The 46

How many earthquakes have we had on the CSZ, with magnitudes of 7.8 or above during the past 10,000 years? 46. Where do we find that data? It took me a while to dig, so for anyone who is curious, here is the journey that led me to ‘The 46’.

In early 2017, all I could find for the dates of past CSZ earthquakes was a single Oregonian article. https://projects.oregonlive.com/maps/earthquakes/timeline. If you haven’t seen it, I highly recommend playing with their graph. You can hover over the bars for more info or download data! However, it only lists 40 earthquakes for the 10,000-year period (which is the number reported when the article was written).

I later found sites like OEM claiming there had been 41, not 40. https://www.oregon.gov/OEM/hazardsprep/Pages/Cascadia-Subduction-Zone.aspx. (They still list 41). I started looking for the date of the missing quake, to no avail.

August 5, 2017: Oregon State University releases an article stating there have, in fact, been 43 quakes, but of course doesn’t provide the dates for said three… & I’m getting frustrated.  https://today.oregonstate.edu/archives/2016/aug/subduction-zone-earthquakes-oregon-washington-more-frequent-previous-estimates.

John Vidale, Former director of the Pacific Northwest Seismic Network, puts me in touch with Erin Wirth. She is an Affiliate Assistant Professor with UW and Research Geophysicist with USGS. She sent me an Excel workbook entitled Goldfinger_etal_2016_Tables.

In the Excel workbook, Erin explained that the list contains the 43 earthquakes in the past ~10,000 years (plus older ones too!). The mean age is given relative to the year 1950, so you’ll have to take the numbers in column ‘B’ and add 70 years in order to get the age of the earthquake (for 2020). Erin explains that the dates are estimates, based on carbon-dating. (If anyone would like to see the worksheet, let me know. I’m happy to email it.)

I had the spreadsheet from Seth Stein’s lab to work with. Seth Stein is a Northwestern University Professor. It took me a headache of a time to find a place online that would show me earthquake probability models, so here’s that link https://sites.northwestern.edu/sethstein/presentations/educational-material/earthquake-probabilities/

I worked with the 43 earthquake dates until I thought my eyes were going to bleed from staring at the screen too long. I did this week after week & failed, repeatedly. I still couldn’t come up with a 37% risk I kept hearing about.

It was driving me nuts!

Then, in response to an email inquiry, Bob Yeats (Former Professor at OSU & author of Living with Earthquakes in the Pacific Northwest
http://oregonstate.edu/instruct/oer/earthquake/00%20front%20rev_color.html) put me in touch with his former student, Chris Goldfinger at OSU (no introduction needed?). He sent me his Excel Sheet with the 37% calculation & I should mention, answered every annoying question I asked. He was awesome.

On his spreadsheet was the Mean 221 & the standard deviation of 121. I now had the history. I had the calculation worksheet. I even had the inputs… but I STILL couldn’t figure out how he’d gotten those numbers for Mean & Standard Deviation with only the 43 quakes.

Chris Goldfinger sent me the list of segments found on my segment page. The red section on each indicates how much of the fault ruptured. The white lettering on the left is a list of the earthquakes that ruptured in that particular way.

Here is where I get “smart” points. I cross-referenced the white lettering earthquakes listed on each segment with the quake names in his Excel sheet & found extras on the segment page. I figured they wouldn’t be on the segment page if the researchers didn’t know about them…

I reached out to Chris Goldfinger again for the ages of the 3 quakes. Rough ages are available to the researchers, but they are not releasing carbon-dates out to the public yet (checking and rechecking data). For this reason, all 46 earthquakes are used in calculating risk, since they all occurred, but only 43 have been released in research papers for date purposes. His hope was to have the other three released this year or next. Hopefully soon.

The names of the quakes are in chronological order with T1 being the most recent & T25 being the oldest. Because of this, I was able to know where the earthquakes fall historically for the calculations. Go me 🙂 (For the most part, if the TX is followed by a lowercase letter, it is a partial rip & those without were full-rips on Segment A).

With all 46 earthquakes, I was finally able to replicate the calculations for the Mean & Standard Deviation that Chris Goldfinger provided. Urika! These are shown on my Calculations page.