To elaborate our models we removed uncertain localities such as 'Bogota skins', those localities with no coordinates and all records with coordinates laying down in sites with estimated elevations below 422 m and above 1,500 m.
The habitat suitability model generated in Maxent showed areas suitable in climatic terms for this hummingbird in a very few areas out from the eastern slope of the Eastern Andes such as west of serrania de San Lucas and in the far east Amazonia near the rio Negro. Those areas are not known to be occupied by this hummingbird and were deleted from our final potential distribution map. An important belt of areas south of Villavicencio at 422-1500 m, which might be possibly suitable for this species were not predicted by our model. Those areas were added to our final distribution map noting their status.
Assuming that the distribution of the species may have filled the complete climatic and elevational models generated, its distribution today in remnants of forest is about 17,163 km2, which corresponds to a loss of 44 % of its potential original distribution due to deforestation. Nonetheless, it is important to highlight that most of the area from where come the few specimens collected for this species near Villavicencio (Meta) and Medina (Cundinamarca) and to the north from them has been heavily deforested (see maps).
Regularized training gain is 3.695, training AUC is 1.000, unregularized training gain is 5.980.
Algorithm converged after 240 iterations (1 seconds).
The follow settings were used during the run:
2 presence records used for training.
10002 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: 0.620, threshold: 1.980, hinge: 0.500
Feature types used: linear
'Equal Training Sensitivity and Specificity' and 'Equate Entropy of Thresholded and Original Distributions' thresholds and omission rates:
0.001-0.025-Fractional predicted area
0.000-0.000-Training omission rate