Processing profiles

A processing profile stores the settings of the CLUSTERING and REFINE panels, the database bar, the QC rule sets and the aggregation rules, so the same setup can be reused, for example one profile per network or region. Profiles are declared in the SeisComP application configuration (catalyst.cfg). Each one points to a .profile file that holds the actual key = value settings.

Declare profiles in catalyst.cfg

Put this in @SYSTEMCONFIGDIR@/catalyst.cfg ($SEISCOMP_ROOT/etc/catalyst.cfg) or in the user configuration @CONFIGDIR@/catalyst.cfg (~/.seiscomp/catalyst.cfg):

# Available processing profiles.
profiles = iceland, ukraine

# Profile applied at start-up. Default: the first one in the list.
defaultProfile = iceland

# Each profile points to a settings file. @SYSTEMCONFIGDIR@, @CONFIGDIR@,
# @DATADIR@ and @ROOTDIR@ are resolved by SeisComP.
profile.iceland.configFile = @SYSTEMCONFIGDIR@/iceland.profile
profile.ukraine.configFile = @SYSTEMCONFIGDIR@/ukraine.profile

At start-up Catalyst reads the profiles list, resolves each configFile path and fills the Profile dropdown in the toolbar. It then applies the profile named by defaultProfile. If that name is not in the list, Catalyst logs a warning and uses the first profile. The profile is applied at the end of start-up, after the locator and travel-time lists have been filled, so the dropdown never shows a profile whose settings are not in effect.

If no profiles are configured, the dropdown shows “(no profiles configured)” and the controls keep their built-in defaults.

Other module settings

Settings that belong to the installation rather than to a profile, such as minPhases, the score plugin or the 3D-view files, are set in catalyst.cfg only. See configuration.

The .profile files

A .profile file is a plain key = value list. Lines starting with # are comments. Shortened example:

# Catalyst processing profile - Iceland (IMO)
nucMethod      = dbscan
phase          = P+S
dbscanVelP     = 7
dbscanVelS     = 4
dbscanMaxDistP = 12
dbscanMaxDistS = 15
minSize        = 6
nucWindowDB    = 90
dupWindow      = 1
mergeDist      = 0.05
mergeTime      = 1
mergeMatchWin  = 0.2
mergeMinShared = 3
interface      = LOCSAT
locProfile     = sil
ttt            = LOCSAT
tttModel       = sil
assocTolerance = 2
maxDistance    = 100
qcZones        = ~/SC_configs/IMO/zones.geojson
cities         = "Reykjavik, Grindavík, Húsavík, Akureyri, Selfoss"

Rules for loading:

• A key that is missing keeps the control’s current value.

• Unknown keys are ignored. Keys from older versions, such as avgVelocity or maxDist, have no effect any more.

• Combo values (phase, interface, locProfile, ttt, tttModel) are applied only if the option exists in that dropdown on this installation. Otherwise the field is left unchanged, a warning is logged, and a dialog lists the settings that could not be applied.

• Booleans accept true or 1. Everything else counts as false.

• A leading ~/ in qcZones is expanded to the home folder.

Using profiles in the GUI

• The Profile dropdown in the toolbar lists the configured profiles.

• Load applies the selected profile. Nothing runs automatically, except that the QC zone file and the bearing catalogue are loaded if their paths are set and the files exist.

• Save as… writes the current settings to a .profile file. The dialog starts with the selected profile’s configFile, so pressing Enter overwrites it. If you don’t have write access to @SYSTEMCONFIGDIR@, save under ~/.seiscomp/ instead. When the target file already exists, keys the GUI does not manage (for example cities) are kept. After creating a new file, add profiles and profile.<name>.configFile entries for it to catalyst.cfg so it appears in the dropdown next time.

Settings keys

Nucleation

Key	Control
nucMethod	Nucleation tab: dbscan, backprojection, gamma or gridsearch
phase	DBSCAN: Phase (P, S, P+S or Is)
dbscanVelP / dbscanVelS / dbscanVelIs	DBSCAN: Vel P / Vel S / Vel Is [km/s]
dbscanMaxDistP / dbscanMaxDistS / dbscanMaxDistIs	DBSCAN: Max dist P / S / Is [s]
minSize	DBSCAN: Min size (picks)
maxDistPctl	DBSCAN: Dist pctl [%] used by Auto dist
pgMaxDist	DBSCAN: Pg max [km] for the apparent-velocity fit (0 = all picks)
nucWindowDB	DBSCAN: Window [s]
use3D	DBSCAN: Use 3D distances
bpGrid	Back projection: Grid [km]
bpDepth	Back projection: Depth [km]
bpMaxRange	Back projection: Max range [km]
minSizeBP	Back projection: Min size
bpPhaseP / bpPhaseS / bpPhaseIs	Back projection: phases used (several = joint search)
bpT0Tol	Back projection: t0 tol [s]
bpAzTol	Back projection: Az tol [deg]
bpMaxAzGap	Back projection: Max az gap [deg] (360 = off)
bpEDT	Back projection: EDT stacking
bpEdtSigma	Back projection: EDT σ [s]
nucWindowBP	Back projection: Window [s] (currently has no effect)
gammaGridKm	GaMMA: Grid [km]
gammaCoarseWin	GaMMA: Coarse window [s]
gammaOtTol	GaMMA: OT tol [s]
gammaWindowSec	GaMMA: Processing window [s]
gammaMinPts	GaMMA: Min size
gammaDepthKm	GaMMA: Depth [km]
gammaMaxRangeKm	GaMMA: Max range [km]
gammaMaxSpread	GaMMA: Max OT spread [s]
gsGridFile	Grid search: Grid file (scautoloc grid.conf)
gsStationFile	Grid search: Station file (scautoloc station.conf)
gsMinStations	Grid search: Min. stations
gsDmax	Grid search: Max. distance [deg]
gsDt0	Grid search: Agreement offset [s]
gsDefaultDt	Grid search: Projected window [s]
gsMaxRadiusFactor	Grid search: Max. radius factor
gsWindowSec	Grid search: Processing window [s]

Pick aggregation and filters

Key	Control
dupWindow	Aggregation: Dup. win [s]
aggMode	Aggregation: Mode (0 = Generate virtual pick, 1 = Keep preferred pick)
aggAcrossPhases	Aggregate picks of different phases together (no GUI control)
pickPrefRules	Aggregation: preferred-pick priority list (format below)
pickIsoMinStations	Isolated-pick filter: minimum stations (no GUI control)
pickIsoWindow	Isolated-pick filter: time window [s] (no GUI control)
pickIsoOverdetectingOnly	Isolated-pick filter: apply only to overdetecting stations (no GUI control)
pickIsTraceVelFilter	Pick filter: trace-velocity filter for Is picks on/off
pickIsTraceVelMin / pickIsTraceVelMax	Pick filter: trace-velocity range [km/s]

Relocation, association and merging

Key	Control
interface	Locator Interface
locProfile	Locator Profile
fixDepth	Fix depth on/off
fixDepthValue	Fix depth [km]
ignoreInitial	Ignore initial location
mcDraws	Monte Carlo rescue: MC draws
mcFraction	Monte Carlo rescue: min set (smallest number of picks in a subset)
ttt	Association: TTT interface
tttModel	Association: TTT model
assocTolerance	Association: Assoc. tol [s]
maxDistance	Association: Max distance [km] (0 = no limit)
mergeDist	Merge: Merge dist [km]
mergeTime	Merge: Merge time [s]
mergeMatchWin	Merge: Pick match [s]
mergeMinShared	Merge: Min shared picks
preferRules	Preferred-origin ranking used when merging (format below)

QC, references and display

Key	Control
qcProfiles	Named QC rule sets (format below)
qcZones	GeoJSON file with the QC zones, loaded with the profile
bearingCatalog	Slow/baz source catalogue (CSV), loaded with the profile
cities	Reference cities in the 3D view, comma-separated in quotes

Database bar

Key	Control
dbPicks / dbOrigins	picks / origins: load all picks / all origins in the window
dbPrefOrigins / dbPrefPicks	pref. origins / pref. picks: load only each event’s preferred origin / only the picks associated with it
dbAggregate	aggregate: aggregate the picks after every load
dbKeepPicks / dbKeepOrigins	keep picks / keep origins: add to what is loaded instead of replacing it
dbStart / dbEnd	Time window, ISO 8601 (for example 2025-11-01T00:00:00Z)

Rule formats

qcProfiles

Rule sets are joined with |. Each one is name=chain, and a chain is a ;-separated list of conditions:

qcProfiles = Gap zones=0,2,2,180,0;0,14,4,1,0|Weak=0,1,0,6,0;1,3,2,1,0

Each condition is conn,field,op,threshold,indent:

• conn: how the condition joins the one above. 0 = AND, 1 = OR. AND binds more tightly than OR. Ignored for the first condition.

• field:

  Code	Field	Code	Field
  0	Depth [km]	9	S cov (0 = good)
  1	Phase count	10	Fault Q (0 = good)
  2	Azimuthal gap [deg]	11	Min sta dist [km]
  3	RMS [s]	12	Mode (0 = automatic, 1 = manual)
  4	Mismatch	13	Status (0 = preliminary, 1 = confirmed, 2 = reviewed, 3 = final, 4 = rejected, 5 = reported)
  5	Move-out 0–1	14	Zone
  6	DD ref ties	15	Clean Q (0 = good)
  7	DD residual [s]	16	Score
  8	DD qual (0 = good)

• op: 0 <, 1 ≤, 2 >, 3 ≥, 4 =, 5 ≠

• threshold: number

• indent: grouping depth (0 = top level)

The custom (“(custom)”) query is scratch space and is not saved. It always starts empty when a profile is loaded. Loading a profile that contains qcProfiles replaces all named rule sets. The built-in defaults are added back if they are missing.

preferRules

Preferred-origin ranking, ;-separated, in priority order. Each rule is field,value. For numeric fields (the QC field codes above), value is 1 to prefer higher values or 0 to prefer lower ones. For Author (1000) and Method (1001), value is the author or method name to prefer.

preferRules = 12,1;3,0;1000,scautoloc

pickPrefRules

Preferred-pick priority list for “Keep preferred pick”, in the same field,value format:

Code	Field
2000	Manual (manual beats automatic)
2001	Status (more vetted wins)
2002	Time uncertainty [s]
2003	Time (earliest / latest)
2004	Author (value = author name)
2005	Method (value = methodID)

Not stored in profiles

These settings currently live only in the GUI session: the DBSCAN Min sta setting, the aggregation time mode, cell size and carried measurements, the swarm, template and fingerprint matching parameters, the map clustering settings, and the pick and origin filters except the trace-velocity filter.