摘要
该研究基于生态位模型理论对我国建兰地理分布及气候影响因素进行研究。以建兰在国内的现有地理分布数据为基础,通过运用地理信息系统(GIS),建立了基于最大熵模型(Maxent)理论的建兰国内适生区预测模型,并对影响建兰地理分布的关键生物气候影响因子做出分析。结果表明:经ROC验证,模型的训练AUC为0.977,测试AUC为0.944,数据检验显示所建立模型的预测精度良好;刀切法检验结果表明,最冷季均温、最旱季均温、年均气温、最旱月降水量为影响建兰潜在分布的最关键生物气候因子;中国建兰潜在分布区主要位于广东北部、广西东北、江西及湖南南部的南岭山脉地区,福建北部武夷山南段地区以及广东莲花山以南等地。通过对模型数据与预测结果进行分析,为未来国内建兰种质资源保护、引种驯化与规模化室外种植研究提供了科学建议与理论基础。
This research studied the geographical distribution and climatic factors of Cymbidium ensifoliumin China based on the theory of ecological niche model.The research established the suitable distribution area prediction model and analyzed the key climatic factors of Cymbidium ensifolium with geographic information system,based on maximum entropy model theory and existing domestic geographical distribution data of Cymbidium ensifoliumin China.The results showed that ROC test,the training data AUC was 0.977,testing data AUC was 0.944,indicated the model had good accuracy in prediction,with high practical value.Jackknife test showed that the key biological climatic factors for potential distribution of Cymbidium ensifolium were mean temperature of coldest quarter,mean temperature of driest quarter,annual mean temperature,and precipitation of driest month.The major potential distribution areas of Cymbidiumensifoliumin China were Nanling Mountains area located in northern Guangdong,northeastern Guangxi,and southern part of Jiangxi and Hunan,southern part of Wuyi Mountain located in northern Fujian and south of Lianhua Mountain in Guangdong.With the analysis of model data and predictions,the aim was to provide scientific advice and theoretical basis for the future research of germplasm resources protection,introduction and domestication and large-scale outdoor cultivation for Cymbidium ensifoliumin China.
作者
梁春
罗清
陆祖正
谢振兴
覃茜
黄欣怡
LIANG Chun LUO Qing LU Zuzheng XIE Zhenxing QIN Qian HUANG Xinyi(Guangxi Subtropical Crops Research Institute,Nanning,Guangxi 530001)
出处
《北方园艺》
CAS
北大核心
2017年第9期199-204,共6页
Northern Horticulture
基金
广西科技计划资助项目(桂科AB16380060)
广西壮族自治区亚热带作物研究所基本科研业务费专项资助项目(桂热研201610)
关键词
建兰
最大熵模型
地理信息系统
地理分布预测
生物气候因子
Cymbidium ensifolium
maximum entropy model
geographic information system
geographical distribution prediction
biological climatic factor
