摘要
【目的】预测我国酸枣适生区对气候变化的响应,为气候变化背景下酸枣种质资源开发与利用提供科学依据。【方法】基于GIS技术和R语言,筛选出酸枣121个地理调查数据和34个环境因子,利用R语言的ENMeval数据包,调整MaxEnt模型默认参数,并运用优化后的MaxEnt模型分析酸枣当前适生区的地理分布;采用Person相关分析和VIF方差膨胀因子筛选出参与建模的所需因子,刀切法筛选该适生区主导环境因子,根据IPCC发布的第5次气候模型,探讨在不同气候情景下酸枣未来地理分布的变化。【结果】模型优化结果表明,当模型参数中特征组合为线性、二次型与片段化,以及调控倍频为3.5时,MaxEnt模型的复杂度最低,受试者工作特征曲线分析法的AUC为0.946,模型预测的可靠性极高、精度极好;刀切法结果显示,年均气温、最湿季平均气温、最湿月降雨量、降水量变异系数、海拔和表层(0~30 cm)土壤盐基饱和度是影响酸枣分布的主要环境因子;当前酸枣适生区主要集中在陕西、河南北部、山西中南部、河北、宁夏北部、内蒙古东南部、四川东北部、辽宁西部和山东境内,在未来气候变化情景下,酸枣适生区分布会有不同程度变化,到2050s和2070s,酸枣适生区面积有一定增加,在RCP4.5和RCP8.5情景下响应敏感且呈递增趋势,而在RCP2.6情景下的递增趋势较其他2种情景小。【结论】气候、土壤和地形因子共同影响酸枣的潜在地理分布,其中气候因子所占权重最大,是最有可能引起酸枣地理分布区迁移的环境变量;模型经过参数优化能准确模拟预测出酸枣潜在地理分布区;未来全球变暖引起的气候变化将会改变酸枣的分布格局。MaxEnt模拟结果显示,适生区总面积趋于增加,且扩张区域主要位于中高纬度地区,减少区域主要位于低纬度地区;在3种气候变化情境下,2050s和2070s的酸枣适生区质心呈往高纬度迁移的趋势,酸枣适生区域整体迁移至华北平原和东北平原一带。
【Objective】In order to provide a scientific basis for the development and utilization of Ziziphus jujuba var.spinosa germplasm resources,the suitable range for growing Z.jujuba var.spinosa was predicted under the context of climate change.【Method】Based on the GIS technology and R language,data of 121 geographical concurrence points and 34 environmental factors were selected for the study.The default parameters of MaxEnt model were adjusted by using ENMeval package in R language,and the geographical distribution of suitable areas for Z.jujuba var.spinosa was predicted by using the optimized MaxEnt model.Person correlation analysis and VIF variance expansion factor were used to select the required factors for modeling,and the jackknife method was used to select the dominant environmental factors of the suitable area.According to the fifth climate model released by IPCC,the changes of future geographical distribution of Z.jujuba var.spinosa under different climate scenarios were discussed.【Result】The model optimization showed that the MaxEnt model has the lowest complexity and the best when the feature combination is linear,quadratic and fragmented,and the regulation frequency multiplier is 3.5.The AUC of the working characteristic curve analysis method of the subjects was 0.946,indicating that the prediction model has high reliability and excellent accuracy.The jackknife test showed that the main environmental factors affecting the distribution of Z.jujuba var.Spinose were the annual mean temperature,the mean temperature of wettest quarter,the precipitation of wettest month,the precipitation seasonality,the elevation,and the base saturation of topsoil(0-30 cm).At present,the suitable growing areas for Z.jujuba var.spinosa in China are mainly concentrated in Shaanxi,northern Henan,central and southern Shanxi,Hebei,northern Ningxia and southeastern Inner Mongolia,northeastern Sichuan,western Liaoning and Shandong.Under the different climate scenarios in the future,the suitable areas for Z.jujuba var.spinosa will change in varying degrees.By the 2050s and the 2070s,the suitable areas for Z.jujuba var.spinosa will increase to a certain extent.The suitable areas under RCP 4.5 and RCP 8.5 scenarios respond sensitively,showing an increasing trend,and the increasing trend is weaker under RCP 2.6 scenario than under the other two scenarios.【Conclusion】Climate,soil and terrain factors all affect the potential geographical distribution of Z.jujuba var.spinosa,among which climate factors account for the largest weight,which is most likely to cause the migration of geographical distribution of Z.jujuba var.spinosa.The optimized model can accurately simulate and predict the potential geographical distribution area of Z.jujuba var.spinosa.In the future,the climate change of global warming will change the distribution pattern of Z.jujuba var.spinose,the total suitable growing areas will increase,the expansion will be mainly in the middle and high latitudes,and the reduction will be mainly in the low latitudes.Under the three climate change scenarios,the center of suitable areas for Z.jujuba var.spinosa in the 2050s and the 2070s will migrate to high latitudes.The suitable area for Z.jujuba var.spinosa will migrate to the North China Plain and the Northeast Plain as a whole,and the priority should be given to arrange new planting areas in this region,so as to reduce the loss caused by climate change.
作者
赵光华
崔馨月
王智
荆红利
樊保国
Zhao Guanghua;Cui Xinyue;Wang Zhi;Jing Hongli;Fan Baoguo(College of Life Science, Shanxi Normal University ,Linfen 041000;Agriculture College, Guangxi University ,Nanning 530000;Shanwei Middle School ,Shanwei 516600)
出处
《林业科学》
EI
CAS
CSCD
北大核心
2021年第6期158-168,共11页
Scientia Silvae Sinicae
基金
山西省科学技术发展计划项目(20140311017-7)。
关键词
酸枣
MaxEnt模型
ENMeval数据包
气候变化
适生区预测
Ziziphus jujuba var.spinosa
MaxEnt model
ENMeval package
climate change
prediction of suitable growing areas
