Stefan,

ARTSCAT-3 throws away line mixing data, which is kept in ARTSCAT-5.  It
also throws away the Einstein coefficient and the statistical weights of
the lines.  I am unsure if it even parses quantum numbers.  All of these
things are needed in advanced calculations and provided for in HITRAN.

(Alex ran into the former problem with the oxygen simulations yesterday.  I
guess he (or I) wanted to store O2 of LBLRTM to save reading speed but
forgot to change the pseud-class index of the lines so all the line mixing
data was missing upon reading).

Since ArrayOfLineRecord is a psuedo-class, we cannot store the required
level of catalog for some lines and not for others.  The argument before
was that we should therefore save disk space/reading time for users only
needing high level calculations, and assume that if you did more advanced
calculations you would know the internals of ARTS and use the conversion to
ARTSCAT-5 before saving the catalog.

I think this argument was valid then but no more.  ARTSCAT-5 is slower to
read and takes up more disk space but the features are not only used by
folks that know the internals of ARTS anymore.

(This does not negate that we need an overall faster method of
storing/reading ArrayOfLineRecord but such a method could be forced to
operate on ARTSCAT-5 only, since it is the catalog that most closely
represents what is going on in the internals of ARTS.)

With hope,
//Richard

2017-09-15 17:23 GMT+09:00 Stefan Buehler <[email protected]>:

> Dear Richard,
>
> > The reason to make the change is that I often run into problems of lost
> data because ARTSCAT-3 simply throws it away.
>
> Please help me understand the problem, which data are thrown away?
>
> Best wishes,
>
> Stefan
>
>
> _______________________________________________
> arts_dev.mi mailing list
> [email protected]
> https://mailman.rrz.uni-hamburg.de/mailman/listinfo/arts_dev.mi
>
>
_______________________________________________
arts_dev.mi mailing list
[email protected]
https://mailman.rrz.uni-hamburg.de/mailman/listinfo/arts_dev.mi

Reply via email to