0
$\begingroup$

I'm confused with the GFS model data products, presented at NCEP Products Inventory website. There are two types of products for analysis: ANL ("**.anl") and FH000 ("*.f000"). As the inventory suggests, they are both analysis, but ANL containing less data, than FH000 (in partucular, ANL has no data for altitudes 2 meters and 10 meters, while FH000 has). enter image description here

Both ANL and FH000 have some shared data, but the fact is there are differences between them within the shared datasets.

For example, I've taken similar datasets from ANL and FH000 to check whether they are the same or not. The datasets are Temperature at altitudes 80 m and 100 m. The subsets for 80 m temperature showed, that the data arrays are different (ANL ("**.anl") on the left, FH000 ("*.f000") on the right): enter image description here

Could you please help find out, what are the ANL and FH000 products, what are the differences between them, and why are the seemingly common datasets diferent? Which data to rely on in research?

New contributor
Outlaw is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.
$\endgroup$
8
  • 1
    $\begingroup$ Does earthscience.stackexchange.com/a/10142/6298 particularly the second part help with your question? $\endgroup$ Jun 9 at 17:01
  • 1
    $\begingroup$ Yeah, I had to deal with such things like 15 years ago. I don't miss them :-( Sorry, wish I had more input, know it can be obnoxious to navigate :-/ $\endgroup$ yesterday
  • 1
    $\begingroup$ My only other guesses focus around ANL either starting later by eliminating some variables (and so including more "late" data), which would be somewhat in line with the quote... $\endgroup$ yesterday
  • 1
    $\begingroup$ ...or perhaps it's related to spinup (I don't know if GFS still does spinup, but many models have done), where basically it mixes the previous model run with current data, then with some smoothing/scaling to get stable initial conditions so that things that may be dynamically unstable at the resolution of the model aren't kept (even if realistic)? And so maybe analysis (and then GDAS FNL) are focused on representing the true current data??? Just a wild guess... $\endgroup$ yesterday
  • 1
    $\begingroup$ Though it's weird to me that 2 meter temperature (the standard "surface" level) isn't in the ANL. I guess the focus is on the higher atmosphere??? But then, things like SBCAPE are very dependent on 2m temperature, so how does it have that but not 2m T. Blah. $\endgroup$ yesterday

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.