Efficiency of distance sampling methods in estimation of biometric characteristics of wild pistachio (pistachio atlantica subsp. mutica) open stands in Zagros

Background and objectives: Distance sampling methods are a group of commonly-used methods in estimation of quantitative characteristics of forest, although each method and related estimators have different efficiencies due to different factors (such as environmental conditions, stand density, and distribution of trees). Therefore, it seems necessary to know how these methods operate in open forests regarding to their importance and their role in estimation of quantitative characteristics with low cost and time. The objective of this study was efficiency assessment of 12 important distance sampling methods to obtain biometric characteristics of wild pistachio trees (density and canopy closure) in Zagros open stands. This study investigated the distance methods in two single- and multi-distance groups. Moreover, it was aimed to select the appropriate estimator of each method and the efficiency of these methods were compared in the same conditions.
Materials and methods: A wild pistachio stand with low density (19.44 trees ha -1) was selected in Bane Research Site, Firoozabad, Fars province, south western Iran. Using a 100 m × 100 m mesh, 46 sample points were located within the boundary. Six single-distance and six multi-distance sampling methods were applied to estimate the characteristics. Root mean squared error was applied to evaluate the precision and accuracy criterion and comparing the means using one-sample t-test were implemented to assess the accuracy of the results.
Results: Using precision and accuracy criteria, the appropriate estimators were identified in nearest individual ( Cottam et al. ) , nearest neighbour ( Cottam and Curtis ) , and T - square ( Byth ) single - distance methods and point - centered quarter ( Cottam and Curtis ) multi-distance sampling method. To estimate density and canopy closure, nearest individual ( Cottam et al. ) and nearest neighbour ( Cottam and Curtis ) single - distance methods were the most efficient ones. Between multi - distance sampling methods, Byth and Ripley ( accuracy of 3.70 % and p < 0.05 ) , joint ( accuracy of 2.36 % and p < 0.05 ) and shared point ( accuracy of 2.36 % and p < 0.05 ) methods were suitable for density and canopy closure estimation , respectively.
Conclusion: In general , nearest individual and nearest neighbour single - distance methods were efficient enough to estimate the biometric characteristics of wild pistachio trees in the studied open stand precisely and accurately. The effect of sampling network and distribution of trees can be investigated in future studies.