Add leaf lifespan as LMA indep parameter

[fabern]

This is a PR for an open discussion on how to prescribe leaf life span. Should it be hardcode-linked to LMA or should it be a free parameter?

Currently, leaf lifespan is scaled linearly with leaf mass per area (LMA):
leafLS = 28.57143 * LMA = LMA / 0.035 ! years

This is taken over from:

• https://github.com/wengensheng/LM3PPA/blob/3ecc0f070fe16af0157edcb72f0313ac2d9fd494/src/land_lad2/vegetation/vegn_data.F90#L517-L518
• https://github.com/wengensheng/ModelE-BiomeE/blob/17733f2f8523f67b74d5b4000908230be79ea6d1/BiomeE_const.f#L259
• https://github.com/wengensheng/ModelE-BiomeE/blob/17733f2f8523f67b74d5b4000908230be79ea6d1/BiomeE_VEG.f#L554

That means the provided parameter params_species$leafLS is effectively inactivated since above relationship is hardcoded in the code.
This PR would be an attempt to activate this parameter.

There is some evidence that leaf life span scales with LMA. Below articles do linear regression between LMA and LLS. However, this is done in the log-log transformed space:

• Osnas 2013 https://www.science.org/doi/10.1126/science.1231574
• Wang 2023 https://doi.org/10.1126/sciadv.add5667

[fabern]

As a linked, but separate issue:

With the status prior to this PR:
If LMA < 0.035 kgC/m2, then LS becomes smaller than 1 and the phenotype switch enforces deciduous behavior (phenotype=0). This is a hardcoded limit. It expresses the approach of the leaf economic spectrum: are leaves an expensive or cheap investment?

Issue:
According to TRY-DB, Picea Abies has specific leaf areas of 17.5 mm2/mg = 17.5 m2/kg (p05: 11.5 to p95: 25.5). From that follows a leaf mass area of 0.057 kg/m2, which can be transformed to LMA of 0.028 kgC/m2, assuming 50% of the leaf's mass is carbon.

The issue is now that the hardcoded phenotype switch prevents simulating Picea Abies as evergreen species (forcing phenotype=0).

Potential solutions are:

• a) (reject this PR#329): keep code as-is, keep hardcoded limit of 0.035 gC/m2 and distrust the estimate of Picea Abies LMA of 0.028 gC/m2 and find improved estimates (see. e.g [1] )
• b) (reject this PR#329): keep code as-is, only fix by decreasing LMA hardcoded limit to 0.027 gC/m2 – not supported by e.g. [2] that
• c) (reject this PR#329): specify phenotype as input parameter (0 for deciduous, 1 for evergreen) remove link between LMA or leafLS and phenotype (note BiomeEP internally still needs values for leafLS and alpha_L). Allows to enforce Norway Spruce as evergreen, independently, but gets rid of the leaf economic spectrum approach.
• d) (accept this PR#329): i.e. specify both leafLS and LMA as input parameters. In this case, alpha_L and phenotype are both derived from leafLS. This allows to decouple LMA and leafLS. Is also getting rid of the leaf economic spectrum approach.

@stineb, @wengensheng : do you have any thoughts on these issues (message 1 and 2)?
Linking derived traits to LMA is quite central to approach using the leaf economic spectrum. It simplifies large scale simulation by reducing the number of needed input parameters.

I suggest to use solution a), given alternative estimates between 0.062 and 0.166 kgC/m2 [1].

[1] Bellan, M., Krejza, J., & Marková, I. (2019). Estimation of newly grown needle area in Norway spruce 
based on simple biometrical tree parameters. Austrian Journal of Forest Science/Centralblatt Für Das 
Gesamte Forstwesen, 136(1). 
https://www.forestscience.at/content/dam/holz/forest-science/2019/01/CB1901_Art1.pdf

Their figure 6 shows decreasing SLA (i.e. increasing LMA) with tree height. (Or rather the distance from 
top?) They express this relationship as:
SLA = 338.79 * H_m^(-0.816197) where H_m is height of branch above ground.

Approximate values are (again using 50% C of mass of leaf):
SLA: at 5m above ground: 30 cm2/g = 3 mm2/mg = 3 m2/kg  ==> LMA of 0.333 kg/m2 => approx 0.166 kgC/m2
SLA: at 20m above ground: 80 cm2/g = 8 mm2/mg = 8 m2/kg  ==> LMA of 0.125 kg/m2 => approx 0.062 kgC/m2

[2] “We interpret leaves with MA values of <87 g m 2 to have leaf life spans of <1 year, and leaves with MA values >129 g m 2 to have leaf life spans of >1 year (see ‘‘Results’’ and ‘‘Discussion’’).” (Royer et al., 2007, p. 583)

[2] Royer, D. L., Sack, L., Wilf, P., Cariglino, B., Lusk, C. H., Wright, I. J., Westoby, M., Jordan, G. J., Niinemets, 
Ü., Coley, P. D., Cutter, A. D., Labandeira, C. C., Palmer, M. B., Johnson, K. R., Moles, A. T., & Valladares, F. (2007). 
Fossil leaf economics quantified: Calibration, Eocene case study, and implications. Paleobiology, 33(4), 
574–589. https://doi.org/10.1666/07001.1