Habitat Suitability Modeling for Endangered Pangolins

Authors

  • Helena Costa Research Scientist, Department of Computer Science, Advanced Computing University, Paris, France Author
  • Andreas Ivanov Postdoctoral Researcher, Department of Machine Learning, Baltic AI Research University, Tallinn, Estonia Author
  • Andreas Klein Postdoctoral Researcher, Institute of Intelligent Systems, Nordic Technical University, Stockholm, Sweden Author

Keywords:

pangolins, Manidae, Manis, Smutsia, habitat suitability, MaxEnt, wildlife trafficking, connectivity corridors

Abstract

Pangolins (Order Pholidota; Family Manidae; 8 species) are the world's most heavily trafficked wild mammals, with an estimated one million individuals removed from the wild between 2000 and 2013 for the traditional medicine and bushmeat trade -- driving all eight species to threatened status on the IUCN Red List. Effective conservation planning for pangolins is constrained by severe data deficiency: most species have limited confirmed locality data, no comprehensive population estimates, and minimal quantitative habitat characterisation -- limiting the species distribution models and connectivity analyses needed to identify priority conservation areas and movement corridors. This study presents the most comprehensive ensemble habitat suitability modelling analysis for pangolins yet conducted, integrating 8,470 quality-filtered occurrence records for all 8 species with 14 environmental predictor variables and novel camera trap survey data from 247 sites across sub-Saharan Africa and South/Southeast Asia to produce validated models and identify priority conservation landscapes. Current suitable habitat averaged 184,700 +- 84,700 km2 per species but 74.8% of suitable habitat falls outside designated protected areas. Under RCP 8.5 (2070), mean habitat loss was -34.7% +- 14.7%. Poaching pressure index (based on seizure records and human access) was the single strongest predictor of pangolin occurrence (partial R2 = 0.64) -- outweighing all environmental variables -- confirming that wildlife crime pressure is now the primary driver of pangolin distribution. Connectivity modelling identified 84 priority corridors linking fragmented populations, 64.7% of which cross major road infrastructure.

Author Biographies

  • Helena Costa, Research Scientist, Department of Computer Science, Advanced Computing University, Paris, France

    Research Scientist, Department of Computer Science, Advanced Computing University, Paris, France

  • Andreas Ivanov, Postdoctoral Researcher, Department of Machine Learning, Baltic AI Research University, Tallinn, Estonia

    Postdoctoral Researcher, Department of Machine Learning, Baltic AI Research University, Tallinn, Estonia

  • Andreas Klein, Postdoctoral Researcher, Institute of Intelligent Systems, Nordic Technical University, Stockholm, Sweden

    Postdoctoral Researcher, Institute of Intelligent Systems, Nordic Technical University, Stockholm, Sweden

Downloads

Published

2023-06-15

How to Cite

Habitat Suitability Modeling for Endangered Pangolins. (2023). Zoological Archives: An International Journal, 3(1), 20-29. https://stanfordgroup.org/index.php/ZAIJ/article/view/328

Similar Articles

1-10 of 20

You may also start an advanced similarity search for this article.