There are two records from Nevado del Ruiz in Biomap and one from Putumayo that had wrong coordinates and was placed in the pacific lowlands. These records are clearly erroneous and were removed from the samples dataset used to model the habitat of this species. In spite of that, the habitat suitability model generated in Maxent showed areas that might be possibly occupied by this species west of the Andes in Sierra Nevada de Santa Marta, serrania del Baudo, the Pacific slope of the Western Andes and the northern end of the three cordilleras. These areas, are not occupied by the species and were removed from the potential distribution map.
Assuming that the distribution of the species may have filled the complete climatic model generated in the Amazon region, its distribution today in remnants of forest is about 420,431 km2, which corresponds to a loss of 9 % of its potential original distribution due to deforestation in that region. Overall, this estimate possibly may be higher if we account for other areas possibly making part of its distributions such as those in the eastern slope of the Eastern Andes.
Regularized training gain is 0.713, training AUC is 0.877, unregularized training gain is 1.133.
Algorithm converged after 600 iterations (17 seconds).
The follow settings were used during the run:
29 presence records used for training.
10029 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: 0.269, categorical: 0.250, threshold: 1.710, hinge: 0.500
Feature types used: linear quadratic hinge
'Equal Training Sensitivity and Specificity' and 'Equate Entropy of Thresholded and Original Distributions' thresholds and omission rates:
0.202-0.49-Fractional predicted area
0.207-0.069-Training omission rate