On Estimation in k-tree Sampling

—The plot design known as k-tree sampling involves taking the k nearest trees from a selected sample point as sample trees. While this plot design is very practical and easily applied in the field for moderate values of k, unbiased estimation remains a problem. In this article, we give a brief introduction to the history of distance-based techniques in forest inventory sampling, present a new and simple approximation technique for estimation, and describe how to eventually develop a design-unbiased estimator. This article draws on two manuscripts that were recently published (Kleinn and Vilcko 2006a, Kleinn and Vilcko 2006b), in which more details are elaborated.