Verify Rotation with hodoplot.m

There are basically two ways to confirm proper determination of the tool orientation by PCA, and the application of that determination by rotating the horizontal components to a standard orientation. One can select a geophone station depth and plot a hodogram of the particle motion before rotation and after. The other evidence comes later when we plot all the horizontal data and look for any abrupt changes in the direct arrival waveform which would suggest a possible rotation error.

The former is done with the Octave program, hodoplot.m (see 6.7.4). Start an Octave session and execute from within the Octave text window the following command (segyinfo.m, bsegin.m must also be in the path)

hodoplot;

The Octave session must be started in the same directory as the data to be plotted, and of course, a copy or link to hodoplot.m would also be in that directory. A number of obvious dialog GUI's come up, most of which can be defaulted. The two that are important are the selection of which trace to plot on which axis, and the time window selection. The standard BSU orientation is to have the down-hole components on channels 1,2, and 3 (V-comp, R-comp, T-comp). In that case, one select channel 2 for the x-axis, and channel 3 for the y-axis. The default time window GUI shows 0 to tmax seconds. One benefits from looking only at the direct arrival, and can step through the motion by progressively increasing the maximum time (the program loops until you click on cancel). For this example, the max times to plot were chosen as follows:

• 0    to .065 sec
• .065 to .070 sec
• .070 to .075 sec
• .075 to .078 sec

Figure 15 shows the result of plotting the data, first as recorded (A),and then a second time with the rotated data set (B).

Figure 15: Plots produced by hodoplot.m confirms that data were rotated as desired