> 3dFWHM doesn't want an entire time series which is what the
> errts files is, so should there be a specific time point I choose from
> the time series or should I calculate the fwhm and sigma values for
> each errts time point and get averages for AlphaSim?
Calculating FWHM based on one specific time point of the residual errors is too biased. You can calculate the standard deviation of the residual time series by running 3dTstat -stdev, and then estimate FWHM of the standard deviation.
Gang