From songyun at ualberta.ca Fri Jan 10 22:53:44 2020 From: songyun at ualberta.ca (Songyun Huang) Date: Fri, 10 Jan 2020 14:53:44 -0700 Subject: [rf] Srf calculation Message-ID: Hello sir, I am Songyun from the University of Alberta, and now I work on S receiver function as a graduate student. So sorry to bother you but I have some questions about the deconvolution process in rf package. After I running the code and getting Srf in a stream, the component I should focus on is L component of Srf based on my understanding, and in the documentation, there is no L component mentioned, so I just feel confused whether I pay attention to the wrong component of Srf? The next question is whether I need to reverse the amplitude polarity of Srf? I know the coefficients of Sp and Ps conversions have different signs so I think I should reverse Srf polarity to make it comparable with Prf. In the documentation, it mentions that Q and T components are multiplied by -1 to get a positive phase for positive velocity contrast. Does it mean I still need to reverse Srf sign again by myself? And all L components are not reversed? Thank you for your time. Have a nice day. -- Sincerely, Songyun Huang -------------- next part -------------- An HTML attachment was scrubbed... URL: From tom.eulenfeld at uni-jena.de Sat Jan 11 00:16:44 2020 From: tom.eulenfeld at uni-jena.de (Tom Eulenfeld) Date: Sat, 11 Jan 2020 00:16:44 +0100 Subject: [rf] Srf calculation In-Reply-To: Message-ID: <20200111001644.Horde.xFfSb7Ccccki6fwixpgTQ9j@webmail.uni-jena.de> Hi Songyun, welcome to the mailing list! > After I running the code and getting Srf in a stream, the component I > should focus on is L component of Srf based on my understanding, and in the > documentation, there is no L component mentioned, so I just feel confused > whether I pay attention to the wrong component of Srf? Yes, you are correct, you should pay attention to the L component in the Srf. The Q component should resemble more or less a peak. In the docs only Prfs are described. > The next question is whether I need to reverse the amplitude polarity of > Srf? I know the coefficients of Sp and Ps conversions have different signs > so I think I should reverse Srf polarity to make it comparable with Prf. In > the documentation, it mentions that Q and T components are multiplied by -1 > to get a positive phase for positive velocity contrast. Does it mean I > still need to reverse Srf sign again by myself? And all L components are > not reversed? Thanks for raising this question. Sorry, I fear the documentation is not up to date after I dealt with issue #4 (https://github.com/trichter/rf/issues/4). At the moment only the Q component is polarity reversed *before* the deconvolution for Prf as well as Srf. (This is not the case for the R component when using ZRT rotation.) I think with this procedure the polarity should be correct for Q component in Prf and L component in Srf. This could be easily checked with synthetic data. Best regards! Tom Zitat von Songyun Huang : > Hello sir, > > I am Songyun from the University of Alberta, and now I work on S receiver > function as a graduate student. So sorry to bother you but I have some > questions about the deconvolution process in rf package. > > After I running the code and getting Srf in a stream, the component I > should focus on is L component of Srf based on my understanding, and in the > documentation, there is no L component mentioned, so I just feel confused > whether I pay attention to the wrong component of Srf? > > The next question is whether I need to reverse the amplitude polarity of > Srf? I know the coefficients of Sp and Ps conversions have different signs > so I think I should reverse Srf polarity to make it comparable with Prf. In > the documentation, it mentions that Q and T components are multiplied by -1 > to get a positive phase for positive velocity contrast. Does it mean I > still need to reverse Srf sign again by myself? And all L components are > not reversed? > > Thank you for your time. Have a nice day. > > -- > Sincerely, > Songyun Huang From crmpeter at gmail.com Wed Jan 15 03:51:25 2020 From: crmpeter at gmail.com (Petr Predein) Date: Wed, 15 Jan 2020 10:51:25 +0800 Subject: Qopen for local earthquakes (less then 70 km) Message-ID: Hello, I am new to this mailing list, and very interesting in using Qopen package for calculating intrinsic and scattering attenuation. And my question to respected community is, can I use earthquakes with epicenter very close to some station, with magnitudes no more then 3-3.5, for example. Normally, I read in (Sens-Schönfelder C., Wegler U., 2006: Radiative transfer theory for estimation of the seismic moment, p. 3) that hypocentral distance should be greater than 60 km, because it is inappropriate to use mean velocity of S-waves about 3.5 km/s for shorter distances (upper layers of the Earth crust?) But it is also shown that Qopen uses improved method, unlike original (Sens-Schönfelder C., Wegler U., 2006). Also, I understand that it's incorrect one cannot use earthquakes with epi distances less then several km up to 10 km, depending on frequency range. Finally, can I use Qopen for small earthquakes with epicentral distance from 10 to 70 km? (average depth is in range 15-20 km, hovewer). And, if yes, which setting should I use in conf.json file to avoid big errors, with stations on free surface. Thanks in advance! -------------- next part -------------- An HTML attachment was scrubbed... URL: From tom.eulenfeld at uni-jena.de Wed Jan 15 10:10:22 2020 From: tom.eulenfeld at uni-jena.de (Tom Eulenfeld) Date: Wed, 15 Jan 2020 10:10:22 +0100 Subject: [Qopen] for local earthquakes (less then 70 km) In-Reply-To: References: Message-ID: Hello Petr, please use "[Qopen]" keyword in the subject when posting a message to the mailing list. Then, it its categorized correctly and automatically accepted. The choice of Sens-Schönfelder and Wegler (2006) is comprehensive in their setting. Still, it is IMHO absolutely no problem to use shorter distances. I have done it myself. Your dataset appears suitable. Please take the following in mind: 1. v0 can be set to another velocity, it should be a "mean" value for the used model, also time windows need to be adapted. 2. The resulting Q-values for scattering and intrinsic attenuation will be "mean" values over the sampled volume. This means Q-values will depend on the scale of the dataset and used time windows. 3. For local datasets with small earthquakes, results for low frequencies will not be useful. Setting the time windows in your config is a bit of trial and error. The direct S wave window should cover the the complete S wave train, and the coda window best starts at the end of the S wave window. For your convenience, I append the config files I used for the USArray (Eulenfeld and Wegler, 2017) and the config files for the different regions investigated in Eulenfeld and Wegler (2016). Even for a successful inversion, have a look at the log file and watch out for errors. I hope you will find the package and these hints useful. Best regards! Tom On 15.01.20 03:51, Petr Predein wrote: > Hello, I am new to this mailing list, and very interesting in using > Qopen package for calculating intrinsic and scattering attenuation. > And my question to respected community is, can I use earthquakes with > epicenter very close to some station, with magnitudes no more then > 3-3.5, for example. > Normally, I read in (Sens-Schönfelder C., Wegler U., 2006: Radiative > transfer theory for estimation of the seismic moment, p. 3) > that hypocentral distance should be greater than 60 km, because it is > inappropriate to use mean velocity of S-waves about 3.5 km/s for shorter > distances (upper layers of the Earth crust?) > But it is also shown that Qopen uses improved method, unlike original > (Sens-Schönfelder C., Wegler U., 2006). > Also, I understand that it's incorrect one cannot use earthquakes with > epi distances less then several km up to 10 km, depending on frequency > range. > Finally, can I use Qopen for small earthquakes with epicentral distance > from 10 to 70 km? (average depth is in range 15-20 km, hovewer). > And, if yes, which setting should I use in conf.json file to avoid big > errors, with stations on free surface. > Thanks in advance! > -------------- next part -------------- A non-text attachment was scrubbed... Name: conf_us.json Type: application/json Size: 2361 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: conf_insheim.json Type: application/json Size: 2755 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: conf_landau.json Type: application/json Size: 2695 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: conf_molasse.json Type: application/json Size: 2480 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: conf_vogtland.json Type: application/json Size: 2393 bytes Desc: not available URL: From crmpeter at gmail.com Wed Jan 15 15:48:49 2020 From: crmpeter at gmail.com (Petr Predein) Date: Wed, 15 Jan 2020 22:48:49 +0800 Subject: [Qopen] for local earthquakes (less then 70 km) In-Reply-To: References:

Message-ID: Thank you so much for detailed answer and very useful instruction with examples. I really appreciate it! I also find Qopen package works for me and hope the result will not put me to shame. Thanks! ср, 15 янв. 2020 г., 17:10 Tom Eulenfeld : > Hello Petr, > > please use "[Qopen]" keyword in the subject when posting a message to > the mailing list. Then, it its categorized correctly and automatically > accepted. > > The choice of Sens-Schönfelder and Wegler (2006) is comprehensive in > their setting. Still, it is IMHO absolutely no problem to use shorter > distances. I have done it myself. Your dataset appears suitable. Please > take the following in mind: > > 1. v0 can be set to another velocity, it should be a "mean" value for > the used model, also time windows need to be adapted. > 2. The resulting Q-values for scattering and intrinsic attenuation will > be "mean" values over the sampled volume. This means Q-values will > depend on the scale of the dataset and used time windows. > 3. For local datasets with small earthquakes, results for low > frequencies will not be useful. > > Setting the time windows in your config is a bit of trial and error. The > direct S wave window should cover the the complete S wave train, and the > coda window best starts at the end of the S wave window. > > For your convenience, I append the config files I used for the USArray > (Eulenfeld and Wegler, 2017) and the config files for the different > regions investigated in Eulenfeld and Wegler (2016). > > Even for a successful inversion, have a look at the log file and watch > out for errors. > > I hope you will find the package and these hints useful. > Best regards! > > Tom > > > > > On 15.01.20 03:51, Petr Predein wrote: > > Hello, I am new to this mailing list, and very interesting in using > > Qopen package for calculating intrinsic and scattering attenuation. > > And my question to respected community is, can I use earthquakes with > > epicenter very close to some station, with magnitudes no more then > > 3-3.5, for example. > > Normally, I read in (Sens-Schönfelder C., Wegler U., 2006: Radiative > > transfer theory for estimation of the seismic moment, p. 3) > > that hypocentral distance should be greater than 60 km, because it is > > inappropriate to use mean velocity of S-waves about 3.5 km/s for shorter > > distances (upper layers of the Earth crust?) > > But it is also shown that Qopen uses improved method, unlike original > > (Sens-Schönfelder C., Wegler U., 2006). > > Also, I understand that it's incorrect one cannot use earthquakes with > > epi distances less then several km up to 10 km, depending on frequency > > range. > > Finally, can I use Qopen for small earthquakes with epicentral distance > > from 10 to 70 km? (average depth is in range 15-20 km, hovewer). > > And, if yes, which setting should I use in conf.json file to avoid big > > errors, with stations on free surface. > > Thanks in advance! > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From crmpeter at gmail.com Thu Jan 16 07:24:06 2020 From: crmpeter at gmail.com (Petr Predein) Date: Thu, 16 Jan 2020 14:24:06 +0800 Subject: [Qopen] for local earthquakes (less then 70 km) In-Reply-To: References:

Message-ID: Sorry to bother you again, but now I'm running Qopen with different settings. And it seems that such parameters as velocities, mean density or use_picks parameter (whether I have P- and S-onset or it’s calculated), will not change the result dramatically. But, I noticed, setting of coda_window length, is another matter. With setting by default, "coda_window": ["S+20s", ["S+150s", "3SNR"]] that means window will start 20 second after time of S-wave, with length = 130s (or while more then 3*SNR)? and will start just after "bulk_window" = ["S-2s", "S+20s"] But if I want coda_window to be no more then 40 second, for example, with "bulk_window" = ["S-0.2s", "S+5s"], if I set coda_window: ["S+5s", ["S+45s", "3SNR"]] With this setting I get g0 values (and finally Q scattering) with absolutely different order, then with first one. Is it correct, config format and results? It's hard for me to conclude. And to see time where directly coda will start and end, I searched output files, and didn’t succeed. I attach config file with my settings, just in case. ср, 15 янв. 2020 г. в 22:48, Petr Predein : > Thank you so much for detailed answer and very useful instruction with > examples. I really appreciate it! I also find Qopen package works for me > and hope the result will not put me to shame. Thanks! > > ср, 15 янв. 2020 г., 17:10 Tom Eulenfeld : > >> Hello Petr, >> >> please use "[Qopen]" keyword in the subject when posting a message to >> the mailing list. Then, it its categorized correctly and automatically >> accepted. >> >> The choice of Sens-Schönfelder and Wegler (2006) is comprehensive in >> their setting. Still, it is IMHO absolutely no problem to use shorter >> distances. I have done it myself. Your dataset appears suitable. Please >> take the following in mind: >> >> 1. v0 can be set to another velocity, it should be a "mean" value for >> the used model, also time windows need to be adapted. >> 2. The resulting Q-values for scattering and intrinsic attenuation will >> be "mean" values over the sampled volume. This means Q-values will >> depend on the scale of the dataset and used time windows. >> 3. For local datasets with small earthquakes, results for low >> frequencies will not be useful. >> >> Setting the time windows in your config is a bit of trial and error. The >> direct S wave window should cover the the complete S wave train, and the >> coda window best starts at the end of the S wave window. >> >> For your convenience, I append the config files I used for the USArray >> (Eulenfeld and Wegler, 2017) and the config files for the different >> regions investigated in Eulenfeld and Wegler (2016). >> >> Even for a successful inversion, have a look at the log file and watch >> out for errors. >> >> I hope you will find the package and these hints useful. >> Best regards! >> >> Tom >> >> >> >> >> On 15.01.20 03:51, Petr Predein wrote: >> > Hello, I am new to this mailing list, and very interesting in using >> > Qopen package for calculating intrinsic and scattering attenuation. >> > And my question to respected community is, can I use earthquakes with >> > epicenter very close to some station, with magnitudes no more then >> > 3-3.5, for example. >> > Normally, I read in (Sens-Schönfelder C., Wegler U., 2006: Radiative >> > transfer theory for estimation of the seismic moment, p. 3) >> > that hypocentral distance should be greater than 60 km, because it is >> > inappropriate to use mean velocity of S-waves about 3.5 km/s for >> shorter >> > distances (upper layers of the Earth crust?) >> > But it is also shown that Qopen uses improved method, unlike original >> > (Sens-Schönfelder C., Wegler U., 2006). >> > Also, I understand that it's incorrect one cannot use earthquakes with >> > epi distances less then several km up to 10 km, depending on frequency >> > range. >> > Finally, can I use Qopen for small earthquakes with epicentral distance >> > from 10 to 70 km? (average depth is in range 15-20 km, hovewer). >> > And, if yes, which setting should I use in conf.json file to avoid big >> > errors, with stations on free surface. >> > Thanks in advance! >> > >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: conf_BR.json Type: application/json Size: 4196 bytes Desc: not available URL: From crmpeter at gmail.com Tue Jun 23 09:05:53 2020 From: crmpeter at gmail.com (Petr Predein) Date: Tue, 23 Jun 2020 15:05:53 +0800 Subject: [Qopen] Using Qopen for coda Q determination or coda normalization Message-ID: Hello Tom! I’m very grateful to you for your response and for further developing Qopen, it’s very helpful. And I noticed that you updated Qopen package, as I can see there are some changes in the imaging process, and also coda Q determination and coda normalization added, isn’t it? And I would like to test this features. So, I tried to update Qopen package, and for coda normalization I see, the global change is that now total spectral energy density is normalized by mean value of coda-window 20 length (from 180 to 200 seconds in your example, or am I misunderstood?). But I get almost the same values for `g0` and `b` parameter but site amplifications differs, compared to results with original options, for the Tutorial dataset So, could you please explain to me how to use Qopen to compare with coda normalization. Also, unfortunately coda Q determination (these 4 options) qopen fails with error: AttributeError: module 'qopen.rt' has no attribute 'G_diffapprox3d' ps. I must say, also it’s my first experience to make CodaNorm package using a similar method. That script was a bit trivial and only calculates RMS amplitudes in fixed windows for bulk wave and for coda-wave (the further calculations was not my task, however, but they showed a large scatter of results depending on the geometric spreading value, for example). Thanks in advance, cheers! Без вирусов. www.avast.ru <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2> -------------- next part -------------- An HTML attachment was scrubbed... URL: From tom.eulenfeld at uni-jena.de Wed Jun 24 10:52:13 2020 From: tom.eulenfeld at uni-jena.de (Tom Eulenfeld) Date: Wed, 24 Jun 2020 10:52:13 +0200 Subject: [Qopen] Using Qopen for coda Q determination or coda normalization In-Reply-To: References: Message-ID: <4b3ced49-5c7a-3f90-f809-1f3d422da087@uni-jena.de> Hello Petr! you are welcome. On 23.06.20 09:05, Petr Predein wrote: > Hello Tom! > I’m very grateful to you for your response and for further developing > Qopen, it’s very helpful. > And I noticed that you updated Qopen package, as I can see there are > some changes in the imaging process, and also coda Q determination and > coda normalization added, isn’t it? > And I would like to test this features. So, I tried to update Qopen > package, and for coda normalization I see, the global change is that now > total spectral energy density is normalized by mean value of coda-window > 20 length (from 180 to 200 seconds in your example, or am I misunderstood > But I get almost the same values for `g0` and `b` parameter but site > amplifications differs, compared to results with original options, for > the Tutorial dataset > So, could you please explain to me how to use Qopen to compare with coda > normalization. Yes, coda normalization is a new option that is turned off by default. It was added to allow a better comparison to MLTWA (multiple lapse time window analysis) which employs coda normalization. That you get the same values for g0 and b is actually a good sign. If it produced a different result you would need to have a closer look. As a side effect, site amplifications and source spectra are useless when using coda normalization. The only advantage of coda normalization is that the number of unknowns in the optimization is reduced, but if the original method gives similar results I would prefer this one. > Also, unfortunately coda Q determination (these 4 options) qopen fails > with error: > AttributeError: module 'qopen.rt' has no attribute 'G_diffapprox3d' These new options are only available in the development version. How did you install? I suggest to install qopen-dev with pip install git+https://github.com/trichter/qopen.git Please run "qopen-runtests" or all tests with "qopen-runtests -a". Does the error persist? Note that this coda Q in Qopen is also only for comparison purposes. The result is an average coda Q. Often coda Q is calculated separately for each station and event. This is not possible with Qopen. > ps. I must say, also it’s my first experience to make CodaNorm package > using a similar method. That script was a bit trivial and only > calculates RMS amplitudes in fixed windows for bulk wave and for > coda-wave (the further calculations was not my task, however, but they > showed a large scatter of results depending on the geometric spreading > value, for example) > Thanks in advance, cheers! I have seen this code on github, but never used it. But it makes sense that the result depends on the geometric spreading factor. You have to make some assumptions (single scattering vs diffusion) for calculation and interpretation of coda Q. The "G_diffapprox3d" from Qopen uses the diffusion approximation. Other Green's functions are possible and have its impacts on the results. Cheers! Tom From tom.eulenfeld at uni-jena.de Thu Jul 23 11:53:06 2020 From: tom.eulenfeld at uni-jena.de (Tom Eulenfeld) Date: Thu, 23 Jul 2020 11:53:06 +0200 Subject: [rf] [yam] [qopen] Discuss topics at ObsPy discourse forum Message-ID: Dear all, ObsPy opened a new self-hosted discourse forum. I would like to move the sporadic discussions from this mailing list there. I expect the forum under the ObsPy umbrella to be future-proven and easy-discoverable. Please post new topics to the category "ObsPy Related Projects" at https://discourse.obspy.org/c/obspy-related-projects If you want to search for earlier discussions, the archive of this mailing list is at https://lserv.uni-jena.de/pipermail/seistools/ Best regards, Tom