Replies: 6 comments 2 replies
-
This looks like a pretty good idea to me. Would the idea be to add some simple code to handle assigning the associated From a historical perspective, my assumption is that current convention was adopted by fates based on/inspired by the host land models vegetation properties type. Is that correct? |
Beta Was this translation helpful? Give feedback.
-
I was actually thinking about refactoring the checks in the code and having a single Sorry, I don't know the source of the current convention... |
Beta Was this translation helpful? Give feedback.
-
This is great @mpaiao . I like the idea of defining ivergreen, etc. EDTypesMod or FatesConstantsMod would both work to hold the definitions. Both modules have similar types of definitions. |
Beta Was this translation helpful? Give feedback.
-
Great idea @mpaiao. I'm excited about semi deciduosness :) On the historical front, @glemieux, FATES didn't use to have any drought deciduousness, so yes, it was generally added to mirror the HLM convention (which was actually JULES at that point!) |
Beta Was this translation helpful? Give feedback.
-
Thanks all for the feedback! Perhaps to keep things a bit more tractable, I may submit two pull requests, one with the changes in drought-deciduous phenology (#801, though I may need to supersede that with another PR) and one with the parameter change. |
Beta Was this translation helpful? Give feedback.
-
In the parameter files, we have three PFT-dependent parameters to describe which type of leaf phenology strategy is associated with each PFT:
fates_phen_evergreen
,fates_phen_season_decid
,fates_phen_stress_decid
. The code checks and prevents a PFT to have more than one of these types simultaneously active.I am actually adding a drought semi-deciduous leaf phenology to FATES, and I am wondering if we should simplify this by creating a single PFT-dependent parameter that describes the leaf phenology (
fates_phen_type
or something similar), with options like:This would allow us to add more strategies without multiplying the number of parameters and internal tests. I am not sure if there would be any downsides to this, though.
Beta Was this translation helpful? Give feedback.
All reactions