Seward Peninsula Volcanic Fields

image002Mount Bendeleben, Alaska

Earlier this year there was a significant earthquake swarm in Northwestern Alaska, north of the Seward Peninsula. While this wasn’t a swarm in the normal volcanic sense of a swarm (hundreds to thousands of smaller earthquakes in a period of hours to days), it was enough to get the attention of the locals who weren’t used to that level of activity. The largest quake was June 16, with a magnitude of 5.7, which is quite strong in that part of the state.

image003Espenberg region, Alaska

So what does an amateur volcanaholic do when earthquakes start breaking out in the middle of nowhere? This one takes a look to see if that area had a history of volcanic activity. It turns out that a look there are at least three volcanic fields some 150 km south and east of the area of the earthquakes, which was a surprise. Some of the activity may have been within the last several thousand years.

Checked with the Alaska Volcano Observatory to see if there was any excitement afoot. They wrote back that the quakes were tectonic in nature; all connected to local fault lines that had decided to wake up over the spring and early summer. The faults are mostly buried and do not manifest themselves real well on the surface. The region is remote, with few people on the ground, and permafrost ranges to some 60 M below the surface. The AIEC database shows the region fairly active for the last 15 years or so.

image005Approach into Buckland, Alaska,_Alaska#mediaviewer/File:Buckland_Alaska_aerial_view.jpg

But the notion that there were volcanic fields in the region, relatively recently active intrigued me so here is a bit of a science report on a part of the state I never visit.


The volcanic fields are located some 150 – 200 km north and east of Nome, Alaska. Nome sits on the southern side of the Seward Peninsula in the far western part of the state. This is the part of the state closest to Siberia.

Nome is the destination of the largest sporting event in Alaska, our yearly dog race, the Iditarod, which celebrates the successful delivery of diphtheria serum from Anchorage to Nome by dogsled in 1925. The event was recreated as a race starting in 1967. The trail is around 1,800 km long and these days a winning team takes some 8 – 9 days to make the trip. The start is in Anchorage the last weekend in February, so temperatures are generally cold.

Nome was also the site of a couple gold rushes around the turn of the 20th Century and went through the typical boom and bust cycles of the Western US during the 19th Century. It is one of the largest cities in the region with a population of around 3,800.


image006Volcanic Fields on the Seward Peninsula  Image courtesy AVO/ADGGS, Janet Shaefer,

The three volcanic fields are the Espenberg, Imuruk Lake and Koyuk-Buckland Volcanic fields. They are part of the larger Bering Sea Volcanic Province, which has been sporadically active over the last 30 MY. The activity in the province can be traced back as far as 30 MY starting in the Koyuk – Buckland volcanic field. This is also the least well known volcanic field in the region, mostly due to its age and remoteness. Activity in the Bering Sea Volcanic Province picked up significantly some 6 MY ago beginning in the Imuruk volcanic field and continued More recent activity has been higher volume eruptions in and around the Bering Sea west and south of Nome. The majority of eruptive material is basalts.

There is more than a little head scratching about the source of magma for this region. It sits some 250 to over 1200 km to the north of the Aleutian Arc and the associated volcanic activity to be a standard back-arc volcanic province. Models tying the magma to rifting or mantle plume / hot spot powered volcanism have not been conclusively demonstrated either.

As you travel east across Alaska at the same latitude from the three volcanic fields to Fairbanks, you traverse a region with geothermal energy potential. It is called the Central Alaskan Hot Springs Belt. There are several locations with hot springs that express surface activity. Major fault lines are mostly along an east–west axis, relieving energy delivered into the state via the collision with the Pacific Plate along the Denali Fault. The hot spring belt is not thought to be tied to volcanism in any meaningful way.

image008Earthquake and fault map of Alaska

The Koyuk – Buckland volcanics is by the far the largest of these fields. It is also the oldest. At least one researcher connects it to the Imuruk Lake volcanic field, though others do not. It sits some 720 km NW Anchorage and is composed primarily by basalt out of over 40 cinder cones and small shield volcanoes. Some of the basalt measures 150 m thick.

image010Map of Koyuk – Buckland Volcanic Field

The Imuruk Lake volcanic field was active for some 850,000 years starting some 6 MY ago. It deposited some 110 km3 over an area of 2,000 km2 over that time. More recent activity has deposited another 300 km2 with the newest flow being some 1,655 years old. It appears to have the lowest rate of eruptive material of all the volcanic fields studied in the Bering Sea Volcanic Province. It is comprised primary of basalts out of some 75 cinder cones surrounded by lava flows. There are a few small shield volcanoes complete with subsidence calderas and heights in the 100 m range.

image011Cinder cone and lava flow in Imuruk Lake Volcanic Field

image013Map of Imuruk Lake Volcanic Field

The furthest west field is the Espenberg Volcanic Field which is marked by four of the largest maars known to exist. The reason for the large size is the interaction of magma with permafrost and water. To put the size of these maars in perspective, they are similar in diameter with calderas formed after the Katmai – Novarupta eruption in 1912, Crater Lake, Oregon and Kilauea.

ArticleMap of Maars of Espenberg Volcanic Field

In order to understand what happened, we first need to understand what happened with the onset of the most recent round of ice ages starting some 2 MY ago. Not only the onset lead to the formation of glaciers which at times have covered much of the North American continent, but also led to the formation of permafrost over large areas of Siberia, Alaska, Canada and Scandinavia. Interesting enough, most of central, western and northern Alaska did not have glacial coverage during the last glaciation. Instead, the state got permafrost, permanent today in the northern third of the state. Depths of the permafrost range from some 100 M in the vicinity of the volcanic fields in question to nearly 800 M in the Prudhoe Bay oil fields, making the extraction of shallow well oil difficult at best. The other thing about this part of the state is that the soil is very wet, with ice comprising over 40% of the permafrost.

image015aDevil Mountain Maars

The newest is called Devil Mountain Lake and it measures some 8 km in diameter. It was dug via series of blasts which appear to have taken place as melted permafrost laden material slumped into the original explosive pit. There is evidence of pyroclastic flows during the eruptive sequence, making it more complex than the typical one-shot maar.

The paper referenced in the Additional Reading section on the Espenberg maars describe steep sides of these maars.  Depth is measured today at the deepest at around the 100 m, the depth  of the permafrost layer, though it appears that the craters were excavated to at least twice that depth into the older bedrock and sediment.  Additionally, it is also not uncommon for methane to accumulate under permafrost and the possibility exists that the explosions were also fueled by methane being torched off.
Older activity in the field includes at least five small shield volcanoes and basalt lava flows.  These date from 1.6 MY to more recent.  The maars all appear to have formed after the basalt activity, perhaps indicating when large quantities of permafrost were finally present for the upwelling magma to interact with.


One oddity on these three fields it the switch from rather run of the mill basalt cinder cone, small shield volcano activity to maars, as Espenberg is the only one of these fields known to have maars, though a pair of older lakes in the Imuruk Volcanic Field are suspected to also have maar-like activity.  The switch appears to be related entirely to the presence of permafrost brought about by the cyclic ice ages of the last 2 MY.

One final thought:  In these days of highly politicized climate science here in the US, it is getting increasingly difficult to find actual data on permafrost without wading through hundreds of recent articles on the warming, melting arctic and the horrors it will bring.  On a personal basis, I would rather my SciFi be labeled as such.

image017Imuruk Lake Volcanic Field



1,700 thoughts on “Seward Peninsula Volcanic Fields

  1. Herðubreið seems to be waking up again while Tungna’s going back to sleep ! How it can sleep with Barda banging away like that I do not know.

  2. I’m going to postulate a guess as to what happened during this spike, and say it was a subglacial phreatic explosion in the caldera. Given the fact that the tremor level was largest on the Vonarskard station (when compared to Dynjguhals and Grimsfjall), it’s fairly evident that this took place in or very close to the Caldera.

    Additionally, this was the largest unfiltered spike in tremor we’ve seen so far. Larger than when the dike was first opening, larger than the ice-cauldron subglacial eruptions, and larger than when the initial eruption started out at Holuhraun. So there was a lot of energy behind this event, suggesting more than magma simply opening up a dike.

    Another thing worth mentioning is that unlike when the dikes were initially forming, this was a quick event without sustained tremor as we saw during the start of this eruption. This once again, points towards an explosion of some sort, which would likely be phreatic at this stage in the game.

    The final piece of the puzzle is the associated quake, which was a 4.7 magnitude quake at a very shallow depth of .2km at the same time as the tremor spike, indicative of a shallow event of some sort (which fits with a phreatic detonation).

    And then a time-chart across the entire eruptive period to show the magnitude.

    • And to provide my input, you would expect to see phreatic eruptions before seeing more of a significant eruptive event. From a caldera subsidence perspective, you would get one of two scenarios.

      1. Water seeps down in the structurally damaged caldera lid, and comes into contact with extremely hot rock or even fresh magma, resulting in a phreatic detonation.

      2. Magma moves up into the cracks caused by the caldera subsidence, and comes into contact with hydrothermal system beneath the glacier, resulting in a phreatic detonation.

      This is just my postulation, and in zero ways an expert opinion, I’m not a scientist who studies this stuff like IMO does, so look out for any messages on their behalf.

  3. Recently, some of you have expressed concerns over quakes not getting checked and corrected. My thoughts.

    IMO’s responsibility is to the people of Iceland, not us. If something serious is potentially out there, it would only make sense that they would redirect staff to work out the details of that threat rather than manually fixing the waveform picks off the seismos.

    If you read between the lines, there might be something serious afoot.

    No insider data here, just an observation.

    • Geo, even if it is not something serious maybe the resources are just better used somewhere else rather than keeping the lists updated. I am sure they could have an intern or PA do this offline for later reference but what is the real value to them to keep up the lists?

    • Personal value? Almost nil. They are the experts and have access to the raw data. The list is just a thing that interested parties can look at. (we fall into that category).

      The other day someone posted a mostly mangled Gurgle translation of discussion points dealing with the hydropower stations. The tone of that seemed to be a bit alarmed.

      Also, someone posted a fantastic photoset of Holuhraun and mentioned out of hand that the pilot was forbidden from flying in the vicinity of Barda. That makes sense considering the alert status. Think about that for a moment. What are Bardabunga and Grimsvotns method of eruption? Smolder a bit? No, not at all. When they go, they go full on with almost no ramping up stage. It’s all or nothing.

      I’m just a watcher, but it seems like there is quite a bit to keep an eye on.

      Don’t sit with your back to the door…

  4. Wording of this mornings (9:30) Science advisory report has changed a bit:

    . Three scenarios are considered most likely:
    . The eruption on Holuhraun declines gradually and subsidence of the Bardarbunga caldera stops.
    . Large-scale subsidence of the caldera occurs, prolonging or strengthening the eruption on Holuhraun. In
    this situation, it is likely that the eruptive fissure would lengthen southwards under Dyngjujokull,
    resulting in a jokulhlaup and an ash-producing eruption. It is also possible that eruptive fissures could
    develop in another location under the glacier.
    . Large-scale subsidence of the caldera occurs, causing an eruption at the edge of the caldera. Such an
    eruption would melt large quantities of ice, leading to a major jokulhlaup, accompanied by ash fall.
    Other scenarios cannot be excluded.

  5. New summary map. Sources: IMO, IES and NASA with gratitude to all and admiration for the work you do and the effort put in to keep us informed.

      • Hi Simi! How are you? Missed you and Kat and Motsfo last night (my time) when we had a rather lively discussion about the new earthquake craze: Strictly Come Quaking. It’s worth going back two pages for those comments. 😀

        • Hi! I’m OK, I have just been working hard on my garden (planting the winter garden and the garlic for next year), yard (holes and trenches to dig and fill in for sewer replacement, almost done!) and house (painting and caulking the outside before the winter rains set in. I finished caulking 10 pm last night (by flashlight!) and it started raining in the night. It will mostly rain for the next few months.

          • You have been working hard! Great timing on the caulking, just as the rains begins, but I’m most impressed by your working by flashlight to finish the job! 😮

            • I understand – that would have been a powerful motivator for me too! Must have felt good, lying in bed that night, knowing the job had been finished before that rain arrived… nice one! 🙂

            • It is possible that standing on a ladder with a flashlight in one hand and a tool in the other hand isn’t the safest thing, but I am a gal who does what needs to be done!

    • When we only see 35 to 40 cm a day, you don’t realise how it’s rapidly accumulating over time. I was surprised myself that it is this much already!

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s