Near-normal, empirical, and identity yield tables for estimating stand growth

Abstract

Historically, forest growth was estimated using a normal or near-normal yield table as a density standard, and a relative density change equation to estimate approach to the standard. Although normal yield tables have come under intense criticism, critics have generally ignored the relative density change equation. If a yield table captures the "true" relations between volume, age, and site for a species, the relative density change equation can be a simple function of initial relative density and age. If a yield table does not capture the true relations between volume, site, and age, the inadequacy can be overcome by developing more complex relative density change equations, i.e., by transferring representation burden to the change equation. Introduced in the present paper is the concept of an identity yield table (all entries are one), wherein the entire burden of representing the relations between volume, site, and age is transferred from a density standard to a relative density change equation. Modern whole stand (net) growth models are equivalent to historical relative density change equations based on identity yield tables. The conjecture of a continuum of methods to estimate growth from near-normal to empirical to identity yield tables, each with an appropriate relative density change equation, and each equally accurate, is tested on Wisconsin jack pine (Pinusbanksiana Lamb.). The empirical yield table and its relative density change equation were more biased than near-normal and identity-based projection systems.