细叶桉在中国的潜在适生区预测

Potential Suitable Area of Eucalyptus tereticornis in China

本文使用细叶桉的实地分布点数据,结合气候、地形、土壤数据集,运用最大熵模型(MaxEnt)预测了细叶桉在我国的潜在适生区,分析了影响细叶桉分布的主要生态因子。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0.867;细叶桉适生区集中在东南沿海,最适生区总面积168 880 km^2,集中在广东大部分地区,广西南部和海南北部;总适生面积467 278 km^2,占研究区总面积的16.19%。刀切法分析结果表明,年均温、海拔、坡度、昼夜温差与年温差比值、最暖季度平均温度、土壤阳离子交换量6个因子是影响细叶桉分布的主导生态因子,累积贡献率达86.6%。我国适生区的年均温、最暖季度平均温度与自然分布区的相似性较强;与自然分布区相比,我国适生区的昼夜温差与年温差比值、海拔、坡度更低,利于细叶桉的生长。

A MaxEnt model was developed for predicting potentially suitable geographic areas for Eucalyptus tereticornis.This model was developed using data from locations of successful plantations of this species in China,along with climate,soil and topographic data.The dominant ecological factors affecting the potential distribution of E.tereticornis were also analyzed.The model developed in this study proved to have high precision the correspondence between the training data and test data was more than 0.867.The area suitable for this species is concentrated in the southeast coastal region of China,with the highly suitable areas being concentrated in the coastal Guangdong,southern regions of Guangxi and norther regions of Hainan.Highly suitable areas for E.tereticornis in China totaled more than 168 880 km^2,while the suitable area totals 467 278 km^2,which covers 16.19% of the research area.By the jackknife test,the dominant ecological factors influencing the potential geographic distribution of E.tereticornis are:annual mean temperature,altitude,slope,isothermality,mean temperature of the warmest quarter and soil cation exchange capacity.Together these factors account for 86.6%of the variation in suitable areas.Annual mean temperature and mean temperature of the warmest quarter proved to be similar to that found in native habitats of E.tereticornis.Compared with the latter,suitable areas in China feature lower isothermality,altitudes and slopes.