Dominant/Codominant Height and Site Index Models for Aspen in the Lake States

Aspen (Populus spp.) is one of the most important commercial forest types in the Lake States of Michigan, Minnesota, and Wisconsin. Popular site index models for aspen in the Lake States were developed in the 1970s and based on data collected in the 1920s. However, new theories and approaches to site index modeling emerged in the early 2000s and as recently as 2021. This study assessed models derived from such approaches to identify a suitable dominant/codominant height and site index model for aspen in the Lake States. Data from the USDA Forest Service (Forest Service), Forest Inventory and Analysis program were used to fit six popular models derived from the Generalized Algebraic Difference Approach (GADA) and four models derived from the Unified Theory of Algebraic Differences Approach (UTADA). A base-age invariant, site-specific consecutive reiteration methodology was used to fit each model. Models were tested for fit and validation statistics and assessed for desirable characteristics. A model based on GADA was identified for its theoretical grounding, fit and validation accuracy, and reasonable model behavior. The resulting model can be applied as a height prediction model or as a traditional site index model, providing accuracy and flexibility to aspen managers in the Lake States.