Another Hot Spot: Mt. KARTHALA, Comoros

The pit crater Changouméni (created 1918?) in the northern lobe of the Karthala caldera is filled to the brim with new lava. 17/01/2007. (© OVK)

Just a humpback emerging from the Indian Ocean: Mt. Karthala (left) on Grande Comore. On the right is the inactive La Grille volcano. (Google Earth)

You might have heard the name before, but could you point out the Comoro Islands on the map, offhand? This handful of volcanic islands is strewn into the Indian Ocean between the island of Madagascar and Mozambique on the African continent. The youngest and largest in the nation of Comoros, Grande Comore (also known as Ngazidja), measures 27km by 67km and boasts no less than two massive shield volcanoes. The southern two-thirds of the island are dominated by the huge and better-known Mt. Karthala that created a stunning dark lava landscape offset by incredibly white beaches and turquoise seas. A plateau averaging 650m high connects Mt. Karthala with La Grille volcano in the north. The oldest rocks from a (or “the”?) pre-Karthala volcano are found at the lower SE rift zone and form the Massif du Badjini, a peninsula at the SE tip of the island.

The Comores islands are roughly W-NW aligned and extend over 270 km in the Western Indian Ocean, on the northern edge of the Mozambique channel. The channel has ben formed by the opening of the Somali and Mozambique Basins in Middle-Jurassic to Upper-Cretaceous times. This tectonic rifting dislocated the micro continent Madagascar to the south, away from the African continental plate.The Comoros hotspot has been recently proposed to be one of 11 “primary plumes” on Earth. It overlaps with the offshore East African Rift System (EARS).

The Comoros hotspot in relation to the main faults of the western and eastern branches of the East African Rift System (from Chorowicz, 2005 and Macgregor, 2015)

Less than 200 km away from the offshore EARS, the Comoros archipelago formed simultaneously as a northwestward prograding volcano chain, younging in the opposite direction compared to the offshore rift. This chain of volcanic islands and seamounts becomes progressively older from the still-active volcanism at Grande Comoros toward the 10–7 Ma old mafic volcanic rocks in northern Madagascar. Although there are different interpretations of the origin of the Comoros volcanism, it is commonly regarded as hotspot related.

Similar to Hawaii’s journey across the Pacific, this hotspot migrated from the Seychelles over Northern Madagascar towards its present position underneath the island of Grande Comore. However, the chemical composition of the magmas is somewhat different: it appears that portions of lithospheric mantle, remaining after the southward Madagascar migration, must be involved in the genesis of magma underneath the Comores. The Seychelles – N-Madagascar – Comores alignment might have a magmatic melting mechanism that is similar to volcanic hot lines, like the ‘Cameroon Hot Line’, that cannot be explained by pure Hawaiian hot spot migration, but where mantle plume involvement is undisputed.

Simplified volcano-tectonic map of Grande Comore island with historical lava flows and fissure systems. The black box marks the Karthala Caldera Complex. (Adapted from Savin et al., 2005) Sam Poppe (PDF)

La Grille volcano
The northern volcano, La Grille, is a basaltic shield volcano and is variously described as not active, or even extinct. However, the GVP description says that it shows “recent lava flows, some perhaps as young as a few hundred years”. The volcano lacks a summit caldera like its neighbor to the south. The La Grille volcano also contrasts with Karthala in its abundance of pyroclastic cones up to 800 m in height. These were erupted along fissures paralleling the summit ridge, which has an irregular profile and is elongated in a N-S direction, and from radial fissures that reach as far as the coast.

MT. KARTHALA

Karthala is a Hawaiian-style basaltic shield volcano; with 3,000 m submerged plus 2361 m a.s.l. it is one of those very high rising colossuses. The volcano is abt. 30 km long and 15 km wide, and is topped by a caldera complex created by repeated collapse. Large and long cracks (=elongated rift zones) extend from the summit to the NNW and SE. Many lava flows have reached the sea on both sides of the island, including eruptions from the summit caldera and vents on the northern and southern flanks. An 1860 lava flow from the summit caldera even traveled ~13 km to the NW, reaching the western coast north of the capital city of Moroni. Karthala usually emits lava, but phreatic explosions with large plumes of ash and tephra have also been known – and have caused deaths and property damage in the past. During the last half century it appears that the eruption type is changing from effusive – mainly along the rift zones – towards phreato-magmatic or purely phreatic activity within the summit caldera complex. Thus, Karthala might be developing an elevated risk of explosive activity that is different from your typical benign Hawaiian-type effusive activity.

 

THE CALDERA COMPLEX

The Karthala Caldera Complex: a. Aerial photograph, note the steep outer caldera walls and the pit craters Changouméni and Chahalé; b. Caldera outline with inferred caldera units, which would have collapsed chronologically from a to h. (After Strong & Jacquot, 1972) Sam Poppe (PDF)

Karthala calderas. The best Google Earth image is from 06/05/2005

Mt. Karthala features one impressive set of eight (8) or more overlapping calderas, nested pit craters and several terraces.within a rectangle of 2.6 x 3.5 km. They are aligned in a sort of diamond shape, supporting the idea that they have been constructed along a cross of two intersecting rifting lines. The center of the caldera complex is occupied by the crater “Choungou Chahalé”, the older main crater, which consists now of two inner pits. For scientists, the walls of Chahalé expose an excellent ~200 m deep window into the volcanoes past – meter-thick basalt lava flows, feeder dykes, ancient caldera floor and other features of the volcanoes “inner workings”. Another crater is “Choungou Changouméni”, a small circular pit crater of 220 m diameter in the northern lobe of the caldera, next to a small scoria cone from 1972. Since the pit was (re-?)created in 1918 its depth has decreased gradually: it was 150m in 1936, 60m in 1952, 5m in 1980 and filled up completely in the last eruption in 2007. The caldera walls are sub-vertical and cut into a series of sub-horizontal layers of basalt and slag; its base is generally flat but it is covered with blocks of lava and scree.

View to the Southern wall of Choungou Chahalé in July 2011. Note the bench feature with a substantial pyroclastic cover and deep erosion gulleys, and to the immediate East of it a subvertical scarp inferred to be an ancient caldera floor that is now completely filled by tabular lava flows. The most bright rock column in the center of the picture is an ancient eruptive vent. (© Sam Poppe (PDF))

View into the northern Caldera. (© Marie & Olivier)

Interesting is the way the magma is thought to be rising from its deep source – imagine it like climbing up the stairs in a sky-scraper and stopping at each floor for a long breath: After a short ascend the magma is collected in large pockets. Within the residing magma pocket, cooling results in fractionation*, after that, a part of the fractionated magma migrates rapidly upwards – only to become arrested again at a higher level within the lithosphere. Again fractionation happens in a magma chamber. In this manner, the magma ascends and evolves gradually. In the finally erupted lava from Karthala olivine and pyroxene crystals are the cumulate minerals from the last fractionation phase in the shallowest magma pocket.

*Fractionation: in each cooling-phase some of the minerals will crystallize out of the liquid magma (according to their specific melting temperature). As they become solid, they can’t move on with the liquid but remain in the pocket. Thus the liquid is now devoid of that mineral and therefore has a different chemical composition and properties. From the final composition of the lava scientists can tell what cooling stages a magma has gone through.

Panoramic view of the western and northern crater walls of Choungou Chahalé. Walls are ~200 m high. Note the onlapping lava flow in the center (the northern wall of Chahalé), and the substantial pyroclastic debris fans against the wall feet. The 1991 explosive crater is visible in the Northeast. (© Sam Poppe (PDF))

HISTORICAL ERUPTIONS

34 eruptions have been recorded since 1808, most of them with a VEI of 0-2, but the ones in 1918 and Nov. 2005 registered with a VEI 3. So, we have on average an eruption every 6 years – quite a busy volcano! Monitoring the activity of Karthala has begun in 1986 with the measuring of surface deformations, and with the opening of the Karthala Volcano Observatory a seismic network was implemented in 1988. Here are some of the more notable and some recent eruptions:

2006 & 2007: Two phreato-magmatic eruptions, one creating a lava lake in the main crater (2006) and the other completely filling up the 1918-pit Changouméni crater (image left).

 

2005: 17/04 to 18/04/2005: Although seismicity had been high from the previous day on, it was impossible to say whether an eruption had begun. Now two of the three seismic sensors installed at the top of Karthala are not transmitting signals. The signs were there: a lot of “dust” being blown about, a river running “very heavy greyish water identical to concrete mortar”, incandescence at the crater seen by pilots the night before.

The aircraft used in a survey flight  😯 over Karthala volcano , April 18, 2005 (© M’dé)

But no visual observation was possible. It became obvious during the day, though, that the dust could only be ash. A survey flight next day confirmed the eruption and stated that the main crater had been enlarged considerably and a lava lake was visible in it. The ash caused much damage in the villages, mainly by contaminating the water tanks. This was a phreato-magmatic eruption, after 14 years of quiet. It completely vaporized the crater lake from 1991, which soon after began to fill again.

Masses of ash on the morning of 25/11/2005 (© Ahmed Aboubacar – CNDRS)

25/11 to 10/12/2005: The next event wasn’t far away – another violent phreato-magmatic explosion had caused a panic among the people the first night. The southern half of Grande Comore was covered with a layer of ash 2-5cm thick, and Moroni, the capital, was paralyzed. All offices and shops remained closed. In the Chahalé crater again a lava lake had been established that begun fountaining after a few days: on 1/12 two fountains of 15m hight rose with a frequency of every 15-20 seconds. This eruption lasted 14 days and fortunately had not caused any fatalities.

Karthala. Left: During lava solidification, a glow is visible at the site of future hornitos in the Chahalé (main crater). Nov. 30, 2005(© F.Sauvestre/KVO). Right: Same view as in the previous photo with the solidified hornitos (© F.Marcel-Asselin/GVP).

Karthala crater in Febr. 2002. The green lake had appeared for the first time after the 1991 eruption.

1991: This eruption marks the beginning of a style change in Karthala’s behaviour. It was also the first monitored by the new observatory. The eruption followed three months of increasing seismicity and inflation. After several hours of relative calm a sizeable phreatic explosion further enlarged Choungou Chahalé and lowered its floor. In a visit to the summit two weeks after the 11 July explosion people heard a “fountaining” sound, but no lava fountains or any other source were visible. It turned out later that this would have been the sound of the forceful arrival of water into the new crater, forming a crater lake for the first time. Scientists believe now, that this explosion had changed the hydrothermal system below the volcano, as all following eruptions were phreato-magmatic in nature.

Pahoehoe lava on the beach at Chindini on the island of Grande Comore. 2013. (© David Stanley, via Wikimedia)

1972: A month-long strombolian eruption begun in September; lava was emitted from four fissures within the caldera and possibly an external fissure. During the initial two days a flow of aa lava was emitted along 600 m of an old fissure zone.

1918: This eruption began suddenly in the night of August 11-12, with intense light flashes signalling the opening of a crevasse on the north flank of Karthala at an altitude of 1980 m, about 600 m above the source of a 1904 eruption. Seismic activity was felt on the flanks of the volcano until 23rd August, when a violent earthquake was felt on all of Grande Comore and the neighbouring island of Moheli. Violent explosive activity occurred within the caldera and created what is called the Changouméni today, a round pit crater. Although it is thought by some scientists that this crater had existed in the past but been obscured by overflowing lava.

1903: A poorly reported gas release event that might have killed as many as 17 people and would have originated from the summit caldera.

Lava flows off Mount Karthala have reached the Indian Ocean at many points on Grande Comore (© David Stanley, via Wikimedia)

1862: A lava flow that breached the Northern summit caldera wall and rushed down to the western coast at Itsandra, immediately to the north of the islands’ capital Moroni.

 

LANDSLIDE HAZARDs

Among the obvious volcanic hazards like eruptions there is one which is not always taken seriously enough: the landslide. A sharp landslide scarp constitutes the northern limit of the Badjini massif at the SE end of Grande Comore. This landslide scarp is evidence of a large failure of the E flank of Mt. Karthala. Amphitheatre-shaped escarpments have now been recognized as large sector collapses of the volcanoes’ flanks due to instability of the steep volcanic edifice. Large flank landslides represent a major volcanic risk for populated areas at the base of volcanic edifices. If such landslides truncate a volcanic edifice high enough around its summit, they may cause an unloading effect with consequent expansion of the underlying magma chamber and the facilitation of magma ascent.
Models have also shown that the caldera-bounding ring fault itself represents a plane of weakness, that might control or trigger catastrophic failure and flank collapse. Normally, this might be calculated and anticipated, but what if a former caldera and its ring fault is completely buried by new volcanic materials? At least one buried caldera was proposed to exist around the summit of Karthala. Considering the on average > 20° sloping unstable eastern flank of the volcano, this could indeed pose a risk of future flank failures with possible devastating consequences.

In some countries you can learn the geology if you are an avid writer of letters! 😉 (Images from mountainstamp.com)

Disclaimer: I am not a scientist, all information in this (and any of my other posts) is gleaned from the www and/or from books I have read, so hopefully from people who do get things right! 🙂 If you find something not quite right, or if you can add some more facts, please leave a comment.

Enjoy! – GRANYIA

SOURCES & FURTHER READING

– GVP, Karthala
– Geography of the Comoros Wikipedia
– Caldera collapse on basaltic shield volcanoes: […] Karthala caldera complex (2012, PDF)
– Project strengthening the capacity of the Volcano Observatory Karthala (OVK)
– Comores Online – Karthala
– The offshore East African Rift and the Comoros hotspot
– Earth of Fire, Karthala (Blog)

Advertisement