CCPP Framework Meeting Minutes 2019 08 15

NCAR CPF/CCPP transition

• NCAR uses CCPP transition to clean up physics
• starting with boundary layer scheme
  • use the 2-dim. PBL scheme inside the 3-dim. WRF wrapper
• overlap of schemes with what is already in CCPP (we have YUS)

Standard names

• dimensions
• vertical dimension: how to avoid confusion?
  • layer? level? interface? midpoint? cell average?
  • seek input from scientists
  • look at CF discussion pages as well
  • formula terms should be in the metadata
  • straw-man to go out to scientists: "interface" + "layer" (with additional qualification midpoint/average?), or "interface" + "midpoint + "cell average" (with documented alias "layer")
  • still need to define and standardize the variable families
• TODO FV3 cleanup (Dom, Grant):
  • vertical_dimension_plus_one --> vertical_interface_dimension
  • horizontal_dimension --> horizontal_loop_extent, ix -> im
• TODO Steve to talk to CGD folks, Dave to MMM folks, Ligia to Jack/Moorthi; all: Friday August 23 at PI call
• TODO Dave to put together a document (with nice figures :-) ) to share
• Where to put the list of standard names?
  • new repository (organization-independent)
  • initially based on xml -> convert to html
  • TODO Steve to get started with it
  • the repository/website will allow feedback (GitHub issues page)
• standardization of families: lean on grib standard (has families)? or ignore on purpose?
• future work: define grib code table so that CCPP variables can easily be identified in/converted to grib variables (for a subset?)
• future work, in general: provide mapping to other standards