Climate change risks and conservation needs of Lophira lanceolata in West Africa

---

Résumé

Lophira lanceolata is a multipurpose fruit tree species widely distributed across West Africa, but its populations are declining due to climate change and anthropogenic pressures. This study assessed the current and future distribution of L. lanceolata across West Africa under two Shared Socioeconomic Pathways - SSP2-4.5 and SSP5-8.5- for 2050 and 2070. An ensemble modelling framework combining five algorithms - Generalized Linear Models (GLM), Generalized Additive Models (GAM), Random Forest (RF), MaxEnt, and Boosted Regression Trees (BRT) - was employed. Habitat suitability projections and conservation priority areas were mapped and compared across models and scenarios. Results indicate that climate change will likely cause shifts and contractions in suitable habitats, with a general southward retreat under high-emission conditions. BRT predicted the largest extent of priority conservation areas (6–7 % of the study area), followed by GAM (5–5.4 %), GLM (~4%), MaxEnt (~3.6 %), and RF (~3–3.3 %). Under SSP5-8.5, BRT and GLM showed reductions in distribution areas, whereas GAM, RF, and MaxEnt predicted slight increases or stability. By 2070, BRT projected a significant decline of 5.38 % in suitable areas, while GAM indicated an increase of 2.80 %. These discrepancies highlight the importance of using multiple algorithms to capture model uncertainty. The spatial variability in priority zones identifies key areas for long-term conservation. Findings support strategies such as legal protection, assisted migration, and ecological restoration to enhance the resilience of L. lanceolata. Overall, this study underscores the value of integrative modelling for evidence-based biodiversity conservation in tropical drylands.

Mots-clés

Modelling ; Lophira lanceolata ; Threatened species ; Theoretical reference ; Sustainable management