Download (PDF 491 KB): https://research.fs.usda.gov/download/treesearch/14403.pdf
Abstract
In forest inventories, missing plots are caused by hazardous terrain, inaccessible locations, or denied access. Maryland had a large number of denied-access plots in the latest periodic inventory conducted by the Northeastern Forest Inventory and Analysis unit. The denial pattern, which can introduce error into the estimates, was investigated by dropping the 1999 denied-access plots in the 1986 periodic inventory. The denied-access plots represented the population in terms of percentages of forest and nonforest, ownership, land use, and cubic-foot volume. Board-foot volume was less representative. Several single imputation group means??Euclidean type? distance measures, multiple regression imputation, and listwise deletion with the adjustment of the stratum weights?are compared for estimating the missing cubic- and board-foot volume on forest land. Information on the forested condition of the denied-access plot can be found only through photointerpretation (PI) or satellite imagery such as Multi-Resolution Land Characteristics, (MRLC). Results were inconclusive following an examination of the standard and sampling errors for the state or the root mean square errors for the denied-access plots. As a result, 2 to 12 percent of the data in increments of 2 percent were dropped randomly in a simulation study; the missing plot attributes estimated using each technique. The best simulation study procedure for PI-based forest/nonforest stratification is PI stratum classification. The best simulation study procedure for satellite-based forest/nonforest stratification is the listwise deletion alternative.
Parent Publication
Citation
King, Susan L. 2005. Comparison of Imputation Procedures for Replacing Denied-access Plots. In: McRoberts, Ronald E.; Reams, Gregory A.; Van Deusen, Paul C.; McWilliams, William H.; Cieszewski, Chris J., eds. Proceedings of the fourth annual forest inventory and analysis symposium; Gen. Tech. Rep. NC-252. St. Paul, MN: U.S. Department of Agriculture, Forest Service, North Central Research Station. 155-162
