TABLE 2.

Description of the 15 predictors implemented. Environmental covariates (0 denotes no used, while 1 denotes used), selection method of environmental covariates (C= Pearson´s correlation and B=Boruta). TC denotes threshold correlation and this takes values of 0.3, 0.4, 0.5, 0.6, and 0.7. The largest TC value was evaluated first and in case that any covariate satisfied this TC value was used the second largest and so on.

Model	Predictor	Environmental covariates	Selection method	Average of covariates	Correlation
M0	K.e + K.g + K.ge	0	-	0	-
M1	K.e + K.g + K.ge+ ${\mathit{X}}_e$	1	-	0	>0
M2	K.ec + K.g + K.gec	1	C	0	TC
M3	K.e + K.g + K.ge+ ${\mathit{X}}_e$	1	C	0	TC
M4	K.ec + K.g + K.gec+ ${\mathit{X}}_e$	1	C	0	TC
M5	K.e + K.g + K.ge+ ${\mathit{X}}_{e2}$	1	C	0	TC
M6	K.e + K.g + K.ge+ ${\mathit{X}}_{e.avg}$	1	C	1	TC
M7	K.ec + K.g + K.gec+ ${\mathit{X}}_{e.avg}$	1	C	1	TC
M8	K.ec + K.g + K.gec+ ${\mathit{X}}_{e.avg}$	1	C and B	1	TC
M9	K.e + K.ge + X.g.ec	1	B	0	-
M10	K.e + K.g + K.ge+ ${\mathit{X}}_e$ (Tenative true)	1	B	0	-
M11	K.ec + K.g + K.gec+ ${\mathit{X}}_{e.avg}$ (Tentative false)	1	B	1	-
M12	K.ec + K.g + K.ge+ ${\mathit{X}}_{e.avg}$ (Tenative true)	1	B	1	-
M13	K.ec + K.g + K.gec+ ${\mathit{X}}_{e.avg}$ (Tenative true)	1	B	1	-
M14	K.ec + K.g + K.gec+ ${\mathit{X}}_{e.avg}$ (Tenative False)	1	B	1	-