IPMFunctions

Functions required to run the CzIPM.R script. This script contains the description of the growth, survival and fecundity functions used to build the IPMs.1. The best-fitted model for survival (σ) was a logistic mixed effect model including size as fixed factors and population nested in years as a random factor. 2. For growth (γ), the best-fitted model was a linear mixed effect model, with size as fixed factor and population nested in year as random factor. 3. Fertility (φ(z)), was estimated as the relation between reproductive status (reproductive vs. non-reproductive) and size with a binomial regression. 4. Recruitment per capita (δ(N)) is density-dependent in C. zosteroides (Capdevila et al., 2015), so a generalized linear model with Poisson error distribution and a log-link function was fitted, correlating the recruit:adult ratio as a function of the adult density. 5. To model the effect of temperature on the probability of settlement (ε) we used a generalized linear mixed models (GLMM), with a Poisson error distribution and a logit link function, the independent variable was the number of zygotes, temperature was treated as a fixed variable and we used the ID of each quadrat of the Petri dishes as a random variable. 6. To model the effect of temperature and time (fixed factors) on germling survival (σs), we used a GLMM with a binomial error distribution and a logit link function, with the ID of each quadrat of each Petri dish as a random variable to deal with the lack of independence between observations repeated at different times and a binomial error distribution was assumed to deal with the binary response variable (survive vs. die). 7. The size distribution of recruits was estimated as a normal probability function. In addition, the function required to project the density-dependent and stochastic IPMs is provided.