forestgeo/fgeo.data

Update census tables.

maurolepore opened this issue · 3 comments

maurolepore commented

Census tables tree and stem are out of date. They don't match with the output of rtbl(). Try recreate them with rtbl().

maurolepore commented

Suzanne (via), following this post:

The R tables for Luquillo do not have agb nor ba, since I decided to ignore the scripts for creating these variables for now (told Valentine to ignore them). However, if you do want to include agb or ba, you are welcome to do so, since they should not be hard to calculate. However, for now, I think we should just concentrate on getting the rest of the columns in the R tables correctly. Also, many of the users prefer using their own equations to estimate agb.

With respect to the other variables, MeasureID should change to DBHID.

Whether you want to keep StemTag, CensusID, pom, and DFstatus in the Tree tables is up to you (or Stuart?).

And whether you want to keep CensusID, pom and countPOM in the Stem tables is again up to you.

These variables will probably never be used in any of the functions in the CTFSRPackage.

I guess Rick refined the creation of the R tables from ViewFullTable. You can easily change your scripts to reflect the same thing. It is really up to you and the rest of the ForestGeo team.

My suggestions is to eliminate these variables from the Stem table since they will not be used in any analayses.

In the Tree tables, I suggest keeping StemTag, so that the user knows whether the largest stemtag changed from one census to the other.

maurolepore commented

TREE

  • Remove agb and ba:

  • Remove StemTag, CensusID, pom, and DFstatus (These variables will probably never be used in any of the functions in the CTFSRPackage).

  • Change MeasureID to DBHID.

STEM

  • agb.

  • Remove CensusID, pom and countPOM (These variables will probably never be used in any of the functions in the CTFSRPackage).

  • Change MeasureID to DBHID.

  • Keep StemTag

  • Create a function to convert from stem to tree table, instead of having an entire -- almost duplicated -- dataset.

  • From stem table, remove nostems and other calculated variables.

  • For sanitizing code, maybe extract all possible variables from each type of table from all possible datasets -- new and old? Then sanitize to keep the structure described here.

maurolepore commented
  • Check that documentation matches updated tables.