Introduction
The tomato leafminer, Tuta absoluta (Meyrick) (Lepidoptera: Gelechiidae), is among the most destructive invasive pests of tomato worldwide. Since its emergence from its native South American range, it has rapidly spread across Europe, Africa, Asia, and the Middle East, causing severe economic losses in both open-field and protected tomato production systems (Desneux et al., Reference Desneux, Wajnberg, Wyckhuys, Burgio, Arpaia, Narváez-Vasquez, González-Cabrera, Ruescas, Tabone, Frandon, Pizzol, Poncet, Cabello and Urbaneja2010, Reference Desneux, Luna, Guillemaud and Urbaneja2011; Abbes et al., Reference Abbes, Harbi and Chermiti2012; Biondi et al., Reference Biondi, Guedes, Wan and Desneux2018; Mansour et al., Reference Mansour, Brévault, Chailleux, Cherif, Grissa-Lebdi, Haddi, Mohamed, Nofemela, Oke, Sylla, Tonnang, Zappalà, Kenis, Desneux and Biondi2018; Han et al., Reference Han, Bayram, Shaltiel-harpaz, Sohrab, Saji, Esenali, Jalilov, Ali, Shashank, Ismoilov, Lu, Wang, Zhang, Wan, Biondi and Desneux2019). Its exceptional invasive capacity poses a serious threat to regional food security and global tomato production (Mansour et al., Reference Mansour, Brévault, Chailleux, Cherif, Grissa-Lebdi, Haddi, Mohamed, Nofemela, Oke, Sylla, Tonnang, Zappalà, Kenis, Desneux and Biondi2018; Han et al., Reference Han, Bayram, Shaltiel-harpaz, Sohrab, Saji, Esenali, Jalilov, Ali, Shashank, Ismoilov, Lu, Wang, Zhang, Wan, Biondi and Desneux2019; Colmenárez et al., Reference Colmenárez, Vásquez, De Freitas Bueno, Cantor, Hidalgo, Corniani and Lagrava2022).
Larvae of T. absoluta feed cryptically within the leaf mesophyll, stems, and fruits, causing extensive leaf mining, fruit tunnelling, and marked reductions in photosynthetic capacity. This endophytic behaviour, by creating a protected microenvironment within plant tissues, not only limits exposure to chemical control agents but also reduces the effectiveness of some natural enemies. Under favourable conditions and in the absence of effective management, infestations may result in yield losses approaching 100% (Desneux et al., Reference Desneux, Luna, Guillemaud and Urbaneja2011; Urbaneja et al., Reference Urbaneja, Arnó, González-cabrera, Pinto and Peña2013; Biondi et al., Reference Biondi, Guedes, Wan and Desneux2018). This concealed feeding behaviour reduces the efficacy of contact insecticides and predisposes plants to secondary infections, thereby amplifying overall crop damage (Cherif et al., Reference Cherif, Attia-barhoumi, Mansour, Zappalà and Grissa-lebdi2019; Halder et al., Reference Halder, Kushwaha, Rai and Singh2019).
Protected cultivation systems, while enhancing yield stability and fruit quality, often create microclimatic conditions that favour rapid population growth and continuous generations of T. absoluta. This species thrives optimally at temperatures between 25°C and 30°C combined with relative humidity (RH) above 70%, conditions that are frequently met inside greenhouse structures (Özgökçe et al., Reference Özgökçe, Bayındır and Karaca2016; Cherif et al., Reference Cherif, Attia-barhoumi, Mansour, Zappalà and Grissa-lebdi2019). Elevated temperatures, reduced pressure from natural enemies, and year-round host availability can accelerate pest development and intensify infestation levels, making management particularly challenging under greenhouse conditions (Harbi et al., Reference Harbi, Abbes and Chermiti2012; Cherif et al., Reference Cherif, Attia-barhoumi, Mansour, Zappalà and Grissa-lebdi2019; Campos et al., Reference Campos, Amiens-desneux, Béarez, Soares, Ponti, Biondi, Harwood and Desneux2021; Jat et al., Reference Jat, Dangi and Reddy2022).
Management of T. absoluta has traditionally relied on intensive applications of synthetic insecticides. However, this approach has become increasingly unsustainable due to the rapid evolution of resistance to multiple insecticide classes, including pyrethroids, spinosyns, diamides, and organophosphates (Roditakis et al., Reference Roditakis, Vasakis, Garcia-vidal, Martínez-Aguirre, Rison, Haxaire-Lutun, Nauen, Tsagkarakou and Bielza2018; Grant et al., Reference Grant, Jacobson, Ilias, Berger, Vasakis, Bielza, Zimmer, Williamson, Ffrench-Constant, Vontas, Roditakis and Bass2019; Guedes et al., Reference Guedes, Roditakis, Campos, Haddi, Bielza, Siqueira, Tsagkarakou, Vontas and Nauen2019). In addition to reduced field efficacy, excessive insecticide use disrupts beneficial arthropod communities, increases production costs, and raises concerns regarding environmental contamination and human health (Desneux et al., Reference Desneux, Decourtye and Delpuech2007; Biondi et al., Reference Biondi, Zappalà, Desneux, Aparo, Siscaro, Rapisarda, Martin and Tropea Garzia2015; Colmenárez et al., Reference Colmenárez, Vásquez, De Freitas Bueno, Cantor, Hidalgo, Corniani and Lagrava2022), highlighting the urgent need for alternative, environmentally sound management strategies.
Integrated pest management (IPM) offers a sustainable framework for controlling T. absoluta by combining monitoring tools, cultural practices, biological control agents, and selective physical or behavioural interventions (Kogan and Bajwa, Reference Kogan and Bajwa1999; Mansour et al., Reference Mansour, Brévault, Chailleux, Cherif, Grissa-Lebdi, Haddi, Mohamed, Nofemela, Oke, Sylla, Tonnang, Zappalà, Kenis, Desneux and Biondi2018; Desneux et al., Reference Desneux, Han, Mansour, Arnó, Brévault, Campos, Chailleux, Guedes, Karimi, Konan, Lavoir, Luna, Perez-Hedo, Urbaneja, Verheggen, Zappalà, Abbes, Ali, Bayram, Cantor, Cuthbertson, De, Erler, Firake, Haddi, Hajjar, Ismoilov, Jaworski, Kenis, Liu, Madadi, Martin, Mazih, Messelink, Mohamed, Nofemela, Oke, Ramos, Ricupero, Roditakis, Shashank, Wan, Wang, Wang, Zhang and Biondi2022). Within IPM programmes, trap-based approaches play a pivotal role by enabling early pest detection, monitoring population dynamics, and supporting timely, evidence-based decision-making (Benvenga et al., Reference Benvenga, Fernandes and Gravena2007; Caparros Megido et al., Reference Caparros Megido, Haubruge and Verheggen2013).
Pheromone traps are widely used for monitoring T. absoluta populations due to their high specificity and sensitivity, enabling accurate detection of population build-up and seasonal flight peaks (Ferrara et al., Reference Ferrara, Vilela, Jham, Eiras, Picanço, Attygalle, Svatos, Frighetto and Meinwald2001; Chermiti et al., Reference Chermiti, Abbes, Aoun, Othmane, Ouhibi, Gamoon and Kacem2009; Aksoy and Kovanci, Reference Aksoy and Kovanci2016). The female-emitted sex pheromone, consisting primarily of (3E,8Z,11Z)-tetradecatrien-1-yl acetate with a minor component of (3E,8Z)-tetradecadien-1-yl acetate, has been effectively synthesised and deployed as a cornerstone of IPM for monitoring and mass trapping (Griepink et al., Reference Griepink, van Beek, Posthumus, de Groot, Visser and Voerman1996; Svatoš et al., Reference Svatoš, Attygalle, Jham, Frighetto, Vilela, Saman and Meinwald1996; Puigmartí et al., Reference Puigmartí, Bosch and Guerrero2015). Synthetic pheromone traps facilitate early pest detection, inform treatment thresholds, and can reduce mating success by depleting local male populations (Roda et al., Reference Roda, Brambila, Barria, Euceda and Korytkowski2015; Cherif et al., Reference Cherif, Harbaoui, Zappalà and Grissa-lebdi2018; Mansour et al., Reference Mansour, Cherif, Attia-barhoumi, Zappalà and Grissa-Lebdi2019). However, because pheromone traps capture only males, their capacity to suppress overall population growth and particularly egg-laying by females is inherently limited when deployed in isolation (Chermiti and Abbes, Reference Chermiti and Abbes2012; Lobos et al., Reference Lobos, Occhionero, Werenitzky, Fernandez, Gonzalez, Rodriguez, Calvo, Lopez and Oehlschlager2013).
In contrast, light traps and coloured sticky traps rely on visual cues and can attract both male and female moths, complementing pheromone-based monitoring within IPM programmes (Mangrio et al., Reference Mangrio, Gilal, Rajput, Hajano and Gabol2023; Zhang et al., Reference Zhang, Zhang, Zhao, Wang, Huang, Lü, Li, Liu, Xian, Zhao, Li, Wan, Liu and Wang2023; Sadique et al., Reference Sadique, Ishtiaq, Qayyum, Alkherb, Abbasi, Arshad, Ullah, Faried, Akram and Rebouh2025). Building on these findings, recent studies have shown that specific visual cues, particularly light wavelength and trap colour, substantially influence trap performance, with yellow or golden light sources often outperforming blue or darker spectra (Lobos et al., Reference Lobos, Occhionero, Werenitzky, Fernandez, Gonzalez, Rodriguez, Calvo, Lopez and Oehlschlager2013; Braham, Reference Braham2014; Shiberu and Getu, Reference Shiberu and Getu2017; Mangrio et al., Reference Mangrio, Gilal, Rajput, Hajano and Gabol2023; Sadique et al., Reference Sadique, Ishtiaq, Qayyum, Alkherb, Abbasi, Arshad, Ullah, Faried, Akram and Rebouh2025). Nevertheless, the extent to which improved trap attractiveness translates into reduced infestation levels on different tomato plant organs remains insufficiently explored, particularly under protected cultivation conditions in the Middle East.
Trap performance and pest population dynamics are further influenced by environmental factors. Temperature, especially nocturnal minimum temperature, plays a critical role in adult flight activity, mating behaviour, and population development, whereas RH and other microclimatic variables often exert secondary effects (Özgökçe et al., Reference Özgökçe, Bayındır and Karaca2016; Cherif et al., Reference Cherif, Attia-barhoumi, Mansour, Zappalà and Grissa-lebdi2019; Campos et al., Reference Campos, Amiens-desneux, Béarez, Soares, Ponti, Biondi, Harwood and Desneux2021). Incorporating these factors into monitoring programmes is therefore essential for improving the reliability of trap-based forecasting systems.
Therefore, although T. absoluta has been present in the Eastern Mediterranean region since 2009 (Seplyarsky et al., Reference Seplyarsky, Weiss and Haberman2010), information on its non-chemical monitoring and management under greenhouse conditions in Palestine remains scarce. Accordingly, this study was undertaken to evaluate the effectiveness of pheromone traps, golden and blue light traps, and black and yellow (AMARILLO) sticky traps commonly adopted by local growers under greenhouse conditions for monitoring and managing T. absoluta in protected tomato production systems.
Methodology
Experimental location and greenhouse characteristics
The study was conducted under protected cultivation conditions at five commercial greenhouse locations in Tulkarm City, West Bank, Palestine. At each location, five separate greenhouses (one per trapping treatment) were used as experimental units. Greenhouses within each location were of similar construction (polythene-covered steel-frame structures, approximately 500 m2 each) and were managed identically with respect to irrigation, fertilisation, and agronomic practices throughout the growing seasons. A minimum distance of 50 m was maintained between adjacent greenhouses within each location to prevent cross-attraction or interference between pheromone and light treatments. The region is characterised by a relatively coastal Mediterranean climate and represents one of the major greenhouse tomato production areas in Palestine. The experiment was carried out over two consecutive growing seasons (2024 and 2025). In both seasons, tomato cultivation extended from February to August, which corresponds to the common greenhouse tomato production period in the region.
Cultivation of tomato
Tomato (Solanum lycopersicum L.) variety Julia was grown under protected-crop conditions. Planting density was 2.4 plants m−2, with 40 cm spacing between plants within the row. Transplanting was carried out on raised beds, with 50 cm between rows on the same bed and 120 cm between adjacent beds. All agronomic practices were applied uniformly according to standard greenhouse tomato production guidelines in accordance with recommendations of the Palestinian Ministry of Agriculture for protected-crop tomato production.
Trapping treatments
Pheromone traps
Delta-type plastic pheromone traps (white colour), complete with a sticky insert and hanger (Alpha Scents, USA), were used in the study. The traps were fitted with removable sticky sheets coated with a non-drying adhesive at the base. Each trap was supplied with a rubber septum dispenser containing 0.8 mg of synthetic sex pheromone for T. absoluta (Bioyome Ltd). Traps were positioned according to the growth of the tomato crop, i.e. set up to 60 cm above the average plant height (Ferrara et al., Reference Ferrara, Vilela, Jham, Eiras, Picanço, Attygalle, Svatos, Frighetto and Meinwald2001), and pheromone lures were replaced every 40 days to maintain trapping efficiency.
Light traps
The traps were fitted with LED bulbs emitting blue (360–400 nm) and golden (570–580 nm) wavelengths, each with a rated power of 5 W. Each light source was installed above an insect-collecting water tray placed directly on the ground, at a height of approximately 20 cm above the ground level. The water tray contained a dilute detergent solution (2–3 drops of liquid detergent per litre of water) to reduce surface tension and prevent insect escape; the solution was replenished weekly or whenever evaporation reduced the volume by more than 50%. The collection tray (34 cm in diameter and 4.7 cm in height) contained water with 0.2% detergent to reduce surface tension and prevent moth escape. The water surface was maintained approximately 1 cm below the tray opening, with a distance of about 15 cm between the light source and the water surface to ensure effective attraction and capture of adult moths.
Sticky traps
Black sticky traps consisted of polyethylene adhesive tapes (30 cm width) (Flora IPM, Turkey), while yellow sticky traps (AMARILLO) were prepared by applying AMARILLO adhesive (Luxembourg Industries Ltd) on shrink-wrap sheets. Traps were installed between adjacent greenhouse columns 3.3 m per span, positioned parallel to tomato rows at heights of 50–100 cm above ground level. Sticky surfaces were replaced weekly.
Weather parameters measurement
Digital thermo-hygrometers (ThermoPro TP50 or equivalent model, ASIN B01H1R0K68) were deployed to record daily minimum temperature (Tmin), maximum temperature (Tmax), and RH under greenhouse conditions. These sensors feature a temperature measurement range of −50°C to 70°C (−58°F to 158°F) with ±1°C accuracy and humidity range of 10–99% RH with ±2–3% accuracy, updating readings every 10 seconds. Devices were positioned 1.5 m above ground level in a shaded, representative central location within the greenhouse, with data logged manually every 24 hours at 8:00 AM; maximum/minimum values were reset daily to capture diurnal fluctuations influencing T. absoluta activity. Battery-powered units with LCDs and tabletop/magnetic mounting ensured continuous operation throughout the 2024–2025 cropping season, enabling correlation analysis between microclimatic variables and trap captures.
Experimental design and sampling
At each location, five trapping treatments were evaluated: pheromone traps, golden light traps, yellow sticky traps (AMARILLO), black sticky traps, and blue light traps. Traps were installed at the time of seedling transplantation and remained in place throughout the experimental period. The experiment was conducted over two consecutive growing seasons (2024 and 2025), providing replicated temporal data across similar agroclimatic conditions. Monitoring was conducted for 28 consecutive weeks in each season, and data were recorded on a weekly basis.
Each experimental greenhouse received one trap type only (two traps per greenhouse of 500 m2), ensuring that a single trap type was present within each experimental unit to avoid interference or mutual attraction effects among treatments. At this density, traps served the dual purpose of population monitoring and mass trapping for population suppression. A minimum distance of 10 m was maintained between traps to ensure independence of observations. No chemical insecticides, biological control agents, or additional trapping devices were used during the entire experimental period, allowing the exclusive evaluation of the trapping treatments.
The experiment was arranged according to a randomised complete block design (RCBD), with each location considered as a block and the five trapping treatments randomly assigned within each block. This design accounted for spatial variability among sites while enabling valid statistical comparisons among treatments.
Adult T. absoluta populations were monitored weekly by counting the number of moths captured in each trap. Captured adults were sexed under a hand lens (10×) based on external morphological characteristics following Halder et al. (Reference Halder, Kushwaha, Rai and Singh2019): males were identified by the presence of a pair of claspers (valvae) on the terminal abdominal segments, while females lacked these structures. Weekly mean captures of males, females, and total moths (males + females) were calculated for each trapping treatment.
In parallel, infestation levels were assessed weekly by randomly selecting 10 tomato plants from each replicated experimental unit. For each plant, all leaves, stems, and fruits present at the time of sampling were examined in their entirety. Plant parts were classified as healthy or infested based on characteristic symptoms of T. absoluta infestation, including leaf mines with dark frass accumulation within galleries and visible entry holes or tunnelling damage in stems and fruits. Only active mines and galleries containing live larvae were recorded; old or empty galleries were excluded from the assessment.
The mean percentage infestation per plant organ was calculated for each of the ten sampled plants and then averaged across plants to yield a plot-level mean. The percentage infestation of T. absoluta was calculated for each plant organ using the following equation:
\begin{equation*}{\text{Infestation}}\left( \% \right) = \left( {\frac{{{\text{Number of infested plant parts}}}}{{{\text{Total}}\,{\text{number}}\,{\text{of}}{\text{examined}}\,{\text{plant}}\,{\text{parts}}}}} \right) \times 100\end{equation*}Data collection continued weekly from transplanting until final harvest in each growing season.
Statistical analysis
All data were subjected to statistical analysis using IBM SPSS Statistics software (version 26.0; IBM Corp., Armonk, NY, USA). Two analytical approaches were applied depending on the response variable and figure. For the weekly capture and infestation data (figs 1–3 and 5–7), a repeated-measures ANOVA was conducted under an RCBD structure, with trapping treatment as the between-subject factor, week as the within-subject (repeated) factor, and greenhouse site as the blocking factor. The treatment effect was tested against the treatment × site interaction as the error term, yielding degrees of freedom of F(4, 16) for five-treatment comparisons and F(3, 12) for female captures (pheromone traps excluded). For the overall seasonal mean data (figs 4 and 8), a classical RCBD ANOVA was performed on seasonal means, with df = F(4, 16). Prior to analysis, the normality of residuals and homogeneity of variances were formally verified using the Shapiro–Wilk test and Levene’s test, respectively. Both tests confirmed that ANOVA assumptions were met for all variables; no data transformation was therefore required. When significant treatment effects were detected, mean separation was performed using the least significant difference (LSD) test at P ≤ 0.05.
Weekly mean capture of Tuta absoluta males in five trapping treatments (pheromone, golden light, blue light, black sticky, and AMARILLO) under greenhouse conditions. Values are means ± SE across sites (LSD₀.₀₅ = 3.4290, P < 0.05).
Figure 1 Long description
The horizontal axis label is, Observation weeks. The horizontal axis range is 0 to 28. The vertical axis label is, Mean weekly capture. The vertical axis range is 0 to 70. Legend labels: Pheromone, Golden, AMARILLO, Black sticky, Blue. Pheromone series coordinate pairs: (2, 2), (4, 4), (6, 6), (8, 7), (10, 13), (12, 14), (14, 28), (16, 35), (18, 40), (20, 52), (22, 50), (24, 52), (26, 45), (28, 50). Golden series coordinate pairs: (10, 5), (12, 8), (14, 12), (16, 14), (18, 20), (20, 22), (22, 24), (24, 26), (26, 21), (28, 22). AMARILLO series coordinate pairs: (10, 4), (12, 6), (14, 10), (16, 12), (18, 16), (20, 18), (22, 20), (24, 18), (26, 15), (28, 14). Black sticky series coordinate pairs: (10, 3), (12, 5), (14, 8), (16, 10), (18, 12), (20, 14), (22, 16), (24, 15), (26, 13), (28, 12). Blue series coordinate pairs: (10, 2), (12, 4), (14, 6), (16, 8), (18, 10), (20, 12), (22, 14), (24, 18), (26, 12), (28, 12).
Pearson correlation analysis was conducted to evaluate the relationships between weekly mean trap captures and mean weekly infestation levels on tomato leaves, stems, and fruits. Correlation coefficients (r) and associated probability values were calculated to assess the strength and significance of these relationships. Additional correlation analyses were performed to examine the association between selected environmental variables and weekly mean trap captures.
Results
Trap performance: adult moth captures
The weekly mean capture of male T. absoluta in pheromone, light, and sticky traps across both growing seasons (2024 and 2025) is presented in fig. 1. A highly significant difference was observed among trapping treatments in the number of male moths captured throughout the experimental period. A repeated-measures RCBD ANOVA revealed a highly significant treatment effect (F(4, 16) = 51.95, P < 0.001). Overall, the highest numbers of T. absoluta males were consistently captured in pheromone traps throughout the study period, followed by golden light traps, whereas blue, black sticky, and AMARILLO traps were comparatively less effective.
During the first week after transplanting, mean male captures were 0.60 ± 0.19 males/trap in pheromone traps and 0.10 ± 0.10 males/trap in blue light traps, while no males were recorded in golden, black sticky, or AMARILLO traps. Male captures increased progressively and reached a peak during the full vegetative and fruiting growth stages of tomatoes, i.e. the 23rd week after transplanting, when the highest weekly capture was recorded in pheromone traps (52.60 ± 2.03 males/trap). In comparison, the maximum weekly male captures recorded in golden, AMARILLO, black sticky, and blue traps were 26.60 ± 0.62, 24.40 ± 2.98, 18.70 ± 1.06, and 18.10 ± 1.93 males/trap, respectively (fig. 1).
Various light and sticky traps used in the study attracted female T. absoluta, whereas pheromone traps recorded zero female captures by design (male-specific attraction) and were therefore excluded from the female-treatment comparison. A highly significant difference was observed among trapping treatments in the number of females captured throughout the experimental period (F(3, 12) = 30.78, P < 0.001). Golden light traps consistently recorded the highest female captures on a week-by-week basis, whereas blue, AMARILLO, and black sticky traps showed comparatively lower levels of attraction, with clear temporal variation among trap types across the observation weeks (fig. 2).
Weekly mean capture of Tuta absoluta females in golden light, blue light, black sticky, and AMARILLO traps under greenhouse conditions. Pheromone traps captured no females by design. Values are means ± SE across sites (LSD₀.₀₅ = 2.1810, P < 0.05).
Figure 2 Long description
Legend labels: Pheromone, Golden, AMARILLO, Black sticky, Blue. Horizontal axis label: Observation weeks. Tick labels shown: 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28. Range: 2 to 28. Vertical axis label: Mean weekly capture. Tick labels shown: 0, 5, 10, 15, 20, 25, 30. Range: 0 to 30. Line identification by legend markers: each series is drawn as a line with small point markers and vertical error bars at each week. Pheromone series: a flat line on the baseline at mean weekly capture 0 across the full observation weeks range. Golden series: rises from near 0 in early weeks to above 10 by the mid period, then reaches its highest level in the early to mid 20s weeks. The highest visible point is slightly above 20 and below 25 around week 23. After that peak, the line drops to just below 20 around week 24 and then stays near 19 to 20 through week 28. AMARILLO series: increases from near 0 in early weeks to around 10 by the mid period, then climbs to its highest level in the early to mid 20s weeks. The highest visible point is below 20 and above 15 around week 23. After the peak, the line declines to near 10 to 12 by the late 20s weeks. Black sticky series: increases from near 0 in early weeks to around 5 to 10 by the mid period, then reaches a local high around the early 20s weeks near 12 to 13. After that, the line trends down slightly and ends near 11 to 12 by week 28. Blue series: increases from near 0 in early weeks to around 5 to 10 by the mid period, then reaches a local high around the early 20s weeks near 12 to 13. After that, the line drops more sharply than Black sticky around weeks 24 to 26 and ends below 10 by week 28. Purpose statement based on visible relationships: the Golden series stays above the other nonzero series for most weeks after the mid period, while Pheromone remains at 0 throughout.
During the early weeks after transplanting, females were detected in all non-pheromone traps, and captures increased progressively as the season advanced. Female captures reached a peak during the vegetative and early fruiting growth stages of tomatoes, i.e. around the 23rd week after transplanting, when the highest mean weekly capture was recorded in golden light traps (22.50 ± 1.05 females/trap). In comparison, the maximum weekly female captures recorded in AMARILLO, black sticky, and blue traps were 17.80 ± 2.10, 13.50 ± 0.85, and 12.40 ± 1.12 females/trap, respectively. Thereafter, a gradual decline in female captures was observed towards the later stages of crop development (fig. 2).
A highly significant difference was observed among trapping treatments in the weekly mean capture of total T. absoluta moths (males + females) throughout the study period (F(4, 16) = 46.77, P < 0.001; fig. 3). Owing to the male-specific nature of pheromone traps, significantly higher numbers of moths per week were consistently captured in pheromone traps, followed by golden light traps, which attracted both male and female moths.
Weekly mean capture of total Tuta absoluta adults (males + females) in pheromone, light, and sticky traps under greenhouse conditions. Values are means ± SE across sites (LSD₀.₀₅ = 3.0190, P < 0.05).
Figure 3 Long description
The graph compares mean weekly capture for five trap types over 28 observation weeks. The horizontal axis represents observation weeks, ranging from 2 to 28. The vertical axis represents mean weekly capture, ranging from 0 to 70. The lines are visually distinguished by color: Pheromone (red), Golden (yellow), AMARILLO (green), Black sticky (black) and Blue (blue). Pheromone shows the highest capture rates, peaking around week 22 at approximately 50 captures. Golden follows closely, peaking at week 24 with about 50 captures. AMARILLO peaks at week 24 with around 40 captures. Black sticky and Blue have lower capture rates, peaking at weeks 24 and 20 respectively, with captures around 30. The graph illustrates the differences in capture efficiency among the trap types, with Pheromone consistently leading, followed by Golden and AMARILLO, while Black sticky and Blue show similar lower trends.
The weekly overall maximum mean number of moths was recorded in pheromone traps (52.60 ± 2.03 moths/trap) during the 23rd week after transplanting, whereas golden light traps reached 49.00 ± 0.61 moths/trap in the same week. The maximum mean weekly captures recorded in the remaining traps AMARILLO, black sticky, and blue were 42.20 ± 5.40, 32.00 ± 1.26, and 30.40 ± 2.50 moths/trap, respectively (fig. 3).
Overall seasonal mean capture of T. absoluta adults differed highly significantly among trapping treatments for males (F(4, 16) = 172.05, P < 0.001), females (F(3, 12) = 29.14, P < 0.001), and total moths (F(4, 16) = 46.78, P < 0.001) (fig. 4).
Overall seasonal mean capture of Tuta absoluta males, females, and total adults in pheromone, light, and sticky traps under greenhouse conditions. Bars represent means ± SE across sites. Different letters indicate significant differences among treatments (LSD₀.₀₅ at P ≤ 0.05; Males = 1.4768; Females = 1.2035 (pheromone excluded); Total = 2.5292).
Overall, the highest mean capture of males was observed in pheromone traps (27.64 ± 0.47 males/trap), followed by golden (13.04 ± 0.41 males/trap) and AMARILLO (10.54 ± 1.16 males/trap) traps, which did not differ significantly from each other, while black sticky (9.39 ± 0.99 males/trap) and blue (8.00 ± 0.98 males/trap) traps recorded significantly fewer males and were not significantly different from each other.
Furthermore, golden traps captured the highest mean number of females (11.27 ± 0.28 females/trap), followed by AMARILLO (7.85 ± 0.92 females/trap), black sticky (6.96 ± 0.66 females/trap), and blue (5.74 ± 0.64 females/trap) traps, while pheromone traps recorded zero female captures by design.
Overall, the highest total mean capture of T. absoluta moths was recorded in pheromone traps (27.64 ± 0.47 moths/trap), followed by golden (24.31 ± 0.61 moths/trap), AMARILLO (18.38 ± 2.07 moths/trap), black sticky (16.35 ± 1.63 moths/trap), and blue (13.74 ± 1.62 moths/trap) traps.
Effects on plant infestation levels
A highly significant effect of trapping treatment on the mean percentage infestation of T. absoluta on tomato leaves was detected (F(4, 16) = 65.34, P < 0.001; fig. 5). Leaf infestation was negligible during the early observation weeks and increased progressively as the season advanced. Overall, tomato plants grown under pheromone and golden trap treatments generally exhibited lower weekly infestation levels than the other treatments. Peak infestation occurred around the 23rd week after transplanting, when the highest mean weekly infestation was recorded in the blue treatment (25.01 ± 2.44%), followed by black sticky (23.23 ± 2.72%) and AMARILLO (22.59 ± 2.92%), whereas lower peak levels were observed under pheromone (19.56 ± 3.03%) and golden (17.05 ± 3.16%) treatments. Thereafter, leaf infestation declined towards the end of the cropping cycle (fig. 5).
Weekly mean percentage infestation of Tuta absoluta on tomato leaves under five trapping treatments during the cropping season. Values are means ± SE across sites (LSD₀.₀₅ = 1.1221, P < 0.05).
Figure 5 Long description
Mean percentage infestation. Legend: Pheromone, Golden, AMARILLO, Black sticky, Blue. Horizontal axis label: Observation weeks. Range: 2 to 28. Vertical axis label: Mean percentage infestation. Range: 0 to 30. Pheromone series coordinate pairs: (6, 2), (8, 3), (10, 5), (12, 7), (14, 9), (16, 12), (18, 14), (20, 15), (22, 16), (24, 18), (26, 12), (28, 8). Golden series coordinate pairs: (6, 2), (8, 3), (10, 5), (12, 7), (14, 10), (16, 13), (18, 14), (20, 17), (22, 16), (24, 17), (26, 13), (28, 7). AMARILLO series coordinate pairs: (6, 4), (8, 6), (10, 8), (12, 12), (14, 18), (16, 21), (18, 23), (20, 24), (22, 23), (24, 24), (26, 18), (28, 6). Black sticky series coordinate pairs: (6, 4), (8, 7), (10, 11), (12, 16), (14, 22), (16, 23), (18, 24), (20, 25), (22, 24), (24, 24), (26, 16), (28, 10). Blue series coordinate pairs: (6, 5), (8, 8), (10, 13), (12, 18), (14, 23), (16, 24), (18, 24), (20, 24), (22, 24), (24, 25), (26, 20), (28, 15). Error bars are shown at the plotted points for each series.
As compared with leaves, tomato stems suffered a relatively low level of T. absoluta infestation; however, a highly significant difference was observed among trapping treatments over the entire study duration (F(4, 16) = 26.88, P < 0.001). The maximum weekly stem infestation was recorded in the blue trap treatment during the 14th week after transplanting (3.79 ± 0.16%). The highest mean weekly infestation observed in AMARILLO, black sticky, and golden trap treatments reached 3.65 ± 0.16% (15th week), 3.56 ± 0.30% (15th week), and 3.19 ± 0.23% (15th week), respectively, whereas the maximum infestation recorded under the pheromone treatment was 2.28 ± 0.22% (14th week) (fig. 6). Thereafter, stem infestation showed a progressive decline towards the later stages of the cropping cycle, reaching negligible levels during the final observation weeks (fig. 6).
Weekly mean percentage infestation of Tuta absoluta on tomato stems under five trapping treatments during the cropping season. Values are means ± SE across sites (LSD₀.₀₅ = 0.1487, P < 0.05).
Figure 6 Long description
Mean percentage infestation. Legend: Pheromone, Golden, AMARILLO, Black sticky, Blue. Horizontal axis label: Observation weeks. Range: 2 to 28. Vertical axis label: Mean percentage infestation. Range: 0 to 5. Pheromone series coordinate pairs: (4, 0), (6, 0.5), (8, 1.0), (10, 0.9), (12, 1.6), (14, 2.3), (16, 2.0), (18, 1.6), (20, 1.3), (22, 0.5), (24, 0), (26, 0), (28, 0). Golden series coordinate pairs: (4, 0), (6, 0.6), (8, 0.9), (10, 1.4), (12, 1.9), (14, 2.4), (16, 3.2), (18, 2.0), (20, 1.2), (22, 0.3), (24, 0), (26, 0), (28, 0). AMARILLO series coordinate pairs: (4, 0), (6, 0.4), (8, 1.2), (10, 2.0), (12, 2.4), (14, 3.0), (16, 3.6), (18, 2.8), (20, 1.6), (22, 0.4), (24, 0), (26, 0), (28, 0). Black sticky series coordinate pairs: (4, 0), (6, 0.3), (8, 1.0), (10, 2.1), (12, 2.6), (14, 3.1), (16, 3.5), (18, 3.2), (20, 1.8), (22, 0.6), (24, 0), (26, 0), (28, 0). Blue series coordinate pairs: (4, 0), (6, 0.8), (8, 1.8), (10, 2.3), (12, 3.2), (14, 3.9), (16, 3.6), (18, 3.4), (20, 2.0), (22, 0.7), (24, 0), (26, 0), (28, 0).
Similar to leaves and stems, a highly significant difference was recorded in the mean weekly infestation of T. absoluta on tomato fruits over the entire study period (F(4, 16) = 35.68, P < 0.001), with infestation becoming evident following fruit set (fig. 7). Although weekly mean fruit infestation exhibited pronounced fluctuations among trapping treatments, tomato plants grown under blue and sticky trap treatments generally experienced higher levels of damage. The highest mean weekly percentage infestation was recorded in the blue trap treatment (10.26 ± 0.69%) during the 23rd week after transplanting, whereas the maximum mean weekly infestation in black sticky traps reached 9.80 ± 0.59% during the 21st week. The maximum weekly infestation recorded in AMARILLO, pheromone, and golden trap treatments was 9.00 ± 0.53%, 7.40 ± 0.68%, and 6.97 ± 0.68%, respectively (fig. 7).
Weekly mean percentage infestation of Tuta absoluta on tomato fruits under five trapping treatments during the cropping season. Values are means ± SE across sites. Sites (LSD₀.₀₅ = 0.3921, P < 0.05).
Figure 7 Long description
The horizontal axis label is, Observation weeks. The horizontal axis range is 2 to 28. The vertical axis label is, Mean percentage infestation. The vertical axis range is 0 to 12. Legend labels: Pheromone, Golden, AMARILLO, Black sticky, Blue. Pheromone series coordinate pairs: (8, 0.5), (10, 1.0), (12, 1.5), (14, 2.5), (16, 3.5), (18, 4.0), (20, 4.0), (22, 5.5), (24, 4.5), (26, 3.0), (28, 1.5). Golden series coordinate pairs: (8, 0.5), (10, 1.5), (12, 2.0), (14, 2.5), (16, 3.5), (18, 4.5), (20, 5.0), (22, 6.0), (24, 5.0), (26, 4.0), (28, 2.0). AMARILLO series coordinate pairs: (8, 0.5), (10, 2.0), (12, 3.0), (14, 4.0), (16, 5.0), (18, 6.0), (20, 7.5), (22, 8.0), (24, 9.0), (26, 5.0), (28, 2.5). Black sticky series coordinate pairs: (8, 0.5), (10, 2.0), (12, 3.0), (14, 4.5), (16, 5.5), (18, 7.0), (20, 8.0), (22, 9.5), (24, 8.0), (26, 5.5), (28, 3.0). Blue series coordinate pairs: (8, 0.5), (10, 2.5), (12, 3.5), (14, 4.5), (16, 6.0), (18, 8.0), (20, 9.0), (22, 9.5), (24, 10.5), (26, 5.0), (28, 2.5).
Overall seasonal mean percentage infestation of T. absoluta on tomato leaves, stems, and fruits is presented in fig. 8. A significant treatment effect was detected for overall infestation on tomato leaves (F(4, 16) = 4.13, P = 0.0074), stems (F(4, 16) = 77.36, P < 0.001), and fruits (F(4, 16) = 126.47, P < 0.001). In general, tomato leaves exhibited the highest overall infestation, followed by fruits, whereas stems showed the lowest infestation levels. The highest overall mean infestation was recorded in the blue treatment on leaves (15.96 ± 1.75%), stems (1.44 ± 0.05%), and fruits (4.44 ± 0.19%), with similarly high levels observed in black sticky traps (14.67 ± 1.67%, 1.38 ± 0.05%, and 4.09 ± 0.19%, respectively). Intermediate infestation levels were observed under AMARILLO on leaves (12.52 ± 1.65%), stems (1.27 ± 0.05%), and fruits (3.60 ± 0.17%). In contrast, the lowest overall infestation levels were recorded under pheromone and golden treatments across all plant parts (leaves: 8.98 ± 1.28% and 8.85 ± 1.45%; stems: 0.79 ± 0.04% and 0.98 ± 0.05%; fruits: 2.47 ± 0.13% and 2.81 ± 0.18%, respectively) (fig. 8).
Overall seasonal mean percentage infestation of Tuta absoluta on tomato leaves, stems, and fruits under different trapping treatments. Bars represent means ± SE. Different letters indicate significant differences among treatments within each plant organ (LSD₀.₀₅ at P ≤ 0.05; Leaves = 4.5668; Stems = 0.0897; Fruits = 0.2131).
The relationship between mean weekly trap capture of T. absoluta and its infestation on tomato leaves, stems, and fruits is presented in table 1. Mean weekly trap capture exhibited a highly significant, moderate-to-strong positive correlation with infestation levels on tomato leaves and fruits across all trapping treatments. Correlation coefficients ranged from r = 0.81 to 0.93 for tomato leaves and from r = 0.89 to 0.95 for tomato fruits (P < 0.001), indicating a consistent association between adult moth capture and infestation intensity on these plant parts throughout the observation period. Among the tested treatments, relatively stronger correlations were observed on tomato leaves in the Golden trap (r = 0.9281, P < 0.001) and on tomato fruits in the AMARILLO trap (r = 0.9455, P < 0.001). In contrast, correlations between mean weekly trap capture and stem infestation were weak, negative, and not statistically significant across all trapping treatments (r = −0.09 to −0.02; P > 0.05), reflecting the generally low and variable infestation levels recorded on tomato stems (table 1).
Pearson correlation coefficients (r) between weekly mean trap capture of Tuta absoluta adults and percentage infestation on tomato leaves, stems, and fruits under greenhouse conditions
Table 1 Long description
Pearson correlation coefficients and P values are listed for weekly mean trap capture versus percent infestation on tomato leaves, stems, and fruits, for five trap types. For pheromone traps, leaves have r 0.8933 with P less than 0.001, stems have r minus 0.0885 with P 0.6543, and fruits have r 0.8906 with P less than 0.001. For golden traps, leaves have r 0.9281 with P less than 0.001, stems have r minus 0.0179 with P 0.9280, and fruits have r 0.9245 with P less than 0.001. For amarillo traps, leaves have r 0.8556 with P less than 0.001, stems have r minus 0.0393 with P 0.8427, and fruits have r 0.9455 with P less than 0.001. For black sticky traps, leaves have r 0.8143 with P less than 0.001, stems have r minus 0.0416 with P 0.8335, and fruits have r 0.9208 with P less than 0.001. For blue traps, leaves have r 0.8587 with P less than 0.001, stems have r minus 0.0218 with P 0.9124, and fruits have r 0.9222 with P less than 0.001.
Environmental correlates of trap performance
The relationship between selected environmental parameters and the performance of pheromone and coloured traps in attracting T. absoluta moths is presented in table 2. Among the examined weather variables, minimum temperature (Tmin) exhibited a strong and highly significant positive correlation with the capture of males, females, and total moths across all trap types (r = 0.849–0.935, P < 0.0001). In contrast, maximum temperature (Tmax) showed a weak to moderate but statistically significant relationship with moth capture (r = 0.363–0.395, P < 0.0001). RH was also weakly to moderately correlated with moth captures across trap types, with all correlations being statistically significant (P < 0.01). Overall, the results indicate that moth activity under greenhouse conditions was more closely associated with minimum temperature than with maximum temperature or RH (table 2).
Pearson correlation coefficients (r) between weekly mean trap capture of Tuta absoluta adults and selected environmental variables (minimum temperature, maximum temperature, and relative humidity) under greenhouse conditions
Table 2 Long description
Weekly mean trap capture of Tuta absoluta adults is summarized as Pearson correlation coefficients r with Tmin, Tmax, and RH for five trap types, with a P value listed under each row of r values. For Pheromone, r is 0.934 with Tmin, 0.363 with Tmax, and 0.400 with RH, and each P value is less than 0.0001. For Golden, r is 0.935 with Tmin, 0.391 with Tmax, and 0.420 with RH, and each P value is less than 0.0001. For Amarillo, r is 0.888 with Tmin, 0.391 with Tmax, and 0.330 with RH, and each P value is less than 0.0001. For Black sticky, r is 0.919 with Tmin, 0.395 with Tmax, and 0.328 with RH, and each P value is less than 0.0001. For Blue, r is 0.849 with Tmin, 0.382 with Tmax, and 0.243 with RH, with P less than 0.0001 for Tmin and Tmax and P 0.0038 for RH.
Discussion
The present study provides a comprehensive evaluation of the performance of pheromone, light, and sticky traps for monitoring and mass trapping of T. absoluta, and their relationship with infestation dynamics on tomato plants under protected cultivation. By integrating adult capture data, infestation levels on leaves, stems, and fruits, and key environmental parameters, the study offers robust evidence supporting trapping systems as core components of IPM programmes.
Pheromone traps consistently recorded the highest captures of male T. absoluta throughout the cropping cycle, with a clear seasonal peak reaching 52.60 ± 2.03 males per trap during the 23rd week after transplanting. This pattern confirms the high sensitivity of sex pheromone traps for detecting population build-up and monitoring male flight activity, as widely documented under both greenhouse and field conditions (Aksoy and Kovanci, Reference Aksoy and Kovanci2016; Bayram et al., Reference Bayram, Duman, Buyuk and Mutlu2017; Mangrio et al., Reference Mangrio, Gilal, Rajput, Hajano and Gabol2023). The male-biased response observed in the present study is biologically expected, given that pheromone-mediated mate-searching behaviour is the primary driver of male activity in T. absoluta (Witzgall et al., Reference Witzgall, Kirsch and Cork2010; Biondi et al., Reference Biondi, Guedes, Wan and Desneux2018).
In contrast, coloured light and sticky traps attracted both males and females, with pronounced differences among trap types. Golden light traps consistently outperformed other non-pheromone traps, recording peak captures of 22.50 ± 1.05 females and 26.60 ± 0.62 males per trap. Similar superiority of yellow–golden light spectra has been reported in previous studies, where such wavelengths enhanced attraction of T. absoluta, particularly females, compared with blue or darker colours (Braham, Reference Braham2014; Sridhar and Kumaran, Reference Sridhar and Kumaran2018; Ardeh et al., Reference Ardeh, Kheirkhah and Askari Seyahooei2021). The ability of golden traps to capture females is of particular importance from a management perspective, as female removal directly reduces oviposition pressure and subsequent larval infestation.
Total moth captures closely followed trends observed for male populations, with pheromone traps recording the highest overall means, followed by golden light traps. Although pheromone traps are male-specific by design, their high efficiency resulted in elevated total captures compared with other trapping systems, a pattern also reported by Bayram et al. (Reference Bayram, Duman, Buyuk and Mutlu2017) and Mangrio et al. (Reference Mangrio, Gilal, Rajput, Hajano and Gabol2023).
Infestation levels varied markedly among tomato plant organs. Leaves consistently exhibited the highest infestation, followed by fruits, whereas stems showed minimal and irregular damage throughout the cropping cycle. This distribution reflects the known feeding and oviposition preferences of T. absoluta, which primarily targets leaf mesophyll and fruit tissues, while stem infestation remains sporadic and of limited economic relevance (Urbaneja et al., Reference Urbaneja, Arnó, González-cabrera, Pinto and Peña2013; Biondi et al., Reference Biondi, Guedes, Wan and Desneux2018).
Plants grown under pheromone and golden trap treatments exhibited substantially lower infestation levels on leaves and fruits compared with those under blue and black sticky traps. Seasonal mean leaf infestation remained below 9% under pheromone and golden treatments, whereas it exceeded 15% under blue traps. Similar reductions in infestation associated with intensive trapping or mass-trapping strategies have been reported in greenhouse systems, highlighting the potential of trapping to contribute to population suppression when integrated with other control measures (Cocco et al., Reference Cocco, Deliperi and Delrio2012; Cherif et al., Reference Cherif, Harbaoui, Zappalà and Grissa-lebdi2018).
Correlation analysis revealed strong and highly significant positive relationships between mean weekly adult captures and infestation levels on tomato leaves (r = 0.81–0.93) and fruits (r = 0.89–0.95), confirming that adult trap captures are reliable predictors of subsequent crop damage. This close association supports previous findings that trap-based monitoring can be effectively used to forecast infestation risk and optimise the timing of management interventions against T. absoluta (Caparros Megido et al., Reference Caparros Megido, Haubruge and Verheggen2013; Biondi et al., Reference Biondi, Guedes, Wan and Desneux2018; Mangrio et al., Reference Mangrio, Gilal, Rajput, Hajano and Gabol2023). In contrast, correlations between adult captures and stem infestation were weak and non-significant, reflecting the low and variable nature of stem damage.
Environmental conditions significantly influenced adult moth activity under protected cultivation. Minimum temperature showed a strong positive association with captures of males, females, and total moths, whereas maximum temperature and RH exhibited weaker relationships. These results are consistent with the thermal biology of T. absoluta, which shows optimal development and activity within moderate nocturnal temperature ranges (Campos et al., Reference Campos, Amiens-desneux, Béarez, Soares, Ponti, Biondi, Harwood and Desneux2021; González-zamora et al., Reference González-zamora, Amiens-desneux, Béarez, Ponti, Biondi and Desneux2022). Similar patterns have been reported in greenhouse environments, where minimum temperature was identified as a key driver of adult flight and trap encounter rates (Mangrio et al., Reference Mangrio, Gilal, Rajput, Hajano and Gabol2023).
From an IPM perspective, the present findings support the complementary use of pheromone and golden light traps. Pheromone traps provide precise and sensitive monitoring of male populations and population thresholds, while golden light traps contribute to the removal of both sexes, thereby directly reducing reproductive potential. Such dual-trap strategies align with IPM frameworks emphasising reduced reliance on chemical insecticides and increased use of physical and biological control tools (Zappalà et al., Reference Zappalà, Biondi, Alma, Al-jboory, Arnò, Bayram, Chailleux, El-Arnaouty, Gerling, Guenaoui, Shaltiel-Harpaz, Siscaro, Stavrinides, Tavella, Vercher-Aznar, Urbaneja and Desneux2013; Mansour and Biondi, Reference Mansour and Biondi2021).
Importantly, early and continuous deployment of traps from the beginning of the cropping cycle substantially suppressed T. absoluta populations and reduced seasonal infestation pressure. It is noteworthy that infestation levels on plant organs declined during the final weeks of the season (weeks 25–28) despite adult trap captures remaining elevated. This apparent paradox likely reflects end-of-season plant senescence: as foliage aged and was progressively removed during routine cultural operations, fewer suitable sites were available for oviposition and larval development, even though adult populations remained high. Additionally, senescent tissue may show reduced symptomatic expression, making infestation harder to detect visually. These dynamics highlight that trap captures and visible infestation do not always track each other synchronously and that both metrics are needed for a complete picture of pest pressure. This reduction not only lowers reliance on chemical insecticides but also supports the conservation of natural enemies and environmental sustainability, which are critical objectives in modern tomato IPM systems (Cherif et al., Reference Cherif, Harbaoui, Zappalà and Grissa-lebdi2018; Chouikhi et al., Reference Chouikhi, Assadi, Lebdi and Belkadhi2022).
This study provides an integrated evaluation linking trap efficiency, sex-specific capture patterns, infestation dynamics across plant organs, and microclimatic drivers of T. absoluta under protected cultivation. When integrated with complementary cultural and biological control measures, trap-based strategies offer a practical and environmentally sound foundation for sustainable tomato production systems.
Conclusion
This study confirms that pheromone, light, and sticky traps differ significantly in their effectiveness for monitoring T. absoluta under protected tomato cultivation. Pheromone traps were the most efficient for attracting male moths, whereas golden light traps captured both males and females and outperformed other non-pheromone traps. Adult trap captures showed a strong positive relationship with infestation on tomato leaves and fruits, while the association with stem infestation was weak and negligible. Among environmental factors, minimum temperature exerted the strongest influence on moth activity, whereas maximum temperature and RH played minor roles. Therefore, the early and continuous use of pheromone and golden light traps can enhance monitoring accuracy, contribute to population suppression, and reduce reliance on chemical insecticides, supporting sustainable IPM strategies in protected tomato production.
Competing interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Open access
