Using small area estimation and Lidar-derived variables for multivariate prediction of forest attributes

This article is part of a larger document. View the larger document here.

  • Authors: Mauro, F.; Monleon, Vicente; Temesgen, H.
  • Publication Year: 2015
  • Publication Series: General Technical Report (GTR)
  • Source: In: Stanton, Sharon M.; Christensen, Glenn A., comps. 2015. Pushing boundaries: new directions in inventory techniques and applications: Forest Inventory and Analysis (FIA) symposium 2015. 2015 December 8–10; Portland, Oregon. Gen. Tech. Rep. PNW-GTR-931. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. p. 73.

Abstract

Small area estimation (SAE) techniques have been successfully applied in forest inventories to provide reliable estimates for domains where the sample size is small (i.e. small areas). Previous studies have explored the use of either Area Level or Unit Level Empirical Best Linear Unbiased Predictors (EBLUPs) in a univariate framework, modeling each variable of interest at a time, and not considering their potential correlation. Yet most forest inventory variables such as basal area (G) and volume (V) are strongly correlated. In this situation, EBLUPs for multivariate responses can improve the quality of the estimates. In this study, we apply multivariate SAE techniques in a LiDAR assisted forest inventory. We compare the resulting estimates to those obtained using traditional univariate SAE techniques and other synthetic estimates widely used in forest inventories. The study area is a set of Bureau of Land Management (BLM) and Bureau of Indian Affairs (BIA) owned forest lands in Southwestern Oregon. The small areas are the subsets of the BLM/BIA lands in the study area contained in each 12 level Hydrologic Unit Codes (HUC12). Variables of interest were G and V. A total of 899, 0.125 acre plots were measured in the field. Univariate and multivariate fixed effects and mixed effects regression models were developed. Preliminary results show that correlation between HUC12 level random effects for different variables is moderate while residuals for different variables are highly correlated.

  • Citation: Mauro, F.; Monleon, V.J.; Temesgen, H. 2015. Using small area estimation and Lidar-derived variables for multivariate prediction of forest attributes. In: Stanton, Sharon M.; Christensen, Glenn A., comps. 2015. Pushing boundaries: new directions in inventory techniques and applications: Forest Inventory and Analysis (FIA) symposium 2015. 2015 December 8–10; Portland, Oregon. Gen. Tech. Rep. PNW-GTR-931. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. p. 73.
  • Posted Date: January 20, 2016
  • Modified Date: September 20, 2016
  • Print Publications Are No Longer Available

    In an ongoing effort to be fiscally responsible, the Southern Research Station (SRS) will no longer produce and distribute hard copies of our publications. Many SRS publications are available at cost via the Government Printing Office (GPO). Electronic versions of publications may be downloaded, printed, and distributed.

    Publication Notes

    • This article was written and prepared by U.S. Government employees on official time, and is therefore in the public domain.
    • To view this article, download the latest version of Adobe Acrobat Reader.