PSN-L Email List Message

Subject: Re: Standards use can follow
From: Branden Christensen branden.christensen@...........
Date: Fri, 27 Jul 2012 04:03:34 +0000


Barry:

Angel's partner here. I generally do not weigh in on these discussions but
you are hitting on an area of seismology I am passionate about: metadata.

Fortunately, much of what you so keenly propose already exists. Take the
two biggest registries for example: IRIS amd GFZ. In recent times a few
industry standards have emerged:

Data Exchange Protocol: SeedLink
Waveform Protocol: MiniSeed
Metadata protocol: Dataless Seed

Together the miniseed and dataless create a complete Seed volume. MiniSeed
files contain waveform data and a header to identify the Station Network
Channel Location or SCNL, etc. The dataless is really amazing. It contains
the station SCNL, lat lon alt depth, description, serial numbers, sensor
and digitizer flavor, acquisition start and end dates/ times and the full
response information: poles, zeros and gain- or exactly the required
information to construct the systems transfer function and remove the
instrument response. Most real time software packages such as Earthworm,
Earlybird and SeisComP and post processing systems such as Seisan
automatically use this dataless information to remove the instrument
response. This permits the software to produce magnitudes and event
locations or stuff sensative to the real amplitudes and phase of the input
signal.

All of this seed information is passed around via SeedLink Servers in real
time.

If you are thinking of how the amateur community might contribute, then
Angel made a good point about the need for detecting local seismicity below
4. But also consider the civil defense applications of seismology.
Magnitude and locations are really an academic pursuit- pretty much
meaningless to the general public who want to know 1) did it shake and, if
so, 2) how much. So lots and lots of amateur neural networks of
accelerometers could make a difference. Imagine piping all of that data
into a real time system that created shake maps with color intensities
displaying how hard the ground shook and where. These maps benefit from
lots and lots of nodes, be they amateur or professional.

Great thread, keep it coming.

Best,

Branden Christensen
On Jul 26, 2012 8:11 PM, "Barry Lotz"  wrote:

> Hi Angel
> Thank you. The links you listed shed light on the USGS station name I have
> been monitoring. Should the network have some significance in the name?
> Does "WE" refer to something? Maybe the network registry could eventually
> contain the station's sensor specifics. Maybe a some of these station
> properties could be listed which would show organization , professionalism,
> &  completeness. Maybe some items might be:
> - type of sensor ( feedback or not? , homemade? ,  mfg new?, mfg restored
> or modified?)
> - calibrated or not
> - method of timing
> - if calibrated, is there a frequency response plot available.
> - some kind of  station noise quality. Sensor noise is a subject in it's
> self, but maybe a combined noise plot or related value could be determined
> which would the combined site/sensor noise, which is ultimately what
> affects the sensors output. Maybe something comparing the station sensor to
> the NLNM or NHNM. It probably should be similar to how the professional
> stations are evaluated.
> - availability of digital data?
> - ?
> Just some thoughts to keep the thread live.
>
> Regards
> Barry
> www.seismicvault.com
>
>
>
> ------------------------------
> *From:* "sismos@.............." 
> *To:* psnlist@..............
> *Sent:* Wed, July 25, 2012 6:03:49 AM
> *Subject:* Standards use can follow
>
> HI Barry,
>
> I might have to take this in parts.  I think that there are some very easy
> steps we can take that cost nothing.  We can follow the channel naming
> convention.
>
> All of the modern softwares and mainly earthworm and seiscomp refer to
> channels as SCNL
>
> Station
> Channel
> Network
> Location
>
> STATION  name can be up to 5 characters, letters and numbers, no special
> characters and can not start with a number.  My main station for the last
> 12 years is BRU2.  If the station name are to be shared the need to be
> unique.  There is an agency that keeps a big station book.  We could
> probably come up with a scheme that was unique to us, for example all or
> our station names would end in 99  so maybe a new site for me would be
> ALR99.
>
> CHANNEL name is a bit harder but but more or less defined here.
>
> http://www.iris.washington.edu/manuals/SEED_appA.htm
>
> NETWORK code we would have to apply for one, if we haven't, I could do it,
> it would be easy, They are assigned by the FDSN, Federation of Digital
> Seismic Network.  We might ask for one, maybe "WE", as far as I can tell
> numbers are allow in this field, so maybe we could get a pair of numbers.
>
> http://www.fdsn.org/station_book/
>
> LOCATION is just a two character field, numbers and letter. Normally 00
> for the main station.  The usefulness is that if we have many instruments
> co-located we can tell them apart.  Dave has many vertical at the same
> place so they would all have the same station name, the same channel name
> and the same network code and we could tell them apart by the location
> code. So...
>
> My main station has SCNLs as follow
>
> BRU2 HHZ PA 00
> BRU2 HHN PA 00
> BRU2 HHE PA 00
>
> Dave's might be
>
> DFN99 HHZ WE 00
> DFN99 HHZ WE 10
> DFN99 HHZ WE 20
>
> This would be the SCNL for three co-located vertical high gain broadbands
> from the station DFN99 sampled at more than 80 samples per second in the
> network WE.
>
> Other easy things we can do is to have our station coordinates
> Then the time thing, GPS is best but WWV if fine and cheap and there is
> NTP network time which I also think is ok as long as we can regulate the
> computer time to less than 100 milliseconds for regional locations. Small
> tight networks need better time.
>
> Many latin american country networks have nothing more, They only pick P
> and S report coda magnitudes.  You do not need to calibrate your
> instruments to do that.  To calculate other magnitudes we would need to
> calibrate our instruments.
>
> Most of use have Larry's digitizer which easily integrates into earthworm
> and once we had our data in earthworm sharing our data in real time would
> be easy.  We would need an earthworm ID but that would be easy and we could
> all share the same ID.
>
> More latter, questions and comments welcome.  Who wants to be our
> registrar??
>
> Saludos,
>
> Angel
>
>
>
>
> On 07/17/2012 02:19 AM, Barry Lotz wrote:
> > Angel & Dave
> >  I agree. How do we move forward?
> > My 2 cents:  I work for a structural testing and inspection lab. Our lab
> and inspectors comply with certain standards so that our
> results/inspections are credible ( eg NIST, ASTM, AWS, ICC). Maybe it would
> be possible to have an accepted protocol for our stations with sensor(s). I
> would like to comment on  the items you mentioned.
> > >"We have noisy instruments"
> >    It seems possible to determine the instrument noise by maybe
> "nesting"? Site noise could be evaluated over "X" time during day,night or
> both. Could it mirror the evaluation of professional systems? It seems a
> threshold could be determined for credibility station/sensor noise limits.
> Do we use the NLNM graph with an envelope of limits?
> > >"We do not calibrate"
> >    Could a accepted standardized procedure be described for horizontal
> and vertical sensors, that all could use? I am familiar with methods that
> Dave and Brett use for our FBV's.
> > >"We do not have accurate time"
> >    Would , as an example, larry's SDR program and ADC unit with GPS time
> be sufficient? What would be an accepted variational allowance?
> > >"We do not use a standard format for data exchange"
> >      Could a documented conversion program(s) be used to convert from
> say psn to an "standard"  format? Maybe it already exists.
> > > "We do not use standard naming conventions"
> >      Should we have a described procedure for this?
> > I think Dr Weilandt has programs which could address some of the noise
> and calibration issues above. I haven't completely read it but is the NMSOP
> what professionals use? Could we have something similar with more nuts and
> bolts procedures and info?
> >      General agreement maybe the biggest hurdle, but I agree we should
> make an effort to have more credible stations and sensors if we desire.
> Maybe we just needs some agreed upon documented standards to try to achieve.
> >
> > Regards
> > Barry
> >
> >
> >
> >
> > ------------------------------------------------------------------------
> > *From:* "sismos@.............." 
> > *To:* psnlist@..............
> > *Sent:* Sat, July 14, 2012 12:48:51 PM
> > *Subject:* Re: diamagnetic levitation seismometer possibility
> >
> > On 07/14/2012 12:38 AM, Thomas Dick wrote:
> > > On 7/13/2012 5:55 PM, Dave Nelson wrote:
> > >> The number of possible seismic instrument configurations which will
> provide some response  to seismic motions is vast. The question is the
> practicality/utility of a given configuration.
> > >> The key figure of merit for any instrument configuration  is the
> instrument self noise and response as a function of frequency. This
> directly determines the minimum seismic motion the instrument is capable of
> detecting and then providing useful data for analysis.
> > >> If one is willing to wait for that rare magnitude 7 or 8 event the
> simplest / noisiest instrument may do the job in some contexts, such as
> classroom demonstrations .
> > >> The amateur astronomer community has evolved to the point where it
> provides useful  ( if not essential) information to the astronomy
> scientific community. I believe the amateur seismology community could do a
> similar service but not with inadequate instrumentation.
> > >>  The goal should be to develop amateur instruments with
> characteristics near the performance of professional instruments and then
> operating  them in reasonably low noise sites. (An instrument in a
> residential basement  will work reassembly well if carefully done.)
> > >> Larry Cochrane has  already provided us with excellent equipment to
> handle the sensor data and connect it to a network. Some work needs to be
> done in this area but we have a good start.
> > >> _*My challenge is to include instrument self noise and generator
> constant, both as a function of frequency, as a FIRST PRIORITY when
> evaluating  the utility of an instrument concept. *_
> > >> Just another gadget that will respond if you shake it is not where we
> want to spend our efforts and resources.
> > >> I do  NOT  mean to imply there are not some truly innovative and
> possibly revolutionary ideas out there but we should  look at each of them
> carefully to determine early whether they justify significant effort or
> belong in the "that was interesting" stack.
> > >> Just where determination is made is a personal choice but it should
> be based on some form of analysis and/or test.
> > >> Comments Please.
> > >> Dave Nelson
> > >> Rolling Hills Estates, California
> > > My impression is that most academia and professional seismologists
> hold the amateur in very low esteem.
> >
> > Yes, they do hold us in low esteem and this is our own fault.
> >
> > We have noisy instruments
> > We do not calibrate
> > We do not have accurate time
> > We do not use a standard format for data exchange
> > We do not use standard naming conventions
> >
> > The academic and professional seismologists can already locate and
> characterize (within a few minutes) all events over about 4.2 Mb, They
> don't need us for that.  Where we could excel and make a meaningful
> contribution is in the seismicity of our own backyards, the small events
> less than one degree from our instruments. Recording those is a bit harder
> than picking up the squiggles from a 6.5 Mb 10 degrees away.
> >
> > These are just a few things we do and do not do and until we do we will
> just be amateurs.
> >
> > Just my two cents
> >
> > Angel
> >
> >
> >
> >
> > __________________________________________________________
> >
> > Public Seismic Network Mailing List (PSNLIST)
> >
> > To leave this list email PSNLIST-REQUEST@..............  PSNLIST-REQUEST@..............> with the body of the message (first line
> only): unsubscribe
> > See http://www.seismicnet.com/maillist.html for more information.
>
>
> __________________________________________________________
>
> Public Seismic Network Mailing List (PSNLIST)
>
> To leave this list email PSNLIST-REQUEST@.............. with the body of
> the message (first line only): unsubscribe
> See http://www.seismicnet.com/maillist.html for more information.
>

Barry:

Angel's partner here. I generally do not weigh in on these discussio= ns but you are hitting on an area of seismology I am passionate about: meta= data.

Fortunately, much of what you so keenly propose already exists. Take the= two biggest registries for example: IRIS amd GFZ. In recent times a few in= dustry standards have emerged:

Data Exchange Protocol: SeedLink
Waveform Protocol: MiniSeed
Metadata protocol: Dataless Seed

Together the miniseed and dataless create a complete Seed volume. MiniSe= ed files contain waveform data and a header to identify the Station Network= Channel Location or SCNL, etc. The dataless is really amazing. It contains= the station SCNL, lat lon alt depth, description, serial numbers, sensor a= nd digitizer flavor, acquisition start and end dates/ times and the full re= sponse information: poles, zeros and gain- or exactly the required informat= ion to construct the systems transfer function and remove the instrument re= sponse. Most real time software packages such as Earthworm, Earlybird and S= eisComP and post processing systems such as Seisan automatically use this d= ataless information to remove the instrument response. This permits the sof= tware to produce magnitudes and event locations or stuff sensative to the r= eal amplitudes and phase of the input signal.

All of this seed information is passed around via SeedLink Servers in re= al time.

If you are thinking of how the amateur community might contribute, then = Angel made a good point about the need for detecting local seismicity below= 4. But also consider the civil defense applications of seismology. Magnitu= de and locations are really an academic pursuit- pretty much meaningless to= the general public who want to know 1) did it shake and, if so, 2) how muc= h. So lots and lots of amateur neural networks of accelerometers could make= a difference. Imagine piping all of that data into a real time system that= created shake maps with color intensities displaying how hard the ground s= hook and where. These maps benefit from lots and lots of nodes, be they ama= teur or professional.

Great thread, keep it coming.

Best,

Branden Christensen

On Jul 26, 2012 8:11 PM, "Barry Lotz" = <barry_lotz@.............> wrote:
Hi Angel
Thank you. The links you listed shed light on the USGS= station name I have been monitoring. Should the network have some signific= ance in the name? Does "WE" refer to something? Maybe the network= registry could eventually contain the station's sensor specifics. Mayb= e a some of these station properties could be listed which would show organ= ization , professionalism, &=A0 completeness. Maybe some items might be= :
- type of sensor ( feedback or not? , homemade? ,=A0 mfg new?, mfg restored= =A0 or modified?)
- calibrated or not
- method of timing
- if= calibrated, is there a frequency response plot available.
- some kind = of=A0 station noise quality. Sensor noise is a subject in it's self, bu= t maybe a combined noise plot or related value could be determined which would the combined site/sensor noise, which is ultimately what affec= ts the sensors output. Maybe something comparing the station sensor to the = NLNM or NHNM. It probably should be similar to how the professional=A0 stat= ions are evaluated.
- availability of digital data?
- ?
Just some thoughts to keep the t= hread live.

Regards
Barry
www.seismicvault.com




From: "sismos@.............." <sismos@..............<= /a>>
To:
psnlist@..............
Sent: Wed, July 25, 2012 6:03:49 AMSubject: Standards use can= follow

HI Barry,

I might have to take this in parts.=A0 I think that there = are some very easy steps we can take that cost nothing.=A0 We can follow th= e channel naming convention.

All of the modern softwares and mainly = earthworm and seiscomp refer to channels as SCNL

Station
Channel
Network
Location

STATION=A0 name can be= up to 5 characters, letters and numbers, no special characters and can not= start with a number.=A0 My main station for the last 12 years is BRU2.=A0 = If the station name are to be shared the need to be unique.=A0 There is an = agency that keeps a big station book.=A0 We could probably come up with a s= cheme that was unique to us, for example all or our station names would end= in 99=A0 so maybe a new site for me would be ALR99.

CHANNEL name is a bit harder but but more or less defined here.

= http://www.iris.washington.edu/manuals/SEED_appA.htm

NETWORK code we would have to apply for one, if we haven't, I could= do it, it would be easy, They are assigned by the FDSN, Federation of Digi= tal Seismic Network.=A0 We might ask for one, maybe "WE", as far = as I can tell numbers are allow in this field, so maybe we could get a pair= of numbers.

h= ttp://www.fdsn.org/station_book/

LOCATION is just a two c= haracter field, numbers and letter. Normally 00 for the main station.=A0 Th= e usefulness is that if we have many instruments co-located we can tell the= m apart.=A0 Dave has many vertical at the same place so they would all have= the same station name, the same channel name and the same network code and= we could tell them apart by the location code. So...

My main station has SCNLs as follow

BRU2 HHZ PA 00
BRU2 HHN PA 00
BRU2 HHE PA 00<= br>
Dave's might be

DFN99 HHZ WE 00
DFN99 HHZ WE 10
DFN= 99 HHZ WE 20

This would be the SCNL for three co-located vertical hi= gh gain broadbands from the station DFN99 sampled at more than 80 samples p= er second in the network WE.

Other easy things we can do is to have our station coordinates
Then = the time thing, GPS is best but WWV if fine and cheap and there is NTP netw= ork time which I also think is ok as long as we can regulate the computer t= ime to less than 100 milliseconds for regional locations. Small tight netwo= rks need better time.

Many latin american country networks have nothing more, They only pick = P and S report coda magnitudes.=A0 You do not need to calibrate your instru= ments to do that.=A0 To calculate other magnitudes we would need to calibra= te our instruments.

Most of use have Larry's digitizer which easily integrates into earthworm and once we had our data in earthworm sharing ou= r data in real time would be easy.=A0 We would need an earthworm ID but tha= t would be easy and we could all share the same ID.

More latter, que= stions and comments welcome.=A0 Who wants to be our registrar??

Saludos,

Angel




On 07/17/2012 02:19 AM, Barry = Lotz wrote:
> Angel & Dave
>=A0 I agree. How do we move for= ward?
> My 2 cents:=A0 I work for a structural testing and inspection= lab. Our lab and inspectors comply with certain standards so that our resu= lts/inspections are credible ( eg NIST, ASTM, AWS, ICC). Maybe it would be = possible to have an accepted protocol for our stations with sensor(s). I wo= uld like to comment on=A0 the items you mentioned.
> >"We have noisy instruments"
>=A0 =A0 It seems poss= ible to determine the instrument noise by maybe "nesting"? Site n= oise could be evaluated over "X" time during day,night or both. Could it mirro= r the evaluation of professional systems? It seems a=A0 threshold could be = determined for credibility station/sensor noise limits. Do we use the NLNM = graph with an envelope of limits?
> >"We do not calibrate"
>=A0 =A0 Could a accepted st= andardized procedure be described for horizontal and vertical sensors, that= all could use? I am familiar with methods that Dave and Brett use for our = FBV's.
> >"We do not have accurate time"
>=A0 =A0 Would , a= s an example, larry's SDR program and ADC unit with GPS time be suffici= ent? What would be an accepted variational allowance?
> >"We = do not use a standard format for data exchange"
>=A0 =A0 =A0 Could a documented conversion program(s) be used to convert= from say psn to an "standard"=A0 format? Maybe it already exists= ..
> > "We do not use standard naming conventions"
>=A0 =A0 =A0 Should we have a described procedure = for this?
> I think Dr Weilandt has programs which could address some= of the noise and calibration issues above. I haven't completely read i= t but is the NMSOP what professionals use? Could we have something similar = with more nuts and bolts procedures and info?
>=A0 =A0 =A0 General agreement maybe the biggest hurdle, but I agree we= should make an effort to have more credible stations and sensors if we des= ire. Maybe we just needs some agreed upon documented standards to try to ac= hieve.
>
> Regards
> Barry
>
>
>
>
= > ----------------------------------------------------------------------= --
> *From:* "sismos@.............." <sismos@..............>
> *To:* psnl= ist@..............
> *Sent:* Sat, July 14, 2012 12:48:51 PM
&g= t; *Subject:* Re: diamagnetic levitation seismometer possibility
> > On 07/14/2012 12:38 AM, Thomas Dick wrote:
> > On 7/13/2012 5= :55 PM, Dave Nelson wrote:
> >> The number of possible seismic = instrument configurations which will provide some response=A0 to seismic mo= tions is vast. The question is the practicality/utility of a given configur= ation.
> >> The key figure of merit for any instrument configuration=A0 i= s the instrument self noise and response as a function of frequency. This d= irectly determines the minimum seismic motion the instrument is capable of = detecting and then providing useful data for analysis.
> >> If one is willing to wait for that rare magnitude 7 or 8 event the simplest / noisiest instrument may do the = job in some contexts, such as classroom demonstrations .
> >> T= he amateur astronomer community has evolved to the point where it provides = useful=A0 ( if not essential) information to the astronomy scientific commu= nity. I believe the amateur seismology community could do a similar service= but not with inadequate instrumentation.
> >>=A0 The goal should be to develop amateur instruments with cha= racteristics near the performance of professional instruments and then oper= ating=A0 them in reasonably low noise sites. (An instrument in a residentia= l basement=A0 will work reassembly well if carefully done.)
> >> Larry Cochrane has=A0 already provided us with excellent equi= pment to handle the sensor data and connect it to a network. Some work need= s to be done in this area but we have a good start.
> >> _*My c= hallenge is to include instrument self noise and generator constant, both as a function o= f frequency, as a FIRST PRIORITY when evaluating=A0 the utility of an instr= ument concept. *_
> >> Just another gadget that will respond if= you shake it is not where we want to spend our efforts and resources.
> >> I do=A0 NOT=A0 mean to imply there are not some truly innovat= ive and possibly revolutionary ideas out there but we should=A0 look at eac= h of them=A0 carefully to determine early whether they justify significant = effort or belong in the "that was interesting" stack.
> >> Just where determination is made is a personal choice but it = should be based on some form of analysis and/or test.
> >> Comm= ents Please.
> >> Dave Nelson
> >> Rolling Hills Es= tates, California
> > My impression is that most academia and professional seismologist= s hold the amateur in very low esteem.
>
> Yes, they do hold us in low esteem and this is our own fault.
= >
> We have noisy instruments
> We do not calibrate
>= We do not have accurate time
> We do not use a standard format for d= ata exchange
> We do not use standard naming conventions
>
> The academi= c and professional seismologists can already locate and characterize (withi= n a few minutes) all events over about 4.2 Mb, They don't need us for t= hat.=A0 Where we could excel and make a meaningful contribution is in the s= eismicity of our own backyards, the small events less than one degree from = our instruments. Recording those is a bit harder than picking up the squigg= les from a 6.5 Mb 10 degrees away.
>
> These are just a few things we do and do not do and until we = do we will just be amateurs.
>
> Just my two cents
> > Angel
>
>
>
>
> __________________________________________________________
>
>= ; Public Seismic Network Mailing List (PSNLIST)
>
> To leave t= his list email PSNLIST-REQUEST@.............. <mailto:PSNLIST-REQUEST@............. OM> with the body of the message (first line only): unsubscribe
> See http://www.seismicnet.com/maillist.html for more information.

__________________________________________________________

Pub= lic Seismic Network Mailing List (PSNLIST)

To leave this list email PSNLIST-REQUEST@.............. with the body of t= he message (first line only): unsubscribe
See http://www.seismicnet.com/maillist.html for = more information.

[ Top ] [ Back ] [ Home Page ]