Quake Watch 2012

Originally posted by aorAki
deep canterbury

A nice deep one (134.44 km) off Banks Peninsula. Just a baby though.

quake circles

I like the way this data is presented. Still trying to convince a friend to do 3D projections (location, depth) for the recent (since 2010) Canterbury series.

My seismologist friend is certain that the 16000 year recurrence interval isn't correct and that it's more 'regular' than that.

Its not even a baby yet, still in the womb
I see Canterbury Quake Live are using Geonet Rapid now, caution advised , that quake hasn't made the official Geonet database, is an automatic reading and hasn't been manually checked (see the R on the list?)
very deep and out of character for the area which is 1-30km normally.

Still in drought, only 3 confirmed listed yesterday for the whole of NZ, 2 down your way and one north of me at Managahao reservior east of Levin again.

What I want to know is......

Well lots of things but mainly why to the USGS hide earthquakes behind different scales?

:RANT ON

I know we have been over this is a hundred or even a thousand times but it still irritates the hell out of me when they do it.

Here is an example.

Magnitude 5.6 - NEAR THE NORTH COAST OF PAPUA, INDONESIA

Hey, no problemo it is a 5.6. Um no, that is mb - it is a 6.0 Mw. Who sez? EMSC sez. Geofon almost sez - they have it at 5.9 Mw, but USGS have it as a 6.0 Mw themselves - The Global CMT project - but I accept that they did not calculate that but they are promulgating it and therefore one has to assume that they accept it. LDEO are simply ignoring it.

The point is that at some stage it WILL get converted to Mw most likely, and end up in the ANSS catalogue.

It is in there as Mb at present

Date Time Lat Lon Depth Mag Magt Nst Gap Clo RMS SRC Event ID
----------------------------------------------------------------------------------------------
2012/04/21 01:16:52.84 -1.6030 134.2740 16.00 6.60 Mw 323 1.15 NEI 201204211001
2012/04/21 01:19:28.23 -35.1890 -16.1440 10.00 5.00 Mb 27 1.07 NEI 201204211002
2012/04/21 01:25:12.49 -1.5770 134.2050 10.10 5.60 Mb 146 1.23 NEI 201204211003
2012/04/21 11:04:30.65 3.2960 93.7300 10.00 5.20 Mb 115 0.88 NEI 201204211017
2012/04/21 11:36:04.70 -14.8230 -71.4520 105.60 5.40 Mb 412 0.80 NEI 201204211019
2012/04/21 12:36:10.96 -21.8590 170.1440 39.50 5.20 Mb 53 0.86 NEI 201204211021
2012/04/21 13:23:39.30 1.6880 120.9200 51.00 5.00 ML 46 1.33 NEI 201204211024
2012/04/22 00:35:06.75 23.2130 143.5220 48.90 5.00 ML 163 0.67 NEI 201204221002
2012/04/22 08:13:57.93 32.4180 47.1620 40.00 5.10 ML 52 0.63 NEI 201204221006
2012/04/23 02:14:24.11 -23.9890 -175.9220 42.80 5.10 Mb 164 0.77 NEI 201204230005
2012/04/23 03:08:01.74 -17.9600 -175.1200 192.40 5.00 Mb 32 0.90 NEI 201204230007

Source: ANSS

Why to they persist in making themselves look stupid and inept?

There is absolutely no reason why they cannot calculate the Mw value for ALL quakes and post ONLY Mw values (JV will probably jump on me with a good reason why not ) and it would bring some consistency to the face that USGS is showing to the public.

Anything over 5 or maybe 5.5 Mb should be converted. Some one give me a good reason why not. Is it any wonder the public think that USGS is lying? No, because they don't understand the scales and see the changes as shenanigans.

:RANT OFF

looks like they felt that small one in la 30 min ago
latimesblogs.latimes.com...

reply to post by stanguilles7


I got notice 13s after the origin time from my nifty earthquake early warning app. Estimated M3.5 vs the final M3.9, and depth of 9 vs the current 3.7km. Not bad.

It also correctly forecast that I would feel no shaking from the waves arriving in several minutes.

reply to post by PuterMan


I didn't notice the magnitude rant until just now. You linked the old page, this is the new, official page for that earthquake

earthquake.usgs.gov...

Mwp = 5.9

Clicking down to the technical page says who derived it and how, and clicking further down (for more significant events, not much to see for this one) to contributed solutions lists all the entities that tossed in estimates.

You can click still further down, to toggle data visibility, under magnitude, and see the magnitude readings for individual stations.

Mw is the name of a magnitude derived from a moment, but still the moment could be derived from the P waves, from the P and S body waves, or from the surface waves (or even from the normal modes).

This link (you need to cut and paste in your browser) www.google.com... 2FLee%2FMyDoc_Backups%2FMyDocs_20120309%2FMISC%2FProjects%2FGEM%2FMagnitude%2520Nomenclature%2520Standard.doc&ei=W9CVT9yQCajjiAKosuiJCg&usg=AFQjCNGkuF xETCF3XL7cEDSWN3t0cpUHwQ&sig2=uOPMwfL5GIfaRx1aGn9OXg shows a list of magnitudes the NEIC might cite.

unformatted version:
Phase(s) Algorithm External Name Chicxulub Long Code Chicxulub Short Code NEIC Standard Code
R&G CMT Mw(CMT) Mw(UCMT) MW(UCMT) Mwc
P&S teleseismic MT Mw(MT) Mw MW Mwb
P&S regional MT Mw? Mw? MW? Mwr
W-phase LP amplitude Mw(W) - - Mww
R LP stack Mw? - - Mws
P, pP, sP BB depth Mw? - - Mwd
P SP coda fit Mw(coda) - - Mwca
P moment Mwp Mwp MP Mwp
P SP amplitude mb mb MB mb
P BB amplitude mB(BB) - - mBBB
P energy Me Me ME Me
R LP amplitude Ms Ms MS Ms
R BB amplitude Ms(BB) - - MsBB
R multi-band amp Ms(VMAX) - - MsVX
P&S SP coda amplitude ML ML ML ML
LG SP amplitude mb(Lg) mbLg LG mbLg
P&S coda duration Md Md MD Md
P&S coda amplitude Mca Mca MC Mca


Mwp is calculated from only the P waves for rapidity, but sometimes deeper small events generate ratty surface waves and marginal S waves, and Mwp is preferable.

I don't know the full logic tree NEIC uses, but to search for a single one-measure captures everything magnitude is a Sisyphean task.

reply to post by JohnVidale


Thanks for that John, but that sent me off on one again since it is the 5.6mb that is coming out in the feeds. So the feeds and the web site are not coordinated. Would not surprise me one little bit.

Just checked it out. You linked to the wrong quake John. This is the one and that is still showing 5.6 in all the listings.


For those of you who use GEE there is a new version out today

We have made a new release of GEE! Version 2.1.5 can be downloaded here. There are only a few noticable changes, but still significant.

New installer - We have a new installer for windows. It is a little clunky at present, but is free. Hopefully this will enable us to put out future releases more easily, and hence more frequently.
New servers - To deal with some network firewall issues that were blocking access to the servers that GEE needs, we have switched to a separate set of public servers. This should allow GEE to get data that was being blocked in the previous version.
Bug fixes in libraries - While we have not done much work on GEE proper, GEE depends on many code libraries that have had significant work done in the years since the last release. This fixes several commonly encountered bugs.

GEE download page

WARNING to dialup users: This file is 51 MB

Download the file - a jar file. Double click it and keep clicking to the end!! Installs just fine. Looks much the same as the old one.

reply to post by PuterMan


I got this email 3/9 - "The beta version has been available on the event pages for several months,
and has been rapidly evolving based on public feedback and input from GHSC and ESC."

And this went out 3/30 - "Barring any unforeseen issues today, we're excited to inform you that we'll be launching new earthquake maps and event pages on Monday, April 2nd. You'll also see a new EHP homepage and new Earthquakes section main page."

So the new page is just out.

John

[edit - I did link the wrong one. You're quite right, they used Mb, and MW from Lamont is 6.0. The two are measuring different passbands, Mb being shorter period. On the other hand, the Mb average 92 stations from around the world, so it might be more stable than the MW estimate.]

Do we have the Molucca Sea at 5.9 logged yet ?

www.emsc-csem.org...

& Just a kwiki to post Crete having a burp - 4.8

www.emsc-csem.org...

reply to post by JohnVidale


Geofon clocked it at 5.9 Mw as I said with 57 stations

Unfortunately the Russians only have th mb value this time. Sod's Law I guess!

Ye have to admit they do a nice map!

Oz clocked it as 5.7 mb

Indonesia has it as 5.7 ML

Earthquakes from around the world!! Wicked.

hmmm, today seems like the march of the 5.8/5.9- however you want to measure them earthquakes. Similiar sized quakes that have shown up in the last half-dozen hours. Almost regular like. Starting in the south and migrating north.

hmmm

Kermadec 17:36 -117
Molucca 21:21 0.321
Kuril 22:40 154.779

The Molucca Sea earthquake interests me very much. I was wondering aloud earlier about Asia getting squeezed. I think this is the case and Molucca kinda supports my case. Here's a bit about the micro plate and it's unique tectonic forces. It's literally being squeezed and sunk by other plates.

en.wikipedia.org...

reply to post by PuterMan


Personally, I always use the Lamont number, if it is available, as Goran has included lots of nice corrections and carefully calibrated it - he's an artist. But for fast, good numbers, the USGS page is great, plus I can call them with questions, should they arise.

But really, the difference between 5.6, 5.9, and 6.0 is not so much for any practical purpose, so long as one is consistent, and not cherry-picking between estimates to try to fudge marginal statistics.

reply to post by JohnVidale

But really, the difference between 5.6, 5.9, and 6.0 is not so much for any practical purpose, so long as one is consistent, and not cherry-picking between estimates to try to fudge marginal statistics.

Very true, and exactly the reason why when I do my reports I only use USGS data. This is also however why it is important to have as many as possible of the Mag 5+ magnitudes on the same scale otherwise statistics can be skewed.

It is statistical male cow excrement to have different scales in use for the earthquakes that matter and better to have all as Mw even if some are based on fewer stations. The difference between the estimate based on 57 stations for Geofon and ?? for the Global CMT project cannot be great and the difference between 5.9 and 6.0 is much smaller than the figure of 5.6 and 5.9.

In fact earthquakes should ONLY be described in energy terms. Magnitudes are just rubbish unless they all sing from the same hymn sheet.

reply to post by PuterMan


The complication is that different numbers are more robust in different magnitude and distance ranges, and one can't legislate what other networks, and especially other countries are going to use.

In the Pacific Northwest, we use duration magnitude up to 2 or 3, then ML up to about 4 or so, then MW, once we've computed a moment tensor, above that. There is simply no long-period energy above the noise to compute MW at small magnitudes.

This is also why ML was predominant for so many years. With a velocity sensor good out to only about 20s period, it could capture magnitude well from M2 up to about M7 before saturating.

None of these are energy. Energy calculations are notoriously unreliable because they heavily count high frequencies, which in term rely on correcting properly for attenuation of the waves along the path. Attenuation structure is highly variable, and focusing and defocusing also causes wide variations in high-frequency amplitude, as does near-station geology.

reply to post by PuterMan


I agree with what you are saying, but my thoughts are perhaps a bit at odds with current thinking, if consistancy is what you are after why not just use mb?
As you know mb are just about always available for the larger quakes from USGS in the phase data, but that might not help with your automatic programme unable to pick that out?
The Russians always have mb as well, so a good cross checker. I have been a fan of RAS for about 5 years because of that, consistancy by primarily using mb.
What will happen though if you (or I for that matter) just use mb, is the numbers will be lower than now (Mw)

Even just using ML for New Zealand I have used the micro, very minor, minor , light, moderate, strong, very strong and major tags, rather than Mag 1,2,3,4,5,6,7,8 , although the numbers behind the coloured markers on my maps do match the word descriptions.
I have been wondering if it would be more accurate to go the japanese way and go half magnitudes for the ranges, for example a "moderate" quakes should be 4.5-5.5 rather than 5.0-5.9, and a "strong" quakes should be 5.5 to 6.5, a lot of 5.7's are what I'd call strong quakes.But thats not what you were talkng about, , me just rambling on again.

reply to post by muzzy

if consistancy is what you are after why not just use mb?

Because Mw is the ONLY scale that is consistent across all magnitudes and mb is saturated beyond about 7 max. Mb only has a use as an initial assessment and seeing as it is based on a max 5 seconds slice of an incoming P wave it cannot be accurate (not that any scale is but one would hope to be as accurate as possible)



For precisely the same reason this Mwp scale that seems to be creeping in is also inaccurate.

I have no problem with initial estimates of the magnitude being in mb, ml, mi, or whatever but WITHIN 7days (not after) it shuld be possible to give an Mw value for all earthquakes. If it might be considered that there are not enough stations to create a firm Mw value then do it and say so.

I would rather see both mb and mw quoted side by side than have some using one and some using another. That would be consistent.

I haven't seen this mentioned yet:

Magnitude mb 5.4
Region CARLSBERG RIDGE
Date time 2012-04-24 09:51:03.0 UTC
Location 5.61 N ; 61.43 E
Depth 40 km
Distances 1829 km E Mogadishu (pop 2,587,183 ; local time 12:51:03.6 2012-04-24)
1499 km SE Salalah (pop 163,140 ; local time 13:51:03.6 2012-04-24)
1241 km SE Xafun (pop 10,919 ; local time 12:51:03.6 2012-04-24)

link: www.emsc-csem.org...

reply to post by JohnVidale

None of these are energy.

The moment of magnitude can be used to quantity both shallow and deep earthquakes on the basis of wave energy radiated and provides a uniform scheme.

(Emphasis by me)

Magnitude scale and quantification of earthquakes (PDF 700kb)
Kanamori. H. 1983

In addition read pages 189 and 190 of that document (4 to 6 of 15 physical page numbers) where it is clearly described that the mb scale is at odds with other magnitude scales and was developed for seeking out nuclear explosion signatures, and the remainder of the document compares the different scales. There is little doubt in my mind that not only is Mw measuring energy but that as a scale it transcends all others except Ms as an accurate determination of magnitude, although as I say I would prefer to see M0 in dynes or joules.

I really really hate to quote Wikipedia but.....

The moment magnitude scale (abbreviated as MMS; denoted as MW) is used by seismologists to measure the size of earthquakes in terms of the energy released.

Source: Moment magnitude scale

Since the Mw scale is based on M0 which is the magnitude of the seismic moment in dyne centimetres the Mw scale IS based on energy and nothing else. It is far and away a better measure for ALL earthquakes over Mag 5, or perhaps Mag 5.5 but in any event any quake that may be classified in that level or above should be calculated as Mw.

If it is possible to perform a moment tensor then the scale to be used should be Mw. It is complete nonsense to perform a moment tensor and then continue to quote the earthquake in mb.

I have already accepted that it may not be possible to get an Mw value for the smaller quakes.

More Mw reading for the avid quake nerd

Eartyhquake Magnitude Scales: James P McCaplin (Undated)(PDF 1.5Mb)
USGS: Magnitudes

Of special interest is the definition of global relations converting different magnitude scales to the most reliable and useful scale of magnitude, the moment magnitude, Mw

.
Empirical global relations converting MS and mb to moment magnitude(Scribd)
E.M. Scordilis Department of Geophysics, School of Geology, Aristotle University, Thessaloniki 54124, Greece. Received 28 December 2005; accepted in revised form 12 January 2006

Unfortunately the above document is pay-walled if you want to download. You can read for free however.

An Evaluation of the SCSN Moment Tensor Solutions: Robustness of the Mw Magnitude Scale, Style of Faulting, and Automation of the Method
John F. Clinton, Egill Hauksson, Kalpesh Solanki

From the above link:

We have generated moment tensor solutions and moment magnitudes (Mw) for > 1700 earthquakes of local magnitude (ML) > 3.0 that occurred from September 1999 to November 2005 in southern California. The method is running as an automated real-time component of the Southern California Seismic Network (SCSN), with solutions available within 12 mins of event nucleation. For local events, the method can reliably obtain good quality solutions for Mw with ML > 3.5, and for the moment tensor for events with ML > 4.0.

Earthquake magnitudes

Empirical relationships for magnitude and source-to-site distance conversions using recently compiled turkish strong-ground motion database. (PDF 463kb)
Note in the above:

The most common magnitude scale published by the agencies is Md and this is followed by mb. Although the most reliable magnitude scale is moment magnitude (it does not suffer from saturation as in the case of other scales), it is the least existing magnitude scale among the others.

I guess that is enough to be going on with.

reply to post by megabogie


Unfortunately, moment is not energy, although it has the same dimensions. I'm surprised you're so confident about this issue.

It is obviously not energy radiated, the quantity represented in Me, moment based on energy, as an earthquake could happen so gradually that no energy is radiated, which is called quasi-static. This case would still have moment but an Me of zero.

It is also not total energy released, which goes to radiation, cracking, and heat. To see that, consider a highly stressed fault and low stress fault. If they both move with the same slip across the same area, the more highly stressed fault will release more energy - just like moving a spring, energy is the product of resisting force times distance moved. It takes more work to move a stiff spring than a soft spring. But both earthquakes will have the same moment.

In order to measure energy, one either has to know the absolute stress levels and the motion, or measure the radiated energy and the energy expended in cracking and heat. The input or the output. Only a small, unknown fraction of the energy is radiated and we don't know the absolute stress, which could range from less than the earthquake stress drop (overshoot) to 10 or 20 times the stress lab (which is observed in lab experiments on rock).

reply to post by PuterMan


Hmmm, my reply above is directed to the wrong person.

To sum up, if all earthquakes had stress drop equivalent to the pre-quake absolute stress, perhaps energy would generally be proportional to moment, although I expect geometric factors would vary for different kinds of faulting, normal, reverse, strike-slip and also for vary with hypocentral depth relative fault length.

But as the ratio of stress drop to absolute stress varies widely, relating energy to moment has much more uncertainty than the disadvantages in measuring earthquake size with varying magnitude measures, depending on the frequency and path of seismic waves available to measure.

reply to post by JohnVidale


I agree with you that the energy being measured is not the total energy being radiated by the earthquake, of that I am fully aware, however it seems to me from my investigations into the scales that the Mw scale represents the best estimate of magnitude based on the measurable portion of energy. Not only that it seems that most seismologists are in agreement about this.

Reading back through this we may possibly be talking at cross purposes. Mw as I understand it refers to the radiated energy of the earthquake and thus other energy considerations do not come into it.

Of course nothing is ideal, but I will continue to maintain that the Mw scale should be used at every opportunity and mb, which is not efficient, should not be used by the USGS to cloak the fact of another Mag 6, for example, where the Mw scale has been calculated by the Global CMT Project, in other words a moment tensor calculation has been made.

I can guarantee you that the earthquake in question will suddenly magically become a Mag 5.9 Mw or probably Mag 6.0 Mw later if things run their normal course.

You probably don't spend much time tracking changes that are made but I do and I have seen and commented on this behaviour by the USGS on many occasions. It used to be that the second they were off the 7 day listing the change was made. Maybe now it is a 30 day listing it will be 30 days, but which ever it is you can be assured I WILL be watching.