Validation of FIBER 3.0 for tolerant hardwood stands in Ontario
- Download PDF (1498407)
- This publication is available only online.
Forest Resaerch Information Paper. Issue 124. 1995. pp. 1-32
Growth and yield projections aid foresters in assessing timber management opportunities and in making management decisions. With these uses, questions arise about the reliability and limits of growth and yield simulators. Using long-term studies of hardwood stands in Ontario the growth simulator FIBER 3.0 has been tested. Short-term (5 years) projections of stand growth are reliable (modelling efficiency from 70% to 90%). Errors for individual stand projections, however, range from -30% to 40%. The magnitude of error in individual stand projections is influenced by species composition and size of sample plots. Small sample sizes do not fully account for the variation in stand structure. The best results (prediction errors below 20%) occur when projecting stands that have 50% to 70% sugar maple and are sampled using plots > 0.2 ha. When the length of the projection period increases, the reliability of FIBER 3.0 decreases and errors increase. Longterm (15- and 20-year) projections underestimate key stand characteristics and the performance of the model is less satisfactory. An examination of various site class options suggests that the site class component does not significantly influence model performance. A comparison of predicted and observed diameter distributions indicates that FIBER 3.0 is more accurate in uneven-aged stands than in even-aged. The model is more reliable in basal area prediction of small sawlog classes than of larger sawlog classes.
Bankowski, Jacek; Dey, Daniel C.; Woods, Murray; Rice, Jim; Boysen, Eric; Batchelor, Brian; Miller, Roj. 1995. Validation of FIBER 3.0 for tolerant hardwood stands in Ontario. Forest Resaerch Information Paper. Issue 124. 1995. pp. 1-32