基于MaxEnt生态学模型的毛姜花潜在园林引种区预测

以温度和降雨为环境因子指标,采用MaxEnt模型软件对毛姜花Hedychium villosum wall.进行潜在引种区预测。模型的评价采用ROC (Receiver operating characteristic)曲线分析进行,区划图的判读采用引种栽培数据进行辅助。ROC曲线检验显示,所建立的毛姜花MaxEnt预测模型可靠性高;经引种栽培数据与预测结果的比较分析,在适生性被划分为0~1的13级时,大于0.08的区域内即可成功引种。本方法适用于以近自然林为营建目标的引种区预测,且操作简便,便于推广应用。  

Prediction of Potential Hedychium Viilosum Wall. Introduction Area Based on MaxEnt Ecologic Niche Model

The study use the MaxEnt ecological model to predict the potential introduction area of Hedychium villosum wall, with temperature and precipitation as the environment factors. The model - based evaluation was carried out with Receiver operating characteristic （ROC）, and the cultivation data were used to assist in interpreting the regionaliza- tion map. ROC curve test showed that the MaxEnt model constructed for H. villosum had high reliability. By compa- ring the predicted results with the cultivation data, it was concluded that the species could be introduced successfully in the area where the degree exceeded 0.08 when the suitability degrees were divided into thirteen categories between 0 - 1. The method described in this study is suitable for predicting the introduction regionalization with close - to - nature forest establishment as the objective, and besides it is easy to operate and convenient to apply.