Abstract Vitellaria paradoxa, the shea tree, an economically important fruit-tree species native to savanna regions is threatened in Burkina Faso due to overexploitation and changing land-use. Furthermore, it remains unclear how climate change will influence its frequency and distribution. We investigated the impact of climate change on the projected spatial distribution of favorable habitats for V. paradoxa. Species distribution modeling techniques implemented in MaxEnt combined with GIS were used to forecast the current and future distribution of V. paradoxa. We selected two climatic scenarios (RCP4.5 and RCP8.5) and two global climate models (MPI-ESM-MR and HadGEM2-ES) to encompass the full range of variation in the models. Presence records of the species were collected and linked to bioclimatic and edaphic variables. The most characteristic and least correlated variables were selected for modeling after a collinearity test. Under current climatic conditions, ~51% of the national area was found to be favorable for cultivation and conservation of the species. Under future climate projections, our models predict that favorable habitats of this species will decline by 12% (RCP4.5) and 13% (RCP8.5) by 2070. The predictive modeling approach presented here may be applied to other economically important tree species.

中文翻译：

---

气候变化减少了布基纳法索乳木果树（Vitellaria paradoxa CF Gaertn.）的分布面积

摘要 Vitellaria paradoxa，乳木果树，一种原产于热带稀树草原地区的经济上重要的果树物种，由于过度开发和土地利用的变化，在布基纳法索受到威胁。此外，尚不清楚气候变化将如何影响其频率和分布。我们调查了气候变化对 V. paradoxa 有利栖息地的预测空间分布的影响。MaxEnt 中实施的物种分布建模技术与 GIS 相结合，用于预测 V. paradoxa 的当前和未来分布。我们选择了两个气候情景（RCP4.5 和 RCP8.5）和两个全球气候模型（MPI-ESM-MR 和 HadGEM2-ES）来涵盖模型中的全部变化范围。收集了该物种的存在记录并将其与生物气候和土壤变量联系起来。在共线性测试后，选择最具特征和最不相关的变量进行建模。在目前的气候条件下，大约 51% 的全国面积被发现有利于该物种的种植和保护。根据未来的气候预测，我们的模型预测，到 2070 年，该物种的有利栖息地将减少 12% (RCP4.5) 和 13% (RCP8.5)。此处介绍的预测建模方法可能适用于其他具有重要经济意义的树种.