New #OpenAccess paper in @SeismoSocietyAm "The Seismic Record": The 27 February 2022 Lop Nor Earthquake: Detectability, Location, and Discrimination. How forensic #seismology can demonstrate that a seismic disturbance near a #nuclear test-site is natural.…
I wish to thank my co-authors, Esteban J Chaves and Mark Fisk, for their amazing help, cooperation, and enthusiasm! The paper is quite compact as the journal's philosophy demands, but here is a little TL-DR for anybody who is curious about forensic seismology.
On 27 February 2022, a magnitude 4.8 #seismic event took place near the northern part of the Chinese Lop Nor #nuclear test site. Forensic #seismology has to identify the source of such seismic disturbances. #LopNor #forensicseismology #nucleartesting
If we consider the signals at one station, AAK in Kyrgyzstan, the February 2022 event is clearly very close to the site of previous nuclear tests and has the same sequence of seismic arrivals with similar time-delays.
Comparing the signal at the same station from a nuclear test a little further away, the signal looks quite a bit different. But this reminds us that the shape of a seismic signal depends not only on the type of source, but also the location. We need to look more closely.
At another station, NIL in Pakistan, the signal from the February 2022 event (top) looks significantly different from the signals from these nuclear tests. So the difference in the shape of the signals changes at different distances and directions!
It is useful to look at how the Signal-To-Noise ratio (SNR) varies at different seismic stations around the world. This can also give clues as to what kind of seismic event it was – and it can indicate parts of the world where stations best detect Lop Nor events.
The signal to noise ratio on stations in many parts of the world was very poor: many places under or barely over the detection threshold. We best observe the event on stations in Central Asia – but significantly also in Alaska on the @AKearthquake seismic network.
Many new seismic stations in Alaska that were in the USArray Transportable Array recorded fantastic signals from this event. Most stations showed a second arrival about 8 seconds after the first: a possible clue that it was a deep event: too deep for human involvement!
Zooming into the topography near the Lop Nor nuclear test site we can look at the different location estimates for the 27 February event relative to previous underground nuclear tests. There is location uncertainty but it is clearly within 30 km of earlier explosions.
We know the locations of the Chinese nuclear tests with great accuracy thanks to so-called precision seismology and satellite imagery. (Image from Fisk, 1992)…
The Bayesloc probabilistic seismic location program allows us to compare the accuracy with which we can locate each seismic event relative to the known locations. The 2022 event location has a similar uncertainty to that of each of the nuclear explosions.
However, the Bayesloc program can locate many events simultaneously and correct for errors in the predicted seismic traveltimes. The "uncertainty clouds" are now smaller and closely match the known locations relatively, even if with a "bias offset".
But another cool thing about Bayesloc is that, if you know where one or more seismic events actually took place, you can impose this information as a "prior". We can have high confidence in the "calibrated" location estimates for the 2022 event.
The signals on three stations at so-called regional distances (in China and Kazakhstan) are very strong and we can use them to solve for the "seismic mechanism" and the most likely depth. This study indicates an earthquake at 20-25km depth.
The most likely form of the earthquake has a so-called angle of strike which is very similar to previous earthquakes observed near Lop Nor – and the angle makes sense in the context of local geomorphological features.
However, we need to remember that key stations are on Chinese territory. These stations were simply switched off at the times of Chinese nuclear tests! Just try getting hold of the data! Here is a whole day's data from station IC.XAN. A nuclear test took place during that gap.
So we need to ensure that we can also identify the event as an earthquake using stations much further away. The stations in Alaska that showed this "double pulse" in 2022 didn't for the nuclear explosions. The second phase is a so-called depth-phase. A clear earthquake.
We conclude that the 27 February 2022 seismic event at Lop Nor was an earthquake, consistent with earlier events in the same reason, and at a depth where we can rule out it being the consequence of any anthropogenic activity. Forensic seismology.
All of the waveform data used was obtained via @IRIS_EPO and all of the input data for the location estimates is freely available from…

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Dr. Steven J. Gibbons

Dr. Steven J. Gibbons Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!


0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy


3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!