To elaborate our model we removed uncertain localities such 'Bogota skins', those localities with no coordinates and all records with coordinates laying down in sites with estimated elevations below 2,700 m and above 3,750 m.
The habitat suitability model generated in Maxent showed some areas suitable in climatic terms for this species in the northern half of the Western Andes, the northern end of the Central Andes and in the most southern portion of the Eastern Andes. Those areas are not supported by any records, however, we decided to leave them in our final potential distribution map since those do not represent extensive areas from where it is not known the species and can be potential areas to look for the species, particularly in the southern Eastern Andes.
Assuming that the distribution of the species may have filled the complete climatic model generated, its distribution today in remnants of forest is about 9,476 km2, which corresponds to a loss of 76 % of its potential original distribution due to deforestation. Nonetheless, this species favours edges and therefore possibly deforestation has not negatively affected greatly its populations.
Regularized training gain is 3.101, training AUC is 0.988, unregularized training gain is 3.340.
Algorithm converged after 1160 iterations (38 seconds).
The follow settings were used during the run:
96 presence records used for training.
10096 points used to determine the Maxent distribution (background points and presence points).
Environmental layers used (all continuous): bio10co bio11co bio12co bio13co bio14co bio15co bio16co bio17co bio18co bio19co bio1co bio2co bio3co bio4co bio5co bio6co bio7co bio8co bio9co
Regularization values: linear/quadratic/product: 0.079, categorical: 0.250, threshold: 1.040, hinge: 0.500
Feature types used: hinge product linear threshold quadratic
'Equal Training Sensitivity and Specificity' and 'Equate Entropy of Thresholded and Original Distributions' thresholds and omission rates:
0.049-0.045-Fractional predicted area
0.052-0.052-Training omission rate