# To get a rough idea on how to use the inverison code, run ./mcd -help # To get a better idea on how to use it, check out the information in the TESTING/ sub-directory. # To understand what the inversion is actualy doing, read my thesis. Here's a more descriptive list than the one printed with ./mcd -help, including both short and long versions of the keywords: -k --kernel_set file name for the kernel set. This should be a binary file. Look around for an idl .pro file that converts the old .fits version into a binary file. -z --kernel_set_depth another binary file related to kernels, this contains the physical depths of each layer in Mm. Also has delta-z. Look around for idl .pro code to make this -i --input filename for the main data input. Should be a binary file, again, look for idl code to make this. Probably something like prep_ola.pro. -d --depths one or more depths to target independently. Examples: -d 0 -d 0,1,2,3,4 -d 0.0,15.0,30.0 -n --n-values optional restriction on radial order for inversion. listed n-vals will be used, everything else thrown away during read-in Examples: -n 0 (f-mode inversion) -n 8 (good for quick testing) [nothing] (use all modes) -o --output base filename for output. Will be appended to with details about depth, regularization, etc -l --lambda log10 value of lambda, typical values between 0 and -5. allows a comma-deparated list -L --lambda-z additional log10 lambda to add per Mm in depth, typically 0 to -1. allows list -m --mu log 10 value of mu, typically 0 to -5. allows list -M --mu-z additional log10 mu to add per Mm in depth, typically 0 to -1. allows list -A --apod amount of data apodization to use, useful for smoothing out results. allows list -p --padding amount of zero-padding to use around data, in degrees. allows list, but might crash... -a --avgker base filename for averaging kernel output. leave out to prevent avgker from being computed at all -t --threads number of openmp threads to use.