基于最大熵（MaxEnt）模型对西北地区2种红砂属植物潜在适宜区预测分析

基于黄花红砂与五柱红砂的现有分布数据，利用地理信息系统（ArcGIS）技术与最大熵（MaxEnt）模型软件，对2种红砂属植物在3个时期的潜在适宜分布区进行模拟。结果表明:1）五柱红砂的潜在适宜分布区主要在青海地区、新疆南疆地区以及甘肃部分地区；黄花红砂的潜在适宜分布区主要集中在宁夏地区、内蒙古地区、甘肃地区以及新疆与青海部分地区。2种红砂属植物在3个时期的潜在适宜分布区有所增加。黄花红砂的潜在适宜区朝着东北方向移动，五柱红砂的潜在适宜区出现收缩现象。2种红砂属的低适宜区相比较高适宜区与较适宜区变化较大。2）黄花红砂与五柱红砂通过MaxEnt模型运算出的AUC值均>0.9，表明MaxEnt模型预测精度很高。可以用于2种红砂属植物的潜在适宜分布区的预测。3）控制黄花红砂潜在分布的关键环境变量为最湿月降水量、最暖季度降水量和最湿季度降水量，影响五柱红砂潜在分布的关键环境变量为海拔、等温性和温度季节性变化。上述研究结果对2种红砂属植物资源保护利用与未来分布趋势提供了重要的指导意义。

Prediction and Analysis of Two Potential Suitable Areas of Reaumuria Plants in Northwest China Based on the Maximum Entropy (MaxEnt) Model

Based on the existing distribution data,by using geographic information system (ArcGIS) technology and maximum entropy (MaxEnt) model software,the potentially suitable distribution areas (PSDAs) of Reaumuria trigyna and R.kaschgarica over three periods were simulated in.The results showed that 1) PSDAs for R.kaschgarica,were mainly in Qinghai,southern Xinjiang,and parts of Gansu,and for R.trigyna,PSDAs were mainly in Ningxia,Inner Mongolia,Gansu,Xinjiang,and some parts of Qinghai.PSDAs increased in three periods.For R.trigyna,PSDAs would move toward the northeast,and for R.kaschgarica,the areas would shrink.The low-suitable areas of the two specieswould be relatively large compared to the higher-suitable areas and the more suitable areas.2) The AUC values calculated by the MaxEnt model of the two species were both greater than 0.9,indicating that the MaxEnt model had high prediction accuracy,and could be used for the prediction.3) The key environmental variables controlling the potential distribution of Reaumuria trigyna would be the precipitations in the most moist month,in the warmest quarter,and the most moist quarter.The key environmental variables affecting PSDAs would be elevation,isothermal and temperature seasons.The above research results would provide important guiding significances for the conservation and utilization of plant resources of two Reaumuria species and future distribution trends.