The Validation of the Mixedwood Growth Model (MGM) for Use in Forest Management Decision Making

Abstract

We evaluated the Mixedwood Growth Model (MGM) at a whole model scale for pure and mixed species stands of aspen and white spruce in the western boreal forest. MGM is an individual tree-based, distance-independent growth model, designed to evaluate growth and yield implications relating to the management of white spruce, black spruce, aspen, lodgepole pine, and mixedwood stands in Alberta, British Columbia, Saskatchewan, and Manitoba. Our validation compared stand-level model predictions against re-measured data (volume, basal area, diameter at breast height (DBH), average and top height and density) from permanent sample plots using combined analysis of residual plots, bias statistics, efficiency and an innovative application of the equivalence test. For state variables, the model effectively simulated juvenile and mature stages of stand development for both pure and mixed species stands of aspen and white spruce in Alberta. MGM overestimates increment in older stands likely due to age-related pathology and weather-related stand damage. We identified underestimates of deciduous density and volume in Saskatchewan. MGM performs well for increment in postharvest stands less than 30 years of age. These results illustrate the comprehensive application of validation metrics to evaluate a complex model, and provide support for the use of MGM in management planning.

Keywords

• forest modeling
• model evaluation
• individual tree model
• boreal forest