Minimal adequate multiple-predictor Generalized Linear Model (GLM) and Spatial Linear Model (SLM) results for all grid sizes.

The predictors are shallow water area (SW), coastline length (CL), habitat diversity based on area (HDIa), habitat diversity based on number of patches (HDIn), sea surface temperature (SST), and net primary productivity (NPP). Values under predictor variables are t-values for GLM and z-values for SLM. Asterisks indicate significance value of P: *(<0.05), **(<0.01); ***(<0.001); ns (not significant). Grid sizes are as follows: Small = 23,000 km2; Medium = 92,000 km2; Large = 368,000 km2; UTM = 617,000 km2; Extra large = 1,470,000 km2; Largest = 5,100,000 km2. R2 values are for GLM results, while pseudo-R2 (p-R2) values are for SLM results. The lower AIC values between SLM and GLM of the same grid size per group are highlighted in boldface. Lower AIC values suggest that SLM models are preferred in most cases than its GLM counterparts.