Auxiliary network sampling for multi-objective national forest inventories: a practical examination of search distances
Building upon the work of Roesch and McWilliams (2007), this paper discusses additional examples of an auxiliary sampling strategy that allows National Forest Inventory (NFI) practitioners to efficiently estimate the attributes of rare events or species. When the rare event is immediately identifiable in the field, such as the tree species itself, a trivial simplification of theory is evident, that being that clusters of networks of trees (of that species) are being sampled by the area-based plot, rather than clusters of individual trees. When this is the case one does not have to think of the design as adaptive but rather as simply auxiliary. This can be of some help in the presentation of a large multi-objective inventory. This paper focuses on a practical examination of network search distances for this special case of an auxiliary mechanism for NFIs such as the Forest Inventory and Analysis (FIA) sample design in the United States.