EVALUATION OF EURASIAN AND HYBRID WATERMILFOIL ACCESSIONS FOLLOWING EXPOSURE TO DIFFERENT ENVIRONMENTAL CONDITIONS AND HERBICIDES By JENS P BEETS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PA RTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2018
2018 Jens P Beets
To the memory of Dr. Mike Netherland
4 ACKNOWLEDGMENTS I would like to thank my committee for their support: Dr. Mike Netherland, Dr. Stephen Enloe, Dr. Bill Haller, and Dr. Dail Laughinghouse. I would especially like to thank Dr. Netherland for all of his support and guida nce He gave me the opportunity to work for him and be his stu dent and he helped me tremendously since I started working for him both professionally and personally. Dr. Netherland dedicated so much time and energy into my work and personal development, an d for that I will be forever grateful. His influence and suppo rt kept me in the aquatic plant management field and helped guide me to pursue a doctoral program. Dr. Jay Ferrell has also been invaluable in his support and guidance. Conversations with Dr. Fe rrell have allowed me to look for new approaches and justifica tions in my work Thank you to everyone else that has helped and guided me in my work at UF especially Carl Della Torre, Chetta Owens, Dean Jones, Sherry Bostick, Cody Lastinger, and Joshua Woo d. I am also thankful for the organizations that helped fund m y research: Florida Fish and Wildlife Conservation Commission, Aquatic Ecosystem Research Foundation, US Army Corp s of Engineers Research and Development Center, and SePRO, especially Dr. Mark H eilman who has provided support and feedback in several studie s. Finally, I would like to thank my friends and family for their support and guidance. My parents and my sister Kalmia, have been extremely supportive, helpful and are a continuing inspiration My cousin, Jeremy convinc ed me to make the move to Florida and get started in the field of aquatics, that I have grown to enjoy and appreciate
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 Eurasian Watermilfoil ................................ ................................ .............................. 14 Northern Watermilfoil ................................ ................................ .............................. 16 Hybrid Watermilfoils ................................ ................................ ................................ 17 Management ................................ ................................ ................................ ........... 19 Response of EWM to Herbicides ................................ ................................ ............ 21 Auxin Mimic Herbicides ................................ ................................ ........................... 22 Florpyrauxifen benzyl (ProcellaCOR) ................................ ................................ ..... 23 Growth Chambers and OECD ................................ ................................ ................ 24 2 GROWTH CHAMBER EVALUATION OF FIVE AUXIN MIMIC HERBICIDES AGAINST EURASIAN AND HYBRID WATERMILFOIL ................................ .......... 27 Materials and Methods ................................ ................................ ............................ 29 Results and Discussion ................................ ................................ ........................... 31 3 GROWTH CHAMBER EVALUATION OF SUBSTRATE TYPE AND FERTILIZATION ON GROWTH OF EURASIAN WATERMILFOIL AND HYDRILLA ................................ ................................ ................................ .............. 38 Materials and Methods ................................ ................................ ............................ 39 Results and Discussion ................................ ................................ ........................... 40 4 LARGE SCALE MESOCOSM EVALUATION OF FLORPYRAUXIFEN BENZYL ON EURASIAN AND HYBRID WATERMILFOIL AND SEVEN NATIVE SUBMERSED AQUATIC PLANTS ................................ ................................ ......... 48 Materials and Methods ................................ ................................ ............................ 51 Results and Discussion ................................ ................................ ........................... 52 Milfoil Efficacy ................................ ................................ ................................ ... 52 Native Species ................................ ................................ ................................ 53
6 5 EVALUATION OF HERBICIDE RESPONSE AND GROWTH BETWEEN EURASIAN AND FOUR HYBRID WATERMILFOIL ACCESSIONS ....................... 60 Materials and Methods ................................ ................................ ............................ 62 Molecular Confirmation of Hybrids ................................ ................................ ... 62 DNA Analysis ................................ ................................ ................................ ... 63 Small scale Growth Chamber Study ................................ ................................ 63 Mesocosm Growth Study ................................ ................................ ................. 64 Mesocosm CET Comparison ................................ ................................ ............ 65 Results and Discussion ................................ ................................ ........................... 66 Mol ecular Confirmation of Milfoil Populations ................................ ................... 66 Small scale Growth Chamber Study ................................ ................................ 67 Mesocosm Growth Study ................................ ................................ ................. 67 Mesocosm CET Comparison ................................ ................................ ............ 68 6 CONCLUSIONS ................................ ................................ ................................ ..... 79 APPENDIX: SOURCES OF MATERIALS ................................ ................................ ..... 81 LIST OF REFERENCES ................................ ................................ ............................... 82 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 90
7 LIST OF TABLES Table page 2 1 Overview of five auxin mimic herbicides tested at eight concentrations on EWM and HWM in each g rowth chamber trial ................................ .................... 34 2 2 Average EC 50 (g a.i. L 1 ) values based on dry biomass with 95% confidence intervals of EWM and HWM for the five auxins tested at eight concentrations ... 35 2 3 Average EC 50 (g a.i L 1 ) values based on total plant length with 95% confidence intervals of EWM and HWM for the five auxins tested at eight concentratio ns ................................ ................................ ................................ .... 36 3 1 Soil analysis of substrates used in growth assay prior to addition of fertilizer .... 44 3 2 Available Nitrogen and Phosphate of fertilizer additions used in growth cha mber study. ................................ ................................ ................................ ... 45 4 1 Mean (SE) florpyrauxifen benzyl concentration (g L 1 ) collected at hours after treatment (HAT) and days after treatment (DAT) intervals following treatme nt ................................ ................................ ................................ ............ 55 5 1 Treatment rates for mesocosm trial at four exposure times with herbicides applied in mesocosm trial at LAERF on 4/12/201 7 ................................ ............. 71 5 2 Dry biom ass effective concentration of florpyrauxifen benzyl (g L 1) for four accessions of hybrid watermilfoil ................................ ................................ ........ 72
8 LIST OF FIGURES Figure page 1 1 Line drawing of Eurasian watermilfoil ................................ ................................ 26 2 1 Dose response curves for mean dry biomass for five auxin mimic herbicides tested on EWM and HWM ................................ ................................ .................. 37 3 1 Influence of growth media and fertilizer on mean (SE) biomass accumulation of monoecious hydrilla among fertilizer treatment levels .............. 46 3 2 Influence of a) fertilizer and b) substrate on mean (SE) biomass accumulation of EWM ................................ ................................ ......................... 47 4 1 Mean ( SE) dry aboveground biomass at 30 and 60 days after treatment (DAT) with florpyrauxifen b enzyl on EWM and HWM ................................ ......... 56 4 2 Mean ( SE) dry aboveground biomass at 30 and 60 days after treatment (DAT) with florpyrauxifen benzyl on American pondweed and Illinois Pondweed ................................ ................................ ................................ .......... 57 4 3 Mean ( SE) dry aboveground biomass at 30 and 60 days after treatment (DAT) with florpyrauxifen benzyl on elodea and Heteranth era ........................... 58 4 4 Me an ( SE) dry aboveground biomass at 30 and 60 days after treatment (DAT) with florpyrauxifen benzyl on Southern vallisneria and Northern vallisneria ................................ ................................ ................................ ........... 59 5 1 Results of Principal Coordinate Ana lysis of milfoil DNA samples combined with reference samples from Thum database. ................................ .................... 73 5 2 Dose response curves of average biomass fo r four accessions of HWM after exposure to florpyrauxifen benz yl in growth chambers ................................ ....... 74 5 3 Mean plant dry biomass (SE) a boveground and belowground at each harvest period (7, 8, and 9 months after planting) for all HWM accessions in the 6,700 L mesocosm growth study at Lewisville, TX ................................ ....... 75 5 4 Mean plant dry biomass (SE) aboveground and belowground at for each milfoil biotype across all harvest periods in 6,700 L mesocosm growth study at Lewisv ille, TX ................................ ................................ ................................ .. 76 5 5 Mean reduction in dry biomass (SE) for (a) 300 g L 1 2,4 D 7 day, (b) 300 g L 1 2,4 D+750 g L 1 endothall 7 da y, and (c) 1200 g L 1 2,4 D+300 g L 1 endothall 6 hr CET sce narios 60 days after treatment in mesocosms.. ........... 77
9 5 6 Mean r eduction in dry biomass (SE) for (a) 1.5 g L 1 florpyrauxifen benzyl 7 day, (b) 3 g L 1 florpyrauxifen benzyl 6 hr, (c) 6 g L 1 f lorpyrauxifen benzyl 6 hr, and (d) 12 g L 1 florpyrauxifen benzyl 3 hr CET scenario 60 days after treatment in mesocosms. ................................ ................................ ... 78
10 LIST OF ABBREVIATIONS 2,4 D 2,4 D [2,4 dichlorophenoxy] acetic acid AI Active Ingredient CAIP Center for Aquatic and Invasive Plants CET Concentration exposure time DAT Days after treatment DW Dry Weight HAT Hours after treatment LAERF Lewisville Aquatic Ecosystem Research Facility SE Standard Error
11 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EVALUATION OF EURASIAN AND HYBRID WATERMILFOIL ACCESSIONS FOLLOWING EXPOSURE TO DIFFERENT ENVIRONMENTAL CONDITIONS AND HERBICIDES By Jens Beets December 2018 Chair: Michael Netherland Cochair: Stephen Enloe Major: Agronomy Eurasian watermilfoil (EWM) and Hybrid watermilfoil (HWM) are problematic submer s ed aquatic invasive plants in many waterways of the n orthern United S tates. Auxin mimic herbicides, such as 2,4 D and triclopyr, are commonly used herbicides to manage to control invasive populations of EWM and HWM. The arylpicolinate herbicide florpyrauxifen benzyl provides a new tool to augment control op tions of proble matic aquatic weedy species including EWM and HWM. Experiments were conducted in growth chambers and large s cale mesocosm s to better understand the efficacy of florpyrauxifen benzyl on EWM and HWM and differences between HWM accessions T h e a im s of these s tudies were to: 1) evaluate HWM and EWM response to several auxin mimic herbicides under static, environmentally controlled conditions, 2) evaluate the growth of two submersed plant species in OECD recommended sediment, 3 ) investigate the po tential for inc reased herbicide tolerance of HWM, 4 ) evaluate a wide range of CET conditions to determine the effect of florpyrauxifen benzyl on well established EWM, HWM, and several native aquatic plant species under large scale mesocosm condition s and 5 ) document differences in biomass and response to
12 herbicides between populations of NWM, EWM and HWM Growth chamber results indicate strong response to florpyrauxifen benzyl in both EWM and HWM, with differences in response between EWM and HWM for the au x in mimic herbicides tested Similar to growth chamber results large scale studies demonstrated significant reduction in EWM and HWM in several CET scenarios. In addition to growth differences between HWM accessions, there were differences in herbicide res p onse between accessions.
13 CHAPTER 1 LITERATURE REVIEW Invasive species are a recognized threat to biodiversity and economic function s of many ecosystems worldwide. Invasive species are often non native, and in vasive plants have several key life history tr aits that make them better competitors compared to native species of the ecosystem they invade. Some of these traits include sexual and asexual reproduction, rapid growth, phenotypic plasticity, and tolerance to envi ronmental he t erogeneity (Sakai et al 20 01). Invasive species often lack the predators or pathogens that control the populations in their native habitats, increasing resource availability for these invasive species and their ability to invade. There are an estimated 25,000 introduced plant speci es in the United States, costing nearly $35 b illion between losses, damages, and control Of this, aquatic weeds cost $110 million (Pimentel et al. 2005). Similar to terrestrial systems, there many native aquatic pla nts desirable to ecosystems because of t heir benefit to biotic and abiotic factors within a lake. Aquatic plant species serve not only as food and habitat to fish and other aquatic organisms, but also improve diversity, stabilize sediment, and improve wate r clarity (Madsen 2014). Invasive aquati c species such as Eurasian watermilfoil ( Myriophyllum spicatum L.) can destabilize the biodiversity as well the as economic and ecologic function of water bod ies There are vascular plant species that dominate aquat ic ecosystems and the majority or entire ty of their life cycle occurs while submersed in water. These plants display a wide range of adaptations to submersed life, dependent on the conditions where they evolved. Most of these macrophytes have terrestrial a ncestors, evidenced
14 by vestigial structu res such as thin cuticles, and poorly developed or functionless stomata and xylem (Sculthorpe 1967). Haloragaceae, the watermilfoil family, primarily includes marsh or water herbaceous eudicots distributed worldwide, with the one shrubby genus Haloragoden dron endemic to Australia. Haloragaceae is comprised of eight genera with 145 recorded species. Defining characteristics include f lowers that are often unisexu al, although some species have bisexual flowers. Th e genus of most interest in the family is the aquatic Myriophyllum with approximately 68 species (Moody and Les 2010). There are fourteen milfoil species present in the United States, including both native and invasive species. Ther e is confirmed hybridization between native and non native species w ith reported hybrids of M. heterophyllum Michx. x M. pinnatum (Walter) Br itton et al. and M. spicatum L. x M. sibiricum Kom. confirmed (Moody and Les 2002). Eurasian W atermilfoil Eurasian Watermilfoil, hereafter referred to as EWM is a submersed eudicot na tive to Eurasia first reported in the United States in the 1940s Since t hen EWM has spread throughout aquatic ecosystems the Northern third of the country. There are at least two distinct genetic lineages of EWM present in North America (Zuellig and Thum 2012). EWM is listed as a noxious weed in several southern states includi ng Florida, North Carolina, South Carolina, and Alabama. Despite a variety of management options, EWM has proven to be a persistent problem in lakes and rivers (Smith and Barko 1990, Madsen et al. 1991). EWM is a rooted perennial plant with slender, heavil y branched stems and grayish green, whorled leaves with a feathery appearance and 24 or more leaflets (Figure 1 1; Godfrey and Wooten 1981). EWM rapidly establishes in a variety of
15 ha bitats, including impounded and natural freshwater, as well as both brack ish and spring waters. One factor that contributes to the invasive potential of EWM in displacing native vegetation is its ability to spread via vegetative growth as well as fragmenta tion. Milfoil species have a high propensity to autofragment after peak biomass is attained. Autofragmentation is a self induced abscission of apical stems that have formed roots (Madsen and Smith 1997). This allows for a more rapid rate of long distance spread than stolon expansion. Autofragmentation accounts for up to 26% of EWM expansion with stolon growth remaining as the primary means of local spread (Madsen and Smith 1997). EWM can also use allofragmentation, or mechanical breakage, as a means of fr agment spread (Madsen and Smith 1997). Invasive milfoils have a longe r growing period and photosynthetic tissue configuration that allows for growth a wider range of light conditions than many common native species such as Northern watermilfoil ( Myriophy llum sibiricum Kom.), coontail ( Ceratophyllum demersum L.), and Vallisn eria spp. (Grace and Wetzel 1978). Peak biomass (200 2000 g m 2 DW) occurs from June to September (Adams and McCracken 1974, Stanley et al. 1975), with biomass remaining unchanged duri ng early winter. Flowering and seed production coincide with peak bioma ss production during the warmer summer months, with fluctuations in growth occurring after flowering due to fragmentation (Stanley et al. 1975, Adams and McCracken 1974, Smith and Barko 1990). Overwintering generally occurs in an evergreen form, however, n ew unexpanded shoots attached to rootstocks also aid overwintering (Stanley et al. 1975). Dormancy may also play a role in preventing localized extinction events (Grace and Wetzel 1978)
16 EWM is typically found in water 1 to 10 m deep, with growth to the w ater su r face commonly observed in water 3 to 5 m deep (Aiken et al. 1979). This is due to a low light compensation point, which allows it to inhabit deeper regions of water bodies depen ding on water clarity (Grace and Wetzel 1978). As milfoils reach the w ater surface the stems tend to branch profusely, while the lower leaves slough off, thus forming a dense surface canopy that has the potential to shade out native vegetation and subsequ ently reduce species diversity (Grace and Wetzel 1978, Madsen et al. 19 91). Buoyancy is provided via vascular lacunae that oxygenate the roots (Moody and Les 2010). Like many submersed macrophytes, EWM shows a preference for nutrient rich sediments and ten ds to grow better on fine textured inorganic sediments (Li et al. 2015, Smith and Barko 1990). Despite having some aspects of Krantz anatomy that are typical of C 4 plant s milfoil is a C 3 plant (Stanley and Naylor 1972). Like several other aquatic species EWM photorespiration is much lower than that of terrestrial C 3 plants however, the ratio of photosynthesis to photorespiration is more similar to terrestrial C 3 plants (Van et al. 1976). Northern Watermilfoil Northern watermilfoil hereafter referred to as NWM is a native species with a range spanning from Alaska throughou t the continental United States excluding the southeastern United States (USDA 2011). Much like EWM, NWM is a rooted perennial with leaves in whorls around the stem with six to eleven l eaflets per leaf. While it can produce high plant biomass, it is not g enerally recognized as a canopy forming plant. In addition to reproducing vegetatively and via seed, NWM produces turions at the terminal node (Berger 2011). NWM provides a high quality habitat for aquatic wildlife
17 serving as both cover and a food source making it integral to proper ecosystem function. Hybrid Watermilfoils Hybridization between species is recognized as important and frequent in the of evolution and speciation of plant s. Hybridization can contribute to genetic diversity, adaptations, and the formation of ecotypes and species (Rieseberg et al. 1993 ). Hybridization has also been implicated in contributing to and promoting invasions (Rieseberg et al. Ellstrand 1993). Plant Communities w h ere hybridization can occur between inv asive and native species may be more susceptible to invasion and extinction of a parental species may occur (Sakai et al. 2001) From a management standpoint, hybridization can further complicate identi fication due to the ability to present characteristics similar to either parent or even novel morphologies. Hybridization in invasive plant management creates two major concerns: 1) gene contamination via outbreeding depression or genetic assimilation that can drive the native parent species extinct and 2) hy bridization that leads to heterosis or hybrid vigor (Moody and Les 2002). Thompson (1991) reported that the hybridization of a native and invasive Spartina sp. has the potential to create recombinant ge nomes with genotypes superior in fitness to parental s pecies. Heterosis is of particular concern in species that spread through vegetative means as the dissolution of heterosis is less likely and can spread indefinitely (Moody and Les 2002). Human disturb ance may lead to the creation of novel niches that hyb rids are better suited to fill compared to their parent species (Ellstrand and Schierenbeck 2000). One example of this is chemical applications to control invasive species in water bodies, where an herb icide treatment may control more susceptible parent pl ants and create the
18 opportunity for hybrids to spread within a lake. There is even the possibility for hybridization to stimulate the evolution of invasive potential (Ellstrand and Schierenbeck 2000). H ybrid Watermilfoil ( Myriophyllum spicatum L. x Myrioph yllum sibiricum Kom.), hereafter called HWM, was first documented by Moody and Les (2002). Due to morphological similarities between HWM and its parental species it is suspected that hybrids went unnoti ced for several years or possibly decades and were onl y definitively identified as hybrids based on ribosomal DNA analysis (Moody and Les 2002). HWM may pose a greater threat due to higher invasive potential and lower sensitivity to herbicides than EWM (La rue et al. 2013, Berger et al. 2015). The genetic mech anisms for these concerns are not well understood but could be due to heterosis or increased genetic variation (Larue et al. 2013). There may be significant differences between HWM in different lakes d ue to the hybrid populations arising independently (St urtevant et al. 2009). Several studies have indicated there are populations of HWM that show differences in response to several herbicides compared to EWM. The HWM population in Townline Lake (Big Rapid s, MI) has shown reduced and variable response to the herbicides fluridone, 2,4 D, and triclopyr (Berger et al. 2012, Berger et al. 2015, Glomski and Netherland 2010, LaRue et al. 2013, Thum et al. 2012). HWM can be more abundant in lakes historically trea ted with 2,4 D than either parental species (LaRue et al. 2013). Resource managers in northern states have reported differential herbicide response in several northern lakes following treatment. While significant efforts have been placed on identifying hy brid
19 watermilfoils, there has been much less informati on generated regarding the potential for hybrid vigor, invasive potential, and response to different herbicides. Management Aquatic weeds that form surface canopies or mats, are known to negatively aff ect water quality by reducing dissolved oxygen in the water column below the mat and leading to high variation in temperatures and pH (Bowes et al. 1979). In addition to creating severe daily fluctuations in environmental conditions, mat formation by invas ive species such as EWM also shades out other often l ower growing submersed species present which reduces plant species diversity and promot es a monoculture. Large scale aquatic invasive species infestations have multiple economic and ecological impacts such as interrupting navigation, providing habitat for disease vectors, impeding water flow, altering macroinvertebrate diversity, and nitrogen and phosphorus loading from plant degradation (Madsen et al. 1991) C ontrol measures to mitigate impacts on rec reation, fisheries, and wildlife diversity in water bodies is often necessary due to the combination of negative water quality impacts, water flow reduction/blockage, and native species displacement T here are several non chemical methods of control used t o manage aquatic invasive submersed plant species Mechanical harvesting can provide large scale control of invasive aquatic species though it is usually non selective and results are often temporary M echanical control ha s the potential to spread fragmen ts and may create mats of decomposing fragmented plant matter in the water that pose health risks such as mosquito breeding habitat (Madsen 2000, Grace and Wetzel 1978). Costs of submersed wee d harvesting vary and are dependent on location and the species of submersed plants Harvesting usually requires the harvested plant biomass to be
20 transported off site which further increases the cost of mechanical control. M ilfoil harvest cost s have ranged from $300 to $600/acre and hydrilla ( Hydrilla verticillata [ L .f .] Royle) a pproximately $455/acre (University of Minnesota 2018 Haller and Jones 2012). Drawdowns are a common method used to desiccate milfoil and other evergreen perennial aquatic species (Tarver 1980). Drawdowns can be particu larly effective in wint er, and can provide long term control but require six to eight weeks for drying and can interfere with the use of the water body as well as increase the potential for annual plants to spread ( Haller 2014) While drawdowns are gener ally feasible on reserv oirs, they are much less likely to be an option for control in natural lakes. Biological control has been used with varying levels of success. Grass carp ( Ctenopharyngodon Idella Val enciennes ) can provide a long term and cost effect ive means of invasive p lant management, especially when combined with chemical management practices (Eggeman 1994). They are a popular control method in isolated water bodies and grass carp preferentially feed on specific aquatic plant species such as hydr illa (Madsen 2000). How ever, grass carp have a lower preference for EWM than other invasive plant species and may in fact release EWM infestations from hydrilla and other targeted species (Richardson 2008). For this reason, grass carp are rarely used in p ublic waters of norther n states for EWM control. The milfoil weevil ( Euhrychiopsis lecontei D. ) is native to North American with the ability to reduce milfoil growth at insect densities above 25 m 2 (Newman and Inglis 2009). E uhrychiopsis lecont e i ha s the ability to reduce root biomass as well as aboveground biomass (Newman et al 1996). Detached biomass increases with weevil density, while the total above ground biomass
21 does not significantly differ from populations not affected by the weevil which i ndicat es insect damage is of ten too low to provide effective control (Newman et al. 1996). Response of EWM to Herbicides EWM has a long history of control with herbicide treatments, such as 2,4 D which has been used since the 1950s (Gallagher and Haller 1990). Understanding concent ration and exposure time (CET) is critical to proper control of EWM and other aquatic plants (Netherland and Getsinger 1992). Laboratory and field studies have proven invaluable in developing effective herbicide use patte rns for control of EWM (Berger et a l. 2012, Poovey et al. 2007). Poovey et al. (2007) suggested that the effectiveness of herbicide applications is dependent on several factors, including plant growth, age and density. Seasonal timing of triclopyr treatm ents can provide selective control of EWM (Netherland and Glomski 2014) Triclopyr and 2,4 D have been used to control EWM in both small scale spot treatments and whole lake treatments at lower dosages (Glomksi and Netherland 2010, Green and Westerdahl 199 0, Nault et al. 2014, Wersal et al. 2010). Low dose long term exposure treatments across an entire water body, while initially highly effective, have been suggested to provide selection pressure that could result in displacement of more sensitive parental genotypes and selection for a more tolerant hybrid. Several other herbicides and herbicide combinations have been used to control EWM. Fluridone has historically provided selective control with large scale, low concentration applications (Madsen et al. 2002 ) and have the added benefit of multiple year control and cost effectiveness (Berger et al. 2012, Madsen et al. 2002). Carfentrazone has also been shown to reduce EWM biomass, however, more effective control is attained when used in combination with 2, 4 D (Gray et al. 2007). Diquat can
22 provide control of EWM even with relatively short half lives (Skogerboe et al. 2006). Similarly, endothall provides selective control of EWM and is dependent on rate and timing (Skogerboe and Getsinger 2002). The contact herbicides d iquat and endothall ten d to provide only short term control, causing applicators to favor the use of triclopyr, 2,4 D, and fluridone for long term control of EWM (Berger et al. 2012). Due to reduced efficacy of herbicide treatments on HWM and r epeat applications of fluridone lea ding to resistance in hydrilla (Netherland and Jones 2015) there is a need for the development of new herbicide chemistry and a better understanding of differences between HWM and EWM. Auxin M imic H erbicides Auxin mimic herbicides, such as 2,4 D and tric lopyr, are commonly used management tools to control invasive populations of EWM and HWM (Netherland and Getsinger 1992; Poovey et al. 2007; Wersal et al. 2010). These herbicides provide selective and systemic control of many aquatic invasive species. Auxi n hormones are involved in root initiation, shoot growth, and development, among other plant growth processes (Grossman 2010). Auxin mimic herbicides simulate auxin overdose in plants ; however, synthetic auxins are more s table than natural auxins, making t he synthetic auxins more resistant to inactivation by the plant (Grossman 2010, Richardson et al. 2016). Auxin mimic herbicide damage acts in three successive phases: 1) stimulation, where abnormal plant growth is due to uncontrollable cell division 2) i nhibition, where plant growth is stunted, and physiological responses are suppressed and 3) decay due to cell wall degradation (Richardson et al. 2016). Auxin herbicides are perceived by the TIR1/AFB auxin receptors, inactivating Aux/IAA repressors and de pressing auxin response factors (Grossman 2010) This causes an overexpression of the genes
23 responsible for ethylene and abscisic acid (ABA) biosynthesis. Excess ethylene in shoots leads to epinasty and permanent over expression of ABA. ABA distribution th roughout the plant mediates stomatal closure and reactive oxygen species (ROS) are overprodu ced along with limited transpiration and carbon assimilation. ABA also limits cell division and combined with ethylene cause chloroplast damage and destroy cell mem branes. Ultimately, this leads to growth inhibition, tissue desiccation, and plant death (Gr ossmann 2010). Florpyrauxifen benzyl (ProcellaCOR) Florpyrauxifen benzyl is a new herbicide chemistry developed by SePRO Corporation (Carmel, Indiana USA) in part nership with Dow Agrosciences (Indianapolis, Indiana USA ) The herbicide is part of a new c lass of synthetic auxins, the arylpicolinates, that differ in binding affinity compared to other auxins registered for aquatic use such as 2,4 D and triclopyr (Bell et al. 2015, Lee et al. 2013) Florpyrauxifen benzyl has a binding affinity more similar to the terrestrial ly registered herbicide aminopyralid (Epp et al. 2016). Florpyrauxifen benzyl is translocated to grow ing points within affected plants, however pre vious studies show little translocation to the roots (Miller and Norsworthy 2018). Dow Agro Science s has developed this chemistry for use in rice to control multiple herbicide resistant barnyard grass ( Echinochloa crus galli [L.] P Beauv.). Florpyrauxifen benzyl shows high efficacy on barnyard grass at label rates, with no difference between plan ts resistant or susceptible to other herbicides (Duy et al. 2018 ; Miller et al. 2017 ). In small scale laboratory screening, florpyrauxifen benzyl is also shown to b e active on several aquatic weed species including crested floating heart ( Nymphoides cristata [Roxb.] Kuntze ), hydrilla ( both dioecious and monoecious
24 biotypes), and Eurasian watermilfoil ( Beets and Netherland 2018a, Netherland and Richa rdson 2016, Rich ardson et al. 2016). These studies suggest rapid uptake and activity under static conditions at low concentrations from 1 to 27 g L 1 as well as selectivity against several desirable native species Growth Chambers and OECD Growth chamb ers provide high ly controlled, uniform conditions allowing for increased repetition and eliminat ion of external variables such as shading, herbivory, and temperature fluctuations While growth chambers have their own set of limitations, such as algal inter ference, they ha ve small space requirements and provide a controlled environment for testing (Netherland and Getsinger 2018). P rotocols developed by the European Organisation for Economic Co operation and Development (OECD) have been adopted t o further sta ndardize studies in growth chambers and allow for direct comparisons between trials performed at different times and by different laboratories The OECD developed several protocols for testing non target damage by chemicals, especially herbicides, on aquatic species. The protocol was historically used on Lemna spp. and algae, but more recently was adopted for Myriophyllum spp to allow for meaningful testing of different modes of ac tion (OECD 2014). The protocol provides specific guidelines for light and temperature condi tions, water quality, sediment, and experimental design to ensure consistency between replicates OECD protocol allow s for greater replication in small scale stud ies than commonly used methods such as aquaria or small mesocosms may allow. Although the pro tocol was designed for risk assessment of chemicals on rooted dicotyledons, it has been shown to be effective for testing new herbicides (Netherland and Richardson 2 016) as well as bioassays of contaminated sediments (Feiler et al. 2004). Consistency of p urchased sediments can
25 often be problematic in comparisons between trials, creating a need for a standardized alternative, which OECD protocols provide. The OECD rec ommended substrate requires mixing of components, which can be time intensive and costly to have the components shipped. The overall goal of this research is to evaluate differences in growth and herbicide response between HWM and its parental genotypes (EWM and NWM) as well as possible variation between HWM populations The first objective wa s to evaluate auxin herbicide response in EWM and HWM in a series of small scale experiments The second objective was to evaluat e differences in submersed species growth in various substrate types and fertilization levels, to compare the OECD recommended growth media to other growth media in small scale studies The third objective was to evaluate florpyrauxifen benzyl activity on well established milfoils and seve ral native submersed species in large scale mesocosm evaluations This was based on high acti vity of florpyrauxifen benzyl seen in small scale studies. The final objective was to better understand growth and herbicide efficacy differences between HWM popul ations, as well as populations of its parental species This was done by m icrosatellite analy sis of several HWM accessions, as well as small and large scale evaluations of herbicide response. This research can provide a better understanding of how results from small scale studies translate to studies performed in larger, more realistic systems Th ese results can inform management practices and improve the knowledge of invasive milfoils.
26 Figure 1 1. Line drawing of Eurasian watermilfoil (EWM; M yri ophyllum spicatum L.).
27 CHAPTER 2 GROWTH CHAMBER EVALUATION OF FIVE AUXIN MIMIC HERBICIDES AGAINST EU RASIAN AND HYBRID WATERMILFOIL There is increasing concern by aquatic plant managers regarding apparent decreased herbicide efficacy on hybrid watermilfoil ( M. spicatum x M. sibiricum Kom. ; HWM ) compared to Eurasian watermilfoil ( Myriophyllum spicatum L. ; EWM) control. For example, 2,4 D tolerance has been observed in the HWM population from Hayden Lake, Idaho (Tom Woolf, personal communication Taylor et al. 2017 ). Thes e anecdotal observations by plant managers on the apparent differential responses of EWM and HWM require additional stu dies. Studies have compared auxin efficacy on EWM and HWM under different conditions, but there is a lack of direct comparisons between these two milfoils to auxin mimics tested under uniform laboratory conditions (Glomski an d Netherland 2010, Netherland a nd Willey 2017, Poovey et al. 2007). EWM and HWM submersed aquatic invasive plants that have spread across the northern United States and are problematic in several water bodies such as Hayden Lake, ID and Lake Minnetonka, MN Both genotypes quickly displa ce native vegetation by often forming dense surface mats which outcompete other submersed species, impede water flow, alter macroinvertebrate diversity, and can lead to nitrogen and phosphorus loading from plant degradation (Madsen et al. 1991) This can alter water quality parameters such as DO, temperature, and pH, by stratification similar to that observed for invasive hydrilla (Bowes et al. 1979). The combination of negative water quality impacts, water flow reduction/blo ckage, and native species displ acement often necessitate s control to mitigate negative impacts on recreation, fisheries, aesthetics, and wildlife diversity in water bodies.
28 Auxin mimic herbicides, such as 2,4 D and triclopyr, are commonly used management tools to control invasive popul ations of EWM and HWM (Netherland and Getsinger 1992; Poovey et al. 2007; Wersal et al. 2010). These herbicides provide selective and systemic control of many aquatic invasive dicotyledons, including EWM and HWM. Auxin mimic herbicides simulate auxin overd ose in plants, however, these are more stable than natural auxins, making the synthetic auxins more resistant to inactivation by the plant. Auxin hormones are involved in root initiation, shoot growth, and development, among other processes (Grossman 2010). Auxin mimic herbicide damag e occurs in three successive phases: 1) stimulation, where abnormal plant growth occurs due to uncontrollable cell division; 2) inhibition, where plant growth is stunted, and physiological respons es are suppressed; and 3) decay due to cell wall degradation ( Grossman 2010, Sterling and Hall 1997 ). The development of the arylpicolinate herbicide florpyrauxifen benzyl provides a potential new product to augment control options of problematic aquati c weed species. Florpyrauxifen benzyl is part of a new class of synthetic auxins, the arylpicolinates, that differ in binding affinity compared to currently registered auxins such as 2,4 D and triclopyr (Bell et al. 2015, Lee et al. 2013). In small scale l aboratory evaluations florpyrauxifen benzyl has shown high activity on EWM (Netherland and Richardson 2016, Richardson et al. 2016). Aminocyclo p yrachlor and aminopyralid have also shown high activity on many invasive species. Bukun et al. (2010) reported that aminocyclopyrachlor has the potential for greater biolo gical activity than other auxin herbicides due to higher absorption. Aminopyralid has shown comparable or improved efficacy for control of Canada thistle ( Cirsium arvense ) (Enloe et al. 2007). It is unknown
29 if this high level of terrestrial activity transl ates into aquatic systems on a plant with known auxin mimic susceptibility but studies have shown aminocyclopyrachlor efficacy on several aquatic floating plants, including water hyacinth ( Eichho rnia crassipes [Mart.] Solms) (Israel 2011). The objectives of these experiments were to 1) evaluate HWM and EWM response to five auxin mimic herbicides in static, environmentally controlled conditions, using EC values derived from length and biomass of t reated plants, and 2) i nvesti gate decreased herbicide efficacy on HWM and provide insight concerning tolerance to a specific auxin mimic herbicide or tolerance to the family of herbicides. This study will focus on a confirmed HWM from a single population t hat has been reported to be t olerant to 2,4 D (Tom Woolf, personal communication), however, it is important to consider that hybrid populations arise independently, and herbicide response may vary greatly between populations due to inherited traits. Materi als and Methods The efficacy of five auxin mimic herbicides was evaluated on EWM and HWM in growth chambers at the University of Florida Center for Aquatic and Invasive Plants (CAIP) Gainesville, FL All chambers were kept at 25 C on a 16 hour light cycle Trial 1 was initiated on 7/ 21/15 for EWM and 5/11/16 for HWM, and repeated (Trial 2) on 10/11/16 for HWM and 2/26/18 for EWM. Five auxin herbicides we r e tested on EWM and HWM, with an untreated control in each trial All plant material was taken from cul ture tanks at CAIP (EWM was o riginally collected from Crystal River, FL and HWM originally from Hayden Lake, ID) Two 10 cm apical stems were collected from the EWM and HWM stock tanks and planted separately in soil in 250 mL beakers hen placed in 2L beake rs containing nutrient soluti on in the growth chambers. The plants were
30 allowed to grow for seven days prior to treatment to allow root development Each experimental unit consisted of a 2L beaker containing a 250 mL beaker with 200 mL of Organisation for Economic Co operation and Dev elopment (OECD) sediment with two apical stems of EW M or HWM in growth solution as described in Smart and Barko (1985) (OECD 2014) Each of the five auxin herbicides was tested at eight concentrations with three replications in trial 1 and five replications in trial 2 (Table 2 1). Each beaker was randomly assigned a n herbicide concentration In addition five beakers were randomly selected for pretreatment harvest. Due to space constraints in growth ch ambers, treatments were tes ted sequentially, with 2,4 D 1 aminocyclopyrachlor 2 and aminopyralid 3 tested first on HWM followed by triclopyr 4 and florpyrauxifen benzyl 5 This treatment plan was then repeated on EWM. Two weeks after each trial was initiated all plant root and shoots were harvested and combined then sprayed with water to remove necrotic material, and total length of each plant (sum of main stem and all branching stems) measured to the nearest 0.1 cm. Treatment placement was randomized on each shelf for all trials. Tota l plant length and plant dry biomass were used to determine half maximal effective concentration (EC 50 ) values. Samples were placed in labeled bags in a forced air drying oven at 60 C for 72 hours to determine dry biomass to near est 0.01 g. Statistical a na lysis : To account for temporal differences between treatments as well as variation in growth chamber temperature and light source fluctuations, each shelf contained one replication of each treatment in a Randomized Complete Block Design (RCBD). Each shelf was be treated as a separate block. Data between the repeated trials were not statistically different at the 5% level and therefore combined for
31 analysis Analysis was performed in R using the drc package based on dry viable biomass and total length (Kneze vic et. al 2007). The drc package was also used to create the dose response figures. Results and Discussion Symptoms typical of auxin herbicides were ob served within two days of treatment with epinasty and necrosis being the most prevalent. Florpyrauxifen benzyl was extremely active on EWM based upon dry biomass reduction with an EC 50 value of 0.001 g L 1 ai, 5600 x less than the next lowest EC 50 value, 5 .6 g L 1 ai for aminocyclopyrachlor (Table 2 2). EWM sensitivity to t riclopyr, aminocyclopyrachlor, a nd 2,4 D did not significantly differ based on EC 50 values for dry biomass (Table 2 2). Similar trends were observed for EC 50 values based on length, alt hough the drc package was not able to determine an EC 50 value for florpyrauxifen benzyl due to high mor tality (Table 2 3) Aminopyralid did not differ in EC 50 values from the other auxin mimics tested, despite its EC 50 value being noticeably greater than the other EC 50 values. The high variation in aminopyralid EC 50 values was likely due to two replicates b eing observably less robust at the 1 and 27 g L 1 ai treatments than other plants at the same rate. The results from these trials indicat e that florpyrauxifen benzyl is highly active on EWM, and at concentrations up to four orders of magnitude lower than other auxin mimic herbicides. Florpyrauxifen benzyl was also highly active on HWM with a biomass EC 50 value of 0.38 g L 1 ai, 38x lower than the triclopyr EC 50 value of 14.7 g L 1 ai. Triclopyr had a significantly lower biomass EC 50 value than 2,4 D, am inocyclopyrachlor, and aminopyralid. Aminocyclopyrachlor, 2,4 D, and aminopyralid did not have different EC 50 values for HWM based on biom ass (Table 2 2). H erbicides fell into three groups in
32 terms of efficacy on HWM biomass with florpyrauxifen benzyl exh ibiting the highest efficacy followed by triclopyr, with efficacy not differing between 2,4 D, aminocyclopyrachlor, and aminopyralid. A similar trend was observed in EC 50 values based on length, with variation in the magnitudinal differences between compou nds compared to biomass values (Table 2 3). EC values for florpyrauxifen benzyl increase d from 0.001 for EWM to 0.38 g L 1 ai for HWM indicating HWM is less sensitive to florpyrauxifen benzyl (Figure 2 1 a; Table 2 2). HWM response to triclopyr did not si gnificantly differ from EWM while HWM was 5.8x less sensitive to 2,4 D than EWM (Figure 2 1c). EWM was 17x more sensitive to aminocyclopyrachlor than HWM (Figure 2 1d) but sensitivity to aminopyralid did not significantly differ between EWM and HWM (Figur e 2 1e). These differences indicate that Hayden HWM is not only 2,4 D tolerant but also exhibits differences in herbicide sensitivity across auxin mimic herbicides in short term static conditions. Florpyrauxifen benzyl shows a high degree of activity on bo th EWM and HWM at concentrations significantly lower than the other herbicides tested, s uggesting it has promise as an effective tool for managers with EWM and HWM infestations. While aminocyclopyrachlor and aminopyralid did provide control at use rates similar to 2,4 D and triclopyr, they were not as effective as florpyrauxifen benzyl. Due to high herbicidal activity noted in terrestrial systems, it is interesting that this is not observed in aquatic systems, or at least with milfoils. There may be a bia s issue in the protocol used. Based on the data from these trials, it could be predict ed that florpyrauxifen benzyl would have a strong effect in terrestrial systems while aminocyclopyrachlor and aminopyralid would have less of an effect, which is not wha t
33 has been observed with aminocyclopyrachlor ( Enloe et al. 2007, Minogue et al. 201 1 ). I t is apparent that florpyrauxifen benzyl is more toxic to milfoils based on activity at much lower concentrations than the other auxin mimic herbicides tested However, there are also efficacy differences between EWM and this accession of HWM with some of the other auxin mimic herbicides.
34 Table 2 1. Overview of five auxin mimic herbicide s tested at eight concentrations on EWM and HWM in each growth chamber trial (n=3 Trial 1, n =5 Trial 2). Herbicide Concentrations (g a.i. L 1 ) Florpyrauxifen benzyl 0, 0.1, 0.3, 1, 3, 9, 27, 81 Triclopyr 0, 1, 3, 9, 27, 81, 243, 729 2,4 D 0, 1, 3, 9, 27, 81, 243, 729 Aminocyclopyrachlor 0, 1, 3, 9, 27, 81, 243, 729 Aminopyralid 0, 1, 3, 9, 27, 81, 243, 729
35 Table 2 2. Average EC 50 (g a.i. L 1 ) values based on dry biomass with 95% confidence intervals of EWM and HWM for the five auxins tested at eight concentrations (pooled; n=8). Herbicide EWM EC 50 HWM EC 50 Florpyrau xifen benzyl 0.001 [0, 0.006] 0.38 [0.10, 0.65] Triclopyr 9.9 [6.2, 13.5] 14.7 [7.4, 22.1] 2,4 D 10.4 [4.9, 15.8] 60.3 [25.0, 95.7] Aminocyclopyrachlor 5.6 [2.0, 9.2] 94.1 [38.0, 150] Aminopyralid 66.3 [0, 149] 73.7 [34.7, 113]
36 Table 2 3. Averag e EC 50 (g a.i L 1 ) values based on total plant length with 95% confidence intervals of EWM and HWM for the five auxins tested at eight concentrations (pooled; n=8). Herbicide EWM EC 50 HWM EC 50 Florpyrauxifen benzyl N/D* 0.19 [0.01, 0.37] Triclopyr 5.5 [1.48, 9.6] 14.2 [2.0, 26.4] 2,4 D 8.3 [2.80, 13.7] 45.7 [22.3, 69.0] Aminocyclopyrachlor 6.4 [0.50, 12.2] 111 [44.8, 176] Aminopyralid 25.4 [0, 68.2] 92 .9 [42. 3 143] The drc package was not able to determine an EC 50 based on length due to high mort ality with florpyrauxifen benzyl.
37 Figure 2 1. Dose response curves for mean dry biomass for five auxin mimic herbicides tested on EWM and HWM : a) florpyrauxifen benzyl, b) triclopyr, c) 2,4 D, d) aminocyclopyrachlo r, and e) aminopyralid. Each symb ol represents mean values ( standard error, n=8). Solid lines are fits for Eurasian Watermilfoil (EWM) and dashed lines are fits for Hybrid Watermilfoil (HWM).
38 CHAPTER 3 GROWTH CHAMBER EVALUATION OF S UBSTRATE TY PE AND FERTILIZATION ON GROWTH OF EURASIA N WATERMILFOIL AND HYDRILLA Eurasian watermilfoil ( Myriophyllum spicatum L.; EWM) and monoecious hydrilla ( Hydrilla verticillata [ L.f ] Royle) are problematic, submersed aquatic invasive species that have invaded water bodies across the northern region of the United States. Like many submersed macrophytes, watermilfoil shows a preference for nutrient rich fine textured inorganic sediments (Smith and Barko 1990, Li et al. 2015). Monoecious hydrilla has been shown to prefer sediments low in organic matter (M cFarland and Barko 1987). Both species grow as rooted, submersed plants, and detached fragments may spread to form new rooted populations withi n a water body (True Meadows et al. 2016; Smith and Barko 1990). The Organisation for Economic Cooperation and D evelopment (OECD) developed a protocol for testing non target damage by chemicals, especially herbicides, on aquatic species (OECD 2014) The protocol provides specific guidelines for light and temperature, water quality, s ubstrate and experimental design to ensure consistency between replicates and allow for higher replication in small scale stud ies than common research metho ds conducted in aquaria or small mesocosms. Although the test was designed for risk assessment of chemicals on rooted dicotyledons, it has been shown to effective for testing new herbicides (Netherland and Richardson 2016) as well as bioassays of contamin ated sediments (Feiler et al. 2004). A modified version of substrate recommended by the OECD for herbicide testing in growth cham bers has been used in herbicide testing on EWM and hydrilla to standardize small scale studies. Although it allows for consi stency in s ubstrate parameters such as pH and soil composition, the substrate requires mixing of specific
39 components according to st rict guidelines. This is costly and more time consuming to prepare than using commercially available potting soils or sand. Nutrients, especially nitrogen and phosphorus, are necessary for plant growth, but in aquatic systems can also lead to increased al gal populations. A higher fertilizer rate than that recommended by the OECD protocol could increase short term growth or al low longer term studies, or higher fertilizer rates may creat e prohibitive algal blooms or inhibit growth due to root burn. It would save significant time and work in study preparation for evaluating response of hybrid watermilfoils, hydrilla, and other su bmersed species to various herbicide treatments i f a commercial potting soil provides similar or improved growth compared to the OE CD substrate Therefore, the objectives of this study were: 1) determine if the OECD recommended substrate increases short term growth compared to commercially available potting soils or builders sand amended with nutrients and 2) determine if the OECD re commended fertilizer rate and substrate promotes the greatest growth Materials and Methods The experiment to evaluate the e fficacy of OECD on aquatic plant growth was conducted using t wo trials carried out in four growth chambers at the University of Flor ida Center for Aquatic and Invasive Plants (CAIP). This experiment had 2 factors: substrate type, and fertilizer rate. Hydri lla and EWM w ere planted in either OECD s ubstrate (OECD 2014), commercial potting soil historically used at the Lewisville Aquatic E cosystem Research Facility (LAERF) in Lewisville, TX (Hapi gro Hope Agri Products, INC. top soil) 6 commercial potting soil used at CAIP (Margo Garden Products professional top soil) 7 or pure sand (Table 1). All four soil media w ere then amended with eith er 200 mg kg 1 (1x), 400 mg kg 1 (2x), 1000 mg kg 1 (5x) ammonium chloride
40 and sodium phosphate mixed with deio nized water or 3g kg 1 15 9 12 Osmocote 8 The 3g kg 1 15 9 12 Osmocote is commonly used in fertilization of submersed aquatic species for stock cultivation (Mudge 2018) This is equivalent to 2.25x more phosphate and 7.5x more nitrogen than the OECD proto col requires (Table 3 2). The experiment was conducted as a randomized complete block design with treatments arranged as a 4x 4 factorial comparin g 4 substrate types and 3 rates of ammonium chloride and sodium phosphate and the Osmocote The experiment cons isted of two identical trials t rial 1 was initiated on 2/8/17, and trial 2 was initiated 6/28/18. All chambers were kept at 25 C on a 16 hour li ght cycle. All of the plant material was collected from culture tanks at the CAIP, the EWM originated from Crys tal River, FL while the monoecious hydrilla originated from Lake Harding, GA. Establishment in the growth chamber began by collecting 10 cm apica l stems from the stock tanks. Each experimental unit consisted of a 250 mL beaker with 200 mL of s ubstrate with two 10 cm apical stems of EWM or one sprouted tuber o f monoecious hydrilla. The 250 mL beaker was then lowered into a 2 L beaker containing Smar t and Barko (1985) culture solution. Each treatment contained four replicates and the trial was repeated Due to algal growth, all beakers were flushed and replaced with new Smart and Barko solution 15 days after planting. A destructive harvest was conduct ed following the 30 day growth period All plants were rinsed, sorted and placed in a forced air drying oven to obtain dry biomass Two way ANOVA tests were performe d in R on dry biomass data with Tukey HSD test for multiple comparisons. Results and Discus sion N o s ubstrate type or fertilizer rate indicated a clearly advantageous medium for short term growth of hydrilla or EWM in these growth chamber trials There was a
41 significant interaction between substrate type and fertilizer treatment on biomass accumu lation of monoecious hydrilla (p = 0.04; Figure 3 1). Few differences were observed between substrate and fertilizer combinations, however there were some notable tr ends. Monoecious hydrilla biomass planted in OECD substrate was never larger than hydrilla planted in other substrate types. Also monoecious hydrilla planted in Margo potting soil had increased biomass compared to initial biomass with all fertilizer treat ments (p<0.05) At the 5x rate of OECD fertilizer, hydrilla grown in sand did not significa ntly increase in biomass compared to initial biomass and was significantly smaller than hydrilla planted in potting soil at the same fertilizer rate. This may be ind icative of root burning by the fertilizer in sand, while organic matter content in the Marg o potting soil prevented root burn via redox reactions (Reddy and DeLaune 2008). There was no significant interaction between substrate type and fertilizer treatmen t on EWM biomass, therefore, one way ANOVAs were run on the main effects (Figure 3 2). Simi lar to monoecious hydrilla, no fertilizer rate or substrate type provided a clear advantage to EWM biomass production. No fertilizer rate resulted in significantly d ifferent biomass from the 1x fertilizer rate, indicating increased fertilizer does not prov ide an advantage to EWM biomass production in small scale growth chamber studies (Figure 3 2a). EWM biomass accumulation was 34% lower in sand than plants in the OE C D substrate but biomass did not differ between this treatment and that of the two potting s oils (Figure 3 2b). These data suggest that EWM and monoecious hydrilla can grow just as well, if not better, in commercially available potting soil compared to the substrate required by the OECD protocol. Although monoecious hydrilla growth does differ i n substrate s
42 depending on fertilizer, biomas s of the plants grown in Margo potting soil was not different from biomass of plants grown in OECD s ubstrate at every fertilizer treatment. Plants grown in Margo potting soil also resulted in significantly more b iomass than the initial biomass even at fert ilizer rates in which the OECD s ubstrate did not increase Monoecious hydrilla and EWM grown in pure sand had reduced biomass in several treatments and did not produce increases in biomass. Which may be a result of burn from the fertilizer at increased rat es, or competition due to the higher algal density observed in some trials. Both root damage and algae growth are common problems in propagation of submersed aquatic vegetation (Mudge 2018). Although it was not q uantified, algae were observably more abunda nt in the 5x fertilizer rate and sand treatments. Reduced growth in these treatments may be attributed to reduced light, nutrient competition, or root burn, or a combination of all three factors. I t is surprising that EWM and hydrilla did not grow signific antly more in OECD required substrate or Hapi gro potting soil d ue to their reported preference for inorganic substrates (McFarland and Barko 1987; Smith and Barko 1990) Both species produced equal or greater b iomass in the Margo top soil compared to OEC D substrate which was not expected, given the greater organic matter content of the Margo top soil (Table 3 1). Increased fertilizer rate s do not seem to have a consistently significant positive effect on plant b iomass, therefore the rate recommended by th e OECD protocol appeared sufficient and may help prevent algal competition from the lower nitrogen and phosphorus concentrations in the OECD sediment S imilar plant biomass growth in potting soil with higher orga nic matter compared to the OECD required substrate indicates that these potting soil mixes can be used in place of the OECD required
43 substrate for small scale growth studies potentially reducing costs and study setup times It should be noted that there is variation in commercially purchased soils, and composition varies. OECD required substrate may be beneficial when performing studies requiring standardized substrate conditions, such as studies with new herbicide chemistri es, but the OECD required substra te does not provide better growth compared to that in commercial potting soils
44 Table 3 1. Soil analysis of substrate s used in growth assay prior to addition of fertilizer. Substrate Total P (mg kg 1 ) pH Total N (mg kg 1 ) Organic Matter (%) Margo 189. 9 7.67 390.6 35.09 Hapi gro 49.91 8.28 391.3 5.31 Sand 1.02 8.28 10.58 0.13 OECD 0 7 0 4.5
45 Table 3 2. Available Nitrogen and Phosphate of fertilizer additions used in growth chamber study. Fertilizer Total N (g kg 1 ) Phosphate (g kg 1 ) OECD (1X) 0.06 0.12 OECD (2X) 0.12 0.24 OECD (5X) 0.3 0.5 Osmocote (3g kg 1 ) 0.45 0.27
46 Figure 3 1. Influence of growth media and fertilizer on mean (SE) biomass accumulation of monoecious hydrilla among fertilizer treatment levels (pooled; n= 8) 30 day s after planting. Horizontal black line indicates mean initial biomass. Asterisks indicate biomass significantly higher than average initial growth. Means with the same letter are not significantly different at (p=0 05). 1 2 3 4 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 1x 2x 5x Osmocote Dry Biomass (g) Fertilizer Hapi-gro Top Soil Sand OECD Margo Top Soil abc c* bc* abc abc bc* bc* ab ab ab a bc* abc abc abc bc*
47 Figure 3 2. Influence of a) fertilizer and b) substrate on mean (SE) biomass accumulation of EWM (pooled; n= 8) 30 d ays after planting. Horizontal black line indicates average initial biomass. Significant increase in biomass was observed across all fertilizer rates and sediment typ es compared to initial. Means with the same letter are not significantly different at (p=0 .05). 1 2 3 4 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 1x 2x 5x Osm Dry biomass (g) Fertilizer A a b ab ab 1 2 3 4 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Hapi-gro Top soil OECD Margo Top Soil Sand Dry Biomass (g) Substrate B ab a a b
48 CHAPTER 4 LARGE SCALE MESOCOSM EVALUATION OF FLORPYRAUXIFEN BENZYL ON EURASIAN AND HYBRID WATERMILFOIL AND SEVEN NATIVE SUBMERSED AQUATIC PLANTS Eurasian watermilfoil ( Myriophyllum spicatum L.; EWM) and Hybrid Eurasian watermilfoil ( M. spicatum L x M. sibiricum Kom ; HWM) are problematic submer s ed aquatic invasive plants in many North American waterways. Auxin mimic herbicides, such as 2,4 D and triclopyr are commonly used for selective of control of invasive populations of EWM HWM and other dicotyledonous species by stimulating auxin overdose (Netherland and Getsinger 1992; Poovey et al. 2007; Wersal et al. 2010). Differences in response to 2,4 D betwe en EWM and HWM has led to discussion if this response is specific to 2,4 D or au xin mimics in general. These synthetic auxins are more stable in their binding to auxin receptors than natural hormones making the synthetic auxins more resistant to inactivat ion by the plant (Grossman 20 10 ). Moody and Les (2002) documented hybrid populat ions of watermilfoil, previously thought to be EWM, using nuclear ribosomal DNA analysis. Due to their highly similar morphology, DNA analysis is the most accurate method for d iscerning between EWM and HWM. The potential for inherited traits in HWM, such a s increased invasiveness, hybrid vigor, or increased tolerance to herbicides presents additional concerns for aquatic weed control programs (Ellstrand and Schierenbeck 2000, Mo ody and Les 2002, Thompson 1991). Chemical applications have the potential to cr eate niche habitats for HWM if herbicides have reduced efficacy (LaRue et al. 2013). In this situation, EWM could be drastically reduced or eliminated by exposure to auxin her bicides, while HWM survives to spread and repopulate treated sites (Ellstrand an d Schierenbeck 2000). However, it is important to consider that hybrid populations can
49 arise independently, and herbicide response may vary greatly between hybrid populations d ue to different inherited traits. Development of a new class of synthetic auxin s, the arylpicolinates, has resulted in production of a new herbicide called florpyrauxifen benzyl and it may herbicide provide a tool to augment control options of problemati c aquatic weedy species. The arylpicolinates differ in binding affinity compared to currently registered auxins such as 2,4 D and triclopyr (Bell et al. 2015, Lee et al. 2013). In small scale laboratory studies florpyrauxifen benzyl has been shown to be ac tive on several aquatic weed s including crested floating heart ( Nymphoides cris tata [Roxb.] Kuntze ) hydrilla ( Hydrilla verticillata [ L.f ] Royle both dioecious and monoecious biotypes) and EWM (Netherland and Richardson 2016, Richardson et al. 2016). R esults from these studies suggested that concentrations of florpyrauxifen benzyl had activity on EWM well below typical use rates for 2,4 D and triclopyr. C oncentration and exposure time (CET) requirements are key factors in evaluation of a new herbicid e to determin e use pattern s CET represents the amount of time that various herbicide concentrations are in contact with a plant and describes how an aquatic herbicide should affect a given plant species (Getsinger and Netherland 1997, Getsinger and Nether land 2018). Under operation al herbicide use, a wide range of potential CET scenarios may occur due to various factors such as treatment scale, water flow or exchange, application rate, adsorption, degradation, and diffusion (Nault et al. 2014, Netherland a nd Jones 2015, Green and We sterdahl 1990, Netherland and Glomski 2014, Glomski and Netherland 2010, Glomski and Netherland 2014, Glomski et al. 2009, Skogerboe et al. 2006). CET is species dependent and can play an important
50 role in herbicide selectivity. There has been considerable research conducted to define the CET requirements for control of EWM with the herbicides 2,4 D (Green and Westerdahl 1990, Nault et al. 2014) and triclopyr (Netherland and Getsinger 1992, Netherland and Glomski 2014, Netherland and Jones 2015). Further in vestigation of CET requirements is needed to evaluate the efficacy and use patterns of the new compound, florpyrauxifen benzyl. Large mesocosms allow for the inclusion of more plant species in a single unit and are less prone t o the plants rapidly reachi ng carrying capacity compared to small scale studies When large mesocosms are planted in the late summer or fall and treated the following spring, they better represent plant phenology and field conditions and provide a more rea listic environment than new ly planted small mesocosms or growth chambers (Netherland and Glomski 2014). The large scale mesocosms using more robust plants are generally used to confirm results from small scale studies (Netherland and Richardson 2016 ). The goal of this research was to evaluate a wide range of CET conditions to determine the effect of florpyrauxifen benzyl on well established EWM, HWM, and several native submersed species. Our objectives with this experiment were to determine the most effect ive CET combinations for EW M and HWM control and to observe the effect of these CET scenarios on native species. Native submersed species from North America included: American pondweed ( Potamogeton nodosus Poir. ), elodea ( Elodea canadensis Michx.) water s targrass ( Heteranthera dubi a [Jacq.] MacMill. ) Illinois pondweed ( Potamogeton illinoensis Morong ), as well as vallisneria
51 ( Vallisneria americana Michx.) from southern (Gainesville, FL) and northern (NY) locations. These species are considered desirable a nd less problematic than EW M. Materials and Methods Plants were established on 9/15/2015 from apical stems or root nodes ( Vallisneria ) at the U.S. Corps of Engineers Lewisville Aquatic Ecosystem Research Facility (LAERF) in Lewisville, TX. Each 6,700L mes ocosm was planted with two 3 L pots for each species of American pondweed, Illinois pondweed, elodea, water stargrass, EWM, HWM, and two populations of vallisneria fr om southern and northern locations. S pecimens of HWM with reported tolerance to 2,4 D were used from a single population (Hayden Lake, Idaho), (Beets and Netherland 2018 b Taylor et al. 2017 ) All plants were established in topsoil amended with Forestry Su pply 9 20 10 5 fertilizer tablets (4.5 g kg 1 ). Plants were allowed to establish from 9/201 5 to 4/2016, and then were treated with herbicide as noted below One treated and one control tank contained HOBO 10 data loggers to observe daily temperature fluctuat ions during the study period. Florpyrauxifen benzyl treatments were applied at concentrat ions of ( 0, 3, 9 and 27 g a.i. L 1 ) for 6 and 24 hour water exchange half lives as well as two concentrations (3 and 9 g L 1 ) as static treatments with no water exc hange U ntreated water was circulated through the mesocosm s at appropriate times to provid e nominal target water exchange half lives (Netherland and Glomski 2014). Each of the nine treatments had three replications randomly assigned to mesocosms Water s amples were collected from representative treatments and analyzed via liquid chromatography and tandem mass spectroscopy to determine actual herbicide concentrations (EPA 2015). Harvests were conducted at 30 and 60 d ays after treatment by collecting aboveground standing crop of plants. Samples were dried in a forced air dryer at 70 C for two week s and then
52 weighed to the nearest 0.1 g Results were analyzed using separate one way ANOVA tistical differences in aboveground biomass (p=0.05) among treatments at each harvest period. Heteroscedascity (unequal variance in predicted vs residual data) was an issue and data for EWM and HWM were square root transformed to meet assumptions of norma lity and equal variance. Nontransformed data are presented. Results and Discussion Temperature in the mesocosms ranged from 16.6 to 26.9 C with a mean temperature of 21.7 C during the study period. Herbic ide analysis determined florpyrauxifen benzyl degrad ation was within expectations based on dilution scenarios driven breakdown (Table 4 1; WA Dept. of Ecology 2017). Sample concentration fluctuations are likely due to a combination of herbi ci de photolytic degradation (0.6 day half life) plant uptake, and limitations in analysis due to herbicide solubility in wate r (10 to 15 g L 1 ) Milfoil Efficacy Florpyrauxifen benzyl provided near complete reduction of EWM and HWM biomass for up to 60 days following treatment even at the lowest concentrations and exposure times evaluated (Figure 4 1a and b ). EWM biomass was s ignificantly reduced by all CET scenarios, whereas, untreated control biomass showed a n increase between harvest per iods ( Figure 4 1 a). A ll exposure scenarios resulted in large reduction s in HWM biomass t hirty and sixty days after treatment compared to th e untreated control However, 30 days after treatment HWM biomass in the 3 g L 1 6 hour treatment (the lowest scenario) was greater than HWM biomass in the other CET treatments (Figure 4
53 2b). Differences in herbicide sensitivity between EWM and HWM have b een anecdotally observed in the field and seen in small scale studies (Beets and Netherland 2018 b Taylor et al 2017 ). These use rates were also two orders of magnitude below the use rates for currently registered herbicides such as triclopyr and 2,4 D ( Green and Westerdahl 1990, Nault et al. 2014, Netherland and Getsinger 1992) and suggest the potential use of f lorpyrauxifen benzyl for milfoil control programs. Native Species Overall, florpyrauxifen benzyl had minimal effect on the native species evalu ated in this study. It had no significant effect on American pondweed or Illinois pondweed biomass (Figure 4 2 a and b ) and some treatment s of Illinois pondweed had greater biomass than the untreated control at 30 days. Increases in growth in treated mesoc osms compared to untreated controls may be indicative of a lack of competition from the controlled milfoil. Elo dea was not significantly affected by time or treatment (Figure 4 3a ) and Heteranthera showed the most treatment related variability, with one tr eatment (3 g L 1 /6 hr) showing a large increase in biomass and another (9 g L 1 static) showing injury symptoms (Figure 4 3b) Given its sensitivity to 2,4 D, Heteranthera may be a plant that requires further refinement of CET for selective milfoil treat ments and did not grow well in this study No treatment scenario resulted in a si gnificant reduction in southern vallisneria (Figure 4 4a ). Northern vallisneria growth was minimal, however, northern vallisneria biomass in the 9 g L 1 /24 hr and 27 g L 1 /2 4 hr scenarios after 60 days was greater than the untreated control after 30 days (p < 0.001; Figure 4 4 b ). Overall, this study confirms preliminary studies indicating a high level of activity on EWM and HWM by florpyrauxifen benzyl. In addition, exposure requirements were
54 much shorter than expected, as evidenced by the strong control of EWM and HWM at the 3 g L 1 /6 hr water exchange scenario. This information is promis ing for selective control of target milfoil populations when compared to the lack of re sponse by native plants in the majority of CET scenarios. EWM and HWM were completely controlled in the 3 g L 1 static treatments and also scenarios with higher herbici de concentrations, whereas, native species exhibited variable but largely in significant responses to higher concentration as well as in both static treatments While low rate, static treatments are often used in targeting invasive aquatic species, hydrodyn amic processes can greatly alter CET and therefore herbicide treatment efficacy. Static applications such as whole pond treatments have the potential to lack selectivity depending on the initial application rate. However, based on these results florpyrauxi fen benzyl provides selective control of EWM and HWM under multiple CET scenarios. In s pecies rich areas, the ability to use low use rates to control milfoil invasions and allow the spread of native species via post treatment regrowth and sustained control of EWM and HWM is vital to management. This study also indicated that prior small scal e trials were useful predictors of use patterns for larger scale studies. Given the level of sensitivity of both EWM and HWM to the rates and exposures evaluated, the qu estion of potential treatment related differences between EWM and HWM was not adequatel y addressed. Although there is some evidence of increased tolerance by HWM, further trials (with this and additional strains of HWM) to determine if there are real diff erences in response to florpyrauxifen benzyl are warranted.
55 Table 4 1. Mean ( SE ) flor pyrauxifen benzyl concentration (g L 1 ) collected at hours after treatment (HAT) and days after treatment (DAT) intervals following treatment (n=3) Dashes indicate time periods where no sample was collected. CET scenario 1 HAT 6 HAT 24 HAT 48 HAT 72 HA T 7 DAT 10 DAT 14 DAT 27 g L 1 6 hr 16. 2 (2.8) 8.1 (0.66) 3.9 (2.9) 1.3 (0.21) 27 g L 1 24 hr 14. 3 (2. 1 ) 8. 9 (0.72) 9.0 (0.31) 8.6 (2.48) 2.0 (0.42) 0.84 (0.21) 3 g L 1 static 2. 2 (0.2) 1.6 (0.67) 0.77 (0.43 ) 0.10 (0.03) 0.07 (0.03) 9 g L 1 static 6. 9 (0.5) 2.6 (0.04) 1.2 (0.27) 0.25 (0.04) 0.08 (0.04)
56 Figure 4 1 Mean ( SE) d ry aboveground biomass at 30 and 60 days after treatment (DAT) with florpyrauxifen benzyl at 3 g L 1 for 6 hr, 24 hr and static water exchange half lives 9 g L 1 for 6 hr, 24 hr and static water exchange half lives and 27 g L 1 for 6 and 24 hr water exchange half lives on (a) EWM and (b) HWM ( n=3). Letters above bars represent differences between tre atment s 60 day harvest dates that were analyzed separately 0 40 80 120 160 Control 3 g/L 6 hrs 3 g/L 24 hrs 3 g/L Static 9 g/L 6 hrs 9 g/L 24 hrs 9 g/L static 27 g/L 6 hrs 27 g/L 24 hrs Biomass (g DW) Concentration (g L 1 )/ Time (Hrs) A 30 DAT 60 DAT A a b B b B b B b B b B b B b B b B 0 40 80 120 160 Control 3 g/L 6 hrs 3 g/L 24 hrs 3 g/L Static 9 g/L 6 hrs 9 g/L 24 hrs 9 g/L static 27 g/L 6 hrs 27 g/L 24 hrs Biomass (g DW) Concentration (g L 1 )/ Time (Hrs) B 30 DAT 60 DAT a A b B c B c B c B c B c B c B c B
57 Figure 4 2. Mean ( SE) dry aboveground biomass at 30 and 60 days after treatment (DAT) with florpyrauxifen benzyl at 3 g L 1 fo r 6 hr 24 hr and static water exchange half lives, 9 g L 1 for 6 hr, 24 hr and static water exchange half lives, and 27 g L 1 for 6 and 24 hr water exchange half lives on (a) American pondweed and (b) Illinois Pondweed (n=3) Letters above bars represen t diff Differences in mean biomass between 60 day treatments were not observed. 0 20 40 60 80 100 Control 3 g/L 6 hrs 3 g/L 24 hrs 3 g/L Static 9 g/L 6 hrs 9 g/L 24 hrs 9 g/L static 27 g/L 6 hrs 27 g/L 24 hrs Biomass (g DW) Concentration (g L 1 )/ Time (Hrs) A 30 DAT 60 DAT 0 40 80 120 Control 3 g/L 6 hrs 3 g/L 24 hrs 3 g/L Static 9 g/L 6 hrs 9 g/L 24 hrs 9 g/L static 27 g/L 6 hrs 27 g/L 24 hrs Biomass (g DW) Concentration (g L 1 )/ Time (Hrs) B 30 DAT 60 DAT b b ab a ab ab ab ab
58 Figure 4 3. Mean ( SE) dry aboveground biomass at 30 and 60 days after treatment (DAT) with florpyrauxifen benzyl a t 3 g L 1 for 6 hr, 24 hr and static water exchange half lives, 9 g L 1 for 6 hr, 24 hr and static water exchange half lives, and 27 g L 1 for 6 and 24 hr water exchange half lives on (a) e lodea and (b) Heteranthera (n=3) Differe nces in mean biomass we re not observed between treatments at 30 and 60 DAT. 0 100 200 300 Control 3 g/L 6 hrs 3 g/L 24 hrs 3 g/L Static 9 g/L 6 hrs 9 g/L 24 hrs 9 g/L static 27 g/L 6 hrs 27 g/L 24 hrs Biomass (g DW) Concentration (g L 1 )/ Time (Hrs) A 30 DAT 60 DAT 0 20 40 60 Control 3 g/L 6 hrs 3 g/L 24 hrs 3 g/L Static 9 g/L 6 hrs 9 g/L 24 hrs 9 g/L static 27 g/L 6 hrs 27 g/L 24 hrs Biomass (g DW) Concentration (g L 1 )/ Time (Hrs) B 30 DAT 60 DAT
59 Figure 4 4 Mean ( SE) dry aboveground biomass at 30 and 60 days after treatment (DAT) with florpyrauxifen benzyl at 3 g L 1 for 6 hr, 24 hr and static water exchange half lives, 9 g L 1 for 6 hr, 24 hr and static water exchange half lives, and 27 g L 1 for 6 and 24 hr water exchange half lives on (a) Sou thern v allisneria and (b) Nort hern v allisneria (n=3) Differences in mean biomass were not observed between treatments at 30 and 60 DAT for S. v a llisneria or 60 DAT for N. vallisneria 0 10 20 30 40 Control 3 g/L 6 hrs 3 g/L 24 hrs 3 g/L Static 9 g/L 6 hrs 9 g/L 24 hrs 9 g/L static 27 g/L 6 hrs 27 g/L 24 hrs Biomass (g DW) Concentration (g L 1 )/ Time (Hrs) A 30 DAT 60 DAT 0 1 2 Control 3 g/L 6 hrs 3 g/L 24 hrs 3 g/L Static 9 g/L 6 hrs 9 g/L 24 hrs 9 g/L static 27 g/L 6 hrs 27 g/L 24 hrs Biomass (g DW) Concentration (g L 1 )/ Time (Hrs) B 30 DAT 60 DAT a a a ab a a a a b
60 CHAPTER 5 EVALUATION OF HERBICIDE RESPONSE AND GROWTH BETWEEN EURASIAN AND FOUR HYBRID WATERMILFOIL ACCESSIONS Hybridizatio n in plants can be problematic to invasive plant management. Hybridization between native an d invasive species is of particular interest due to two main issues: 1) communities where hybridization between native and invasive species occurs may be more susce ptible to invasion leading to extinction of parental species (Sakai et al. 2001), and 2) hyb ridization can complicate identification due to the ability to present characteristics similar to either parent or novel morphologies. A primary example of this in aquatic plant management is hybridization between invasive Eurasian watermilfoil ( Myriophyll um spicatum L.; EWM) and native Northern watermilfoil ( Myriophyllum sibiricum Kom.; NWM). Concerns of hybrid watermilfoil ( M. spicatum L. x M. sibiricum Kom. ; HWM) have existed for decades, however, the highly similar morphology of EWM and suspected HWM co mplicated identification. Techniques for molecular confirmation advanced and the use nuclear ribosomal DNA (nrDNA) analysis proved to be an accurate method of ident ification and p opulations of HWM previously thought to be EWM were documented (Moody and Le s 2002). New or inherited characteristics that are problematic for management, such as hybrid vigor, increased invasiveness, or increased herbicide tolerance could potentially occur in hybrid populations If HWM populations have increased herbicide toleran ce compared to parental species, herbicide treatments for control of EWM may create niche habitats for HWM a s described by Ellstrand and Schierenbeck (2000) EWM wa s drastically reduced or eliminated by exposure to auxin mimic herbicides, while HWM survive d to spread and repopulate former sites of EWM Significant differences between hybrid populations in different lakes, or even within
61 lakes may occur due to hybrid populations likely arising i n dependently (Sturtevant et al. 2009). A few studies have indic ated there are populations of HWM that exhibit differential responses to several herbicides compared to EWM. The HWM population from Townline Lake, MI has shown red uced and variable response to fluridone, 2,4 D, and triclopyr (Berger et al. 2012, Berger et al. 2015, Glomski and Netherland 2010, LaRue et al. 2012, Thum et al. 2012). HWM is often more abundant in lakes historically treated with 2,4 D than either parent al species (LaRue 2012, Moody and Les 2007). Differential herbicide response in HWM has been reported by resource managers following treatment of several lakes in Northern states (Nault et al. 2014). HWM populations are genetically diverse and distinct po pulations can occupy the same water body (Taylor et al. 2017). This creates a need for ident ification of problematic populations as well as molecular identification of milfoils when studies are performed. Techniques have been developed to identify genetically distinct hybrid watermilfoil populations using several molecular techniques, including m icrosatellite analysis ( Thum et al. 2006, Wu et al. 2013 ) Microsatellite analysis is particularly effective in comparing genetic diversity among population s with more recent ancestry, such as independently arising HWM populations (Thum 2018). While signif icant efforts have been invested in identifying hybrid watermilfoils, there is much less information generated regarding the potential for hybrid vigor, inv asive potential, and response to herbicides. There are also relatively few comparisons between diffe rent hybrid watermilfoil populations. Florpyrauxifen benzyl has shown high activity on milfoil and several other invasive plant species however, studies wi th florpyrauxifen benzyl have been limited to
62 small scale systems ( Beets and Netherland 2018a, Nethe rland and Richardson 2016 Richardson et al. 2016 ). Small scale studies can provide rapid evaluation of herbicide activity as well as dose response measurem ents. H owever, they control abiotic factors and do not allow a realistic analysis of exposure requir ements in field conditions. Con centration exposure time (CET), the amount of time that a concentration of herbicide is in contact with a plant, is an extrem ely critical factor in aquatic herbicide treatment efficacy (Getsinger and Netherland 1997). Exposur e time and concentration of herbicide treatments are essential factors to aquatic plant management and allow better understanding for potential use rates w hich may differ between genetically distinct HWM populations The goal of this research was to eval uate potential differences in herbicide tolerance, growth of hybrids and parental genotypes, and variation among hybrid accessions. Our objectives with this study were to: 1) use molecular techniques to confirm that the accessions of Northern w atermilfoil ( Myriophyllum sibiricum Kom.; NWM), EWM, and HWM were genetically distinct, 2) obtain herbicide dose response data for four HWM accessions and compare conce ntration and exposure time combinations of florpyrauxifen benzyl on EWM NWM, and HWM populations 3 ) observe seasonal growth (biomass allocation, surface matting, etc.) in four HWM populations and its parental species (NWM and EWM), and 4 ) compare the CET scenarios for florpyrauxifen benzyl to similar rates of 2,4 D and 2,4 D + endothall. Materials and Methods Molecular Confirmation of Hybrids Plant samples were collected from apical shoots of milfoil plants in the field as well as stock culture tanks at the Center for Aquatic and Invasive Plants (CAIP),
63 Gainesville, FL. F ield sites were sampled at Ham Lake, MN, Minnetonka Lake, MN, and Big Cornelian Lake, MN. Samples were collected from the stock tanks at CAIP originated from Crystal River, FL, Ham Lake, MN, Alpine Lake, WI, Lake Minnetonka, MN, and Hayden Lake, ID. Tissue from 2 3 a pical meriste m s was cut from separate stems and flash frozen with liquid nitrogen before storing in a deep freezer at 80 C with five replications from each site Additional apical meristems were planted in mesocosms at Montana State University (MSU) for future refer ence an d redundancy. DNA Analysis Plant processing and analysis was completed at Montana State University in Dr. using DNEasy Plant Mini Kits (Qiagen). The internal transcr ibed sp acers (ITS) were amplified using universal primers ITS1 and ITS4 before subjecting to PCR reactions (Thum et al. 2006). T hermal cycling was completed at 94 C for 2 minutes followed by 25 cycles of: 94 C 1 minute, 56 C 30 seconds, 72 C 1 minute, with a fina l extension at 72 C for 8 minutes before holding at 4 C. The PCR product was then run on a 1% agarose gel to check for correct size and purity. Seven microsatellite markers were used to genotype the samples: Myrsp1, Myrsp 5, Mysrp 9, Myrsp 12, Myrsp 13, My rsp 15, and Myrsp 16 (Thum et al. 2017, Wu et al. 2013). Microsatellite data were scored using GeneMapper and POLYSAT was used to distinguish clones based on microsatellite loci. Unique clones were identified using a Principal Coordinates Analysis. Small scale Growth Chamber Study Experiments were carried out in four growth chambers at the University of Florida Center for Aquatic and Invasive Plants (CAIP). All chambers were kept at 25 C on a 16 hour light cycle. This experiment was conducted twice, with t rial one initiated on 5/22/17
64 and harvested 6/9/17, and trial two began on 8/13/18 and harvested 9/4/18. Florpyrauxifen benzyl was tested on four accessions of HWM (Hayden Lake, ID; Ham Lake, MN; Minnetonka Lake, MN; Alpine Lake, WI). All plant mate rial wa s collected from stock culture tanks at the CAIP Ten cm apical stems were harvested and planted in 250 mL beakers placed in 2 L beakers containing nutrient solution in the growth chambers. Each experimental unit consisted of a 250 mL beaker with 20 0 mL of Organisation for Economic Co operation and Development (OECD) substrate (OECD 2014) with two apical stems of EWM (Crystal River, FL) or HWM (Hayden Lake, ID) in a 2 L beaker containing growth solution as described in Smart and Barko (1985). Five h erbicid e concentrations were randomly tested on each accession with six replications for each treatment. Each beaker was randomly assigned an herbicide concentration. Prior to treatment, four beakers of each accession were harvested to collect pretreatment data. A ll plants were harvested s eventeen days after trial initiation and shoots and roots were washed with water to remove necrotic material and residual sediment and then combined. Samples were sorted based on treatment and accession, then placed in labeled b ags in a forced air drying oven at 7 0 C for 72 hours to determine dry biomass. Data were analyzed using the drc package in R to determine EC 50 and EC 90 values as well a s derive dose response curves (Knezevic et. al 2007). Plant dry biomass (roots and shoot s combined) was used to determine EC 50 and EC 90 values for each accession. Trials were not statistically different at the 5% level and were pooled to improve statistic al analysis. Mesocosm Growth Study Apical stems were planted on 9/28/16 and allowed to e stablish for 7 months until 4/12/17 at the Lewisville Aquatic Ecosystem Research Facility (LAERF) in
65 Lewisville, TX. All plants were grown in commercial top soil amended with Forestry Supply 9 20 10 5 fertilizer tablets (4.5 g kg 1 ). Three 6,700 L mesocos ms were plante d with nine 3 L pot s containing 10 cm apical stems of each biotype of NWM (Minnetonka, MN) EWM (Crystal River, FL and Lake Minnetonka, MN) and HWM (Hayden Lake, ID; Ham Lake, MN; Minnetonka MN; Alpine Lake WI) and harvested at study initiatio n on 4/12/17 Three additional mesocosms were planted with 18 3 L pots containing 10 cm apical stems of each biotype of NWM, EWM, and HWM biotype s on 9/28/16 Each pot contained plants from a single location. Harvests were performed 30 and 60 days after in itiation by harvesting one of the two pots of each plants in the three mesocosms at each harvest time Above and belowground samples were collected from plants at each harvest date and washed to remove sediment then dried in a forced air dryer at 7 0 C un ti l desiccated and weighed. Data were analyzed using two one way ANOVAs and biomass (p=0.05) between harvest dates. Mesocosm CET Comparison Twenty one 6,700 L mesocosms were set up at LAERF si mi lar manner to those described above on 9/28/16 with each the nine milfoil accession s or biotype s planted in a single 3 L pot Plants were allowed to establish for 5 months and herbicide treatments were applied on 4/12/2017, when they were treated. Meso co sm treatments were applied as follows: florpyrauxifen benzyl 5 at 3 and 6 g L 1 six hour water exchange half life ; and 12 g L 1 six hour water exchange half life; 1.5 g L 1 for seven day water exchange half life ; 0.3 mg L 1 2,4 D 11 seven day water excha nge half life 1.2 mg L 1 2,4 D + 3.0 mg L 1 endothall 12 six hour water exchange half life and 0.3 mg L 1 2,4 D + 0.75 mg L 1 endothall water exchange half life (Table 1). Exposure times were
66 achieved by circulating untreated water through the mesocosm to provide nominal water exchange at the target retention time ( Netherland and Glomski 2014). Each treatment was randomly applied to three mesocosms ( replicate s ) A destructive harvest was performed 60 days after treatment when a boveground biomass was collec ted, and samples were dried in a forced air dryer at 70 C until desiccated and weighed. Data were analyzed using one differences in aboveground biomass between treated and untreated milfoil 60 D AT (p = 0.0 5). One treated and one control tank contained HOBO 10 data loggers to observe temperature fluctuations during the study period. Results and Discussion Molecular Confirmation of Milfoil Populations The HWM accessions and EWM and NWM biotypes used in this st udy were confirmed to be genetically distinct based on the seven microsatellite loci tested. Principal coordinates analysis distinguished the different populations according to variant (Figure 5 1). The reference samples provided by the Thum database were used to validate the identification. Large symbols in Figure 5 1 indicate samples collected and processed in this study and s amples with the same x and y coordinates are considered genetically identical (Thum et al. 2017). Fresh fi eld sa mples from Ham Lak e and Lake Minnetonka were identical to and CAIP culture stock. with one exception; one sample from Minnetonka CAIP stock was identical to that from Crystal River. This is likely due to sample contamination during preparation or tip collection from stock t anks. It also highlights the importance of verifying the accession on which herbicide assays are performed. This also implies that there is no genetic drift in tanks, despite cultures being maintained for several years.
67 The two samp les from Big Cornelian ( squares) were identified as NWM based on location within cluster of NWM samples (Figure 5 1). Samples from Ham, Alpine, Minnetonka, and Hayden lakes clustered together in the middle, and were classified as HWM Hayden HWM has a diff erent EWM parental linea ge than the other HWM accessions tested, as indicated by its documented genetic distance from other lineages (Figure 5 1). This parental lineage is indicated by the grouping of EWM from the Thum reference samples The samples from C rystal River CAIP stock were distinguished as EWM (Figure 5 1). Small scale Growth Chamber Study Florpyrauxifen benzyl was highly active on all HWM accessions tested with EC 50 values from 0.11 to 0.57 g L 1 ai (Table 5 2). The HWM accession from Hayden, ID had an EC 50 value of 0.11 g L 1 ai, 3 to 5x lower than the other HWM accessions tested. HWM from Minnetonka, Alpine, and Ham lakes did not have significantly different EC 50 values (Table 5 2; Figure 5 2). EC 90 values were not significantly different be tween any HWM accession tested. While the Hayden accession is more sensitive to florpyrauxifen benzyl than the other accessions tested based on EC 50 values, control of all accessions may be achieved at similar EC 90 concentrations. These results are in consi stent with field operati ons where plant managers have suspected differential response of HWM to other auxin herbicides Further research should be conducted to identify problematic populations and determine proper management strategies with florpyrauxifen benzyl Mesocosm Growth Study N o significant interaction was observed between harvest date and species among the untreated controls and there was no significant difference in aboveground
68 biomass between harvest dates ( p>0.05; Figure 5 3a). There was a 39% increase in belowgroun d biomass between the 7 month and 9 month harvests (Figure 5 3b). The increase in belowground biomass likely due to plants being well established at the time of initial harvest and having little room for continued aboveground growth within the tanks, but s till having space within pots for root growth. There were significant differences in above and belowground biomass between milfoil biotypes (Figure 5 4). Both NWM populations, the Florida EWM, and Hayden HWM had statistically similar aboveground biomass T hese milfoil biotypes had significantly less aboveground biomass than the other HWM accessions as well as the Minnetonka EWM populations. Belowground biomass production followed similar trends of growth b etween biotypes The differences observed between ac cessions indicated HWM accessions have variation not only in morphology (Moody and Les 2002; LaRue et al. 2013) but also in biomass production. HWM demonstrated a capacity to attain higher biomass than it s native parent, NWM, although this does appear limi ted to certain accessions. HWM accessions also produced higher biomass than EWM, dependent on the HWM accession and EWM population being compared. Mesocosm CET C omparison Temperature in the mesocosms ra nged from 15.4 to 29.6 C with a mean temperature of 23 C during the 60 day study period. The CET experiment indicated that there are differences in response between different accessions of HWM in certain concentration exposure scenarios. The biomass in u nt reated controls of both NWM populations was extremel y variable and result ed in non significant reductions in all treatment scenarios despite 100% reduction in biomass
69 The 300 g L 1 7 day 2,4 D CET scenario resulted in 3 0 8 7 % reduction of milfoil biomass, however, only EWM biotypes were significantly reduce d (Figure 5 5a). This is comparable to whole lake treatments with 2,4 D, which result in 40 88% control, dependent on hydrologic conditions (Nault et al. 2014, Wersal et al. 2010). When endothall was adde d to the 2,4 D CET scenario biomass reduction incre ased, ranging from 58 99%, with significant reductions of EWM biotypes as well as HWM from Alpine and Ham Lakes (Figure 5 5b). When rates of 2,4 D were increased, but endothall concentration and retention time was decreased effica cy was reduced on HWM, wi th no significant reductions in biomass (Figure 5 5c) This is indicative of decreased 2,4 D efficacy on HWM compared to EWM in short exposure scenarios, and either combinations with other fast acting herbicides or longer herbicide exposure times may be re quired spot treatments and treatments in hydrologically fluctuating systems A 65 99% reduction in biomass was observed on both NWM and EWM (Figure 5 5c) Biomass control on HWM ranged from 0% on Alpine HWM to 73% on Hayden HWM (Fi gure 5 5c). These data cl early show that HWM is more tolerant to 2,4 D treatments as has been noted in the field by aquatic applicators. Milfoil biomass reduction due to florpyrauxifen benzyl treatment varied between CET scenarios. Reduction in biomass was only achieved in the 1.5 g L 1 7 day florpyrauxifen benzyl treatment with Florida EWM, one population of Minnetonka EWM, and Alpine HWM (Figure 5 6a). When concentration was increased to 3 g L 1 and exposure time was decreased to 6 hours, significant b iomass reduction was only achieved with the Florida EWM population (Figure 5 6b). The 6 g L 1 6 hour and 12 g L 1 3 hour treatments resulted in significant reductions of biomass for all EWM and
70 HWM (Figure 5 6c and d). NWM was also completely controlled in these CET scenarios bu t were not significantly different from controls These results further indicate differences in herbicide sensitivity between HWM accessions as well as differences between EWM populations. E fficacy differences between biotypes were not observed at high CET and total control was achieved for all HWM accessions tested This has promising management implications since florpyrauxifen benzyl has shown selective control for milfoil in the presence of native species and further corrobor ates that milfoil control can be achieved with florpyrauxifen benzyl at relatively low use rates with sufficient concentration exposure time. However, low dose treatments should be approached with caution as they may select for more herbicide tolerant HWM accessions. This highligh ts that CET for the entire water body must be taken into consideration even in spot treatments.
71 Table 5 1. Treatment rates for mesocosm trial at four exposure times with herbicides applied in mesocosm trial at LAERF on 4/12/2017 (n=3). Treatment Rate Ex posure Time 2,4 D 300 g L 1 7 days 2,4 D + Endothall 300 g L 1 (as acid) + 750 g L 1 (as salt) 7 days 2,4 D + Endothall 1200 g L 1 (as acid) + 300 g L 1 (as salt) 6 hours Florpyrauxifen benzyl 1.5 g L 1 7 days Florpyrauxifen benzyl 3 g L 1 6 ho urs Florpyrauxifen benzyl 6 g L 1 6 hours Florpyrauxifen benzyl 12 g L 1 6 hours
72 Table 5 2. Dry biomass effective concentration of florpyrauxifen benzyl (g L 1) for four accessions of hybrid watermilfoil (n=10; pooled) in growth chamber study 17 DAT. Values in brackets indicate 95% confidence intervals. V alues that share the same letter within an EC are not significantly different at the 5% level. Accession EC 50 ( g L 1 ) EC 90 ( g L 1 ) Hayden 0.11 [0.04, 0.17] a 1.1 [0.37, 1.7] b Alpine 0.57 [0.30, 0.84] b 1.4 [0.46, 2.4] b Minnetonka 0.37 [0.22, 0.53] b 0.83 [0.25, 1.4] b Ham 0.32 [0.14, 0.49] ab 1.1 [0.36, 1.9] b
73 Figure 5 1. Results of Principal Coordinate Analysis of milfoil DNA samples combined with reference samples from Thum da tabase. Squares indicate NWM, circles HWM, and triangles EWM. Small symbols indicate reference samples from Thum database, large symbols are samples from this study. Symbols are based on repli cations of each sample (n = 5). Genetically identical samples ha ve overlaid symbols. -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 EWM (Thum) HWM (Thum) NWM (Thum) Big Cornelian NWM 1 Big Cornelian NWM 2 Alpine HWM Crystal River EWM Ham HWM Hayden HWM Minnetonka HWM
74 Figure 5 2. Dose response curves of average biomass for four accessions of HWM after exposure to florpyrauxifen benzyl in growth chambers. Each symbol represents mean v alues ( standard error, n = 10).
75 Figure 5 3. Mean plant d ry biomass (SE) (a) aboveground and (b) belowground at each harvest period (7, 8, and 9 months after planting) for all HWM accessions in the 6,700 L mesocosm growth study at Lewisville, TX (n=27). Means with the same letter are not significantly different at (p = 0.05). 0 50 100 150 200 250 7 MAP 8 MAP 9 MAP Biomass (g DW) Harvest Date A a a a 0 10 20 30 7 MAP 8 MAP 9 MAP Biomass (g DW) Harvest Date B a b ab
76 Figure 5 4. Mean plant dry biomass (SE) (a) aboveground and (b) belowground at for each milfoil biotype across all harvest periods in 6,700 L mesocosm growth study at Lewisville, TX (n=27). Means with the same letter are not signifi cantly different at (p = 0.05). 0 100 200 300 400 Biomass(g DW) Milfoil A a a a a b bc b bc c 0 10 20 30 40 50 Biomass (g DW) Milfoil B a a ab bc bcd bcd cd cd d
77 Figure 5 5. Mean reduction in dry biomass (SE) for (a) 300 g L 1 2,4 D 7 day, (b) 300 g L 1 2,4 D+750 g L 1 endothall 7 day, and (c) 1200 g L 1 2,4 D+300 g L 1 endothall 6 hr CET scenarios 60 days after treat ment in mesocosms. Green bars represent NWM biotypes, orange bars represent EWM biotypes, and black bars represent HWM accessions. Asterisks indicate significant redu ction in biomass compared to untreated controls (n = 3). 0 20 40 60 80 100 % Reduction Accession A * 0 20 40 60 80 100 % Reduction Accession B * * 0 20 40 60 80 100 % Reduction Accession C *
78 Figure 5 6. Mean reduct ion in dry biomass (SE) for (a) 1.5 g L 1 florpyrauxifen benzyl 7 day, (b) 3 g L 1 florpyrauxifen benzyl 6 hr, (c) 6 g L 1 florpyrauxifen benzyl 6 hr, and (d) 12 g L 1 florpyrauxifen benzyl 3 hr CET scenario 60 days after treatment in mesocosms. Green bars represent NWM biotypes, orange bars represent EWM biotypes, and black bars represent HWM accessions. Asterisks indicate significant reduction in biomass compare d to untreated controls (n = 3 except (c) where n=2 ). Several treatments resulted in 100% biomass reduction in all replicates, as indicated by no error bars 0 20 40 60 80 100 % Reduction Accession A * 0 20 40 60 80 100 % Reduction Accession B 0 20 40 60 80 100 % Reduction Accession C * * * 0 20 40 60 80 100 % Reduction Accession D * * *
79 CHAPTER 6 CONCLUSIONS Hybrid milfoil populations are likely to arise independently through sexual reproduction (Thum and McNair 2018) and can have genet ic variation (i.e. different accessio ns) within a lake (Taylor et al. 2017). This is problematic for management efforts, as HWM have a higher invasive potential and show lower sensitivity to several currently used herbicides (LaRue et al. 201 3 Berger 2011 ). It is important to properly identi fy problematic HWM accessions and distinguish those with herbicide sensitivity differences to make better informed management decisions. It appears that culture tanks of milfoil can be maintained without concern for gen etic drift compared to field samples. Management of problematic milfoil populations is imperative to prevent spread within a water body as well as to other, nearby water bodies. Invasive watermilfoils can displace desirable native species, as well as impac t economic functions of lakes such as recreation, fishing, and transport. Impacts on ecologic functions such as species richness, abundance and diversity can also be observed in milfoil monocultures (Madsen et al. 1991). Both small scale growth ch amber studies, and large scale mesocosm studi es are effective methods for evaluation of herbicide sensitivity. Small scale studies provide data on sensitivity differences, while large scale mesocosm studies allow for more complex manipulations of potential hydrologic conditions in a more realistic sy stem. As expected, testing the efficacy of florpyrauxifen benzyl, 2,4 D, and endothall showed variation in efficacy among HWM accessions confirming anecdotal reports of aquatic plant managers Concentration expo sure time also proved to be instrumental, wit h accessions showing variation in control dependent on the CET scenario. Care should
80 be taken in determining necessary CET, as low dose treatments could result in selecting for herbicide tolerant HWM accessions. From these studies, florpyrauxifen benzyl as well as 2,4 D/endothall combinations were effective for milfoil control with appropriate CET.
81 APPENDIX : SOURCES OF MATERIALS 1 DMA 4 IVM, Dow AgroSciences LLC. 9330 Zionsville Road Indianapolis, IN 46268. http://www.cdms.net/ldat/ld4JS003.pdf 2 Met hod 240SL, Bayer Environmental Science, Bayer CropScience LP, 2 T.W. Alexander Drive Research Triangle Park, NC 27709. http://www.dot.stat e.oh.us/Divisions/ContractAdmin/Contracts/PurchDocs/375 18/CWCChem01/OHIO%20BID%20 %20LABELS%20and%20SDS/Method%20240S L.%20LABEL.pdf 3 Milestone, Dow AgroSciences LLC. 9330 Zionsville Road Indianapolis, IN 46268. https://assets.greenbook.net/19 51 50 19 03 2018 D02 879 007_Milestone_Specimen_Label.pdf 4 Renovate 3, SePRO Corporation. 11550 North Meridian Street, Suite 60 0 Carmel, IN 46032. https://sepro.com/Documents/Renovate 3_Label.pdf 5 SX 1552 SePRO Corporatio n. 11550 North Meridian Street, Suite 600 Carmel, IN 46032. 6 Hapi gro Top Soil, Hope Agriproducts, Inc. 2400 Old Lewisville Road Hope, AR 71801. 7 Professional Topsoil, Margo TM Garden Products. 50 N Laura Street Suite 2550 Jacksonville, FL 32202. 8 Osmo cote Plus Smart release Plant Food 15 9 12. Marysville, OH 9 20 5 10 Planting Tablets, Forestry Suppliers, Inc. 205 West Rankin Street Jackson, MS 39201. 10 H OBO Water Temp erature Pro v2. U22 001, Onset Computer Corporation. 470 MacArthur Blvd. Bourne MA 02532. 11 2,4 D Amine, Alligare. Alligare, LLC 13 N. 8 th Street Opelika, AL 36801. http://alligare.com/wp content/uploads/2018/05/24 d amine specimen_pa_mk approved 20150928.pdf. 12 Aquathol K, United Phosphorus, Inc. 630 Freedom Business Center, Suite 402 King of Prussia, PA 19406. http ://www.cdms.net/ldat/ld195006.pdf.
82 LIST OF REFERENCES Adams MS McCracken MD. 1974. Seasonal production of the Myriophyllu m component of Lake Wingra, Wisconsin. J. Ecol. 62:457 467. Aiken SG, Newroth PR, Wile I. 1979. The biology of Canadian weeds. 34. M yriophyllum spicatum L. Can. J. Plant Sci. 59: 201 215. Beets J, Netherland MD. 2018a. Mesocosm response of crested floa ting heart, hydrilla, and two native emergent plants to florpyrauxifen benzyl: a new arylpicolinate herbicide. J. Aquat. Plant Manage. 56:57 62. Beets J, Netherland MD. 2018 b Laboratory and mesocosm evaluation of growth and herbicide response in Eurasian watermilfoil and four accessions of hybrid watermilfoil [Abstract]. In: Proceedings of the Aquatic Plant Management Society Annual Me eting. APMS, Buffalo, NY. http://www.apms.org/wp/wp content/uploads/Full Program and Abstracts.pdf. Accessed August 1, 201 8. Bell JL, Schmitzer R, Weimer MR, Napier RM, Prusinska JM. 2015. Mode of action analysis of a new arylpicolinate herbicide [Abstra ct]. In: Proceedings of the Weed Science Society of America Annual Meeting. WSSA, Lexington, KY: Weed Science Society of A merica http://wssaabstracts.com/public/30/abstract 290.html. Accessed August 13, 2017 Berger S. 2011. Characterization of a suspected herbicide tolerant hybrid watermilfoil ( Myriophyllum spicatum x M. sibiricum ). University of Florida Master of Science Thesis. Agronomy Department. Berger ST, Netherland MD, MacDonald GE. 2012. Evaluating fluridone sensitiv ity of multiple hybrid and Eurasian watermilfoil accessions under mesocosm conditions. Journal of Aquatic Plant Management. 50: 135 144. Berger ST, Ne therland MD, MacDonald GE. 2015. Laboratory documentation of multiple herbicide tolerance to Fluridone, Norflurazon, and Topramazone in a hybrid watermilfoil ( Myriophyllum spicatum x M. sibiricum ) population. Weed Sc i 63(1): 235 241. Bowes G Holaday AS, Haller WT. 1979. Seasonal variation in the biomass, tuber density, and photosynthetic metabolism of hydrilla in three Florida lakes. J. Aquat. Plant Manage. 17:61 65. Bukun B, Lindenmayer RB, Nissen SJ, Westra P, Shaner DL, Brunk G. 2010. Ab sorption and translocation of aminocyclopyrachlor and aminocyclopyrachlor methyl ester in Canada thistle ( Cirsium arvense ). Weed Sci 58:96 102
83 Duy L, Chon NM, Mann RK, Kumar BV, Morell MA. Efficacy of Rinskor TM (florpyrauxifen benzyl ester) on herbicid e resistant b arnyardgrass ( Echinochloa crus galli ) in rice fields of Mekong Delta, Vietnam. J. Crop. Sci. Biotech. 21(1): 75 81. Eggeman D. 1994. Integrated hydrilla management plan utilizing herbicide and triploid grass carp in Lake Istokpoga. In: Procee dings of the Grass Carp Symposium. U.S. Army C. Engineer, Waterway Experiment Station, Vicksburg, MS. pp. 164 166. Ellstrand NC, Schierenbeck KA. 2000. Hybridization as a stimulus for the evolution of invasiveness in plants?. Proc Nat l. Acad Sci U.S.A. 97:13:7043 7 050. Enloe SF, Lym RG, Wilson R, Westra P, Nissen S, Beck G, Moechnig M, Peterson V, Masters RA, Halstvedt M. 2007. Canada thistle ( Cirsium arvense ) control with aminopyralid in range, pasture, and non crop areas. Weed Technol 21:890 894. EPA. 2015. Pe sticide analytical methods: ECM for florpyrauxifen benzyl & degradates in water MRID 49677722. https://www.epa .gov/sites/production/files/2017 09/documents/ecm_ _florpyrauxifen benzyl_degradates_in_water_ _mrid_49677722.pdf Epp JB, Alexander AL, Bakko TW, Buysse AM, Brewster WK, Bryan K, Daeuble JF, Fields SC, Gast RE, Green RA, Irvine NM, Lo WC, Lowe CT, Renga JM, Richburg JS, Ruiz JM, Satchivi NM, Schmitzer PR, Sidall TL, Webster JD, Weimer MR, Whiteker GT, Yerkes CN. 2016. The discovery of Arylex TM active and Rinskor TM active: tw o novel auxin herbicides. Bioorg. Med. Chem. Lett. 24: 362 371. F eiler U, Kirch esch I, Heininger P. 2004. A New Plant based Bioassay for Aquatic Sediments. J Soils Sediments 4(4): 261 266. Gallagher JE, Haller WT. 1990. History and development of aquatic weed control in the United States. Rev in Weed Sci 5:115 192. Getsinger KD, Netherland MD. 1997. Herbicide concentration/exposure time requirements for controlling submersed aquatic plants: summary of research accomplishments. US Army Corps of Engineers Aquatic Plant Control Program. Misc Paper A 97 2. 27 pp. Getsinger KD, Nether land MD. 2018. Use of herbicides in areas of high water exchange: practical considerations. J. Aquat. Plant Manage. 56s: 39 43.
84 Glomski LM, Netherland MD, Nelson LS. 2009. Evalu a tion of 2,4 D and triclopyr against spatterdock and bul rush. APCRP Technica l Notes Collection (ERDC/TN APCRP CC 10). Vicksburg, MS. U.S. Army Engineer Research and Development Center, Vicksburg, MS. Glomski LM Netherland MD 2010. Response of Eurasian and hybrid watermilfoil to low use rates and extended exposures of 2,4 D and triclopyr. J. Aquat. Plant Manage. 48:12 14. Glomski LM, Netherland MD. 2014. Response of waterlily, spatterdock, and hardstem bulrush to liquid and granular triclopyr treatments. J. Aquat. Plant Manage. 52: 81 84. Godfrey RK, Wooten JW. 1981. Aquatic a nd wetland plants of Southeastern Un ited States: dicotyledons. The University of Georgia Press. Grace JB Wetzel RG. 1978. The production biology of E urasian watermilfoil ( Myriophyllum spicatum L.): A review. J. Aquat. Plant Manage. 16:1 11. Gray CJ, Mad sen JD, Wersal R, Getsinger KD. 2007. Eurasian watermilfoil and parrotfeather control using carfentrazone ethyl. J. Aquat. Plant Manage. 45:43 46. Gre en WR Westerdahl HE. 1990. Response of Eurasian watermilfoil to 2,4 D concentration and exposure times. J. Aquat. Plant Manage. 28:27 32. Grossman K. 2010. Auxin herbicides: current status of mechanism and mode of action. Pest. Manag. Sci. 66(2):113 120. Haller WT. 2014. Chapter 7: Mechanical Control of Aquatic Weeds, pp.43 50. In: Gettys LA, Haller WT, Pe tty DG. ( eds. ). Biology and control of aquatic plants: A best management handbook. Third Ed. Aquatic Ecosystem Restoration Foundation, Marietta, GA. 25 2 pp. Haller WT, Jones KD. 2012. Technology and improved efficacy of mechanical control of hydrilla. Aqu atics. Fall 2012:17 19. Heath E, Hoyman T, Nault M, Skogerboe J. 2014. Field research of early season whole lake herbicide strategies for control of hybrid Eurasian watermilfoil [ A bstract]. 54 th Annual Meeting of the Aquatic Plant Management Society, 13 1 6 July 2014, Savannah, GA. p. 30 31. Is rael TD. 2011. Aminocyclopyrachlor efficacy and behavior in selected aquatic plants. MS Thesis. Raleigh, NC: North Carolina State University. 105 p p
85 Knezevic SZ, Streibig JC, Ritz C. 2007. Utilizing R software pack age for dose response studies: the conce pt and data analysis. Weed Technol. 21:840 848. LaRue EA, Zuellig MP, Netherland MD, Heilman MA, Thum RA. 2013. Hybrid watermilfoil lineages are more invasive and less sensitive to a commonly used herbicide than the ir exotic parent (Eurasian watermilfoil). Evol Appl 6:462 471. Lee S, Sundaram S, Armi tage L, Evans JP, Hawkes T, Kepinski S, Ferro N, Napier RM. 2013. Defining binding efficiency and specificity of auxins for SCFTIR1/AFB Aux/IAA co receptor complex for mation. ACS Chem Biol 9:673 682 Li F, Zhu L, Xie Y, Jiang L, Chen X, Deng Z, Pan B. 2 015. Colonization by fragments of the submerged aquatic macrophyte Myriophyllum spicatum under different sediment type and density conditions. Sci. Rep. 5:11821 Madse n JD. 2015. Chapter 1: Impact of invasive aquatic plants on aquatic biology pp 1 8. In: Gettys LA, Haller WT, Petty DG. ( eds. ). Biology and control of aquatic plants: A best management handbook. Third Ed. Aquatic Ecosystem Restoration Foundation, Marietta GA. 252 pp. Madsen JD, Smith DH. 1997. Vegetative spread of Eurasian watermilfoil col onies. J. Aquat. Plant Manage. 35: 63 68 Madsen JD, Sutherland JW, Bloomfield JA, Eichler LW Boylen CW. 1991. The decline of native vegetation under dense Eurasian watermilfoil canopies. J. Aquat. Plant Manage. 29:94 99 Madsen JD 2000. Advantages and disadvantages of aquatic plant management techniques. APCRP Technical Notes Collection (ERDC/EL APCRP MP 00 1). Vicksburg, MS. U.S. Army Engineer Research and Develop ment Center, Vicksburg, MS. 38 pp. Madsen JD, Getsinger KD, Stewart RM, Owens CS 2002. Whole lake fluridone treatments for selective control of Eurasian watermilfoil: II. Impacts on submersed plant communities. Lake and Reservoir Manage. 18:3: 191 200. Madsen JD, Wersal RM, Getsinger KD, Skogerboe JG. 2010. Combinations of endothall w ith 2,4 D and triclopyr for Eurasian watermilfoil control. APCRP Technical Notes Collection (ERDC/TN APCRP CC 14). Vicksburg, MS. U.S. Army Engineer Research and Development Center, Vicksburg MS. 11 pp. McFarland DG, Barko JW. 1987. Effects of temperatur e and sediment type on growth and morphology of monoecious and dioecious hydrilla. J. Freshw. Ecol. 4(2): 245 252.
86 Miller MR, Norsworthy JK, Sco t t RC. 2017. Evaluation of f lorpyrauxifen benzyl on herbicide resistant and herbicide susceptible barnyargrass accessions. Weed Technol. 32(2):126 134. Miller MR, Norsworthy JK. 2018. Influence of soil moisture on absorption, translocation, and metabolism of florpyrauxifen benzyl. W eed Sci. 66(4):418 423. Minogue PJ, Enloe SF, Osiecka A, Lauer DK. 2011. Compariso n of aminocyclopyrachlor to common herbicides for kudzu ( Pueraria montana ) management. Inv. Plant Sci Manage. 4:419 426. Moody ML Les DH. 2002. Evidence of hybridity in i nvasive watermilfoil ( Myriophyllum ) populations. P. Natl. Acad. Sci. USA 99:14867 1 4871. Moody ML, Les DH. 2007. Geographic distribution and genotypic composition of invasive hybrid watermilfoil ( Myriophyllum spicatum x M. sibiricum ) populations in North America. Biol. Invasions. 9:559 570. Moody ML Les DH. 2010. Systematic s of the aquatic angiosperm genus Myriophyllum (Haloragaceae). Syst. Bot. 35:121 139. Mudge CR. 2018. Propagation methods of submersed, emergent and floating plants for research. J. Aquat. Plant Manage. Research Methods 56:2 9. Nault ME, Netherland MD, Mikulyuk A, Skogerboe JG, Asplund T, Hauxwell J, Toshner P. 2014. Efficacy, selectivity, and herbicide concentrations following a whole lake 2,4 D application targeting Eurasian waterm ilfoil in two adjacent northern Wisconsin lakes Lake Reserv Manage 30: 1 10. Netherland MD, Getsinger KD 1992. Efficacy of triclopyr on Eurasian watermilfoil: concentration and exposure time effects. J. Aquat. Plant Manage. 30:1 5. Netherland MD Glo mski LM. 2014. Mesocosm evaluation of triclopyr on Eurasian watermilfoil and three native submersed species: the role of treatment timing and herbicide exposure. J. Aquat. Plant Manage. 52: 57 64. Netherland MD, Jones D. 2015. Fluridone resistant Hydr illa ( Hydrilla verticillata ) is still dominant in the Kissimmee chain of Lakes, FL. Invas. Plant Sci. Manage. 8(2): 212 218 Netherland MD Richardson RJ. 2016. Evaluating sensitivity of five aquatic plants to a novel arylpicolinate herbicide utilizing an Org anization for Economic Cooperation and Development Protocol. Weed Sci 64(1):181 190.
87 Netherland MD Willey L. 2017. Mesocosm evaluation of three herbicides on Eurasian watermilfoil ( Myriophyllum spicatum ) and hybrid watermilfoil ( Myriophyllum spicatum x Myriophyllum sibiricum ): Developing a predictive assay. J. Aquat Plant Manage 55: 38 41 Newman RM, Holmberg KL, Biesboer DD, Penner BG.1996. Effects of a potential biocontrol agent, Euhrychiopsis lecontei on Eurasian watermilfo il in experimental tank s. Aquat. Botany 53:131 150. Newman RM Inglis WG. 2009. Distribution and abundance of the milfoil weevil Euhrychiopsis lecontei, in Lake Minnetonka and relation to milfoil harvesting. J. Aquat. Plant Manage. 47:21 25. OECD. 2014. Test No. 239: Water sed iment Myriophyllum spicatum toxicity test, OECD Guidelines for the Testing of Chemicals, Section 2. Paris: OECD Publishing. DOI: http://dx.doi.org/10.1787/978926424155 en Pa rks SR, McNair JN, Haus ler P, Tyning P, Thum RA. 2016. Divergent responses of cryptic invasive watermilfoil to treatment with auxinic herbicides in a large Michigan lake. Lake Reserv. Manag e 32(4): 366 372. Pimentel D, Zuniga R, Morrison D. 2004. Update on the environmental a nd economic costs associated with alien invasive species in the United States. Ecological. Economics 52:273 288. Poovey AG, Slade JG, Netherland MD. 2007. Susceptibility of Eurasian watermilfoil ( Myriophyllum spicatum ) and a milfoil hybrid ( M. spicatum x M. sibiricum ) to triclopyr and 2, 4 D amine. J. Aquat. Plant Manage. 45:111 115. Reddy KR, DeLaune RD. 2008. Biogeochemistry of wetlands. CRC Press, Taylor & Francis Group. 806 pp. Richardson RJ. 2008. Aquatic plant manageme nt and the impact of emerging herbicide resistance issues. Weed Technology. 22(1):8 15. Richardson RJ, Haug EJ, Netherland MD. 2016. Response of seven aquatic plants to a new arylpicolinate herbicide. J. Aquat. Plant Manage. 54:26 31 Rieseberg LH, Ellst rand NC, Arnold M. 1992. What can molecular and morphological markers tell us about plant hybridization? Crit. Rev. Plant Sci. 12(3): 213 241. Sakai AK, Allendorf FW, Jolt JS, Lodge DM, Molofsky J, With KA, Baughman S, Cabin IM, Thompson JN, Weller SG. 2001. The population biology of invasive species. Annu. Rev. Ecol. Syst 32:305 332.
88 Sculthorpe CD. 1967. The biology of aquatic vascular plants. Edward Arnold (Pub.) Lts. London. 610 pp. Skogerboe JG, Getsinger KD. 2002. Endo thall species selectivity evaluation: Northern latitude aquatic plant community. J. Aquat. Plant Manage. 40:1 5. Skogerboe JG, Getsinger KD, Glomski LM. 2006. Effica cy of diquat on submersed plants treated under simulated flowing water conditions. J. Aqua t. Plant Manage. 44: 122 125. Smart RM, Barko JW. 1985. Laboratory culture of submersed freshwater macrophytes on natural sediments. Aquat Bot. 21(3): 251 263. Smi th CS Barko JW. 1990. Ecology of Eurasian watermilfoil. J. Aquat. Plant Manage. 28:55 64. Stanley RA Naylor AW. 1972. Photosynthesis in Eurasian watermilfoil ( Myriophyllum spicatum L.). Plant Physiol. 50:149 151. Stanley RA, Shackelford E, Wade D, Warr en C. 1975. Effects of season and water depth on Eurasian watermilfoil. J. Aquat. Plant Ma nage. 14:32 36. Sterling TM, Hall JC. 1997. Mechanism of action of natural auxins and the auxinic mimic herbicides. Crit. Rev Toxicol. 1(3):111 141 Sturtevant AP, Hatley N, Pullman GD, Sheick R, Shorez D, Bordine A, Mausolf R, Lewis A, Sutter R, Mortim er A. 2009. Molecular characterization of Eurasian watermilfoil, northern watermilfoil, and the invasive interspecific hybrid in Michigan lakes. J. Aquat. Plant Manag e. 47:128 135. Tarver DP. 1980. Water fluctuation and the aquatic flora of Lake Miccosuke e. J. Aquat. Plant Manage. 18:19 23. Taylor LL, McNair JN, Guastello P, Pashnick J, Thum RA. 2017. Heritable variation for vegetative growth rate in ten distinct genotypes of hybrid watermilfoil. J. Aquat. Plant Manage. 55: 51 57. Thompson JD. 1991. The biology of an invasive plant. Bioscience. 41:393 401. Thum RA. 2018. Genetic variation and aquatic plant management: key concepts and practical implications. J. Aquat. Plant Manage. 56s:101 106. Thum, RA, Heilman M, Hausler P, Huberty L, Tyning P, Wcisel D, Zuellig M, Berger S, Netherland M. 2012. Field and laboratory documentation of reduced fluridone sensitivity of a hybrid watermilfoil biotype ( Myriophyllum spicatum x Myriophyllum sibiricum ). J. Aquat. Plant Manage. 50:141 146.
89 Thum R, Lennon JT, Conn or J, Smagula AP. 2006. A DNA fingerprinting approach f or distinguishing native and non native milfoils. Lake Reserv Manage. 22:1: 1 6. Thum RA, McNair JN. 2018. Inter and intraspecific hybridization affects germination and vegetative growth in Eurasian watermilfoil. J. Aquat. Plant Manage. 56: 24 30. Thum R, Newman R, Fieldseth E. 2017. Occurrence and distribution of Eurasian, Northern and hybrid watermilfoil in Lake Minnetonka and Christmas Lake: genetic analysis phase II. AIS Grant Report, Hennepin C ounty, MN. 15 pp. True Meadows S, Haug EJ, Richardson RJ. 2016. Monoecious hydrilla A review of the literature. J. Aquat. Plant Manage. 54:1 11. University of Minnesota Aquatic Invasive Species Research Center. 2018. Eurasian watermilfoil control options https://www.maisrc.umn.edu/ewm control USDA, NRCS (United States Department of Agriculture, Natural Resources Conservation Service). 2011. The PLANTS database. (http://plants.usda.gov, 2 February 2011). National Plant Data Center, Baton Rouge, LA 7087 4 4490 USA. Van TK, Haller WT, Bowes G. 1976. Comparison of the photosynthetic characteristics of three submersed aquatic plants. Plant Physiol. 58:761 768. Washington State Department of Ecology. 2 017. Supplemental environmental impact statement for st ate of Washington aquatic plant and algae management. Washington Dept. of Eco. Pub. No. 17 10 020. 205 pp Wersal RM, Madsen JD, Woolf TE, Eckberg N. 2010. Assessment of herbicide efficacy on Eurasian watermilfoil and impacts to the native submersed plant community in Hayden Lake, Idaho, USA. J. Aquat. Plant Manage. 48:5 11. Wu Z, Yu D, Xu XW. 2013. Development of microsatellite markers in the hexaploidy aquatic macrophyte, Myriophyllum spicatum (Halo ragaceae). Appl. Plant Sci. 1(2) : 12000230 Zuellig MP Thum R. 2012. Multiple introductions of invasive Eurasian watermilfoil and recurrent hybridization with northern watermilfoil in North America. J. Aquat. Plant Manage. 50: 1 19.
90 BIOGRAPHICAL SKETCH Jens Beets was bor n in Richmond, Virginia in 1992 to Dr. Lisa Muehlstein and Dr. Jim Beets, both professors at University of Richmond at the time After graduating from Hilo High School, he attended University of Puget Sound and graduated with a Bachelor o f Science degree, majoring in b iology. After completing his degree Jens moved to Gainesville to work as an OPS employee for Dr. Mike Netherland. This developed his interest in aquatic invasive plant management and led to pursuing a Master of Science degre e under Dr. Nether land. In addition to his thesis studies, Jens has performed several studies involving the native grass Paspalidium geminatum (Firssk.) Stapf. With Florida Fish and Wildlife Conservation Commission funding. After completion of his Master of Science, Jens in tends to enroll at North Carolina State University and pursue a Doctor of Philosophy with Dr. Rob Richardson in Fish and Wildlife.