Closed eviatarbach closed 11 months ago
Thanks for pointing this out @eviatarbach ! Yes, both of these are issues and the solution you gave (hopefully) can solve things. Can you provide the traditional notation for the solutions for omega?
Great! Is this clear?
This has been revised to the equation above. Thanks @eviatarbach !!
Compared to the code in calc_omega.R, I get a different expression for the domain of validity for the two solutions of
omega
. By solving Eq. 14 in Smith et al. (2019) for omega, assuming thatm
,I
,c
,theta
, andphi
are all greater than 0, and requiring thatomega >= 0
andA_j >= 0
, Mathematica returns the following (see notebook here):Minus solution:
((8 c > m && 4 c < m && m/c < 8 theta && theta < 1) || (4 c > m && theta > 1))
Plus solution:((8 c < m && theta < 1) || (m/c > 8 theta && 8 c > m && 4 c < m))
It seems that this is equivalent to the code in
calc_omega.R
in some cases, not so in others. For one thing,theta
cannot be greater than 0.85 incalc_omega.R
due to the computation of thequad_root
variable, whereas the above does not have this limitation. Also,calc_omega.R
can return negative values, for example withtheta=0.85, c=0.053, m=0.2
, which I don't think makes sense, since it leads to a negative Vcmax.