site: they had just been send the two rocks and descriptions. A lack of local knowledge, and possibly some inaccurate information, had obscured the facts.
So if it wasn’t geothermal heat and if it wasn’t the electricity cable, what had happened? The earthquake had likely been a red herring: earthquakes this small size (M3.9) are frequent in the region, none have ever produced magma, and the 2011 event did not relate to an earthquake. There is one other source of energy available, and although it does not normally produce magma, it was the only likely candidate. The magma was a fulgurite.
The word fulgurite comes from latin: it describes fossil lightning. (Fulgur means lightning.) It may not be obvious how lightning can fossilize, but this can happen when lightning hits the ground. The bolt can travel into it and fuse sand or other substance along its path. The fused material forms the fulgurite fossil. In some cases the fused tubes can reach meters deep.
The suggestion was made that at Tor Zawar, the lightning had struck the pylon and traveled through the metal structure and support cable into the ground. Rather than most of the energy being used near the surface, the metal structure channeled the bolt towards the foundations or the mounting point: all the energy was deposited there. And that was a considerable energy. An average lightning bolt carries around 3 giga-Joule, while the strongest can carry 100 times more. Temperatures along the path of a lightning bolt can reach 30,000 C (although that is in the atmosphere, not the solid ground).
Melting 10m3 of basaltic rock requires around 25 GJ. This was the approximate volume of the 2010 eruption. So it can have been caused by a strong lightning bolt if all the energy was transferred underground, with no way of escape other than into the rock.
Fulgurite fused rock remains where it forms, along the path of the lightning strike. It does not form flowing lava. But in this case, all the melt happened 1 meter below the ground and the molten rock reacted as any magma would do: look for an escape, upward. Hence the fissures and hence the lava flow.
So why did the analysis of the rocks point at an origin 60 km underground, rather than the 1 meter where it really formed? The pylons were mounted on outcroppings of the ancient volcanic rocks, on the edge of the Bibai belt. The rocks that were melted were those volcanic rocks (perhaps with some added cement), and the lava that came out was the old solidified pillow lava, remelted by lightning. When the extruded lava was analyzed, it correctly found that it had come from 60 km depth. But that had happened not in 2010, but 70 million years earlier. This lava was originally produced at the bottom of a long vanished ocean; it predated the arrival of India by 20 million years. Even the Deccan traps were still in the future, and dinosaurs were still roaming the earth. The scientists got it spot on – just with a minor deviation on the date.
We know that volcanic eruptions can induce lightning. The dust column attracts lightning like flames attract moths. But now we have learned from Tor Zawar that lightning can also induce a volcanic eruption. Who would have thought.
the volcano that wasn’t?
But was this a volcano? If we stick with the definition that a volcano is where molten rock from below reached the surface, then this is a volcano. If we require the magma to be produced deep underground by geothermal heat (something not normally specifically listed), then it isn’t. We may also think of volcanoes to erupt repeatedly, but there are many volcanoes that only erupt once. So call this a volcano if you want. Let Pakistan have its first eruption since the Pleistocene.
Albert, March 2023
Eruption of basaltic magma at Tor Zawar, Balochistan, Pakistan on 27 January 2010: geochemical and petrological constraints on petrogenesis: A. Kerr et al. Mineralogical Magazine, December 2010, Vol. 74(6), pp. 1027–1036
Comments on the eruption of basaltic magma at Tor Zawar, Balochistan, Pakistan on 27 January 2010, with a discussion of the geochemical and petrological constraints on its petrogenesis : A. Kassi et al. Mineralogical Magazine, June 2012, Vol. 76(3), pp. 717–723
Occurrences of rock-fulgurites associated with steel pylons of the overhead electric tr erhead electric transmission line at T ansmission line at Tor Zawar or Zawar, Ziarat District at District and Jang Tor Ghar, Muslim Bagh, Pakistan: A. Kassi et al. Turkish Journal of Earth Sciences, 2013, Volume 22 Article 5
Bulletin of the Global Volcanism Network, vol. 37, no. 3 (March 2012) https://volcano.si.edu/showreport.cfm?doi=10.5479/si.GVP.BGVN201203-600600