willvieira / forest-IPM

Integral projection model for trees species of eastern north America
MIT License
3 stars 0 forks source link

sim: coexistence #15

Closed willvieira closed 1 year ago

willvieira commented 1 year ago

Calculate invasion growth rate ($\lambda$) for pair of competing species in two different conditions:

Both lambdas are then used to calculate the sensitivity ($S_i$) of the invading species $i$ to the resident species $j$:

$$ Si = \frac{\lambda{i,Nj = 0} - \lambda{i,Nj^\ast}}{\lambda{i,N_j = 0}} $$

With the sensitivity of both species ($S_i$ and $S_j$) to the other, we can finally compute the two coexistence metrics: niche difference (ND) and relative fitness difference (RFD) following Carroll et al. 2011 and Narwani et al. 2013

$$ ND = 1 - \sqrt{S_i S_j} $$

$$ RFD = \sqrt{\frac{S_i}{S_j}} $$

Note that both ND and RFD can be computed with total population size ($N^\ast$) or total biomass ($BA^\ast$) instead of the invasion growth rate. Also note that both ND and RFD are in function of temperature and precipitation set to the optima value of the resident species $j$. Finally, they are also dependent on plot random effects that are defined to zero for the moment.

Output

Questions:

willvieira commented 1 year ago

Sensitivity of invader species i over resident speciesj at equilibrium within its optimal environment. I computed the sensitivity using three different metrics: i) initial lambda, ii) population size at equilibrium (N), and iii) population biomass at equilibrium (BA).

Screen Shot 2023-07-25 at 12 32 35 PM Screen Shot 2023-07-25 at 12 31 30 PM Screen Shot 2023-07-25 at 12 30 25 PM
willvieira commented 1 year ago

Using BA among the three metrics (reason to be described), next figure shows the distribution of BA sensitivity across species and their shade tolerance.

Screen Shot 2023-07-26 at 3 30 28 PM
willvieira commented 1 year ago

BA sensitivity can be seen as the interspecific competition parameter $\alpha{ij}$. Using the following formula, we can compare the BA sensitivity to $\alpha{ij}$:

$$ \alpha{ij} = log( \frac{BA{i,Nj^\ast}}{BA{i,N_j=0}}) $$

Screen Shot 2023-07-27 at 1 30 28 AM
willvieira commented 1 year ago

Using the carrying capacity ($K = BA_{i,Nj = 0}$) to define $\alpha{ii} = 1/K$:

Screen Shot 2023-07-26 at 3 59 12 PM