Primary magnitudes¶
Primary magnitudes are computed from amplitudes and station-event distances. Currently the following primary magnitude types are implemented.
Calculates magnitudes of different types.
The purpose of scmag is to compute magnitudes from pre-computed amplitudes. Instead it takes amplitudes and origins as input and produces StationMagnitudes and (network) Magnitudes as output. It does not access waveforms. The resulting magnitudes are sent to the “MAGNITUDE” group. scmag doesn’t access any waveforms. It only uses amplitudes previously calculated.
The purpose of scmag is the decoupling of magnitude computation from amplitude measurements. This allows several modules to generate amplitudes concurrently, like scautopick or scamp. As soon as an origin comes in, the amplitudes related to the picks are taken either from the memory buffer or the database to compute the magnitudes.
scmag makes use of the fact that origins sent by scautoloc, scolv or other modules include the complete set of arrivals, which reference picks used for origin computation. The picks in turn are referenced by a number of amplitudes, some of which are relevant for magnitude computation.
Read the scamp documentation for more details on amplitude measurements.
Primary magnitudes are computed from amplitudes and station-event distances. Currently the following primary magnitude types are implemented.
Duration magnitude as described in HYPOINVERSE (Klein [51]).
Mjma is computed on displacement data using body waves of period < 30s.
Local (Richter) magnitude calculated on the horizontal components using a correction term to fit with the standard ML (Richter [58]).
Local custom magnitude calculated on the horizontal components according to Hessian Earthquake Service and Stange [63]
Local magnitude calculated on the horizontal components according to SED specifications.
Local magnitude calculated on the vertical component using a correction term to fit with the standard ML.
Local magnitude calculated from MLv amplitudes based on GNS/GEONET specifications for New Zealand (Ristau et al. [59]).
Nuttli magnitude for Canada and other Cratonic regions (Nuttli [54]).
Additionally, scmag derives the following magnitudes from primary magnitudes:
Estimation of the moment magnitude Mw based on mB using the Mw vs. mB regression of Bormann and Saul [39]
Estimation of the moment magnitude Mw based on Mwp using the Mw vs. Mwp regression of Whitmore et al. [66]
Summary magnitude, which consists of a weighted average of the individual magnitudes and attempts to be a best possible compromise between all magnitudes. See below for configuration and also scevent for how to add the summary magnitude to the list of possible preferred magnitudes or how to make it always preferred.
More details are given in the section Summary magnitude.
Estimation of the moment magnitude Mw based on a weighted average of other magnitudes, currently MLv, mb and Mw(mB), in future possibly other magnitudes as well, especially those suitable for very large events. The purpose of Mw(avg) is to have, at any stage during the processing, a “best possible” estimation of the magnitude by combining all available magnitudes into a single, weighted average. Initially the average will consist of only MLv and/or mb measurements, but as soon as Mw(mB) measurements become available, these (and in future other large-event magnitudes) become progressively more weight in the average.
If an amplitude is updated, the corresponding magnitude is updated as well. This allows the computation of preliminary, real-time magnitudes even before the full length of the P coda is available.
Station magnitudes of a particular magnitude type are calculated based on measured amplitudes considered by this magnitude type and the distance between the origin and the station at which the amplitude was measured. Typically, epicentral distance is used for distance. Magnitudes may support configurable distance measures, e.g., MLc. The relation between measured amplitudes, distance and station magnitude is given by a calibration function which is specific to a magnitude type and configurable for some magnitudes.
Note
Usually station magnitudes use amplitudes of the same type. However, some magnitude consider amplitudes of another type. E.g. MLr uses amplitudes computed for MLv.
Depending on the geographic region in which events, stations or entire ray paths
are located, different calibration functions and constraints may apply. This is
called “magnitude regionalization”. The region is defined by a polygon stored in
a region file. For a particular magnitude, regionalization can be configured by
global parameters, e.g., in $SEISCOMP_ROOT/etc/global.cfg
.
Add magnitude type profile to the magnitudes parameters. The name of the profile must be the name of the magnitude type.
Add the profile-specific parameters.
Example for MLc in $SEISCOMP_ROOT/etc/global.cfg
the polygon with name
test defined in a BNA file:
magnitudes.MLc.regionFile = @DATADIR@/spatial/vector/magnitudes/regions.bna
magnitudes.MLc.region.test.enable = true
magnitudes.MLc.region.test.A0.logA0 = 0:-1.3, 60:-2.8, 100:-3.0, 400:-4.5, 1000:-5.85
The network magnitude is a magnitude value summarizing several station magnitudes values of one origin. Different methods are available for forming network magnitudes from station magnitudes:
Method |
Description |
---|---|
mean |
The usual mean value. |
trimmed mean value |
To stabilize the network magnitudes the smallest and the largest 12.5% of the station magnitude values are removed before computing the mean. |
median |
The usual median value. |
median trimmed mean |
Removing all station magnitudes with a distance greater than 0.5 (default) from the median of all station magnitudes and computing the mean of all remaining station magnitudes. |
Configure the method per magnitude type by magnitudes.average
.
Default values apply for each magnitude type which are defined by the magnitude
itself.
In the scolv Magnitudes tab the methods, the
stations magnitudes and other parameters can be selected interactively.
scmag can compute a summary magnitude as a weighted sum from all available
network magnitudes.
This magnitude is typically called M as configured in
summaryMagnitude.type
.
It is computed as a weighted average over the available magnitudes:
The coefficients a and b can be configured per magnitude type by
summaryMagnitude.coefficients.a
and summaryMagnitude.coefficients.b
, respectively.
Furthermore each magnitude type can be specifically added to or excluded from the
summary magnitude calculation
as defined in summaryMagnitude.whitelist
or
summaryMagnitude.blacklist
, respectively.
The preferred magnitude of an event is set automatically by scevent or interactively in scolv. It can be any network magnitude or the summary magnitude.
etc/defaults/global.cfg
etc/defaults/scmag.cfg
etc/global.cfg
etc/scmag.cfg
~/.seiscomp/global.cfg
~/.seiscomp/scmag.cfg
scmag inherits global options.
Default: MLv,mb,mB,Mwp
Type: list:string
The magnitude types to be calculated. Station magnitudes are computed from their amplitudes, network magnitudes from their station magnitudes.
Default: 0.5
Type: double
The minimum weight of an arrival for an associated amplitude to be used for calculating a magnitude.
Note
magnitudes.* General parameters for computing magnitudes. Others are configured by global binding parameters for specific magnitude types.
Default: default
Type: list:string
The methods for computing the network magnitude from station magnitudes. Exactly one method per magnitude can be configured. To define the averaging method per magnitude type append the type after colon, e.g.: "magnitudes.average = default, MLv:median"
default: Compute the mean if less than 4 contributed station magnitudes exist. Otherwise apply trimmedMean(25), trimmed mean with 25%.
Default: 1
Type: int
Unit: s
Interval between 2 sending processes. The interval controls how often information is updated.
Note
summaryMagnitude.* The summary magnitude is the weighted average from all defined network magnitude types: Single network magnitude values are multiplied with their magnitude-type specific weight and summed up. The resulting sum is divided by the sum of all weights.
Default: true
Type: boolean
Enables summary magnitude calculation.
Default: M
Type: string
Define the type/name of the summary magnitude.
Default: 1
Type: int
This is the minimum station magnitude required for any magnitude to contribute to the summary magnitude at all. If this is set to 4, then no magnitude with less than 4 station magnitudes is taken into consideration even if this results in no summary magnitude at all. For this reason, the default here is 1 but in a purely automatic system it should be higher, at least 4 is recommended.
Default: true
Type: boolean
Allow computing the summary magnitude even if only one single network magnitude meeting the other criteria is available. Unselecting this parameter will suppress computing summary magnitudes if only one network magnitude is available.
Type: list:string
Define the magnitude types to be excluded from the summary magnitude calculation.
Type: list:string
Define the magnitude types to be included in the summary magnitude calculation.
Note
summaryMagnitude.coefficients.* The coefficients defining the weight of network magnitudes for calculating the summary magnitude. Weight = a * magnitudeStationCount + b.
Default: 0, Mw(mB):0.4, Mw(Mwp):0.4
Type: list:string
Define the coefficients a. To define the value per magnitude type append the type after colon. A value without a type defines the default value.
Default: 1, MLv:2, Mw(mB):-1, Mw(Mwp):-1
Type: list:string
Define the coefficients b. To define the value per magnitude type append the type after colon. A value without a type defines the default value.
Show help message.
Show version information.
Use alternative configuration file. When this option is used the loading of all stages is disabled. Only the given configuration file is parsed and used. To use another name for the configuration create a symbolic link of the application or copy it. Example: scautopick -> scautopick2.
Load given plugins.
Run as daemon. This means the application will fork itself and doesn’t need to be started with &.
Enable/disable self-shutdown because a master module shutdown. This only works when messaging is enabled and the master module sends a shutdown message (enabled with --start-stop-msg for the master module).
Set the name of the master-module used for auto-shutdown. This is the application name of the module actually started. If symlinks are used, then it is the name of the symlinked application.
Set the name of the master-username of the messaging used for auto-shutdown. If "shutdown-master-module" is given as well, this parameter is ignored.
Time span in hours after which objects expire.
Verbosity level [0..4]. 0:quiet, 1:error, 2:warning, 3:info, 4:debug.
Increase verbosity level (may be repeated, eg. -vv).
Quiet mode: no logging output.
Limit the logging to a certain component. This option can be given more than once.
Use syslog logging backend. The output usually goes to /var/lib/messages.
Path to lock file.
Send log output to stdout.
Execute in debug mode. Equivalent to --verbosity=4 --console=1 .
Use alternative log file.
Overrides configuration parameter connection.username
.
Overrides configuration parameter connection.server
.
Overrides configuration parameter connection.timeout
.
Overrides configuration parameter connection.primaryGroup
.
A group to subscribe to. This option can be given more than once.
Overrides configuration parameter connection.contentType
.
Set sending of a start and a stop message.
List all supported database drivers.
The database connection string, format: service://user:pwd@host/database. "service" is the name of the database driver which can be queried with "--db-driver-list".
The config module to use.
Load the inventory from the given database or file, format: [service://]location .
Do not use the database at all
Defines an event parameters XML file to be read and processed. This implies offline mode and only processes all origins contained in that file. It computes station magnitudes for all picks associated with an origin where amplitudes are available and the corresponding network magnitudes. Station and network magnitudes having the evaluation status set are ignored. Use --reprocess to include those magnitudes. It outputs an XML text adding the station- and network magnitudes to the input XML file.
Reprocess also station and network magnitudes with an evaluation status set but do not change original weights. New contributions are added with weight 0.
With that flag all existing station magnitudes are recomputed based on their associated amplitudes. If an amplitude cannot be accessed, no station magnitude is updated. Network magnitudes are recomputed based on their station magnitude contributions. No new objects will be created in this mode, it only updates values and weights. The method to accumulate the station magnitudes to form the network magnitude will be read from the existing object and replicated. If it cannot be interpreted, then the configured default for this magnitude type will be used instead. Weights of station magnitudes will be changed according to the accumulation method of the network magnitude.
Keep the original weights in combination with --static.