双角草在我国的潜在适生区预测及其生态特征分析

Estimation of Potential Suitable Distribution Areas and Ecological Characteristics of Diodia virginiana L.in China

双角草是一种繁殖能力极强的阔叶杂草,现分布于北美洲、亚洲等地区,该杂草在中国的适生范围尚不明确。本研究采用R语言软件的Kuenm语言包来优化Max Ent模型参数,利用Maxent软件对双角草在我国潜在适生区进行预测,使用Arc GIS软件制图和适生等级分类,依据受试者工作特征(Receiver Operating Characteristic Curve,ROC)曲线评价Maxent模型预测的准确性,根据刀切法检验、环境变量建模贡献率和置换重要值选择影响双角草分布的主导环境变量,根据环境变量响应曲线计算主导环境变量的适生阈值。选择最优模型参数(delta AICc=0、AUC=0.9091、5%数据遗漏率=0.0497)建模,结果表明双角草在我国的潜在适生区面积总计为95.57万km^(2),约占国土总面积的9.96%,适生区相对集中且主要分布在我国东南低山丘陵平原地区,鉴于其较强的适生能力、繁殖能力和经济重要性,应引起高度关注。上述预测结果可为有关部门对该杂草制定检疫和防控措施提供参考。

Diodia virginiana L.is a highly reproductive broad-leaved weed distributed in North America and Asia,yet its suitability region in China remains unclear.This study utilized the R language kuenm package to optimize the MaxEnt parameters,with ArcGIS software used for mapping and classification of suitability levels.The accuracy of the Maxent model prediction was evaluated based on the Receiver Operating Characteristic Curve.We selected dominant environment variables that affect the distribution of Diodia virginiana through the knife cut test,environmental variable modeling contribution rate,and permutation importance value analysis.We calculated the fitness threshold for dominant environment variables based on the response curve.Optimal model parameters(DeltaAICc=0,AUC=0.909,Omission rate at 5%=0.0497)were selected for modeling.Results indicate a total potential suitable area of 955700 km^(2)for Diodia virginiana growth in China,accounting for approximately 9.96%of the total land area,with concentrations in the low hills and plains region of Southeast China.Considering its strong adaptability,reproductive ability,and economic importance,increased attention to this weed is warranted.Moreover,the predictions offer valuable insights for relevant departments to formulate quarantine and control measures.