Earthquake Report: 1700 Cascadia subduction zone 317 year commemoration

Today (possibly tonight at about 9 PM) is the birthday of the last known Cascadia subduction zone (CSZ) earthquake. There is some evidence that there have been more recent CSZ earthquakes (e.g. late 19th century in southern OR / northern CA), but they were not near full margin ruptures (where the entire fault, or most of it, slipped during the earthquake).
I have been posting material about the CSZ for the past couple of years here and below are some prior Anniversary posts, as well as Earthquake Reports sorted according to their region along the CSZ. Below I present some of the material included in those prior reports (to help bring it all together), but I have prepared a new map for today’s report as well.

On this evening, 317 years ago, the Cascadia subduction zone fault ruptured as a margin wide earthquake. I here commemorate this birthday with some figures that are in two USGS open source professional papers. The Atwater et al. (2005) paper discusses how we came to the conclusion that this last full margin earthquake happened on January 26, 1700 at about 9 PM (there may have been other large magnitude earthquakes in Cascadia in the 19th century). The Goldfinger et al. (2012) paper discusses how we have concluded that the records from terrestrial paleoseismology are correlable and how we think that the margin may have ruptured in the past (rupture patch sizes and timing). The reference list is extensive and this is but a tiny snapshot of what we have learned about Cascadia subduction zone earthquakes. Brian Atwater and his colleagues have updated the Orphan Tsunami and produced a second edition available here for download and here for hard copy purchase (I have a hard copy).

Here is a map of the Cascadia subduction zone, modified from Nelson et al. (2006). The Juan de Fuca and Gorda plates subduct northeastwardly beneath the North America plate at rates ranging from 29- to 45-mm/yr. Sites where evidence of past earthquakes (paleoseismology) are denoted by white dots. Where there is also evidence for past CSZ tsunami, there are black dots. These paleoseismology sites are labeled (e.g. Humboldt Bay). Some submarine paleoseismology core sites are also shown as grey dots. The two main spreading ridges are not labeled, but the northern one is the Juan de Fuca ridge (where oceanic crust is formed for the Juan de Fuca plate) and the southern one is the Gorda rise (where the oceanic crust is formed for the Gorda plate).

Today I prepared this new map showing the results of shakemap scenario model prepared by the USGS. I prepared this map using data that can be downloaded from the USGS website here. Shakemaps show what we think might happen during an earthquake, specifically showing how strongly the ground might shake. There are different measures of this, which include Peak Ground Acceleration (PGA), Peak Ground Velocity (PGV), and Modified Mercalli Intensity (MMI). More background information about the shakemap program at the USGS can be found here. One thing that all of these measures share is that they show that there is a diminishing of ground shaking with distance from the earthquake. This means that the further from the earthquake, the less strongly the shaking will be felt. This can be seen on the maps below. The USGS prepares shakemaps for all earthquakes with sufficiently large magnitudes (i.e. we don’t need shakemaps for earthquakes of magnitude M = 1.5). An archive of these USGS shakemaps can be found here. All the scenario USGS shakemaps can be found here.
I chose to use the MMI representation of ground shaking because it is most easily comparable for people to understand. This is because MMI scale is designed based upon relations between ground shaking intensity and observations that people are able to make (e.g. how strongly they felt the earthquake, how much objects in their residences or places of business responded, how much buildings were damaged, etc.).
The MMI ground motion model is based upon a computer model estimate of ground motions, different from the “Did You Feel It?” estimate of ground motions that is actually based on real observations. More on the MMI scale can be found here and here.

Here is the USGS version of this map. The outline of the fault that was used to generate the ground motions that these maps are based upon is outlined in black.

I prepared an end of the year summary for earthquakes along the CSZ. Below is my map from this Earthquake Report.

  • Here is the map where I show the epicenters as circles with colors designating the age. I also plot the USGS moment tensors for each earthquake, with arrows showing the sense of motion for each earthquake.
  • I placed a moment tensor / focal mechanism legend in the lower left corner of the map. There is more material from the USGS web sites about moment tensors and focal mechanisms (the beach ball symbols). Both moment tensors and focal mechanisms are solutions to seismologic data that reveal two possible interpretations for fault orientation and sense of motion. One must use other information, like the regional tectonics, to interpret which of the two possibilities is more likely.
  • In some cases, I am able to interpret the sense of motion for strike-slip earthquakes. In other cases, I do not know enough to be able to make this interpretation (so I plot both solutions).

    I include some inset figures in the poster.

  • In the upper left corner is a map of the Cascadia subduction zone (CSZ) and regional tectonic plate boundary faults. This is modified from several sources (Chaytor et al., 2004; Nelson et al., 2004)
  • Below the CSZ map is an illustration modified from Plafker (1972). This figure shows how a subduction zone deforms between (interseismic) and during (coseismic) earthquakes. Today’s earthquake did not occur along the CSZ, so did not produce crustal deformation like this. However, it is useful to know this when studying the CSZ.
  • To the lower right of the Cascadia map and cross section is a map showing the latest version of the Uniform California Earthquake Rupture Forecast (UCERF). Let it be known that this is not really a forecast, and this name was poorly chosen. People cannot forecast earthquakes. However, it is still useful. The faults are colored vs. their likelihood of rupturing. More can be found about UCERF here. Note that the San Andreas fault, and her two sister faults (Maacama and Bartlett Springs), are orange-red.
  • To the upper right of the Cascadia map and cross section is a map showing the shaking intensities based upon the USGS Shakemap model. Earthquake Scenarios describe the expected ground motions and effects of specific hypothetical large earthquakes. The color scale is the same as found on many of my #EarthquakeReport interpretive posters, the Modified Mercalli Intensity Scale (MMI). The latest version of this map is here.
  • In the upper right corner I include generalized fault map of northern California from Wallace (1990).
  • To the left of the Wallace (1990) map is a figure that shows the evolution of the San Andreas fault system since 30 million years ago (Ma). This is a figure from the USGS here.
  • In the lower right corner I include the Earthquake Shaking Potential map from the state of California. This is a probabilistic seismic hazard map, basically a map that shows the likelihood that there will be shaking of a given amount over a period of time. More can be found from the California Geological Survey here. I place a yellow star in the approximate location of today’s earthquake.

This figure shows how a subduction zone deforms between (interseismic) and during (coseismic) earthquakes. We also can see how a subduction zone generates a tsunami. Atwater et al., 2005.

Here is a version of the CSZ cross section alone (Plafker, 1972).

Here is an animation produced by the folks at Cal Tech following the 2004 Sumatra-Andaman subduction zone earthquake. I have several posts about that earthquake here and here. One may learn more about this animation, as well as download this animation here.

Here is a graphic showing the sediment-stratigraphic evidence of earthquakes in Cascadia. 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.

Here is a photo of the ghost forest, created from coseismic subsidence during the Jan. 26, 1700 Cascadia subduction zone earthquake. Atwater et al., 2005.

Here is a photo I took in Alaska, where there was a subduction zone earthquake in 1964. These tree snags were living trees prior to the earthquake and remain to remind us of the earthquake hazards along subduction zones.

This shows how a tsunami deposit may be preserved in the sediment stratigraphy following a subduction zone earthquake, like in Cascadia. Atwater et al., 2005. If there is a source of sediment to be transported by a tsunami, it will come along for the ride and possibly be deposited upon the pre-earthquake ground surface. Following the earthquake, tidal sediment is deposited above the tsunami transported sediment. Sometimes plants that were growing prior to the earthquake get entombed within the tsunami deposit.

The NOAA/NWS/Pacific Tsunami Warning Center has updated their animation of the simulation of the 1700 “Orphan Tsunami.”
Source: Nathan C. Becker, Ph.D. nathan.becker at

Below are some links and embedded videos.

  • Here is the yt link for the embedded video below.
  • Here is the mp4 link for the embedded video below. (2160p 145 mb mp4)
  • Here is the mp4 link for the embedded video below. (1080p 145 mb mp4)

  • Here is the text associated with this animation:

    Just before midnight on January 27, 1700 a tsunami struck the coasts of Japan without warning since no one in Japan felt the earthquake that must have caused it. Nearly 300 years later scientists and historians in Japan and the United States solved the mystery of what caused this “orphan tsunami” through careful analysis of historical records in Japan as well as oral histories of Native Americans, sediment deposits, and ghost forests of drowned trees in the Pacific Northwest of North America, a region also known as Cascadia. They learned that this geologically active region, the Cascadia Subduction Zone, not only hosts erupting volcanoes but also produces megathrust earthquakes capable of generating devastating, ocean-crossing tsunamis. By comparing the tree rings of dead trees with those still living they could tell when the last of these great earthquakes struck the region. The trees all died in the winter of 1699-1700 when the coasts of northern California, Oregon, and Washington suddenly dropped 1-2 m (3-6 ft.), flooding them with seawater. That much motion over such a large area requires a very large earthquake to explain it—perhaps as large as 9.2 magnitude, comparable to the Great Alaska Earthquake of 1964. Such an earthquake would have ruptured the earth along the entire length of the 1000 km (600 mi) -long fault of the Cascadia Subduction Zone and severe shaking could have lasted for 5 minutes or longer. Its tsunami would cross the Pacific Ocean and reach Japan in about 9 hours, so the earthquake must have occurred around 9 o’clock at night in Cascadia on January 26, 1700 (05:00 January 27 UTC).

    The Pacific Tsunami Warning Center (PTWC) can create an animation of a historical tsunami like this one using the same too that they use for determining tsunami hazard in real time for any tsunami today: the Real-Time Forecasting of Tsunamis (RIFT) forecast model. The RIFT model takes earthquake information as input and calculates how the waves move through the world’s oceans, predicting their speed, wavelength, and amplitude. This animation shows these values through the simulated motion of the waves and as they race around the globe one can also see the distance between successive wave crests (wavelength) as well as their height (half-amplitude) indicated by their color. More importantly, the model also shows what happens when these tsunami waves strike land, the very information that PTWC needs to issue tsunami hazard guidance for impacted coastlines. From the beginning the animation shows all coastlines covered by colored points. These are initially a blue color like the undisturbed ocean to indicate normal sea level, but as the tsunami waves reach them they will change color to represent the height of the waves coming ashore, and often these values are higher than they were in the deeper waters offshore. The color scheme is based on PTWC’s warning criteria, with blue-to-green representing no hazard (less than 30 cm or ~1 ft.), yellow-to-orange indicating low hazard with a stay-off-the-beach recommendation (30 to 100 cm or ~1 to 3 ft.), light red-to-bright red indicating significant hazard requiring evacuation (1 to 3 m or ~3 to 10 ft.), and dark red indicating a severe hazard possibly requiring a second-tier evacuation (greater than 3 m or ~10 ft.).

    Toward the end of this simulated 24-hours of activity the wave animation will transition to the “energy map” of a mathematical surface representing the maximum rise in sea-level on the open ocean caused by the tsunami, a pattern that indicates that the kinetic energy of the tsunami was not distributed evenly across the oceans but instead forms a highly directional “beam” such that the tsunami was far more severe in the middle of the “beam” of energy than on its sides. This pattern also generally correlates to the coastal impacts; note how those coastlines directly in the “beam” have a much higher impact than those to either side of it.

    Offshore, Goldfinger and others (from the 1960’s into the 21st Century, see references in Goldfinger et al., 2012) collected cores in the deep sea. These cores contain submarine landslide deposits (called turbidites). These turbidites are thought to have been deposited as a result of strong ground shaking from large magnitude earthquakes. Goldfinger et al. (2012) compile their research in the USGS professional paper. This map shows where the cores are located.

    Here is an example of how these “seismoturbidites” have been correlated. The correlations are the basis for the interpretation that these submarine landslides were triggered by Cascadia subduction zone earthquakes. This correlation figure demonstrates how well these turbidites have been correlated. Goldfinger et al., 2012.

    This map shows the various possible prehistoric earthquake rupture regions (patches) for the past 10,000 years. Goldfinger et al., 2012. These rupture scenarios have been adopted by the USGS hazards team that determines the seismic hazards for the USA.

    Here is an update of this plot given new correlations from recent work (Goldfinger et al., 2016).

    Here is a plot showing the earthquakes in a linear timescale.

    I combined the plot above into another figure that includes all the recurrence intervals and segment lengths in a single figure. This is modified from Goldfinger et al. (2012).

7 thoughts on “Earthquake Report: 1700 Cascadia subduction zone 317 year commemoration

    1. really depends upon the location….
      based upon what i saw from the current modeling (to be released soon I hope), it looks like maybe 5 meters on the open coast +-. but it is strongly dependent on location… and when I was looking at the maps, I was focused on the northern North Spit.
      basically, the 2009 CGS maps are overestimates of the run up… given what is possible from a subduction zone that last ruptured 317 yrs ago (ie. the CGS may present maps that represent a much longer strain accumulation, which is not physically possible. but we need to wait and see what they present.)

    2. highly unlikely
      1. large portion of locked zone under land, so less seafloor deformation compared to further north
      2. water is shallow where locked zone is
      3. shorter recurrence here, so slip less per event
      this is what i have been talking about for years, but the tsunami models are finally showing the lower runups i expected.
      but we cannot predict any of this, just forecast with some level of certainty. ;-) tho if one stays out of the yellow regions on the RCTWG tsunami evac maps, they will be safe from tsunami (but NOT from ground shaking). note in my map above (higher res on my report) the high ground shaking in our region!!! very high in humboldt.

  1. 317 … can’t even divide a 10 into it?….or 115,705 day anniversary? Be careful to not oversaturate on a single event that is not too likely to happen soon, otherwise it may have the opposite effect. A southern San Andreas rupture on the statistical likelihood scale is MUCH more likely to happen and yet all I hear about is Cascadia since “the” article. A tough balance, but maybe we are driving that one hazard too hard at the expense of more likely scenarios — and I haven’t mentioned drowning in a flood or being burnt up in a wildfire. All important but I think we’ve maybe gone of the skids :>) Maybe celebrate every 5th year—ha!

    1. I feel a responsibility to my general education classes to produce a summary of this, with something new added (used arcgis to plot MMI scenario this year). I understand. But since this hazard is at my feet, I remember it every day. So you are welcome I don’t post about it that frequently. Heheheh. To give you credit where it is deserved (and for my under represented students that likely have families that live closer to these higher probability faults, I included s shake map scenario for some other faiths in my digital presentations today. Almost as if I anticipated (predicted? Heheh. NOT) your comment. Love you bro.

Leave a Reply to Jason Robert Patton on Facebook Cancel reply

Your email address will not be published. Required fields are marked *