While reviewing the section of the paper on the model structure function, I came to the conclusion that the way we were dealing with the finite map size was not right.
The equation A1 for beta(r0) is fine, but it does not make sense to apply this to the sigma^2 and r0 parameters in the model structure function. Instead, it gives the reduction in the empirically determined sigma^2 and r0, but this is purely because of truncation of the large scales, not because of any change in the strucfunc at a given separation.
Note that the curves in Fig A1 are all exactly the same in the ranges where they overlap.
Therefore, we should go back to bfunc03s instead of bfunc04s.
I have edited all the CI*.ipynb notebooks to reflect this, and have re-run them.
While reviewing the section of the paper on the model structure function, I came to the conclusion that the way we were dealing with the finite map size was not right.
The equation A1 for beta(r0) is fine, but it does not make sense to apply this to the sigma^2 and r0 parameters in the model structure function. Instead, it gives the reduction in the empirically determined sigma^2 and r0, but this is purely because of truncation of the large scales, not because of any change in the strucfunc at a given separation.
Note that the curves in Fig A1 are all exactly the same in the ranges where they overlap.
Therefore, we should go back to
bfunc03sinstead ofbfunc04s.I have edited all the
CI*.ipynbnotebooks to reflect this, and have re-run them.