To elaborate our model we removed uncertain localities such as those localities with no coordinates and all records with coordinates laying down in sites with estimated elevations below 318 m and above 1,100 m. Also two records in BioMap from 'rio Putumayo', which represent a locality with a highly uncertain georeferencing in the database were deleted before modelling.
The habitat suitability model generated in Maxent showed extensively areas suitable in climatic terms for this species in the Pacific, the southern Amazon and the head of the Magdalena valley . those areas are not known to be occupied by the species and were excluded from the potential distribution map of this hummingbird.
Assuming that the distribution of the species may have filled the complete climatic model generated, its distribution today in remnants of forest is about 5,053 km2, which corresponds to a loss of 33 % of its potential original distribution due to deforestation.
This near-endemic species has been catalogued by BirdLife International (2016) as Vulnerable (VU) because it is believed that its population will decline quickly in the next three generations given the known models of deforestation of the Amazon basin. It is considered to be uncommon and local (McMullan & Donegan 2014) or rare (HBW Alive 2016) and its extent of occurrence has been estimated in 30,700 km2 (BirdLife International 2016). Our maps suggest its extent of occurrence just in Colombia is about 7,589 km2. Although, forested areas have not been greatly degraded in is potential original geographical distribution as our analyses suggest, forest degradation and fragmentation continues along its potential range and therefore we believe it must be considered as Endangered (EN) at national level in Colombia. This species is little known and needs urgently research of its range and general ecology.
Regularized training gain is 1.386, training AUC is 0.960, unregularized training gain is 1.899.
Algorithm converged after 220 iterations (1 seconds).
The follow settings were used during the run:
5 presence records used for training.
10004 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: 0.575, threshold: 1.950, hinge: 0.500
Feature types used: linear
'Equal Training Sensitivity and Specificity' and 'Equate Entropy of Thresholded and Original Distributions' thresholds and omission rates:
0.096-0.250-Fractional predicted area
0.000-0.000-Training omission rate