Current Pest Status and Management Practices for Systena frontalis (Coleoptera: Chrysomelidae) in Ornamental Plants in the Eastern United States: An Online Survey


 Systena frontalis (F.) (Coleoptera: Chrysomelidae) is an important pest in ornamental plant nurseries in the United States. Information on current pest status and management practices employed by nurseries, garden centers, and landscape care operations are crucial to developing an effective research and extension program for ornamental crops. An online survey was developed and administered by the research team in 2020 to gather data on pest status and current pest management practices. The questionnaire included three focus areas: 1) participant (location, industry type, or operating area); 2) pest status indicators (incidence, affected crops, and estimated loss); and 3) common pest management practices. The questionnaire was distributed to stakeholders via e-mail lists, newsletters, and blogs. Seventy-five responses were received; 82.6% of which were from wholesale nursery operators in 19 U.S. states and 1 Canadian province. For most respondents (72%), damage recurred yearly in the past 10 yr and persisted from April to October. About 56% of respondents reported damage on more than five host plant species (with Hydrangea spp. being the most frequently identified) representing approximately 25% of the total number of plants grown in the facilities. Presence of S. frontalis is being monitored mainly through visual inspection of foliage for adult presence or foliage damage (100%), with scouting occurring mainly at weekly intervals (57%). The majority of respondents used broad-spectrum insecticides (such as pyrethroids) for adult (89%) and larval control (47%). We estimated that a grower spends USD$1,637/ha/yr on insecticides and labor for monitoring and implementing S. frontalis management.

Systena frontalis (F.) (Coleoptera: Chrysomelidae) is a serious insect pest in outdoor, containerized production of ornamental plants (Mahr 2005, Lauderdale 2017, Cloyd and Herrick 2018, Joseph and Hudson 2020. Systena frontalis is a native species of central and eastern North America (Riley et al. 2003, Mahr 2005, Lauderdale 2017, Cloyd and Herrick 2018, Joseph and Hudson 2020. It is referred to as the redheaded flea beetle (e.g., Mahr 2005) or the cranberry flea beetle (e.g., Dittl 1988) (not ESAapproved common names), reflecting its pest status on ornamental plants and cranberry (Vaccinium oxycoccos L.) crops (Dittl 1988, Mahr 2005, Averill and Sylvia 2011, Maine Cooperative Extension [MCE] 2020. It is also documented as a pest of potato (Solanum tuberosum L.) (Storch et al. 1979), corn (Zea mays L.) Barton 1969, Jacques andPeters 1971), and blueberry (Vaccinium L.) (Maltais and Ouellette 2000). Adults skeletonize upper and/ or lower leaf surfaces (Fig. 1); such damage appears rapidly and can be severe, depending on the density of adults. Moderate-tosevere foliar damage often renders ornamental plants unmarketable. Larvae feed on roots, but their impacts on plant growth and appearance are unknown.
Systena frontalis adults are small (3.5-6.25 mm in length), shiny, oval-shaped, with metallic, black-colored bodies, and red-tinted heads ( Fig. 1A; Mahr 2005, MCE 2020). Females are slightly larger than males. A pair of light brown serrate antennae, with dark brown distal segments, originates below the eyes (Joseph and Hudson 2020). The last pair of femora is swollen, which enables S. frontalis to jump (Mahr 2005). Mature females lay pale-yellow eggs singly in soil or substrate (Mahr 2005). Larvae are approximately 0.5-1 cm in length, creamy-white, and with pale brown head capsule and legs. The last larval abdominal segment is oriented upward with a prominent seta (Mahr 2005). Larvae develop through three instars before pupating in the soil.
Systena frontalis overwinters as eggs in containerized nurseries (Lauderdale 2017, Herrick andCloyd 2020), but the overwintering stage has not been confirmed in other ecosystems. Adult feeding injury typically is first observed in late April or early May in Georgia (Chong, unpublished data); however, timing may fluctuate based on local weather conditions. Surveys in North Carolina found first-generation larvae to emerge between 250 and 480 GDD 50 (Growing Degree Day with a base temperature of 50°F or 10°C) using 1 January as the biofix date (Lauderdale 2017). There is only one generation in Maine (MCE 2020). All life stages may be found on the same containerized plant in the summer (Joseph and Hudson 2020). Plant phenological indicator for larval development varies among geographical locations (Lauderdale 2017).
Despite S. frontalis' economic impact on ornamental plant nurseries, there is limited knowledge on its phenology and habits on ornamental plant species in the United States. Adoption of integrated pest management by growers and landscape maintenance operators has not been assessed. We document the current pest status and management practices to provide the basis for development of research and extension priorities in the areas of economic impact and integrated pest management.

Materials and Methods
A survey questionnaire was developed to collect information on the pest status and management of S. frontalis in the ornamental industry, which include wholesale and retail nurseries, retail garden centers, and landscape installation and maintenance companies (Table 1). In recent years, university researchers and extension personnel had received queries about S. frontalis mainly from ornamental plant nurseries; it was not clear if this insect is considered a pest in garden centers and commercial and residential landscapes. Hence, some questions in the survey were tailored to obtain specific information from garden centers and landscape installation and care operations. The questionnaire was developed and distributed through e-mail lists, newsletters, and blogs to the target respondents in their respective states by a group of collaborating researchers and extension personnel from Cornell University, University of Delaware, University of Maryland, Virginia Polytechnic Institute and State University, North Carolina State University, Clemson University, University of Tennessee, University of Georgia, Auburn University, Louisiana State University, University of Florida, and Texas A&M University.
Questions were organized into three sections. The first group of questions gathered information on participant characteristics, including the type, location, and size of the respondents' operations. No personal or demographic information of the respondents was collected; hence, this questionnaire was reviewed and exempted from approval by the Institutional Review Board at the University of Georgia (approval number PROJECT00002513). The second group of questions focused on the prevalence, damage, and host plants affected by S. frontalis. The third group of questions focused on the current monitoring and management practices conducted by various operations. The questionnaire was converted into an online survey tool developed by Qualtrics (Provo, UT) under subscription purchased by the University of Georgia. If the respondents selected the option that they never had a S. frontalis issue in their facility, the survey ended at that point. The respondents could select multiple options for certain questions (Table 1) specific instructions were provided in parentheses. Similarly, if the respondents selected the option that they never used insecticides for S. frontalis control in their facility, the survey ended at that point. A couple of questions (17 and 18) included an option for respondents to type an answer(s) or leave a comment. The questionnaire was distributed via e-mails, newsletters, and extension publications from the participating institutions and trade organizations representing the green industry. The newsletters are distributed beyond state lines and reach a broad audience. An estimated 23,437 entities, including wholesale and retail nurseries (owners, growers, and managers), retail garden centers, landscape installation and maintenance companies, greenhouse growers or managers, university extension agents, state department of agriculture employees, and private pesticide license holders were contacted. Survey responses received via Qualtrics between 24 June and 2 September 2020 were included in this analysis.

Participant Type and Facilities
Seventy-five entities or respondents from 19 U.S. states and 1 Canadian province responded to the survey (Fig. 2). All responses (except for one response each from the states of Minnesota and Washington, and Ontario, Canada) came from the eastern United States, with respondents from North Carolina (11 counties), Georgia (7), Tennessee (7), New Jersey (6), and Maryland (3) submitted the greatest numbers of responses. Some counties had multiple responses, such as Wake (six respondents), Johnston (four), Wayne (three), and New Hanover (three) Counties in North Carolina, Baltimore County (four) in Maryland, Cumberland County (three) in New Jersey, and McDuffie County (three) in Georgia. Participant response rates to specific questions are listed in Table 1.
The respondents (n = 75) identified themselves as a grower/manager/owner, representing a wholesale nursery (82.6%), retail nursery (17.3%), retail garden center, and landscape installation (17.3%), and landscape maintenance company (12%) involved in making pest management decisions at their respective operations (Fig. 3A). A total of 13 respondents were involved in multiple industry sectors, such as both retail garden centers and landscaping companies, both retail garden centers and wholesale nurseries, or all three industry sectors. These respondents were categorized as a wholesale nursery if this sector was part of the operation; if not, they were considered retail nurseries. Of the 61 wholesale nursery respondents, 44.3% operated > 20.3 ha, and 26.2% operated between 8.2 and 20.2 ha ( Table 2). Among the 13 retail nursery respondents, 30.7% of respondents had 0.4-2.0 ha, 30.7% had 8.2-20.2 ha, and approximately 7.7% had > 20.3 ha under operation. Of the nine landscape maintenance respondents, 66.7% had between 0.4 and 2.0 ha of the service area. A total of 11 retail garden centers participated in the survey; 72.2% of them were independent centers operating from one location (Fig. 3B). Fifty-five respondents provided information on their annual sales in the last couple of years, of which 47% indicated Are there any restrictions for you to use neonicotinoids (such as imidacloprid, i.e., Merit, or dinotefuran, i.e., Safari)? 72 a This questionnaire was focused on redheaded flea beetle in the ornamental industry. The respondents included farm managers and growers associated with nurseries, garden centers and landscape installation and maintenance companies. One participant did not provide location information. Individuals involved in academia or agrochemical companies did not participate in the survey.

b
Respondents could check more than one choice.

Systena frontalis Pest Status
Among 72 responses to the question about the frequency of infestation, 72% found recurrent yearly infestation in the past 10 yr, 14% reported infestation every 2-3 yr, and 14% reported no infestation at all (Fig.  5A). Forty-eight respondents who reported yearly infestation (92% of this group) were exclusively wholesale producers or had wholesale nurseries as part of their operations. Respondents who had never experienced an infestation of this pest operated wholesale nurseries (seven responses or 11% of responses from this sector) and landscape maintenance companies (three responses or 33%) in Georgia (two responses or 22% of responses from this state), New Jersey (one response or 11%), North Carolina (two responses or 8%), Pennsylvania (one response or 50%), and Tennessee (four responses or 57%).
Sixty-one respondents provided information on host plants attacked by S. frontalis (most likely based on observations of adult feeding damage on leaves); 56% observed attacks on more than five host plant species (Fig. 5B). More than half of the respondents suffered 10-25% damage to their crops (Fig. 5C). Respondents documented attacks on 36 plant genera (Table 3). The top five plant hosts were Hydrangea spp., Itea virginica, Weigela spp., Ilex spp., and Rosa spp. (Table 3). Few respondents responded to the question about crop losses. Three respondents reported estimated crop losses that ranged from $501 to $5,000 per infestation, and two respondents reported plant replacement costs of < $500-$1,000 per infestation (Table 1).

Monitoring and Management
Among 58 respondents of this group of questions, 62% indicated that they have adequate knowledge of the biology and behavior of S. frontalis to manage this pest. There are, however, differences in perceived inadequacy of knowledge among groups of respondents. Small operations with sales < $100,000 and those between $500,000 and $1 million generally did not feel they had adequate knowledge (Fig. 4B). Even larger operators had only slightly higher confidence in their knowledge level, with approximately 40% of respondents with sales > $2 million indicated inadequate knowledge. Of the 51 wholesale nursery respondents, 33% of them expressed inadequate knowledge of S. frontalis biology (Fig. 4C). Moreover, approximately 50% of the respondents (mostly wholesale nurseries) who expressed inadequate knowledge suffered approximately 10-25% damage to plants (Fig. 5D).
Fifty-six respondents provided multiple responses to the questions on monitoring (Table 1; questions 15 and 17). Respondents observed adult beetle activity as early as February-March (3% of 118 responses) but most observed activity from late May to August (> 60%) and persisted into September and October (35%) (Fig. 6). All respondents monitored beetle activity visually, but some (13 respondents) also employed other monitoring techniques, including canopy shaking, sticky card, using growing degree-day model to predict adult emergence, and root ball dissection to sample for larvae (Fig. 7A). Most respondents conducted weekly scouting (56% of 57 respondents), but 1 respondent who reported infestation never monitored beetle activity (Fig. 7B).
Fifty-eight respondents provided 88 responses to the multiplechoice question on management tactics (Table 1; question 18). All except 3 of 58 respondents managed adults and/or larvae (Fig. 8A). Application of insecticides against adults is the major management approach against S. frontalis (89% of 88 responses), followed by larval control (47%) and both adult and larval control (48%) (Fig.  8A). A small subset of respondents used entomopathogenic nematodes (11% of 88 responses) and entomopathogenic fungi (2%) for larval control. The most frequently used insecticides are the neonicotinoids (36% of 88 responses), followed by carbaryl (23%; indicated in 'others'), pyrethroids (21%), organophosphates (15%), and diamides (6%) (Fig. 8B). When asked if they have an adequate number of effective insecticides to manage S. frontalis, 41 respondents responded, and 54% of them expressed a need for more effective insecticides, whereas the rest were satisfied with currently available insecticides.
Thirty-five (66%) respondents reported no restriction on the use of neonicotinoids for S. frontalis control in their operations (Fig. 9). However, approximately 11 and 15% of respondents refrained from using neonicotinoids because of the perceived risk of consumer dissatisfaction or the company's no-neonicotinoid-use policy, respectively. These no-neonicotinoid respondents were almost exclusively large wholesale nurseries (sales > $1 million).

Cost of S. frontalis Management
Fifty-four respondents reported spending < $100-$500,000 per year on S. frontalis management. Among these respondents, 22 and 24% spent approximately $100-$1,000 and $2,000-$5,000 per year, respectively (Fig. 10). Among 18 respondents with > $2 million in sales, 61% of them expressed that they did not have adequate insecticides available for S. frontalis control (Fig. 11A). Operations with > $1.5 million in sales tended to spend greater amounts on S. frontalis control (Fig. 11B). By dividing the amount spent on management (including insecticides and monitoring and application labor costs) by the operation area reported for each respondent, we estimated that the 54 respondents spent an average of $1,637/ha annually for the management of S. frontalis.

Discussion
Our survey results show that S. frontalis is a serious and persistent pest for some of the most commonly grown and sold plant taxa    Average sales (USD) in last two years

Respondents (%)
Average sales (USD) in last two years Industry type in nurseries and garden centers across the eastern United States.
Most of the respondents represented wholesale nurseries, several of which had over 20.3 ha in production and $2 million in sales. Respondents reported that S. frontalis infestation was previously detected in wholesale and retail nurseries, as well as retail garden centers. Operations that include both nursery/garden center and landscape installation/maintenance reported S. frontalis infestations, but none of the operations that perform only landscape installation/maintenance indicated an issue with the pest. This result has two important implications. First, S. frontalis may not be a pest in ornamental landscapes; thus, future research and extension efforts should focus on nurseries. Since this conclusion is based on small sample size (n = 8), comprehensive state-level surveys of S. frontalis prevalence and economic impacts are needed to corroborate this survey's findings. Second, unidentified factors (environment factors, production or maintenance practices, etc.) have prevented the dispersal, establishment, and damage of S. frontalis in ornamental landscapes. Recent S. frontalis outbreaks in cranberry fields of the Northeast (MCE 2020) suggested that this species is capable of building damaging populations on established hardy plants in the fields. There are some unidentified differences in environment and production/maintenance practices among nurseries, landscapes, and fields that may lead to different   . 7. Percentage of survey respondents responding to (A) monitoring tactic(s) deployed to determine S. frontalis incidence (n = 80) and (B) the monitoring interval (n = 57) in their operations. The respondents had the opportunity to respond to more than one tactic used for scouting S. frontalis. *Sticky cards on any color; **damage or adult beetles; ***light traps with LED, blacklight, UV light, or other techniques; and ****other tactics used include monitoring growing degree days for S. frontalis and scouting for S. frontalis larvae in the root balls of containers.
pest status of S. frontalis among these systems. An understanding of the underlying mechanisms is crucial to the development of a sustainable and effective management program for S. frontalis in ornamental plant and cranberry production.
Systena frontalis is a polyphagous pest. This survey resulted in a list of 39 host plant taxa (Table 3), with respondents identified species of Hydrangea. Itea, Weigela, Ilex, Rosa, and Rhododendron being the most frequently damaged plant taxa. Lauderdale (2017) reported that S. frontalis fed upon anise tree (Illiucium spp.), goldenrod (Solidago spp.), joe pye weed (Eutrochium spp.), and zinnia (Zinnia spp.), but our survey respondents did not identify these species as hosts. Adult S. frontalis was reported to attack weeds growing in and around nurseries, such as jewelweed (Impatiens  (Lauderdale 2017;Chong, unpublished data). This wide host range suggests that S. frontalis can feed on virtually any broadleaf woody or herbaceous plant growing inside or around nurseries, and presents a significant challenge to the management of S. frontalis. Management of weed species that may serve as potential hosts of S. frontalis in and around nurseries may be important in mitigating adult dispersal and damage in nurseries.   insecticides that are used in their operations (n = 53). *Entomopathogenic nematodes; **entomopathogenic fungi; ***pyriproxyfen to reduce larvae; and ****seven respondents who indicated the use of carbaryl, and one respondent used pyriproxyfen. The respondents had the opportunity to respond to more than one control tactic for S. frontalis in their facility.
The majority of respondents monitored adult S. frontalis activity by visually inspecting for beetle presence or foliage damage. Some respondents complemented visual inspection with other monitoring tactics, such as deploying sticky cards and shaking the canopy to dislodge the adults then quantify them. Although visual inspection and sticky cards are routinely used to monitor adult flea beetles such as Phyllotreta cruciferae Goeze and Phyllotreta striolata (Fabricius) in canola (Brassica napus L.) or Blepharida rhois (Forster), Chaetocnema denticulata (Illiger), Epitrix subcrinita (Leconte), Systena blanda (Melsheimer), and many others in vegetables (Bunn et al. 2015, Knodel et al. 2017, sticky cards and shaking the canopy may not be reliable monitoring tactics for S. frontalis (Alabama Cooperative Extension System [ACES] 2020) because these tactics only detect beetles that are already present on the plants. In fact, visual inspection detected beetle presence and damage 1 mo before the first beetles were captured on sticky cards (Chong, unpublished data). Visual inspection for beetle presence and/or damage may not be the most effective monitoring method. Tolerance for S. frontalis feeding damage is essentially zero in ornamental plant production system where crops are valued for their esthetics. Therefore, other monitoring tools and tactics that can provide advance warnings of beetle presence or potential damage are needed. Some survey respondents used growing degree-day model to determine beetle activity (Kunkel and Colon 2013), but other monitoring tools or tactics are not currently available (such as light or semiochemical traps) or not commonly used (such as plant phenological indicators). The survey results indicate that the development of reliable monitoring or predictive tools for S. frontalis based on hostderived volatiles, pheromones, or visual cues is warranted. Before these tools become available, the majority of survey respondents scouted for adult activity weekly or even daily in order to provide sufficient advance information prior to damage becoming unacceptable. The monitoring should be conducted on the most preferred ornamental plant and weed species starting in early to mid-May.
Survey respondents invested a large sum of money for the management of S. frontalis, with operations generally spending more on pesticides, equipment, and labor for scouting and application. A wholesale nursery often spends $30,000 USD or more per year on direct costs related to scouting and application of insecticides (B. Jernigan, personal communication). Respondents used a wide range of (mainly contact) insecticides, with neonicotinoids, pyrethroids, organophosphates, and carbamate being the most commonly used. Because S. frontalis is a problem throughout the growing season from April to October (reported in this survey) and that currently available contact insecticides typically lack long-term residual activity against this pest, repeated foliar applications of insecticides are essential to suppress adult populations and prevent damage (Lauderdale 2017, ACES 2020, Joseph and Hudson 2020. Spray coverage of the entire canopy is critical against the highly mobile beetles Herrick 2018, Joseph andHudson 2020). Research suggests that insecticides applied to foliage and potting soil reduced the adult S. frontalis feeding damage on foliage (Herrick and Cloyd 2020). Insecticides commonly used by the respondents are broad-spectrum (organophosphates, carbamate, and pyrethroids) and can indirectly affect nontarget organisms, such as beneficial arthropods and pollinators. Over half of the respondents did not think they have sufficiently effective long-residual insecticides to manage S. frontalis for the entire growing season. This outcome suggests that more research on new active ingredients with prolonged residual activity and minimal effects on the environment and nontarget organisms is warranted. The survey suggests that more respondents targeted the adults than larvae with management sprays.
Most of the respondents (66%) indicated that they do not have any reservations about using neonicotinoids for S. frontalis control. The respondents who reported discontinued use of neonicotinoids due to company policy or customer preference were almost exclusively large wholesale growers, who likely supply their products to major national retail outlets that had committed to reduced or discontinued the use of neonicotinoids (Friends of the Earth 2017). Growers who have discontinued the use the neonicotinoids are actively seeking alternative management options, including the use of biological control agents, for S. frontalis and other pests (Joseph 2020).
We estimated that the respondents spent an average of $1,637/ ha/yr for the management of S. frontalis. This amount includes the costs of pesticides, labor, and equipment. We were unable to estimate the cost of crop losses because few respondents responded to questions about crop loss in this survey. The poor response rate is the result of a unique combination of pest impacts and the nursery business model. Even 1 or 2 d of feeding by S. frontalis can create unsightly foliar damage that renders a plant unmarketable in most instances. These damaged plants were often held back from sales, pruned, and held until the new flush of foliage develops. These plants are sold once they recover from the damage; thus, growers often do not consider and report these plants as completely 'lost' to the pest. This perception, however, masked the true cost of S. frontalis infestation. Growers incur increased opportunity costs when they remove the damaged plants from their availability list and wait for the plants to recover. The increased labor, space, and time inputs increase the production cost of the recovered plants by 10% of the original selling price (B. Jernigan, personal communication). This increased production cost is often not accounted for in the survey as a measurement of crop loss since the recovered plants are eventually sold. Future surveys of pest damage and cost of control should also include this opportunity cost in their analysis of a pest's impact in any agricultural sector.
In summary, the survey respondents, mostly wholesale growers, indicated that the S. frontalis infestation and damage recur annually in the eastern United States, with beetle activities most prevalent and damaging on some of the most commonly grown plant taxa from April through October. The respondents mostly monitor the adults by visual inspection at weekly intervals. This suggests that monitoring can be developed with a better understanding of the hostuse pattern, spatial and temporal movement through seasons, and semiochemical and other cues that can be used in traps. Because of a near-zero threshold for S. frontalis feeding damage, the narrow marketing window of ornamental plants, and the persistence of adult S. frontalis throughout the growing season, respondents apply broad-spectrum insecticides several times and spend a substantial amount for S. frontalis adult and larval control. The management of S. frontalis could be improved with timely insecticide application targeting larvae or adults. To implement, timely applications require an enhanced understanding of S. frontalis biology, dispersal, behavior, and ecology. A refined and sensitive monitoring tool could help improve the timeliness of insecticide application. Development of cost-effective and reliable nonchemical options for S. frontalis control such as entomopathogenic nematode and fungi is warranted. via e-mails and newsletters. We also thank B. Jernigan (McCorkle Nurseries, Dearing, GA) for in-depth discussions on redheaded flea beetle management strategies, production practices, and economics in the nursery industry. An IPM Working Group Grant from the Southern Integrated Pest Management Center (grant number 2018320016 to SVJ and JHC) provided supports for the online survey and the formation of a S. frontalis working group, of which all coauthors of this manuscript belong. All coauthors were involved in the design and administration of the survey; SVJ, JHC, and BC curated and analyzed data; SVJ and JHC prepared the original draft of this report; and all coauthors edited this report.