气候变化情境下木荷潜在地理分布及生态适宜性分析  

Potential geographical distribution and ecological suitability of Schima superba under the situation of climate change

作　　者：文国卫 黄秋良 吕增伟 赖文峰 施晨阳 朱晓如 蒋天雨 张国防[1] WEN Guowei;HUANG Qiuliang;LÜZengwei;LAI Wenfeng;SHI Chenyang;ZHU Xiaoru;JIANG Tianyu;ZHANG Guofang(College of Forestry,Fujian Agriculture and Forestry University,Fuzhou 350002,China)

机构地区：[1]福建农林大学林学院,福州350002

出　　处：《生态学报》2023年第16期6617-6626,共10页Acta Ecologica Sinica

基　　金：福建农林大学科技创新基金(CXZX2019046);福建农林大学高峰学科项目(71201802404)。

摘　　要：木荷(Schima superba)是中国南方常见的速生阔叶耐火树种之一,在生态修复和维持环境稳定方面具有重要作用,因而预测其潜在适生区变化具有重要意义。基于Maxent模型,结合158条木荷在中国的有效分布记录和筛选后的8个生物气候变量,预测木荷现代和2041—2060年3种气候排放浓度情景下木荷的潜在分布区。Maxent模型的可靠性以受试者工作特征曲线来判断,综合气候因子的贡献率、置换重要值和刀切法检验,来分析并探讨制约木荷地理分布的重要气候因子。结果表明:(1)Maxent模型训练和测试数据的AUC值分别为0.936±0.001和0.943±0.008,表明模型的预测精度极高;(2)现代木荷潜在地理分布的总适生区面积198.87×10^(4)km^(2),占中国国土面积的20.7%,主要位于在中国长江以南的大部分地区,包括福建、浙江、中国台湾、江西、湖南、广东、海南、广西、贵州、云南、重庆,以及四川、河南和安徽部分区域;(3)制约木荷潜在地理分布的重要生物气候因子为降水因子(最干月降水量26—214 mm、年降水量800—2500 mm和最湿季降水量590—1030 mm)和温度因子(气温年变化8.5—28℃);(4)使用Maxent模型模拟的木荷适生区面积大于其实际分布面积,暗示着木荷人工林在未来有较大的发展潜力。本研究结果可为木荷种质资源收集和人工林种植奠定理论基础,对维持区域性生态系统安全和稳定、缓解气候变化和2060年实现碳中和目标具有战略性意义。Schima superba is one of the common fast-growing and broad-leaved fire-resistant trees in southern China.It plays an important role in ecological restoration and maintaining environmental stability.Therefore,it is of great significance to predict the change of its potential suitable area.Based on Maxent model,combined with 158 effective distribution records of S.superba in China and 8 selected bioclimatic variables,this study predicts the potential distribution areas of S.superba are predicted under three climate emission concentration scenarios of modern S.superba and 2041-2060.The reliability of Maxent model is judged by the area of under the receiver operator characteristic curves.The important climate factors restricting the geographical distribution of S.superba were analyzed and discussed by integrating the percent contribution,permutation importance and Jackknife test of climate factors.The results show that:(1)the AUC values of Maxent model training and test data are 0.936±0.001 and 0.943±0.008,respectively,indicating that the prediction accuracy of the model is very high.(2)The total suitable area of potential geographical distribution of modern S.superba is 198.87×10^(4) km^(2),accounting for 20.7%of China's total land area.It mainly locates in most areas south of the Yangtze River,including Fujian,Zhejiang,Taiwan,china,Jiangxi,Hunan,Guangdong,Hainan,Guangxi,Guizhou,Yunnan,Chongqing,and some areas of Sichuan,Henan and Anhui.(3)The important bioclimatic factors restricting the potential geographical distribution of S.superba are precipitation factors and temperature factors.The annual precipitation is 800-2500 mm.The precipitation of the driest month is 26-214 mm and the precipitation of the wettest quarter is 590-1030 mm.The range of annual temperature is 8.5-28℃.(4)The suitable area of S.superba simulated by Maxent model is larger than its actual distribution area,suggesting that S.superba plantation has great development potential in the future.The results of this study can provide a theoretical

关 键 词：木荷 Maxent模型 气候变化 潜在适生区 生态适宜性 

分 类 号：S792.99[农业科学—林木遗传育种]