Experimental Evaluation of Koala Scat Persistence and Detectability with Implications for Pellet-Based Fauna Census

Table 4

Model selection for explaining scat decay rate during 36 weeks, all models were fitted with interval censored survival model with the 48 sites as frailty effect.

Models	AICc	ΔAICc	AICc weight	Evidence ratio
Rain + litter + flooded	3633.4	0	0.259	1
Rain + litter + flooded + elevation	3633.5	0.1	0.246	1.05
Rain + litter + detritivores + flooded + elevation	3633.6	0.2	0.234	1.11
Rain + litter + detritivores + flooded	3633.9	0.5	0.202	1.28
Vegetation + rain + litter + detritivores + flooded	3639.6	6.2	0.012	22.20
Vegetation + rain + litter + detritivores + flooded + elevation	3639.8	6.4	0.011	24.53
Rain + detritivores + flooded	3639.9	6.5	0.010	25.79
Vegetation + rain + litter + flooded	3640.5	7.1	0.007	34.81
Rain + detritivores + flooded + elevation	3640.6	7.2	0.007	36.60
Vegetation + rain + litter + flooded + elevation	3641.3	7.9	0.005	51.94
Vegetation + rain + detritivores + flooded + elevation	3644.3	10.9	0.001	232.76
Vegetation + rain + detritivores + flooded	3644.9	11.5	0.001	314.19
Rain + flooded	3644.9	11.5	0.001	314.19
Vegetation + rain + litter + detritivores + elevation	3645.3	11.9	0.001	383.75
Rain + litter + elevation	3645.7	12.3	0.001	468.72

AICc: akaike information criterion corrected for small sample size, ΔAICc: AICc differences, see descriptions in text (models with a weight <0.001 are not shown).