群落相似性和聚类分析.pptxVIP

Estimating Community Parameters;第四章 群落相似性和聚类分析 第一节 相似性测量 在群落研究中，生态学家经常会得到某一群落的物种组成和数量。例如在保护区研究中，我们经常要回答的问题是这几个保护区他们在区系组成上有什么不同？哪些更相似，哪些差异较明显？要回答群落分类的这样复杂问题，我们先以测量两个群落的相似性着手。 4.1.1 Binary Coefficients 4.1.2 Distance Coefficients 4.1.3 Correlation Coefficients 4.1.4 Morisita’s Index of Similarity;Binary Coefficients;;where =Euclidean distance between samples and =Number of individuals (or biomass) of species in sample =Number of individuals (or biomass) of species in sample =Total number of species ;Euclidean distance increases with the number of species in the samples, and to compensate for this, the average distance is usually calculated: where = Average Euclidean distance between samples j and k = Euclidean distance (calculated in equation 11.5) n = Number of species in samples Both Euclidean distance and average Euclidean distance vary from 0 to infinity; the larger the distance, the less similar the two communities.;One of the simplest metric functions is called the Manhattan, or city-block, metric: where = Manhattan distance between samples j and k = Number of individuals in species i in each sample j and k n = Number of species in samples This function measures distances as the length of the path you have to walk in a city—hence the name. Two measures based on the Manhattan metric have been used widely in plant ecology to measure similarity.;Bray-Curtis Measure Bray and Curtis (1957) standardized the Manhattan metric so that it has a range from 0 (similar) to 1 (dissimilar). where B = Bray-Curtis measure of dissimilarity = Number of individuals in species i in each sample (j, k) n = Number of species in samples Some authors (e.g., Wolda 1981) prefer