MaxEnt模型预测气候变化下野鸦椿在中国的潜在地理分布

Prediction of the Potential Geographical Distribution of Euscaphis japonica in China Under Climate Change Using the MaxEnt Model

筛选野鸦椿(Euscaphis japonica)的288条分布记录及10个环境因子,采用MaxEnt和ArcGIS模拟探究野鸦椿当前、未来2041—2060年、未来2081—2100年3种气候变化情景下的潜在区分布,使用环境变量贡献率、置换重要值、刀切法检验综合评估影响当代野鸦椿分布的主导环境因子,对野鸦椿潜在适生区的变化趋势进行分析。结果表明:MaxEnt模型的评价指标受试者工作特征曲线下面积(AUC)训练集和测试集平均值分别为0.943、0.928,说明模型的模拟结果精准度高。最干月降水量、最冷月最低温、坡度、等温性是限制野鸦椿地理分布的4个主要因子。野鸦椿的适宜栖息地面积为141.33×104km^(2),横跨长江以南大部分亚热带地区。随着未来气候变暖,野鸦椿的分布区域呈现沿秦岭-淮河线向北移的趋势。未来需在水热条件及地形层面进行野鸦椿培育,以增加野鸦椿的适生性。

Screening 288 distribution records of Euscaphis japonica and 10 environmental factors,the present study employed MaxEnt and ArcGIS to simulate the potential distribution of E.japonica under three climate change scenarios,including the current,future 2041-2060,and future 2081-2100.Utilizing the contribution rate of environmental variables,permutation importance,and jackknife test to comprehensively evaluate the dominant environmental factors affecting the contemporary distribution of E.japonica,and analyzing the trend of potential suitable habitat changes for E.japonica.The results showed that the average area under the receiver operating characteristic curve(AUC)values of the MaxEnt model were 0.943 and 0.928 for the training and test datasets,respectively,indicating high simulation accuracy.The four main factors limiting the geographical distribution of E.japonica were the precipitation of the driest month,the minimum temperature of the coldest month,slope,and isothermality.The suitable habitat area of E.japonica was 141.33×104km 2,spanning most of the subtropical region south of the Yangtze River.With future climate warming,the distribution of E.japonica tends to shift northward along the Qinling Mountains-Huaihe River line.Future efforts should focus on cultivating E.japonica under favorable water-heat conditions and topographical factors to increase its habitat suitability.