Search

Insect outbreaks, such as the current mountain pine beetle (Dendroctonus ponderosae Hopkins) outbreak in lodgepole pine (Pinus contorta Dougl. ex Loud. var. latifolia Engelm.) forests in British Columbia, are major disturbances in many forests. After an insect outbreak, the advance regeneration typically forms a new canopy, which may be adequate for timber objectives in some stands. Our purpose was to quantify and then model the abundance and spatial distribution of advance regeneration (trees <10.0 m tall). We sampled understory and overstory trees in 28 lodgepole pine stands in south-central British Columbia at two spatial scales: 0.1 ha plots and 25 m2 subplots. We developed models predicting advance regeneration abundance and spatial distribution. Density of advance regeneration averaged 2689 trees/ha (range 120 to 23 000 trees/ha), most of which were <1 m tall. Although advance regeneration was clumped, 75% of the subplots contained at least one individual. Models indicated negative relationships of advance regeneration abundance to overstory basal area and density. Over half the stands had enough advance regeneration to form new stands of adequate density, indicating that use of advance regeneration is a viable option in this mountain pine beetle outbreak and probably other insect disturbances. [ABSTRACT FROM AUTHOR]

Growth response functions (GRFs) that relate the growth of a population to the climate of the sites in which it is tested are gaining attention for their ability to predict impacts of climate change on tree growth. However, nonclimatic site to site variation introduces error into GRFs. Using data from a large lodgepole pine (Pinus contorta Dougl. ex Loud.) provenance test in British Columbia and the Yukon, Canada, a technique is presented that accounts for the effect of nonclimatic variation in GRFs. The mean height of the “local” provenances at each test site was used to predict “site height” from site climate variables in multiple regression. Residuals from the site height equation provided an index of the nonclimatic effect for each site and were included as a covariate in quadratic GRFs that related provenance height at each test site to mean annual temperature at each test site. Inclusion of the nonclimatic index in the model resulted in a moderate or large displacement of GRFs for 25% of the provenances, while increasing mean R2 values for 138 of 140 provenances and decreasing the root mean squared error for 113 of 140 provenances. These results suggest that inclusion of the nonclimatic index in GRF models could substantially affect height predictions for some provenances and reduce prediction error for most provenances. [ABSTRACT FROM AUTHOR]