Introduction
Weedy flowers provide nectar and pollen rewards to pollinators in urban landscapes (Hicks et al. Reference Hicks, Ouvrard, Baldock, Baude, Goddard, Kunin, Mitschunas, Memmott, Morse, Nikolitsi, Osgathorpe, Potts, Robertson, Scott and Sinclair2016). An extensive study conducted by Lowenstein et al. (Reference Lowenstein and Matteson2019) for 3 yr showed that white clover (Trifolium repens L.) and Convolvulus spp. were the most visited by 24 pollinator morphotypes among 106 plant taxa including weedy and ornamental flowers. With the high likelihood of weedy flower–pollinator interactions in urban landscapes, the risk associated with a flower’s exposure to systemic insecticide treatments and transfer of toxic residues to pollinators also increases (Gels et al. Reference Gels, Held and Potter2002; Larson et al. Reference Larson, Redmond and Potter2013). Synthetic insecticides such as neonicotinoids and pyrethroids are widely utilized for turfgrass insect pest management, but also pose a significant threat to pollinators and pollination services (Desneux et al. Reference Desneux, Decourtye and Delpuech2007; Held and Potter Reference Held and Potter2012; Stanley et al. Reference Stanley, Garratt, Wickens, Potts and Raine2015).
Recent threats to biodiversity and attention to pollinator services are steering innovation to safeguard pollinators from harmful pesticides (Basu et al. Reference Basu, Ngo, Aizen, Garibaldi, Germmill-Herren, Imperatriz-Fonseca, Klein, Potts, Seymour and Vanbergen2024; Biesmeijer et al. Reference Biesmeijer, Roberts, Reemer, Ohlemuller, Edwards, Peeters, Schaffers, Potts, Kluekers, Thomas, Settele and Kunin2006; Vasiliev and Greenwood Reference Vasiliev and Greenwood2020). Larson et al. (Reference Larson, Dale, Held, McGraw, Richmond, Wickings and Williamson2017) recommended several best practices to minimize the negative effects of pesticides on pollinators. Best practices include mowing or applying herbicides to the treatment area to remove weedy flowers before insecticide application (Godara et al. Reference Godara, Williamson, Koo and Askew2023; Larson et al. Reference Larson, Redmond and Potter2015); improving insecticide formulations (Gels et al. Reference Gels, Held and Potter2002); and using safer insecticide chemistries (Larson et al. Reference Larson, Dale, Held, McGraw, Richmond, Wickings and Williamson2017). Floral features are crucial in attracting pollinators to flowers. UV patterns, while invisible to humans, impact pollinator foraging behavior on a variety of plant species (Koski and Ashman Reference Koski and Ashman2014; Papiorek et al. Reference Papiorek, Junker, Alves-Dos-Santos, Melo, Amaral-Neto, Sazime, Wolowski, Freitas and Lunau2016; Rae and Vamosi Reference Rae and Vamosi2013). Peitsch et al. (Reference Peitsch, Fietz, Hertel, de Souza, Ventura and Menzel1992) documented that hymenopteran insects have a trichromatic color vision system with photoreceptor peaks at 340 nm, 430 nm, and 535 nm. In a previous study, honeybee (Apis mellifera L., Hymenoptera: Apidae) evacuation from herbicide-treated turfgrass areas was affected by rapid changes in UV reflection on treated weedy flowers, but ultimately driven by a reduction in floral density (Godara and Askew Reference Godara and Askew2024). An underexplored potential contribution to pollinator foraging behavior is exogenous bioactive residues that may transiently alter floral UV reflectance and foraging cues on treated weedy flowers thus potentially reducing pollinator exposure via deterrence. Application of UV-absorbing cream to star lily (Hypoxis camerooniana Baker) flowers reduced pollinator visitation and also affected associated landing behavior (Klomberg et al. Reference Klomberg, Dywou Kouede, Bartos, Mertens, Tropek, Fokam and Janecek2019). Similarly, after the African potato (Hypoxis hemerocallidea Fisch. & C.A. Mey.) flower surface was coated with sunscreen composed of UV-absorbing compounds (Cole et al. Reference Cole and Shyr2015), East African lowland honeybee (Apis mellifera scutellata) visitation was reduced significantly (Johnson and Andersson Reference Johnson and Andersson2002).
Synthetic pigments and colorants are commonly used in turfgrass systems to improve aesthetic appearance and mitigate stress (Ervin et al. Reference Ervin, Zhang and Fike2004; Russell et al. Reference Russell, Karcher and Richardson2019). McCall et al. (Reference McCall, Sullivan, Zhang, Martin, Wong and Ervin2021) observed lower UV-A (315 to 400 nm) reflectance from creeping bentgrass (Agrostis stolonifera L.) turf treated with synthetic phthalocyanine pigment–containing products. Azoxystrobin reduced bumblebee (Bombus terrestris, Hymenoptera: Apidae) foraging performance and pollination services on lacy phacelia (Phacelia tanacetifolia Benth.) flowers by 16% and 26%, respectively (Tamburini et al. Reference Tamburini, Pereira-Peixoto, Borth, Lotz, Wintermantel, Allan, Dean, Schwarz, Knauer, Albrecht and Klein2021). The mechanism underlying these observed effects remains unclear due to limited research (Tamburini et al. Reference Tamburini, Pereira-Peixoto, Borth, Lotz, Wintermantel, Allan, Dean, Schwarz, Knauer, Albrecht and Klein2021), but exogenous bioactive residues could affect floral reflectance and contribute to reduced foraging performance of bumblebees. Fungicides, such as dodine, nitrothal-isopropyl, penconazole, and triadimefon, which have a repellency index of >88%, were utilized as honeybee repellent in apple (Malus domestica Borkh.) orchards (Solomon and Hooker Reference Solomon and Hooker1989). Several fungicides are currently formulated with synthetic pigments to enhance turf color, stress tolerance, and turf quality along with disease control (Anonymous 2025a; Fefer et al. Reference Fefer, Liu, Ruo and Hevia2013; Vandenberg et al. Reference Vandenberg, Rees and Hanrahan2015). Goulson et al. (Reference Goulson, Martinez, Hughes and Williams2000) observed that an optical brightener, Tinopal CBS, applied at 0.1% v/v reduced the insect visitation on common comfrey (Symphytum officinale L.) florets due to the ability of optical brightener to absorb incident UV radiation and emit back a blue portion of the visible spectrum. A premix of synthetic auxin herbicides and topramezone did not affect the UV reflectance of T. repens inflorescences but reduced the UV floral reflectance and reflecting area of dandelion (Taraxacum officinale F.H. Wigg.) and foxglove beardtongue (Penstemon digitalis Nutt. ex Sims) flowers (Godara and Askew Reference Godara and Askew2024). However, information is lacking on how fungicides, turf colorants, and optical brighteners influence UV floral patterns of turfgrass weeds and associated insect foraging.
The elucidation of how bioactive residues influence insect foraging, especially as transient deterrents, offers a unique opportunity to enhance pesticide formulations to protect pollinators from insecticide exposure without destroying long-term food resources. Because a few products have been shown to alter floral UV features, we hypothesized that pigmented fungicide, turf colorant, optical brightener, and sunscreen will alter the UV reflectance of weedy flowers in cool-season turf and could transiently affect insect visitation. Our objective was to evaluate the aforementioned product classes on the UV floral features of and pollinator visitation to plant species having UV-absorbing inflorescences, UV-reflecting petal apices with bullseye patterns, and UV-reflecting petals with contrasting reproductive parts.
Materials and Methods
These studies were conducted in 2023 at the Virginia Tech Glade Road Research Facility, Virginia Tech Turfgrass Research Center, and a private residence in Montgomery County, VA (Table 1). Three UV floral classifications were evaluated in the study: “UV-absorbing inflorescence,” represented by T. repens; “UV-reflecting at the petal apex with bullseye pattern,” represented by T. officinale and bulbous buttercup (Ranunculus bulbosus L.); and “UV-reflecting petals with contrasting reproductive parts,” represented by P. digitalis (Godara and Askew Reference Godara and Askew2023, Reference Godara and Askew2024). Bullseye pattern flowers exhibited over three times greater UV-reflectance intensity at the petal apex compared with the petal base (Godara and Askew Reference Godara and Askew2023). The study employed a randomized complete block design with a single factor (treatment), four replications, and two temporal runs for each floral classification as detailed in Table 1. Temporal runs for each floral classification were conducted at least 2 to 4 wk apart and were maintained at a buffer zone of at least 20 m between experimental sites (Table 1). Due to the use of resident weeds in managed turfgrass, each classification was studied separately. The floral density of P. digitalis was too low to assess insect foraging visits, so this species was evaluated only for the impact of exogenous residues on floral reflectance features (Table 1).
Table 1. Research experiments conducted in Blacksburg, VA, to evaluate the effect of optical brightener, pigmented fungicide, and turfgrass colorant on pollinator visitation and floral reflectance of three different ultraviolet (UV) floral classes of weeds.
a Abbreviations: GRRF, Glade Road Research Facility; TRC, Turfgrass Research Center; PR, Private Residence
b Penstemon digitalis flowers were only evaluated for the transient effect of exogenous residues on floral reflectance and UV-reflecting area in the studio. All others were assessed in both field and studio to study the impact on floral reflectance and insect foraging visits.
Plots measuring 3.2 m2, 3.2 m², and 6.5 m² were chosen for R. bulbosus, T. repens, and T. officinale, respectively, to achieve densities of 10, 30, and 5 flowers m⁻² at study initiation. The floral density for each weed species was based on preliminary research to ensure enough pollinator visitation at study initiation (data not shown). A 3-m buffer was maintained between plots, and a 6-m buffer between blocks, to prevent influences from adjacent areas on pollinator visitation. A comprehensive list of treatments, including active ingredients, trade names, manufacturer details, and application rates, is detailed in Table 2. All treatments were applied to weedy flowers at 8:00 AM, 1 d after experiment initiation (Tables 1 and 2). Applications were made using a CO2-pressurized backpack sprayer fitted with four extended-range flat-fan XR11006 nozzles (TeeJet® Technologies, Wheaton, IL, USA), set to deliver 468 L ha⁻¹ at 176 kPa, except for the positive control, which used a sunscreen applied with a spray atomizer. The conventional spray tips produced medium-sized droplets ranging from 226 to 325 µm (Anonymous 2025b), whereas the atomizer produced very fine droplets of 40 to 138 µm (Liu et al. Reference Liu, Rua, Wokovich, Guo and Keire2019), offering enhanced coverage and potentially altering floral UV reflection more significantly. The experiments were conducted for 16 d, started 1 d before treatment application, and concluded at 14 d after treatment (DAT). During this period, mowing and fertilization of the cool-season turfgrass was suspended to maintain consistent flower density and pollinator activity.
Table 2. List of treatments with active ingredients, trade names, manufacturers, and rates evaluated. a
a All treatments were applied through a conventional CO2-pressurized backpack sprayer delivering medium-sized droplets (226 to 325 µm), except for the atomized sunscreen treatment, which was applied with a spray atomizer delivering fine droplets (40 to 138 µm).
Three flowers were randomly collected from each plot at all experimental sites (Table 1) and placed in test tubes (Falcon™ 15-ml centrifuge tube, Corning, Tewksbury, MA, USA), each containing 5 ml of Miracle-Gro solution for fresh cut flowers (Scotts, Miracle-Gro, Marysville, OH, USA). These flowers were then transported to an indoor studio at Virginia Tech’s Glade Road Research Facility in Blacksburg, VA, where floral reflectance was evaluated using radiometry and the UV-reflecting area was quantified via UV photography at 4 h after treatment (HAT). Spectral reflectance data were collected using a QEPRO-XR spectrometer (Ocean Insight, Orlando, FL, USA), which measures wavelengths from 200 to 950 nm with a resolution of 1.6 nm (Figure 1). Illumination for these measurements was provided by a PX-2 Pulsed Xenon light source (Ocean Insight), emitting light from 200 to 750 nm (Figure 1), directed through a UV-sensitive fiber optic probe (QR400-7-SR, Ocean Insight). The probe was positioned at a 45° angle and 2 cm above the flower surface. Reflectance was measured at the petal apex, petal base, and reproductive structures of each flower (Camargo et al. Reference Camargo, Lunau, Batalha, Brings, de Brito and Morellato2019; Tunes et al. Reference Tunes, Camargo and Guimaraes2021).
Figure 1. An ultraviolet (UV)-sensitive fiber optic probe connected to a spectrometer and light source was used to collect floral reflectance data via radiometry (left). Normal photography (top) and UV photography (bottom) to quantify the UV-reflecting area of flowers. A WS-1 diffuse reflectance standard (A), a barium sulfate standard (B), a barium sulfate and charcoal mixture (C), and a charcoal standard (D) along with a standard ruler (E) and color standard were utilized.
For UV photography, a modified Canon EOS 5D Mark IV digital camera (Canon, Tokyo, Japan) was used, where the internal hot mirror was replaced with a UV band-pass filter (Life Pixel Infrared, Mukilteo, WA, USA). This camera was equipped with a vintage, non–UV coated Levnoc Helios 44-mm lens and connected to a remote digital monitoring screen (Shinobi 4K HDMI Monitor, Atomos, Carlton, Australia). UV images were taken under UV illumination from a UV light source (Everbeam®, Surrey, Canada) emitting 315 to 405 nm with a peak at 365 nm (Figure 1). Comparative floral images were also captured with a non-modified Canon EOS 5D Mark IV under white light (Craftsman®, Towson, MD, USA), with a light intensity range of 400 to 700 nm (Figure 1). Reflectance standards were used, including a WS-1 (Ocean Insight) diffuse reflectance standard, which reflects >98% of light ranging from 250 to 1,500 nm; a custom-made barium sulfate standard reflecting 100% of the UV light; a custom-made barium sulfate and charcoal mixture, uniformly reflecting 50% of the UV light; and a custom-made charcoal standard absorbing 100% UV light (Garcia et al. Reference Garcia, Greentree, Shrestha, Dorin and Dyer2014) (Figure 1). Additionally, a standard ruler and a color standard (DGK-Pro Multifunction Color Chart, DGK Color Tools, Boston, MA, USA) were used for scale and color calibration (Figure 1). Detailed methodologies are further discussed by Godara and Askew (Reference Godara and Askew2024).
Floral density, floral quality, and pollinator visitation data (excluding P. digitalis) were assessed at 1 d before and 0, 1, 2, 3, 4, 7, 10, and 14 DAT. Flower density was evaluated by manually counting all flowers in each plot. Digital images were captured for three representative flowers in each plot to measure the flower quality. Flower images from the field sites, UV images, and visible-light images of weedy flowers from the studio were subjected to batch processing using an object selection tool in Adobe Photoshop (Adobe, San Jose, CA, USA) to remove background elements. Floral images were then analyzed using Turf Analyzer (Green Research Services, Fayetteville, AR, USA) to quantify UV-reflecting areas based on hue and saturation thresholds steered by radiometric assessment of different floral parts in preliminary studies. Insect visitation was evaluated by counting unique foragers, which are physically interacting with flowers for 1 min in each plot three times each day (∼10:00 AM, ∼12:00 PM, and ∼02:00 PM). Flower quality and insect visitation data were expressed as a percentage of nontreated control within each block per evaluation timing. Treatment was considered a fixed effect, while experimental run and replication were treated as a random effect in the model statement. All response variables were analyzed using ANOVA with the PROC GLM procedure in SAS v. 9.3 (SAS Institute, Cary, NC, USA). The mean square of the treatment effect for all response variables was evaluated using the mean square of the random variable, with experimental run as the error term (MacIntosh Reference McIntosh1983). Data were presented separately by experimental run if significant run interactions occurred; otherwise, data were pooled over runs. Appropriate mean comparisons were conducted using Fisher’s protected LSD (α = 0.05). Microsoft Excel (Microsoft Office, Redmond, WA, USA) was used to fit the linear correlation between pollinator visitation and the UV reflectance of bullseye flowers.
Results and Discussion
Floral Reflectance of UV-reflecting Petals with Contrasting Reproductive Organs
The treatment effect was significant for UV-A (P = 0.0103) and UV-B reflectance (P = 0.0211) from reproductive parts of P. digitalis flowers at 4 HAT and not dependent on the experimental run (P > 0.05) (Table 3). Nontreated P. digitalis flowers reflected 3.9% and 1.6% of potential UV-A and UV-B, respectively, from reproductive parts based on the barium sulfate standard reflectance (Table 3). An optical brightener at both rates and sunscreen applied via atomizer reduced UV-A and UV-B reflectance of P. digitalis floral reproductive parts by at least 31% compared with the nontreated control (Table 3).
Table 3. Effect of exogenous residues on ultraviolet (UV) reflectance of flowers having UV-reflecting petals with contrasting reproductive organs assessed via radiometry and digital image analysis at 4 h after treatment.a,b
a Abbreviations: UV-A (315–400 nm), UV-B (280–315 nm), and UV-C (215–280 nm).
b Values followed by the same letter within each column are not different based on Fisher’s protected LSD (α = 0.05) and reflect the average of two trials (P-value of trial by treatment interaction is > 0.05) conducted in June and July of 2023 in Blacksburg, VA, that each included Penstemon digitalis flowers.
c Radiometric reflectance in the UV-C range from reproductive parts was not statistically significant
d Optical brightener was applied at 20 g L−1 rate.
e Optical brightener was applied at 30 g L−1 rate.
The treatment main effect was significant (P <0.0001) for UV-A, UV-B, and UV-C reflectance from P. digitalis petals at 4 HAT, and not dependent on experimental run (Table 3). Nontreated P. digitalis petals reflected 29%, 11%, and 8% of UV-A, UV-B, and UV-C, respectively (Table 3). In previous studies, Godara and Askew (Reference Godara and Askew2024) also documented 29% UV-A reflectance from nontreated P. digitalis petals. All treatments reduced UV-A, UV-B, and UV-C floral reflectance of P. digitalis petals (Table 3). Optical brightener and sunscreen treatments reduced the UV-A, UV-B, and UV-C reflectance of P. digitalis petals by at least 49%, 42%, and 46%, respectively, compared with nontreated petals (Table 3). Optical brightener reduced the UV reflectance of P. digitalis flowers at 4 HAT, regardless of the rate (Table 3). Optical brighteners absorb incident UV radiation ranging from 260 nm to 400 nm and emit back a visible spectrum ranging from 400 nm to 575 nm (Mustalish Reference Mustalish2000), responsible for reducing the UV reflectance of P. digitalis flowers. Sunscreen treatment applied via atomizer reduced the UV reflectance of P. digitalis petals by at least 80% at 4 HAT (Table 3). Rae and Vamosi (Reference Rae and Vamosi2013) also showed in field experiments that the UV-reflectance of yellow monkeyflower (Mimulus guttatus DC) was reduced by at least 80% after sunscreen treatment comprised of similar active ingredients as evaluated in our study.
The treatment main effect was significant (P = 0.0329) for digitally assessed UV-reflecting area of P. digitalis flowers at 4 HAT (Table 3) and not dependent on experimental runs. Nontreated P. digitalis flowers had a 0.86 cm2 UV-reflecting area (data not shown). Optical brightener at both evaluated rates and sunscreen reduced the digitally assessed UV-reflecting area of P. digitalis flowers to <57% at 4 HAT in agreement with radiometric data (Table 3). Sunscreen applied via atomizer and the high rate of optical brightener reduced the digitally assessed UV-reflecting area of P. digitalis to ≤35% (Table 3). Similar trends between digitally assessed UV-reflecting areas and radiometric data suggest that digital image analysis could be a viable cost-effective alternative for assessing exogenous residue effects on floral UV-reflecting features. Penstemon digitalis provides nectar and pollen to a wide range of pollinators (Clinebell and Bernhardt Reference Clinebell and Bernhardt1998) and plays a crucial role in pollinator-friendly landscapes (Anonymous 2024). It typically grows to a height of 76 cm, with inflorescences reaching 13 cm in length (Parachnowitsch and Kessler Reference Parachnowitsch and Kessler2010), and is generally found in pollinator gardens and urban areas. Although our experiments were focused on assessing the transient effect of treatments on UV floral reflectance, these results could have broader implications, as P. digitalis flowers are potentially vulnerable to insecticide drift in urban settings.
Floral Reflectance of UV-reflecting Petals with Bullseye Pattern
The treatment effect was significant for UV-A (P = 0.0227) and UV-B (P = 0.044) reflectance from the petal base of bullseye pattern flowers at 4 HAT and not dependent on experimental runs (Table 4). Only sunscreen treatments reduced the UV reflectance from the petal base, as UV-A and UV-B reflectance was ≤3.5% and ≤2.9%, respectively, compared with the 5% and 3.8% reflectance from the petal base of nontreated flowers (Figure 2; Table 4).
Table 4. Effect of exogenous residues on ultraviolet (UV) reflectance from flowers having UV-reflecting petals with bullseye pattern assessed via radiometry and digital image analysis at 4 h after treatment.a,b
a Abbreviations: UV-A (315–400 nm), UV-B (280–315 nm), and UV-C (215–280 nm).
b Values followed by the same letter within each column are not different based on Fisher’s protected LSD (α = 0.05). Radiometric data averaged over two trials (trial by treatment interaction is not significant, P-value > 0.05) conducted in Blacksburg, VA, in April and May of 2023, involved trials on T. officinale or R. bulbosus flowers. Digital image analysis presented separately due to trial by treatment interaction (P-value = 0.019).
c Radiometric reflectance in the UV-C range from reproductive parts was not significant
d Optical brightener was applied at 20 g L−1 rate.
e Optical brightener was applied at 30 g L−1 rate.
Figure 2. Impact of turf colorant, optical brightener, pigmented fungicide, and sunscreen on ultraviolet (UV) reflectance in Taraxacum officinale flowers. Left: UV-modified camera image under UV light, illustrating spray residue effects on UV reflectance. Right: Standard camera image under UV light, revealing visible optical brightener residues on the flower.
Similarly, the treatment effect was significant for UV-A (P = 0.0072), UV-B (P = 0.0006), and UV-C (P = 0.0025) reflectance from the petal apex of bullseye flowers at 4 HAT (Table 4). Turf colorant–treated bullseye flowers had equivalent or higher reflectance from petal apex and base at 4 HAT (Figure 2; Table 4). McCall et al. (Reference McCall, Sullivan, Zhang, Martin, Wong and Ervin2021) also observed increased UV-A reflectance of turf colorants in solidified water agar compared with nontreated. It was speculated that zinc oxide and titanium dioxide possibly reflect UV light and act as photoprotectors by redirecting the incident light away from the target. All treatments except turf colorant reduced the UV reflectance of the bullseye petal apex at 4 HAT (Figure 2; Table 4). Optical brightener reduced UV reflectance of the bullseye petal apex by at least 36% (Figure 2) compared with nontreated petals, regardless of the rate (Table 4). However, sunscreen applied through an atomizer was more effective in reducing the floral UV reflectance of bullseye flowers compared with sunscreen applied through a conventional sprayer (Figure 2; Table 4). Sunscreen applied via atomizer reduced the UV-A, UV-B, and UV-C reflectance by 90%, 86%, and 73%, respectively, compared with sunscreen treatment applied through a conventional sprayer, which only reduced reflectance by 56%, 53%, and 41%, respectively, from the petal apices (Figure 2; Table 4). The conventional backpack sprayer produced medium-sized droplets ranging from 226 to 325 µm, whereas the spray atomizer generated very fine droplets between 40 and 138 µm, resulting in improved product coverage after atomizer application (Anonymous 2025b; Liu et al. Reference Liu, Rua, Wokovich, Guo and Keire2019). Optical brightener at both rates and sunscreen applied through a conventional sprayer reduced the UV reflectance of the bullseye petal apex similarly (Figure 2; Table 4). Pigmented fungicide also reduced the UV-A, UV-B, and UV-C reflectance of the bullseye petal apex by 27%, 33%, and 24%, respectively (Figure 2; Table 4). Previous studies focused solely on UV-A concerning effects on insect foraging (Camargo et al. Reference Camargo, Lunau, Batalha, Brings, de Brito and Morellato2019; Peitsch et al. Reference Peitsch, Fietz, Hertel, de Souza, Ventura and Menzel1992), and our findings indicate that UV-B and UV-C responses to exogenous residues do not provide additional insights beyond those explained by UV-A (Tables 3 and 4).
For the UV-reflecting area assessed by digital image analysis, the treatment by experimental run was significant (P = 0.019), likely caused by differences in UV-reflecting areas of T. officinale and R. bulbosus flowers after treatment with pigmented fungicides (Table 4). Nontreated T. officinale and R. bulbosus flowers had 1.94 cm2 and 0.76 cm2 digitally assessed UV-reflecting areas, respectively (data not shown). Pigmented fungicide reduced the UV-reflecting area of T. officinale flowers to 21% but had a minimal impact on the UV-reflecting area of R. bulbosus flowers (Table 4). This disparity could be attributed to the epidermal cell morphology differences between these species, as the upper epidermis of R. bulbosus flowers acts as a strong light reflector toward the center of the floral reproductive region, potentially shadowing the effect of pigmented fungicide (van der Kooi et al. Reference van der Kooi, Elzenga, Dijksterhuis and Stavenga2017). All other treatments caused changes in floral UV reflectance that were consistent between species and generally in agreement with radiometric data (Table 4). Optical brightener treatments at both rates and sunscreen treatments reduced the digitally assessed UV-reflecting areas of T. officinale (Figure 2) and R. bulbosus to ≤47% and ≤54%, respectively (Table 4).
Floral Reflectance of UV-absorbing Inflorescences
The treatment main effect was significant for UV-A reflectance (P = 0.0183) from T. repens inflorescences (data not shown). Trifolium repens is classified as a UV absorber, as it reflects <3% of the incident UV (Godara and Askew Reference Godara and Askew2024). Trifolium repens inflorescences reflected <1.8% of UV-A after pigmented fungicide and atomized sunscreen treatment compared with 2.3% UV-A reflectance from nontreated flowers (data not shown). Pigmented fungicide, turf colorant, or atomized sunscreen is likely to cause a marginal reduction in UV-A reflectance of T. repens inflorescences due to exogenous residue deposition after treatment delivery.
Treatment Influence on Pollinator Visitation
The main effect of treatment significantly influenced pollinator visitation on bullseye flowers at 1 DAT (P = 0.0108) and 2 DAT (P = 0.0226) (Table 5). Taraxacum officinale flowers were mainly visited by honeybees and flies (Diptera: several families), whereas R. bulbosus flowers attracted solitary bees (Chelostoma florisomne, Hymenoptera: Megachilidae) and flies (data not shown). At 1 DAT, the application of an optical brightener at 20 g L−1 and 30 g L−1 reduced pollinator visitations to 64% and 40% of the nontreated control, respectively (Table 5). Similar reductions in pollinator visitation were noted by Goulson et al. (Reference Goulson, Martinez, Hughes and Williams2000), who observed a 65% decrease in visitation by three bumblebee species on S. officinale flowers at 1 HAT with a 0.1% v/v optical brightener. When sunscreen was applied via an atomizer, it reduced pollinator visitation to 10%, which was 3.4 times less than the same treatment applied using a conventional sprayer (Table 5). A study on blackeyed Susan (Rudbeckia hirta L.) also found a >90% reduction in pollinator visitation post–sunscreen treatment via atomizer at 1 HAT (Horth et al. Reference Horth, Campbell and Bray2014).
Table 5. Effect of treatments on pollinator visitation at 1 and 2 d after treatment (DAT) a
a Means within each column followed by the same letter are not significantly different based on Fisher’s protected LSD (α = 0.05). The experimental run by treatment effect was not significant (P-value > 0.05) for UV-reflecting at petal apex with bullseye pattern flowers (Taraxacum officinale and Ranunculus bulbosus) and UV-absorbing inflorescence (Trifolium repens), so data are combined by experimental runs for each UV floral classification
b Honeybees, solitary bees, and flies were primary visitors, >90% of total visitors on UV-reflecting at petal apex with bullseye pattern flowers.
c Honeybees, bumblebees, and solitary bees accounted for >95% of the total visitors to UV-absorbing inflorescence sites
The application of pigmented fungicide led to a transient reduction in pollinator visitation, with 19% and 15% fewer visits compared with nontreated flowers on 1 and 2 DAT, respectively (Table 5). A similar transient effect was observed by Tamburini et al. (Reference Tamburini, Pereira-Peixoto, Borth, Lotz, Wintermantel, Allan, Dean, Schwarz, Knauer, Albrecht and Klein2021), who reported a 16% decrease in bumblebee visitations on P. tanacetifolia after azoxystrobin application, akin to trifloxystrobin’s mode of action. At 2 DAT, pollinator visitation on flowers treated with 30 g L−1 optical brightener and sunscreen was ≤57% compared with nontreated flowers (Table 5). There is a positive correlation between pollinator visitation and UV reflectance of bullseye flowers at 1 DAT, where 61% of the variation in pollinator visitation could be explained by changes in UV reflectance (Figure 3). Pollinator visitation to flowers is influenced by multiple factors including flower color (Streisfeld and Kohn Reference Streisfeld and Kohn2007), nectar rewards (Pyke Reference Pyke2016), and scent emission (Dobson and Bergstrom Reference Dobson and Bergstrom2000), potentially driving foraging behavior separately from the effects of UV reflection.
Figure 3. Correlation between pollinator visitation and ultraviolet reflectance of flowers having ultraviolet (UV)-reflecting petals with a bullseye pattern at 1 d after treatment.
The treatment effect was significant for pollinator visitation on UV-absorbing inflorescent T. repens at 1 DAT (P = 0.0397) and 2 DAT (P = 0.0464) and was not dependent on the experimental run (Table 5). Honeybees, bumblebees, and solitary bees accounted for more than 95% of the total visitors to UV-absorbing inflorescences (data not shown). Turf colorant, pigmented fungicide, and atomized sunscreen slightly reduced pollinator visitations, yet foraging visits were at least 70% as frequent as nontreated inflorescences at 1 DAT and 80% as frequent at 2 DAT (Table 5). Other evaluated treatments did not affect bee visitation on T. repens inflorescences (Table 5). Pollinators also detect light radiation in the visible spectrum (Peitsch et al. Reference Peitsch, Fietz, Hertel, de Souza, Ventura and Menzel1992), and this may be an important factor for pollinators to locate floral rewards (Streisfeld and Kohn Reference Streisfeld and Kohn2007). Likewise, turf and T. repens leaves reflect UV radiation (McCall et al. Reference McCall, Sullivan, Zhang, Martin, Wong and Ervin2021) and our treatments may have similarly altered their appearance and appeal to pollinators in ways not measured in this experiment. After 2 d, turf colorant, optical brightener, and pigmented fungicide did not significantly affect pollinator visitations on bullseye or UV-absorbing flowers, indicating a transient effect (data not shown). Despite changes to floral UV-reflectance (Tables 3 and 4) and transient effects on pollinator foraging visits (Table 5), these treatments did not impact field-scale floral density and floral quality at any time during the assessment period for bullseye-classified species, T. officinale and R. bulbosus, or the UV-absorber classified as T. repens (data not shown).
In conclusion, treatments like application of optical brightener, pigmented fungicide, and sunscreen significantly altered the reflectance of the UV-reflecting flowers tested, and impacted associated pollinator visitations. The effects of these chemicals are most pronounced within 2 d after treatment, with UV reflectance and pollinator visits on UV-reflecting species potentially reduced up to 90% on the first day after treatment. These changes in UV reflectance correlate with pollinator behavior, explaining 61% of visitation variation. Sunscreen and optical brightener appear to show the most promising potential as candidates for transient pollinator deterrence without affecting the floral density of turfgrass weeds. The sustained floral density of T. officinale, R. bulbosus, and T. repens suggests that these products, when formulated with insecticides, can potentially deter pollinators from treated areas without eliminating food resources. Limiting the direct contact of pollinators with insecticide-treated plant tissue in lawns will ultimately reduce any potential negative effects on exposed individuals. Additionally, the enhanced UV suppression and pollinator deterrence by atomized sunscreen compared with sprayed sunscreen indicates potential avenues to further optimize insect deterrence for all products tested, using different application methods. However, the incomplete deterrence across all treatments suggests that additional factors also influence pollinator foraging behavior. Our findings add tools for transient pollinator deterrence without affecting floral density and associated food resources. This transient deterrence of pollinator visitations could be paired with applications of insecticide with low residuality on plant tissue, such as some pyrethroids (Carroll et al. Reference Carroll, Carson and Held2022). Research on extended deterrence of pollinator visitations in lawns warrants some further investigation.
Acknowledgments
The authors want to thank Jeremy Leichner, John Hinson, and the Virginia Tech Turfgrass Research Center staff for technical assistance. The authors also thank Alejandro Del Pozo-Valdivia (Virginia Tech, Entomology), Jacob Barney, and Michael Goatley (Virginia Tech, School of Plant and Environmental Sciences) for editing an earlier version of this article.
Funding statement
The authors want to thank the PBI Gordon Corporation for partially funding this study under the “2022 PBI-Gordon Turfgrass Pest Management Research Grant Program” and the Virginia Agricultural Council for partial funding.
Competing interests
The authors declare no conflicts of interest.
Open access
