James McCarter, University of Washington, firstname.lastname@example.org (Presenter)
Large-scale assessments of forested landscapes are becoming more common as we demand more from our natural resources. A wealth of inventory data is available, but each source presents issues with data quality, consistency across data multiple data sets, and area represented by each plot. Formatting the inventory information for use in growth and yield models is only the first challenge when projecting the future condition of forested landscapes. The inventory information does not always have the correct coding of location and site information that is used by the growth model. Extensive databases of code mappings need to be developed for handling the translation between data and modeling systems. In some cases species have to be re-coded for use in the growth models. Each of these decisions can have significant influence on the results that may vary 50% or more with improper selections. In addition the growth models often need to be constrained in growth estimates to prevent over predictions. Preforming estimates of biomass or carbon from forest inventory presents an additional challenge with a wide variety of published biomass equations available. Lessons learned from three large scale projects (state wide, regional, and national) will be presented along with current solutions.