摘要
通过物种分布模型预测入侵物种适生区,优先选择物种最可能扩散的地区,提早进行防控,是生物入侵管理的有效手段。南美蟛蜞菊是一种具有严重危害性的入侵植物,已在中国南方部分地区形成入侵。基于南美蟛蜞菊在美洲和中国海南的分布记录和19个生物气候变量,应用MaxEnt模型对其适生区进行建模,并通过计算AUC(受试者工作特征曲线下的面积)/AICc(小样本下校正的赤池信息标准)指数选择MaxEnt的最佳设置,用于预测南美蟛蜞菊在中国当前及未来气候变化情形下的适生区。结果表明,海南、广东、广西、云南、福建、台湾等地区是南美蟛蜞菊的适生区,且在未来气候情形下,其适生区范围向北扩张。为减小南美蟛蜞菊入侵可能导致的损失,需要根据南美蟛蜞菊当前分布情况,结合其适生区分布图,优先对现有分布区周边的适生区进行防控。
Predicting the suitable habitat of invasive species through species distribution modeling, and selecting the area that the species would spread with high potential for prevention, is the most effective effort of bio-invasion management. Wedelia trilobata （L.） Hitehc. is an invasive plant with strong detriment and has invaded parts of South China. The aim of this study was to predict the suitable habitat of W. trilobata in China. Based on distribution records from America and Hainan Island in China, together with 19 climate variables, the present study used the MaxEnt for distribution modeling of W. trilobata. The optimal configuration of MaxEnt was selected by AUC[the area under the receiver operating characteristic curve （ROC）] and AICe （sample size corrected akaike information criteria） values of models, and was then used to predicting the suitable habitat of W. trilobata in China under both current and future climate scenario. The results indicated that, there exists suitable habitat of W. trilobata in Hainan, Guangdong, Guangxi, Yunnan, Fujian and Taiwan. Under future climate, suitable habitat of W. trilobata expands northward. In order to minimize the damage caused by invasion of W. trilobata, preventive actions should be taken with top priority focusing on the suitable habitat around the current distribution of W. trilobata.
出处
《广东农业科学》
CAS
CSCD
北大核心
2013年第14期174-178,共5页
Guangdong Agricultural Sciences
基金
国家自然科学基金(30970556
31170386)
高等学校博士学科点专项科研基金(20093227110004)
