TY - JOUR
AU - Humble, Leland, M
AB - Abstract Addressing the risk from pests present in wood and wood products destined for international trade is an essential step towards minimizing the movement, introduction and establishment of invasive species. One method of managing the pest risk associated with wood commodities is the use of a systems approach that incorporates multiple independent measures applied along a production pathway. However, quantifying the reduction of risk can be difficult because the approach requires raw material infested with the pest of interest at a sufficient density to be able to quantify changes in pest abundance. We tested a systems approach for the production of sawn wood using green ash, Fraxinus pennsylvanica Marshall (Lamiales: Oleaceae), infested with emerald ash borer, Agrilus planipennis Fairmaire (Coleoptera: Buprestidae), by quantifying the change in pest density during the milling process and the precise effect of heat treatment on insects in situ. Greater than 90% of emerald ash borer were removed at the first step of the milling process (debarking) and >99% were removed before the production of green sawn wood. No insects survived kilning or heat treatment. All life stages of emerald ash borer were killed at 56°C and above. Heat, however, had no sublethal effect on emerald ash borer performance. These results show that the application of a systems approach to mitigate emerald ash borer in heat-treated, sawn wood is effective. Moreover, the model-system approach developed in this study can be a template for investigating the effect of systems approaches for other phloem-feeding insects. emerald ash borer, phytosanitary, systems approach, heat treatment Addressing the risk associated with pests present in wood and wood products destined for international trade is an essential step towards minimizing the movement, introduction, and establishment of invasive species. Manufacturers can use a variety of techniques to minimize the risk from pests, but heat treatment and fumigation are the most common single-treatment measures used to meet trade requirements for wood products (Allen et al. 2017). These techniques are developed on a pest- or commodity-specific basis under laboratory conditions by determining the combination of time and temperature, or time and concentration, that will kill the target organism thus rendering the commodity ‘safe’ (Mushrow et al. 2004, Haack and Petrice 2009, Myers and Bailey 2011, Yokoyama 2011, Mayfield et al. 2014, e.g., Chen et al. 2016, Pawson et al. 2019, Wang et al. 2019). For heat treatment the time-temperature combination that achieves a minimum temperature of 56°C for a minimum duration of 30 consecutive minutes (‘56/30’) throughout the entire profile of the wood (IPPC 2019) is accepted by the International Plant Protection Convention (IPPC) in the International Standards for Phytosanitary Measures No. 15 (ISPM 15). Eighty countries also recognize the 56/30 standard in their own phytosanitary regulations (IPPC 2019). While 56/30 has been implemented to limit the spread of invasive wood-boring species associated with wood packaging material (Haack et al. 2014), the ‘systems approach’ concept that integrates multiple phytosanitary measures along the production and shipping paths of a commodity has been recognized as an alternative strategy (Allen et al. 2017, NAPPO 2018). A systems approach minimizes phytosanitary risk through a combination of measures which address one or more pest risks associated with a commodity. A systems approach is defined as: a pest risk management option that integrates different measures, at least two of which act independently, with cumulative effect (FAO 2019). The development of a systems approach involves understanding the nature of the pest risk, the commodity production pathway, where and when measures can be applied and an assessment of both the feasibility and impact of a proposed system (NAPPO 2018). Guidance on the use of integrated pest risk management to mitigate risks associated with the movement of wood and wood commodities in North America is given in the Regional Standards for Phytosanitary Measures 41 (RSPM 41) (NAPPO 2018). RSPM 41 provides a framework for National Plant Protection Organizations within which they may negotiate the details of systems approaches for specific commodities. These approaches may include, among other things, verification and methods for quantifying risk. Quantifying the change in risk of each element of a systems approach is integral to demonstrating that an approach is effective. Thus, to assess the effectiveness of a systems approach, we must obtain data about how each step of the production process affects risk. For example, how the abundance of a potentially invasive pest is changed as the raw material it infests is transformed into a commodity. With this information, we can measure the change in risk that occurs during production, and determine which steps in the production pathway should be incorporated into a systems approach. Obtaining the data needed to quantify a change in risk can be difficult. In the specific case of the production of sawn wood (i.e., lumber), a pest must infest the raw material (i.e., a tree) at high densities so that any reduction in density resulting from the production process can be detected and quantified. Moreover, the investigator needs to find a sufficient number of samples all with similar high initial pest densities so as to permit an estimation of variability. Finally, there must be a production facility (i.e., a mill) available that can convert the raw material to the finished product. Meeting the first two criteria can be difficult because most species exist at low densities in their native range making homogeneous, high-density populations of potentially invasive insects hard to find. The only exception to this being, perhaps, those species that occasionally experience significant outbreaks (e.g., bark beetles), but these are often short-lived. Meeting the final criterion can only occur when the infested material is located near-enough to a mill to make the production of the commodity economically-viable. This means testing a systems approach for a generalized commodity requires a unique combination of a pest insect experiencing an outbreak on a high-value host growing near a manufacturing facility. The emerald ash borer, Agrilus planipennis Fairmaire (Coleoptera: Buprestidae), is a beetle native to Asia introduced to North America sometime in the 1990s (Siegert et al. 2014). Since that time, the insect has spread throughout most of eastern North America and as far west as Colorado (USDA APHIS 2019). In Canada, the insect has a more-or-less contiguous distribution through southern Ontario and Quebec, with satellite infestations in northern Ontario, Manitoba, New Brunswick, and Nova Scotia (CFIA 2019b). A typical infestation begins in the crown of the tree, progressing downwards over a period of 2–5 yr, or more. The larvae of the insect feed on the phloem layer of the tree, which cuts off the flow of resources to the upper branches, eventually strangling and killing the tree. The result of this invasion is that within its contiguous range emerald ash borer has killed tens of millions of ash trees, with a typical infested green ash Fraxinus pennsylvanica Marshall (Lamiales: Oleaceae) tree able to produce 89 beetles per m2 of bark surface area over the course of an infestation (McCullough and Siegert 2007). We used green ash trees infested by emerald ash borer to quantify the effect of both milling and heat treatment on risk reduction in sawn wood. This involved two separate experiments. In the first experiment, we evaluated the effect of milling on emerald ash borer incidence by simulating the production of ash sawn wood at a commercial sawmill. In the second experiment, we tested the effect of heat treatment on both survival and sublethal effects of emerald ash borer larvae in situ using ash slabs sectioned from infested ash trees. We show that almost all emerald ash borer are removed during the initial stages of the milling process and confirm that heat treatment is effective at killing larvae and pupae of the insect in wood. However, we also show that emerald ash borer mortality occurs at lower temperatures than previously reported. Methods Trees All trees for both studies were harvested from ash woodlots located near Tillsonburg, Ontario, Canada (42.86, −80.73). The trees were located and inspected in the fall to determine whether they displayed symptoms that suggested they were infested by emerald ash borer (e.g., thinning crowns, presence of exit holes, woodpecker feeding damage). The trees were harvested in the following winter using chainsaws and marked with a unique number. In 2016, five trees were harvested for use in the heat treatment study. In 2017, we selected and cut 50 trees for use in the milling study and the heat treatment study; in 2018, another 50 trees were cut for the milling study only. We recorded the diameter at breast height (dbh; ~130 cm above ground) for all trees harvested in 2016 and 2017. Effect of Milling To determine the effect of milling on emerald ash borer survival, we replicated the milling process of turning raw round wood with bark into sawn wood and sequentially removed samples of each product and byproduct of milling from a sawmill production line (Supp Fig. 1 [online only]). These samples were round wood with bark (i.e., the controls), debarked round wood, slabs with bark (hereafter ‘slabs’), cants, edges, and sawn wood. Round wood, debarked wood, and sawn wood were defined as in FAO (2019). Slabs are the byproduct created when the natural rounded surface of the round wood is removed; cants are the roughly square pieces of sawn wood created when the slabs are removed from the round wood; edges are the byproduct created when the round or imperfect outside sections that are removed in the production of sawn wood of a desired width or quality. We also subjected some sawn wood to heat treatment or heat treatment followed by moisture reduction (i.e., kiln drying). We did not retain the bark. We subsequently refer to each of these products and byproducts as the experimental treatment groups. After harvesting each tree, a 2.4 m (8 foot) section of round wood was identified by the harvester and removed. The harvester was instructed by the investigators to select the ‘best’ 2.4 m section of each tree, with ‘best’ defined as what would typically be selected by the harvester for delivery to the sawmill for the production of ash sawn wood. The result was the experimental sections of round wood were usually selected from the mid to upper bole of the tree and were generally straight with few lateral deflections. The pieces of round wood were then transported to the edge of the stand and then trucked to a sawmill (Townsend Lumber, Tillsonburg, Ontario, Canada) where they were stored prior to milling (4 or 5 wk). The debarked round wood was created using a Rosser head debarker and all other milling was done with an automated saw operated by an experienced sawyer. We instructed the sawyer to mill the round wood samples that we provided in the same manner and to the same standards as they would regular production sawn wood destined for commercial sale. In both years, samples of round wood from 7 or 8 individual trees were assigned to each of the six treatment groups (Table 1; Supp Fig. S1 [online only]). All subsequent samples that were created were labeled with the tree number as they were produced. The assignment of round wood from each tree to the treatments was haphazard but was contingent on the order in which the round wood entered the sawmill. That is, the first round wood placed on the production line was debarked, the next had their slabs removed and the cants were retained, and the following 24 pieces of round wood were turned into green sawn wood from which the slabs, sawn wood, and edges were retained. Table 1. Summary statistics for trees and samples in the milling experiment Year . Milling treatment . ntrees . Total surface Area, m2 . Surface area, m2 (mean ± 1 SD) . Total volume, m3 . Volume, m3 (mean ± 1 SD) . Bark remaining, % (mean ± 1 SD) . 2017 Round wood 8 16.93 2.12 ± 0.13 1.19 0.15 ± 0.02 90.64 ± 24.42 Debarked wood 8 17.01 2.13 ± 0.32 1.22 0.15 ± 0.05 21.04 ± 11.23 Slabs 32 147.99 4.62 ± 1.28 11.23 0.35 ± 0.13 11.75 ± 12.69 Edges 23 42.56 1.85 ± 0.77 0.34 0.01 ± 0.01 7.91 ± 6.03 Cants 8 33.24 4.15 ± 1.56 1.14 0.14 ± 0.05 3.13 ± 3.97 Green sawn wood 8 35.67 4.46 ± 2 0.44 0.06 ± 0.02 0 Heat-treated sawn wood 7 45.24 6.46 ± 1.6 0.53 0.08 ± 0.02 0 Heat treated and kiln-dried sawn wood 7 36.52 5.22 ± 1.97 0.42 0.06 ± 0.02 0 2018 Round wood 8 24.61 3.08 ± 0.29 2.48 0.31 ± 0.04 87.41 ± 23.31 Debarked wood 8 26.70 3.34 ± 1.05 2.71 0.34 ± 0.17 4.86 ± 5.37 Slabs 15 85.08 5.67 ± 1.67 7.20 0.48 ± 0.18 1.83 ± 1.29 Edges 23 91.39 3.97 ± 2.24 0.77 0.03 ± 0.02 1.61 ± 1.44 Cants 9 58.38 6.49 ± 1.61 1.53 0.17 ± 0.08 0.47 ± 0.36 Green sawn wood 7 96.05 13.72 ± 9.38 1.74 0.25 ± 0.29 0.27 ± 0.34 Heat-treated sawn wood 8 87.13 10.89 ± 5.41 1.03 0.13 ± 0.07 0.11 ± 0.16 Heat-treated and kiln-dried sawn wood 8 89.15 11.14 ± 2.58 1.05 0.13 ± 0.03 0.22 ± 0.32 Year . Milling treatment . ntrees . Total surface Area, m2 . Surface area, m2 (mean ± 1 SD) . Total volume, m3 . Volume, m3 (mean ± 1 SD) . Bark remaining, % (mean ± 1 SD) . 2017 Round wood 8 16.93 2.12 ± 0.13 1.19 0.15 ± 0.02 90.64 ± 24.42 Debarked wood 8 17.01 2.13 ± 0.32 1.22 0.15 ± 0.05 21.04 ± 11.23 Slabs 32 147.99 4.62 ± 1.28 11.23 0.35 ± 0.13 11.75 ± 12.69 Edges 23 42.56 1.85 ± 0.77 0.34 0.01 ± 0.01 7.91 ± 6.03 Cants 8 33.24 4.15 ± 1.56 1.14 0.14 ± 0.05 3.13 ± 3.97 Green sawn wood 8 35.67 4.46 ± 2 0.44 0.06 ± 0.02 0 Heat-treated sawn wood 7 45.24 6.46 ± 1.6 0.53 0.08 ± 0.02 0 Heat treated and kiln-dried sawn wood 7 36.52 5.22 ± 1.97 0.42 0.06 ± 0.02 0 2018 Round wood 8 24.61 3.08 ± 0.29 2.48 0.31 ± 0.04 87.41 ± 23.31 Debarked wood 8 26.70 3.34 ± 1.05 2.71 0.34 ± 0.17 4.86 ± 5.37 Slabs 15 85.08 5.67 ± 1.67 7.20 0.48 ± 0.18 1.83 ± 1.29 Edges 23 91.39 3.97 ± 2.24 0.77 0.03 ± 0.02 1.61 ± 1.44 Cants 9 58.38 6.49 ± 1.61 1.53 0.17 ± 0.08 0.47 ± 0.36 Green sawn wood 7 96.05 13.72 ± 9.38 1.74 0.25 ± 0.29 0.27 ± 0.34 Heat-treated sawn wood 8 87.13 10.89 ± 5.41 1.03 0.13 ± 0.07 0.11 ± 0.16 Heat-treated and kiln-dried sawn wood 8 89.15 11.14 ± 2.58 1.05 0.13 ± 0.03 0.22 ± 0.32 Open in new tab Table 1. Summary statistics for trees and samples in the milling experiment Year . Milling treatment . ntrees . Total surface Area, m2 . Surface area, m2 (mean ± 1 SD) . Total volume, m3 . Volume, m3 (mean ± 1 SD) . Bark remaining, % (mean ± 1 SD) . 2017 Round wood 8 16.93 2.12 ± 0.13 1.19 0.15 ± 0.02 90.64 ± 24.42 Debarked wood 8 17.01 2.13 ± 0.32 1.22 0.15 ± 0.05 21.04 ± 11.23 Slabs 32 147.99 4.62 ± 1.28 11.23 0.35 ± 0.13 11.75 ± 12.69 Edges 23 42.56 1.85 ± 0.77 0.34 0.01 ± 0.01 7.91 ± 6.03 Cants 8 33.24 4.15 ± 1.56 1.14 0.14 ± 0.05 3.13 ± 3.97 Green sawn wood 8 35.67 4.46 ± 2 0.44 0.06 ± 0.02 0 Heat-treated sawn wood 7 45.24 6.46 ± 1.6 0.53 0.08 ± 0.02 0 Heat treated and kiln-dried sawn wood 7 36.52 5.22 ± 1.97 0.42 0.06 ± 0.02 0 2018 Round wood 8 24.61 3.08 ± 0.29 2.48 0.31 ± 0.04 87.41 ± 23.31 Debarked wood 8 26.70 3.34 ± 1.05 2.71 0.34 ± 0.17 4.86 ± 5.37 Slabs 15 85.08 5.67 ± 1.67 7.20 0.48 ± 0.18 1.83 ± 1.29 Edges 23 91.39 3.97 ± 2.24 0.77 0.03 ± 0.02 1.61 ± 1.44 Cants 9 58.38 6.49 ± 1.61 1.53 0.17 ± 0.08 0.47 ± 0.36 Green sawn wood 7 96.05 13.72 ± 9.38 1.74 0.25 ± 0.29 0.27 ± 0.34 Heat-treated sawn wood 8 87.13 10.89 ± 5.41 1.03 0.13 ± 0.07 0.11 ± 0.16 Heat-treated and kiln-dried sawn wood 8 89.15 11.14 ± 2.58 1.05 0.13 ± 0.03 0.22 ± 0.32 Year . Milling treatment . ntrees . Total surface Area, m2 . Surface area, m2 (mean ± 1 SD) . Total volume, m3 . Volume, m3 (mean ± 1 SD) . Bark remaining, % (mean ± 1 SD) . 2017 Round wood 8 16.93 2.12 ± 0.13 1.19 0.15 ± 0.02 90.64 ± 24.42 Debarked wood 8 17.01 2.13 ± 0.32 1.22 0.15 ± 0.05 21.04 ± 11.23 Slabs 32 147.99 4.62 ± 1.28 11.23 0.35 ± 0.13 11.75 ± 12.69 Edges 23 42.56 1.85 ± 0.77 0.34 0.01 ± 0.01 7.91 ± 6.03 Cants 8 33.24 4.15 ± 1.56 1.14 0.14 ± 0.05 3.13 ± 3.97 Green sawn wood 8 35.67 4.46 ± 2 0.44 0.06 ± 0.02 0 Heat-treated sawn wood 7 45.24 6.46 ± 1.6 0.53 0.08 ± 0.02 0 Heat treated and kiln-dried sawn wood 7 36.52 5.22 ± 1.97 0.42 0.06 ± 0.02 0 2018 Round wood 8 24.61 3.08 ± 0.29 2.48 0.31 ± 0.04 87.41 ± 23.31 Debarked wood 8 26.70 3.34 ± 1.05 2.71 0.34 ± 0.17 4.86 ± 5.37 Slabs 15 85.08 5.67 ± 1.67 7.20 0.48 ± 0.18 1.83 ± 1.29 Edges 23 91.39 3.97 ± 2.24 0.77 0.03 ± 0.02 1.61 ± 1.44 Cants 9 58.38 6.49 ± 1.61 1.53 0.17 ± 0.08 0.47 ± 0.36 Green sawn wood 7 96.05 13.72 ± 9.38 1.74 0.25 ± 0.29 0.27 ± 0.34 Heat-treated sawn wood 8 87.13 10.89 ± 5.41 1.03 0.13 ± 0.07 0.11 ± 0.16 Heat-treated and kiln-dried sawn wood 8 89.15 11.14 ± 2.58 1.05 0.13 ± 0.03 0.22 ± 0.32 Open in new tab The end product of the milling process was green sawn wood from 23 or 24 of the initial emerald ash borer-infested round wood samples that entered the sawmill. The sawn wood from 7 (2017) or 8 (2018) of these round wood samples was collected ‘green’ (i.e., not subject to drying or moisture reduction). We refer to the sawn wood that was not subject to heat treatment as ‘green sawn wood’. The sawn wood from the remaining 14 (2017) or 16 (2018) samples were then further subjected to heat treatment to 56°C for 30 min or heat treatment followed by moisture reduction. All treatments were applied using the commercial kilns at Townsend Lumber. Following milling and heat treatment, all the products and byproducts were transported to a secure rearing site. In 2017, the rearing site was a large shed located on the grounds of the Ontario Ministry of Natural Resources and Forestry Seed Plant in Angus, Ontario. In 2018 the rearing site consisted of two large tents set up on the grounds of the Great Lakes Forestry Centre (GLFC) in Sault Ste. Marie, Ontario, Canada. In both years, the products and byproducts that were held out-of-doors were placed in cardboard drums (36 cm diameter × 104 cm tall; Timms et al. 2006). Products and byproducts that were too long to fit into a drum were cut to <104 cm in length, and all pieces were placed in the same drum. In 2017, the products and byproducts were placed into the drums in early February and March and remained there until mid-August, with emergence beginning around mid-June and continuing until approximately mid-July. In 2018, the products and byproducts were stored in a sea container on the grounds of the GLFC (Fick and MacQuarrie 2018) until early May, then transferred to the drums. Emergence began in mid-June and continued until mid-July. Emerald ash borer that were in the products and byproducts emerged and were contained in the sealed drums. After emergence had completed the drums were opened and all the insects were collected. In 2017 and 2018, some of the control and debarked round wood were too large to fit into the rearing drums (i.e., >30 cm diam.). These large pieces of wood were instead placed into plastic rearing boxes (76 × 46 × 37 cm; Rubbermaid, Newell Brands Inc., Atlanta, GA) or cardboard tubes capped at either end with ridged plastic caps (76 cm diameter × 122 cm tall, Sonotube, Sonoco, Hartsville, SC). These large pieces of wood were set up indoors in rearing rooms at the GLFC (Fick and MacQuarrie 2018). As with the products and byproducts set up out of doors, all emerald ash borer that emerged were contained in the tubes or rearing boxes and when emergence had completed the containers were opened and the insects collected. After adult emergence was completed, we dissected all products and byproducts to remove any remaining emerald ash borer life stages that had not emerged. Dissection involved the removal of all remaining bark and phloem tissue, and inspection of every piece for emerald ash borer (e.g., within pupal chambers). The insects we found were either adults that failed to emerge from the sample, pupae that failed to complete development to adulthood, prepupae, or larvae (i.e., were in a 2-yr life cycle; Cappaert et al. 2005). Prior to dissection, we measured all the ash products and byproducts in order to determine their volume and surface area and estimated the percentage of the sample that still retained any phloem tissue or bark (Supp Fig. S1 [online only]). For all the insects that were extracted from the samples, we first determined if they were emerald ash borer. If so, the developmental stage was determined by either inspection (for adults, pupae, and prepupae) or by measuring the width of the head capsule (for larvae). Analysis We fit a generalized linear model (GLM) with a Poisson distribution to test for the effect of milling treatment on the number of emerald ash borer that emerged from the samples. The experimental unit for this experiment was considered to be the original 2.4 m sample of round wood cut from the infested ash tree, so for this analysis, we expressed emergence on a per-sample basis. That is, the total number of insects that emerged from all products and byproducts created from each original sample of round wood. We tested the GLM for issues with overdispersion and heterogeneity of variance. Finding both issues to exist, we fit a negative binomial distribution to the data, which was successful in accounting for the overdispersion and heterogeneity. We then tested for the statistical significance of the main effect via analysis of variation (ANOVA) and when we detected a significant main effect we tested the separation of means using Tukey’s Honestly Significant Difference (HSD) test (Tukey 1949). Effect of Heat Treatment Trees that had been identified in the falls of 2015 and 2016 were harvested in winters of 2016 and 2017; a 1 m section of round wood was removed from each harvested tree and then transported to the GLFC. In 2016, these pieces of round wood were cut from some of the same trees that were harvested for use in the milling experiment. In February of 2016 and 2017, the 1 m pieces of round wood were placed on a portable sawmill (Board Master, Bonter Midway Manufacturing Inc., Marmora, Ontario, Canada) and one, 6.35 cm thick (2.5 inch) slab (as defined above) was removed from each of four sides. Each slab was marked with a unique identification number. In 2016, we also noted the orientation of each slab (i.e., the cardinal direction the bark was facing before the tree was harvested) to determine whether aspect had an effect on total emerald ash borer abundance (Sensu Timms et al. 2006). Aspect did not affect total emerald ash borer abundance in the 2016 trees and so we did not record orientation in 2017. After being produced, the slabs were placed in sealed cardboard tubes (as above) and then transported to Québec, Quebec for heat treatment. All slabs were heat-treated to one of six target temperatures in a kiln at the FPInnovations research laboratory in Québec, Quebec, Canada. The kiln was a 2.4 m (8 foot) track kiln (Cathild Industrie SA, Mansigné, France) with indirect heating that had been retrofitted with a Séchoir MEC kiln controller (MEC, Victoriaville, Quebec, Canada). This particular kiln was designed to develop kiln schedules for wood and wood products and was capable of maintaining precise temperatures for extended periods. Humidity in the kiln was maintained at 75% RH by steam injection and atmospheric venting. In 2016, we treated the slabs to 48, 56, and 71°C. In 2017, we treated the slabs to 50, 52, 54, and 56°C. We selected the target 2016 temperatures to bracket previously identified temperatures that killed emerald ash borer (McCullough et al. 2007, Nzokou et al. 2008, Myers et al. 2009, Goebel et al. 2010, Sobek et al. 2011). We selected the 2017 temperatures to bracket those temperatures where emergence did and did not occur from the slabs treated in 2016. Prior to being placed in the kiln, all the slabs were held at ambient, indoor conditions (18–20°C) for at least 1 h. Then, each slab was fitted with a type ‘T’ thermocouple (range 0–350°C; Special limits of error ± 0.5°C or 0.4%; part FF-T-24-SLE, Omega Engineering Inc., Stamford, CT) to measure its inner temperature during treatment. The thermocouple was inserted into the flat side of the slab in a hole drilled to a depth equal to 2/3 of the slab’s maximum thickness (6.4 ± 0.6 cm; mean ± 1 SD) placed along the midline at a point equal to twice the thickness of the slab, as measured from one end. A bead of silicone was then applied to hold the thermocouple in place. Both the optimal depth and location of the thermocouple were determined in a series of test heat treatments done with a set of extra ash slabs created for this purpose (Supp Fig. S2 [online only]). Once fitted with a thermocouple, the slab was then placed into the kiln and heated to a target core temperature (i.e., 48–71°C, as above). The thermocouple was connected to a data logger (Digi-Sense 12 Channel Scanning Benchtop Thermometer, model 92000-00, Barnant Company, Division of Cole-Parmer Instrument Company, Montreal, Quebec, Canada) which was connected to a computer running ScanLink (Barant Company 2001) which provided real-time reporting of each slab’s core temperature. Slabs were randomly assigned to a target temperature, but we took care to ensure that no more than two slabs from the same tree were assigned to any given temperature. When the core temperature for a given slab reached the target temperature, the slab was then held at that temperature for a further 30 min. After 30 min, we removed the slab from the kiln, removed the thermocouple, and allowed the slab to cool to room temperature. In both years, we created a set of control slabs that were not fitted with a thermocouple and not heated but were otherwise treated in the same manner as the test slabs. Following the heat treatments, the slabs were again put in the sealed drums and transported to the GLFC. At the GLFC all slabs were placed in a 4°C cooler for 60–90 d and then placed individually into 19.1–31.8 cm diameter (7.5–12.5 inch) by 61.0–81.0 cm tall cardboard tubes (Sonotube, Sonoco, Hartsville, South Carolina) to allow any surviving emerald ash borer to complete pupation and emerge as adults. The cardboard tubes were sealed at the bottom with a rigid cap, but were open at the top and covered with mesh screen (1 × 1 mm) held in place with a large rubber band. The cardboard tubes containing the treated ash slabs were checked every 24–48 h and all the newly emerged emerald ash borer adults were removed. In 2016, a small number of individuals escaped from the rearing drums holding the control and 48°C treatment slabs when the rubber bands were improperly reinstalled following the daily inspections (30 insects or 5% of the total insects collected). These insects were retained but not used in our analyses (MacQuarrie et al. 2019). All the other live adults were sexed and retained for the following experiments. Sublethal Effects of Heat Treatment Following emergence, all adult emerald ash borer were reared to determine whether there were sublethal effects on adult performance. We used three metrics of performance—female viability, adult longevity, and female fecundity. After emergence, each adult was placed into a plastic rearing cup provisioned with ash foliage and a water source then held in a climate-controlled rearing chamber (16:8 (L:D) h, 60–80% RH; MacQuarrie 2019). The ash foliage in all rearing experiments was Fraxinus udhei (Wenz.) Lingelsh. from a greenhouse-raised source maintained at GLFC. Up to eight beetles were placed in each rearing cup with insects from the same heat treatment and slab held together in the same cup, but with males and females housed in separate cups. The cups were further segregated so that all individuals in a cup were the same post-emergence age ± 1 d. This method ensured that all the adults used in the subsequent experiments were the same age. Males and females were kept separate for 14 d to allow them to complete maturation feeding before being transferred to one of the following experiments. Any individuals that died during the 14-d-period were retained. Positive Test for Female Viability (Single-Pair Mating) We conducted a positive test for fecundity by mating adult female emerald ash borer that emerged from the heat-treated slabs to males that had experienced no heat treatment. These emerald ash borer males were obtained from infested ash bolts that had not been subject to any of the experimental manipulations To obtain these males, we collected infested ash bolts from emerald ash borer-infested stands in southern Ontario, Canada in the fall of 2015 and 2016. These bolts were transported to the GLFC, stored in a sea container until spring and then in the 4°C cooler until they were needed. Previous work has shown these storage conditions have no effect on adult emerald ash borer viability (Fick and MacQuarrie 2018). A selection of bolts was then placed into sealed boxes in a rearing room (see Fick and MacQuarrie 2018 for description) at the same time the ash slabs were established in the cardboard tubes, thus emergence from the two sets of ash samples was approximately synchronous. All male emerald ash borer were retained and placed in separate rearing cups and held as above until they were needed while all female emerald ash borer that emerged were used in other experiments, or incorporated into the GLFC emerald ash borer colony (Roe et al. 2018). We then placed 20 females from each heat treatment into individual rearing cups and added a single male from the untreated ash bolts. Each cup was provisioned with ash foliage and a water source. Each cup was then sealed by placing a piece of black window screen mesh and then a paper coffee filter over the top of the cup and securing both with a rubber band. This mesh-filter combination (Duan et al. 2011) acts as an artificial oviposition substrate for the emerald ash borer female, who inserts her ovipositor between the openings in the screen and deposits her eggs on the coffee filter. The coffee filters were replaced every 48–72 h and transferred to a clear plastic box. This box was then housed in the same rearing chamber as the adult emerald ash borer to allow the eggs to mature and hatch. We then counted the total number of eggs that were laid on each coffee filter, and the total number of eggs that hatched. Each rearing cup was maintained until the female died, at which point she was collected and preserved and her total lifespan recorded. If the male died before the female, he was replaced with a similarly aged male if one was available. Within-Treatment Test of Viability (Multiple-Pair Mating) We tested the effect of heat treatment on the lifespan and fecundity of adult emerald ash borer by rearing and mating individuals from the same heat treatment. In this experiment, adult emerald ash borer were reared under the same condition as in the positive test of female viability, with the following adjustments: 1) Males and females emerged from the same heat treatment and were placed in rearing cups containing 1, 2 or 3 pairs of insects. Care was taken not to create cups with males and females that emerged from the same slab to reduce the likelihood of sibling mating. These cages were also provisioned with ash foliage, a water source, and an oviposition substrate. 2) All the pairs of insects were maintained in the cup until they died, at which point they were removed, sexed, and the day of death recorded, from which we computed lifespan. 3) All the eggs collected on the oviposition substrates were maintained until neonates emerged and then counted. However, in this experiment, these eggs were the product of all females present in the cage. Following the end of all experiments, we measured the body length, elytra length and thorax width of each individual adult emerald ash borer using an optical micrometer mounted in the eyepiece of a dissecting microscope. Analysis: We completed independent analyses to test for the effect of heat treatment temperature on 1) the number of adult emerald ash borer that emerged from each slab, 2) the number of emerald ash borer adults that failed to emerge, 3) the number of prepupae found in each slab after emergence had completed, 4) the number of larvae found in each slab after emergence had completed, and 5) the total number of emerald ash borer in each slab at the start of the experiment. For these analyses, we fit GLMs with a Poisson distribution since all the data were counts. We also nested year within treatment since different temperatures were tested in the 2 yr of the experiment. We tested the initial models for overdispersion using methods from Cameron and Trivedi (1990) and for homogeneity of variance using visual inspection of the residuals. In all analyses of the effect of heat treatment, we found evidence of overdispersion and so we either modeled the effect of dispersion using a negative bionomial distribution or, when that approach failed, we fit a hurdle model (Zeileis et al. 2008). The hurdle model was a two-part approach whereby we first tested for the presence or absence of emergence of insects from a slab using a logistic regression approach and then, for those slabs where emergence did occur, we fit a model with a Poisson distribution as above. This method thereby estimated both the probability (log-odds) that an adult emergence would occur from a slab heated to a given treatment and an estimate of the difference in overall emergence among the different treatments. Our approach was modified somewhat for the second part of the hurdle model, in that we had to limit the analysis to just those temperature treatments where adults emerged because there were no data for adults for the other treatments with which we could fit the model. When we detected a significant effect of heat treatment we compared the separation of means among the different treatment temperatures using Tukey’s HSD. When we detected a significant effect of heat treatment and year, we first re-fit the model with year as an additive effect before testing the separation of means in order to average the potential effect of year-over-year differences. To test for the effect of heat treatment on fertility and fecundity we first standardized the total number of eggs by expressing it in terms of eggs female−1 day−1. We did this to account for the fact that the number of females present in a cup changed over the course of the experiment. We tested for the effect of heat treatment, sex and year (nested within heat treatment) on lifespan by fitting linear models followed by ANOVA. Finally, we used linear models followed by ANOVA to test for the effect of heat treatment (nested within year) and year on body length, elytra length, and thorax width of adult emerald ash borer. In this analysis, we treated males and females separately, as female emerald ash borer are known to be larger than male emerald ash borer and thus allowed us to fit a simpler statistical model (Fick and MacQuarrie 2018, MacQuarrie 2019). All analyses for the milling and heat treatment experiments were done in the R statistical computing language (R Core Team 2019). We fit and tested the linear models and GLMs using function in the stats (R Core Team 2019) package and used functions in the AER (Kleiber and Zeileis 2008), MASS (Venebles and Ripley 2002), nlme (Pinheiro et al. 2018), and pscl (Jackman 2017) packages to fit and test the GLMs and hurdle models. The Tukey HSD tests were done using functions in the emmeans package (Lenth 2019). We computed the odds ratios for the presence–absence portion of the hurdle model using functions in the oddsratio package (Schratz 2017). We used a non-parametric Cox proportional hazards model to determine whether time to emergence was influenced by heat treatment (sensu Fick and MacQuarrie 2018, MacQuarrie 2019). This model was fit using procedures from Harrell (2015) and functions in the survival (Therneau 2015) and rms (Harrell 2019) packages. All data and R code for all analyses can be obtained from the Dryad digital repository (MacQuarrie et al. 2019). Results Milling Experiment The trees used in the 2018 replicate of the milling experiment were larger than those used in 2017 replicate (48.1 ± 6.2 cm dbh in 2018 vs 33.7 ± 4.0 cm dbh in 2017). We selected larger trees for use in 2018 so as to more closely mimic the size of trees typically used in the production of ash sawn wood for export. This meant that we sampled a larger amount of ash material in 2018 than in 2017 (Table 1) We collected a total of 725 (2017) and 794 (2018) live emerald ash borer adults from the treatments in the milling experiment (Fig. 1a and b) with most insects (69% in 2017, 99% in 2018) emerging from the un-milled round wood (i.e., the controls). The remaining insects emerged from the debarked round wood, slabs, and edges. In 2017, seven adult emerald ash borer emerged from cants and green sawn wood, but none emerged from the cants or green sawn wood in 2018. No emergence was observed from the heat-treated sawn wood or heat-treated and kiln-dried sawn wood (Fig. 1a and b). This pattern of emergence was reflected in the statistical analyses where we observed a significant effect of both treatment (deviance = 464.14; df = 7; dfresidual = 179; P < 0.001) and year (deviance = 44.00; df = 1; dfresidual = 178; P < 0.001) in the analysis of deviance when we examined emergence from each sample (Fig. 2). We also found that there was more emergence from the controls than from all the other milling treatments (bars in Fig. 2). In both years there was a substantial number of insects that either failed to emerge (Fig. 1a and b; gray portions of bars) in the control treatments, the majority of which were either too young to complete development (larvae) or failed to pupate (prepupae) or were unable to successfully emerge. These insects were also all removed by milling treatments, with the exception of seven prepupae that were collected from green sawn wood in 2018. Fig. 1. Open in new tabDownload slide Total emergence (black) of adult emerald ash borer from seven milling treatments and controls (round wood) in 2017 (a) and 2018 (b), and percent of insects remaining in all treatments (c). Gray shading in (b) and (c) show the counts of insects that were resident in the samples before milling but did not emerge (unemerged adults and pupae) or were in a stage that required additional development before being able to emerge (prepupae, larvae). Note the difference in the scale of the y-axes between a and b. Numbers below give the exact sample sizes for each stage (a, b) or the percentage of all insects remaining relative to the number found in the controls (c). Fig. 1. Open in new tabDownload slide Total emergence (black) of adult emerald ash borer from seven milling treatments and controls (round wood) in 2017 (a) and 2018 (b), and percent of insects remaining in all treatments (c). Gray shading in (b) and (c) show the counts of insects that were resident in the samples before milling but did not emerge (unemerged adults and pupae) or were in a stage that required additional development before being able to emerge (prepupae, larvae). Note the difference in the scale of the y-axes between a and b. Numbers below give the exact sample sizes for each stage (a, b) or the percentage of all insects remaining relative to the number found in the controls (c). Fig. 2. Open in new tabDownload slide Number of adult emerald ash borer emerging from wood pieces produced during the milling process in 2017 (upper panel) and 2018 (lower panel). Treatments arranged left to right from round wood (controls) to end products (sawn wood). There was no emergence from the heat-treated sawn wood or heat-treated and kiln-dried sawn wood and so these two treatments were not included in these analyses. Treatment pairs connected by a bar are significantly different from each other (Tukey’s Honestly Significant Difference test, alpha = 0.05). Note the difference in the scale of the y-axes. Fig. 2. Open in new tabDownload slide Number of adult emerald ash borer emerging from wood pieces produced during the milling process in 2017 (upper panel) and 2018 (lower panel). Treatments arranged left to right from round wood (controls) to end products (sawn wood). There was no emergence from the heat-treated sawn wood or heat-treated and kiln-dried sawn wood and so these two treatments were not included in these analyses. Treatment pairs connected by a bar are significantly different from each other (Tukey’s Honestly Significant Difference test, alpha = 0.05). Note the difference in the scale of the y-axes. To estimate the effect of milling on reduction in population size we summed the total number of insects recovered from each of the milling treatments, and the round wood controls. We then expressed each of these sums as a percentage of the insects found in all the round wood control treatment in both years of the experiment (Fig. 1c). Assuming that the population of emerald ash borer in the sample trees was approximately equal prior to their being harvested, this analysis shows that 90% of emerald ash borer were removed at the debarking stage, and 99.5% of all emerald ash borer originally present in the sample were removed before the production of the sawn wood. The percentage shown for slabs is higher than that of debarked wood because this represents emergence from all the slabs from the original round wood samples that were milled into cants, green sawn wood, heat-treated sawn wood, and heat-treated and kiln-dried sawn wood (Table 1). Heat Treatment Experiment The heat-treated slabs typically spent a total of 150–200 min in the treatment kiln (Supp. Fig. S3a [online only]). The maximum temperature reached for each treatment temperature averaged 1.4°C above the target temperature for all treatments (range 1.02–1.69°C above target; Supp. Fig. S3b [online only]). The time to achieve the maximum temperature varied somewhat and did not correlate with the treatment temperature (Supp. Fig. S3c [online only]). Almost all slabs spent at least 30 min at the treatment temperature, and none more than 35 min (Supp. Fig. S3d [online only]). We observed a decrease in emergence with increasing treatment temperature and no emergence from slabs in the 54°C and 56°C temperature treatments (Fig. 3a). A single insect was recorded as being collected from 1 slab treated to 71°C (Fig. 3a), but we attribute this to an experimental error (see discussion). Insects that failed to emerge in the treatment temperatures that were <52°C were typically adults (Fig. 3b) or, to a lesser extent, prepupae (Fig. 3c). Insects that failed to emerge in the temperature treatments that were >52°C were almost all prepupae (Fig. 3c), indicating that the heat treatment had arrested development in the prepupal stage. Emerald ash borer arrested in the prepupal stage (Fig. 3c) were likely dead as any live insects would have continued development and emerged as adults. A small number of insects in all treatment slabs were younger than prepupae (Fig. 3d) and there were slightly more of these insects in the 71°C treatment than in all the other treatment temperatures. Emerald ash borer found in the larval stage would not have been viable (Fig. 3d) as the larval stage of the insect does not develop in cut wood. When we combine the total emerged and unemerged populations of emerald ash borer in the treatment slabs we found that there was no difference among the treatment temperatures (Fig. 3e) indicating that all the temperature treatments began the experiment with the same mean number of insects. The probability of finding a given emerald ash borer stage (an emerged adult, unemerged adult, prepupae, and larvae) was influenced by the temperature treatments (Table 2). Increasing heat decreased the probability of an adult emerald ash borer emerging by 85% in the 50°C treatment, relative to the control, by 99.6% in the 52°C treatment and by 99–100% in the 54, 56, and 71°C treatments. Year had no effect on the probability of finding a life stage, except for larvae where there was an effect of year. When we examine the effect of year and treatment on the number of insects we found that treatment had an effect on the number of emerged adults, prepupae and larvae, and the total number of insects but not the number of unemerged adults (Table 2). Year had an effect on the number of prepupae and larvae and on the total number of insects (Table 2). Table 2. Analysis of deviance for effect of year and heat treatment on the presence and absence of emerald ash borer and the number of emerald ash borer in each of four stages (a-d) and overall (e) . Response variable . Probability of finding stage (Presence/absence) . . . . . Numbera . . . . . Stage . Model term . df . Deviance . Residual df . Residual deviance . P . df . Deviance . Residual df . Residual deviance . P . a) Emerged adults Null model 189 260.843 82 103.268 Year 1 2.326 188 258.516 0.127 1 0.052 81 103.216 0.819 Year (treatment)b 7 208.531 181 49.984 <0.001 3 15.454 78 87.762 0.001 b) Unemerged adultsc Null model 189 192.893 36 40.555 Year 1 0.035 188 192.857 0.849 1 0.726 35 39.829 0.394 Year (treatment) 7 54.757 181 138.100 <0.001 3 3.738 32 36.091 0.291 c) Prepupae Null model 189 162.354 160 309.727 Year 1 0.495 188 161.859 0.481 1 19.091 159 290.636 <0.001 Year (treatment) 7 19.436 181 142.422 <0.001 7 132.926 152 157.709 <0.001 d) Larvae Null model 189 235.418 130 161.614 Year 1 12.130 188 223.288 <0.001 1 14.282 129 147.332 <0.001 Year (treatment) 7 22.938 181 200.349 0.002 7 32.512 122 114.819 <0.001 e) Totald Null model 178 197.111 Year 1 3.859 177 193.252 0.049 Year (treatment) 7 3.413 170 189.839 0.844 . Response variable . Probability of finding stage (Presence/absence) . . . . . Numbera . . . . . Stage . Model term . df . Deviance . Residual df . Residual deviance . P . df . Deviance . Residual df . Residual deviance . P . a) Emerged adults Null model 189 260.843 82 103.268 Year 1 2.326 188 258.516 0.127 1 0.052 81 103.216 0.819 Year (treatment)b 7 208.531 181 49.984 <0.001 3 15.454 78 87.762 0.001 b) Unemerged adultsc Null model 189 192.893 36 40.555 Year 1 0.035 188 192.857 0.849 1 0.726 35 39.829 0.394 Year (treatment) 7 54.757 181 138.100 <0.001 3 3.738 32 36.091 0.291 c) Prepupae Null model 189 162.354 160 309.727 Year 1 0.495 188 161.859 0.481 1 19.091 159 290.636 <0.001 Year (treatment) 7 19.436 181 142.422 <0.001 7 132.926 152 157.709 <0.001 d) Larvae Null model 189 235.418 130 161.614 Year 1 12.130 188 223.288 <0.001 1 14.282 129 147.332 <0.001 Year (treatment) 7 22.938 181 200.349 0.002 7 32.512 122 114.819 <0.001 e) Totald Null model 178 197.111 Year 1 3.859 177 193.252 0.049 Year (treatment) 7 3.413 170 189.839 0.844 Presence and absence data were fit to a binomial distribution; abundance data were fit to a negative binomial distribution to address issues with overdispersion in the raw data (except where noted). The Null model in all analyses is an intercept-only model lacking all terms of interest, P-values < 0.05 are indicated in bold. aThese statistical models were fit using data from samples where the number of emerald ash borer was > 0; therefore, treatments where no emerald ash borer of a given stage were found are not included in these models; data from 71°C treatment were included in the analysis, see discussion. bTreatment was nested within year because of unequal replication of treatments between years. cAbundance model fit with a Poisson distribution. dNo presence/absence model was fit for these data. Open in new tab Table 2. Analysis of deviance for effect of year and heat treatment on the presence and absence of emerald ash borer and the number of emerald ash borer in each of four stages (a-d) and overall (e) . Response variable . Probability of finding stage (Presence/absence) . . . . . Numbera . . . . . Stage . Model term . df . Deviance . Residual df . Residual deviance . P . df . Deviance . Residual df . Residual deviance . P . a) Emerged adults Null model 189 260.843 82 103.268 Year 1 2.326 188 258.516 0.127 1 0.052 81 103.216 0.819 Year (treatment)b 7 208.531 181 49.984 <0.001 3 15.454 78 87.762 0.001 b) Unemerged adultsc Null model 189 192.893 36 40.555 Year 1 0.035 188 192.857 0.849 1 0.726 35 39.829 0.394 Year (treatment) 7 54.757 181 138.100 <0.001 3 3.738 32 36.091 0.291 c) Prepupae Null model 189 162.354 160 309.727 Year 1 0.495 188 161.859 0.481 1 19.091 159 290.636 <0.001 Year (treatment) 7 19.436 181 142.422 <0.001 7 132.926 152 157.709 <0.001 d) Larvae Null model 189 235.418 130 161.614 Year 1 12.130 188 223.288 <0.001 1 14.282 129 147.332 <0.001 Year (treatment) 7 22.938 181 200.349 0.002 7 32.512 122 114.819 <0.001 e) Totald Null model 178 197.111 Year 1 3.859 177 193.252 0.049 Year (treatment) 7 3.413 170 189.839 0.844 . Response variable . Probability of finding stage (Presence/absence) . . . . . Numbera . . . . . Stage . Model term . df . Deviance . Residual df . Residual deviance . P . df . Deviance . Residual df . Residual deviance . P . a) Emerged adults Null model 189 260.843 82 103.268 Year 1 2.326 188 258.516 0.127 1 0.052 81 103.216 0.819 Year (treatment)b 7 208.531 181 49.984 <0.001 3 15.454 78 87.762 0.001 b) Unemerged adultsc Null model 189 192.893 36 40.555 Year 1 0.035 188 192.857 0.849 1 0.726 35 39.829 0.394 Year (treatment) 7 54.757 181 138.100 <0.001 3 3.738 32 36.091 0.291 c) Prepupae Null model 189 162.354 160 309.727 Year 1 0.495 188 161.859 0.481 1 19.091 159 290.636 <0.001 Year (treatment) 7 19.436 181 142.422 <0.001 7 132.926 152 157.709 <0.001 d) Larvae Null model 189 235.418 130 161.614 Year 1 12.130 188 223.288 <0.001 1 14.282 129 147.332 <0.001 Year (treatment) 7 22.938 181 200.349 0.002 7 32.512 122 114.819 <0.001 e) Totald Null model 178 197.111 Year 1 3.859 177 193.252 0.049 Year (treatment) 7 3.413 170 189.839 0.844 Presence and absence data were fit to a binomial distribution; abundance data were fit to a negative binomial distribution to address issues with overdispersion in the raw data (except where noted). The Null model in all analyses is an intercept-only model lacking all terms of interest, P-values < 0.05 are indicated in bold. aThese statistical models were fit using data from samples where the number of emerald ash borer was > 0; therefore, treatments where no emerald ash borer of a given stage were found are not included in these models; data from 71°C treatment were included in the analysis, see discussion. bTreatment was nested within year because of unequal replication of treatments between years. cAbundance model fit with a Poisson distribution. dNo presence/absence model was fit for these data. Open in new tab Fig. 3. Open in new tabDownload slide Number of adult emerald ash borer emerging from untreated control ash slabs and ash slabs heated to a target treatment temperature (a), and the number of emerald ash borer adults (b), prepupae (c), and larvae (d) found in the same slabs after adult emergence had completed, as well as the total number of emerald ash borer in each slab before the experiment was begun (e). Bars at the top of panels in (a), (c), and (d) indicate a significant difference between pairs of treatment means (Tukey’s HSD test, alpha < 0.05). Note the difference in scale in the y-axes in (a–e). Values in brackets below the target treatment temperatures are the average maximum observed temperatures (±1SE) recorded during the heat treatment. Fig. 3. Open in new tabDownload slide Number of adult emerald ash borer emerging from untreated control ash slabs and ash slabs heated to a target treatment temperature (a), and the number of emerald ash borer adults (b), prepupae (c), and larvae (d) found in the same slabs after adult emergence had completed, as well as the total number of emerald ash borer in each slab before the experiment was begun (e). Bars at the top of panels in (a), (c), and (d) indicate a significant difference between pairs of treatment means (Tukey’s HSD test, alpha < 0.05). Note the difference in scale in the y-axes in (a–e). Values in brackets below the target treatment temperatures are the average maximum observed temperatures (±1SE) recorded during the heat treatment. Male emerald ash borer emerged before female emerald ash borer in all treatment temperatures and we observed slightly earlier emergence from the 2017 cohort of insects compared to those reared in 2016 (Fig. 4). Though we did not quantify it, this year-over-year difference in emergence is probably a result of small differences in the temperature of the room where the cardboard tubes were held. Regardless, we also found that emergence was slightly delayed among insects in the 50°C treatment, but advanced in the 52°C treatment (Fig. 4). The result is that we found a significant effect of sex (deviance = 17.05; df = 1; dfresidual = 7; P < 0.001), treatment (deviance = 76.6776; df = 3; dfresidual = 4; P < 0.001), year (deviance = 162.16; df = 1; dfresidual = 3; P < 0.001) and an interaction between sex and treatment (deviance = 7.99; df = 3; dfresidual = 0; P = 0.001). Fig. 4. Open in new tabDownload slide Emergence proportion over time of adult emerald ash borer from heat-treated ash slabs or untreated control ash slabs at 26°C. Male emerald ash borer (right) emerge earlier than female emerald ash borer (left), as did insects from slabs subjected to higher heat treatments (bottom two rows). Lines show the model estimates for the proportion of the total population still not emerged on a given day. Experiments were repeated over 2 yr (legend). Fig. 4. Open in new tabDownload slide Emergence proportion over time of adult emerald ash borer from heat-treated ash slabs or untreated control ash slabs at 26°C. Male emerald ash borer (right) emerge earlier than female emerald ash borer (left), as did insects from slabs subjected to higher heat treatments (bottom two rows). Lines show the model estimates for the proportion of the total population still not emerged on a given day. Experiments were repeated over 2 yr (legend). As expected, male emerald ash borer were smaller than female emerald ash borer (Fig. 5). In general, we observed only small reductions in size associated with higher treatment temperatures, and these were more pronounced in males than in females (Fig. 5). There was an effect of treatment on thorax width but not on body length nor elytra length in females, while for males we observed an effect of heat treatment on all three measures of body size (Table 3). There was also a significant effect of year in all analyses of males and females (Table 3). Table 3. Analysis of variation for effect of year and heat treatment on body length, elytra length, and thorax width of adult emerald ash borer; P-values < 0.05 are indicated in bold Body Part . Sex: . Female . . . . . Male . . . . . . Model Term . df . Sum of squares . Mean square . F . P . df . Sum of squares . Mean square . F . P . Body length Year 1 65.30970 65.309695 97.4314 <0.001 1 4.66530 4.665300 8.8497 0.003 Year (treatment) 3 3.76652 1.255507 1.8730 0.133 3 9.42495 3.141649 5.9595 <0.001 Residuals 553 370.68398 0.670315 786 414.35495 0.527169 Elytra length Year 1 43.41286 43.412857 94.1573 <0.001 1 1.32133 1.321330 3.6828 0.055 Year (treatment) 3 2.75163 0.917210 1.9893 0.114 3 6.09053 2.030177 5.6585 <0.001 Residuals 554 255.43129 0.461067 786 282.00556 0.358786 Thorax width Year 1 6.98536 6.985363 121.2364 <0.001 1 0.91286 0.912856 21.7454 <0.001 Year (treatment) 3 0.51977 0.173256 3.0070 0.030 3 0.76460 0.254866 6.0713 <0.001 Residuals 554 31.92022 0.057618 786 32.99567 0.041979 Body Part . Sex: . Female . . . . . Male . . . . . . Model Term . df . Sum of squares . Mean square . F . P . df . Sum of squares . Mean square . F . P . Body length Year 1 65.30970 65.309695 97.4314 <0.001 1 4.66530 4.665300 8.8497 0.003 Year (treatment) 3 3.76652 1.255507 1.8730 0.133 3 9.42495 3.141649 5.9595 <0.001 Residuals 553 370.68398 0.670315 786 414.35495 0.527169 Elytra length Year 1 43.41286 43.412857 94.1573 <0.001 1 1.32133 1.321330 3.6828 0.055 Year (treatment) 3 2.75163 0.917210 1.9893 0.114 3 6.09053 2.030177 5.6585 <0.001 Residuals 554 255.43129 0.461067 786 282.00556 0.358786 Thorax width Year 1 6.98536 6.985363 121.2364 <0.001 1 0.91286 0.912856 21.7454 <0.001 Year (treatment) 3 0.51977 0.173256 3.0070 0.030 3 0.76460 0.254866 6.0713 <0.001 Residuals 554 31.92022 0.057618 786 32.99567 0.041979 Open in new tab Table 3. Analysis of variation for effect of year and heat treatment on body length, elytra length, and thorax width of adult emerald ash borer; P-values < 0.05 are indicated in bold Body Part . Sex: . Female . . . . . Male . . . . . . Model Term . df . Sum of squares . Mean square . F . P . df . Sum of squares . Mean square . F . P . Body length Year 1 65.30970 65.309695 97.4314 <0.001 1 4.66530 4.665300 8.8497 0.003 Year (treatment) 3 3.76652 1.255507 1.8730 0.133 3 9.42495 3.141649 5.9595 <0.001 Residuals 553 370.68398 0.670315 786 414.35495 0.527169 Elytra length Year 1 43.41286 43.412857 94.1573 <0.001 1 1.32133 1.321330 3.6828 0.055 Year (treatment) 3 2.75163 0.917210 1.9893 0.114 3 6.09053 2.030177 5.6585 <0.001 Residuals 554 255.43129 0.461067 786 282.00556 0.358786 Thorax width Year 1 6.98536 6.985363 121.2364 <0.001 1 0.91286 0.912856 21.7454 <0.001 Year (treatment) 3 0.51977 0.173256 3.0070 0.030 3 0.76460 0.254866 6.0713 <0.001 Residuals 554 31.92022 0.057618 786 32.99567 0.041979 Body Part . Sex: . Female . . . . . Male . . . . . . Model Term . df . Sum of squares . Mean square . F . P . df . Sum of squares . Mean square . F . P . Body length Year 1 65.30970 65.309695 97.4314 <0.001 1 4.66530 4.665300 8.8497 0.003 Year (treatment) 3 3.76652 1.255507 1.8730 0.133 3 9.42495 3.141649 5.9595 <0.001 Residuals 553 370.68398 0.670315 786 414.35495 0.527169 Elytra length Year 1 43.41286 43.412857 94.1573 <0.001 1 1.32133 1.321330 3.6828 0.055 Year (treatment) 3 2.75163 0.917210 1.9893 0.114 3 6.09053 2.030177 5.6585 <0.001 Residuals 554 255.43129 0.461067 786 282.00556 0.358786 Thorax width Year 1 6.98536 6.985363 121.2364 <0.001 1 0.91286 0.912856 21.7454 <0.001 Year (treatment) 3 0.51977 0.173256 3.0070 0.030 3 0.76460 0.254866 6.0713 <0.001 Residuals 554 31.92022 0.057618 786 32.99567 0.041979 Open in new tab Fig. 5. Open in new tabDownload slide Size of adult emerald ash borer females (left) and males (right) emerging from ash slabs subjected to one of three heat treatments, and those emerging from untreated controls. Within the panels, treatment pairs connected by a bar are significantly different from each other (Tukey’s Honestly Significant Difference test, alpha = 0.05). Note the difference in scale in the y-axes. Experiments were repeated over 2 yr, the 48°C treatment was completed in 2016, the 50°C and 52°C were completed in 2017. Fig. 5. Open in new tabDownload slide Size of adult emerald ash borer females (left) and males (right) emerging from ash slabs subjected to one of three heat treatments, and those emerging from untreated controls. Within the panels, treatment pairs connected by a bar are significantly different from each other (Tukey’s Honestly Significant Difference test, alpha = 0.05). Note the difference in scale in the y-axes. Experiments were repeated over 2 yr, the 48°C treatment was completed in 2016, the 50°C and 52°C were completed in 2017. We were only able to assess lifespan for individuals emerging from control, 48, 50, and 52°C treatments, as no adult insects emerged from the other treatment temperatures (Fig. 3). When testing for the effect of ‘experiment’ we considered all the insects in the two mating experiments (single- and multiple-pair mating). We also included the lifespans of all individuals that died prior to being placed into one of the two experiments. In doing so we found no effect of heat treatment (F3,1235 = 1.54, P = 0.20) or sex (F1,1235 = 0.444, P = 0.50) on lifespan of adult emerald ash borer. We did however detect an effect of both year (F1,1235 = 91.69, P < 0.001) and mating experiment (F2,1235 = 59.60, P < 0.001). We attribute the year-over-year affect to improvements in our rearing procedures between the two cohorts that contributed to increased success of adults in our rearing (Fig. 6). The differences among mating experiments resulted in slightly better longevity among insects in the multiple-pair mating treatment. Fig. 6. Open in new tabDownload slide Lifespan of adult emerald ash borer from heat-treated ash slabs or untreated control slabs reared under laboratory conditions. Unmated individuals are those that perished before being transferred to the single-pair or multiple-pair mating experiments. Experiments were repeated over 2 yr, the 48°C treatment was completed in 2016, the 50°C and 52°C were completed in 2017; controls from both years have been pooled. Fig. 6. Open in new tabDownload slide Lifespan of adult emerald ash borer from heat-treated ash slabs or untreated control slabs reared under laboratory conditions. Unmated individuals are those that perished before being transferred to the single-pair or multiple-pair mating experiments. Experiments were repeated over 2 yr, the 48°C treatment was completed in 2016, the 50°C and 52°C were completed in 2017; controls from both years have been pooled. There were limited data available to test for the effect of fertility and fecundity because of the low numbers of insects that emerged in most of the heat treatments. Therefore, this analysis was limited to a comparison of individuals in the control and 48°C treatment from the 2017 cohort, only. In this analysis, we did see a significant effect of treatment (F1,50 = 4.8341, P = 0.032) on fecundity, but the overall effect was weak, with just 1.1 ± 0.6 more eggs female−1 day−1 (estimate ± 1 SE) in the control than the 48°C treatment. There was no effect of the mating experiment (F1,50 = 0.0019, P = 0.965). The difference between the heat treatments in fertility amounted to 0.83 ± 0.41 more viable eggs female−1 day−1 (estimate ± 1 SE) in the control, but there was no statistically significant effect of either mating experiment (F1,50 = 0.4515, P = 0.5047) or heat treatment (F1,50 = 3.6025, P = 0.063) on fertility. Discussion The incorporation of multiple production steps during the production of sawn wood from infested round wood reduces the incidence of pests. These steps may be included as phytosanitary measures within a systems approach that can remove phloem-feeding insects from finished wood products. Debarking removes a large proportion of the insects that are associated with the bark and underlying tissue. In this experiment, debarking removed >90% of the insects present in the round wood before it entered the sawmill. Almost all of the remaining insects were removed during the slabbing process such that 99.7% of all insects had been removed by the time the squared sawn wood (the cant) was produced. The final steps in the production process (heat treatment and kiln drying) kills any remaining insects present in the sawn wood. Although it should be noted, that in the milling experiment we conducted there were no insects—dead or otherwise—present in any of the sawn wood that was heat-treated, or heat-treated and then kiln-dried. Emerald ash borer larvae are easily recovered during dissection, even in a desiccated state. Thus, that we did not recover any insects suggest this sawn wood was free of emerald ash borer before it entered the heat treatment process. Our explicit test of heat treatment temperatures recovered no insects above the target temperature of 54°C. This is a much lower temperature than would be experienced by the insect when being treated to 56°C for 30 min throughout the profile of the wood. We recovered a single adult emerald ash borer from a slab heat-treated to 71°C. One interpretation of this recovery is that this insect survived the treatment temperature and emerged as a viable adult, but this is inconsistent with the results from the lower treatment temperatures, where no insects emerged. Therefore, we investigated the provenance of this insect to determine whether it was a ‘true’ survivor. The heating profile of the slab indicated that the insect should have received 71°C through the profile of the wood and so we can not reject inadequate heating as a cause. We then investigated the information recorded in the laboratory notebooks. These records indicate that this single insect was the only one reared from the slab, labeled ‘72’ in the notebook. As we described in the methods, all slabs for this experiment were maintained in a common rearing room to elicit emergence. At the same time, as slab 72 was in the rearing room, slab ‘27’ from the untreated control set was also present in the rearing room. Slab 27 is recorded as having produced adult emerald ash borer in the days immediately prior to that when slab 72 produced its single insect, and in the days afterward. Moreover, slab 27 did not produce any insects on the day slab 72 did. For this reason, we attribute the insect recorded from slab 72 as actually having emerged from slab 27, and that a transposition error was made when entering the emergence data in the laboratory notebook. We offer the following observations to support our contention: Of the 1,414 insects recovered after dissecting the slabs in the 54, 56, and 71°C treatments only 5 (0.35%) were in the adult stage and only 1 (the 71°C survivor) was ‘alive’ (0.07%). Recovering 4 other dead adults from the slabs is consistent with the number of dead adults recovered from slabs in the other treatments. These adults are likely insects that failed to emerge in the years prior to when the trees were harvested. All the other insects in the 54, 56, and 71°C treatments were arrested in the prepupal or larval stage (Fig. 3c), indicative of overwintering emerald ash borer, which are the stages the insects would have been in when the trees were harvested in February in Canada. This evidence strongly suggests that heat treatment above 54°C causes mortality of emerald ash borer and thus, the single insect is a not true survivor. Unfortunately, without additional evidence to support this hypothesis, we did not feel it was appropriate to reject this single insect as being the product of a known demonic intrusion (Hurlbert 1984) and so it was retained in all our analyses. Systems approaches have been adopted around the world in many agricultural production systems (Quinlan and Ikin 2009) but are not common for forest products (Allen et al. 2017). RSPM 41 (NAPPO 2018) provides a framework and guidance for developing forest product systems approaches as a practical option for risk management when a single risk management measure is not available or practical. This study has provided some of the first empirical evidence for the effectiveness of a forest products systems approach. We have shown that the combination of multiple measures, applied through the production process of sawn ash wood, mitigate the risk of emerald ash borer. We had hypothesized that heat treatment may also contribute to biological mitigation of pest risk by reducing the viability of insects that survive heat treatment. However, the emerald ash borer in this experiment treated to sublethal temperatures (<54°C) did not show evidence of biologically meaningful reductions in size, fertility or fecundity. To our knowledge, this is the first study to quantify the effect of sawn wood production processes on risk reduction in a wood-boring insect. Previous work has focused on the effect of heat treatment in production systems (Mushrow et al. 2004, Myers and Bailey 2011, Mayfield et al. 2014, Lazarescu et al. 2015, Naves et al. 2019, Pawson et al. 2019, Wang et al. 2019). Work on physical impacts on pest survival has focused on the effect of chipping wood in risk mitigation (e.g., McCullough et al. 2007, Flo et al. 2014). This lack of attention to the effect of milling on pest risk reduction may stem from the difficulty in obtaining infested roundwood with consistent densities of the target of interest. This difficulty in obtaining sufficient sample sizes is a legitimate challenge to assessing the effects of heat in Probit 9 studies (Schortemeyer et al. 2011). It may be possible, however, to infer the process of milling using other data that is more-easily obtained. For instance, the effect of debarking on a generic wood borer can be inferred by measuring the amount of material removed during the debarking process and comparing this to the measured depth of where the insect feeds inside the wood or phloem (Romo et al. 2018). For some wood-boring insects, this depth may be equivalent to that of the phloem layer of the tree, for others, like emerald ash borer, it may be the depth of the pupation chamber (Wei et al. 2007). It is still somewhat laborious to obtain a representative sample of ‘habitation depths’ for rare wood-boring insects but the tactic has the advantage of not relying on a fortuitous outbreak, as we do in this study. Heat treatment did not appear to result in sublethal effects in emerald ash borer. This suggests that heat treatment that does not kill the insect will also not mitigate the risk of invasion. We did observe a small reduction in lifespan for insects that survived heat treatment to 52°C where some insects died before they would have been expected to complete their obligatory pre-oviposition feeding (Fig. 6). However, those insects that survived the pre-oviposition period were still able to mate and produce viable eggs. Thus, insufficient heat treatment (i.e., <56°C) does not provide a reliable mitigating effect against the transport of a potentially invasive insect. We had hypothesized that sublethal temperatures may have induced sublethal effects on the developing emerald ash borer. These effects were not observed. However, heating to sublethal temperatures can affect subsequent tolerance to heating by inducing the production of heat-shock proteins in the insect (Lurie and Jang 2007). One effect of this is that insects that have experienced a period of pre-conditioning (e.g., by sublethal heat treatment) could require more heat in order to cause mortality than insects that were not subject to conditioning (Tang et al. 2007). Previous studies have assessed the effect of heat treatment on survival and subsequent emergence of emerald ash borer. Our results are similar to those reported earlier, which suggests that the thermal tolerance of the population we sampled is similar to those populations used in the other studies (Nzokou et al. 2008, Myers et al. 2009, Goebel et al. 2010, Sobek et al. 2011). Emerald ash borer mortality has been reported at lower temperatures (40°C vs 54°C; McCullough et al. 2007), but when the wood was treated for longer than in our study (50 h vs 0.5 h). Ours is also the first study to assess survival in simulated sawn wood. All previous work has examined the effect on survival in bark or wood chips (McCullough et al. 2007), logs (Nzokou et al. 2008), firewood (Myers et al. 2009, Goebel et al. 2010) or excised larvae and pupae (Sobek et al. 2011). In general, in situ heat treatment experiments like this one are likely to be difficult to complete for most potentially invasive wood-boring insects. One solution to this limitation is to test the effect of heat on insects using specialized water baths (Lurie and Jang 2007) or heat blocks (Neven et al. 2012) to treat insects extracted from a tree. Obtaining large numbers of insects for these kinds of experiments will be a challenge, however, though many insects respond to baits hung in live-traps or kariomones produced by wounded trees (e.g., Mercader et al. 2013) and this behavior could be exploited to concentrate insects within a tree that can then be extracted for use in heating experiments. Our results are specific to the emerald ash borer, but the effect of milling can be extended to other phloem-feeding insects that are at risk of becoming invasive pests. For instance the bronze birch borer, Agrilus anxius Gory (Coleoptera: Buprestidae) that attacks birch, Betula L. (Fagales: Betulaceae) in North America poses a risk to European birch species (Muilenburg and Herms 2012), other Agrilus species have recently been recorded as significant pests of oak, Quercus L. (Fagales: Fagaceae) in North America (Coleman and Seybold 2008, Haack et al. 2009) as have species of Pityophthorus (Coleoptera: Cuculionidae) attacking walnut, Juglans L. (Fagales: Juglandaceae) (Tisserat et al. 2009). The methods we have employed here could be applied to assessing the potential effectiveness of the systems approach to limit the risk of export of these species. Mitigating the risk posed by potentially invasive species is essential to maintaining trade in valuable wood commodities. Systems approaches are effective for managing the pest risk in agricultural commodities and have been applied to other commodities, including wood products such as Christmas trees and green sawn wood products (NAPPO 2012, CFIA 2019a). The limitation in providing scientific evidence to support new forest product systems approaches will be the difficulty in obtaining reliable data to quantify the effectiveness of phytosanitary measures. We have shown that one way this limitation can be addressed is by using a model system to generate data on the effect of different steps in a production pathway. This model system has provided a template from which we can extrapolate the effect of similar production processes on mitigating the risk by other wood-boring insects. We have also confirmed that heat is an effective measure that reliably prevents the emergence of any insects that may survive the milling process. Moreover, at least for insects like the emerald ash borer that inhabit the outer part of the tree, this heat requirement could perhaps be tempered from the current regulatory requirements by some countries (e.g., 60°C for 60 min; USDA 2016). Doing so would still provide reliable protection against insect introductions, but would also have the added effect of reducing the energy needed to produce the commodity. Supplementary Data Supplementary data are available at Journal of Economic Entomology online. Supplemental Figure S1: Flow chart describing the milling process (left) and rearing process (right) used in the milling experiment. Outputs (parallelograms) of the milling process are the wood products and milling by-products used as the experimental treatments and are placed in the diagram to indicate where they were extracted from the milling process. Supplemental Figure S2: Temperatures recorded over one hour from thermocouples inserted into two test ash slabs to determine the optimal placement of thermocouples for the heat treatment experiment. One slab was used to test the effect of varying the location of the thermocouples relative to the end of the slab (a) and another was used to test the effect of varying the location of the thermocouples relative to the end, and the depth (b) of the slab. All thermocouples reported the same temperature after ca. 30 minutes. Supplementary Figure S3: Summary of temperature observations from ash slabs subjected to heat treatment in a kiln. Rows in the plot correspond to the temperature treatments (48 – 71 °C). Observed temperatures (a) for each slab over time (grey lines) and the average over all slabs (black line). Each slab achieved a maximum temperature (b) above that of the intended treatment and took different lengths of time (c) to achieve the desired temperature. The time spent at the treatment temperature (d) was intended to be 30 minutes, but varied among slabs. Box and whisker plots in b, c & d show the median (horizontal line), the mean (*), the 25th and 75th percentiles (box), 1.5 times the interquartile range (whiskers), and outliers (points). Note that in a and b scales vary among panels; in b, c & d light gray indicates data taken in 2016, black indicates data taken in 2017. Acknowledgments We thank J. Allard, C. Emilson, G. Jones, D. Fromme, J. Skillings, K. Scarfone, T. Ladd, G. Roth, and R. Scharbach, for field and lab assistance with this project; M. Penner and the staff of Townsend Lumber, Tilsonburg, Ontario for harvest, transportation and milling, and guidance in the preparation of the samples used in the milling experiment; D. Oliver for milling of the experimental ash slabs used in the heat treatment experiment; FPInnovations, Quebec for assistance and technical advice for the experimental kiln treatments; the Ontario Ministry of Natural Resources and Forestry Seed Plant in Angus, Ontario for providing insect rearing facilities; and E. Allen, G. Blackburn, J. Tanney, and two anonymous reviewers for valuable comments on earlier drafts of the manuscript. This work was supported by the Quebec Wood Export Bureau, OntarioWood, the Canadian Food Inspection Agency, and Natural Resources Canada. References Cited Allen , E. , M. Noseworthy , and M. Ormsby . 2017 . Phytosanitary measures to reduce the movement of forest pests with the international trade of wood products . Biol. Invasions 19 : 3365 – 3376 . Google Scholar Crossref Search ADS WorldCat Barant Company . 2001 . ScanLink, version 2.0 . Barrant Company , Montreal, Canada . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Cameron , A. C. , and P. K. Trivedi . 1990 . Regression-Based tests for overdispersion in the Poisson model . J. Econometrics 46 : 347 – 364 . Google Scholar Crossref Search ADS WorldCat Canadian Food Inspection Agency [CFIA] . 2019a . D-17-04: Canadian green sawn wood certification program . http://www.inspection.gc.ca/plants/plant-pests-invasive-species/directives/date/d-17-04/eng/1546882362007/1546882362522. Canadian Food Inspection Agency [CFIA] . 2019b . Emerald ash borer regulated areas in Canada . http://www.inspection.gc.ca/plants/plant-pests-invasive-species/insects/emerald-ash-borer/areas-regulated/eng/1347625322705/1367860339942. Cappaert , D. , D. G. McCullough , T. M. Poland , and N. W. Siegert . 2005 . Emerald ash borer in North America: a research and regulatory challenge . Am. Ent . 51 : 152 – 165 . Google Scholar Crossref Search ADS WorldCat Chen , Z. , M. S. White , and R. Mack . 2016 . Evaluating vacuum and steam process on hardwood veneer logs for export . Eur. J. Wood Wood Prod . 75 : 911 – 918 . Google Scholar Crossref Search ADS WorldCat Coleman , T. W. , and S. J. Seybold . 2008 . Previously unrecorded damage to oak, Quercus spp., in southern California by the goldspotted oak borer, Agrilus coxalis Waterhouse (Coleoptera: Buprestidae) . Pan-Pac. Entomol. 84 : 288 – 300 . Google Scholar Crossref Search ADS WorldCat Duan , J. J. , T. Watt , and C. B. Oppel . 2011 . An alternative host plant-based method for laboratory rearing of emerald ash borer to produce larval parasitoids for biological control, pp. 3 . In G. Parra , D. Lance , V. Mastro , R. Reardon and C. Benedict (eds.), Proceedings, Emerald Ash Borer Research and Technology Meeting , 10–23 October 2011 . USDA Forest Service/APHIS/FHTET-2011-06 , Wooster, OH . Fick , W. E. , and C. J. K. MacQuarrie . 2018 . An artificial delay in emergence influences the number but not the fitness of adult emerald ash borer emerging from infested ash wood . Entomol. Exp. Appl . 166 : 171 – 182 . Google Scholar Crossref Search ADS WorldCat Flo , D. , P. Krokene , and B. Okland . 2014 . Importing deciduous wood chips from North America to northern Europe - the risk of introducing bark-and wood-boring insects . Scand. J. Forest. Res . 29 : 77 – 89 . Google Scholar Crossref Search ADS WorldCat Food and Agriculture Organization of the United Nations [FAO] . 2019 . Glossary of phytosanitary terms. International Standard for Phytosanitary Measures No. 5 . Published by FAO on behalf of the Secretariat of the International Plant Protection Convention (IPPC) , Rome, Italy . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Goebel , P. C. , M. S. Bumgardner , D. A. Herms , and A. Sabula . 2010 . Failure to phytosanitize ash firewood infested with emerald ash borer in a small dry kiln using ISPM-15 standards . J. Econ. Entomol . 103 : 597 – 602 . Google Scholar Crossref Search ADS PubMed WorldCat Haack , R. A. , and T. R. Petrice . 2009 . Bark- and wood-borer colonization of logs and lumber after heat treatment to ISPM 15 specifications: the role of residual bark . J. Econ. Entomol . 102 : 1075 – 1084 . Google Scholar Crossref Search ADS PubMed WorldCat Haack , R. A. , T. R. Petrice , and J. E. Zablotny . 2009 . First report of the european oak borer, Agrilus sulcicollis (Coleoptera: Buprestidae), in the United States . Great Lakes Entomol . 42 : 1 . OpenURL Placeholder Text WorldCat Haack , R. A. , K. O. Britton , E. G. Brockerhoff , J. F. Cavey , L. J. Garrett , M. Kimberley , F. Lowenstein , A. Nuding , L. J. Olson , J. Turner , et al. 2014 . Effectiveness of the International Phytosanitary Standard ISPM No. 15 on reducing wood borer infestation rates in wood packaging material entering the United States . PLoS One 9 : e96611 . Google Scholar Crossref Search ADS PubMed WorldCat Harrell , F. E. , Jr. 2015 . Regression modeling strategies , Springer , New York, NY . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Harrell , F. E. , Jr. 2019 . rms: Regression Modeling Strategies, R package, version 5.1-3 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hurlbert , S. H . 1984 . Pseudoreplication and the design of ecological field experiments . Ecol. Monogr . 54 : 187 – 211 . Google Scholar Crossref Search ADS WorldCat International Plant Protection Convention [IPPC] . 2019 . Regulation of wood packaging material in international trade . Food and Agriculture Organization of the United Nations , Rome, Italy . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Jackman , S . 2017 . pscl: Classes and methods for R developed in the Political Science Computational Laboratory, version R package version 1.5.2 . Sydney, Australia . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kleiber , C. , and A. Zeileis . 2008 . Applied econometrics with R . Springer-Verlag , New York, NY . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Lazarescu , C. , C. Breuil , and S. Avramidis . 2015 . Phytosanitation of mountain pine beetle infected lodgepole pine using dielectric fields at radio frequencies . Maderas-Cienc. Technol . 17 : 221 – 228 . OpenURL Placeholder Text WorldCat Lenth , R . 2019 . Emmeans: estimated marginal means, aka least-squares means , R package, version 1.3.3 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lurie , S. , and E. Jang . 2007 . The influence of heat shock proteins on insect pests and fruits in thermal treatments, pp. 269 – 290 . In J. Tang , E. Mitcham , S. Wang and S. Lurie (eds.), Heat treatments for postharvest pest control: theory and practice . CAB International , Wallingford, United Kingdom . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC MacQuarrie , C. J. K . 2019 . The effect of host condition on adult emerald ash borer (Agrilus planipennis) performance . Forestry . doi:10.1093/forestry/cpz008. OpenURL Placeholder Text WorldCat MacQuarrie , C. J. K. , M. Gray , R. Lavallée , M. Noseworthy , M. Savard , and L. M. Humble . 2019 . Data from: assessment of the systems approach for the phytosanitary treatment of wood infested with wood-boring insects . Dryad online repository . doi:10.5061/dryad.q573n5tf9. OpenURL Placeholder Text WorldCat Mayfield , A. E. , S. W. Fraedrich , A. Taylor , P. Merten , and S. W. Myers . 2014 . Efficacy of heat treatment for the thousand cankers disease vector and pathogen in small black walnut logs . J. Econ. Entomol . 107 : 174 – 184 . Google Scholar Crossref Search ADS PubMed WorldCat McCullough , D. G. , and N. W. Siegert . 2007 . Estimating potential emerald ash borer (Coleoptera: Buprestidae) populations using ash inventory data . J. Econ. Entomol . 100 : 1577 – 1586 . Google Scholar Crossref Search ADS PubMed WorldCat McCullough , D. G. , T. M. Poland , D. Cappaert , E. L. Clark , I. Fraser , V. Mastro , S. Smith , and C. Pell . 2007 . Effects of chipping, grinding, and heat on survival of emerald ash borer, Agrilus planipennis (Coleoptera: Buprestidae), in chips . J. Econ. Entomol . 100 : 1304 – 1315 . Google Scholar Crossref Search ADS PubMed WorldCat Mercader , R. J. , D. G. McCullough , and J. M. Bedford . 2013 . A comparison of girdled ash detection trees and baited artificial traps for Agrilus planipennis (Coleoptera: Buprestidae) detection . Environ. Entomol . 42 : 1027 – 1039 . Google Scholar Crossref Search ADS PubMed WorldCat Muilenburg , V. L. , and D. A. Herms . 2012 . A review of bronze birch borer (Coleoptera: Buprestidae) life history, ecology, and management . Environ. Entomol . 41 : 1372 – 1385 . Google Scholar Crossref Search ADS PubMed WorldCat Mushrow , L. , A. Morrison , J. Sweeney , and D. Quiring . 2004 . Heat as a phytosanitary treatment for the brown spruce longhorn beetle . Forest. Chron . 80 : 224 – 228 . Google Scholar Crossref Search ADS WorldCat Myers , S. W. , and S. M. Bailey . 2011 . Evaluation of a heat treatment schedule for the Asian longhorned beetle, Anoplophora glabripennis (Coleoptera: Cerambycidae) . Forest Prod. J . 61 : 46 – 49 . Google Scholar Crossref Search ADS WorldCat Myers , S. W. , I. Fraser , and V. C. Mastro . 2009 . Evaluation of heat treatment schedules for emerald ash borer (Coleoptera: Buprestidae) . J. Econ. Entomol . 102 : 2048 – 2055 . Google Scholar Crossref Search ADS PubMed WorldCat Naves , P. , M. L. Inacio , F. Nobrega , E. Sousa , and M. Michielsen . 2019 . Pinewood nematode presence and survival in commercial pallets of different ages . Eur. J. Wood Wood Prod . 77 : 301 – 309 . Google Scholar Crossref Search ADS WorldCat Neven , L. G. , S. J. Wang , and J. Tang . 2012 . An improved system to assess insect tolerance to heated controlled atmosphere quarantine treatment . Entomol. Exp. Appl . 143 : 95 – 100 . Google Scholar Crossref Search ADS WorldCat North American Plant Protection Organization [NAPPO] . 2012 . RSPM 37 Integrated measures for the trade of Christmas trees, NAPPO Regional Standards for Phytosanitary Measures . The Secretariat of the North American Plant Protection Organization , Ottawa, Canada . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC North American Plant Protection Organization [NAPPO] . 2018 . RSPM 41 Use of systems approaches to manage pest risks associated with the movement of forest products, NAPPO Regional Standards for Phytosanitary Measures . The Secretariat of the North American Plant Protection Organization , Raleigh, NC . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Nzokou , P. , S. Tourtellot , and D. P. Kamdem . 2008 . Kiln and microwave heat treatment of logs infested by the emerald ash borer (Agrilus planipennis Fairmaire) (Coleoptera: Buprestidae) . Forest Prod. J . 58 : 68 – 72 . OpenURL Placeholder Text WorldCat Pawson , S. M. , M. K. F. Bader , E. G. Brockerhoff , W. J. B. Heffernan , J. L. Kerr , and B. O’Connor . 2019 . Quantifying the thermal tolerance of wood borers and bark beetles for the development of Joule heating as a novel phytosanitary treatment of pine logs . J. Pest Sci . 92 : 157 – 171 . Google Scholar Crossref Search ADS WorldCat Pinheiro , J. , D. Bates , S. DebRoy , D. Sarkar , and R Core Team . 2018 . nlme: linear and nonlinear mixed effects models. R package, version 3 . 1 –137 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Quinlan , M. M. , and R. Ikin . 2009 . Enhancements of pest risk analysis techniques: A review of the application of systems approach to risk management in plant health. EU Framework 7 Research Project, Enhancement of Pest Risk Analysis Techniques . Report number PD No. 4.2. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC R Core Team . 2019 . R: A language and environment for statistical computing, version 3.5.3 . Vienna, Austria . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Roe , A. D. , M. Demidovich , and J. Dedes . 2018 . Origins and history of laboratory insect stocks in a multispecies insect production facility, with the proposal of standardized nomenclature and designation of formal standard names . J. Insect Sci . 18 : 1 . Google Scholar Crossref Search ADS WorldCat Romo , C. M. , M. K. F. Bader , and S. M. Pawson . 2018 . Measuring wood-boring depth of Arhopalus ferus and Prionoplus reticularis in Pinus radiata logs informs the development of phytosanitary treatments . Entomol. Exp. Appl . 166 : 823 – 833 . Google Scholar Crossref Search ADS WorldCat Schortemeyer , M. , K. Thomas , R. A. Haack , A. Uzunovic , K. Hoover , J. A. Simpson , and C. A. Grgurinovic . 2011 . Appropriateness of probit-9 in the development of quarantine treatments for timber and timber commodities . J. Econ. Entomol . 104 : 717 – 731 . Google Scholar Crossref Search ADS PubMed WorldCat Schratz , P . 2017 . R package ‘oddsratio’: Odds ratio calculation for GAM(M)s & GLM(M)s, R package, version 1.0.2 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Siegert , N. W. , D. G. McCullough , A. M. Liebhold , F. W. Telewski , and H. MacIsaac . 2014 . Dendrochronological reconstruction of the epicentre and early spread of emerald ash borer in North America . Divers. Distrib . 20 : 847 – 858 . Google Scholar Crossref Search ADS WorldCat Sobek , S. , A. Rajamohan , D. Dillon , R. C. Cumming , and B. J. Sinclair . 2011 . High temperature tolerance and thermal plasticity in emerald ash borer Agrilus planipennis . Agr. Forest Entomol . 13 : 333 – 340 . Google Scholar Crossref Search ADS WorldCat Tang , J. , S. Wang , and J. A. Johnson . 2007 . Biology and thermal death kinetics of selected insects, pp. 133 – 161 . In J. Tang , E. Mitcham , S. Wang and S. Lurie (eds.), Heat treatments for postharvest pest control: theory and practice . CAB International , Wallingford, UK . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Therneau , T . 2015 . A Package for Survival Analysis in S, R package, version 2.38 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Timms , L. L. , S. M. Smith , and P. de Groot . 2006 . Patterns in the within-tree distribution of the emerald ash borer Agrilus planipennis (Fairmaire) in young, green-ash plantations of south-western Ontario, Canada . Agr. Forest Entomol . 8 : 313 – 321 . Google Scholar Crossref Search ADS WorldCat Tisserat , N. , W. Cranshaw , D. Leatherman , C. Utley , and K. Alexander . 2009 . Black walnut mortality in Colorado caused by the walnut twig beetle and thousand cankers disease . Plant Health Progress . 10 . OpenURL Placeholder Text WorldCat Tukey , J. W . 1949 . Comparing individual means in the analysis of variance . Biometrics . 5 : 99 – 114 . Google Scholar Crossref Search ADS PubMed WorldCat United States Department of Agriculture [USDA] . 2016 . United States Department of Agriculture: Treatment Manual , 2nd ed. Washington, DC . https://www.aphis.usda.gov/import_export/plants/manuals/ports/downloads/treatment.pdf. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC United States Department of Agriculture Animal Plant Health Inspection Service [USDA APHIS] . 2019 . Cooperative emerald ash borer project initial county EAB detections in North America . https://www.aphis.usda.gov/plant_health/plant_pest_info/emerald_ash_b/downloads/MultiState.pdf. Venebles , W. N. , and B. D. Ripley . 2002 . Modern applied statistics with S , 4th ed. Springer , New York, NY . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Wang , L. C. , J. L. Yu , D. J. Yu , Z. J. Chen , W. F. Zhang , W. Lin , and L. Xu . 2019 . Evaluation of low pressure and vapor heat as a phytosanitary treatment for Chlorophorus annularis (Coleoptera: Cerambycidae) in postharvest bamboo poles . J. Econ. Entomol . 112 : 119 – 126 . Google Scholar Crossref Search ADS PubMed WorldCat Wei , X. I. A. , Y. U. N. Wu , R. Reardon , T-H. Sun , M. I. N. Lu , and J.-H. Sun . 2007 . Biology and damage traits of emerald ash borer (Agrilus planipennis Fairmaire) in China . Insect Sci . 14 : 367 – 373 . Google Scholar Crossref Search ADS WorldCat Yokoyama , V. Y . 2011 . Approved quarantine treatment for Hessian fly (Diptera: Cecidomyiidae) in large-size hay bales and Hessian fly and cereal leaf beetle (Coleoptera: Chrysomelidae) control by bale compression . J. Econ. Entomol . 104 : 792 – 798 . Google Scholar Crossref Search ADS PubMed WorldCat Zeileis , A. , C. Kleiber , and S. Jackman . 2008 . Regression models for count data in R . J. Stat. Softw . 27 : 1 – 25 . OpenURL Placeholder Text WorldCat © Her Majesty the Queen in Right of Canada, as represented by the Minister of Natural Resources, 2019. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Assessment of the Systems Approach for the Phytosanitary Treatment of Wood Infested With Wood-Boring Insects
JO - Journal of Economic Entomology
DO - 10.1093/jee/toz331
DA - 2020-04-06
UR - https://www.deepdyve.com/lp/oxford-university-press/assessment-of-the-systems-approach-for-the-phytosanitary-treatment-of-db6hN5K1Xd
DP - DeepDyve
ER -