Open Collections

A hybrid model to estimate natural recruitment and growth in stands following mountain pine beetle disturbance

University of British Columbia

A method of linking SORTIE-ND and PrognosisBC was developed for the purpose of predicting natural regeneration and forecasting future stand conditions in mountain pine beetle (Dendroctonus ponderosae Hopkins - MPB) attacked stands in the Interior Douglas-fir (IDF) and Sub-Boreal Spruce (SBS) biogeoclimatic ecosystem zones of central and southeastern British Columbia. PrognosisBC, a spatially-implicit growth model, lacked a submodel suitable for predicting natural regeneration in unsalvaged MPB-disturbed stands. To fill this gap, estimates of regeneration (trees <7.5 cm diameter at breast height - DBH) were supplied to PrognosisBC using the light-mediated forest dynamics model SORTIE-ND and the linked model was used to forecast future stand conditions. In order to improve results, a density-dependent system of crown allometry equations to predict crown depth and crown radius was developed and then added to SORTIE-ND. The equations used stand-level measures of stems ha-¹, basal area (m² ha-¹), and the basal area of trees taller than the target tree to explicitly account of the effects of crowding on the crown axes. Additionally, crown radius and crown depth were used as dependent regressors. The equations were fit using a nonlinear three-stage least squares estimator and generally provided good estimates of crown depth and crown radius for lodgepole pine (Pinus contorta var. latifolia), hybrid spruce (Picea engelmannii x glauca (Moench) Voss), Douglas-fir (Pseudotsuga menziesii var. glauca (Beissn.) Franco) and trembling aspen (Populus tremuloides Michx.). Tests of the hybrid model with the improved system of crown allometry equations were performed using reconstructed plot data collected from natural stands disturbed by MPB 25-years ago. The hybrid model provided good estimates (small mean bias and low root mean square error) for the basal area of advance regeneration (2 < DBH < 7.5 cm) for lodgepole pine (Pinus contorta var. latifolia). The best estimates were achieved when trees <7.5 cm DBH were transferred from SORTIE-ND to PrognosisBC 15-years after MPB-disturbance. For trees <2 m in height, poor estimates of stems ha-¹ where obtained. Despite the shortcomings with respect to trees <2 m tall, the results suggest that linking SORTIE-ND and PrognosisBC is an effective method of building a hybrid model capable of being used in MPB-disturbed forests. However, full parameterization of the SORTIE-ND model is likely needed to obtain accurate estimates for all sizes of natural regeneration.

Thesis/Dissertation

application/pdf

2009-01-20

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/3819

Forestry

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/