基于主成分聚类分析的美国黄松引种区划方法

Optimization of Divisional Methods Based on Principal Composition Cluster Analysis on Suitable Ecological Region for Introduced Pinus ponderosa

对我国42个美国黄松引种地的年均温、最热月均温、最冷月均温、极端最高温、极端最低温、≥10℃积温、无霜期、年均降雨量、年蒸发量、平均风速、年日照时数、平均海拔12个气候指标进行了主成分分析和聚类分析,划分为最适宜区、适宜区、次适宜区和不适宜区4种类型,并结合美国黄松的生长状况对2种方法区划的结果进行对比。结果表明:主成分-聚类分析法既可以对多变量数据进行合理地分类,又能对各类优劣程度做出综合评价,能充分反映适宜美国黄松生长的气候情况,与美国黄松实际生长情况比较后,验证了该方法是切实可行的。

Principal component analysis and cluster analysis were used to study 12 climate indicators of Pinus ponderosa in 42 places in China, such as annual average temperature, temperature in the hottest and coldest month, extreme maximum and minimum temperature, ≥10°C accumulated temperature, frost-free period, average annual rainfall and evaporation, the average wind speed, the average wind speed and annual sunshine hours and average altitude. They were divided into 4 types, the most suitable area, suitable area, sub-suitable area and unsuitable area, and comparisons were made among two analytic methods combined with the actual growth of Pinus ponderosa. The results showed that the principal component-cluster analysis not only can classify multivariate data, but also is able to make a comprehensive evaluation; it correctly reflects the most suitable areas for Pinus ponderosa. After testing by the actual growth of Pinus ponderosa, we confirm the method was practicable.