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 1,100 m and above 2,390 m. Also four records in BioMap, three from Ricaurte (Cundinamarca) and one from Santander, which represent corrupted data in the database were deleted. Records from Ricaurte (Cundinamarca) possibly confused with Ricaurte (Nariño) and therefore assigned the wrong locality.
The habitat suitability model generated in Maxent exhibited low specificity, showing a vast extension of areas suitable in climatic terms for this species along the Andes east of the Pacific slope, except the eastern slope of the Eastern Andes (not predicted as suitable). Also were predicted as suitable adjacent areas in the Pacific lowlands, lowlands in the southwestern Caribbean and in the far east Amazon. Those areas are not known to be occupied by this species and were deleted from our final potential distribution map.
Assuming that the distribution of the species may have filled the complete climatic model generated, its distribution today in remnants of forest is about 5,592 km2, which corresponds to a loss of 62 % of its potential original distribution due to deforestation.
This species has been categorised as Near Threatened (NT) according to BirdLife International (2016) because despite the small distribution and suspected low numbers allegedly its range is not very fragmented or limited to a few locations. BirdLife does not have an estimation of the extent of occurrence of this species. Our maps suggest that just in Colombia the extent of occurrence of the species is 14,772 km2, which is greater than what had been previously assumed. Contrary to what has been suggested by BirdLife, our analysis suggest the forested habitat of this species has been severely transformed and fragmented and given the fact that the extent of occurrence is below the threshold of 20,000 km2, this species must be recategorised as Vulnerable (VU).
Regularized training gain is 1.286, training AUC is 0.940, unregularized training gain is 1.651.
Algorithm converged after 140 iterations (1 seconds).
The follow settings were used during the run:
8 presence records used for training.
10008 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: 0.530, threshold: 1.920, hinge: 0.500
Feature types used: linear
'Equal Training Sensitivity and Specificity' and 'Equate Entropy of Thresholded and Original Distributions' thresholds and omission rates:
0.125-0.276-Fractional predicted area
0.125-0.000-Training omission rate