To elaborate the habitat suitability model we removed a few records with uncertain localities. Also, coordinates from the record 302831, a sighting from La Paya, in DatAves were moved west a few centecimals of a degree to make them fit in the mask of Colombia.
The habitat suitability model generated in Maxent showed several areas that are suitable in climatic terms for this species west of the Andes. These areas are not known to be occupied by this chachalaca and were excluded from its potential distribution map. On the other hand, there are vast areas in the Orinoco and Amazon regions that were not predicted as suitable by our model, likely because of the lack of data from those zones. Nonetheless, it is suspected the species occurs throughout because of the existence of records in the central northwestern Orinoco region and in the border with Ecuador, in the southern portion of the Colombian Amazon (EBIRD, 2015).
Assuming that the distribution of the species may have filled the complete climatic model generated in the Colombian Amazon, its potential distribution today in remnants of forest is about 469,734 km2, which corresponds to a loss of 9 % of its potential original distribution in the region due to deforestation.
Regularized training gain is 1.055, training AUC is 0.905, unregularized training gain is 1.541.
Algorithm converged after 1200 iterations (32 seconds).
The follow settings were used during the run:
36 presence records used for training.
10035 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: 0.233, categorical: 0.250, threshold: 1.640, hinge: 0.500
Feature types used: linear quadratic hinge
'Equal Training Sensitivity and Specificity' and 'Equate Entropy of Thresholded and Original Distributions' thresholds and omission rates:
0.194-0.348-Fractional predicted area
0.194-0.028-Training omission rate