Table 3.

Summary statistics for the top MLPE regression models

Predictors	logLik	AIC	ΔAIC	Weight	ρ
Temperature annual range***	22,813.91	−45,619.8	0.00	1	0.27
Inverted forest cover***, altitude***, terrain roughness*	22,807.03	−45,602.1	17.76	0	0.23
Inverted forest cover***	22,576.63	−45,145.3	474.56	0	0.25
Geographic distance	22,521.73	−45,035.5	584.37	0	0.27

All models contained interindividual genetic relatedness as response variable and the different landscape resistance distances as predictors. Log‐likelihoods are followed by the Akaike information criterion (AIC), ΔAIC, model weight, and the MLPE correlation coefficient rho (ρ). Isolation by geographic distance was included here for comparison.

Likelihood ratio tests: *p < 0.05, **p < 0.01, * p < 0.001