Volcano tales: Bogoslof

Location of Bogoslof. Adapted from Beget et al., 2005. From https://avo.alaska.edu/images/image.php?id=103411

At any given day, there are about 20 volcanoes erupting around the globe. Some of them are notorious for frequently showering nearby cities with ash, others wake up and blow without any warning – but all of them have their own intriguing story. For this blog, I will regularly pick one of the active volcanoes reported by the Smithsonian/USGS Weekly Volcanic Acitvity Report, and take a closer look at their history and their peculiarities.


Bogoslof, Fox Islands

On 20 December 2016, an explosive eruption at Bogoslof volcano produced an ash plume that rose to 10.3 km above sea level. It was the first activity reported since 1992, but the volcano seems to have awakened now: all through January, Bogoslof has continued to erupt explosively every 1-4 days, disrupting local air traffic by sending lightning-garnished plumes of ash, ice, and gases up to 11 km into the atmosphere. Meanwhile, pyroclastic surges, tephra and lava bombs more than doubled the size of the island itself. Not a bad record for an island that originally measured 0.3 km2, is it?


Well, Bogoslof Island, located some 40 km behind the Aleutian arc, is not your average island: It is the lively uppermost tip of a ~1500 m tall submarine stratovolcano. What reaches the surface here are generation after generation of lava domes, creating a transient island morphology in constant battle against the sea. Russian explorers first described a ‘small rocky prominence’ in 1768 (Miller et al., 1998), before an eruption in 1796 pushed a 10 x 10-6 m3 hornblende-bearing andesitic lava dome above sea level to form Bogoslof Island, or ‘Castle Rock’. In 1883, a second lava dome, a hornblende-bearing basaltic trachyandesite, advanced to the surface about 1 km NW of Castle Rock, to form ‘Fire Island’. Both of these lava domes have resisted weathering, and they quite literally bracketed eruptions since: The 1906-1910 and 1926-1928 eruptions all had their vents in between Castle Rock and Fire Island, and during both periods lava and pyroclastic products, combined with general uplift, temporarily joined up all subaerial portions of Bogoslof to form one big island. The main active vent usually took shape as a ‘hot lagoon’ in the course of the eruptions, occasionally isolated from the sea, as shown here in a morphological summary of Bogoslof from 1826-1907 by Jaggar (1908).


Most of the morphology sketched above has been engulfed by the sea since, and previous to the 2016 event, only Castle Rock, Fire Island, and the 1927 and 1992 domes had been withstanding erosion.

So how did surface morphology change at Bogoslof this time, 100 years after Jaggar’s sketches? Here is Alaska Volcano Observatory’s timeline of morphology changes since December 2016:

After initial explosions in the vent area removed parts of the previous island in December 2016, the further evolution shows strikingly similar patterns to previous eruptive cycles. Even though the current eruption hasn’t joined up Fire Island with Castle Rock (yet), the island surface area has almost tripled and the vent has become isolated from the sea as of 31 January. The accompanying eruptions of 30-31 January had a higher content of ash than previous explosions, some of which reached the neighbouring, populated Unalaska island.


The increase in ash content and the physical isolation from the sea indicate that the vent has risen above sea level, hence interaction with sea water plays a less crucial role. Consistently, explosivity and activity in general have decreased since 31 January, with the exception of four successive explosive events with column heights of 8 – 11 km from 17-19 February. The next few weeks will show whether the volcano’s activity will further wane or whether it will enter a stage of dome-building, similar to other historic eruptions.


Today, we can handily observe further developments at Bogoslof using remote monitoring techniques such as satellites and seismic and infrasound stations in neighbouring, populated islands. – But back in 1907, geologists had to embark on a perilous journey to the island and set foot onto the actual steaming lava dome to make observations. Jaggar’s colourful impressions of the lava dome (a spine) and the surrounding hot lagoon, published 1908 in the Bulletin of the American Geographical Society, are a documentation of true scientific spirit in an intense environment:

‘New activity began in March, I906, and by the midsummer of I907 two new hills had risen from the sea midway between Bogoslof and Grewingk, continuous land united all into one island, and a hot salt-water lagoon encircled the newer hill which was steaming like a pudding. The water lapped the sliderock slopes directly, without any intervening beach, the tumble of fragments under water was stained bright orange, and all the water was turbid and hot, steaming silently, at a temperature of about 90° degrees. […] There was no noise. The pinnacle on the summit was like a great parrot’s beak, rounded and smooth on the west, but making an overhanging cliff forty feet high on the east. The steam-vents gave temperatures varying from 94° to 212°, and the hottest vent, at the foot of the parrot cliff, was adjacent to rock practically incandescent, for here a piece of paper burst into flames.’


This account sounds as fascinating as it sounds terrifying, yet countless sea-lions and birds didn’t seem to mind the ongoing eruption that much – some were even nesting on the active lava dome:

‘Indeed these hardy birds seem to luxuriate in the volcanic warmth-perhaps it aids the hatching of their eggs, for even on the venomous-looking lava projections of McCulloch Peak, in the midst of steam and heat and sulphur, the birds crowded, as elsewhere, and laid their eggs in the hollows of the rocks and there reared their chicks.’


That didn’t always go well: Hunnicott (1943) found ‘bird skeletons […] in great numbers lying about the island where they had been veritably roasted alive’, others note that sea lions got ‘scalded so that their hair had come off’, and Dall (1884) simply notes: ‘A piece of seal meat suspended in a crevice was thoroughly cooked in a short time.’


Archaic and dangerous as this sounds, I do envy the scientists on these exhibitions for one thing: They could collect unlimited rock samples! And it seems like we could use some of these today, because literature on the geochemistry and petrology of Bogoslof is sparse. Arculus et al. (1977) characterised Bogoslof rocks as hornblende-bearing andesites and basalts and noted a conspicuous decrease in silica content from the 1796 (61 wt%) to the 1927 (46 wt%) eruption at decoupled REE-patterns and Sr isotope signatures. These features, combined with its setting behind the Aleutian arc and large depths to the slab (Plank & Van Keken 2013), make Bogoslof an intriguing study subject, yet no comprehensive study has been published since 1977. So as Bogoslof keeps on redefining its subaerial shape, there are many more worlds to explore with this volcano!