Query on reorientation of accelerometer records

[mnaguit]

Query:

I am attempting to resolve the Wellington Airport accelerometer readings into a true north true east configuration. The instruments are labels as N34E and S56E and I have the acceleration records for those orientations for the Kaikoura earthquake.

Can you find someone who can point me to the methodology I should use to combine those records to provide a true north acceleration simulated record.

[mnaguit]

I got your message sent to Richard. I am one of the seismologists at GeoNet so should be able to help you.

I’ll assume you are using Python, obspy and FDSN. If not I highly recommend these. We have a number of tutorials about how to use these, the one for accessing waveform data is https://github.com/GeoNet/data-tutorials/blob/main/FDSN/FDSN_dataselect.ipynb

To do the rotation I’d recommend using the obspy.signal.rotate.rotate2zne function: https://docs.obspy.org/packages/autogen/obspy.signal.rotate.rotate2zne.html

Here you can enter the data streams and known orientations into the function and get oriented three component records out. I am not sure if you can use this as a function of a stream obspy, but you could easily decompose the stream into the function using [0], [1], and [2] indices to get the traces and then trace.data to get the raw data.

Let me know if this works for you.

Thanks, Sam