Abstract
Species diversity is one of the most important indices used to evaluate the sustainability of forest communities. The sampling method and the number of plots are factors affecting the estimation of plant biodiversity. In the present study, effects of different inventory grids on estimation of tree species diversity were compared in semi-arid forests of Iran. There were selected 50 hectares of these forests representing the regional forests. Sampling procedures were carried out on circular plots (1,000m2) within inventory grids, with dimensions of 50 × 50 m (200 plots), 100 × 50 m (100 plots), 100 × 100 m (50 plots), 200 × 50 m (50 plots), 200 × 100 m (25 plots), and 250 × 200 m (10 plots). For each plot, the type of the species and the number of trees were recorded. Simpson (1-D), Hill (N2), Shannon-Wiener (H’), Mc Arthur (N1), Smith-Wilson (Evar) and Margalef (R1) indices were used to estimate the tree species diversity. The inventory grid was evaluated based on the precision and cost criteria (E%2 × T). The obtained sampling error values showed that the inventory grid consisting of 200 plots exhibited more accuracy for estimating the biodiversity indices. But based on the results of E%2 × T, the inventory grid with 25 plots was selected as the most appropriate one for estimating the tree species diversity in semi-arid forests. The results of this study can also serve to estimate the tree species diversity in other semi-arid forests of Iran.