This seems to be semi-documented on page 45 of the 3-PG user manual:
This “species” parameter is actually a site-specific parameter.
Though I suspect the sentence is perhaps meant to be read "These two "species" parameters are actually site-specific parameters."
However, the transpiration discussion in section 11.7.3 of the manual can be interpreted in favor of tree species specific values of Qa and Qb, which is consistent with these two variables being declared real(kind=8), dimension(n_sp), intent(in) :: Qa, Qb and used as such in s_transpiration_3pgpjs(). That's is also mostly the case in s_transpiration_3pgmix().
But 3-PGmix also calculates soil evaporation as if soil is the first tree species defined. From a software design perspective I'd normally expect to see these two values in d_site rather than coming from the first tree species in d_parameters.
Since different parts of the manual and different parts of the code seem to have different notions about the meaning of Qa and Qb I thought it might be interesting to note this for discussion here.
I tend to see Qa and Qb around the middle of the parameter cloud in sensitivity analysis, enough to make them sometimes eligible for inclusion in model calibration. So I'm a bit curious whether there's model design intent for using just one value each for Qa or Qb across all species, if it's expected Qa and Qb vary on a per species basis (presumably with distinct soil Qa and Qb values), or if this is something that's not yet well understood.
This seems to be semi-documented on page 45 of the 3-PG user manual:
Though I suspect the sentence is perhaps meant to be read "These two "species" parameters are actually site-specific parameters."
However, the transpiration discussion in section 11.7.3 of the manual can be interpreted in favor of tree species specific values of Qa and Qb, which is consistent with these two variables being declared
real(kind=8), dimension(n_sp), intent(in) :: Qa, Qband used as such ins_transpiration_3pgpjs(). That's is also mostly the case ins_transpiration_3pgmix().But 3-PGmix also calculates soil evaporation as if soil is the first tree species defined. From a software design perspective I'd normally expect to see these two values in
d_siterather than coming from the first tree species ind_parameters.Since different parts of the manual and different parts of the code seem to have different notions about the meaning of Qa and Qb I thought it might be interesting to note this for discussion here.
I tend to see Qa and Qb around the middle of the parameter cloud in sensitivity analysis, enough to make them sometimes eligible for inclusion in model calibration. So I'm a bit curious whether there's model design intent for using just one value each for Qa or Qb across all species, if it's expected Qa and Qb vary on a per species basis (presumably with distinct soil Qa and Qb values), or if this is something that's not yet well understood.