Among the wood resources potentially available from forest areas are “other removals”. Other removals is a broad category that includes the unutilized wood volume from cut or otherwise killed growing stock, wood from cultural operations (e.g. pre-commercial thinning), or wood from timberland conversion to other uses (e.g., crop land, roads, and urban development but excluding timber volumes removed from the forest inventory due to reclassification of timberland to productive reserved forest land). Other removals data is available through the Timber Product Output database (http://www.fia.fs.fed.us) from the USDA Forest Service Forest Inventory and Analysis Program (FIA) which tracks and reports on the fate of wood harvested from U.S. forests. The FIA collects forest residue data through surveys and questionnaires sent to forest owners and harvesters, and by on-site measurements of the material felled, removed, or left during harvest.
The TPO database provides estimates of other removal quantities by county, by inventory source (i.e., total and growing stock), by tree species, and by land ownership class, in cubic feet and board feet. Estimated quantities for the RPA (Resource Planning Act) survey year 2007 by ownership class and tree class are presented in table 1, and their distribution is shown in figure 1. About 1.66 billion ft3 (~ 24.7 million dry tons) were generated in RPA survey year 2007 in the 48 contiguous states, with about 26% from softwood tree species and the remainder from hardwood species. Less than 9% of the material came from National Forests and other public lands.
The FIA data form the basis of the estimated available other removal quantities found in several studies such as Perlack, 2005; Southern States Energy Board, 2006; Western Governors’ Association, 2006; Encyclopedia of Southern Bioenergy Resources, 2006; 25x25 Initiative, 2006; and Walsh, 2007.
The TPO database is limited to historical data and does not project potential quantities of other removals for future time frames. The Forest and Rangeland Renewable Resources Planning Act (RPA) of 1974 directs the Forest Service to conduct a study every 10 years that projects future costs and availability of timber products, evaluates emerging issues, and examines impacts of policy options. The RPA assessment (Haynes, 2003) projects timberland acres will decline from 503.8 million acres in 1997 to 489 million acres in 2050, mostly as a result of urban development, with the greatest decline projected to be in the Northeast, the Southeast, and the Lake States due to increases in population and income. More recent estimates suggest that the forest acres lost to development could be higher due to the combination of increasing population and an increase in the area of land developed per person (Thompson, 2006). Wood resources removed from these acres could be used for bioenergy and bioproducts, however, quantities may be more unpredictable than other forest operations, and they may not be a sustainable resource (i.e., forest land converted to agricultural uses can potentially be converted back to forestry, but land converted to urban development cannot be easily reverted). Also, the economic feasibility of recovery is reduced by the scattered and often unplanned volumes removed from urban areas, or from the non-commercial aspect of removals from the pre-commercial thinning of small trees that are uneconomical to remove from plantations.
Using the USDA Forest Service FIA data, Perlack (2005) estimated that other removal quantities will increase from the current level of 9.2 million dry tons to 17.4 million dry tons by mid-century, but conducted no economic analysis and provided insufficient information to determine how the rate of growth was determined.
Few studies attempt to estimate supply curves (i.e., quantities available as a function of price) for other removals either for current or future time frames. The Antares Group (1999) estimated the availability 72.2 million green tons of forest residues at prices of less than $4.00/MMBtu, but did not separate the forest resources in their report. Walsh (2000) estimated state forest residue supply curves (including other removals) and summed the state data to obtain total U.S. forest residue quantities at several prices (e.g., 44.9 million dry tons available at delivered prices of $50/dt respectively).
Walsh (2007) estimated other removal collection cost schedules for the years 2007, 2010, 2015, 2020, and 2025 (table 2). The analysis estimates collection costs only--market prices will be higher. Future quantities were estimated by applying projected changes in county housing units (estimated as a function of projected state population increases (U.S. Census Bureau) and state average number of occupants per household and apportioned to counties based on the rate of change of county housing starts between 2000 and 2005) to FIA county data. Estimated harvest/collection costs are based on a regional distribution of non-growing stock harvest costs obtained from a somewhat updated version of a model originally developed by McQuillan (1988), which includes forest inventory data, logging and chipping costs, wood types, site accessibility, site slope, and equipment operability constraints to estimate nine regional supply schedules for softwood and hardwood chips for the base (1983) and future years. The analysis suffers from the inability to fully update the original economic model to account for changes in technology and to fully account for many factors that will affect local housing growth and/or pre-commercial thinning operations over the time frame of the analysis.
