Pheromone‐based mating disruption of the tea tussock moth, Euproctis pseudoconspersa , in Camellia japonica street trees

Abstract

BACKGROUND

The tea tussock moth, Euproctis pseudoconspersa, is a nuisance and defoliating pest of Camellia japonica, widely planted as street and windbreak trees. Pheromone‐based mating disruption (MD) is a low‐hazard alternative for urban settings; however, its performance in tall, dense canopies remains elusive. Herein, we investigated seasonal flight activity, vertical flight distribution and MD efficacy across 3 years (2023–2025) in roadside C. japonica stands in Jeju, Republic of Korea.

RESULTS

Pheromone monitoring of male moths showed two distinct flight periods per year. Furthermore, racemic 10,14‐dimethylpentadecyl isobutyrate consistently attracted males. Male capture inhibition (MCI) was 34.8% when MD dispensers were deployed only at 1.5 m. Contrastingly, vertically distributing dispensers within the canopy (≈1, 3 and 5 m) increased MCI to 84.2–95.8% across subsequent trials. In 2025, leaf damage was decreased from 10.5% (control) to 2.0% (MD), corresponding to 81.0% damage reduction. Dispensers released pheromone for >60 days, whereas each male flight period lasted for ≈30 days. This indicates potential for decreasing pheromone loading to enhance cost‐efficiency without compromising disruption.

CONCLUSION

MD can effectively suppress E. pseudoconspersa in tall C. japonica street‐tree systems when dispensers are deployed at several canopy heights. Optimizing dispenser loading and enhancing deployment technologies (e.g. assisted canopy installation) may further improve feasibility in urban and forest environments. © 2026 Society of Chemical Industry.

Summary of the key finding: Deployment of racemic pheromone dispensers at multiple canopy heights markedly enhanced mating disruption efficacy against E. pseudoconspersa in C. japonica street trees, achieving ≤95.8% male capture inhibition and substantial leaf‐damage reduction.

1. INTRODUCTION

The genus Camellia (Family Theaceae), including Camellia japonica and Camellia sinensis, is traditionally cultivated for several applications, including essential oil production, tea and ornamental use. ¹ Between the two, C. japonica, an evergreen shrub, is native to Asia, including southern China, Japan and the Korean peninsula, and is now cultivated worldwide, including Europe, North and South America, southern Africa and Oceania. ² , ³ In Korea, C. japonica is planted as a windbreak in coastal areas of Jeju Island. Additionally, its seed oil is utilized for edible oil and cosmetic production. ⁴ It is widely planted as an ornamental plant in parks and arboreta in southern coastal areas of Korea. Moreover, its ornamental value has recently been further enhanced following its designation as ‘Our Ornamental Plant’ by the Korea National Arboretum, Korea Forest Service. ⁵ Additionally, its importance as an early spring food resource for honeybees has increased. ⁶ The distribution of C. japonica in Korea is currently limited primarily to the southern coast regions and Jeju; however, it is expected to expand inland and northward under climate change within the next 40 years. ⁷ Such alterations are likely to increase the ecological and economic criticality of C. japonica and may intensify the effect of its associated insect pests.

The tea tussock moth (TTM), Euproctis pseudoconspersa (Lepidoptera: Erebidae), is distributed throughout China, Japan, Taiwan and Korea. ⁸ , ⁹ The species exhibits a relatively narrow host range, feeding primarily on Camellia spp., including economically important tea plant, C. japonica and C. sinensis. ¹⁰ , ¹¹ Larvae are gregarious, and larval feeding results in severe foliar damage, ranging from leaf skeletonization to complete defoliation under outbreak conditions. ¹² The larva exhibits notorious urticating hairs that cause allergic reactions upon contact with human skin, ¹³ , ¹⁴ posing forestry and public health concerns. Although chemical insecticides are generally the most effective means of controlling this pest, their use has crucial drawbacks, including potential risks to human health and adverse effects on nontarget organisms and the environment. ¹⁵ In particular, the application of chemical insecticides may substantially influence humans because camellia seed oil is utilized as an edible raw material. Additionally, C. japonica trees are widely planted as urban street and ornamental trees in parks, where people frequently encounter them. For controlling TTM in tea plantations, chemical insecticides such as imidacloprid are used ¹⁶ despite concerns noted above, and pheromone‐based mass trapping using pheromone has been investigated. ¹¹ However, the application of these approaches to C. japonica has not yet been explored. Therefore, there is a growing demand for alternative control methods that are nonhazardous, effective and socially acceptable.

Pheromone‐based mating disruption (MD) is considered one of the most suitable techniques to meet these requirements. ¹⁷ , ¹⁸ , ¹⁹ MD is a pest control strategy that uses synthetic sex pheromones to interfere with mate‐finding behavior, thereby reducing successful mating and subsequent population growth. It works by saturating the environment with pheromone signals, which confuses males, inhibits their ability to locate females and ultimately suppresses reproduction. ¹⁷ The sex pheromone of the TTM has been identified as (R)‐10,14‐dimethylpentadecyl isobutyrate. ⁹ Additionally, males are attracted to (S)‐enantiomer and racemic mixtures. ²⁰ , ²¹ Our preliminary assays confirmed the racemic pheromone's suitability for field‐scale MD application as it successfully attracted male TTMs.

Effective MD requires that sufficient airborne pheromone concentrations be maintained throughout the temporal window of adult activity and across the spatial domain in which males search for females. Thus, in this study, we (i) characterized seasonal occurrence of the TTM, (ii) evaluated its vertical flight range within camellia canopies and (iii) assessed the field efficacy of MD across multiple seasons.

2. MATERIALS AND METHODS

2.1. Experimental site

All experiments were conducted over 3 years (2023–2025) on the roadsides of Seogwipo, Jeju (33° 19′ 26” N, 126° 43′ 57″  E). Camellia japonica trees were planted as roadside windbreaks along both the northern and southern sides of a 12‐m‐wide road, extending approximately 1 km east and 0.5 km west from the intersection. The trees were 7–10 m in height with crowns of ≈4 m wide, and spacing was narrow owing to their windbreak function (Supporting information, Fig. S1).

2.2. Attraction efficacy of the racemic pheromone and seasonal occurrence

The racemic sex pheromone (1) successfully attracted male TTMs in the preliminary field assays performed between September and November, 2022. Pheromone traps were deployed to confirm the attractiveness of racemic sex pheromone over a year and to investigate the seasonal occurrence of the TTM. Rubber septa (10 mm outer diameter × 18 mm; Wheaton Industries, Millvilles, NJ, USA) were utilized as dispensers. One milligram of the racemic sex pheromone, 10,14‐dimethylpentadecyl isobutyrate (1), which the authors synthesized, was applied to each rubber septa. The baited septa were placed in red Delta traps (AD Corp., Andong, Korea) at 1.5 m aboveground and attached to the branches of C. japonica. Monitoring was performed over 3 consecutive years (2023–2025) and trap inspection was conducted every 2 weeks. The pheromone lure was replaced every 4 weeks, and the adhesive sheet was replaced with a new sheet at each inspection. All of the adhesive sheets were brought to the laboratory for identification and counting. Only the male capture in control plot was used to determine the timing of adult flight periods per year.

2.3. Vertical flight range assessment

Pheromone traps were installed at three different heights to understand the vertical flight range of TTMs: ≈1 m (lower), 3 m (middle) and 5 m (upper) aboveground within the same tree. The traps were installed on 26 June 2024, and spaced ≈220 m apart. The same pheromone lure and trap types as described previously were used. Inspections were conducted every 2 weeks until 2 October 2024. Mean numbers of the captured males at each height were compared to assess the vertical stratification of male flight activity.

2.4. Chemicals and MD dispensers

2.4.1. Synthesis and chemical analysis

The sex pheromone, 10,14‐dimethylpentadecyl isobutyrate (1), was synthesized from two starting materials, 1,7‐heptanediol (2) and achiral 3,7‐diemthyloctan‐1‐ol (5) (Scheme 1). The Wittig reagent, (7‐hydroxyheptyl)triphenylphosphonium bromide (4), was prepared from 1‐bromoheptan‐1‐ol (3) and triphenylphosphine. The aldehyde 3,7‐dimethyloctanal (6) was obtained by oxidation of 3,7‐dimethyloctanol (5). A Wittig reaction between aldehyde 6 and Wittig reagent 4 yielded 10,14‐dimethypentadec‐7‐en‐1‐ol (7). Hydrogenation of the monoenol (7) followed by esterification with isobutyl chloride afforded the final desired pheromone (1) (Scheme 1). Mass spectra were obtained either on an 5975C mass‐selective detector (MSD) series apparatus coupled with a 7890N gas chromatography (GC) system (both Agilent, Santa Clara, CA, USA), equipped with an HP‐Innowax column (30 m × 0.25 mm inner diameter × 0.25 mm; J&W Scientific/Agilent). Infrared (IR) spectra were obtained using a Frontier FTIR spectrometer (PerkinElmer Co., Ltd, Shelton, CT, USA). Nuclear magnetic resonance (NMR) spectra were recorded in chloroform‐d (CDCl₃) with tetramethylsilane (TMS) as an internal standard using a Avance III HD spectrometer (Bruker, Billerica, MA, USA) at 600 MHz for ¹H spectra and at 150 MHz for ¹³C spectra and DEPT (135°) at NICEM, Seoul National University. The spectroscopic data confirming the structure of compound 1 are summarized in Table 1.

Scheme 1.

Synthetic processes for the preparation of the compound 1. Abbreviation: PPH₃, triphenylphosphine; PDC, pyridinium dichromate; 5% Pd/C, 5% palladium on carbon; MeOH, methanol, TEA, triethylamine

Table 1.

Spectroscopic data of compound 1 (10,14‐dimethylpentadecyl isobutyrate)

Technique	Spectral data
GC–MS (m/z, %)	326 (7.5, M+), 283 (3.0), 210 (0.9), 168 (4.3), 153 (8.3), 140 (5.2), 126 (15.0), 111 (21.9), 97 (35.6), 89 (100), 88 (41.6), 71 (66.9), 57 (51.8), 43 (55.3)
IR (cm−1, neat)	2926 (s), 1739 (s), 1468 (m), 1192 (m)
1H NMR (600 MHz, CDCl₃, δ ppm)	4.057 (t, J = 6.7 Hz, 2H), 2.537 (sept, J = 6.0 Hz, 1H), 1.608 (sex, J = 5.0 Hz, 2H), 1.523 (sept, J = 5.5 Hz, 1H), 1.397–1.187 (m, 17H), 1.166 (d, J = 6.0 Hz, 6H), 1.154–1.020 (m, 4H), 0.8685 (d, J = 5.5 Hz, 3H), 0.867 (d, J = 5.5 Hz, 3H), 0.840 (d, J = 5.0 Hz, 3H)
13C NMR (151 MHz, CDCl₃, δ ppm)	177.194 (C=O), 64.377 (CH₂), 39.368 (CH₂), 37.316 (CH₂), 37.093 (CH₂), 34.045 (CH), 32.759 (CH), 29.964 (CH₂), 29.586 (CH₂), 29.513 (CH₂), 29.241 (CH₂), 28.655 (CH₂), 27.966 (CH), 27.058 (CH₂), 25.898 (CH₂), 24.783 (CH₂), 22.691 (CH₃), 22.602 (CH₃), 19.693 (CH₃), 18.997 (CH₃ × 2)

Abbreviations: s = strong; m = medium; d = doublet; t = triplet; sept = septet; sex = sextet; m = multiplet.

2.4.2. MD dispenser

Matind disruption was conducted using the sex pheromone 1 and a dispenser consisting of an acetate filter impregnated with pheromone, enclosed in a low‐density polyethylene (LDPE) pouch (0.06 mm thickness, 25 × 70 mm). The MD dispenser was prepared as follows: The pheromone was dissolved in hexane and absorbed onto the acetate filters (5.0 outer diameter × 30 mm). Following hexane evaporation, the acetate filters were sealed into the pouch. Each dispenser contained 50 mg of the pheromone.

2.5. Pheromone residue daily release amount in the MD dispenser

Mating disruption pouches containing 50 mg of the pheromone were placed in a shaded outdoor area to measure the pheromone residue. The pheromone amount in the pouch was weighed every 10 days for the first 20 days and subsequently every 7 days up to 58 days in 2023 (20July–9 September) using an electronic balance (ML204; Mettler‐Toledo, Columbus, OH, USA). The percentage of the residue in the MD dispenser was calculated by dividing the pheromone amount on Day (D)n by that on D0. Daily release rate was calculated by subtracting the weight on Dn from that on Dm and dividing the interval days.

2.6. Field assay of MD

Table 2 summarizes the key information on the field bioassays conducted in Experiments A–D, including MD installation periods, experimental size, and pheromone application amounts.

Table 2.

Summary of MD deployment parameters, plot characteristics and pheromone application rates in field experiments conducted from 2023 to 2025

MD experiment	Periods of MD installation	Treatment	Area (m2)	Volume (m3)	Rate of MD application
A	20 September– 28 November 2023	MD (East)	2000 (100 × 20)	16 000	20 g (100 g ha−1, 1.25 mg m−3)
		Control (East)	2000 (100 × 20)	16 000	‐
		MD (West)	2000 (100 × 20)	16 000	20 g (100 g ha−1, 1.25 mg m−3)
		Control (West)	2000 (100 × 20)	16 000	‐
B	2 October – 24 November 2024	MD	4000 (200 × 20)	32 000	42 g (105 g ha−1, 1.3 mg m−3)
		Control	4000 (200 × 20)	32 000	‐
C	27 May– 10 September, 2025	MD	4000 (200 × 20)	32 000	42 g (105 g ha−1, 1.3 mg m−3)
		Control	4000 (200 × 20)	32 000	‐
D	30 September– 20 November 2025	MD	4000 (200 × 20)	32 000	42 g (105 g ha−1, 1.3 mg m−3)
		Control	4000 (200 × 20)	32 000	‐

Note: Area: length 100 m × width 20 m [road width (12 m) and tree crown width 4 m × 2 sides], volume: area × 8 m (average height of tree).

2.6.1. Experiment A

The MD dispensers (3–4 per tree) were hung on branches at 1.5 m aboveground. The MD‐treated plots were located at the east and west ends of a 100‐m road segment, whereas the non‐MD control plot (c. 200 m in length) was situated between them. A total of 40 g pheromone was applied on both sides of the road (40 g/16 000 m³; length 100 m × width 20 m × height 8 m). The distances between the MD‐treated and control plots were ≈470 m (east) and 180 m (west). Two pheromone traps were installed at 1.5 m aboveground in each plot (Fig. S2) and spaced ≈50 m apart.

2.6.2. Experiment B

The MD dispensers (three per tree) were installed at ≈1 m (lower), 3 m (middle) and 5 m (upper) aboveground using an arborist throw line. The MD‐treated plot was located at the east end of a c. 200‐m road segment, and the control plot was situated ≈220 m apart with a similar length. A total of 42 g pheromone was applied (42 g/32 000 m³; length 200 m × width 20 m × height 8 m). In each plot, pheromone traps were installed at the same heights as the MD dispensers, and spaced ≈50 m apart (Fig. S2).

2.6.3. Experiment C

The MD dispensers were vertically installed during first flight season in 2025. The installation method and experimental design were the same as those used in Experiment B. The MD‐treated plot was located at the east end of a c. 200‐m road segment, and the control plot was ≈600 m apart. A total of 42 g pheromone was applied. Pheromone traps were installed at corresponding vertical positions as in Experiment B and spaced ≈50 m apart (Fig. S2).

2.6.4. Experiment D

The experiment was performed during the second flight season in 2025. The experimental setup, including dispenser placement, plot arrangement and pheromone application, was identical to that of Experiment C.

2.6.5. Evaluation of male capture inhibition (MCI) and MD efficacy

Male capture inhibition (MCI) for each year was calculated using the following equation ²² : MCI (%) = 100 × (C − D)/C (C, number of captures in the control plots; D, number of captures in the MD plots). The analysis excluded zero catches, as recorded by the MD‐treated and control traps.

The MD efficacy was evaluated using the following equation: MD efficacy (%) = 100 × (C′ − D′)/C′ (C′, leaf‐damage rate of the control plot; D′, leaf‐damage rate of the MD plot). Fifty leaves per tree were randomly collected from 50 trees in both MD‐treated and nontreated plots (2500 leaves each) on 19 and 20 August to calculate the leaf‐damage rates. Among the damaged leaves, feeding damage caused by TTMs was identified based on their characteristic feeding behavior. Early instar larvae gregariously feed on the leaf's underside, consuming only the abaxial surface, whereas late instar larvae feed individually, leaving the leaf veins intact (Fig. S3).

2.7. Statistical analyses

Vertical flight range was evaluated by comparing the trap capture data according to the vertical position. The capture number of males were transformed to log (χ + 1) and analyzed using one‐way analysis of variance (ANOVA). Statistical analyses were performed using JMP v9.0.2 (SAS Institute, Cary, NC, USA). Mean (± SE) values of the untransformed data are reported.

Generalized linear mixed models (GLMMs) were used for the analysis of seasonal occurrence dynamics (SOD) and MD. In both analyses, the response variable included the number of captured males, assuming a Poisson error distribution with a log link function. For SOD, a fixed explanatory variable included the flight season (first versus second flight season), with year treated as a random effect. For the MD analysis, treatment (MD versus control) was included as a fixed explanatory variable, with trap included as a random effect. The effects of explanatory variables were assessed using type II Wald chi‐square tests, and post hoc pairwise comparisons were conducted using Tukey's honestly significant difference (HSD) test. All GLMM analyses were performed using lme4, car and emmeans packages in R (v4.3.3).

3. RESULTS

3.1. Attraction efficacy of racemic pheromone and seasonal occurrence

The racemic pheromone successfully attracted only TTMs through the years. In 2023, 725 males were captured in the pheromone traps. Two distinguishable flight seasons were present through the years and a similar temporal pattern was demonstrated every year. Figure 1 presents the seasonal occurrence of TTMs for 3 years. The first flight season occurred from 14 June to 24 July 2023, 25 June to 23 July 2024, and 2 to 16 July 2025. The second flight season occurred from 4 October to 14 November 2023, 2 October to 13 November 2024, and 10 October to 6 November 2025. The first peaks occurred on 13 July 2023, 23 July 2024 and 16 July 2025, and the second peaks occurred on 31 October 2023, 30 October 2024 and 22 October 2025. The capture numbers between first and second flight season were not significantly different (χ² = 1.4255, P = 0.2325).

Figure 1.

Seasonal occurrence of male E. pseudoconspersa captured by pheromone traps (mean ± SE) in 2023 (A, N = 4), 2024 (B, N = 3) and 2025 (C, N = 3).

3.2. Vertical flight range assessment

A difference in capture number was noted according to the trapping period. Therefore, the capture data were divided by week and compared (Fig. 2). No significant differences in male captures were observed among vertical positions (Week 1: F _2,9 = 0.9418, P = 0.46; Week 2: F _2,9 = 0.7821, P = 0.53).

Figure 2.

Vertical distribution of male E. pseudoconspersa captures at c. 1 m (lower), c. 3 m (middle) and c. 5 m (upper) aboveground. Bars with the same letters indicate statistically insignificant differences (P < 0.05).

3.3. Release profile of the experimental dispensers

Figure 3 shows the pheromone residue dynamics and daily release rates of MD dispensers. The residual pheromone content was calculated as a relative value, taking the amount of pheromone before treatment as 100%. The percentage of the residual pheromone in the MD pouch was 67.48% after 27 days and 48.78% after 58 days. The daily release amount ranged from 0.068 to 0.929 mg day⁻¹ per MD pouch, and averaged at 0.385 mg day⁻¹ per MD pouch, which corresponds to 0.771 g ha⁻¹ day⁻¹.

Figure 3.

Pheromone residue dynamics and daily release rates of mating disruption dispensers over 58 days of field aging.

3.4. MD field assay

The MD treatments affected the number of attracted males through the seasons (Table 3, MD Expt A: χ² = 7.6383, P = 0.00571; Expt B: χ² = 47.337, P < 0.0001; Expt C: χ² = 35.310, P < 0.0001; Expt D: χ² = 22.158, P < 0.0001). In 2023, the MCI was 34.8% (Expt A). Contrastingly, the MCIs in the second flight season (Expt B) in 2024 and both flight seasons (Expt C and Expt D) in 2025 were 87.3, 84.2, and 95.8%, respectively (Table 3). The percentage of damaged leaves by TTMs was 2.0% in the MD plot and 10.5% in control plot, and the MD efficacy following MD treatment was 81.0% in 2025 (Table 3).

Table 3.

Male capture inhibition, statistical outcomes and leaf‐damage reduction resulting from mating disruption treatments during field trials from 2023 to 2025

MD experiment	Periods male caught	Total number of males caught		Statistical values			MCI1 (%)	Damaged leaves (%)		Control efficacy
		MD	Control	χ2	df	P		MD	Control
A	14 October– 14 November 2023	257	394*	7.638	1	0.0057	34.8	‐	‐	‐
B	14 October– 13 November 2024	13	102***	47.337	1	< 0.0001	87.3	‐	‐	‐
C	2–16 July 2025	12	76***	35.310	1	< 0.0001	84.2	‐	‐	‐
D	22 October– 6 November 2025	3	71***	22.158	1	< 0.0001	95.8	50 (2.0)	263 (10.5)	81.0

Note: ¹MCI, male capture inhibition, ² Significant difference using Tukey's pairwise test at P < 0.01 (*) and P < 0.0001 (***).

4. DISCUSSION

The TTM showed geographic variation in voltinism and seasonal occurrence. It is bivoltine in the main island of Japan, whereas it is observed to have six annual generations in Taiwan. ²³ In Southern China, the TTM has been shown to complete two separate generations a year, ¹¹ whereas 2–4 generations per year are reported in the tea‐producing regions of China. ²⁰ In the present study, TTMs exhibited obvious bivoltine behavior in Korea. These geographic differences in voltinism are primarily driven by climatic conditions, particularly temperature, which directly influences developmental rates and the number of generations per year. Warmer regions, such as Taiwan and southern China, provide sufficient thermal accumulation to support multiple generations, whereas cooler regions, such as Korea, restrict development to fewer generations. Additionally, temperature‐dependent development and seasonal environmental factors, including humidity and rainfall, can affect stage‐specific survival and population dynamics, thereby shaping seasonal occurrence patterns. Such climatic influences on voltinism and phenology have been widely reported in forest lepidopteran insects, where increasing temperature can induce shifts from univoltine to bivoltine or multivoltine life cycles. ²⁴ These differences have important implications for MD strategies because the timing and duration of pheromone release must coincide with male flight activity. Therefore, MD deployment can be optimized to target discrete flight periods in bivoltine regions rather than requiring continuous season‐long pheromone release.

Although the sex pheromone of TTMs and its absolute configuration has been examined, the moths were attracted to a racemic pheromone which influenced male capture in this study. Additionally, field attract bioassay revealed no significant difference in the number of male captures despite the (R)‐enantiomer of the TTM sex pheromone evoking stronger EAG responses than the (S)‐enantiomer. ²⁵ In China, attractive efficacies of (R)‐ and (S)‐enantiomers was not significantly different. ²¹ Furthermore, several previous studies based in Japan and China have reported that TTMs were caught in traps containing a racemate. ⁹ , ¹¹ In the lepidopteran pheromone system, optical isomers play crucial roles. Leucoptera scitella L. (Lepidoptera: Lyonetidae) was highly attracted to the (S,S)‐enantiomer. The (+)‐enantiomer of cis‐7,8‐epoxy‐2‐methyloctadecane was several times more effective in eliciting trap capture of the male gypsy moth, Lymantria dispar (L.), than the racemic mixture owing to the antagonistic effects of the (−)‐enantiomer. ²⁶ Nevertheless, some previous reports have shown that racemates attract male moths, with the racemic pheromone mixtures of Rusicada privata, Lambdina pellucidaria and Lambdina athasaria attracting male moths. ²⁷ , ²⁸ , ²⁹

In the present study, racemates were used for MD and successfully inhibited male capture. Racemic pheromone formulations have been successfully applied for MD across multiple insect taxa, including Lepidoptera, Diptera, Hemiptera and Coleoptera. In species such as gypsy moth (Lymantria dispar), ³⁰ grapevine mealybug (Planococcus ficus) ³¹ and the grub beetle Dasylepida ishigakiensis, ³² racemic blends have shown satisfactory results in field‐scale disruption. This indicates that strict enantiomeric purity is not always necessary for effective MD. It has been reported that racemic mixtures appeared to be equally effective in preventing copulation under laboratory conditions for MD development. ³³ The high cost of synthesizing pure enantiomers has been a major barrier. Racemic blends, containing natural and non‐natural stereoisomers, may be advantageous for MD owing to their low production cost.

In MD Experiment A, MD dispensers were deployed only at a single height (1.5 m aboveground). Although the MD treatment substantially decreased male captures compared with the control, MCI was only 34.8%, which did not meet the registration standard required in Korea. ³⁴ We hypothesized that a mismatch between the dispenser installation height and the vertical flight range of the TTM might have caused the low MD efficiency. Thus, we assessed the vertical flight behavior of the moth. Male captures did not significantly differ among trap heights within the tree crown. Based on these results, we installed MD dispensers vertically across varying heights, which substantially enhanced the MCI to 84.2%–95.8% in Experiments B, C and D. These findings provide evidence that installing MD dispensers at several vertical positions in tall trees can improve MD efficacy, which is likely to be the result of maintaining more homogeneous pheromone concentrations throughout the canopy airspace. Vertical stratification of moth flight within tree canopies has been documented in different orchard and forest systems. Moreover, previous studies have suggested that tree height and the vertical distribution of the target pest influence the effectiveness of MD. ³⁵ , ³⁶ , ³⁷ Consistent with our findings, de Lame and Gut ³⁸ recommended installing MD dispensers at two vertical positions—particularly, in the upper and lower thirds of the canopy—in trees taller than 4.5 m to achieve adequate MD.

Herein, the mean release rate of the MD formulation was estimated to be 770 mg ha day⁻¹, which lies above the typical range of 20–500 mg ha day⁻¹ reported as sufficient for MD in several Lepidopteran pests. ³⁹ However, camellia crowns in this study were 7–10 m tall and formed dense, wind‐modified structures owing to their role as roadside windbreaks. Such canopy characteristics are likely to have increased pheromone dilution and vertical mixing, thereby requiring higher emission rates to maintain effective pheromone concentrations throughout the active airspace of male moths. A similar requirement for higher release rates has been observed in other tall‐canopy systems, where canopy architecture and airflow dynamics influence dispenser performance. ⁴⁰ Additionally, upper‐range emission values of 300–500 mg ha day⁻¹ reported for codling moth and European grapevine moth ⁴¹ , ⁴² support the fact that relatively high emission levels are necessary under structurally complex or ventilated environments. Despite the relatively high release rate, the effective flight window of TTMs lasts no more than c. 30 days per generation, whereas the MD dispensers continued releasing pheromone for >60 days. This mismatch suggests that pheromone loading could be decreased in future formulations to enhance cost‐efficiency without compromising disruption efficacy.

Taken together, the strong inhibition of male captures and the considerable decrease in leaf damage show that MD is a promising management strategy for TTMs. Nonetheless, manually deploying dispensers using an arborist throw line is labor‐intensive and impractical for large‐scale implementation. Therefore, development of alternative application methods, such as UAV‐based deployment, will be necessary for enhancing operational efficiency and facilitating broader adoption of MD in urban and forest environments.

5. CONCLUSION

MD using a racemic pheromone effectively suppressed E. pseudoconspersa (TTM) in tall C. japonica street‐tree systems when dispensers were deployed across multiple canopy heights. Vertical placement within the canopy was a key determining factor for MD efficacy. Furthermore, the pheromone release duration exceeded the male flight period, suggesting that pheromone loading could be reduced to improve cost‐efficiency. These findings demonstrate the potential of racemic pheromone‐based MD as an effective and environmentally compatible strategy for managing TTMs in urban and forest landscapes.

CONFLICT OF INTEREST

Authors, S.J., Y.U.K. and K.B.K. were employed by the company AD Corp. The remaining author, J.K. declares no conflicts of interest.

AUTHOR CONTRIBUTIONS

Conceptualization, J.K.; methodology, J.K., S.J., Y.U.K. and K.B.K.; validation J.K. and S.J.; formal analysis, J.K. and S.J.; investigation, J.K., S.J., Y.U.K. and K.B.K.; data curation, J.K.; writing—original draft preparation, J.K.; writing—review and editing, J.K., S.J., Y.U.K. and K.B.K.; project administration, J.K.; funding acquisition, J.K. and S.J. All authors have read and agreed to the published version of the manuscript.

Supporting information

Figure S1. Roadside view of the experimental site of C. japonica street trees: (A) eastern section and (B) western section of the study area.

Fig. S2. Field bioassay experimental site in Seogwipo, Jeju. Mating disruption experiments in 2023 (A), 2024 (B) and 2025 (C), and vertical flight range assessment in 2024 (D). Circles and triangles in red indicated the pheromone traps.

Fig. S3. Diagnosis of damage leaf damage on C. japonica.

(A) Typical damage caused by the tea tussock moth (E. pseudoconspersa), (B) enlarged view of (A), and damage caused by other factors including microorganisms (C) and other insect pests (D).

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.