Scholarlink

<p>Understanding the impact of climate change on species distribution is crucial for effective conservation planning. This study assessed the habitat suitability of <i>Vitex pseudo-negundo</i> under current and future climatic conditions using species distribution models based on multiple machine learning algorithms, including random forest (RF), support vector machine, generalized linear model, and multivariate adaptive regression splines. The RF model provided the most accurate predictions (area under the curve = 0.995), making it the most reliable approach for modeling species distribution. The most influential factors of <i>V. pseudo-negundo</i> distribution included BIO9 (“mean temperature of the driest quarter”) and soil electrical conductivity, whereas topographic variables such as topographic wetness index, aspect, and plan curvature had the least impact. Habitat suitability modeling indicated that 47.98% of the study area was classified as low suitability, whereas only 8.07% was highly suitable. Future climate projections (shared socioeconomic pathways (“SSP1-2.6” and “SSP5-8.5” scenarios) revealed significant habitat loss in the central region, whereas the western areas may become more suitable. By 2090, unsuitable habitats were projected to increase to 88.12% under the “SSP5-8.5” scenario, with only approximately 3% of the study area remaining suitable under both scenarios. These findings highlight the potential range shifts driven by climate change and reinforce the importance of integrating predictive modeling into conservation planning to ensure the long-term survival of <i>V. pseudo-negundo</i>. Given its medicinal and industrial value in pharmaceuticals and agroforestry, preserving <i>V. pseudo-negundo</i> in the face of climate change is crucial for its ecological and economic sustainability.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning-Based Projections of Medicinal and Industrial Plant Species Distribution: Conservation Insights for a Changing Climate

  • 期刊论文

  • 发表日期：

    2026年6月25日

查看全文
  • Musa Neyestani,
  • Atiyeh Amindin,
  • Soroor Rahmanian,
  • Gholamabbas Ghanbarian,
  • Roja Safaeian,
  • Hamid Reza Pourghasemi

摘要

Understanding the impact of climate change on species distribution is crucial for effective conservation planning. This study assessed the habitat suitability of Vitex pseudo-negundo under current and future climatic conditions using species distribution models based on multiple machine learning algorithms, including random forest (RF), support vector machine, generalized linear model, and multivariate adaptive regression splines. The RF model provided the most accurate predictions (area under the curve = 0.995), making it the most reliable approach for modeling species distribution. The most influential factors of V. pseudo-negundo distribution included BIO9 (“mean temperature of the driest quarter”) and soil electrical conductivity, whereas topographic variables such as topographic wetness index, aspect, and plan curvature had the least impact. Habitat suitability modeling indicated that 47.98% of the study area was classified as low suitability, whereas only 8.07% was highly suitable. Future climate projections (shared socioeconomic pathways (“SSP1-2.6” and “SSP5-8.5” scenarios) revealed significant habitat loss in the central region, whereas the western areas may become more suitable. By 2090, unsuitable habitats were projected to increase to 88.12% under the “SSP5-8.5” scenario, with only approximately 3% of the study area remaining suitable under both scenarios. These findings highlight the potential range shifts driven by climate change and reinforce the importance of integrating predictive modeling into conservation planning to ensure the long-term survival of V. pseudo-negundo. Given its medicinal and industrial value in pharmaceuticals and agroforestry, preserving V. pseudo-negundo in the face of climate change is crucial for its ecological and economic sustainability.