Skip to main content
. 2017 Oct 31;7:14424. doi: 10.1038/s41598-017-14774-2

Figure 4.

Figure 4

Species presence/absence data (1) and trait information (2) were used together with a GLMM to determine the presence/absence of each freshwater fish species. The final GLMM indicated that only few traits determined the probability of species prevalence (3). Predicted probabilities of presence/absence were input into a community assembly model that encompasses these trait values forming different resulting community assemblages (4). Each community represents differences in information entropy and community state, i.e. stochastic or deterministic (5).