عنوان مقاله [English]
This study extracts diagnostic species of plant communities of Khybus protected forest based on statistical analysis of phytosocilogical data. The affinity of vascular plants was calculated using a statistically defined coefficient of fidelity. Additionally, with the development of Total Phi Fidelity Index, TPFI, as a new criterion in fidelity attention with sharpness and uniqueness indices the evaluation of plant communities was created. It is shown that there are 33 diagnostic species in the plant communities without box tree (Buxus hyrcana Pojark.) based on Φ coefition in which the contribution of herbaceous species were 76.7% (22 species) and the contribution of woody species were 33.3% (11 species), while in the list of diagnostic species in the box tree plant communities (with 7 species) the herbaceous species were never presented. Chi- square goodness of fit test (χ2= 234.97, df= 25) in reviewing the consistence of TPFI and TWINSPAN are not independent and they are applicable together with 99% probability. In this case, conformity and correlation between the two mentioned methods were assessed 78.9% and 93.5% from basis of Kappa index and Spearman correlation coefficient. Also, quality assessment of TPFI and TWINSPAN results based on sharpness and uniqueness showed that the mean values of these two indices in the provided vegetation units by the TPFI method (Sh=33.76, U=88.74) is consistently higher than the submitted plant communities by TWINSPAN (Sh=29.98, U=71.51). Totally, we are concluded that TPFI method can be used as a new approach in improvement of TWINSPAN results in plant community classification.