Penelope purpurascens


To elaborate our model we removed uncertain localities such as 'Bogota skins' and all records with coordinates laying down in sites with estimated elevations above 2,300 m.

The habitat suitability model generated in Maxent showed a few areas that are suitable in climatic terms for this species far east of the Andes in eastern Guainia, eastern Vaupes and southeastern Amazonas. These areas are not known to be occupied by the species and were excluded from the potential distribution map of this guan. Otherwise, there are several areas predicted in the model in the eastern slope of the Eastern Andes from where there are not known records, but that might be likely since there are records north and south of this belt; and therefore these areas were included in the final distribution map.

Distribution of specimens, according to BioMap, suggests a possible area (≈ 4,393 km2) of intergradation between subspecies aequatorialis and brunnescens in between the mid-low Magdalena and low Cauca valleys in southern Bolivar and northeastern Antioquia. It might be that this area extends further to the east. Additionally, it is interesting to note that there is one specimen from La Candela, Huila identified as subspecies brunnescens, which may represent high individual variation in the populations of the Magdalena valley or indicate that intergrades reach far into this valley. However, with the distributional data we have at this time we cannot add anything else here; and this will need further revision in the future.

Assuming that the distribution of the species may have filled the complete climatic model generated, its distribution today in remnants of forest is about 158,878 km2, which corresponds to a loss of 67 % of its potential original distribution due to deforestation.


Regularized training gain is 0.984, training AUC is 0.893, unregularized training gain is 1.311.

Algorithm converged after 1220 iterations (37 seconds).

The follow settings were used during the run:

75 presence records used for training.

10073 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: 0.121, categorical: 0.250, threshold: 1.250, hinge: 0.500

Feature types used: linear quadratic hinge

responsecurves: true

jackknife: true

maximumiterations: 2000

'Equal Training Sensitivity and Specificity' and 'Equate Entropy of Thresholded and Original Distributions' thresholds and omission rates:


28.011-10.531-Cumulative threshold

0.386-0.158-Logistic threshold

0.187-0.375-Fractional predicted area

0.187-0.053-Training omission rate