Evaluation of Monitoring Traps and Lures for Drosophila suzukii (Diptera: Drosophilidae) in Berry Plantings in Florida

Drosophila suzukii (Diptera: Drosophilidae) is an invasive insect pest that was detected in Florida in August 2009 in Hillsborough County. Very limited information is available for berry growers to properly detect and monitor this serious pest in southern highbush blueberry (hybrids of Vaccinium corymbosum L. × V. darrowi Camp), rabbiteye blueberry (Vaccinium virgatum L.), and blackberry (Rubus fruticosus L.) production systems. We compared several D. suzukii traps and lures/baits at two sites in Florida. The traps evaluated included Trécé, Scentry, and a standard homemade cup trap. These traps were compared with various baits and lures, including Trécé lure, Scentry lure, yeast bait, and Suzukii trap, under Florida production systems. Early detection is important to develop an effective monitoring system so management action can be taken before economic damage occurs. Data were recorded as overall trends, as well as in 4–5 trapping periods from early to late season. Overall, the Scentry trap baited with Scentry lure, the Trécé trap baited with Trécé lure + yeast, and the Trécé trap baited with Scentry lure were the best performing traps. Yeast-based traps were also attractive to D. suzukii early in the season, but they did not provide consistent captures as the season progressed. The Scentry trap with yeast bait, the Scentry trap with Scentry lure, the Trécé trap with Trécé lure + yeast bait, and a cup trap with yeast bait caught most of the flies during the first trapping period in 2015 and 2016 in the rabbiteye blueberry. In the southern highbush blueberry, the population of D. suzukii was much lower than in the rabbiteye blueberry planting, and the Scentry trap with Scentry lure captured the highest number of flies during the first trapping period in 2016. In the blackberry, the Scentry trap with Scentry lure numerically had the highest captures during the first trapping period, but this was not significantly different from the cup trap with yeast bait, the Trécé trap baited with Suzukii trap, and the Trécé trap with Trécé lure. Overall, the Scentry trap with Scentry lure was the most consistent trap that captured D. suzukii flies throughout the season in the three production systems—rabbiteye blueberry, southern highbush blueberry, and blackberry. Growers in low pressure systems that are similar to Florida can use the Scentry trap with Scentry lure to monitor D. suzukii populations.


Introduction
The production of berry fruits, especially blueberries and blackberries, is increasing in Florida and other southern States in the United States. In 2017, 33,953 hectares of blueberries with a value of $720 million were cultivated in the United States. During the same year, Florida's growers produced

Study Sites
In 2015 and 2016, a rabbiteye blueberry and a blackberry planting were investigated in Citra, Florida. In 2016, a southern highbush blueberry planting was investigated in Tallahassee, Florida (Figure 1, Site-a and Site-b). The southern highbush plants were 4 years old, and the bushes werẽ 0.9 m tall. The blackberry plants were 7 years old, and the bushes were 1.4 m tall.
Insects 2019, 10, x 3 of 13 D. suzukii in Florida. The purpose of this study was to evaluate the effectiveness of traps and lures in low pressure rabbiteye and southern highbush blueberry in open berry fields in Florida. Evaluation studies were also conducted in a blackberry planting.

Study Sites
In 2015 and 2016, a rabbiteye blueberry and a blackberry planting were investigated in Citra, Florida. In 2016, a southern highbush blueberry planting was investigated in Tallahassee, Florida (Figure 1, Site-a and Site-b). The southern highbush plants were 4 years old, and the bushes were ~0.9 m tall. The blackberry plants were 7 years old, and the bushes were 1.4 m tall.

Experimental Design
The randomized complete block design was used for the experiments with 4 replicates at each site. All trap treatments were blocked according to the variety of blueberry and blackberry. Therefore, all treatments were exposed to the same varieties to prevent any varietal effects. All traps were deployed in the canopy of the growing blueberry and blackberry bushes. The traps were placed in the field when 5% of the berries showed signs of turning blue. All traps in each experiment were serviced once per week. This included rotating the trap positions in the field to prevent any positional bias within each block. Servicing the traps also included replacing the fly drowning solution with respective treatments and using Tupperware (Tupperware Corporation, Orange Blossom Trail, FL, USA) to collect the solution from the traps. The traps were then taken back to the laboratory to separate the adults of D. suzukii from the other insect species and to sex the flies. For the experiments that were carried out in Citra during 2015 (blueberry and blackberry), the different trap contents were assessed for comparing males and females relative to the overall mean capture totals that are reported in the figures.

Traps and Lures
The Trécé trap (Great Lakes IPM, Vestaburg, MI, USA) is marketed as a monitoring device for D. suzukii. The Trécé trap is a 0.95 L clear container that comes with a lid and wire hanger (

Experimental Design
The randomized complete block design was used for the experiments with 4 replicates at each site. All trap treatments were blocked according to the variety of blueberry and blackberry. Therefore, all treatments were exposed to the same varieties to prevent any varietal effects. All traps were deployed in the canopy of the growing blueberry and blackberry bushes. The traps were placed in the field when 5% of the berries showed signs of turning blue. All traps in each experiment were serviced once per week. This included rotating the trap positions in the field to prevent any positional bias within each block. Servicing the traps also included replacing the fly drowning solution with respective treatments and using Tupperware (Tupperware Corporation, Orange Blossom Trail, FL, USA) to collect the solution from the traps. The traps were then taken back to the laboratory to separate the adults of D. suzukii from the other insect species and to sex the flies. For the experiments that were carried out in Citra during 2015 (blueberry and blackberry), the different trap contents were assessed for comparing males and females relative to the overall mean capture totals that are reported in the figures.

Traps and Lures
The Trécé trap (Great Lakes IPM, Vestaburg, MI, USA) is marketed as a monitoring device for D. suzukii. The Trécé trap is a 0.95 L clear container that comes with a lid and wire hanger ( The commercial lures (Trécé and Scentry) used in the study are marketed to last 6 to 8 weeks for the best results. Neither the blueberry nor blackberry field experiments extended beyond this time so there was no need to replace the lures during the trap servicing. The manufacturers of both of these products do not provide the chemical contents embedded in the lures and describe them as "proprietary". The lures contain several components, including acetic acid, ethanol, acetoin, and mathionol (information obtained directly from company representatives). The lure used in our study known as "Suzukii trap" was not commercially available during the time of the study and was only used for research purposes. This product is composed of 7% hydrolyzed protein and 2% organic acid. The remaining ingredients are proprietary and assigned to the manufacturers by law. The yeast bait was brewer's yeast (Saccharomyces cerevisiae)-a one-celled fungus. The detergent solution was 98% water with 2% odorless detergent (Publix Super Market, Gainesville, FL, USA). The yeast bait and the detergent solution were replaced weekly. The combination of different types of traps and lures were tested to determine their efficiencies in trapping D. suzukii adults in berry plantings. The details of the various treatments used in the study are provided in Table 1.  The commercial lures (Trécé and Scentry) used in the study are marketed to last 6 to 8 weeks for the best results. Neither the blueberry nor blackberry field experiments extended beyond this time so there was no need to replace the lures during the trap servicing. The manufacturers of both of these products do not provide the chemical contents embedded in the lures and describe them as "proprietary". The lures contain several components, including acetic acid, ethanol, acetoin, and mathionol (information obtained directly from company representatives). The lure used in our study known as "Suzukii trap" was not commercially available during the time of the study and was only used for research purposes. This product is composed of 7% hydrolyzed protein and 2% organic acid. The remaining ingredients are proprietary and assigned to the manufacturers by law. The yeast bait was brewer's yeast (Saccharomyces cerevisiae)-a one-celled fungus. The detergent solution was 98% water with 2% odorless detergent (Publix Super Market, Gainesville, FL, USA). The yeast bait and the detergent solution were replaced weekly. The combination of different types of traps and lures were tested to determine their efficiencies in trapping D. suzukii adults in berry plantings. The details of the various treatments used in the study are provided in Table 1.

Effectiveness of Nine Trapping Systems to Detect D. suzukii in Rabbiteye Blueberry in Citra, Florida
The experiment was conducted in rabbiteye blueberries (Vaccinium virgatum) during the 2015 and 2016 fruiting season from May to June. The blueberry bushes were approximately 8 years old. During this study, the blueberry plants were approximately 2 m in height and 1 m apart with 1.5 m between rows that were approximately 50 m long. The experimental blocks were approximately 5 m apart. The treatments were blocked by rabbiteye cultivars in a randomized block design. Four rabbiteye cultivars were included-"Climax", "Brightwell", "Premier", and "Powderblue". In this experiment, we determined the effectiveness of nine trapping systems for monitoring D. suzukii (Table 1).

Effectiveness of Seven Trapping Systems to Detect D. suzukii in Southern Highbush Blueberry (SHB) in Tallahassee, Florida
The experiment was conducted in southern highbush blueberries (Vaccinium corymbosum L. × V. darrowi Camp) during the 2016 fruiting season (April-May) in Tallahassee, Florida, at the Florida A&M University's Center for Viticulture and Small Fruits Research Center. The blueberry bushes in Tallahassee were four years old. The experimental plots of blueberry bushes comprised 8 rows. Each row consisted of 24 southern highbush blueberry plants. The experimental blocks were approximately 5 m apart, and the blueberry rows were approximately 3 m apart. Each row was approximately 100 m long, and the bushes were spaced 1.5 m apart. On average, the plants were 1.5 m tall. The cultivars used in this experiment consisted of "Meadowlark", "Star", "Emerald", "Farthing", "Sweetcrisp", and "Abundance". The data for Experiment 2 were collected over the course of 7 weeks. In this experiment, we determined the effectiveness of 7 trapping systems for monitoring D. suzukii (Table 1).

Effectiveness of Six Trapping Systems to Detect D. suzukii in Blackberry Planting in Citra
This experiment was conducted during May to June in Citra, Florida, in blackberries. The rows of blackberries were approximately 100 m long. The plants were on a trellis system and were approximately 2 m in height. These bushes were approximately 7 years old. The traps were installed within the bushes away from direct sunlight. In Florida, blackberries typically ripen during May and June. The blackberry varieties used in this study included "Arapaho", "Chickasaw", "Kiowa", and "Ouachita". The data were collected over the course of 4 weeks during the blackberry growing season. In this experiment, we determined the effectiveness of 6 trapping systems for monitoring D. suzukii (Table 1).

Data Analysis
The data from each experiment were analyzed separately using a two-way repeated measure analysis of variance (ANOVA) [17]. The counts of D. suzukii were transformed to ln (x + 0.5) to normalize the distribution and homogenize the variances before analysis. The male and female data were separated during the counts and analyzed separately to determine the effects of the treatments on each sex of D. suzukii. The treatment means were separated by Tukey's HSD (honestly significant difference) test [18].

Effectiveness of Nine Trapping Systems to Detect D. suzukii in Rabbiteye Blueberry Planting
In 2015, the Scentry traps with Scentry lures and the Trécé traps with Trécé lures + yeast bait captured the highest overall mean number of flies. These two trap-and-lure systems did not differ significantly from each other (Figure 3). The Trécé trap with Trécé lure + yeast treatment was not different from any other trapping systems, except the cup trap with Suzukii trap and the Scentry trap with Suzukii trap, which did not differ significantly from one another (Figure 3).
In 2015, the Scentry traps with Scentry lures and the Trécé traps with Trécé lures + yeast bait captured the highest overall mean number of flies. These two trap-and-lure systems did not differ significantly from each other (Figure 3). The Trécé trap with Trécé lure + yeast treatment was not different from any other trapping systems, except the cup trap with Suzukii trap and the Scentry trap with Suzukii trap, which did not differ significantly from one another (Figure 3).  Table 2). The lowest number of flies (male and female) was recorded on the cup trap with Suzukii trap, the Scentry trap with Suzukii trap, and the cup trap with Scentry lure, which did not differ significantly from one another ( Table 2).  Table 2). The lowest number of flies (male and female) was recorded on the cup trap with Suzukii trap, the Scentry trap with Suzukii trap, and the cup trap with Scentry lure, which did not differ significantly from one another ( Table 2). For the Citra blueberries in 2016, the Scentry trap with Scentry lure, the Trécé trap with Scentry lure, and the Scentry trap with yeast bait captured significantly more flies than the cup trap with Suzukii trap and the Scentry trap with yeast bait (F = 6.2; df = 8, 27; p = 0.03). None of the other treatments were significantly (p > 0.05) different from each other (Figure 4).
For the Citra blueberries in 2016, the Scentry trap with Scentry lure, the Trécé trap with Scentry lure, and the Scentry trap with yeast bait captured significantly more flies than the cup trap with Suzukii trap and the Scentry trap with yeast bait (F = 6.2; df = 8, 27; p = 0.03). None of the other treatments were significantly (p > 0.05) different from each other (Figure 4). With the exception of the Trécé trap with Trécé lure + yeast bait and the Trécé trap with Scentry lure, more male D. suzukii were captured on the Scentry trap with Scentry lure than any other trapping systems evaluated (Table 3). With regards to female D. suzukii, the cup trap with yeast bait captured more flies than the cup trap with Suzukii trap, the cup trap with Scentry lure, and the Scentry trap with Suzukii trap (Table 3). With the exception of the Trécé trap with Trécé lure + yeast bait and the Trécé trap with Scentry lure, more male D. suzukii were captured on the Scentry trap with Scentry lure than any other trapping systems evaluated (Table 3). With regards to female D. suzukii, the cup trap with yeast bait captured more flies than the cup trap with Suzukii trap, the cup trap with Scentry lure, and the Scentry trap with Suzukii trap (Table 3).

Effectiveness of Seven Trapping Systems to Detect D. suzukii in Southern Highbush Blueberry (SHB) in Tallahassee, Florida
The overall pooled data (18 April to 23 May) collected in southern highbush blueberry in 2016 Tallahassee, Florida, indicate that the Scentry trap with Scentry lure captured significantly more flies columns with different letters are significantly different (p ≤ 0.05; Tukey's HSD test) from each other. Yt = Yeast bait, T lure = Trécé lure, and Sc lure = Scentry lure.

Effectiveness of Seven Trapping Systems to Detect D. suzukii in Southern Highbush Blueberry (SHB) in Tallahassee, Florida
The overall pooled data (18 April to 23 May) collected in southern highbush blueberry in 2016 Tallahassee, Florida, indicate that the Scentry trap with Scentry lure captured significantly more flies than all other trapping systems (F = 5.37; df 2, 21; p = 0.0017) ( Figure 5). None of the other trapping systems were significantly different (p > 0.05) ( Figure 5). There were significantly more male and female flies captured in the Scentry trap with Scentry lure (F = 3.89; df = 34, 105; p = 0.0001) when compared with the other six treatments. There were no significant differences in male captures for all treatments except the Scentry trap with Scentry lure ( Table 4). The Scentry trap with Scentry lure captured more female D. suzukii compared with other treatments (F = 5.19; df = 34, 105; p = 0.0001). Significantly more female D. suzukii were captured in the Scentry trap + yeast bait when compared to all treatments (Table 4). There were significantly more male and female flies captured in the Scentry trap with Scentry lure (F = 3.89; df = 34, 105; p = 0.0001) when compared with the other six treatments. There were no significant differences in male captures for all treatments except the Scentry trap with Scentry lure (Table 4). The Scentry trap with Scentry lure captured more female D. suzukii compared with other treatments (F = 5.19; df = 34, 105; p = 0.0001). Significantly more female D. suzukii were captured in the Scentry trap + yeast bait when compared to all treatments (Table 4).

Treatments
Mean ± SE per Trap

Efficiency of Six Trapping Systems in Blackberry Planting in Citra
The Scentry trap with Scentry bait captured significantly more flies than the cup trap with Suzukii trap and the Scentry trap with Suzukii trap (F = 3.20; df = 5, 18; p = 0.0307). However, there were no significant differences between the cup trap with yeast bait, the Trécé trap with Suzukii trap, and the Trécé trap with Trécé lure (Figure 6). The cup trap with Suzukii trap and the Scentry trap with Suzukii trap captured the fewest D. suzukii, and there was no significant difference between these two  (Table 5).

Efficiency of Six Trapping Systems in Blackberry Planting in Citra
The Scentry trap with Scentry bait captured significantly more flies than the cup trap with Suzukii trap and the Scentry trap with Suzukii trap (F = 3.20; df =5, 18; p = 0.0307). However, there were no significant differences between the cup trap with yeast bait, the Trécé trap with Suzukii trap, and the Trécé trap with Trécé lure (Figure 6). The cup trap with Suzukii trap and the Scentry trap with Suzukii trap captured the fewest D. suzukii, and there was no significant difference between these two treatments. The Scentry trap with Scentry bait captured significantly more flies than other treatments (F = 1.06; df =5, 18; p = 0.0412). The cup trap/yeast trap, Trécé trap/Suzukii bait, Trécé trap/ Trécé lure, and the cup trap/Suzukii bait captured the lowest numbers of male flies. The Scentry trap/Scentry bait treatment captured the most female D. suzukii compared to all other treatments (F = 2.28; df = 23, 72; p = 0.0043) ( Table 5).

Discussion
Identifying effective monitoring tools is important when developing management programs for invasive pests. Overall, our findings indicate that the Scentry trap baited with Scentry lure, the Trécé trap baited with Trécé lure + yeast, and the Trécé trap baited with Scentry lure were the best performing traps during the 2015 and 2016 studies conducted in rabbiteye blueberries. Traps baited with yeast also performed well initially, and the captures were similar to the Scentry and Trécé traps during the first trapping period in 2015 and the first and second trapping periods in 2016. The problem with the yeast was that the D. suzukii captures fell off quickly by the end of the second trapping period during the typical production season. The changes in the attraction to yeast may be related to the reproductive status of the flies. Flies caught early in the season (during the first or second trapping periods) may not be reproductively developed and so may be more attracted to food-based baits, such as yeast [19]. Later in the season (from the third to the fifth trapping period) there were a lot more berries in the field and an abundance of fruit volatiles that ovipositing D. suzukii flies may have been attracted to as opposed to yeast.
The attractiveness of yeast early in the season has been previously reported [13,17]. In a recent study, yeast species, strain, and growth media were found to influence the attraction of D. suzukii flies to baited traps [20]. Yeast-based baits tend to attract many nontarget insects, including beneficials [13]. These differences in selectivity were quite noticeable when sorting through the samples, but this was not documented. Overall, the authors found that yeast-based baits were often cloudy and had sediments that made fly identification extremely difficult.
The Trécé trap with Trécé lure + yeast was also attractive, but growers would have to purchase all three components (trap, lure, and yeast), and this could become a cumbersome and time-consuming process to combine all three components in the field before deploying the traps.
The Scentry trap and Scentry lure system provided the most consistent captures of D. suzukii flies in all three production systems (rabbiteye, southern highbush, and blackberry). This trap-and-lure system could be particularly important for monitoring flies, especially in low pressure systems, such as Florida, where fly populations rarely exceed 20 flies per trap.
In our studies in southern highbush blueberry in which the highest D. suzukii population was 3.0 ± 1.1 and 4.0 ± 0.7 in the first and second trapping period, respectively, the Scentry trap and Scentry lure captured significantly more flies than all the other trap-and-lure systems evaluated, including the Trécé trap with Trécé lure and an additional yeast bait booster.
The attractiveness of the Scentry trap and Scentry lure may be related to both trap design and lure. The trap is red in color (Figure 2, Trap-3) with several entry holes. During trap evaluation studies, red color traps or darker colored traps performed better than clear traps [21][22][23]. In addition, lure and entry holes were more important in determining attraction than trap color [11,12]. Scentry lure is potentially a mixture of fermented wine vinegar, ethanol, and acetic acid; this information is proprietary for Scentry Biologicals Inc. as indicated by the company representative.
The homemade cup trap did not catch as many flies as the Scentry and Trécé traps in both years, but when baited with Scentry lure or yeast, the cup captures were not significantly different to either of these traps. The performance of the homemade cup trap could be related to the structure and transparency of the trap. The holes may have been insufficient entry pathways to capture enough D. suzukii flies. Redesigning the trap by making more suitable fly entry holes and changing the color of the trap may improve its performance.
All the traps required preparation work before deploying them in the field. The ease of implementation of the trap varied among the different traps. We noticed that when servicing the Scentry trap (with a screw lid) it proved easier to detach in comparison with the Trécé trap and the cup trap, which both had lids that popped off but would often become stuck together. Overall, the cup trap/Suzukii trap and the Scentry trap/Suzukii trap performed poorly, indicating that Suzukii trap is ineffective in our system irrespective of the trap used.
In the 2016 blueberry planting in Citra, significant differences were found in the numbers of male and female flies captured. The highest number of male flies was captured in the Scentry trap with Scentry lure; however, when the Trécé trap was equipped with Scentry lure and the Trécé trap with Trécé lure + yeast booster, there were no significant differences in male captures, indicating the efficiency of Scentry lure. It is apparent that Scentry lure out-competes the volatiles emitted in blueberries, because the trap-and-lure system continued to perform well during the fourth and fifth trapping periods when the blueberries were in their peak ripening period. Similar to the 2015 fruiting season, the cup trap with yeast bait captured a high percentage of females, but this was not different to the Scentry with Scentry lure, the Trécé trap with Trécé lure + yeast bait, and the Trécé trap with Scentry lure.
The overall trap captures for 2016 were lower than 2015. This may be due to the fluctuation in weather patterns among the two growing seasons. The mean temperature in Citra, Florida in 2015 was 23 • C, while the mean temperature in 2016 was 26 • C [24]. The optimal temperature for the development of the egg-to-adult stage of D. suzukii has been found to be 28.2 • C [6].
The lowest number of male flies was captured in the cup trap with Suzukii trap; however, this was not significantly different to the cup trap with Scentry lure and the Scentry trap with Suzukii trap. Similarly, the cup trap with Suzukii trap caught the lowest number of females numerically but was not significantly different to the Scentry trap with Suzukii trap and the cup trap with Scentry lure. These findings indicate the poor performance of Suzukii trap bait for both sexes. Another disadvantage in using this product is that the 200 mL of Suzukii trap often evaporated almost completely by the seventh day leaving behind a brown, thick, syrup-like residue. To solve this issue, we added a small amount of water to dilute the residue to release and record the trapped flies.

Conclusions
Drosophila suzukii is a relatively new and serious pest of berry plantings in Florida. Several traps, lures, and their combinations were evaluated over two years in three planting systems. A Scentry trap and Scentry lure provided the most consistent captures of D. suzukii throughout the entire season, attracting both male and female D. suzukii. Other traps, including a Trécé trap and Trécé lure + yeast bait, performed well but could be difficult and time consuming to deploy in the field. The effective performance of the Scentry trap and Scentry lure system could be due to the physical characteristics of the trap, including color, selectivity, and the attractiveness of the lure.