论文部分内容阅读
应用主成分分析，因子分析和逐步回归分析技术对黑河地区天然落叶松幼中龄林进行聚类分析和差别分析得到研究地区森林经营的４个类型组，８个亚类和１６个立地类型。对各类技术指标进行频数分析后得到４个类型组的宏观经营区划结果。结果表明，上述分类及区划技术克服了立地类型的地域分异缺点，避免了单一多元样本分类方法所产生的局限性和不足之处。
Principal component analysis, factor analysis and stepwise regression analysis were used to cluster and analyze the young middle-aged larch in Heihe region. Four types, eight sub-categories and 16 site-based types of forest management were studied. After analyzing the frequency of all kinds of technical indexes, the results of macro-management division of 4 types of groups were obtained. The results show that the above classification and zoning technology overcomes the shortcomings of geographical differentiation and avoids the limitations and shortcomings of single multivariate sample classification methods.