As part of the development of the ISC data management system it is intended that all data will be maintained on the ISC's Oracle database. The database development process began with the development of a suitable schema, ISC1.0, to contain all required and useful data. The second stage which is described in this document was to parse all of the previously published data from the ISC bulletin into the database. This data was available from 1904 to the present day, though phase data was available only from the start of 1964. These data are in two file formats: historical files ( prior to 1964 ) and Fixed Format Bulletin ( FFB ) files.
The FFB and historical files are the only readily accessible source of electronic data covering the entire period over which the ISC has analysed and published data. There are some limitations in the content of this data, principally the absence of any un-associated phase data and phase data only associated with the primary hypocentre of any event. However this data is available and a project to extract the un-associated phase data from the ISC's tape archive exists. In the future as new data is loaded in the database rather than SIP all of the phase associations reported to the ISC will be preserved.
The loading of the database consists of 6 main stages:
The first 5 stages are run from the script run-allx in the directory /export/home/oracle/ISCDB/LOAD. The final stage is carried out by the script insertx, in the same directory. As the scripts stand they require only 2 command line parameters, the year ( 4 digits ) and month ( 2 digits ) and will then run to completion.
ksh> run-allx 1997 06
ksh> insertx 1997 06
However they rely on the FFB being in the default directory of /export/iscftp/Bulletin/year. A few simple checks are made at the start of any run to ensure that the year and month in question have not been previously loaded.
A log file is kept of the output from each stage of the process and must be checked for error messages prior to the data being inserted into the main tables. A description of the important points of the log file will be given for each stage of the process. The log file for a particular job will be written to /export/home/oracle/ISCDB/LOAD/LOG/YEAR/YEARMONTH. After a month has been inserted the log file is made read-only to prevent accidental modifications.
For access to the main tables the ISC data administrator username and password are required to run any program that modifies the database tables. Although the code could be run for individual users to load their own tables. The scripts use environment variables to hold the username ISCUSER and password ISCPASS.
Notes:
All of the scripts were written for a Korn shell environment.
Where appropriate, files containing data or code have been named after their associated database table. With the exception of the pub_comments table which was originally called published where files are named after the original table.
On occasion, an FFB other than those for published data will need to be read and loaded into a set of tables. To do this a number of parameters in the scripts need to be changed manually and insertx should never be run. The two parameters that need to be changed are the environment variables FILE ( location of the FFB ) and OUTFILE ( log file ).
The scripts in which this needs to be done are: /export/home/oracle/ISCDB/LOAD/readx /export/home/oracle/ISCDB/FFB/gethypx /export/home/oracle/ISCDB/FFB/getphx /export/home/oracle/ISCDB/LOAD/PROGS/bothx
Examples of the alternative settings for the variables will be found in the files.
The parsing scripts were developed from the existing FFB search script developed for use on the ISC WEB site and use the nawk programming language. Nawk is a language designed to process sequential text files and as such was well suited to the fixed format text in the FFB files.
The following files are required to run stage 1 of the loading process and are called from run-allx:
File Location Purpose
readx ~/oracle/ISCDB/LOAD Script to call main and set the
command line parameters.
main ~/oracle/ISCDB/LOAD/PROGS nawk programme to control the
parsing of the FFB files.
read.nawk ~/oracle/ISCDB/LOAD/PROGS nawk subroutines for reading the
various parts of the FFB file.
insert.nawk ~/oracle/ISCDB/LOAD/PROGS nawk subroutines for writing the
values extracted from the FFB
into files for SQL*Loader.
phase_numbers ~oracle/ISCDB/LOAD/DATA Data file with ISC Phase ids.
The output from these programs is in 9 files, shown below, in the directory /export/home/oracle/ISCDB/LOAD/DAT. These do not need to be examined unless there is a problem with SQL*Loader and data has to be modified manually rather than through improvements to the nawk scripts.
published.dat association.dat stamag.dat amplitude.dat event.dat netmag.dat hypoc_err.dat phase.dat hypocenter.dat
The FFB files are sequentially written text files containing all of the data that was published in the printed bulletin. Data within the FFB is contained within 11 different records or formats:
Format 0 File Header Format 1 Epicentre Format 2 Epicentre Continuation Format 3 Epicentre Comment Format 4 Epicentre Comment Continuation Format 5 Initial Phase Format 6 Later Phase Format 7 Phase Comment Format 15 Initial Phase ( 5 character station code ) Format 90 Agencies Format 91 Stations
A detailed description of the FFB can be found online at http://www.isc.ac.uk/doc/misc/fform.html or on the ISC CDs.
No information from the Header record is used for loading the database.
For the FFB the order in which the data is written is:
Header Agencies All agency numbers and information currently available Stations Only station information relevant to data within the FFB Then for each event: Secondary hypocentres Formats 1 or 3 Primary hypocentre Format 1 Phase data Formats 5, 6, 7 and 15.
Each format 1 line may also have associated format 2, 3 records, and format 3 records may have continuation lines.
The data from the FFB is extracted and inserted into the database tables described briefly below. A complete description of the ISC schema is given in the document Database Schema, available online at http://www.isc.ac.uk/doc/database/schema.html.
Table Contents event prime hypid hypocenter information relating to the details of all epicentres hypoc_err information relating to the accuracy of epicentre estimates netmag magnitude data from all associated phase data association data detailing the link between an epicentre and phase data phase station or phase data (associated and un-associated ) stamag magnitudes from each phase data amplitude amplitude data from each phase data pub_comments comments, Format 3, 4 and 7 data remark Non-FFB data detailing any necessary about the data loadedAppendix 1 details the relationship between the FFB variables and their destination attribute in the database tables. Importantly any modifications of the data or algorithms used to modify the data in the FFB are shown here.
All data within the database in identified by a unique ID number, this number provides the basic association or external key between all of the tables.
Event identification, evid, is incremented on each occurrence of a new event, indicated where the last primary epicentre flag is A', Format 1 position 26. This method is required because the primary epicentre is the last entry in an event and the FFB file is read sequentially.
Hypocentre IDs are incremented when a Format 1 line is read.
N.B. Though comments ( Format 3 ) can be included as an epicentre estimate in the FFB they contain no data beyond a time and a comment and as such were not entered into the hypocenter table. They are preserved in the pub_comments table as a unique comment associated with the appropriate event. prime_hyp is taken as the appropriate hypid when the prime epicentre flag is read.
Individual phases are identified by phid, which is incremented at each occurrence of a Format 5,6 or 15 line. A reading, a group of phases from the same station and event, is linked by rdid which is incremented at each Format 5 or 15 line.
magid, ampid and remid are incremented only when the appropriate data has been extracted from a phase line.
A large number of magnitudes are contained within the comments ( format 3 and 4 ). These magnitudes have been extracted and added to the netmag table. These magnitudes typically consist of only a magnitude, type and reporter. It is useful to note that the author for these netmag entries will be of the form 'NEIC;LYU' where the first part is the comment reporting agency and the second, if present the magnitude reporter.
For both hypocentres ( the prime must be within a specific month ) and phase dates where the date is outside of the month in question, the month is not incremented, but an extra day is created, e.g. 32nd January. This is not acceptable to general use and these dates have been identified and months and years incremented accordingly.
It has been important to differentiate between null values and 0 when parsing the data.
Log output from Stage 1 is as shown below:
Thursday September 16 10:07:30 BST 1999 Date and time of start of run 601B ISC is not prime hypocentre 608B 920B 927B Number of Hypocentres = 1217 Number of events = 661 Number of Readings = 12439 Number of phases = 19820 01 1964 Date of historical file header Number of events = 661 Number of Hypocentres = 1217
The features which need to be examined are the number of events and hypocenters in the FFB, first section, and in the historical file, second section. This is a rough guide only to the accuracy of the extraction program, discrepancies though are noted and are described later in this report. The non-prime ISC hypocentres are only present for information purposes.
A number of parameters in the database tables have to be set up after the FFB data has been loaded and a number of checks have to made on the data. Due to the size of the database the only efficient way to carry out these procedures was to have a set of tables containing only data from the month in question. Also this means that the main database contains only data that has already been checked.
A tablespace, load, was created solely for these temporary tables and the tables were created using scripts generated using make from controlled source files. All the SQL scripts for generating the load tables are in directory /export/home/oracle/ISCDB/CREATE, and the source files are in the sub-directory DESCRIP. The make command needs to be run from the directory DESCRIP.
The script create_loadx simply calls each SQL script in turn for each table.
The log file need only be examined for Oracle error messages. If all has gone well a series of 9 messages of Table dropped and 9 of Table created should be present. Typically only a message saying that a table could not be created is important, as if a run has been aborted at some point the tables may have been dropped but not recreated.
The script loadx, in directory /export/home/oracle/ISCDB/LOAD, calls the Oracle utility SQL*Loader for each of the 9 tables being loaded. A control file is required to provide information about the parameters read from the data files and their destination in the database. SQL*Loader produces log files describing the loading process and files containing any bad data. All of the SQL*Loader files are in the directory /export/home/oracle/ISCDB/LOAD/CTL.
SQL*Loader uses control files to set parameters such as the input file and database destinations. These files are named after their associated table with file extension .ctl. A log file, extension .log and a file containing any flawed data, extension .bad, will be created when SQL*Loader is run.
In the main log file the number of bad data lines and the data in question will be included. If there is any bad data the run will have to be made again with either the bad data, or parsing program modified as appropriate.
A number of modifications need to be made to the tables which were not possible using the nawk program, essentially because the reading of the FFB is done sequentially rather than randomly. The various SQL and PL/SQL scripts are run by the script fixupx, as before from the directory /export/home/oracle/ISCDB/LOAD. The various scripts called by fixupx are detailed below in the order in which they are executed. All scripts are in the directory /export/home/oracle/ISCDB/LOAD/FIXUP.
non_prime-isc.sql (PL/SQL)
This update sets the phase association to be to the ISC hypocentre where the ISC hypocentre is not the prime. The ISC hypocenter is typically non-prime when a fixed location is know for the event, for instance if the event is a nuclear test.
Note though that a number of problems occur where on old data the number of observations contributing to an ISC magnitude does not match the number of stamag entries. This is because the algorithm used to calculate a network magnitude when the data was published has now changed. For instance, data used previously by the ISC is not now due typically to changes in distance constraints. There is typically little or no documentation about when and what changes were made to the code.
The log file contains a small description of this update and should be followed by the entry "PL/SQL procedure successfully completed.". If not then the reason for the failure should be investigated though at no point in loading of the database has an error occurred at this stage.
set_id_num.sql (PL/SQL)
All primary ID numbers used in the database have to be unique. This script sets all IDs in the temporary tables to start from the next available ID number from the main tables. It is not necessary to use a unique number generator because only 1 user has write access to the main tables and as long as standard scripts are used a non-unique ID number cannot be generated.
As each table is updated a comment to that effect is printed in the log. Then, as with the previous update the log file should have the entry "PL/SQL procedure successfully completed.".
index_load (SQL)
Despite the small size of the temporary tables a number of indexes are required to ensure efficient running of a number of later scripts. Though some further indexes are created and dropped as required by later scripts.
Consideration of the value of indexes has cut the time taken by some updates by as much as a factor of 10 reducing the time taken to run the updates from around 4 hours to just under 1 hour. Little is printed in the log file beyond confirmation of the index creation and the time taken to do so.
hypocenter.sql (SQL)This SQL script carries out a number of updates to the hypocenter table. In a number of these cases the SQL code relies upon the id numbers being sequential. This assumption should not normally be relied upon, though for the load tables it is valid.
Each stage of the update script is indicated in the log file, followed in each case by the number of rows updated, the time taken and confirmation of the commit.
| 1 | The hypocenter magid is set from the first appropriate entry in netmag. |
N.B. Typically this will be mag1 from format 1. However please note that if the magnitude has come from a comment line the author in netmag will be in the form 'code;code', where the second code will be the hypocenter author. |
|
| 2 | The hypocenter magnitude and magtype are set from netmag using the magid set in the previous stage. |
| 3 | For events where there is no ISC hypocentre and ndef is not set then ndef is set to nass. |
| 4 | srn and grn were set for non-ISC hypocentres from the ISC entries. |
N.B. This update is no longer used. |
|
| 5 | The velocity model used is set to 'JB' for all ISC hypocentres. |
| 6 | ndp and depdp are set null where both are 0. |
| 7 | For ISC hypocenters nass is set by counting the number of associated phases. |
hypoc_err.sql
| 1 | sdobs is set to null where it has been input as 0. |
| 2 | sdepth is set to null where it has been input as 0. |
| 3 | stime is set to null where it has been input as 0. |
N.B. This update is no longer used. |
|
| 4 | smajax, sminax and strike are set null where smajax ( i.e. either ) is 0. |
For the hypoc_err updates each stage is not labelled but the log file has the number of rows updated, the time taken and confirmation of the commit.
stamag.sql
This is a fairly major update that attempts to correctly associate all stamag magnitudes with the netmag magnitude that is calculated from them. The magtype of the stamag magnitudes also has to be generated using the appropriate algorithms. The association is not maintained in the FFB and the algorithm used in the ISC code had to be extracted and used to help in the association.
The magtype is 'mb': if period is <= 3 seconds or null and period > 0 or log( amplitude/period) is not null and distance >= 21 and distance <= 100
N.B. Notes ( From Robin ) suggest that prior to the October 1975 Bulletin that distance was not used in the 'mb' algorithm. However two exceptions were encountered:
October 1975 the distance had to be <= 100 to correctly associate the stamag entries.
June 1974 the distance had to be > 21 to correctly associate the stamag entries.
The magtype is 'MS': if period >= 10 and period <= 60 seconds and distance > 20 and distance <= 160
N.B. Depth <= 60 occasionally had to be used as an extra constraint.
N.B. Parts of the ISC association code suggest a minimum distance of 5 degrees was used, however other parts of the code eliminated magnitudes with a distance below 20. The algorithm was essentially tuned to suit the data where particular cases showed potential errors in the code. For instance phase magnitudes where the distance = 20 also had to be eliminated to make the associations correct.
MS magnitudes are calculated using the method shown below. Magnitude from reading: MREADING = ( MZMLR + MZ + MN +ME ) / ( No. magnitude components ) Typically only a magnitude from the Z component would be present. Network Magnitude: MS = ( MREADING1 + ... MREADINGn) / ( No. Reading magnitudes )
The above methods were used to calculate a netowrk magnitude from the associated stamag values, this magnitude is compared to the network magnitude from the FFB as part of the data integrity checks detailed later in the report.
For mb magtypes the magid is set for the first 'mb' magnitude in each reading. This is why the uncertainty column in stamag is set to the reading id of its appropriate phase line. The stamag uncertainty is not in the FFB. It was also important to remove any 0 magnitudes as these make the SQL impossible to construct.
MS magtypes are easy to associate as they are all required to be associated with the netmag entry.
For both of these magtypes a test is made that the association is correct, though only by number of associations and a relatively close magnitude, see test section later.
The script is carried out in a number of stages, with the log file containing a comment on which stage is being run and the usual output. The individual stages are described briefly below.
| 1 | Uncertainty and magid are set to NULL ( should be the case anyway ). |
| 2 | Type of uncertainty is changed to hold a larger number. |
| 3 | Uncertainty is set to the reading id ( rdid ) of the corresponding phase id ( phid ). |
| 4 | Magnitude is set NULL where it is 0. |
| 5 | mb magtype is assigned where appropriate. |
| 6 | MS magtype is assigned where appropriate. |
N.B. A temporary table is created with the phid of the first mb magnitude in each reading. | |
| 7 | MS magids are associated from the netmag table. |
| 8 | mb magids are associated from the netmag table |
Though as much effort as possible has gone into the accurate and automatic extraction of data from the FFB files, some errors or inconsistencies were found in the data which could not be resolved. These were shown up by a number of tests carried out on the table after all updates had been made.
A number of netmag magnitudes could not be correctly associated with their stamag values, typically there were not enough values but in 3 cases there were too many suitable values. As a fraction of the total ( 0.04% ) the number of errors is small. The most recent error is in 1988 data and most are in the 1960s data suggesting either a change in the algorithm used to calculate netmag or improvements in data management as computing resources improved. All of these errors are shown in Appendix 2 and are identified in the database by a remid of 1 in netmag.
The second problem with netmag associations occurred where the netmag calculated from the correctly associated stamag entries did not agree to within 0.1 of the value of magnitude 1 or 2 in the FFB. This is only searched for where the network magnitudes have been associated correctly. The differences are still very small and likely to be simply rounding errors as only 1 decimal place is recorded for the stamag entries. All the occurrences of this error have a remid of 8225 for the netmag entry and a listing of these is given in Appendix 3.
In all cases where the netmag and stamag values gave an error an examination of the data was made to ensure no errors were present in the program used for association. To facilitate this the log file contains details of each incorrect netmag. Despite detailed investigations a number of errors could not be solved by modifying the program or by modification of the associations as sufficient data was not available in the FFB to indicate what was wrong.
16-11-1996 07:48:23, hypid 2157458, had an ML magnitude of 27. This is clearly wrong and appears to be a duplicate of the depth value. The database value has been set null with a remark, remid 8238 made to record the problem. The ML 27 value was in the published bulletin.
A number of operator phase IDs in the FFB are essentially rubbish, but these were generally allowed to go into the database as checking every possible phase type was not possible. The simple and oft repeated errors were eradicated automatically. A typical error was where a phase had a number at the end of the string e.g. 'P 4', the number was removed from the phase before entering it into the database.
However some, shown below had to be removed by hand as the speech marks in the string affect the function of SQL*Loader. All had the operator phase ( phase table ) set to null in the database.
Date phid Operator id Reporting Station Database id 24-07-1976 06:22:21 22825445 2&"& MAW Null 16-05-1976 22:21:11 22927145 )&"& MAW Null 28-12-1975 15:37:41 23197144 2>"& TOO Null 14-01-1979 12:40:19 21306116 S"P BRS S'P'
From 1996 the fake P phases supplied from the NEIC were identified in the FFB with an operator ID number of 111 and typically an alphanumeric ID of "P". In the database these phases have been given an ID of "PFAKE" to ensure their correct identification. A detailed description of the nature of fake P phases can be seen at http://www.isc.ac.uk/doc/analysis/1998p01.
Only 7 occurances of non P phase fake Ps were found, these are shown below. But it has been confirmed that all should have an operator ID of P. The ID number in all cases was 111 and they are clearly fake P from the printed bulletin.
DAY MSEC ISC_PHAS OP_PHASE STA PHID ------------------- ---------- -------- -------- ------ ---------- 11-01-1996 04:05:00 0 NULL pP TUC 29572290 11-01-1996 04:05:20 0 NULL pP ALQ 29572318 11-01-1996 04:10:20 0 NULL pP OBN 29572346 11-01-1996 04:10:30 0 NULL pP BOSA 29572385 30-01-1996 22:18:50 0 NULL PKIKP CCM 29667534 30-01-1996 22:18:50 0 NULL PKIKP FVM 29667538 30-01-1996 22:18:50 0 NULL PKIKP SLM 29667541
As part of the testing procedure the historical file is compared to the FFB, this has shown up a number of inconsistencies between the two files.
One frequent inconsistency occurred where there were two near identical hypocentre entries in the FFB and one in the historical file. The only difference is that one entry would have a comment attached. . As all of these hypocentres were published both have been entered into the database, though a remark has been added. In a certain respect these errors do not affect the database as there is no inconsitency with the data in the same way as network magnitudes. However there is some possibility that the duplicates may need to be deleted at some point. A list of these hypocentres is given in Appendix 5.
However the opposite to the previous problem has also been encountered: in February 1969 the UPP epicentre ( 21-02-1969 10:58:24, hypid 1790034 ) appears twice in both the historical file and the printed bulletin but has been removed from the FFB file. It is this form of undocumented editing of the files which make it difficult to guarantee an exact match between data in the database and that in the printed bulletin.
The following 4 discrepancies occurred where an event consisted of a single hypocentre with no phase data. All of the hypocentres/events were deleted from the historical files.
DATE msec Agency hypid remid 31-10-1979 23:29:34 100 ADE 1457395 8237 31-01-1979 23:41:43 0 ADE 1487619 8237 31-03-1973 23:50:24 800 CAN 1691306 8245 30-09-1975 22:00:36 800 PAS 1597140 8241
| February 1975 | The ISC epicentre for 09-02-1975 21:18:10 has not been included in the historical file and the LAO epicentre has not been included in the printed bulletin. This particular epicentre should not have had the ISC epicentre published because there are only 4 associated readings, ISC epicentres usually have at least 5 readings. The most likely possibility is that after the first editing pass some readings were deleted but not the epicentre. Hypid 1615048, remid 8242 |
| February 1969 | In both the printed bulletin and the historical file the event with an ISC hypocentre at 21-02-1969 08:41:06 had an epicentre from UPP, though clearly in the wrong event. But unlike most other similar errors the incorrect epicentre has been deleted from the FFB. It does highlight the problem of knowing exactly what is being extracted from the FFB and how it compares to the printed bulletin. |
| April 1964 | A completely spurious event appears in the FFB where the hypocenters, including the ISC hypocentre, appear to have become associated with a few phases from the two surrounding events. The printed bulletin does not contain these data and to load it correctly into the database it was deleted temporarily from the FFB file. Part of the FFB is shown in Appendix 4. |
| March 1964 | An event with an ISC epicentre at 31-04-1964 00:44:52 also has a USGS epicentre in the printed bulletin at a time of 00:44:53. The FFB has no entry for the USCGS epicentre and the historical file shows the epicentre with a time of 03:44:53. As the epicentre is not in the FFB it must have at some time been deleted, as there is no explanation as to why there will be no entry in the database. Though there is no explanation of how the printed bulletin could be correct and the FFB from which it was created wrong. |
| December 1967 | A hypocentre in the FFB and printed bulletin has been removed from the historical files ( From NOU at 31-12-1967 04:52:14 ). There is another NOU hypocentre at the same time but slightly different location, so the most reasonable explanation is that this was considered a bad duplicate, though only after publication of the data. It has been left in the database with a comment. The questionable hypocentre has hypid 1818835 with remid 8248 added to indicate some doubt about it's reliability. |
| July 1974 | The event evid 746383 on 27-07-1974 21:24:33 had been deleted from the historical file. The hypocentres are shown below. |
DATE Agency hypid remid 27-07-1974 21:24:33 BCIS 1648828 8244 27-07-1974 21:24:36 HFS2 1648829 8244 27-07-1974 21:24:39 NEIS 1648830 8244 27-07-1974 21:24:39 ISC 1648833 8244 27-07-1974 21:24:41 MOS 1648831 8244 27-07-1974 21:24:43 LAO 1648832 8244
| September 1976 | ndef set to 0 in the FFB, not supposed to be a valid entry for this variable, set to null in the database. |
| August 1976 | ndef set to 0 in the FFB, not supposed to be a valid entry for this variable, set to null in the database. |
| July 1976 | ndef set to 0 in the FFB, not supposed to be a valid entry for this variable, set to null in the database. |
ISC hypocentre for 05-06-1994 18:36:00 where there is no phase data in the FFB, nsta and nass are null but ndef is 17. The ISC epicentre has not been included in the historical file. ISC hypid 374624.
This stage of the process involves a simple SQL script which first counts the number of rows in each load table and then inserts the whole load table into the main database table. The count of the rows is a simple check to tally against the number of rows inserted by the SQL statement, both of which are printed in the main log file. At this stage any failure to insert the rows would be messy to undo. Typically failures would only occur if there was insufficient space in the table or failure of the Oracle Instance.
Once loaded the log file is made read-only to prevent accidental modification or removal.
During the early stages of loading the main database tables were analysed after each month was loaded but this had to be reduced to each full year as the tables got bigger and finally dropped. This was because the time taken to estimate the statistics was too large an overhead on the loading procedure.
FFB value Database tables and column
Record category NA
Next record category NA
Year hypocenter.day
Month hypocenter.day
Day hypocenter.day
Hours hypocenter.day
Minutes hypocenter.day
Seconds hypocenter.day
hypocenter.msec
Precision of time NA
Agency number hypocenter.author
hypoc_err.author
pub_comments.author
netmag.author
Prime estimate flag event.prime_hyp
Latitude hypocenter.lat
Precision of latitude NA
Longitude hypocenter.lon
Precision of longitude NA
Depth in kilometres hypocenter.depth
Precision of depth NA
Magnitude 1 hypocenter.magnitude
netmag.magnitude
Magnitude 1 (end of range) NA
Magnitude 1 precision NA
Type of magnitude 1 hypocenter.magtype
netmag.magtype
No. observations contributing to mag 1 netmag.nsta
Standard error for magnitude 1 netmag.uncertainty
Precision of standard error for mag 1 NA
Geographical region number hypocenter.grn
Seismic region number hypocenter.srn
Number of observations hypocenter.nsta
Standard deviation of one observation hypoc_err.sdobs
Precision of standard deviation of one ob NA
Obs contributing to standard deviation hypocenter.ndef
Magnitude type is set as shown below for any magnitude type in format 1,2,3 and 4:
FFB magtype in Database
S MS
W MW
L mL
All other FFB magnitude types are made lower case and prefixed by 'm'.
FFB value Database tables and column
Record category NA
Next record category NA
Year NA
Month NA
Magnitude 2 netmag.magnitude
( If no mag 1 ) hypocenter.magnitude
Magnitude 2 (end of range) NA
Magnitude 2 precision NA
Type of magnitude 2 netmag.magtype
( If no mag 1 ) hypocenter.magtype
Number of obs. contributing to mag 2 netmag.nsta
Standard error for magnitude 2 netmag.uncertainty
Precision of s.e. for magnitude 2 NA
Standard error for origin time hypoc_err.stime
Precision of s.e. for origin time NA
Standard error latitude (degrees) see below
Precision of s.e. for latitude NA
Standard error longitude (degrees) see below
Precision of s.e. for longitude NA
Standard error for depth (km) hypoc_err.sdepth
Precision of s.e. for depth NA
Explosion/effects flag: hypocenter.evtype
Mantissa of explosion charge size in tons NA
Exponent of explosion charge size NA
Precision of explosion charge size NA
Number of pP-P observation hypocenter.ndp
Std. deviation of 1 pP-P observation NA
Depth from pP-P observations (km) hypocenter_depdp
Std. error of depth from pP-P observations(km) NA
Maximum intensity NA
Intensity scale NA
Distance in degrees to closest observation hypocenter.mindist
Degrees to most distant observation hypocenter.maxdist
The values of standard error for latitude and longitude are used as shown below:
if standard error latitude > standard error longitude
hypoc_err.smajax = 111.0 * standard error latitude
hypoc_err.sminax = 111.0 * standard error longitude
hypoc_err.strike = 0.0;
else
hypoc_err.smajax = 111.0 * standard error longitude
hypoc_err.sminax = 111.0 * standard error latitude
hypoc_err.strike = 90.0;
The explosion effects flag ( evtype ) is set to follow IMS1.0 as shown below:
Flag evtype
D de
F fe
R kr
N kn
H kh
The default is uk for unknown.
FFB value Database tables and column
Record category NA
Next record category NA
Year NA
Month NA
Days NA
Hours NA
Minutes NA
Seconds NA
Agency number pub_comments.author
Prime-estimate flag NA
Comment pub_comments.pubcomment
netmag.magnitude
netmag.magtype
FFB value Database tables and column
Record category NA
Next record category NA
Year NA
Month NA
Serial number NA
Comment pub_comments.pubcomment
netmag.magnitude
netmag.magtype
FFB value Database tables and column
Record category NA
Next record category NA
Year phase.day
Month phase.day
Station code phase.sta
association.sta
Station number NA
Network code NA
Source code stamag.author
Format received code NA
Local - Teleseismic flag NA
Azimuth in degrees from epicentre to station association.esaz
Distance in degrees association.delta
Number of phases in this observation NA
Days phase.day
Hours phase.day
Minutes phase.day
Seconds phase.day
phase.msec
Precision of time phase.deltime
Operator's ID (numerical) NA
Operator's ID (characters) phase.phase
Operator's residual NA
ISC ID (numerical) ( as string ) association.phase
ISC residual association.timeres
Direction of first motion see below
Instrument type see below
Component phase.chan
Sharpness see below
Signal-to-noise ratio NA
Log A/T amplitude.logat
Precision of log A/T NA
Mantissa of amplitude amplitude.amp
Exponent of amplitude "
amplitude units amplitude.delamp
Period in seconds amplitude.per
Precision of period NA
Magnitude stamag.magnitude
column 5 of station code ( Format 15 only )
The direction of first motion is set as shown below:
phase.lp_fm = direction of first motion when instrument type is B, b, L or l and
phase.sp_fm = null
otherwise
phase.lp_fm = null and
phase.sp_fm = direction of first motion
Sharpness is used to set the following variables:
phase.emergent = sharpness if sharpness is e or E and
phase.impulsive = null
otherwise
phase.emergent = null and
phase.impulsive = sharpness if sharpness is I or I
phase.chan is set to '?' if component is null, and will always prefixed by '??' according to IMS1.0 format for S/H/I Channel Codes.
phase.deltime = 10^(-tim_prec) stamag.author is set to NEIS if source code is 'U' and JMA if 'J'. amplitude.delamp = 10^(-ampl_prec)
FFB value Database tables and column
Record category NA
Next record category NA
Year phase.day
Month phase.day
Phase count NA
Days phase.day
Hours phase.day
Minutes phase.day
Seconds phase.day
phase.msec
Precision of time phase.deltime
Operator's ID (numerical) NA
Operator's ID (characters) phase.phase
Operator's residual NA
ISC ID (numerical) ( as string ) association.phase
ISC residual association.timeres
Direction of first motion as format 5
Instrument type as format 5
Component phase.chan
Sharpness as format 5
Signal-to-noise ratio NA
Log A/T amplitude.logat
Precision of log A/T NA
Mantissa of amplitude amplitude.amp
Exponent of amplitude "
Precision of amplitude amplitude.delamp
Period in seconds amplitude.per
Precision of period NA
Magnitude stamag.magnitude
See format 5/15 comments above for any data manipulation.
FFB value Database tables and column Record category NA Next record category NA Year NA Month NA Count of comment records for this station NA Comment pub_comments.pubcomment
DATE HYPID MAGID AUTHOR MAGTYP MAG NSTA STAMAGS ------------------- ---------- ---------- ------ ------ ----- ---- ------- 13-05-1988 15:34:59 963721 538618 ISC mb 4.8 20 19 20-06-1987 00:53:04 1023335 564573 ISC mb 6.1 145 144 20-06-1987 00:53:04 1023335 564574 ISC MS 4.5 4 5 15-12-1975 01:08:46 1588615 855336 ISC mb 5.2 18 27 14-06-1975 19:14:02 1603944 862428 ISC mb 5.2 44 43 17-05-1973 16:00:02 1684562 900303 ISC mb 5.1 22 23 06-11-1971 21:59:56 1724241 917877 ISC mb 6.6 69 69 02-10-1969 22:06:01 1773113 938604 ISC mb 6.4 65 43 26-04-1968 15:00:01 1811308 956595 ISC mb 6.2 46 45 10-12-1967 19:30:01 1817728 959343 ISC mb 4.8 14 6 18-10-1967 14:30:02 1821427 961074 ISC mb 5.7 26 20 07-09-1967 13:45:03 1822407 961529 ISC mb 5 8 5 27-09-1967 17:00:02 1823617 962082 ISC mb 5.7 41 34 18-08-1967 20:12:30 1824740 962572 ISC mb 4.6 4 1 31-08-1967 16:30:00 1825487 962954 ISC mb 5 6 1 22-07-1967 10:59:03 1826732 963537 ISC mb 4.9 9 5 27-07-1967 13:00:02 1827125 963718 ISC mb 5 5 3 26-06-1967 16:00:02 1829144 964640 ISC mb 5.1 6 2 29-06-1967 11:25:03 1829289 964716 ISC mb 4.6 4 2 10-05-1967 13:40:03 1829979 965001 ISC mb 4.9 7 2 20-05-1967 15:00:01 1830619 965298 ISC mb 5.8 44 36 23-05-1967 14:00:02 1830825 965397 ISC mb 5.7 28 23 26-05-1967 15:00:05 1831008 965475 ISC mb 5.4 15 12 23-02-1967 18:34:00 1836348 968036 ISC mb 4.4 3 1 23-02-1967 18:50:02 1836350 968037 ISC mb 5.6 31 25 19-01-1967 16:45:02 1838051 968881 ISC mb 5.3 20 17 20-01-1967 17:40:05 1838111 968910 ISC mb 5.3 9 6 13-12-1966 21:00:02 1839298 969569 ISC mb 4.6 5 3 20-12-1966 15:30:01 1839613 969726 ISC mb 6.3 50 45 12-09-1966 15:29:59 1844201 971979 ISC mb 4.6 3 1 07-07-1966 09:59:57 1847771 973610 ISC mb 4.1 8 2 08-07-1966 08:29:56 1847808 973631 ISC mb 3.9 9 1 08-07-1966 09:29:56 1847810 973633 ISC mb 4.1 8 2 09-07-1966 08:29:56 1847847 973651 ISC mb 4 9 2 09-07-1966 09:29:56 1847849 973653 ISC mb 4.2 7 1 10-07-1966 08:29:56 1847895 973679 ISC mb 4.2 8 1 10-07-1966 09:29:56 1847897 973681 ISC mb 4 10 3 11-07-1966 08:29:56 1847947 973703 ISC mb 4 9 2 12-07-1966 08:29:55 1848015 973736 ISC mb 4.1 8 1 12-07-1966 09:29:56 1848017 973738 ISC mb 4.1 8 1 13-07-1966 08:29:57 1848061 973761 ISC mb 4.1 8 2 13-07-1966 09:29:56 1848068 973766 ISC mb 4 10 3 14-07-1966 08:29:57 1848113 973793 ISC mb 4 9 2 14-07-1966 09:29:56 1848115 973795 ISC mb 4.1 8 1 15-07-1966 08:29:56 1848160 973821 ISC mb 3.9 6 2 18-07-1966 08:59:56 1848297 973897 ISC mb 4.1 9 2 18-07-1966 09:59:57 1848301 973901 ISC mb 4 10 3 19-07-1966 09:29:57 1848355 973931 ISC mb 4 8 2 20-07-1966 08:29:57 1848400 973957 ISC mb 4.1 8 1 20-07-1966 09:29:57 1848404 973960 ISC mb 4 8 2 21-07-1966 08:29:56 1848480 974004 ISC mb 4 7 2 21-07-1966 09:29:56 1848487 974009 ISC mb 4.1 7 1 22-07-1966 08:29:57 1848529 974029 ISC mb 4.2 7 1 22-07-1966 09:29:57 1848531 974031 ISC mb 4.2 7 2 23-07-1966 08:29:56 1848572 974055 ISC mb 4.4 5 1 23-07-1966 09:29:56 1848576 974058 ISC mb 4.2 6 2 24-07-1966 08:29:57 1848639 974096 ISC mb 4.2 6 1 24-07-1966 09:29:57 1848647 974104 ISC mb 4.1 7 2 25-07-1966 08:29:57 1848688 974122 ISC mb 4.1 7 1 25-07-1966 09:29:57 1848694 974128 ISC mb 4 8 2 26-07-1966 09:29:56 1848736 974149 ISC mb 4 6 2 27-07-1966 08:29:56 1848779 974177 ISC mb 4.1 8 3 27-07-1966 09:29:56 1848781 974179 ISC mb 3.9 7 3 28-07-1966 08:29:57 1848850 974210 ISC mb 4.1 8 2 28-07-1966 09:29:57 1848852 974212 ISC mb 4 8 2 02-06-1966 15:30:01 1849133 974340 ISC mb 5.6 34 29 03-06-1966 14:00:02 1849184 974364 ISC mb 5.7 33 29 30-06-1966 22:15:02 1850989 975272 ISC mb 6.1 39 36 05-05-1966 14:00:03 1851243 975411 ISC mb 4.4 3 2 06-05-1966 15:00:03 1851331 975457 ISC mb 5.4 22 18 12-05-1966 19:37:26 1851698 975652 ISC mb 4.3 4 1 13-05-1966 13:30:02 1851737 975674 ISC mb 5.6 25 20 19-05-1966 13:56:30 1852039 975829 ISC mb 5.9 36 29 27-05-1966 20:00:03 1852568 976114 ISC mb 5 7 3 06-04-1966 13:57:19 1853086 976375 ISC mb 4.4 5 2 14-04-1966 14:13:46 1853538 976610 ISC mb 5.4 9 5 25-04-1966 18:38:03 1854143 976938 ISC mb 4.5 4 3 24-02-1966 15:55:10 1857425 978613 ISC mb 5 5 2 18-01-1966 18:35:03 1858391 979102 ISC mb 5.2 5 4 03-12-1965 15:13:04 1859299 979524 ISC mb 5.6 22 18 16-12-1965 19:15:02 1859745 979766 ISC mb 5.3 12 10 29-10-1965 21:00:03 1862331 981235 ISC mb 5.8 50 49 01-09-1965 20:07:59 1862422 981285 ISC mb 4.2 3 2 10-09-1965 17:12:02 1862732 981447 ISC mb 5.1 8 5 23-07-1965 17:00:02 1865706 983146 ISC mb 5.4 20 17 21-04-1965 22:00:03 1870186 985932 ISC mb 5 3 2 05-12-1964 21:15:03 1876895 990640 ISC mb 4.8 8 4 05-11-1964 15:00:01 1878413 991269 ISC mb 4.8 4 3 02-10-1964 20:03:00 1879669 991921 ISC mb 4 3 2 09-10-1964 14:00:02 1879962 992061 ISC mb 4.8 4 2 22-10-1964 15:59:59 1880664 992426 ISC mb 4.6 7 3 24-04-1964 20:10:01 1889936 997416 ISC mb 5.2 4 2 29-04-1964 20:47:00 1890158 997547 ISC mb 4.1 3 1 20-02-1964 15:30:02 1893060 999297 ISC mb 4.8 6 4 16-01-1964 16:00:02 1894058 999790 ISC mb 4.9 11 5 16-01-1964 18:08:41 1894065 999795 ISC mb 4.9 4 1 23-01-1964 16:00:01 1894377 999957 ISC mb 3.9 3 1
DATE HYPID MAGID AUTHOR MAGTYP MAG NSTA ------------------- ---------- ---------- ------ ------ ---- ---- 04-04-1991 18:59:59 727506 425568 ISC mb 5.7 75 13-04-1985 03:00:04 1163684 630356 ISC MS 6.8 34 09-03-1976 14:00:01 1579504 851503 ISC mb 5.8 78 14-02-1976 11:30:01 1582844 852761 ISC mb 5.8 56 20-11-1975 15:00:01 1591915 856853 ISC mb 5.9 76 28-10-1975 14:30:02 1594691 858070 ISC mb 6.2 66 19-06-1975 13:00:00 1604612 862742 ISC mb 5.9 74 14-05-1975 14:00:01 1607223 863900 ISC mb 5.9 64 28-02-1975 15:15:00 1616570 867998 ISC mb 5.6 58 30-08-1974 14:59:59 1632152 874525 ISC mb 5.6 52 14-10-1969 07:00:06 1773781 938922 ISC mb 6.3 62 16-09-1969 14:30:02 1775907 939954 ISC mb 6.1 49 07-05-1969 13:45:01 1783430 943857 ISC mb 5.5 43
1 21964 4241430 600-1 4B 291000-1 1304000-1 0 0 99 99 99 2 11964 4 99 99 900-1 600-2 1800-2 99 99 1 11964 4241430 720-1171C 293000-1 1301000-1 310 0 480 -1B 99 60-1 14 1 11964 42414301000 0 15D 297000-1 1299000-1 33099 99 99 99 1 21964 42414301190-1 1A 292500-2 1299600-2 710 0 460 -1B 4 99 238 20 29 130-2 29 2 51964 4 99 99 240-2 420-3 770-3 58-1 99 1 91 5 61964 4YKS 570 1 21 128 22414303300 0101P/PKP 9999 0 -15 99 99 99 6 51964 4 22414305100 0 35S -129999999 99 99 99 5 61964 4NGS 344 1 359 347 2241431 800 0101P/PKP 9999 0 32 e 99 99 99 . . . 5 51964 4ORV 366 1 47 8464 12414423840-1101P/PKP 9999 0 -5 i 99 99 99 5 51964 4FBC 139 1 8 8619 12414424500 0101P/PKP 9999 0 -10C e 99 99 99 5 11964 4UBO 528 1 41 9146 12414431300 0101P/PKP 9999 0 14 e 7-1 99 9946 1 11964 42414402830-1171B 133000-1 -888000-11580 0 510 -1B 039-2 100-1 44 1 11964 42414404200 0 19C 128700-2 -883500-2 800 0 99 99 99 1 21964 42414403020-1 1A 136500-2 -885800-21530 0 420 -1B 1 99 73 6 4 145-2 102 2 51964 4 99 99 150-2 410-3 440-3 41-1 99 14911 418 1 52 5 61964 4SSS 473 1 273 60 22414405050-1101P/PKP 9999 0 -23 i 99 99 99 6 51964 4 22414441100 0 35S 17959999999 99 99 99 5 51964 4NCS 340 1 308 74 12414405220-1101P/PKP 9999 0 -14 i 99 99 99 5 51964 4UBO 528 1 329 3232 12414431300 0101P/PKP 9999 09999 e 7-1 99 9942 5 11964 4FBC 139 1 11 5196 12414424500 0101P/PKP 9999 09999C e 99 99 99 1 11964 42414402830-1171B 133000-1 -888000-11580 0 510 -1B 039-2 100-1 44 1 11964 42414404200 0 19C 128700-2 -883500-2 800 0 99 99 99 1 21964 42414403020-1 1A 136500-2 -885800-21530 0 530 -1B 17 99 73 6 112 145-2 102 2 51964 4 99 99 150-2 410-3 440-3 41-1 99 14911 418 1163 5 61964 4SSS 473 1 273 60 22414405050-1101P/PKP 9999 0 -23 i 99 99 99 6 51964 4 22414441100 0 35S 17959999999 99 99 99 5 51964 4NCS 340 1 308 74 12414405220-1101P/PKP 9999 0 -14 i 99 99 99 5 61964 4LPS 287 1 319 85 22414405380-1101P/PKP 9999 0 -5 i 99 99 99 6 51964 4 22414412200 0 35S 849999999 99 99 99 5 51964 4BHP 51 1 117 1000 12414425100-1101P/PKP 9999 0 0 i 16-1 99 99 5 61964 4TAC 483 1 301 1167 52414431500-1101P/PKP 9999 0 20 i 99 99 99 . .
DATE msec Agency hypid remid 14-05-1977 03:39:23 100 LDG 1543028 8239 14-05-1977 03:39:23 100 LDG 1543029 8239 07-03-1977 09:41:39 500 SPGM 1547496 8240 07-03-1977 09:41:39 500 SPGM 1547497 8240 07-06-1980 18:35:02 800 LDG 1426142 8236 07-06-1980 18:35:02 800 LDG 1426143 8236 07-03-1981 00:51:36 700 TRN 1396067 8235 07-03-1981 00:51:36 700 TRN 1396068 8235 22-06-1981 10:21:01 700 PEK 1387317 8232 22-06-1981 10:21:01 700 PEK 1387318 8232 04-10-1981 00:01:28 800 MOS 1350451 8229 04-10-1981 00:01:28 800 MOS 1350452 8229 14-01-1982 11:36:05 300 PEK 1341371 8228 14-01-1982 11:36:05 300 PEK 1341372 8228 26-02-1983 03:01:25 400 MOS 1292029 8227 26-02-1983 03:01:25 400 MOS 1292030 8227 20-02-1983 10:49:50 300 MOS 1291206 8224 20-02-1983 10:49:50 300 MOS 1291207 8224 25-02-1973 20:04:28 0 WEL 1692982 8246 25-02-1973 20:04:28 0 WEL 1692983 8246 12-12-1970 10:03:00 0 TAP 1747049 8247 12-12-1970 10:03:00 0 TAP 1747050 8247 20-04-1967 01:42:42 0 BUL 1832550 8249 20-04-1967 01:42:42 0 BUL 1832551 8249 03-07-1981 17:14:50 200 PEK 1381758 8231 03-07-1981 17:14:50 200 PEK 1381759 8231 30-06-1981 21:55:50 600 MOS 1388342 8233 30-06-1981 21:55:50 600 MOS 1388343 8233 31-12-1967 04:52:14 0 NOU 1818834 8248 31-12-1967 04:52:14 0 NOU 1818835 8248 09-11-1974 10:12:57 900 PAS 1622242 10437 09-11-1974 10:12:57 900 PAS 1622243 10437