DATASET: TESTING NICHE EQUIVALENCE IN AMPHIDROMOUS FISH POPULATIONS

Presence records of Galaxias maculatus were collected from 12 locations across five river basins in central-southern Chile during March, May, August, and November of 2019 as part of a study on fish sampling and processing (Ramírez-Álvarez et al. 2022. doi.org/10.1038/s41598-022-06936-8)

Isotopic niches defined using a standard ellipse area (SEAc) in isotopic space, represented by a 2D ellipsoidal space (δ13C - δ15N) (see Supporting Information: Standard ellipse area functions - Ramírez-Álvarez et al. 2024. doi:10.1007/s10750-024-05738-5) - Empirical Bayesian Kriging

Database of varibles used for niche modelling: isotopic niche and seven abiotic variables selected from 23 predictor variables: (1) 19 climate variables representing 1950–2000 climate averages from WorldClim (http://www.worldclim.org/). (2) Four spatially continuous topographic and hydrological variables from the EarthEnv Project adjusted to the HydroSHEDS river network (http://www.earthenv.org/) (Domisch et al. 2015). Selection of variables was accomplished by analysis of covariance and multicollinearity, using the ENMeval R package (Muscarella et al. 2014): (1) principal component analysis (PCA) to explore relationships among all predictors, evaluating the composition of components (component variables) that accounted for ≥65% of variance explained, (2) pairwise comparisons  to detect pairs of variables with strong correlations (Pearson correlation coefficients <0.8), groups with a correlation of less than 0.8 were considered independent, and (3) variance inflation factor (VIF) <10, to reduce the effect of collinearity between predictors (Listed below). A VIF greater than 10 indicates collinearity problems in the model. The vifcor and vifstep functions were employed by calculating two different strategies to exclude highly collinear variables using a stepwise procedure (Muscarella et al. 2014).

Domisch, S., G. Amatulli & W. Jetz, 2015. Near-global freshwater-specific environmental variables for biodiversity analyses in 1 km resolution. Scientific Data 2(1):150073 doi:10.1038/sdata.2015.73.

Muscarella, R., P. J. Galante, M. Soley‐Guardia, R. A. Boria, J. M. Kass, M. Uriarte & R. P. Anderson, 2014. ENM eval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods in ecology and evolution 5(11):1198-1205