基于MaxEnt模型的细尤犀金龟在中国的适生分布区预测  

Prediction of suitable distribution area of Eupatorus gracilicornis Arrow in China based on MaxEnt model

作　　者：张鸿辉 舒旭 余新林 朱恩骄 和秋菊 赵敏[4] 易传辉[5] ZHANG Hong-hui;SHU Xu;YU Xin-lin;ZHU En-jiao;HE Qiu-ju;ZHAO Min;YI Chuan-hui(College of Forestry,Southwest Forestry University,Kunming 650224,China;Tengchong Branch of Baoshan Management and Protection Bureau of Yunnan Gaoligong Mountain National Natural Reserve,Tengchong,Yunnan 679100,China;Yunnan Forestry Technological College,Kunming 650224,China;Institute of Highland Forest Science,Chinese Academy of Forestry,Kunming 650224,China;Yunnan Academy of Biodiversity,Southwest Forestry University,Kunming 650224,China)

机构地区：[1]西南林业大学林学院,昆明650224 [2]云南高黎贡山国家级自然保护区保山管护局腾冲分局,云南腾冲679100 [3]云南林业职业技术学院,昆明650224 [4]中国林业科学研究院高原林业研究所,昆明650224 [5]西南林业大学云南生物多样性研究院,昆明650224

出　　处：《西南农业学报》2024年第7期1601-1610,共10页Southwest China Journal of Agricultural Sciences

基　　金：云南省科技厅科技计划-农业联合专项重点项目(202301BD070001-004)。

摘　　要：[目的]对濒危物种细尤犀金龟适生区进行预测,为其保护提供理论依据。[方法]基于地理分布记录和环境变量数据,采用MaxEnt模型,利用27个有效分布点和19个环境因子,借助ArcGIS软件模拟细尤犀金龟在中国当前气候条件和未来(2050s,2070s)RCP2.6与RCP8.5此2种气候条件下的适生分布区,并筛选出影响分布的主导环境因子。[结果]对19个环境因子采用MaxEnt模型进行10次重复运算,计算出AUC值为0.986,表明MaxEnt模型精度高,同时去除掉贡献率和置换重要值较低的环境因子。采用MaxEnt模型中的“刀切法”筛选得到6个环境因子并进行分析,最终根据贡献率得出影响细尤犀金龟潜在地理分布的主导环境因子前4位依次为温度季节性(BIO4,53.7%)、最热季度的降水(BIO18,13.5%)、年降水量(BIO12,12.2%)和等温性(BIO3,11.8%)。在当前气候条件下,细尤犀金龟高适生区主要位于云南南部、西藏南部区域以及台湾地区中部零星区域。在2050年RCP2.6与2070年RCP8.5排放情景下,高适生区面积在台湾地区、云南和西藏部分地区虽有小幅度增加,但其它排放情景下的面积均有所减少。[结论]温度季节性、最热季度的降水、年降水量、等温性、最冷月份的最低温度和年温差是影响细尤犀金龟在中国潜在分布的主导环境因子。细尤犀金龟适生区的预测结果与其当前在中国的实际分布情况基本一致,但由于栖息地丧失、破碎化、退化以及生境的严重破坏,其未来的实际适生分布区存在进一步缩小和破碎化的趋势,濒危物种细尤犀金龟生存环境不容乐观。[Objective]The prediction of the habitat area of the endangered species Eupatorus gracilicornis was made to provide a theoretical basis for its protection.[Method]Based on geographical distribution records and environmental variable data,27 effective distribution points and 19 environmental factors were utilized,with the help of ArcGIS simulating the E.gracilicornis adaptive distribution areas in China of RCP2.6 and RCP8.5 under the current and future climate conditions(2050s,2070s),MaxEnt model was used and the dominant environ-mental factors affecting the distribution were screened out.[Result]The MaxEnt model was used to repeat 10 operations for 19 environmental factors,and the calculated AUC value was 0.986,indicating that MaxEnt model had high accuracy.Six environmental factors were selected by the knife cutting'method in MaxEnt model and analyzed.Finally the top four environmental variables affecting the potential geograph-ical distribution were temperature seasonality(BI04,53.7%),precipitation of the warmest quarter(BI018,13.5%),annual precipitati-on(BI012,12.2%)and isothermality(BI03,11.8%).The top four contributing rates of the dominant environmental variables affecting the potential geographical distribution of E.gracilicornis were temperature seasonality(BI04,53.7%),precipitation in the hottest quarter(BI018,13.5%),annual precipitation(BI012,12.2%),and isothermality(BI03,11.8%).Under the current climatic conditions,the high suitable areas of E.gracilicornis were mainly located in the southern part of Yunnan province,the southern part of Xizhang Autono-mous Region,and the central part of Taiwan province.Under the 2050 RCP2.6 and 2070 RCP8.5 emission scenarios,the area of the high suitable area increased slightly in parts of Taiwan,Yunnan and Xizhang,but decreased under the other emission scenarios.[Conclusion]Temperature seasonality,precipitation of warmest quarter,annual precipitation,isothermality,min temperature of coldest month,tempera-ture annual range are the main environmental factors affecting t

关 键 词：细尤犀金龟 潜在适生区 濒危 MAXENT ARCGIS 

分 类 号：S891.8[农业科学—特种经济动物饲养]