44 Imidacloprid Residues Five additional branchlet samples from each of the four branches were combined, generating ~25 g of hemlock foliage for imida- cloprid residues. The foliage was placed into plastic bags, and stored at -18°C until imidacloprid residues were run. Foliage was sent out to independent labs for residue testing. Foliage sent to Bayer Environmental Science determined the 2007–2008 imi- dacloprid residues; the USDA FS (Pineville, Louisiana, U.S.) conducted the 2009–2010 imidacloprid residues. Samples sent to Bayer were ground in liquid nitrogen and extracted using a modi- fied QuEChERS procedure prior to HPLC-MS/MS analysis (Le- hotay et al. 2010). Imidacloprid metabolites that have insecticidal activity include olefinic-, dihydroxy-, and hydroxy-imidacloprid (Sangha and Machemer 1992; Suchail et al. 2001). Imidaclo- prid, imidacloprid-olefin, and 5-OH imidacloprid residues were reported. Each sample was analyzed in duplicate, and the mean reported. Hemlock foliage sent for the 2009–2010 analyses con- ducted by the USDA FS were dried, ground in a Wiley mill prior to solvent extraction. For extraction, 1 g of dried hemlock foliage was placed in a 15 ml vial to which 10 ml of methanol was added, and the vial was agitated overnight on a shaker table. The sample was allowed to settle and supernatant drawn for ELISA analyses (Fischer et al. 2009). Polyclonal ELISA (EnviroLogix™ ; Portland, Maine, U.S.) kit was used in these analyses. Plate absorbance was read using a Biotek ELx808 plate reader (Biotek, Inc., Winooski, Vermont, U.S.) and results calculated using Biotek Gen5 soft- ware. Samples outside specified calibration of >6.0 ppb were di- luted with deionized water and reanalyzed. Each sample was ana- lyzed in duplicate, and the mean was reported. These (HPLC and ELISA) methods are similar; however the ELISA method is semi- quantitative for imidacloprid and will overestimate imidacloprid because reported residues include imidacloprid plus metabolites (Montfort et al. 1994; Fischer et al. 2009; Frank Byrne pers. comm.). Researchers report the “imidacloprid plus metabolites” of the HPLC and ELISA analyses as imidacloprid residues for the purposes of comparison. Two or three untreated hemlock sam- ples were used for residue analyses and were therefore excluded from statistical analyses. For imidacloprid efficacy, researchers used the published LC50 for HWA of 0.300 µg/g (Cowles et al. 2006) as a reference value. The means are presented in Table 1. Precipitation Data Onken (1994, unpublished data) reported a direct relation- ship between the amount of rainfall that trees receive in a given year and the amount of new growth in the following year. Monthly weather summaries were obtained from on- line data provided by the National Weather Service Fore- cast Office, Greenville-Spartanburg, South Carolina, U.S. (NOAA), ~80 km from the study site. Monthly total precipita- tion from 2006 to 2009 was compared to the norm (average). Statistical Analyses Data were transformed before carrying out analysis of variance (ANOVA) in order to achieve approximate homogeneity of vari- ances across treatments. The arcsine transformation was applied to the proportion of terminals showing growth. A log transformation was applied to HWA/cm and imidacloprid residues, the specific transformations being Log(HWA_cm + 0.005) and Log(Residue + 0.05), respectively, due to 0 values for each measurement. ©2012 International Society of Arboriculture Doccola et al.: Imidacloprid Treatments for HWA A repeated ANOVA was performed using Proc Mixed (SAS, Inc., Cary, North Carolina, U.S.) on the data across years for each of the three (transformed) responses. The fixed effects in the mod- el were Block, Treatment, Year, and Treatment*Year interaction. Based on preliminary analyses, Block*Treatment was included as an additional fixed effect for LogHWA only. To allow for serial correlations and heterogeneity of variances across years, the error covariance was assumed to be a first-order heterogeneous autore- gressive structure using the REPEATED statement with option ARH(1). Comparisons among treatment means averaged over years were performed if the Treatment*Year interaction was not important. If interaction was statistically significant and biologi- cally important, treatment comparisons were carried out sepa- rately for each year (in MINITAB Version 15, Minitab, Inc., State College, Pennsylvania, U.S.). Comparisons were carried out us- ing the protected LSD procedure at significance level 0.05. Trend analyses were plotted for each imidacloprid treatment applying a quadratic trend model, which was the best fit for the residue data. RESULTS The amount of time needed to make an application varied with the method used, although all three methods were rela- tively efficient. Mean application times were 2, 5, and 10 min- utes for the QUIK-jet, Kioritz, and Tree I.V., respectively. Growth Digital scans from 2007 to 2010 documented the temporal chang- es in tree condition (Figures 1). Year effects were highly signifi- cant (F = 231.24, DF = 3, 126, P < 0.001) reflecting an increase in growth over time for all treatments. Treatment and Year*Treatment interaction were both significant (F = 10.02, DF = 5, 41, P < 0.001 and F = 3.02, DF = 15, 126, P < 0.001, respectively). A graph of treatment means against year showed that trends were roughly similar across treatments except that the rate of increase was slower for the UTC and soil alone treatments, thus treatments comparisons are based on means across years (Table 1; Figure 2). Log HWA Treatment, Year and Treatment*Year interaction were all signifi- cant (F = 9.46, DF = 12.49, P < 0.001; F = 12.49, DF = 3,126, P < 0.001; and F = 2.14, DF = 15,126, P = 0.011, respective- ly). Block and Block*Treatment effects were also significant (F = 16.01, DF = 1, 36, P < 0.001 and F = 5.28, DF = 5, 36, P = 0.001, respectively) mainly because of some high values for soil treatment in Block II. Trends across years are similar for the treatments 2–5 (tree injections +/- soil), but different from UTC and imidacloprid soil treatment, so treatment compari- sons are reported separately for each year (Table 1). Means of HWA/cm across years for treatments are presented in Figure 3. Log Residue Year effects are highly significant (F = 147.36, DF =3, 105, P < 0.001), Treatment and Treatment*Year interaction are also sig- nificant (F = 12.52, DF = 4, 34, P < 0.001 and F = 7.38, DF = 12, 105, P < 0.001, respectively). Trends across years were not consistent across treatments and treatment comparisons are reported separately for each year (Table 1). Means for imida- cloprid residues for 70 to 1165 DAT are presented in Figure 4.
March 2012
| Title Name |
Pages |
Delete |
Url |
| Empty |
Ai generated response may be inaccurate.
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.
Downloading PDF
Generating your PDF, please wait...
This process might take longer please wait
