Lepidopyga lilliae


To elaborate our model we removed uncertain localities such as those localities with no coordinates and all records with coordinates laying down in sites with estimated elevations above 51 m.

The habitat suitability model generated in Maxent showed a few areas suitable in climatic terms for this species in the interior of the Alta Guajira, these areas are not known to be occupied by the species and were excluded from the potential distribution map of this hummingbird.

Assuming that the distribution of the species may have filled the complete climatic model generated, its distribution today in remnants of forest is about 202 km2, which corresponds to a loss of 94 % of its potential original distribution due to deforestation.

This Colombian endemic has been catalogued as Critically Endangered (CR) by BirdLife International (2017) given its very small range and rarity. Its habitat has been severely degraded and it may continue being disturbed. It range size has been estimated in 43,700 km2, which may be an error. Our maps suggest a much smaller extent of occurrence of 3,698 km2, though not including the new area where the species has been apparently recorded in Bocas del Atrato in Uraba (BirdLife International, 2017).

This species is not well known and understood and some authors believe it may be a race, morph or an age stage of the more common Lepidopyga coeruleogularis (McMullan & Donegan 2014; HBW Alive, 2017). It needs urgently studies of its ecology and genetics to define better threats to its populations and its taxonomic status.


Regularized training gain is 4.872, training AUC is 0.997, unregularized training gain is 5.201.

Algorithm converged after 540 iterations (20 seconds).

The follow settings were used during the run:

18 presence records used for training.

10018 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: 0.481, categorical: 0.250, threshold: 1.820, hinge: 0.500

Feature types used: hinge linear quadratic

responsecurves: true

jackknife: true

maximumiterations: 2000

'Fixed Cumulative Value 10%', 'Equal Training Sensitivity and Specificity' and 'Equate Entropy of Thresholded and Original Distributions' thresholds and omission rates:


10.000-4.670-9.130-Cumulative threshold

0.076-0.009-0.048-Logistic threshold

0.007-0.027-0.008-Fractional predicted area

0.056-0.000-0.056-Training omission rate