Hazelnut tolerance and weed response to soil-applied pulsed electric field

Abstract

Weed management in orchards often relies on herbicides, increasing the risk of resistance and highlighting the need for effective nonchemical alternatives. Pulsed electric field (PEF) technology delivers short, high field-strength pulses that disrupt cell membranes; however, the efficacy of soil-applied PEF for orchard weed control remains largely untested. This study evaluated soil-applied PEF for weed management and hazelnut (Corylus avellana L.) tolerance under field conditions and assessed how soil incubation time influences weed susceptibility to PEF under controlled conditions. In a young hazelnut orchard, treatments included a nontreated control, a hand-weeded control, and soil-applied PEF at 1, 10, and 100 J cm^-³, applied once using opposing vertical-pin soil electrodes. Weed control was assessed up to 60 days after treatment, and crop response was evaluated through visual injury and physiological measurements. In parallel container experiments, tubers of Cyperus esculentus and seeds of Echinochloa crus-galli and Amaranthus palmeri were planted and incubated for 0, 3, 5, or 7 days prior to PEF application at 0, 25, 50, or 200 J cm^-³ using a parallel-plate applicator. In the field, soil-applied PEF caused minimal crop injury and had no consistent effect on physiological parameters, indicating high crop tolerance; however, weed suppression was limited even at 100 J cm^-³. In contrast, under controlled conditions, PEF reduced weed biomass, with efficacy strongly influenced by incubation time and species identity. Increasing incubation time by 3 to 7 days increased susceptibility across species and reduced the energy required for effective control, with substantial biomass reductions observed at 25–50 J cm^-³. These results demonstrate that soil-applied PEF provides limited weed control under current field application method, likely due to inefficient energy delivery. In contrast, the parallel-plate application in containers effectively controlled weed propagules. Improving applicator design and optimizing energy delivery pathways will be critical to enhance the performance of PEF technologies in orchard systems.

1 Introduction

Hazelnut (Corylus avellana L.) is a high-value perennial nut crop with major production concentrated in Turkey and Italy, and expanding production in the United States and Chile (FAO, 2026). Hazelnuts are economically important for domestic consumption and export markets and are a key raw ingredient for confectionery products, including chocolate-based goods (Bento et al., 2016; Dağ and Sarica, 2025). In the Pacific Northwest of the US, hazelnut is a high-revenue specialty crop that supports many growers and rural communities (Asci and Devadoss, 2021).

In hazelnut orchards, weed management is a persistent challenge because tree rows are permanent, mechanical access is limited because it can damage the root system, and weeds compete with trees, reducing growth, productivity, and long-term orchard performance (Mennan et al., 2006; Kaya-Altop et al., 2016; Mennan et al., 2020). Effective in-row weed management is critical during the first years after planting to ensure rapid canopy development and to maintain the clean orchard floor required for efficient nut harvest (Pedroso and Moretti, 2022; Sauk and Beyhan, 2024). Consequently, weed control programs rely on repeated herbicide applications in the planted row combined with flail mowing the alleyways (Mennan et al., 2020; Moretti, 2021). This continued reliance on herbicides intensifies selection pressure for resistant weed populations, increasing the likelihood of control failures and reducing the longevity of effective modes of action. In Oregon hazelnut orchards, herbicide-resistant weeds have been documented, including Italian ryegrass (Lolium perenne L. ssp. multiflorum (Lam.) Husnot) and annual bluegrass (Poa annua L.) (Miranda and Moretti, 2025; Heap, 2026; Miranda et al., 2026). Recent work in Oregon hazelnut production has also identified multiple herbicide-resistance cases in annual bluegrass (Poa annua), underscoring the vulnerability of perennial orchard systems to resistance evolution under repeated chemical applications (Avila-Garcia et al., 2012; Bobadilla et al., 2021; Miranda and Moretti, 2025; Miranda et al., 2026). The evolution of herbicide resistance highlights the need for nonchemical tools capable of diversifying weed management programs.

One promising area is electric weed control, which is commonly applied to weed foliage. In aboveground contact systems, electrical energy is delivered through electrodes that contact plant tissues, allowing current to pass through the weed and into the soil (Baccin and Moretti, 2025b; Borger and Slaven, 2025; Baccin and Moretti, 2026). The most common approach uses alternating current (AC) delivered through metal applicators that contact the plant canopy (Diprose et al., 1985; Vigneault and Benoît, 2001). Electricity flowing through the plant increases tissue temperature, disrupts cellular integrity, and can result in plant death. These systems are generally most effective on young, herbaceous, and well-hydrated weeds (Baccin and Moretti, 2025b; Borger and Slaven, 2025; Baccin and Moretti, 2026). However, aboveground electric weeders can be bulky, energy-demanding, and nonselective; therefore, careful application is required to avoid crop injury (Schreier et al., 2022).

An emerging technology in electric weed control is pulsed electric field (PEF) (Bloomer et al., 2023; Yudaev et al., 2023; Yudaev and Daus, 2023; Gafni et al., 2025). Unlike AC and DC systems, which deliver energy continuously, PEF delivers short, high field-strength electrical pulses capable of disrupting cellular membranes through electroporation (Kotnik et al., 2019; Riga et al., 2020). Field strength, defined as voltage per unit distance between electrodes (V mm^-¹ or V cm^-¹), is determined by the applied voltage and electrode spacing. In food processing, electric fields< 100 V cm^-¹ are considered low intensity, 0.1–1 kV cm^-¹ moderate, and > 5 kV cm^-¹ high intensity (Bazhal et al., 2003; Lebovka et al., 2005).

Although PEF is widely adopted in food processing and biotechnology to inactivate microorganisms (Garner, 2019; Cavalcanti et al., 2023), soil-applied PEF has recently demonstrated agronomic potential in soil pest management (Bozorov, 2020; Moretti et al., 2024). Soil-applied PEF differs from aboveground contact electrical weed control because electrical pulses are delivered into a defined soil volume rather than directly through weed foliage. In this approach, electrodes are inserted into the soil or placed on opposite sides of the soil volume, and short electrical pulses are discharged between electrodes of opposite polarity. Current moves through conductive soil-water pathways, and biological targets are exposed when seeds, propagules, roots, or emerging seedlings intersect the effective electric field. Riga et al. (2020) reported effective suppression of plant-parasitic nematodes and soilborne pathogens following soil-applied PEF treatments, showing that a lethal electrical dose can be delivered to below-ground biological targets. These findings suggest that soil-applied PEF can function as a pest management tool and that similar mechanisms could be exploited for weed suppression. Electroporation is the disruption of cell membranes caused by an external electric field. When the induced membrane disruption is temporary, cells may recover; however, when the disruption is severe or prolonged, electroporation can become irreversible, leading to loss of membrane function, cellular leakage, and tissue death.

From a biological perspective, germinating weed seeds and vegetative propagules may be suitable targets for soil-applied PEF because germination activates membrane systems, increases tissue hydration, and establishes conductive plant–soil pathways that may facilitate current flow through living tissues (Leong et al., 2016; Kotnik et al., 2019). Electroporation theory predicts that metabolically active cells are more susceptible to irreversible membrane destabilization when critical transmembrane potentials are reached (Kotnik et al., 2019). Consequently, seed and propagule susceptibility to PEF is likely governed by germination stage, hydration status, tissue conductivity, and applied energy (Evrendilek et al., 2019). Several abiotic factors that regulate weed seed germination may also influence the effectiveness of soil-applied PEF, including soil temperature, moisture, light exposure, nitrate availability, pH, and aeration (Travlos et al., 2020). Therefore, soil-applied PEF may be effective for weed management if energy delivery to plant tissues is optimized and applications coincide with susceptible developmental stages. However, there is currently no integrated evaluation linking orchard-scale crop tolerance and preemergence weed efficacy. Therefore, the objectives of this study were to (i) quantify hazelnut physiological tolerance to soil-applied PEF under orchard conditions, (ii) evaluate the efficacy of soil-applied PEF for weed control, and (iii) determine how soil incubation time influences weed susceptibility to PEF.

2 Materials and methods

The field study was conducted at the Oregon State University Lewis-Brown Research Farm near Corvallis, OR (44.564° N, -123.262° W) on a Malabon silty clay loam soil with 5% organic matter, and a pH of 6.0. The study was conducted in a 6-month-old experimental hazelnut orchard planted with “Polly-O” trees spaced at 0.9 m apart in a single row. Each experimental plot consisted of three adjacent trees. The study was organized as a randomized complete block design with four blocks and five treatments: hand weeding, nontreated control, and soil-applied PEF at 1, 10, and 100 J cm^-3. PEF treatments were applied once in July 2025, at an air temperature of 34 °C and a relative humidity of 56%. The field studies were conducted during the summer of 2025. Representative images of the young hazelnut orchard, field application, and vertical-pin electrode configuration are shown in Figure 1.

Figure 1

2.1 Pulsed electric field in a young hazelnut orchard

Pulsed electric field treatments were applied using the Direct Energy System (Lisi Global, Richland, WA, USA). The system generates millisecond-duration electrical pulses that are discharged into the soil, with energy flow confined between two electrodes of opposite polarity. Pulse frequency was maintained at 100 Hz and pulse duration at 0.005 s. The electric field strength was 200 V mm^-1. The total energy delivered in each treatment was determined by the number of pulses applied. The energy of a single pulse corresponds to the energy stored in the capacitor prior to discharge and was calculated as

where E is the energy of a single pulse (J), C is soil capacitance, and V is the peak voltage applied across the electrodes. The number of pulses required to reach each treatment level was calculated by dividing the target energy by the energy of a single pulse.

In all cases, a treatment consisted of a single application. PEF was delivered either through a vertical-pin in the field study or parallel-plates in the container studies. The vertical-pin apparatus was a stationary electrode frame consisting of two rows of 12 pins of positive and negative polarities (Figure 1). Rows were spaced 30 cm apart, and opposing polarities were assigned by row. Metal pins were 6.35 mm in diameter and were 10 cm in length. Pins were spaced 5.08 cm apart within a row, creating a 60 cm swath and treating an area of 0.18 m² per placement. Based on the pin insertion depth of 10 cm, the estimated soil volume exposed to the electric field at each placement was approximately 0.018 m³. The frame treated one tree at a time and was moved sequentially along the planting row until all three plants per plot were treated.

2.1.1 Crop response

Crop response was evaluated using visual injury ratings and physiological measurements. Visual crop injury was assessed using the Canadian Weed Science Society 0–100 crop phytotoxicity rating scale, where 0% represents no visible phytotoxicity relative to the untreated control and 100% represents complete plant death (Canadian Weed Science Society, 2018). Ratings were based on general crop phytotoxicity, including visible chlorosis, necrosis, stunting, canopy reduction, and plant mortality. Following the CWSS-SCM scale, injury from 0 to 9% was considered slight discoloration and/or stunting, 10% was considered the upper threshold of acceptable injury, 11 to 30% was considered not acceptable, and injury greater than 30% was considered severe (Canadian Weed Science Society, 2018). Leaf chlorophyll content was measured using a fluorescence-based meter (CCM-300, OptSciences, Hudson, NH, USA). The equipment measures the chlorophyll fluorescence emission at 700 nm to far-red emission at 735 nm, and estimates chlorophyll based on the ratio of F735/F700 using regression equations (Gitelson et al., 1999).

Photosynthetic quantum yield of photosystem II (ΦPSII) and stomatal conductance were measured simultaneously on the same leaf using a porometer/fluorometer (LI-600; LI-COR Biosciences, Lincoln, NE, USA). The maximum fluorescence under light-adapted conditions (Fm′) was obtained using a high-intensity, multiphase saturating flash following established fluorescence measurement procedures (Demmig-Adams and Adams, 1996). The ΦPSII was calculated by the LI-600 algorithm: ΦPSII = (Fm′ − Fs)/Fm′. Fs is the steady-state fluorescence under ambient light conditions. Measurements were collected at 1, 3, 7, 14, 30, and 60 days after treatment (DAT). Measurements were taken from three fully expanded leaves located mid- canopy of each tree within each plot.

2.1.2 Weed assessments

Weed response was evaluated at 1, 3, 7, 14, 30, and 60 days after treatment (DAT) using weed control around hazelnut trees in the row. Weed control (%) was visually estimated on a 0 to 100% scale, where 0 means no control relative to the nontreated area and 100 means complete control.

2.2 Influence of soil incubation time on weed susceptibility to pulsed electric field

Greenhouse experiments were conducted at Oregon State University, Corvallis, OR, to evaluate the effect of soil incubation time and PEF energy density on weed control efficacy. Greenhouse conditions were maintained at 28 ± 2 °C during the day and 21 ± 2 °C at night, with 75 ± 5% relative humidity and a 14/10 h light/dark photoperiod. Natural light was supplemented with 400-W LED lamps, and photosynthetic photon flux density averaged 850–1200 µmol m^-² s^-¹.

Plant propagules included Cyperus esculentus L. tubers and seeds of Echinochloa crus-galli L. P. Beauv. and Amaranthus palmeri S. Wats. Propagules were soaked in water for 12 h at 5 °C prior to planting. Tubers of C. esculentus were planted at a depth of approximately 2 cm, and seeds of E. crus-galli and A. palmeri were planted at approximately 1 cm depth.

Propagules were planted in square plastic containers with a top width of 8.9 cm, a bottom width of 6.4 cm, and a depth of 8.9 cm. Containers were filled with approximately 500 cm³ (5 × 10^-4 m³) of sterilized substrate consisting of a 1:1 mixture of pasteurized sand and Willamette loam soil. Because soil electrical conductivity and texture influence current distribution between electrodes, soil physical and chemical properties were analyzed prior to the experiment. Soil analysis indicated a sandy texture (78% sand, 16% silt, and 6% clay), pH of 7.2, electrical conductivity of 0.12 dS m^-¹, cation exchange capacity of 13 meq 100 g^-¹, and organic matter content of 0.5%. Soil moisture was maintained at approximately 20% volumetric water content throughout the study. Six C. esculentus tubers or approximately 0.035 g of seeds of the other species (≈16 seeds of E. crus-galli and ≈100 seeds of A. palmeri) were planted per container. The experimental unit consisted of approximately 500 cm³ of soil exposed to the electric field during each PEF treatment. In the container experiments, PEF was applied once after propagules had been planted and incubated in moist soil for 0, 3, 5, or 7 days. The 0-day incubation treatment was treated immediately after planting, before visible germination or emergence, whereas the 3-, 5-, and 7-day incubation treatments were treated after progressively longer periods of germination and early seedling development. This timing was used because actively growing tissues were expected to be more susceptible to membrane disruption caused by pulsed electric fields.

Containers were treated with PEF at 0, 25, 50, or 200 J cm^-³ using a parallel-plate applicator (Figure 2). The system delivered pulses at a field strength of 200 V mm^-¹, a frequency of 100 Hz, and a pulse duration of 0.005 s. The parallel-plate applicator consisted of two aluminum electrodes (5 × 5 × 0.31 cm). One electrode was placed at the bottom of the container before filling with soil and remained in place throughout the experiment; the second electrode was placed on the soil surface and secured using a spring-loaded mechanism to ensure consistent contact with the soil surface. The distance between electrodes corresponded to the soil depth in the container (8.9 cm). This configuration directed current vertically through the soil profile, exposing the entire soil column to the applied electric field.

Figure 2

The experiment was arranged as a two-factor factorial in a randomized complete block design with four blocks. Factors included soil incubation time (0, 3, 5, and 7 days before PEF application) and PEF energy density (0, 25, 50, or 200 J cm^-³). Plants were harvested 28 days after treatment. Shoots and roots were separated, dried, and weighed. Each species experiment was conducted twice, and species were analyzed as independent experiments.

2.3 Statistical analysis

All analyses were conducted in R (version 4.5.3; R Core Team, Vienna, Austria). For the field study, repeated measures ANOVA models included treatment, DAT, and treatment × DAT interaction. When treatment effects were significant, means were separated using Sidak-adjusted comparisons (α = 0.05) by the emmeans and multcomp packages. Model assumptions were evaluated using residual diagnostics, and transformations were applied when necessary.

The container study design had two factors: incubation time and applied energy density. Each treatment combination was replicated four times, and the experiment was repeated. Each weed species was conducted and analyzed as an independent experiment rather than as a species factor within a single combined factorial model because species differed in propagule type, planting density, growth habit, and biomass magnitude. As an exploratory multivariate summary, redundancy analysis (RDA) was implemented to assess weed biomass responses to PEF across species, energy density, and incubation time (Milberg and Hallgren, 2002). RDA was performed using square root–transformed shoot and root biomass as response variables and energy density and incubation time as explanatory variables, with species included as a conditioning factor to account for interspecific variation. The analysis was implemented in the vegan package in R, with response variables scaled prior to ordination (Oksanen et al., 2013).Because each weed species was conducted as an independent experiment, shoot and root biomass were analyzed separately by species. For each species, shoot and root biomass were analyzed using linear models that included incubation time, PEF energy density, and their interaction as fixed effects. Thus, six biomass models were fitted in total: one shoot biomass model and one root biomass model for each of the three species. Incubation time was treated as a continuous variable to estimate biomass response slopes across incubation periods, whereas PEF energy density was treated as a categorical factor. Linear trends describing the change in biomass as a function of incubation time within each PEF energy density were estimated using the emtrends function in the emmeans package. Ninety-five percent confidence intervals were used to assess whether slopes differed from zero within each species, response variable, and energy density. Species-specific biomass analyses were interpreted within species and were not used to make direct statistical comparisons among species. Model assumptions were evaluated by visual inspection of residual plots.

3 Results

3.1 Pulsed electric field in a young hazelnut orchard

Soil-applied PEF at 1, 10, and 100 J cm^-³ caused little to no measurable crop response and did not provide effective weed control. Across the response variables, treatment effects were detected for weed control, primarily because of the hand-weeded treatment, whereas hazelnut injury and physiological measurements did not show consistent treatment responses over time. (Table 1; Figure 3).

Figure 3

A significant treatment effect (P = 0.0033) was observed for weed control, but no effect of DAT and its interaction were observed (Table 1). The hand-weeding treatment was the only treatment that controlled weeds, primarily C. esculentus, whereas PEF had no effect. Hand- weeding consistently resulted in the highest weed control at every evaluation date, reaching 100% across all evaluation dates (Table 2). PEF at 1 and 10 J cm^-3 were comparable to the nontreated control, with 0% control at every assessment (Table 2). PEF at 100 J cm^-3 produced minor numerical suppression, reaching 25% visual weed control at 1 and 3 DAT; however, this response was not statistically different from the nontreated control and declined to 12.5% at 7 and 14 DAT, remaining below 13% at 30 and 60 DAT (Table 2). Crop injury was negligible during the entire study (Table 2), suggesting that PEF is selective and safe for hazelnut at the tested energy densities in this study. Injury ratings were 0% for the control, hand-weeding, and the 1 J cm^-3 treatment at all evaluation dates. Negligible injury (2.5%) was observed later in the season for the 10 J cm^-3 and 100 J cm^-3 treatments at 14, 30, and 60 DAT, but these values did not differ among treatments.

Based on the physiological responses measured, PEF did not injure the crop. Leaf chlorophyll content increased during the study time for all treatments and then stabilized, which is typical as leaves mature and canopy conditions change through the season (Figure 3A). There was a strong DAT effect for chlorophyll (P <0.001), while treatment and DAT × treatment were not significant (Table 1). In the figure, the treatment lines generally increased from the earliest dates and remained close to one another from 7 to 30 DAT, with only small spreading at 60 DAT. Numeric differences did not form a consistent pattern that aligned with energy density.

Photosystem II quantum efficiency also showed no response to PEF treatment (Figure 3B), including a slight and brief decrease at specific dates, but the overall shape of the response was similar across treatments and did not indicate sustained stress in PEF-treated trees. This was supported by the ANOVA, which did not detect significant effects of DAT, treatment, or their interaction for photosystem efficiency (Table 1). Stomatal conductance increased quickly after treatment and then remained within a similar range among treatments through 60 DAT (Figure 3C). Even small daily differences were evident among treatments; no patterns were associated with PEF. These results indicate that, in a young hazelnut orchard, soil-applied PEF was safe at the tested energy densities but not effective for weed control.

3.2 Influence of soil incubation time on weed susceptibility to pulsed electric field

Redundancy analysis indicated that, after accounting for species effects, energy density and incubation time explained a small but significant portion of the multivariate variation in shoot and root biomass (Figure 4; partial RDA: F = 4.13, P = 0.005). Species accounted for most of the total variation in biomass response, with 72.1% of the total inertia conditioned by species. After this species effect was removed, the constrained component associated with energy density, incubation time, and their interaction represented 1.33% of total inertia, equivalent to 4.76% of the remaining variation. Incubation time was the strongest experimental predictor in the partial RDA (F = 7.08, P = 0.011), followed by energy density (F = 5.08, P = 0.025), whereas the energy density × incubation time interaction was not significant (F = 0.23, P = 0.691). The primary constrained axis captured most of the treatment-related structure (RDA1 = 92.9% of constrained variation; F = 11.50, P = 0.008), whereas RDA2 was not significant (F = 0.89, P = 0.349). The direction of the response vectors indicated that increasing incubation time and energy density were associated with reductions in shoot and root biomass. Variance partitioning further confirmed that species identity explained the majority of variation in biomass responses (unique adjusted R² = 0.675), while the unique fractions explained by energy density and incubation time were small but detectable (unique adjusted R² = 0.005 and 0.004, respectively). Thus, the RDA supports the interpretation that biomass response was strongly species-dependent, while incubation time and energy density contributed smaller but significant reductions in biomass.

Figure 4

Biomass responses to PEF varied among the independently analyzed species (Tables 3, 4; Figure 5). In A. palmeri, shoot and root biomass were generally low across treatments and incubation times, typically less than 0.15 g, which limited visual separation among energy densities. Although treatment differences were not consistent across all incubation times, PEF tended to reduce biomass relative to the nontreated control. The effect of incubation time was modest overall; however, a significant negative relationship between incubation time and biomass was detected at 50 J cm^-³ for both shoots and roots, with slopes of −0.0159 (95% CI: −0.0226 to −0.0092) and −0.0097 g day^-¹ (95% CI: −0.0150 to −0.0045), respectively, indicating increased susceptibility with increased incubation time.

In contrast, C. esculentus tubers exhibited the greatest biomass and the strongest response to PEF. In the nontreated control, shoot and root biomass remained high across all incubation times. PEF treatments substantially reduced biomass at 3, 5, and 7 days after planting, with shoot and root biomass often approaching zero regardless of energy density (Figure 5). This pattern was supported by slope analysis, which showed strong negative relationships between incubation time and biomass at 25 and 50 J cm^-³. At 25 J cm^-³, shoot and root biomass declined by −0.098 and −0.174 g day^-¹, respectively, while at 50 J cm^-³, slopes were −0.098 and −0.179 g day^-¹ (Table 4), indicating a pronounced increase in susceptibility with delayed exposure. In contrast, no clear incubation-time effect was observed at 200 J cm^-³, suggesting that this energy density was sufficient to cause high levels of control regardless of tuber incubation time. These results indicate that while high energy densities can overcome developmental differences, allowing tubers to incubate reduced the energy required for effective control, with substantial suppression achieved at 50 J cm^-³ following short incubation periods. At 0 days, treatment differences among energy densities were more pronounced, with intermediate energies occasionally resulting in greater biomass than the highest energy density, indicating that dose effects were more detectable immediately after emergence.

For E. crus-galli, biomass was intermediate relative to the other species, and PEF reduced both shoot and root biomass at several incubation times. The response varied with both energy density and incubation timing. At 0 days after planting, biomass reduction was only evident at 200 J cm^-³, whereas lower energy density produced limited suppression. As incubation time increased (3–7 days), biomass declined across multiple energy densities, indicating an increased sensitivity to PEF energy as incubation increased. Slope analysis indicated significant reductions in biomass with increasing incubation time at 0 and 25 J cm^-³ for shoots, with slopes of −0.0115 and −0.0085 g day^-¹, respectively (Table 4). For roots, a significant negative slope was observed at 0 J cm^-³ (−0.0149 g day^-¹), while slopes at higher energy levels were not significantly different from zero. These results indicate that biomass declined with increasing incubation time, but the magnitude of the response was smaller and less consistent across energy density than observed for C. esculentus, suggesting a weaker dependence on incubation time for PEF efficacy in this species. Across species, increasing incubation time generally enhanced PEF efficacy, particularly at intermediate energy density (25 and 50 J cm^-³). This pattern indicates that delayed application, coinciding with early seedling development, increased treatment effectiveness. In contrast, responses at the highest energy density (200 J cm^-³) were less sensitive to incubation timing, suggesting that high energy densities reduced the relative importance of plant developmental stage.

4 Discussion

In the hazelnut orchard, soil-applied PEF applied at 1, 10, and 100 J cm^-³ produced little to no crop response. Hazelnut injury remained near zero throughout the evaluation period, and physiological indicators (chlorophyll index, photosynthetic efficiency, and stomatal conductance) did not show consistent treatment effects over time. These observations suggest that the PEF doses and application configuration used in this study did not cause measurable damage to young hazelnut trees, and most temporal variation in physiological measurements was likely driven by seasonal and environmental factors rather than the electrical treatment itself. The limited crop response observed in the present study is consistent with previous reports indicating that electric pulse treatments can be applied to orchard trees with minimal injury when parameters are properly selected (Bozorov, 2020). Soil-applied PEF has shown promise for other belowground targets, including plant-parasitic nematodes and soilborne pathogens (Riga et al., 2020) and certain insect pests (Kusakari et al., 2020; Jin et al., 2025).

Despite the promising efficacy for controlling soilborne pests, PEF had negligible effect on weed control in the present study. The 1 and 10 J cm^-3 energy densities were similar to the nontreated control, and the 100 J cm^-3 treatment provided only minor numerical suppression early after application, which was not statistically different from the nontreated control (Table 2). An interpretation of these findings is that the soil-applied PEF method did not deliver energy at a level that caused detectable effects. Instead, this limited response may also reflect constraints of the applicator design, as the vertical-pin configuration likely produced a highly non-uniform electric field with limited electrode–soil contact, reducing effective energy transfer into the soil profile. Previous work in food processing has documented that electrode geometry strongly influences electric field distribution and treatment efficacy, with non-uniform configurations resulting in reduced biological effects with localized regions of high intensity near electrodes and large portions of the treated volume receiving sublethal exposure (Jeyamkondan et al., 1999). Variations in soil moisture, texture, and organic matter, as well as weed size and inconsistent pin–plant contact, can alter current pathways and reduce the effective energy received by target tissues. Because electric current follows the path of least resistance, heterogeneous soil conditions and irregular plant distribution result in uneven energy delivery within the treated volume (Rhoades et al., 1989; Corwin and Lesch, 2005). As a result, soil-applied PEF is unlikely to provide the uniform plant tissue exposure required for consistent weed control, particularly under variable weed density and growth stage, which is consistent with the limited biological response observed in the hazelnut orchard. The transient suppression of weeds observed at 100 J cm^-³ may also be consistent with incidental direct contact between the applicator and weed foliage, rather than an effect of electricity moving through the soil to the roots. Electrical weed control is known to be highly effective when energy is delivered directly to plant tissues. For example, alternating current (AC) systems can strongly suppress C. esculentus when electricity is applied to foliage (Baccin and Moretti, 2025b). Similarly, foliar-applied pulsed electric field systems have demonstrated control of multiple weed species, including perennials, when electrical energy is delivered through aboveground tissues (Bloomer et al., 2023; Gafni et al., 2025).

The contrasting responses between the field and container studies were likely driven by differences in applicator configuration, treated soil volume, and electric field distribution. The field study used a vertical-pin applicator, whereas the container study used a parallel-plate applicator. These represent fundamentally different energy-delivery configurations. Recent work with soil-applied PEF demonstrated that parallel-plates generate a more uniform electric field distribution than vertical-pin, and that this greater uniformity was associated with stronger biological suppression of weeds, plant-parasitic nematodes, and soilborne pathogens (Benedetti et al., 2026). In the present field study, the vertical-pin configuration treated a larger and more heterogeneous soil volume, where soil moisture, bulk density, organic matter, roots, and weed position likely altered current pathways. Because current follows the path of least resistance, portions of the treated soil volume may have received sublethal exposure. In contrast, the container experiment used one electrode below the soil and one electrode on the soil surface, forcing current vertically through a smaller and more uniform soil column. This configuration increased electrode contact area and improved the likelihood that germinating seeds, tubers, roots, and emerging shoots intersected the effective electric field. The difference in energy range should also be considered: the container experiment included 200 J cm^-³, whereas the highest energy density evaluated in the field was 100 J cm^-³. Therefore, greater field suppression might have occurred at higher energy densities; however, increasing energy density in field soil would also increase energy demand and may reduce field-scale practicality unless energy delivery becomes more targeted and efficient. Thus, the container results demonstrate that weed propagules can be susceptible to soil-applied PEF when exposure is uniform, whereas the field results indicate that the vertical-pin configuration used in the orchard did not distribute energy uniformly enough to provide consistent weed control. This interpretation is consistent with electroporation concepts: when the electric field across cell membranes is sufficiently intense and applied for enough time, membrane function is disrupted, and tissues fail to recover (Kotnik et al., 2019; Cavalcanti et al., 2023).

Scalability remains a major limitation for soil-applied PEF in orchard systems. In the present study, energy treatments were expressed as nominal applied energy density, but real-time operational energy consumption, equipment efficiency, duty cycle, treatment speed, and field work rate were not measured. Therefore, field-scale energy requirements cannot be fully estimated from these data. Actual orchard energy demand would depend on treated band width, electrode geometry, soil moisture, soil electrical conductivity, contact efficiency, and the proportion of orchard floor treated. These limitations suggest that soil-applied PEF may be better suited for targeted orchard applications, such as localized treatment of persistent weed patches or belowground pest targets, rather than broadcast treatment of large soil volumes. Alternatively, future development may need to explore other electricity-delivery pathways, including foliar application of electrical energy, where direct contact with aboveground weed tissues may improve energy-use efficiency and biological response.

Weed susceptibility increased with incubation time, likely due to changes in electrical properties during early seedling development (Figure 5). As seeds imbibe water and tissues differentiate, dielectric constant and electrical conductivity increase, facilitating current flow through plant tissues. Conductivity in germinating seeds has been shown to increase several-fold as water content and tissue connectivity increase. For example, in seeds of Cardiospermum halicacabum L., conductivity remained low during early imbibition but increased several-fold (approximately 5–10×) between days 7 and 14 of germination as tissues developed and water content increased (Kurinobu and Okazaki, 1995). A similar response was reported in snap bean (Phaseolus vulgaris L.), where electrical impedance spectroscopy showed that increasing seed moisture content from approximately 25% to 50% resulted in substantial increases in conductivity (mS m^-¹) and corresponding reductions in electrical resistance (kΩ), indicating enhanced ionic conduction and tissue connectivity (Repo et al., 2002). In the present study, weed propagules incubated in the soil for 3 to 5 days were generally the most affected, aligning with the idea that developing seedlings provide more continuous conductive pathways for the electrical current. In contrast, ungerminated propagules (0-day) were less affected at lower energies, likely due to limited conductive pathways. Species-specific responses were also expected, as differences in seed size, morphology, and early growth patterns influence the formation of conductive pathways and recovery potential following injury (Hu et al., 2017; Gafni et al., 2025).

Figure 5

Overall, these results indicate that the primary limitation of soil-applied PEF for weed control is not necessarily the biological sensitivity of weeds, but the ability to deliver sufficient and uniform energy under field conditions. The container study demonstrates that growth suppression is achievable when exposure is uniform, whereas the orchard study shows that a single soil application with the vertical-pin configuration did not provide consistent weed control. Improving field efficacy will likely require application strategies that increase the probability of current passing through target tissues, including modified electrode designs, repeated applications to compensate for variability in contact, or integration with other nonchemical approaches (Baccin and Moretti, 2025a, 2026). The high level of crop safety observed also supports further exploration of soil-applied PEF for belowground targets. Although soil-applied PEF was ineffective for weed control under the orchard conditions tested, it may become a viable component of integrated weed management in perennial horticultural systems if energy delivery and biological targeting are better aligned.

References

• 1

  AsciS.DevadossS. (2021). “ Trends and issues relevant for the US tree nut sector,” in Choices (Milwaukee, WI: Agricultural & Applied Economics Association).

  • Google Scholar
  • View reference in article

• 2

  Avila-GarciaW. V.Sanchez-OlguinE.HultingA. G.Mallory-SmithC. (2012). Target-site mutation associated with glufosinate resistance in Italian ryegrass (Lolium perenne L. ssp. multiflorum). Pest Manag. Sci.68, 1248–1254. doi: 10.1002/ps.3286

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 3

  BaccinL.MorettiM. L. (2025a). Electrical weed control in organic highbush blueberry: influence of operational speed and number of applications on weed control. Front. Agron. 7. doi: 10.3389/fagro.2025.1688961

  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  BaccinL. C.MorettiM. L. (2025b). Yellow nutsedge (Cyperus esculentus) management with electrical weed control. HortScience60, 2382–2387. doi: 10.21273/hortsci18995-25

  • CrossRef
  • Google Scholar
  • View reference in article

• 5

  BaccinL.MorettiM. L. (2026). Cirsium arvense management with electrical weed control and clopyralid. Pest. Manage. Sci. 82, 6575–6584. doi: 10.1002/ps.70736

  • CrossRef
  • Google Scholar
  • View reference in article

• 6

  BazhalM.NgadiM.RaghavanG.NguyenD. (2003). Textural changes in apple tissue during pulsed electric field treatment. J. Food Sci.68, 249–253. doi: 10.1111/j.1365-2621.2003.tb14147.x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 7

  BenedettiT.Dos SantosP. M.ZasadaI. A.WeilandJ. E.CrispJ.MorettiM. L. (2026). Field strength and frequency-dependent effects of pulsed electric fields on soilborne nematodes, pathogens, and weeds. Sci. Rep. 16. doi: 10.1038/s41598-026-56820-y

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  BentoA.CaboP.MalheiroR. (2016). “ Nuts: agricultural and economic importance worldwide,” in Wild Plants, Mushrooms and Nuts: Functional Food Properties and Applications. Eds. FerreiraI. C. F. R.MoralesP.BarrosL. (Chichester, UK: John Wiley & Sons, Ltd.), 353–376. doi: 10.1002/9781118944653.ch10

  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  BloomerD. J.HarringtonK. C.GhanizadehH.JamesT. K. (2023). The electric spatula: Killing weeds with pulsed microshocks from a flat-plate electrode. Agronomy13, 2694. doi: 10.3390/agronomy13112694

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 10

  BobadillaL. K.HultingA. G.BerryP. A.MorettiM. L.Mallory-SmithC. (2021). Frequency, distribution, and ploidy diversity of herbicide-resistant Italian ryegrass (Lolium perenne spp. multiflorum) populations of western Oregon. Weed. Sci.69, 177–185. doi: 10.1017/wsc.2021.2

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 11

  BorgerC. P.SlavenM. J. (2025). Electric weed control—how does it compare to conventional weed control methods? Weed. Sci.73, e46. doi: 10.1017/wsc.2025.18

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 12

  BozorovE. (2020). Electric pulse treatment of trees as an environmentally friendly mechanism for protection of orchards. IOP Conf. Ser. Earth Environ. Sci.614, 012043. doi: 10.1088/1755-1315/614/1/012043

  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  Canadian Weed Science Society (2018). Description of 0-100 Rating Scale for Herbicide Efficacy and Crop Phytotoxicity. Available online at: https://weedscience.ca/cwss_scm-rating-scale/#:~:text=The%200%2D100%20scale%20is,and%20100%20%3D%20complete%20plant%20death.&text=The%20assessment%20of%20herbicide%20action,with%20an%20untreated%20control%20plot (Accessed March 10, 2026). [Online].

  • Google Scholar
  • View reference in article

• 14

  CavalcantiR. N.BalthazarC. F.MargalhoL. P.FreitasM. Q.Sant’AnaA. S.CruzA. G. (2023). Pulsed electric field-based technology for microbial inactivation in milk and dairy products. Curr. Opin. Food Sci.54, 101087. doi: 10.1016/j.cofs.2023.101087

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 15

  CorwinD. L.LeschS. M. (2005). Apparent soil electrical conductivity measurements in agriculture. Comput. Electron. Agric.46, 11–43. doi: 10.1016/j.compag.2004.10.005

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 16

  DağM. M.SaricaD. (2025). Global hazelnut export dynamics: assessment the role of environmental and economic factors with panel NARDL model. Appl. Fruit Sci.67, 302. doi: 10.1007/s10341-025-01536-y

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 17

  Demmig-AdamsB.AdamsW. W. (1996). The role of xanthophyll cycle carotenoids in the protection of photosynthesis. Trends Plant Sci.1, 21–26. doi: 10.1016/s1360-1385(96)80019-7

  • CrossRef
  • Google Scholar
  • View reference in article

• 18

  DiproseM. F.FletcherR.LongdenP. C.ChampionM. J. (1985). Use of electricity to control bolters in sugar-beet (Beta vulgaris L) - a comparison of the electrothermal with chemical and mechanical cutting methods. Weed. Res.25, 53–60. doi: 10.1111/j.1365-3180.1985.tb00617.x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  EvrendilekG. A.KaratasB.UzunerS.TanasovI. (2019). Design and effectiveness of pulsed electric fields towards seed disinfection. J. Sci. Food Agric.99, 3475–3480. doi: 10.1002/jsfa.9566

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 20

  Food and Agriculture Organization of the United Nations [FAO] (2026). FAOSTAT Statistical Database. Available online at: http://www.fao.org/faostat/en/#data/QC (Accessed September 1, 2026). [Online].

  • Google Scholar
  • View reference in article

• 21

  GafniR.BecharA.BakshianH.SmirnovE.RosenfeldL.LatiR. N. (2025). Shocking results: interspecific variation in response to low‐energy electrocution for weed control at various phenological stages. Pest. Manage. Sci.81, 6928–6939. doi: 10.1002/ps.70044

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  GarnerA. L. (2019). Pulsed electric field inactivation of microorganisms: from fundamental biophysics to synergistic treatments. Appl. Microbiol. Biotechnol.103, 7917–7929. doi: 10.1007/s00253-019-10067-y

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  GitelsonA. A.BuschmannC.LichtenthalerH. K. (1999). The chlorophyll fluorescence ratio F735/F700 as an accurate measure of the chlorophyll content in plants. Remote Sens. Environ.69, 296–302. doi: 10.1016/S0034-4257(99)00023-1

  • CrossRef
  • Google Scholar

• 24

  HeapI. M. (2026). International Herbicide-Resistant Weed Database. Available online at: http://www.weedscience.com (Accessed February 13, 2026). [Online].

  • Google Scholar
  • View reference in article

• 25

  HuX. W.ZhangR.WuY. P.BaskinC. C. (2017). Seedling tolerance to cotyledon removal varies with seed size: A case of five legume species. Ecol. Evol.7, 5948–5955. doi: 10.1002/ece3.3169

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  JeyamkondanS.JayasD.HolleyR. (1999). Pulsed electric field processing of foods: a review. J. Food Prot.62, 1088–1096. doi: 10.4315/0362-028x-62.9.1088

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  JinS.ZhangQ.TangB.ZhuX.LiuL.ZhangX. (2025). Lethal effect of pulsed electric fields on Tribolium castaneum: Optimization and mechanistic insight into electro-neurotoxicity. Agriculture16, 4. doi: 10.3390/agriculture16010004

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  Kaya-AltopE.HaghnamaK.SariaslanD.PhillippoC. J.MennanH.ZandstraB. H. (2016). Long-term perennial weed control strategies: Economic analyses and yield effect in hazelnut (Corylus avellana). Crop Prot.80, 7–14. doi: 10.1016/j.cropro.2015.10.022

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 29

  KotnikT.RemsL.TarekM.MiklavčičD. (2019). Membrane electroporation and electropermeabilization: mechanisms and models. Annu. Rev. Biophys.48, 63–91. doi: 10.1146/annurev-biophys-052118-115451

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 30

  KurinobuS.OkazakiY. (1995). “ Dielectric constant and conductivity of one seed in germination process”, in: Proceedings title: IAS ’95: Conference Record of the 1995 IEEE Industry Applications Conference, Thirtieth IAS Annual Meeting, (Piscataway, NJ, USA: IEEE Industry Applications Conference), 2, 1329–1334. doi: 10.1109/IAS.1995.530455

  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  KusakariS.OkadaK.ShibaoM.ToyodaH. (2020). High voltage electric fields have potential to create new physical pest control systems. Insects11, 447. doi: 10.3390/insects11070447

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  LebovkaN. I.PraporscicI.GhnimiS.VorobievE. (2005). Does electroporation occur during the ohmic heating of food? J. Food Sci.70, E308–E311. doi: 10.1111/j.1365-2621.2005.tb09969.x

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  LeongS. Y.BurrittD. J.OeyI. (2016). Electropriming of wheatgrass seeds using pulsed electric fields enhances antioxidant metabolism and the bioprotective capacity of wheatgrass shoots. Sci. Rep.6, 25306. doi: 10.1038/srep25306

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 34

  MennanH.BozogluM.BaserU.BrantsI.BelvauxX.Kaya-AltopE.et al. (2020). Impact analysis of potential glyphosate regulatory restrictions in the European Union on Turkish hazelnut production and economy. Weed. Sci.68, 223–231. doi: 10.1017/wsc.2020.10

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 35

  MennanH.NgouajioM.IsikD.KayaE. (2006). Effects of alternative management systems on weed populations in hazelnut (Corylus avellana L.). Crop Prot.25, 835–841. doi: 10.1016/j.cropro.2005.11.007

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 36

  MilbergP.HallgrenE. (2002). Highlighting differential control of weeds by management methods using an ordination technique. Weed. Technol.16, 675–679. doi: 10.1614/0890-037x(2002)016[0675:Hdcowb]2.0.Co;2

  • CrossRef
  • Google Scholar
  • View reference in article

• 37

  MirandaJ. W.GainesT. A.MorettiM. L. (2026). Poa annua becomes the first weed to evolve resistance to indaziflam applied preemergence and early‐postemergence. Pest. Manage. Sci.82, 539–551. doi: 10.1002/ps.70214

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 38

  MirandaJ. W.MorettiM. L. (2025). Herbicide resistance survey of annual bluegrass (Poa annua) in Oregon’s hazelnut production. Weed. Sci.73, 1–24. doi: 10.1017/wsc.2025.10064

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 39

  MorettiM. L. (2021). POST control of Italian ryegrass in hazelnut orchards. Weed. Technol.35, 638–643. doi: 10.1017/wet.2021.41

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 40

  MorettiM. L.BenedettiT.MedeirosP.CrispJ.McombJ.WeilandJ. E.et al. (2024). Watts up! Exploring soil-applied pulse electric field as an alternative to methyl bromide. Crops Soils57, 16–19. doi: 10.1002/crso.20373

  • CrossRef
  • Google Scholar
  • View reference in article

• 41

  OksanenJ.BlanchetF. G.KindtR.LegendreP.MinchinP. R.O’haraR.et al. (2013). Package ‘vegan’. Community Ecol. Package. Version.2, 1–295. doi: 10.32614/CRAN.package.vegan

  • CrossRef
  • Google Scholar

• 42

  PedrosoR. M.MorettiM. L. (2022). Hazelnut growth and weed control in response to selected preemergence herbicides. Weed. Technol.36, 1–6. doi: 10.1017/wet.2022.58

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 43

  RepoT.PaineD.TaylorA. (2002). Electrical impedance spectroscopy in relation to seed viability and moisture content in snap bean (Phaseolus vulgaris L.). Seed. Sci. Res.12, 17–29. doi: 10.1079/Ssr200194

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 44

  RhoadesJ. D.ManteghiN. A.ShouseP. J.AlvesW. J. (1989). Soil electrical conductivity and soil salinity: New formulations and calibrations. Soil Sci. Soc Am. J.53, 433–439. doi: 10.2136/sssaj1989.03615995005300020020x

  • CrossRef
  • Google Scholar
  • View reference in article

• 45

  RigaE.CrispJ. D.McCombG. J.WeilandJ. E.ZasadaI. A. (2020). Directed energy system technology for the control of soilborne fungal pathogens and plant-parasitic nematodes. Pest. Manage. Sci.76, 2072–2078. doi: 10.1002/ps.5745

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 46

  SaukH.BeyhanM. (2024). The effect of different orchard ground conditions on the performance of a mechanical hazelnut harvesting machine. Anadolu. Tarım. Bilimleri. Dergisi.39, 351–363. doi: 10.7161/omuanajas.1443157

  • CrossRef
  • Google Scholar
  • View reference in article

• 47

  SchreierH.BishM.BradleyK. W. (2022). The impact of electrocution treatments on weed control and weed seed viability in soybean. Weed. Technol.36, 481–489. doi: 10.1017/wet.2022.56

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 48

  TravlosI.GazoulisI.KanatasP.TsekouraA.ZannopoulosS.PapastylianouP. (2020). Key factors affecting weed seeds' germination, weed emergence, and their possible role for the efficacy of false seedbed technique as weed management practice. Front. Agron.2, 1. doi: 10.3389/fagro.2020.00001

  • CrossRef
  • Google Scholar
  • View reference in article

• 49

  VigneaultC.BenoîtD. L. (2001). “ Electrical weed control: theory and applications,” in Physical Control Methods in Plant Protection. Eds. VicentC.PannetonB.Fleurat-LessardF. ( Springer-Verlag Berlin Heidelberg, New York, NY, USA), 174–188.

  • Google Scholar
  • View reference in article

• 50

  YudaevI.DausY. V. (2023). Characteristics of changes in the lectrical conductivity properties of weed plant tissues under electropulse damage. Surf. Eng. Appl. Electrochem.59, 329–336. doi: 10.3103/S1068375523030171

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 51

  YudaevI.DausY.PanchenkoV.BolshevV. (2023). Influence of factors determining weeds’ plant tissue reaction to the electric pulse damage impact. Agriculture13, 1099. doi: 10.3390/agriculture13051099

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article