Research report: Integrating MaxEnt and MaxNet for climate-resilient habitat restoration: Predicting the ecological niche of Eucalyptus tereticornis in semi-arid landscapes

Choosing the right algorithm can shape how conservationists restore degraded landscapes, and a new comparative study shows just how much that choice matters. Researchers modeled the ecological niche of Eucalyptus tereticornis, a widely planted tree in semi-arid agro-climatic regions, using two prominent presence-only species distribution models: Maximum Entropy (MaxEnt) and its machine-learning counterpart MaxNet. The team drew on 138 georeferenced occurrence records supplemented by 25 pseudo-absence points, incorporating environmental variables such as wind speed, temperature range, precipitation, and root zone soil wetness.

Both models were optimized and evaluated with accuracy, precision, recall, F1-score, and the Area Under the Curve (AUC). MaxEnt outperformed MaxNet on most metrics, achieving an accuracy of 87.5 percent and an AUC of 0.956, compared with MaxNet's accuracy of 75 percent and AUC of 0.926. The comparison also revealed how algorithm design influences ecological inference.

Variable importance differed between the models: MaxEnt highlighted surface pressure as the key driver of predicted suitability, while MaxNet emphasized wind speed. Probability density plots added a further nuance. MaxEnt distinctly identified high-suitability zones, whereas MaxNet produced a more balanced, conservative distribution across the landscape.

The study underscores the importance of model selection and careful tuning in species distribution modeling. Because MaxEnt and MaxNet weight environmental responses differently, integrating outputs from both can yield more robust insights for habitat suitability assessment than either model alone. The findings carry practical weight for afforestation planning under changing climatic conditions.

Data-driven suitability maps can guide where plantations of Eucalyptus tereticornis are most likely to establish successfully, supporting land-use management and biodiversity conservation in semi-arid agro-ecological zones. The researchers note that their approach offers valuable guidance for sustainable plantation efforts in similar environments, while the observed performance gap between the two models serves as a caution: predictions of where a species can thrive depend not only on environmental data but on the statistical machinery used to interpret it.

Subject of Research: Comparative species distribution modeling of Eucalyptus tereticornis habitat using MaxEnt and MaxNet in a semi-arid region

Article Title: Integrating MaxEnt and MaxNet for climate-resilient habitat restoration: Predicting the ecological niche of Eucalyptus tereticornis in semi-arid landscapes

Article References: Kumar, P., Paul, M., Kumari, P., Dobriyal, M., Lavania, P., Srivastav, M., Pandey, P. C., & Turyasingura, B. (2026). Integrating MaxEnt and MaxNet for climate-resilient habitat restoration: Predicting the ecological niche of Eucalyptus tereticornis in semi-arid landscapes. PLOS One, 21(10), e0360385. https://doi.org/10.1371/journal.pone.0360385

Image Credits: AI Generated

DOI: 10.1371/journal.pone.0360385

Keywords: species distribution modeling, MaxEnt, MaxNet, Eucalyptus tereticornis, semi-arid landscapes, ecological niche, afforestation, habitat suitability, AUC, climate resilience, Integrating, climate-resilient