مجله جنگل ایران (May 2017)
Comparison of direct Ordination methods in vegetation analysis in Jamand district of Noshahr Forests
Abstract
This study was performed in order to investigate the changes in vegetation in compartment No.317 (managed forest) and No.318 (control forest) of Jamand district of Golband forest in Nowshar, Mazandaran. The aim of this study is to compare the direct gradient analysis to express the basic structure of environmental data. For this purpose, random-systematic sampling method with regular grid of 100×200 m was used to locate the sampling plots. The plot size was 400 m2 considering to tree and shrub species and in the center of each main plot, one 100 m2 subplot was laid out for herb species. The cover percentage of each plant species was estimated by using Braun-Blanquet scale. The performance of two direct ordination techniques (CCA, RDA) were evaluated with different criteria like length of gradient, Kaiser-Guttman, eigenvalues greater than 0.4 and Procrustean analysis. CANCOO 4.52 and R software were used. Result of Procrustean analysis showed that the correlation between ordination methods were different. According to the length of gradient criteria, linear modal is proper and CCA and RDA analysis cover Guttman-Kaiser criterion with two axis. None of the analysis has more than 0.4 eigenvalue. Percentage of eigenvalue in Redundancy Analysis is more than Canonical Correspondence Analysis. Regarding to Jamand districtdataset, due to the linear structure of Redundancy Analysis, this analysis is a proper choice.
