Dear R-Sig Colleagues, I am performing a bioregionalization of the Cerrado woody flora in central Brazil.
I am following the steps Produce a dissimilarity matrix from species x site matrix Perform a NMDS on the dissimilarity matrix Interpolate each NMDS axis with Kriging Partition the stacked interpolated rasters with K-means I am willing to change the k-means by the ISO DATA method implemented in ArcGis, which takes an initial number of groups (the same used for the k-means classification). The ISODATA algorithm has some further refinements by splitting and merging of clusters. Clusters are merged if either the number of members (pixel) in a cluster is less than a certain threshold or if the centers of two clusters are closer than a certain threshold. Clusters are split into two different clusters if the cluster standard deviation exceeds a predefined value and the number of members (pixels) is twice the threshold for the minimum number of members. It implements a set of rule-of-thumb procedures that have been incorporated into an iterative classification algorithm. Many of the steps used in the algorithm are based on the experience obtained through experimentation. The ISODATA algorithm is a modification of the k-means clustering algorithm and is supposed to overcome the disadvantages of k-means. The classification produced preserved the main patterns found in kmeans while reducing the number of groups. However, I have not found anything in the literature of bioregionalization using ISO DATA, only kmeans. Do you know of any reasons I should not use ISO DATA for this aim? Thank you very much in advance for any inputs, Best, Alexandre -- Dr. Alexandre F. Souza Professor Associado Chefe do Departamento de Ecologia Universidade Federal do Rio Grande do Norte CB, Departamento de Ecologia Campus Universitário - Lagoa Nova 59072-970 - Natal, RN - Brasil lattes: lattes.cnpq.br/7844758818522706 http://www.esferacientifica.com.br https://www.youtube.com/user/alexfadigas http://www.docente.ufrn.br/alexsouza orcid.org/0000-0001-7468-3631 <http://www.docente.ufrn.br/alexsouza> [[alternative HTML version deleted]] _______________________________________________ R-sig-ecology mailing list R-sig-ecology@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
