Evaluation of Saflufenacil on Glyphosate-Resistant And

ABSTRACT

EVALUATION OF SAFLUFENACIL ON GLYPHOSATE- RESISTANT AND GLYPHOSATE-PARAQUAT RESISTANT HAIRY FLEABANE (Conyza bonariensis)

Hairy fleabane [Conyza bonariensis (L.) Cronq.] is a problematic weed in crop and non-crop areas of California. This problem has been further aggravated by the discovery of herbicide-resistant biotypes. Three experiments were conducted to determine the efficacy of saflufenacil (TreevixTM), a fairly new herbicide, on glyphosate-susceptible (GS), glyphosate-resistant (GR), and glyphosate-paraquat-resistant (GPR) hairy fleabane plants. The studies evaluated the efficacy of saflufenacil when applied alone or in combination with glyphosate at: a) three growth stages (5- to 8-leaf seedling, rosette, and bolting); b) three temperature regimes (15/10ºC, 25/20ºC, 35/30ºC at day/night); and c) three light regimes (100%, 50%, 30%, 0% of full sun). Results differed between experiments conducted in the spring and fall. Saflufenacil-alone was more effective in the fall than in spring. All the GS, GR and GPR plants were controlled by saflufenacil- alone at the 5-to 8-leaf seedling and rosette stage but level of control declined at the bolting stage. Better control was obtained at the 15/10ºC and 25/20ºC than at the 35/30ºC temperature regime. Light regime had no effect on the efficacy of saflufenacil. Efficacy of saflufenacil-alone was inconsistent in spring and varied between the biotypes. Therefore, saflufenacil-alone can provide excellent control of hairy fleabane plants prior to the bolting stage in the fall; but in spring, it will be more effective when applied with glyphosate.

Michelle Dennis May 2015

EVALUATION OF SAFLUFENACIL ON GLYPHOSATE- RESISTANT AND GLYPHOSATE-PARAQUAT RESISTANT HAIRY FLEABANE (Conyza bonariensis)

by Michelle Dennis

A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Plant Science in the Jordan College of Agricultural Sciences and Technology California State University, Fresno May 2015

We, the undersigned, certify that the thesis of the following student meets the required standards of scholarship, format, and style of the university and the student's graduate degree program for the awarding of the master's degree.

Michelle Dennis Thesis Author

Anil Shrestha (Chair) Plant Science

John Bushoven Plant Science

Kurt Hembree University of California, Cooperative Extension

For the University Graduate Committee:

Dean, Division of Graduate Studies AUTHORIZATION FOR REPRODUCTION OF MASTER’S THESIS

X I grant permission for the reproduction of this thesis in part or in its entirety without further authorization from me, on the condition that the person or agency requesting reproduction absorbs the cost and provides proper acknowledgment of authorship.

Permission to reproduce this thesis in part or in its entirety must be obtained from me.

Signature of thesis author: ACKNOWLEDGMENTS I would like to dedicate this thesis to my two children, Rio Renee and Ty “Ox” Dennis. If fate ever leads you to read this when you are grown, realize that it is not the printed words providing clarity to well thought-out studies that created this document. It is all of the mistakes that were made, the trials that were thrown out, and the pages of gibberish that needed to be rewritten that brought me to the completion of this document and the true discoveries. To my husband, Scott, thank you for supporting my journey. For the first time in our 15 years of marriage, I truly can’t find the words. To my parents, thank you for having faith in me every step of the way. To my committee, especially my advisor, Anil Shrestha, thank you for not accepting anything less than 100%. Your insights, questions and critiques during this process have given me more confidence in my own abilities going forward. The world of research was foreign to me and now I have a roadmap to follow. To my fellow graduate student, Sonia Rios, I couldn’t have finished this project without the help and constant motivation that you provided to everyone in the weed science program. I wish you nothing but success in your future. I would also like to thank all of the students, faculty and staff of the Jordan College of Agricultural Sciences and Technology at California State University, Fresno that contributed to this study and the California State University Agricultural Research Institute for funding this project. TABLE OF CONTENTS Page

LIST OF TABLES ...... vii

INTRODUCTION ...... 1

LITERATURE REVIEW ...... 2

Hairy Fleabane Biology and Dispersal ...... 2

Evolution of Herbicide Resistant Hairy Fleabane in the Central Valley ...... 3

Need for Immediate Alternative Herbicides ...... 5

PPO Inhibitors ...... 6 Interaction of Herbicides with Plant Physiology and Environmental Factors ...... 9

Objectives ...... 11 EXPERIMENT 1 – EFFECT OF GROWTH STAGE OF HAIRY FLEABANE ON THE EFFICACY OF ALTERNATIVE HERBICIDES ...... 13

Methods and Materials ...... 13

Results and Discussion ...... 16

EXPERIMENT 2 – TEMPERATURE ...... 27

Methods and Materials ...... 27

Discussion ...... 31

EXPERIMENT 3 – LIGHT INTENSITY ...... 40

Methods and Materials ...... 40

Results and Discussion ...... 43

CONCLUSION ...... 51

REFERENCES ...... 54 LIST OF TABLES

Page

Table 1. Transplanting and herbicide application dates of the different targeted growth stages of glyphosate-susceptible, glyphosate+paraquat resistant, and glyphosate-resistant hairy fleabane plants in 2012 and 2013...... 14 Table 2. Mortality of glyphosate-susceptible (GS), glyphosate+paraquat resistant (GPR), glyphosate-resistant (GS), and aboveground biomass (mean of all three biotypes) of hairy fleabane plants treated with herbicides at the 5- to 8-leaf stage in Fall 2012 and Spring 2013...... 18 Table 3. Mortality and aboveground biomass of glyphosate-susceptible (GS), glyphosate+paraquat resistant (GPR), glyphosate-resistant (GS) of hairy fleabane plants treated with herbicides at the rosette stage...... 21 Table 4. Mortality of glyphosate-susceptible (GS), glyphosate+paraquat resistant (GPR), glyphosate-resistant (GS) in fall 2012 and spring 2013, and aboveground biomass (mean of all three biotypes) in fall 2012 and each biotype in spring 2013 of hairy fleabane plants treated with herbicides at the bolting stage...... 23 Table 5. Transplanting and herbicide application dates of glyphosate- susceptible, glyphosate-resistant, and glyphosate+paraquat resistant hairy fleabane plants in 2012 and 2013...... 28 Table 6. Analysis of variance (ANOVA) table showing the main effects and interactions for plant mortality and aboveground plant biomass of the glyphosate-susceptible, glyphosate-resistant, and glyphosate-paraquat resistant populations of hairy fleabane exposed to three different temperature regimes and treated with various herbicides in 2012 and 2013...... 32 Table 7. Plant mortality and aboveground biomass of hairy fleabane plants (averaged for glyphosate-susceptible, glyphosate-resistant, and glyphosate+paraquat resistant biotypes) after treatment and exposure to 15/10˚C temperature...... 34 Table 8. Plant mortality and aboveground biomass of hairy fleabane plants (averaged for glyphosate-susceptible, glyphosate-resistant, and glyphosate+paraquat resistant biotypes) after treatment and exposure to 25/20˚C temperature...... 35

viii viii Page

Table 9. Plant mortality of glyphosate-susceptible (GS), glyphosate-resistant (GR), and glyphosate+paraquat resistant (GPR) hairy fleabane plants in 2012 and 2013 (averaged for the three biotypes) and aboveground biomass (averaged for the three biotypes in both 2012 and 2013) after treatment and exposure to 35/30˚C temperature...... 37 Table 10. Transplanting and herbicide application dates of glyphosate- susceptible, glyphosate-resistant, and glyphosate+paraquat resistant hairy fleabane plants in 2013 and 2014...... 41 Table 11. Analysis of variance (ANOVA) table showing the main effects and interactions for plant mortality and aboveground plant biomass in spring (average of 2013 and 2014) and fall 2013 of glyphosate- susceptible, glyphosate-resistant, and glyphosate-paraquat resistant populations of hairy fleabane treated with various herbicides and exposed to four different light intensities...... 44 Table 12. Mortality of glyphosate-susceptible (GS), glyphosate+paraquat resistant (GPR), glyphosate-resistant (GS), of hairy fleabane plants treated with herbicides at the 5- to 8-leaf stage in spring (average of 2013/14) and exposed to different light regimes...... 45 Table 13. Above ground biomass of hairy fleabane plants (averaged for the light regimes and three biotypes) treated with herbicides at the 5- to 8-leaf stage and exposed to different light regimes in spring 2013/14. ... 47 Table 14. Mortality of hairy fleabane plants treated with various herbicides and exposed to various light regimes at the 5- to 8-leaf stage in fall 2013...... 48 Table 15. Biomass of hairy fleabane plants treated with herbicides at the 5- to 8-leaf stage in fall 2013...... 49

INTRODUCTION

Hairy fleabane [Conyza bonariensis (L.) Cronq.] is a common and problematic weed in California. In the past 10 years, it has emerged as a strong competitor in both perennial and annual cropping systems as well as in non-crop systems such as roadsides and canal banks (Shrestha et al. 2008). There are two possible reasons for the increased prevalence of this species in perennial crop and non-crop areas. The first is that hairy fleabane prefers areas with less soil disturbance. The other possibility is the evolution of herbicide-resistant populations of hairy fleabane, particularly glyphosate-resistant (GR) and glyphosate+paraquat-resistant (GPR) biotypes. Glyphosate-resistant hairy fleabane was first documented in California in the Central Valley in 2008 (Shrestha et al. 2008). An additional GPR biotype was discovered in the same region in 2010 (Moretti et al. 2013). Glufosinate was identified as an immediate alternative for successful control of these herbicide-resistant biotypes (Moretti et al. 2010). However, in order to prevent the onset of the evolution of glufosinate- resistant populations, alternative strategies are needed to combat this increasingly difficult-to-control weed species. Therefore, mode of actions of active ingredients other than glyphosate, paraquat, and glufosinate are required for a good resistance management plans. Equally important is the need for developing an understanding of the biology of this weed and its response to herbicide applications at various growth stages and effects of environmental factors at the time of treatment on efficacy of alternative herbicides.

LITERATURE REVIEW

Hairy Fleabane Biology and Dispersal The exact origin of hairy fleabane is unknown but it was first described in Argentina; therefore, it most likely originated in Central or South America (Michael 1977). This species belongs to the family Asteraceae, the largest dicot family on earth. The genus Conyza is composed of approximately 60 species of annual herbaceous plants that prefer tropical and subtropical environments (Nesom 1990), but hairy fleabane prefers temperate zones and open habitats. Conyza spp. have extensive reproductive and dispersal capabilities and are found on every continent and terrestrial environment except Antarctica (Cronquist 1980). In the United States, hairy fleabane is distributed primarily in the southwestern and southern states. Hairy fleabane is considered to be a summer annual plant. However, in California, plants overwinter as a rosette and can have multiple generations per year with greater emergence in spring and fall than during other times of the year (Shrestha et al. 2008). Active root growth can continue during the winter, and hairy fleabane plants have been found to have winter taproots at levels >35 cm deep in the soil, enabling it to survive dry soil conditions (Wu et al. 2007). Hairy fleabane plants usually grow to 0.5 to 1 m tall. Shrestha et al. (2014) reported that hairy fleabane plants from the Central Valley of California produced 89,000 to 103,000 seeds per plant. An average of 119,000 from a single plant with 80% viability has been reported in Australia (Wu et al. 2007). Seeds of this species are equipped with a feathery pappus and have a low settling velocity, thus increasing their wind dispersal potential (Andersen 1993). Horseweed (Conyza canadensis L. Cronq.), a close relative of hairy fleabane with 3 3 a similar seed structure, has seeds that have been found to ascend up to 120 – 140 m above ground level, showing its ability to rise above the atmosphere surface layer into the planetary boundary layer where it may be carried for hours prior to descent (Dauer 2009; Shields 2006). Although the percentage of total seed dispersal over distances of 100 m is small, studies have shown that horseweed seeds can travel at least 500 m from their source during a dispersal event (Mortensen et al. 2006; Mortensen 2007). This shows the potential magnitude of long-range dispersal capabilities via wind of both horseweed and hairy fleabane.

Evolution of Herbicide Resistant Hairy Fleabane in the Central Valley In recent years, researchers have attempted to define traits that make a weed invasive as weed populations are discovered that have evolved to become invasive, even in their native habitats (Van Kluenen 2010). Another concern is a weed’s ability to compete with agricultural crops and persist in both managed and unmanaged environments. Factors that ensure the persistence of a weed depends on their densities, the extent of their reproductive output, the surface area they cover, the range of habitats and their potential for putting future generations in a position to continue through time (Baker 1974). Plants belonging to the genus Conyza are a recent example of weed species increasing in prominence globally. Minimum soil disturbance associated with fallows, perennial cropping systems, and reduced tillage management practices of annual crops has provided a favorable niche for the ecological adaptation of Conyzas (Murphy 2006). Reduced tillage and zero tillage systems have a favorable impact on the environment; these systems are known to conserve soil and water resources, as well as improve air quality (Warnert 2012; Madden et al. 2008). These practices have caused weed species such as Conyzas that were not considered a problem in conventional tillage 4 4 practices to rapidly increase in conservation tillage systems. One reason for adaptation of Conyzas to conservation or zero-tillage systems could be that their seeds germinate from the surface of the soil and very little light is needed for germination (Nandula et al. 2006). Hairy fleabane is commonly found in the Central Valley in vineyards, roadsides, canal banks, and fallowed fields. All of these areas are similar in that they are not regularly tilled for planting or weed control. When the seeds are not buried by tillage, the system serves as a niche for Conyzas. However, California is one of the last states to incorporate conservation tillage as a specific farming practice (Horowitz 2010). For this reason, it is more likely that the increased invasiveness of hairy fleabane in Central California is due to an increase in herbicide selection rather than a trend toward conservation tillage. One of the first areas that herbicide-resistant Conyza spp. were found in the Central Valley was on a canal bank that had been sprayed repeatedly with glyphosate over multiple years (Shrestha et al. 2007). The extensive and repeated use of herbicides with similar modes of action (MOA) has rapidly increased the natural selection for herbicide-resistant weeds that cannot be easily managed. Defined simply, herbicide resistance is the inherited ability of a plant to survive and reproduce after exposure to a dose of an herbicide that would be normally lethal to the wild type (Prather et al. 2000). Although herbicides are not known to induce mutations within the plant, continual use allows herbicide-resistant survivors to become the dominant population. Resistance to glyphosate and other common herbicides in several weed species now threatens the continued reliability and utility of these herbicides. Recent reports show 238 unique cases of weeds representing 31 species that are resistant to glyphosate herbicides worldwide (Heap 2015). In California, the first case of herbicide resistance was reported in 5 5

1981 on common groundsel (Senecio vulgaris), which showed resistance to herbicides in the triazine chemical class (Holt 1988). California now has 30 cases of resistant weeds impacting 8 different herbicide sites of action (Heap 2015). Weeds in the genus Conyza have been identified as belonging to the top 10 worst herbicide-resistant weeds worldwide (Heap 2011). Hairy fleabane is currently considered the most difficult weed to control in cropping systems in Australia (Widderick 2012). Horseweed, a close relative of hairy fleabane with similar dispersal capabilities, was the first dicotyledonous plant to evolve resistance to glyphosate (VanGessel 2001). It has since become the most widespread GR weed globally (Heap 2011). Its rapid distribution has been attributed to its dispersal abilities. Hairy fleabane has evolved resistance to bipyridilium, glycine, sulfonylurea, and triazine herbicides in at least 11 countries worldwide. Cases such as these have resulted in renewed local and global interest in investments in new herbicide chemistry.

Need for Immediate Alternative Herbicides The herbicide industry seems to be showing some signs of comeback after several years of downturn after the discovery of glyphosate (Duke 2012). This comeback is partly fueled by the need for new chemistries to control GR weed biotypes. Alternate postemergence herbicides are also being sought for the control of GR biotypes of Conyza spp. including hairy fleabane. For immediate management of herbicide-resistant weed biotypes, alternative herbicides need to be identified as a short-term solution. One such new herbicide recently introduced in California is saflufenacil (Treevix ™), a protoporphyrinogen oxidase (PPO) inhibiting herbicide. Although there are several PPO inhibiting herbicides such saflufenacil, flumioxazin, carfentrazone, pyraflufen, and oxyfluorfen registered for 6 6 use in permanent crops in California with the same MOA, their main chemical compositions differ (Vencill 2002). Of the PPO herbicides listed above, saflufenacil is currently the only one that includes postemergence control of hairy fleabane on its label while pyraflufen (Venue ®) mentions control of hairy fleabane in a supplemental label when tank-mixed with glyphosate. The active ingredient saflufenacil has been registered for postemergence control of broadleaf weeds in citrus, nuts, and pome fruits. Preliminary studies at California State University, Fresno showed that saflufenacil was effective against GR horseweed, another problematic weed in the Central Valley (Shrestha and Moretti 2011). Although in 2000 there were no known cases of resistance to PPO inhibitors world-wide (Prather et al. 2000), by 2015 there are 6 species in 5 countries that have documented cases of resistance to PPO inhibitors, including several cases of multiple herbicide resistance in these species (Heap 2015). Two of these species are located in the United States, prompting a closer look at the properties of PPO-inhibiting herbicides.

PPO Inhibitors Protoporphyrinogen oxidase (PPO) is an enzyme that is located in the chloroplast thylakoid and mitochondria of plants. It is a precursor enzyme involved in the dual biosynthesis of the products heme and chlorophyll. Normally, PPO catalyses the oxidation of protoporphyrinogen to protoporphyrin by molecular oxygen. The inhibition of the PPO enzyme causes an accumulation of the product protoporphyrin, rather than the substrate (Matringe et al. 1989). The herbicidal effects are light dependent, meaning if a PPO inhibited plant is maintained in a dark area the effects will not be seen (Falk et al. 2006). In the 7 7 presence of light, protoporphyrin is excited to the triplet state and interacts with molecular oxygen (O2) to produce singlet oxygen (Hess 2000). This leads to light dependent peroxidation which targets the double bonds of fatty and amino acids within the plant, causing herbicidal effects (Hess 2000). Damage is obvious, in the form of necrotic lesions that are highly visible on the plant tissue. Not all plants respond identically to the application of PPO inhibiting herbicides. Sherman et al. (1991) compared several plants’ reactions to applications of acifluorfen (another PPO-inhibiting herbicide) and found that mustard (Brassica hirta Moench.) and spinach (Spinacia oleracea L.) did not have a high level of protoporphyrin build up after application. It was concluded that this was most likely due to a limited capacity to synthesize substrate in these species, because large amounts were seen after the addition of 5-aminole-vulinic acid, (ALA) plus acifluorfen (Sherman et al 1991). An alternative explanation for tolerance to PPO inhibiting herbicides is the plant’s ability to detoxify singlet oxygen. Studies have also been conducted in other light-dependent herbicide MOAs; a study by Vaughn et al. (1989) that used paraquat, an herbicide that causes cellular damage by generating singlet oxygen, showed that paraquat- resistant hairy fleabane also had a higher level of resistance (10-fold) to singlet oxygen generators diquat and rose bengal. The same study also compared resistance levels of hairy fleabane to other toxic oxygen-generating compounds such as morfamquat, metronidazole, and juglone in which little to no resistance was observed. A third mechanism of resistance to PPO inhibiting herbicides showed a codon deletion in a gene designated as PPX2L which allowed a mutation in a single gene to manifest resistance in both plastid and mitochondrial PPO isoforms (Patzoldt et al. 2006). This is unique in that it was an amino acid deletion, rather than substitution that conferred resistance. 8 8

It is a common practice for growers to use tank-mix applications of herbicides with different MOAs as a resistance management strategy. There has been some work done to explore effects of PPO inhibitors when applied as a tank- mix with glyphosate. Some studies have suggested that the mixture of saflufenacil and glyphosate can create a change in the efficacy of each herbicide. For example, Eubank et al. (2013) reported an additive effect when using a combination of glyphosate with saflufenacil on horseweed. Absorption and translocation effects of the combination were mixed with increased absorption of glyphosate in glyphosate-susceptible (GS) biotypes, reduced absorption in GR biotypes, and reduced glyphosate translocation in both. In a separate study comparing the efficacy of a saflufenacil-glyphosate combination on GS and GR canola (Brassica napus L.) varieties, it was found that the combination reduced translocation of saflufenacil in GS varieties. However, in GR varieties, the combination did not affect the translocation of saflufenacil (Ashigh and Hall, 2010). Additionally, Waggoner et al. (2011) showed a greater benefit to using a mix of saflufenacil and glyphosate on horseweed versus saflufenacil-alone in no-till cotton (Gossypium sp.). Shrestha and Moretti (2011) showed no additional benefit for control of GR horseweed with a tank-mix of saflufenacil and glyphosate compared to saflufenacil-alone. These discoveries show that the mechanism of resistance to PPO inhibitors as well as the causes of resistance development within weed species is not well understood. Therefore, a knowledge-based approach is needed to deal with the continuing evolution of the plant-herbicide relationship including the environmental factors that affect this relationship. 9 9 Interaction of Herbicides with Plant Physiology and Environmental Factors

Developmental Stage Growth stage was shown to be a major factor in the level of control of horseweed by glyphosate (Shrestha et al. 2007; VanGessel et al. 2009), and thus this factor could also influence the efficacy of other postemergence herbicides on hairy fleabane. A greenhouse study by Gonzalez-Torralva et al. (2010), using hairy fleabane seeds that had been collected from areas not previously exposed to glyphosate showed significant differences in susceptibility when treated with glyphosate at three developmental stages (rosette, bolting, and flowering), with early developmental stages being much more sensitive. Although the MOA of glyphosate and saflufenacil are not similar, it should be ascertained if similar effects of plant growth stage can be expected with saflufenacil or a tank-mix of glyphosate and saflufenacil.

Temperature An environmental factor of interest influencing herbicide efficacy is temperature. A study conducted in Israel compared the response of several populations of hairy fleabane and horseweed to glyphosate at day/night thermo- periods of 16/100C, 22/160C, 28/220C, and 34/280C. The results showed a significant negative linear correlation between rising temperature and plant response to glyphosate in terms of effective dose (ED50) values. Plants grown at higher temperatures were 2- to 10-fold more tolerant to glyphosate than at lower temperatures (Rubin et al. 2011). However, it is unknown if there will be an effect of temperature on control of GR and GPR hairy fleabane biotypes of the Central Valley with glyphosate or saflufenacil. This must be ascertained because hairy fleabane emerges at different times of the year, primarily in spring and fall 10 10 months, but also in late spring and early summer. Herbicide efficacy could be greatly affected by its application during these periods of considerable temperature variations.

Light Light intensity can also have a major effect on the activity and efficacy of several postemergence herbicides including saflufenacil. Plants exhibit rapid, light-dependent foliar necrosis when exposed to PPO inhibitors, including saflufenacil (Liebel 2009). Herbicides with this MOA are light-dependent (Hess 2000), and their effectiveness may be enhanced or inhibited by application during different light levels. For example, preliminary studies by Mellendorf et al. (2015) reported that the efficacy of saflufenacil on horseweed was greater under low light (300 µmol m2 sec-1) than under high light (1000 µmol m2 sec-1) intensities. This suggests that there may be a problem with translocation of the herbicide when it is applied to Conyza spp. in high light settings. Another study observing different hairy fleabane biotypes exposed to high light intensity after paraquat treatments showed less photoinhibition in the resistant versus the susceptible biotype (Jansen et al. 1989). Therefore, the effect of light intensity during the time of saflufenacil application should be studied. The evolution of herbicide-resistant weeds has led to a need for knowledge- based integrated weed management (IWM) programs (Sanyal et al. 2008; Swanton et al. 2008). Recommendations include use of cultural practices, rotating herbicides with alternative MOAs, and application timing based on the biology of the target pest. To increase herbicide efficacy, our knowledge should involve both biology of the pest and interactions with environmental factors. A knowledge- based approach to weed management will reduce the overall cost to the grower, 11 11 increase efficacy of the pest management system, and reduce the impact on the environment. To ascertain the biological and environmental factors necessary to achieve the greatest benefit from the use of saflufenacil as an effective herbicide against all biotypes of hairy fleabane, it is essential to explore the effect of factors such as plant growth stage, light intensity, and temperature at the time of herbicide application. Environmental factors and biotype could also have an effect on herbicide efficacy when they are applied alone or as tank-mixes. Saflufenacil has been tested as an alternative to glyphosate for control of GR horseweed in several cropping systems in the US (Eubank et al. 2013; Owen et al. 2011; Waggoner et al. 2011) and on GR hairy fleabane of the Central Valley (Shrestha and Moretti, 2011). However, saflufenacil’s effect on GR and GPR hairy fleabane in response to growth stage, light, and temperature are unknown. Therefore, the effects of growth stage of the weed and environmental (light and temperature) conditions during saflufenacil applications alone or in combination with glyphosate need to be tested.

Objectives 1. Evaluate the efficacy of saflufenacil on glyphosate-resistant (GR), glyphosate and paraquat resistant (GPR), and glyphosate-susceptible (GS) hairy fleabane biotypes at three different growth stages of the plants: 5- to 8-leaf stage, rosette stage (15- to 20-leaf stage), and initial bolting stage. 2. Evaluate the effect of temperature on the efficacy of saflufenacil and other herbicides on GR, GPR, and GS hairy fleabane biotypes. 12 12

3. Evaluate the effect of light intensity on the efficacy of saflufenacil and other herbicides/herbicide combinations on GR, GPR, and GS hairy fleabane biotypes.

EXPERIMENT 1 – EFFECT OF GROWTH STAGE OF HAIRY FLEABANE ON THE EFFICACY OF ALTERNATIVE HERBICIDES

Methods and Materials An experiment was conducted at two different periods, fall 2012 and spring 2013, in an open field near the Ornamental Horticulture Unit at California State

University, Fresno, CA (36.816335 N, -119.734500 W).

Plant Material Seeds of GR, GPR, and GS hairy fleabane were obtained from populations originating from various locations of the Central Valley of California (GR (36˚29’15.00”N; 119˚24’10.00” W), GPR (36°35'48.33"N; 119°30'50.45"W), and GS (36°47’58.00 N; 119°57’16 W). These locations are within a 50 km radius in Fresno County, CA. The biotypes were previously verified as GR, GPR, and GS by Moretti et al. (2010). Individual seeds were planted, with the aid of forceps, in seedling trays of 100 separate cells each (52 cm by 25 cm by 6 cm) that had been prefilled with a moist potting mix (Sunshine Mix #31). Plants were placed in a no- hole catchment tray and water was added to the catchment tray for sub-irrigation. The trays were kept in a greenhouse set at 25°/18˚ C day/night temperature with ambient lighting for germination. Once a 2- to 3-leaf seedling was established, plants of the same size were selected for transplanting. Some of the seedlings were transplanted into 5.7 cm by 5.7 cm by 8.25 cm plastic pots and some were transplanted into 8.9 cm by 8.9 cm by 12.7 cm plastic pots containing the same commercial potting mix used for germination. The former pot size was assigned to plants to be sprayed at the 5- to 8-leaf stage and the latter pot size was assigned

1 Sun Gro ®. Horticulture. Sacramento, CA 95814 www.sungrow.com 14 to plants to be sprayed at the rosette or the bolting stage. The seeding and transplanting dates for the targeted growth stages are shown in Table 1. The plants were kept in the greenhouse and watered until they reached the following three growth stages: 5- to 8-leaf stage, rosette stage, and initial bolting.

Table 1. Transplanting and herbicide application dates of the different targeted growth stages of glyphosate-susceptible, glyphosate+paraquat resistant, and glyphosate-resistant hairy fleabane plants in 2012 and 2013. Seeding Transplant Growth Stage Treatment Year Date Date when treated Date 2012 7/2/2012 7/25/2012 Initial Bolting 10/24/2012

7/30/2012 8/25/2012 Rosette 10/24/2012

8/27/2012 9/24/2012 5- to 8 Leaf Stage 10/24/2012

2013 1/14/2013 2/26/2013 Initial Bolting 5/15/2013

2/11/2013 3/18/2013 Rosette 5/15/2013

3/11/2013 4/17/2013 5- to 8 Leaf Stage 5/15/2013

Treatments and Experimental Design The experimental design was a split-split-plot with growth stage (5- to 8- leaf stage, rosette stage, bolting stage) as the main plot; the biotype (GR, GPR, GS) as sub-plot; and the herbicide treatments as the sub-sub-plot. The herbicide treatments were 0 (non-treated control), 0.25x, 0.5x, 1x, and 2x of saflufenacil (Treevix ®2, where x = 70 g ha -1), 1.1 kg ae ha-1 of glyphosate (Roundup WeatherMax® formulation3), and a tank-mixture of 1.1 kg ae ha-1 glyphosate + 70g ha -1 saflufenacil. A surfactant, methylated seed soil (MSO) 1%v/v, was

2 BASF Corporation, 26 Davis Drive, Research Triangle Park, NC 27709 USA 3 Monsanto Company, St Louis, MO 63167, USA 15 added to the saflufenacil treatments. Ammonium sulfate was added to all herbicide applications at the rate of 2% w/v as recommended by the label. Each treatment was replicated five times and the experiment was conducted twice (fall 2012 and spring 2013). Each plant in an individual pot was considered an experimental unit. Plants were moved to an open field outside the greenhouse during herbicide application. Herbicide applications were made with a CO2-pressurized backpack sprayer equipped with a 3-nozzle (flat fan nozzle TeeJet 8002) boom at 30 cm spacing and a spray volume of 374.76 L ha-1. Spray height was maintained at 0.5 m using a PVC frame. The herbicides were applied on the dates shown in Table 1. The maximum and minimum temperatures during the application dates were 19.4/4.4˚C and 30.5/13.3˚C in fall 2012 and spring 2013, respectively. After the herbicide applications, plants were grouped into trays of 15 plants each by herbicide treatment (5 replicates by 3 biotypes) and kept outdoors at the treatment site for 30 days after treatment (DAT). The plants in each tray continued to be sub-irrigated on a weekly basis by filling each catchment tray ¼ full up to 30 DAT. The average daily maximum and the daily minimum temperatures during the 30-day period after herbicide application were 21.1/6.1˚C and 30.0/14.4˚C in fall 2012 and spring 2013, respectively. Plant mortality, based on visual observations of herbicide phytotoxicity for each treatment, was evaluated at weekly intervals up to 30 DAT. The evaluation was based on a scale of 0 to 100% (where alive = 0% and dead=100%). A plant was considered alive if there was any active green leaf tissue at the center of the plant for the seedling and rosette stage or the center or side of the main stem in the initial bolting stage. The plants were harvested at 30 DAT by clipping them at the soil surface and placing them into individual paper bags. The paper bags 16 containing the plants were dried in a forced-air oven at 60˚C for 48h and then their dry weights were recorded.

Statistical Analysis Data for the two experimental runs were subjected to Levene’s test for homogeneity of variance and Shapiro-Wilk’s test for normality to verify if the assumptions of analysis of variance (ANOVA) were met. Data did not pass the normality test and all possible transformations failed to normalize the data. Therefore, analysis was conducted on non-transformed data. Data were analyzed using the general linear model procedures (PROC GLM) of SAS version 9.4. Means were separated using Fisher’s Least Significant Difference (LSD) test when the ANOVA showed significant differences between the treatments at α = 0.05.

Results and Discussion

Plant Mortality and Aboveground Biomass Interactions (P < 0.05) occurred between the year and the treatments for both plant mortality and aboveground biomass. Therefore, data for each experimental year were analyzed separately. Furthermore, within each year, there were interactions (P < 0.05) between growth stage and biotype, growth stage and herbicide treatment, and biotype and herbicide treatment. Therefore, data were analyzed separately for each growth stage as well. In the next section, the effects are discussed for each growth stage of hairy fleabane in each year of the experiment. 17 Effect of the Herbicide Treatments at the 5- to 8-Leaf Stage of Hairy Fleabane The herbicide treatments had differential effects on the mortality of hairy fleabane biotypes in the 2 years of the study. In fall 2012, there was no interaction between biotype and herbicide treatment and all the herbicide treatments controlled all the plants of all three biotypes, with the exception of glyphosate on the GPR biotype (Table 2). However, in spring 2013, there was an interaction between the biotype and herbicide treatment. Saflufenacil applied at 70 ga ha-1 (1x rate) did not provide adequate control of any of the biotypes (Table 2). Saflufenacil at 140 g ha-1 (2x rate) controlled all of the GS and GPR plants and 80% of the GR plants (Table 2). The mixture of saflufenacil and glyphosate controlled 100% of the GS, GPR, and GR plants. Although glyphosate-alone controlled all the GS plants, it only controlled up to 60% of herbicide-resistant (GR and GPR) plants. An interaction (P < 0.05) occurred between the year and herbicide treatment for aboveground plant biomass. However, there was no interaction between the biotype and herbicide treatment in either fall 2012 or spring 2013. Therefore data from biomass for the three biotypes were combined and analyzed. In fall 2012, since all the herbicide-treated plants were killed, the above ground biomass was similar in all the herbicide treatments as they represented dead plants (Table 2). However, in spring 2013, the above ground biomass was different between some of the treatments. The saflufenacil + glyphosate and the glyphosate-alone treatments resulted in the least biomass as many of the plants in these treatments were killed (Table 2). Although lower rates (<140 g ha-1) of safluefenacil did not kill all the plants, the treatments suppressed the growth of the plants as the saflufenacil-treated plants had less biomass than the non-treated control plants.

Table 2. Mortality of glyphosate-susceptible (GS), glyphosate+paraquat resistant (GPR), glyphosate-resistant (GS), and aboveground biomass (mean of all three biotypes) of hairy fleabane plants treated with herbicides at the 5- to 8-leaf stage in Fall 2012 and Spring 2013. Mortality (%)1 Biomass (g plant-1)1 Fall 2012 Spring 2013 Fall 2012 Spring 2013 Biotype Herbicide2 Rate GS GR GPR GS GR GPR Mean

Control - 0 0 0 0 a 0 a 0 a 0.08 a 0.17 a

Saflufenacil 17.5 g ha -1 100 100 100 0 a 0 a 0 a 0 b 0.10 b

Saflufenacil 35 g ha -1 100 100 100 0 a 0 a 0 a 0 b 0.11 b

Saflufenacil 70 g ha -1 100 100 100 20 a 60 bc 20 a 0 b 0.11 b

Saflufenacil 140 g ha -1 100 100 100 100 b 80 cd 100 c 0 b 0.10 b

Saf. + Gly 70 g ha -1 /1.1 kg 100 100 100 100 b 100 d 100 c 0 b 0.06 c ae ha-1

Glyphosate 1.1 kg ac ha-1 100 100 80 100 b 40 b 60 b 0.01b 0.06c 1 Means within a column followed by the same letter are not significantly different at the 0.05 level of significance according to Fisher’s LSD 2 Saf = saflufenacil; Gly = glyphosate. 3 Mean of all three biotypes.

19 Effect of the Herbicide Treatments at the Rosette Stage of Hairy Fleabane: Similar to the 5- to 8-leaf stage, an interaction (P < 0.05) also occurred between the biotype and herbicide treatment for plant mortality at the rosette stage; but unlike the 5- to 8-leaf stage timing, the interaction occurred in both years of the study. Therefore, data were analyzed separately for each biotype separately within each year. In fall 2012, all the plants of the three population types were completely killed by saflufenacil at rates of 35 g ha-1 or higher (Table 3). This result was similar to that for mortality at the 5- to 8-leaf stage. Glyphosate-alone controlled 20% of the GS plants but did not control any of the GR or GPR plants. This result was different than at the 5- to 8-leaf stage timing where glyphosate killed most of the plants including the resistant types but when glyphosate was tank-mixed with saflufenacil it controlled all the plants from all the three population types at this growth stage. The results in spring 2013 were very different than in fall 2012. Unlike fall 2012, none of the saflufenacil-alone treatments were effective against any of the population types in spring 2013 (Table 3). Although the 140 g ha-1 saflufenacil treatment controlled 40% of the GS plants, it provided very little control of the GR and no control of the GPR plants. Glyphosate-alone provided better control of the GS plants than in the fall 2012 study, but it did not provide good control of the GR and GPR types. The control was better when glyphosate was tank-mixed with saflufenacil than when the herbicides were applied alone, but the control of the GR and GPR plants with this treatment was less effective in spring 2013. Herbicide treatments had differential effects on above ground plant biomass in fall 2012, but there was no interaction between the herbicide treatment and the 20 biotype. Therefore, data for above ground biomass were combined for the three biotypes. All the saflufenacil-treated plants, including the mixture with glyphosate, had the least amount of aboveground biomass as all the plants were killed in these treatments and the biomass included dead plants (Table 3). The biomass of the glyphosate-treated plants was greater than that in the saflufenacil treatments but less than that of the non-treated plants. This was because most of the resistant (GR and GPR) plants had survived glyphosate applications and were alive. In spring 2013, unlike in fall 2012, an interaction occurred between the herbicide treatments and biotypes for above ground biomass. Therefore, data were analyzed separately for each biotype. The above ground biomass was, in general, consistent with plant mortality. The least amount of biomass in the GS plants were recorded in the glyphosate and saflufenacil + glyphosate treatments (Table 3). Again most of the saflufenacil-treated GS plants had less biomass than the non-treated control indicating some level of suppression of the plants by this herbicide. Since most of the GR plants, except in the saflufenacil + glyphosate treatment, survived the herbicide treatments, the aboveground biomass was similar in these treatments but lower than that of the non-treated control plants (Table 3). Similar results were observed in the aboveground biomass of the GPR plants. Again, most of the herbicide-treated GPR plants had survived contributing to a greater biomass. However, the biomass was lower than that of the non-treated control thus indicating that the herbicide treatments suppressed the growth of the GPR plants to some extent even though they did not kill them (Table 3).

Table 3. Mortality and aboveground biomass of glyphosate-susceptible (GS), glyphosate+paraquat resistant (GPR), glyphosate-resistant (GS) of hairy fleabane plants treated with herbicides at the rosette stage. Mortality (%)1 Biomass (g/plant) 1 Fall 2012 Spring 2013 Fall 2012 Spring 2013 Biotype Biotype Herbicide2 Rate GS GR GPR GS GR GPR Mean3 GS GR GPR

Control 0 g ha -1 0a 0a 0a 0a 0a 0a 0.55a 0.78a 0.66a 0.87a

Saflufenacil 17.5 g ha -1 60b 100b 100b 0a 0a 0a 0.17c 0.58b 0.54b 0.60bc

Saflufenacil 35 g ha -1 100c 100b 100b 0a 0a 0a 0.16c 0.51bc 0.47bc 0.74ab

Saflufenacil 70 g ha -1 100c 100b 100b 0a 0a 0a 0.18c 0.49bc 0.46bc 0.62bc

Saflufenacil 140 g ha -1 100c 100b 100b 40b 20a 0a 0.19c 0.57b 0.40c 0.56cd

Saf. + Gly. 70 g ha -1 /1.1 100c 100b 100 100c 80b 60b 0.17c 0.35d 0.47bc 0.47d kg ae ha-1

Glyphosate 1.1 kg ac ha-1 20a 0a 0a 100c 0a 40b 0.31b 0.40cd 0.52b 0.50cd 1 Means within a column followed by the same letter are not significantly different at the 0.05 level of significance according to Fisher’s LSD 2 Saf = saflufenacil; Gly = glyphosate. 3 Mean of all three biotypes.

22 Effect of the Herbicide Treatments at the Bolting Stage of Hairy Fleabane An interaction (P < 0.05) between the biotype and herbicide treatment occurred at this growth stage for plant mortality in both years of the study. In fall 2012, saflufenacil rates of 70 g ha-1 or lower did not control any of the plants of any population types (Table 4). The 140 g ha-1 rate of saflufenacil controlled most of the GS and GR plants but was not effective against the GPR plants. Neither glyphosate-alone nor the tank-mix of saflufenacil and glyphosate provided adequate control of any of the populations in fall 2012 at this growth stage. In spring 2013, none of the saflufenacil-alone treatments, including the 140 g ha-1 rate, controlled any of the plants (Table 4). However, when saflufenacil was tank-mixed with glyphosate, it controlled 80% of the GS plants but did not control any of the GR or GPR plants. At this growth stage, consistent with the previous growth stages, glyphosate-alone controlled only the GS plants. The results for aboveground plant biomass at this growth stage was generally similar to that in the rosette stage in that there was no interaction between the herbicide treatment and biotype in fall 2012 and there was an interaction between these two factors in spring 2013. Similar to the data in the rosette stage, aboveground biomass data were combined for the three biotypes in fall 2012 but analyzed separately for each biotype in spring 2013 (Table 4). In fall 2012, when compared to the non-treated control plants, the aboveground biomass was lower in the highest rate of saflufenacil (140 g ha -1), saflufenacil + glyphosate, and glyphosate-alone treatments. This is most likely because these treatments were effective in controlling some of the GS and GR plants (Table 4). However, unlike at the rosette stage, the lower rates of

Table 4. Mortality of glyphosate-susceptible (GS), glyphosate+paraquat resistant (GPR), glyphosate-resistant (GS) in fall 2012 and spring 2013, and aboveground biomass (mean of all three biotypes) in fall 2012 and each biotype in spring 2013 of hairy fleabane plants treated with herbicides at the bolting stage. Mortality(%)1 Biomass (g/plant) 1 Fall Spring Fall Spring Biotype Biotype Herbicide2 Rate GS GR GPR GS GR GPR Mean3 GS GR GPR

Control 0 g ha -1 0a 0a 0a 0a 0a 0a 0.77a 2.18a 3.91a 2.38a

Saflufenacil 17.5 g ha -1 0a 0a 0a 0a 0a 0a 0.70ab 2.04ab 2.43b 1.58c

Saflufenacil 35 g ha -1 0a 0a 0a 0a 0a 0a 0.61bc 2.12a 2.38b 1.69b

Saflufenacil 70 g ha -1 0 0a 0a 0a 0a 0a 0.68ab 2.07a 2.23b 1.62bc

Saflufenacil 140 g ha -1 80c 100c 0a 0a 0a 0a 0.52cd 1.64bc 2.41b 1.38d

Saf. + Gly. 70 g ha -1 /1.1 kg ae 20ab 60b 0a 80b 0a 0a 0.45d 0.90d 1.62c 1.35d ha-1

Glyphosate 1.1 kg ae ha-1 60bc 0a 20a 100c 0a 0a 0.59bcd 1.25cd 2.03bc 1.65bc 1 Means within a column followed by the same letter are not significantly different at the 0.05 level of significance according to Fisher’s LSD 2 Saf = saflufenacil; Gly = glyphosate. 3 Mean of all three biotypes.

24 saflufenacil (<140 g ha -1) did not suppress the plants as the biomass of the treated plants was similar to that of the non-treated control at this growth stage. In spring 2013, similar to that observed at the rosette stage, the GS plants treated with 140 g ha -1 rate of saflufenacil, glyphosate-alone, and saflufenacil + glyphosate had lower aboveground biomass than the other treatments (Table 4). This result again was influenced by plant mortality as mentioned earlier. The GS plants had similar biomass as the non-treated control at saflufenacil rates less than 140 g ha -1, indicating no suppression of this biotype by saflufenacil at this growth stage. In the GR plants, the least biomass, but similar to glyphosate-alone, occurred in the saflufenacil + glyphosate treatment (Table 4). The biomass was similar in the GPR plants, except the plants treated with the 140 g ha -1 rate of saflufenacil had similar biomass as the saflufenacil + glyphosate. Again some suppression of the GPR plants was observed with all the herbicide treatments compared to the non-treated control (Table 4). None of the GR or GPR plants were controlled by these treatments at this growth stage. These results demonstrated that the control of the different hairy fleabane populations with the various herbicide treatments was affected by growth stage and the year (seasonal timing). The level of control with the herbicides for all three population types, in general, diminished when applications were made at the later growth stages. Although good control of all three populations was obtained at the 5- to 8-leaf and at the rosette stages with saflufenacil applied in the fall, the level of control was low and inconsistent when the same treatment was applied in spring. At the bolting stage, the saflufenacil-alone treatments were not effective against the GR or the GPR plants either in fall or spring. There were some morphological differences noticed in the various growth stages between each season that may have been a factor in herbicide efficacy. Plant leaves in the spring 25 25 were much more narrow and upright versus being more broad and prostrate in fall, especially at the rosette stage. These changes were similar among all three biotypes. Other studies have shown similar differential effects with the control of horseweed by glyphosate-alone at different growth stages (Shrestha et al. 2007; VanGessel et al. 2009; Gonzalez-Torralva et al. 2010). All of these studies showed that early developmental stages were much more sensitive to glyphosate than later stages of growth. However, this study showed that growth-stage of fleabane also influenced the efficacy of glyphosate and perhaps even to a greater degree, the efficacy of saflufenacil. Of further interest was the fact that seasonal effects were evident in the control of the hairy fleabane plants with glyphosate-alone. When applied in fall, glyphosate-alone provided good control of all the 5- to 8-leaf stage plants of all the populations. However, the level of control of the GR and GPR plants was 60% or less when applied in spring. Such seasonal effects in the control of GR and GPR hairy fleabane plants by glyphosate-alone were also reported by Moretti et al. (2013). The authors reported that, although the level of control was somewhat similar in fall and spring, the control was very poor in summer. Ge et al. (2011) reported that GR horseweed was sensitive to glyphosate at low temperatures. Although the role of seasonal effects on the control of the Conyza species with glyphosate-alone have been demonstrated by these aforementioned studies, this study further demonstrated that the efficacy of saflufenacil-alone was also affected by seasons. Further, this study showed that the efficacy of a tank-mix of saflufenacil and glyphosate was affected by the growth stage with better control at the rosette stage or earlier. At the bolting stage, although some of the GS plants were 26 26 controlled, this treatment was not effective on the GR and GPR plants. Therefore, hairy fleabane plants of all three biotypes can be effectively controlled by saflufenacil-alone at 70 g ha-1 or a tank-mix of saflufenacil and glyphosate-alone at the 5- to 8-leaf stage. At the rosette stage, saflufenacil-alone or a tank-mix of saflufenacil and glyphosate may provide better control, particularly of the GS plants when applied in the fall than in the spring. At the bolting stage, there was no control or very poor control of the GR and the GPR plants in either season with any of the treatments tested in the study and the control was generally low for the GS plants as well. Therefore, the best strategy would be to apply saflufenacil- alone or in mixture with glyphosate before the rosette stage for effective control of all three biotypes of hairy fleabane in either fall or spring. Delaying treatment beyond this stage will give variable responses based on the growth stage of the plants, the biotype, and the season of application. In regards to plant biomass, with the exception of the GS biotype at the bolting stage during the spring experiment, biomass of all the treated plants was reduced in comparison to the non-treated control plants regardless of whether or not mortality was observed. Also, in the spring experiments, most of the plants were observed to have plant damage after initial spray including yellowing of leaves and necrosis of leaf tissue, typically referred to as herbicide “burn down” effects, but plants recovered and continued to grow and resulted in various levels of recovery of initial losses in biomass when they were collected 30 DAT.

EXPERIMENT 2 – TEMPERATURE

Methods and Materials This experiment was conducted initially in spring 2012 and repeated in fall 2013. The spring 2012 study was conducted in the greenhouse and growth chambers at California State University, Fresno, CA (36.816335 N, -119.734500 W). In fall 2013, the plants were grown in Fresno and the growth chamber portion of the 2013 trials were conducted in Sacramento, CA (38.480956 N, -121.469079 W) due to a malfunction of the growth chambers in Fresno.

Plant Material Seeds of GR, GPR, and GS hairy fleabane were obtained from populations originating from various locations of the Central Valley of California. The GR population was collected from a roadside in Fresno County (36˚29’15.00”N; 119˚24’10.00” W), the GPR population was collected from a roadside in Reedley, CA (36°35'48.33"N; 119°30'50.45"W), and the GS population was collected from a vineyard in Fresno County, CA (36°47’58.00 N; 119°57’16 W). The biotypes were verified as GR, GPR, and GS by Moretti et al. (2010) as described earlier. All these locations are within a 50 km radius in Fresno County, CA. Individual seeds were planted, with the aid of forceps, in seedling trays of 100 separate cells each (52 cm by 25 cm by 6 cm) that had been prefilled with a moist potting mix (Sunshine Mix #31). Plants were placed in a no-hole catchment tray and water was added to the catchment tray for sub-irrigation. In the spring 2012 run and the first run in fall 2013, the trays were kept in a greenhouse set at 25°/18˚ C day/night temperature with ambient lighting for germination. In the last two trials of fall

1 Sun Gro ®. Horticulture. Sacramento, CA 95814 www.sungrow.com 28 28 2013 trays were placed in a growth chamber (Conviron®, Model #MTR 262) at California State University, Fresno, Biology Department Building. The chamber was set to a temperature of 25°/20˚ C with a 12h day/night photoperiod and PAR -2 -1 measured at 800 μmol m s with a Decagon Ceptometer (Decagon Devices, Pullman, Washington). This was done because the temperature and lighting in the greenhouse did not suffice for germination of the seeds. Once a 2- to 3-leaf seedling was established, plants of the same size were selected for transplanting (Table 5). The seedlings were transplanted into 5.7 cm by 5.7 cm by 8.25 cm plastic pots containing the same commercial potting mix used for germination. After transplanting, all the plants were kept in the greenhouse at California State University that was set at 25°/18˚ C day/night temperature with ambient lighting, until they reached 5- to 8-leaf stage

Table 5. Transplanting and herbicide application dates of glyphosate-susceptible, glyphosate-resistant, and glyphosate+paraquat resistant hairy fleabane plants in 2012 and 2013. Planting Transplant Treatment Year Date Date Date 2012 7/30/2012 8/15/2012 9/12/2012

2013 8/15/2013 8/30/2013 10/07/2013

9/19/2013 9/30/2013 11/05/2013

10/20/2013 10/31/2013 12/05/2013

Treatments and Experimental Design The experiment was a split-split plot consisting of three temperature regimes (15/10° C, 25/20° C, and 35/30°C) as the main plot, three biotypes (GS,

2 Conviron ®, Winnepeg Canada, www.conviron.com 29 29 GR, and GPR) as the sub-plot, and four herbicide treatments as the sub-sub plot. Once the plants reached the 5- to 8-leaf stage, the plants were grouped into three trays and randomly assigned to one of the three growth chambers programmed at the aforementioned temperature regimes. The sub-sub plot treatments were non-treated control for each biotype, 70 g ha -1 of saflufenacil (Treevix ®3), 1.1 kg ae ha-1 of glyphosate (Roundup WeatherMax® formulation4), and a tank-mixture of 70 g ha -1 saflufenacil + 1.1 kg ae ha-1 glyphosate. A surfactant, methylated seed soil (MSO) 1%v/v, was added to the saflufenacil treatments. Ammonium sulfate was added to all herbicide applications at the rate of 2% w/v as recommended by the label. In spring 2012 treatments were replicated five times. In fall 2013 treatments were replicated three times. Each individual pot was an experimental unit. The experiment was repeated 4 times (once in spring 2012 and three times in fall 2013). The growth chambers ( Conviron®, Model #MTR 265 in spring 2012) and (Percival®, Model #I-30BLL6 in fall 2013) were programmed to day/night temperatures of 15/10° C, 25/20° C, and 35/30°C, with a 12h day/12h night period. The photosynthetically active radiation (PAR) level in each of the chambers was set at 600 µmol m-2 s-1 which was confirmed with a Ceptometer (Decagon Devices, Pullman, WA). In each growth chamber, plants of the GS, GR, and GPR types were placed and acclimated to the temperature regimes for 72h prior to being treated with the herbicides. In the spring 2012 studies, each of the three temperature regimes included 60 plants (5 sub-samples x 3 biotypes x 4 herbicide treatments) for a total of 180 plants. In

3 BASF Corporation, 26 Davis Drive, Research Triangle Park, NC 27709 USA 4 Monsanto Company, St Louis, MO 63167, USA 5 Conviron ®, Winnepeg Canada, www.conviron.com 6 Percival Scientific, Perry, IA 50220, www.percival-scientific.com 30 30 the fall 2013 studies, due to a smaller chamber size, each of the three temperature regimes included 36 plants each (3 sub-samples x 3 biotypes x 4 herbicide treatments) for a total of 108 plants. After acclimation to the respective temperature regimes, the plants were removed from the chambers, separated into groups for the appropriate herbicide treatment, and sprayed outdoors. Herbicide applications were made with a CO2-pressurized backpack sprayer equipped with a 3-nozzle (flat fan nozzle TeeJet 8002) boom at 30 cm spacing and a spray volume of 374.76 L ha-1. Spray height was maintained at 0.5 m using a PVC frame. The plants were immediately returned to the designated growth chambers after the herbicide applications and kept in the chambers for 7 additional days. The plants were then returned to the greenhouse located at California State University, Fresno, set at 25°/18°C with ambient lighting and observed for an additional 23 days, i.e. 30 days after treatment (DAT). Mortality of the plants was evaluated at 30 DAT by taking visual ratings on a scale of 0 to 100% (where 0% = alive and 100% = dead). A plant was considered alive if there was any active green leaf tissue. The plants were then harvested at 30 DAT by clipping them at the soil surface and placing them into individual paper bags. The paper bags containing the plants were dried in a forced-air oven at 60˚C for 72h and then their dry weights were recorded.

Statistical Analysis Data on plant mortality and aboveground biomass for the 2 years of the study were subjected to Levene’s test for homogeneity of variance and Shapiro- Wilk’s test for normality to verify if the assumptions of analysis of variance (ANOVA) were met. Data that failed to meet the assumptions of ANOVA were log-transformed prior to analysis. Data were then subjected to ANOVA using the general linear model procedures (PROC GLM) of SAS version 9.4. Year and 31 31 replication were considered as random effects, and temperature, biotype, and herbicide treatment were considered as fixed effects. Interactions between the various factors were also tested. The significance level for the analysis was set at α = 0.05. When the ANOVA indicated significant differences at the 0.05 level, the means were separated using Fisher’s Least Significant Difference (LSD) test.

Discussion

Plant Mortality and Aboveground Biomass There was an interaction (P < 0.05) between the year and the main and sub- plot effects for both plant mortality and aboveground biomass; therefore, data for each year was analyzed separately (Table 6). Furthermore, within each year, there was an interaction (P < 0.05) between biotype and herbicide treatment for plant mortality in 2012 and biomass in both 2012 and 2013. For mortality, there was also a three-way interaction between temperature, biotype, and herbicide in both 2012 and 2013 (Table 6). Therefore, data were further analyzed separately for each temperature regime and when the interaction between biotype and herbicide was significant within each temperature, results were separated further by biotype.

Plant Mortality and Above Ground Plant Biomass at 15/10°C At this temperature regime, the effects of the herbicide treatments on plant mortality differed between the 2 years of the study. However, as indicated earlier, there was no interaction between the biotype and herbicide treatment (Table 6). Therefore, the discussion in this section is for plant mortality and aboveground biomass averaged over the three biotypes. In spring 2012, saflufenacil-alone and the mixture of saflufenacil + glyphosate controlled all the plants, whereas glyphosate controlled only about 71% of the plants (Table 7). The

Table 6. Analysis of variance (ANOVA) table showing the main effects and interactions for plant mortality and aboveground plant biomass of the glyphosate-susceptible, glyphosate-resistant, and glyphosate-paraquat resistant populations of hairy fleabane exposed to three different temperature regimes and treated with various herbicides in 2012 and 2013. Plant Mortality (%) Aboveground plant biomass (g plant-1) 2012 2013 2012 2013 Main effects P-value Temperature <0.0001 0.0759 0.0048 0.0223 Biotype 0.0014 1.000 0.6929 0.3827 Herbicide <0.0001 <0.0001 <0.0001 <0.0001 Interactions Temperature x Biotype 0.9448 0.4471 0.4159 0.4789 Temperature x Herbicide <0.0001 0.1971 <0.0001 0.0010 Temperature x Biotype x Herbicide 0.0007 0.0568 0.1424 0.6554

33 salflufenacil-alone and saflufenacil + glyphosate treated plants were killed and had no measurable biomass; however, the glyphosate-alone treated plants had some measurable biomass as some of the plants had survived this application as mentioned earlier. In 2013, all the herbicides, including glyphosate-alone, provided greater than 95% control of the plants and had no measurable biomass left at 30 DAT (Table 7).

Plant Mortality and Above Ground Plant Biomass at 25/20°C Contrary to the results at 15/10°C, an interaction (P < 0.05) occurred between herbicide and biotype for mortality in spring 2012 so the data were separated further by biotype (Table 8). However, there were no interactions between the factors in fall 2013 (Table 6); therefore, the data were averaged for the three population types in 2013. Similar to above ground biomass at the 15/10°C temperature, there was no interaction between the biotype and herbicide treatment in either 2012 or 2013. Therefore data for above ground biomass for the three biotypes were combined and analyzed (Table 8).

Plant Mortality and Above Ground Plant Biomass at 35/30°C Similar to that observed in 25/20°C, an interaction (P < 0.05) occurred between herbicide and biotype for mortality in spring 2012 so the data were separated further by biotype. In fall 2013, there was no interaction for mortality between temperature and herbicide or between herbicide and hence the data were combined for the three biotypes (Table 9). Similar to biomass at the 15/10°C and 25/20°C temperatures, there was no interaction between the biotype and herbicide treatment in either spring 2012 or fall 2013. Therefore biomass data for the three biotypes were combined and analyzed.

Table 7. Plant mortality and aboveground biomass of hairy fleabane plants (averaged for glyphosate-susceptible, glyphosate-resistant, and glyphosate+paraquat resistant biotypes) after treatment and exposure to 15/10˚C temperature. Plant Mortality (%)1 Above ground plant biomass (g plant-1)1 Spring 2012 Fall 20131 Spring 2012 Fall 2013

Herbicide2 Rate Mean3 Control - 0 a 0 a 0.21 c 0.07 b

Saflufenacil 70 g ha -1 100 c 96.30 b 0.00 a 0.00 a

Saf. + Gly. 70 g ha -1 / 1.1 kg ae ha-1 100 c 100 b 0.00 a 0.00 a

Glyphosate 1.1 kg ae ha-1 71.43 b 96.30 b 0.03 b 0.00 a

1 Means within a column followed by the same letter are not significantly different at the 0.05 level of significance according to Fisher’s LSD 2 Saf = saflufenacil; Gly = glyphosate. 3 Mean of all three biotypes.

Table 8. Plant mortality and aboveground biomass of hairy fleabane plants (averaged for glyphosate-susceptible, glyphosate-resistant, and glyphosate+paraquat resistant biotypes) after treatment and exposure to 25/20˚C temperature. Plant Mortality (%)1 Above ground plant biomass (g plant-1)1 Spring 2012 Fall 2013 Spring 2012 Fall 2013

Biotype Herbicide1 Rate GS GR GPR Mean3 Mean3 Mean3

Control - 0 a 0 a 0 a 0 a 0.20 c 0.10 b

Saflufenacil 70 g ha -1 100 b 100 b 100 c 96.30 b 0.01 a 0.00 a

Saf. + Gly.2 70 g ha -1 / 1.1 kg ae ha-1 100 b 100 b 100 c 100 b 0.02 a 0.00 a

Glyphosate 1.1 kg ae ha-1 100 b 25 a 60 b 100 b 0.05 b 0.00 a

In 2012, similar to the results observed at 25/20°C, the saflufenacil + glyphosate treatment controlled all the plants of all three biotypes (Table 9). However, in contrast to the results observed at 25/20°C, neither saflufenacil-alone nor glyphosate-alone provided satisfactory control of any of the biotypes at this temperature regime. Glyphosate-alone controlled only 60% of the GS plants and did not control any of the GR or GPR plants. Saflufenacil-alone did not control any of the GR plants but it controlled some of the GS and GPR plants. However, the results were very different in 2013 as all the herbicides provided 85 to 100% control of all the biotypes (Table 9). The trends in the aboveground biomass data in 2012 were dissimilar to that observed in the 25/20°C temperature regime. Although there was no interaction between the biotype and the herbicide treatment, the aboveground biomass varied between the treatments (Table 9) because of the differences that occurred in plant mortality of the different biotypes. There was virtually no above ground biomass in the saflufenacil + glyphosate treatment because all the plants had died in this treated. The aboveground biomass of glyphosate-alone and saflufenacil-alone followed the trends of plant mortality (Table 9). Similar to the results obtained in the 25/20°C temperature regime, in 2013, since all the herbicide-treated plants had died, there was no measurable biomass in any of the herbicide-treated plants (Table 9). These results demonstrated that control of the different hairy fleabane populations with the various herbicide treatments may be affected by temperature and/or season. Although the results of the spring 2012 study were not replicated in fall 2013, it was hypothesized that results would be similar between the 2 years, as these were in-part growth chamber studies. The exposure to environmental conditions between the time of removing the plants from the cabinet to 30 DAT

Biotype Herbicide1 Rate GS GR GPR Mean3 Mean3 Mean3

Control - 0 a 0 a 0 a 0 a 0.17 d 0.06 b Saflufenacil -1 70 g ha 20 ab 0 a 40 b 92.59 c 0.11 c 0.00 a Saf. + Gly.2 70 g ha -1 / 1.1 kg ae ha-1 100 c 100 b 100 c 100 c 0.00 a 0.00 a

Glyphosate 1.1 kg ae ha-1 60 b 0 a 0 a 85.19 b 0.07 b 0.01 a 1 Means within a column followed by the same letter are not significantly different at the 0.05 level of significance according to Fisher’s LSD 2 Saf = saflufenacil; Gly = glyphosate. 3 Mean of all three biotypes.

38 may possibly have had an impact on the results. Therefore, it is important to discuss some of the circumstances that may have influenced these results. The photoperiod was different during the spring and the fall studies in 2012 and 2013, respectively. Other studies have also suggested seasonal affects with herbicide resistance in Conyza spp (Moretti 2013), but have not suggested what exactly the factors of seasonal influences are. This study strongly suggested that photoperiod and light levels may be an important factor for the variability observed between seasons as temperature was kept constant in the greenhouse. As discussed earlier, such variability between the seasons was also observed in the growth stage experiment. Therefore, the photoperiod may have had an influence on the physiological development of the hairy fleabane plants during seedling growth. Whether this affected the results after the plants were removed from the chambers and moved to the greenhouse is a possible answer, but would need to be explored further in another study. The seasonal differences in control measures may also clear up the discrepancy in the control of horseweed between studies of Waggoner et al. (2011) showing increased benefit of using saflufenacil mixed with glyphosate; and no differences between saflufenacil alone and the mixture of saflufenacil and glyphosate shown by Shrestha and Moretti (2011). The possibility that temperature makes a difference in control of hairy fleabane supports both determinations (saflufenacil-alone or tank-mix with glyphosate) depending on the environmental factors that were taking place when the study was performed. It is important to note that although results from spring and fall trials in the 2 years differed in this experiment, there were consistencies in plant behavior, mortality and treatment efficacy. Changes in leaf position were noted after plants were acclimated in chambers prior to treatment during both seasons. Leaves were 39 more upright after exposure to 35/30ºC and more prostrate after exposure to 15/10ºC with 25/20 having little to no movement. The mixture of saflufenacil + glyphosate consistently controlled 100% of the plants of all 3 biotypes in both spring and fall trials and under all three temperature regimes. The efficacy of saflufenacil-alone and glyphosate-alone was reduced in spring at the 35/30˚C regime, although their efficacy was better at the lower temperature regimes. In the fall study, the temperature regimes did not make a difference as these treatments provided good control of all three hairy fleabane types. Therefore, these results suggest that any of the herbicide treatments used in this study can be effective on any of the hairy fleabane plants of all three biotypes. The 5-8 leaf stage (as was the stage used in this study), can be effectively controlled by saflufenacil alone and glyphosate-alone may provide good control of all the hairy fleabane biotypes if applications are made on 5- to 8-leaf plants at a day/night temperature range of 15˚C/10˚C - 25˚C/20˚C. At higher temperatures, a tank-mix of saflufenacil + glyphosate may provide more consistent control of all three fleabane population types.

EXPERIMENT 3 – LIGHT INTENSITY

Methods and Materials An experiment was conducted in spring and fall of 2013 and spring 2014 in an open field near the Ornamental Horticulture Unit at California State University,

Fresno, CA (36.816335 N, -119.734500 W).

Plant Material Seeds of GR, GPR, and GS hairy fleabane were obtained from populations originating from various locations of the Central Valley of California (GR (36˚29’15.00”N; 119˚24’10.00” W), GPR (36°35'48.33"N; 119°30'50.45"W), and GS (36°47’58.00 N; 119°57’16 W). As mentioned earlier, these locations are within a 50 km radius in Fresno County. The biotypes were verified as GR, GPR, and GS by Moretti et al. (2010). Individual seeds were planted, with the aid of forceps, in seedling trays of 100 separate cells each (52 cm by 25 cm by 6 cm) that had been prefilled with a moist potting mix (Sunshine Mix #31). Plants were placed in a no-hole catchment tray and water was added to the catchment tray for sub-irrigation. The trays were kept in a greenhouse set at 25°/18˚ C day/night temperature with ambient lighting for germination. Once a 2- to 3-leaf seedling was established, plants of the same size were selected for transplanting. The 40

seedlings were transplanted into 5.7 cm by 5.7 cm by 8.25 cm plastic pots. The seeding and transplanting dates are shown in Table 10. The plants were kept in the greenhouse and watered until they reached the 5- to 8-leaf stage.

1 Sun Gro ®. Horticulture. Sacramento, CA 95814 www.sungrow.com 41 Table 10. Transplanting and herbicide application dates of glyphosate-susceptible, glyphosate-resistant, and glyphosate+paraquat resistant hairy fleabane plants in 2013 and 2014. Planting Transplant Treatment Year Date Date Date 2013 3/11/2013 4/17/2013 5/15/2013 8/15/2013 8/30/2013 9/22/2013 2014 4/6/2014

Treatment and Experimental Design The experiment was a split-split plot design consisting of four light gradients (100% sun, 50% sun, 30% sun, and 0% sun - complete darkness, 50% shade, 70% shade, and 100%-full sun) in the main plot, three biotypes (GS, GR, and GPR) in the sub-plot, and five herbicide treatments in the sub-sub plot. When the plants reached the 5- to 8- leaf stage, they were grouped into trays and randomly assigned to one of the four light intensities. The light intensities were created by using shade-cloth applied over 30 cm tall hoop house type structures made with PVC frames. The herbicide treatments were 0 (non-treated control for each biotype), 70 g ha -1 of saflufenacil (Treevix ®2), 1.1 kg ae ha-1 of glyphosate (Roundup WeatherMax® formulation3), 6 liters ha-1 glufosinate (Rely®4) or 0.3 L ha-1 pyraflufen (Venue®5). The saflufenacil treatments were supplemented with methylated seed oil (MSO) surfactant at a rate of 1%v/v. Ammonium sulfate was added at 2% w/v ammonium sulfate for saflufenacil and glyphosate treatments, and 1% w/v for glufosinate treatments per label recommendations. Pyraflufen treatments were supplemented with 1% v/v crop oil concentrate (COC).

2 BASF Corporation, 26 Davis Drive, Research Triangle Park, NC 27709 USA 3 Monsanto Company, St Louis, MO 63167, USA 4 Bayer Crop Science, Research Triangle Park, NC 5 Nichino America Inc.Wilmington, Delaware 19808 42

Treatments were replicated four times and the study was repeated three times (spring 2013, fall 2013 and spring 2014). Herbicide applications were made with a CO2-pressurized backpack sprayer equipped with a 3-nozzle (flat fan nozzle TeeJet 8002) boom at 30 cm spacing and a spray volume of 374.76 L ha-1. Spray height was maintained at 0.5 m using a PVC frame. Immediately after treatment, plants were exposed for 48h to either of the four different light intensities in an open field setting. The plants were then returned to the greenhouse for survival evaluation for an additional 28 days after treatment (DAT). The light intensity under the shade cloth was verified with a ceptometer15. Mortality of the plants was evaluated at 30 DAT by taking visual ratings on a scale of 0 to 100% (where 0% = alive and 100% = dead). A plant was considered alive if there was any active green leaf tissue at the center of the plant that appeared to be growing. The plants were then harvested at 30 DAT by clipping them at the soil surface and placing them into individual paper bags. The paper bags containing the plants were dried in a forced-air oven at 60˚C for 72h and then the above ground biomass (dry weight) was recorded.

Statistical Analysis Data on plant mortality and aboveground biomass for the 2 years of the study were subjected to Levene’s test for homogeneity of variance and Shapiro- Wilk’s test for normality to verify if the assumptions of analysis of variance (ANOVA) were met. Data that failed to meet the assumptions of ANOVA were log-transformed prior to analysis. Data were then subjected to ANOVA using the general linear model procedures (PROC GLM) of SAS version 9.4. Year and blocks were considered as random effects, and temperature, biotype, and herbicide treatment were considered as fixed effects. Interactions between the various 43 factors were also tested. The significance level for the analysis was set at α = 0.05. When the ANOVA indicated significant differences at the 0.05 level, the means were separated using Fisher’s Least Significant Difference (LSD) test.

Results and Discussion Similar to the growth stage and temperature study, there was an interaction (P < 0.05) between the season and the treatments. However, there were no interactions between the year and treatments in the studies conducted in spring 2013 and spring 2014. Therefore, the data for spring 2013 and 2014 were combined and presented as spring data. Data were then analyzed separately for the spring and fall data due to the interaction mentioned above. Furthermore, there was a light by biotype by herbicide treatment interaction in the spring study (Table 11) and this will be discussed in the following section.

Plant Mortality and Above Ground Plant Biomass in Spring 2013/2014 In the spring trials, light had no effect on the mortality and aboveground biomass of the hairy fleabane plants, but biotype and herbicide treatment had an effect on both mortality and above ground biomass (Table 11). Although there were no interactions between light and biotype, light and herbicide, and biotype and herbicide; there was a three-way interaction between light, biotype, and herbicide for plant mortality. Due to this interaction, data were analyzed separately for each light regime and results for plant mortality are reported by biotype and herbicide treatment within each light regime (Table 12). However, such an interaction was not observed for aboveground plant biomass; therefore, data for this parameter were analyzed for each light regime (Table 13).

Table 11. Analysis of variance (ANOVA) table showing the main effects and interactions for plant mortality and aboveground plant biomass in spring (average of 2013 and 2014) and fall 2013 of glyphosate-susceptible, glyphosate- resistant, and glyphosate-paraquat resistant populations of hairy fleabane treated with various herbicides and exposed to four different light intensities. Plant Mortality (%) Aboveground plant biomass (g plant-1) Spring 2013/14 Fall 2013 Spring 2013/14 Fall 2013 Main effects P-value Light 0.2603 0.0040 0.5081 0.0168 Biotype <0.0001 0.3154 0.0015 0.0028 Herbicide <0.0001 <0.0001 <0.0001 <0.0001 Interactions Light x Biotype 0.3455 0.7365 0.9670 0.9352 Light x Herbicide 0.1920 0.0672 0.1431 0.0013 Light x Biotype x Herbicide <0.0001 0.9870 0.0978 0.1557

Table 12. Mortality of glyphosate-susceptible (GS), glyphosate+paraquat resistant (GPR), glyphosate-resistant (GS), of hairy fleabane plants treated with herbicides at the 5- to 8-leaf stage in spring (average of 2013/14) and exposed to different light regimes. Light 100% Sun Light 50% Sun Light 30% Sun Light 0% Sun

Biotype

Herbicide Treatment GS GR GPR GS GR GPR GS GR GPR GS GR GPR

Plant Mortality (%)1 Control 0a 0a 0a 0a 0a 0a 0a 0a 0a 0a 0a 0a

Saflufenacil 0a 37.5b 12.5 a 12.5 a 0a 0a 25b 25b 12.5a 12.5a 0a 25b

Pyraflufen 0a 0a 0a 0a 0a 0a 0a 0a 0a 0a 0a 0a

Glufosinate 100 b 100 c 100 b 87.5b 100 b 100 b 87.5 c 100 c 75b 100 b 100 b 100 c

Glyphosate 100 b 0a 12.5 a 100b 0a 0a 100 c 0a 0a 100 b 0a 0a

1 Means within a column followed by the same letter are not significantly different at the 0.05 level of significance according to Fisher’s LSD

Results showed that glufosinate consistently provided the best level of control of all three hairy fleabane biotypes at all the light regimes tested (Table 12). Although the plant mortality levels with glufosinate was reduced to about 87.5% in the 30% and 50% light treatments, glufosinate remained the most effective treatment for all the hairy fleabane population types. Mortality of the saflufenacil-treated plants of all three biotypes was very low in all of the light regimes with the best control being only 37.5% of the GR population in full sun (no shade). Pyraflufen, another PPO inhibitor was completely ineffective at controlling any of the hairy fleabane biotypes in the spring. Glyphosate provided excellent control of the GS biotype with 100% mortality under every light regime, but provided little to no control of the GR or GPR hairy fleabane plants. The aboveground biomass results reflected the plant mortality trends (Table 13). As mentioned earlier, there was no effect of light or biotype on the aboveground biomass of the hairy fleabane plants. The only effect observed for this parameter was the herbicide type. The glufosinate-treated plants had the least aboveground biomass because most of the plants in this treatment were killed. A reduction in biomass for the glyphosate-treated plants, compared to the non- treated plants, was observed which was again mainly due to the mortality of the GS plants. Some reduction in the average aboveground biomass was observed in the saflufenacil-treated plants, which again was due to mortality of some of the plants in the various light regimes as discussed earlier. Pyraflufen-treated plants resulted in similar average aboveground biomass as the non-treated control plants as there was no plant mortality observed with this treatment.

47 Table 13. Above ground biomass of hairy fleabane plants (averaged for the light regimes and three biotypes) treated with herbicides at the 5- to 8-leaf stage and exposed to different light regimes in spring 2013/14. Herbicide Treatment Average above ground biomass1 g plant -1 Control 0.29 a

Saflufenacil 0.13 c

Pyraflufen 0.26 a

Glufosinate 0.06 d

Glyphosate 0.16 b 1 Means within a column followed by the same letter are not significantly different at the 0.05 level of significance according to Fisher’s LSD

Plant Mortality and Above Ground Plant Biomass in Fall 2013 Contrary to the results in spring, the light regime had an effect on plant mortality as the herbicide treatment did in the fall (Table 11). However, there were no interactions between the any of the factors and the three-way interaction was not significant either. Therefore, the data were combined for the three biotypes and the results of each herbicide treatment under each light regime are presented in Table 14 to show the main effect of light and herbicide treatment. This was done because the light by herbicide treatment was significant at P = 0.0672 and the same interaction was highly significant (P = 0.0013) for aboveground plant biomass. Unlike the spring trials, saflufenacil performed better during the fall and controlled 100% of the plants of all biotypes at all light levels. Similar to spring treatment, glufosinate provided excellent control of all the hairy fleabane plants in the fall, except that plant mortality was reduced to 91.67% in the 50% light regime which could be more due to experimental error 91.67%. Consistent with the growth stage and the temperature studies, glyphosate provide 48 excellent control of all the hairy fleabane biotypes with 100% control at the 0, 30 and 50% light regime and up to 75% at the full (100%) light regime. Pyraflufen, which did not show any mortality of any of the hairy fleabane plants during the spring, showed differential levels of control at the various light intensities with up to 16.7%, 25%, 8.33%, and 50% at the 100, 50, 30, and 0 light regimes, respectively.

Table 14. Mortality of hairy fleabane plants treated with various herbicides and exposed to various light regimes at the 5- to 8-leaf stage in fall 2013. Herbicide Light 100% Light 50% Light 30% Sun Light 0% Sun Treatment Sun Sun

Plant Mortality (%)1 Control 0a 0a 0a 25a

Saflufenacil 100d 100d 100b 100c

Pyraflufen 16.67b 25b 8.33a 50b

Glufosinate 100d 91.67c 100b 100c

Glyphosate 75c 100d 100b 100c 1 Means within a column followed by the same letter are not significantly different at the 0.05 level of significance according to Fisher’s LSD

The results for aboveground plant biomass was pooled for the biotypes but analyzed separately for the herbicide treatments under each light regime because of the light by herbicide interaction (Table 11). Biomass results reflected the levels of mortality observed in the spring and fall studies (Table 15). The aboveground biomass of the glufosinate- and glyphosate-treated plants was negligible in all the light regimes. Contrary to spring, in the fall, the aboveground biomass of the saflufenacil-treated plants was also negligible (Table 15) because of increased plant mortality. The pyraflufen-treated plants also had less 49 aboveground biomass than the non-treated control plants (Table 15), mainly due to the mortality of some of the plants treated with pyraflufen (Table 14).

Table 15. Biomass of hairy fleabane plants treated with herbicides at the 5- to 8- leaf stage in fall 2013. Herbicide Light 100% Light 50% Light 30% Light 0% Sun Treatment Sun Sun Sun

Above ground biomass (g plant-1)1 Control 0.107a 0.075 a 0.080 a 0.052 a

Saflufenacil 0 c 0 c 0 c 0 c

Pyraflufen 0.041b 0.046 b 0.058 b 0.032 b

Glufosinate 0 c 0.001 c 0 c 0.008 c

Glyphosate 0.006 c 0.002 c 0.006 c 0 c 1 Means within a column followed by the same letter are not significantly different at the 0.05 level of significance according to Fisher’s LSD

Although differences in performance of the PPO herbicides under different light regimes were expected, as in the growth stage and the temperature study, the results were overshadowed by the huge effect of seasons (primarily spring vs fall). The results with saflufenacil was also unexpected as contrary to the temperature study it did not provide good control of the hairy fleabane plants in spring but provided better control in the fall, regardless of the light intensity. Although the performance was poor in the spring trials, saflufenacil was most efficacious at the 100% (full sun) intensity, which was in line with the hypothesis regarding the expected PPO inhibitor mode of action. The slight increase in mortality of saflufenacil at the 30% sun level during the spring with the GS and GR biotype as well as the 0% sun level with the GPR biotype show that there may be some physiological mechanisms occurring with absorption or translocation when the 50 herbicide is applied at lower light levels, similar to what was reported by Mellendorf et al. (2015). But contrary to Jansen’s (1989) observations with paraquat to the resistant biotypes, there was more impact on the GR biotype after treatment with saflufenacil and exposure to high light intensity but the impact on the GPR biotype was weak. Pyraflufen, the other PPO inhibitor used in this study is only recommended for us on hairy fleabane when tank mixed with glyphosate. Although it showed some level of control in the fall study, it may not be useful to apply it alone for control of hairy fleabane. The best control by pyraflufen (up to 50% mortality) in the fall was achieved in the 0% sun (full darkness) light regime which was consistent with the hypothesis that PPO inhibitors would be more efficacious when applied under 100% light (full sun). .

CONCLUSION

This study showed that environmental factors can have an impact on the control of GS, GR, and GPR hairy fleabane plants with postemergence herbicides. The efficacy of the postemergence herbicides on hairy fleabane plants was dominated by the time of the year (spring vs. fall) they were applied, particularly due to differences in temperature and probably photoperiod. Although light intensity had some influence on the mortality of the hairy fleabane plants with the different postemergence herbicides, the effect was more pronounced in the fall than in the spring. In each of the different studies (growth stage, temperature, and light), plant mortality with the different postemergence herbicides was generally greater in the fall than in the spring. Among the herbicides, glufosinate provided the best control of all hairy fleabane plants under all the environments and growth stages tested, except at the bolting stage. An interesting finding throughout the study was the seasonal influence on each experiment. In each experiment, there was in increase in mortality in the fall compared to spring. Spring trials demonstrated that hairy fleabane can be very difficult to control during these months with saflufenacil-alone and glyphosate-alone, even at the seedling stage. This phenomenon is somewhat supported in literature. Moretti, et al. (2013) showed decreased mortality in Conyza sp during the warmer summer months when compared to similar herbicide (glyphosate) rates in fall/winter trials. Rubin, et al. (2011) observed that plants grown at higher temperatures show more tolerance to glyphosate versus plants grown in lower temperatures. Sharkhuu et al. (2014) reported that a glyphosate-resistant Aradopsis thaliana phenotype was caused by a dysfunctional phytochrome B, which is a red and far-red light receptor. 52 52

Although the discovery of mechanisms of resistance was not the goal of this study, some speculations can be made on the method of resistance based on the differences in plant mortality, with saflufenacil-alone and glyphosate-alone, between spring and fall in all three experiments. The first speculation is that the effects may be related to the phytochrome system within the plant. Phytochromes are photoreceptors synthesized in an inactive form within the plant and then activated by light and converted to the biologically active form, allowing the initiation of germination, growth and flowering responses (Chory et al. 1996). According to Jiao et al. (2007) the conversion of phytochrome to the active form allows them to translocate from the cytoplasm to the nucleus and communicate with several genetic regulators downstream of the photoreceptors. Quality of light has been known to induce or repress gene transcription in response to stimuli and the integration of temperature and light both play a part in the programming of plant growth and other genetic changes (Jiao et al. 2007). In the Central Valley, light intensity and temperature vary widely between the spring and fall seasons. Throughout this study, there are indications that seasonal differences are causing an “activation” of resistance mechanisms during the spring. The strongest indicator of this is the lack of difference in behavior of GS hairy fleabane biotype compared to its herbicide-resistant counterparts. The GS population’s response to glyphosate remained the same in both seasons. During the fall, glyphosate-alone was lethal to GS, GR and GPR at the 5- to 8-leaf stage of hairy fleabane. During the spring, glyphosate-alone was still lethal to the GS biotype, but not to the GR and GPR biotypes. Glyphosate reduced the biomass of these plants but had very low mortality. Glyphosate-alone also controlled 100% of the GS biotype when applied at the rosette and bolting stage of hairy fleabane in the spring studies. Therefore, the GS biotype does not seem to have the same response that the GR 53 53 and GPR biotypes have that confers resistance to glyphosate in the spring. Furthermore, it can be speculated that these mechanisms may also provide some level of tolerance to saflufenacil during the spring. Reduced light intensity and temperature in fall and winter months compared to spring and summer can also cause morphological differences within the plant. Hairy fleabane can be highly plastic depending on environmental conditions. Differences were visible in leaf orientations during the spring and fall. The leaves were more upright with a thicker base and reduced leaf surface area in the spring, whereas in the fall, the leaves were broader and more prostrate allowing for a greater leaf surface area and little biomass at the base. This difference can allow variability in plant coverage with foliar herbicides. These observations were noticed in all three different biotypes and should be considered as a factor for variability in herbicide efficacy in spring versus fall months. Although there is a need for further exploration of mechanisms of resistance within hairy fleabane plants, several important conclusions can be made from this study. These include that saflufenacil-alone was more effective in the fall than in spring. All the GS, GR and GPR plants were controlled by saflufenacil-alone at the 5- to 8-leaf and rosette stage but level of control declined at the bolting stage. Better control was obtained at the 15/10ºC and 25/20ºC than at the 35/30ºC temperature regime. During the fall, the different light regimes had no effect on the efficacy of saflufenacil. In the spring, the efficacy of saflufenacil- alone was inconsistent and mortality was variable between the biotypes. Therefore, saflufenacil-alone can provide excellent control of GS, GR and GPR hairy fleabane plants prior to the bolting stage in the fall. In the spring, saflufenacil will be more effective when applied as a tank-mixture with glyphosate and applied prior to the rosette stage.

REFERENCES

Andersen MC (1993) Diaspore morphology and seed dispersal in several wind- dispersed Asteraceae. Am. J. Bot. 80:487-492

Ashigh J, Hall JC (2010) Bases for interactions between saflufenacil and glyphosate in plants. J Agric Food Chem 58:7335-7343

Baker HG (1974) The Evolution of Weeds. Annual Review of Ecology and Systematics 5:1-24.

Chory, J, Chatterjee M, Cook RK, Elieh T, Fankhauser C, Li J (1996) From Seed Germination to Flowering, Light Controls Plant Development via the Pigment Phytochrome. Pages 12066-12071 in Proceedings of the National Academy of Sciences of the United States of America: National Academy of Sciences

Cronquist A (1980) Vascular flora of the southeastern United States, Vol I Asteraceae. Chapel Hill, North Carolina, USA: University of North Carolina Press

Dauer JT, Mortensen DA, Luschei EC, Isard SA, Shields E. Van-Gessel MJ (2009) Conyza canadensis seed ascent in the lower atmosphere. Agric For Meteorol 149:526-534

Duke SO (2012) Why have no new herbicide modes of action appeared in recent years? Pest Manag Sci 68:505-512

Eubank TW, Nandula VK, Reddy KN, Poston DH, Shaw DR (2013) Saflufenacil efficacy on horseweed and its interaction with glyphosate. Weed Bio Manag 13: 135–143

Falk JS, Al-Khatib K, Peterson D.E (2006) Rapid assay evaluation of plant response to Protoporphyrinogen Oxidase (Protox)-inhibiting herbicides. Weed Technol 20: 104-112

Ge X, d'Avignon DA, Ackerman JJ, Duncan B, Spaur MB, Sammons RD (2011) Glyphosate-resistant horseweed made sensitive to glyphosate: low- temperature suppression of glyphosate vacuolar sequestration revealed by 31P NMR. Pest Manag Sci 67: 1215-1221 55 55 González-Torralva F, Cruz-Hipolito H, Bastida F, Mülleder N, Smeda RJ, De Prado R (2010) Differential susceptibility to glyphosate among the Conyza weed species in Spain. J Agric Food Chem 58: 4361-4366

Heap IM (2011) Top 10 worst herbicide resistant weeds globally. Abstract #319 Weed Science Society of America 51st Annual Meeting. Portland, OR: Weed Science Society of America

Heap IM (2015) International Survey of Herbicide Weeds. Herbicide Resistance Action Committee and Weed Science Society of America. www.weedscience.com. Accessed Monday, February 16, 2015

Hess DF (2000) Light-dependent herbicides: an overview. Weed Sci 48:160-170

Holt JS (1988) Reduced growth, competitiveness, and phyotosynthetic efficiency of triazine-resistant Senecio vulgaris from California. J Appl Ecol: 307-318

Horowitz J, Ebel R, Ueda K (2010) No-Till Farming Is a Growing Practice. United States Department of Agriculture, Economic Research Service. Economic Information Bulletin Number 70

Jansen MAK, Yoseph S, Kazzes D, Canaani O, Malkin S, Gressel J (1989) Increased tolerance to photoinhibitory light in paraquat-resistant Conyza bonariensis measured by photoacoustic spectroscopy and 14CO2-fixation. Plant Physiol 91:1174-1178

JiaoY, Lau OS, Deng XW (2007) Light-regulated transcriptional networks in higher plants. Nature reviews. Genetics 8: 217-230

Liebel RA, Westberg DE, Bowe SJ (2009) Saflufenacil: Discovery and mode of action of a new broadleaf herbicide. Page 177 in North Central Weed Science Society Proceedings Volume 64. Kansas City, MO: North Central Weed Science Society

Madden NM, Southard RJ, Mitchell JP (2008) Conservation Tillage Reduces PM. Atmos Environ 16: 3795-3808

Matringe M, Camadro JM, Labbe P, Scalla R (1989) Protoporphyrinogen oxidase as a molecular target for diphenyl ether herbicides. Biochem J 260: 231-235

Mellendorf, TG, Young JM, Matthews JL, Young BG (2015) Influence of application variables on the foliar efficacy of saflufenacil on horseweed. Weed Sci In-Press 56 56 Michael PW (1977) Some weedy species of Amaranthus (amaranths) and Conyza/Erigeron (fleabanes) naturalized in the Asian-Pacific region. Pages 87-95 in Proceedings of the 6th Asian-Pacific Weed Science Society Conference, Volume 1. Indonesia: Asian-Pacific Weed Science Society

Moretti ML, Hanson BD, Hembree KJ, Shrestha A (2010) Multiple-resistant biotypes of hairy fleabane (Conyza bonariensis) documented in the San Joaquin Valley. Abstract #O-236 Weed Science Society of America 50th Annual Meeting. Denver, CO: Weed Science Society of America

Moretti M.L., Bradley D. Hanson, Kurt J. Hembree and Anil Shrestha. 2013. Glyphosate Resistance is more variable than paraquat resistance in a multiple-resistant hairy fleabane (Conyza bonariensis) population. Weed Sci 61(3): 396-402

Mortensen DA, Humston R (2006) Controlled experiments to predict horseweed (Conyza canadensis) dispersal distances. Weed Sci 54: 484-489

Mortensen DA (2007) Temporal and spatial dynamics of long-distance Conyza canadensis seed dispersal. J of Appl Ecol 44: 105-114

Murphy CE, Lemerle D (2006) Continuous cropping systems and weed selection. Euphytica 148: 61-73

Nandula VK, Eubank TW, Poston DH, Koger CH, Reddy KN (2006) Factors affecting germination of horseweed (Conyza canadensis). Weed Sci 54:898- 902

Nesom GL (1990) Further definition of Conyza (Asteraceae:Astereae). Phytologia 68: 229-233

Owen LN, Mueller TC, Main CL, Jason B, Steckel LE (2011) Evaluating rates and application timings of saflufenacil for control of glyphosate-resistant horseweed (Conyza canadenis) prior to planting no-till cotton. Weed Technol 25: 1-5

Patzoldt WL, Hager AG, McCormick JS, Tranel PJ (2006) A codon deletion confers resistance to herbicides inhibiting protoporphyrinogen oxidase. PNAS 103:12329-12334

Prather TS, Ditomaso JM, Holt JS (2000) Herbicide Resistance: Definition and management strategies. University of California Division of Agriculture and Natural Resources: Publication # 8012 57 57 Rubin B, Kleinman Z, Ben-Ami G, Sibony M (2011). Response of Conyza spp. to glyphosate is temperature and plant age dependent. Abstract #364. Weed Science Society of America 51st Annual Meeting. Portland, OR: Weed Science Society of America

Sanyal D, Bhowmik PC, Anderson RL, Shrestha A (2008) Revisiting the perspective and progress of Integrated Weed Management. Weed Sci 56: 161-167

Sharkhuu A, Narasimhan ML, Merzaban JS, Bressan RA, Weller S, Gehring C (2014) A red and far-red light receptor mutation confers resistance to the herbicide glyphosate. Plant J 78:916-926

Sherman TD, Becerril JM, Matsumoto H, Duke MV, Jacobs JM, Jacobs NJ, Duke SO (1991) Physiological basis for differential sensitivities of plant species to protoporphyrinogen oxidase inhibiting herbicides. Plant Physiol 97: 280-287

Shrestha A, Hanson BD, Hembree KJ (2008) Glyphosate-resistant hairy fleabane documented in the Central Valley. Calif Agric 62: 116-119

Shrestha A, Moretti ML (2011) Effect of saflufenacil on glyphosate-resistant and susceptible horseweed (Conyza canadensis) biotypes. Abstract #104. Weed Science Society of America 51st Annual Meeting. Portland, OR: Weed Science Society of America

Shrestha A, Steinhauer K, Moretti M, Hanson B, Jasieniuk M, Hembree K, Wright S (2014) Distribution of Glyphosate-resistant and Glyphosate-susceptible Hairy Fleabane (Conyza Bonariensis) in Central California and Their Phenological Development. J of Pest Sci 87.1: 201-209

Shrestha A, Swanton CJ (2007) Parameterization of the phenological development of select annual weeds under noncropped field conditions. Weed Sci 55: 446- 454

Shields EJ, Dauer JT, VanGessel MJ, Neumann G (2006) Horseweed (Conyza canadensis) seed collected in the planetary boundary layer. Weed Sci 54: 1063-1067

Swanton CJ, Mahoney KJ, Chandler K, Gulden RH (2008) Integrated Weed Management: Knowledge-based weed management systems. Weed Sci 56: 168-172

VanGessel MJ (2001) Glyphosate resistant horseweed from Delaware. Weed Sci 49: 703-705 58 58 VanGessel MJ, Scott BA, Johnson QR, White-Hansen SE (2009) Influence of glyphosate-resistant horseweed (Conyza canadensis) growth stage on response to glyphosate applications. Weed Sci 23: 49-53

Van Kleunen M, Weber E, Fischer M (2010) A meta-analysis of trait differences between invasive and non-invasive plant species. Ecol Lett 13: 235-245

Vaughn KC, Vaughn MA, Camilleri P (1989) Lack of cross resistance of paraquat resistant hairy fleabane (Conyza bonariensis) to other toxic oxygen generators indicates enzymatic protection is not the resistance mechanism. Weed Sci 37: 5-11

Vencill WK, ed (2002) Herbicide Handbook. 8th ed. Lawrence, KS: Weed Science Society of America.

Waggoner BS, Mueller TC, Jason AB, Steckel LE (2011) Control of Glyphosate- resistant horseweed (Conyza canadensis) with saflufenacil tank mixtures in no-till cotton. Weed Technol 25: 310-315

Warnert J (2012) Conservation tillage achieves record acreage, yields. Calif Agric 66(2):54

Widderick M, Walker S, Cook T (2012) Flaxleaf fleabane (Conyza bonariensis) - Strategic solutions using best management practices. J. Weed Sci Res 18 (Special Issue): 687-93

Wu H, Walker S, Rollin MJ, Tan DKY, Robinson G, Werth J (2007) Germination, persistence, and emergence of flaxleaf fleabane (Conyza bonariensis [L.] Cronquist). Weed Bio Manag 7(3):192-99.

Fresno State

Non-Exclusive Distribution License (to archive your thesis/dissertation electronically via the library’s eCollections database)

By submitting this license, you (the author or copyright holder) grant to Fresno State Digital Scholar the non-exclusive right to reproduce, translate (as defined in the next paragraph), and/or distribute your submission (including the abstract) worldwide in print and electronic format and in any medium, including but not limited to audio or video.

You agree that Fresno State may, without changing the content, translate the submission to any medium or format for the purpose of preservation.

You also agree that the submission is your original work, and that you have the right to grant the rights contained in this license. You also represent that your submission does not, to the best of your knowledge, infringe upon anyone’s copyright.

If the submission reproduces material for which you do not hold copyright and that would not be considered fair use outside the copyright law, you represent that you have obtained the unrestricted permission of the copyright owner to grant Fresno State the rights required by this license, and that such third-party material is clearly identified and acknowledged within the text or content of the submission.

If the submission is based upon work that has been sponsored or supported by an agency or organization other than Fresno State, you represent that you have fulfilled any right of review or other obligations required by such contract or agreement.

Fresno State will clearly identify your name as the author or owner of the submission and will not make any alteration, other than as allowed by this license, to your submission. By typing your name and date in the fields below, you indicate your agreement to the terms of this distribution license.

Embargo options (fill box with an X).

Make my thesis or dissertation available to eCollections immediately upon X submission.

Embargo my thesis or dissertation for a period of 2 years from date of graduation.

Embargo my thesis or dissertation for a period of 5 years from date of graduation.

Michelle Dennis

Type full name as it appears on submission

April 6, 2015

Date