Content uploaded by Gaetan Pelletier
Author content
All content in this area was uploaded by Gaetan Pelletier on May 21, 2019
Content may be subject to copyright.
A preview of the PDF is not available
FACTORING IN FORM AND RISK: USING NHRI’S TREE CLASSIFICATION SYSTEM TO REFINE MODELS
Abstract
a summary of 3 projects exploring the use of the NHRI tree classification system
ResearchGate has not been able to resolve any citations for this publication.
Northern hardwood species display a variety of forms and defects that can reduce stem quality and complicate their timber management. However, for the most part, growth and yield models do not account for the influence of stem form and damage. This study determined the influence of stem form and damage on growth, survival, and projected future sawlog value among several northern commercial hardwood species. To accomplish this, hardwood trees on 112 permanent plots across three long-term research sites in Maine were assigned stem form and risk classes using a tree classification system developed in New Brunswick. A highly significant influence of stem form and risk on annualized individual-tree diameter increment and survival was found. Inclusion of these equations into a regional growth and yield model highlighted the importance of stem form and defects on long-term simulations as projected stand-level future value was significantly reduced by over 17%, on average (range of 13% to 28%), when compared with projections that did not include that tree-level information. The results highlight the importance of stem form and defects, as well as the need to account for them, in growth and yield applications that assess the forecasted value of commercially important hardwood stands.
The occurrence (probability) and extent (proportion) of tree internal stem decay are important attributes influencing potential wood quality and value, but variation in decay by species, tree size and geographic range are rarely evaluated and modelled. In this analysis, we used 1246 destructively sampled trees across 33 species in the northeastern United States and New Brunswick, Canada to determine the factors influencing the combined probability and proportion of decay. In the process, we evaluated three modelling approaches including a two-part conditional model, multinomial model and generalized additive model for location, scale and shape (gamlss) that simultaneously predicted both probability and proportion of decay. Predictive capability for all three methods were nearly identical when classifying decay occurrence. Compared with the other methods, the gamlss model had a lower mean bias and root mean square error (RMSE) when predicting decay extent. Tree diameter to height ratio (ratio of diameter at breast height to total height), height, crown ratio, species tolerance to flooding and drought, leaf longevity, and an assessment of perceived tree risk of mortality (risk class) were selected as predictors in the best overall model for decay occurrence. For predicting decay extent, the best model included risk class, crown ratio and the last freezing date of spring. Further analysis identified significant species differences, which we used to develop functional species groupings based on decay occurrence and extent. Despite these observed relationships, a high degree of unexplained variation remained, highlighting the challenges of modelling decay in trees of different species across a range of growing environments.
Northern hardwood trees display a wide variety of stem forms and defects, which can substantially reduce their financial value and also complicate their silviculture. While attributes of stem form and defect have been incorporated into tree classification systems, their ability to assess product value and recovery in standing trees has not been adequately tested. To address this issue, we classified stem form and risk using a system developed by the Northern Hardwoods Research Institute (NHRI) for four species across several locations in Maine, New Hampshire, and New Brunswick: sugar maple (Acer saccharum Marsh.), red maple (Acerrubrum L.), yellow birch (Betula alleghaniensis Britton), and northern red oak (Quercus rubra L.). Using these data, we (i) quantified interspecific and regional variation in stem form and damage, (ii) related potential sawlog recovery to tree size, form, and risk, and (iii) compared the efficacy of the NHRI system with a commonly used classification system and a continuous measure of stem quality. High variation in both stem form and damage among the species was found, with red maple showing the largest range. A simplified NHRI system including three form classes proved to be sufficient in differentiating sawlog potential in individual trees, while a model using a continuous measure of stem quality (estimated merchantable sawlog height) performed best.
A Tree Classification System for New Brunswick. Version 2.0. Northern Hardwoods Research Institute
- G Pelletier
- D Landry
- M Girouard
Pelletier, G., D. Landry and M. Girouard. 2016. A Tree Classification
System for New Brunswick. Version 2.0. Northern Hardwoods
Research Institute. Edmundston, New Brunswick.