Figure 2

UPRLIMET (a) generalized development workflow and (b) prediction workflow. We constructed and compared 26 models to select the top performing model, termed UPRLIMET, based on the lowest overall error between observed and predicted upper extent of trout distributions across western Oregon. (a) Generalized development workflow for UPRLIMET, a single logistic regression model fit to trout occurrence observation data. Stage 1 involved fitting the 4-variable logistic regression to the occurrence observation data. Stage 2 included implementing Stopping Rule #1 (Fig. S1). (b) Generalized prediction workflow where the two-stage UPRLIMET prediction process is applied to all HUC12s in our study area producing a trout distribution map. The four environmental predictor variables in a and b are characterized at the scale of the individual reach (5–7 m) and derived from a 5-m LiDAR-derived digital elevation model (Data S1).