高盆樱桃与钟花樱桃的地理分布模拟及生态特征比较分析

Modeling geographical distribution pattern and comparison of ecological characteristics between Cerasus cerasoides and C. campanulata

该研究以高盆樱桃( Cerasus cerasoides )及其近缘种钟花樱桃( C. campanulata )为对象,通过收集其实际分布的地理坐标,运用BIOCLIM模型模拟现代适生区范围并预测其未来气候下(CCM3,2100)潜在分布区的变化;结合主成分分析和相关性分析确定其主导气候因子,并比较两者在气候限制因子方面的差异;利用受试者工作特征曲线(Receiver operation characteristic, ROC)评估模型预测效果。结果表明:(1)高盆樱桃和钟花樱桃主要分布于我国长江以南大部分省区,两者现代分布中心分布位于云贵高原,武夷山脉和南岭山脉。(2)未来气候变化情境下(CCM3),高盆樱桃和钟花樱桃的适生范围将缩小。二者在中国西南部(高盆樱桃)和东南部(钟花樱桃)的适生范围可能大幅减小,而钟花樱桃在湖南西部的适生区可能增加。(3)主成分(PCA)及相关性分析表明,年降水量(bio12)、最湿季降水量(bio16)、最暖季降水量(bio18)、温度季节变化方差(bio4)是影响高盆樱桃及钟花樱桃当下适生区的主要气候因子,“热量变异幅度”是造成二者分布存在差异的最主要环境因子。(4)钟花樱桃(0.816)和高盆樱桃(0.799)的AUC值均高于随机测试(0.500),说明BIOCLIM模型可以准确预测高盆樱桃及钟花樱桃的分布。这为高盆樱桃和钟花樱桃资源保护、物种鉴定和谱系地理学的研究提供重要指导。

BIOCLIM model was used to simulate the range of modern suitable areas by collecting the geographical coordinates of its actual distribution and predict the change of their potential distribution under climate change (CCM3) for 2 100 s. The dominant climatic factors were determined by principal component analysis and correlation analysis. The differences between two species in climate limiting factors were analyzed. In addition, the receiver prediction effect was evaluated using the receiver operating characteristic (ROC). The results were as follows:(1) The actual distributions of C. cerasoides and C. campanulata covered most areas of the South of the Yangtze River in China. The contemporary distribution centers were located in the Yunnan-Guizhou Plateau, the Wuyi Mountain Range and the Nanling Mountains.(2) Under the future climate change situation (CCM3), the range of potential suitable areas of C. cerasoides and C. campanulata would decrease. The suitable habitat in Southwest ( C. cerasoides ) and Southeast China ( C. campanulata ) would be significantly reduced, while that of C. campanulata in West Hunan would increase.(3) Principal component analysis (PCA) and correlation analysis indicated that the annual precipitation (bio12), precipitation of the driest quarter (bio16), precipitation of the warmest quarter (bio18) and temperature seasonality (bio4) were the major factors influencing the geographical distribution of C. cerasoides and C. campanulata .“Radiator variability” was the most important environmental factor causing the difference in the distribution of two species.(4) Both of the AUC values of C. campanulata (0.816) and C. cerasoides (0.799) were higher than that of random test (0.500), which indicated that the BIOCLIM model could accurately predict the distribution of C. cerasoides and C. campanulata . Our results will be an important guide for the study of conservation, identification and phytogeography of C. cerasoides and C. campanulata .