A proposed sampling design for a timber inventory of a Tennessee forest

The purpose of this investigation was threefold. First, an extensive review of aerial forest parameters was undertaken. This was done in order to establish which parameters might serve as realistic indicators of volume for the Central Forest Region into which most of Tennessee's forests are classified. The four aerial forest parameters that were selected were forest type, crown closure, crown diameter, and stand height. Second, a sampling design was proposed for the Central Forest Region of Tennessee based on the aerial forest volume parameters above, small scale photography (1:120,000) available from the National Aeronautics and Space Administration (NASA), medium altitude photography, and some previous knowledge and subsequent determination of timber volume from the ground. The contributions of such a design are that for the first time: (a) small scale photography is proposed for a timber inventory in Tennessee; (b) photogrammetric parameters at any other scale are proposed as an integral part of a timber inventory in Tennessee; and (c) "3-P" statistical sampling theory is the basis for determination of volume at the ground level of an aerial timber volume inventory. Third, the problem of optimization of the proposed sampling design was considered with respect to the two conventional optimization methods: (a) maximization of precision for a given cost or (b) minimization of cost for a given precision. It was suggested that a new and different form of optimization method may evolve as a consequence of the possibility of several, equal results of minimum cost for a given, precision or as a consequence of the possibility of several, equal results of maximum precision for a given cost.