Plant initial stem diameter is arbitrary

[davidorme]

The initial plant stem diameter during recruitment is entirely arbitrary - we need to find some kind of initial DBH and stochiometry from the carbon and nutrients available in the propagule.

We could set that as a trait, but it would be better to work it out from first principles using the T Model.

Originally posted by @davidorme in #827 (comment)

[arne-exe]

We could set that as a trait, but it would be better to work it out from first principles using the T Model.

I agree with this. The current carbon mass per propagule does not distinguish between PFTs, and its representativeness should be taken with a grain of salt.
Linking to the T model, we could use the PFT specific allometry (height-dbh relationship + T model) to calculate the carbon mass needed for a particular dbh. We could then calculate the carbon required for all integer dbh values smaller than 5 cm (which is the smallest dbh in the cohort distribution). We can then assign the largest dbh that is possible to achieve with the carbon contained in the current propagule. If propagule carbon mass changes in the future (i.e., become PFT specific) this approach should still work.

I could see how starting from the propagule carbon mass instead could potentially work by also linking it to the PFT allometry, but I guess the issue here is that the allometry can vary both on the dbh and height axis. It would also rely a lot more on the accuracy of the current propagule carbon mass, and assume that it is representative across PFTs.

In terms of stoichiometry, I would probably keep it the same as the "adult" cohorts, although in reality there will be changes here (for example the new recruits will have less dense wood). But on the other hand we are also not taking into account other woody tissues at the moment, so it wouldn't be the first assumption here.
Another approach is that we account for this by using a different wood density, so you'd have 2 wood densities per PFT (one as recruit and one as adult).

[arne-exe]

We could then calculate the carbon required for all integer dbh values smaller than 5 cm (which is the smallest dbh in the cohort distribution). We can then assign the largest dbh that is possible to achieve with the carbon contained in the current propagule. If propagule carbon mass changes in the future (i.e., become PFT specific) this approach should still work.

I've just checked this, the smallest dbh measured in the SAFE tree census data is 99 mm, or 0.099 m. The smallest cohort distribution contains everything from 0-0.1 m, and all individuals within this range are given a cohort dbh of 0.05 m.
We could raise the floor of the smallest adult cohort dbh from 0 to 0.05 m, and reserve dbh values below 0.05 m for new recruits.
Depending on which recruit cohort new recruits are assigned into (e.g., 0-0.01, 0.01-0.02, 0.02-0.03, 0.03-0.04, 0.04-0.05) we could let them grow to larger recruit cohorts, before being merged with the adult cohorts once they grow past dbh of 0.05.

[davidorme]

I could see how starting from the propagule carbon mass instead could potentially work by also linking it to the PFT allometry, but I guess the issue here is that the allometry can vary both on the dbh and height axis. It would also rely a lot more on the accuracy of the current propagule carbon mass, and assume that it is representative across PFTs.

Allometry in the T Model only has the DBH axis: everything, including height, is driven by that. We'd need to calculate the DBH achievable from the propagule mass.