Fig 1.
Georeferenced locations of 96 presence points (black dots) and the predicted distribution (gray shading) of Ornithodoros hermsi and Borrelia hermsii in western North America under current climate conditions.
Created in ArcMap (10.2, ESRI, Redlands, CA; http://desktop.arcgis.com/en/arcmap/) using environmental data from WorldClim version 1.4 (http://www.worldclim.org) [42].
Fig 2.
Jackknife analysis of the top environmental predictors used in the model.
Jackknife tests of variable importance for 10 replicate models were performed using only single variables (light gray bars) and without each variable (dark gray bars). The corresponding decrease in AUC was measured; large decreases in AUC illustrate model dependence on variables. BIO5 = maximum temperature of the warmest month, BIO6 = minimum temperature of the coldest month, BIO8 = mean temperature of the wettest quarter, BIO7 = annual temperature range, BIO15 = precipitation seasonality, BIO18 = precipitation of the warmest quarter, elv = elevation.
Table 1.
Estimates of the relative contribution of elevation and six uncorrelated environmental variables used to model the distribution of Ornithodoros hermsi and Borrelia hermsii in western North America.
Fig 3.
Logistic response curves for the occurrence of Ornithodoros hermsi and Borrelia hermsii in western North America.
The mean (dark gray line) ± one standard deviation (light gray shading) of the 10 replicate Maxent models.
Fig 4.
Model consensus for the predicted distribution of Ornithodoros hermsi and Borrelia hermsii under climate change in 2050.
The map shows model consensus among three global climate models (GCMs)—ACCESS1-0, HadGEM2-ES, and CCSM4—and two estimates of greenhouse gas (GHG) concentration trajectories: A) RCP 4.5, a medium estimate of GHG concentrations, and B) RCP 8.5, a high estimate of GHG concentrations. Created in ArcMap (10.2, ESRI, Redlands, CA; http://desktop.arcgis.com/en/arcmap/) using environmental data from WorldClim version 1.4 (http://www.worldclim.org) [42].
Fig 5.
Comparison of current versus future distribution of Ornithodoros hermsi and Borrelia hermsii predicted by Maxent.
The map shows the current predicted distribution in comparison to the consensus of all three global climate models (GCMs)—ACCESS1-0, HadGEM2-ES, and CCSM4—under two scenarios of greenhouse gas (GHG) concentrations, RCP 4.5 and 8.5, representing medium and high estimates of GHG concentrations, respectively. Created in ArcMap (10.2, ESRI, Redlands, CA; http://desktop.arcgis.com/en/arcmap/) using environmental data from WorldClim version 1.4 (http://www.worldclim.org) [42].
Fig 6.
Change in distribution of elevation ranges predicted by each model.
Global climate models for 2050 exhibit a shift of suitable climate to higher elevations.
Table 2.
Total amount of land area predicted as suitable for Ornithodoros hermsi and Borrelia hermsii in western North America under current climate conditions, as well as global climate modeled under two greenhouse gas (GHG) concentration trajectories for the year 2050.
Three global climate models (GCMs)—ACCESS1-0, HadGEM2-ES, and CCSM4—and two GHG concentration pathways, RCP 4.5 and 8.5, representing medium and high estimates of GHG concentrations, respectively, were utilized.