Fig 1.
Map showing Mountain Tapir (Tapirus pinchaque) unique records (n = 155), overlaid with IUCN distribution and MaxEnt calibration and projection area.
Training localities (black dots) and validation localities (white dots) used to generate and validate the models. Dark brown color represents area with altitudes of up 1,000 masl.
Table 1.
Summary of the selected, no-correlated, environmental variables with relative contributions (%) to the model of Tapirus pinchaque.
Fig 2.
Potential distribution model of Tapirus pinchaque in Ecuador.
(A) Potential distribution model is shown with the threshold value of Fixed Omission Value 10 (FOV10, dark blue); (B and C) Remnant potential distribution model with natural forests (green areas), perturbed areas (red) and principal roads in Andes of Ecuador (black lines); (D) Remnant potential distribution model predicted for Protected Areas of Ecuador (yellow areas bordered by black) and perturbed areas (red). Training localities (black dots) and validation localities (white dots) used to generate models are shown in (A). Numbers in (D) correspond to: Cayambe—Coca National Park (1); Sumaco Napo-Galeras National Park (2); Antisana Ecological Reserve (3); Llanganates National Park (4); Sangay National Park (5); Podocarpus National Park (6); Chimborazo Faunistic Reserve (7); Los Illinizas Ecological Reserve (8); Cotopaxi National Park (9); and El Angel Ecological Reserve (10). Note an important reduction (~17%, in green) in the western Andes of the best-predicted potential distribution model (FOV10 threshold) when filtered to only include areas of natural forests (B), and a reduction of 52% when it was filtered to only include Protected Areas in Ecuador (D). Dark brown color represents area with altitudes of up 1,000 masl. The model was generated with the no-correlated environmental variables (Table 1).
Table 2.
Potential distribution and ecological niche models for Tapirus pinchaque, with percentage loss of extent of occurrence under differing global climate models (RCP 4.5/ RCP 8.5 scenarios), natural forest and Protected Areas (PAs) in the tropical Andes of Ecuador.
Table 3.
Potential distribution model of the Mountain Tapir, Tapirus pinchaque, in km2 (percentages), predicted for Protected Areas in the Ecuadorian Andes.
Table 4.
Ecosystem availability (km2) and suitability areas according to the potential distribution model (km2) and relative importance of ecosystem availability (REI = suitability/availability) calculated for Tapirus pinchaque in Ecuador.
Fig 3.
Potential ecological niche model of Tapirus pinchaque in two future scenarios of climate change based on the Global Climate Models of ACCESS 1.0.
Potential ecological niche model for the year 2050, under (A) RCP 4.5 (optimistic) and (B) RCP 8.5 (pessimistic) climatic change scenarios. Numbers correspond to those referred to for Protected Areas estimation in Fig. 1. Note a reduction (~22–38%, in red) of the best-predicted potential distribution model (FOV10 threshold). In both scenarios, suitability areas for Tapirus pinchaque tend to critically reduce, especially in the southern regions. Shifts are depicted from their centroids in the ecological niche-space.
Fig 4.
The role of Protected Areas in two future scenarios of climate change (based in ACCESS 1.0).
Potential distribution model in the year 2050, under RCP 4.5 (optimistic) and RCP 8.5 (pessimistic) climatic change scenarios. In both scenarios, we observed a reduction in the predicted geographic range for Tapirus pinchaque.
Fig 5.
Generalized Additive Models (A-C) and scatter plot for the relation between suitability probability and altitude, in climate change scenarios (D).
Non-parametric smoothers in GAM showing the multiple humped relationship between suitability probability and altitude, highlighting the possibility of minimum thresholds of altitude over ≈ 3,000 m (A-C). The etchings (small black lines) on the x axis in GAMs (A-C) indicate the density of probabilistic values located along the altitude, whereas centroids for the scattered data are represented as squares in (D).