Adaptive strip sampling in forest inventory of scattered species of Ulmus glabra in Hyrcanian Forests, northern Iran

Abstract

Abstract. Fazlollahi Mohammadi M, Ahmadlo F, Keshtiban SM. 2015. Adaptive strip sampling in forest inventory of scattered species of Ulmus glabra in Hyrcanian Forests, northern Iran. Nusantara Bioscience 7: 48-54. Some populations of tree species may be sparse but they are highly geographically clustered. A simple random sample of plots could then yield mainly empty plots. Adaptive Cluster Sampling (ACS) is an efficient design for tallying of these populations. Presented study is designed to estimate the abundance of Wych elm (Ulmus glabra) as a scattered tree species in mountain forests of Iran. This study looks at design-based methods for a forest inventory survey, with the goal of improving estimation when the target species has a patchy spatial distribution. In particular, we examined the efficiency and feasibility of modified adaptive cluster sampling in combination with strip sampling (SACS) as a new sampling design in Hyrcanian forests, Iran for scattered species. To collect data, we recorded a number of 845 trees to represent locations of individuals in a spatially aggregated population by using GPS 60CX model. In overall, results showed that SACS was underestimated in estimating coefficient of variance (CVµ) and (Varµ) while it was overestimated in estimating mean of tree density at all quadrate size except for size 15 m square in compared to conventional strip sampling (CSS). Results evidenced that HH estimator was more efficient than HT ones in estimating mean, (CVµ) and variance (Varµ) in compared to (CSS) approach. Practically results of our study using SACS confirmed that a 15 m square sample size of quadrate with four adjacent units was optimized network size to sampling of such populations. Future studies using SACS may be required, especially, in combine to two-phase sampling strategies when cover of these species is of concern.