search

lizzieinvancouver / ospree

Budbreak review paper database
3 stars 0 forks source link

Unravel ranges data mystery #437

Open dbuona opened 2 years ago

dbuona commented 2 years ago

Hi Rangers,

I've started to try and unravel the ranges data issues. Here are my thoughts:

I checked European data first by comparing the newer STV adjusted file: Synthesis_climate_EUsps_STV.csv to the previous version: Synthesis_climate_EUspsw.csv. The values for all of the environmental variable (not just STV) are substantially different between the 2 versions. This would explain why the full species and EU only model output look really different but now why the North American only models also diverged. @MoralesCastilla any idea why these data look so different for thing like mean GDD, chill portions etc?

Next I looked at North America be checking Cat's Jan 4 commit. There it looks like only the STV values changed, except...correct me if I am wrong, @cchambe12 you fixed the variable "SDev.Tmins" and I believe STV is actually calculated from the "MeanTmins" (either Temp.SD or Geo.SD column of the variable MeanTmins").

Still not sure why that would have affect the North American only models for things like Mean.Chill.Portions so much, but maybe we'll find out.

Anyway, I am not sure how hard or easy this will be to fix so @MoralesCastilla and @cchambe12 let me know how I can be helpful

cchambe12 commented 2 years ago

@dbuona Sorry for the delay on this but yes... I think that is correct. Do I need to do something to fix this or can you relabel the columns?