论文部分内容阅读
本文以人参为研究对象,基于人参分布点位数据和22个气候环境因子数据,运用BioMod2平台10个物种分布模型对当前我国东北地区人参潜在生境分布进行预测.以受试者工作特征曲线(ROC)为权重集成10个模型的模拟结果,构建组合模型,并基于该模型预测了IPCC第五次评估报告中RCP 8.5、RCP 6.0、RCP 4.5和RCP 2.6等4种排放情景下21世纪50和70年代人参潜在分布范围.结果表明:在基准气候条件下,人参适宜生境面积占研究区总面积的10.4%,此类地区主要分布于研究区东北部长白山地区以及小兴安岭东南部区域的森林地带.在未来不同的排放情景下研究区人参的适宜生境变化显著,总体上分布范围将有一定程度的缩小.同时参与建模的10种模型在统计学精度、预测结果以及变量权重上都有差异.模型精度计算结果表明,MAXENT模拟效果最好,GAM、RF和ANN次之,SRE模拟精度最低.本文构建的组合模型在一定程度上提高了现有物种分布模型的预测精度,从而使模拟效果更优.
Based on the ginseng distribution point data and 22 climatic and environmental factors data, 10 potential species distribution models of BioMod2 platform were used to predict the potential habitat distribution of ginseng in northeastern China.According to the receiver operating characteristic curve (ROC ) As the weight of the simulation results of the integration of 10 models to build a combined model and based on the model to predict the IPCC Fifth Assessment Report RCP 8.5, RCP 6.0, RCP 4.5 and RCP 2.6 emissions under the circumstances of the 21st Century 50 and 70 The results showed that the appropriate habitat area of ginseng accounted for 10.4% of the total area of the study area under the reference climate conditions, which mainly distributed in the Changbai Mountain area in the northeast of the study area and the forest area in the southeastern part of the Xiaoxing’an Mountains. In different future scenarios, the suitable habitat changes of the ginseng in the study area are significant, and the distribution range will be narrowed to a certain extent. Meanwhile, the 10 models involved in the modeling have differences in the statistical accuracy, the prediction results and the variable weights. The accuracy calculation results show that MAXENT has the best simulation results, followed by GAM, RF and ANN, and has the lowest SRE simulation accuracy. Model to a certain extent to improve the prediction accuracy of the existing distribution model species, so that better simulation effect.