基于MaxEnt模型的云南干热河谷适生树种选择  

Application of MaxEnt model for selection of suitable tree species in dry⁃hot valley of Yunnan

作　　者：谢孟 张学星 罗燕 马永鹏[5] 李伟 杨琳祥 柳文 赵培仙 李正红[1] 马宏 XIE Meng;ZHANG Xuexing;LUO Yan;MA Yongpeng;LI Wei;YANG Linxiang;LIU Wen;ZHAO Peixian;LI Zhenghong;MA Hong(Institute of Highland Forest Science,Chinese Academy of Forestry,Kunming 650233,China;College of Landscape Architecture,Nanjing Forestry University,Nanjing 210037,China;Yunnan Academy of Forestry and Grassland,Kunming 650204,China;Forestry and Grassland Bureau of Yuanmou,Yuanmou 651300,China;Yunnan Key Laboratory for Integrative Conservation of Plant Species with Extremely Small Populations,Kunming Institute of Botany,Chinese Academy of Sciences,Kunming 650201,China;Yunnan Jicheng Landscape Technology Co.,Ltd.,Mile 652399,China;Key Laboratory of Breeding and Utilization of Resource Insects,National Forestry and Grassland Administration,Kunming 650233,China)

机构地区：[1]中国林业科学研究院高原林业研究所,昆明650233 [2]南京林业大学风景园林学院,南京210037 [3]云南省林业和草原科学院,昆明650204 [4]元谋县林业和草原局,元谋651300 [5]云南省极小种群野生植物综合保护重点实验室,中国科学院昆明植物研究所,昆明650201 [6]云南吉成园林科技股份有限公司,弥勒652399 [7]国家林业和草原局资源昆虫培育与利用重点实验室,昆明650233

出　　处：《生态学报》2024年第9期3689-3707,共19页Acta Ecologica Sinica

基　　金：云南省重点研发计划“特色优良乡土观赏树种选择及应用研究”(202202AE090012);云南省生物种业和农产品精深加工专项(202302AE090018);兴滇人才支持计划青年人才专项(YNWR-QNBJ-2019-010)。

摘　　要：基于MaxEnt模型识别和预测云南干热河谷适生树种,对于改善和恢复其生态治理能力具有重要意义。收集40种具有代表性的潜在适生树种地理分布数据,结合气候、地形和土壤等环境因子,利用MaxEnt模型筛选适生树种。预测当前和2021—2040年四种气候情景下(SSP126、SSP245、SSP370和SSP585)适生树种适生区的分布格局,划分优先种植区域,并明确MaxEnt模型用于树种选择的可行性。结果表明:(1)当前气候情景下,影响干热河谷潜在适生树种分布的主导因子是气候因子,其次是海拔、植物归一化指数、地表太阳辐射量和人类足迹。(2)未来,24种适生树种适生区稳定,发生概率与海拔关系呈单峰分布且高海拔下适生树种丰富度将降低。(3)干热河谷适生树种优先种植区域沿干热河谷呈狭长分布;实际调查发现,元谋县适生树种实际分布区域面积略高于最佳种植区域面积。应用MaxEnt模型筛选干热河谷适生树种选择是可行的,但在应用之前必须通过实地调查来验证树种实际生存情况与预测结果的差异。在干热河谷生态修复造林时,可优先考虑白枪杆、车桑子等24种树种。Based on MaxEnt model,identifying and predicting suitable tree species in the dry⁃hot valley of Yunnan are important for improving and restoring its ecological management capacity.In this study,we collected data on the geographic distribution of 40 representative potentially suitable tree species,and screened suitable tree species using the MaxEnt model in combination with the environmental factors such as climate,topography and soil.To predict the distribution pattern of suitable tree species suitable areas for the current and 2021—2040(SSP126,SSP245,SSP370,and SSP585),to delineate the priority planting areas,and to clarify the feasibility of the MaxEnt model for tree species selection.The results show that:(1)The dominant factors affecting the distribution of potentially suitable tree species in the dry⁃hot valley under the current climate scenario are climatic factors,followed by elevation(elev),normalized difference vegetation index(ndvi),surface solar radiation(rs)and human footprint(hfp).(2)In the future,suitable areas of the 24 suitable tree species are stable,and the probability of occurrence has a unimodal distribution in relation to elevation.The abundance will decrease at higher elevations.(3)The priority planting areas for suitable tree species in the dry⁃hot valleys are narrowly distributed along the banks of the dry⁃hot valley.The actual survey found that the area of the actual distribution area of the suitable tree species in Yuanmou County was slightly higher than the area of the optimal planting area.It is feasible to apply the MaxEnt model to screen the selection of suitable tree species in the dry⁃hot valleys,but the difference between the actual survival of the tree species and the predicted results must be verified through field surveys before application.When reforesting the ecological restoration of the dry⁃hot valley,the priority can be given to 24 species of trees such as Fraxinus malacophylla and Dodonaea viscosa.

关 键 词：云南干热河谷 MaxEnt模型 气候变化 树种选择 优先种植区 

分 类 号：S725.1[农业科学—林木遗传育种]