Answers are grounded in this paper and cite their sources.
Machine learning predicts habitat loss and increased extinction risk for Pachypodium species, indicating urgent conservation needs.
Global climate change poses unprecedented challenges to the maintenance and survival of biodiversity, with endemic species in particular regions facing an exceptionally high risk of extinction. Pachypodium , a genus endemic to South Africa and Madagascar, exhibits strong habitat specificity, yet the impacts of climate change on its distribution patterns remain not fully understood. This study employs the Biomod2 package in R to predict changes in the distribution patterns of 20 Pachypodium species under climate change scenarios (SSP2‐4.5 and SSP5‐8.5). Additionally, machine learning methods have been applied to assess the extinction risk of these species. The results indicate that climate change will severely impact the distribution of the genus Pachypodium . Suitable habitat areas for 15 species within the genus are projected to shrink significantly in the future, with the most pronounced habitat loss occurring in central and eastern Madagascar, eastern Namibia, and central and northern South Africa. Annual precipitation and precipitation seasonality are the main factors influencing these habitat changes. A reassessment of the IUCN categories for Pachypodium reveals that the number of threatened species will increase from 7 to 13. Alarmingly, three species are predicted to face a risk of extinction in the wild due to climate change. Moreover, the current protected areas have proven ineffective in safeguarding the habitats of Pachypodium , with protected habitats expected to decrease by 30.39% under the influence of climate change. These findings provide strategic insights for the conservation of Pachypodium species and highlight the necessity for reforms and adaptive adjustments to current protected area networks to address the challenges posed by climate change.
No takes yet. Share an insight, caveat, or question.
Chen et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: