https://orcid.org/0000-0001-8874-188X
Abstract
The sweetpotato whitefly Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae) is challenging to control using chemical pesticides owing to its resistance to many insecticides. Thus, there has been an increasing demand for alternative control measures. Thus, this study evaluated the efficacy of a newly designed pest suction machine to manage whiteflies on tomato plants (Solanum lycopersicum L.) (Solanales: Solanaceae) in greenhouses over 2 seasons. The suction machine comprised a battery-powered cart with a mounted suction unit, an ultrasonic device, and green lights. Ultrasonic irradiation provided non-contact vibration, facilitating the movement of adult whiteflies away from the plants, and green lights attracted them to the suction device. This combination effectively captured whitefly adults, even with a weak suction force, saving electricity consumption. The efficacy of suction machine was further evaluated by measuring the number of whitefly adults caught by the machine and the number of adults and nymphs remaining on the tomato leaves. The whitefly population was considerably lower in the treated blocks than in the non-treated blocks in the autumn trial. The machine reduced the density of whitefly adults without using chemical pesticides. Although a lot of optimizations would be required, suction control is an additional and alternative strategy that may be incorporated in the integrated pest management of whiteflies on greenhouse tomato plants.
Introduction
Insect pests cause substantial losses in agricultural production (Oerke 2006). In addition, greenhouses provide ideal conditions for the reproduction of pests, leading to the increased use of chemical pesticide sprays and high pesticide resistance (van Lenteren 2000). For example, the sweetpotato whitefly, Bemisia tabaci (Gennadius) (Hemiptera: Aleyrodidae), is a critical pest in greenhouse horticulture worldwide and is listed among the 100 of the world’s worst invasive alien species (Lowe et al. 2000).
Whitefly control is especially crucial for tomato production (Cohen and Harpaz 1964, Jones 2003, Jones 2009, Ghanim 2014). Tomatoes (Solanum lycopersicum L.) (Solanales: Solanaceae) are mainly grown in greenhouses in Japan for commercial purposes. Several varieties of tomatoes (e.g., large, medium, and cherry) are cultivated in Japanese greenhouses with different methods (e.g., soil cultivation and drip fertigation) for variable duration (10–40 weeks). Whiteflies not only damage plants directly but also transmit plant viruses (e.g., tomato yellow leaf curl virus: Cohen and Harpaz 1964, Lapidot and Polston 2010, Ban et al. 2021; tomato chlorosis virus: Wisler et al. 1998, Wintermantel 2010; and tomato infectious chlorosis virus: Duffus et al. 1996, Wintermantel 2010). Tomatoes infected with tomato yellow leaf curl virus stop growing at point elongation (Kato et al. 1998), and the abnormal coloration of tomatoes is caused by a high density of this whitefly population (Schuster et al. 1990, Matsui 1992). In addition, sooty mold caused by honeydew excreted by whitefly induces fruit staining and inhibits leaf photosynthesis (Stansly and Natwick 2010). The consequent decreases in crop yield result in significant economic losses for farmers (Moriones and Navas-Castillo 2010). Moreover, whiteflies have developed resistance to many pesticides, making them difficult to control (Horowitz et al. 2020, Rossitto De Marchi et al. 2021, Perier et al. 2022). Given this scenario, the demand for alternative integrated pest management (IPM) strategies is increasing (Horowitz et al. 2011).
Suction control has been effectively used for controlling various agricultural insect pests in open-field crops (Boiteau et al. 1992, Vincent and Lachance 1993, Weintraub et al. 1996, Weintraub and Horowitz 1999). Boiteau et al. (1992) demonstrated the effectiveness of the “Beetle Eater,” a commercial-scale insect suction device for the Colorado potato beetle (Leptinotarsa decemlineata (Say)), on open-field potatoes in North America. Weintraub et al. (1996) used the “field-scale vacuum” (FSV) to control whiteflies and leafhoppers (Empoasca spp.) on open-field celery, and sweetpotato whiteflies (B. tabaci (Gennadius)) and south American leaf-miner (Liriomyza huidobrensis (Blanchard)) on open-field potatoes in Israel. FSV can reduce the population of whiteflies, leafhoppers, and aphids by 50–75% within weeks. Leafminers also decreased substantially, but only temporarily. Furthermore, FSV used once a week in open-field melons could successfully control sweetpotato whiteflies at the same level as insecticide spraying (Weintraub and Horowitz 1999). Vincent and Lachance (1993) used the “Biovac” prototype on open-field strawberries in Canada and found that it could considerably reduce Tarnished plant bug (Lygus lineolaris (Palisot de Beauvois)) populations 7 out of 15 times. Furthermore, some of these machines have been a commercial success. In open-field strawberries in California, USA, tractor-towed pest vacuum “bug vacuums” are now widely used (Wells et al. 2020), and pest vacuuming has become an established means of Lygus spp. control.
However, these suction machines cannot be used for crops with large heights. The machines commercially used currently are installed to straddle the ridges of crops and remove pests from crops by suction air blown downward from both sides of the unit. Therefore, the target crops should be of low height (Boiteau et al. 1992, Vincent and Lachance 1993, Weintraub et al. 1996, Wells et al. 2020).
In contrast to open fields, crops are grown to a higher height (i.e., more than 3 m) in greenhouses to ensure high yields in a limited area. Therefore, a completely different apparatus is required to use suction in greenhouses. Owing to this limitation of suction machines, auxiliary means are needed to remove pests efficiently in greenhouses.
Thus, in the present study, we designed a suction machine with a completely new framework and 2 additional devices: an ultrasonic focusing device and a green light-emitting diode (LED). Green light is known to attract adult whiteflies (Macdowall 1972, Vaishampayan et al. 1975, Affeldt et al. 1983, Chu et al. 2000, 2004, Jahan et al. 2014). We evaluated the efficiency of this suction machine for controlling whiteflies on tomato plants in a greenhouse.
Materials and Methods
Pest Suction Machine
The pest suction machine used in this study comprised a suction unit installed on a traveling cart (length, 1.05 m; width, 0.6 m; height, 0.4 m; and mass, 86.3 kg), which was equipped with a 53 Ah battery (Somic Ishikawa Co., Ltd., Hamamatsu, Japan). The pest suction unit in turn was composed of an ultrasonic device, a green LED flashlight, a suction device, a computer, and a battery (Fig. 1). The ultrasonic device caused non-contact vibrations on tomato leaves, causing whitefly adults to fly off the plants. The flashlight and rope-shaped LEDs were wrapped around the suction port to attract whitefly adults to the suction port. The suction device was equipped with an electric blower and had a funnel-shaped suction port to suck insects, which were captured and killed using 2 sticky plates inside it. The installation height of the suction device could be adjusted according to the height of the tomato plants. Power for the pest suction machine was supplied by a small, large-capacity battery (AP-SS-0091 (Monster X), 1,700 Wh; Allpower Industry International Limited, Hong Kong, China). The machine was continuously operated on a single charge for over 3 h.
Diagonally forward view of the pest suction machine. (1) An ultrasonic device (the square device between the wheels) causes non-contact vibration on tomato leaves, removing whitefly adults from the plants. (2) A green LED flashlight and green LED rope attract whitefly adults to the suction port. (3) Suction port: whitefly adults attracted to the suction port by a green light were sucked in. (4) Sticky plate in the box: whitefly adults were captured and killed. This pest suction unit (1–4) was mounted on a traveling cart (5), and the suction unit and cart were powered by one battery each. Therefore, the machine could move freely in a greenhouse. The rope LEDs were covered by black cloth to avoid prevent diffused reflection during the operation.
Ultrasonic focusing device was developed by Hoshi et al. (2014), and its specifications were described by Shimizu et al. (2015). Ultrasonic focusing devices cause non-contact vibrations in plants (Shimizu et al. 2015) and have been shown to release B. tabaci adults from plants (Urairi et al. 2022). Shimizu et al. (2015) used a total of 285 ultrasonic transducers (T4010A1, Nippon Ceramic Co., Ltd., Tokyo, Japan; 40 kHz resonance frequency, 1 cm diameter) arranged in a 170 × 170 mm square with non-contact resonance frequency to generate a maximum force of 16 mN. The phase of each transducer was appropriately controlled to create a focal point and scan it in a space with an area of 20 × 20 cm and a depth of 20–40 cm to stimulate multiple points in the tomato plant. To apply the non-contact vibration generated by the ultrasonic device over a wider range, an electric actuator was attached to the ultrasonic device, and a swinging mechanism was provided.
Green light is known to attract adult whiteflies (Macdowall 1972, Vaishampayan et al. 1975, Affeldt et al. 1983, Chu et al. 2000, 2004, Jahan et al. 2014). Thus, we used a commercial green LED for attracting whitefly adults to the suction ports. The bulb part of the Maglite 2D model (Mag Instrument, Inc., Ontario, CA, USA) was changed to a 6 W Green LED power module for Maglite green light (peak wavelength, 518 nm; FusionLite, Hong Kong, China). The Maglite was pointed toward tomato plants. In addition, to attract whitefly adults to the suction port more efficiently, a 5-m-long green rope-shaped LED (LUX-TUBELIGHT RGB Multi-Color; S&I Co. Ltd., Toyama, Japan) was wound from the outside of the funnel-shaped suction port with transparent PET resin 33 cm in diameter. The green LED covered approximately half of the funnel-shaped suction port. When the 2 light sources were turned on, the light intensity on the tomato plant (50 cm from the wheel) was 1.63 E + 19 photons/m2/s.
Suction was performed laterally using an electric blower (EC-H04-R313; Showa Denki Co., Ltd., Osaka, Japan) with a maximum air volume of 16 m3/min. The wind speed near the suction port inlet was approximately 1 m/s. The insects sucked were captured using 2 blue sticky plates (514 mm × 200 mm/2 plates; Horiver Blue, Arysta LifeScience Corporation, Tokyo, Japan) installed in the box by inertial separation (Fig. 2). The exhaust was discharged to the rear of the fuselage.
Overview of the inside of the pest suction box.
Experimental Setup
Experiments were conducted on greenhouse tomato plants (cv. ‘Momotaro Hope’) in soil cultures over 2 seasons. Tomatoes were planted on 21 April 2021 (spring) and 14 September 2021 (autumn). Each trial was conducted in 4 test blocks, which were set in a greenhouse (7.7 × 22.6 × 4.2 m) at Shizuoka Agricultural and Forestry Research Institute, located in Iwata, Shizuoka, Japan. Two small mesh cages (2.3 × 19.0 × 2.0 m; 0.4 mm mesh size; Sun Sun Net Soflight SL4200, Nihon Widecloth Co., Ltd., Kashiwara, Japan) were set in the greenhouse, and each mesh cage was separated at the center by a zippered insect repellent net (0.4 mm mesh size; Dio FN-04, Dio Kasei Co., Ltd., Tokyo, Japan), preventing the movement of insects between the separated regions. In the 2 blocks (hereafter referred to as the “suction blocks”), suction treatment was performed daytime, 6 times a week until the end of the experiment at 11 weeks; in the other 2 blocks (“untreated blocks”), suction treatment was not performed. The blocks were arranged alternately. The blocks contained one row (6.0 × 0.6 m), and 15 tomato seedlings were planted at interval of 0.4 m in each row. In the suction blocks, the pest suction machine was radio-controlled and moved manually. It paused and sucked whiteflies for 5 s for each tomato plant and then moved again. Because the pest sucker ran on both sides of the tomato row, suction was performed for 10 s per tomato plant per run (Fig. 3).
Image diagram of the pest suction machine running. In the suction block, the pest suction machine paused and sucked for 5 s for each tomato plant and then advanced. The suction treatment was performed for 10 s per tomato plant per run since suction was operated on both sides of the tomato row.
Approximately 60 B. tabaci (Q biotype) were released 4 times into each block at 1-week intervals after planting in each season. The insects were originally collected from squash (Cucurbita maxima L.) in Kumamoto, Japan, in 2004 and reared for more than 10 generations on cabbage (Brassica oleracea (L. var. capitata)) in a laboratory under 16L8D photoperiod and 25 °C conditions.
No insecticide was sprayed after planting the tomatoes, but flowable sulfur diluted 500 times was sprayed with a power sprayer every 2 weeks to prevent diseases and the tomato rust mite (Aculops lycopersici (Massee)). The side windows of the greenhouse were always kept open, and the skylights were set to open above 25 °C. The temperature during the trial was recorded by a temperature logger, TR-52i (T&D Corporation, Matsumoto, Japan), which was installed at approximately 140 cm in the northwest block of the greenhouse.
Counting the Number of Whiteflies
The number of B. tabaci adults and nymphs on the upper, middle, and lower compound leaves of all plants in each block was counted at 1-week intervals after planting 11 times in total.
The number of whiteflies was inclusively analyzed among the treatments using a generalized linear mixed model (GLMM) in the R software version 4.3.2 (R Core Team 2023) for all 4 trials in spring and autumn. The whitefly population was assumed to follow a Poisson distribution, with the numbers of adult and nymph whiteflies as response variables. The logarithm of the link function was used. As explanatory variables, treatments were considered a fixed effect, and tomato plants, number of weeks after the first whitefly release, block location, trial season, and survey leaves (upper, middle, and lower leaves) were considered random effects.
We also statistically analyzed the number of adults and nymphs after Yeo-Johanson transformation between treatments for each season and performed an analysis of variance. Dates were also treated as independent variables as with the treatments.
Number of Whitefly Adults Captured and Killed by the Suction Machine
The insects were sent to the box through a flexible duct and captured using 2 sticky blue plates. The number of whitefly adults captured and killed by the 2 sticky blue plates was counted visually. The whiteflies were removed for each run in each block.
Effect of the Whitefly Suction Machine on Tomato Fruits
The number and weight of the fruits harvested at the appropriate time were counted for each block. The proportion of excellent-grade fruits was calculated only for the spring season using the chi-square test. All statistical analyses were performed using R version 4.3.2 (R Core Team 2023).
Results
Changes in the Number of B. tabaci
Figures 4 and 5 show changes in the number of B. tabaci during the trial period in spring and autumn, respectively. In both seasons, the number of whiteflies gradually increased in the suction and untreated blocks. The average temperature (± standard deviation) in the greenhouse during the trial period was 24.9 °C ± 6.6 °C in spring and 22.1 °C ± 8.3 °C in autumn. The maximum and minimum temperatures were 44.2 °C and 10.3 °C in spring and 44.0 °C and 4.0 °C in autumn, respectively.
Population dynamics of whiteflies in the spring of 2021. The number of whitefly (A) adults and (B) nymphs per tomato compound leaf. Closed circles and open squares represent the average number of insects in the suction and untreated blocks, respectively. Error bars indicate standard error. Closed downward triangles indicate whitefly release time points.
Population dynamics of whiteflies in the autumn of 2021. The number of whitefly (A) adults and (B) nymphs per tomato compound leaf. Closed circles and open squares represent the average number of insects in the suction and untreated blocks, respectively. Error bars indicate standard error. Closed downward triangles indicate whitefly release time points. Asterisks indicate significant (P < 0.05) difference in the number of insects between suction and untreated blocks on the day.
Table 1 shows the statistical analysis results of the treatment and control groups. The suction machine significantly decreased the number of adult whiteflies in the suction blocks (GLMM; P < 0.0001).
Statistical results of a generalized linear mixed model, assuming a Poisson distribution, with the number of whiteflies present per compound leaf as the response variable, suction pest control machines in operation set as explanatory variables, and weeks post-whitefly release, blocks, experimental season, leaf position, and individual tomatoes set as random effects
| Estimate | Standard error | Z value | Probability (> |z|) | |
|---|---|---|---|---|
| Intercept | −1.3274 | 1.6558 | −0.802 | 0.423 |
| Treatment | −0.3495 | 0.0227 | −15.398 | <2e-16 *** |
| Estimate | Standard error | Z value | Probability (> |z|) | |
|---|---|---|---|---|
| Intercept | −1.3274 | 1.6558 | −0.802 | 0.423 |
| Treatment | −0.3495 | 0.0227 | −15.398 | <2e-16 *** |
***P < 0.001.
Statistical results of a generalized linear mixed model, assuming a Poisson distribution, with the number of whiteflies present per compound leaf as the response variable, suction pest control machines in operation set as explanatory variables, and weeks post-whitefly release, blocks, experimental season, leaf position, and individual tomatoes set as random effects
| Estimate | Standard error | Z value | Probability (> |z|) | |
|---|---|---|---|---|
| Intercept | −1.3274 | 1.6558 | −0.802 | 0.423 |
| Treatment | −0.3495 | 0.0227 | −15.398 | <2e-16 *** |
| Estimate | Standard error | Z value | Probability (> |z|) | |
|---|---|---|---|---|
| Intercept | −1.3274 | 1.6558 | −0.802 | 0.423 |
| Treatment | −0.3495 | 0.0227 | −15.398 | <2e-16 *** |
***P < 0.001.
In spring, at the end of the trial on 8 July, the average numbers of whitefly nymphs and adults were 2.4 and 2.2, respectively, per compound leaf in the suction blocks, whereas in the untreated blocks, they were 3.6 and 2.5, respectively (Fig. 4). However, statistical significance was observed only in dates but not for the numbers of adults and nymphs (P = 0.289 and P = 0.262 for adults and nymphs, respectively).
In autumn, at the end of the trial on 30 November, the average numbers of whitefly nymphs and adults were 23.9 and 0.4 per compound leaf in the suction blocks, which was 93.4% and 28.6% of those in the untreated blocks (25.6 and 1.4, respectively) (Fig. 5). No statistical significance was found in the number of nymphs (P = 0.791), but it was found in the number of adults (P = 0.026). As a result of a simple main effect test adjusted by Holm’s method, significant differences were found between suction and untreated blocks on dates 9, 17, and 30 November (P = 0.040, 0.004, and 0.036, respectively).
Whitefly Adults Captured and Killed by the Suction Machine
The number of whitefly adults captured and killed by the suction machine increased as the number of whitefly adults on tomato plants increased. In total, 200 and 225 adults in each suction block in spring and 424 and 299 adults in autumn were killed during the experimental period.
Tomato Plants Production
In spring, by 13 July, 193.5 ± 7.5 (33.3 ± 1.7 kg) and 160.5 ± 5.5 (24.0 ± 2.7 kg) tomatoes (± standard error) were harvested in the suction and the untreated blocks, respectively. The average cumulative fruit numbers and weights were 1.2 and 1.4 times higher in the suction blocks than in the untreated blocks. The average proportion of excellent-grade fruits was 77.5% in the suction block and 75.3% in the untreated block, which were not significantly different (χ2 test, P = 0.7511). In autumn, ripe tomato fruits were harvested between 24 December and 4 February after the trial. The average (± standard error) cumulative number and weight of harvested tomatoes were 97.5 ± 1.5 fruits and 19.0 ± 2.7 kg in the suction blocks and 53.5 ± 10.5 fruits and 8.8 ± 1.9 kg in the untreated blocks, respectively. The average cumulative fruit count and weight in the suction blocks were 1.8 and 2.2 times higher, respectively.
Discussion
We demonstrated that the pest suction machine developed in this study could have the potential to control whiteflies on greenhouse tomato plants. The numbers of whitefly adults in autumn on plants in the treated blocks were significantly lower than those in the untreated blocks. Although the operational methods need to be more optimized, this new physical control method can contribute to reducing the application of insecticides and preventing the development of insecticide resistance in whiteflies. In addition, the cumulative number of tomatoes harvested was higher in the suction blocks than in the untreated blocks. Non-contact vibration using the ultrasonic device may have promoted pollination on tomatoes, as reported by Shimizu and Sato (2018) on strawberries, non-contact vibration using an ultrasonic device may have promoted pollination in tomatoes, although this machine should be tested on tomatoes that are free of whiteflies in future.
The importance of IPM that does not solely rely on pesticides is increasing because of the environmental burden caused by chemical pesticides, the emergence of insecticide-resistant pests, and consumer preference for reduced pesticide use. Methods other than pesticide control are also required for greenhouse cultivation. Compared with open-field cultivation, greenhouse cultivation is less susceptible to ultraviolet rays, wind, and rain and has a relatively stable temperature. Therefore, pests can quickly multiply after they invade greenhouses (van Lenteren 2000). Under conventional chemical sprays in greenhouses, pests are continuously exposed to pesticides, and pesticide resistance is more likely to develop in greenhouses than in open-field cultivation (Roditakis et al. 2005). This situation has created a demand for alternative control methods, including physical control methods (Vincent et al. 2003). The pest suction machine newly developed and used in this study is one of the most crucial physical control methods.
In this study, whitefly adults were captured and killed in the machine under different environmental conditions during 4 trials conducted in spring and autumn. The suction machine reduced the number of whiteflies on tomato plants by 25% and 10% in spring and autumn, respectively; the number of whitefly adults on tomato plants in autumn was significantly lower in the treatment blocks than in the untreated blocks. Thus, this machine has the potential to control whitefly adults for months.
Pest suction machines which have been developed for open-field crop cultivation so far blow pests away by the airflow, requiring strong wind and negative pressure to detach pests from plants, which in turn requires large engines. However, because greenhouses are closed environments, whiteflies brown off will return to the tomato plants. The pest suction machine tested in this study drives pests away from the plant using vibration, attracting expelled pests to the suction port by light and suction, and finally killing the insects. This novel machine had the following advantages over existing pest suction machines. First, because the machine did not require strong airflow or wind pressure, it was compact and battery-powered. Greenhouses generally have narrow aisles, making it difficult for large machines to operate. Second, battery-powered machines do not exhaust gases into the enclosed environment of greenhouses. Thus, the suction mechanism of our machine may be applied to tomato plants and other crops frequently infested by whiteflies. Finally, the wavelength of LEDs can be easily changed depending on the pest species. By changing the wavelength of light, it may be possible to control whitefly adults and other insect pests (such as leafminer flies) using this machine. Although this machine was manually radio-controlled in this study, it was originally designed to be controlled automatically. Therefore, the system can be used more frequently, contributing to a much higher efficacy for whitefly control. If we install a larger battery than the one we are using now to extend the operating hours and run at night, avoiding the need for manual labor.
IPM requires the integrated use of all possible control measures economically. Although the effect shown by the suction machine used in this study was limited, Future optimization is expected to improve effectiveness. Physical control can be an essential component of IPM programs, where chemical pesticides are reduced. If the machine is used with other control methods, such as biological control agents, it can contribute to the establishment of new environmentally friendly IPM programs.
Acknowledgments
We thank Dr. Katsuyuki Kohno of the Institute of Vegetable and Floriculture Science, NARO, Japan, for providing statistical advice. Dr. Haruki Katayama gave us the opportunity to do this study. We are grateful to Mr. Isao Hayashi, Mr. Nariichi Suzuki, Mr. Masaaki Nakayama, Mr. Shizuo Ohtsuka, and Ms. Takeko Suzuki for helping in field management and fruit harvesting. TERADA SEISAKUSHO Co., Ltd. assisted in manufacturing this machine. We would also like to acknowledge the Kyusyu Okinawa Agricultural Research Center, NARO, Japan, for providing Bemisia tabaci biotype Q. This research was supported by the research program on development of innovative technology grants (JPJ007097) from the Project of the Bio-oriented Technology Research Advancement Institution (BRAIN).
Author Contributions
Chiharu Saito (Conceptualization [lead], Investigation [lead], Methodology [lead], Writing—original draft [lead], Writing—review & editing [lead]), Eiichi Makita (Investigation [supporting]), Suguru Yamane (Investigation [supporting]), Chihiro Urairi (Conceptualization [supporting], Resources [lead]), Takayuki Hoshi (Resources [equal], Software [lead]), Makoto Doi (Conceptualization [supporting], Methodology [supporting]), Suzuka Yoshizaki (Methodology [supporting]), and Norihide Hinomoto (Formal analysis [lead], Writing—original draft [supporting], Writing—review & editing [supporting])
