1 MANA GEMENT STRATEGIES TO INCREASE EFFECTIVENESS OF INSECTICIDES USED TO MANAGE ASIAN CITRUS PSYLLID, DIAPHORINA CITRI KUWAYAMA, IN CITRUS By CHRISTINE ELIZABETH WEAVER A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVE RSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 201 3
2 2013 Christine Elizabeth Weaver
3 To my parents
4 ACKNOWLEDGMENTS Funding for t his research was provided by Citrus Research and Development Foundation (CRDF). I thank my advisor and graduate committee chair, Dr. Michael Rogers, for his advice, guidance, and support. I also thank my other committee members, Dr. Lukasz Stelinski and Dr. Megh Singh for lending their expertise and advice. I thank Dr. Timothy Ebert for his assistance with my statistical analysis. I thank the members of the lab, Rhonda Schumann, Harry Anderson, Percivia Mariner, Guoping Liu, Emily Collins, and Ki Duk Ki m for their help with field work, data collection, and acquisition of ACP. I also thank Greg Sapp and company for providing the equipment and labor needed to complete the low volume application portion of my research. I thank my mother, father, stepmothe r and grandparents for their love, support, and encouragement.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ 4 LIST OF TABLES ................................ ................................ ................................ ........... 7 LIST OF FIGURES ................................ ................................ ................................ ........ 8 LIST OF ABBREVIATIONS ................................ ................................ .......................... 10 ABSTRACT ................................ ................................ ................................ .................. 11 CHAPTER 1 LITERATURE REVIEW ................................ ................................ ......................... 13 Introduction ................................ ................................ ................................ ............ 13 Asian Citrus Psyllid and Huanglongbing ................................ ................................ 13 Insect Life Cycle and Description ................................ ................................ .... 14 Insect Dispersal ................................ ................................ ............................... 14 Host Plants of Asian Citrus Psyllid ................................ ................................ .. 16 Current Managemen t Strategies for Asian Citrus Psyllid ................................ ........ 17 Biological Control ................................ ................................ ............................ 17 Chemical Control ................................ ................................ ............................. 20 Factors Affecting Pesticide Efficacy ................................ ................................ ....... 23 Application Metho ds ................................ ................................ ........................ 23 Use of Adjuvants ................................ ................................ ............................. 28 Environmental Conditions ................................ ................................ ............... 33 Justification ................................ ................................ ................................ ............ 34 Goals and Hypotheses ................................ ................................ .......................... 35 Specific Objectives ................................ ................................ ................................ 35 2 SEASONAL EFFECTS ON INSECTICIDE APPLICATIONS FOR D. CITRI CONTROL ................................ ................................ ................................ ............. 36 Introduction ................................ ................................ ................................ ............ 36 Materials and Methods ................................ ................................ .......................... 38 Ins ecticide Applications ................................ ................................ ................... 38 Caging of Adult D. citri ................................ ................................ ..................... 39 Statistical Analyses ................................ ................................ ......................... 40 Results ................................ ................................ ................................ .................. 41 Discussion ................................ ................................ ................................ ............. 44 3 ADDITION OF SURFACTANTS TO INSECTICIDES USED TO MANAGE D. CITRI ................................ ................................ ................................ ..................... 58
6 Introduction ................................ ................................ ................................ ............ 58 Materials and Methods ................................ ................................ .......................... 60 Insecticide Application ................................ ................................ ..................... 60 Caging of Adult D. citri ................................ ................................ ..................... 60 Statistical Analyses ................................ ................................ ......................... 61 Results ................................ ................................ ................................ .................. 61 Discussion ................................ ................................ ................................ ............. 62 4 COMPARISON OF LOW VOLUME APPLICATION OF INS ECTICIDES TO STANDARD AIRBLAST APPLICATION OF INSECTICIDES TO MANAGE D. CITRI. ................................ ................................ ................................ .................... 68 Introduction ................................ ................................ ................................ ............ 68 Materials and Methods ................................ ................................ .......................... 70 Insecticide Applications ................................ ................................ ................... 70 Caging of Adult D. citri ................................ ................................ ..................... 71 Statistical Analyses ................................ ................................ ......................... 72 Results ................................ ................................ ................................ .................. 73 Discussion ................................ ................................ ................................ ............. 74 LIST OF REFERENCES ................................ ................................ .............................. 78 BIOGRAPHICAL SKETCH ................................ ................................ ........................... 85
7 LIST OF TABLES Table page 2 1 Weather data collected during caging periods. Compiled from Florida automate d w e ather n etwork (FAWN). ................................ ................................ 47 2 2 ANOVA results for adult D.citri mortality for insecticide treatments testing seasonal effects on insecticide applications for D.citri control ............................ 48 2 3 ANOVA results for comparing different approaches to the analysis of the insecticide effects ................................ ................................ .............................. 48 3 1 ANOVA results for adult D.citri mortali ty for insecticide treatments testing addition of surfactants to insecticides used to manage D. citri ........................... 64 4 1 ANOVA results for adult D.citri mortality for insecticide treatments using low volu me vs airblast application equipment applied during October and December ................................ ................................ ................................ ......... 76
8 LIST OF FIGURES Figure page 2 1 Mean ( SE) percent mortality of D. ci tri adults caged on trees each week following an application of insecticides in January 2012. Within each sampling date, means with the same letter do not dif fer significantly ............... 51 2 2 Mean ( SE) percent mortality of D. citri adults caged on trees each week following an application of insecticides in April 2012. Within each sampling date, means with the same letter do not dif fer significantly ............................... 51 2 3 Mean ( SE) percent mortality of D. citri adults caged on trees each week following an application of insecticides in September 2012. Within each sampling date, means with the same letter do not differ significant ly ................. 53 2 4 Mean ( SE) percent mortality of D. citri adults caged on trees each week following an application of insecticides in November 2012. Within each sampling date, means with the same letter do not dif fer sig nificantly ............... 54 2 5 Mean ( SE) percent mortality of D. citri adults caged on trees each week following an application of insecticides in January 2013. Within each sampling date, means with th e same letter do not dif fer significantly ................. 54 2 6 Rainfall recorded during the caging period 1 day after treatment for each month. Compiled from Florida automated weather network (FAWN). ................ 55 2 7 Mean ( SE) percent mortality of adult D. citri caged on imidacloprid treated trees 1 DAT for each month trials were conducted. ................................ ........... 55 2 8 Mean ( SE) percent mortality of adult D. citri caged on dimethoate treated trees 1 DAT for each month trials were conducted. ................................ ........... 56 2 9 Mean ( SE) percent mortality of adult D. citri cag ed on fenpropathrin treated trees 1 DAT for each month trials were conducted. ................................ ........... 56 2 10 Temperatures recorded during the first week psyllids were caged on insecticide treated and untreated trees each month. Weather data compiled from the Florida automated weather network. ................................ .................... 57 3 1 Mean ( SE) percent mortality of adult D. citri caged on trees 1 day after treatment. Means with the same HSD, P > 0.05). ................................ ................................ ................................ 64 3 2 Mean ( SE) percent mortality of adult D. citri caged on trees 8 days after treatment. Means with the same letter do not diff HSD, P > 0.05). ................................ ................................ ................................ 65
9 3 3 Mean ( SE) percent mortality of adult D. citri caged on trees 15 days after treatment. Means with the same letter do not differ significantly HSD, P > 0.05). ................................ ................................ ................................ 66 3 4 Mean ( SE) percent mortality of adult D. citri caged on trees 22 days after HSD, P > 0.05). ................................ ................................ ................................ 67 4 1 Mean ( SE) percent mortality of adult D. citri caged on trees 1, 5, and 8 DAT with insecticides using low volume vs airblast applicatio n equipment applied in October. ................................ ................................ ................................ ......... 76 4 2 Mean ( SE) percent mortality of adult D. citri caged on trees 1, 4, and 7 DAT with insecticides using low volume vs airblast application equipment appl ied in December. ................................ ................................ ................................ ..... 77
10 LIST OF ABBREVIATIONS ACP Asian citrus psyllid CHMA Citrus Health Management Areas COC Crop oil concentrate DAT Days after treatment EPA Environmental Protection Agency GPA Gallons per acre HLB Huanglongbing HMO Horticultural mineral oil IRAC Insecticide Resistance Action Committee Las Candidatus Liberibacter asiaticus LD Lethal dose LLR Lowest label rate LV Low volume MSO Methylated seed oil PTO Power take off ULV Ultra low volume VOC Vegetable oil concentrate
11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science MANAGEMENT STRATEGIES TO INCREASE EFFECTIVENESS OF INSECTICIDES USED TO MANAGE ASIAN CITRUS PSYLLID, DIAPHORINA CITRI KUWAYAMA, IN CITRUS By Christine Elizabeth Weaver August 2013 Chair: Michael Rogers Major: Entomology and Nematology Asian citrus psyllid (ACP), Diaphorina citri Kuwayama is a known vector of Candidatus Liberibacter asiaticus, which is the putat ive causal agent of citrus greening disease, also known as huanglongbing (HLB). Insecticides are commonly used for D. citri control but the duration of residual control provided is poorly understood. Thus, field trials were conducted to determine whether there are seasonal effects on insecticides used for D. citri control. Regardless of time of year when applied fenpropathrin did not differ from dimethoate (organophosphate) or imidacloprid (neonicotinoid) in initial knock down of adult ACP caged on treate d trees. There were statistically significant differences in the residual activity provided by insecticides that were likely affected by other factors such as rainfall. However, the residual activity present beyond 1 week after application was far lower t han would be considered acceptable control of D. citri The addition of various surfactants did not provide any observable benefit in terms of extending the residual activity of insecticides examined. Comparison of the residual activity of three insecticid es using both low volume and airblast equipment showed no difference in residual activity within 4 DAT. Overall, these studies suggest that the true
12 duration of D. citri control provided by foliar applications of insecticides is much shorter in duration th an previously described.
13 CHAPTER 1 LITERATURE REVIEW Introduction The United States has over 330,000 ha planted to citrus, with Florida leading the citrus producing states with over 260,000 ha (USDA NASS 2011, USDA NASS 2009b). The citrus industry is ver y important to Florida and many other states, including California, Texas, and Arizona (USDA NASS 2009b). The United States citrus industry had a market value of over 3 billion dollars in 2007 (USDA NASS 2009a). Because of the importance of the citrus in dustry to the United States and the state of Florida, it is vital to manage insect and disease problems in an economical and environmentally safe way. Asian Citrus Psyllid and Huanglongbing The Asian Citrus Psyllid, Diaphorina citri Ku wa yama (Hemiptera: Ps yllidae) was first described in Taiwan in 1907 ( Halbert and Manjunath 2004 ) Currently, D. citri is the most important arthropod pest of citrus in Florida due to its status as the vector of citrus huanglongbing. Citrus huanglongbing (HLB), also known as citrus greening disease, is a very damaging disease to citrus trees. HLB is associated with the putative causal agent, Candidatus Liberibacter asiaticus (Las). Las is a bacterial pathogen that is phloem limited, gram negative, and has yet to be cultured on artificial media (hence the Candidatus designation). There are three species of bacteria that are associated with citrus greening: Candidatus Liberibacter asiaticus (Asian species), Candidatus Liberibacter africanus (African species), and Candidatus Li beribacter americanus (South American species). Candidatus Liberibacter asiaticus is the species that affects the United States, Asia, and South America. Candidatus Liberibacter americanus is
14 known only to affect Brazil (Teixeira et al. 2005). Candidatu s Liberibacter africanus is the species that occurs in Africa and is transmitted by Trioza erytreae (del Guercio) (Jagoueix et al. 1994). Symptoms of HLB include yellow/chlorotic mottling of leaves, twig dieback, corked veins, and the eventual death of th e tree. HLB can cause fruit to be misshapen, not properly colored (remain green), bitter tasting, and drop prematurely. HLB will eventually cause the premature death of the tree (da Graca 1991). Insect Life Cycle and Description The Asian citrus psyllid life cycle includes the following stages: egg, five nymphal stages and adult. Females are capable of laying 800 eggs in a lifetime, and only oviposit on young developing citrus leaves, often referred to as new flush yellow when they are laid and measure approximately 0.3 mm, and will turn orange as they mature before hatching (Childers and Rogers 2005). Eclosion occurs 2 4 d after eggs are laid. The average time to complete n ymphal development is 11 15 d but can take as much as 47 d dependi ng on temperature (Liu and Tsai 2000). The nymphal instars measure between 0.25 to 1.7 mm in length, and adults measure 3 4 mm in length (Childers and Rogers 2005). Both males and females are reproductively mature 2 3 d after reaching adulthood (Wenninge r and Hall 2007). In order to achieve maximum reproductive output, females need to be mated multiple times throughout their lives (Wenninger and Hall 2008). Females typically begin to oviposit 1 d after mating (Wenninger and Hall 2007). Insect Dispersal Adult D. citri have been shown to actively disperse between groves (Boina et al. 2009) They move bi directionally between managed and unmanaged groves, but net movement is from unmanaged to managed groves. Unmanaged or abandoned groves
15 are those that d o not take any measures to prevent the spread of Las and therefore serve as potential sources of inoculum (Tiwari et al. 2010). There were approximately 55,252 ha of abandoned ci trus reported in 2012 (USDA NASS. http://www.nass.usda.gov/Statistics_by_State/Florida/Publications/Citrus/aban/CitAA12. pdf Viewed: May 15, 2013.) When moving from unmanaged to managed groves, psyllids typically move from th e inside rows of unmanaged groves to the outside rows of managed groves (Tiwari et al. 2010). Within 3 d, D. citri can disperse between such groves at a distance of 60 100 m (Boina et al. 2009). However, additional s tudies demonstrated that psyllids have the capability to disperse 2000 m in 12 d, and possibly farther (Lewis Rosenblum 2011). Because psyllids have weak flight muscles and have not flown particularly long distances in wind tunnel studies it is likely that long distance dispersal consist s of repeated short distance flights (Sakamaki 2005, Arakawa and Mivamolo 2007). Transmission of Candidatus Liberibacter asiaticus For successful pathogen transmission to occur, a psyllid must first acquire the pathogen by ingesting phloem from an infected ho st plant. Correlation of successful pathogen acquisition with phloem ingestion was first demonstrated by Bonani et al. (2010). Both adults and nymphs of D. citri are capable of acquiring the Las pathogen, however, pathogen acquisition occurs at a much high er rate for nymphs compared to adults (Pelz Stelinski et al. 2010) After acquiring the pathogen, a latency period is thought to be necessary before a psyllid is capable of inoculating a plant with the pathogen. Latency periods ranging from 1 to 25 d have been reported ( Xu et al. 1988, Roistacher 1991). Pathogen inoculation is hypothesized to occur during the process of
16 phloem salivation, the point in the psyllid feeding process immediately preceding phloem ingestion (Bonani et al. 2010, Serikawa et al. 2 012). Pathogen inoculation by adult D. citri is probably the most likely route of pathogen spread since adults are the dispersing stage of the psyllid. While one psyllid can inoculate a healthy plant with the pathogen, trees with 100 or more D. citri are most likely to become successfully inoculated with Las (Pelz Stelinski et al. 2010). Inoculation of a healthy plant by D. citri nymphs may also be a possible, but to a lesser extent. L ow rates of transovarial transmission (3.6%) have been reported (Pelz S telinski et al. 2010 ) which could result in Las infected nymphs feeding on Las negative plants. Additionally, low rates of sexual transmission (2 3%) of Las between conspecific adult D. citri have been reported ( Mann et al. 2011). Host Plants of Asian C itrus Psyllid The subfamily Aurantioideae, within the family Rutaceae contains host plants of D. citri in the following genera: Atalantia, Citrus, Clausena, Euodia, Fortunella, Murraya, and Poncirus (He 2000, Halbert and Manjunath 2004, Yang et al. 2006). While it is important to know the host plants of D. citri it is equally important to know what plants that are related to citrus but are not suitable hosts for breeding efforts. Those that are Citrus jambhiri Lushingto Citrus aurantifolia (Christm.) Swingle] (Nehru et al. 2004). Species that have shown to have decreased oviposition, development, and survival of D. citri [ Citrus sunki ( Citrus reshni Hort. ex Tan) (Nava et al. 2007, Tsagkarakis and Rogers 2010). D. citri avoids colonization on trifoliate orange, Poncirus trifoliate (L.) and will not colonize on white sapote ( Casimiroa edulis Llave et Lex) (Westbrook et al. 2011).
17 Curr ent Management Strategies for Asian Citrus Psyllid Biological Control Biological control is an effective method of management for many invasive insect species (Qureshi et al. 2009). In an effort to keep psyllid management sustainable and utilize integrate d pest management, many different organisms have been investigated for their effectiveness against D. citri A parasitoid that has been of major focus in biological control is Tamarixia radiata Waterson. T. radiata is an ectoparasit ic wasp that parasitiz es 5 th instar D. citri and adult females also feed on younger stage nymphs (Chu and Chien 1991, Qureshi et al. 2009). T. radiata has been successful in lowering incidences of D. citri in other citrus producing countries in which it was introduced (Gomez T orres et al. 2012). It was introduced in Florida as part of a classical biological control program in 1999 2001 from Taiwan and South Vietnam prior to the introduction of HLB. Studies have shown that parasitism rates are lower in Florida than they are in other countries where this method has been more successful. Because the parasitism rates have been low, T. radiata has not proven to be a very successful biological control agent in Florida (Tsai et al. 2002, Michaud 2004, Qureshi et al. 2009). Studies have shown that T. radiata was only responsible for 0.2 1.3% of D. citri nymph mortality in central Florida (Michaud 2004). Reasons that have been proposed to explain this lower rate include more extreme weather in Florida, predation of nymphs containing w asp eggs, and poor overwintering of the host (Qureshi et al. 2009). Tamarixia radiata did establish in Florida after its release, however, another host specific parasitoid wasp, Diaphorencyrtus aligarhensis that was released at the same time, did not. D. aligarhensis is an endoparasitoid that was also released in Florida in 1999 from Taiwan and Vietnam and had to be re released beginning in 2007 from China
18 to re establish the population (Rohrig et al. 2012). When compared with T. radiata female D. alig arhensis cause half the mortality of psyllid nymphs in her lifetime. Additionally, T. radiata have shorter generation times while having higher reproductive rates, and observations have been made that support if both parasitoids oviposit a D. citri nymph the competition is in favor of T. radiata. Re release efforts of D. aligarhensis have not proven successful in Florida as of 2012 (Rohrig et al. 2012). Several species of entomopathogenic fungi have been evaluated for their effectiveness against D. citr i for biological control as well. Isaria fumosorosea Wize and Hirsutella citriformis Speare have received the most attention, likely because they are naturally occurring in Florida. I. fumosorosea has been the focus of several studies because it is alrea dy commercially produced as a bioinsecticide for other pests and it was found naturally occurring in a Florida citrus grove and was thought to be able to survive in citrus grove environments (Hall et al. 2012, Stauderman et al. 2012). In laboratory studie s, I. fumosorosea had high mortality rates and killed 100% of adults treated. The time it took to kill psyllids in the study varied depending on the dosage applied, the quickest being 3.5 days. In greenhouse studies the mortality dropped to 50%, which wa s compared to 100% mortality in the greenhouse with drench imidacloprid treatments (Stauderman et al. 2012). In another laboratory study that looked at I. fumosorosea effect on psyllid feeding, the earliest it reduced feeding was 4 days post treatment (Av ery et al. 2011). Use of biological control against D. citri is not a viable option against preventing disease spread. Natural enemies have not been successful for the most part because those that were introduced for classical biological control have not been able to
19 establish high enough population levels to keep psyllid levels low. This is likely due in large part to the use of broad spectrum insecticides that are repeatedly applied and have a known high toxicity to T. radiata (Hall and Nguyen 2010). Additionally, use of natural enemies would not be a feasible form of disease management through psyllid management because in order to sustain the natural enemy population, some of the host population must be retained each year to support reproduction (van Emden and Service 2004). In the case of a disease vector such as D. citri the goal is to completely rid a grove of the pest because even small vector populations will result in pathogen spread This means the threshold for D. citri is very low, basical ly zero therefore making use of natural enemies in commercially managed groves an unsuccessful means of control. Additionally, entomopathogenic fungi are not likely to prove to be a successful control method for a multitude of reasons. The highest leve ls of mortality were observed in the laboratory, and even in protection of greenhouse studies have shown decreased mortality (only approximately 50% for I. fumosorosea ) (Stauderman et al. 2012). While it would seem that this method could be used in conjunc tion with other chemical control measures, a problem arises where typical grove management methods could inhibit the fungal pathogen. A study on H. citriformis in a commercially managed grove showed that on average only 23% of D. citri observed on leaves were killed by the fungus. These numbers were higher in fall and winter, but lower in spring. Summer of 2006 showed an abundance of psyllid mummies killed by the fungus, but summer of 2007 did not. This prompted the researchers to do a lab study observi ng the toxicity of copper and oil sprays to H. citriformis thus showing its compatibility with commercial
20 management that relies on these methods for other disease and secondary pest management. The results of the study showed that copper hydroxide, petr oleum oil, and elemental sulfur reduced the pathogenicity of the fungus (Hall et al. 2012). These pesticides are widely used in citrus production to protect against insects, diseases, and mites (FL citrus pest management guide 2013). In addition to fungal populations being reduced by conventional management, as stated before the shortest amount of time required for I. fumosorosea to kill D. citri was approximately 4 d, often taking longer than that. In laboratory studies, H. citriformis took approximately 9 10 d to kill 100% of the population (Meyer et al. 2007). This allows more time for psyllids to transmit the Las pathogen. While much research is currently being done to discover the best integrated pest management strategies so as to not rely so heav ily on insecticide use, insecticides remain the best option for providing the level of control necessary. This will not continue to be true for long, as continued back to back applications of insecticides increase the risk of D. citri developing resistanc e to these insecticides. There are reports this is already beginning to occur, and will likely only continue to intensify (Grafton Cardwell et al. 2013). Chemical Control Historically, the use of insecticides to manage psyllid populations has been, and remains, one of the key components of HLB management programs (Bove 2006 ). The primary rationale for insecticide use for psyllid control is to maintain psyllids at as low of levels possible in order to slow pathogen spread (Rogers et al. 2013 ). Of the 28 d ifferent known modes of action described by the Insecticide Resistance Action Committee (IRAC MoA Classification 2012 ), only 8 of these MoAs have been shown to
21 provide some control of psyllid populations. These groups include pyrethroids, organophosphates and carbamates, neonicotinoids, some insect growth regulators, horticultural oils, spinosyns, tetronic acid derivatives and avermectins. Of these 8 MoA groups, the insect growth regulators, tetronic acid derivatives, avermectins and horticulatural oils are more selective in activity and only provide control of the nymphal stage of psyllids. The pyrethroids, organophosphates and carbamates, neonicotinoids, and spinosyns are considered more broad spectrum in activity and control both adult and nymphal stage psyllids. Because control of adult psyllids (the dispersal stage) is needed to minimize disease spread, the use of these broad spectrum insecticides are recommended for providing the best control of psyllid populations and thus are more commonly used by F lorida citrus growers (Rogers et al. 2013). While the use of broad spectrum insecticides can provide rapid kill of all life stages of psyllids, long lasting control of D. citri populations is often not obtained In Brazil for example, citrus growers make u p to 28 insecticide applications per season for psyll id control (Belasque et al. 2009 ). In China, the number of insecticide applications can be as high as 52 times per year, or every week just to keep psyllid populations at low levels (Beattie and Holfor d 2009 ). In Florida, citrus growers currently average 8 12 applications per season with varying success in terms of level of D. citri control provided (Rogers, personal communication). The two primary factors attributed to the failure to keep psyllid popul ations under control despite the use of broad spectrum insecticides are 1) the short duration of residual activity provided by insecticides and 2) psyllid dispersal behavior. Previous field trials examining the duration of psyllid control provided by diff erent insecticides
22 have yielded varying results with the reported duration of control lasting from as little as a few days to several weeks or longer (Childers and Rogers 2005, Qureshi and Stansly 2007). However, in a study where adult psyllids were caged on trees at varying dates following applications of broad spectrum insecticides, none of the insecticides tested provided control of adult psyllids 12 d after applications were made (Okuma and Rogers, unpublished) Thus, field trials suggesting that psylli d control can last many weeks may be more of a measurement of the time required for psyllids to recolonize a tree instead of an actual measure of the true residual control provided by an insecticide. When on e considers the short residual control provided by insecticides, combined with the fact that D. citri can be active dispersers (as discussed above), it becomes evident that the success in maintaining psyllid populations at low levels will depend to some extent on the population levels in surrounding ar eas. For example, if a grower sprays their grove but the adjacent grove owner does not spray, within a week psyllids dispersing from the unsprayed grove may recolonize the treated grove, hence the need for repeated pesticide applications. With the goal of enhancing the effectiveness of insecticide applications for psyllid control, area wide management programs have been developed throughout the citrus growing regions of Florida. Referred to as Citrus Health Management Areas (CHMAs ), growers work together in an effort to manage the spread of HLB by coordinating the timing of their psyllid control sprays. The main purpose of this coordination is to ensure a larger area of control when spray applications are made, and thus decrease the likelihood that psyllids will quickly recolonize recently treated areas. The CHMA efforts thus far have proven successful
23 with s couting results indicating a 67% reduction in psyllid populations statewide during the period of 2011 to 2012 (Rogers et al. 2012). Factors Affecting Pe sticide Efficacy The challenges of controlling D. citri have resulted in Florida citrus growers rethinking how they approach their pest management programs, particularly with regards to the timing and application of insecticides. Traditionally, pesticide applications have been made using high (1, 169 2 ,338 L/ha ) spray volumes and were primarily only needed during the late spring or summer months. In order to control D. citri however, pesticide applications are needed year round and as should be applied as quickly as possible. Factors such as equipment used, tank mixing of pesticides, and environmental conditions can all affect the efficacy of pest control programs. Application Methods Beginning around 2008, growers started considering use of low volume (L V) application technology in the place of traditional airblast sprayers The terms ultra low volume and low volume seem to be relatively interchangeable in many papers, and also seem to be changing over time as the spray volumes get ever more reduced. In a survey of growers in 2002, very low volume w as classified as 234 to 327 L/ha, low vol ume was 935 to 1590 L/ha middle vol ume was 1870 to 3553 L/ha and high vol ume was 4207 to 7012 L/ ha. These classifications were put together based on the technology that growers were using at the time the survey was conducted (Stover et al. 2002). It was not until 2008 that growers were beginning to decrease the application volume further, but not many insecticides were registered for low volume use at that time. In August of 2008, label changes were being petitioned with EPA to be able to apply more insecticides as low volume sprays For the purposes of this paper, low
24 volume will be any spray volume from 1 to 25 gallons per acre (GPA) (9.3 5 to 234 L/ha), and ultra low volumes will be volumes less than 1 GPA (less than 9.3 5 L/ha), which is typically only used in a erial applications such as mosquito control The benefits of using low volume applications include more rapid application at a reduced cost to the grower (Atwood and Stelinski 2008). Another benefit associated with low volume applications is that the concentration of active ingredient per droplet is higher with more droplets hitting the target than in traditional standard airblast (Atwood and Stelinski 20 08). Growers have observed that low volume applications provide similar psyllid control as traditional applications (Atwood and Stelinski 2008). A study conducted in 2008 compared the efficacy of three different LV sprayers and standard airblast applicat ion for psyllid control (the authors referred to this as ULV at that time). The LV applications were made at 1 pint (1.2 L/ha) (when there was no carrier added) or 2 GPA (18.7 L/ha) depending on the LV equipment used and standard airblast applications wer e made at 100 GPA (935 L/ha), all applying a pyrethroid at 1 pint (1.2 L/ha). These treatments were compared with an untreated control. There were no significant differences found within the treatments, however they were significantly different from the control in number of psyllid egg, nymph, and adults that were captured. They also compared two LV sprayers with standard airblast during spring flush. They applied a pyrethroid at 4.3 oz per acre (314 mL/ha) at volumes of 1 pint per acre (1.2 L/ha) and 2 GPA (18.7 L/ha) for the low volume applications, and 100 GPA (935 L/ha) for the standard airblast. They saw an increase in psyllids captured 7 days after treatment, but sample dates after that showed a decrease in captures with the insecticide treatments showing no difference from each other. The number of immature
25 psyllids was reduced to zero for two weeks following application for all insecticide treatments. They also compared different modes of action (pyrethroid, organophosphate, and insect growth re gulator) when applied at low volumes. The organophosphate and pyrethroid treatments significantly decreased adult populations whereas the insect growth regulator did not. All three insecticide treatments provided significant reductions in D. citri nymph populations. Typically, low volume applications have smaller spray droplet sizes than larger volume application technology. Because of this, spray drift is a problem that is associated with low volume applications. When approval was granted for labelin g of insecticides for use with low volume, it was directed that the droplet size applied had to have a volume median diameter of 90 m or greater, which is less likely to result in drift problems (Hoffman et al. 2010). Also to reduce the likelihood for dri ft, low volume applications are made at night when wind conditions are calm, usually beginning at 10 pm and continuing until dawn (Atwood and Stelinski 2008). Airblast sprayer technology has not changed much over the years, and consists of a single axial f an that is used to blow the material out from the nozzles (between 7 40 nozzles on each side), which are configured radially to the target trees. They are either powered by separate engines when engine driven, or tractor power take off (PTO) powered. LV sprayers are also air assist sprayers but can have different construction. One type uses squirrel cage fans in order to generate the high velocity air for air shear nozzles, while another uses six to eight cross flow fans with rotary atomizers stacked in a tower. What makes the applications faster than airblast is that manufacturers recommend they be made at ground speeds of 3 miles per hour (mph) (4.8 km/ha)
26 versus standard airblast which has a recommended ground speed of 1 to 2 mph (1.6 to 3.2 km/ha). Additionally, time is saved in fill ing tanks as small as 2 gal (7.58 L) for low volume versus the standard 100 125 gal (378.5 473 L) for standard airblast (Stover et al. 2003). The insecticides currently labeled for use in citrus at LV applications are aba mectin, carbaryl, diflubenzuron, fenpropathrin, zeta cypermethrin, dimethoate, malathion, and spinetoram. Use of low volume (LV) spray applicators, which apply a volume of approximately 17.3 to 19.7 L/ha, have become increasingly more popular method for application of insecticides in psyllid control. This is because LV sprayers can make applications faster, and thereby less expensive than standard airblast applications that apply a volume of 500 to 1000 L/ha. LV sprayers can apply insecticides to appro ximately 100 ha in a night, which as a result decreases cost by 6 to 7 fold in comparison with standard airblast applications A study showed that LV sprayers have increased effectiveness when fenpropathrin was applied in lab settings, using a piezo elect ric droplet generator to spray plants. The study discovered that smaller droplet size meant an increased concentration of the insecticide that the psyllid would come in contact with. The 40.5 m and 52.0 m droplets that were tested showed the highest mo rtality rates, and these sizes are associated with what is produced by an Ultra Low volume (ULV) hand held sprayer. The results also showed that 90 m droplet sizes, which is similar to that produced by LV sprayers, still had high mortality, although not as high as the ULV like droplet diameters The low volume sprayer still had a higher mortality rate than the larger droplet sizes produced by standard airblast sprayers. The conclusion of this study was that application of fenpropathrin was more effectiv e in smaller droplet sizes;
27 however this study only observed up to 7 days of mortality, and was also done in laboratory settings (Boina et al. 2012). While lower volume applications are often associated with smaller droplet sizes, there is no specific ch aracterization of droplet size that is considered to be linked to volume applied, as this is based on the construction of the spray equipment technology and many other factors that determine droplet size. That being said, larger volumes of spray material are required to produce larger droplet sizes, simply because more volume is being dispensed in each droplet and therefore a larger volume must be applied to cover the entire spray area. Stover and others (2003) discussed this as well as retention of spray material on the leaf surface based on different droplet sizes. Larger droplets tend to have more kinetic energy than smaller droplets, which allows them to be able to reach the target leaf surface better due to a lessened tendency to evaporate, be divert ed by wind, and are more able to penetrate the boundary layer of the tree. However, if the droplet size is too big (i.e. above 300 m), runoff can also occur. Therefore, it can be said that smaller droplets can have better retention than larger droplets, as long as they are large enough to be able to reach the inside of the 1984, Salyani 1995) conclusions that there is a great deal of variability of coverage for both large and small droplet sizes, based large droplet sizes ability to coalesce and runoff leaf surfaces, and small droplet sizes inability to reach the inside of the canopy. However, these authors did not mention droplet sizes in a range of 90 200 m, as was describ ed to be produced by current low volume technology (Hoffman et al. 2010),
28 which have been described to slowly settle into the tree canopy after application and suggested to have better coverage because of this (Stelinski et al. 2009). Use of Adjuvants Ad juvants are a broad category of materials that can be added to pesticides in order to give additional properties that are often added to help improve spray applications. These properties include buffering, spreading, sticking, emulsifying, dispersing, wet ting, and can reduce evaporation, foaming, spray drift and volatilization of spray materials. There are two types of adjuvants: a formulation adjuvant which is added by the manufacturer and is already in the pesticide and a spray adjuvant which is a separ ate product added to the tank by the applicator. Within the spray adjuvant category there are two more categories: activator adjuvants which are used to improve the activity of the pesticide, often helping their absorption and special purpose/utility adjuv ants. Activator adjuvants include surfactants, oils, and nitrogen based fertilizers. surf ace act ing a ge nts Surfactants improve the wetting and spreading properties of the pesticide material, meaning that they help to we t the foliage and spread the material on the leaf better. This helps to improve the biological activity of the pesticide in that it has a better chance of an insect coming into contact with the material. There are three types of surfactant oils: crop oils crop oil concentrate, and vegetable oil concentrate. Crop oils are 95 to 98 percent paraffin or naptha based petroleum that also has 1 to 2 percent surfactant or emulsifier T hese type products help the pesticide to penetrate the leaf or insect cuticle. Crop oil concentrates (COCs) consist of 80 to 85 percent emulsifiable petroleum oil and 15 to 20 percent non ionic surfactant, which give them penetration and spreading properties. Vegetable oil concentrates (VOCs) consist of 80 to 85 percent of some typ e of seed oil that is crop
29 derived, such as cotton, linseed, or sunflower oil. The remaining 15 to 20 percent of the formulation is non ionic surfactant. This category also includes methylated seed oils (MSOs) that have undergone a process called esterif ication to increase their penetrating ability. Nitrogen based fertilizers are mainly used with herbicides to increase their systemic properties. Surfactants are classified by the way in which the ions separate, or ionize. The classifications are cationi c (positive charge), anionic (negative charge), non ionic (no charge), and organosilicone surfactants. Non ionic surfactants help the product penetrate the plant and therefore are often used with systemic pesticides. Organosilicone surfactants help with spreading, reduces the surface tension of the liquid material, and improves rainfastness. Crop oil concentrates, non ionic surfactants, and organosilicone surfactants were used in this project. Special purpose/utility adjuvants include compatibility agen ts, buffering and conditioning agents, deposition agents, defoaming agents, drift control agents and thickeners (Hock et al. 2011) These types of products will not be further explained because they were not used in this project. There are also petroleum oils that are often used for horticultural purposes, which are also known as horticultural oils. They are refined petroleum products that are applied at a dilution in water of 2% material added to the volume of water. Several adjuvants have been studie d for their effects against ACP based on their effects on other insect pests. This includes a specific organosilicone, Silwet L 77, which has demonstrated toxicity against many other arthopods (Srinivasan et al. 2008); and petroleum oil, which is known to be toxic to small and immobile arthropod pests (which the nymphal stage of ACP is) due to its suffocating effects (Rae et al. 2010).
30 Several studies have been conducted to observe if any additional toxicity can be observed from using certain adjuvants a lone or in combination with insecticides. One study compared the uses of Provado (imidacloprid) alone, Provado at different rates with Induce (non ionic surfactant that is a wetter/spreader), Movento (spirotetramat) with 435 oil, Movento with Induce, Move nto with MSO seed oil, Movento with Kinetic (non ionic surfactant that is a wetter/spreader/penetrant), Movento in combination with Provado and Induce, and Sevin (carbaryl). Their results indicated that Provado with Induce had similar toxicity to Provado alone. There was a difference shown by the addition of 435 oil or Induce to Movento, this showed an increase in control of psyllid nymphs at 14 days after treatment. Additionally, Movento with 435 oil was the only treatment to show difference from the co ntrol at 24 days after treatment. The authors stated that Induce and 435 oil were the most effective adjuvants for Movento when compared with the MSO seed oil and Kinetic. Another experiment demonstrated that 435 oil was a successful adjuvant for use wit h Agri mek (abamectin). They also observed that Danitol applied with 435 oil showed good control of nymphs at 3 and 7 days after treatment (Qureshi et al. 2009). Another study was conducted to observe the effects of different organosilicone surfactants ( Kinetic and Silwet L 77) alone and combined with insecticides to ACP mortality. Their review of literature revealed that multiple other arthropod pests have demonstrated toxicity to Silwet L 77 because it is part of a class of surfactants called trisiloax ane surfactants that have shown activity like that of hyperactive soaps that essentially make it so the insect cannot breathe by infiltrating the respiratory system and blocking the gaseous exchange. The authors conducted laboratory, greenhouse, and
31 field studies with Silwet L 77 and Kinetic. They determined that Kinetic is less effective at causing mortality itself than Silwet L 77. When they used Silwet L 77 with one fourth and one half the lowest label rate (LLR) of imidacloprid increased mortality o f adults was observed. Silwet in combination with lower than label rates of either imidacloprid or abamectin showed an increase in mortality of adults. Laboratory, greenhouse, and field trials showed that 0.05% solution of Silwet al.one did suppress nymp hs. When they combined one tenth the LLR of imidacloprid with Silwet, it was significantly less effective than the LLR of imidacloprid alone. It is still possible that Silwet can reduce the use of insecticides, because another field trial within this stu dy did not find and significant difference between the treatments of Silwet al.one, Silwet with one tenth the LLR, Silwet with the LLR of imidacloprid (Srinivasan et al. 2008). A study was conducted after the aforementioned study to determine the effecti veness of Silwet and Kinetic combined with petroleum oil and copper hydroxide on psyllid mortality, and also to compare the toxicity of these with the toxicity of imidacloprid and abamectin to the natural enemy of ACP, T. radiata. They confirmed that Silw et has more insecticidal effects on ACP than Kinetic in screenhouse studies. Silwet in combination with the higher rate of petroleum oil (2%) tested showed to suppress all stages of ACP, but the most control was observed with eggs (81%). While imidaclopr id and abamectin showed to have high toxicity to T. radiata, the use of Silwet with petroleum oil (or Silwet al.one) did not show toxicity to T. radiata The authors suggested this would be a good method to be compatible with biological control, and can b e part of an IPM strategy (Cocco and Hoy 2008). Studies conducted on application of Silwet L 77 alone to pecan orchards for pecan aphid showed that the
32 higher concentrations of Silwet applied were very effective for killing the pecan aphids, however they did not demonstrate any residual activity (Wood and Tedders 1997). A study by Childers and Rogers that was discussed earlier for its experiments on many different insecticides also conducted experiments on the addition of petroleum oils (also called horti cultural mineral oils or HMOs). This study was conducted from a stand point of keeping psyllid popu lations low, as it was before L as had been detected in Florida, so the expectations of a psyllid management program were different than they are today. Tha Agri mek combined with a HMO had similar levels of control as Danitol, Lorsban, and Provado at 5 days after treatment, and again in another experiment at 4 DAT. In the second experiment they were also able to observe good egg and nymph suppression at 14 DAT with Agri mek and HMO. They also applied several different HMOs alone and in combination w ith Kinetic or Citru Film non ionic adjuvants. They were not able to observe control of adults with any of these treatments, however egg and nymph numbers were reduced significantly at 9 DAT for several of the HMOs. This reduction was not observed by 13 DAT for any of the HMOs. They suggested further research should be done with HMOs to better determine control under different conditions and in different locations (Childers and Rogers 2005). In another study of petroleum oils used for psyllid control in China, it was determined that the petroleum oil treatments were as effective as organophosphate and insect growth regulator treatments against early instar nymphs, but were ineffective against eggs. The authors stated that the control of eggs with petrol eum oil was not
33 successful because the sprays were not timed correctly. They explained this could have been because the amount of time between the final spray and the data collection was long enough that new shoots could have grow n to 0.4 mm in length (wh at is required for oviposition) between the time of spray and the time of sampling. A study was conducted in Vietnam with the expectation of finding additional residual activity or rainfastness, as well as toxicity to ACP from the application of an adjuva nt and a mineral oil. They were not able to observe any additional longevity from either of these treatments (Ichinose et al. 2010). Environmental Conditions There are many different factors that make season in which application is made an important comp onent of longevity of the control gained from the application. For ACP control, winter applications are exceptionally important because they are aimed at controlling the overwintering population, and if this is done successfully then reduction in populati ons have been reported for many months during field trials (Qureshi and Stansly 2010). Temperature, which can range from 10 35C in Florida, has been shown to have an impact on toxicity of insecticides in laboratory experiments. Pyrethroids have been obs erved to have a negative correlation with temperature in many different studies (Scott 1995, Musser and Shelton 2005, Satpute et al. 2007, Boina et al. 2009). Organophosphates, however, have been shown to have a positive correlation with temperature (Scot t 1995, Satpute et al. 2007, Boina et al. 2009). These correlations are a general condition of the specified class, and certain members within the class can have different relationships than what was stated (mixed, positive, or negative) (Sparks et al. 19 82, Toth and Sparks 1990). Not much research has been conducted to determine a relationship for neonicotinoids and temperature. A study by Boina et al.
34 (2009) evaluating different insecticide classes determined the relationship between each insecticide a nd its toxicity to D. citri under different temperatures in laboratory or growth chamber settings. They found that dimethoate and chlorpyrifos (organophosphates), carbaryl (carbamate), and abamectin (avermectin) had positive temperature correlations, mean ing as temperature increased the toxicity to D. citri increased (from 17 37C). The pyrethroids that were tested (fenpropathrin, zeta cypermethrin, and lamba cyhalothrin) showed negative temperature correlations, with bifenthrin being an exception and dem onstrating a positive temperature correlation as temperature increased from 27 37C. Three neonicotinoids were tested, and two of those (imidacloprid and thiamethoxam) had a mixed response to temperature increases while the third (acetamiprid) exhibited a positive temperature correlation (Boina et al. 2009). Justification The experience of Florida citrus growers has been that D. citri is a very difficult pest to control for an extended period of time. Thus, repeated use of insecticides is common throughout Florida in order to maintain psyllids at acceptable levels. However, the long term use of increased insecticide applications is not economically viable for growers and also has other potential negative impacts such as the development of pesticide resistan ce and non target environmental impacts. Furthermore, changes in spray applications and effects of environmental conditio ns on insecticide efficacy may a ffect the level of psyllid control provided. Therefore f urther studies examining the effects these fac tors can have on duration of effective psyllid control provided is warranted.
35 Goals and Hypotheses The goal of this project was to determine the residual control of D. citri provided by insecticides under different application conditions. Utilizing three classes of insecticides commonly used to manage D. citri I examined factors that could potentially increase or decrease residual activity of these insecticides. More specifically, I tested the hypothesis that the addition of adjuvants/surfactants would increase the effectiveness of insecticides. I also applied insecticides with different types of spray equipment : low volume sprayer and a standard airblast sprayer to test my hypothesis that the standard airblast sprayer would provide a longer duration res idual activity as a result of better spray coverage Finally, I applied insecticides during different seasons of the year to determine if there would be a difference among the insecticides in terms of level of efficacy provided I hypothesized that the e fficacy of the insecticides tested would not vary between seasons Specific Objectives 1. Determine whether season in which the insecticide is applied has an impact on residual activity. 2. Determine what effect addition of surfactants to selected insecticide s will have on residual activity. 3. Test which spray application method, low volume or standard airblast sprayer, provides the longest residual activity.
36 CHAPTER 2 SEASONAL EFFECTS ON INSECTICIDE APPLICATIONS FOR D. CITRI CONTROL Introduction D. citri are very difficult to control because they have a high rate of reproduction, short generation times, multiple generations within a year, reach sexual maturity quickly, and can disperse easily between groves (Liu and Tsai 2000, Wenninger and Hall 2007, Boina et al. 2009, Tiwari et al. 2010, Lewis Rosenblum 2011). They also overwinter in the citrus groves, so the population does not truly ever go away. While fine feather flush is required for females to feed on to mature eggs, for oviposition as well as nymphal development adult ACP can survive on mature leaves for several months during the winter. Flush production occurs at different times in the year, the major flush periods in Florida occur during the late winter and early spring, the late spring/early summ er and a final flush during the fall No major flushes typically occur in the late fall or winter and thus is the time often referred to as the dormant season in Florida citrus production. The timing of these flushes will influence ACP populations, thereby influencing insecticide application timing (Qureshi and Stansly 2010). These factors make it very important to appropriately time applications throughout the year. There are many variables that can affect residual activity of an insecticide application that change from season to season. These variables include, but are not limited to: temperature, rainfall, relative humidity, dew point, and solar radiation. These variables can have an effect on the degradation of insecticides post application. The wa y these variables affect insecticidal activity often varies depending on the insecticide. There has been some research conducted on the effect of temperature on different insecticides and their toxicity to D. citri There are relationships that have been
37 demonstrated through other studies between temperature and insecticides, but can vary on a pest by pest basis and can also vary within an insecticide class. Therefore, in order to gain insight into the temperature toxicity relationship, insecticides must be tested individually on the target insect. Previous laboratory studies on other insects have determined some general temperature toxicity trends. Pyrethroids have been observed to have a negative correlation with temperature through many different st udies (Scott 1995, Musser and Shelton 2005, Satpute et al. 2007, Boina et al. 2009). Organophosphates, however, have been shown to have a positive correlation with temperature (Scott 1995, Satpute et al. 2007, Boina et al. 2009). These correlations are a general condition of the specified class, and certain members within the class can have different relationships than what was stated (mixed, positive, or negative) (Sparks et al. 1982, Toth and Sparks 1990). Not much research has been conducted to determ ine a relationship for neonicotinoids and temperature. In a laboratory study focused on D. citri there was a positive temperature correlation for dimethoate and chlorpyrifos (organophosphates), carbaryl (carbamate), and abamectin (avermectin), meaning as temperature increased the toxicity of these insecticides to D. citri also increased (from 17 37C). The pyrethroids that were tested (fenpropathrin, zeta cypermethrin, and lamba cyhalothrin) showed negative temperature correlations, with bifenthrin being an exception and demonstrating a positive temperature correlation as temperature increased from 27 37C. Three neonicotinoids were tested, and two of those (imidacloprid and thiamethoxam) had a mixed response to temperature increases, and the third (acet amiprid) demonstrated a positive temperature correlation (Boina et al. 2009).
38 With this knowledge about temperature toxicity under controlled settings, it is important to determine if the same trends could be observed in the field, where temperature and humidity are not constant, and there are other factors that could influence the toxicity such as rainfall and solar radiation. Additionally, while there are studies that sampled the natural population after insecticide applications, this experiment was im portant because it utilized caged psyllids, which demonstrates the true duration of residual control provided not readily apparent when sampling the natural population (Qureshi and Stansly 2007, Qureshi and Stansly 2009, Qureshi and Stansly 2010). In the case of the later, duration of control provided can be confounded by the psyllid population level in the surrounding area which influences the rate of reinfestation of previously treated trees. Materials and Methods Insecticide Applications The efficacy o f insecticide a pplications were evaluated four times during the calendar year. The times chosen for evaluation were based on anticipated weather conditions typical of the four seasons of the year in Florida and included : January, April, September, and Dece mber 2012, and January 2013. The field experiments were conducted in Lake Alfred, FL in a block of Hamlin Citrus sinensis (L.) Osb.) growing on Carrizo citrange ( C. sinensis (L.) x Poncirus trifoliate L.) rootstock The tree spacing was 4.6 x 7.6 meters or approximately 290 trees/ ha The plot design was 4 rows across, 5 trees deep, making each plot about 42.5 hectares in size. For the following insecticides were evaluated : fepropathrin (Danitol 2.4 EC, V alent USA Corporation, Walnut Creek, CA) at a rate of 1.168 L/ha dimethoate (Dimethoate 4E, Cheminova Inc., Research Triangle Park, NC) at a rate of
39 1.168 L/ha imidacloprid (Provado 1.6F, Bayer CropScience LP, Research Triangle Park, North Carolina) at a rate of 0.73 L/ha There was also an untreated control. Treatments were applied using a Pul Blast PTO sprayer with a tank size of 1514 liters Applications were made at a speed of approximately 4km per hour calibrated to de liver 1,168 L/ha spray volume. There were 5 replicate plots for each of the 4 treatments for a total of 20 plots. Following application of insecticides, the sprays were allowed to dry for 24 h prior to initiating psyllid caging. Caging of A dult D. citri Beginning o ne day after treatment (DAT) adult psyllids were caged on branches of insecticide treated and untreated trees to assess duration of control provided by the treatments. In each plot, psyllids were caged on one branch of the two central most tre es of the plot. On each branch used, a mesh sleeve cage was placed over the branch and the open end of the sleeve cage was secured with flagging tape. Prior to placing the sleeve cage on the branch, a plastic vial containing 20 adult D. citri collected fr om a laboratory maintained colony, was placed in the sleeve cage. After the cages were securely attached to the branch, the vials were opened to release ACP into the sleeve cage to allow direct contact with the insecticide residues on the leaves The psyll ids remained caged on the trees for one week, after which time each branch with a cage was cut off the tree just above the sleeve cage so that the caged branch containing psyllids could be returned to the lab for closer examination As each set of cages w as removed a new set of cages containing psyllids were caged on each tree for a second week (to represent 8 DAT) using the same process. Once caged branches were returned to the lab the sleeve cages were carefully removed and the number of live and
40 dead psyllids on each branch was recorded This was repeated until 4 weeks of data (22 DAT) w as collected. Statistical Analyses For each treatment, the number of live and dead ACP was totaled to attain the percent mortality that was recorded. This number wa s also used to weight the mortality. This was necessary because there were not always 20 ACP in each cage, and it corrected for that by not allowing a cage with more psyllids to have more of an influence than a cage with fewer The weighting was achieved by dividing the total number of live Each replication had two pseudo replications, so the weights for each were averaged as were the percent mortality for each. Sta tistical Analysis System (SAS) was used to transform the average mortality using an Arcsin transformation. Proc GLM was used what was used for comparing the mortality rates for each insecticide. All of the treatments of all data collection dates were put into a model to determine if there was an effect of any of the following factors: season/date, days after treatment, temperature at 60 cm (average and maximum), temperature at 2 m (average and maximum) and rainfall. These variables were transformed with a log transformation, except for rainfall. The best models were attained using continuous variables instead of categorical for variables such as season/date and days after t reatment. Season/date was turned into a continuous variable by utilizing the Julian day system in which each calendar day is given a number, starting with 1 for January 1 and counting sequentially after that. Therefore, each date within the treatments wa s assigned a number that was added into the model in place of the actual date. Using continuous variables allows the model to
41 better predict outcomes even if there is not data for that date. A backwards selection was used to remove non significant variab les until the model was significant. Results Within a given season of the year, mortality of D. citri was not significantly different between insecticides tested 1 DAT. However, a s time progressed, all insecticides rapidly lost efficacy, some quicker than others (Figures 2 1 to 2 5). For caging experiments conducted in January 2012 and 2013, there were significant differences in D. citri mortality between treatments 1 and 8 DAT but there were no significant differences between any treatments 15 and 2 2 DAT (Table 2 2). At 1 DAT in January of both years, D. citri mortality was significantly higher for all insecticides compared to control plots but there was no significant difference in mortality between treatments in either year (Figs. 2 1, 2 5). There were significant differences between insecticide treatments at 8 DAT in both years. In January 2012, mortality 8 DAT was highest for fenpropathrin, which was significantly different from imidacloprid and dimethoate, the la t ter of which did not differ sign ificantly (Fig. 2 1). In January 2013, mortality rates were significantly higher for fenpropathrin and imidacloprid compared to dimethoate (Fig. 2 5). However, the magnitude of control provided by all insecticides 8 DAT was much lower in January 2013 comp ared to 2012. In April 2012, there were significant differences in mortality rates between treatments 1 and 8 DAT, but not at 15 and 22 DAT (Table 2 2). At 1 DAT, D. citri mortality was significantly higher for all insecticides compared to control plots bu t there was no significant difference in D. citri mortality between insecticides (Fig. 2 2). At 8 DAT, only imidacloprid had mortality rates significantly higher than that of control plots, however, there were not statistically significant differences in m ortality rates between
42 insecticides ( Fig. 2 2). During the final two caging dates in April 2012, t here were no significant differences in D. citri mortality between treatments both 15 and 22 DAT (Table 2 2). In September 2012, the only date in which there w ere significant differences between treatments in D. citri mortality was at 1 DAT (Table 2 2). At 1 DAT, D. citri mortality was significantly higher for all insecticides compared to control plots but there was no significant difference in mortality between insecticides (Fig. 2 3). Overall, mortality rates for all insecticides 1 DAT were also much lower (<70%) compared to other months in which trials were carried out (Figs. 2 7, 2 8, 2 9). This could be attributed to greater amounts of rainfall recorded dur ing the month of September (Table 2 1), especially since there was so much rainfall during the week of caging associated with 1 DAT (Figure 2 6). November 2012 was the only month when there were significant differences in D. citri mortality between treat ments on all 4 caging dates 1, 8, 15 and 22 DAT (Table 2 2). At 1 DAT, D. citri mortality was significantly higher for all insecticides compared to the control plots with no significant difference between insecticides. Psyllid mortality rates were approxim ately 90% or greater for all insecticides (Fig. 2 4). By 8 DAT, overall psyllid mortality rates dropped to less than 50%. While there were no significant differences between mortality rates between insecticides, mortality rates in imidacloprid and fenpropa thrin treated plots were still significantly higher than the control plots (Fig. 2 4). A similar trend in mortality rates was present 15 DAT with imidacloprid and fenpropathrin having significantly higher mortality rates compared to the control and dimetho ate as well (Fig. 2 4). At 22 DAT, only fenpropathrin was significantly different
43 from the control but was not statistically different from the other insecticides examined (Fig. 2 4). The statistical model that included log of DAT and rainfall as variable s to attribute for the mortality observed showed rainfall to be a significant factor for all treatments ( except the control ) and for all months tested. The associated P values were all less than 0.0001 for both variables, and the R 2 values were improved c ompared to other models (Table 2 3, model 4). This would suggest that the presence or absence of rainfall (whatever the case was for the month tested) had an influence on the variability observed that month. All of the models utilized log of DAT as a var iable, because the majority of the variation demonstrated by residual activity could be explained by the number days since the treatment was made. There was also a model that used only log of DAT as a variable, which was a significant variable in all treat ments except the control (Table 2 3, model 3). When the log of maximum temperature recorded at 60 cm was added to the model, it was a significant model for all treatments except the control ( P < 0.0001). The variable of temperature was most significant f or the fenpropathrin treatment ( P = 0.0005) and increased the R 2 for this model. It was significant ( P = 0.0389) albeit less, for imidacloprid as well (Table 2 3, model 1). Another model that showed significance was the addition of Julian day to log of DAT. This model was significant for all the treatments, but the Julian day variable was most significant for the fenpropathrin treatment ( P = 0.0183) with an increase in the R 2 value (Table 2 3, model 2). Overall the models that had the most significan t effect on D. citri mortality for insecticides examined were the log of DAT alone, and the log of DAT with rainfall.
44 Discussion The results of this study suggest that the time of year in which an insecticide is applied will have little impact on the amou nt of control of D. citri a grower achieves. Immediately following application, all insecticides provided a similar level of D. citri control, regardless of the season of the year in which they were applied. By 8 DAT, while there were statistically signifi cant differences in D. citri mortality between some treatments which varied throughout the year, the actual level of control provided (<50% on average) was far less than a grower would deem as acceptable control. Thus, i t would be more important for a grow er to be mindful of insecticide resistance management and therefore ensure proper rotation of pesticide modes of action than to be concerned with apply ing a given pesticide class based on the time of year the application is being made. That being said, dim ethoate proved to have less residual activity in every month. There was never a situation in which dimethoate treatments had the highest average mortality at 8 to 22 DAT; however, during January and April 2012 there was no significant difference between d imethoate and imidacloprid, and between dimethoate and fenpropathrin, respectively. This would suggest that while dimethoate may not demonstrate much residual activity, it would be difficult to predict which of the other two insecticides tested would prov e to be more effective than the other. In fact, that was the case that was observed in each month, there was no clear fenpropathrin. Again, this is when spray schedules and resistance management would be the most important factor for selecting an insecticide, as imidacloprid is also used for young tree protection as a drench application, it might be necessary for a grower to choose dimethoate even if it has not proven to be effective for very long periods of time.
45 These data do not support or dispute what was discovered in the previous laboratory experiments regarding temperature correlation. It will be difficult to compare these results with laboratory results because there are many more variables in the field than what occurs in the lab. In addition, temperature experiments in the lab were conducted at constant temperature, whereas in the field there are natural fluctuations that will occur throughout the day and nigh t. When considering these results and the lab results, one also should consider that the lab experiments were observing initial LC 50 whereas this experiment observed percent mortality as it decreased over time. The field results showed that initial knock down created by the insecticides showed no statistically significant differences. However, these results would suggest that the results attained from the lab experiments are not applicable to prediction of toxicity behavior of insecticides in the field o ver time. Temperature showed to have a significant effect on the imidacloprid and fenpropathrin treatments, but not dimethoate and control treatments. Even with this information, it is difficult to say how temperature affected these treatments. However, it is likely that some of the variability in our trials can be explained by temperature variations. Figure 2 10 compares the averages of the minimum, maximum, and average temperatures during the first week of caging and shows a somewhat bell shaped relati onship of temperatures throughout the year. January 2013 shows slightly higher minimum and average temperatures than those observed in January 2012, which we can only speculate if this explains the differences observed in the average mortality rates betwe en these months. Because winter dormancy sprays are so important to target over wintering ACP, it is important to note that the efficacy of all the insecticides was greater and longer in
46 these months than in the summer months. Other variables that have b een collected but have yet to be considered for their impact on residual activity include relative humidity, dew point, and solar radiation (Table 2 1). Further research could include additional replications of each season to see if there is a pattern th at can be observed. When these replications are conducted it could be of benefit to be certain the applications of the insecticides are without fault; for example, placing cards in the trees to document spray coverage, as well as calibrating the sprayer p rior to each use. In addition, in order to be able to make better predictions about how this would apply to growers, sampling the natural populations when cages activ ity lasts (caging) versus how long it takes for psyllids to re infest (natural population). Based on th ese data, lab experiments could be conducted to determine if the difference between the actual mortality and re infestation is caused by sub lethal or r epellency type effects, and if so, how long these effects could last as well as what level of residue is required to observe this. Additionally, experiments on the amount of necessary drying time for complete rainfastness could be conducted to clarify the results observed with rainfall effects.
47 Table 2 1. Weather data collected during caging periods. Compiled from Florida automated weather n etwork (FAWN). The minimum and maximum temperatures were the averages of the minimums and maximums observed duri ng the specified caging period. Dates Rainfall (cm) Average temperature (C) Minimum temperature (C) Maximum temperature (C) Average relative humidity (%) Average solar radiation (w/m 2 ) January 24 31, 2012 0.1778 18.16 7.63 29.94 77 167.46 January 31 February 7, 2012 0 19.81 9.67 29.58 80 146.54 February 7 14, 2012 1.0414 15.01 0.62 28.25 72 171.2 February 14 21, 2012 0.4826 19.12 8.37 30.38 78 164.5 April 10 17, 2012 0 22.55 13.97 31.87 66 247.83 April 17 24, 2012 1.397 21.14 8.14 31.76 68 247.65 April 24 May 1, 2012 0.0762 22.5 8.14 33.02 65 301.67 May 1 8, 2012 0.1524 25.7 18.91 35.02 71 284.96 September 13 20, 2012 2.9718 25.3 20.96 34.5 85 166.9 September 20 27, 2012 2.3114 24.76 17.79 33.85 81 180.58 September 27 October 4, 2012 0.7366 2 5.71 21.01 34.43 85 183.44 November 1 8, 2012 0.0254 17.76 7.26 30.28 75 169.12 November 8 15, 2012 0 17.58 5.76 28.83 78 140.32 November 15 22, 2012 0.0254 15.96 6.25 26.46 81 120.59 November 22 29, 2012 0 15.28 5 28.3 72 160.36 January 9 16, 2013 0. 8636 21.37 14.87 29.81 82 139.98 January 16 23, 2013 0.5334 16.09 3.62 29.4 76 114.53 January 23 30, 2013 0.1778 17.29 4.33 29.65 75 164.4 January 30 February 6, 2013 0 15.06 3.01 29.61 65 179.38
48 Table 2 2. ANOVA results for adult D.citri mortality f or insecticide treatments testing seasonal effects on insecticide applications for D.citri control Date DAT DF F P>F R 2 1 3, 16 133.68 0.0001 0.96 8 3, 16 17.62 0.0001 0.76 15 3, 15 2.85 0.07 0.36 22 3, 16 0.4 0.76 0.069 1 3, 16 9.920 0.0006 9.920 8 3, 16 7.00 0.0032 0.56 15 3, 16 0.47 0.7067 0.0811 22 3, 16 1.38 0.2845 0.2845 1 3, 16 18.72 0.0001 0.77 8 3, 16 0.130 0.94 0.023 15 3, 16 0.95 0.94 0.1512 1 3, 15 37.28 0.0001 0.88 8 3, 15 7.88 0.0022 0.611 15 3, 15 12.61 0.0002 0.716 22 3, 15 4.95 0.0139 0.4975 1 3, 16 10.90 0.0004 0.6715 8 3, 16 8.9 0.0011 0.62 15 3, 16 1.32 0.303 0.1982 22 3, 16 1.75 0.1976 0.2468 Table 2 3 ANOVA results for comparing different approaches to the analysis of the insecticide effects Model Treatment Source DF Sums of squares F P>F R 2 1 Imidacloprid Model 2 1.2961293 111.9 <0.0001 0.708671 Error 92 0.53282725 Corrected Total 94 1.82895655 Log DAT 1 1.29591752 223.76 <0.0001 Log T 1 0.0254234 4.39 0.0389 Dimethoate Model 2 1.83530045 134.82 <0.0001 0.745606 Error 92 0.62618662 Corrected Total 94 2.46148706 Log DAT 1 1.81109363 266.09 <0.0001 Log T 1 0.00409394 0.6 0.44 Fenpropathrin Model 2 1.03 378568 63.59 <0.0001 0.580271 Error 92 0.74777105 Corrected Total 94 1.78155673 Log DAT 1 1.00949453 124.2 <0.0001 Log T 1 0.10707378 13.17 0.0005
49 Table 2 3 Continued Untreated Model 2 0.00944905 0.72 0.4878 0.016365 Error 87 0.56795738 Corrected Total 89 0.57740644 Log DAT 1 0.00889459 1.36 0.2463 Log T 1 0.00006022 0.01 0.9237 2 Imidacloprid Model 2 1.28223711 107.89 <0.0001 0.701076 Error 92 0.54671944 Corrected Total 94 1.82895655 Log Julian Day 1 0.0115312 1.94 0.167 Log DAT 1 1.23627502 208.04 <0.0001 Dimethoate Model 2 1.84596728 137.96 <0.0001 0.74994 Error 92 0.61551978 Corrected Total 94 2.46148706 Log Julian Day 1 0.01476077 2.21 0.1409 Log DAT 1 1 .79982709 269.02 <0.0001 Fenpropathrin Model 2 0.97718749 55.88 <0.0001 0.548502 Error 92 0.80436924 Corrected Total 94 1.78155673 Log Julian Day 1 0.05047558 5.77 0.0183 Log DAT 1 0.89784589 102.69 <0.0001 Untreated Model 2 0. 04492669 3.67 0.0295 0.077808 Error 87 0.53247975 Corrected Total 89 0.57740644 Log Julian Day 1 0.03553786 5.81 0.0181 Log DAT 1 0.00683395 1.12 0.2936 3 Imidacloprid Model 1 1.2707059 211.69 <0.0001 0.694771 Error 93 0.55825064 Corrected Total 94 1.82895655 Log DAT 1 1.2707059 1.27 < 0 .0001 Dimethoate Model 1 1.83120651 270.2 <0.0001 0.743943 Error 93 0.63028055 Corrected Total 94 2.46148706 Log DAT 1 1.83120651 270.2 <0.0001 Fenpropathrin M odel 1 0.9267119 100.82 <0.0001 0.52017 Error 93 0.85484483 Corrected Total 94 1.78155673 Log DAT 1 0.9267119 100.82 <0.0001
50 Table 2 3 Continued Untreated Model 1 0.00938883 1.45 0.231 0.01626 Error 88 0.5680176 Corrected T otal 89 0.57740644 Log DAT 1 0.00938883 1.45 0.231 4 Imidacloprid Model 2 1.38296064 142.64 <0.0001 0.756147 Error 92 0.44599591 Corrected Total 94 1.82895655 Log DAT 1 1.37850503 284.36 <0.0001 Rainfall 1 0.11225474 23.16 < 0.0001 Dimethoate Model 2 1.94451963 173.02 <0.0001 0.789978 Error 92 0.51696744 Corrected Total 94 2.46148706 Log DAT 1 1.94394785 345.95 <0.0001 Rainfall 1 0.11331312 20.17 <0.0001 Fenpropathrin Model 2 1.16474677 86.86 <0.00 01 0.65378 Error 92 0.61680995 Corrected Total 94 1.78155673 Log DAT 1 1.11798581 166.75 <0.0001 Rainfall 1 0.23803487 35.5 <0.0001 Untreated Model 2 0.01978361 1.54 0.2195 0.034263 Error 87 0.55762282 Corrected Total 89 0 .57740644 Log DAT 1 0.01416873 2.21 0.1407 Rainfall 1 0.01039478 1.62 0.2062
51 Figure 2 1. Mean ( SE) percent mortality of D. citri adults caged on trees each week following an application of insecticides in January 2012. Within each sam pling P > 0.05). Figure 2 2. Mean ( SE) percent mortality of D. citri adults caged on trees each week following an application of insecticides in April 2012. Within each sampli ng
52 P > 0.05).
53 Figure 2 3. Mean ( SE) percent mortality of D. citri adults caged on trees each week following an application of insecticides in September 2012. Within each sa HSD, P > 0.05).
54 Figure 2 4. Mean ( SE) percent mortality of D. citri adults caged on trees each week following an application of insecticides in November 2012. Within eac h HSD, P > 0.05). Figure 2 5. Mean ( SE) percent mortality of D. citri adults caged on trees each week following an application of insecticides in January 2013. Within eac h sampling
55 P > 0.05). Figure 2 6. Rainfall recorded during the caging period 1 day after treatment for eac h month. Compiled from Florida automated weather n etwork (FAWN). Figure 2 7. Mean ( SE) percent mortality of adult D. citri caged on imidacloprid treated trees 1 DAT for each month trials were conducted.
56 Figure 2 8. Mean ( SE) percent mortality of adult D. citri caged on dimethoate treated trees 1 DAT for each month trials were conducted. Figure 2 9. Mean ( SE) percent mortality of adult D. citri caged on fenpropathrin treated trees 1 DAT for each month trials were conducted.
57 Figure 2 10. Temperatures recorded during the first wee k psyllids were caged on insecticide treated and untreated trees each month. Weather data compiled from the Florida automated weather n etwork.
58 CHAPTER 3 ADDITION OF SURFACTANTS TO INSECTICIDES USED TO MANAGE D. CITRI Introduction Adjuvants are a broad cat egory of materials that can be added to pesticides in order to give additional properties that are often added to help with spray application problems. These properties include buffering, spreading, sticking, emulsifying, dispersing, wetting, and can redu ce evaporation, foaming, spray drift and volatilization of spray materials. Activator adjuvants include surfactants, oils, and nitrogen based surf ace act ing a ge nts Surfactants improve the wetting and spread ing properties of the pesticide material, meaning that they help to wet the foliage and spread the material on the leaf better. This helps to improve the biological activity of the pesticide in that it has a better chance of an insect coming into contact w ith the material. The surfactants tested in this study included a methylated seed oil (MSO concentrate, Loveland Products Inc., Greely, CO.), a nonionic surfactant (Induce, Helena Chemical Company, Collierville, TN), a petroleum oil (IAP 435 All Purpose sp ray oil, Independent Agribusiness Professionals, Fresno, CA), and an organosilicone surfactant (Freeway, Loveland Products Inc, Greely, CO). M ethylated seed oils (MSOs) are part of a category of surfactants called v egetable oil concentrates (VOCs) that co nsist of 80 to 85 percent of some type of seed oil that is crop derived, such as cotton, linseed, or sunflower oil. The remaining 15 to 20 percent of the formulation is non ionic surfactant. MSOs have undergone a process called esterification to increase their penetrating ability. Non ionic and organosilicone are classified as such because of the way in which the ions separate, or ionize in the material The classifications are cationic (positive charge), anionic (negative charge),
59 non ionic (no charge ), and organosilicone surfactants. Non ionic surfactants help the product penetrate the plant and therefore are often used with systemic pesticides. Organosilicone surfactants help with spreading, reduces the surface tension of the liquid material, and im proves rainfastness. Petroleum oils are part of the broader category of horticultural oils. They are refined petroleum products that are applied at a dilution in water of 2% material added to the volume of water. Historically, they have been used for th breathing. This makes them useful because it is a physical aspect to their toxicity and not something to which insects can become resistant. Several adjuvants have been studied for their effects against ACP based on their effects on other insect pests. This includes a specific organosilicone, Silwet L 77, which has demonstrated toxicity against many other arthopods (Srinivasan et al. 2008); and 435 petroleum oil, which is known to be toxic to small and immobile arthropod pests ( similar to the nymphal stage of ACP) due to its suffocating effects (Rae et al. 2010). MSO and Induce (a non ionic surfactant) ha ve been previously tested for efficacy of D. citri and found that MSO and 435 petroleum oil provided the best coverage when combined with Movento (spirotetramat, Bayer CropScience, Research Triangle Park, North Carolina) (Qureshi et al. 2009) Residual control of psyllids provided by foliar applications of pesticides is probabl y much shorter than previously believed (see Chapter 2). Any means by which growers can extend the residual duration of psyllid control provided by foliar sprays will be of value. Thus, t he purpose of this study was to determine if the addition of
60 surfacta nts with two insecticides (dimethoate and fenpropathrin) commonly applied as foliar sprays for psyllid control, will improve the efficacy and duration of control provided. Materials and Methods Insecticide Application Experimental plots were established i n July of 2012 in a 40 acre grove of mature Valencia on Swingle trees located in Lake Placid, FL The applications were made beginning July 9 and were completed by July 11 Plots consisted of 50 trees ( 5 rows by 10 trees in length) with 5 r eplicate plots per treatment. Treatments were applied in a spray volume of 100 GPA (935 L/Ha) using a standard airblast sprayer with a 500 gallon (1,895 L) tank (Durran d Wayland, Inc. LaGrange, GA) The two insecticides chosen for evaluation in this trial included feprop athrin (Danitol 2.4 EC, Valent USA Corporation, Walnut Creek, CA) at a rate of 1.168L/ha and dimethoate (Dimethoate 4E, Cheminova Inc., Research Triangle Park, NC) at a rate of 1.168L/ha Both insecticides were applied alone and in combination with t he fo llowing adjuvants: methylated seed oil (MSO concentrate, Loveland Products Inc., Greely, CO.), a nonionic surfactant (Induce, Helena Chemical Company, Collierville, TN), a petroleum oil (IAP 435 All Purpose spray oil, Independent Agribusiness Professionals Fresno, CA), and an organosilicone surfactant (Freeway, Loveland Products Inc, Greely, CO). There were also 5 replicate plots left untreated as a control for comparison. Caging of Adult D. citri The day following the completion of insecticide treatment s, caging of adult D. citri began to assess efficacy of the treatments. Since each plot had 5 rows, caging was conducted on the middle row of trees to reduce the possibility of contamination (via pesticide drift) from neighboring treatments. For each plot, 20 adult psyllids were caged
61 on on branch of two separate trees using sleeve cages as previously described in Chapter 2. The psyllids remained caged on the trees for one week, after which time the branch was cut off the tree just above the net sleeve so t hat the branch with the caged ACP could be returned to the lab to assess psyllid mortality As the branches were cut off, new sleeve cages containing psyllids were placed on each tree for a second week of evaluations This was repeated until 4 weeks of dat a (22 DAT) were collected. Statistical Analyses For each sleeve cage returned to the lab, psyllid mortality was assessed by making counts of the number of live and dead psyllids in each cage. P ercent mortality was determined by add ing the number of live a nd dead psyllids and dividing the number of dead psyllids. The percent mortality had to be weighted because there were some cages that either had more or fewer than 20 psyllids. The weighting was used to ensure that samples with more psyllids did not inf luence the data more than the samples with the intended amount. The weighting was achieved by dividing the total number of live Each replication had two pseudo repli cations, so the weights for each were averaged as were the percent mortality for each. Statistical Analysis System (SAS) was used to transform the average mortality using an Arcsin transformation. Proc GLM was used used to compare the means. Results There were significant differences in D. citri mortality between treatments at 1 DAT (Table 3 1). For the Danitol treatments, the addition of adjuvants did not improve efficacy with none of the treatments providing signi ficantly better control than Danitol alone (Fig. 3 1). However, mortality of D. citri in Danitol + Freeway treatments were
62 significantly lower than Danitol alone and not different from the untreated control ( Fig. 3 1). Similarly, for dimethoate, the addit ion of adjuvants did not improve control of D. citri There were no significant differences between any of the dimethoate treatments, all of which were not significantly different from D. citri mortality observed in the untreated plots (Fig. 3 1). During t he remaining caging periods (8 22 DAT ) there were no significant differences between any of the treatments (Table 3 1). As shown in Figures 3 2, 3 3 and 3 4, the mean percent mortality continued to decrease over time with no significant effects between t he treatments. Discussion The results gained from this field trial do not provide adequate information to draw conclusions. The results did not show differences between treatments in a way that would allow conclusions to be determined about the efficacy of specific insecticide and surfactant combinations. This could be because of unobserved problems with the application s or perhaps due to weather conditions However, the results were somewhat comparable to the results described in Chapter 2 in which the initial control was good, but there was a lack of residual activity. It is possible that the results of this trial are indeed accurate and suggest that the adjuvants tested have little effect for increasing the true residual control of D. citri provided by these insecticides. In order to gain more clarity the experiment should be repeated several times, and also under different conditions. In further studies, the organosilicone surfactant, Freeway, should be replaced with the organosilicone surfactant Sil wet L 77. Silwet L 77 is the only organosilicone surfactant currently known to have insecticidal activity against ACP because it is part of a class of surfactants called trisiloaxane surfactants that have shown activity like that of hyperactive soaps that essentially make it so the insect cannot
63 breathe by infiltrating the respiratory system and blocking the gaseous exchange It would also be beneficial to use smaller plot sizes to reduce the time it takes to make the insecticide applications, thereby pos sibly reducing unintended variability that may have occurred by applying over three days. Additionally, use of imidacloprid with these surfactants could be useful as well, since some of the surfactants are intended to increase penetrability and imidaclopr id has shown translaminar effects. The benefits of this addition to experiment would have to be weighed, since imidacloprid has already been tested with some of the surfactants; however it has not been tested in this manner or in comparison with these co mmonly used insecticide classes. The problem associated with this addition to the experiment would be that this would cause the trial to be a very large field trial, and therefore require large amounts of ACP to be reared or collected.
64 Table 3 1. ANOVA results for adult D.citri mortality for insecticide treatments testing addition of surfactants to insecticides used to manage D. citri Date DAT DF F P>F R 2 July 19, 2012 1 10, 42 7.72 <.0001 0.6477 July 26, 2012 8 10, 41 0.92 0.5248 0.18325 August 2, 2012 15 10, 42 0.5 0.88 0.10645 August 8, 2012 22 10, 43 1.53 0.1622 0.26233 Figure 3 1. Mean ( SE) percent mortality of adult D. citri caged on trees 1 day after P > 0 .05).
65 Figure 3 2. Mean ( SE) percent mortality of adult D. citri caged on trees 8 days after P > 0.05).
66 Figure 3 3. Mean ( SE) percent mortality of adult D. citri cag ed on trees 15 days after P > 0.05).
67 Figure 3 4. Mean ( SE) percent mortality of adult D. citri caged on trees 22 days after treatment. Means with the same letter do not d P > 0.05).
68 CHAPTER 4 COMPARISON OF LOW VOLUME APPLICATION OF INSECTICIDES TO STANDARD AIRBLAST APPLICATION OF INSECTICIDES TO MANAGE D. CITRI. Introduction Because D. citri are difficult to control, growers have had to r esort to multiple, back to back applications (between 6 15 applications per year in Florida) of insecticides to attempt to attain lowest possible population levels (Grafton Cardwell et al. 2013). This can become very expensive for growers using traditiona l application methods. Standard airblast equipment typically applies around 100 125 gallons per acre (935 1,168 L per hectare) of water (or oil). Most standard airblast applicators used currently hold 500 gallons (4,675 L). This makes the application ta ke a long time, as they person applying will have to go back and refill the tank approximately every 5 acres, which can mean a lot of return trips if the acreage to treat is large. These applications also have to be made at 1 2 miles per hour. Not only d oes it take a lot of time, but it will have a high cost because of fuel costs (number of trips to back to the water source and also more fuel to transport a larger load) and the cost of paying someone to make the time consuming application. To overcome t hese obstacles, growers have beg u n to make applications at lower volumes, with the same amount of insecticide added in. This prompted application for label changes in order to make these applications legal. There is now a Special Local Needs permit that a llows low volume applications because of the threat of citrus greening; however it specifies that the spray droplet size must be above 90 m to help reduce the risk of drift (Hoffman et al. 2010). Also to combat drift problems, low volume applications are made at night, when wind is reduced. Low volume applications can go as low as 2 gallons per acre (19 L/hectare), which amounts to lower cost from less
69 water, to less fuel from a lighter load and less trips to the water source. There have been reports fr om some applicators that they have been able to cover 250 acres (101 hectares) in a single night when weather conditions made for perfect application conditions (Atwood and Stelinski 2008). There are many benefits of utilizing low volume application techno logy, but it is important to determine if its usage results in similar control of D. citri as is associated with standard airblast technology. Growers have observed that low volume applications provide similar psyllid control to traditional applications ( Atwood and Stelinski 2008). In a study conducted in 2008 that compared three different LV sprayers and a standard airblast application for efficacy no significant difference was found between the different application technologies. Because other studies have shown that droplet size can affect toxicity of insecticides, but also differs based on the insecticide used, Boina et al (2009) tested the effect of droplet size of fenpropathrin on D. citri and determined it to be more effective at smaller droplet s izes (40 50 m). This was conducted in laboratory settings. The purpose of this study was to observe the actual residual activity associated with both types of spray equipment technology, using three commonly used insecticides with different modes of ac tion under field conditions. To gain actual residual activity, instead of observing when psyllids re colonize the treated area, ACP were caged on the trees so that mortality could be directly assessed This type of experiment allows for a different kind of observation on the effect of an insecticide treatment on ACP, as it looks at the toxicity of the material on the leaf as it is being exposed to field conditions.
70 Materials and Methods Insecticide Applications This experiment was conducted once in Octob er of 2012, and then repeated in December of 2012. Both field trials were conducted in a 40 acre grove in Lake Placid, FL on mature Valencia on Swingle citrus trees. For both experiments the insecticides used were fepropathrin (Danitol 2.4 EC, Valent US A Corporation, Walnut Creek, CA) at a rate of 1.168L/ha dimethoate (Dimethoate 4E, Cheminova Inc., Research Triangle Park, NC) at a rate of 1.168L/ha and imidacloprid (Provado 1.6F, Bayer CropScience LP, Research Triangle Park, North Carolina) at a rate of 0.73 L/ha The traditional applications were made in 100 gallons of water per acre using a standard airblast sprayer with a 500 gallon tank (Durran d Wayland, Inc. LaGrange, GA) T his equipment was used both in October and December. Two different types of low volume application equipment were used for the two experiments. The October 2012 experiment used a truck mounted LV sprayer Dyna Fog Ag Mister LV 8 (Curtis Dyna Fog, Westfield, IN). The LV sprayer had four 5.5 gallon tanks, and had a spray boom wit h eight nozzles, four on each side of the boom stacked vertically. The LV applications were applied at approximately 7 gallons per acre, with the same amount of insecticide as listed above for the standard airblast application This experiment was conduc ted on 40 acres, but only approximately 28 acres were utilized. The plots were 6 rows across, and consisted of half the row being one plot, and half the row being the next treatment plot (each row has 105 trees). Therefore, the plots were 6 trees across by 52 trees long This allowed each plot to be approximately 2 acres. In order to get five replications, e ach treatment had two plots, one plot having 2 replications and the other plot having 3 replications. All 6 rows were not treated; the
71 outside rows were left for buffer as LV has the capacity to drift, and contamination between treatments was undesired. The airblast treatments were made during the day, and because conditions were calm, the LV application s were made in the early evening. LV applicati on speeds were approximately 5 miles per hour, and standard airblast speeds were approximately 2.5 miles per hour. The treatments were: fenpropathrin applied with LV, dimethoate applied with LV, imidacloprid applied with LV, fenpropathrin applied with sta ndard airblast, dimethoate applied with standard airblast, imidacloprid applied with standard airblast and untreated control. Caging of Adult D. citri Insecticide applications were allowed to dry overnight and the following day adult ACP were caged on tw o interior trees in the plots. For each plot, 20 adult psyllids were caged on one branch of two separate trees using sleeve cages as previously described in Chapter 2. ACP were left on the branches within the cages for three days (Wednesday to Friday) to allow them to be exposed to the insecticide treatments. On the third day, the branches were cut off with clippers and placed in coolers to be transported. Once back at the lab, the number of live and dead ACP were counted and recorded for each cage. At the same time as the branches were being removed from trees, new cages with 20 live ACP were caged on the same trees. For this data collection, the ACP were retained on trees for four days (Friday to Tuesday), and the branches were removed using the same process detailed above, and live and dead psyllids were counted and recorded again. When ACP were removed on Tuesday, they were again replaced by new ACP and cages and remained on the trees for three days (Tuesday to Friday). On Friday, the branches wer e removed for the last time, and no new ACP were caged.
72 This experiment was repeated in December 2012 due to incongruous results from the October field trial. Some changes were made in this trial, including the plot layout and the LV sprayer used, as wel l as timing of application. A P600 series Proptec LV sprayer (Curtec, Vero Beach, FL) was used. It was tractor pulled, and had a 100 gallon tank attached, and the sprayer was calibrated to apply 8 gallons per acre. The same standard airblast sprayer was used as the October field trial. For this experiment, there were 5 replicate plots for each treatment, with 2 pseudo replications for each plot. These plots consisted of 6 rows across by 15 trees long, and the interior rows were the only ones treated (t he row middle then the other side of the middle was treated on both sides). This made each plot size approximately 0 .65 acres, and a total of 31 acres of the 40 acre block were used. The same block that consisted of mature Valencia on Swingle trees was u sed. The same method of caging that was previously described was used. For this experiment, the days of caging were slightly different. Because of the large amount of psyllids needed for each caging date (1,400 ACP), average percent mortality was calcul ated on the same day the branches were removed prior to caging new psyllids. This allowed for comparisons to be made between the treatments, so that as soon as the treatments did not show any significant difference from one another the trial could be stop ped. Thus, if there was still a significant difference, ACP would be caged the day after the branches had been removed and mortality counted. Statistical Analyses When psyllids were counted at the end of each week, live and dead psyllids were recorded. Using this information a percent mortality was gained by add ing the number of live and dead psyllids and dividing the number of dead psyllids into that total. The percent mortality had to be weighted because there were some cages that either had
73 more or l ess than 20 psyllids. The weighting was used to ensure that samples with more psyllids did not influence the data more than the samples with the correct amount. The weighting was achieved by dividing the total number of live and dead psyllids of each tre had two pseudo replications, so the weights for each were averaged as were the percent mortality for each. Statistical Analysis System (SAS) was used to transform the average mo Studentized Range test was used to compare the means. Results In the trial conducted in October, the re were significant differences between the treatments 1 DAT (Table 4 1). For Danito l treatments, low volume application provided significantly higher rates of mortality compared to the untreated control (50% vs. <10%) but was not significantly different from airblast application of Danitol, the latter of which was not different from the control (Fig. 4 1). Regardless of application method, neither dimethoate nor imidacloprid were significantly different from the untreated control 1 DAT (Fig. 4 1). At 5 DAT there were no significant differences between any treatments (Table 4 1) with perce nt mortality less than 40% for all treatments evaluated (Fig. 4 2). While there were significant differences at the 8 DAT caging (Table 4 1), the difference is probably not meaningful as the trend was for significantly higher rates of mortality across all treatments (including the untreated control) with the exception of the low volume dimethoate application (Fig. 4 3). In the December trial the re were significant differences between treatments at 1, 4 and 7 DAT (Table 4 1). At 1 DAT, D. citri mortality was significantly higher for both low volume and airblast applications of dimethoate compared to the untreated control
74 (Fig. 4 4). While mortality was generally higher in both Danitol treatments, neither Danitol treatment was significantly different from t he untreated control, as was the low volume application of imidacloprid 1 DAT. For imidacloprid, only the airblast application was significantly different from the untreat ed control but did not differ significantly from the low volume application of imidac loprid (Fig. 4 4). At 4 DAT, all insecticide treatments provided significantly higher mortality (75 85%) compared to the untreated control (< 10%), but there were no differences between insecticides, regardless of method of application used (Fig. 4 5). At 7 DAT, the mortality of D. citri dropped to less than 40% for all treatments examined (Fig. 4 6). There was no significant difference between the Discussion The results obtained in the October experim ent were extremely unusual and difficult to gain insight from, which was the reason for repeating the experiment in December. To begin with, the initial knockdown (or percent mortality at 1 DAT) of all insecticide treatments was much lower (<50%) than nor mally observed. The standard airblast results observed in this experiment were much lower than those found during the seasonality results discussed in Chapter 2 In addition, the average percent mortality for both imidacloprid application types was below 10%, which is extremely low (Figure 4 1). Even more unusual, the following caging period showed an increase in average percent mortality for both application types of imidacloprid, to between 10 20% (Figure 4 2). This could have been because of the observ ed differences in natural psyllid death while in cages, which has been shown to vary in the control treatments in all cases. The following caging period also had confounding results, and showed an extremely high control mortality rate, which was higher th an all the mortality rates for the
75 insecticide treatments in this caging period. Again, there was an incredible increase in imidacloprid mortality for both application types during this last caging period. It is difficult to conclude much from this incre ase, as the mortality rates in these treatments were close to the mortality rates observed in the high control mortality rates (Figure 4 3). Essentially, these results give no conclusive information about any of the treatments tested. In December, the i nitial knockdown results were still slightly lower that what has been discussed in Chapter 2 While the results obtained at 1 DAT are much easier to understand than those gained in the October trial, they still do not provide a clear cut conclusion to be drawn about one application method being superior to the other (Figure 4 4). At 4 DAT, there was no difference between the treatments observed which suggests that at this time period, the residues from either application method were providing equivalent le vels of residual control of D. citri (Figure 4 5). By the third caging period (7 DAT), there was no difference between the treatments and the control (Figure 4 6). M ortality rates dropped significantly from 4 DAT which could have been a result of rainf all during this period which totaled 7.874 cm during those 3 days. This wou ld support the findings for Chap ter 2 which suggest that rainfall is an important factor affecting residual activity of insecticides.
76 Table 4 1. ANOVA results for adult D.citri mortality for insecticide treatments using low volume vs airblast application equipment applied during October and December Date DAT DF F P>F R2 10/12/2012 1 6, 28 6.2 0.0003 0.570744 10/16/2012 5 6, 28 2.39 0.0543 0.338793 10/19/2012 8 6, 27 3.87 0.00 65 0.462194 12/7/2012 1 6, 21 6.64 0.0005 0.654788 12/10/2012 4 6, 21 7.04 0.0003 0.668038 12/13/2012 7 6, 21 3.41 0.0166 0.493297 Figure 4 1. Mean ( SE) percent mortality of adult D. citri caged on trees 1, 5, and 8 DAT with insecticides using l ow volume vs airblast application equipment applied in October. Within each sampling date, m eans with the same letter do P < 0.05).
77 Figure 4 2 Mean ( SE) percent mortality of adult D. citri caged on trees 1, 4, and 7 DAT with insecticides using low volume vs airblast application equipment applied in December. Within each sampling date, m eans with the same letter P < 0.05).
78 LIST OF REFERENCES Arakawa, K., and K. Miyamoto 2007. Flight ability of Asiatic citrus psyllid, Diaphorina citri Kuwayama (Homoptera: Psyllidae), measured by a flight mill. Res. Bull. Plant Prot. Jpn. 42: 23 26. Atwood, R. and L. Stelinski. 2008. Is there a future for low volume application for psyll id control? Citrus Ind. 89(11): 16. Avery, P. B., V. W. Wekesa, W. B. Hunter, D. G. Hall, C. L. McKenzie, L. S. Osborne, C. A. Powell, and M. E. Rogers. 2011. Effects of the fungus Isaria fumosorosea (Hypocreales: Cordycipitaceae) on reduced feeding and mo rtality of the Asian citrus psyllid, Diaphorina citri (Hemiptera: Psyllidae). Biocontrol Sci. Tech 21(9): 1065 1078. Beattie, G. A. C. an d P. Holford. 2009 Current HLB situation and industry perspective in Asia. Proceedings of the International Research C onference on Huanglongbing, Dec. 2008. 1: 94. Belasque Jr. J., R. B. Bassanezi, P. T. Yamamoto, S. A. Lopes, A. J. Ayres, J. C. Barbosa, A. Tachibana, A. R. Violante, A. Tank Jr., C. L. Giorgetti, F. Di Giorgi, F. E. A. Tersi, G. M. Menezes, J. Dragone, L F. Catapani, R. H. Jank Jr., and J. M. Bove. 2009. Factors associated with control of huanglongbing in Sao Paulo, Brazil: a case study. Proceedings of the International Research Conference on Huanglongbing, Dec. 2008. 1: 337 341. Boina, D. R., M. Salyani S. Tiwari, K. Pelz Stelinski, and L. L. Stelinski. 2012. Spray droplet size affects efficacy of fenpropathrin against Asian citrus psyllid. J. ASTM Int. Boina, D.R., W.L. Meyer E.O. Onagbola, and L.L. Stelinski. 2009. Quantifying dispersal of Diaphori na citri (Hemiptera: Psyllidae) by immunomarking and potential impact of unmanaged groves on commercial citrus management. Environ Entomol. 38: 1250 1258. Bonani, J. P., A. Fereres, E. Garzo, M. P. Miranda, B. Appezzato Da Gloria, and J. R. S. Lopes. 2010. Characterization of electrical penetration graphs of the Asian citrus psyllid, Diaphorina citri in sweet orange seedlings. Entomol. Exp. Appl. 134 : 35 49. Bove, J. M. 2006. Huanglongbing: A destructive, newly emerging, century old disease of citrus. J. P lant Pathol. 88 : 7 37. Bukovac, M.J. 1984. Low volume application of plant growth regulators; performance and limitations. Proc. Annu. Meeting Plant Growth Reg. Soc. Amer. 11:143 150.
79 Childers, C. C., and M. E. Rogers. 2005. Chemical control and management approaches of the Asian citrus psyllid, Diaphorina citri Kuwayama (Homoptera: Psyllidae) in Florida citrus. Proc. Fla. State Hort. Soc. 118: 49 53. Chu, Y. I., and C. C. Chien. 1991. Utilization of natural enemies to control psyllid vectors transmitting citrus greening. Integrated control of plant virus diseases. Food and fertilizer technology center for the Asian and Pacific region, Taipei, Taiwan: 135 145. Cocco, A., and M. A. Hoy. 2008. Toxicity of organosilicone adjuvants and selected pesticides to the Asian citrus psyllid (Hemiptera: Psyllidae) and its parasitoid Tamarixia radiata (Hymenoptera: Eulophidae). Fla Entomol 91(4): 610 620. da Graca, J. V. 1991. Citrus greening disease. Annu. Rev. Phytopathol. 29(1): 109 136. Futuko, T. R. 1990. Mechanism of action of organophosphorus and carbamate insecticides. Environ. Health Perspect. 87: 245 254. Gomez Torres, M. L., D. E. Nava, and J. R. Postali Parra. 2012. Life table of Tamarixia radiata (Hymenoptera: Eulophidae) on Diaphorina citri (Hemiptera: Psy llidae) at different temperatures. J. Econ. Ent. 105(2): 338 343. Grafton Cardwell, E. E., L. L. Stelinski, and P. A. Stansly. 2013. Biology and management of Asian citrus psyllid, vector of the Huanglongbing pathogens. Annu. Rev. Entomol. 58: 413 432. Ha lbert, S.E., and K.L. Manjunath. 2004. Asian citrus psyllids (Sternorrhyncha: Psyllidae) and greening disease of citrus: a literature review and assessment of risk in Florida. Fla Entomol. 87: 330 353. Hall, D. G., and R. Nguyen. 2010. Toxicity of pesticid es to Tamarixia radiata, a parasitoid of the Asian citrus psyllid. BioControl 55(5): 601 611. Hall, D. G., M. G. Hentz, J. M. Meyer, A. B. Kriss, T. R. Gottwald, and D. G. Boucias. 2012. Observations on the entomopathogenic fungus Hirsutella citriformis at tacking adult Diaphorina citri (Hemiptera: Psyllidae) in a managed citrus grove. Biol. Control 57(5): 663 675. He, T. 2000. Citrology. Beijing: China Agriculture Press. Hock, W. K., K. H. Richards, and S. I. Gripp. 2011. Pesticide safety fact sheet: Spray adjuvants. Penn State Cooperative Extension Pesticide Education Program. Found at: http://pubs.cas.psu.edu/freepubs/pdfs/uo221.pdf Viewed: May 15, 2013. Code #: UO221. Ichinose, K., K. Miyaz i, K. Matsuhira, K. Yasuda, Y. Sadoyama, D. H. Tuan, and D. V. Bang. 2010. Unreliable pesticide control of the vector psyllid Diaphorina citri
80 (Hemiptera: Psyllidae) for the reduction of microorganism disease transmission. J Environ. Sci. 45(5): 466 472. I nsecticide Resistance Action Committee. 2012. IRAC MoA Classification Scheme. http://www.irac online.org/content/uploads/MoA classification.pdf Viewed: May 15, 2013. Jagouei x, S., J. M. Bove, and M. Garnier. 1994. The phloem limited bacterium of Int. J. Syst. Bact. 44(3): 379 386. Lewis Rosenblum, H. 2011. Movement patterns and long range dispersal of Asian citrus psyllid, Diaphorina citri M.S. Thesis, University of Florida, Florida. L iu Y. H., and J. H. T sai 2000. Effects of temperature on biology and life table parameters of the Asian citrus psyllid, Diaphorina citri Kuwayama (Homoptera: Psyllidae). Ann. Appl. Biol. 137(3): 201 206. Mann, R. S., K. Pelz Stelinski, S. L. Hermann, S. Tiwari, and L. L. Stelinski. 2011. Sexual transmission of plant pathogenic bacterium, Candidatus Liberibacter asiaticus, between conspecific insect vectors during mating. PloS. ONE. 6(12): e29197. Meye r, J. M., M. A. Hoy, and D. G. Boucias. 2007 Morphological and molecular characterization of a Hirsutella species infecting the Asian citrus psyllid, Diaphorina citri Kuwayama (Hemiptera: Psyllidae), in Florida. J Invertebr. Pathol. 95(2): 101 109. Michau d, J. P. 2004. Natural mortality of Asian citrus psyllid (Homoptera: Psyllidae) in central Florida. Biol Control 29(2): 260 269. Musser, F. R., and A. M. Shelton. 2005. The influence of post exposure temperature on the toxicity of insecticides to Ostrinia nubilalis (Lepidoptera: Crambidae). Pest Manag. Sci. 61: 508 510. n, C., B. Fritz, D. Martin, R. Atwood, T. Hurner, M. Ledebuhr, M. Tandy, J. L. Jackson, G. Wisler. 2010. Evaluation of low volume sprayers used in Asian citrus psyllid control applications. Hort Technol 20(3): 632 638. Narahashi, T. 1982. Modification of nerve membrane sodium channels by the insecticide pyrethroids. Comp. Biochem. Physiol. C Comp. Pharmacol. 72(2): 411 414. Nava, D. E., G. M. L. Torres, M. D. L. Rodrigues, J. M. S. Bento, an d J. R. P. Parra. 2007. Biology of Diaphorina citri (Hemiptera, Psyllidae) on different host plants at different temperatures. J. Appl. Entomol. 131:709 15.
81 Nehru, R. K., K. C. Bhagat, and V. K. Koul. 2004. Influence of citrus species on the development of Diaphorina citri. Ann. Plant Proc. Sci. 12: 436 438. Nehru, R. K., K. C. Bhagat, and V. K. Koul. 2004. Influence of citrus species on the development of Diaphorina citri Ann. Plant Prot. Sci. 12: 436 438. Pelz Stelinski, K.S., R.H. Brlansky, T.A. Ebert, and M.E. Rogers. 2010. Transmission parameters for Candidatus Liberibacter asiaticus by Asian citrus psyllid (Hemiptera: Psyllidae). J Econ Entomol. 103: 1531 1541. Qureshi, J. A., and P. A. Stansly. 2007. Integrated approaches for managing the Asian cit rus psyllid Diaphorina citri (Homoptera: Psyllidae) in Florida. Proc. Fla. State Hort. Soc. 120:110 115. Qureshi, J. A., and P. A. Stansly. 2009. Insecticidal control of Asian Citrus Psyllid Diaphorina citri (Hemiptera: Psyllidae). Proc. Fla. State Hort. Soc. 122: 172 175. Qureshi, J. A., and P. A. Stansly. 2010. Dormant season foliar sprays of broad spectrum insecticides: an effective component of integrated management for Diaphorina citri (Hemiptera: Psyllidae) in citrus orchards. Crop Prot. 29: 860 86 6. Qureshi, J. A., and P. A. Stansly. 2010. Dormant season foliar sprays of broad spectrum insecticides: An effective component of integrated management for Diaphorina citri (Hemiptera: Psyllidae) in citrus orchards. Crop Prot. 29(8): 860 866. Qureshi, J. A., B. Kostyk, and P. A. Stansly. 2009. Control of Diaphorina citri (Hemiptera: Psyllidae) with Foliar and Soil Applied Insecticides. Proc. Fla. State Hort. Soc. 122:189 193. Rae, D. J., W. G. Liang, D. M. Watson, G. A. C. Beattie, and M. D. Huang. 1997. Evaluation of petroleum spray oils for control of the Asian citrus psylla Diaphorina citri (Kuwayma) (Hemiptera: Psyllidae), in China. 43(1): 71 75. Rogers, M. E., G. Carlton, and T. D. Riley. 2012. us Industry 93 (7): 12 16. Rogers, M. E., P. A Stansly, and L. L. Stelinski. 2013. Asian citrus psyllid and citrus leafminer, pp. 39 44. In M. M. Dewdney and M. E. Rogers (eds.), Florida Citrus Pest Management Guide. UF/IFAS Extension. Lake Alfred, FL. Ro hrig, E. A., D. G. Hall, J. A. Qureshi, and P. A. Stansly. 2012. Field Release in Florida of Diaphorencyrtus aligarhensis (Hymenoptera: Encyrtidae), an Endoparasitoid of Diaphorina citri (Homoptera: Psyllidae), from Mainland China. Fla. Entomol. 95(2): 47 9 481. Roistacher, C. N. 1991. Techniques for biological detection of specific citrus graft transmissible diseases. pp. 35 45. Greening. Food Agric. Organ., Rome.
82 Sakamaki, Y. 2005. Possible migration of the Asian citrus psyllid Diaphorina citri Kuwayama ( Homptera: Psyllidae), between and within the islands. Occas. Pap. Kagoshima Univ. Res. Cent. 42: 121 125. Salyani, M. 1995. Spray deposition and drift from airblast sprayers used in citrus applications. Citrus Ind. 76(1):35 39. Salyani, M., and C. W. McCo y. 1989. Deposition of different spray volumes on citrus trees. Proc. Fla. State Hort. Soc. 102:32 36. SAS Institute. 2011. Satpute, N. S., S. D. Deshmukh, N.G.V. Rao, S. N. Tikar, M. P. Moharil, and S A. Nimbalkar. 2007. Temperature dependent variation in toxicity of insecticides against Earias vitella (Lepidoptera: Noctuidae). J. Econ. Entomol. 100: 357 360. Scott, J. G. 1995. Effects of temperature on insecticides toxicity, pp. 111 135. In R. M. Roe and R. J. Kuhr (eds.), Reviews in pesticide toxicology, vol. 3. Toxicology Communications Inc., Raleigh, NC. Serikawa, R. H., E. A. Backus, and M. E. Rogers. 2012. Effects of soil applied imidacloprid on Asian citrus psyllid (Hemiptera: Psyllidae) feeding behavior J. Econ. Entomol. 105 (5): 1492 1502. Sparks, T. C., M. H. Shour, and E. G. Wellemeyer. 1982. Temperature toxicity relationships of pyrethroids on three lepidopterans. J. Econ. Entomol. 75: 643 646. Srinivasan, R., M. A. Hoy, R. Singh, and M. E. Rogers. 2008. Laboratory and field evaluations of silwet L 77 and kinetic alone and in combination with imidacloprid and abamectin for the management of the Asian citrus psyllid, Diaphorina citri (Hemiptera: Psyllidae). Fla Entomol 91(1): 87 100. Stauderm an, K., P. Avery, L. Aristizabal, and S. Authors. 2012. Evaluation of Isaria fumosorosea (Hypocreales: Cordycipitaceae) for control of the Asian citrus psyllid, Diaphorina citri (Hemiptera: Psyllidae). Biocontrol. Sci. Tech. 22(7): 747 761. Stelinski, L., and R. Atwood. 2008. Evaluation of low volume application technologies for Asian citrus psyllid ( Diaphorina citri Kuwayama) control: initial results. Proc. Fla. Hort. Sci. Stelinski, L., M. Rogers, J. Yates, M. Dewdney, and T. Spann. 2009. Low volume appl ications: An emerging technology in the citrus industry. UF/IFAS Extension. Found at: http://www.crec.ifas.ufl.edu/extension/greening/low_volume_app.shtml Viewed: May 15, 2013.
83 Stover, E., D. Scotto, and J. Salvatore. 2002. Pesticide spraying in Indian River grapefruit: I. Survey of current practices. HortTechnol 12(3): 454 460. Stover, E., D. Scotto, C. Wilson, and M. Salyani. 2003. Pesticide spraying in Indian River grap efruit: II. Overview of factors influencing spray efficacy and off target deposition. Hort Technol. 13(1): 166 177. Teixeira, D.C., J.L. Danet, S. Eveillard, E.C. Martins, W.C. de Jesus Jr., P.C Yamamoto, S.A. Lopes, R.B. Ganier, M., and J.M. Bove. 1993. C itrus greening disease: citrus greening disease and the greening bacterium. Twelfth IOCV Conference. 212 219. Tiwari, S., H. Lewis Rosenblum, K. Pelz Stelinski, and L. L. Stelinski. 2010. Incidence of Candidatus Liberibacter asiaticus infection in abandone d citrus occurring in proximity to commercially managed groves. J. Econ. Ent. 103(6): 1972 1978. Tiwari, S., K. Pelz Stelinski, and L. L. Stelinski. 2011 Effect of Candidatus Liberibacter asiaticus infection on susceptibility of Asian citrus psyllid, Diaphorina citri, to selected insecticides. Pest manage. sci. 67(1): 94 99. Tiwari, S., N. Killiny, and L. L. Stelinski. 2013. Dynamic Insecticide Susceptibility Changes in Florida Populations of Diaphorina citri (Hemiptera: Psyllidae). J Econ. Entomol. 106(1): 393 399. Toth, S. J., and T. C. Sparks. 1990. Effect of temperature on toxicity and knockdown activity of cis perm ethrin, esfenvalerate, and lambda cyhalothrin in the cabbage looper (Lepidoptera: Noctuidae). J. Econ. Entomol. 83: 342 346. Tsagkarakis, A. E., and M.E. Rogers. 2010. host plant for Diaphorina citri (Hemiptera: Psy llidae). Fla. Entomol. 93:451 53. Tsai, J. H., J. Wang, and Y. Liu. 2002. Seasonal abundance of the Asian citrus psyllid, Diaphorina citri (Homoptera: Psyllidae) in southern Florida. Fla. Entomol. 85(3): 446 451. UF/IFAS Extension. Citrus Health Management Areas. Found at: http://www.crec.ifas.ufl.edu/extension/chmas/chma_overview.shtml Viewed: May 15, 2013. van Emden, H.F., M.W. Service. 2004. Biological control, pp 151 153. In Pest and Vector Control, Cambridge University Press, Cambridge, United Kingdom. Wenninger, E.J., and D.G. Hall. 2007. Daily timing of mating and age at reproductive maturity in Diaphorina citri (Hemiptera: Psyllidae). Fla Entomol. 90: 715 722.
84 Wenninge r, E.J., L.L. Stelinski, and D.G. Hall. 2008. Behavioral evidence for a female produced sex attractant in Diaphorina citri. Entomol. Exp. Appl. 128(3): 450 459. Westbrook, C. J, D.G. Hall, E. Stover, and Y. P. Duan. 2011. Colonization of Citrus and Citrus related germplasm by Diaphorina citri (Hemiptera: Psyllidae). Hort. Sci. 46:99 7 1005. Wood, B. W., W. L. Tedders, and J. Taylor. 1997. Control of pecan aphids with an organosilicone surfactant. Hort. Sci. 32(6): 1074 1076. Xu, C. F., Y. Xia, K.B. Li, and C. Ke. 1988. Further study of the transmission of citrus Huanglungbin by a psyl lid, Diaphorina citri Kuwayama. pp. 243 248. In Proceedings, 10th Conference of the International Organization of Citrus Virologists. IOCV. Riverside, CA. Yang, Y., M. Huang, G.A.C Beattie, Y. Xia, G. Ouyang, and J. Xiong. 2006. Distribution, biology, ecol ogy and control of the psyllid Diaphorina citri Kuwayama, a major pest of citrus: a status report for China. Int J Pest Manage. 52: 343 352.
85 BIOGRAPHICAL SKETCH Christine Weaver attended Auburn University in 2006. She completed a Bachelor of Science in agronomy and soils in December 2010. Post graduation she moved to Gainesville, Florida and enrolled in a Master of Science program in entomology and nematology Upon completion of some coursework, she moved to Lake Alfred to continue her studies and conduct her researc h at the IFAS Citrus Research and Education Center, under the direction of Dr. Michael Rogers.