Table 2. Model choice procedure of the ABC approaches used for comparing demographic scenarios of A. ostoyae in the Landes de Gascogne.

	Scenario 1	Scenario 2	Scenario 3	Scenario 4	Scenario 5	Scenario 6	Scenario 7
ABC–RF
Votes (s.d.)	54.5 (± 5.6)	102.8 (± 7.7)	66.6 (± 8.0)	40.4 (± 6.4)	27.8 (± 3.9)	198.5 (± 8.9)	9.4 (± 2.8)
Post. Prob. (s.d.)	–	–	–	–	–	0.22 (± 0.05)	–
Prior error rate (s.d.)	–	–	–	–	–	73.8% (± 0.0%)	–
ABC-LDA
Post. Prob. (95% CI)	0.15 (0.15–0.16)	0.28 (0.27–0.28)	0.07 (0.07–0.08)	0.1 (0.09–0.10)	0.11 (0.10–0.11)	0.27 (0.26–0.27)	0.03 (0.03–0.03)
Performance
D1	15.2%	12.4%*	8.9%	6.9%	12.2%	10.3%§	5.6%
D2	12.1%†	23.0%	4.9%†	10.5%†	6.5%†	18.3%†§	2.1%†
D3	11.1%	5.7%*	13.1%	5.2%	12.2%	6.9%§	11.1%
D4	15.2%	24.7%*	11.4%	49.4%	6.3%	15.2%§	12.2%
D5	15.3%	9.70%*	16.1%	2.3%	31.2%	21.2%§	9.7%
D6	8.8%‡	11.5%*‡	4.7%‡	4.2%‡	8.3%‡	16.4%	2.0%‡
D7	22.3%	13.00%*	40.9%	21.5%	23.3%	11.7%§	57.3%

Abbreviations: ABC, approximate Bayesian computation; ABC-RF, ABC random forest; D, proportion of case in which the model choice procedure was able to select a scenario as the most probable with non-overlapping confidence intervals of the posterior probabilities of each scenario; Post. Prob., relative posterior probability for each scenario; s.d., s.d. over 10 replicate analyses.

The number of random forest votes of each scenario was averaged over 10 replicate analyses. For the ABC-RF analysis, the posterior probability and prior error rates of the best scenario were averaged over 10 replicate analyses. Type I or α error rates (risk to exclude the focal scenario when it is the true one) for scenario 2 and scenario 6 are indicated with ‘†’ and ‘‡’, respectively. Type II or β error rates (risk under scenario 2 (here: 9.1%) is marked with ‘*’ and under scenario 6 ‘§’. Demographic scenarios are shown in Figure 2.