基于MaxEnt模型对大花序桉 在我国南方的适生区预测

大花序桉材性优良是我国重要的实木利用树种,本文对大花序桉在我国南方十省(区)的潜在分布区进行预测,并对其适生区进行划定和分析.通过收集大花序桉的地理信息数据,利用最大熵模型(MaxEnt)和ArcGIS空间分析技术,综合相关气候因子,预测划定大花序桉在我国南方十省(区)的潜在地理分布区.结果表明:采用MaxEnt模拟大花序桉潜在地理分布的准确性较高,模型预测准确性的衡量指标训练子集和验证子集AUC值均大于0.88.大花序桉在我国南方的发展空间较大,其中最适宜分布区主要集中在广东、广西、海南、福建沿海,面积依次为广东(11.25×104 km2),广西(9.19×104 km2),海南(2.00×104 km2),福建(1.91×104 km2);适宜区集中在广西、广东、重庆、福建省中部,面积依次为广西(5.87×104 km2)、广东(4.51×104 km2)、重庆(3.64×104 km2)、福建(3.02×104 km2).刀切法(Jackknife)分析结果显示,影响大花序桉潜在适生区分布的主导气候因子为年平均气温、≥10℃积温、最冷月最低温、最冷月均温和极端低温.结合最大熵模型(MaxEnt)和ArcGIS技术,预测和分析了大花序桉在我国南方十省(区)的潜在适生区,阐明了影响其分布的主导气候因子,为该树种在我国的有序推广种植提供科学依据.

Predicting Potential Distribution of Eucalyptus cloeziana in Southern China Using a MaxEnt Model

This study was undertaken to determine the distribution of areas climatically suitable for Eucalyptus cloeziana plantations in ten southern Chinese provinces – this species is considered potentially important for production of solid wood in China on account of its superior wood properties. For this study, the current distribution of E. cloeziana and related climatic factors were collected. The potential suitable distribution range of E. cloeziana in ten southern Chinese provinces was then determined using Maximum Entropy Modeling and ArcGIS spatial analysis technology. The results showed that the accuracy of the MaxEnt model for predicting the potential distribution of E. cloeziana was high, with the correlation between a training dataset and test data being over 0.88. The area with suitable climates for E. cloeziana plantations in southern China was found to be extensive, and is mainly concentrated in the coastal provinces of Guangdong, Guangxi, Hainan and Fujian. The highly suitable area in each of these provinces amounts to: Guangdong (11.25×104 km2), Guangxi (9.19×104 km2), Hainan (2.00×104 km2) and Fujian (1.91×104 km2). Almost all the most suitable environments for E. cloeziana plantations are concentrated in the central regions of Guangxi, Guangdong, Chongqing and Fujian province, and include: Guangxi (5.87×104 km2), Guangdong (4.51×104 km2), Chongqing (3.64×104 km2) and Fujian (3.02×104 km2). Jackknife analysis showed that for regions climatically suitable for E. cloeziana plantations, the factors of annual mean temperatures, accumulated temperature of 10oC or more, average lowest temperature of the coldest month, average temperature of the coldest month and extreme lowest temperature are the dominant climatic factors affecting the potential distribution of E. cloeziana. The results from this study provide a reference for rationale future expansion of E. cloeziana plantation areas in China.