Questions about the inversion of site effects

[wangzb0329]

Hi, Claudio
I've been using SourceSpec lately and encountered an issue: when calculating source parameters, it is necessary to consider the influence of path effects and site effects. My understanding is that SourceSpec inverts path effects but does not take site effects into account. If I want to invert for the site effects of stations, do you have any suggestions? Or how should I modify the code, such as adding site effect parameters to the inversion parameters?

[claudiodsf]

Dear @wangzb0329, thanks for using SourceSpec!

Site effects can be included as spectral corrections, through the parameter residuals_filepath in the config file.

This is the path to a HDF5 file with the mean residuals per station in magnitude units. This file is generally created using the command source_residuals on a previous run of SourceSpec. The idea is to accumulate station residuals from a first run of SourceSpec, using several earthquakes recorded at the same station. So, this depends on the fact that you have a sufficient number of analysed events (the default for source_residuals is at least 20 events per station).

This is an example of spectral station residuals, plotted with the --plot option of source_residuals: #68 (comment)

Alternately, you can build the station residual file yourself, from your own information on site effects. The file format is currently not documented, but I can send you an example if you wish to go that route.

[wangzb0329]

Dear @claudiodsf , thank you for your reply, this solved my problem!