UBC Theses and Dissertations
Forest sampling on two occasions with partial replacement of sample units See, Thomas Elton
Forest sampling is conducted to determine current conditions and trends of change. When current conditions are estimated, most of the commonly used sampling designs specify the spatial distribution of sample units. When estimation of change is desired, several schemes may be employed. Some are combined with current condition inventories; some are independent. The former are relatively imprecise; the latter relatively expensive. A system of temporal distribution of sample units, sampling on successive occasions with partial replacement of sample units, has been developed for simultaneous estimation of current conditions and trends. As the emphasis is on time, rather than area, this system operates with conventional sampling designs to increase their efficiency. This study investigated the theory of sampling with partial replacement to establish the validity of the claims of increased efficiency in comparison with conventional systems. Three cases are examined, through example, for estimation of mean volume per acre and growth in volume per acre. Sample sizes and costs are developed for the situation of simple random sampling of both finite and infinite populations. The comparisons are favorable to the proposed system. The possibilities of using this system with two recent developments in cruising techniques are explored. Finally, the applicability of this system to British Columbia forest surveys is examined. A case is made for transition of the existing provincial system to sampling with partial replacement.
Item Citations and Data