A Technique for Predicting Merchantable Tree Heights for use in Conjunction with Forest Inventories

Abstract

A merchantable tree height regression model has been developed which can predict heights and compute resultant volumes for the northern hardwood species group. Test data was comprised of successional continuous forest inventory files supplied by a northern Wisconsin paper manufacturing company. Percent comparisons between actual (measured) and predicted heights and computed volumes were within acceptable accuracy limits. Model flexibility can accomodate multiple silvicultural consider­ at s including major regional soil groups, stocking levels, and site classifications.