La Soufriere St. Vincent: Background

 

La Soufriere St. Vincent last erupted in 1979 with a volcanic explosivity index (VEI) of 3, causing evacuation of 20,000 people. In 1902-1903 it had a VEI4 eruption that caused ~1500 deaths. Ironically, this eruption was somewhat overshadowed because it was one day before the catastrophic eruption of Mt. Pelee in Martinique, an eruption that destroyed St. Pierre.** There were also eruptions in 1718 and 1812, giving a rough cyclicity of explosive eruptions every 77-94 years.

(left figure) Soufriere St. Vincent is shown at the north end of the island in brown shading. It rises to a height of 1204 m. 

The island developed as a result of subduction of the Atlantic/North American plate beneath the Caribbean plate. The oldest dated rocks on the island are 2.74 Ma and Soufriere's activity began around 0.69 Ma. 

    At present no one knows or is predicting how this eruption cycle will turn out: what kind of eruptions? what kind of magma? how long will the eruptions persist.  But let's look at one possibility, the well documented 1902-1903 eruptions. There were three main phases in that eruption: May 6-7th; September-October 1902; and March 1903.      

    The first eruption began on May 6th, 1902, after 13 months of felt seismicity. Eruptions involved a pre-existing crater lake in a series of steamy explosions, some reaching 1 km in height.  Incandescence (=lava) was observed on the evening of May 6th. Tephra fallout began around 11:00 a.m. on May 7th, with fine ash changing to larger particles ("lapilli") around noon. The paroxysmal stage occurred around 2:00 p.m. with the formation of the "Great Black Cloud," caused by pyroclastic density currents (PDC's) travelling down all flanks of the volcano killing >1500 people. In places the PDC's reached the ocean and traveled across the sea for several kilometers.  Further explosive activity occurred in October 1902 and March 1903. Short summary of eruption history is available here.

    The 1979 eruption began with an explosive phase of 11 Vulcanian type explosions over 13 days, and the eruption columns from these explosions reached 18 km above sea level.  The April 17th eruption generated a PDC that reached out 2 km from the crater.

    In December 2020, lava began oozing quiescently from the summit, posing no serious threat to the 110,000 residents of the island of St. Vincent. However at the end of March, seismicity increased and at 8:41 a.m. on April 9, several major explosions occurred. 20,000-30,000 people living north of the volcano were endangered and required evacuation either by ship or by roads. The north end of the island is connected to the south end by only a single road. Evacuation by ship has been complicated because of the pandemic and evacuees must prove that they have been vaccinated against COVID-19 before they can get on a ship.

    The magma is a basaltic andesite. In contrast to the basalts that erupt in Hawaii or Iceland, for example, this magma is more viscous and thus the magma doesn't degas easily. The dissolved gas is then available to power explosive, rather than gently effusive, eruptions. Volcanologists couldn't predict exactly when an eruption would occur and were thus apparently relieved when the Prime Minister ordered people in the northern sectors to immediately evacuate on April 8.  Early on the morning of April 9, a huge summit explosion sent ash 10 km into the atmosphere; a second blast happened a few hours later. As of today, April 14, water shortages have been developing because ash is contaminating all surface sources. Imagery from the eruption can be seen here.

    As of Wednesday, April 14, pyroclastic flows have reached stretches of coastline and strong explosions are continuing.

**This material is taken from P.D. Cole, R.E.A. Robertson, L. Fedele, and C. Scarpati, Explosive activity of the last 1000 years at La Soufriere, St. Vincent, Lesser Antilles.

A "groovy instability" (and its role in pyroclastic density currents)

 A small volcano, Volcán Bárcena,  480 km off the west coast of Mexico, displays an unusual set of grooves that were carved during an eruption in 1952. The volcano came to life, grew to 335 m elevation, and ended eruptions all in the short span of 7 months.  There were three phases in this eruption: (1) formation of its base by eruptions starting on August 1; (2) creation of a large crater in the summit; and (3) eruption of lava at the base.  The eruption was only intermittently observed and those observations were documented by Adrian Richards, a scientist who was working in the region in 1952 and undertook to document the "terrestrial and submarine geology" of the area.  His observations were only the second observations of the evolution of a new volcano following the observations of Paricutin, Mexico, which erupted from 1943 to 1952 through a cornfield in Mexico.

      Richards argued that the straightness (lack of sinuousity) and lack of dendritic pattern of the grooves indicated that they were not formed by rain (seev^^^below), but they were formed by erosion by "tephra avalanches." He avoided the use of the term "pyroclastic flow" because he did not see the incandescence implied by the term "pyro."

It is not clear when the grooves were formed, but almost certainly as one of the last stages of the eruption sequence. Fortunately, Adrian Richards  listed seven features of the grooves that must be addressed by any theory for their origin: (1) their origin close to the crater rim; (2) U-shaped profiles in cross section; (3) the straightness of the grooves; (4) the (generally) non-coalescing nature of the grooves; (5) increasing width downstream; (6) "turbulent patterns" on the cone, which are interpreted to be a herring bone pattern; and (7) local deposition at the lower ends.

In a new paper (March 2021) Kieffer et al. (2021)*** propose that the grooves were carved into an erodible substrate (volcanic ash, cinders) during passageof a high-energy pyroclastic density current (PDC) that contained streamwise vortices. In a survey of the volcanic literature, we found several volcanoes that have streamwise grooves, some of which--but not all--might have been carved the the same mechanism that we propose for the grooves at Volcán Bárcena. 

We suggest that the grooves terminated in the downstream direction by passage of the supercritical flow in which they were embedded through a hydraulic jump. The jump caused the flow to decelerate, resulting in the formation of dunes.

 

***Kieffer, S.W., Meiburg, E., Best, J., and Austin, J. The mysterious grooves of Volcán Bárcena: a review of the role of streamwise counter-rotating vortices during erosion by dilute pyroclastic density currents. Bulletin of Volcanology v. 83, article number 26 (2021). 

^^^Footnote regarding the effect of rain on the grooves: The two images below show the effect of rain on the initially straight channels, with one image taken in ~1952 and the other in 2005 or 2006 (Google Earth). 

  (left) Barcena from 1952 (Richards) and (right) 10/27/2006 (Google Earth).