Arboriculture & Urban Forestry 41(2): March 2015 As part of our sample design, we placed each sampled tree into one of three categories. Level I trees were visually determined to be uninfested but within 0.5 km of a known infested tree (N = 387 trees); these trees could be considered as either negative or false negative for EAB. Level II trees were apparently infested for 1–2 years (N = 288 trees) and visually symptomatic with the presence of characteristic D-shaped adult emergence holes or frass-filled serpentine tunneling, if the tree was exposed due to bark loss. Level III trees were known to be infested for >2 years (N = 44 trees); some of these trees had already been marked for removal due to EAB. Infestation records for both Level II and III trees were based on information from local authorities, land managers, county extension agents, landowners, and state and federal records. For each tree surveyed, researchers recorded sev- eral signs and symptoms that are already associated with the presence of EAB (Table 1). The proportion of canopy thinning or loss was determined by esti- mating lost foliage from branches in relation to a full canopy (Smitley et al. 2008). Researchers also noted the presence of epicormic branching, bark loss, and cracks in scaffold branches. Trees were thoroughly inspected for the presence of distinctive EAB galleries and exit holes from the basal area up the stem and as far up as possible (to at least 2 m) and including the first scaffold branch if it occurred within that height. The stem was also divided into three sections (lower, middle, and upper third); the presence of branch fractures was noted. Because branches that fail leave stubs behind, the length of the remaining branch portion was also used to estimate the frac- ture point location on the branch. The fracture point was placed into one of three categories: 1) a collar breakpoint, if failure occurred near the union with the stem; 2) a center of gravity breakpoint, if branch failure occurred leaving a stub at least one-third the length of the branch; and 3) a tip, if the fracture occurred beyond the center of gravity and closer to one-third the distal end of the branch. Each tree was inspected from the ground in four cardinal directions and from a distance that gave full view of the tree. Statistical analyses were conducted using the epicalc package (Chongsuvivatwong 2012) in R (R Development Core Team 2012). In the epicalc pack- age, researchers estimated the risk ratio and corre- sponding 95% confidence intervals (CI) of a specific 105 symptom being associated with trees in which EAB was positively detected relative to trees in which EAB was not positively detected. A risk ratio whose confidence intervals include 1 would not indicate a significant association, as it would imply that the symptom is associated with trees that are both posi- tively or not positively identified as an EAB-infested tree. This statistical approach is analogous to epide- miological studies that address, for example, com- monality in foods consumed within a population, in which some individuals succumb to a food-borne outbreak. Thus, the use of this approach allowed the study authors to develop a set of symptoms that were likely to be associated with trees that ultimately were shown to be infested with EAB (from an anthropo- genic perspective) without quantifying the condi- tional probability of infestation given a symptom or suite of symptoms, which would be challenging given the potentially high rate of EAB false negatives. Table 1. Host tree symptoms and recorded values. Symptom Values Bark loss Epicormic branching Canopy loss (%) Scaffold crack Location of fractured branch Fracture point Presence or absence Presence or absence 0%–100% Presence or absence No fractured branch, lower one-third, middle third, or upper third Collar, center of gravity, or branch tip RESULTS The presence of bark loss (risk ratio = 2.7; 95% CI = 2.3, 3.2) and scaffold cracks (risk ratio = 2.5; 95% CI = 2.2, 2.9) were both significantly associated with trees positively identified with EAB (Figure 2A; Figure 2B). In addition, branch fractures only within the upper one-third of the canopy (risk ratio = 1.5, 95% CI = 1.2, 1.9) were significantly associ- ated with trees positively identified with EAB, while branch fractures in the lower third of the canopy (risk ratio = 0.7, 95% CI = 0.5, 0.9) were signifi- cantly associated with trees that were not positively identified with EAB (Figure 2C). Branch fractures in the middle third of the canopy (risk ratio = 1.2; 95% CI = 0.9, 1.5), or the absence of branch frac- tures (risk ratio = 1.0; 95% CI = 0.9, 1.1), were not significantly associated with either trees positively identified with EAB or not (risk ratio = 1.2; 95% CI = 0.9, 1.5). Lastly, branch fractures located closer to the union with the stem were significantly associ- ated with trees positively identified with EAB (risk ©2015 International Society of Arboriculture
March 2015
| Title Name |
Pages |
Delete |
Url |
| Empty |
Search Text Block
Page #page_num
#doc_title
Hi $receivername|$receiveremail,
$sendername|$senderemail wrote these comments for you:
$message
$sendername|$senderemail would like for you to view the following digital edition.
Please click on the page below to be directed to the digital edition:
$thumbnail$pagenum
$link$pagenum
Your form submission was a success. You will be contacted by Washington Gas with follow-up information regarding your request.
This process might take longer please wait
