Salix Alba L. Leaf Volatiles Attract Nematus Hequensis Xiao Adults: Sex-Specific Responses and Optimized Field Trapping Using (E)-2-Hexenal and O-Xylene

1. Introduction

Salix alba L. (white willow) is widely planted worldwide for riparian restoration, biomass production, and ornamental use, with a natural range spanning Europe, Asia, and North Africa, and introductions on all continents except Antarctica. However, its expansion has facilitated the spread of specialist herbivores. Among them, sawflies (Hymenoptera: Tenthredinidae) are particularly damaging due to gregarious defoliation during their larval stages, which can completely strip host trees and reduce growth, timber quality, and ecosystem services. One emerging pest is Nematus hequensis Xiao, first described from the Hequ region of China (Xiao, 1990). Over the past three decades, it has gradually expanded its range and now causes increasing damage in several Chinese provinces, including Liaoning, Inner Mongolia, Beijing, and most recently the Tibet Autonomous Region (Lei et al., 2021). Although currently reported only in China, the global distribution of Salix hosts—including S. alba, Salix matsudana, and hybrid poplars—raises concerns about its potential as an invasive pest, especially under climate change scenarios that may expand suitable habitat at higher latitudes and altitudes. This highlights the urgent need for effective, low-environmental-impact monitoring and control tools that can be deployed preemptively.

Belonging to the order Hymenoptera and family Tenthredinidae, N. hequensis is a phytophagous pest that feeds on the leaves of S. matsudana Koidz, S. alba Tristis, S. alba L., Populus × beijingensis W. Y. Hsu, and other plants (Du 2022). The insect was discovered and named in 1990 (Xiao 1990). Currently, N. hequensis is extensively found in Liaoning (Zhanggutai), Inner Mongolia (Hulunbeier), Beijing, and Xizang (Lhasa), among other places in China (Wang et al. 2015). The insect damages host plants by feeding on their leaves during the larval stage, and it exhibits characteristics that include rapid reproduction, high population density, significant food consumption, and causes substantial damage (Lei et al. 2021). Although currently reported only in China, the global distribution of S. alba hosts raises concerns about its potential as an invasive pest, highlighting the need for effective, low-environmental-impact monitoring tools. In September 2016, N. hequensis was first discovered in Lhasa, Xizang, and subsequently had significant economic, social, and environmental impacts in this region (Lei et al. 2019). Current control relies almost exclusively on chemical insecticides, mainly organophosphates and pyrethroids (Feng et al. 2012). Although no resistance has been reported to date, long-term pesticide use inevitably leads to environmental pollution, non-target effects (e.g., on pollinators and natural enemies), resistance development, and health risks to applicators and nearby residents (Zhu et al. 2024; Zhang et al. 2024). There is therefore a pressing need for alternative, sustainable management methods.

Plant-based attractants offer a greener alternative. The use of host plant volatiles to lure pests into traps—known as attract-and-kill or monitoring—has been successfully developed for many insect groups, including weevils, moths, and fruit flies (Li et al., 2019). The efficacy of such lures depends on multiple factors: volatile composition and concentration, lure type (release rate, longevity), trap design (color, shape, entrance size), trap density and placement height, as well as environmental conditions (temperature, wind, humidity). For sawflies, however, very few plant-volatile attractants have been developed, and none exist for N. hequensis. Previous studies in our laboratory identified 15 volatile compounds from S. alba leaves with relative contents ≥1% and demonstrated that several of these compounds elicited electroantennographic and behavioral responses in N. hequensis adults under laboratory conditions (Zhenhao et al., 2025). However, field validation and optimization of trapping parameters have not been conducted.

The specific objectives of this study were: (1) to screen the 15 S. alba leaf volatiles at four concentrations for field attractiveness to N. hequensis adults; (2) to determine the effects of trap type (large boat-shaped vs. triangular), lure core type (slow-release bottle vs. eraser vs. bag), hanging height (1.0 vs. 1.5 m), and trap spacing (10 vs. 15 m) on capture efficiency; (3) to identify the optimal combination for each sex; and (4) to provide a practical, environmentally friendly monitoring tool that can be used for early warning and potential mass trapping of this emerging sawfly pest.

2. Materials and Methods

2.1. Field-Trapping Experiment Locations and Times

The field-trapping experiment was set up at the Agricultural Research Institute of the Lhasa City Agricultural and Forestry Science Academy, Tibet Autonomous Region (China; 29°38′26″ N, 91°2′5″ E), at an elevation of 3650 m above sea level. The site covers an area of approximately 4 ha. To ensure accuracy, no insecticides or control measures were used in this area before or during the experiment. The experiment was conducted from August 25–September 28, 2024, during the peak outbreak period of adult N. hequensis (the adult emergence period may vary in different years, regions, or areas due to the region, altitude, temperature, humidity, and light). During the experimental period ( August 25–September 28, 2024), the mean temperature was 14.2 °C (range 8.5–21.0 °C), mean relative humidity 63%, and total rainfall 48 mm (data from Lhasa Meteorological Bureau, weather.cma.cn).

2.2. Preparation of Attractant Lures

This study builds on our previous research (Zhenhao et al., 2025). Fifteen volatile compounds (purity ≥97%, suppliers listed in Table 1) were each diluted in n-hexane (≥99.5%; Tianjin Kermel Chemical Reagent Co., Ltd., Tianjin, China) to four concentrations: 100, 10, 1, and 0.1 μg/μL. For each concentration, 10 μL of the diluted volatile was mixed with 10 μL of natural honey (from Lepidotrigona arciferal, provided by the Agricultural Research Institute, Lhasa) as a slow-release carrier. The mixture (20 μL total) was injected into a polyethylene bottle lure (slow-release bottle core, model YL-16ml, Zhongjie Sifang Co., Beijing) using a microsyringe. Blank control lures received 20 μL of n-hexane + honey mixture without any volatile compound. All lures were individually wrapped in aluminum foil, sealed with Parafilm, and stored at −20 °C for a maximum of 7 days. Lures were brought to room temperature 30 min before field deployment. Preliminary laboratory measurement at 25 °C indicated an average release rate of 0.42–0.68 μg/day for the tested compounds at 10 μg/μL (data not shown).

2.3. Information on Blank Attractant Cores and Traps

As shown in Table 2. Three types of traps were used: large boat-shaped, triangular, and bucket-shaped traps (Zhongjie Sifang Co., Ltd., Beijing, China; model numbers YL-LWT-1, YL-DT-2, and BT-L-P2.0, respectively). The attractant cores included eraser-type slow-release, slow-release bag, and slow-release bottle cores (Zhongjie Sifang Co., Ltd.; model numbers Y, YL, and YL-2ml, respectively).

2.4. Field Trapping Experimental Design

The experiment followed a randomized complete block design with three blocks (replicates), each block separated by at least 200 m to avoid interference. Within each block, traps were arranged in a grid with a spacing of either 10 m or 15 m (two density treatments). A 5 m buffer zone of untreated S. alba trees surrounded each block.

For each volatile compound, concentration, trap type, lure core type, and hanging height (1.0 m or 1.5 m) were tested. Blank control traps (containing no volatile) were included in all configurations to account for physical attraction.

Traps were rotated weekly within each block to minimize position effects. Sticky boards were replaced weekly, and water in bucket traps replenished. The experiment lasted four consecutive weeks ( August 25–September 28, 2024), with counts recorded daily during the first three days of each week, then averaged per week. Each treatment combination was therefore replicated three times (once per block) and over three weekly measurements.

2.5. Field Population Monitoring of N. Hequensis Adults

From July 22 to September 23 in 2023 and 2024, adult N. hequensis abundance was monitored in the same experimental site. Six plots (each 0.5 ha) were established. In each plot, five S. alba trees (10–15 years old, 5–8 m height) were randomly selected. At each tree, a sweep net (40 cm diameter) was used to capture adults within a 5 m radius around the trunk for 5 minutes. Captured adults were counted and sexed. Sampling was conducted weekly in the morning (China, 10:00–12:00) under sunny conditions, with five repetitions (five separate sampling rounds) per monitoring year.

2.6. Statistical Analysis

Data are presented as mean ± standard error (SE). One-way analysis of variance (ANOVA) was used to test the effects of volatile compound, concentration, trap type, lure type, hanging height, and spacing on the number of captured adults (separately for females and males). When significant differences were detected (P <0.05), post-hoc comparisons were performed using Duncan’s new multiple range test (Wang et al. 2023). All analyses were conducted using SPSS v.20.0 (IBM, Armonk, NY, USA).

3. Results

3.1. Effect of Attractant Concentration on Trapping Amount

Different concentrations of plant-derived attractants did not cause a concentration-dependent increase in trapping efficiency (Figure 1). However, the trapping quantity at a concentration of 10 μg/μL (119 individuals per day on average) was significantly higher than that at other concentrations.

The concentration of attractants significantly affected daily capture (ANOVA: F₃,₈= 12.7, P < 0.001). At 10 μg/μL, the mean capture was 119 ± 9.3 adults/day (n=3), which was significantly higher than at 100 μg/μL (32 ± 5.1), 1 μg/μL (28 ± 4.6), and 0.1 μg/μL (41 ± 6.2) (Duncan’s test, P <0.05). Notably, the highest concentration (100 μg/μL) did not yield the highest capture, indicating a non-linear dose-response relationship (Figure 1).

3.2. Effect of Trap Density on Trapping Amount

Trap spacing (10 m vs. 15 m) did not significantly affect daily capture for either females (F₁,₄= 2.31, P = 0.20) or males (F₁,₄= 1.89, P = 0.24), with average captures of 49 and 33 females per trap, and 175 and 142 males per trap, respectively (Figure 2). Given the lack of statistical difference and lower material cost, a spacing of 10 m (one trap per 40 m²) is recommended for practical application.

3.3. Effect of Trap Suspension Height on Trapping Amount

Hanging height had a strong effect on capture (F₁,₄= 34.5, P = 0.004). Traps suspended at 1 m captured 216 ± 18.2 adults/day, whereas those at 1.5 m captured only 14 adults/day (Figure 3). The manufacturer’s recommended height (1.5 m) was therefore ineffective for this species under local conditions.

3.4. Effects of Different Lures and Traps on Trapping Amount

3.4.1. Large Boat-Shaped Trap

As shown in Figure 4, at a concentration of 100 μg/μL, dimethyl sulfide attracted 48 male and 3 female adults. At a concentration of 10 μg/μL, (E)-2-Hexenal had the strongest attractive effect on male adults, attracting 265 individuals, while also attracting a single female adult. 2-Hydroxybenzaldehyde attracted 71 male and 1 female adults, and o-Xylene attracted 62 female and 2 male adults. At a concentration of 1 μg/μL, most volatiles exhibited weak attractive effects on both female and male adults. At a concentration of 0.1 μg/μL, butanoic acid, (Z)-3-hexenyl ester exerted the strongest attractive effect on male adults, attracting 78 individuals, while also attracting two female adults. Overall, at a concentration of 10 μg/μL, the number of attracted female and male adults was the highest, with 107 female and 483 male adults. This indicates that at higher concentrations, the large boat-shaped trap exhibited good trapping effects on both female and male adults.

1: dimethyl sulfide; 2: (E)-2-Hexenal; 3: ethylbenzene; 4: p-Xylene; 5: o-Xylene; 6: styrene; 7: benzaldehyde; 8: furan, 2-pentyl-; 9: 3-hexen-1-ol, acetate, (Z); 10: benzyl alcohol; 11: 2-hydroxybenzaldehyde; 12: 1,6-dioxaspiro[4.4]nonane, 2-ethyl; 13: butanoic acid, 3-hexenyl ester, (Z); 14: cis-3-hexenyl isovalerate; 15: pentadecanal. Note: An asterisk ‘*’ indicates a significant difference (p < 0.05, independent samples t-test), while double asterisks ‘**’ indicate a highly significant difference (p < 0.01, independent samples t-test).

3.4.2. Triangle Trap

Compared with the control group, in the experimental group: at a concentration of 100 μg/μL, all volatiles exhibited weak trapping effects on both female and male adults (Figure 5). At a concentration of 10 μg/μL, ethylbenzene had the strongest trapping effect on male adults, attracting 13 individuals, with no female adults. At a concentration of 1 μg/μL, all volatiles exerted weak trapping effects on both female and male adults. At a concentration of 0.1 μg/μL, benzaldehyde had the strongest attractive effect on male adults, attracting 30 individuals, while attracting no female adults. Overall, at a concentration of 0.1 μg/μL, a relatively high number of male adults were trapped (67 individuals), whereas the number of female adults was lower. This indicates that the triangular trap had a stronger trapping effect on male adults at this concentration, whereas the effect it had on female adults was relatively weaker.

1: dimethyl sulfide; 2: (E)-2-Hexenal; 3: ethylbenzene; 4: p-Xylene; 5: o-Xylene; 6: styrene; 7: benzaldehyde; 8: furan, 2-pentyl-; 9: 3-hexen-1-ol, acetate, (Z); 10: benzyl alcohol; 11: 2-hydroxybenzaldehyde; 12: 1,6-dioxaspiro[4.4]nonane, 2-ethyl; 13: butanoic acid, 3-hexenyl ester, (Z); 14: cis-3-hexenyl isovalerate; 15: pentadecanal. Note: An asterisk ‘*’ indicates a significant difference (p < 0.05, independent samples t-test), while double asterisks ‘**’ indicate a highly significant difference (p < 0.01, independent samples t-test).

3.5. Dynamic Monitoring of the N. Hequensis Adult Population in the Field

The survey results of the population dynamics of N. hequensis adults in the test area (Figure 6) showed that during the monitoring period in 2023, one major peak (August 12) and one smaller peak (August 19) were observed, which occurred on August 12th and August 19th, respectively. This may be related to the high temperatures in the region during this period (Website for Weather Inquiry in Lhasa City: weather.cma.cn). During the monitoring period in 2024, only one small peak period of emergence was observed on September 15th (compared with that in 2023). The survey found that during the N. hequensis emergence period, the region experienced continuous rainfall and low-temperature weather, which might have been responsible for the absence of a peak emergence period. Overall, the population density in 2024 was significantly lower than that in 2023, and duration of the emergence period was shorter. Despite the lower overall density in 2024, the relative attractiveness among different volatile treatments remained consistent (The pre-experimental results in 2023 are compared with the experimental results in 2024; data not shown), indicating that population size did not bias treatment comparisons within the same year.

4. Discussion

4.1. Non-Linear Concentration-Response and Ecological Implications

The non-linear dose-response (peak attraction at 10 μg/μL, decline at 100 μg/μL ) observed in this study aligns with the sensory adaptation and saturation hypothesis documented in many phytophagous insects (Li et al., 2019; Hu et al., 2018). For N. hequensis adult, the olfactory system may become saturated at concentrations above a certain threshold, leading to reduced antennal sensitivity or even avoidance behavior. High concentrations of a single volatile might also disrupt the natural blend ratio that sawflies use to recognize host plants. In nature, S. alba leaves emit a complex mixture of volatiles; the synthetic attractant we tested uses individual compounds in isolation. At 100 μg/μL, the compound likely overwhelms the peripheral olfactory receptors. This finding underscores the importance of concentration optimization in attractant development: simply increasing dosage does not linearly increase trap catch, and an intermediate concentration (10 μg/μL) is optimal. Interestingly, for some compounds (e.g., (Z)-3-Hexen-1-ol acetate), the lowest concentration (0.1 μg/μL) also showed moderate attraction (78 males), suggesting a possible bimodal response that may reflect different ecological roles at trace versus high doses. This has been reported in other insects where low doses act as long-distance cues and high doses as short-range arrestants.

4.2. Sex-Specific Attraction: Chemical Ecology of Host Finding

The strong attraction of N. hequensis female adults to o-Xylene (12.4% relative content in leaf volatiles) is a key discovery. o-Xylene is an aromatic hydrocarbon that may serve as an oviposition cue, guiding gravid N. hequensis female adults to suitable host foliage. Female N. hequensis adults use contact chemoreception and olfaction to select oviposition sites; the presence of specific host volatiles indicates leaf quality and low competition. In contrast, male N. hequensis adults were most attracted to (E)-2-Hexenal, a ubiquitous green leaf volatile released from damaged or healthy leaves. This compound may indicate the presence of a host plant with active feeding (by conspecifics or other herbivores) and could serve as a rendezvous signal for mating. Such sex-specific divergence in odorant preference has been documented in other tenthredinids, e.g., Hoplocampa testudinea (apple sawfly) where female adults respond to floral volatiles and male adults to leaf volatiles. The practical implication is that we can develop gender-specific lures: o-Xylene for N. hequensis female adult biased trapping (which could reduce future population growth by removing reproductive N. hequensis female adults), and (E)-2-Hexenal for N. hequensis male adult monitoring (which is useful for early detection and phenology prediction). The N. hequensis male adult biased trap catch overall (4.5:1 ratio) reflects the natural sex ratio and suggests no strong behavioral bias beyond what is expected from population demography.

4.3. Trap Type Comparison

The large boat-shaped trap outperformed the triangular trap by a factor of 4.8. This difference is likely due to two design features: (1) larger sticky board area (800 cm² vs. 450 cm²), (2) four open entrances allowing approach from all directions vs. only two ends. The poor performance of triangular traps suggests that N. hequensis adults are not strongly attracted to the physical shape or color of triangular traps, and rely more heavily on olfactory cues. However, because triangular traps are cheaper (US$0.7 vs. US$1.5) and easier to install and maintain, they could be deployed in large numbers for low-density monitoring if the target is only males and the user accepts low catch rates.

4.4. Effect of Hanging Height

The experimental results showed that the combination of a 1.5 m suspension height, bucket-type traps, eraser-type slow-release lures, and slow-release bag lures did not significantly attract adult N. hequensis in the field. The total number of captured adults was <30, and none of the individual experimental groups showed any significant differences. Although these results deviated from the experimental expectations, they provide valuable references for future experimental designs and a deeper understanding of the biological characteristics of adult N. hequensis. When the traps were suspended at a height of 1.5 m, the total number of adult N. hequensis captured was <70. This is partly due to the weak flying ability of adult N. hequensis and because they tend to be active at lower heights owing to food sources, habitat preferences, or strategies to avoid predators. In some field-trapping experiments, the number of target insects captured by the traps gradually decreased with an increasing suspension height, and in some cases, the traps could not effectively capture the target insects. For example, in studies targeting Chilo suppressalis (Walker), the number of captured adults gradually decreased with an increasing trap suspension height (Zhou et al. 2020). Lure effectiveness is influenced by the suspension height. For example, in field-trapping experiments using lures containing plant volatiles for Monochamus alternatus, the attractants released by lures at higher heights could not be sufficiently dispersed or did not reach areas where the pests were active. Consequently, the pests could not detect the presence of attractants and were thus not attracted to the traps (Hu et al. 2018). In addition, higher lure positions are more susceptible to wind effects, which can alter the concentration and diffusion range of attractants, making it difficult for pests to be lured into the traps (Wang et al. 2011).

4.5. Effects of Environmental Factors on the Volatile Components of Host Plants

The composition of volatiles from the same plant species can vary in terms of composition and content depending on the geographical environment, season, growth stage, and organ location. Bai’s research showed that seasonal changes and daily variations can affect the release of volatile substances in subtropical bamboo forests (Bai et al. 2016); Wu’s analyses revealed significant differences in the types and content of volatile substances in the tender shoots of Citrus reticulata across the different months (Wu 2019); and the composition of volatiles in Zizyphus jujuba differs significantly during the young leaf, flowering, and young fruit stages. The variety of volatiles during the young fruit stage is much greater than that during the young leaf stage, and each developmental stage harbors its own unique volatiles (Han et al. 2010). Considering the above factors, volatile samples were collected in the present study during the adult occurrence period of N. hequensis. The collection period was from 10:00–15:00, which is the peak activity period for N. hequensis adults. The collected samples were mature, healthy tender leaves sourced from an area in which they exhibited long-term occurrence.

4.6. Influence of Environmental Factors on the Effect of Field Trapping

Environmental factors have a significant impact on the field-trapping efficiency of insects. Temperature, humidity, light, wind speed, vegetation coverage, and hanging height are important influencing factors (Su et al. 2020). Suitable temperature and humidity can promote the activity and foraging behavior of insects, thereby enhancing trapping efficiency (Xu et al. 2024). Light intensity and photoperiod significantly affect the activity and orientation of certain insects, and appropriate lighting conditions can significantly increase the capture rate of traps (Shang et al. 2024). High wind speeds may hinder the flight of insects and affect trapping efficiency, whereas under lower wind speed conditions, the capture rate of traps is higher (Chen et al. 2010). Vegetation cover and plant species have important effects on the habitat and environmental activity of insects. Suitable vegetation cover can enhance the capture rate of traps (Song et al. 2025). Therefore, in practical applications, comprehensively considering the impact of various environmental factors and choosing appropriate trapping strategies and methods to improve trapping efficiency are crucial.

4.7. Comparison with Other Sawfly Attractant Studies

Only a handful of studies have developed plant volatile attractants for sawflies. For the Neodiprion sertifer (pine sawfly), (E)-β-farnesene and other terpenes have been used. For Hoplocampa testudinea, pear ester has been tested. Our study is, to our knowledge, the first to report a sex-specific attractant for a Nematus species using host leaf volatiles. The combination of (E)-2-Hexenal (for males) and o-Xylene (for females) offers a flexible toolkit. The fact that both compounds are relatively inexpensive and commercially available increases the potential for rapid adoption.

4.8. Study Limitations and Future Directions

Several limitations must be acknowledged. First, all experiments were conducted in a single location (Lhasa, 3650 m altitude). Responses to volatiles may differ in other populations due to local adaptation or different environmental conditions (Yuan, 2010; Zhu, 2014). Multi-site validation across the species range is needed. Second, we did not measure real-time volatile release rates under field temperature and wind conditions; release rates likely vary with temperature, affecting effective concentration at the trap. Third, we did not test blend effects (mixtures of compounds). In nature, insects respond to complex blends, and a synthetic blend might be even more attractive than single compounds. Fourth, we did not assess the longevity of lures under field conditions beyond four weeks; longer-duration studies are needed for operational use. Fifth, the bucket traps and eraser/bag lures performed poorly in preliminary tests, but we did not systematically optimize them; there may be alternative designs or release materials that work better.

Future research should: (1) test the optimal combinations in other regions (e.g., lower altitudes, different climates); (2) evaluate blends, especially a 1:1 mixture of (E)-2-Hexenal and o-Xylene; (3) develop a controlled-release formulation with extended field life (e.g., 8-12 weeks); (4) assess non-target effects on beneficial insects; and (5) integrate attractant-based monitoring into decision-supported Integrated Pest Management (IPM) programs.

5. Conclusions

This study successfully developed and field-optimized plant volatile-based attractants for N. hequensis adults. The key findings and recommendations are: (1) Optimal configuration for general monitoring (both sexes): large boat-shaped traps, hanging height 1.0 m, spacing 10 m (one per 40 m²), slow-release bottle lures, with a choice of attractant depending on target sex; (2) For male-focused monitoring (early detection, phenology): use 10 μg/μL (E)-2-Hexenal. Expected capture: 265 males per trap over 4 weeks; (3) For female-focused monitoring (population suppression): use 10 μg/μL o-Xylene. Expected capture: 62 females per trap over 4 weeks; (4) Low-cost alternative for males only: triangular trap with 0.1 μg/μL ethylbenzene, but captures will be 30 males per 4 weeks, only 11% of the boat trap efficiency; (5) Non-linear concentration response: 10 μg/μL is optimal; higher concentrations are less effective.

These results provide the first plant-volatile attractant for a Tenthredinid sawfly (Hymenoptera: Tenthredinidae) and offer a practical, environmentally friendly monitoring tool for N. hequensis. The sex-specific responses also shed light on the chemical ecology of host location in sawflies. With further validation across regions and development of longer-lasting lures, this technology could be integrated into IPM programs for willow pests worldwide.