1.- Requests #104 & #105 (D. Oppenheimer - PENDING): ==================================================== ----------------------------------------------------------------- --> Expand "event types" definitions. --> Add "ecategory" field. ----------------------------------------------------------------- a) After consulting with Peggy & Allan, the following list of event types has been approved: ------------------------------------------------------------------------ ETYPE NAME DESCRIPTION CATEGORY ===================================================================================== le local local earthquake earthquake re regional regional earthquake earthquake ts teleseism teleseismic earthquake earthquake se slow slow earthquake earthquake lp longperiod long period volcanic earthquake earthquake to tornillo tormillo wavelet earthquake tr nv-tremor non-volcanic tremor tremor vt v-tremor volcanic tremor tremor nt nuclear nuclear test explosion qb quarry quarry blast explosion ce calibr calibration explosion ex explosion generic chemical blast explosion sh shot refraction/reflection survey shot explosion sn sonic sonic shockwave atmospheric th thunder thunder atmospheric ve eruption volcanic eruption earth movement co m-collapse mine/tunnel collapse earth movement df debris debris flow/avalanche earth movement av snow-ice snow/ice avalanche earth movement ls landslide landslide earth movement rb rockburst rockburst earth movement rs rockslide rockslide earth movement bc bldg building collapse/demolition impact pc plane plane crash impact mi meteor meteor/comet impact impact st trigger subnet trigger unknown uk unknown unknown type unknown ot other other miscellaneous unknown ------------------------------------------------------------------------ ------------------------------------------------------------------------ ==> Approved. Ellen to come up with a DDL script for the constraint on Event.etype. ------------------------------------------------------------------------ b) Ellen came up with a proposal for the event type/category tables: ------------------------------------------------------------------------ Here is a proposed model for storing categories of event types. EventType: ------------------------------------------------- ETYPE NOT NULL VARCHAR2(2) like 'le' NAME or LABEL NOT NULL VARCHAR2(16) like 'local' DESCRIPTION VARCHAR2(80) like 'local earthquake inside net' Primary key is EType, in Event type there is a foreign key constraint to this key. Category: ----------------------------------------------------- CATID NOT NULL NUMBER - Unique numerical identifier CATEGORY NOT NULL VARCHAR2(25) - Category name DESCRIPTION VARCHAR2(80) Primary key is CatID, unique constraint on Category AssocEvCat: ------------------------------- CATID (foreign key to Category) ETYPE (foreign key to EventType) Primary key is CatId and Etype ------------------------------------------------------------------------ ------------------------------------------------------------------------ ==> Approved. The tables should be named: EventType, EventCategory & AssocTypeCat. Ellen to come up with DML & DDL scripts for those tables. ------------------------------------------------------------------------ c) The fixed parameters that Allan requested should go into a different table (AP Schema): ------------------------------------------------------------------------ I would like to request in addition to a catalog "name" or "label" string tag column one or more columns related to the fixing of depth, like "fixDepth = 0 or 1 and if =1 "trialDepth=0.0" ETYPE NOT NULL VARCHAR2(7) like 'le' NAME or LABEL NOT NULL VARCHAR2(16) like 'local' CATEGORY NOT NULL VARCHAR2(25) like 'earthquake' DESCRIPTION VARCHAR2(80) like 'local earthquake inside net' FIXDEPTH NOT NULL VARCHAR2(1) like '1' TRIALDEPTH NUMBER(5,2) like 0.1 ------------------------------------------------------------------------ ------------------------------------------------------------------------ ==> Allan agreed to store those parameters in the jiggle properties file. ------------------------------------------------------------------------ 2.- Request #120 (D. Given - PENDING): ====================================== ----------------------------------------------------------------- --> Add new CubeMessageText table to application schema. Cube message codes need site-specific info to compose event emails. This is used by the Java HeaderMessage and CubeFormat classes. ----------------------------------------------------------------- ------------------------------------------------------------------------ Allan emailed the table structure to the group: CUBEMESSAGETEXT Name Null? Type ----------------------- -------- ---------------- SRCNET NOT NULL VARCHAR2(2) TEXTTYPE NOT NULL VARCHAR2(40) LINENO NOT NULL NUMBER(8) TEXT NOT NULL VARCHAR2(132) Nobody seems to object about this table except for its name. What about renaming it to MESSAGETEXT? ------------------------------------------------------------------------ ------------------------------------------------------------------------ ==> Approved. The table will be renamed MESSAGETEXT. ------------------------------------------------------------------------ 3.- Request #119 (E. Yu - PENDING): =================================== ----------------------------------------------------------------- --> Remove amp02 constraint (amplitude > 0) from amp table. Paul Friberg has asked this constraint be disabled on HVO servers. I am wondering if this should apply to SCEDC/NCEDC as well. According to Paul, the meaning of a negative amplitude is fabs() was not run on it. ----------------------------------------------------------------- ------------------------------------------------------------------------ I spoke with Paul Friberg and he said the application that does not do an fabs() call is localmag. The others (RAD, TriMag?) do make such a call. I am not sure if we use that here at the SCSN. (I'ved cc'ed Paul for confirmation) He also said is it possible that future networks adopting CISN software may have their own amp programs that do not call fabs() and therefore insert negative amplitudes. I will next find out what uses amplitudes in our system. ------------------------------------------------------------------------ ------------------------------------------------------------------------ ==> Discussed. Pete said: "I recommend against changing the amplitude constraint: amplitudes as used in seismology and most other fields of physics are non-negative quantities. If localmag neglects to take the absolute value of some "peak" quanitity, it is a bug. And since I wrote localmag, I feel somewhat qualified to say this." SCWG wants to have Standards group opinion. ------------------------------------------------------------------------ 4.- Request #10 (D. Given - PENDING): ===================================== ----------------------------------------------------------------- --> Add new column "nread" to NETMAG table. ----------------------------------------------------------------- ------------------------------------------------------------------------ Every magnitude in the dbase has an attribute called "NSTA". Most programs write the number of *observations* (amps, codas) contributing to the magnitude in this field. One program (TriMag) writes the number of *stations*. So the field name doesn't match its most common usage. This will be confusing to future developers. The WG considered just changing TriMag to write # of observations and leaving the misleading column name (NSTA) as it is. This may cause difficulties as the software propagates out to other networks. The WG is now proposing to "fix" the situation by doing the following: * add a new field called NOBS for # observations * retroactively rewrite NSTA and populate NOBS in the dbase to reflect this change * change all codes to use the new NOBS field and interpret NSTA as # stations This change would require changes to several codes. For example, TriMag, data import and load programs, Jiggle, STP, dbselect, EWmag2CISN, magPref package, alarm rules, and more. QUESTION: Do you think it is worth the effort to change the code to fix the 'misnamed' NSTA field and record both the number of observations and the number of stations? ------------------------------------------------------------------------ 5.- Request #19 (D. Given - PENDING): ===================================== ----------------------------------------------------------------- --> Add "timequality" to WAVEFORM. Need to store waveform time quality (Currently must retrieve SEED timeseries data and read headers to get this info). ----------------------------------------------------------------- ------------------------------------------------------------------------ Doug G. to go back to Egill for more information. ------------------------------------------------------------------------