On the use of climate covariates in aquatic species distribution models: are we at risk of throwing out the baby with the bath water?

• DOI: 10.1111/ecog.03134
• Publication Date: 2017-06-05T06:06:47Z
• Authors: Daniel J. McGarvey; Mitra Menon; Taylor Woods; Spencer Tassone; Jessica Reese; Marie Vergamini; Erik Kellogg
• Keywords: Species distribution

Species distribution models (SDMs) in river ecosystems can incorporate climate information by using air temperature and precipitation as surrogate measures of instream conditions or independent water hydrology to link habitat. The latter approach is preferable but constrained the logistical burden developing models. We therefore assessed whether regional scale, freshwater SDM predictions are fundamentally different when data versus used covariates. Maximum entropy (MaxEnt) SDMs were built for 15 fishes one two covariate sets: 1) (climate variables) combination with physical habitat variables; 2) temperature, (instream Three procedures then compare results from vs First, equivalence tests average pairwise differences (site‐specific comparisons throughout each species’ range) among Second, ‘congruence’ determined how often same stream segments assigned high suitability Third, Schoener's D Warren's I niche overlap statistics quantified range‐wide similarity predicted Equivalence revealed small, between (mean MaxEnt raw scores all species < 3 × 10 –4 ). Congruence showed a strong tendency predict at (median congruence = 68%). reflected margin 0.78, median 0.96). Overall, we found little support hypothesis that covariates model fish distributions scale Columbia Basin.