Abstract
The protection of high-value trees against bark beetles and the development of alternatives to bole sprays is a priority for the tree manager. The objective of this study was to evaluate stem-injected TREE-äge® (emamectin benzoate [EB]) as a protective treatment for western white pines (Pinus monticola Dougl. ex D. Don) against mountain pine beetle (MPB, Dendroctonus ponderosae Hopkins). Treatment efficacy was based solely on tree mortality as per Shea protocols (i.e., ≥ 60% check vs. ≤ 20% treated tree mortality). Our first experiment was installed in 2007 and included trees stem-injected with TREE-äge and untreated controls. Bole application of S-(-)-verbenone and green leaf volatile (GLV) blend was included for observational comparison. Pressure from MPB was heavy, as indicated by the number and timing of control tree mortality (90%). Strip attacks by MPB in TREE-äge trees indicated that the impacts of EB, and by inference its distribution, were inconsistent. In 2009, the injection protocol was revised to improve EB distribution in the phloem via closer injection points. In the 2009 TREE-äge-treated trees, adult beetle mining stopped when they contacted phloem and was insufficient to cause tree death by girdling. Blue-stain fungi colonized the sapwood of trees in both studies. Isolates from autopsied trees treated with TREE-äge alone were subsequently identified as Grosmannia clavigera and Leptographium longiclavatum (Ophiostomatales: Ascomycota), species that can incite tree mortality. In 2013, we revised our protocol to include GLV plus verbenone or propiconazole with TREE-äge, wherein these treatments proved effective in protecting trees against MPB and their associated pathogenic fungi.
- Bark Beetles
- Blue-Stain Fungi
- Grosmannia clavigera
- Leptographium longiclavatum
- Propiconazole
- Tree Injection
- Verbenone
INTRODUCTION
White pines, including western white (Pinus monticola Dougl. ex D. Don), sugar (P. lambertiana Dougl.), limber (P. flexilis James), and whitebark (P. albicaulis Engelm.), are a component of more than 28 Society of American Foresters (SAF) Cover Types (Burns and Honkala 1990). They are important throughout the western USA for timber, aesthetics, recreation, water quality, and ecosystem functioning and are favored hosts for the mountain pine beetle (MPB), Dendroctonus ponderosae Hopkins (Coleoptera: Curculionidae: Scolytinae), the major insect threat to western white pines (Furniss and Carolin 1977; Burns and Honkala 1990). Effective treatments (i.e., as defined by Shea et al. 1984) for the protection of individual pines from MPB attack are limited to insecticides applied to the outside of the bole prior to beetle attack. Shea’s protocol is a management tool that defines treatment effectiveness under conditions where beetle pressures incite ≥ 60% of control tree mortality, but cause ≤ 20% of treated tree mortality. Active ingredients (products) with demonstrated efficacy include carbaryl (Sevin™), which is often preferred due to its longevity (reliably two seasons), bifenthrin (Onyx™), and permethrin (Astro™), both of which typically provide one season of protection (Shea et al. 1984; Haverty et al. 1985; Fettig et al. 2006). If bole-sprayed insecticides are undesirable for a particular application (e.g., close to water, poor access for equipment, incompatible with management objectives), there are no effective alternatives yet identified.
Verbenone is an anti-aggregation pheromone derived from α-pinene and used to deter beetles from attacking trees (Fettig et al. 2009). Green leaf volatiles (GLVs) are organic compounds (including alcohols, esters, and aldehydes) that are released when plants are wounded; these reduce bark beetle attraction to trees and are sometimes used in conjunction with verbenone as a combined strategy to protect trees (Gillette et al. 2014). Development of semiochemical treatments continues to be an active area of research, but consistency in effective tree protection has yet to be realized (Progar 2005; Kegley and Gibson 2009).
TREE-äge® (Syngenta Crop Protection, Greensboro, NC, USA) is a formulation of emamectin benzoate (EB, 4% [wt/wt]) designed for systemic application against tree pests. Formulations of EB have been shown to inhibit propagation of the pine wilt nematode, Bursaphelenchus xylophilus (Steiner and Buher) Nickle, in pine shoot tissues (Takai et al. 2003) and impact a variety of insect pests, most notably Coleoptera and Lepidoptera that attack crown tissues, including leaves and cones (Grosman et al. 2002; Grosman and Upton 2006). The product is effective (for achieving management objectives) against the emerald ash borer, Agrilus planipennis Fairmaire (Coleoptera: Buprestidae)(Smitley et al. 2010; McCullough et al. 2011); however, treatments applied to protect trees from bark beetles have produced mixed results. Previous applications against D. ponderosae have been ineffective, as have those against D. rufipennis (Grosman et al. 2010). However, effective results have been attained against D. brevicomis and in one year against D. frontalis (Grosman et al. 2009, 2010).
To be useful, tree protective treatments must succeed against beetle pressures that are equal to or greater than those that trigger management interventions. Beetle pressures are characterized experimentally by the observed mortality of control trees in the experimental population; evaluation under sufficient beetle pressure is critical for determining treatment utility and cost-to-benefit considerations. Previous studies with bark beetles and TREE-äge have predominantly lacked sufficient pest pressures to determine efficacy via the Shea protocols; however, outbreak populations of MPB in the Warner Mountains area of northeastern California during the late 2000s provided an opportunity to develop and evaluate tree protection treatments under conditions of heavy MPB pressure. The objectives of this experiment were: (1) to evaluate TREE-äge for its efficacy to protect trees against high levels of pressure from the mountain pine beetle in western white pine; (2) to determine EB residues in the phloem; and (3) to evaluate and improve upon existing tree-injection protocols. Three distinct field studies were installed in 2007, 2009, and 2013.
MATERIALS AND METHODS
This study was conducted from fall 2007 to fall 2015 in the Warner Mountains Ranger District, Modoc National Forest, California, USA. Our study site included stands that were mixed coniferous forests, predominantly lodgepole pine (Pinus contorta Dougl. ex Loud. variety murrayana) with western white pine (SAF Cover Type 218). Surveys first noted increased tree mortality from D. ponderosae in 2005, apparently peaking in 2008 (USDA Forest Service 2008). During the study period, over 60% of the pine component in the area was killed by MPB. The study consisted of three experiments; experiment 1 began with tree injections in fall 2007, experiment 2 in fall 2009, and experiment 3 in spring 2013. In each experiment, trees were selected in transects along roads, far enough to avoid the creation of road hazards from dead trees, but close enough for servicing. Trees were at least 100 m apart, each being considered an independent, experimental unit. The population of experimental trees consisted of 10 trees per treatment in a completely randomized design.
Experiment 1
Experiment 1 evaluated prophylactic bole-injection with TREE-äge for its ability to protect western white pines from MPB-caused tree mortality. On 17 to 18 September 2007, 10 trees were injected with TREEäge using the Tree I.V. system (Arborjet, Inc., Woburn, MA, USA), and 10 trees were set aside as untreated controls. Trees averaged 30.9 ± 2.0 cm (mean ± 1 SD) diameter at breast height (DBH), and treated trees were injected with the high label rate of 10 mL of product per 2.5 cm DBH. The Tree I.V. is a micro-infusion delivery system comprised of a 1-L pressure bottle configured with a delivery line which splits into 4 injector tips via an in-line manifold. TREE-äge was mixed 1:1 v/v with bottled, distilled water. Four injection points were made into each tree near the root collar, spacing them about equally apart. The average dose of active ingredient (AI) injected was 4.8 g of EB per tree.
Experiment 2
Trees in experiment 2 were treated on 5 November 2009. Ten trees were injected with TREE-äge using the QUIK jet micro-injector (Arborjet, Inc.) and ten trees were set aside as untreated controls. The QUIK-jet micro-injector is designed with a 5-cc capacity barrel and applied at one point at a time. This micro-injection method made it possible to increase the number of injection points around the tree, but at a lower per tree dose. Study trees (n = 20) averaged 28.8 ± 2.0 cm (mean ± 1 SD) DBH. Those treated with TREE-äge (n = 10; mean DBH = 28.4 ± 2.0 cm) were injected with 7.5 mL of product per 2.5 cm DBH, a rate intermediate to the medium and high label use rates. To promote better distribution of insecticide across the phloem tissue, injections were made about every 9 cm of circumference at the root collar (mean number of sites = 11.4 per tree). The average dose of AI was 3.4 g of EB per tree.
Experiment 3
Trees in experiment 3 were treated on 11 June 2013. Ten trees each were treated with (a) TREE-äge using the air-hydraulic micro-injector (Arborjet, Inc.) plus GLV/verbenone (Synergy Semiochemicals, Corp., Burnaby, BC, Canada; Product No. 3413), the latter to deter beetle attack; (b) TREE-äge and Alamo (14.3% wt/wt propiconazole, ME Syngenta Crop Protection, Greensboro, NC, USA)(PPZ); or (c) were untreated controls. The study trees (n = 30) averaged 25.6 ± 2.5 cm DBH. TREE-äge was applied once every 10 cm of basal circumference at 6.7 mL per injection (equivalent to the medium label rate of 5 mL per 2.5 cm DBH) in treatments (a) and (b). In addition, treatment (a) received one pouch each of verbenone and GLV affixed on the north-facing bole at 182 cm from ground level. In addition, treatment (b) received PPZ at 8 mL per injection site (label rate of 6 mL PPZ per 2.5 cm DBH). The mean cm DBH of injected trees and number of injection points were 26.0 cm and 10, respectively. The mean dosages of EB and PPZ were 2.08 g and 8.8 g, respectively. Table 1 provides a summary of the three experiments.
Summary of experiments 1–3 reflects the modification of protocols over time. In the first experiment, TREE-äge was injected into Pinus monticola (western white pine) using the Tree I.V. at relatively wide spacing at the 10 mL/2.5 cm label rate. Subsequent experiments used the modified closer injection spacing using as low as the 5 mL/2.5 cm label rate. In the last experiment, either verbenone and green leaf volatile (V/GLV) or 14.3% propiconazole (PPZ), a systemic fungicide, was applied in conjunction with TREE-äge.
Beetle Challenge
Challenge from MPB was promoted in the experiments using commercially available attractant lures consisting of exo-brevicomin, trans-verbenol, and a 1:1 mixture of myrcene and terpinolene (Synergy Semiochemicals, Corp., Burnaby, BC, Canada; Product No. 3093). This is the combination lure identified as most attractive for MPB in northeastern California (Strom et al. 2004). Attractant lures were attached to poles approximately 1 m away from the tree bole at 1.2-m height (Strom et al. 2004) and installed once roads were passable during the early flight season for MPB in the study area. In experiment 1, lures were deployed on 1 July 2008, replaced on 1 August 2008, and removed on 4 September 2008. In experiment 2, trees were baited with attractant lures on 7 July 2010, and lures were removed on 4 November 2010. In experiment 3, trees were baited 11 July 2013 through 21 May 2014, with lures attached to a stake 1 m from the tree bole on the north side; the GLV/verbenone pouch (Synergy Semiochemicals, Burnaby, BC, Canada; Synergy Shield Pouch, Product No. 3413) was installed directly on the tree. Depending on temperature, MPB adults may be active from May through October (USDA Forest Service 2015).
Treatment and Tree Evaluations
Trees were evaluated similarly for each experiment. The basic procedure was to treat in fall (except for experiment 3, which was treated in spring), bait with semiochemical attractants the following summer, conduct initial tree evaluations (crown fade and count attacks) in the fall (approximately 1 year post-treatment), and perform a final, crown-based evaluation of tree mortality the next fall (approximately 2 years post-treatment and 1 year post–beetle-attack-period). Because the majority of precipitation at our site is received during the winter and spring seasons, and MPB attacks were promoted into the fall, we made our crown assessments for fade after approximately 1 year had passed from the end of the baiting period. A similar protocol was used in Idaho by Haverty et al. (1998). Crown evaluations placed each tree into one of three categories: green (live), fading (stressed or dying), or faded (dead), with the latter category being indicated by red-brown crowns. Pitch tubes were counted at least once after baiting. In experiment 3, pitch tubes were counted on two sides of the bole within a rectangular area approximately 2812 cm2 at breast height.
Prior to beetle emergence in the second year after treatment, two emergence cages (aluminum window screening 37 × 76 cm) were installed on opposite sides of each TREE-äge or control tree at about breast height to capture emerging beetles (Bentz 2006). In each case, cages were installed prior to brood emergence. In experiment 3, emergence cages were installed 20 to 21 May 2014.
During fall of 2014 (approximately 16 months after treatment), emergence cages were removed, MPB collected, and the number of D. ponderosae attacks and emergence holes in the bark counted within the caged area (2 sides, each 2812 cm2; Table 2). Attacks were made prior to cage installation, and cages provided a well-delineated area for counting them. Bark samples (19 cm wide and 12 cm long) were removed from TREE-äge and control trees to estimate MPB activity by laboratory dissection. Using this method, attacks, adult galleries, larval galleries, and emergence holes were determined (Table 3). On 25 August 2013, lures and GLV/verbenone repellents were changed. On 18 June 2014, phloem samples were again taken from the trees, and on 19 August and 15 October 2014, canopy assessments were conducted.
Mean number of mountain pine beetle (Dendroctonus ponderosae) pitch tubes present (± SEM) in Pinus monticola (western white pine) among treatments observed on 30 July 2013 in experiment 3, 19 days after baiting. Pitch tubes were counted on two sides of the bole within a rectangular area (37 × 76 cm) 2812 cm2 at breast height. The untreated controls had the greatest number of pitch tubes, which differed significantly from the combination treatments. The lowest number of pitch tubes was observed on the TREE-äge and verbenone/green leaf volatile (V/GLV) treatment. The purpose of V/GLV is to decrease mountain pine beetle (MPB) attack by acting as an anti-aggregation pheromone. The dual purpose of the TREE-äge/PPZ treatment was to prevent MPB adults from successfully mining the phloem and to inhibit the development of pathogenic blue-stain fungi.
Evaluation of TREE-äge-treated and untreated western white pine (Pinus monticola) use by Dendroctonus ponderosae in three experiments in the Warner Mountains of California (mean ± SEM). Bark samples were collected at breast height (1.4 m) and were approximately 210 cm2. Data are based on 10 trees per treatment, however, actual sample size varied due to the number of trees killed in each treatment and the adequacy of bark samples for measuring response variables. Larval data from experiment 2 were determined from 7 untreated trees; 3 were discarded due to a lack of discernible MPB activity caused by secondary agents in the phloem. Trees in experiment 1 were treated on 17 to 18 September 2007, baited with attractant semiochemicals to promote attack by MPB from 1 July to 4 September 2008, and had bark samples removed on 11 to 12 August 2009. Trees in experiment 2 were treated on 5 November 2009, baited from 7 July to 4 November 2010, and had bark samples removed on 27 October 2011. Trees in experiment 3 were treated on 11 June 2013, baited from 11 July 2013 to 21 May 2014, and sampled 15 October 2014.
Tree Autopsies
Fading and faded trees (9 untreated controls and 7 TREE-äge-treated) from experiment 2 (injected 5 November 2009) were felled for autopsy on 27 October 2012, 2 years after they were baited to promote attack by MPB. Cross-sectional bole samples (approximately 10 cm thick) were collected from 3 to 12 m high in 3-m intervals using a chain saw. Samples (n = 63) were debarked, changes in the physical appearance of the vascular tissues were noted, and percentage of sapwood circumference blue-stained was calculated. Samples of xylem and phloem tissues were collected for culturing and identification.
To quantify the relationship between injection frequency (circumferential distance between injection points) and colonization pattern by MPB, we evaluated trees from the first two experiments on 27 October 2012 and the third experiment on 20 October 2015. For each tree, we measured the circumference at the base and breast height and the distance between injection points. This was followed by removal of the outer bark to expose the inner bark or sapwood face. Areas of exposed tissue were categorized as either colonized by MPB (at least one gallery > 2.5 cm in length) or not, and the circumferential distance of category types was recorded using a flexible measuring tape. For each tree, two sums were determined, i.e., the circumferential distances colonized and un-colonized by MPB. In experiment 3, a subset of 21 trees (7 from each treatment) were felled for autopsy.
Insecticide Residues
Residues of EB in phloem tissue were determined using enzyme-linked immunosorbent assay (ELISA) kits (HORIBA Biotechnology Co., Ltd., Kyoto, Japan). Phloem samples were field collected using a 5-cm hole saw and kept frozen until processing. Samples were collected from two compass directions (North and South) at each sampling period. In experiment 1, trees were sampled on 1 July 2008 and 7 July 2009. In experiment 2, trees were sampled on 7 July 2010, 4 November 2010, and 27 July 2011. In experiment 3, trees were sampled on 11 July 2013, 18 June 2014, and 15 October 2014. To allow the evaluation of height effects on movement of EB, the first sample collection (1 July 2008) included height (1.2 and 2.4 m) as a factor in addition to direction. All other samples were removed at breast height while avoiding previously sampled areas.
Phloem samples were kept frozen (−19 °C) until preparation for residue determinations. They were oven-dried at 38 °C for at least 48 hours, then ground in a Wiley mill (Thomas Scientific, Swedesboro, NJ, USA, mesh size 20). A 0.5-g portion of each ground sample was extracted in 10 mL of chromatography grade methanol (Burdick and Jackson, Muskegon, MI, USA) by overnight horizontal agitation. After removal from the agitator, sediment was allowed to settle, and 1.5 mL from each sample was transferred to a 12 × 32 mm auto-sampler vial with a Teflon-lined screw cap (SUN SRI, Rockwood, TN, USA). Samples were refrigerated (4 °C) until determinations of residue concentrations were made.
For ELISA kits, all manufacturer instructions were followed. Duplicate wells were used for each sample and their mean recorded. The useable range of the Horiba EB kits is 0.3 to 3.0 ng/mL, so sample extracts were diluted serially with 10% (v/v) methanol/water until a result within this range was achieved. A Biotek ELx808 plate reader (Biotek, Inc., Winooski, VT, USA) was used to measure absorbance at 450 nm and results were calculated using 4-parameter curve fitting software (Biotek Gen5, Biotek, Inc.). Both matrix-only blanks and EB spikes (0.6 μg/g EB) were included with each plate to provide an estimate of matrix interferences and to ensure kit performance; the mean value of the tray blanks in the same run was subtracted from each treated sample value to provide the data used for presentation and analysis.
STATISTICAL ANALYSES
The response variable of primary interest was tree mortality as indicated by crown fade at the conclusion of the experiment. We used the definition of treatment efficacy provided by the Shea protocols (Shea et al. 1984) and applied their evaluation procedure using percentage of tree mortality; at least 60% of untreated trees must die for a valid experimental test, and no more than 20% of treated trees may die for treatment efficacy to be attained. The effect of treatment on MPB utilization of phloem was measured from bark samples (number of attacks, adult gallery number and length, larval gallery number and length). These values were averaged for each tree, and resulting means were subjected to t-tests; when unequal variances between treatments were indicated (Brown-Forsythe test, P < 0.10), Welch’s t-test was employed. Ordered categorical data were arranged into contingency tables, and equality of treatments was tested via a Kruskal-Wallis statistic, with ties replaced by mid-ranks (KW) and P-value based on the exact permutation distribution (Px)(StatXact, V. 10, Cytel, Inc., Cambridge, MA, USA). In experiment 3, treatment differences in crown fading over time were evaluated by subjecting the number of trees tallied in each of three crown-fade categories (green, fading, faded) and compared by treatment using Fisher’s exact test with permutation P-value. Each crown observation period was analyzed separately.
Insecticide residues were evaluated by subjecting the first set of samples (collected 1 July 2008) to mixed model ANOVA for a randomized complete block design (RCBD) with a 2 × 2 factorial treatment structure. In this analysis, each tree was considered a block (and random) with sample height (1.2 and 2.4 m) and compass direction (East and West) as fixed factors. Phloem samples thereafter were collected only at breast height (1.4 m), and tree residue means, by date of collection, were subjected to repeated measures ANOVA and Tukey’s HSD to determine effects based on experiment and sample date. This provided 5 means for pairwise comparisons (2 and 3 sampling dates for trees in experiments 1 and 2, respectively, Figure 1). Residue concentrations were transformed by In(y + 0.03) prior to analysis to reduce variance heteroskedasticity that was apparent in residual plots.
Mean concentration of emamectin benzoate (EB) in phloem tissue of western white pines in experiments 1 (injected 2007) and 2 (injected 2009). First number before the dash indicates the year of treatment, and the number after the dash indicates the number of days post-treatment when sample was collected. Samples collected at 244 and 287 days post-treatment were collected at the time of tree baiting and indicate residue concentrations present when trees were initially challenged by mountain pine beetle, Dendroctonus ponderosae. Different letters above standard error bars indicate means that are significantly different (Tukey’s HSD, P < 0.05).
The effect of the centimeter distance between injection sites in the 2 experiments (wide spacing in 2007 and close spacing in 2009) on the percentage of circumference protected from MPB (galleries > 2.5 cm) was evaluated by regression analysis. All statistical analyses used P < 0.05 to determine significance.
RESULTS AND DISCUSSION
Tree Mortality
Our primary response variable was tree mortality, as indicated by crown fade, and our primary evaluation criterion was treatment efficacy as per the Shea criteria (Shea et al. 1984). We allowed a second season for final crown assessments. Our experience and observations made in previous studies (e.g., Grosman et al. 2009) indicate that trees injected with TREE-äge may fade more slowly than control trees. In experiment 1, none of the treatments was efficacious, with mortality of untreated trees being 9 of 10 and TREE-äge-treated trees 9 of 10. In experiment 2, 9 of 10 untreated trees died, and 7 of 10 TREE-äge trees died at our final evaluation on 27 October 2011. Trees treated in 2013 (experiment 3) received their final evaluation of change in crown condition on 21 October 2015. Sufficient mortality of control trees was achieved (10 of 10) to provide a valid experimental challenge to treatments by beetles (Shea et al. 1984).
MPB Colonization
We observed successful MPB colonization between areas roughly coinciding with injection points in experiment 1, leading us to inject at closer spacing in experiment 2 (approximately every 10 cm of basal circumference). MPB adults mine the phloem along the axis of the tree bole. This linear pattern of mining differs markedly from other scolytid beetles (such as spruce beetle and western pine beetle) whose mining galleries are more sinuous. For scolytids that create serpentine mines, the standard injection spacing of 15 cm should suffice, as the probability of encountering the injected chemistry is higher.
EB Residues
At the time of baiting (7 July 2010, 244 days post-injection), the mean concentration of EB in the phloem was 31.9 ± 9.2 μg/g dry weight, and on 27 July 2011 (629 days post-injection), the mean concentration of EB was 24.7 ± 8.3 μg/g (Figure 1). These values are, respectively, 13.2 and 5.6 times higher than the equivalently timed samples from experiment 1 (Figure 1). In experiment 2, residues were also determined 364 days post-treatment (mean = 28.5 ± 12.6 μg/g), at which time they were between those found at 244 and 629 days. An additional benefit of the closer spacing was that lower per tree doses of EB were applied (i.e., the medium label rate of 5 mL/2.5 cm DBH)(Table 1), and the absence of an effect of time for either injection method suggests that the lower amount of AI is sufficient for preventing adult gallery construction for the time period covered in this experiment.
Injection Point Spacing
To estimate an optimal distance between injection points for future applications, a regression of the percentage of circumference at breast height successfully protected from MPB (galleries > 2.5 cm) and the centimeter distance between injection sites was conducted for the two experiments (Figure 2). This evaluation requires an assumption of linearity of response between 10-and 25-cm injection distances, a test of which will require additional experimentation because we did not inject at intermediate distances. However, the prediction provides a starting point to guide future experiments. In the first experiment, injection sites were spaced an average (± 1 SD) of 32.2 ± 4.7 cm apart using the Tree I.V. In the second experiment, the interval between injections was decreased to 8.9 ± 0.5 cm. The analysis suggests that decreasing the distance between injection sites to approximately 10 cm around the basal circumference may be most effective for the lateral distribution of the injected chemistry. Interpolation was based on the regression equation y = 1.14 – 0.016x (Figure 2). We applied this closer spacing for tree injection in experiment 3. Micro-injection applications (in this experiment, using the air hydraulic) delivered product in less time compared to the TREE I.V., with realized savings in both materials and labor.
Regression based on in situ measurements of the number of basal tree injections of TREE-äge and the percentage of circumference at breast height that was successfully attacked by MPB (galleries > 2.5 cm) in the two experiments. In experiment 1, injection sites were spaced (mean ± 1 SD) 32.2 ± 4.7 cm apart (standard for Tree I.V.), whereas in experiment 2, the spacing between injections was reduced to 8.9 ± 0.50 cm. This analysis suggests that decreasing the distance between injection sites to approximately 10 cm around the tree circumference may be sufficient for complete lateral distribution of chemistry injected into the tree bole.
MPB Success and Blue-Stain
In spite of the increased concentration of residues and better coverage in the 2009 treatments (experiment 2), 7 of 10 treated trees died. Our sampling procedures and whole-tree autopsies showed that treated tree boles were not successfully attacked by MPB. We observed no adult gallery construction beyond the area of attack and no reproduction (Table 3). Trees were, however, colonized by blue-staining fungi. Injection procedures used in experiment 2 were apparently effective for delivering lethal concentrations of EB to MPB adults across more phloem area than in our first experiment, a result that suggests something other than girdling from adult beetle gallery construction caused tree mortality. The number of MPB attacks was not reduced, yet crowns of our injected trees faded more slowly (Table 3), further suggesting that mortality was not due to insect attack per se.
Initially, we had not identified the agent of mortality in experiment 2. However, many associates are introduced by Dendroctonus beetles (e.g., Hofstetter 2011; Klepzig and Hofstetter 2011). Fungal symbionts have been shown to cause tree mortality by themselves (Mathre 1964; Basham 1970; Shrimpton 1973; Strobel and Sugawara 1986; Yamaoka et al. 1995). Blue-staining fungi effectively colonized the xylem, cambium, and phloem tissues of experimental trees attacked by MPB. In our lab, we identified two isolates morphologically from experiment 2 as Grosmannia clavigera (Rob.-Jeffr. & R.W. Davidson) Zipfel, Z.W. de Beer, & M.J. Wingf. (Ophiostomatales, Ascomycota) and Leptographium longiclavatum (S.W. Lee, J.J. Kim, & C. Breuil)(Ophiostomatales, Ascomycota). These isolates were subsequently confirmed by DNA sequencing conducted by Drs. Wingfield and de Beer, now maintained in the culture collection (CMW) of FABI under Grosmannia clavigera #38988 and Leptographium longiclavatum #38989. In a greenhouse study, these isolates, when inoculated into western white pine seedlings, resulted in plant mortality independent of the presence of MPB (Wyka et al. 2016), demonstrating their potential for pathogenicity.
Fungal Infection and Tree Interaction
In transverse section, the colonization of the ophiostomatoid fungi appears blue; however, when viewed in tangential aspect (bark removed) the sapwood appears streaked and dark (brownish-purple or black) in color, indicative of necrosis of the lateral vascular tissues (Figures 3 and 4). The discoloration of xylem tissue at the point of MPB attack is shown in Figure 4. Pinus species have constitutive defensive chemistries (terpenes, resins, and phenols) that are induced to greater concentrations in response to injury and infection (e.g., Franceschi et al. 2005). Affected areas darken with oxidation (Shigo 1991) and may appear streaked with oxidized phenols (Shigo 1989). These areas are clearly distinguishable from areas without attacks. The percentage of discolored xylem at 3 m was consistently 100%, whereas at 12 m, the mean percentage (± 1 SEM) of discolored xylem was 36.7% ± 8.9%. The difference in pattern of colonization and discoloration of the vascular tissues with height may have been due to higher MPB attack densities and thus greater inoculum loads lower in the tree where diameters and bark roughness are greater (Shepherd 1965). Dead trees had on average 98.2% ± 1.8% (SE) of their circumference bluestained compared to 16.8% ± 5.3% SE of live trees (in experiment 3). Differential field autopsies of felled trees compared strip attacks with height and circumference of bole discoloration in experiment 3 by treatment. Though not statistically significant, the untreated controls had 6.2 m2 ± 1.4 SE of tangential sapwood surface discolored compared to 4.9 m2 ± 0.6 SE and 3.0 m2 ± 0.9 SE of TREE-äge/PPZ and TREEäge plus V/GLV, respectively. However, the extent of circumferential colonization of phloem by these fungi (as reflected by brown-purple necrotic lesioning) may have played a role in reduced tree survival. The mean square meter area of live and dead (3.2 ± 0.6 SE and 6.4 ± 0.9 SE, respectively) phloem was significant. Data of autopsied trees in each treatment are presented in Table 4.
Brownish necrotic lesions in sapwood (bark excised in tangential section) associated with ophiostomatoid infection above and below the mountain pine beetle attack points. Illustrated are two individual attacks by MPB in western white pine. Mountain pine beetles introduce fungal spores when they mine into the tree’s vascular tissues.
Transverse sections of western white pine attacked by mountain pine beetle (MPB), where (a) one-third of the sapwood column is discolored by blue-stain, typical of autopsied live trees, and (b) the entire sapwood column is discolored by blue-stain, typical of autopsied faded or dead trees. Two isolates obtained from autopsied trees in experiment 2 were subsequently identified as Grosmannia clavigera (Rob.-Jeffr. & R.W. Davidson) Zipfel, Z.W. de Beer, & M.J. Wingf. (Ophiostomatales, Ascomycota) and Leptographium longi-clavatum (S.W. Lee, J.J. Kim, & C. Breuil)(Ophiostomatales, Ascomycota), both of which are pathogenic to western white pine.
Summarized data of felled western white pine in experiment 3 as of 21 October 2015, where 7 of 10 trees per treatment were autopsied. The percent of bole height to percent circumference attacked by MPB and stained phloem was calculated. The controls had the greatest staining with a mean of 55.6% of tree height, which differed significantly from the TREE-äge/V/GLV trees with only 21.2% of bole height stained. The TREE-äge/PPZ-treated trees were intermediate, with 38.3% of bole height stained. Likewise, the controls had the greatest mean square meter of phloem blue-stained, followed by TREE-äge/PPZ and TREE-äge/V/GLV, but these were not significant. Of the dead (n = 10) trees, 49.9% of bole height was blue-stained, which was significantly different from the live (n = 11) trees with 27.9% of bole height stained. Furthermore, mean square meter area of live and dead trees (3.2 ± 1.9 SD and 6.4 ± 2.9 SD, respectively) was statistically significant. Grouping information used the Tukey method and at the 95% confidence level.
CONCLUSIONS
In addition to the increased tree mortality observed over time, observations of Grosman et al. (2009) agree with those made in this study. Attacking beetles produced short galleries, did not reproduce, and colonization of associated blue-staining fungi appeared similar to untreated trees at 5 m. Crown fade was also slower in treated trees than in control trees in both studies.
In our study, movement of AI into phloem tissue was confirmed with both biological and chemical evaluations. Pressure by MPB was high, providing an excellent opportunity to examine phloem for areas with biologically active or inactive concentrations of EB. In experiment 2, we did not observe signs of successful MPB reproduction. Residue concentrations were high, and adult MPB galleries were mostly absent except for initial attacks. This observation suggests that the injection procedures used in 2009 and in 2013 may be useful for assessing the relative contributions of different organisms to the death of pines attacked by MPB.
High levels of attack pressure are the most likely conditions to require management interventions with insecticides and they are also the conditions during which efficacy is least likely to be attained. This is particularly noteworthy for systemics, because each attacking insect must contact some amount of phloem tissue, and usually ingest it, for insecticides to be debilitative or lethal.
We have determined from isolates of trees autopsied in experiment 2 that two species of fungi, G. clavigera and L. longiclavatum, can negatively impact tree health, and our observations and those of Grosman et al. (2009) suggest that bark beetle associated fungi successfully colonize trees treated with TREEäge even without beetle success. Whether or not combination treatments can be developed to address these factors (e.g., TREE-äge injection plus a fungicide like propiconazole to reduce fungal establishment) was the focus of experiment 3. Increasing treatment effectiveness was dependent upon identifying the organisms that contribute to tree death (bark beetles and fungi), developing treatments that impact them, and on improving our knowledge of the relationship between treatment efficacy and quantifiable beetle pressure.
In experiment 3 (evaluated 21 October 2015), 2 years and 4 months after treatment, none of the untreated controls survived, whereas 80% of the TREE-äge- and GLV/verbenone-treated trees and the TREE-äge/PPZ-treated trees remained alive (Table 5). For each assessment period (August 2014, October 2014, and October 2015), differences were highly significant among controls, TREE-äge plus PPZ, or V/GLV (P = 1.83E-07, 8.67E-08, and 1.99E-06, respectively). Comparisons of the TREE-äge treatments with PPZ or V/GLV, however, were not statistically different for each of the three evaluation periods (P = 0.5000, 0.2368, and 0.2090, respectively). Treatments using TREE-äge with the systemic fungicide propiconazole or with an anti-aggregation semiochemical satisfied the requirements for efficacy used in this study (Table 5, Figure 5).
In experiment 3, trees were treated on 11 June 2013. The treatments included TREE-äge plus propiconazole (PPZ), TREE-äge with verbenone and green leaf volatile (V/GLV) to deter MPB attack, and untreated controls. Change in tree condition and percent tree mortality were observed over time in this experiment. Evaluations of trees included rating the canopy qualitatively as green (live), fading (stressed or dying), or faded (dead). Evaluations were conducted at 14 months (on 19 August 2014), at 16 months (on 15 October 2014), and at 28 months after treatment (on 21 October 2015). Final evaluations were conducted on 21 October 2015.
Crown fade of western white pines in experiment 3 at the final evaluation period. Fade categories were the same as in the previous experiments and consisted of green crown (live), fading (stressed or dying), and faded (dead). TREE-äge and propiconazole (EB/PPZ) or verbenone/green leaf volatile (EB/V/GLV) were applied on 11 June 2013; the final crown evaluations were made on 21 October 2015, where no more than 20% of the treated trees failed compared to 100% of the controls, meeting the requirements for efficacy as defined by Shea et al. (1984).
ACKNOWLEDGMENTS
We thank the Warner Mountains Ranger District, Modoc National Forest, especially Anne Mileck, whose cooperation was essential for the execution of the study. We thank Ben Parpart, James Parpart, and Steven Walters (Southern Research Station, Pineville, LA), and Erin Frolli, Rueben Mahnke, Danny Cluck, and Sherry Muse (Forest Health Protection, Susanville, CA) for technical assistance. We also thank Stephen Wyka (Arborjet, Inc., R&D) for technical assistance in tree autopsies, and Cavell Brownie, professor emeritus, NCSU, Raleigh for providing guidance on the statistical analyses. A previous version of the manuscript was reviewed by Don Grosman (Arborjet, Inc.) and David Cox (Syngenta, Inc., Fresno, CA). Funding for this study was provided by USDA Forest Service, R5 Forest Health Protection, Vallejo, CA, the Southern Research Station, Pineville, LA, and Arborjet, Inc., Woburn, MA. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the United States Government.
Footnotes
Conflicts of Interest:
Joseph J. Doccola reports being the Director of Research & Development for Arborjet, Inc., Woburn, MA, which distributes TREE-äge® and Propizol. The remaining authors made no disclosures.
- © 2020, International Society of Arboriculture. All rights reserved.
LITERATURE CITED
Jump to section
Related Articles
Cited By...
- No citing articles found.