Key messages

  • Behavioral manipulation of the greenhouse whitefly was studied to optimize insect mass trapping

  • Blue and UV LEDs provided strong stimuli on disturbing whiteflies from the host plant

  • Tailored green LED traps attracted substantially more whiteflies than common yellow sticky traps

  • Combining disturbing and attractive visual cues contributes to novel mass trapping methods in IPM

Introduction

The greenhouse whitefly Trialeurodes. vaporariorum (Westwood) (Hemiptera: Aleyrodidae) belongs among the most prominent pests of glasshouse crops in Europe, and classifies as one of the most important threats of glasshouse tomato (Solanum lycopersicum) crops (Nauen et al. 2014; Perring et al. 2018). Although whitefly control still heavily relies on conventional insecticides (George et al. 2015), whiteflies have shown resistance to the most important of them (Wardlow et al. 1976; Gorman et al. 2002, 2007). As the European Union (EU) has implemented regulations toward the reduction of chemical insecticides (Hillocks 2012), the adoption of alternative biocontrol methods, that promote a more sustainable approach to pest control, has become increasingly important (Prokopy and Kogan 2009; Stenberg 2017; IOBC-WPRS, IBMA, PAN-Europe 2019). In our study, we aimed on the development of such a method by visually deterring greenhouse whiteflies from their host, in order to lure them toward light traps, and selectively eliminate them using laser applications, reducing thus their population in the crop.

Whiteflies largely rely on visual cues, such as color and shape, to identify suitable host plants from a distance, and upon landing, they deploy olfactory and gustatory senses to confirm the suitability of the plant for feeding or oviposition (Prokopy and Owens 1983; Finch and Collier 2000; Bleeker et al. 2011; Dixon 2012). T. vaporariorum visual orientation is dependent on wavelength (wavelength-specific behavior) and light intensity (Stukenberg and Poehling 2019; Athanasiadou and Meyhöfer 2023), and is essential for early host plant detection, settlement, and dispersal (Vaishampayan et al. 1975; Coombe 1981; Moericke et al. 1966). This indicates that optical manipulation of whiteflies can be used to reduce their settlement and numbers on a crop, acting, thus, as an important integrated pest management (IPM) tool. Early research on the visual orientation of T. vaporariorum showed a “settling” response on yellow-reflecting surfaces (Lloyd 1921; Moericke et al. 1966; Vaishampayan et al. 1975), which led to the use of yellow sticky traps (YST) to monitor populations, ensure prompt targeted action (Muvea et al. 2014; Sampson et al. 2021; Van Tol et al. 2012), and contribute to mass trapping as a supplementary control measure (Reitz et al. 2020; Dearden et al. 2024). Subsequent studies unveiled the preference for monochromatic green–yellow light peaking at 550 nm (MacDowall 1972; Coombe 1981, 1982; Mellor et al. 1997), and Chu et al. (2004) demonstrated that significantly more whiteflies were caught on a YST equipped with lime green (530 nm) LEDs than an unlit one. Stukenberg et al. (2015) and Stukenberg and Poehling (2019) showed whitefly preference for green (550 nm) over yellow LEDs (574 and 590 nm) and YST. This finding raised the assumption that the highly attractive green LED trap could contribute to the development of a more effective mass trapping, especially when combined with light that disperses whiteflies from their host. Some ecological factors that contribute to the dispersal of whiteflies from their habitat are the quality deterioration of the host plant due to physiological changes, insufficient nutrients for feeding and oviposition, competition with other pests, and presence of natural enemies (Sehgal et al. 2006). Deterrent light serves as an additional factor in whitefly disturbance signaling low resource quality and, according to previous studies, blue light has shown this property (Athanasiadou and Meyhöfer 2023). Additionally, attractiveness to green light was hindered when simultaneously combined with blue light (447 and 469 nm) and blue light alone showed no attractivity for adult whiteflies (Stukenberg et al. 2015), confirming earlier research on its inhibitory impact on T. vaporariorum (Vaishampayan et al. 1975; Coombe 1981; Affeldt et al. 1983). Based on these findings, a trichromatic visual system was proposed for T. vaporariorum, with a blue–green opponent mechanism, where an overstimulated blue photoreceptor inhibits a neighboring green one, responsible for settling on the leaf (Stukenberg and Poehling 2019), a process already identified in other herbivorous insects (Döring and Chittka 2007; Döring and Röhrig 2016). Significantly less whiteflies settled on broccoli and lettuce plants when their soil was covered with white foil reflecting in the 400–490 nm spectrum compared to plants growing on bare soil (Niemann et al. 2021), and presence of blue LEDs below lettuce and poinsettia plants led to reduced number of T. vaporariorum landing on them, compared to plants without blue LEDs (LUH patent DE 10 2018 208 424 B3) (Niemann and Poehling 2022). While these studies investigated the settlement of whiteflies on already blue illuminated plants and used blue light as a preventive control technique, Athanasiadou and Meyhöfer (2023) demonstrated that blue light (465 nm) repelled significantly more whiteflies already settled on tomato leaves compared to non-exposed to light leaves, testing it as a curative control technique.

These findings can be utilized to manipulate lighting conditions in a crop setting in order to reduce whitefly infestation and crop damage. In our study, disturbing LED light was used to visually startle whiteflies from host plants, causing them to disperse, and subsequently land on an attractive surface other than the plant i.e., a green LED-enhanced YST, contributing thus to the development of a new push–pull strategy (Athanasiadou and Meyhöfer 2023). After landing on the trap, whiteflies can get eliminated by use of technological means, such as identification by a camera system and application of low power, lethal, laser beams, a technique developed in the joint research project “LichtFalle”, funded by the German Ministry of Food and Agriculture (www. hortico40. de).

This is the first investigation on the effect of disturbing lights on greenhouse whiteflies settled on a host plant (push) and the contribution of a trap on the whitefly recapture (pull), in laboratory and greenhouse assays. For our experiments, we took into account the findings of Athanasiadou and Meyhöfer (2023) who showed the property of blue and blue + UV light to efficiently disturb whiteflies from their host when illuminating continuously and in high intensity. Therefore, we explored the effect of blue and blue + UV LEDs as repelling cues illuminating tomato plants infested with whiteflies, while a YST or a green LED-enhanced YST (further referred to as “green LED trap”) were used as attractive cues, in a climate chamber experiment. Additionally, in a scaling-up experiment in the greenhouse, in order to examine if the position of blue LEDs had an effect on the whitefly disturbance, we tested them in different arrangements around whitefly infested tomato plants, in the presence of green LED traps.

Materials and methods

Rearing of the greenhouse whiteflies

All greenhouse whiteflies (T. vaporariorum) used in this study were reared on tomato plants (Solanum lycopersicum L. cv. Brioso) in a gauze cage (90 × 60 × 60 cm), in a climate chamber, at 23 ± 3 °C, 50% ± 5% relative humidity (RH) and light:dark 16:8 h, at the Institute of Horticultural Production Systems, Leibniz Universität Hannover, Germany. Tomato plants were grown in pots (h × d = 10 × 12 cm) under greenhouse conditions, before introduction to the rearing and experiments.

Technical LED setup

For the experiment in the climate chamber, single chip LEDs (1.0 × 1.0 cm) were purchased (Luminotrix® LED-Technik GmbH, Hechingen, Germany) in blue (465 nm peak wavelength, 1 W, 3 V, Nichia NCSB219B-V1) and UV (365 nm peak wavelength, 2 W, 4 V, Nichia NCSU275). Aluminum plates (5 × 5 × 0.1 cm) were constructed, and four LEDs were glued with thermal conductive double-sided adhesive tape (Ak-tt12-80, Akasa Ltd., Greenford, UK) on them. The deterrent LED colors tested were blue i.e., four blue LEDs attached on a plate, and blue + UV (mix) i.e., two blue and two UV LEDs attached on a plate. Four blue or four mix plates were placed on a square metallic frame (25 × 25 cm), one plate in the middle of each side (Fig. 1). Maximum intensity of LEDs was used and it was measured by placing the sensor in the center of an aluminum plate with four LEDs on it, at 15 cm distance, in darkness. The intensity of four blue LEDs was 50 μmol/m2/s and was measured with the LI 250 Light Meter and the LI 190 Quantum Sensor (LI-COR Biosciences GmbH, Bad Homburg, Germany), a suitable sensor to measure broadband photosynthetic active radiation (PAR, 400–700 nm). The intensity of four mix LEDs was 30 μmol/m2/s when measured with the same device and 12 μmol/m2/s when measured with the datalogger Almemo® 2390–5 in combination with the UV-A Sensor type 2.5 (Ahlborn GmbH, Bodenwerder, Germany), suitable only for measuring UV radiation. This resulted in a total intensity of 42 μmol/m2/s (30 + 12) for the mix. All LEDs were used at maximum intensity, at 350 milliamperes (mA) operating current. The intensity of UV LEDs in W/m2 was converted into μmol/m2/s using the LED spectra, Planck’s constant, and Avogadro’s number.

Fig. 1
figure 1

Top view of the experimental setup in the climate chamber, showing the blue LEDs treatment (left side) and the blue + UV LEDs treatment (right side): 1: tomato plant, 2: blue LED (single chip), 3: UV LED (single chip), 4: aluminum plate, and 5: metallic frame

Full size image

For the experiment in the greenhouse, spotlights with blue LEDs (456 nm, 17.8 × 22 cm, 30 W, 100–240 V, 80° reflection angle, LED IP FL-30 SMD, Thomann GmbH, Burgebrach, Deutschland) were mounted on metallic stands and used in different arrangements i.e., horizontal, vertical and double vertical (Fig. 2). Details on the arrangement of the spotlights are explained below. The overall intensity of the spotlights for each arrangement was measured with the LI 250 Light Meter and the LI 190 Quantum Sensor (LI-COR Biosciences GmbH, Bad Homburg, Germany). The intensity of a single blue spotlight was 80 μmol/m2/s at 40 cm (plant’s position) and 510 μmol/m2/s at 15 cm distance, measured by placing the sensor at the center of the spotlight, in darkness. The intensity of three spotlights in the horizontal, three spotlights in the vertical, and six spotlights in the double vertical arrangement was 108 μmol/m2/s, 115 μmol/m2/s, and 186 μmol/m2/s, respectively, with the sensor placed at 40 cm distance and facing the spotlights.

Fig. 2
figure 2

Scheme of the vertical (left side) and horizontal (right side) arrangement of the blue spotlights in the greenhouse experiment. The double vertical arrangement consisted of 2 vertical stands (6 spotlights in total). 1: blue spotlight and 2: metallic stand

Full size image

The green LED trap used in our experiments was a modified model of the trap used by Stukenberg (2018). Enclosed in an aluminum frame (24 × 20 cm) was a non-sticky yellow card (22 × 18 cm) (IVOG biotechnical systems GmbH, Vogelsang, Germany) and eight green LED chips (520 nm peak wavelength, 1.0 × 1.0 cm, 1 W, 3 V, Nichia NCSG219BT-V1, Luminotrix® LED-Technik GmbH, Hechingen, Germany), facing inwards (2 per side) (Fig. 3). The size of the illuminating area of the trap was 16 × 12 cm. Transparent plastic film (Rico Design GmbH & Co. KG, Brakel, Germany) was attached on the top of the trap and coated with insect glue (Insektenleim, Temmen GmbH, Hattersheim, Germany) for whitefly capturing. The intensity of the light trap was 5 μmol/m2/s and was measured with the LI 250 Light Meter and the LI 190 Quantum Sensor (LI-COR Biosciences GmbH, Bad Homburg, Germany) 25 cm (plant’s position) from the center of the trap, in darkness. All green LEDs were used at maximum intensity, at 350 mA operating current. Regarding the conventional YST, a YST (IVOG biotechnical systems GmbH, Vogelsang, Germany) was cut into the same size as the illuminating area of the LED trap, and only one side of it was accessed by the whiteflies.

Fig. 3
figure 3

Scheme of the green LED trap: a Cross-section showing the aluminum frame enclosing the green LEDs. All the layers from the bottom: PVC plate, yellow card, LED scattering acrylic glass, black cardboard frame and transparent acrylic glass cover and b top view scheme showing trapping surface, black cardboard frame and layout of green LEDs inside the aluminum frame

Full size image

Experimental design and procedure

Two experiments were conducted in sequence to study the disturbing effect of selective LEDs against T. vaporariorum settled on a host, and the contribution of a trap on their recapture: a climate chamber and a greenhouse experiment, both at the Institute of Horticultural Production Systems, Leibniz Universität Hannover, Germany. In the climate chamber, wavelengths of blue and mix LEDs were tested as deterrent factors in the presence of a YST and a green LED trap, both used as attracting factors. In the greenhouse experiment, blue LED spotlights in different arrangements and intensities were tested as deterrent factors along with a green LED trap as attracting factor. The green LED trap showed higher efficacy than the YST in the chamber experiment and was therefore chosen for the greenhouse experiment. A general overview of the experimental designs is given in Table 1. In the following paragraphs, the specifics of each experiment are described.

Table 1 Overview of the design and characteristics of the experiments
Full size table

Experiment under controlled conditions

This experiment was conducted in a climate chamber, at 23 ± 3 °C, 50% ± 5% RH, light:dark 16:8 h. A 7-week-old tomato plant (55–60 cm) was placed on a table, 1.2 m below the fluorescent tubes of the chamber. The luminance of the fluorescent tubes in the chamber was 6500 ± 200 lx and was measured from 1.2 m distance (plant’s position) with the datalogger HOBO U12-012 (Datenlogger-Store, Eichstetten, Germany). A square metallic frame with blue or mix LEDs was placed as deterrent light source around the tomato plant at soil level, with the LEDs in a horizontal arrangement, facing upward toward the plant canopy (Fig. 1). The distance between the plant and each side of the metallic frame was 13 cm. A YST or a green LED trap was positioned at 30 cm height, facing toward the plant, with a distance of 25 cm in between. For both traps, whiteflies were caught only on the side of the trap facing toward the plant. The whole setup was enclosed in a gauze cage (90 × 60 × 60 cm). To establish the pests on the plant, 40 adult greenhouse whiteflies were released from a glass vial (h × d = 5 × 2.5 cm) into the cage about 24 h before the LEDs were switched on, to allow their settlement on the plant. Prior to switching on the LEDs, the glass vial was removed and whiteflies on the plant were counted. Whiteflies found on other surfaces were manually removed out of the cage. All the LEDs, including the green LED trap, were switched on for 2 h, which was the duration of each trial, and after that, whiteflies were counted on the plant and the trap. The whiteflies that flew off the plant were considered disturbed and were located either glued on the trap or on other surfaces within the cage. After counting, whiteflies on the plant and other surfaces within the cage were manually removed and discarded, and the plants were replaced every two trials. The same setup, without any deterrent LEDs but with a trap present, was included as control. Fifteen replicates were performed consecutively (three replicates per day) for each LED color and trap, in a randomized block design, and with thick carton sheets (90 × 60 cm) separating the cages to avoid interference of neighboring LEDs.

Experiment under greenhouse conditions

This scaling-up experiment was conducted between January and February 2023 (calendar weeks 3–8), in a greenhouse cabin (15 × 7 m), at 22–25 °C and 40% ± 5% RH, measured with the datalogger HOBO U12-012 (Datenlogger-Store, Eichstetten, Germany). Supplementary light from sodium vapor lamps (Son-t Agro 400 W) was applied from 6 am to 10 pm. Eight-week-old tomato plants in 10 L pots with Klasmann TS1 soil (Klasmann–Deilmann GmbH, Geeste, Germany) were placed into the greenhouse cabin, approx. 2.7 m below the greenhouse lamps. They were irrigated 3 times per day, for 3 min, at 8-h intervals, and 0.01% Ferty 3 MEGA fertilizer (Nitsch & Sohn GmbH & Co. KG, Kreuztal, Germany) was mixed in the irrigation water. The plants reached 1.5 m height when the experiment started and were maintained at this height during the experiment by cutting their main shoot. Moreover, fruits and side shoots were always removed. The plants were placed at 3 m distance and separated by a black polyethylene foil, to avoid interference of neighboring LEDs. Each plant had a distance of 40 cm from the blue spotlights and 50 cm from the green LED trap, which was placed on the opposite side of the spotlights, at 76 cm height. The blue spotlights were arranged in a horizontal, vertical, and double vertical way. In the horizontal arrangement, spotlights were positioned at 25 cm height facing upward, in a 45° angle with the plant. The distance between each spotlight was 56 cm. In the vertical arrangement, three spotlights were attached on a metallic stand above each other at 50, 100, and 150 cm height from the floor, respectively. In the double vertical arrangement, there were two metallic stands with three spotlights each, placed in the same way as the vertical arrangement. The distance between the two stands was 60 cm (Fig. 4). Thirty adult greenhouse whiteflies were released 20 h before switching on the LEDs, in three different heights of a plant, by placing three cage-clips (10 whiteflies/cage-clip) in the bottom, middle, and top of the plant canopy, on those sides of the plant facing the spotlights and not the trap. After 24 h, whiteflies found on leaves without clips were manually removed and discarded; then, the clips were removed and whiteflies were counted on the respective leaves. Afterward, blue spotlights and green LED traps were switched on for 4 h, which was the duration of the experiment, starting always at 12:00. The duration of this experiment was extended to 4 h instead of 2 h, in order to increase the whitefly recapture rate; in preliminary greenhouse trials exposure to LEDs for less than 4 h did not result in a substantial number of trapped whiteflies. After the LEDs were switched off, whiteflies were counted on the plant and the green LED traps. The whiteflies that flew off the plant and did not return to it after take-off were considered disturbed. After counting, whiteflies were manually removed from all surfaces and discarded. The same setup, without blue LEDs but with a trap present, was included as control. The plants were arranged in two rows, each facing the east and west, respectively, and all blue light arrangements were repeated in both sides. In this experiment, 30 replicates were performed consecutively (one replicate per day) for each blue light arrangement.

Fig. 4
figure 4

Scheme of the experimental setup in the greenhouse: a horizontal arrangement of three blue spotlights surrounding the plant, b vertical arrangement of three blue spotlights placed above each other on a stand, and c double vertical arrangement of six blue spotlights on two stands (three placed above each other on one stand). 1: tomato plant, 2: blue spotlight, and 3: green LED trap

Full size image

Statistical analysis

Data were statistically analyzed with R version 4.2.1 (R Core Team, 2019). For the climate chamber experiment, to compare the number of whiteflies that left the plant from the bottom and top canopy height (response variable) between the different treatments (explanatory variable), one-way analysis of variance (ANOVA) followed by Tukey’s HSD (α = 0.05) was used, since the data met the assumptions of normality and homogeneity. For the greenhouse experiment, the number of whiteflies that left the plant (response variable) between three canopy heights (explanatory variable) was compared using Kruskal–Wallis rank-sum test with Dunn’s post hoc and Holm-adjusted p-values (α = 0.05), since the data did not meet assumptions of normality and homogeneity (Pohlert 2018). To compare the number of whiteflies on the plant (response variable) on each canopy height between before and after the experiment (explanatory variable), one-way analysis of variance (ANOVA) followed by Tukey’s HSD (α = 0.05) was used, since the data met the assumptions of normality and homogeneity. The rate of whiteflies that were disturbed, trapped out of disturbed, and trapped out of released (response variables) for different LED treatments and types of traps (explanatory variables) was analyzed using generalized linear model (GLM), assuming a quasibinomial distribution (count data with overdispersion) (McCullagh and Nelder 1989; Cox et al. 2019). A deviation analysis (F-test) running on the logit link was fitted to determine influences of the explanatory variables on the number of whiteflies settled on tomato plants (McCullagh and Nelder 1989; Demétrio et al. 2014). Subsequent Tukey-type multiple pairwise comparisons at α = 0.05 using the R-package “emmeans” (Lenth 2019) were conducted to clarify which treatment differed from another (mean value differences) in each of the experiments. All figures showing results are boxplots and were made using R (version 4.2.1) and the ggplot2 package (Wickham 2016).

Results

Experiment under controlled conditions

The percentages of disturbed and trapped out of disturbed whiteflies represent the percentage of T. vaporariorum that took-off the plant and were caught on the trap after take-off, respectively, during the 2 h experiment. The percentage of trapped out of released whiteflies demonstrates the amount of the initially released whiteflies that were caught on the trap by the end of the experiment. Regarding the percentage of disturbed whiteflies, the results showed that blue and mix wavelength had an effect on whitefly disturbance both in the presence of YST (F2,42 = 4.57, p = 0.01) and green LED trap (F2,42 = 22.41, p < 0.001). In the trials with YST, blue and mix wavelength disturbed 2.5 and 2.7 times more whiteflies, respectively, compared to the control (both p < 0.05). This shows that approximately eight and three out of 40 insects flew off the plant in the presence of deterrent light and the control, respectively (Fig. 5). In the trials with green LED trap, blue and mix wavelength deterred about 14 and 15 out of 40 whiteflies, respectively, whereas the control only about six (both p < 0.001) (Fig. 5). Therefore, the disturbing LEDs induced a significant take-off behavior compared to the control regardless of the trap used. However, the type of the trap played an important role on the mobility of whiteflies since, with the green LED trap, take-off behavior was significantly higher compared to YST for blue (F1,28 = 10.73, p = 0.001), mix (F1,28 = 10.16, p = 0.002) and the control (F1,28 = 8.85, p = 0.004) (Fig. 5).

Fig. 5
figure 5

Disturbed whiteflies that left the tomato plant during 2 h of exposure to blue and blue + UV LEDs, with a yellow sticky trap or a green LED trap present, under controlled conditions. Means are indicated with red dots and labeled with numbers. Significant differences between the effect of the treatments and the trap type are indicated by asterisks (GLM, pairwise mean comparisons: *0.01 < 0.05, **0.001 < 0.01, ***p < 0.001), n = 15

Full size image

Regarding the trapped out of disturbed whiteflies, no effect of the deterrent wavelengths was observed when using YST (F2,42 = 0.13, p = 0.91) or green LED trap (F2,42 = 0.07, p = 0.93). In the presence of green LED trap, more than 87% of disturbed whiteflies were caught in all treatments (Fig. 6). The type of the trap had an effect on the catching rate of whiteflies, as the green LED trap caught 29% and 50% more whiteflies compared to the YST, under blue (F1,28 = 5.32, p = 0.02) and mix (F1,28 = 6.86, p = 0.01) light, respectively. In the absence of deterrent LEDs (control), no significant difference between the efficacy of the traps was noted (F1,28 = 3.04, p = 0.09) although green LED trap caught 2.3 times more whiteflies compared to YST (Fig. 6).

Fig. 6
figure 6

Whiteflies caught on a yellow sticky trap or a green LED trap after they left the tomato plant, during 2 h of exposure to blue and blue + UV LEDs, under controlled conditions. Means are indicated with red dots and labeled with numbers. Significant differences between the effect of the treatments and the trap type are indicated by asterisks (GLM, pairwise mean comparisons: *0.01 < 0.05), n = 15

Full size image

Blue and mix wavelength had an effect on the trapped out of released whiteflies, both in the presence of YST (F2,42 = 3.97, p = 0.03) and green LED trap (F2,42 = 22.98, p < 0.001). In the trials with YST, blue and mix led to about six trapped out of 40 released whiteflies, while the control showed less than one trapped whitefly; however, only blue differed significantly (p = 0.04). In the trials with green LED trap, blue and mix wavelength led to about 12 and 13 caught out of 40 released whiteflies, respectively, while in the control, only six were caught (both p < 0.001). Therefore, the deterrent LEDs led to higher recapture rate compared to the control regardless the trap used. However, the type of the trap had an effect on the number of trapped whiteflies since green LED trap caught significantly more of the released whiteflies compared to YST for blue (F1,28 = 7.6, p = 0.01), mix (F1,28 = 10.15, p = 0.003), and the control (F1,28 = 15.46, p < 0.001) (Fig. 7).

Fig. 7
figure 7

Whiteflies caught on a yellow sticky trap or a green LED trap from the initially released population, during 2 h of exposure to blue and blue + UV LEDs, under controlled conditions. Means are indicated with red dots and labeled with numbers. Significant differences between the effect of the treatments and the trap type are indicated by asterisks (GLM, pairwise mean comparisons: *0.01 < 0.05, **0.001 < 0.01, ***p < 0.001), n = 15

Full size image

The LED wavelength had an effect on the number of whiteflies that left the lower half of the plant during the experiment, both in the presence of a YST (F2,42 = 5.07, p = 0.01) and a green LED trap (F2,42 = 13.14, p < 0.001). Significantly more whiteflies left the lower half canopy in the presence of blue and mix wavelength compared to the control, regardless of the trap type (all p < 0.05). However, no differences were observed between the treatments on the number of whiteflies that took-off from the upper half of the plant canopy, both in the presence of a YST (F2,42 = 1.27, p = 0.29) and a green LED trap (F2,42 = 1.43, p = 0.25) (Fig. 8, 9).

Fig. 8
figure 8

Whiteflies on the lower (bottom) and upper (top) half of the plant canopy, before and after 2 h of exposure to blue and blue + UV LEDs, with a yellow sticky trap present, under controlled conditions. Means are indicated with red dots and labeled with numbers. Significant differences between the three treatments on the reduction of whiteflies during the experiment from the bottom and the top canopy height are indicated by asterisks. There were no significant differences between the treatments on the reduction of whiteflies from the top canopy height (ANOVA followed by Tukey’s HSD, α = 0.05: *0.01 < 0.05, **0.001 < 0.01, ***p < 0.001), n = 15

Full size image
Fig. 9
figure 9

Whiteflies on the lower (bottom) and upper (top) half of the plant canopy, before and after 2 h of exposure to blue and blue + UV LEDs, with a green LED trap present, under controlled conditions. Means are indicated with red dots and labeled with numbers. Significant differences between the three treatments on the reduction of whiteflies during the experiment from the bottom and the top canopy height are indicated by asterisks. There were no significant differences between the treatments on the reduction of whiteflies from the top canopy height (ANOVA followed by Tukey’s HSD, α = 0.05: *0.01 < 0.05, **0.001 < 0.01, ***p < 0.001), n = 15

Full size image

Greenhouse experiment

The presence of blue spotlights had an effect on the disturbance of whiteflies (F3,116 = 140.71, p < 0.001), since all treatments with blue light induced significantly higher take-off behavior compared to the control (all p < 0.001). Moreover, there were significant differences between the arrangements, with the double vertical being the most effective, as it repelled 12.8 times more whiteflies than the control (p < 0.001) and nearly 15 out of 30 settled whiteflies. The vertical and horizontal arrangements followed, repelling about eight and six out of 30 whiteflies, respectively, whereas the control only about one (both p < 0.001) (Fig. 10). Concerning the trapped out of disturbed whiteflies, no effect of blue light arrangement was demonstrated (F3,116 = 1.25, p = 0.29). The highest number of disturbed whiteflies was caught in the horizontal arrangement, but without differing significantly from the other treatments (Fig. 11). However, the presence of blue light showed a significant effect on the percentage of trapped out of released whiteflies, regardless the arrangement (F3,116 = 196.57, p < 0.001), with most efficient the double vertical, resulting to 11.7 times more trapped whiteflies than the control (p < 0.001), which corresponds to approximately eight trapped out of 30 released whiteflies (Fig. 12).

Fig. 10
figure 10

Disturbed whiteflies that left the tomato plant during 4 h of exposure to blue LEDs in different arrangements, with a green LED trap present, under greenhouse conditions. Means are indicated with red dots and labeled with numbers. Significant differences between the effect of the treatments are indicated by different letters (GLM, pairwise mean comparisons, α = 0.05), n = 30

Full size image
Fig. 11
figure 11

Whiteflies caught on the green LED trap after they left the tomato plant, during 4 h of exposure to blue LEDs in different arrangements, under greenhouse conditions. Means are indicated with red dots and labeled with numbers. Differences between the effect of the treatments were not statistically significant, hence, represented by the same letters (GLM, α = 0.05), n = 30

Full size image
Fig. 12
figure 12

Whiteflies caught on the green LED trap from the initially released population, during 4 h of exposure to blue LEDs in different arrangements, under greenhouse conditions. Means are indicated with red dots and labeled with numbers. Significant differences between the effect of the treatments are indicated by different letters (GLM, pairwise mean comparisons, α = 0.05), n = 30

Full size image

Since whiteflies were released at three different heights on each plant, we compared the disturbed whiteflies between each plant height for all treatments. The results showed that only in the horizontal arrangement, there were notable differences between the number of disturbed whiteflies from the bottom, middle, and top canopy height (F2,87 = 97.48, p < 0.001), as significantly more whiteflies were disturbed from the bottom compared to the middle and top (both p < 0.001) (Fig. 13). This result was confirmed by the number of whiteflies found on the bottom of the plant after the experiment, which was significantly lower than their number before the experiment (F1,58 = 155.5, p < 0.001) (Fig. 14a). No significant changes on the number of whiteflies on the middle (F1,58 = 2.91, p = 0.09) and top (F1,58 = 3.23, p = 0.08) part were detected after the experiment in the horizontal arrangement (Fig. 14a). Contrastingly, in the vertical and double vertical arrangements, the number of whiteflies decreased significantly across all canopy heights after the experiment (all p < 0.001) (Fig. 14b, c), whereas no changes appeared in the control (all p > 0.05) (Fig. 14d).

Fig. 13
figure 13

Disturbed whiteflies that left from the bottom, middle, and top plant canopy height during 4 h of exposure to blue LEDs in different arrangements, with a green LED trap present, under greenhouse conditions. Means are indicated with red dots and labeled with numbers. Significant differences between the canopy heights are indicated by different letters for each treatment (Kruskal–Wallis followed by Dunn’s post hoc, α = 0.05), n = 30

Full size image
Fig. 14
figure 14

Whiteflies on the bottom, middle, and top plant canopy height, before and after 4 h of exposure to three blue spotlights in horizontal (a) and vertical (b) arrangement, six blue spotlights in double vertical (c) arrangement, and the control (d), with a green LED trap present, under greenhouse conditions. Means are indicated with red dots and labeled with numbers. Significant differences on the number of whiteflies on the plant between before and after the experiment for each canopy height individually are indicated by asterisks (ANOVA followed by Tukey’s HSD, α = 0.05: ***p < 0.001), ns: not significant differences, n = 30

Full size image

Discussion

The findings of this study reveal the property of blue light to efficiently disturb T. vaporariorum from a host plant and lead them toward a trap, in both laboratory and greenhouse conditions. Additionally, they confirm the broad field of the whitefly visual system suggested by Athanasiadou and Meyhöfer (2023), since in our experiments, whiteflies responded to disturbing LEDs illuminating upwards and attractive green LEDs illuminating sideways. Our findings validate that the green LED trap exhibits higher attractivity than the YST and provides additional evidence that the presence of blue light enhances the attractivity of both traps.

The results under controlled conditions in the climate chamber demonstrate the quality of blue and blue + UV light to repel significantly more whiteflies from a host plant compared to the control, regardless of using a green LED trap or a YST. This suggests the property of blue color to disrupt the perception of green color—reflecting from the leaves-by concealing it, causing the whiteflies to take-off. Given that whiteflies were already settled and feeding on leaves, it can be proposed that the interference of blue color overrides the feeding signal, even when the whitefly’s piercing mouthparts are already inserted into the leaf tissue. Insects’ photoreceptors employ different mechanisms to adjust their responses to perceived light against ambient light (Laughlin and Hardie 1978; Arshavsky 2003). This technique prevents photoreceptor overstimulation and functions as a method to maintain a consistent perception of color (Foster 2011; Kemp et al. 2015). In our experiment, exposure to blue and blue + UV light for 2 h led to significantly more disturbed whiteflies compared to the control, suggesting the difficulty of whitefly photoreceptors to adapt to the perceived deterrent light, which induced their overstimulation (Howard et al. 1987; Ignatova et al. 2021). This result corroborates with the time threshold of 2 h proposed by Athanasiadou and Meyhöfer (2023), who demonstrated that significantly more whiteflies took-off from leaves illuminated with blue and UV light for 2 h, compared to the ones illuminated for a shorter time period. Beyond this threshold, deterrent lights evidently indicate poor host plant quality or potential danger, forcing the whiteflies to abandon their valuable resource and seek an alternative one, regardless of feeding or ovipositing. This indicates the importance of visual signals in eliciting whitefly behavioral responses at short distances, which is also known for other herbivores (Hawkes 1974; Prokopy and Owens 1983; Ren et al. 2020). In this experiment, the plants were closer to the roof lights compared to the plants in the greenhouse experiment, and as a result, the upper half of them was stronger illuminated than the lower one. Therefore, since whiteflies avoid exposure to intense light, as evidenced by their tendency to settle on the underside of leaves (Lenteren and Noldus 1990), they accumulated mostly on the lower half of the canopy over the 24 h before the experiment. As seen from our results, more whiteflies were deterred from the lower half of the plant compared to the upper one. This could be attributed to the fact that, since the deterrent LEDs were situated below the plant, the whiteflies settled there perceived the light in higher intensity than the ones settled on the upper canopy half, where the leaves were exposed in fewer rays and lower light intensity, as light was partially blocked by the canopy in between. In a natural setting, whiteflies are only slightly exposed to blue wavelength on the underside of a leaf, where they settle. Hence, it is suggested that introduction of blue light above a specific time threshold and intensity creates unfavorable or potentially hazardous conditions for whiteflies, prompting them to take-off from their host as a survival mechanism (Athanasiadou and Meyhöfer 2023). In the presence of blue and blue + UV LEDs, both the green LED trap and the YST captured a significantly higher number of released whiteflies compared to their absence. This indicates that the presence of these deterrent lights leads to higher attractivity and therefore efficacy of both traps, by increasing the number of whiteflies that disperse from the host. Significantly higher number of settled and disturbed whiteflies was attracted to the green LED trap than to the YST, regardless the treatment. In preliminary greenhouse trials, we allowed a whitefly population to build up over 2 months on a tomato crop, with both a green LED trap and a YST present. The results showed that the green LED trap captured a significantly higher number of whiteflies (8,930 individuals) compared to the YST (920 individuals). Additionally, Stukenberg et al. (2015) and Stukenberg (2018) showed that preference for the green LED trap over the YST was apparent only in choice experiments or when the whitefly population on the plants was very high (> 400 individuals). However, in our no-choice scenario, with a smaller whitefly population, there was a clear preference for the green LED trap over the YST, suggesting the possibility to use a YST enhanced with green LEDs for effective monitoring and/or mass trapping even when whitefly populations on plants are still low. Moreover, the significantly higher trapping performance of the green LED trap compared to the YST in the control suggests the property of the former to trigger whitefly movement from the canopy, serving as an alternative attractive platform; a whitefly, after take-off, is more likely to land on a surface illuminating at 520 nm—which correlates closely to the peak reflectance of green leaves—than a YST, which has a broad reflection spectrum (Stukenberg et al. 2015). Sensitivity peak at the green region is also known for other herbivores; photoreceptors of the green peach aphid Myzus persicae and pea aphid Acyrthosiphon pisum peaked at 490–530 nm and 518 nm, respectively (Kirchner et al. 2005; Döring et al. 2011). This suggests that the green-sensitive photoreceptor is a common feature among herbivores, indicating its importance in host plant detection (MacDowall 1972; Prokopy and Owens 1983; Döring et al. 2009; Kelber and Osorio 2010). Apart from the fitted wavelength of the green LED trap, its light intensity is consistent irrespectively of the sunlight conditions, compared to the reflected light intensity of the YST, which is highly dependent on the ambient light conditions (Johansen et al. 2011; Zhang et al. 2020; Cruz‐Esteban et al. 2020). As seen in previous studies, higher light intensity of a green LED trap increased the number of trapped whiteflies (Stukenberg et al. 2015) and the attraction of T. vaporariorum to a YST decreased with darker or less saturated shades of yellow (Vaishampayan et al. 1975). These characteristics render the green LED trap ideal for mass trapping in the greenhouse, where the ambient light conditions vary. No significant difference was observed between the effect of blue and blue + UV on the whitefly disturbance, even though the combined intensity of blue + UV was slightly lower than blue alone (42 and 50 μmol/m2/s, respectively). This corroborates with the results of Poushand et al (2017), who showed that increasing UV light exposure from 0.5 to 12 min on whiteflies settled on bean leaves led to a higher mortality rate 24 h later. Moreover, they align with the findings of Athanasiadou and Meyhöfer (2023), who demonstrated that 93% more whiteflies took-off from tomato leaves illuminated with blue + UV for 2 h, compared to non-illuminated ones, proposing that the presence of these lights conceals the green color of leaves and indicates lack of food resources. Given the reliance of whiteflies on UV for take-off and migratory behavior (Coombe 1982; Mellor et al. 1997; Antignus et al. 2001; Mutwiwa et al. 2005; Kumar and Poehling 2006; Gulidov and Poehling 2013), it is assumed that the disturbing effect of this combination is attributed, in part, to the role of UV radiation on dispersal and, secondly, the repelling quality of blue light. It is important to note that the inhibitory effect of blue light on whiteflies was consistent whether it was emitted from the same side with green light (Stukenberg et al. 2015) or from different sides, as in our setup. This reveals that the deterrent property of blue light and, hence, the validity of the whitefly blue-green opponent mechanism, remains consistent regardless of the positioning and direction of the two light sources.

Regarding our experiment in the greenhouse, the results showed that blue light disturbed significantly more whiteflies compared to the control, regardless the arrangements of the blue LEDs. More specifically, the double vertical arrangement exhibited the highest deterrence, possibly due to LEDs covering a wider canopy surface of the plant and the overall high intensity of blue light resulting from six spotlights, in which whiteflies have shown susceptibility (Athanasiadou and Meyhöfer 2023). This is supported by the fact that the horizontal arrangement, which illuminated only the lower canopy, was the least effective and the vertical arrangement was the second most effective, covering the height of the plant however only from one side. This indicates that ensuring light coverage at various heights of the canopy is more efficient against whiteflies than covering only the lower part from several sides, as in the horizontal arrangement. Moreover, coverage of the upper plant parts with blue light should enhance whitefly disturbance, considering that whiteflies tend to migrate to younger plant tissue for egg laying (Lenteren and Noldus 1990; Lynch and Simmons 1993) and would therefore be exposed to it after emergence. Based on our results, the difference in the disturbing efficacy between the double vertical and vertical arrangement was more pronounced than that between the vertical and horizontal arrangement. This suggests that the intensity effect of blue light appears to be more influential on whitefly disturbance than the positional effect. Our findings indicate that the whitefly population decreased evenly across the various heights of the plant in both the vertical and double vertical arrangements, suggesting that whiteflies did not migrate toward lower canopy layers, which provided greater shadow from the natural light, or upper layers, which have fresher and higher quality leaves (Noldus et al. 1986). The only arrangement in which there was a notable difference in the number of disturbed whiteflies between the canopy heights was the horizontal one, where significantly more whiteflies were disturbed from the bottom compared to the two upper levels. This is consistent with our results from the chamber experiment and can be attributed to the fact that blue LEDs illuminated in higher intensity and from several sides the lower leaves, directly on their underside, where whiteflies frequently settle. Given that whiteflies are intensity dependent (Athanasiadou and Meyhöfer 2023), they took-off from the lower leaves in search of an alternative resource, such as higher canopy layers, which were more protected from blue light.

Green LED traps captured significantly more of the released whiteflies in the presence of blue light compared to the control, regardless of the spotlights’ arrangement. This validates the findings of our chamber experiment, which showed that blue light not only induced whitefly flight behavior but also increased their susceptibility to get caught on a green LED trap—a strong directional stimulus within their environment. Blue LEDs in double vertical arrangement resulted in the highest recapture rate of released whiteflies compared to the other treatments. Since a strong correlation was therefore observed between the number of disturbed and subsequently trapped whiteflies, it can be proposed that a high number of disturbed insects lead to a high number of trapped ones, underlying the contribution of disturbing blue LEDs on whitefly recapture, in the greenhouse. An important observation is that, in our experiments, the blue photoreceptor of the greenhouse whitefly exhibited sensitivity at 465 and 456 nm, expanding the proposed peak range between 480 and 490 nm by Stukenberg and Poehling (2019), and revealing that using narrow-band LEDs with slight variations in the blue region shows no influence on the behavioral response of whiteflies. Regarding the fate of whiteflies after disturbance in the greenhouse, more than half of them landed on a green LED trap, regardless of the treatment. This validates the green LED trap’s ability to attract whiteflies post-disturbance and suggests that whiteflies’ response to green light might be lower while sitting on a plant, where the signal to engage in foraging activities such as feeding and oviposition is strong, in comparison with behaviors during flight, such as host searching and dispersal. The green LED traps in our greenhouse were almost never exposed to direct sunlight, since greenhouse shading was activated at a threshold of 40 kilolux, imitating the shaded conditions in a commercial greenhouse. Therefore, it can be proposed that the green LED traps could be effectively implemented in commercial greenhouses for whitefly monitoring and mass trapping. An important observation is that, overall, less of the disturbed whiteflies were caught on the green LED trap in the greenhouse compared to the chamber experiment. This can be attributed to a number of factors: (1) the trap—plant distance in the greenhouse was larger than in the chamber, (2) the trap’s dimensions remained the same, and (3) the greenhouse experimental arena was larger and not restricted, allowing the whiteflies to fly or land elsewhere. Additionally, the greenhouse environment may have had more competing visual stimuli and hiding places, making it harder for the whiteflies to locate the trap. Furthermore, the increased air circulation in the greenhouse could have caused the whiteflies to disperse further, reducing their chances of encountering the trap.

Conclusion and outlook

Our findings revealed that blue and blue + UV LEDs efficiently deter whiteflies from their host and enhance the catching rate of both a green LED trap and a YST, proposing a new push–pull approach in IPM. Due to the absence of light fluctuations in controlled conditions, the results offer valuable insights into whitefly photoreceptor sensitivities. The quality of blue light to disturb nearly 50% of settled whiteflies in a greenhouse highlights its potential as a sustainable and effective whitefly biocontrol tool, particularly in the early stages of infestation. The superior efficacy of the green LED trap over the YST in capturing whiteflies demonstrates its ability for efficient monitoring, which could lead to mass trapping by incorporating blue light as a deterrent factor. Further research is needed to assess whether this push–pull approach can be effective with higher population densities in the greenhouse. Since blue light is crucial for plant photosynthesis and growth (McCree 1971; Hogewoning and Trouwborst 2010), its implementation in agricultural crops should not endanger the plants. Following research should focus on evaluating the impact of these lights on important natural enemies used for whitefly biocontrol, and therefore, harming them should be avoided.

Author contribution statement

MA conceived, designed, and conducted the experiments, analyzed data, provided supervision, wrote the draft manuscript, and did reviewing and editing. RS conducted the experiments and collected data. RM provided resources, supervision, project administration, acquired funding, conceived and designed the experiments, did data supervision, reviewing, and editing. All of the authors read and approved the manuscript.