Methods

Study Site

Figure 2. Map of treatment blocks located at Robb, Alberta.

In 2008, a randomized block design was established near Robb, Alberta, in the Upper Foothills natural subregion (Natural Regions Committee 2006) to study the effects of simulated mountain pine beetle (MPB) outbreaks on lodgepole pine forests. Study sites were dominated by 110- to 120-year-old lodgepole pine trees and were located on brunisolic, grey, luvisolic soils (McIntosh and Macdonald 2013a).

Three replicate blocks were established, each containing four 0.48 ha (60 m x 80 m) plots that were randomly assigned to one of four treatments: control, 50% MPB mortality, 100% MPB mortality, and salvage logging (Figure 2; McIntosh and Macdonald 2013a).

Glyphosate (N-(phosphonomethyl) glycine) was applied in 2009 to achieve ~50% (moderate) and 100% (high) simulated MPB-induced mortality treatments by chemical girdling. Salvage logging was achieved by clear-cutting to mimic the management option that is done on many MPB attacked forests (McIntosh and Macdonald 2013a).

Data Collection

Data collected one year prior to treatment as a reference (2008), in the year of treatment (2009; green attack), one year post-treatment (2010; red attack) were previously reported by McIntosh and Macdonald (2013a; 2013b). Five years post-treatment (2014; grey attack), and seven years post-treatment (2016; grey attack) data were reported by Steinke et al (2020). In the summer of 2025 new data on overstory, canopy cover, downed woody material, understory vegetation, and forest floor nutrient availability was collected (Fig. 3).

Figure 3. Experimental framework of simulated mountain pine beetle (MPB) mortality and salvage treatment.

Within each treatment plot, nine sampling points were established in a systematic grid with a minimum spacing of 20-m (Fig. 4). This resulted in a total of 108 sampling points across the three blocks. At each sampling point, trees were measured in 8-m radius plots. Downed woody material was sampled using the line intersect method which run from each sampling point out 8-m in a randomly selected direction. In the center a 1x1-m quadrat was established to assess canopy cover, understory vegetation and soil nutrient availability.

Figure 4. Representation of the experimental study design, in which each block contains a 0.64-hectare (60 x 80 m) plot for each treatment and each plot contains nine nested sampling points (including 8-m radius overstory plots) in a systematic grid. A 1 x 1-m quadrat is set up at the center of each sampling point, and a transect of downed woody material is set up as shown by the 8-m arrow above.

Canopy cover

Hemispherical canopy photographs were captured at each sampling point using a digital Nikon Coolpix 4500 equipped with an FC-E8 fisheye lens. The camera was leveled on a tripod approximately 1.3 meters above the forest floor with the bottom oriented towards north. Autobracketing was used at −3, 0, and +3 to achieve optimal lighting along with a timer, to prevent movement during the photo shoot.

Hemispherical canopy photographs were processed using the Spot Light Intercept Model (SLIM, Version 3.02a). The resulting gap fraction ratios were converted to percent canopy cover by subtracting the gap fraction from 100 at each individual sampling point.

Overstory trees

In 8-m radius circular plots (0.02 ha) data from each overstory tree was recorded at each sampling point. For each tree, species, live/dead status and diameter at breast height (DBH) was recorded. Basal area (m²/ha) and tree density (trees/ha) was assessed by live/dead status and calculated from individual tree DBH measurements and stem counts using an expansion factor of 50, corresponding to the ratio between a full hectare and the 200 m² circular plot area.

Downed woody material

The line-intersect method was used to measure downed woody material (DWM) along an 8-meter transect originating from each sampling point at a randomly selected angle to guard against a possible orientation bias (Van Wagner 1968, 1982; Brown 1974). The diameter of the DWM was measured at each point of intersection with calipers and then sorted into diameter size class categories: 0-0.5 cm, 0.5-1.0 cm, 1-3 cm, 3-5 cm, 5-7 cm, and > 7 cm (McRae et al. 1979; Van Wagner 1982). Pieces in the smaller diameter size classes of 0–0.5, 0.5–1, and 1–3 cm were counted along the first 2 m length of each transect. Size classes of 3–5 and 5–7 cm were recorded along the first 4 m, while length, diameter, and decay class were recorded along the full 8 m for all pieces > 7 cm.

Total biomass (Mg/ha) of DWM was calculated for small (0–7 cm) and large (> 7 cm) pieces following Van Wagner (1982) and Nalder et al. (1997). For small pieces, size-class-specific composite factors for Pinus contorta were used, while for large pieces, specific gravity values were derived from decay classes as outlined by Delisle and Woodard (1988), distinguishing between sound (decay class ≤ 3) and rotten wood (decay class > 3).

Understory plant community

Understory plant community was sampled within 1 x 1-m quadrats at the center of each sample points. Percent cover (0-100) of each species/taxa was estimated to the nearest 1/10th percent for species with <1% cover and to the nearest 1% for species with > 1% cover. Total cover could exceed 100% due to vertical overlapping of some species. Species scientific names were confirmed using the USDA Plants database (http://plants.usda.gov/). Cover estimates were also recorded for litter, tree/snag boles, downed woody material, exposed mineral soil, and rock.

Soil nutrients

To measure soil nutrient availability, ion exchange membranes called Plant Root Simulators (PRS®, Western Ag Innovations, Inc., SK, Canada) were buried in the ground during the growing season (mid-June to mid-September). Cation probe (purple) adsorbed NH4-N , K, Ca, Mg, Fe, Mn, Cu, Zn, Pb, Al and Cd; and anion probe (orange) adsorbed NO3-N, SO4, P and BO3. One pair (1 cation and 1 anion) of probes was installed vertically at each of the four corners of the quadrats, except in 2025 where they were only installed in sample points corners of each plot (n = 48). After they were removed, they were cleaned with deionized water and shipped to Western Ag for analysis.

Not all variables were measured across all treatments plots and years (Table 1). Overstory structure (basal area, tree density, and canopy cover) was not measured in the Salvage treatment between 2009 and 2016, as the clearcut removed all overstory trees and the regenerating stand had not yet reached a measurable size. Soil nutrient availability in 2025 was measured only at plot corners due to logistical constraints.

Table 1. Sampling design summary showing variables measured across treatments plots and years. Green cells indicate complete sampling across all four treatments (Control, 50-killed, 100-killed, and Salvage) at all 108 sampling points. Amber cells indicate partial sampling limited to Control, 50-killed, and 100-killed treatments plots (81 points), or reduced sampling points. Soil nutrients in 2025 were measured only at plot corners of each treatment plots (n = 48). Dashes indicate years where the variable was not sampled.

All analyses were conducted in R (version 4.4.1) using the vegan, lme4, lmerTest, and emmeans packages. All figures were produced in R using ggplot2. Treatment and year colors were selected from the RColorBrewer package (Neuwirth 2022) to ensure colorblind-friendly visualization.

Report abuse