To elaborate the habitat suitability model we added during the modelling exercise a record of a sighting from 'isla Miraña, Amazonas' (Renjifo et al., 2002).
The habitat suitability model generated in Maxent showed extensively areas that are suitable in climatic terms for this species in many areas of tropical Colombia. Using the known localities and the literature we limited the final potential distribution map to the areas predicted as marginally suitable, suitable and highly suitable that are south of the rio Caqueta.
Assuming that the distribution of the species may have filled the complete climatic model generated, its potential distribution today in remnants of forest is about 114,627 km2, which corresponds to a loss of 5 % of its potential original distribution due to deforestation. It is important to note that this model is limited due to the very small sample of localities from where this species is known in Colombia (three localities). Some authors limit its distribution in the country to the belts closer to the Caqueta and Putumayo rivers and the Amazon Trapezium (McMullan & Donegan, 2014), while others just to the surrounding areas near each of the three known localities (HBW Alive, 2015). The Colombian Amazon region still remains highly unexplored ornithologically, even more areas such as the southern Amazon region. Thus, it might be that further exploration will show many other localities where this curassow this threatened and rare curassow is present in Colombia within the distribution we have highlighted.
Regularized training gain is 0.625, training AUC is 0.913, unregularized training gain is 1.147.
Algorithm converged after 180 iterations (1 seconds).
The follow settings were used during the run:
3 presence records used for training.
10003 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: 0.605, threshold: 1.970, hinge: 0.500
Feature types used: linear
'Equal Training Sensitivity and Specificity' and 'Equate Entropy of Thresholded and Original Distributions' thresholds and omission rates:
0.155-0.535-Fractional predicted area
0-0-Training omission rate