gps class

class csi.gps(name, utmzone=None, ellps='WGS84', lon0=None, lat0=None, verbose=True)

A class that handles a network of gps displacements

Args:

• name : Name of the dataset.

Kwargs:

• utmzone : UTM zone (optional, default=None)

• lon0 : Longitude of the center of the UTM zone

• lat0 : Latitude of the center of the UTM zone

• ellps : ellipsoid (optional, default=’WGS84’)

• verbose : Speak to me (default=True)

addstation(station, lon, lat, vel, err, synth=None, los=None)

Add a station to a network.

Args:

• station : name of the station

• lon : Longitude

• lat : Latitude

• vel : velocity (3 numbers)

• err : uncertainty (3 numbers)

Kwargs:

• synth : Synthetics (3 numbers)

• los : Line-of-sight projection (1 number)

Returns:

• None

buildCd(direction='en')

Builds a diagonal data covariance matrix using the formal uncertainties in the GPS data.

Kwargs:

• direction : Direction to take into account. Can be any combination of e, n and u.

Returns:

• None

buildsynth(faults, direction='sd', poly=None, vertical=True, custom=False)

Takes the slip model in each of the faults and builds the synthetic displacement using the Green’s functions.

Args:

• faults : list of faults to include.

Kwargs:

• direction : list of directions to use. Can be any combination of ‘s’, ‘d’ and ‘t’.

• vertical : True/False

• include_poly : if a polynomial function has been estimated, include it.

• custom : if some custom green’s function was used, include it.

Returns:

• None

combineNetworks(gpsdata, newNetworkName='Combined Network', mergeType=None)

Combine networks into a new network.

Args:

• gpsdata : List of gps instances.

Kwargs:

• newNetworkName : Name of the returned network

• mergeTypeStyle of merging

  None: duplicates stations sum: if two identical stations, sum the displacements

Returns:

• None

compute2Dstrain(fault, write2file=False, verbose=False)

Computes the 2D strain tensor stored in the fault given as an argument.

Args:

• fault : Instance of a fault class

Kwargs:

• write2file : Write to a file

• verbose : talk to me

Returns:

• None

computeBestHelmert(components=2, data='data')

Fits a full Helmert transform to the network

Kwargs:

• components : Take the 2 horizontal (default) or 3 enu

• data : Can be ‘data’, ‘synth’, ‘res’, or ‘transofmration’

Returns:

• None

computeCustom(fault)

Computes the displacements associated with the custom green’s functions.

Args:

• fault : A fault instance.

computeHelmertTransform(fault, verbose=False)

Removes the Helmert Transform stored in the fault given as argument.

Args:

• fault : Instance of a fault class

Kwargs:

• verbose: talk to me

Returns:

• None

computeTransformation(fault, verbose=False, custom=False)

Computes the transformation that is stored with a particular fault. Stores it in transformation.

Args:

• fault : An instance of a fault class

Kwargs:

• verbose : talk to me

• custom : Do we have custom green’s functions

Returns:

• None

compute_rotation(elon, elat, omega)

Removes a rotation from the lon, lat and velocity of a rotation pole.

Args:

• elon : Longitude of the rotation pole

• elat : Latitude of the rotation pole

• omega : Amplitude of the rotation (in rad/yr).

Returns:

• None

createNetwork(lonc, latc, expand, nstation, prefix='S')

Builds a GPS network with randomly dropped stations within the box. Box is specified by the center (lonc, latc) and its half-size in degrees (expand). {nstation} gives the number of stations randomly chosen.

Args:

• lonc : Longitude of the center of the box

• latc : Latitude of the center of the box

• expand : Half-size in degrees of the box

• nstation : Number of stations in the box

• prefix : Station name prefix (number of the station is added)

deletestation(station)

Removes a station from the network

Args:

• station : Name of the station

Returns:

• None

distance2fault(faults)

Computes the distance between the GPS stations and a list of faults.

Args:

• faults : list of faults

Return:

• d : distance

extractTimeSeriesOffsets(date1, date2, destination='data')

Puts the offset from the time series between date1 and date2 into an instance of self.

Args:

• date1 : datetime object.

• date2 : datetime object.

Kwargs:

• destination : if ‘data’, results are in vel_enu, if ‘synth’, results are in synth

get2DstrainEst(strain=True, rotation=True, translation=True, computeNormFact=True)

Returns the matrix to estimate the full 2d strain tensor. Positive is clockwise for the rotation. Only works with 2D.

When building the estimator:

• First column is translation along the x-axis

• Second column is translation along the y-axis

• Third column is the Epsilon_xx component

• Fourth column is the Epsilon_xy component

• Fifth column is the Epsilon_yy component

• Sixth column is the Rotation term

Kwargs:

• strain: True/False

• rotation: True/False

• translation: True/False

• computeNorFact: Recompute normalizatin factor.

Returns:

• 2D array

getHelmertMatrix(components=2, meanbase=None, center=None)

Returns a Helmert matrix for a gps data set.

Kwargs:

• components: How many components (can be 2 or 3)

• meanbase: float for baseline length normalization

• center: tuple of float for the center o the network

Returns:

• 2d array

getMisfit()

Computes the summed misfit of the residuals

Returns:

getNetworkAtDate(date, verbose=True)

Returns a GPS object which displacements are the values at the desired date.

Args:

• date : datetime.datetime instance

Return:

• gps : a GPS instance

getNumberOfTransformParameters(transformation)

Returns the number of transform parameters for the given transformation. Strain is only computed as an aerial strain (2D). If verticals are included, it just estimates a vertical translation for the network.

Args:

• transformation : String. Can be ‘strain’, ‘full’, ‘strainnorotation’, ‘strainnotranslation’, ‘strainonly’

Returns:

• Integer

getRMS(vertical=False)

Computes the RMS of the data and if synthetics are computed, the RMS of the residuals

Returns:

• dataRMS, synthRMS: 2 floats

getSubNetwork(name, stations)

Given a list of station names, returns the corresponding gps object.

Args:

• name : Name of the returned gps object

• stations : List of station names.

Returns:

• gps : Instance of the gps class

getTransformEstimator(transformation, computeNormFact=True)

Returns the estimator for the transform.

Args:

• transformation : String. Can be ‘strain’, ‘full’, ‘strainnorotation’, ‘strainnotranslation’, ‘strainonly’, ‘translation’ or ‘translationrotation’

Kwargs:

• computeNormFact: compute and store the normalizing factor

Returns:

• 2d array

getVariance()

Computes the Variance of the data and if synthetics are computed, the RMS of the residuals

Returns:

• dataVariance, synthVariance: 2 floats

geterr(station)

Gets the errors enu for the station.

Args:

• station : name of the station.

Returns:

• vector : 3D vector of uncertainties

getprofile(name, loncenter, latcenter, length, azimuth, width, data='data')

Project the GPS velocities onto a profile. Works on the lat/lon coordinates system.

Args:

• name : Name of the profile.

• loncenter : Profile origin along longitude.

• latcenter : Profile origin along latitude.

• length : Length of profile.

• azimuth : Azimuth in degrees.

• width : Width of the profile.

Kwargs:

• data : Do the profile through the ‘data’ or the ‘synth’etics.

Returns:

• None: Profiles are stored in self.profiles

getstation(station)

Gets informations for a station.

Args:

• station : name of the station

Returns:

• lon, lat, vel, err, synth, los

getvelo(station, data='data')

Gets the velocities enu for the station.

Args:

• station : name of the station.

Kwargs:

• data : which velocity do you want (‘data’ or ‘synth’)

Returns:

• vel : 3D vector

importNetwork(gpsdata, iftwo='keep')

Adds stations from gpsdata to the current network. If station is already in here, there is several options:

• if iftwo == ‘keep’: Keep both measures

• if iftwo == gpsdata.name: Keep the incomcing measure

• if iftwo == self.name: Keep the current one

Args:

• gpsdata : A gps instance

Kwargs:

• iftwo : same station policy

Returns:

• None

initializeTimeSeries(start=None, end=None, stationfile=False, sqlfile=None, time=None, interval=1, verbose=False, los=False, factor=1.0)

Initializes a time series for all the stations.

Kwargs:

• start : Starting date

• end : Ending date

• interval : in days (default=1).

• stationfile : Read the time series from the station file

• sqlfile : Red the time series from a sqlfile

• time : time would be taken from this array

• verbose : talk to me

• los : Los vector

• factor : scaling factor

Returns:

• None

intersectProfileFault(name, fault)

Gets the distance between the fault/profile intersection and the profile center.

Args:

• name : name of the profile.

• fault : fault object from verticalfault.

Returns:

• distance : float

keep_stations(stations)

Keeps only the stations on the arg list.

Args:

• stations : list of stations to keep.

Returns:

• None

lonlat2xy()

Pass the position of the stations into the utm coordinate system.

makeDelaunay(plot=False)

Builds a Delaunay triangulation of the GPS network.

Kwargs:

• plot : True/False(default).

Returns:

• None

plot(faults=None, figure=135, name=False, legendscale=10.0, scale=None, legendunit='', figsize=None, plot_los=False, drawCoastlines=True, expand=0.2, show=True, error=True, title=True, colorbar=True, cbaxis=[0.1, 0.2, 0.1, 0.02], cborientation='horizontal', cblabel='', landcolor='lightgrey', seacolor=None, shadedtopo=None, Map=True, Fault=True, zorder=None, vertical=False, verticalsize=[30], verticalnorm=None, box=None, width=0.005, headwidth=3, headlength=5, headaxislength=4.5, minshaft=1, minlength=1, data=['data'], color=['k'], titleyoffset=1.1, alpha=1.0)

Plot the network

Kwargs:

• faults : list of instances of faults

• data : list of data to plot (can be ‘data’, ‘synth’, ‘res’ or ‘transformation’)

• vertical : plot verticals (True/False)

• verticalsize : size of the dots for vertical plots (list as long as data)

• color : lits of color specifications as long as data

• name : plot the name of the stations

• legendscale : size of the legend (default is 10)

• scale : scale of the arrows

• ref : can be ‘utm’ or ‘lonlat’

• drawCoastlines: True/False

• expand : Expand the map (in degrees)

• show : plot to screen

• figure : number of the figure.

• faults : List of fault objects to plot the surface trace of a fault object (see verticalfault.py).

• plot_los : Plot the los projected gps as scatter points

• box : Lon/lat box [lonmin, lonmax, latmin, latmax]

Returns:

• None

plotprofile(name, legendscale=10.0, fault=None, data=['parallel', 'normal', 'vertical'], show=True, figsize=None, colorbar=False, cbaxis=[0.1, 0.2, 0.1, 0.02], cborientation='horizontal', cblabel='')

Plot profile.

Args:

• name : Name of the profile.

Kwargs:

• legendscale : Length of the legend arrow.

• fault : Add a fault on the plot

• data : list of type of data to use

• show : Show me

Returns:

• None

project2InSAR(los=None, incidence=None, heading=None)

Projects the GPS data into the InSAR Line-Of-Sight provided.

Args:

• los : list of three components of the line-of-sight vector.

• incidence : incidence angle (single float)

• heading : heading (single float)

Returns:

• None

readStat(station_file, loc_format='LL')

Read simple station ascii file and populate station attributes

If loc_format=’XY’, then the file should be as

STNAME	X_COORD	Y_COORD

If loc_format=’LL’, then the file should be as

STNAME	LON	LAT

Args:

• station_file: station filename including station coordinates

Kwargs:

• loc_format: station file format (default= ‘LL’)

Returns:

• None

read_from_ICM(velfile, factor=1.0, header=1)

Reading velocities from an ICM (F. Ortega’s format) file. Maybe obsolete now.

Args:

• velfile : Input file

Kwargs:

• factor : multiplication factor for velocities

• header : length of the file header

Returns:

• None

read_from_en(velfile, factor=1.0, minerr=1.0, header=0, error='std')

Reading velocities from a en file formatted as:

StationName	Lon	Lat	e_vel	n_vel	e_err	n_err

Args:

• velfile : File containing the velocities.

Kwargs:

• factor : multiplication factor for velocities

• minerr : if err=0, then err=minerr.

• header : length of the file header

• error : if ‘variance’, then np.sqrt(err) is taken. Default is ‘std’

Returns:

• None

read_from_enu(velfile, factor=1.0, minerr=1.0, header=0, checkNaNs=True)

Reading velocities from a enu file formatted as

StationName	Lon	Lat	e_vel	n_vel	u_vel	e_err	n_err	u_err

Args:

• velfile : Input file

Kwargs:

• factor : multiplication factor for velocities

• minerr : if err=0, then err=minerr.

• header : length of the header

• checkNaNs : If True, kicks out stations with NaNs

Returns:

• None

read_from_midas(velfile, stationfile, factor=1.0, minerr=1.0, header=0, checkNaNs=True)

Reading velocities from a file produced by the midas tool. Details are available here: http://geodesy.unr.edu/

Args:

• velfile : Input file

• stationfile : File of the list of stations

Kwargs:

• factor : multiplication factor for velocities

• minerr : if err=0, then err=minerr.

• header : length of the header

• checkNaNs : If True, kicks out stations with NaNs

Returns:

• None

read_from_sopac(velfile, coordfile, factor=1.0, minerr=1.0)

Reading velocities from Sopac file and converting to mm/yr. Format is as sopac was providing it in 2013.

Args:

• velfile : File containing the velocities.

• coordfile : File containing the coordinates.

Kwargs:

• factor : Scaling factor

• minerr : If err is lower than minerr, err is set to minerr

Returns:

• None

read_from_unavco(velfile, factor=1.0, minerr=1.0, header=37)

Reading velocities from a unavco file. This follows the unavco format as it was in 2013.

Args:

• velfile : Input file

Kwargs:

• factor : multiplication factor for velocities

• header : length of the file header

• minerr : If the error is lower than minerr, set it to minerr

Returns:

• None

reference(station, refSynth=False)

References the velocities to a single station.

Args:

• station : name of the station or list of station names.

• refSynth : Apply referencing to synthetics as well (default=False)

Returns:

• None

reference2network(network, components=2)

Removes a Helmert transform that best references the velocity field from self to that of network.

Args:

• network : gps instance

• components: Number of components to use

Returns:

• None

reject_stations(station)

Reject the stations named in stations.

Args:

• station : name or list of names of station.

Returns:

• None

reject_stations_awayfault(dis, faults)

Rejects the pixels that are {dis} km away from the faults

Args:

• dis : Threshold distance.

• faults : list of fault objects.

Returns:

• None

reject_stations_fault(dis, faults)

Rejects the pixels that are dis km close to the fault.

Args:

• dis : Threshold distance.

• faults : list of fault objects.

Returns:

• None

remove2Dstrain(fault)

Computess the 2D strain and removes it.

Args:

• fault : Instance of a fault class

Returns:

• None

removeBestHelmert(components=2, data='data')

Fits a Helmert transform to the network and removes it.

Kwargs:

• components : Take the 2 horizontal (default) or 3 enu

• data : Can be ‘data’, ‘synth’, ‘res’, or ‘transofmration’

Returns:

• None

removeHelmertTransform(fault)

Computes the Helmert and removes it.

Args:

• fault : Instance of a fault class

Returns:

• None

removeSynth(faults, direction='sd', poly=None, custom=False)

Removes the synthetics from a slip model.

Args:

• faults : list of faults to include.

Kwargs:

• direction : list of directions to use. Can be any combination of ‘s’, ‘d’ and ‘t’.

• poly : if a polynomial function has been estimated, include it.

• custom : if some custom green’s function was used, include it.

Returns:

• None

removeTransformation(fault, custom=False)

Removes the transformation that is stored in a fault.

remove_euler_rotation(eradius=6378137.0, stations=None, verbose=True)

Removes the best fit Euler rotation from the network.

Kwargs:

• eradius : Radius of the earth (should not change that much :-)).

• stations : List of stations on which rotation is estimated. If None, uses all the stations.

• verbose : talk to me

Returns:

• None

scale(factor)

Scales the gps velocities by a factor.

Args:

• factor : multiplication factor (float)

Returns:

• None

scale_errors(scale)

Scales the errors (in-place multiplication)

Args:

• scale : float

Returns:

• None

select_stations(minlon, maxlon, minlat, maxlat)

Select the stations in a box defined by min and max, lat and lon.

Args:

• minlon : Minimum longitude.

• maxlon : Maximum longitude.

• minlat : Minimum latitude.

• maxlat : Maximum latitude.

Returns:

• None

setGFsInFault(fault, G, vertical=True)

From a dictionary of Green’s functions, sets these correctly into the fault object fault for future computation.

Args:

• fault : Instance of Fault

• G : Dictionary with 3 entries ‘strikeslip’, ‘dipslip’ and ‘tensile’. These can be a matrix or None.

Kwargs:

• vertical : Do we set the vertical GFs? default is True

Returns:

• None

setStat(sta_name, x, y, loc_format='LL', initVel=False)

Set station names and locations attributes

Args:

• sta_name: station names

• x: x coordinate (longitude or UTM)

• y: y coordinate (latitude or UTM)

Kwargs:

• loc_format: location format (‘LL’ for lon/lat or ‘XY’ for UTM)

• initVel: initialize a vel_enu attribute with zeros

Returns:

• None

setStatFromFile(filename, initVel=False, header=0)

Set station names and locations attributes. File should be formatted as

Station Name	lon	lat
TAGA
POUT	14.5	23.2

Args:

• filename : name of the station list

Kwargs:

• initVel : Intialize a vel_enu vector or not

• header : Length of the file header

Returns:

• None

setvelo(station, vel, data='data')

Sets the velocity enu for a station.

Args:

• station : name of the station

• vel : 3 comp list or np.array

Kwargs:

• dataCan be ‘data’, ‘synth’ or any

  attribute of the size of vel_enu

Returns:

• None

simulateTimeSeriesFromCMT(sismo, scale=1.0, verbose=True, elasticstructure='okada', sourceSpacing=0.1)

Takes a seismolocation object with CMT informations and computes the time series from these.

Args:

• sismo : seismiclocation object (needs to have CMTinfo object and the corresponding faults list of dislocations).

Kwargs:

• scale : Scales the results (default is 1.).

• verbose : talk to me

• elasticstructure : can be okada or edks

• sourceSpacing : spacing of sources in case edks is chosen

Returns:

• None

simulateTimeSeriesFromCMTWithRandomPerturbation(sismo, N, xstd=10.0, ystd=10.0, depthstd=10.0, Moperc=30.0, scale=1.0, verbose=True, plot='jhasgc', elasticstructure='okada', relative_location_is_ok=False)

Runs N simulations of time series from CMT. xtsd, ystd, depthstd are the standard deviation of the Gaussian used to perturbe the location The moment will be perturbed by a fraction of the moment (Moperc). if relative_location_is_ok is True, then all the mechanisms are moved by a common translation.

Args:

• sismo : seismiclocation object

• N : Number of perturbed models

Kwargs:

• xstd : std dev in longitude (km)

• ystd : std dev in latitude (km)

• depthstd : std dev in depth (km)

• Moperc : maximum perturbation of the seismic moment (%)

• scale : Scaling factor

• verbose : talk to me

• plot : name of the station to plot

• elasticstructure : okada or edks

• relative_location_is_ok : a common perturbation for all mechanisms

Returns:

• None

write2GMTfile(namefile=None, data='data', outDir='./', vertOrHoriz='horiz')

Writes GPS data to the GMT psvelo format file, converting units from m to cm Args:

• namefile : Name of the output file.

• data : data, synth, strain, transformation.

• vertOrHoriz : write vertical or horizontal data to file (Default horiz)

write2file(namefile=None, data='data', outDir='./')

Write the data to a file. If namefile is None, then the output file will be in the form outDir/self.name.dat

Kwargs:

• namefile : Name of the output file.

• data : data, synth, strain, transformation.

• outDir : Output directory

Returns:

• None

writeEDKSdata()

This routine prepares the data file as input for EDKS.

writeProfile2File(name, filename, fault=None)

Writes the profile named ‘name’ to the ascii file filename.

Args:

• name : Name of the profile to write out

• filename : Name of the output file

Kwargs:

• fault : Add the location of a fault (uses the fault trace)

Returns:

• None

writeTimeSeries(verbose=False, outdir='./', steplike=False)

Writes the time series of displacement in text files. Filenames are entirely determined from the name of the station example: STAT.dat, COPO.dat, ISME.dat …

Kwargs:

• verbose : talk to me

• outdir : output directory

• steplike : write 2 dots per day

Returns:

• None

xy2lonlat()

Convert all stations x, y to lon lat using the utm transform.