Managing Ambient Noise Tomography module

[Abrisot24]

Hello, When i access Seismic ambiant noise in isp, add my metadata xml and mseed files everything seems fine but when i run the pre processing step, i got the error : 'list index out of range ' how can i fix that ? thanks in advance

[aolivarc]

Hi Abrisot24, thanks for reporting. Could you please share with us the full output of the error from your terminal? So we can have a better idea of the problem. Thanks!

[Abrisot24]

[rcabdia]

Ok, the problem that you are experienced it is related to the workflow process. So please find her a detailed explanation plus a screenshots to guide you. I recommend download two stations from IRIS server, let's see 1. month per stations, one channel and the metadata. This will help you a lot to make a toy test. I use to download ANMO and TUC, channels BHZ, BH1 and BH2 (https://service.iris.edu/fdsnws/dataselect/docs/1/builder/). To split the big downloaded mseed files into DAILY mseed files I use to use msedcut from (https://www.gfz.de/en/section/geophysical-imaging/infrastructure/geophysical-instrument-pool-potsdam-gipp/software/gipptools)

Please strictly follow this process

1. Create a project:
   First Set the folder to your DAILY mseed files, Second Set the path to your metadata (needs to include the response of your instrument and thestations coordinates). Third set the output path. Finally fill the parameters box

Then, click on create a project. See the terminal to check that is working correctly. Then Save the project (this is a good practice because for making later more pre-processing tests will save you all previous steps).

Now, ready to click in pre-preocessing. Track the progress on thee terminal. This can take some minutes. Check also that after processing all mseeds of every station is writting the output matrix inside the expected output folder.

Finally, after finishing the pre-processing, click at cross and stack. this action will compute the cross corrlation of all pairs of matrix inside the output folder in which you are pointing out. The output of cros and stack are three folders with the empirical green functions (stack, rotated, and daily "needs more than 1 month of data") ready to be plot.

Now you can make a dispersion curve analysis. of your EGFs.

Good luck, please report us if you are succeed

Best regards,

Roberto Cabieces

[Abrisot24]

Thanks you for your answer,
I have resolved index out of range problem but for my data for an example of samples, creating processing is good load metadata is good; run processing is good, i have something for stack (format HDF), stack daily but nothing for stack rotated but when i want to plot nothing happen, if i want to plot daily correction i have different issue for some it's error 'H' or index out of range maybe it's because is not one month data but when i do with a sample of one month for each station is to heavy for the computer.

[rcabdia]

What I understood is that everything is working but you do not have output on radial transversal neither in daily. For radial transversal you need to have numerical matrix output from preprocessing for North and East component lets say BHN and BHE. Of course in the example I saw befor I had data from IRIS for BH1, BH2 and BHZ.

Second point, why you do not have daily output ? because a daily output can be generated from a minimum of 20 days cross correlation of noise. Indeed the next daily is ten days after the initial daily. So for example for a time spam of 90 days you will have a daily EGFs for the 20, 30, 40, 50, 60, 70 and 80 centered days.

This answer your questions?