Mountain Pine Beetle - ualberta.campb/MPB Literature Review April 2005.pdf · MPB Review April 28, 2005 Mountain Pine Beetle Population Dynamics: Insights from empirical evidence

MPB Review April 28, 2005

Mountain Pine Beetle Population Dynamics:

Insights from empirical evidence and modeling1

William A. Nelson*, Alex Potapov†, Anina Hundsdorfer‡, Mark Lewis°, Allan Carroll˜& Fangliang He§

*Centre for Mathematical Biology & Department of Biological Sciences, University of Alberta, Edmonton, AB T6G 2E9, Canada. [email protected]

†Centre for Mathematical Biology, University of Alberta, Edmonton, AB T6G 1G1, Canada. [email protected]

‡Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2H1, Canada. [email protected]

°Departments of Mathematical and Statistical Sciences, plus Biological Sciences, University of Alberta, Edmonton, AB T6G 1G1, Canada. [email protected]

˜Natural Resources Canada, Pacific Forestry Centre, 506 West Burnside Rd., Victoria, BC V8Z 1M5, Canada. [email protected]

§Department of Renewable Resources, University of Alberta, Edmonton, AB T6G 2H1, Canada. [email protected]

1 Literature review on the spatial modeling of Mountain Pine Beetles (report) for NRC project # 8.19 “Modeling Spatio-temporal Patterns of MPB Infestations”

1


INTRODUCTION The mountain pine beetle (Dendroctonus ponderosae Hopkins) is an aggressive bark-

beetle that occurs naturally throughout much of western North America (Logan & Powell

2001). Beetles breed in the phloem tissue of recently dead host trees and, given sufficient

beetle densities, coordinate attacks that can overwhelm the defenses of even the healthiest

trees. While able to breed in many species of pine, mountain pine beetles are most often

found in stands of lodgepole pine (Pinus contorta) and ponderosa pine (Pinus

ponderosae). Insects pests are the largest cause of forest disturbance in terms of both area

affected and economic impact (Logan et al. 2003). Mountain pine beetle outbreaks have

become more intense during the last few decades, and have spread to new areas (Logan &

Powell 2001; Carroll & Safranyik 2003). For example, mountain pine beetle infestations

in Canada have escalated into the largest insect epidemic in the history of the province of

British Columbia, covering millions of hectares (Ebata 2004), and are rapidly spreading

into the province of Alberta (Ono 2003).

Control and prevention of outbreaks requires an understanding of the mechanisms

that determine population dynamics and dispersal. For over half a century, researchers

have studied beetle phenology (Amman & Cole 1983; Amman & Bartos 1991; Bentz et

al. 1991), behavior (Ryker & Rudinsky 1976), outbreak development (Bentz et al. 1993;

Logan & Bentz 1999), host defenses (Reid & Gates 1970; Raffa & Berryman 1982b) and

factors determining stand and individual tree susceptibility (Bartos & Amman 1989;

Elkin & Reid 2004). Several models have been developed to estimate the probability of

attack (Berryman 1978; Mitchell & Preisler 1991; Powell et al. 1996; Powell et al. 2000;

Heavilin et al. 2005). However, the ability to reliably predict where and when outbreaks

will occur remains a problem.

In this report, we review the ecological and behavioral mechanisms that influence

the population dynamics and dispersal of the mountain pine beetle. We review the

statistical and mechanistic models that have been developed for the mountain pine beetle,

and assess the successes and failures of each. Through this contrast between empirical

studies and modeling efforts, we identify important voids in our current ecological

understanding, suggest areas where the application of statistical models could provide

greater insight, and propose an approach to model development that will hopefully better

2


predict when and where outbreaks may occur, as well as further our understanding of the

ecological processes that are germane to infestations.

MOUNTAIN PINE BEETLE ECOLOGY

General Life Cycle

Mountain pine beetles are typically univoltine, completing a single generation each year.

Recently developed adults emerge from their host trees in mid to late summer to search

for and attack new hosts (Figure 1). Attacking beetles bore through the outer bark of

potential hosts into the phloem tissue. Healthy trees can resist small numbers of attacks

by producing resin that slows down or stops beetles from constructing galleries in the

phloem tissue (Raffa & Berryman 1983). In response, attacking beetles emit aggregation

pheromones that can recruit additional beetles into a mass attack on the focal tree. If

sufficient beetles are available, resin defenses will be overwhelmed and beetles

successfully construct galleries (Raffa & Berryman 1983). To avoid intraspecific

competition, attack density is thought to be regulated by the production of anti-

aggregation pheromones and a reduction in the emission of aggregation pheromones

(Raffa & Berryman 1983; Borden et al. 1987), both of which signal the depletion of host

defenses (Geiszler et al. 1980). When a focal tree is saturated with anti-aggregation

pheromones, newly arriving beetles switch to nearby trees (Mitchell & Preisler 1991).

Once host defenses have been overwhelmed, beetles construct vertical galleries in

the phloem tissue to lay eggs. Mountain pine beetles infect the host with an aggressive

fungal symbiont that helps to resist host defenses. Males speed up gallery construction by

moving frass out of the way of the female into bottom of the gallery (Reid 1958). Males

die in the galleries soon after mating (Reid 1958), but females may re-emerge in fall for a

second flight and create a second gallery (Reid 1962). The recently laid eggs quickly

develop into larvae (Bentz et al. 1991). Larvae create feeding galleries at right angles to

the egg gallery and collectively girdle the tree. Beetle populations often over-winter as

third or fourth instar larvae, and resume development in the spring (figure 1). Pupation

occurs in early spring (Reid 1962) and teneral adults mature by feeding under the bark for

3-4 weeks (Raffa 1988) before emerging as fully developed adults in late July or August

3


(Amman 1973). The third and fourth instars, as well as the pupae, require higher

temperatures (>15ºC) for development than eggs, first or second instar, providing an

inherent mechanism for synchronizing larval instar molts and adult emergence. This is

because higher temperature thresholds in the third and fourth instars prevent development

to advanced life-stages less cold tolerant and allow late-hatching eggs to catch up (Bentz

et al. 1991). Coincident adult emergence is important for successful mass attacks.

To better understand these ecological processes, the mountain pine beetle life-cycle

can be divided into two phases: a within-tree reproductive phase and an among-tree

dispersal phase. The within-tree phase determines the density of newly emerging adults

that result from the previous years attacks. The among-tree phase governs how these

newly emerging adults will disperse and aggregate to attack new hosts.

Within-Tree Phase

The engraved galleries that remain on the host reveal a wealth of ecological information

about the mountain pine beetles. They provide a rare opportunity to directly observe the

rates of reproduction and mortality that determine population growth. Commonly

collected data include the number of successful attacks, the number of eggs laid, the rate

of development through the stages, the causes of mortality, and the number of new adults

that emerge the following season. Based on this information, it is known that the density

of emerging beetles is largely governed by attack density, host characteristics, and

temperature (Safranyik 1978; Raffa & Berryman 1983; Berryman et al. 1985; Raffa

2001).

Attack density

The density of attacking beetles has both a positive and negative effect on the density of

new adults that emerge the following year. The positive influence results from the need to

overwhelm host tree defenses. At low attack densities, a healthy tree can successfully

repel all attacks (Raffa & Berryman 1983; Mulock & Christiansen 1986). As the attack

density (attacks per area of bark) increases, the proportion of attacks that successfully

produce offspring also increases (Raffa & Berryman 1983; Elkin & Reid 2004). When

averaged over all trees in a stand, the shape of this relationship is a smooth monotonically

4


saturating function (Raffa & Berryman 1983; Elkin & Reid 2004); a pattern observed

across a number of aggressive bark-beetle species (Berryman 1974) (Figure 2a).

However, it is also known that the minimum density required to overwhelm a host tree

increases with the vigor of the host (Mulock & Christiansen 1986), which implies that

each host of a different vigor has a separate relationship between attack density and

proportion of successful attacks. As a result, the relationships reported for the average of

all trees in a stand represent the average of many curves reflecting both the underlying

process of host defense and the heterogeneity of host vigor within the stand.

A common field method used to study bark beetles is to set out fresh log bolts (cut

trees with ends sealed in wax) and observe the pattern of beetle attacks. Beetles respond

to these logs in a similar manner as live trees, but without having to contend with host

defenses. In the absence of tree defenses there is an almost linear decline in the number

of new adults produced per attack as attack density increases (Cole 1962; Berryman

1974; Raffa & Berryman 1983; Amman & Pasek 1986) (Figure 2b). This negative

density-dependence begins at even the lowest of possible attack densities (Raffa 2001);

indicating that, in the absence of tree defenses, the maximum per capita growth rates are

attained by a single attack. At high attack densities, the strength of the negative density-

dependence can be sufficient to reduce the per capita emergence rate to below one (Cole

1962; Raffa & Berryman 1983). This likely results from a combination of reduced egg

laying rates at higher gallery densities and increased mortality from greater intraspecific

competition (Cole 1962).

The combined effects of both positive and negative density-dependence on

reproductive success creates a unique tension for aggressive bark beetles. Any increase in

beetle density required to overcome host defenses translates into a direct reduction in the

per-capita reproductive success. There is good evidence to suggest that the positive and

negative density-dependent mechanisms operate independently of one another, such that

the number of new adults emerging from each attack can be described by the simple

product of per-capita reproductive success and proportion of successful attacks

(Berryman 1974; Raffa & Berryman 1983; Raffa 2001) (Figure 2b). The advantage from

using this perspective is that the influence of other factors, such as temperature and host

5


vigor, on overall reproductive success is readily understood by considering their

independent effects on each of the two density-dependent mechanisms.

Estimates of attack densities are usually based on the average over a tree (Raffa &

Berryman 1983). If host defenses have been completely overwhelmed, such that beetles

have fully colonized the tree, then the average density accurately reflects the local attack

densities within the tree. However, strip-attacks are common in infected stands, where

only localized spots on a tree have been successfully attacked (Raffa &Berryman 1983;

Mulock & Christiansen 1986; Bentz et al. 1996). These localized attacks successfully

produce new adult beetles, but do not kill the host tree (also referred to as strip-kills).

Averaging the attack densities over the entire bark area of a strip-attacked tree would

underestimate the true attack densities that succeeded in overwhelming host defenses.

The factors behind the development of strip-attacked trees versus fully-attacked trees

may be related to the local availability of beetles. The within-tree distribution of attacks

is clumped when the average attack density on the tree is low, but becomes random as the

average attack density on the tree increases (Raffa & Berryman 1983). The change in

distribution suggests that attacks are focused within local areas on a tree, and that fully-

attacked hosts may be the result of multiple strip-attacks. This perspective may be

important when considering the endemic state of a beetle population. If there are

insufficient beetles to attack an entire tree, the population could persist by concentrating

attacks on local spots. Since strip-attacked trees do not die, they will not be evident in

aerial surveys.

Host characteristics

Tree characteristics can have a large influence on the reproductive success of bark

beetles. Larger trees have a greater surface area of bark and thicker phloem. Thicker

phloem provides a greater volume of food that can be utilized by the growing beetle

larvae, which in turn supports greater egg densities and more emerging adults

independent of the initial attack density (Amman & Cole 1983; Amman & Pasek 1986).

As a result, larger trees give rise to more beetles emerging on both an absolute and per-

unit-area scale (Reid 1963). Thin-phloemed trees cannot support large densities of beetles

(Berryman 1976). Within a tree, phloem thickness decreases with increased height above

6


the ground, and determines the surface area of the bole that can be colonized by bark

beetles (Langor 1989). It has also been suggested that phloem thickness and surface area

may account for much of the differences in MPB reproductive success among host

species (Langor 1989).

The greatest challenge for aggressive bark beetles is to overcome host defenses and

gain access to the phloem habitat. Pine trees produce a range of chemical and physical

products that are thought to provide defense against bark beetles (Reid et al. 1967;

Berryman 1969, 1972; Dunn & Lorio 1992). However, the primary defense mechanism is

resin. Defenses are initiated as soon as burrowing begins because the attacking beetles cut

into special ducts, causing resin to flow out of the wound. If the flow is sufficient, it can

flush an attacker out of the entrance hole. If the attacking beetle is not flushed out,

flowing resin can plug the entrance hole and prevent the emission of aggregation

pheromones (Raffa & Berryman 1983). Hosts also initiate a secondary resin response by

producing oleoresin flow in the galleries under construction. If the host tree has sufficient

energy and resources, this secondary response can kill all of the freshly laid eggs (Reid &

Gates 1970; Christiansen et al. 1987). As discussed above, the successful defense against

attack depends on both the density of attacking beetles and host vigor.

Larger trees are less vigorous than smaller trees, when vigor is measured as the

stem growth per leaf area (Shrimpton & Thompson 1982; Larsson et al. 1983). A number

of studies have show that more vigorous trees are more resistant to attack, requiring a

greater density of beetles to overwhelm tree defenses (Waring & Pitman 1983; Mulock &

Christiansen 1986; Christiansen et al. 1987). From the perspective of the attacking beetle,

host vigor is perceived through resin flow (Raffa & Berryman 1983; Dunn & Lorio 1992;

Lorio et al. 1995). Trees with greater vigor have greater reserves of carbohydrates and

can produce greater amounts of resin (Christiansen et al. 1987). Support for the

relationship between the ability of a host to defend against attacking beetles and the

reserves of energy to produce resin is evident in the temporal patterns of resin flow. Resin

flow declines rapidly in hosts that become completely overwhelmed, whereas resin flow

remains high in hosts that successfully prevent attacks (Raffa & Berryman 1983; Lorio et

al. 1995).

7


Temperature

Temperature has a strong influence on both survival and development in the mountain

pine beetle. Winter temperatures cause significant mortality, ranging from 20% during

moderate winters to 100% during very cold winters (Amman 1973; Amman & Cole

1983; Amman 1984; Langor 1989). In fact, it has been argued that the northern expansion

of the mountain pine beetle range is prevented by temperature induced mortality at a

latitude where the average minimum temperatures is less than -40ºC (Safranyik 1978).

However, the ability to survive cold winters is not shared equally among the different

life-stages. Laval stages have the greatest cold tolerance, in particular the 3rd and 4th stage

(Amman 1973), which can survive temperatures of -34ºC (Safranyik 1978; Bentz &

Mullins 1999). In contrast, the egg, pupae and adult stages are particularly vulnerable to

temperature and do not appear to survive winters colder than ~ -18ºC (Reid 1962; Reid &

Gates 1970; Amman 1973; Safranyik 1978). The result is that winter mortality depends

not only on temperature, but also on the timing between the minimum temperatures and

the stage of beetle development.

Development rates in all insects are strongly tied to temperature. Provided that food

is sufficient, development through each stage of the mountain pine beetle can be reliably

predicted by the number of degree-days (Reid & Gates 1970; Bentz et al. 1991). The

implication is that lower than average temperatures will extend the length of time it takes

to complete one generation, and higher than average temperatures will reduce it. At the

extremes, temperature could determine whether a population is bivoltine, univoltine or

semivoltine (Logan & Powell 2001). Empirical support for this idea can be found across

elevations. Mountain pine beetle populations at high elevations often require two years to

complete a single generation, whereas populations at lower elevations require only one

(Amman 1973).

A more subtle influence of temperature on beetle development is the control over

which stage will be forced to over-winter. Cooler summer temperatures may not be

sufficient for eggs to develop into larvae before winter sets in, preventing the population

from surviving because of the low cold tolerance for eggs (Amman & Cole 1983). At

lower elevations, Amman (1973) found that populations developed into the 2nd or 3rd

8


instar larvae prior to the onset of winter. In contrast, at higher elevations where the

temperature is colder, development was much slower and most populations entered

winter as eggs. The result was significantly greater mortality at the higher elevations. In

addition, larvae that grew at warmer temperatures were able to develop longer galleries,

which means that they attained a larger mass before the winter set in than larvae at cooler

temperatures, which is thought to increase survival at colder temperatures (Amman

1973). The reverse may also be true. Warmer summer temperatures may cause the

population to develop through the larval stage into pupae and adults before winter sets in

(Reid 1962). Since these stages also have low cold tolerance, this could cause elevated

mortality rates.

Among Tree Phase

The attack phase of the mountain pine beetle is probably the most famous part of the life-

cycle—yet it is also the least understood. This phase is characterized by individual

dispersal from recently depleted hosts, followed by strong aggregation that can

overwhelm the defenses of even the most vigorous trees. Pheromones are generated by

beetles during the process of attacking a host. Aggregation is coordinated by the

chemotactic response of individuals to these pheromones. Reproductive success depends

significantly on the ability to recruit sufficient adults into mass-attacks. However, as

discussed above, too great of an attack density will also reduce reproductive success

because of larval competition within the phloem. From the perspective of the bark-

beetles, the optimal attack density is the minimum required to overwhelm host defenses.

The pheromone system appears to be well tuned to generate these optimal attack

densities.

Non-pheromone directed dispersal

There are two situations when beetles move without the influence of pheromones; when

they are insensitive to pheromones because of recent emergence and when they are the

‘pioneer’ beetles emerging into an environment without other attacking beetles. Newly

emerging adults require a period of flight exercise before they respond to pheromones

cues (Borden et al. 1986; Raffa et al. 1993). While the length of time for this flight period

9


is not well understood in natural stands, it may be sufficient to disperse greater than 50m

(Borden 1993; Carroll & Safranyik 2004).

The term pioneer beetles refers to the first wave of beetles that emerge in a flight

season. Pioneer beetles disperse into an environment without any guidance from

pheromone cues. Such movement appears to be a random search for suitable hosts

(Safrinyik et al. 1974; Burnell 1977; Turchin & Theony 1993), perhaps influenced by a

preference for larger trees (Cole & Amman 1969). Dispersal of these pioneer beetles can

be on the order of several hundred meters (Turchin & Theony 1993). Unfortunately, there

is little data to determine how much beetles are assessing potential hosts during this

dispersal phase, or how they respond to other physical factors such as wind direction and

slope. Once beetles have landed on a tree, they must decide whether or not to initiate

gallery construction. Beetles assess host quality by chewing into the bark (Raffa &

Berryman 1982a). If the host is not perceived to be suitable, then beetles continue flight.

The behavior of these pioneer beetles can have a large impact on the spatial dynamics of

an infestation because these beetles will establish the first pheromone plumes that will

focus aggregation.

Pheromone directed dispersal

As with all aggressive bark-beetles, mountain pine beetles show a strong chemotactic

response to pheromones. Two broad classes of pheromones have been identified; those

that aggregate and those that repulse (Purswaran et al. 2000; Raffa 2001). While a

number of the chemicals responsible for the aggregate and anti-aggregate behavior have

been identified (Purswaran et al. 2000), it is not possible to quantify the spatial or

temporal pattern of these chemicals in a natural stand. As a result, the proposed

mechanisms behind the pheromone system are based on the temporal patterns of beetle

attacks in relation to the host response.

The temporal sequence of attacks on a focal host is characterized by a rapid

increase in the number of beetles landing, followed by a rapid decrease, within four to

five days (Raffa & Berryman 1983; Bentz et al. 1996). We distinguish the two phases of

the sequence because it has been proposed that each are the result of different

mechanisms. The initial aggregation is the result of aggregation pheromones that are

10


derived from the host materials (reviewed in Raffa 2001). If subsequent attacks to the

host are experimentally prevented (Raffa & Berryman 1983), or the host is able to resist

attack (Lorio et al. 1995) then attraction to the host can go on for a very long time,

sometimes for the entire flight period. It has suggested that this is because pheromones

are specifically derived from resin. As long as the host can produce resin defenses,

beetles will have a substrate from which to produce pheromones (Raffa & Berryman

1983; Borden et al. 1987). The advantage to this perspective is that when a host has been

overwhelmed, resin flow stops and the aggregation pheromones are no longer produced

(Raffa 2001). This simple mechanism does not require any sophisticated behavior from

the beetles, but may successfully prevent the over-colonization of the host.

While some researchers believe that the end of the host attack is initiated by a

decline in aggregation pheromones (Raffa & Berryman 1983; Raffa 2001), others believe

that it is initiated by anti-aggregation pheromones (Bentz et al. 1996; Pureswaran et al.

2000; Borden et al. 2003). Anti-aggregation pheromones are produced by both male and

females, but it has been hypothesized that some of these are only produce by males in

response to crowded conditions (Pureswaran et al. 2000). In many ways, the debate

between the two alternative mechanisms may not be important because the mechanisms

have a similar effect. However, a real distinction has to do with the perception of

crowding. If anti-aggregate pheromones are responsible for terminating the attack, then

there is the possibility that local crowding may not be tightly coupled to the state of the

host, allowing for sub-optimal colonization strategies.

When a focal tree is well under attack, secondary attacks begin to appear on the

surrounding trees (Bentz et al. 1996). It was once thought that this was a result of spill-

over because the focal tree had been completely colonized (Geiszler & Gara 1978).

However, temporal observations have revealed that secondary attacks appear long before

the focal tree is fully colonized (Bentz et al. 1996), suggesting that secondary attacks are

pheromone driven rather than directly driven by attack densities on the focal tree. The

pattern of secondary host attacks surrounding the focal tree appears to be random

(Geiszler & Gara 1978).

11


Host interference

As discussed above, the primary defense response of hosts is to produce resin, which

inhibits gallery construction by the attacking beetles (Reid et al. 1967; Raffa & Berryman

1982b; Raffa & Berryman 1983). If beetles succeed in developing galleries, resin flow

can seal the entrance hole and prevent the emission of pheromones (Raffa & Berryman

1983). It has been proposed that the interference of pheromone production by resin may

be critical for outbreaks because it provides an opportunity for vigorous trees to avoid

aggregated attacks, even when sufficient beetles are available (Raffa & Berryman 1983;

Raffa 2001). While not discussed, this hypothesis also implies that hosts with high vigor

would likely only come under mass-attack if there were enough random landings, by

either pioneer or secondary attacks, to reduce resin flow sufficiently and allow

pheromone emission.

Long range dispersal

The dispersal mechanisms discussed above are the most common behavior in mountain

pine beetle infestations. However, a small proportion of the population may undergo long

range dispersal, primarily above the tree canopy. Beetles have been observed to travel

distances of up to 20-30kms during long distance dispersal (Safranyik 1978). It is not

known if such dispersal is the result of being carried on wind currents (Furniss & Furniss

1972) or from sustained flight (Safranyik 1978).

Competition with non-aggressive bark beetles

Secondary bark beetles, such as the pine engraver (Ips pini), are less successful at

attacking live hosts. Most often, these beetles colonize trees that are weak or have been

killed by aggressive species such as the mountain pine beetle. These secondary bark

beetles often respond to aggregation pheromones emitted by aggressive beetle species

(Raffa 2001), which provides the potential for strong interspecific competition. For

example, secondary bark beetles that colonized hosts attacked by spruce beetles caused

reduced brood densities through interspecific competition (Poland & Borden 1998).

However, field studies reveal that pine engravers colonize marginal areas on trees

attacked by mountain pine beetles, reducing interspecific competition (Safranyik et al.

12


1999). It is not clear whether interspecific competition has an important role in regulating

mountain pine beetle dynamics.

MODEL INSIGHTS

A number of models have been developed for mountain pine beetle infestations. This

effort has covered a wide range of approaches from simple strategic models (Heavilin et

al. 2005) to complex simulation models (Safranyik et al. 1999), from a focus on dispersal

(Preisler 1993) to a focus on population growth (Berryman 1979), and from statistical

models (Mitchell & Preisler 1991) to predictive models (Powell et al. 1996).

Our primary goal is to understand the factors that contribute to population outbreaks

in the mountain pine beetle, and thus increase our ability to predict where and when

outbreaks occur, as well as the resulting spatial-temporal dynamics. Thus, we focus our

review efforts on the modeling work that has attempted to understand either the spatial or

temporal patterns of beetle dynamics.

Statistical models

The most obvious effect of aggressive bark beetles is their impact on the host tree. As a

result, numerous studies have recorded the number of beetle-killed trees over space and

time. While some of these are at a spatial-scale of the individual host tree (hereafter

referred to as stand-level), the longest time-series that cover the widest areas are at a

spatial-scale of the forest (hereafter referred to as landscape-level) where the observations

are often simply the presence or absence of beetle-attacked trees somewhere in the stand.

Data on host mortality is only one component of the beetle-host interaction. However, it

can provide valuable insight into which factors have the greatest spatial and temporal

correlation with patterns of attacked trees.

At the stand-level, the detailed observations on attacking beetles suggest that attack

success depends on a complex set of processes involving behavioral aggregation and host

vigor. The qualitative prediction is that larger and older trees, which are less vigorous,

will be successfully attacked at lower densities than smaller and younger trees. While

there is ample qualitative supported for this in stand-level data (Waring & Pitman 1983;

Mulock & Christiansen 1986; Christiansen et al. 1987), quantitative statistical models of

13


spatial attack patterns can provide further insight (Mitchell & Preisler 1991; Preisler

1993; Preisler & Mitchell 1993). These analyses use auto-logistic regression methods to

model the probability of host attack as a function of host size, age, vigor, and proximity

to other attacked trees. They found that host size and vigor have a significant influence on

the probability of attack. Larger trees are attacked more often than smaller ones,

particularly when host vigor is low. The analyses have also found a significant influence

of proximity to attacked trees. Small trees were attacked more often if they were near

other attacked trees, suggesting that spatially explicit processes may be important for

understanding attack patterns. The results do not appear to be influenced by changing

beetle densities (Mitchell & Preisler 1991).

Spatial heterogeneity in factors such as average temperature, elevation or forest

management practices are larger at the landscape-level than at the stand-level, providing

an opportunity to study how such factors correlate with the probability of attack. While

much landscape-level data exists for the mountain pine beetle, statistical models of

landscape-level attack patterns have only been applied to the southern pine beetle

(Gumpertz et al. 2000; Zhu et al. 2004). These studies also used auto-logistic regression

methods to study the correlation between landscape-level patterns of environmental

factors and the spatial-temporal autocorrelation of attack patterns. The analyses found

that the probability of host attack by the southern pine beetle correlates with spatial

proximity to previous attacks, host factors such as size, and environmental factors such as

temperature.

Mechanistic models

Early models of the mountain pine beetle focused on the population dynamics without

explicitly considering space (Berryman 1978, 1979). The models incorporate such things

as host size, age and vigor, as well as beetle density (Raffa & Berryman 1986; Berryman

et al. 1989). Beetle dynamics are determined by the ability to successfully attack the

available hosts using the well established relationships for host resistance and beetle

reproduction. Beetles are assumed to only attack trees whose defenses can be

successfully overwhelmed by the current density. These models often include many

details, such as the influence of temperature, stage-specific mortality, and management

14


practices such as pheromone traps and forest thinning (Raffa & Berryman 1986;

Safranyik et al. 1999). The primary conclusion from these models is that the population

growth rate is bimodal as a function of beetle density (Figure 3). Population growth is

positive at low densities of beetles, but becomes negative at intermediate densities

because larval competition is intense and vigorous trees remain inaccessible until

population densities are higher (Raffa & Berryman 1986). If beetle densities can increase

above this point, then the population growth is again positive, reflecting the ability of the

population to attack vigorous trees. The conclusion from these models is that outbreaks

result from environmental factors (e.g. drought) that cause host vigor to decrease and

allow endemic densities of beetles to successfully attack the usually inaccessible high

quality hosts (Raffa & Berryman 1986).

The outbreak model proposed by Berryman (1979) (Figure 3) has some parallels

with other insect outbreak models. For example, the now classic spruce budworm

outbreak models can display the same three equilibria; stable endemic state, unstable

transition and stable outbreak state (Ludwig et al. 1978). However, the ecological

mechanisms behind these models are different. While resource limitation limits the

outbreak states of both models, the endemic state of the spruce budworm model is

maintained by predation whereas in the endemic state of the mountain pine beetle model

it is maintained by host vigor. A more subtle difference is the time scale for resource

regeneration. Trees regenerate at a much slower rate than they are killed during beetle

outbreaks. Numerical simulations that explicitly account for this (Raffa & Berryman

1986) reveal that there is no stable outbreak state. Rather, the beetle population erupts

rapidly depleting the available hosts, and then crashes to a level lower than the endemic

state. As the forest regenerates, the beetle population returns to the endemic state

abundance. Simulations such as this indicate that population growth is likely not well

described by beetle density alone.

Recent work has expanded the detailed models of Raffa & Berryman (1986) to

include spatially explicit dispersal (Powell et al. 1996; White & Powell 1997; Logan et

al. 1998; Powell et al. 2000). The relationships that govern the local beetle-host

interactions are very similar, but the pheromone system is explicitly modeled. Pheromone

synthesis is assumed to depend on resin capacity, as related to host vigor, and diffuses

15


through the stand (Powell et al. 1996). Beetle dispersal can have both random movement

as well as the strong chemotactic pheromone response. These models predict a similar

transition from endemic to epidemic beetle infestations, which the authors refer to as

phase transitions (White & Powell 1997). Furthermore, a similar relationship is found

between host vigor and beetle density, where the focal trees are generally weaker in the

endemic state than in outbreak states. The conclusion from these models is that outbreaks

can be caused by the chance spatial arrangement of weak trees, which leads to a locally

high density of emerging beetles and the initiation of large-scale infestations (White &

Powell 1997).

Most models of mountain pine beetles have focused on the emergence and

propagation of infestations through a forest on time scales that are fast enough to ignore

stand regeneration. However, since mountain pine beetles have been part of a natural

cycle of disturbance and regeneration with their hosts for centuries (Logan & Powell

2001), the ecological interactions at the stand scale must also allow for coexistence at the

landscape scale. Heavilin et al. (2005) developed a landscape-level model that provides

for forest regeneration as well as considering how infestations spread over spatial scales

larger than a stand. The tree component of the model is stage-structured to allow for

heterogeneity in host vigor, and beetle population dynamics are determined by the

number of attacked trees from the previous year. Beetle dispersal is continuous across

space, but does not allow for aggregation behavior. While such a model is useful for

understanding how outbreaks progress, it will likely fail to capture the dynamics of the

endemic population. Endemic populations may exist in strip-attacks, which do not kill the

host. Thus, at low densities, it is unlikely that the beetle population could be

approximated by the number of dead trees.

CONCLUSIONS

A great deal of effort has focused on understanding the basic ecology of mountain pine

beetles. Since reproductive effort, development and mortality are relatively easy to

observe, this effort has resulted in a unique depth of ecological understanding. For

example, much is known about how live hosts defend against attacks, how temperature

interacts both directly and indirectly with development and survival, what attack densities

16


are required to overwhelm hosts of different vigor, how intraspecific competition reduces

reproductive success, how pheromones are produced, and how chemotactic behavior

allows for aggregated mass-attacks. Yet, we have been largely unsuccessful at predicting

when and where infestation outbreaks will develop. In the paragraphs below, we discuss

areas where further empirical information and modeling efforts would be most beneficial.

While many parts of the mountain pine beetle life-cycle are well understood, there

are some key processes where little is known. Most of these have to do with dispersal.

The behavior of pioneer beetles is not well understood in terms of both direction and

dispersal distance. Since these beetles are the first to sample available hosts and initiate

attacks, the resulting spatial dynamics of an infestation may largely depend on how

beetles respond in an environment without pheromone cues. Furthermore, a finer

quantitative assessment of how beetles respond to pheromone cues in natural stands

would increase our understanding of aggregation behavior, in particular how it relates to

secondary attacks.

The second area where further data are important is the attack pattern at all levels of

infestations. For example, few data exist on the endemic state of the beetle population.

While some studies have recorded the occurrence of strip-attacks, there is little data to

assess how important these are for the persistence of endemic populations. An alternative

is that the endemic population only exists as spot infestations that appear in aerial surveys

as clumps of dead trees (Borden 1993). Since strip-attacks can only be detected by

targeted ground surveys, such data would help distinguish between the importance of the

local endemic population versus long distance dispersal for initiating outbreaks observed

by aerial surveys. Observations on the spatial distribution of attacks, both successful and

unsuccessful, would also provide important independent data for model validation.

Statistical analyses of the attack patterns within a stand reveal that proximity to

attacked trees has a large influence on the probability of being attacked. This provides

important insight into the patterns that emerge from the complex interaction between

beetle behavior and pheromone production. Similarly, spatial-temporal landscape-level

analyses, which has been conducted other aggressive bark-beetles, can identify the

environmental factors and host characteristics that have the strongest correlation with the

probability of infestation. Applying these landscape-level analyses to the mountain pine

17


beetle data would be invaluable to help guide model development. The statistical

methods discussed above all focus on the probability of host mortality (binary data),

ignoring the density of beetles. When sufficient data are available, a significant

improvement would be to develop these models for non-binary data such as beetle

density.

We feel that the inability to predict infestation outbreaks is largely because beetle

dynamics are an emergent property from a number of interacting and sometimes complex

processes. There is no doubt that mathematical models provide the only means to

understand how outbreaks emerge from these interacting processes, but such predictions

may be sensitive to how the models are developed and parameterized. Previous modeling

efforts suggest that a minimal model needs to explicitly model both the within-tree

reproductive phase and the among-tree dispersal phase. We believe that key to creating a

robust and meaningful model of mountain pine beetle infestations depends a careful

balance of strategic development guided by independent validation.

References Cited Amman, G.D. 1973. Population changes of the mountain pine beetle in relation to

elevation. Environmental Entomology 2: 541-547.

Amman, G.D. 1984. Mountain Pine-Beetle (Coleoptera, Scolytidae) Mortality in 3 Types of Infestations. Environmental Entomology 13: 184-191.

Amman, G.D. & Bartos, D.L. 1991. Mountain Pine-Beetle Offspring Characteristics Associated with Females Producing 1st and 2nd Broods, Male Presence, and Egg Gallery Length. Environmental Entomology 20: 1562-1567.

Amman, G.D. & Cole, W.E. 1983. Mountain pine beetle dynamics in lodgepole pine forests. Part II: Population dynamics. USDA Forest Service General Technical Report INT-145, 59pp.

Amman, G.D. & Pasek, J.E. 1986. Mountain Pine-Beetle in Ponderosa Pine - Effects of Phloem Thickness and Egg Gallery Density. USDA Forest Service Intermountain Research Station Research Paper INT-367, 7pp.

Bartos, D.L. & Amman, G.D. 1989. Microclimate: an alternative to tree vigor as a basis for mountain pine beetle infestations. USDA Forest Service Intermountain Research Station Research Paper INT-400.

18


Bentz, B. J., Amman, G.D. & Logan, J.A. 1993. A critical assessment of risk classification systems for the mountain pine beetle. Forest Ecology and Management 61: 349-366.

Bentz, B.J., Logan J.A. & Amman G.D. 1991. Temperature-Dependent Development of the Mountain Pine-Beetle (Coleoptera, Scolytidae) and Simulation of its Phenology. Canadian Entomologist 123: 1083-1094.

Bentz, B.J. & Mullins, D.E. 1999. Ecology of mountain pine beetle (Coleoptera : Scolytidae) cold hardening in the intermountain west. Environmental Entomology 28: 577-587.

Bentz, B.J., Powell, J.A. & Logan J.A. 1996. Localized spatial and temporal attack dynamics of the mountain pine beetle in lodgepole pine. USDA Forest Service Intermountain Research Station Research Paper INT-RP-494, 7pp.

Berryman, A.A. 1969. Response of Abies grandis to attack by Scolytus ventralis (Coleoptera: Scolytidae). Canadian Entomologist 98: 285-288.

Berryman, A.A. 1972. Resistance of conifers to invasion by bark beetle-fungal associations. BioScience 22: 598-602.

Berryman, A.A. 1974. Dynamics of bark beetle populations- Towards a general productivity model. Environmental Entomology 3: 579-585

Berryman, A.A. 1976. Theoretical Explanation of Mountain Pine Beetle Dynamics in Lodgepole Pine Forests. Environmental Entomology 5: 1225-1233.

Berryman, A.A. 1978. Towards a theory of insect epidemiology. Research on Population Ecology 19: 181-196

Berryman, A.A. 1979. Dynamics of bark beetle populations: analysis of dispersal and redistribution. Bulletin de la Societe Entomologique Suisse 52: 227-234.

Berryman, A.A., Dennis, B., Raffa, K.F. & Stenseth, N.C. 1985. Evolution of optimal group attack, with particular reference to bark-beetles (Coleoptera, Scolytidae). Ecology 66: 898-903.

Berryman, A.A., Raffa, K.F., Millstein, J.A. & Stenseth, N.C. 1989. Interaction dynamics of bark beetle aggregation and conifer defense rates. Oikos 56: 256-263

Borden, J.H. 1993. Uncertain Fate of Spot Infestations of the Mountain Pine-Beetle, Dendroctonus-Ponderosae Hopkins. Canadian Entomologist 125: 167-169.

Borden, J.H., Chong, L.J., Earle, T.J. & Huber, D.P.W. 2003. Protection of lodgepole pine from attack by the mountain pine beetle, Dendroctonus ponderosae (Coleoptera: Scolytidae) using high doses of verbenone in combination with nonhost bark volatiles. Forestry Chronicle 79: 685-691.

19


Borden, J.H., Chong, L.J. & Lacey, T.E. 1986. Pre-Logging Baiting with Semiochemicals for the Mountain Pine-Beetle, Dendroctonus-Ponderosae, Coleoptera, Scolytidae, in High Hazard Stands of Lodgepole Pine. Forestry Chronicle 62: 20-23.

Borden, J.H., Ryker, L.C., Chong, L.J., Pierce, H.D., Johnston, B.D. & Oehlschlager A.C. 1987. Response of the mountain pine-beetle, Dendroctonus pondersosae Hopkins (Coleoptera-Scolytidae), to 5 semiochemicals in British-Columbia lodgepole pine forests. Canadian Journal of Forest Research 17: 118-128

Burnell, D.G. 1977. Dispersal-Aggregation Model for Mountain Pine Beetle in Lodgepole Pine Stands. Researches on Population Ecology 19:99-106.

Carroll, A.L. & Safranyik, L. 2004. The bionomics of the mountain pine beetle in lodgepole pine forests: establishing a context. Pages 19-30 in T.L. Shore, J.E. Brooks, and J.E. Stone, editors. Mountain pine beetle Symposium: Challenges and Solutions. October 30-31 2003, Kelowna, British Columbia, Canada.

Christiansen, E., Waring, R.H. & Berryman, A.A. 1987. Resistance of conifers to bark beetle attack: searching for general relationships. Forest Ecology and Management 22: 89-106.

Cole, W.E. 1962. The effects of intraspecific competition within mountain pine beetle broods under laboratory conditions. USDA Forest Service Intermountain Research Station Research Paper INT-97.

Cole, W.E. & Amman, G.D. 1969. Mountain pine beetle infestations in relation to lodgepole pine diameters. USDA Forest Service Intermountain Research Station Research Paper INT-95.

Dunn, J.P. & Lorio, P.L. 1992. Effects of bark girdling on carbohydrate supply and resistance of loblolly pine to southern pine beetle (Dendroctonus frontalis Zimm.) attack. Forest Ecology and Management 50: 317-330.

Ebata, T. 2004. Current status of mountain pine beetle in British Columbia. Pages 52-56 in T.L. Shore, J.E. Brooks, and J.E. Stone, editors. Mountain pine beetle Symposium: Challenges and Solutions. October 30-31 2003, Kelowna, British Columbia, Canada.

Elkin, C.M. & Reid, M.L. 2004. Attack and reproductive success of mountain pine beetles (Coleoptera : Scolytidae) in fire-damaged lodgepole pines. Environmental Entomology 33:1070-1080.

Furniss, M.M. & Furniss, R.L. 1972. Scolytids (Coleoptera) on snowfields above timberline in Oregon and Washington. Canadian Entomologist 104: 1471-1477.

20


Geiszler, D.R., Gallucci, V.F. & Gara, R.I. 1980. Modeling the dynamics of mountain pine beetle aggregation in a lodgepole pine stand. Oecoligia 46: 244-253.

Geiszler, D.R. & Gara, R.I. 1978. Mountain pine beetle attack dynamics on lodgepole pine. in Theory and practice of mountain pine beetle management in lodgepole pine forests (D.M. Baumgartner, ed.), pp. 182-187. Pullman: Wash. State Univ. Coop. Ext. Serv.

Gumpertz, M.L.,Wu, C.T. & Pye, J.M. 1999. Logistic Regression for Southern Pine Beetle Outbreaks with Spatial and Temporal Autocorrelation. Forest Science 46:95–107.

Heavilin, J., Powell, J. & Logan, J. 2005. Modeling disturbance and recovery of lodgepole forest due to mountain pine beetle outbreaks on landscape scales. in: Plant Disturbance Ecology: The Process and the Response, AP/Elsevier

Langor D. W. 1989. Host Effects on the Phenology, Development, and Mortality of Field Populations of the Mountain Pine-Beetle, Dendroctonus-Ponderosae Hopkins (Coleoptera, Scolytidae). Canadian Entomologist 121:149-157.

Larsson, S., Oren, R., Waring, R.H. & Barrett, J.W. 1983. Attacks of mountain pine beetle as related to host vigor of ponderosa pine. Forest Science 29: 395-402.

Logan, J.A. & Bentz, B.J. 1999. Model analysis of mountain pine beetle (Coleoptera: Scolytidae) seasonality. Environmental Entomology 28: 924-934.

Logan, J.A., & Powell, J.A. 2001. Ghost forests, global warming, and the mountain pine beetle. American Entomologist 47: 160-173.

Logan, J.A., Regniere, J. & Powell, J.A. 2003. Assessing the impacts of global warming on forest pest dynamics. Frontiers in Ecology and the Environment 1: 130-137.

Logan, J.A., White, P., Bentz, B.J. & Powell, J.A. 1998. Model analysis of spatial patterns in mountain pine beetle outbreaks. Theoretical Population Biology 53: 235-255.

Lorio P.L., Stephen, F.M., & Paine, T.D. 1995. Environment and ontogeny modify loblolly pine response to induced acute water deficits and bark beetle attack. Forest Ecology and Management 73: 97-110

Ludwig, D., Jones, D.D. & Holling, C.S. 1978. Qualitative analysis of insect outbreak systems: the spruce budworm and forest. Journal of Animal Ecology 47: 315-332.

Mitchell R.G. & Preisler, H.K. 1991. Analysis of spatial patterns of lodgepole pine attacked by outbreak populations of the mountain pine-beetle. Forest Science 37: 1390-1408.

21


Mulock, P. & Christiansen, E. 1986. The threshold of successful attack by Ips typographus on Picea abies: a field experiment. Forest Ecology and Management. 14: 125-132.

Ono, H. 2003. The mountain pine beetle: scope of the problem and key issues in Alberta. Pages 62-66 in T.L. Shore, J.E. Brooks, and J.E. Stone, editors. Mountain pine beetle Symposium: Challenges and Solutions. October 30-31 2003, Kelowna, British Columbia, Canada.

Poland, T.M. & Borden J.H. 1998. Semiochemical-induced competition between Dendroctonus rufipennis and two secondary species, Ips tridens and Dryocoetes affaber (Coleoptera : Scolytidae). Journal of Economic Entomology. 91: 1142-1149.

Powell, J.A., Kennedy, B., White, P., Bentz, B.J., Logan, J.A. & Roberts, D. 2000. Mathematical elements of attack risk analysis for mountain pine beetles. Journal of Theoretical Biology 204: 601-620.

Powell, J.A., Logan, J.A. & Bentz, B.J. 1996. Local projections for a global model of mountain pine beetle attacks. Journal of theoretical biology 179: 243-260.

Preisler, H.K. 1993. Modelling spatial patterns of trees attacked by bark-beetles. Applied Statistics 42: 501-514.

Preisler, H.K. & Mitchell, R.G. 1993. Colonization patterns of the mountain pine beetle in thinned and unthinned lodgepole pine stands. Forest Science 39: 528-545.

Pureswaran, D.S., Gries, R., Borden, J.H. & Pierce, H.D. 2000. Dynamics of pheromone production and communication in the mountain pine beetle, Dendroctonus ponderosae Hopkins, and the pine engraver, Ips pini (Sag) (Coleoptera : Scolytidae). Chemoecology 10: 153-168.

Raffa, K.F. 1988. Host orientation behavior of Dendroctonus ponderosae: Integration of token stimuli and host defenses. pp 369-390 in W.J. Mattson, J. Leveux, C. Bernard-Dagan, editors. Mechanisms of Woody Plant Resistance to Insects and Pathogens. Springer-Verlag, New York.

Raffa, K.F. 2001. Mixed messages across multiple trophic levels: the ecology of bark beetle chemical communication systems. Chemoecology 11: 49-65

Raffa, K.F. & Berryman, A.A. 1982a. Gustatory cues in the orientation of Dendroctonus ponderodase (Coleoptera: Scolytidae) to host trees. Canadian Entomologist114: 97-104.

Raffa, K.F. & Berryman, A.A. 1982b. Physiological Differences between Lodgepole Pines Resistant and Susceptible to the Mountain Pine-Beetle (Coleoptera,

22


Scolytidae) and Associated Microorganisms. Environmental Entomology 11: 486-492.

Raffa, K.F. & Berryman, A.A. 1983. Physiological-Aspects of Lodgepole Pine Wound Responses to a Fungal Symbiont of the Mountain Pine-Beetle, Dendroctonus-Ponderosae (Coleoptera, Scolytidae). Canadian Entomologist 115: 723-734.

Raffa, K.F., Berryman, A.A. 1986. A Mechanistic Computer-Model of Mountain Pine-Beetle Populations Interacting with Lodgepole Pine Stands and its Implications for Forest Managers. Forest Science 32: 789-805.

Raffa, K.F., Phillips, T.W. & Salom, S.M. 1993. Strategies and mechanisms of host colonization by bark beetles. in Beetle-pathogen interactions in conifer forests. Academic Press. pp. 103-128.

Reid, R.W. 1958. The behavior of the mountain pine beetle, Dendrochtonus monticolae Hopk., during mating, egg laying, and gallery construction. Canadian Entomologist 90: 505-509.

Reid, R.W. 1962. Biology of the mountain pine beetle, Dendroctonus monticolae, in the East Kootenay region of British Columbia I. Life cycle, brood development, and flight periods. Canadian Entomologist 94: 531-538.

Reid, R.W. 1963. Biology of the mountain pine beetle, Dendroctonus monticolae, in the East Kootenay region of British Columbia III. Interaction between the beetle and its host, with emphasis on brood mortality and survival. Canadian Entomologist 95: 225-238.

Reid, R.W. & Gates, H. 1970. Effect of Temperature and Resin on Hatch of Eggs of Mountain Pine Beetle (Dendroctonus ponderosae). Canadian Entomologist 102: 617-622.

Reid, R.W., Whitney, H.S. & Watson, J.A. 1967. Reactions of lodgepole pine to attack by Dendrochtonus pondersosae Hopkins and blue stain fungi. Canadian Journal of Botany 45:1115

Ryker, L.C. & Rudinsky, J.A. 1976. Sound production in Scolytidae- aggressive and mating behavior of mountain pine beetle. Annals of the Entomological Society of America 69: 677-680.

Safranyik, L. 1978. Effects of climate and weather on mountain pine beetle populations. in Theory and Practise of Moutnain Pine Beetle Management in Lodgepole Pine Forests (Berryman, A.A., Amman, G.D., Stark, R.W. eds.) pp. 77-89, Symposium Proceedings: Forest, Wildlife and Range Experiment Station, University of Idaho, Moscow.

23


Safranyik, L., Barclay, H., Thomson, A. & Riel, W.G. 1999. A population dynamics model for the mountain pine beetle Dendroctonus ponderosae Hopk. (Coleoptera: Scolytidae). Canadian Forest Service, Pacific Forestry Centre, Information Report BC-X-386.

Safranyik, L., Shrimpton, D.M. & Whitney, H.S. 1974. Management of lodgepole pine to reduce losses from the mountain pine beetle. Environment Canada Forest Service Technical Report 1. Victoria, British Columbia. 24pp

Shrimpton, D.M. & Thomson, A.J. 1983. Growth-characteristics of lodgepole pine associated with the start of mountain pine-beetle outbreaks. Canadian Journal of Forest Research 13: 137-144.

Turchin, P. & Thoeny, W.T. 1993. Quantifying dispersal of southern pine beetles with mark recapture experiments and a diffusion model. Ecological Applications 3: 187-198

Waring, R.H. & Pitman, G.B. 1983. Physiological stress in lodgepole pine as a precursor for mountain pine beetle attack. Journal of Applied Entomology 96: 265-270.

White, P. & Powell, J.A. 1997. Phase transition from environmental to dynamic determinism in mountain pine beetle attack. Bulletin of mathematical biology 59:609-643.

Zhu, J., Huang, H.C. & Wu, P. 2005 Modeling spatial-temporal binary data using Markov Random Fields. Journal of Agricultural, Biological and Environmental Statistics (in press).

24


Janu

ary

February March

April

May

June

July

AugustSeptember

Octobe

rN

ovem

ber

Dec

embe

rLife-cycle of theMountain Pine

Beetle

eggslarvae Ilarvae II

larvae III

larvae IV

Adult

Pupae

period of emergence, flight and attack

Janu

ary

February March

April

May

June

July

AugustSeptember

Octobe

rN

ovem

ber

Dec

embe

rLife-cycle of theMountain Pine

Beetle

eggslarvae Ilarvae II

larvae III

larvae IV

Adult

Pupae

period of emergence, flight and attack

Figure 1. Mountain pine beetle life-cycle. Much of the life-cycle is spent in the phloem habitat, with the exception of when adults emerge and disperse to attack new hosts (hatched area). Eggs are laid in freshly excavated galleries and develop through four larval stages that feed on the phloem tissue. Beetles often over-winter in the larger larval stages, and pupate in late spring. Adults further develop in the phloem tissue and emerge during the brief flight period.

25


26

Attack Density (attacks/m2)

New

adu

lts p

er a

ttack Intraspecific Competition

Intraspecific Competition & Host Resistence

Suc

cess

ful a

ttack

s (%

) a

b

Figure 2. The effect of attack density on intraspecific competition and overcoming host defenses (from Raffa & Berryman 1983). a) the proportion of successful attacks increases with attack density, reflecting the ability of higher beetle densities to overwhelm host defenses. b) the decline in per capita reproduction begins at very low attack densities (dashed line) when host defenses are artificially prevented. The natural distribution of per capita reproduction as a function of attack density reflects the combined effects of intraspecific competition and host resistance (solid line).


loge(Nt)

-8 -6 -4 -2 0 2

log e(

Nt+

1/Nt)

-2

-1

0

1

2

N1 N2 N3

Figure 3. Population growth rate as a function of population density (from Berryman 1979). It is postulated that the population dynamics of the mountain pine beetle has three equilibria. The first is stable (N1), representing the endemic population state, the second (N2) is unstable, and the third (N3) is stable representing the outbreak state. Growth at the endemic state is controlled by host vigor, and growth at the outbreak state is controlled by resource limitation.

27

Documents

Mountain Pine Beetle - ualberta.campb/MPB Literature Review April 2005.pdf · MPB Review April 28, 2005 Mountain Pine Beetle Population Dynamics: Insights from empirical evidence