Proof of Concept for an Approach to a Finer Resolution Inventory

Abstract

This report presents a proof of concept for a statistical framework to develop a timely, accurate, and unbiased fiber supply assessment in the State of Georgia, U.S.A. The proposed approach is based on using various data sources and modeling techniques to calibrate satellite image-based statewide stand lists, which provide initial estimates for a State inventory on a common timeline. The system is based on using Georgia ground inventory data from the forest products industry, enhanced by various geographic information system and remote sensing data, and applied with the k-th ?nearest neighbor? methods to time-series-stratified satellite imagery. The initial estimates are then scaled regionally to the Forest Inventory and Analysis (FIA) summary totals to eliminate potential bias in the initial estimates. The system enhances the FIA inventory data in four significant ways. First, it removes the need for the specific FIA plot coordinates; although, the coordinates, if available, would probably enhance the analysis. Second, it provides a current common timeline of inventory estimates based on the Landsat Thematic Mapper imagery for the given year and season. Third, it provides currently accurate high-resolution area estimates. Last, it uses various auxiliary data available from private and public sources in the State and can easily take advantage of other data as they become available.

Parent Publication

• Proceedings of the fifth annual forest inventory and analysis symposium

Citation

Cieszewski, Chris J.; Iles, Kim; Lowe, Roger C.; Zasada, Michal. 2005. Proof of Concept for an Approach to a Finer Resolution Inventory. In: Proceedings of the fifth annual forest inventory and analysis symposium; 2003 November 18-20; New Orleans, LA. Gen. Tech. Rep. WO-69. Washington, DC: U.S. Department of Agriculture Forest Service. 222p.