J. Wilson White and Jennifer E. Caselle. 2008. Scale-dependent changes in the importance of larval supply and habitat to abundance of a reef fish. Ecology 89:1323–1333.

Appendix C. A summary of the Bayesian analysis of older kelp bass survivorship.

TABLE C1. Summary of model selection for older (1+ year-old) kelp bass (Paralabrax clathratus) survivorship analysis. All models had a constant density-independent (DI) term and some combination of terms for older kelp bass density, giant kelp (Macrocystis pyrifera) stipe density, and the interaction between older kelp bass density and kelp stipe density. Survivorship was modelled as an exponential process; see text for model details. Model selection in this analyses utilized the Deviance Information Criterion, DIC, which is analogous to AIC but more appropriate for use with the output from Markov chain – Monte Carlo procedures such as the Metropolis-Hastings algorithm we used. ΔDIC is the difference in DIC between each model and the model with lowest DIC; Akaike weight, w, estimates the probability that each model is the most parsimonious model.

TABLE C2. Estimated mean values and 95% confidence intervals (CI) for parameters in most parsimonious model (selected using DIC) describing survivorship of older kelp bass. Survivorship was modelled as an exponential process; see text for model details. Kelp bass density is the number of older (1+ year-old) kelp bass per transect and Kelp is the number of giant kelp stipes per transect; both values were estimated in annual benthic surveys at n = 37 site-year combinations.