Improving County Sawlog Removal Population Estimates using Historical and Distance-Based Mill Procurement Data in the Southern US
摘要
The Timber Products Output (TPO) survey, administered by the Forest Inventory and Analysis (FIA) program of the USDA Forest Service, collects data annually from primary wood processing mills. The survey collects data on the volume of roundwood received and residues generated by the primary forest industry by major tree species and counties of harvest. Prior to 2019, the TPO survey was implemented as a full census of all active mills, with varying frequencies across FIA regions. For Southern FIA, TPO surveys of non-pulp mills were delivered every other year. From 2019 forward, TPO transitioned to a national program, with all FIA regions adopting an annual sample survey. The sample methodology involves stratification of the mill population by state, primary roundwood product and mill measure-of-size. Current methodology assigns population county roundwood removal volume estimates only to the counties in which wood was procured by the sampled mills, which can lead to either underestimation or overestimation of actual county volumes. This study evaluates two proposed methods that leverage auxiliary information to improve population county roundwood removal estimates: (1) the Historical Procurement method, which incorporates historical mill-county procurement information and (2) the Radius Procurement method, which incorporates mill-county distance data. We conducted Monte Carlo analysis to test the performance of these two methods, using TPO data from sawmills canvassed in 2011 and 2013 across 11 states in the Southern FIA. Our results showed that both proposed methods achieved improved accuracy in county sawlog removal estimates, compared to the current methodology.