首页 | 本学科首页   官方微博 | 高级检索  
     

气候变化下花榈木在我国的潜在分布区预测
引用本文:宋颖1,2,张港隆2,贾全全1,温强1,莫晓勇2,刘丽婷1. 气候变化下花榈木在我国的潜在分布区预测[J]. 西北林学院学报, 2021, 36(6): 108-115. DOI: 10.3969/j.issn.1001-7461.2021.06.15
作者姓名:宋颖1  2  张港隆2  贾全全1  温强1  莫晓勇2  刘丽婷1
作者单位:(1.江西省林业科学院,江西 南昌 330032;2.华南农业大学 林学与风景园林学院,广东 广州 510642)
摘    要:针对花榈木的种质资源保护利用问题,基于花榈木在中国的188条地理分布记录和9个气候变量,于 SSP1-2.6、SSP2-4.5、SSP5-8.5 3种气候情景模式下,利用ENMeval数据包优化后的MaxEnt模型和ArcGIS对其当前和未来(2050s和2070s)的生态位进行模拟,研究影响花榈木地理分布的主导环境因子,分析其在不同气候情景模式下的空间分布格局和迁移趋势。结果表明,当模型的特征组合(FC)为线性、二次型、片段化、乘积型和阈值性,正则化乘数(RM)为2时,模型复杂度和过拟合程度较低,此时训练集的平均AUC和平均标准偏差分别为0.954和0.004,表明模型预测精度很高。花榈木在当前气候下的潜在分布区主要在我国东南部,核心分布区位于江西、湖南、福建、浙江等省份及其周边区域。刀切法(Jackknife)表明最干季度降水量、最湿季度降水量、最冷季度平均温度是影响其分布的主导气候因子。未来3种气候情景下花榈木总体适生区分布相对稳定,核心适生区主要表现为东西向稳定,于南北向中部迁移,尤以广东中部及江西中部收缩显著,这些迁移的核心适生区是花榈木应对气候变化的敏感区域,需引起重视。

关 键 词:气候变化  潜在分布区  花榈木  MaxEnt模型  模型参数优化

 Prediction of Potential Distribution of Ormosia henryi in China Under Climate Change
SONG Ying1,' target="_blank" rel="external">2,ZHANG Gang-long2,JIA Quan-quan1,WEN Qiang1,MO Xiao-yong2,LIU Li-ting1.  Prediction of Potential Distribution of Ormosia henryi in China Under Climate Change[J]. Journal of Northwest Forestry University, 2021, 36(6): 108-115. DOI: 10.3969/j.issn.1001-7461.2021.06.15
Authors:SONG Ying1,' target="  _blank"   rel="  external"  >2,ZHANG Gang-long2,JIA Quan-quan1,WEN Qiang1,MO Xiao-yong2,LIU Li-ting1
Affiliation:(1.Jiangxi Academy of Forestry,Nanchang 330032,Jiangxi,China; 2.College of Forestry and Landscape Architecture,South China Agricultural University,Guangzhou 510642,Guangdong,China)
Abstract:In view of the protection and utilization of germplasm resources of Ormosia henryi,based on 188 geographical distribution data in China,nine variable factors were selected after correlation analysis,the maximum entropy model (MaxEnt) optimized by ENMeval data package and ArcGIS were used to simulate the current and future (2050s and 2070s) potential distribution under three climate scenarios of SSP1-2.6,SSP2-4.5 and SSP5-8.5.It was found that when the feature combination (FC) of the model was linear,quadratic,hinge,product and threshold,and the regularization multiplier (RM) was 2,the complexity and over fitting degree of the model were low,and the average AUC and average standard deviation of the training set were 0.954 and 0.004,respectively,which indicated that MaxEnt model had high prediction accuracy.The potential distribution area of O.henryi in the current climate was mainly in the southeast of China,the main core suitable areas were concentrated in Jiangxi,Hunan,Fujian,Zhejiang and their adjacent areas.The application of Jackknife method indicated that the dominant factors influencing the distribution were precipitation in the driest month,precipitation of wettest quarter and mean temperature of the coldest quarter.Under the three climate scenarios in the future,distribution of the overall suitable area of O.henryi would be relatively stable with no significant change,the core suitable area would be stable in the east-west direction,and migrate from the south-north direction to the middle,especially in middle of Guangdong and Jiangxi.Attention should be paid to these areas that were sensitive to climate change.
Keywords:climate change  potential distribution  Ormosia henryi Ormosia henryi  MaxEnt model  model parameter optimization
点击此处可从《西北林学院学报》浏览原始摘要信息
点击此处可从《西北林学院学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号