Species, application frequencies, habitat features and combination modes of Rhododendron species in Kunming City were studied through random sampling in different subareas of Kunming City. The results show that Rhodod...Species, application frequencies, habitat features and combination modes of Rhododendron species in Kunming City were studied through random sampling in different subareas of Kunming City. The results show that Rhododendron species were widely applied in parks, residential areas and street-side green spaces. Diversified application patterns presented excellent landscape effects, but only a few species were used, for example, Rhododendron pulchrum Sweet., R. hybridum, R. simsii var. simsii, while R. delavayi Franch. and R. simsii var. mesembrinum Rehd. were occasionally found. Landscaping features of Kunming were characterized by domination of Rhododendron species, advantages and problems of Rhododendron species in landscaping of Kunming City were analyzed. It was proposed that introduction and domestication of wild Rhododendron species should be enhanced, rich germplasm resources of Rhododendron species should be fully used, and more indigenous Rhododendron species applied.展开更多
物种分布与环境因子之间存在着紧密的联系,因此利用环境因子作为预测物种分布模型的变量是当前最普遍的建模思路,但是绝大多数物种分布预测模型都遇到了难以解决的"高维小样本"问题。该研究通过理论和实践证明,基于结构风险...物种分布与环境因子之间存在着紧密的联系,因此利用环境因子作为预测物种分布模型的变量是当前最普遍的建模思路,但是绝大多数物种分布预测模型都遇到了难以解决的"高维小样本"问题。该研究通过理论和实践证明,基于结构风险最小化原理的支持向量机(Support vector machine,SVM)算法非常适合"高维小样本"的分类问题。以20种杜鹃花属(Rhododendron)中国特有种为检验对象,利用标本数据和11个1km×1km的栅格环境数据层作为模型变量,预测其在中国的潜在分布区,并通过全面的模型评估——专家评估,受试者工作特征(Receiver operator characteristic,ROC)曲线和曲线下方面积(Area under the curve,AUC)——来比较模型的性能。我们实现了以SVM为核心的物种分布预测系统,并且通过试验证明其无论在计算速度还是预测效果上都远远优于当前广泛使用的规则集合预测的遗传算法(Algorithm for rule-set prediction,GARP)预测系统。展开更多
基金Supported by Scientific Research Start-up Foundation(A2002160)Yunnan Province Natural Science Foundation(2009CD064)
文摘Species, application frequencies, habitat features and combination modes of Rhododendron species in Kunming City were studied through random sampling in different subareas of Kunming City. The results show that Rhododendron species were widely applied in parks, residential areas and street-side green spaces. Diversified application patterns presented excellent landscape effects, but only a few species were used, for example, Rhododendron pulchrum Sweet., R. hybridum, R. simsii var. simsii, while R. delavayi Franch. and R. simsii var. mesembrinum Rehd. were occasionally found. Landscaping features of Kunming were characterized by domination of Rhododendron species, advantages and problems of Rhododendron species in landscaping of Kunming City were analyzed. It was proposed that introduction and domestication of wild Rhododendron species should be enhanced, rich germplasm resources of Rhododendron species should be fully used, and more indigenous Rhododendron species applied.
文摘物种分布与环境因子之间存在着紧密的联系,因此利用环境因子作为预测物种分布模型的变量是当前最普遍的建模思路,但是绝大多数物种分布预测模型都遇到了难以解决的"高维小样本"问题。该研究通过理论和实践证明,基于结构风险最小化原理的支持向量机(Support vector machine,SVM)算法非常适合"高维小样本"的分类问题。以20种杜鹃花属(Rhododendron)中国特有种为检验对象,利用标本数据和11个1km×1km的栅格环境数据层作为模型变量,预测其在中国的潜在分布区,并通过全面的模型评估——专家评估,受试者工作特征(Receiver operator characteristic,ROC)曲线和曲线下方面积(Area under the curve,AUC)——来比较模型的性能。我们实现了以SVM为核心的物种分布预测系统,并且通过试验证明其无论在计算速度还是预测效果上都远远优于当前广泛使用的规则集合预测的遗传算法(Algorithm for rule-set prediction,GARP)预测系统。
