Open Collections

UBC Undergraduate Research

Determinants of tree susceptibility to attack by the red alder bark beetle, Alniphagus aspericollis (LeConte)… Takaro, Tristan Apr 22, 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52966-Takaro_Tristan_FRST_498_Graduating_Thesis_2013.pdf [ 339kB ]
JSON: 52966-1.0075630.json
JSON-LD: 52966-1.0075630-ld.json
RDF/XML (Pretty): 52966-1.0075630-rdf.xml
RDF/JSON: 52966-1.0075630-rdf.json
Turtle: 52966-1.0075630-turtle.txt
N-Triples: 52966-1.0075630-rdf-ntriples.txt
Original Record: 52966-1.0075630-source.json
Full Text

Full Text

Determinants of tree susceptibility to attack by the red alder bark beetle, Alniphagus aspericollis (LeConte) (Coleoptera: Scolytidae)     Thesis by: Tristan Takaro  Under the supervision of: Dr. Allan Carroll and Dr. Peter Marshall    In partial fulfillment for the requirements of the degree of  Bachelor of Science in Forest Sciences                           The University of British Columbia Faculty of Forestry Department of Forest Sciences  2013  (Submitted April 22nd, 2013) Table of contents - Introduction ??????????..............???? Page 3 - Materials and Methods ???????..............??... Page 9 o Study area ?????????..............??... Page 9 o Plot selection and description ..................??..... Page 9  o Bark beetle attack census ?................................. Page 10 o Statistical analysis ????..............??..?? Page 11 - Results ????????..............???????? Page 12 o Study area ???????????????. Page 12 o Bark beetle attack census ?????????. Page 13 - Discussion ???????..............???????.. Page 16   Index of tables and figures - Table 1 ?.......................................................................... Page 11 - Table 2 ???????............................?????.. Page 14 - Table 3 ?????..............?..............??..??..?.. Page 14 - Table 4 ???..............????................???..?.. Page 14 - Figure 1 ???..............??..?..............????..?. Page 10 - Figure 2 ???....................??..............????..?. Page 15 - Figure 3 ??....................??..............?????..?. Page 15 - Figure 4 ???????????????????. Page 16 Abstract  The immeasurable effect that bark beetles can have on forest ecosystems provides justification for the close study of bark beetle populations.  Bark beetles that attack hardwoods have not been as extensively studied as those that attack coniferous species, and the red alder bark beetle in particular has received very little attention.  Alniphagus aspericollis is a bivoltine ?secondary? bark beetle that specializes on red alder.  In this observational study in Vancouver, British Columbia several tree characteristics are assessed for their relationship with the likelihood of the tree being attacked by the bark beetle using logistic regression analysis.  It was found that diameter at breast height (DBH) did not significantly affect the likelihood of attack, and that the level of injury of a tree is a good predictor of its likelihood of being attacked by the bark beetle.   Keywords: bark beetle, secondary, Vancouver, British Columbia, injury, injuries, likelihood, diameter, DBH.  Introduction Bark beetles are widely recognized as one of the most prominent and pervasive disturbance agents of the temperate forests of the world (Avtzis et al. 2012).  They have been particularly destructive in North America, which is the native home to a majority of the tree-killing species of bark beetles (Stark 1995).  They can devastate entire landscapes of trees during population outbreaks, causing drastic ecological and economic ramifications (Simard et al. 2012).  Even when the trees are not killed, as commonly occurs with low population densities of several bark beetle species, they are commonly weakened by the attacks of the beetle which predisposes them to death by windthrow or pathogens.  The immeasurable effect that bark beetles can have on forest ecosystems provides justification for the close study of bark beetle populations.  The abundance of suitable host material (in terms of stand basal area and the abundance of large-diameter trees) and the stress level of the stand have been cited as key factors affecting the susceptibility of stands to bark beetle infestations (Simard et al. 2012).  Among other factors, the density of attacks on an individual susceptible tree is related to the distance from that tree to a tree with emerging beetles (Cates and Alexander 1982).  For the mountain pine beetle (Dendroctonus ponderosae Hopkins) tree mortality can be a result of stress associated with mechanical damage from construction, logging and vandalism among other stress-linked factors (Haverty et al. 1998).  The plant stress hypothesis and the environmental constraint hypothesis both predict that changes in plant defenses of stressed plants cause decreased resistance of those plants to insect attack (Bleiker et al. 2005).  During population outbreaks bark beetles in general are known to preferentially attack larger-diameter trees (Simard et al. 2012; Cates and Alexander 1982), but some species behave differently. The majority of bark beetles are known to preferentially attack weakened trees but the opposite can also occur in the case of population outbreaks of ?primary? bark beetles (Avtzis et al. 2012).  Bark beetles are considered to be ?primary? if they preferentially attack and kill vigorous trees (Marchetti et al. 2011).  A bark beetle is considered ?secondary? if it is most often observed to reproduce in trees that are weakened or recently killed or in slash (Teen 2012).  High stand and individual tree susceptibility to bark beetle attack has been found to be closely linked with defoliation of the stand or tree, and this pattern has been observed in several bark beetle-conifer systems (Bowers et al. 1996).  In a 2009 study on a secondary bark beetle known as Pseudips mexicanus, every attacked tree was observed to have some amount of scarring, a broken top, or a dwarf mistletoe infection (Smith et al. 2009).  Bark beetle species that infest coniferous trees have been more heavily studied than those that affect hardwood species, possibly due to the fact that the majority of phloemophagous bark beetle species are exclusive to conifers, as are the majority of tree-killing species (Ohmart 1989).  This could be attributed to the qualitative complexity of induced responses of angiosperms to inoculation by fungal associates of bark beetles and the effectiveness of this response, or to the greater difference between costs and nutritive benefits of attacking hardwoods than conifers (Ohmart 1989).       One hardwood-attacking bark beetle, the banded elm bark beetle (Scolytus schevyrewi Semenov), was found to have similar attack success as measured by the density of exit holes in 10-cm and 24-cm (diameter at breast height, DBH) bolts of Siberian elm, Ulmus pumila L. (Lee et al. 2011).  S. schevyrewi is native to Asia, was first observed in North America in 2003, and has since expanded to occur in 28 US states and four Canadian provinces (Lee et al. 2011).   A native elm bark beetle, Hylurgopinus rufipes Eichhoff, was found to occur in very low densities on trees with < 10 cm DBH (Anderson and Holliday 2003).  H. rufipes is one of the most destructive hardwood bark beetles because it is a vector for Dutch elm disease (Anderson and Holliday 2003), and S. schevyrewi may be a vector for it as well (Lee et al. 2011).  Interestingly Anderson and Holliday also found that the densities of entrance holes was very low in trees with DBH < 10 cm but densities plateau at approximately 3 holes per 100 cm2 (300 per m2) once DBH reaches about 20 cm (Anderson and Holliday 2003).  They also found that overwintering beetles were almost always found in the lower 25 cm of the bole, likely due to the thicker bark there providing more protection from the weather (Anderson and Holliday 2003).  Yet another elm bark beetle, Scolytus kashmirensis, has been observed to prefer trees that are older and taller for feeding (Buhroo 2012).  In a study on the dieback of ash in Slovakia bark beetles of the Hylesinus genus (fraxini and crenatus species) were found to preferentially attack ash trees that were weakened by a pathogen that causes patches of bark necrosis (Kunca et al. 2011).   Red alder is a common, short-lived hardwood tree in the Pacific Northwest.  While individual trees can live for over 100 years, stands of red alder typically ?break up? at around 50-70 years of age depending on site productivity (Deal 2006).  In the Pacific Northwest mortality of red alder is said to increase rapidly in stands over 90 years old, and very few red alder trees remain once the stand has reached 130 years (Deal 2006).  Red alder plays an important role as a both a nitrogen-fixer and a pioneer species that is good at colonizing new sites with exposed mineral soil.  The species has traditionally been viewed as having low economic value relative to the species which eventually replace it with succession (Douglas-fir, Western hemlock, Sitka spruce, etc.) in the Pacific Northwest and it was widely considered to be a ?weedy? species (Deal 2006).  In a general technical report by the United States Department of Agriculture (USDA) published in 2006 prices of red alder sawlogs were said to equal or exceed those of Douglas-fir sawlogs (Deal 2006).  While red alder lumber hasn?t been valued very highly in the past, demand for it has continually increased in the recent past (Deal 2006), providing economic incentive to understand more about the insects that infest it.   Much of the study of the insects that infest red alder (Alnus rubra Bong.) has focused on defoliators while bark beetles have not received as much attention (Deal 2006; Niemiec et al. 1995; Markham and Chanway 1998; Myers and Williams 1987).  A species of ambrosia beetle described as new in 1984 and named Gnathotrichus alniphagus was found on Alnus firmifolia in southwest Morelos, Mexico (Wood 1984).  The alder bark beetle (Alniphagus aspericollis LeConte) is a specialist on red alder and has been found to occur throughout the range of its host (Borden 1969).  The species was originally described under the genus Hylesinus by LeConte in 1876 (Chamberlin 1958), and was later changed to Alniphagus by James Malcolm Swaine (Chamberlin 1939).  A. aspericollis has also been found on Alnus incana, and Alnus rhomnifolia as well as Betula occidentalis in Idaho (Furniss and Johnson 2002).  There are two other identified species in the genus, Alniphagus hirsutus which is native to western North America and infests Alnus viridis subsp. sinuata (Sitka alder) and Alnus incana subsp. tenuifolia (Mountain alder) and Alniphagus alni which is native to eastern Asia (Schedl 1949).  A previously classified fourth species in the genus which was named A. africanus was transferred to the Hylesinopsis genus in 1987 (Wood 1987).  A. aspericollis is bivoltine on Burnaby Mountain, British Columbia (Borden 1969), and is therefore likely bivoltine in the city of Vancouver, B.C. as well, with one attack period occurring in late spring and a second in late summer.  The species is known to attack weakened trees, which is typical of secondary bark beetles, sometimes resulting in the death of the tree (Borden 1969).    Female A. aspericollis beetles initiate attack and begin the excavation of entrance tunnels while they await the arrival of a male beetle (Borden 1969). Chamberlin (1958) reported that entrance holes were often found at the base of branches, while Borden (1969) did not observe this.  Beetle attack density can be quite variable, ranging from about 45 to 375 entrance holes per square meter with a mean of 229.7 (Borden 1969).  Special tunnels for hibernation which consist of an irregular chamber with several short chambers branching off of it were observed by Chamberlin, and a photograph of one of these chambers indicates that the angle of entrance into the bark may be quite shallow rather than perpendicular to the bark surface (as is the case for a typical attack entrance hole) (Chamberlin 1958).  Red alder trees in Burnaby Mountain Park, in Burnaby, B.C. that were partially uprooted, likely during a windstorm a few years previous, were found to have attacks and live beetles of both Alniphagus aspericollis and Gnathotrichus retusus, the latter being an ambrosia beetle known as the western pinewood stainer (K?hnholz et al. 2000).  The trees were not observed to have any root diseases (K?hnholz et al. 2000).  Weakened and recently killed trees are the usual targets of attack by the beetle (Chamberlin 1958). Stand and individual tree susceptibility to bark beetle attacks are said to be driven by a combination of environmental and genetic factors, and physical factors including tree characteristics are deemed important drivers at least in conifer bark beetle systems (Bowers et al. 1996).  Borden (1969) noted that areas where scars or cracks exposed the inner bark were often heavily attacked.  Both young trees and old, healthy trees can be killed by the attacks of the beetle if beetle densities are high enough (Chamberlin 1958), but it is difficult to attribute the cause of the tree?s death to the beetle because the state of the tree prior to attack is often unknown (Borden 1969).  The death of patches of bark caused by the attacks of A. aspericollis can increase the likelihood of stem breakage from windthrow or snow accumulation through the rot that occurs in patch killed areas (Borden 1969).  In this study I determined which tree characteristics of red alder are correlated with the increased likelihood of that tree being attacked by the bark beetle.  Various injuries that the trees had as well as the DBH and the height of the tree were recorded.  I predict that trees with more injuries will be more likely to be attacked by A. aspericollis than those with fewer injuries, due to the increased stress associated with injured trees and the correlation between tree stress and attack susceptibility.    Materials and Methods  Study area  The study was conducted in a red alder stand in Pacific Spirit Regional Park on the endowment lands of the University of British Columbia in Vancouver, B.C., Canada.    Alnus rubra was the dominant tree species in the study area, with black cottonwood (Populus trichocarpa) also found in the overstory but in very low abundances.  Other tree species in the study area include western redcedar (Thuja plicata), western hemlock (Tsuga heterophylla), English holly (Ilex aquilifolium), and grand fir (Abies grandis) and were all in the sapling stage except for holly which commonly reached the mid-canopy.  The understory was dominated largely by red elderberry (Sambucus racemosa), salmonberry (Rubus parviflorus), and trailing blackberry (Rubus ursinus).  The elevation of the study area was approximately 80 meters above sea level.   Plot selection and description Three plots were laid out in areas that differed slightly in their understory plant community composition or in the density of red alder stems, in an attempt to capture the variation present in the red alder stand as a whole.  These differences were determined through a combination of walking through the stands and viewing the stands in Google Earth ?.  Two of the three plots were approximately 30 m * 30 m, while the third was slightly larger (plot 1).  This size was selected to ensure that each plot contained approximately 10 attacked trees, based on preliminary surveys of the study area.    Fig. 1. Map of study area displaying approximate locations of the three plots in the red alder stand on the University Endowment Lands in Vancouver, BC. Bark beetle attack census Data collection began in November 9th, 2012 and was completed on March 4th, 2013.  Each tree was assigned a number and that number was either written in permanent marker on the bark of the tree in the case of plot 1, or written on a 3-inch long strip of flagging tape which was placed at the base of the tree covered in leaves in the cases of plots 2 and 3, to increase efficiency of re-visiting specific trees.  A total of 200 trees were sampled across all plots.  The diameter at breast height (DBH) of each tree was measured using a Richter ? 5 m fiberglass diameter tape.  The height of each tree was calculated using a Suunto ? PM-5 / 1520 clinometer and an Eslon ? 30-meter fiberglass measuring tape.  A column in the data sheet titled ?Injury codes? provided a space to write down a suite of numbers corresponding to different tree characteristics (Table 1).  Additional notes on trees were recorded in the field notebook.  At each plot every red alder stem was sampled and assigned a number, including snags.  A Suunto ? increment borer was used to core 3 living trees in each plot to determine the average age of the whole field site.  The cores were taken in March of 2013 after the rest of the data collection had been completed.  These trees were randomly selected using a random integer generator found at  Cores were taken at 1.3 m on the bole and were stored in straws labeled with the year of collection and tree number.  A starter was used with the increment borer to ensure that the borer entered the tree without any wobble.  The starter consisted of a 5 cm-long metal peg attached to two pieces of 2 cm-thick composite board.  The ages of the trees indicated by the cores were averaged to determine the average stand age for the entire field site.  Table 1: Red alder tree characteristics included in the injury coding system Tree characteristics Broken top Large bark fissures Heavy scarring Thin crown Leaning Suppressed Cat-face presence / absence Fungus present Top kill unbroken Forked top   Statistical Analysis The PROC LOGISTIC function was used to create a model that would predict whether or not a tree is attacked based on the tree characteristics and variables that I measured.  The statistical package that was used for the statistical analysis was SAS 9.3 ? (SAS Institute 1999).  The logistic function is well-suited to this type of analysis because the dependent variable, attack status, is binary (attacked = 0, not attacked = 1).    Three different models were assessed using logistic regression analysis.  An injury scoring system named ?number of injuries? was created based on the number of injury codes that each tree had, and this system was used as a predictor is the first logistic model.  The first model was run to determine if the likelihood of the trees being attacked was affected by their DBH, the number of injuries that they had, or the plot in which they were growing.  The second model was run to determine if the height of trees had an effect on their likelihood of being attacked.  Approximately 30 trees of the 200 sampled had a cat-face or similar deep scarring that exposed the inner bark.   Because Borden (1969) noted that attacks were often focused on and around injuries in which the inner bark was exposed, an index called ?cat-face score? was created based on the size of the cat-face scars. This index contained four levels (none, small, medium and large), and was run in the third logistic regression model along with plot to determine if the size of the cat-face had and effect on the likelihood of the tree being attacked.  A multiple linear regression model was then assessed using the PROC REG function with height as the dependent variable and DBH, plot, and a binary variable called ?breakage? for broken tops as predictors.  This model was run to determine if tree height was affected by the DBH, the plot in which the tree was growing, or the presence or absence of a broken top.    Results Study area Three tree cores from living trees in each plot were analyzed to determine that the mean stand age of the study area is 59.7 years.  This age is consistent with the approximation that the area was logged about 50 ? 60 years ago according to the Bird Studies Canada website.   Bark beetle attack census The likelihood of attack of red alder by Alniphagus aspericollis was significantly influenced by the number of injuries on trees, and the plot in which the trees were growing but not by the tree diameter (Table 2).  When interactions between all variables in the model were included none were found to be significant given the selected alpha value of 0.05.  Fig. 2 shows that for each plot the attacked trees had a significantly higher mean number of injury codes than the unattacked trees.  The mean number of injuries per tree ranged from 1.55 to 2.10 in attacked trees and from 0.54 to 0.83 in unattacked trees (Fig. 2).  Twenty-one of 26 trees that had three or more injury codes were attacked.  Attacked trees had almost three times the number of injury codes than unattacked trees on average.  The average number of broken tops was six times greater for attacked trees than unattacked trees.  The average number of broken tops was about 3.8 times greater for trees with DBH less than 25 cm than for trees with DBH greater than 25 cm.  Trees with DBH less than 25 cm had 16% more injury codes on average than trees with DBH greater than 25 cm.  Fig. 3 shows that the height of attacked trees differed significantly from the height of unattacked trees in each plot.  When the interaction between height and plot was included in the model it was not found to be significant.  The height of trees ranged from 6.22 ? 6.90 m in attacked trees and from 8.64 ? 9.40 m in unattacked trees (Fig. 3).  Attacked trees were 27% shorter on average than unattacked trees.  The likelihood of attack of red alder trees by A. aspericollis was also influenced by the height of trees (Table 3).  In the third logistic regression model the tree?s cat-face score was not found to significantly affect attack status (p = 0.3490), while the plot in which the tree was growing was (p = 0.0024).  In the multiple linear regression (MLR) model DBH and breakage were found to significantly affect tree height but plot was not (Table 4).  Fig. 4 shows that the relationship between height and DBH was quite similar for trees with and without broken tops. Black- and rust-colored staining of the outer surface of the bark was observed around many of the beetle entrance holes, as was also observed by Borden (1969), but many holes were also observed to have no staining.    Table 2. Type III analysis of effects. Model 1: Predicting attack status of red alder trees by Alniphagus aspericollis based on number of injuries tree had, the plot in which they were growing, and their diameter at breast height (DBH).  DF = degrees of freedom  Effect DF Wald ?2 Pr > ?2 Number of injuries 4 36.5928 <.0001 Plot 2 6.5619 0.0376 DBH 1 3.5664 0.0590   Table 3. Type III analysis of effects. Model 2: Predicting attack status of red alder trees by Alniphagus aspericollis based on tree height and the plot in which they were growing. DF = degrees of freedom Effect DF Wald ?2 Pr > ?2 Height 1 32.2898 <.0001 Plot 2 8.1514 0.0170  Table 4. Parameter estimates. Multiple linear regression model: Predicting tree height based on the tree?s diameter at breast height (DBH), presence or absence of a broken top, and the plot in which the tree was growing.  DF = degrees of freedom Variable DF Parameter estimate Standard error t value Pr > abs(t) Intercept 1 6.29430 0.68796 9.15 <.0001 DBH 1 0.11229 0.01868 6.01 <.0001 Breakage 1 -3.92913 0.28311 -13.88 <.0001 Plot 1 -0.21597 0.13407 -1.61 0.1088   00.511.522.50 1Attack statusNumber of injuriesPlot1Plot2Plot3 Fig. 2. Comparison of the mean number of injuries of red alder trees that were attacked by Alniphagus aspericollis (0) and unattacked trees (1) in plots 1, 2, and 3. All results are statistically significant (? = 0.05). The data are presented as x? +/- SE  0246810120 1Attack statusTree height (m)Plot1Plot2Plot3 Fig. 3. Comparison of the mean height of red alder trees that were attacked by Alniphagus aspericollis (0) and unattacked (1) trees in plots 1, 2, and 3. All results are statistically significant (? = 0.05). The data are presented as x? +/- SE  y = 0.1168x + 1.8661R2 = 0.1149y = 0.1202x + 5.6753R2 = 0.237902468101214160 10 20 30 40 50DBH (cm)Height (m) Fig. 4. Scatter plot of red alder tree diameter at 1.3 m (DBH) vs. tree height.  Black diamonds represent trees with broken tops, and gray squares represent trees without broken tops   Discussion  The alder bark beetle, Alniphagus aspericollis, is capable of detecting and selectively colonizing red alder trees with impaired vigor.  Trees with greater numbers of putative vigor-impairing injuries were more likely to be attacked than those with fewer injuries.  The injury codes that were most common and therefore likely the drivers of this relationship include the following: broken top, heavy scarring, large bark fissures, fungus present, forked top, and top kill unbroken.  In a study conducted in 1991 in Big Basin Redwoods State Park in California Douglas-fir, Pseudotsuga menziesii (Mirb.) Franco, trees with broken tops and mountain hemlock, Tsuga mertensiana (Bong.) Carr., trees with dead tops were both said to be in states of declining vigor due to their injuries (Singer et al. 1991).  In a 1991 study in an old-growth Norway spruce forest in Sweden growth was said to be impeded for several years in trees that experienced top breakage (Hofgaard et al. 1991).  In a 2006 study 10-year-old seedlings of Douglas-fir, western redcedar, Thuja plicata Donn ex D. Don, and grand fir, Abies grandis (Douglas ex D. Don) Lindley, with various types of damage caused by logging (including broken tops) were all found to have decreased vigor as measured by reduced height growth in the two years following the damage (Newton and Cole 2006).  Absolute height growth was proven to be a good indicator of vigor for balsam fir and lodgepole pine (Ruel et al. 2000).  Mortality of six-year-old Douglas-fir advance regeneration was found to increase with damage severity and number of damage-linked injuries (Ruel et al. 2000).  This behavior of selectively attacking weakened trees is expected of any secondary bark beetle.  Incidences of a secondary bark beetle that infests aspen trees in Colorado were found to be significantly more common in plots in which trees were damaged than in plots with healthier trees (Marchetti et al. 2011).  While it is considered a secondary bark beetle, the aspen stands in which infestations were found to occur ?had many trees? in the 30 to 50 cm range of DBH (Marchetti et al. 2011), and the beetle was more commonly found in trees with a significant amount of living phloem tissue (Petty 1977).  For the birch bark beetle in Russia and Scandinavia healthy trees can be attacked but unhealthy trees are its usual targets (Linnakoski et al. 2009).   A preference of populations of A. aspericollis for attacking trees with reduced vigor could increase the average vigor of the stand by ?weeding out? the least vigorous trees.  This could potentially increase the volume of wood that can be harvested from the stand due to the increased growth experienced by trees that neighbor weaker trees that die from the beetle attacks.  It could also keep pathogens and diseased trees to lower levels in the stand by accelerating the death of weakened trees thereby limited the spread of the pathogens.  The preference for attacking trees with impaired vigor could also limit the ability of beetle populations to exhibit eruptive outbreak behavior due to the low likelihood that entire stands would have impaired vigor.  Managers of red alder stands should focus on limiting logging damage to retained trees during harvesting to reduce susceptibility of individual trees to attack by the bark beetle.   Intriguingly Alniphagus aspericollis does not prefer to focus its attacks on trees of any specific diameter in Vancouver, BC, unlike many other bark beetles.  It is curious that DBH did not have an effect on the likelihood of attack and it could provide some indication as to why the alder trees in this forest are infested with the bark beetle.  One possible explanation for the lack of significance of DBH in the model is that the stand is so old that even many of the larger diameter trees are weakened, meaning that all of the trees are so susceptible to attack that no relationship with DBH could be isolated.  It could be that beetle?s preference has gradually shifted from smaller diameter to larger diameter trees as the resources in the smaller trees became exhausted, but this cannot be known for certain because the time that each tree was attacked was not determined.  Because no preference for attacking larger diameter trees was observed for Alniphagus aspericollis it is very unlikely that populations of this species are capable of eruptive outbreaks.  Successful establishment of individual beetles is likely to be lower on average as a result of their lack of preference because the defenses of larger diameter trees are likely more difficult to overcome than those of smaller diameter trees.  Their lack of preference for any specific diameter may also mean that populations would be able to persist in a range of stands that vary in tree diameter.   Given that the likelihood that a tree was attacked was influenced by the plot in which it was growing suggests that the variation among plots affects some aspect of the beetles? preference for individual trees.  This could be attributed to differences in wind exposure or slope position among the three plots which would lead to differences in the proportion of trees in each plot that have broken tops.  The height difference between attacked and unattacked trees indicates that a tree is significantly more likely to be attacked if it has a broken top than if it does not.  This statement can be made because of the correlation between height and broken tops.  However, this could also simply be due to the fact that dead trees were included in the figure, and dead trees were found to be more likely to have broken tops than living trees and all dead trees were attacked by the beetle.   The fact that the cat-face score variable was not significant in the third logistic regression model suggests that the cat-face score index was likely not a very good indicator of the tree?s injury level.  This finding could be attributed to the limited number of trees that had cat-face scars or to the limited degree to which a red alder tree?s vigor is affected by that type of damage.  The greater height of trees without broken tops than those with broken tops highlights the strong relationship between height and broken tops.  It also reveals that the relationship between DBH and height is similar for trees with and without broken tops because of the similarity in the slopes of the two best fit lines.  This is interesting because it suggests that larger diameter trees with broken tops are more likely to have the break occur at a greater height than smaller diameter trees.  This indicates that the greater stem stiffness of larger diameter trees causes them to break off at greater heights than smaller diameter trees in the event of top breakage.   In this observational study I have shown that Alniphagus aspericollis preferentially attacks red alder trees with impaired vigor.  This information will be of interest to forest managers of red alder stands in which limiting bark beetle-caused damage is a primary objective.  I have also shown that the likelihood of an individual tree being attacked by A. aspericollis is not significantly affected by the DBH of the tree.  These findings suggest that mortality of entire landscapes of red alder due to the bark beetle is unlikely to occur and they pave the way for the development of the study of other aspects of the A. aspericollis-red alder system.    Acknowledgements   Thanks to Lori Daniels for loaning me two increment borers and providing me with WD-40, a starter for the increment borer, a bunch of straws, some chopsticks, and a felt-tip pen. Thanks to my friends Benn Mapes, Alex LaForce, and Martin Lentz for each helping me out with the data collection for one day. Thanks to Jordan Burke for loaning me his Garmin ? eTrex ? 20 GPS device. I would like to thank Dr. Allan Carroll, Dr. Sarah Gergel, Dr. Richard Hamelin, Dr. Monique Sakalidis, Dr. Sally Aitken, Rob Roy McGregor, and Mark Macdonald for valuable discussions.     Literature Cited  Anderson, P.L. and Holliday, N.J. 2003. Distribution and survival of overwintering adults of the Dutch elm disease vector, Hylurgopinus rufipes (Coleoptera: Scolytidae), in American elm trees in Manitoba. Agr. Forest Entomol. 5: 137?144.   Avtzis, D.N., Bertheau, C., and Stauffer, C. 2012. What is next in bark beetle phylogeography? Insects. 3: 453?472. doi:10.3390/insects3020453.   Bleiker, K.P., Lindgren, B.S., and Maclauchlan, L.E. 2005. Resistance of fast- and slow-growing subalpine fir to pheromone-induced attack by western bark beetle (Coleoptera: Scolytinae). Agr. Forest Entomol. 7: 237?244.     Borden, J.H. 1969. Observations on the life history and habits of Alniphagus aspericollis (Coleoptera: Scolytidae) in Southwestern British Columbia. The Can. Ent. 101: 870?878.    Bowers, W.W., Borden, J.H., and Raske, A. 1996. Incidence and impact of Polygraphus rufipennis (Coleoptera: Scolytidae) in Newfoundland. Forest Ecol. Manag. 89: 173?187. doi:10.1016/S0378-1127(96)03850-9.  Buhroo, A.A. 2012. Host selection behavior and incidence of the bark beetle Scolytus kashmirensis (Coleoptera: Curculionidae: Scolytinae) attacking elm (Ulmus spp.) trees in Kashmir. Forestry Studies in China. 14(3): 224?228.   Cates, R.G., and Alexander, H. 1982. Host resistance and susceptibility. In Bark beetles in North American conifers. Edited by J.B. Mitton and K.B. Sturgeon. University of Texas Press, Austin, Texas. pp. 21?45.  Chamberlin, W.J. 1939. The bark and timber beetles of North America. OSC Cooperative Association. Corvallis, Oregon. pp. 1?513.   Chamberlin, W.J. 1958. The scolytoidea of the Northwest. Oregon State University Press. Corvallis, Oregon. pp. 1?207.   Chen, C-Y., Yang, H-C. P., Chen, C-W., and Chen, T-H. 2008. Diagnosing and revising logistic regression models: effect on internal solidarity wave propagation. Engineering Computations. 25(2): 121?139. doi:10.1108/02644400810855940.  Deal, R.L. 2006. Red alder stand development and dynamics. USDA For. Serv. PNW-GTR-669.  Furniss, M.M., and Johnson, J.B. 2002. Field guide to the bark beetles of Idaho and adjacent regions. Agricultural Publications, University of Idaho. Moscow, ID. pp. 1?111.   Haverty, M.I, Shea P.J., Hoffman, J.T., Wenz, J.M., and Gibson, K.E. 1998. Effectiveness of esfenvalerate, cyfluthrin, and carbayrl in protecting individual lodgepole pines and ponderosa pines from attack by Dendroctonus spp. USDA For. Serv. PSW-RP-237.   Hofgaard, A., Kullman, L., and Alexandersson, H. 1991. Response of old-growth montane Picea abies (L.) Karst. Forest to climatic variability in Northern Sweden. New Phytol. 119(4): 585?594.    K?hnholz, S., Borden, J.H., and McIntosh, R.L. 2000. The ambrosia beetle, Gnathotrichus retusus (Coleoptera: Scolytidae) breeding in red alder, Alnus rubra (Betulaceae). J. Entomol. Soc. BC. 97: 103?104.   Kunca, A., Leontovy?, R., Z?brik, M., and Gubka, A. 2011. Bark beetle outbreak on weakened ash trees and applied control measures. EPPO Bulletin. 41: 11?13. doi:10.1111/j.1365-2338.2010.02428.x.  Lee, J.C., Negr?n, J.F., McElwey, S.J., Williams, L., Witcosky, J.J., Popp, J.B., and Seybold, S.J. 2011. Biology of the Invasive Banded Elm Bark Beetle (Coleoptera: Scolytidae) in the Western United States. Ann. Entomol. Soc. Am. 104(4): 705?717. doi:10.1603/AN10150.  Linnakoski, R., de Beer, Z.W., Rousi, M., Solheim, H., and Wingfield, M.J. 2009. Ophiostoma denticiliatum sp. nov. and other Ophiostoma species associated with the birch bark beetle in southern Norway. Persoonia. 23: 9?15. doi:10.3767/003158509X468038.   Marchetti, S.B., Worrall, J.J., and Eager, T. 2011. Secondary insects and diseases contribute to sudden aspen decline in southwestern Colorado, USA. Can. J. For. Res. 41: 2315?2325. doi:10.1139/X11-106.  Markham, J.H., and Chanway, C.P. 1998. Response of red alder (Alnus rubra) seedlings to a wooly alder sawfly (Eriocampa ovata) outbreak. Can. J. For. Res. 28: 591?595.   May, C.A., Petersburg, M.L., and Guti?rrez, R.J. 2004. Mexican spotted owl nest- and roost-site habitat in northern Arizona. J. Wildlife Manage. 68(4): 1054?1064.    Myers, J.H., and Williams, K.S. 1987. Lack of short or long term inducible defenses in the red alder ? western tent caterpillar system. Oikos. 48: 73?78.   Niemiec, S.S., Ahrens, G.R., Willits, S., and Hibbs, D.E. 1995. Hardwoods of the Pacific Northwest. Forest Research Laboratory, Oregon State University, Corvallis. Research Contribution 8. 13 p.   Newton, M, and Cole, E.C. 2006. Harvesting impacts on understory regeneration in two-storied Douglas-fir stands. Western Journal of Applied Forestry. 21(1): 14?18.   Ohmart, C.P. 1989. Why are there so few tree-killing bark beetles associated with angiosperms? Oikos. 54(2): 242?245.   Petty, J.L. 1977. Bionomics of two aspen bark beetles, Trypophloeus populi and Procryphalus mucronatus (Coleoptera: Scolytidae). Great Basin Nat. 37(1): 105?127.   Plank, M.E., Snellgrove, T.A., and Willits, S. 1990. Product values dispel ?weed species? myth of red alder. Forest Products Journal 40(2): 23?28.  Ruel, J., Messier, C., Doucet, R., Claveau, Y., and Comeau, P. 2000. Morphological indicators of growth response of coniferous advance regeneration to overstorey removal in the boreal forest. Forestry Chron. 76(4): 633?642. doi:10.5558/tfc76633-4.   Simard, M., Powell, E.N., Raffa, K.F., Turner, M.G. 2012. What explains landscape patterns of tree mortality caused by bark beetle outbreaks in Greater Yellowstone? Global Ecol. Biogeogr. 21(5): 556?567.  Singer, S.W., Naslund, N.L., Singer, S.A., and Ralph, C.J. 1991. Discovery and observations of two tree nests of the marbled murrelet. Condor. 93(2): 330?339. doi:10.2307/1368948.   Schedl, K.E. 1949. A new species of Alniphagus (Col. Scolytidae) from Canada. Can. Entomol. 81(9): 235?235. doi:10.4039/Ent8123-7.  Smith, G.D., Carroll, A.L., and Lindgren, B.S. 2009. Life history of a secondary bark beetle, Pseudips mexicanus (Coleoptera: Curculionidae: Scolytinae), in lodgepole pine in British Columbia. Can. Entomol. 141(1): 56?69. doi:10.4039/n08-054.   Stark, R.W. 1982. Generalized ecology and life cycle of bark beetles. In Bark beetles in North American conifers. Edited by J.B. Mitton and K.B. Sturgeon. University of Texas Press, Austin, Texas. pp. 21?45.   Teen, E. 2012. Associations of secondary bark beetles with dying and live lodgepole pine in the post-outbreak phase of mountain pine beetle, Dendroctonus ponderosae (Hopkins), in the central interior of British Columbia, Canada. M.Sc. thesis, Department of Biology, University of Northern British Columbia, Prince George, B.C.  Wood, S.L. 1984. New synonymy and new species of American bark beetles (Coleoptera: Scolytidae), Part X. Great Basin Nat. 44(1): 113?119.  Wood, S.L. 1987. Six new Scolytidae (Coleoptera) from Mexico. Western North American Naturalist. 47(4): 547?550.   


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items