search

JackDMcGrath / LOOPY

Corrections for Interferogram unwrapping errors, based on the COMET LiCSBAS software
GNU General Public License v3.0
4 stars 1 forks source link

Cosesimic decision based on parameter uncertainties #49

Closed JackDMcGrath closed 1 year ago

JackDMcGrath commented 1 year ago

Decide whether an inversion is successfull based on whether the parameters are above the error values

JackDMcGrath commented 1 year ago

So what if we find a parameter is unsuccessful? Unclear how to proceed after that point. How do we re-model? Do we flag areas as dodgy?

JackDMcGrath commented 1 year ago

Decided that the best solution is to not remodel - if we're going to start doing that then we should just do a batter Bayesian or Maximum Likelhood inversions. Instead, I shall provide the 'True' model (eg the modelled coseismic displacement is the offset from the long term trend, so with more than 1 earthquake this becomes incorrect. True coseismic is calculated using the difference between inverted dates) and the error maps. If needed, the error maps can be used to mask the results