粗皮桉在我国的潜在适生区与主导生态因子研究

Study on Potential Suitable Area for Eucalyptus pellita in China and Key Factors Influencing This

本文基于粗皮桉的实地栽培点分布数据,结合气候、地形、土壤因子数据集,运用最大熵模型(MaxEnt)预测了粗皮桉在中国的潜在适生区。结果表明:MaxEnt的预测准确性较高,模型预测的训练子集和测试子集AUC值均大于0.855。粗皮桉适生区分布集中在东南沿海,最适生区总面积53522 km^2,集中在广东西部和东部沿海,广西中部,福建和海南沿海;总适生面积145 655 km^2,占研究区总面积的5.05%。刀切法分析结果表明,海拔、温度变化方差、最冷月份最低温度、最暖季度降水量、坡向5个因子是影响粗皮桉分布的主导生态因子,累积贡献率为85.7%。中国适生区的温度变化方差与自然分布区的相似性较强;与自然分布区相比,我国适生区的海拔更低、最冷月份最低温度更高、最暖季度降水量更大,有利于粗皮桉的生长。

The areas geographically suitable for Eucalyptus pellita plantations in China were analyzed using maximum entropy modeling technology in this study, by combining data on climate, topography and soil factors. The results showed that the MaxEnt model is able to predict the potential distribution of E. pellita with a high precision – the correspondence between the training data and test data was over 0.855. The majority of area suitable for this species is concentrated in China＇s southeast coastal region, with most of the highly suitable areasbeing concentrated in the western and eastern coastal regions of Guangdong, central regions of Guangxi and coastal regions of both Fujian and Hainan, covering a total area of 53522 km^2. The total suitable area for the species in China totals 145655 km^2, which is 5.05% of the total research area. Jackknife analysis of MaxEnt showed that the dominant ecological factors influencing the potential geographic distribution of E. pellita are altitude, seasonal temperature variance, minimum temperature of the coldest month, precipitation of the warmest quarter and topographical aspect; together these factors account for 85.7% of the variation in suitable areas. Seasonal temperature variance of the suitable area in China is similar to that found in E. pellita＇s native habitat.But, compared with the latter, suitable areas in China have lower altitudes, higher minimum temperatures of the coldest month and greater precipitation of the warmest quarter – the latter being a trait favorable for the rapid growth of E. pellita.