TY - JOUR
AU - Evenden, Maya, L
AB - Abstract Wheat midge, Sitodiplosis mosellana Géhin, is an invasive pest of wheat, Triticum aestivum L. (Poaceae) throughout Canada and the United States. The applicability of available monitoring tools, including sex-pheromone baited traps, yellow sticky cards, and soil core sample surveys, in the northern-most agroecosystem of its invasive range has not been assessed. In this study, the attraction of male wheat midge to two Delta traps (green and orange) baited with one of three pheromone lures (a flex lure and two red septa lures from different sources) were compared. The efficacy of three yellow sticky cards (7 × 12 cm, 14 × 18 cm, and 14 × 18 cm rolled into a cylinder) for capture of male and female midge was assessed. Larvae were extracted from wheat heads sampled at the same sites to determine relationships with earlier adult trap capture. More male adult midges were captured in pheromone-baited traps with a greater surface area and in traps baited with the Scotts flex lure than the Great Lakes IPM septa lure, which had higher and more variable pheromone release rates. The smaller yellow sticky cards captured more male and female midges than the larger yellow sticky cards, regardless of shape. The number of female midges captured on yellow sticky cards predicted the number of larvae in wheat heads. The number of male midges captured in pheromone-baited traps did not predict larval density. Relationships were found between the number of overwintering cocoons recovered in soil core samples and emerging midges the following spring. Integrated Pest Management, pest monitoring, agriculture, pheromone Integrated Pest Management (IPM) programs are routinely developed and implemented to manage agricultural, forestry, and urban systems (van Lenteren and Woets 1988, Swanton and Weise 1991, Hobbs and Humphries 1995, Lewis et al. 1997, Kogan 1998, Thomas 1999). The basis of IPM programs is the accurate detection of target pest populations to then determine the risk of damage. Monitoring tools in insect pest management include sweep net samples, stationary traps, and visual in-field assessments for either the target insect or characteristic damage. Traps coated with an adhesive agent are relatively inexpensive and capture target insects as they move through the environment. Capture of target pests in unbaited sticky traps has been used to develop action thresholds in many agricultural systems (Prokopy et al. 1982, Stanley et al. 1987, Bechinski et al. 1989, Shipp et al. 2000). Some pest attraction or aversion due to trap color can impact trap catch on unbaited sticky cards (Kirk 1984, Brach and Trimble 1985), but sticky cards can provide less-biased sampling than other trap types through passive pest interception (Southwood 1978). Sticky cards are cheap but can be difficult as management tools due to high bycatch, which interferes with target insect identification by growers and pest managers. Increased capture of target insects or exclusion of non-target bycatch can be manipulated by altering trap color or position (Yee 2013, Chavalle et al. 2019). The supernormal stimulus of yellow sticky traps mimics plant foliage and attracts many herbivorous insects (Prokopy and Owens 1983). The sensitivity and specificity of unbaited traps can be enhanced with the addition of attractants. Traps baited with semiochemicals can attract target insects in the managed ecosystem. Species-specific semiochemicals, such as sex pheromones, can attract target insects even at low population densities and/or when insects are patchily distributed in the habitat (Witzgall et al. 2010). Trap capture in sex-pheromone baited monitoring traps is used for insect pest detection, to develop economic thresholds, and to time control efforts (Witzgall et al. 2010). Many sex pheromones are produced and released by females, and capture of males may not accurately indicate the breeding population and reproductive potential of an insect population (Witzgall et al. 2010). Wheat midge, Sitodiplosis mosellana Géhin, is an invasive pest of wheat, Triticum aestivum L. (Poaceae), in North America. In North Dakota, greater than US$27 million in yield losses were attributed to wheat midge in the mid-1990s (Knodel and Ganehiarachchi 2008). This pest has recently expanded its range into northwestern Alberta (>56°N) to the Peace River region (Western Committee on Crop Pests 2011). Different monitoring tools, including yellow sticky cards and traps, are available to monitor wheat midge, but these tools have not been directly compared and have not been tested in the northern-most region of the wheat midge distribution in North America. Wheat midge larvae feed on the developing ovaries and seed of wheat plants, which decreases grain weight and causes seed abortion if feeding pressure is strong (Lamb et al. 2000). Wheat midge damage not only lowers crop yield but also decreases grain milling quality and grade (Miller and Halton 1961, Dexter and Edwards 2012). Seed infestation of 6% causes quality loss of harvested wheat, whereas 11% seed infestation causes yield loss (Lamb et al. 2000). A threshold based on seed infestation, however, cannot be used to make control decisions because larvae are concealed inside the developing wheat head and are protected from control methods (Ding and Lamb 1999). Wheat midge adults are monitored in the field using yellow sticky cards, sex-pheromone-baited Delta traps, and by scouting fields visually. An action threshold of the capture of four adult midges on 10 yellow sticky cards (each 7.5 × 12.5 cm) over 3 d (Lamb et al. 2002) is used in Manitoba, but effectiveness in other regions varies. In Belgium, yellow pan traps are equally effective for capture of wheat midge as yellow sticky cards (Chavalle et al. 2019), and are recommended to monitor density. Control decisions for wheat midge in Canada are based on in-field counts of adult females on wheat heads performed at dusk. Counts are averaged across multiple locations in the field. In Canada, the action threshold is one midge per 8–10 heads for quality loss and two midges per 8–10 heads for yield loss (Elliott et al. 2002). In Germany, one adult female per three heads is the damage threshold (Basedow and Schütte 1973). Typically, wheat midge fly and lay eggs while wheat heads are emerging from the flag leaf collar. If flight occurs after most wheat heads are flowering, control is not recommended as it is often ineffective and unnecessary. At this stage, most larvae are sheltered inside the wheat head, and larvae that begin development after wheat anthesis have lowered survival (Ding and Lamb 1999). Flight phenology of wheat midge is monitored throughout the Canadian Prairie Provinces (Alberta, Saskatchewan, and Manitoba) with Delta traps baited with a 1 mg racemic mixture of the female-produced sex pheromone, 2,7 nonanediyl dibutyrate (Gries et al. 2000). Male midge capture in sex pheromone-baited traps correlates to larval infestation levels in the United Kingdom (Bruce et al. 2007). In addition, risk forecast maps based on annual fall soil core surveys are produced annually in the Canadian prairies. These maps use the number of overwintering midge cocoons (Doane et al. 1987) in 17 soil cores (1.9 cm diameter × 10.8 cm depth) sampled from wheat fields (Doane et al. 2000). Doane et al. (2000) recommend six larvae per 100 cm2 of soil sampled as a threshold for potential economic infestation. Soil core sampling of cocoons, however, did not predict infestation in Manitoba (Lamb et al. 1999), the United Kingdom (Oakley et al. 1998) or in Germany (Basedow and Schütte 1974). The ability of soil core sampling to predict infestation has not been tested in Northern Alberta. Due to the expansion of wheat midge into the Peace River region of Alberta and the potential for further northwestern expansion, current monitoring techniques were assessed for application in northern regions. Daylength, seasonal and daily temperatures, the length of growing season, and growing practices differ across regions and latitudes. These climatic and cultural differences can impact the effectiveness of monitoring tools and the management goals of producers. In general, the efficacy of monitoring tools for wheat midge in Canada needs to be compared for the development of an economic threshold. The overall objective of this research is to test the sensitivity and predictive ability of multiple monitoring tools for wheat midge in the Peace River region of Alberta, Canada. Materials and Methods Comparisons of Monitoring Techniques for Adult Wheat Midge The effectiveness of different monitoring tools for adult wheat midge was tested in commercial wheat fields (each approximately 160 acres) in the Peace River region of Alberta, Canada (55° to 56°N by −116° to −118°W). Different yellow sticky cards and different pheromone lures and traps were compared in separate experiments. Field sites were selected that were seeded to wheat cultivars susceptible to wheat midge, rather than ‘midge tolerant’ varietal blends, during the monitoring year. Capture of adult male and female wheat midge was compared on different unbaited sticky cards. Fields (n = 9, Supp Table S1 [online only]) were monitored throughout the wheat midge flight in both years of the study (22 June to 23 August 2017, and 18 June to 9 August 2018). Three different types of yellow sticky cards were compared: 1) a large flat card (Alpha Scents, West Linn, OR, 18 × 14 cm); 2) a large flat card (Alpha Scents, 18 × 14 cm) rolled into a cylinder; and 3) a small flat card (Great Lakes IPM, Vestaburg, MI, 12 × 7 cm). The total surface area available for trapping was equalized by varying the number of traps per treatment in each field. To determine if trap capture differences were due to trap type or the number of traps in a field, an additional statistical comparison was done using only the adjacent cards of each type at three locations per field (a total of three cards of each type). Yellow sticky cards were positioned along one north-south transect at each field margin (1 m from field edge), with surfaces facing east-west because of predominantly west winds in region. Ten small flat cards were positioned 10 m apart along each transect. One large flat and one large rolled card were positioned 1 m apart adjacent to the third, sixth, and ninth small cards from the north end of each transect (Fig. 1). Traps were collected and replaced, weather permitting, every 4–10 d. Traps were examined under a dissecting stereomicroscope (Zeiss stemi-C, Oberkochen, Germany) at 8× magnification. Wheat midge adults were identified (Harris 1966), counted, and separated by sex. Fig. 1. Open in new tabDownload slide Arrangement of yellow sticky cards along field edge of commercial wheat fields. Traps were placed in the ditch adjacent to fields in a north-south transect. Ten small flat cards were positioned 10 m apart along each transect. One large flat and one large rolled card were positioned adjacent to the third, sixth, and ninth small cards from the north end of each transect. Fig. 1. Open in new tabDownload slide Arrangement of yellow sticky cards along field edge of commercial wheat fields. Traps were placed in the ditch adjacent to fields in a north-south transect. Ten small flat cards were positioned 10 m apart along each transect. One large flat and one large rolled card were positioned adjacent to the third, sixth, and ninth small cards from the north end of each transect. A second experiment tested capture of adult male wheat midge in different pheromone-baited traps. Traps tested included two historically used traps: green and orange Delta traps (Scotts and Scentry) baited with either a blue flex lure (Scotts, 100 μg pheromone), a red rubber septa lure (Scotts, 1 mg racemic pheromone) or a red rubber septa lure (Great Lakes IPM, 1 mg racemic pheromone), combinations were organized in a full-factorial randomized block design with each commercial site treated as a block. Traps were positioned 50 m apart along a north-south transect at each of 16 field margins (Supp Table S1 [online only]). Traps were collected and replaced, weather permitting, every 4–14 d from 20 June until 17 August in both years of the study (2016–2017). Lures were not changed during the monitoring period and remained at the same randomized position assigned at the beginning of monitoring. Traps were examined under a dissecting stereomicroscope (Zeiss stemi-C) at 8× magnification. Wheat midge adults were identified (Harris 1966), counted, and separated by sex. The pheromone release rate of each of the three lure treatments (Scotts flex, Scotts septa, and Great Lakes IPM septa) was measured through aeration of lures in the lab at ambient temperature. Lures (three per treatment) were aerated individually inside glass chambers (5 × 10 cm OD) at 23.5 ± 0.5°C for 3 d. Air was drawn through the chambers using Dyna pumps (A.O. Smith, Tipp City, OH, USA) at a rate of 0.4 liter/min into a Porapak Q (50–80 mesh; Supelco—part of Sigma–Aldrich, Canada, Oakville, ON) adsorbent material. The Porapak was flushed daily using 2 ml of 50:50 blend of pentane and diethylether with an internal standard of 500 ng Dodecyl-Acetate. Porapak-q extracts were analyzed with a 5890 Hewlett Packard Gas Chromatograph (GC) (Hewlett-Packard, now Agilent Technologies, Santa Clara, CA) with a DB-5 (30 m × 0.32 mm ID) column (Agilent J&W columns, Agilent Technologies). The GC oven was operated with the following program: 50°C held for 1 min, temperature was then increased to 280°C by a rate of 10°C per min. The injector port was set to 250°C, and the flame ionization detector was kept at 280°C. Area counts of the peak of the compound of interest were compared to the area counts with the known amount of the internal standard in order to quantify the amount of pheromone in Porapak extracts. Differences in diffuse spectral reflectance between the two types of yellow sticky cards and the two Delta traps were measured using a dual-beam spectrophotometer operating from 2500 to 250 nm (Cary 500 G UV–Vis–NIR, Varian, Inc., Mississauga, ON). Reflectance was measured in 1 nm intervals. Measurements were calibrated using a 99% Spectralon reflectance standard (Labsphere, North Sutton, NH). Comparisons of Adult Capture to Larval Densities To determine whether adult wheat midge capture in traps can predict larval population density, wheat midge larvae in the subsequent generation were sampled at the same field sites where adult monitoring with unbaited yellow sticky cards and pheromone-baited traps took place. To sample larvae, wheat heads were collected in mid-August of both study years when plants were between early milk (73) and early dough stages (83) (Zadoks et al. 1974). Larval sampling occurred between 2 August and 31 August, depending on growth stages in fields. At sites where sticky card monitoring took place, five samples of 50 wheat heads each were collected along one north-south transect at each field, adjacent to every second sticky card trap (Fig. 1). Wheat heads were clipped 5–10 cm below the head and placed into paper bags for transport to the laboratory for processing. At sites where pheromone-baited trap monitoring took place, 24 samples of 50 wheat heads each were collected along three north-south transects at each field: along the field margin in proximity to the pheromone-baited traps, and 20 and 50 m into the field from the margin. Eight samples were collected along each transect at points parallel to the eight pheromone traps for a total of 24 samples per field. Wheat heads were broken apart with a single-head thresher (Almaco specialized equipment model SV SRE-2, 99 M Ave, Nevada, IA), with the air turned off and vents covered to retain larvae. Samples were hand sorted and the number of larvae per sample was counted. Five heads from each sample were randomly selected for hand-dissection to determine the accuracy of samples processed with the single-head thresher. Since there was no effect of processing method (mixed model analysis of variance [ANOVA] with method as a fixed factor and site as a random factor, χ 2 = 1.210, df = 1, P = 0.271), larval densities used for regression analyses were determined from the remaining 45 heads processed with the single-head thresher. Comparison of Soil Core Samples to Subsequent Adult Wheat Midge Emergence Soil core samples to estimate density of overwintering wheat midge cocoons were collected in commercial wheat fields and compared to subsequent adult wheat midge capture in emergence traps positioned in the same fields the following spring. In 2015 and 2017, four field sites in the Peace River region were monitored to assess cocoons using soil core samples, while five field sites were monitored in 2016 (Supp Table S1 [online only]). Soil cores were collected in the fall before field cultivation (between 25 and 30 September 2015, 3 October and 12 October 2016, and 11 October and 9 November 2017). Soil cores were collected using a steel corer (1.9 cm diameter × 10.8 cm depth). A total of 17 cores (Doane et al. 2000) were collected at each site, positioned 20 m apart along an ‘M’ pattern starting 20 m into the field from the margin. Soil samples were wet sieved (Doane et al. 1987) and cocoons were dissected to determine parasitism rates. The number of viable cocoons was determined by subtracting the number of parasitized and dead larvae from the total number of cocoons collected at each site. The number of viable cocoons was divided by surface area of the soil sampled to determine cocoons per m2. Emergence traps were positioned at these same sites the following spring (Supp Table S1 [online only]). Emergence traps consisted of 2-liter plastic pails (Plastipak) with the bottoms cut off (Doane and Olfert 2008). Each bucket had ventilation holes lined with fine mesh cut into the sides 3 × 2.54 cm in diameter) and in the lid (1 × 5.08 cm in diameter). Emerging midges and parasitoids were collected on a single clear polyethylene strip coated with Stikem special (Scotts) (14 cm × 5 cm sticky surface) that was secured under the lid of each trap. Traps were placed with the bottom 7 cm of each bucket buried in the soil. Emergence traps were checked three times weekly between 20 June and 5 August at four sites in 2016, between 6 June and 9 August at five sites in 2017 and four sites in 2018. Traps were positioned 10 m apart along a north–south transect within the field along the first row of wheat stubble from the margin. In 2016, eight traps were placed in each field. Due to low midge capture in emergence traps in 2016, 20 traps were placed in each field in 2017 and 2018. The number of adults captured in emergence traps was totaled per site and divided by the total surface area sampled by all traps per field. Statistical Analyses All analyses were done in R studio 3.13 (R Core Team 2016). All linear mixed models were fit with package lme4 (Bates et al. 2015). Assumptions of normality and heteroscedasticity were tested visually by examining q–q plots, residual plots, and by conducting Shapiro-Wilks tests on model residuals. Distributions of data were examined visually using histograms and q–q plots against a normal distribution, and different model variance structures were evaluated by comparing AIC values and determinations of residual normality. Nonsignificant interactions were removed from models if ANOVA comparisons between models with and without the interaction were not significant. Type II Wald-χ 2 tests were done on GLMM’s following model selection using the car package (Fox and Weisberg 2011), followed by post-hoc Tukey tests using the package multcomp (Hothorn et al. 2008). Comparisons of Monitoring Techniques for Adult Wheat Midge The capture of adult midges on the different yellow sticky cards was totaled per trap type, site and over the season. Generalized linear mixed models for total male (Gaussian error distribution) and total female (Poisson error distribution) capture included card type, year, and their interaction as a fixed factor and site as a random factor. To determine if trap capture differences were due to trap type or the number of traps in a field (10 small flat and three large flat and rolled cards per field), total female capture was modeled additionally using capture on only three of each trap type at each site for 2017. Total counts were converted to rates of capture per cm2 trapping surface on each trap type by including an offset of the log of the cm2 in the model (O’Hara and Kotze 2010). The log value was included, as a Poisson or negative binomial regression model: log (y) = x + log (e) simplifies to log (y/e) = x. The repeated measures generalized linear mixed model included trap type as a fixed factor and trap ID and site as random factors and had a Poisson error distribution. Season total trap capture of male midges was compared using generalized linear mixed models with lure and card type as fixed factors and site as a random factor. Error distributions were specified as gamma with a log link in 2016 and as negative binomial in 2017. Analysis was repeated with total trap capture standardized for trap surface area as above. These models were structured as above with the same error distributions, except the 2016 data was fit with a negative binomial distribution. Release rate data of the different lures tested in the field was log-transformed to meet the assumption of normality as based on Shapiro-Wilks test (W = 0.674, P < 0.001). Data were analyzed with a repeated-measures linear mixed model with lure type as a fixed factor and date of aeration and subject as random factors. The variance-covariance structure was general positive-definite with log-Cholesky parametrization. Comparisons of Adult Monitoring Tools to Larval Densities The relationship between adult female midge capture on yellow sticky cards and larvae in the subsequent generation was examined with a generalized linear mixed model for 2017 and 2018. Female midge capture on the different card treatments was totaled per site over the growing season. The relationship between adult female capture at each site and the mean number of larvae sampled per site was compared. Collection year was included as a random factor. The relationship between season total adult male midge capture in pheromone-baited traps and larvae in the subsequent generation was examined with repeated measures generalized linear mixed models for 2016 and 2017. Season total male midge capture was used as results were similar between models for season total and models for total midge capture until mid-July, when wheat heads had finished emerging. Transect was included as a fixed factor and site was included as a random factor. For 2017 data, location along the north-south transect was included as a random factor. For 2016 data, the number of larvae was totaled per transect due to overdispersion and poor model fit. Total adult counts were log (x+1) transformed and the model was fit with a Poisson error distribution. Differences among transects were analyzed with a post-hoc Tukey test to determine whether there was variation in distribution of larvae within fields. For both yellow sticky card and pheromone-trap analyses, conditional and marginal R2 values were calculated as in Nakagawa and Schielzeth (2013) using the muMIn package (Barton 2018). Comparison of Soil Core Samples to Subsequent Adult Wheat Midge Emergence The relationship between the number of cocoons per m2 and the number of adults per m2 from emergence traps was analyzed with a linear model. The number of adults was included as a fixed factor, while year was included as a random factor with both random intercepts and random slopes. Marginal and conditional r2 values were calculated as above. Results Comparisons of Monitoring Techniques for Adult Wheat Midge There was a significant interaction between year and trap (χ 2 = 16.17, df = 2, P < 0.001) that affected the number of female midges captured on yellow sticky cards. In 2017, the smaller yellow sticky cards captured more female midges than the larger yellow cards arranged in either a flat or rolled configuration (Fig. 2). In 2018 (at very low midge densities), there were no differences in female midge capture between the large flat and small flat cards. Overall, the type of sticky card influenced season total capture of female midges (χ 2 = 207.77, df = 2, P < 0.001). The analysis of male wheat midge capture is not reported due to low total numbers of males captured on yellow sticky cards. Fig. 2. Open in new tabDownload slide Effect of yellow sticky card type on season total number of female midges captured (16 cards per site). The large flat cards, large rolled cards, and the small flat cards were monitored in commercial wheat fields (four in 2017 and four in 2018). Sixteen traps were positioned along the field margin of each field. The dark midline represents the median and the bottom and top of the box indicate the first and third quartiles, respectively. The vertical lines, or whiskers, indicate the maximum value, or 1.5 times the interquartile range, whichever is smaller. Outliers are represented by points above the whiskers. Fig. 2. Open in new tabDownload slide Effect of yellow sticky card type on season total number of female midges captured (16 cards per site). The large flat cards, large rolled cards, and the small flat cards were monitored in commercial wheat fields (four in 2017 and four in 2018). Sixteen traps were positioned along the field margin of each field. The dark midline represents the median and the bottom and top of the box indicate the first and third quartiles, respectively. The vertical lines, or whiskers, indicate the maximum value, or 1.5 times the interquartile range, whichever is smaller. Outliers are represented by points above the whiskers. To determine whether the difference in capture of females was due to the number of cards per field, counts from only the three cards adjacent to the larger cards were also compared in 2017. The analysis using only three small cards indicated that more female midges were captured on small cards than either configuration of the large card (χ 2 = 20.81, df = 2, P < 0.001). Trap color affected the number of adult male midge captured in the differently colored pheromone traps. Marginally more male wheat midges were captured on orange than green traps in 2016 (χ 2 = 3.58, df = 1, P = 0.058) and significantly more in 2017 (χ 2 = 12.28, df = 1, P < 0.001). When trapping surface area was added to the model, however, there was no effect of trap color on male wheat midge capture in 2016 (χ 2 = 0.64, df = 1, P = 0.425) or 2017 (χ 2 = 0.364, df = 1, P = 0.546). As would be expected, more midges were captured in pheromone-baited traps than in unbaited control traps in both 2016 (χ 2 = 333.21, df = 3, P < 0.001) and 2017 (χ 2 = 349.87, df = 3, P < 0.001). In 2016, all lure types attracted a similar number of male midges (Fig. 3). In 2017, traps baited with the flex lures captured more male midges compared to traps baited with the rubber septa lures from Great Lakes IPM (Fig. 3). Fig. 3. Open in new tabDownload slide Season total number of male midges captured in Delta traps (green or orange) baited with either a rubber septa lure (Great Lakes IPM or Scotts), a flex lure (Scotts) or no lure (control). Traps were positioned in commercial fields (eight in 2016 and nine in 2017) in the Peace River region of AB during wheat midge flight and replaced weekly. Letters indicate significant differences among lure treatments within year (Post-hoc Tukey tests, α = 0.05). The dark midline represents the median and the bottom and top of the box indicate the first and third quartiles, respectively. The vertical lines, or whiskers, indicate the maximum value, or 1.5 times the interquartile range, whichever is smaller. Outliers are represented by points above the whiskers, with the number of midges indicated for extreme outliers. Note: y-axes for 2016 and 2017 have different scales. Fig. 3. Open in new tabDownload slide Season total number of male midges captured in Delta traps (green or orange) baited with either a rubber septa lure (Great Lakes IPM or Scotts), a flex lure (Scotts) or no lure (control). Traps were positioned in commercial fields (eight in 2016 and nine in 2017) in the Peace River region of AB during wheat midge flight and replaced weekly. Letters indicate significant differences among lure treatments within year (Post-hoc Tukey tests, α = 0.05). The dark midline represents the median and the bottom and top of the box indicate the first and third quartiles, respectively. The vertical lines, or whiskers, indicate the maximum value, or 1.5 times the interquartile range, whichever is smaller. Outliers are represented by points above the whiskers, with the number of midges indicated for extreme outliers. Note: y-axes for 2016 and 2017 have different scales. The reflectance spectra of the large and small yellow sticky cards differed in the infrared and visible ranges. The small card had a lower percent reflectance throughout the visible range than the large and had some reflectance in the UV range (Fig. 4). The orange and green Delta traps had distinct spectra in the visible range, particularly in the 300 nm and 500 nm ranges (Fig. 4). Fig. 4. Open in new tabDownload slide Diffuse spectral reflectance measured from different traps tested for adult wheat midge capture. Reflectance was measured at 1 nm intervals from 2500 nm to 250 nm using a dual-beam spectrophotometer. Measurements were calibrated using a 99% Spectralon reflectance standard. Fig. 4. Open in new tabDownload slide Diffuse spectral reflectance measured from different traps tested for adult wheat midge capture. Reflectance was measured at 1 nm intervals from 2500 nm to 250 nm using a dual-beam spectrophotometer. Measurements were calibrated using a 99% Spectralon reflectance standard. The red rubber septa lures from Great Lakes IPM released more pheromone than Scotts flex lures (χ 2 = 7.26, df = 2, P = 0.026). There was high variability in the amount of pheromone released from the septa lures over and among the 3 d aeration from both Great Lakes IPM (mean on day 1 = 461 ng ± 536 SD, mean on day 2 = 807 ng ± 768 SD, mean on day 3 = 1159 ng ± 1226 SD) and Scotts (mean on day 1 = 169 ng ± 129 SD, mean on day 2 = 618 ± 879 SD, mean on day 3 = 533 ng ± 738 SD) compared to the flex lures (mean on day 1 = 37 ng ± 20 SD, mean on day 2 = 43 ng ± 9 SD, mean on day 3 = 43 ng ± 18 SD) (Fig. 5). Fig. 5. Open in new tabDownload slide Average amount of wheat midge pheromone (2,7-nonanediyl dibutyrate) released (log ng) over 3 d from: Scotts flex lure, Scotts septa lure, and Great Lakes IPM septa lures (three lures per treatment). Vertical lines represent standard error of the mean. Fig. 5. Open in new tabDownload slide Average amount of wheat midge pheromone (2,7-nonanediyl dibutyrate) released (log ng) over 3 d from: Scotts flex lure, Scotts septa lure, and Great Lakes IPM septa lures (three lures per treatment). Vertical lines represent standard error of the mean. Comparisons of Adult Capture to Larval Densities The season total number of adult female midges captured on the smaller yellow sticky cards (10) cards was significantly related to the number of larvae found in wheat heads in the same sites (marginal r2 = 0.682, χ 2 = 15.01, df = 1, P < 0.001). This relationship, however, was driven by a single site (Fig. 6). Fig. 6. Open in new tabDownload slide Relationship between the number of female wheat midges captured on small yellow sticky cards and the subsequent number of larvae collected in wheat head samples. Larvae were collected from wheat head samples (five samples of 50 heads each from a total of nine field sites) in 2017 and 2018. The solid line represents the overall fit of the generalized linear mixed model. The cumulative number of female midges captured on small yellow cards was modeled for each site and included as a fixed factor. Fig. 6. Open in new tabDownload slide Relationship between the number of female wheat midges captured on small yellow sticky cards and the subsequent number of larvae collected in wheat head samples. Larvae were collected from wheat head samples (five samples of 50 heads each from a total of nine field sites) in 2017 and 2018. The solid line represents the overall fit of the generalized linear mixed model. The cumulative number of female midges captured on small yellow cards was modeled for each site and included as a fixed factor. The season total number of adult male midges captured in pheromone-baited traps (six per site) did not predict the number of larvae in wheat heads at the same site in 2016 (marginal r2 = 0.08, χ 2 = 0.13, df = 1, P = 0.713) or 2017 (marginal r2 = 0.004, χ 2 = 0.01, df = 1, P = 0.903). This result was consistent with models of 2016 between cumulative male midge counts until mid-July (General linear mixed regression, marginal r2 = 0.09, χ 2 = 0.20, df = 1, P = 0.655) and subsequent larval counts. Similarly, models for 2017 cumulative counts of male midges until mid-July (General linear mixed regression, marginal r2 = 0.003, χ 2 = 0.026 df = 1, P = 0.732) and larval counts were not significant. In 2016, there was a significant effect of the location of the sampling transect on the number of larvae recovered in wheat head samples (χ 2 = 4190.25, df = 2, P < 0.001). The number of larvae found in samples decreased with distance into the field (Fig. 7). In 2017, there was no difference in the number of larvae collected in each of the different sampling transects (χ 2 = 2.64, df = 1, P = 0.104), although there was a weak trend for more larvae in samples from the field edge compared to 50 m into the field (Fig. 7). Fig. 7. Open in new tabDownload slide Total number of larvae in wheat head samples collected along different transects along the field edge, 20 m into the field from the edge, and 50 m into the field from the edge. In 2016, 24 samples of 50 heads each were collected from each of eight fields. In 2017, 16 samples of 50 heads each were processed from each of nine fields. In 2017, samples from the 20 m transect were not processed. Letters above the plot indicate significant differences among groups (post-hoc Tukey test, α = 0.05). The dark midline represents the median, and the bottom and top of the box indicate the first and third quartiles, respectively. The vertical lines, or whiskers, indicate the maximum value, or 1.5 times the interquartile range, whichever is smaller. Outliers are represented by points above the whiskers. Note: y-axes have different scales. Fig. 7. Open in new tabDownload slide Total number of larvae in wheat head samples collected along different transects along the field edge, 20 m into the field from the edge, and 50 m into the field from the edge. In 2016, 24 samples of 50 heads each were collected from each of eight fields. In 2017, 16 samples of 50 heads each were processed from each of nine fields. In 2017, samples from the 20 m transect were not processed. Letters above the plot indicate significant differences among groups (post-hoc Tukey test, α = 0.05). The dark midline represents the median, and the bottom and top of the box indicate the first and third quartiles, respectively. The vertical lines, or whiskers, indicate the maximum value, or 1.5 times the interquartile range, whichever is smaller. Outliers are represented by points above the whiskers. Note: y-axes have different scales. Comparison of soil core samples to subsequent adult wheat midge emergence The number of cocoons per m2 was significantly related to the number of adult midges per m2 found in emergence traps the subsequent spring (marginal r2 = 0.469, χ 2 = 11.246, df = 1, P < 0.001, Fig. 8). Fig. 8. Open in new tabDownload slide Relationships between the number of cocoons collected per m2 of soil and the total number of midge adults per m2 captured in emergence traps at the same fields during the following season. Soil core samples (17 cores per field site) were collected from wheat fields following harvest and monitored with emergence traps the following spring (eight traps per field in 2016, and 20 traps per field in 2017 and 2018). Different lines represent linear regression best fit lines for the relationship during the different years. Mixed model regression was fit with a random intercept and random slope model for different collection years. Fig. 8. Open in new tabDownload slide Relationships between the number of cocoons collected per m2 of soil and the total number of midge adults per m2 captured in emergence traps at the same fields during the following season. Soil core samples (17 cores per field site) were collected from wheat fields following harvest and monitored with emergence traps the following spring (eight traps per field in 2016, and 20 traps per field in 2017 and 2018). Different lines represent linear regression best fit lines for the relationship during the different years. Mixed model regression was fit with a random intercept and random slope model for different collection years. Discussion Wheat midge continues to be a difficult pest to manage. Current economic thresholds are based on in-field counts of midges at dusk (Elliott et al. 2002), which occurs later than 10 pm in the Peace River Region during wheat midge flight. Accurate identification of wheat midge is difficult in the field, as they average 2.5–3.0 mm in wing length (Harris 1966). The development of economic thresholds based on trap capture would provide an accurate and more appealing tool for managers and producers to use. The tools tested in this study are useful to monitor wheat midge activity, but further work is needed to relate midges sampled to economic thresholds. Capture of female midges on yellow sticky cards was indicative of densities of midge larvae in the same fields, although the strong relationship was mainly driven by a single data point. Further research is needed at high and mid-range populations in order to validate this relationship. Economic thresholds developed in Manitoba recommend an action threshold of four or more adult midges per 10 yellow sticky cards to control populations above the larval threshold of two larvae per wheat spike (Lamb et al. 2002). Magnification and illumination of wheat midge adults on yellow sticky cards allows for easier and more accurate identification of wheat midge adults than direct counts conducted under low light conditions in the field. Additionally, physical capture of insects allows for confirmation of species identification. Yellow sticky cards also captured the wheat midge parasitoid, Macroglenes penetrans Kirby (Hymenoptera: Pteromalidae) and this could allow producers to monitor for natural enemies and include conservation of this important parasitoid in their management decisions. Yellow sticky card traps may accurately indicate the reproductive population of wheat midges as they captured mostly females. Yellow sticky cards accurately represented the sex ratio of wheat midge moving into the field: in this study, 97% of the midges captured on the yellow sticky cards were female. In Manitoba and southeastern Alberta, at least 95% of wheat midges entering wheat fields are female (Smith et al. 2007). While sex pheromone-baited traps capture mostly males (99% of wheat midges captured in the current study were male), they were more sensitive than yellow sticky cards and captured a mean of 17 times more midge per cm2 than yellow sticky cards at nearby sites. In Belgium, pheromone-baited traps captured 100 to 1,000 times more wheat midges than sticky or pan traps (Chavalle et al. 2019). The pheromone-baited traps, however, did not indicate low or high populations of midge larvae sampled at the same sites and can only be used to monitor flight phenology. Monitoring seasonal flight phenology may not be a cost-effective way to determine when to conduct in-field scouting, however, as wheat is only susceptible to wheat midge during a short window during wheat head emergence. Differences in capture on the different types of yellow sticky cards tested may be due to differences in color. The large cards have a black grid pattern printed on the plastic while the small cards are plain yellow, and the yellow background hue differs between the two cards. The large card has a higher spectral reflectance through the red and orange regions of the visible range than the small card, and the small card may have some reflectance in the UV range. There are also small differences in the rate at which reflectance decreases at 500 nm, with the large card having a more rapid decrease than the small card. These differences in spectral reflectance between the two cards may be driving the differences in midge capture; however, more research is needed to characterize the visual preferences of female wheat midge. There is evidence that color impacts the orientation of wheat midge adults as female wheat midges lay more eggs on wheat heads placed against a contrasting background than on a similarly colored background (Gharalari 2008). The smaller yellow sticky cards are cheaper than the large cards (US$0.35 compared to US$1.39), making them more attractive to agronomists and producers as a widespread management tool. The green or orange color of pheromone traps did not impact male wheat midge capture in the Peace River region of Alberta. For male wheat midges, olfactory cues may be more important than visual cues for mate location, or visual cues may be important in the absence of olfactory cues. The importance of different colors or hues within the green–red spectrum to wheat midge orientation is unknown, as is the relative importance of olfactory and visual cues. Understanding the point at which color becomes important in host or mate location would benefit trap optimization. Although the pheromone release rate from septa lures was far greater than that from the flex lures, fewer male midges were captured in traps baited with septa than flex lures. The septa lures are loaded with a higher dose of pheromone (1 mg) compared to the flex lures (100 μg). It is unlikely that the higher dose is inhibitory, as dose-dependent responses of male wheat midge to pheromone in other studies show no upper threshold (Bruce et al. 2007). Male wheat midges orient to traps baited with lures that release up to 137 μg of pheromone per day, although a release rate of 0.38 μg per day is sufficient for wheat midge monitoring (Bruce et al. 2007). In this study, the red rubber septa lure released a mean of 0.809 μg of pheromone per day, while the blue flex lure released 0.041 μg of pheromone per day, but were still effective for male wheat midge in the Peace River region. Release of the wheat midge pheromone from polyethylene vials increases >3× by increasing lure loadings from 5 to 10 mg (Bruce et al. 2007). In the United Kingdom, rubber septa lures loaded with 1 or 5 mg of pheromone performed as well as polyethylene vials loaded with 5 mg of pheromone over a 1-mo period; however, rubber septa lures from different sources, in different environmental conditions or over different time periods may perform differently. Seasonal changes in pheromone release from lures likely did not cause the trap capture differences: trends among lures were consistent over the trapping period. Additionally, no differences were found in trap capture between new septa lures and septa lures aged in the field for 2 wk (post-hoc Tukey, Z-value = 0.218, P = 0.996) or 4 wk (post-hoc Tukey, Z-value = 0.532, P = 0.950) (A. Jorgensen et al. unpublished data). High variation in release rates among lures could cause inaccurate indications of wheat midge activity, which could delay management activities and cause unnecessary damage. Pheromone traps used for monitoring wheat midge activity need to be sensitive enough to indicate the start of emergence, so that producers can time in-field scouting. There were no significant relationships between pheromone trap capture and the number of wheat midge larvae in wheat heads. Bruce et al. (2007) found a significant relationship between capture of male wheat midges in Delta traps baited with pheromone released from polyethylene vials and larval infestations, but also found high variability in capture among fields only 100 m apart. In Saskatchewan, Mirciou (2004) found a significant relationship between male midges captured in green Delta traps baited with pheromone released from rubber septa and larval infestations only in wheat fields seeded on wheat stubble. In the Peace River region, many producers rotate wheat and canola on a 2-yr cycle. Sex pheromone traps may not detect potential sites of infestation because wheat midge males mate at the emergence site and females travel to oviposition sites (Smith et al. 2007). It is not known if males disperse after mating, but male midges fly an average of 400 m on flight mills (Hao et al. 2013). In 2016, more wheat midge larvae were present in samples collected at the field edge than at 20 or 50 m into the field, and in 2017 distributions of larvae followed the same trend, but there were no significant differences. Higher numbers of midge larvae are found at field edges in Germany (Basedow 1977), but not in Manitoba (Lamb et al. 1999). Higher larval density at field margins was not consistent between years, therefore restricting monitoring activities to field margins is not recommended. There were significant relationships between the number of cocoons recovered from soil cores and the numbers of adult midges that emerged the following spring. This confirms that in the Peace River region of Canada, soil core samples collected in the fall using methods developed by Doane et al. (2000) can indicate regional densities of wheat midges. This can vary between field sites as well as between years, and the predictive ability of fall core collections will vary depending on spring moisture conditions. Attempts to forecast wheat midge populations from soil samples have been imprecise in Germany (Basedow and Schütte 1974), the United Kingdom (Oakley et al. 1998), and in Manitoba (Lamb et al. 1999). Determination of the precise conditions that enable soil core sampling to accurately predict adult populations in some years while not in others is critical. Wheat midge larvae experience variable mortality over winter (Basedow and Gillich 1982) and require adequate soil moisture to emerge in the spring (Basedow and Schütte 1971). Between 5 (Wise and Lamb 2004) to 96% (Basedow and Gillich 1982) of overwintering midge larvae undergo prolonged diapause if moisture conditions are not adequate in the spring. Effective wheat midge management could be conducted with lowered cost and decreased labor by improving alternate management strategies. Wheat midge risk forecast maps using soil core samples accurately indicate regional population densities of wheat midge in northern Alberta and are a valuable tool to aid producers in determining whether to use a standard cultivar or wheat midge ‘resistant’ cultivars of wheat. Producers can use these maps, in combination with previous midge-related downgrading or yield to determine relative risk for the following year. Pheromone-baited traps are used across Canada and the United States for wheat midge, but cannot be used to make management decisions, and do not indicate local population levels. Further research is needed to develop action thresholds using yellow sticky cards. Monitoring using small, yellow sticky cards just before and during the susceptible crop stage could be a cost-effective tool for making management decisions. Acknowledgments This project was supported by the Canadian Government A-base funding, delivered through the department Agriculture and Agri-Food Canada (WBSE Project J-001303.001.03: New tools for managing wheat midge) and a TA-ship to A. Jorgensen from the University of Alberta. We thank Regine Gries, Shelby Dufton, Dr. Owen Olfert, Scott Meers, Shelley Barkley, Dr. Tyler Wist, Nancy Melnychuk, Dr. Breanne Tidemann, Patty Reid, Dr. Scott Noble, Sheila Wolfe, Holly Spence, Janelle Barbarich, and summer students for help with data collection, processing, and advice. This work was possible due to the generous cooperation of Peace River wheat growers: Marc & Lise Rochon, Mike Caron, Norm Lamoureux, Danny & Wayne Dion, Denis Boucher, Lucien Houle, Garnet Berge, Walter Hill, Wayne Longson, Greg Sears, Norm Wilson, and Greg Newman. References Cited Amarawardana , L . 2009 . The chemical diversity of midge pheromones. PhD dissertation . Univesity of Greenwich , London, United Kingdom . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Barton , K . 2018 . MuMIn: Multi-Model Inference. R package version 1.40.4. https://CRAN.R-project.org/package=MuMIn. Basedow , T . 1977 . Über den Flug der Weizengallmücken Contarinia tritici (Kirby) und Sitodiplosis mosellana (Géhin) (Dipt., Cecidomyidae) in Beziehung zur Windrichtung und zu Weizenf eldern . Zeitschrift für Angewandte Entomologie . 83 : 173 – 183 . Google Scholar Crossref Search ADS WorldCat Basedow , T. , and H. Gillich. 1982 . Untersuchungen zur Prognose des Auftretens der Weizengallmücken Contarinia tritici (Kirby) und Sitodiplosis mosellana (Géhin)(Dipt., Cecidomyidae) . Anzeiger für Schädlingskunde, Pflanzenschutz, Umweltschutz . 55 : 84 – 89 . Google Scholar Crossref Search ADS WorldCat Basedow , T. , and F. Schütte. 1971 . Untersuchungen zum überliegen der weizengallmüken Contarinia tritici (Kirby) und Sitodiplosis mosellana (Géhin) in Norddeutschland im Sommer 1970 . Nachrichtenbl Deut Pflanzenschutzdienstes . 23 : 4 – 8 . OpenURL Placeholder Text WorldCat Basedow , T. , and F. Schütte. 1973 . Neue Untersuchungen über Eiablage, wirtschaftliche Schadensschwelle und Bekämpfung der Weizengallmücken (Dipt.; Cecidomyidae) . Zeitschrift für Angewandte Entomologie . 73 : 238 – 251 . Google Scholar Crossref Search ADS WorldCat Basedow , T. , and F. Schütte. 1974 . Uber die unterschiedliche Anfalligkeit einiger Winterweizensorten gegenuber den beiden Weizengallmuckenarten Contarinia tritici (Kirby) und Sitodiplosis mosellana (Géhin)(Dipt., Cecidomyidae). Nachrichtenblatt des Dtsch . Pflanzenschutzdienstes . 26 : 122 – 125 . OpenURL Placeholder Text WorldCat Bates , D. , M. Mächler, B. Bolker, and S. Walker. 2015 . Fitting linear mixed-effects models using lme4 . J. Stat. Soft . 67 : 51 . Google Scholar Crossref Search ADS WorldCat Bechinski , E. J. , C. D. Mcneal, and J. J. Gallian. 1989 . Development of action thresholds for the sugarbeet root maggot (Diptera: Otitidae) . J. Econ. Entomol . 82 : 608 – 615 . Google Scholar Crossref Search ADS WorldCat Bor , Y. J . 1995 . Optimal pest management and economic threshold . Agric. Syst . 49 : 113 – 133 . Google Scholar Crossref Search ADS WorldCat Brach , E. J. , and R. M. Trimble. 1985 . Effect of adhesive on the spectral reflectance of insect traps . Can. Entomol . 117 : 1565 – 1568 . Google Scholar Crossref Search ADS WorldCat Bruce , T. J. , A. M. Hooper, L. Ireland, O. T. Jones, J. L. Martin, L. E. Smart, J. Oakley, and L. J. Wadhams. 2007 . Development of a pheromone trap monitoring system for orange wheat blossom midge, Sitodiplosis mosellana, in the UK . Pest Manag. Sci . 63 : 49 – 56 . Google Scholar Crossref Search ADS PubMed WorldCat Chavalle , S. , F. Censier, G. San Martin y Gomez, and M. De Proft. 2019 . Effect of trap type and height in monitoring the orange wheat blossom midge, Sitodiplosis mosellana (Géhin) (Diptera: Cecidomyiidae) and its parasitoid, Macroglenes penetrans (Kirby) (Hymenoptera: Pteromalidae) . Crop Prot . 116 : 101 – 107 . Google Scholar Crossref Search ADS WorldCat Dexter , J. E. , and N. M. Edwards. 2012 . The implications of frequently encountered grading factors on the processing quality of common wheat . Assoc. Oper. Millers . https://www.grainscanada.gc.ca/en/grain-research/scientific-reports/pdf/process-quality-wheat.pdf. OpenURL Placeholder Text WorldCat Ding , H. , and R. J. Lamb. 1999 . Oviposition and larval establishment of Sitodiplosis mosellana (Diptera : Cecidomyiidae) on wheat (Gramineae) at different growth stages . Can. Entomol . 131 : 475 – 481 . Google Scholar Crossref Search ADS WorldCat Doane , J. , and O. Olfert. 2008 . Seasonal development of wheat midge, Sitodiplosis mosellana (Gehin) (Diptera: Cecidomyiidae), in Saskatchewan, Canada . Crop. Prot . 27 : 951 – 958 . Google Scholar Crossref Search ADS WorldCat Doane , J. F. , O. O. Olfert, and M. K. Mukerji. 1987 . Extraction precision of sieving and brine flotation for removal of wheat midge, Sitodiplosis mosellana (Diptera: Cecidomyiidae), cocoons and larvae from soil . J. Econ. Entomol . 80 : 268 – 271 . Google Scholar Crossref Search ADS WorldCat Doane , J. F. , M. K. Mukerji, and O. Olfert. 2000 . Sampling distribution and sequential sampling for subterranean stages of orange wheat blossom midge, Sitodiplosis mosellana (Géhin) (Diptera: Cecidomyiidae) in spring wheat . Crop Prot . 19 : 427 – 434 . Google Scholar Crossref Search ADS WorldCat Elliott , B. , L. Mann, M. Braun, and O. Olfert. 2002 . Integrated management of crop pests - abundance of wheat midge and its parasite in different management systems . Proc. Soils Crop . 21 – 22 . OpenURL Placeholder Text WorldCat Fox , J. , and S. Weisberg. 2011 . An {R} Companion to Applied Regression, 2nd ed. Sage , Thousand Oaks, CA . http://socserv.socsci.mcmaster.ca/jfox/Books/Companion. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gharalari , A. H . 2008 . Oviposition behaviour of wheat midge Sitodiplosis mosellana (Géhin) (Diptera : Cecidomyiidae) and inheritance of deterrence resistance . Ph.D dissertation, University of Manitoba , Winnipeg, Manitoba, Canada . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gharalari , A. H . 2008 . Oviposition behaviour of wheat midge Sitodiplosis mosellana (Géhin) (Diptera : Cecidomyiidae) and inheritance of deterence resistance. Gries , R. , G. Gries, G. Khaskin, S. King, O. Olfert, L. A. Kaminski, R. Lamb, and R. Bennett. 2000 . Sex pheromone of orange wheat blossom midge, Sitodiplosis mosellana . Naturwissenschaften . 87 : 450 – 454 . Google Scholar Crossref Search ADS PubMed WorldCat Hao , Y. N. , J. Miao, Y. Q. Wu, Z. J. Gong, Y. L. Jiang, Y. Duan, T. Li, W. N. Cheng, and J. X. Cui. 2013 . Flight performance of the orange wheat blossom midge (Diptera: Cecidomyiidae) . J. Econ. Entomol . 106 : 2043 – 2047 . Google Scholar Crossref Search ADS PubMed WorldCat Harris , K. M . 1966 . Gall midge genera of economic importance (Diptera: Cecidomyiidae) Part 1: introduction and subfamily Cecidomyiinae; supertribe Cecidomyiidi . Trans. R. Entomol. Soc. London . 118 : 313 – 358 . Google Scholar Crossref Search ADS WorldCat Hobbs , R. J. , and S. E. Humphries. 1995 . An integrated approach to the ecology and management of plant invasions . Conserv. Biol . 9 : 761 – 770 . Google Scholar Crossref Search ADS WorldCat Hothorn , T. , F. Bretz, and P. Westfall. 2008 . Simultaneous inference in general parametric models . Biom. J . 50 : 346 – 363 . Google Scholar Crossref Search ADS PubMed WorldCat Kirk , W. D. J . 1984 . Ecologically selective coloured traps . Ecol. Entomol . 9 : 35 – 41 . Google Scholar Crossref Search ADS WorldCat Knodel , J. , and M. Ganehiarachchi. 2008 . Integrated pest management of the wheat midge in North Dakota. https://www.ag.ndsu.edu/publications/crops/integrated-pest-management-of-the-wheat-midge-in-north-dakota. Kogan , M . 1998 . Integrated pest management: historical perspectives and contemporary developments . Annu. Rev. Entomol . 43 : 243 – 270 . Google Scholar Crossref Search ADS PubMed WorldCat Lamb , R. J. , I. L. Wise, O. O. Olfert, J. Gavloski, and P. S. Barker. 1999 . Distribution and seasonal abundance of Sitodiplosis mosellana (Diptera: Cecidomyiidae) in spring wheat . Can. Entomol . 131 : 387 – 397 . Google Scholar Crossref Search ADS WorldCat Lamb , R. J. , J. R. Tucker, I. L. Wise, and M. A. H. H. Smith. 2000 . Trophic interaction between Sitodiplosis mosellana (Diptera : Cecidomyiidae) and spring wheat: implications for yield and seed quality . Can. Entomol . 132 : 607 – 625 . Google Scholar Crossref Search ADS WorldCat Lamb , R. L. , I. L. Wise, J. E. Gavloski, L. A. Kaminski, and O. O. Olfert. 2002 . Making control decisions for Sitodiplosis mosellana (Diptera : Cecidomyiidae) in wheat (Gramineae) using sticky traps . Can. Entomol . 134 : 851 – 854 . Google Scholar Crossref Search ADS WorldCat van Lenteren , J. C. , and J. van Woets. 1988 . Biological and integrated pest control in greenhouses . Annu. Rev. Entomol . 33 : 239 – 269 . Google Scholar Crossref Search ADS WorldCat Lewis , W. J. , J. C. van Lenteren, S. C. Phatak, and J. H. Tumlinson , III. 1997 . A total system approach to sustainable pest management . Proc. Natl. Acad. Sci. U. S. A . 94 : 12243 – 12248 . Google Scholar Crossref Search ADS PubMed WorldCat Miller , B. S. , and P. Halton. 1961 . The damage to wheat kernels caused by the wheat blossom midge (Sitodiplosis mosellana) . J. Sci. Food Agric . 12 : 391 – 398 . Google Scholar Crossref Search ADS WorldCat Mircioiu , L . 2004 . Development of pheromone-based monitoring of orange wheat blossom midge, Sitodiplosis mosellana (Gehin) (Diptera: Cecidomyiidae). M. PM. dissertation . Simon Fraser University , Vancouver, British Columbia, Canada . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Nakagawa , S. , and H. Schielzeth. 2013 . A general and simple method for obtaining R2 from generalized linear mixed-effects models . Methods Ecol. Evol . 4 : 133 – 142 . Google Scholar Crossref Search ADS WorldCat Oakley , J. N. , P. C. Cumbleton, S. J. Corbett, P. Saunders, D. I. Green, J. E. B. B. Young, and R. Rodgers. 1998 . Prediction of orange wheat blossom midge activity and risk of damage . Crop Prot . 17 : 145 – 149 . Google Scholar Crossref Search ADS WorldCat O’Hara , R. B. , and D. J. Kotze. 2010 . Do not log-transform count data . Methods Ecol. Evol . 1 : 118 – 122 . Google Scholar Crossref Search ADS WorldCat Olfert , O. O. , M. K. Y. Mukerji, and J. F. Doane. 1985 . Relationship between infestation levels and yield loss caused by wheat midge, Sitodiplosis mosellana (Géhin) (Diptera: Cecidomyiidae), in spring wheat in Saskatchewan . Can. Entomol . 117 : 593 – 598 . Google Scholar Crossref Search ADS WorldCat Olfert , O. , R. M. M. Weiss, and R. H. H. Elliott. 2016 . Bioclimatic approach to assessing the potential impact of climate change on wheat midge (Diptera: Cecidomyiidae) in North America . Can. Entomol . 148 : 52 – 67 . Google Scholar Crossref Search ADS WorldCat Prokopy , R. J. , and E. D. Owens. 1983 . Visual detection of plants by herbivorous insects . Annu. Rev. Entomol . 28 : 337 – 364 . Google Scholar Crossref Search ADS WorldCat Prokopy , R. J. , G. L. Hubbell, R. G. Adams, and K. I. Hauschild. 1982 . Visual monitoring trap for tarnished plant bug adults on apple . Environ. Entomol . 11 : 200 – 203 . Google Scholar Crossref Search ADS WorldCat R Core Team. 2016 . R: a language and environment for statistical computing . R Foundation for Statistical Computing , Vienna, Austria . https://www.R-project.org/. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Shipp , J. L. , K. Wang, and M. R. Binns. 2000 . Economic injury levels for western flower thrips (Thysanoptera: Thripidae) on greenhouse cucumber . J. Econ. Entomol . 93 : 1732 – 1740 . Google Scholar Crossref Search ADS PubMed WorldCat Smith , M. A. H. , I. L. Wise, and R. J. Lamb. 2007 . Sex ratios of Sitodiplosis mosellana (Diptera: Cecidomyiidae): implications for pest management in wheat (Poaceae) . Bull. Entomol. Res . 94 : 569 – 575 . Google Scholar Crossref Search ADS WorldCat Southwood , T. R. E . 1978 . Ecological methods , 3rd ed. Chapman & Hall , London, United Kingdom . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Stanley , B. H. , W. H. Reissig, W. L. Roelofs, M. R. Schwarz, and C. A. Shoemaker. 1987 . Timing treatments for apple maggot (Diptera, Tephritidae) control using sticky sphere traps baited with synthetic apple volatiles . J. Econ. Entomol . 80 : 1057 – 1063 . Google Scholar Crossref Search ADS WorldCat Swanton , C. J. , and S. F. Weise. 1991 . Integrated weed management: the rationale and approach . Weed Technol . 5 : 657 – 663 . Google Scholar Crossref Search ADS WorldCat Thomas , M. B . 1999 . Ecological approaches and the development of “truly integrated” pest management . Proc. Natl. Acad. Sci. U. S. A . 96 : 5944 – 5951 . Google Scholar Crossref Search ADS PubMed WorldCat Western Forum on Pest Management. 2011 . Western committee on crop pests. http://www.westernforum.org/Documents/WCCP/WCCPMinutes/WCCPMINUTES-2011.pdf. Wise , I. L. , and R. J. Lamb. 2004 . Diapause and emergence of Sitodiplosis mosellana (Diptera: Cecidomyiidae) and its parasitoid Macroglenes penetrans (Hymenoptera: Pteromalidae) . Can. Entomol . 136 : 77 – 90 . Google Scholar Crossref Search ADS WorldCat Witzgall , P. , P. Kirsch, and A. Cork. 2010 . Sex pheromones and their impact on pest management . J. Chem. Ecol . 36 : 80 – 100 . Google Scholar Crossref Search ADS PubMed WorldCat Yee , W. L . 2013 . Preferences by Rhagoletis indifferens (Diptera, Tephritidae) for rectangles of various yellow colours and fluorescence . J. Appl. Entomol . 137 : 255 – 233 . Google Scholar Crossref Search ADS WorldCat Zadoks , J. C. , T. T. Chang, and C. F. Konzak. 1974 . A decimal code for the growth stages of cereals . Weed Res . 14 : 415 – 421 . Google Scholar Crossref Search ADS WorldCat © Her Majesty the Queen in Right of Canada, as represented by the Minister of Natural Resources Canada, 2020. 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 Available Tools for Monitoring Wheat Midge (Diptera: Cecidomyiidae)
JF - Environmental Entomology
DO - 10.1093/ee/nvaa017
DA - 2020-06-13
UR - https://www.deepdyve.com/lp/oxford-university-press/assessment-of-available-tools-for-monitoring-wheat-midge-diptera-LwsC4He2mo
SP - 627
VL - 49
IS - 3
DP - DeepDyve
ER -