PEER is looking for feedback on the newly released BETA version of the PEER Ground Motion Database. Any comments related to the enhancement of the database features and the user interface are requested so that this information can be used to improve the website before the final release. Please enter your comments below, and the PEER staff will respond to your feedback as soon as possible. Thank you in advance for your thoughtful response.
Please note work is already underway to add the following features to the final database: (1) addition of 75% more ground motion records from recent earthquakes since 2003, and (2) incorporation of elastic response spectra for a range of damping ratios.
Hi,
Your beta tool is extremely useful. I have a few tips to help with functionality:
1) the results table is very clunky at times. the biggest problem is it’s too narrow and too tall. when event or station names require a line break the table becomes very long and to scroll right you have to page down, scroll over, then page back up. in the process it is easy to lose track of the line you are interested in since the record number/name are now hidden. being able to reformat column widths or the table size would help.
2) sometimes the search parameters do not seem to be enforced (for scaled searches at least). for example, if i input a Vs30 range of (700, 1000) resulting records are returned with values well outside that range. there could be an option for “soft” (i.e. ballpark) or “firm” (exact) ranges.
3) If there are many results, it appears only the best 30 are listed. it would be nice to at least have the option to view the next 30, or the next X best results.
Thanks
Dear PEER,
Thank you for your useful database.
After scaling to match a target spectrum, we have the table representing 30 records with their parameters including Scale Factor (ScaleF). I have a question: Are the downloaded records from “Save Time Series Records” already multipled by ScaleF or do we have to multiply those records by ScaleF?
I am looking forward to hear from you.
Vui
Hi,
I am using matlab code to calculate velocity, displacement time series and spectrum. time seires and spectrum shape are consistent but amplitude of them quite different. like hundred times bigger then mine. could you tell me that are multiply with 981
How Can I download Bipec?
Hello,
I search the 2000 version of the database whith USGS Site Classification (D < 180 m/s ), compared with the latest version of the database,the average shear wave velocity of some records are larger than 180 m/s (e. g. , NGA1205,Vs30=492.3 m/s ).
Thank you in advance,
Wu Hua
Hello and thank you for your very useful program.
I have a question: which codes the matching criterion is based on?
I am looking forward to hearing from you.
Cheers
Hi
I can not get old ground motion database from the site. You can help me?
Hi.. I need to have 3 -4 chart and horizontal component and soil types are classified (USGS) and position faults and record earthquakes. But I can not get from the site. You can help me?
Hi,
Some NGA sequences have more than 3 time histories. For example NGA 1409 has 156 records. Could you please let me know what is the reason? Which ones are more preferable to use?
Thanks
Here is the list:
NGA # H-Components # Vertical Component
1010 6 3
1409 104 52
1611 4 2
1612 4 2
1613 6 3
1614 6 3
1615 6 3
Hi,
Time histories of NGA 2114 can not be downloaded. As a matter of fact an empty zip file is downloaded.
The spectra and time histories of both horizontal components are shown by the web application but it sees that the vertical component info is missing in your database.
I checked the PEER 2005 database and I found the vertical component:
http://peer.berkeley.edu/nga/data?doi=NGA2114
Could you please verify this issue?
Thanks
Hi,
I tried to download un-scaled spectra but it was failed. I have read the manual carefully and listened to the tutorial video several times to make sure that I don’t miss something. Would you please assist me how to resolve this?
Thanks,
Ronny
Hi.. I need to have 3 -4 chart and record earthquakes. But I can not get from the site. You can help me?
Pingback: URL
The downloaded Search_Results_unscaled.csv file contains a field labeled Comp. (FN, FP, GM, Vert) in the same row as the station number and response spectra data. The downloaded file UnscaledAccelRecords.zip file contains the time history data for the selected(s) station (AT2 files). The AT2 files are named by the station number and component orientation not by the FN, FP designation. Can you include the record orientation into the search results table (i.e. 000 or 090) along with the FN and FP?
Thank you
Chris
I followed the process you described after downloading the unscaled record and read data horizontally. However there seem to be some differences between the plots of the time histories I get and the plot generated on your website. Do the plots on your website correspond to the unscaled record that we can download, or are the records further processed before plotting the time histories? Thanks in advance
I am thankful for your interesting database. I need to some of earthquake records that recommended at FEMA p695. I found a lot of them on your database, but I dont find the Record Sequence Numbers(NGA Flatfile) 848 and 1633 into databases. please help me to find them.
Thanks
Manjil rotated records 1634 and 1638 appear to be either truncated or uncorrected.
hi to all.
I am a geologist and working with EZ-FRISK7.52.
How can i do matchin spectrum by this software.
tank you.
Sorry, please read as NGA # 2114 (in stead of # 2112) for 2002 Denali, which I could not download. Thanks.
I tried to download unscaled record # 2112 for 2002 Denali, Alaska; but in vain. Would I be informed when it is available for download? Thanks.
Seems to me that the fault parallel component NGA_1614DUZCE.1061_FP.acc is not named correctly and two fault normal components (NGA_1614DUZCE.1061_FN.acc and NGA_1614DUZCE.106_FN.acc) are reported for Lamon 1061 station instead. Please have this corrected in the rotated database.
Hi
It is highly recommended it increase the maximum number of search outputs , from 50 .
( Unscaled records > Chart Controls > Total Number output )
This is the best way to download all unscaled records of a specific E.Q.
Thank you
After working with the PGMD, and reading both documents, I’m still confused as to what to expect when saving records selected using the “scaled” search approach. With “scaling” enabled in the search criteria section of the interface, I note that clicking the “Save Time Series Records” offers to save a file named “AccelRecords_rotated.zip” Note that the filename doesn’t mention “scaled.” The 2nd alternative is to click the “Save Original Unscaled Time Series Records” button, implying that the 1st alternative should in fact save scaled (as well as rotated) time series records. Nowhere in the available documentation is there any clear indication that the saved records contained in the “AccelRecords_rotated.zip” file are supposed to be unscaled (or scaled for that matter). Yet my experience thus far is that the records in that file are in fact only rotated, and not scaled. The scaling factor for each component pair is clearly indicated in the “SaveTargetSpectrum.csv” file I get when clicking on the “Save Search Spectra” button, and the associated spectra are definitely scaled, but it appears that it is up to me to apply the the reported scaling factors to my saved records in order to achieve the spectral MSE indicated. Am I doing something wrong? In any case, could you please clarify whether clicking “Save Time Series Records” should result in scaled or unscaled time series?
Marvin, your comments are correct. In order to facilitate other Users in understanding what they are obtaining by clicking on the “Save Time Series Records” button, we are uploading a revised version of the User Manual. In the next version of the database, we can hopefully allow Users to save directly scaled and rotated time series. Thanks again for your constructive comments.
Hello,
One of the parameters computed and reported for each time history is D5-95, the significant duration. To calculate it, you must have also calculated the record’s Arias Intensity. Could you please also add Arias Intensity as both an output parameter and also as a search parameter? In my office, we use AI as a design target, much the same way we use SA as a target. Thanks for your consideration.
Marvin,
your comment is really valuable for us, especially because we are in the process to plan the next revision of this product. Thanks for your feedback!
NGA record number 0879 (Landers) appears to be uncorrected data for both horizonral components, both the rotated and un-rotated versions. The vertical component looks fine.
Thanks for your feedback, we’ll look into it. The data was passed on to us based on previous analysis from a different research group, so I hope I’ll find out what is going on with that recording. Thanks a gain!
Is there a downloadable version of the program somewhere so that users could perform the scaling on their own records which are not in the PEER database?
Tim, the online application is based to some extent on the software DGML which was produced by Geomatrix Consultants, with the collaboration of Dr. Gang Wang. It is a Matlab executable software which includes several routines, and it should be available for distribution as a CD if you inquiry Geomatrix (now AMEC) – Oakland branch. I do not know if it is free of charge or not, sorry. Hope this helps.
Hi,
I’m working on sites located in Eastearn Canada, why is the Database not including any eastern strong motion event, like Virginia 2011, Saguenay 1988 …
Will they be added anytime soon?
Thank you!
Jean-Christophe
Jean-Christophe, as mentioned in the home page of the PGMD, this database includes records from regions associated with shallow crustal seismicity. The events that you quote are associated with stable continental regions (SCR), so they are targeted by a different reseach program, i.e. NGA-East. We have collected those recordings that you mentioned and eventually we are going to incorporate them in a different database, namely dedicated to SCR. I’m sure that there is a way to obtain these records elsewhere, but I do not sure on which institution to inquiry, sorry. Good luck with your work.
Thanke you for your dedication to the Ground Motion Database first.
I have a problem while saving the searching results. For example, there are 100 ground motions matching my options, but the limitaion of display is up to 50. So I wonder may I save all the ground motions matching my options? Is there any way to solve this problem?
Thanke you in anvance.
Wen, you can play with the limitation on how many records to display (in the chart control), but the limitation on the amount of data to download is imposed on us following the copyrights requirements from the various institutions that provided the original data. One possible way of solving your problem could be to impose different narrow range of Magnitudes or Vs30 while performing the search for compatible records, so that you can expand the possible amount of data you can download after repeating your seach different times. Hope it helps.
It would be nice if users could export the search results table as it appears on screen with all the valuable data (scale factors, magnitudes, descriptions, Vs30 , . . . .) to an Excel file.
Timothy, if you perform a scaled search and click on the button “Save Search Spectra”, the Summary of PEER Ground Motion Database Search Criteria as well as its results are saved as a .cvs file called “SaveTargetSpectrum”. It is compatible with Excel and it contains the informations that you quoted in your comment.
It may be a processor probem on my machine, but when I:
1. scale and download a set of data for a particular target spectrum
2. go back to the target spectrum page and define a new target
3. scale the same sequence numbers to the new target
The new set of files which I download – which I assumed would be scaled to the second target spectrum – are actually identical to those scaled to the first target spectrum.
Maybe a setting in the Internet Browser? Not sure what else to try other than closing the browser after every scaling action and re-starting at the PEER home page again.
Timothy, you are correct. The online application accepts unique tokens for defining target spectrum and compatible records, which are selected and listed on the base of their RMS with respect to the target spectrum. If you change the characteristics of your target spectrum, you will have different set of ground motions that are “most” compatible with it. One possible way to solve the problem is to impose that your seach will be performed on the desired NGA Sequence Numbers (or similar) as it is allowed when you use the “Additional Search Option” button.
Hello,
I want to use unscaled earthquake record to do my research. Can you tell me how to get the PGA,PGV and PGD value of each component of an original earthquake record using the PGMD? not the FN and FP component’s PGA,PGV and PGD.
Yoin, you can obtain those information if you perform the unscaled search and you click on the “Save Search Spectra”. It will create a xls file inclusive of several parameters, among which also the PGA, PGV and PGD.
hello.
how can i open file (*.at2) and draw (acceleration ~ time) from this file?
Thank you in advance,
tuannam
Tuannam, you can open the files by taking into accont the presence of some lines in the header (typically 4) and the fact that the acceleration time series data are presented in 5 columns. That being said, you have to skip 4 lines, then read 5 values from each line, and progress to the next line, read 5 values again and so forth until you complete your loop over the available lines. I’m going to send you in private a Matlab script that reads the acceleration data, for your convenience.
The time vector is easily to obtain once you know the sampling rate (second entry in the 4th line of the header) and the number of sampled data in your file. If for instance your sampling rate is 0.01 and you have 3700 samples, you can obtain a time vector using this line (in Matlab).
t=0:0.01:(3700-1)*0.01;
Let me know if this approach is not clear for you.
Hello,
I can’t download the PGMD Search Reportwhen I use the Unscaled Search. Everytime I click the “Save Search Spectra” button, it skip to a page says”We’re sorry, but something went wrong.We’ve been notified about this issue and we’ll take a look at it shortly.” I have tried many times but never succeed. Could you tell me how to solve this problem?
Emma, thanks for letting us know about this problem. Occasionally, once a month or so, the system needs to do some cleanup, and there was one dependency file that was accidentally deleted.
Now the system is fully operational and we’ll keep an eye on it.
I noticed that in this new version of Ground Motion Database, only the acceleration time series can be downloaded. Is there any way I can download the velocity and displacement time series?
Quidan, you are right. In the current version, only the acceleration time series are available for dowload. We might consider your comment/request for the future versions of the database. Thanks for your valuable feedback!
Hello,
I am trying to create response spectra for the research I’ve been doing, using earthquake records from PEER’s database.
It is important for me to know the soil type (S1, S2 etc.) of the site where each ground motion took place. Unfortunately, this information is not given for each station.
Do you know if it is possible to obtain the soil type for each NGA?
Thank you in advance.
Dinos, in the 2010 Beta version, you can serch by ranges of Vs30 values. In case you want to know more about the recording stations and sites, you can find a detailed flatfile following this link: http://peer.berkeley.edu/assets/NGA_Flatfile.xls. Good luck with your research!
hi,i have a question about finding vertical component of record.
how can i find out these components?.i can achieved horizontal componenet only!!! please handle me.thnx
Babak, when you perform a search for recordings, you can save the results by downlading a zip file with the selected ground motions. The files that contain the appendix “DWN” or “UP” are the vertical components. When you perform a search in the scaled section, you can match only the horizontal components (the geometric mean of them), but when you dowload the results, you will see the vertical time histories included in the zip file as well.
Hope it helps! Good luck!
Hi,
How could find the direction of time history records base on fault raupture diectivity? because as you now the accelogrames which is recorded in perpendicular to fault rupture are pulse like.
One of the search parameters is the “Pulse” option. According to this functionality, you can select “any record”, “only pulse-like records” or “no pulse-like records”. I invite you to familiarize with that option while perfoming your queries.
Good luck!
Hello
Could you tell me the method you used to integral the acceleration time series to get the velocity and displacement time series. Because I download the unscaled acceleration seires from your website and use cumulative integral method to compute velocity and displacement series, but I found that the figures of the velocity and displacemt series I got are quite different with the figures shown on your web site. So I want to know what method are you exactly use.
Zhang, the apperent discrepancy is not related to the integration method, but to the signal correction method. The acceleration time series are tapered and zero padded at the 2 edges before filtering, and then they are integrated to obtain velocity and displacement which are not affected by baseline trends. The corrected accelerations presented in the database do not include the zero padding, so when you integrate the time series you cannot obtain velocity and displacement fully compatible with the ones displayed in the web-application. If you would like a list of relevant literature in this regard, please write to peer_center@berkeley.edu, attn. Carola, and I’ll be happy to help you. Best wishes!
Hello,
I am interested in evaluting vertical ground motion records. Is there a way to specifiy “only vertical component”?
Thank you,
Don
Don, if you see the reply to Dinos, you can find a way to obtain the vertical components. Up to this version of the database, the option “only vertical componet” is not included, but we can study its feasibility for future releases. Thanks for your feedback!
We have been trying to download accelerograms recorded in California; however, it appears that for the records two parallel components are available and I am interested to define which is the normal and the parallel to the fault component. For example:
Record: Kern County
Parallel Components:1)NGA_no_12_PEL090.at2
2)NGA_no_12_PEL180.at2
I had used the unscaled procedure for downloading the accelerograms as I said in my previews comment. I have restated my question as it might have been unclear before.
Thanks for your help.
Vasilis
Vasilis, in the “unscaled search” option, you dowload records AS RECORDED, it means with their original orientation. Usually the numbers 090 and 180 refer to the azimuth of the components (the angle from North to the orientation of the sensor component, in clockwise direction). While the majority of the horizontal components are oriented in the the NS (180) and EW (90) direction, you can find cases where the orientation is a little more random.
In the “scaled search” option instead, the Fault Normal and Fault Parallel components are called “FN” and “FP”.
I am trying to download the ground motion data for record number 2114–the 2002 Denali, Alaska EQ recorded at TAPS Pump Sta. #10 CWB for a FEMA P695 analysis.
I can display the data on screen, but when I try to save the data, the resulting zip file is empty.
Thanks
Steve
Steve, you are right. Our database do not contain the original, as recorded, time series that you are looking for. However, we have the FN and FP components of this recording. You can try to perform a “dummy” search in the “scaled” section, by creating a default target spectrum, impose NO scaling, and impose that the matching recordings should come from the Denali event. You will be able to find the ground motion recordings you are looking for. Hope it helps!
Hi,
In current version of the database it is impossible to search based on hypocentral distance from tha source. Provided that some hazard deaggregation data are reported in combination of Moment Magnitude and Hypocentral distance, It would be very useful to have this option in search engine.
Thanks
Morteza, your point is valid. The choice to use finite-fault distance metrics such as Rjb or Rrup is related to the metric used in the NGA models as predictor variable. There is a good amount of literature that indicated statistically robust approaches to ensure distance conversion. If you cannot find these papers, please write at peer_certer@berkeley.edu, attn. Carola, and I will send you the references and/or material. Good luck!
Hi,
Could you please let me know how many ground motions can be shown by a search? I thick the default number is 30. When I change the default value to some thing more, such as 100, the maximum number of shown ground motions is 45. Is the any limitation? How many sequence numbers can be typed in the “Sequence Number” box? If I have the NGA sequence numebrs of something like 170 ground motions, how can I have all of them in one search?
Thanks
Morteza, you can virtually perform a search as wide as possible, but the limitation of 100 units is imposed by the agencies that kindly provided PEER with their proprietary recordings. The copyright requirements are in this way applied. Sorry for the inconvenience. Hope you can find the provided web application useful anyway for your research.
Hi,
I’d like to find Original Unscaled Time Series of the record sequence numbers “848, 1633, 1485″.
I tried to find this files on PEER, but there was zero result.
Could i get this here or not? Or did i try it by wrong way?
Hope your positive opinion
Thanks
Jiwook, these records have been excuded from the database according to some indications received by the original authors of the rotated time series database. According to them, the list of records excluded in described in Appendix B2, available in the “documentation” page.
Hello;
How can I find the difference between the columns in out put record from unscaled SAVE PGMD SEARCH RESULT? there are four columns in the out put file and I could not find any note that define the columns.
Thanks;
Arkan, I’m not quite sure I understand your question, but let me try. If you perform an “unscaled” seach, and you either:
1) save search spectra: you’ll be able to open and xls file with all the desired information, inclusive of summary of research criteria plus the properties of the selected GM with their headers
2) save original unscaled time series record: you’ll be able to save several recordings relative to the seach criteria applied. When you open this recording files, you’ll find an header as well as the acceleration time history in 4 columns. This format is quite common to the available recordings from GM databases and you should be able to open the files with a program which skips the header lines and reads 4 values for each line.
Let me know if I manage to answer to your questions, in the meantime all the best with your reseach!
Thanks;
The problem is related to the second part of your answer (save original unscaled time series record), when I import the text file of time series in Excel, it has 4 columns without any header to describe them, and I could not understand the difference of each column. I am wondering if you explain the exact meaning of these columns, and hoe related to the selected record.
Thanks;
Arkan, as I mentioned in my previous reply to you, these 4 columns are the actual values of the ground motion. You need to open them in a different way rather than xls, so to skip header and read 4 values for each line, go to the next line and read other 4 values, and so forth until the end of the file. This means that, if for instance you have 200 points to read, you have to wrap your 50 rows by 4 columns file into a 1 column file with 200 rows, did it help now?
Hello,
In my thesis, I am going to use only FN components of the pulse-like motions. I want to ask that do I have to know the location of the position of the fault to choose FN component.
Thank you in advance.
Hi, in order to select the FN componets, it is not necessary to know the position of the fault. You can just perform a “scaled” search (if you don’t really want to scale the ground motion, just set the Scaling Factor to 1), save the rotated time series, and consider the records with the name ending in _FN.acc
Hello. Here is my question as well as my feedback. Why one needs to have scaled accelerograms? Earthquake is not like a radio. How can someone decide the amplification or the de-amplification of an accelerogram? Scaling does not affect the duration of the accelerogram. I know current performance-based seismic design needs and targets but scaling procedure is full of doubts. I mean, why not to use entirely artificial motions instead of scaled ones? Thank you in advance.
Hi,
I’ve downloaded a set of 30 time histories with criteria I selected and everything worked fine. I then modified the values of magnitude, distance, etc. for another set but as I try to download the time histories, I only get 21 files in the .zip file. If I look at the list of time histories found, I have 30 of them and the each have names for the files I should be downloading (FN FP and vert) which are not all present in the .zip file. Have you seen this problem before?
Dominic,
the problem you are experiencing might be related to the fact that, in default, the application saves the FIRST 7 recordings that provide a good match with your target spectrum. Being composed of 3 components (2 horizontal and 1 vertical), the total amount of downloaded time series would be 21 following default setting. In order to download whatever time series you desire, please check the box in front of them and move forward with the saving operation. Good luck with your selection!
Hi,
1994 Northridge-06_Jensen Filter Plant Generator earthquake (NGA no 1704) does not have a complete set of data files. There is only one unscaled data of one of it’s horizontal components. Is there a particular reason for this?
Thanks.
Selma, unfortunately there is no specific reason for this lack of data. We are currently performing a check among the available GM recordings to make sure no component results missing, but this improvement will be reflected in the new versions of the database.
Hi,
There is a note at the saved search spectra like below.
“Note: Horizontal Components with star (*) in front are used to compute the mean spectra”
But i couldn’t see the stars anywhere. How can i understand which components used to compute mean spectra ?
Ibrahim, the mean of the selected spectra is computed from the records that are checked in the Spectral Search result page. In default, the application marks the first 7 records, but the user is able to modify this selection. The ground motion component used to achieve the spectral matching with the target spectrum is the Geometric Mean (GM) of the two horizontal components, as rotated to the FN and FP directions.
As for the (*) symbol in the saved search spectra output, you are correct and we will give it proper consideration for the future versions. Thank you.
Hello,
I wonder whether the ground motion records in this database are rotated, as said in the database website. I noticed that the downloaded spectra were actually rotated but not sure whether the time history records were rotated since the calculated spectra using the records in the database are different from those in the database. Would you confirm the orientation of the records in the database?
Hello Cy. Yes, I can confirm that, upon clicking on the “Save Time Series Records” in the “Scaled Search ” section, you will be able to retrieve FN and FP time series.
Dear PEER,
I find your database very comprehensive and useful. I am now using it doing a research to earthquake behavior of large container cranes at the Delft University of Technology in the Netherlands.
I have only one question; do you already have data available of the Japanese Mw9.0 earthquake of March 11?
I would you to have the spectrum to compare it with other earthquakes..
Thank you!
John, the current Beta version of PGMD will not include the 2011 Tōhoku earthquake. Furthermore, the event is associated with a subduction tectonic environment, whereas the PGMD collects only records from active regions with shallow crustal seismicity.
You can retrieve records from that event at the kik-net web site at http://www.kik.bosai.go.jp/
Good luck with your research!
For SHAKE91, Idriss refers to the Diamond Heights(E-W) record for Loma Prieta, Has the station been renamed?
The station that you are quoting is SF – Diamond Heights, the CSMIP station number is 58130.
If you perform a combined search for event name and station name, you’ll be able to retrieve the data associated with your request.
Hope this helps.
Could you please answer my question?
Based on FEMA356, the scaling process for ground motion is:
“For each data set, the square root of the sum of the squares (SRSS) of the5%-damped site-specific spectrum of the scaled horizontal components shall be constructed. The datasets shall be scaled such that the average value of the SRSS spectra does not fall below 1.4 times the 5%-damped spectrum for the design earthquake for periods between 0.2T seconds and 1.5T seconds”
Are PEER scaling process is same as FEMA356? If, yes, I read your documents carefully, but there is not any hint about coefficient “1.4″ and “average value of the SRSS spectra”.
Peyman,
the PGMD web application has the functionality to create different design spectra, one of them being in accordance with Code ASCE/SEI7-05 Reference: “Minimum Design Loads for Buildings and Other Structures”, ASCE, 2006.
The user has the capability to customize his search criteria based on his project’s requirements, including to impose limitations to the range of possible scaling factors. The parameters adopted by the users are beyond our responsibility, because we assume that an informed and professional practitioner will be aware of their meanings and requirements.
For your information, the quoted SRSS is equal to the geometric mean of the two horizontal components that we provide.
Good luck with your project!
Hi,
How can the SRSS and geometric mean be equal? SRSS = sqrt(S1^2+S2^2), while GM=sqrt(S1*S2)
Thanks
Hello,
After each search, the earthquake records that can be downloaded using “Save Original Unscaled Time History Records” button are not in the same coordinate system with the corresponding spectra that can be downloaded using “Save Search Spectra” button. The spectra are in ‘normal-parallel to fault’ coordinate but the time history records are not in those directions. Is there any way to get the records and their corresponding spectra in the same directions?
Where is the Appendix B (Summary of PEER-NGA Records included in and excluded from PGMD Database) you are mentioning in the Technical Report for the PGMD Web Application? It is not in the report, nor anywhere else (or I do not know where to look at).
I need the list of the datas you are providing, which I found in the flatfile. However, I believe that file includes the 369 records excluded from PGMD database. I searched for some of the records I need (which are in the list) but not in the system.
Thanks!
Thank you for helping us improve our service.
The documentation page is now complete and includes the list of Records included or excluded in the current Beta version of the PGMD.
Is there a way to turn off/on the legend in the spectra plot? It sometimes covers up one-quarter of the plot making it difficult to read the spectra in certain situations.
Thanks!
Thanks for you feedback! The “legend-switch” feature is not currently implemented but we received your request and will give it consideration for the next versions if this application.
Hi,
I was wondering what the damping ratio is for the calculated response spectra on the site. It is mentioned that in the near future more damping ratios will be added, but what is the current one? Is it 5%?
Thank you in advance,
Shideh
Shideh, your guessing is correct. The response spectra are computed for 5% damping. If you have further doubts on the parameters adopted in the web application, please refer to the manual and the technical documentation.
Thanks for your interest in the database and its functionality.
Your scroll bars disappear too early to be able to highlight line items when trying to search for earthquakes and stations. Please adjust them.
Others have reported similar kinds of glitches with the interface and it often is the result of using out-of-date browsers. Please note that this web-application works better with browser at least versions:
* Internet Explorer 8
* Firefox 3.6
* Safari 5.0
Do you plan to add any criteria in searching the GMs in terms of ranging the GM IMs such as PGA, PGV, etc.
Actually, this is already available in the previous version. Thank you in advance.
Your comment is really important to us, given the fact that the current BETA Version is under study for future improvement. Thanks for your feedback.
The 2000 version of the database offered an ability to search for records based on pseudo-acceleration response spectrum (would pick all records with spectra passing through a bounding box defined by T_min, T_max, Sa_min, Sa_max). Am I correct that this type of search feature went away in the 2005 version of the database and also does not exist in the 2010 beta version? i.e. Is there (or will there be) anyway to search these later versions of the database for records that satisfy certain criteria on their response spectral accelerations?
Thanks,
Keep up the good work.
Dear Todd, your observations are correct, in the sense that the customizable criteria for searching spectra compatible with the target spectrum do not include explicitly the Sa_min or Sa_max of a bounding box.
However, in the 2010 Beta Version, this option is not only implicit, but also enhanced by the introduction of a “Weighting Function”. This function can be fully customizable: it affects the computation of the degree of similarity between the target spectrum and the candidate spectra, and controls the related scale factor.
I invite you to read the pages 10 and 11 of the Technical Report, where the effect of the Weighting Function is fully described.
Should you have any further doubt, please don’t hesitate to contact us again.
Hello,
What is being used to calculate the response spectrum for each time series? In other words, what numerical solution algorithm is being employed (central difference, Newmark, other) in solving the SDF equation prior to extracting the peak displacement at each period?
Thanks!
~Simon
Hello and thank for your interesting question!
To the best of our knowledge, the code used to compute response spectra implements the method described in Nigam and Jennings, 1969, “Calculation of response spectra from strong-motion earthquake records”, published in the Bulletin of the Seismological Society of America; April 1969; v. 59; no. 2; p. 909-922.