Model selection statistics for Tasmanian devils and swamp rats.

The ten most supported multi-season two-species occupancy models based on AIC (Akaike’s Information Criterion) in program PRESENCE 4.0. Models that did not fall within the top ten that tested specific hypotheses are shown in bold below the top ten. The models were fitted to detection data from two study sites in cool temperate forests in southern Tasmania during standardized surveys from 2009–2011. The terms in parentheses represent the sources of variation in model parameters. ‘s’ denotes study site (Southwest and Tasman Peninsula), ‘h’ denotes habitat (dry eucalypt forest and woodland, wet eucalyptus forest and woodland, rainforest and related scrub, and agricultural exotic), ‘b’ denotes days since baiting, and ‘season’ denotes a model where detection varied with season. ‘.’ indicates a parameter set equal across species and survey times. The probabilities of site occupancy of devils and swamp rats were estimated independently, and the probability of site occupancy, colonization and extinction were constrained to be constant for both species and covariates for all models. Tasmanian devil Sarcophilus harrisii is abbreviated Sh and swamp rat Rattus lutreolus is abbreviated Rl.Delta AIC is the difference in AIC values between each model and the model with the lowest AIC.†AIC wt is the model weight.‡Number of parameters in the model.**Twice the negative log-likelihood.