Alternative models for cold-deciduous phenology?

[mpaiao]

The cold-deciduous phenology implemented in FATES is based on Model 3 of Botta et al. (2000). They indeed found that this was the best model for cool temperate broadleaf trees, but they recommend Model 1 for deciduous needleleaf trees and model 2a for scrublands, savannahs and grasslands (and a few other mixed landscapes). Should we consider implementing these other models at some point?

[rosiealice]

Sounds good. I can confirm that there is no particular reason we didn't do this in the first place. Is this still the best phenology model paper 24 years later?! Kyla Dahlin did some work on tropical phenology in 2017, for example:

https://nsojournals.onlinelibrary.wiley.com/doi/full/10.1111/ecog.02443

[mpaiao]

Just to clarify, Botta et al. (2000) did implement drought deciduous phenology, but the drought deciduous phenology in FATES is not based on Botta et al. (2000). Only the cold-deciduous phenology is based on Botta et al. (2000). But I think previous work with ED2, which also uses Botta et al. (2000) for temperate phenology, showed large uncertainties and biases using this model for Harvard Forest. If I remember correctly, this was David Medvigy's motivation to implement a "satellite phenology" version of ED2. So yes, maybe we should look for alternative phenology models (for all latitudes).

I've seen Kyla's paper before and it may be indeed a good alternative for tropical phenology, my only concern with it is that the model does use continents to modify some parameters. But maybe @JessicaNeedham is planning on adding some continent-specific information that can be used for phenology too.

[rosiealice]

Indeed. I'm not certain there are good philosophical reasons for avoiding continent/latitude specific calibrations...

[adrifoster]

I am exploring adding a few new pfts based on continent and latitude with my calibration! The hard part will be porting that back to an actual global SP run

…

On Sat, Apr 19, 2025 at 1:43 PM Rosie Fisher ***@***.***> wrote: Indeed. I'm not certain there are good philosophical reasons for avoiding continent/latitude specific calibrations... — Reply to this email directly, view it on GitHub <#1390 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADE42ITTCS6UWJUQLLXVXVD22KRNXAVCNFSM6AAAAAB3LDCZ22VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMJWHA2DCMRRG4> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***> *rosiealice* left a comment (NGEET/fates#1390) <#1390 (comment)> Indeed. I'm not certain there are good philosophical reasons for avoiding continent/latitude specific calibrations... — Reply to this email directly, view it on GitHub <#1390 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADE42ITTCS6UWJUQLLXVXVD22KRNXAVCNFSM6AAAAAB3LDCZ22VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQMJWHA2DCMRRG4> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>