To elaborate our model we removed uncertain localities such as all records with coordinates laying down in sites with estimated elevations above 800 m. Also five records in ProAves, three from Antioquia and two from Magdalena, and a further two records in BioMap and DatAves from Choco, which represent corrupted data were removed previously to modelling.
The habitat suitability model generated in Maxent showed areas suitable in climatic terms for this species west of the Andes, particularly in the Pacific Region. These areas are not known to be occupied by the species and were excluded from the potential distribution map. Otherwise, a few areas east of the Andes that potentially are occupied by the species were not predicted by our model, these were highlighted as such in our maps.
Assuming that the distribution of the species may have filled the complete climatic model generated, its distribution today in remnants of forest is about 419,447 km2, which corresponds to a loss of 11 % of its potential original distribution due to deforestation.
Regularized training gain is 0.884, training AUC is 0.900, unregularized training gain is 1.451.
Algorithm converged after 580 iterations (15 seconds).
The follow settings were used during the run:
20 presence records used for training.
10020 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: 0.442, categorical: 0.250, threshold: 1.800, hinge: 0.500
Feature types used: hinge linear quadratic
'Equal Training Sensitivity and Specificity' and 'Equate Entropy of Thresholded and Original Distributions' thresholds and omission rates:
0.171-0.413-Fractional predicted area
0.150-0.100-Training omission rate