Earthquake Report: M 8.2 near Perryville, Alaska

A few days ago, I was passed out on my couch (sleep apnea) and for some reason I awoke and noticed that I had gotten a CSEM notification of a large earthquake offshore of Alaska. Well, after looking into that, I sent my boss, Rick, a text message: “8.2.”
https://earthquake.usgs.gov/earthquakes/eventpage/us6000f02w/executive
Rick Wilson runs the tsunami program at the California Geological Survey (CGS) and works with the California Governor’s Office of Emergency Services (Cal OES) to use official forecasts of tsunami size from the National Tsunami Warning Center (NTWC) to alert coastal emergency managers about the level of potential evacuation that they may want to act upon.
More about this process can be found here. Take a look at the CGS Special Report 236 to learn about the Tsunami Playbooks and the “FASTER” approach for tsunami evacuation guidance. Evacuation is something that is done at the local level, so CGS and Cal OES can only provide recommendations.
Needless to say, we were both at the ready to respond. Rick has hourly phone calls with the NTWC and follows up with phone calls and emails to specific interested parties (e.g. the emergency managers). We each went into tsunami response mode. I manage the Tsunami Event Response Team, which may be activated to collect observations of tsunami inundation or ocean currents.
I started looking at tide gage and DART Buoy data to see how large the tsunami was in the epicentral region. The M 8.2 was in the region of the 1938 M 8.2 earthquake which generated a transoceanic tsunami. I also looked into the literature about the 1938 tsunami, to see what size that tsunami was. The 1938 tsunami had a decimeter scale wave height (peak to trough) for gages in Alaska and in California (Johnson and Satake, 1994). Jeff Freymueller et al. (2021) had also recently worked on the 1938 earthquake source area and tsunami modeling as well.
The nearest tide gage for this 2021 event is at Sand Point, but the nearest gage in 1938 was in Unalaska. So, in order to get a modest comparison between 1938 and 2021, I felt a need to wait for the Unalaska data to trickle in. This may give us some idea whether the 1938 tsunami recorded in Crescent City and San Francisco might be a decent analogue. Of course, we need to get the official forecast from the NTWC prior to sending out any information. But, that process can take hours (over 3 hours in this case). So, we need to get our minds wrangled around the possibilities in the absence of more information.
Earthquake and Tectonic Background:
The plate boundary in the north Pacific is a convergent (pushing together) plate boundary where the Pacific plate on the south ‘subducts’ northwards beneath the North America plate on the north. The Alaska-Aleutian subduction zone forms a deep sea trench which can be seen in maps of the region. The subduction zone fault dips into the Earth, getting deeper to the north.
Between earthquakes (the interseismic period), the megathrust fault is seismogenically coupled (i.e. ‘locked’) just like velcro has the ability to hold together one’s wallet. The plates are always moving towards each other. Because the fault is locked, the crust surrounding the fault bends elastically to accommodate this convergent motion.
As the crust bends and flexes, it stores energy (i.e. tectonic strain). The part of the fault closest to the seafloor (the southernmost part of this subduction zone fault) gets pulled downwards, while the part of the crust further to the north flexes upwards.
The materials along the earthquake fault have properties that resist motion (like the velcro). But, as the plates converge and increase the amount of energy stored, the forces on the fault may exceed the strength of the fault. At this time, the fault slips, causing an earthquake.
The part of the fault that was being pulled downwards gets pushed upwards during the earthquake (the coseismic period), while the crust that was being flexed upwards between earthquakes thus subsides downwards during the earthquake.
The Alaska-Aleutian subduction zone has a history of subduction zone earthquakes and tsunami, plus there exists a prehistory of earthquakes and tsunami in some parts of this plate boundary. Geologists are often asked to determine the potential hazard of future earthquakes and tsunami and their answers are based on what we know from the past (using both historic and prehistoric data).
The 2021 M 8.2 earthquake happened in the same location as a 1938 M 8.2 earthquake, just to the east of a sequence of earthquakes from last year (22 July and 19 October 2020).
Tsunami:
When the earthquake fault slips, and the upper plate deforms, the vertical motion of the plate can elevate (or lower) the overlying ocean water. After the water changes position, it seeks to return to sea-level (an equipotential surface). If elevated, the water drops downwards and then oscillates up and down. This is the process that generates waves that radiate from the area with seafloor deformed by the earthquake.

    Things that make a tsunami larger are [generally]:

  1. More vertical land motion (possibly from larger slip on the fault, e.g. from a larger magnitude earthquake)
  2. Deeper water (deeper water = more volume of water moving = more energy to create larger tsunami waves)

So let’s take a look at the things that may have affected the size of the tsunami from this 2021 M 8.2 earthquake.
First of all, based on the earthquake slip models (estimates of how the earthquake slipped, in meters, and how that slip varied along the fault) suggest that a majority of the largest slip happened beneath the continental shelf. The water depth on the shelf is similar to many shelfs worldwide, shallower than about 200 meters. How does this affect the size of the tsunami?
Well, I guess that is the main point, the ground deformation that generated the tsunami was beneath shallow water.
These slip models are based on a variety of data and most of the data are seismic data. Some tsunami are generated by slow slip (not generating seismic waves) on the shallow part of the fault. These are called tsunami earthquakes.
Because tsunami earthquakes may be generated by slip in this way, slip models using seismic data cannot resolve the location of the slip on the fault that created these tsunami. However, the tsunami from this 2021 M 8.2 earthquake were small. Therefore the updip part of the fault probably did not contribute significantly to the tsunamigenic ground deformation.

Below is my interpretive poster for this earthquake

  • I plot the seismicity from the past 3 months, with diameter representing magnitude (see legend). I also include earthquake epicenters from 1921-2021 with magnitudes M ≥ 7.5.
  • I plot the USGS fault plane solutions (moment tensors in blue and focal mechanisms in orange), possibly in addition to some relevant historic earthquakes.
  • A review of the basic base map variations and data that I use for the interpretive posters can be found on the Earthquake Reports page. I have improved these posters over time and some of this background information applies to the older posters.
  • Some basic fundamentals of earthquake geology and plate tectonics can be found on the Earthquake Plate Tectonic Fundamentals page.
  • I include outlines of the historic subduction zone earthquakes as prepared by Peter Haeussler from the USGS in Anchorage. He appears in the video about the 1964 earthquake below.
  • Some of the tide gage and DART buoy locations are labeled.
  • Note how there are still aftershocks from the 2018 M 7.9 earthquake sequence.

    I include some inset figures. Some of the same figures are located in different places on the larger scale map below. I present 3 posters, each with slightly different information.

  • This is the first poster I prepared.
  • In the upper center is a low-angle oblique view of the plate boundary. Note the oceanic Pacific plate is subducting beneath the continental North America plate. As the plate goes down, the water embedded in the rocks and sediment are released into the overlying mantle wedge. This water causes the mantle to melt, which rises, erupts as lava and forms the volcanic chain we call the Aleutian Islands. I place a green star in the “epicentral” location of the 2021 M 8.2 earthquake.
  • In the upper left corner is part of a figure from Witter et al. (2019) that shows sections of the megathrust fault relative to how much the fault is thought to be locked. This is called the coupling ratio. For a fault that is fully coupled (or locked), the ratio is 1.0. For a fault that is slipping about 50% and accumulating about 50% of the plate motion rate, the coupling ratio is 0.5. Many subduction zones have low coupling ratios of 0.2-0.6. The region of the fault west of the 1938 and 2021 M 8.2 earthquakes is called the Shumagin Gap, thought to be possibly aseismic (with a coupling ratio closer to 0). But the 2020 sequence of M 7.8 and 7.9 earthquakes filled much of this gap.
  • In the upper right corner is a plot showing the earthquake shaking intensity using the Modified Mercalli Intensity Scale (MMI). This is a USGS model based on observations of intensity from thousands of earthquakes. Read more about MMI here.
  • In the center right is a plot showing the aftershocks within a couple hours of the mainshock
  • In the lower right corner is the initial record of the tsunami at the Sand Point tide gage (see map for gage location).
  • I labeled the USGS slab 2.0 slab contours (Hayes et al., 2018). These depth contours represent the depth of the megathrust fault at these locations. The M 8.2 hypocentral depth is 32.2 km and the slab2 depth is about 35 km. Nice!
  • Here is the map with 3 month’s seismicity plotted. There are 3 posters. The first one is something I put together around 2 hours after I awoke on the couch (abt 2am my time). I prepared the 2nd poster an hour later, which includes some information about tsunami prehistory. I prepared the 3rd poster late Sunday evening, about 3 days after the earthquake.


  • This is the second map I prepared and some figures are the same as in the first poster.
  • Below the low-angle oblique map is a slip model from the USGS. The color represents the amount of slip on the fault. Note that the maximum slip is close to the epicenter. This is not always the case, as for the 1938 event, it appears that the maximum slip was not where the mainshock epicenter was.
  • In the upper left corner is a map from Nelson et al. (2015). Those authors studied the prehistoric tsunami records at Chrikof Island, an island about 200 km to the east of the 2021 M 8.2 epicenter. The lower map shows GPS derived plate motion rates.
  • In the lower right corner is also from Nelson et al. (2015). On this plot, the vertical axis represents time with “today” at the top and over 5000 years ago at the bottom. The horizontal axis is space, west to east from left to right. Each colored symbol represents the time of a prehistoric tsunami. The vertical size of these symbols represents the uncertainty (or “error”) associated with those chronologic data. We can take the number of earthquakes or tsunami over a period of time to estimate how frequently those process happen over time.
  • To the left is a more updated version of the Sand Point tide gage, showing a wave height (peak to trough) of about 45 cm. We cannot compare this to the 1938 tsunami as there was not a tide gage at Sand Point in 1938


  • I prepared a 3rd poster, but updated it to this 4th poster.
  • In the Intensity Data area, I added USGS “Did You Feel It?” data, which come from reports from real people. Learn more about dyfi here. The model data are the colored lines labeled in white and the dyfi data are colored polygons labeled in yellow.
  • In the aftershocks plot, I added epicenters from the several days after the mainshock. I also added a transparent overlay of the USGS finite fault model (the slip model). Compare the overlap, or non-overlap, of the slip region and the aftershocks. Why do you think that they are not completely overlapping?
  • In the lower right section are tide gage records from gages in the area included in the poster. I plot the tidal forecast (dark blue), the tide gage observed water surface elevation (medium blue), and the difference between these data (in light blue) which is a record of the tsunami (and other waves, like wind waves). I made a rough approximation estimate of the maximum wave height and labeled this in yellow. The San Point tide gage has a mx wave height of about 0.8 m!
  • I also plot the data from the DART buoy 46403, which is the closest DART buoy to the mainshock epicenter. The DART buoy network is used to help calibrate tsunami forecast models during tsunami events. These are basically pressure transducers on the seafloor that measure changes in pressure caused by waves and atmospheric processes. The data plotted here are not tsunami data, but seismic wave data. One reason we know that this is not a tsunami is that the waveform initiated about 3 minutes after the earthquake. A tsunami would take longer to get to the buoy.
  • In the upper left corner is a pair of maps that show USGS earthquake induced ground failure models. The map on the right shows what areas have likelihood of having landslides triggered by the 2021 M 8.2 earthquake. The panel on the right shows the possibility that areas might experience liquefaction induced by the earthquake.
  • I added aftershocks associated with the 2020 M 7.8/7.5 sequence that filled the Shumagin Gap (green circles) and outlined the aftershock region for both 2020 and 2021 sequences. The 2021 sequence is not yet over. The largest aftershock so far has only been M 6.1. The 1938 M 8.2 event had a M~7 event 5 days after the mainshock. Stay tuned?


    •

Tectonic Overview

Below is an educational video from the USGS that presents material about subduction zones and the 1964 earthquake and tsunami in particular.
Youtube Source IRIS
mp4 file for downloading.

    Credits:

  • Animation & graphics by Jenda Johnson, geologist
  • Directed by Robert F. Butler, University of Portland
  • U.S. Geological Survey consultants: Robert C. Witter, Alaska Science Center Peter J. Haeussler, Alaska Science Center
  • Narrated by Roger Groom, Mount Tabor Middle School

This is a map from Haeussler et al. (2014). The region in red shows the area that subsided and the area in blue shows the region that uplifted during the earthquake. These regions were originally measured in the field by George Plafker and published in several documents, including this USGS Professional Paper (Plafker, 1969).


Here is a cross section showing the differences of vertical deformation between the coseismic (during the earthquake) and interseismic (between earthquakes).


This figure, from Atwater et al. (2005) shows the earthquake deformation cycle and includes the aspect that the uplift deformation of the seafloor can cause a tsunami.


Here is a figure recently published in the 5th International Conference of IGCP 588 by the Division of Geological and Geophysical Surveys, Dept. of Natural Resources, State of Alaska (State of Alaska, 2015). This is derived from a figure published originally by Plafker (1969). There is a cross section included that shows how the slip was distributed along upper plate faults (e.g. the Patton Bay and Middleton Island faults).


Here is a graphic showing the sediment-stratigraphic evidence of earthquakes in Cascadia, but the analogy works for Alaska also. Atwater et al., 2005. There are 3 panels on the left, showing times of (1) prior to earthquake, (2) several years following the earthquake, and (3) centuries after the earthquake. Before the earthquake, the ground is sufficiently above sea level that trees can grow without fear of being inundated with salt water. During the earthquake, the ground subsides (lowers) so that the area is now inundated during high tides. The salt water kills the trees and other plants. Tidal sediment (like mud) starts to be deposited above the pre-earthquake ground surface. This sediment has organisms within it that reflect the tidal environment. Eventually, the sediment builds up and the crust deforms interseismically until the ground surface is again above sea level. Now plants that can survive in this environment start growing again. There are stumps and tree snags that were rooted in the pre-earthquake soil that can be used to estimate the age of the earthquake using radiocarbon age determinations. The tree snags form “ghost forests.


This is a photo that I took along the Seward HWY 1, that runs east of Anchorage along the Turnagain Arm. I attended the 2014 Seismological Society of America Meeting that was located in Anchorage to commemorate the anniversary of the Good Friday Earthquake. This is a ghost forest of trees that perished as a result of coseismic subsidence during the earthquake. Copyright Jason R. Patton (2014). This region subsided coseismically during the 1964 earthquake. Here are some photos from the paleoseismology field trip. (Please contact me for a higher resolution version of this image: quakejay at gmail.com)


This is another video about the 1964 Good Friday Earthquake and how we learned about what happened.

  • Here is a map that shows historic earthquake slip regions as pink polygons (Peter Haeussler, USGS). Dr. Haeussler also plotted the magnetic anomalies (grey regions), the arc volcanoes (black diamonds), and the plate motion vectors (mm/yr, NAP vs PP).


    •

  • Here is the figure from Sykes et al. (1980) that shows the space time relations for historic earthquakes in relation to the map.


    • Above: Rupture zones of earthquakes of magnitude M > 7.4 from 1925-1971 as delineated by their aftershocks along plate boundary in Aleutians, southern Alaska and offshore British Columbia [after Sykes, 1971]. Contours in fathoms. Various symbols denote individual aftershock sequences as follows: crosses, 1949, 1957 and 1964; squares, 1938, 1958 and 1965; open triangles, 1946; solid triangles, 1948; solid circles, 1929, 1972. Larger symbols denote more precise locations. C = Chirikof Island. Below: Space-time diagram showing lengths of rupture zones, magnitudes [Richter, 1958; Kanamori, 1977 b; Kondorskay and Shebalin, 1977; Kanamori and Abe, 1979; Perez and Jacob, 1980] and locations of mainshocks for known events of M > 7.4 from 1784 to 1980. Dashes denote uncertainties in size of rupture zones. Magnitudes pertain to surface wave scale, M unless otherwise indicated. M is ultra-long period magnitude of Kanamori 1977 b; Mt is tsunami magnitude of Abe[ 1979]. Large shocks 1929 and 1965 that involve normal faulting in trench and were not located along plate interface are omitted. Absence of shocks before 1898 along several portions of plate boundary reflects lack of an historic record of earthquakes for those areas.

  • Here is a great illustration that shows how forearc sliver faults form due to oblique convergence at a subduction zone (Lange et al., 2008). Strain is partitioned into fault normal faults (the subduction zone) and fault parallel faults (the forearc sliver faults, which are strike-slip). This figure is for southern Chile, but is applicable globally.


    • Proposed tectonic model for southern Chile. Partitioning of the oblique convergence vector between the Nazca plate and South American plate results in a dextral strike-slip fault zone in the magmatic arc and a northward moving forearc sliver. Modified after Lavenu and Cembrano (1999).

In 2016, there was an earthquake along the Alaska Peninsula, a M 7.1 on 2016.01.24. Here is my earthquake report for this earthquake. Here is a map for the earthquakes of magnitude greater than or equal to M 7.0 between 1900 and today. This is the USGS query that I used to make this map. One may locate the USGS web pages for all the earthquakes on this map by following that link.

Tsunami Data

I plot tide gage data for gages in the north and northeast Pacific Ocean. These data are from NOAA Tides and Currents, though are also available via the eu tide gage website here.

    Each plot includes three datasets:

  1. The tidal forecasts are shown as a dark blue line.
  2. The actual observed water surface elevation is plotted in medium blue.
  3. By removing (subtracting) the tide forecast from the observed data, we get the signal from wind waves, tsunami, and atmospheric phenomena. This residual is plotted in light blue.

The scale for the tsunami wave height is on the right side of the chart.
Note the all tsunami wave height plots are the same vertical scale, except for Sand Point.
I measured the largest wave heights for each site, displayed in yellow.
Alaska














Here are the data from the DART buoy nearest the M 8.2. People often mistake these data for tsunami data, but this is generated by seismic waves.
One way to test one’s hypothesis about whether these buoy data are seismic waves or tsunami waves, one simply need to take a look at the time that the wave begins to be recorded by the DART buoy.
Seismic waves travel through water at about 1.5 kms per second. While tsunami wave velocity (based on the shallow water wave equation) for depths ranging from 200-4000 meters is between ~0.02 to 0.2 kms per second, much slower than seismic waves.

Surface Deformation

Below are surface deformation data generated by the USGS based on their finite fault model. The three panels show surface deformation in the north, east, and vertical directions.
North, East, and Up are positive (blue) while South, West, and Down are negative (red).
Note the upper panel and how the Pacific plate is moving to the north and the North America is moving south. Does this make sense?
The middle panel is interesting too, but skip to the lower panel, vertical. The accretionary prism (forming the continental slope), directly above the aftershocks and mainshock, rises up during the earthquake. The upper North America plate landward of the slip patch subsides. Does this make sense?
Earlier in this report we took a look at the geologic evidence for megathrust subduction zone earthquakes, evidence that records this “coseismic” subsidence.

Shaking Intensity and Potential for Ground Failure

  • Below are a series of maps that show the shaking intensity and potential for landslides and liquefaction. These are all USGS data products.

    • There are many different ways in which a landslide can be triggered. The first order relations behind slope failure (landslides) is that the “resisting” forces that are preventing slope failure (e.g. the strength of the bedrock or soil) are overcome by the “driving” forces that are pushing this land downwards (e.g. gravity). The ratio of resisting forces to driving forces is called the Factor of Safety (FOS). We can write this ratio like this:

    FOS = Resisting Force / Driving Force

    When FOS > 1, the slope is stable and when FOS < 1, the slope fails and we get a landslide. The illustration below shows these relations. Note how the slope angle α can take part in this ratio (the steeper the slope, the greater impact of the mass of the slope can contribute to driving forces). The real world is more complicated than the simplified illustration below.


    Landslide ground shaking can change the Factor of Safety in several ways that might increase the driving force or decrease the resisting force. Keefer (1984) studied a global data set of earthquake triggered landslides and found that larger earthquakes trigger larger and more numerous landslides across a larger area than do smaller earthquakes. Earthquakes can cause landslides because the seismic waves can cause the driving force to increase (the earthquake motions can “push” the land downwards), leading to a landslide. In addition, ground shaking can change the strength of these earth materials (a form of resisting force) with a process called liquefaction.
    Sediment or soil strength is based upon the ability for sediment particles to push against each other without moving. This is a combination of friction and the forces exerted between these particles. This is loosely what we call the “angle of internal friction.” Liquefaction is a process by which pore pressure increases cause water to push out against the sediment particles so that they are no longer touching.
    An analogy that some may be familiar with relates to a visit to the beach. When one is walking on the wet sand near the shoreline, the sand may hold the weight of our body generally pretty well. However, if we stop and vibrate our feet back and forth, this causes pore pressure to increase and we sink into the sand as the sand liquefies. Or, at least our feet sink into the sand.
    Below is a diagram showing how an increase in pore pressure can push against the sediment particles so that they are not touching any more. This allows the particles to move around and this is why our feet sink in the sand in the analogy above. This is also what changes the strength of earth materials such that a landslide can be triggered.


    Below is a diagram based upon a publication designed to educate the public about landslides and the processes that trigger them (USGS, 2004). Additional background information about landslide types can be found in Highland et al. (2008). There was a variety of landslide types that can be observed surrounding the earthquake region. So, this illustration can help people when they observing the landscape response to the earthquake whether they are using aerial imagery, photos in newspaper or website articles, or videos on social media. Will you be able to locate a landslide scarp or the toe of a landslide? This figure shows a rotational landslide, one where the land rotates along a curvilinear failure surface.

  • Below is the liquefaction susceptibility and landslide probability map (Jessee et al., 2017; Zhu et al., 2017). Please head over to that report for more information about the USGS Ground Failure products (landslides and liquefaction). Basically, earthquakes shake the ground and this ground shaking can cause landslides.
  • I use the same color scheme that the USGS uses on their website. Note how the areas that are more likely to have experienced earthquake induced liquefaction are in the valleys. Learn more about how the USGS prepares these model results here.


    •

Some Relevant Discussion and Figures

  • Johnson and Satake (1994) studied tsunami waveforms from the 10 November 1938 Alaska M 8.2 earthquake. Their analysis was designed to estimate the source for the tsunami. Below are some figures from their paper, with figure captions beneath each figure.
  • This first plot shows the tsunami records from tide gages. This is the plot I used to consider the potential impact to the coast from the 2021 M 8.2 tsunami.


    • Digitized marigrams from 1938 Alaskan earthquake recorded in Crescent City, San Diego, and San Francisco. The tidal componenht asn ot beenr emoved.S tartt ime listedf or each record is the time in minutes from the origin time of the earthquaketo the startt ime of the digitizedr ecord.

  • Here is a map that shows the fault model that they used, as well as the amount of slip that they used for each fault element.


    • Location of subfaults used in inversion of tsunami waveforms. Graph shows slip distribution in meters.

  • This is a figure comparing their model results (synthetic = dashed) compared to the tide gage records (solid lines).


    • Observed and synthetic waveforms from inversion for four subfaults. Start time of each record is different. The arrows indicate the parts of the waveforms used for the inversion.

  • Freymueller et al. (2021) also studied the 1938 M 8.2 event, seeking to resolve the slip on the fault using tsunami modeling.
  • Below are figures with their captions in blockquote.
  • Here are some maps showing 2 of the slip distrubutions that they used for their modeling.


    • Example slip distributions for two of the slip models, shallow eastern and shallow far eastern. For each model the slip is the product of a function f(x) representing the along-strike variation and g(y) representing the downdip variation, and then scaled to a constant magnitude MW 8.25. The functions f(x) and g(y) are based on relations in Freund and Barnett [1976]. For the central and western models, the rupture area is the same as for the eastern model, but the area of higher slip is shifted to the west. For the mid-depth and deep models, the main area of high slip is shifted downdip.

  • Here are some maps showing vertical seafloor displacements for some of their tsunami scenarios.


    • Vertical seafloor displacements caused by representative slip scenarios. On the left side, the slip is concentrated in the east and the deep, mid-depth and shallow slip distribution scenarios are shown. On the right, the Western, Central and Far Eastern slip distribution scenarios are shown assuming the shallow rupture. Displacements are in meters. Red contours show depth to the plate interface from 0 to 80 km with a 10 km increment.

  • Here are plots that show some results of their modeling. The tide gage data are plotted in black and their simulated waves are plotted with red and blue lines.


    • Tide gauge data and model predictions for the eastern and far eastern source models.

    Here is an animation from one of the Ferymueller et al. (2021) models for the 1938 M 8.2 tsunami.

  • Nelson et al. (2015) presented their evidence for prehistoric tsunami on Chirikof Island, an island in the forearc in the eastern part of the 1938 earthquake slip patch.
  • They found evidence for many tsunami over a timespan from before 5000 years ago.
  • Below are some figures from their paper, with figure captions in blockquote.
  • This figure shows the tectonic setting and the area of their field study.


    • A) Location of Chirikof Island within the plate tectonic setting of the Alaska-Aleutian subduction zone. Rupture areas for great twentieth century earthquakes on the megathrust are in pink. (B) Velocity field of the Alaska Peninsula and the eastern Aleutian Islands observed by global positioning system (GPS) (Fournier and Freymueller, 2007). Colors show inferred rupture areas for earthquakes in 1788 (green) and 1938 (orange). Both A and B are modified from Witter et al. (2014). The section of the megathrust between Kodiak Island and the Shumagin Islands has been referred to as the Semidi segment (e.g., Shennan et al., 2014b). (C) Physiography of Chirikof Island (Google Earth image, 2012) showing the location of our study area at Southwest Anchorage, a prominent moraine, a fault scarp (facing southeast) that probably records the 1880 earthquake, the New Ranch valley reconnaissance core site, and UNAVCO GPS station AC13 (http:// pbo .unavco .org /station /overview /AC13). In the eighteenth and nineteenth centuries, Chirikof Island was known to native Alutiiq and Russians as Ukamuk Island.

  • Here is a plot that shows the timing for the prehistoric tsunami inferred by these authors. The vertical axis is the time scale, with “today” at the top. Each colored pattern represents the age range for a tsunami deposit.
  • These data are plotted left to right, west to east, so we can compare tsunami records at different locations along the margin. These comparisons are important so that we can test different hypotheses about how subduciton faults may slip over time. In the 2021 case, the slip area was close to the 1938 earthquake. But, did has this always occured here?


    • Age probability distributions for probable (red) and possible (orange) tsunami deposits at Southwest Anchorage (labels as in Fig. 11) compared with age distributions for possible tsunami deposits at Sitkinak Island (Briggs et al., 2014a) and with age estimates for great earthquakes and tsunamis on Kodiak Island (from studies referenced on this figure;
    Fig. 1). Dotted horizontal lines show our correlation of evidence for some younger earthquakes and tsunamis. Times of great earthquakes inferred from episodes of village abandonment determined from archaeological stratigraphy in the eastern Alaska-Aleutian megathrust region are also shown (Hutchinson and Crowell, 2007).

Return to the Earthquake Reports page.