Estimating and circumventing the effects of perturbing and swapping inventory plot locations

Abstract

The Forest Inventory and Analysis (FIA) program of the USDA Forest Service reports data and information about the Nation's forest resources. Increasingly, users request that FIA data and information be reported and distributed in a geospatial context, and they request access to exact plot locations for their own analyses. However, the FIA program is constrained by law from disclosing exact locations and, instead, releases only approximate locations. The results of the study indicate that the effects of using approximate locations are negligible for design-based estimates for circular areas of radii greater than approximately 20 mi (32 km). For model-based estimates, the results are less definitive and depend on the modeling technique, the spatial resolution of units on which variables are observed, spatial correlation in the variables, and the quality of fit of the model to the data. Several methods for circumventing the effects of the approximate locations are discussed.