Mw 7.0 Papua Quake of 06 Apr 2013

[davenn]

hi all

meant to post this earlier ... got busy


the above mentioned Mw 7.0 quake is still visible on my digital seismograph screen at
http://www.sydneystormcity.com/quakedrum.GIF for a few more hours if anyone interested :)

This had very large amplitude Surface Waves compared to the majority of M7 events that I have recorded

cheers
Dave

 

[Astronuc]

So far a mild swarm:

Mag    Location                                Date and Time (UTC)   Long      Lat      Depth(km)
4.5 240km WSW of Abepura, Indonesia           2013-04-06  12:06:26  3.409°S  138.560°E    69.6 
4.6 241km WSW of Abepura, Indonesia           2013-04-06  06:20:46  3.400°S  138.553°E    68.3 
4.7 238km ENE of Enarotali, Indonesia         2013-04-06  10:04:10  3.468°S  138.447°E    86.0 
5.2 235km   E of Enarotali, Indonesia         2013-04-06  07:50:31  3.523°S  138.432°E    71.1 
4.1 220km ENE of Enarotali, Indonesia         2013-04-06  05:57:59  3.522°S  138.300°E    70.6 
7.0 239km   E of Enarotali, Indonesia         2013-04-06  04:42:36  3.526°S  138.466°E    68.0 

5.1  44km   W of Ambunti, Papua New Guinea    2013-04-06  12:27:56  4.194°S  142.444°E   118.6 
4.7  41km   W of Namatanai, Papua New Guinea  2013-04-06  06:56:50  3.723°S  152.064°E     8.4 
5.3  22km SSE of Taron, Papua New Guinea      2013-04-06  10:55:42  4.654°S  153.116°E    66.4 
5.2  38km WNW of Taron, Papua New Guinea      2013-04-05  19:55:56  4.314°S  152.721°E    35.1

It's on the northern edge of the Australian plate. There are separate quakes further east in PNG.

 

[davenn]

The 5.2 aftershock wasnt quite big enough for me to pick up at 3600km from my location
would have had to be at least a 5.7 ;)

only a few other M4 + aftershocks, reasonably small number considering the size of the mainshock
But that's probably partly due to the focal depth at ~ 70 km

Dave

 

[Greg Bernhardt]

davenn what seismograph are you using? Is it turned on and recording most hours?

 

[Borek]

https://www.physicsforums.com/showthread.php?t=646834

 

[davenn]

Thanks Borek :)
I knew that was somewhere in the system

Greg,

Yes, it runs 24/7 the system itself is pretty stable
the only dodgy part is the screen capture program, called Snagit, it is currently set to do a screen capture every 5 minutes and then it ftp's that capture file to my www site addy as listed in my first post. It crashes a couple of times a week and I have to restart it when I get home from work, or get up in the morning

Hopefully within the next month, my new force babanced vertical (FBV) seismometer will be up and running.
I am just awaiting its construction to be completed
have a look here for a PDF of the full description of the unit that is starting to become quite popular amongst amateur seismologists

The FBV is a broadband sensor and one can use software to bandpass filter the bandwidth of interest. I am hoping that I will be seeing signals down to a period of at least 25 - 30 seconds compared to the ~ 10 - 12 seconds that the current Lehman sensor is giving me.

Cheers
Dave

 

[Borek]

Do I understand correctly you are using one program to collect and draw the data on the screen, and other - unrelated one - to do screenshots of the data displayed by the first program (and to upload them to your web page)?

Does it mean there is no software that will do all three things (data collection, picture generation, uploading) in a one sweep?

I especially find the part of making screenshots (when you have a raw data) strange.

 

[davenn]

Do I understand correctly you are using one program to collect and draw the data on the screen, and other - unrelated one - to do screenshots of the data displayed by the first program (and to upload them to your web page)?

No, everything is done on the one dedicated PC

Does it mean there is no software that will do all three things (data collection, picture generation, uploading) in a one sweep?

well, the data collection software is supposed to be able to upload screenshots, but I have never been able to get it to work

I especially find the part of making screenshots (when you have a raw data) strange.

you mis-understand the purpose of making and uploading the screenshots
its purely so that you, me, anyone worldwide can go to that www page and see what my system is recording in near real time ( as I said earlier... 5 minute updates)
Some of the major professional seismological institutions have online seismo displays...

eg. ...
GNS - New Zealand
LISS[/PLAIN] - Live Internet Seismic Server


and many of us amateur/semi-pro guys are just doing the same :) ...

eg. ...
A station on the central east coast of NSW, Australia north of me ~ 150 km
A friend's station 50km inland from me
A collection of stations hosted on a Californian amateur seismo site

cheers
Dave

 

[Borek]

No, everything is done on the one dedicated PC

I know it is on one computer, I was asking if it is several separate programs running on one machine.

well, the data collection software is supposed to be able to upload screenshots, but I have never been able to get it to work

OK, that explains a lot.

you mis-understand the purpose of making and uploading the screenshots

I understand it perfectly. Just displaying the data on the screen to make a screenshot to save it on disk is like scratching you left ear with your right foot - doable, but there are simpler and better methods. Image should be generated directly from the raw data without using screen display as an intermediate. There are libraries and ready tools devoted to such an image generation, which is why I was surprised reading your description of what is going on.

Unless what you call a screenshot is not a real screenshot. But it definitely looks like one, with a title bar, menu and whatnot.

http://earthquake.usgs.gov/monitoring/operations/heliplots_gsn.php

See how their plots are JUST plots - no other stuff around.

 

[davenn]

I know it is on one computer, I was asking if it is several separate programs running on one machine.

There are 2 programs rumming on the PC, the datalogger prog and the screen capture prog

I understand it perfectly. Just displaying the data on the screen to make a screenshot to save it on disk is like scratching you left ear with your right foot - doable, but there are simpler and better methods. Image should be generated directly from the raw data without using screen display as an intermediate.

no, you still are not quite on the same page ;)
NOTE: the "screen capture" is all done within the puter. The screen doesn't have to be left on to do the capture, I turn the monitor off when I go to bed, turn it on in the morning to see any activity during the nite and turn it off again before going to work
The datalogger puter has its screen at my bench above the screen connected to the puter I'm typing this message on.
I as I type this, I am also looking at the datalogger screen seeing the raw data from the sensor in real time! no delays, no filtering. I can do filtering in the datalogger software if I wish, but I prefer ( as others also do) to do filtering on any saved event files. I can sit here and watch quakes roll in in real time.
It is the displayed raw data from that 1 of 4 channels that are being simultaneously recorded by the datalogger. I can display all 4 channels simultaneously if I wish, it just looks too messy, so I just display the long period seismometer channel.
The datalogger can handle 8 channels. At the moment I have 4 channels, 1 long period and 3 short period seismometers. When that other sensor arrives some time soon, it will become channel 5.

There are libraries and ready tools devoted to such an image generation, which is why I was surprised reading your description of what is going on.

not really sure what you mean there ? that is ... why would I need another image generating program ?
The datalogger generates the image to the screen, just as your word processor does so you can see what you are typing

Unless what you call a screenshot is not a real screenshot. But it definitely looks like one, with a title bar, menu and whatnot.

it is a screen capture(shot), just as if you had hit "cntrl and print screen" keys

See how their plots are JUST plots - no other stuff around.

thats just a difference in the capture/display method of the capture software and or the datalogger software, nothing more

cheers
Dave

 

[davenn]

As far as screen capturing goes ...

Now I aint no computer engineer ;) But I recall from the dim distant past that, isn't what is to be displayed on the monitor memory mapped ?
And that memory mapped data is sent as the VGA signal ( whichever resolution) to the monitor.
So all the screen capture software is doing is reading that memory mapped data at an instant in time ?

Dave

 

[Borek]

As far as screen capturing goes ...

Now I aint no computer engineer ;) But I recall from the dim distant past that, isn't what is to be diaplayed on the monitor memory mapped ?
And that memory mapped data is sent as the VGA signal ( whichever resolution) to the monitor.
So all the screen capture software is doing is reading that memory mapped data at an instant in time ?

Yes, screen capture program reads the data from the video memory, after it was put there by Windows GUI. You don't need to put this information on the screen to save it as an image, image can be rendered directly to the disk (actually it will be rendered into a normal RAM first, but it doesn't matter much). Using screenshots to put the data on the disk is in a way similar to calculating something on the computer and then manually copying the results on paper. It can be done, but there are better ways.

not really sure what you mean there ? that is ... why would I need another image generating program ?

To get better control over the image quality, to get rid of the Windows toolbar which is not needed, to make sure you are safe from GUI quirks, to make sure your program generates the same output regardless of whether it is working in the background or foreground, to have control over the resolution and color depth of your images and so on.

Not to mention the fact that using screenshots looks amateurish

The datalogger generates the image to the screen, just as your word processor does so you can see what you are typing

it is a screen capture(shot), just as if you had hit "cntrl and print screen" keys

Right. When you type something into word processor it displays the effect, but when you want to print the document you don't do a screenshot to paste the image into some graphical program to print it from there, you send it directly to the printer. Same with your data - they don't have to go through the screenshot to become an image on the disk, image should be created on the disk directly from the data.

thats just a difference in the capture/display method of the capture software and or the datalogger software, nothing more

Exactly - and I am more then sure they don't go through the screenshots, but they generate images using dedicated software.

 

[davenn]

Using screenshots to put the data on the disk is in a way similar to calculating something on the computer and then manually copying the results on paper. It can be done, but there are better ways.

That isn't being done :)

Same with your data - they don't have to go through the screenshot to become an image on the disk, image should be created on the disk directly from the data.

That isn't being done either :)

you still seem not to be understanding why the screenshot is done. It isn't stored anywhere except on my www site, and its only there for 5 minutes till the next uploaded shot overwrites it
The screenshot images NEVER get saved on the datalogger puter ( at all, not for any length of time). They never get permanently saved anywhere


Dave

 

[Borek]

I am 100% sure that making screenshots and uploading them is not the most reasonable and effective way of presenting the data from your seismograph and there are better techniques, but apparently my English fails me when I try to explain why. I give up.

 

[davenn]

no don't give up :)

If you have/know of a better piece of software that you know about for ftp'ing a snapshot of what is presented on the screen at that instant in time give me a link and I will investigate.
Snagit was the one I finally settled with after trying 4 or 5 different programs.
The only major problem I have with Snagit is that its not reliable, as I said a number of posts ago, it crashes regularly. I would love a prog that more stable :)

Dave

 

[jim mcnamara]

Davenn -

Correct me where I'm wrong.

Seismographs provide a physical picture - linear time graph - of analog data.

Seismographic data is kept, stored and analyzed using the output of pens moving on the surface of slowly moving paper streams. Like an EEG for tracking neural transmissions. The desired and most usable endpoint of your data input stream is the piece of paper with squiggly lines (or a picture of it).

The digitally transformed signal data are of no real use?

Or more naively: to perform data analysis, native data as wiggly lines on a long piece of graph paper is what you go to?

Borek wants to store the immediate predecessor input to the pens, not the emulated output of the pens. As I get it, that is analog data, which is then converted to digital data and amplified. I'd use a sound card. The digitally converted data could be kept for a long time, like using an mp3 format to store analog sound data. Or music if you think of music as data.

So there are points in the seismographic system where digital data transformation is made. Even if they may be transient now.

This may be complete nonsense to you but it sounds feasible to me. Useful? I dunno.

 

[davenn]

Davenn -

Correct me where I'm wrong.

Seismographs provide a physical picture - linear time graph - of analog data.

Seismographic data is kept, stored and analyzed using the output of pens moving on the surface of slowly moving paper streams. Like an EEG for tracking neural transmissions. The desired and most usable endpoint of your data input stream is the piece of paper with squiggly lines (or a picture of it).
The digitally transformed signal data are of no real use?
Or more naively: to perform data analysis, native data as wiggly lines on a long piece of graph paper is what you go to?

Greetings Jim
thanks for taking an interest :)

That's not what is happening. my system IS A DIGITAL system. There are no pen on paper recorders. The only place there is an analog signal is the velocity generated voltage from the sensor through the 300 - 500 gain low noise preamp to the input of the A to D chip. From there on its all digital, the datalogging, storage, and display.
In the late 80's and through the 90's I had a pen on rotating drum recorder. But they are pretty much a thing of the past now.

Borek wants to store the immediate predecessor input to the pens, not the emulated output of the pens. As I get it, that is analog data, which is then converted to digital data and amplified. I'd use a sound card. The digitally converted data could be kept for a long time, like using an mp3 format to store analog sound data. Or music if you think of music as data.
So there are points in the seismographic system where digital data transformation is made. Even if they may be transient now.
This may be complete nonsense to you but it sounds feasible to me. Useful? I dunno.

thats all pretty irrelevant ... I suspected he still wasnt understanding what was going on :) and if what you say is what he was thinking, I can now understand his confusion

I have the raw digital data for my 4 channels on the HDD that I can access any time and replay events, going back months. I can save that event into the analysis software, do FFT filtering, naming the event, putting in the lat and long, depth, date and time so it can calculate magnitude and distance from my station
The datalogging software is specifically designed for datalogging earthquakes. The analysis software is specifically designed to work with the data files from the logging software. It will even allow me to download raw digital records from the USGS/NEIC database of events and I can view that record of a given event along side my own record

this is a screen shot of a quake in the analysis prog...

hope that helps clear up the misunderstandings :)

Dave

 

[billiards]

Dave, I have to agree with Borek.

Rather than send a screenshot, get your computer connected to the data logger to send a seismic trace file (or 4 if you have 4 channels). (What format do you save your seismic data?)

Then have a script to generate a plot from the data. You can be as creative as you like, you could include FFTs in your plot for example. Whatever format you are using there is bound to be something freely available online to display your data. For example, although I haven't used it myself, I am told obspy is a pretty neat and flexible tool if you want to use python.

 

[davenn]

Hi Billiards

I don't think I understand what you are getting at ?

send a trace file ... where ? why? I am not going to be able to display it on the www site
I don't need an interactive or a file or anything like that accessable online. I don't need to be able work with the data remotely. No one else does. I can wait till I get home from work, holiday etc to analyse and save any quake events that the system recorded in my absence
It would be making something very complex that at the moment we are all doing very easily

all we all want to do, the professionals included, is to have our traces viewable online by anyone anywhere.
so they can see any activity over the last 24 hrs or so
Yes, there a bit of variation in the way various people are displaying the trace on their www sites, but the result is the same ... its just a xxx.gif or xxx.png file its all it needs to be

The data is saved in a very specific format. There's over 150+ of us world wide using this system
We all upload our event files to a repository in California, so that anyone ,anywhere can access each others recordings and do comparisons, event locations etc

regards
Dave

 

[billiards]

Hi Dave,

Your data logger can send the trace to the computer. The computer can then generate a plot from the data. The plot can be saved in some convenient format, and sent to the internet server for all to see. The whole process can be automated.

That way you have more control over the image and it should be more stable as you won't have to run snaggit. The only downside is that you would have to write some code to generate the plot, which might be a stumbling block.

 

[davenn]

Hi Billiards
thanks for hanging in there :)

Hi Dave,

Your data logger can send the trace to the computer. The computer can then generate a plot from the data. The plot can be saved in some convenient format, and sent to the internet server for all to see. The whole process can be automated.

but that is basically what is happening now, it is all automated.

That way you have more control over the image and it should be more stable as you won't have to run snaggit. The only downside is that you would have to write some code to generate the plot, which might be a stumbling block.

All I want/would like is a program that is more stable than Snagit ... its pretty crappy ... but that's what you get when its available free haha
I got home from work today and it had again failed about an hour before I got home

Code writing ... Yup big stumbling block ... the closest I get to any sort of code writing these days is HTML for my www pages

I really just need to figure out how to use the ftp function within the datalogger software
from what I am told it is supposed to be able to do that

regards
Dave

 

[billiards]

So the data logger software is running on the datalogger? That sends a plot to your computer which is displayed on screen and then captured using snagit ...?

At no point in this process is the raw data transferred to your external computer?

Your computer is not doing anything other than receiving and displaying the image, capturing it with snagit, and then uploading the file to an ftp site?


or ...

is the datalogger software running on your PC?

 

[davenn]

Greetings

OK ... I have a PC that is dedicated to the seismic datalogging. The datalogger software running on that PC is called WinSDR = Windows Seismic Data Recorder.
The datalogger PC is connected to the internet via my home network router
The A to D electronics and preamp boards are in boxes external to the datalogger computer.
The A to D unit ( an 8 channel, 16 bit system), is designed to run with that above mention software and is fed to the PC via RS232, tho the designer has recently come out ( after many years of RS232) with a USB version
On the datalogger PC, the Snagit screen capture prog. is also running. It captures the screen shots and ftp's them to my www site.

If the system records an earthquake, I can, at my leisure, go into the WinSDR program and do a save file of that event. I just enter a date and time ( that gives me a few minutes before the start of the event) and how long I want to save the file for. For a really big event, that could be up to a couple of hours of data until the trace tails off. for example that file up the page a bit was ~ 45 minutes ( 16hr 56m 23sec UTC to 17hr 41m 23sec), it was recorded at 100 Samples Per Second with a total of 304,200 samples ( see the info at the end of the first main line of text).

Now, WinSDR is capable of automatically saving event files, The save trigger is on data count amplitude, it will save for a nominated length of time before and after the trigger point/time. But when the system is located in a domestic home enviroment, with people walking around, heavy trucks going up and down the road 20 metres from the sensor, auto saves are not viable, there would be too many false triggers. For a quiet site, its ideal.
So hence I do manual saves of recorded events.
Those saved raw data files initially go into a folder on the datalogger PC. I will then access that folder via the network from this (my main) PC, copy them across and do analysis on them in the WinQuake analysis program.
Those files, onced filtered and all the quake info data added, Origin date, time, depth, lat, long etc etc, they will then get saved on this, my main PC. I will also upload those files to the repository in California for others to view and make use of the data.

So, the ONLY raw data that is transferred to my main PC is individual event files. The 24/7 raw daily files are only saved by the logger on the logger PC

Now, just prior to starting to type this reply, I finally figured out how to use the ftp option within WinSDR.
yipeee haha. it is visible here and as you can see , it looks much better and more like all the other loggers that don't use a screen capture program

WOW, been a successful nite, time for bed its 22:34 Hrs Australian Eastern Time

regards
Dave

 

[billiards]

Congratulations on fixing the problem, Dave!

Fingers crossed for no more crashes.

Thanks for explaining your setup a bit more.

Out of interest do you know of any studies that have made use of the public network?

 

[davenn]

Congratulations on fixing the problem, Dave!
Fingers crossed for no more crashes.

Thanks ... it was really all the discussion in this thread that got me to revisit the WinSDR logger and try to figure out how to use its capture and ftp function
So you ALL helped to inspire me

indeed and the display looks so much better :) its nice to be finally get rid of Snagit !

Thanks for explaining your setup a bit more.

no problems, it is an excellent system and Larry Cochrane (the guy who designed it all) has done a great job and is always looking at improvements as the years go by. Gosh the years roll by it has been 17 years since I first started doing digital recording with Larry's system. He came out with it in ~ 1994 or 95. I got my first system from him in 1996 ... back then it was only a 12 bit A to D. in the mid 2000's it went to a 16bit A to D and as of a few months ago he has now released his 24bit A to D hardware

Out of interest do you know of any studies that have made use of the public network?

from a professional point of view, no. The observatories don't seem to be interested in our data
I would suspect that they think our setups are not tightly enough controlled. Timing, accurate knowledge of amplifier gain settings etc ... and to a certain extent, they would be right haha

And although there is a sprinkling of professional within the Public Seismic Net. Pretty much all of us are doing it for the fun and personal learning. There are a number of members who are very competent electronic and mechanical engineers and they use those skills in pusing new boundaries in seismometer sensor designs. Like the FBV unit I have coming soon. I linked to a PDF file about it earlier in this thread.

There are a couple of other datalogging systems out there and Larry has done well to try and make his WinQuake analysis program compatable so that it will read the data files from those loggers as well.

There are a lot of systems in schools around the world and this exposes many children to the world of seismology with the study of P, S and Surface waves, travel times etc.

regards
Dave

 

[Borek]

Good to know its fixed

 

[Astronuc]

How about this one?

M6.6 - 101km W of Panguna, Papua New Guinea
2013-04-14 01:32:23 UTC
http://earthquake.usgs.gov/earthquakes/eventpage/usb000g6lc#summary

and a

M6.8 - 22km ESE of Aitape, Papua New Guinea
http://earthquake.usgs.gov/earthquakes/eventpage/usb000g8my#summary