To elaborate our models we removed records with coordinates laying down in sites with estimated elevations below 470 m and above 2,000 m.
The habitat suitability model generated in Maxent showed a very few areas suitable in climatic terms for this hummingbird in the Pacific slope of the West Andes in the south, and in the far east Amazonia in the region of the rio Negro. These areas are not known to be occupied by this hummingbird and were deleted from our final potential distribution map. Furthermore, an adjunct region climatically suitable in the head of the Magdalena Valley may not be part of the distribution of this species since there are no known records from that area. Nonetheless, given the geographical proximity it was left in our final potential distribution map, although highlighting this situation.
Assuming that the distribution of the species may have filled the complete climatic model generated, its distribution today in remnants of forest is about 4,035 km2, which corresponds to a loss of 42 % of its potential original distribution due to deforestation. It is important to highlight that most of the loss of habitat has occurred in the area of the head of the Magdalena valley, which as we said in the previous paragraph its unlikely part of the distribution of this species in Colombia.
This hummingbird is a near-endemic species, distributed in Ecuador, northeast Peru and south central Colombia. Globally, it is considered Vulnerable (VU). Nevertheless, at country level in Colombia given its very restricted distribution and the projections of deforestation of the Amazon basin (BirdLife International 2016) its category might be Endangered (EN).
Regularized training gain is 4.578, training AUC is 0.998, unregularized training gain is 5.260.
Algorithm converged after 500 iterations (2 seconds).
The follow settings were used during the run:
11 presence records used for training.
10011 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: 0.757, categorical: 0.464, threshold: 1.890, hinge: 0.500
Feature types used: linear quadratic
'Equal Training Sensitivity and Specificity' and 'Equate Entropy of Thresholded and Original Distributions' thresholds and omission rates:
0.007-0.010-Fractional predicted area
0.000-0.000-Training omission rate