基于Biomod2组合模型对气候变化下滇山茶适生区的预测  

Prediction of suitable area of Camellia reticulata under climate change based on the Biomod2 ensemble model

作　　者：高灿 樊智丰[1,2,3] 马长乐 杨建欣[1,2,4] 郭帅龙[1,2] GAO Can;FAN Zhifeng;MA Changle;YANG Jianxin;GUO Shuailong(School of Land-scape Architecture and Horticulture Sciences,Southwest Forestry University,Kunming 650224,China;Southwest Landscape Architecture Engineering Technology Research Center,State Forestry and Grassland Administration,Kunming 650224,China;Kunming University of Science and Technology,Kunming 650093,China;Key Laboratory of Forest Resources Conservation and Utilization in the Southwest Mountains of China Ministry of Education,Key Laboratory of National Forestry and Grassland Administration on Biodiversity Conservation in Southwest China,Yunnan Provincial Key Laboratory for Conservation and Utilization of Inforest Resource,Kunming 650224,China)

机构地区：[1]西南林业大学园林园艺学院,昆明650224 [2]国家林业和草原局西南风景园林工程技术研究中心,昆明650224 [3]昆明理工大学后勤保障服务中心,昆明650093 [4]西南山地森林资源保育与利用教育部重点实验室、西南地区生物多样性保育国家林业和草原局重点实验室、云南省林下资源保护与利用重点实验室,昆明650224

出　　处：《生态学杂志》2024年第11期3526-3536,共11页Chinese Journal of Ecology

基　　金：农业联合专项重点项目(202301BD070001-150);云南省农业联合项目(202101BD070001-100);云南省教育厅科学研究基金项目(2023Y0743和2023Y0757);西南山地森林资源保育与利用教育部重点实验室;西南地区生物多样性保育国家林业和草原局重点实验室和云南省林下资源保护与利用重点实验室开放课题(LXXK-2023M05)资助。

摘　　要：滇山茶是云南珍贵的野生冬花资源,也是国家二级保护植物。滇山茶对自然生境要求比较苛刻,暂未得到充分的园林应用。气候变化会在较大尺度上影响物种的分布,了解滇山茶气候变化下的响应状态,是野生滇山茶资源保护及其开发利用的基础。研究基于Biomod2集成最优组合模型,利用气候、地形、土壤和植被因子分析了从末次冰盛期到本世纪末的滇山茶分布动态。结果表明:推进式回归树模型、柔性判别分析、随机森林和优化后的最大熵模型集成得到的最优组合模型的精度和效果较单一模型更高,比全部模型组合的效果更佳且模型更简单。31个环境因子中,滇山茶适宜生境分布受温度年较差影响最大。当前气候情景下,滇山茶主要分布在西南、华南和台湾地区;其中,高适生区位于云南绝大部分地区、贵州西南部和广西西北部;此外,西藏错那市南部、福建东南部和海南西南部也有分布。滇山茶的冰期避难所在云南中部地区,随气候变暖滇山茶的适宜生境不断向周边四省扩散。未来的气候变化不会引起滇山茶的大规模迁移或灭绝,云南及其与西藏、四川、贵州和广西4个省份交界处始终是滇山茶稳定的适生区。未来全球气候变暖有利于滇山茶的生存繁衍,适生区将向北扩散,未来可重点考虑在云南北部、西藏错那市南部、四川南部、贵州南部和福建沿海地区对滇山茶进行引种驯化和园林应用。Camellia reticulata,a valuable wild winter flower plant species in Yunnan,has been listed as classⅡnational key protected wild species in China.C.reticulata has not been fully utilized in landscape applications due to its strict requirements for natural habitats.Climate change affects the distribution of plant species.Understanding the response of C.reticulata to climate change is the basis for its conservation,development and utilization.The distribution dynamics of C.reticulata from the Last Glacial Maximum to the end of this century was simulated with an optimal ensemble model(EM)in the Biomod2 package,based on climatic,topographic,soil and vegetational factors.The results showed that the optimal EM which composed of Generalized Boosted Models,Flexible Discriminant Analysis,Random Forest and optimized Maximum Entropy Models had better accuracy and performance compared to the single models,and being more simpler than the EM integrating by all model algorithms.Among the 31 envi-ronmental factors,temperature annual range was the main factors affecting the suitable habitat distribution of C.reticulata.Under the current climate scenario,C.reticulata is mainly distributed in the southwest China,south China and Taiwan Region,China,of which the highly suitable distribution region including most areas of Yunnan,southwestern Guizhou and northwestern Guangxi.Southern Cuona of Xizang,southeastern Fujian and southwestern Hainan also have scat-tered distribution of C.reticulata.Central region of Yunnan might be refugia site for C.reticulata in glacial age.With the warming climate,the suitable habitat of C.reticulata continues to spread to the surrounding four provinces.Future climate change would not cause large-scale migration or extinction of C.reticulata.Yunnan and its bordering provinces Xizang,Sichuan,Guizhou and Guangxi have always been stable and suitable areas for C.reticulata.The results indicated that global warming is conducive to the survival and reproduction of C.reticulata,and the suitable areas will spread

关 键 词：末次冰盛期 全新世中期 物种分布模型 模型优化 适宜分布区 

分 类 号：Q948[生物学—植物学]