When did the blast wave of the eruption "hit" your station?

[michael.k...@gmx.at]

As you may have noticed, there was a huge vulcano eruption in the southern pacific. The blast wave hit central europe yesterday n the evening around 20:10 CET.

You can observe a steep rise in pressure at the time:

You can see the spike in weewx stations around the world...

[uajqq]

About 11:00 EST for my location!

[michael.k...@gmx.at]

A closer look shows the shockwave from the other side also:

[Nate Bargmann]

Looking at the past 24 hours there is an interesting "wave" shown just
after midnight (0600 UTC) here. What time did it occur? News reports
refer to "local time" but make no mention of UTC.

- Nate

--
"The optimist proclaims that we live in the best of all
possible worlds. The pessimist fears this is true."
Web: https://www.n0nb.us
Projects: https://github.com/N0NB
GPG fingerprint: 82D6 4F6B 0E67 CD41 F689 BBA6 FB2C 5130 D55A 8819

[Karen K]

Interesting! I immediately looked into my records and - yeah - it's there, too.

[paul.ba...@gmail.com]

Fascinating observations!    It would be very nice to have the precises coordonates for all stations.

Included is a very preliminary plot of the two events.  I suspect we can find other "reverberations".

My coordonates are 46:22:50 N / 6:13:31 E  / 410 m (Nyon in Switzerland)..

The times arre given in UTC hours and the barometer in mbar. Left graph for JAnuary 15, right for January 16.

The station is a Weatherflow Tempest, sampling interval is every 100 s.

Waiting for other observations!     Have a good day,     Paul

[Nate Bargmann]

It may be that the first wave came through much earlier than I have a
graph available. My graphs from the daily view has scrolled past the
point where anything around 1400 to 1500z 15 Jan can no longer be seen.

There is probably some way to query the DB on the WeeWX server to tease
this data out if anyone would be so kind as to share. I use the SQLite
backend.

[Greg Troxel]

It's easy, but then you get to play with awk to get what you really
want.

sqlite3 weewx.sdb "select dateTime, barometer, altimeter, pressure from archive order by dateTime;"

[Rick Kowalewski]

From Perth, Western Australia



Cheers, Rick

[Tim Tuck]

And this video shows amalgamated data and other satellite pics.

https://www.youtube.com/watch?v=zoMRwyNhqJ4

cheers

Tim

[Les Niles]

I had to pull the data out of the database and plot it in a spreadsheet. Taking a second look, I found the long-path pulse also. The second trip around should be about 32-36 hours later; I haven’t looked for that yet.

-Les

[Cameron D]

I did the same, since by the time I looked it had disappeared from the daily plot.

I am 3300km from the volcano (East coast Australia).

Unfortunately my station only updates the barometer value every 15 minutes, so there is only a single point at the "peak", a value back around average, then the next reading is the negative pulse.

I've plotted about a week, to get an idea how much noise there usually is in the data. The first red arrow is the primary pulse and the second wave, a bit less than 30 hours later.

I wondered whether the stuff indicated by the green arrow was more eruptions, but then realised it was a band of thunderstorms passing over..

I also looked at barometer plots from the Australian weather Bureau and did a crude plot of timing vs distance. From Norfolk Island to Perth gives us a range from 2000 and 7000 km from the volcano...

Those results are only reported every 30 minutes, so the error bars would be rather big.

[Les Niles]

OK, this is getting kind of ridiculous: I’ve picked up the first wave on its second pass around the Earth, at 14:20 PST January 16.  From the timing, I’m estimating that the wave propagates at an altitude where the temperature is about -15ºC.  The second pass of the long-path wave should come around tomorrow morning.  The whole saga is at http://www.2pi.org/wx/TongaVolcano.html

  -Les

On 16 Jan 2022, at 19:51, Les Niles <l...@2pi.org> wrote:

I had to pull the data out of the database and plot it in a spreadsheet. Taking a second look, I found the long-path pulse also. The second trip around should be about 32-36 hours later; I haven’t looked for that yet.

 -Les

--
You received this message because you are subscribed to the Google Groups "weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to weewx-user+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/AC7953D3-80D3-4F87-9EC2-D2ACBDE2370B%402pi.org.

On Jan 16, 2022, at 3:44 PM, Nate Bargmann <n0...@n0nb.us> wrote:

It may be that the first wave came through much earlier than I have a

--
You received this message because you are subscribed to the Google Groups "weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to weewx-user+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/20220116234355.5rr4jtrjpjcz32ax%40n0nb.us.

--
You received this message because you are subscribed to the Google Groups "weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to weewx-user+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/AC7953D3-80D3-4F87-9EC2-D2ACBDE2370B%402pi.org.

[Timothy Bryan]

My buddy who also has a station near by  think this is when the wave went by in Hawai'i

[Bob]

For those who find this topic interesting you might want to look at
Hamsci @ Googlegroups.

Bob - wd6dod

[morr...@gmail.com]

Not sure I am really seeing it here in maritime Canada. We're about 13000 km from Tonga.

Attached is a gnuplot command file and its output is:

The pressure pulse should have arrived about 12:00 local time. There is a little pressure increase at about 12:10.

Use the command file to read from weewx database:

rm barometer.png; sqlite3 weewx.sdb "select datetime, barometer from archive where datetime > 1642320000 and datetime < 1642384200 order by dateTime;" | gnuplot barometer.gnup

and modify as needed for your own location.

[Rob Cranfill]

Seattle, Washington! I noticed it the same day, but didn't read y'all's comments till now. Pretty cool. :-)

[storm...@gmail.com]

Check early morning Jan 16, when the wave came the opposite direction. Might of capture that wave.  They say the second wave was stronger on the east coast.

  

[Les Niles]

I think you’re looking at the wrong time period.  The initial, direct wave arrived at the U.S. west coast — about 8400 Km from Tonga — around 4:00am Pacific time on Jan 15.  It should take 3-1/2 to 4 hours to travel the additional 4600 Km to your location.  That, plus the 4 hour time difference, would put it at around 12:00 noon your time on Jan 15, not Jan 16.  The long path wave should’ve arrived at your location something like 13-14 hours later, or shortly after midnight on Jan 16.  The second trip around the Earth of the initial wave would be something like 32-36 hours after the first arrival, or 20:00-24:00 your time on Jan 16.  What you’ve plotted here misses all of those.  

  -Les

[morr...@gmail.com]

I may be starting to obsess a bit, but give the attached python script a try. You may have to install additional modules, e.g. geopy and sqlite3.

1. On line 9, specify your location

2. On line 14, specify the range of hours you want to examine, around the event.

3. On line 25, you may need to adjust the query if your pressure data is not in the barometer column of the archive table

3. Move to the directory containing your weewx.sdb database, or on line 22, specify its full path.

4. Run the script

The script calculates the distance from the eruption to your location, the travel time to your location at the speed of sound, and then queries and plots, and saves the barometer data for a time range around the estimated arrival time. It also remove any linear trend and plots the difference from that linear trend. 

Follow up with any improvements!

Mine looks like this:

[michael.k...@gmx.at]

Obsessed? You're just wrong in the right way :D ;)

I really like this thread, and your script. Here is my result:

distance to eruption 16894.9 km arrival at 1642272882 (2022-01-15 19:54:41)

Google Earth Pro tells me, the distance is 16880.6km. 

[michael.k...@gmx.at]

An improvement could be to also extract data for the 2nd wave from the other direction.

[Karen K]

What I realized: The arrival time is not quite accurate in my case. The script says 19:35, but the records show the peak at 20:25. 

[vince]

Very cool - I had to spin up a gnome desktop lightsail instance and VNC in from my Mac mini but I got there.  Is there a way to run this one headless to generate the image for viewing on a different computer ?

FWIW - on ubuntu you need to: "sudo apt-get install python3-matplotlib python3-numpy python3-geopy" for the requisite pieces of the puzzle.

My data => distance 9195.8 km arrival at 1642248822 (2022-01-15 04:13:41) here about 3 km from the water between Seattle+Tacoma

[morr...@gmail.com]

On Tuesday, January 18, 2022 at 5:15:49 p.m. UTC-4 vince wrote:

Very cool - I had to spin up a gnome desktop lightsail instance and VNC in from my Mac mini but I got there.  Is there a way to run this one headless to generate the image for viewing on a different computer ?

Easiest is just to comment out the "plt.show()"  at the end. I've attached a version which does that, and pretties up the output a bit.

FWIW - on ubuntu you need to: "sudo apt-get install python3-matplotlib python3-numpy python3-geopy" for the requisite pieces of the puzzle.

My data => distance 9195.8 km arrival at 1642248822 (2022-01-15 04:13:41) here about 3 km from the water between Seattle+Tacoma

It looks like your pressure was fairly flat. Maybe expand the time range, so that's obvious?

 

[Cameron D]

Thanks for that obsession. I was wondering how to start learning about plotting in python.  I've not got very far.

What I have done in the attached version is:

• add option towork with mysql db
• plot first 3 pulse times.
• added a background grey bar for 3 min either side of hte expected time.
  
• also a wide-scale plot covering them all.
• as you get closer to the equator, tidal changes dominate the baseline in that timescale - I tried higher order polynomials, but they are next to useless.

This version generates 4 png files.

I was even able to see a small dip at the expected time it was making its second pass around (as seen in the image). The dip far left is the opposite direction pulse.

[morr...@gmail.com]

On Wednesday, January 19, 2022 at 12:42:04 a.m. UTC-4 Cameron D wrote:

• as you get closer to the equator, tidal changes dominate the baseline in that timescale - I tried higher order polynomials, but they are next to useless.

I also had little luck with higher order polynomials to remove the general trend.

I've put the script here: https://github.com/morrowwm/weewx_tonga_browse

[bgra...@umw.edu]

Hello,

Not being a programmer, I probably shouldn't have messed with this, but being curious...

I tried the code posted on github as well as the one by Cameron D. In both cases I got the following error:

```

root@n4mrv:/home/bg/weewx_tonga_browse-main# python3 ./tonga.py   [file from Cameron D]

distance to eruption 12056.6 km arrival at 1642258360 (2022-01-15 09:52:39)
      opposite pulse arrival at 1642308921 (2022-01-15 23:55:21)
      second time around pulse arrival at 1642385471 (2022-01-16 21:11:10)
Traceback (most recent call last):
  File "./tonga.py", line 178, in <module>
    plot_burst( cursor, arrival_time, hour_span, "primary" )
  File "./tonga.py", line 54, in plot_burst
    coeff = np.polyfit(xdata, ydata, background_order )
  File "<__array_function__ internals>", line 180, in polyfit
  File "/usr/local/lib/python3.8/dist-packages/numpy/lib/polynomial.py", line 638, in polyfit
    raise TypeError("expected non-empty vector for x")
TypeError: expected non-empty vector for x

```

I added my lat/lon information but may have missed something else I need to change. Python modules were installed as directed. Copy of weewx.sdb is in the same directory as the program.

Thanks.

Bob

[morr...@gmail.com]

It looks like you retrieved no data. It calculated decent looking arrival times, so no problem with the change you made?

Try just a query for data spanning the first arrival. From a shell prompt:

sqlite3 weewx.sdb "select datetime, barometer from archive where datetime > 1642258360-100000 and datetime < 1642258360+100000 order by dateTime;"

[Cameron D]

yes, definitely looks like there is no data.

I have attached another version of mine, in which the trend line is disabled by default, but I suspect that would just delay the inevitable and it would crash trying to do the plot.

I also fixed up a few plotting errors in my code to do with the mysteries (to me) of layer ordering.

I also had a background bar showing either side of expected arrival - in this version I have now changed that to start at the expected arrival and stop 1 hour later.

[storm...@gmail.com]

Download the latest script from https://github.com/morrowwm/weewx_tonga_browse. Beside installing the Python Modules Vince stated above, I needed to also install  python3-scipy. I do have data between 1642172400 and 1642693629; except for a couple "null" in that time period.   So when I run the script, I get the following error.

raspberrypi:~/Desktop/Tonga $ python3 tonga_barometer.py
distance to eruption 13293.9 km arrival at 1642261629 (2022-01-15 10:47:08)
select datetime, barometer from archive where datetime > 1642172400 and datetime < 1642693629 order by dateTime;

Traceback (most recent call last):

  File "tonga_barometer.py", line 47, in <module>
    coeff = np.polyfit(xdata, ydata, 2)
  File "/usr/lib/python3/dist-packages/numpy/lib/polynomial.py", line 590, in polyfit
    y = NX.asarray(y) + 0.0
TypeError: unsupported operand type(s) for +: 'NoneType' and 'float'

There should be a way to check for "null" data within the time period.

[Cameron D]

To eliminate NULL data points use a query like:

select datetime, barometer from archive where datetime > 1642172400 and datetime < 1642693629 and barometer is not null order by dateTime;
 You could add the part in bold into the query in the python script.

Of course, if they are all null...

[Cameron D]

and add a debug printout for how many result lines there are after ...

    result = cursor.fetchall()

add the line:
    print( "query returned {} data points".format(len(result)))

[storm...@gmail.com]

Modified the query and added the print statement as suggested by Cameron D and here are the results:

Morrowwn Script:

raspberrypi:~/Desktop/Tonga $ python3 tonga_barometer.py
distance to eruption 13293.9 km arrival at 1642261629 (2022-01-15 10:47:08)
select datetime, barometer from archive where datetime > 1642172400 and datetime < 1642693629 order by dateTime;
Traceback (most recent call last):
  File "tonga_barometer.py", line 47, in <module>
    coeff = np.polyfit(xdata, ydata, 2)
  File "/usr/lib/python3/dist-packages/numpy/lib/polynomial.py", line 590, in polyfit
    y = NX.asarray(y) + 0.0
TypeError: unsupported operand type(s) for +: 'NoneType' and 'float'

Cameron D Script:

raspberrypi:~/Desktop/Tonga $ python3 tonga_baro.py
distance to eruption 12255.7 km arrival at 1642258384 (2022-01-15 09:53:04)
      opposite pulse arrival at 1642306911 (2022-01-15 23:21:50)
      second time around pulse arrival at 1642383509 (2022-01-16 20:38:29)
query returned 96 data points
query returned 96 data points
query returned 95 data points
tonga_baro.py:108: RankWarning: Polyfit may be poorly conditioned
  coeff = np.polyfit(xdata, ydata, 5 )

Note: LAT/Long and speed of sound are identical in both scripts.

[Cameron D]

the RankWarning is just a warning, you should still have image files produced.  I expect the polyfit it just ignores higher order terms and just returns a quadratic.

That is because the polynomial is a poor representation of the background, and the spline fit works better, if you get the parameters correct.

Initially you are better just to view the results without trend removal.  If the peaks  do not stand out then  background removal is unlikely to help much.

Also, with my code,  make sure you start with oversampling set to 1. Only adjust it if you see a stair-step effect.

[storm...@gmail.com]

Yes, the image files are being produced. Only thing I noticed was the arrival time for the opposite pulse is off by approx. 20 minutes according to the actual PWS data, this could be due to the wave speeding up or slowing down as it made its way around the globe.  Right now I'm using your default settings.

[Cameron D]

There are certainly uncertainties in the barometric data timings. My system is probably a worst case - it only reports a value every 15 minutes, and I don't know if that is an average of the previous 15 minutes or just a snapshot reading at the transition time. So for me 20 minutes is within the uncertainty.

Also, I don't know if it is expected, but the  pulses that travelled further show only the negative part, which comes after the positive. So there is extra lag in that.

I derived a speed by fitting data to Australian bureau charts, and they only show 30 minute samples, so my speed is hardly definitive.  I would expect it to vary a bit around the globe.  The average speed I derived from the first peak alone was 0.30 m/s, so that is either an indicator of the uncertainty in the fitting or that the average speed increased as it progressed.

[storm...@gmail.com]

If I recall looking @ MADIS 15 minute data, it least for me, the pressure of the opposite wave did drop a some before the major spike. After changing the colors on the charts, it aligns with the first drop in pressure.

[Tim Tuck]

Hi all,

The script works well on my data but the pressure readout is in inHg rather than hPa.

Can an option be inserted to have hPa computed before plotting ?

thanks

Tim

[Cameron D]

I assume your DB is storing in US units - mine is metricWX (I knew one day it would be useful to have done that!)

One easy way is to modify the queries yourself - where the  query (2 queries in my code)  specifies "barometer", replace with 

"barometer * 33.8639 as barometer"

[Tim Tuck]

Hi Cameron,

Beut! - that worked a treat - thanks :)

I've also noticed that the return pulse highlight appears to be off, see pic below. Or is it meant to be like that ?

thanks

Tim

[Cameron D]

I don't understand the physics of it. As per the last few postings the speed of sound may be variable enough to show up here.

I would just tinker with the speed of sound parameter and see if you can make all peaks look sensible. At least you have data at a good frequency.

[bgra...@umw.edu]

Thanks all, bad copy of weewx.sdb. All is well.

[paul.ba...@gmail.com]

Hello,

I did a little experience using LOWESS (also called  LOESS in some places) with success I think. In my experience it is the best method I tried over the years.

This non parametric local smoothing exists for many lingages including Fortran (which I used...), C, Python, R etc.

It use three parameters: a smoothing factor, adjusted between 0 and 1. You have to experiment to find the best according to your problem. In my case .03 was good.. the second is the number of iterations, 1 to...; I adopted 2, the doc says any larger number will not change antthing; the last one is a speed-up factor; I choose 0 (= No Speed-Up). With 3639 data points, the processing time is barely seen (10 years old Intel i3770 processor). 

The upper pannel represent the original data in blue and the smoothed one on red.

The lower pannel represent the difference between the original data and the smoothed one.

The time (T) is given in days, starting at 0:00:00 UTC on January 15th. The last point is at 24:00:00 on January 18th.

The vertical axis is in mbar. My altitude is 410 meters.

The first and second pulses are very clearely visible. There are a few other signals needing interpretation...  Also noticable is the smoother signal before the first spike compared to the rest of the data.
Have a nice week-end,     Paul

[storm...@gmail.com]

Here some graphs plotted from Cameron D script.  

distance to eruption 12255.7 km arrival at 1642260006 (2022-01-15 10:20:06)

opposite pulse arrival at 1642310592 (2022-01-16 00:23:12)

second time around pulse arrival at 1642390434 (2022-01-16 22:33:54)

[Ted Frohling]

Thanks very much for the new python script.  It works.

The other three files looked great as well.

They were all for Tucson, AZ, USA

ted

--
You received this message because you are subscribed to the Google Groups "weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to weewx-user+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/71655988-a5d6-4771-bb56-12fa455f24f2n%40googlegroups.com.

[morr...@gmail.com]

Did an update to my script at: https://github.com/morrowwm/weewx_tonga_browse

Mostly incorporating additions from Cameron (thanks!)

All parameters you likely want to set are near the top of the script.

1. support for mysql (untested, I don't have a mysql based weewx installation)
2. calculation and display of all three initial arrival, arrival from opposite direction, and second arrival after one trip around the globe
3. highlighting of the three events with grey bars
4. cubic spline curve fit of raw data and use of that fit to remove the base pressure variations. The new smoothing_hours parameter can be used to control how close the fit is. 
5. changed the display of data to highlight the extracted events and subdue the original data, other cosmetic changes

I had a data outage about 26 hours after the event, we're quite a distance from Tonga, plus a major storm was clearing out in our area. So not obviously anything unusual happened.

Some scientist might be interested in the arrival times of all this data, though.

[pmcg...@gmail.com]

WOW!  This is some interesting data, here is my location in Northern Utah.  Looks like the reverse may be a little off in calculation but this made it easy to find instead of hunting through the data.

Great Work!

[morr...@gmail.com]

The curve fit (solid black line) isn't matching your data (turquoise dots) really well. Try reducing smoothing_hours to maybe 12, or 6. 

The true model to estimate arrival times accurately is probably a lot more complicated than time = distance / constant_velocity. 

[storm...@gmail.com]

Nice work (Morrowwn  thanks!).  If you have any nulls in the data during the selected timeframe, change the query to read:  (Cameron D thanks!)

query = "select datetime, barometer from archive where datetime > %.0f and datetime < %.0f and barometer is not null order by dateTime;" % (start_time, stop_time)

If you want the bars closer to the rise and falls of the events, need to adjust the speed of sound slightly.

[morr...@gmail.com]

Forgot that, thanks for the reminder. Added now.

[morr...@gmail.com]

Also, I found a bug in calculating the smooth (spline fit) curve which is removed to highlight the eruption events. Added display of the points used to calculate the fitted curve.

[Cameron D]

Thanks morrowwm - much better to work from one source.

If anybody is needing to use the mysql version you should be warned that there is a ridiculously large number of mysql connector packages, and the one I used under Debian 10 is not packaged in Debian 11.

I tried one of the Debian 11 ones and the error handling codes are reported differently, so the easiest way out would be to just delete those lines and just print the print(err).

Drop me a line if you need help on that.

Cameron.

[bgra...@umw.edu]

Hi morrowwm,

I have tried your latest github post and am getting the following error:

```

root@n4mrv:/home/bg/weewx_tonga_browse-main# ./tonga_barometer.py

distance to eruption 12056.6 km

arrival at 1642257762 (2022-01-15 09:42:41)
opposite pulse arrival at 1642307533 (2022-01-15 23:32:13)
second time around pulse arrival at 1642382887 (2022-01-16 20:28:06)
select datetime, barometer from archive where datetime > 1642240800 and datetime < 1642395600 and barometer is not null order by dateTime;
query returned 171 data points

Traceback (most recent call last):

  File "./tonga_barometer.py", line 96, in <module>
    knots = np.linspace(np.min(tdata), np.max(tdata), (stop_time-start_time)/(3600*smoothing_hours), endpoint=True)  # spline knot every N hours
  File "<__array_function__ internals>", line 180, in linspace
  File "/usr/local/lib/python3.8/dist-packages/numpy/core/function_base.py", line 120, in linspace
    num = operator.index(num)
TypeError: 'float' object cannot be interpreted as an integer
```

The earlier version worked correctly and I don't think I have changed anything. I am using the same copy of weewx.sdb and only changed my lat/lon. 

Much appreciated for your work on this software.

Cheers,

Bob

[Tom Hogland]

Mine, from 9400km (Alaska). Not too bad of a match, although reducing speed to .31 brings it closer. Regardless, using .32 lets you pick out the pressure waves. We actually heard the initial eruption up here was well...

[bgra...@umw.edu]

Bingo! Many thanks!  Bob

bg@n4mrv:~/weewx_tonga_browse-main$ sudo python3 ./tonga_barometer.py

distance to eruption 12056.6 km

arrival at 1642257762 (2022-01-15 09:42:41)

opposite pulse arrival at 1642307533 (2022-01-15 23:32:13)

second time around pulse arrival at 1642382887 (2022-01-16 20:28:06)

select datetime, barometer from archive where datetime > 1642240800 and datetime

 < 1642395600 and barometer is not null order by dateTime;

query returned 171 data points

Fitting between 1642240800.0 and 1642395600.0 with 10 knots

[storm...@gmail.com]

Data from both Davis and Acurite PWS

[morr...@gmail.com]

OK, I've put the fix up on github. Thanks for finding this bug.

[morr...@gmail.com]

Interesting, no sign of a pulse from the second trip around.

[Paul Eaton]

Here's one from England. A bit messy after 07:00 on 16-01 but the Initial and Reverse pulses show up very clearly. Best match is with speed reduced to 0.30 km/s.

[Gérard P]

Very interesting thread, using same speed as Paul, and here is a plot from France near Paris.

[DR]

There must be locations where the initial and reverse waves met at the same time.  Was anyone along that line that has tracings to share?

On 1/30/2022 11:55 AM, Gérard P wrote:

Very interesting thread, using same speed as Paul, and here is a plot from France near Paris.

--
You received this message because you are subscribed to the Google Groups "weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to weewx-user+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/4a390999-0824-44ab-aaf0-4c2209206783n%40googlegroups.com.

[Karen K]

daleea...@gmail.com schrieb am Sonntag, 30. Januar 2022 um 20:54:51 UTC+1:

There must be locations where the initial and reverse waves met at the same time.  Was anyone along that line that has tracings to share?

That's somewhere in Niger in Africa. 

 

[Gérard P]

I made a mistake (wrong sign) on longitude, here is the correct graph with better correlation for the three peaks.

[morr...@gmail.com]

There's a wunderground feed from the Diori Hamani International Airport at lat/long 13.4, 2.2, but is doesn't seem to have barometric pressure history.

I put in approxiamte antipodal lat/long of (20.54, 4.7) and it calculate arrivals as:
distance to eruption 20002.7 km
arrival at 1642282593 (2022-01-15 17:36:33)
opposite pulse arrival at 1642282702 (2022-01-15 17:38:21)