Scirtothrips dorsalis (Hood) (Thysanoptera: Thripidae) is a recently identified invasive pest to the Caribbean and poses a significant threat to agriculture and trade in the region. Methods are needed to detect the presence and to monitor populations of this pest so that it can be effectively managed. Three different CC trap base colors (blue, yellow, and white) with or without dichlorvos as a killing agent, and a newly developed and named the Blue-D trap were studied in Taiwan and St. Vincent for attraction and capture of S. dorsalis. In lemons in Taiwan, mean numbers of S. dorsalis caught in Blue-D traps were greater compared with dichlorvos cube modified CC traps. In St. Vincent chili pepper plantings, the Blue-D traps caught more Thrips palmi (Karny), Frankliniella sp., and Microcephalothrips abdominalis (Crawford) than dichlorvos cube modified CC traps. More Frankliniella intonsa (Trybom), Megalurothrips usitatus (Bagnall), T. palmi, Frankliniella sp., and M. abdominalis were caught in blue and white base CC traps than yellow base CC traps. Average captures per CC trap per week were 0.07 and 0.02-0.09 S. dorsalis in Taiwan and St. Vincent, respectively. There were no differences in S. dorsalis captures in white, blue, or yellow base CC traps. The average weekly S. dorsalis catch for yellow sticky card traps was 19.8. CC traps can be used for detection of S. dorsali and collecting intact S. dorsalis for taxonomic and genetic determinations when a few of the species are found in a large commercial production area. Yellow sticky traps can be used for monitoring S. dorsalis populations. A combination detecting system of visual observation, yellow sticky traps, and CC traps may be an effective S. dorsalis population detecting and monitoring system.
Scirtothrips dorsalis (Hood) was described in 1916 as a new species collected from castor bean and chili plants in Coimbatore, Southern India (Hood 1919). S. dorsalis are polyphagous pests that are widespread in habitat ranging from temperate to tropical climate regions in Pakistan, Japan, and Australia (Mound & Palmer 1981). Primary hosts are onion, cashew nut, tea, chili, cotton, tomato, mango, tobacco, and castor bean. The insect has been reported by the Animal Plant Health Inspection Service (APHIS) as one of the thirteen most important pest species that could become a serious threat to United States (US) agricultural crops if it becomes established in the country (USDA-APHIS 2004). The Florida Nurserymen and Growers Association (FNGA) also consider S. dorsalis as an exotic pest with high potential to damage their industry if it becomes established in the state (FNGA 2003). Since 1984, USDA-APHIS inspectors at various US ports of entry have reported finding live S. dorsalis a total of 89 times from imported plant materials of 48 plant taxa (USDA 2003). On July 6, 2003, a Plant Protection and Quarantine (PPQ) officer in Miami, FL intercepted live S. dorsalis on chili peppers shipped from St. Vincent, a Caribbean island nation (Skarlinsky 2003). Subsequently, surveys of St. Vincent and St. Lucia confirmed the presence of S. dorsalis on both islands (Ciomperlik & Seal 2004). Thrips samples collected from both islands were catalogued and submitted to the USDA-Agricultural Research Service (ARS) Systematic Entomology Laboratory, Beltsville, MD, and the Australian Commonwealth Scientific & Industrial Research Organization (CSIRO) Entomology Laboratory, Canberra, Australia.
A pest risk assessment by Venette & Davis (2004) indicates that the potential geographic distribution of S. dorsalis in the U.S. ranges from the northeastern Atlantic area to Minnesota in the northern latitudes and to Texas in the south. Meissner et al. (2005) indicates that permanent establishment would likely be limited to southern and West Coast states. The species also seems capable of spreading throughout the entire Caribbean region. So far, it has become established on the Caribbean islands of St. Lucia, St. Vincent (Ciomperlik & Seal 2004), and Trinidad (USDA Offshore Pest Information System 2004).
Current survey, detection, and monitoring methods for S. dorsalis are laborious and require significant manpower and technical training. Most species in the genus Scirtothrips are pale in color, minute, and must be cleared and slide mounted for species identification. In addition, the genus is confused taxonomically. The names Heliothrips minutissimus (Bagnall), Anaphothrips andreae (Karny), Scirtothrips padmae (Ramakrishna), and S. fragaiae (Girault) appear in the literature but are now considered to be synonyms of S. dorsalis (Mound & Palmer 1981). Ongoing research and development methods that incorporate Polymerase Chain Reaction-Restriction Fragment Length Polymorphism (RFLP-PCR) genetic techniques (Toda & Komazaki 2002) are being conducted to explore the ribosomal ITS2 DNA regions that can be used to rapidly distinguish between thrips species. These methods require that individual insect samples be whole and undamaged, free of foreign contaminating substances, and preferably without contaminant DNA. Based on these requirements, sticky traps often used for thrips detection and monitoring are unsuitable.
To obtain specimens for taxonomic and genetic studies, terminal leaf samples can be collected in ziplock bags and washed with ethanol to obtain full intact specimens (Ciomperlik & Seal 2004). The plastic cup trap (named CC trap) (Fig. 1) also collects intact S. dorsalis specimens. It can be installed in a commercial production area where the pest has been found or suspected and serviced periodically for long periods of time. The CC trap was designed and validated for monitoring sweetpotato whitefly, Bemisia tabaci (Gennadius) B-biotype, populations. Trap design was based on the sweetpotato whitefly attraction to the color yellow, flight patterns approaching plants, and landing behavior (Chu et al. 1995; Chu & Henneberry 1998). The white base CC traps caught more S. dorsalis and Thrips palmi (Karny) in a peanut field in India in 1996 than yellow base traps (Chu et al. 2000). To extend the usefulness of the CC traps for detection and monitoring S. dorsalis, T. palmi, and other thrip species under Caribbean island conditions, we are studying different methods to increase trap efficacy.
White, yellow, and blue colors have been reported attractive to S. dorsalis, T. palmi, or F. occidentalis (Beavers et al. 1971; Gillespie & Vernon 1990; Cho et al. 1995; Tsuchiya et al. 1995; Chu et al. 2000; Hoddle et al. 2002; Chen et al. 2004). Chu et al. (2005) have modified CC whitefly traps for detecting and monitoring western flower thrips.
Objectives of the current study were (1) to evaluate CC thrip trap modification with a killing agent and a specimen preservative, (2) to evaluate a modification of a commercially available dichlorvos strip package for use as a thrips trap, named the Blue-D trap in this report, and (3) to evaluate sticky card traps.
Materials and Methods
Comparison of CC and Blue-D traps
The study was conducted in commercial farms with randomized complete block designs with 15 and five replicates in Taiwan and St. Vincent, respectively. Treatments were re-randomized weekly in St. Vincent, but not in Taiwan. The four CC trap treatments in Taiwan were trap base colors (white, blue, and yellow with a 1 cm2 dichlorvos cube and 15 ml, 10% ethylene glycol) and the Blue-D trap. In St.Vincent, the three trap base colors with or without dichlorvos cube and with or without ethylene glycol made a total of 12 treatments. Blue-D trap was the 13th treatment. The dichlorvos cubes in CC traps were not replaced during the experimental periods. The ethylene glycol treatment was used to preserve thrip specimens. The CC traps were serviced weekly, disassembled in the field, rinsed with 20 ml 85% ethanol to dislodge insects that were retained in the trap bases, and stored in labeled glass vials for identification in the laboratory.
The Blue-D traps were Hot Shot® No-Pest® Strip dispensers (United Industry Corp., St. Louis, Mo.) fitted with plastic bags attached to the dispenser bottoms for collecting dead insects. Vertically oriented blue plastic strips (two 2.5-cm wide strips spaced at 2.5 cm apart) were attached inside the front and back surfaces of the dichlorvos dispensers packages (Fig. 2). The Blue-D traps were replaced weekly.
Experiment 1. Taiwan (dry season). The experiment was conducted in a 0.6-ha lemon (Citrus limon L., cv. Eureka) orchard for eight weeks from 26 November 2004 to 30 January 2005 in Neipu, Taiwan. The traps were suspended in trees 1.9 m above the ground.
Experiment 2. St. Vincent (wet season). The experiment was conducted about one mile inland from Caribbean sea in Georgetown, St. Vincent Island in two (∼0.2 ha each) geographically separated chili pepper (Capsicum chinense L.) fields. Plants were set 1 m within and between row spacings in both fields. Plants in one field were Scotch Bonnet variety and in the other field the West Indies Red variety. Traps were placed 1 m apart on wooden stakes placed within the plant rows. The trap bases were located about 22 cm below the tallest plant terminals. The experiment was conducted for six weeks from 14 October to 29 November, 2004.
Experiment 3. St. Vincent (dry season). The experiment was conducted in two Scotch Bonnet variety (∼0.2 ha each) fields, as described in Experiment 2. The experiment was conducted for six weeks from 23 March to 4 May, 2005.
All thrips in the samples were identified and counted. Scirtothrips dorsalis was readily separated from the other species by the small size (0.7-1.0 mm), pale yellow color, and the presence of microtrichia extending along the median area of the abdominal sternites (Hoddle & Mound 2003). The remaining species were separated from each other by characters found in Nakahara's key to Thripidae (unpublished). Representative individuals of each thrips species were slide mounted to confirm the species identity. Voucher specimens from St. Vincent were deposited in the USDA-ARS Systematic Entomology Laboratory and in the St. Vincent Ministry of Agriculture and Fisheries.
Seasonal Weather during the Wet and Dry Seasons in St. Vincent, 2004-2005
Daily rainfall data and air temperatures during both seasons were obtained from weather station records housed at Rabacca Field Station in Georgetown, St. Vincent. The weather station was located about 0.4 km from the experimental chili pepper fields. CC trap captures were examined in relation to average rainfall and temperature in the wet and dry seasons.
Comparison of Sticky Card Trap Colors (Dry Season)
Experiment 4. The experiment was conducted during the dry season in 2005 in the same commercial pepper fields in St. Vincent described for Experiment 3. The thrips trap captures compared sticky card traps with the CC and Blue-D trap captures. The sticky card traps were 10.0 by 10.5 cm in size. White sticky card traps were constructed by coating both sides with brush-on Tanglefoot® formula (Tanglefoot Co., Grand Rapids, MI). Blue sticky card traps of the same dimensions were Takitrap® obtained commercially (Oecos Ltd., Kimpton, Hertfordshire, England). Yellow sticky card traps also of the same dimensions were custom made commercially (Olson Products, Medina, OH). The sticky card traps were placed in chili pepper rows 1 m apart. Traps were oriented vertically with wire loops attached to 25 cm long wooden stakes. Traps were placed 5-10 cm above the plant terminals.
Statistical Analysis
Numbers of thrips caught were averaged over sampling periods for each experiment. Data were analyzed by t-tests, ANOVA orthogonal comparisons, or three factor factorial analysis (Anonymous 1989). Means were separated by Tukey's HSD.
Results
Experiment 1. Comparison of CC and Blue-D traps in Taiwan (dry season). Blue-D traps caught more S. dorsalis than the CC traps (F = 4.8, df = 1, 58, P = 0.034). Mean captures for F. intonsa or T. hawaiiensis (Morgan) for the two trap types were not statistically different (Table 1). Catches of S. dorsalis by the three different color CC trap bases were not significantly different. White base CC traps caught more T. hawaiiensis (F = 14.5, df = 2, 28, P = <0.001), and blue base CC traps caught more M. usitatus (F = 49.0, df = 2, 28, P = <0.001) compared with other trap base colors. White and blue base CC traps caught more F. intonsa (F = 7.3, df = 2, 28, P = 0.028) compared with yellow base CC traps. The four thrips species are considered economic pests in Taiwan (Chang 1995).
Experiment 2. Comparison of CC and Blue-D traps in St. Vincent (wet season). Blue-D traps caught more T. palmi, Frankliniella sp., and M. abdominalis than CC traps (F = 42.4 – 99.2, df = 1, 58, P = <0.001). Mean captures of S. dorsalis for the two trap types were not significantly different. Captures of S. dorsalis in CC traps for different trap base colors were not significantly different. Blue and white base CC traps caught more T. palmi and Frankliniella sp. (F = 4.1 and 5.3, df = 2, 44, P = 0.024 and 0.009, respectively) than yellow base CC traps. Blue base CC traps caught more M. abdominalis than white and yellow base CC traps (F = 99.2, df = 2, 44, P = 0.003). Dichlorvos cubes in CC traps increased captures of all thrips species (F = 6.0 – 20.6, df = 1, 44, P = 0.018 – <0.001) except S. dorsalis. Ethylene glycol in the CC trap bases had no effect on trap catch numbers. However, thrips captured in ethylene glycol equipped traps were well preserved, with less damage to antennae and less desiccation than traps without ethylene glycol.
There were significant treatment interactions for CC trap base colors and dichlorvos treatments for T. palmi trap catches (F = 6.6, df = 2, 44, P = 0.003), but not for the three other thrips species (Table 2). Thrips captures for the ethylene glycol treatment, or the interactions with CC trap base color and dichlorvos were not significantly different.
Frankliniella sp. captures were identified by Systematic Entomology Laboratory as Frankliniella cephalica (Crawford) and Frankliniella insularis (Franklin).
Experiment 3. Comparison of CC and Blue-D traps in St. Vincent (dry season). Mean numbers of T. palmi, Frankliniella sp., and M. abdominalis caught in Blue-D traps were greater compared with CC traps (F = 8.2 – 10.4, df = 1, 58, P = 0.002 – 0.006). Mean captures of S. dorsali were not significantly different for the two trap types (Table 3). dichlorvos cubes in CC traps increased captures of all four thrips species (F = 8.6 – 72.2, df = 1, 44, P = 0.005 – <0.001). On average, blue and white base CC traps caught more T. palmi, Frankliniella sp., and M. abdominalis than yellow base CC traps (F = 9.5 – 16.7, df = 2, 44, P = 0.002 – <0.001). The addition of ethylene glycol in traps resulted in greater captures of S. dorsalis, Frankliniella sp., and M. abdominalis (F = 7.4 – 10.5, df = 1, 44, P = 0.009 – 0.002), than the mean captures of T. palmi. Blue and white base CC traps with ethylene glycol caught more Frankliniella sp. (F = 4.0, df = 2, 44, P = 0.024), but not the other three thrips species. There were no significant differences in thrips captures for other treatment interactions.
Seasonal Weather Effects on CC Trap Catches during the Wet and Dry Seasons in St. Vincent, 2004-2005
CC trap captures were 4.5 and 6.1 fold greater for S. dorsalis and T. palmi, respectively, during the dry season compared with the wet season. Rainfall averaged 1.1 mm per day during the dry season and 18.0 mm per day during the wet season (Table 4). Air temperatures in the dry season was 1.3°C higher compared with the wet season.
Experiment 4. Comparison of Sticky Card Trap Colors (Dry Season). Significantly more S. dorsalis were caught on yellow sticky card traps compared with blue sticky card traps. Yellow and blue sticky card traps caught more S. dorsalis than white sticky card traps (Table 5). More T. palmi, Frankliniella sp. and M. abdominalis were caught on blue sticky card traps compared with white and yellow sticky card traps.
Discussion
Blue-D traps caught more S. dorsalis than the CC traps in Taiwan, but approximately equal numbers in St. Vincent. Similarly, Blue-D traps consistently caught more of the other three thrips species in the study compared with the CC trap. Overall, the addition of dichlorvos cubes increased CC trap captures of S. dorsalis in St. Vincent. The blue base CC trap with dichlorvos cubes caught more S. dorsalis than the other treatment combinations only during the dry season in St. Vincent. Unfortunately, previous trap studies conducted in India on S. dorsalis did not include blue base CC trap comparisons with white and yellow base traps (Chu et al. 2000). The addition of ethylene glycol to CC traps increased trap catches of S. dorsalis and M. abdominalis during the dry season in St. Vincent. The addition of ethylene glycol resulted in better preserved specimens for taxonomic and genetic studies.
Numbers of S. dorsalis captured in CC traps with dichlorvos were low in both Taiwan and St. Vincent. We reported earlier that blue sticky card traps caught more F. occidentalis in a broccoli field than yellow sticky card traps (Chen et al. 2004). Blue sticky card traps also captured greater numbers of T. palmi, Frankliniella sp., and M. abdominalis than yellow or white traps in the current studies. Our results from St. Vincent indicate that yellow sticky card traps were more attractive to S. dorsalis than white or blue sticky card traps. Similarly, Hoddle et al. (2003) reported that Scirtothrips perseae (Nakahara) was more attracted to yellow than white or blue sticky card traps.
The Blue-D trap did not consistently capture greater numbers of S. dorsalis than CC traps. Its potential toxicity in the environment is of concern. Although the CC trap captures fewer S. dorsalis, the quality of the captured specimens is high. They are easily recovered from the trap and stored in ethanol for later taxonomic and genetic analysis. Yellow sticky traps capture more thrips than the CC traps but they also capture a large number of non-target insects. In addition, thrips that are captured on the sticky trap are not easily removed and stored for later studies. Sticky traps seem to be less labor intensive, require less component assembly and therefore less expertise in trap placement than the CC traps.
Seal et al. (2005) have determined from direct plant sampling that economic damage to chili peppers by S. dorsalis occurs at densities of 0.5 to 2 individuals (larvae or adults) per terminal leaf. This sampling method requires nine samples per 24-48 m2 area in order to achieve the 90% precision level. This method may be too labor intensive to use in large scale survey and detection efforts. Alternatively, visual observation for plant damage symptoms like curled, deformed, or yellow leaves coupled with placing sticky card traps can be utilized as a preliminary detection tool. Positive detections would then be followed by direct plant sampling to capture large number of individual specimens for taxonomic verification. APHIS guidelines for survey are based on the principle of finding one S. dorsalis in a suspected area. The guidelines suggest 2,280 CC traps for initial survey that will be placed in one square mile areas for detecting S. dorsalis (USDA-APHIS, 2004). If one or more S. dorsalis is found the second phase of survey the survey area will be expanded to the eight surrounding square miles. The presence of a single S. dorsalis in the second survey will lead to an expansion of the survey area to 80 surrounding square miles for the third phase of survey. Results of our studies estimate that CC traps would catch 46 and 205 for wet and dry seasons, respectively. These would translate to the captures of 46 and 205 S. dorsalis in CC traps for the initial survey, 148 and 666 for the second survey, and 1069 and 4813 for third phase of survey during wet and dry seasons, respectively, in St. Vincent.
Current methods employed for detection of S. dorsalis are inefficient as demonstrated in the present report. Studies of S. dorsalis behavior, including the development of attractants and pheromones as potential lures, are being conducted to develop more efficient trap systems for detection and monitoring of this insect pest.
During preparation of this manuscript, S. dorsalis was detected on roses in Palm Beach, FL and in multiple retail garden centers on hot pepper seedlings by the Florida Department of Agriculture and Consumer Services (FDACS) (Wayne Dixon, pers. comm.). Further surveys of retail garden centers in the Lower Rio Grande Valley of Texas likewise revealed the presence of this new invasive species on hot pepper seedlings (M. Ciomperlik, unpublished data). Specimens were confirmed as Scirtothrips dorsalis by the USDA ARS Systematic Entomology Laboratory. The current distribution of S. dorsalis in the Caribbean is limited to a few islands. It has recently invaded the US, and is expected to spread over time through agricultural trade and tourism (Venette & Davis 2004; Meissner et al. 2005). These observations also indicate an alarming potential for rapid spread of this pest thrip species through interstate movement of ornamentals and plant seedlings in the US. The potential impact of this thrips on agriculture in the United States alone has been estimated at approximately $3.6 to $6.0 billion a year (Lynn Garrett, USDA APHIS PPQ CPHST, pers. comm.). Effective survey and detection methods are needed to monitor the spread, and manage populations, of S. dorsalis both in the Caribbean and the US.
Acknowledgments
The authors thank Tsung Tao Hsieh of National Pingtung University of Science and Technology in Taiwan for technical assistance during the studies. In addition, we thank Jason Carlson and Juan Rodriguez (USDA, APHIS) for technical support. We also thank Laurence Mound (CSIRO) and Steve Nakahara (USDA, ARS, SEL) for identification of thrips specimens.
References Cited
Table 1.
Seasonal means (±SE) thrips caught in a lemon orchard in Neipu, Taiwan, dry season, 26 November 2004 to 30 January 2005 (Experiment 1).
Table 2.
Seasonal means (±SE) of thrips caught in traps in commercial chili pepper fields, Georgetown, St. Vincent, wet season, 14 October to 29 November (Experiment 2)
Table 3.
Seasonal means (±SE) of thrips caught in commercial chili pepper fields, Georgetown, St. Vincent, dry season, 23 March to 4 May 2005 (Experiment 3).
Table 4.
Seasonal mean (±SE) rainfall, air temperature, and CC trap captures of thrips species during the wet and dry seasons in St. Vincent, 2004-2005.
Table 5.
Seasonal means (±SE) of thrips caught on white, blue, and yellow sticky card traps in commercial chili pepper fields, Georgetown, St. Vincent, dry season, 23 March to 4 May 2005 (Experiment 4).
