Res2Dinv Format Error

[shdlovezxm]

Problem description

Excuse me Sir, I try to load a resisitivity data use format Res2Dinv. The following error occurred. Please help me. thank you very mush.

Your environment

Please provide the output of print(pygimli.Report()) here. If that does not work, please give provide some additional information on your:

Operating system: Windows Python version: 3.9 pyGIMLi version: 1.4.5 Way of installation: Conda package

Steps to reproduce

data = ert.load("RES2DC.txt")
print(data)
...

Actual behavior

05/12/23 - 09:57:26 - pyGIMLi - INFO - could not read unified data format for ERT ... try res2dinv
05/12/23 - 09:57:26 - pyGIMLi - INFO - could not read res2dinv ... try Ascii columns
05/12/23 - 09:57:26 - pyGIMLi - INFO - Failed importing Ascii column file. Consider using pybert.
05/12/23 - 09:57:26 - pyGIMLi - INFO - No electrode positions found!
05/12/23 - 09:57:26 - pyGIMLi - INFO - imported:  Data: Electrodes: 0 data: 0
85  25  3.8 686.649
Data: Electrodes: 0 data: 0

RES2DC.txt

[halbmy]

A simple bug only occuring for Pole-Dipole that was corrected (eedac16b)

[shdlovezxm]

There was an error during terrain import because the Sensors are 73, while the terrain data is only 30. How should I handle this?

print(data) print(data.sensorCount())

Data: Sensors: 73 data: 358, nonzero entries: ['a', 'b', 'm', 'n', 'rhoa', 'valid'] 73

0 1291.784 5 1289.801 10 1287.954 15 1286.057 20 1284.747 25 1282.701 30 1280.152 35 1278.045 40 1276.219 45 1274.407 50 1271.613 55 1270.119 60 1268.588 65 1266.813 70 1264.946 75 1264.023 80 1265.691 85 1265.998 90 1264.944 95 1263.731 100 1259.808 105 1256.49 110 1252.827 115 1249.332 120 1246.578 125 1243.499 130 1240.428 135 1238.206 140 1236.123 145 1234.132

[halbmy]

Generally, topographic data are having a different discretization. However, it seems like the are not read from the res2dinv file. Try reading them with numpy and use the following function for interpolation: https://www.pygimli.org/pygimliapi/_generated/pygimli.meshtools.html#pygimli.meshtools.interpolateAlongCurve

[shdlovezxm]

I understand. Thank you very much for your patient response. Good luck to you.

[halbmy]

We enabled reading topography from res2dinv file format (48ce3720).

Note that the points are read but not yet automatically applied to the electrode positions, which might make sense later in a script. One can use

data.setSensorPositions(pg.meshtools.tapeMeasureToCoordinates(data.additionalPoints(), pg.x(data)))

However, in your case there is something wrong as your electrodes go beyond zero. So either there is something wrong in your data file, or our import routine has a bug for pole-dipole in conjunction for midpoint (is it between M and N or between A and M or the mean of A,M,N?) positions specified. We will have to check but you should do so as well as independent on topography, your data are not correctly read.

[halbmy]

Any reply to my last comment on the wrong data?

Note that we have released version v.1.4.6 supporting res2dinv topography information.

[halbmy]

I've checked the code of the current version v1.4.6 with some other res2dinv files and the following procedure works well:

topo = data.additionalPoints()  # 
elec = pg.meshtools.tapeMeasureToCoordinates(topo, pg.x(data))
data.setSensorPositions(elec)

It assumes that the x coordinates in the data file are along the topography. You can check the positions by

plt.plot(pg.x(topo), pg.y(topo), ".-", label="topo")
plt.plot(pg.x(elec), pg.y(elec), "+", markersize=4, label="electrodes")
plt.grid()
plt.legend()

[halbmy]

In your case the electrodes start at -30m.

If you're sure that your format is correct and the import of Pole-Dipole data is wrong, please open a new issue.