We used a simulation model to investigate possible effects of a severe mountain pine beetle (
The mountain pine beetle (MPB) is currently in the outbreak phase of an infestation cycle throughout much of its range in British Columbia, Canada. Outbreaks are also occurring in other parts of MPB range in western North America and the beetle is expanding into areas previously considered beyond its natural ecological range [
For the first time in recorded history, large numbers of beetles from British Columbia have crossed the Rocky Mountains and are attacking lodgepole pine (
It is possible that Alberta’s nascent outbreak will come to resemble the current epidemic in British Columbia. Many of the factors that led to the outbreak in British Columbia are increasingly prevalent in Alberta: a warming climate, fire suppression leading to extensive mature pine stands, limited access, administrative and economic constraints, and infection sources with the potential for rapid population expansion [
The Government of Alberta has implemented a MBP Management Strategy intended to contain the infestation and maintain the long-term timber supply [
The core elements of the Alberta MPB Management Strategy—control, salvage, and prevention—all involve forest harvesting. Given that the mills in the region have capacity limitations, allocating harvest effort among these management options requires a tradeoff decision. The effects of beetle and management activities on environmental and economic indicators further complicate the decision [
We used an existing simulation model,
In ALCES, the landscape can be stratified into multiple independently tracked classes. For example, a forest land base can be stratified into several stand types and age classes, and different harvest and regeneration strategies can be applied to each stratum. Because we were specifically interested in tracking the effects of the MPB, we used forest inventory data to stratify stands containing pine into three categories: Pure_Pine (
The disturbance and regeneration modules in ALCES are user specified. We used the insect disturbance module to specify the type and age of stands subject to MPB infestation as well as the temporal trajectory of the outbreak (i.e., ha disturbed in each year of the outbreak). The model has the capacity to allow stands to transition to a different stand type after disturbance. The age of stands after disturbance can also be defined by the user (i.e., reset to the zero age class is not necessary).
Our study area is comprised of six forest management areas in the foothills of central Alberta (Figure
The province of Alberta, showing the location of our study area.
We did not model the population dynamics of the MPB per se, but the projected effects of the beetle on pine stands. We assumed, as a worst-case scenario, that 80% of pine stands would be attacked during the outbreak. This is consistent with projections for the MPB outbreaks occurring in many parts of British Columbia [
Temporal trajectory of the simulated MPB outbreak, expressed as the annual percentage of the total area of pine attacked.
All infected stands were salvaged, to the extent that sufficient mill capacity was available (see below). The salvage of infected stands was given priority over scheduled harvesting operations because the removal of infected stands serves as a primary beetle control measure in the Alberta MPB Management Strategy [
In our simulation, the regeneration process was same whether a stand was salvaged or harvested conventionally. Stand age was reset to zero and stands then followed standard growth and yield curves provided by the forest companies in our study area. We assumed that the regeneration efforts applied to salvaged stands would ensure that no changes in stand type would occur.
Infested stands that were not salvaged regenerated naturally. In mixed stands we accounted for the change in internal composition and volume resulting from pine mortality but left the age of the stand unchanged (Table
Transition matrix used to determine the fate of stands after MPB attack in the absence of salvage operations.
Before MPB Attack | After MPB Attack | ||
Forest Type | Area (ha) | Forest Type | Proportio |
Pure_Pine (pine | 637,132 | Pure_Pine | 0.29 |
Low_Pine | 0.71 | ||
High_Pine (pine 50%–79%) | 304,191 | Non-pine Softwood | 0.80 |
Low_Pine | 0.20 | ||
Low_Pine (pine | 366,507 | Hardwood | 0.10 |
Low_Pine | 0.56 | ||
Non-pine Softwood | 0.34 | ||
Hardwood (pine | 770,374 | Hardwood | 1.00 |
Non-pine Softwood (pine | 382,132 | Non-pine Softwood | 1.00 |
We assumed that the MPB epidemic in Alberta would occur within the next 20 years, as an extension of the outbreak currently occurring in British Columbia. Given that the timing of the outbreak cannot be reliably predicted we simulated two versions of the MPB epidemic—one in which the outbreak began immediately and another in which the outbreak was delayed by 20 years. For comparison, we also conducted a simulation in which there was no MPB outbreak. Each of the three MPB scenarios was run in combination with two management scenarios: conventional harvest (CH) and pine strategy (PS). In the CH scenario, conventional harvest rules were applied, as described in the Detailed Forest Management Plans of forestry companies in our study area. The PS scenario was based on the pine strategy component of the Alberta MPB Management Strategy, which seeks to reduce the number of susceptible pine stands by 75% over the next 20 years through directed harvesting by the forest industry [ CH-None: CH with no MPB outbreak. PS-None: PS with no MPB outbreak. CH-Immed: same as CH-None, but with a MPB outbreak in year one. PS-Immed: same as PS-None, but with a MPB outbreak in year one. CH-Delay: same as CH-None, but with a MPB outbreak in year 21. PS-Delay: same as PS-None but with a MPB outbreak in year 21.
The models simulated 100 years of activity. We assumed that MPB attack and forest harvesting were the only forms of disturbance on the landscape (i.e., other forms of industrial development and natural disturbance were not simulated) and that only one MPB outbreak would occur over the course of the simulation. The target AAC for softwoods was set at 4.1 million m3, based on data provided by the forestry companies in our study area. We assumed that mill capacity and AAC could be temporarily increased by 20% to accommodate the surge in wood flow from the pine strategy and from salvage related to the MPB outbreak. Wood volume harvested during the surge cut was applied against long-term even flow requirements, but as per current provincial policy, salvage wood was not. The AAC was not recalculated after the MPB attack.
In both CH-None and PS-None the softwood AAC was achieved throughout the entire 100-year simulation. The relative proportion of forest types did not change in these runs because all harvested stands were regenerated to their original stand type. Fifty-eight percent of the forest was classified as old-growth forest at the start of the simulation (where old-growth is defined as stands older than 80 years for hardwood and older than 100 years for all other stand types). By year 70, the percentage of forest classified as old growth had declined to 22% and 25% in CH-None and PS-None, respectively. The declines in old growth were relatively balanced among forest types in PS-None, but in CH-None the declines were most pronounced in the hardwood and mixedwood forest types (Figure
Representation of old-growth forest in the study area at year 70 of the simulation for each of the management scenarios, by stand type.
PS-None was unable to achieve its objective of reducing the amount of susceptible pine forest by 75%. The area of Pure_Pine and High_Pine older than 60 years (i.e., the highly susceptible stands) was only reduced by 43% at the conclusion of the surge cut in year 20. Over the same period CH-None reduced the area of susceptible stands by 5%.
The CH and PS scenarios were functionally similar when subjected to an immediate MPB attack. Harvesting operations in both cases focused on salvage by year three of the simulation. Once the outbreak subsided, harvesting efforts in both scenarios switched largely to non-pine forest types because little merchantable pine remained on the landscape.
In both immediate outbreak scenarios, the beetle attacked 39% of the forested land base. Constraints on mill capacity meant only 57% of the merchantable pine volume killed by the MPB was salvaged. Stands that were not salvaged were subjected to the model’s transition matrix (Table
Composition of the forest land base in the study area at year 70. Scenarios without MPB are not shown because forest composition does not change in the absence of MPB infestation.
After the outbreak, in both scenarios, the softwood AAC was maintained mostly through the harvest of Low_Pine and Non-pine Softwood stands. This land base could only sustain the original rate of harvest until year 60, at which time a shortage of timber precipitated a 75% decline in softwood harvest volume. Only 7% of the forest was in the old-growth category at the point of collapse, and 5% remained at year 70 (Figure
The proportion of the land base attacked by the MBP in CH-Delay and PS-Delay was 37% and 31%, respectively. Of the merchantable pine volume killed by the beetle, 59% was salvaged in CH-Delay and 97% in PS-Delay. Approximately 30% of the stands attacked by the beetle in PS-Delay were stands regenerating from the pine strategy’s surge harvest. Since these stands were old enough to be attacked by the MBP, but too young to be considered for salvage, the rate of salvage (which only considers eligible stands) was artificially inflated.
Softwood harvest collapsed in both delayed outbreak scenarios, but occurred about ten years later than in the immediate outbreak scenarios. The key features of the temporal dynamics of the PS-Delay scenario are presented in Figure
Temporal dynamics of the PS-Delay scenario showing the total softwood harvest excluding salvage (green tree), softwood salvage, and the remaining area of merchantable pure pine stands (i.e., those older than 80 years).
Our study suggests that the current rate of softwood harvest in our study area cannot be maintained if Alberta’s pine beetle infestation follows a trajectory similar to the outbreak in British Columbia. To do so, forestry companies would have to increase their harvest of non-pine species to levels that are not sustainable. According to our simulations, maintaining current harvest levels after the MPB attack would lead to a general collapse in the softwood timber supply in 60–70 years. If other forms of disturbance such as wildfire and petroleum development had been included in our simulations, the collapse in timber supply would have occurred even earlier.
The proactive pine strategy, meant to reduce the number of highly susceptible stands, could not be effectively implemented in the face of an immediate MPB outbreak. By year three of our simulation of the pine strategy, harvesting operations were focused on salvage instead of green-tree harvest. Following the outbreak, almost no merchantable pine remained to be harvested. The pine strategy was, therefore, no different than conventional harvest in terms of what material was harvested (beetle-killed pine).
Although we did not model it, the endpoint would have been the same if harvesting under the pine strategy had continued to focus on green trees. This is because harvesting operations and the MPB compete for the same target species, and the near-total removal of pine from the land base is the ultimate outcome, regardless of the source of mortality (in our worst-case scenario).
The simulated pine strategy failed to prevent the collapse in wood supply even if the epidemic stage of the infestation was delayed for 20 years, for two main reasons: not enough pine had been removed from the land base to prevent substantial tree mortality from the MPB, and the pine strategy itself contributed to the timber shortfall. Mill capacity constraints limited the pine strategy’s removal of highly susceptible pine to 43%. Additional pine trees remained within Low_Pine stands that were not targeted by the pine strategy. Finally, stands harvested at the beginning of the pine strategy’s surge cut were old enough to be attacked by the peak of the delayed MPB outbreak.
The harvest rules in our conventional harvest scenario were designed to produce an even-aged forest with maximum stand age set by the rotation interval. As a result, the proportion of old-growth forest in the conventional harvest scenario decreased from 58% at the start of the simulation to 22% by year 70. When the MPB was added to the system, the amount of old growth declined to 8% or less in both the conventional harvest and pine strategy scenarios. All stand types were affected because the pine killed by the beetle caused a general shortfall in timber supply that forced the model to increase the rate of harvest of other species to maintain the AAC. A reduction in old-growth forest of this magnitude would significantly reduce habitat supply for species dependent on older forest, potentially reducing their abundance and range.
The harvest rules in our simulation specified that stands be regenerated to their original type. As a result, the composition of the land base in the absence of MPB stayed the same. But when the MPB was added to the system, up to 37% of the Pure_Pine and High_Pine stands transitioned to Low_Pine and Non-pine Softwood. These transitions occurred in stands that were left to regenerate naturally, because they were either too young to be salvaged or there was insufficient mill capacity to process them.
In our simulations, we assumed that all pure and mixed stands containing pine older than 20 years will support the same rate of MPB attack and spread. This assumption is inconsistent with studies that have shown that stand age and pine density are important determinants of susceptibility to MPB [
An important area of research over the next few years will be to determine the actual rate of MBP attack and spread in Alberta and to use this information to undertake more refined projections than those used in our study. It may well be that, in Alberta, the combination of younger stands (via the pine strategy), lower density of pine in mixed forests, and cooler temperatures will collectively serve to slow MPB population growth to a manageable level—or at least, to avoid severe loss of timber.
Our simulations show that forestry companies lack the capacity for fully implementing the surge cut prescribed by the pine strategy. Even though we increased mill capacity by 20% and completely stopped harvesting stands with low pine volume, the 20-year surge cut reduced the amount of susceptible pine by only 43%.
The economic case for increasing mill capacity by even 20% is weak, given the current glut of pine on the market due to the MPB outbreak in British Columbia and the high likelihood of a fall-down in future wood supply [
Given the inability of the pine strategy to achieve its stated objectives, and given that the surge cut itself contributes to the future shortage in wood supply, alternative management options should be explored. For example, consideration should be given to converting the pine land base to mixed forest, in what we call the mixedwood approach, to increase its resistance to the effects of the MPB [
The mixedwood approach offers several advantages over the existing pine strategy. First, by focusing on a smaller land base (pure pine only) it is more achievable with existing mill capacity. Second, if left standing, the non-pine volume in infested mixed stands will help maintain continuity of the timber supply, especially during the midterm [
Given that the MPB is now widely distributed across the eastern slopes of the Rocky Mountains of Alberta and British Columbia, the likelihood of overwintering mortality affecting all beetle populations in Alberta is low [
If the outbreak in Alberta does follow the trajectory observed in British Columbia, then management interventions will have little impact and a catastrophic outcome is likely, at least socioeconomically [
The authors thank their forest industry partners for providing them with forest inventory data and information on their harvesting operations. They also thank John Stadt and colleagues at Alberta Sustainable Resource Development for providing them with the government’s perspectives on MBP management. Funding for this study was provided by the Government of Canada as part of the Mountain Pine Beetle Program administered by Natural Resources Canada and by NSERC-ACR Chair in Integrated Landscape Management at the University of Alberta.