The tdip package holds some functions for importing data, and for modelling IP effects, all based upon pyGIMLi.

Main entry, however, is the class TDIP, to be imported by

from tdip import TDIP

It can be initialized empty or directly with a data file

pro = TDIP("datafile.ohm")

Valid formats are:

• Syscal Pro export file (*.txt)
• ABEM TXT export file (*.txt or raw time series)
• Syscal Pro binary file (*.bin)
• GDD format (*.gdd)
• Ares II format (*.2dm)
• Aarhus Workbench processed data (.tx2 and .dip)
• res2dinv data

You can also use pro.loadData(filename).

As a result, the instance holds an ERT data file pro.data, which can be used and processed with the pyGIMLi ERT module, and a matrix of apparent chargeabilities pro.MA.

For most formats, it should also load the gate times pro.t, if not, they can be set by pro.setGates().

Preprocessing

pro.showRhoa() shows the apparent resistivity. If it is not there yet, it can be created by computing the geometric factor, either analytically or numerically, by ensureRhoa(). This should be done after changing the electrode positions after loading.

.fitDataDecays() fits every decay by a simple Debye (exponential decay) or Cole-Cole (possibly also CPA) model to have a rough guess of the quality of the decay.

The function .showData() can show every property like 'a' or 'rhoa', including he fitted 'm0', 'tau' or the fit quality 'fit'.

You can show the apparent chargeability for individual gates by .showApparentChargeability(). The function pro.generateDataPDF() generates a multipage PDF with all included data fields.

Decays can be shown by pro.showDecay() passing an AB or MN combination (or both). The function pro.generateDecayPDF() generates a multipage PDF with all decays grouped for the same AB. You can also work with individual decays after decay=pro.getDecay(abmn=...)

For further manipulation, the functino getDataIndex(abmn=[...]) can help you identifying specific numbers.

Most important preprocessing tool is pro.filter(). It works either along the data sequence axis (like data.remove()) or along the time axis and removes complete data (including all gates) or gates (for all data). Important keywords are

rmin=0, rmax=1e9, emax=1e9, # apparent resistivity and error
kmax=1e6, electrode=None, forward=False,
umin=0, umax=9e9, nr=[],
a=None, b=None, m=None, n=None, ab=None, mn=None, corrSID=1,
tmin=0, tmax=1e9, # gate times
fitmax=1e9, m0max=None, m0min=None, taumin=-9e99, taumax=9e99, # fitted decays

If you don't want to remove the whole data point, but only disregard some erroneous points (like outliers), you can mask these using .mask() with the keywords mamin and mamax. You can also mask the whole decay by filter(..., mask=True) in order to not to loose the apparent resistivity information.

Inversion

First, you start with inverting the apparent resistivity through pro.invertRhoa()

This triggers an ERTManager() and you can pass all options.

You can invert a single apparent chargeability by pro.invertMa() specifying its number (0 for the fitted m0).

To invert all gates sequentially, use pro.individualInversion() or simultaneously, use pro.simultaneousInversion()

There is a special inversion into Debye models for every cell. pro.invertDebye()

Class instances can be accessed by pro.ERT, pro.invIP

Postprocessing

As a result of apparent chargeability inversion, you have a matrix of chageabilities pro.M for every model cell, matching the inversion mesh pro.pd.

.getModelDecay() returns a decay class instance for a number or position (using getCellID()).

.fitModelDecays()

fitModelDecays() can be used to fit a Cole-Cole model to every model decay, and display m0, tau and c as sections.

showColeColeResults()

saveFigures()

saveResults()

loadResults()

so that you can do postprocessing independent of inversion by

pro = TDIP(filename)
pro.loadResults()
...

Working in 3D

While visualization is designed for 2D profiles, numerical computations should work in 3D as well, provided a 3D mesh is created for the ERT manager.

Conversion to FD

convertToFD() converts the whole data set to a frequency-domain instance FD (package FDIP required).