Guided transect sampling for assessing sparse populations

Abstract

Guided Transect Sampling is suggested for transect based surveys when sampling sparse and geographically scattered populations. It is a two-stage design in which wide strips are selected in the first stage followed by a subsampling procedure guided by prior information in the second stage. As a basis for the second stage, the target area is divided into grid-cells containing covariate data obtained, for example, by remote sensing. The general principle can be coupled with different transect based survey methods such as strip sampling or line transect sampling. Also, different strategies for the second stage guidance can be used, resulting in estimators with different properties. In this study, Guided Transect Sampling was compared with a traditional strip survey under different simulated forest conditions. The comparisons indicated that the proposed method can improve the efficiency of transect based inventories of sparse populations.