To elaborate our model we removed uncertain localities such as those localities with no coordinates, 'Santa Marta skins' and all records with coordinates laying down in sites with estimated elevations below 3,132 m and above 4,800 m.
The habitat suitability model generated in Maxent showed a few areas suitable in climatic terms for this species in Serrania del Perija and a spur north of the Sierra Nevada de Santa Marta. Areas north of the Sierra Nevada de Santa Marta are not known to be occupied by this hummingbird and were deleted from our final potential distribution map. Areas in Perija were left highlighting the fact that there are no known records from there.
This species, sometimes considered a subspecies of the complex Oxypogon guerinii-stubelii-cyanolaemus has been catalogued as Critically Endangered (CR) by BirdLife (2016). According to BirdLife (2016) its habitat has been severely degraded and it is believed its population is very small and may be decreasing. Nevertheless, the especies is very little known and there are no estimations of its population or its extent of occurrence. Our maps suggest an extent of occurrence of 1,444 km2, which is well below the threshold of 5,000 km2 to be categorised as Endangered (EN) and this combined with the fact that the paramo areas in Colombia are extremely fragmented and continue being burned and transformed for crops or cattle breeding, suggests this species must be currently correctly categorised as Critically Endangered (CR). However, there is great need for studies to know better its ecology.
Regularized training gain is 5.018, training AUC is 0.999, unregularized training gain is 5.362.
Algorithm converged after 260 iterations (1 seconds).
The follow settings were used during the run:
6 presence records used for training.
10006 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: 1.000, categorical: 0.560, threshold: 1.940, hinge: 0.500
Feature types used: linear
'Equal Training Sensitivity and Specificity' and 'Equate Entropy of Thresholded and Original Distributions' thresholds and omission rates:
0.002-0.007-Fractional predicted area
0.000-0.000-Training omission rate