A form of two-phase sampling utilizing regression analysis

Abstract

A two-phase sampling technique was introduced and tested on several horizontal point sampling inventories of hardwood tracts located in northern West Virginia and western Maryland. In this sampling procedure species and dbh are recorded for all “in-trees” on all sample points. Sawlog merchantable height was recorded on a subsample of intensively measured (second phase) sample points and these heights were predicted on the non-intensive (first phase) sample points. Regression analysis was used to predict heights on first phase points in order to achieve an estimate of board foot volume per acre for every point. Results indicate an improved estimate of the mean volume per acre when compared to traditional double sampling using basal area as the auxiliary variable. An unbiased sampling error was also achieved in this process.

Parent Publication

• Proceedings, 15th central hardwood forest conference

Citation

Fiery, Michael A.; Brooks, John R. 2007. A form of two-phase sampling utilizing regression analysis. e-Gen. Tech. Rep. SRS–101. U.S. Department of Agriculture, Forest Service, Southern Research Station: 60-68 [CD-ROM].