To elaborate our model we removed uncertain localities such as those localities with no coordinates, 'Bogota skins' and all records with coordinates laying down in sites with estimated elevations below 2,700 m and above 4,067 m. Also two records in BioMap from Caldas in the Central Andes, which represents corrupted information in the database.
The habitat suitability model generated in Maxent made excellent predictions, showing almost no areas suitable in climatic terms for this species out of the central Eastern Andes, except a small area in the southern Central Andes. That area is not known to be occupied by this hummingbird and was deleted from our final potential distribution map.
Regularized training gain is 3.707, training AUC is 0.992, unregularized training gain is 3.848.
Algorithm converged after 480 iterations (14 seconds).
The follow settings were used during the run:
58 presence records used for training.
10057 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: 0.170, categorical: 0.250, threshold: 1.420, 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:
0.023-0.024-Fractional predicted area
0.017-0.017-Training omission rate