University of Florida Thesis Or Dissertation Formatting

INFLUENCE OF MULCH CHEMICAL AND PHYSICAL PROPERTIES ON PREEMERGENCE HERBICIDE INTERACTIONS AND WEED CONTROL

DEBALINA SAHA

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2019

To my beloved husband, Somnath Saha, and to my advisor, Stephen Christopher Marble. I could not have done this without you. Thank you for all your support.

ACKNOWLEDGMENTS

I would like to overwhelmingly express my thankfulness and gratitude towards my advisor, Dr. Stephen Christopher Marble for his invaluable support, and guidance throughout this entire doctoral degree program and shaping my current and future career. I would like to thank my co-advisor, Dr. Héctor E. Pérez for his constant intellectual support and guidance. Thanks to Dr. Brian J. Pearson for serving my committee member, allowing me to become his teaching assistant, and for his constant support and providing feedback on my research. It was not possible without the guidance of Drs. Marble, Pérez, and Pearson. I would also like to thank Dr. Gregory E.

MacDonald and Dr. Dennis Calvin Odero for serving my committee members and providing their feedback to my dissertation. Thanks to Annette Chandler and Rodrigo

Mendez-Boza for helping me in the field, greenhouse, and laboratory experiments.

I would also like to extend my gratitude towards my parents, beloved husband, in-laws, and friends for providing moral support for this entire journey and making it successful.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 7

LIST OF FIGURES ...... 9

LIST OF ABBREVIATIONS ...... 10

ABSTRACT ...... 11

CHAPTER

1 LITERATURE REVIEW ...... 13

Overview of Landscape Weed Management ...... 13 Overview of Nursery Weed Management ...... 14 Mulching as a Weed Control Method ...... 17 Herbicide and Mulch Interaction ...... 19 Physical Properties of Mulch ...... 21 Allelopathic Properties of Mulch ...... 22 Knowledge Gaps ...... 25

2 MULCH TYPE AND DEPTH, HERBICIDE FORMULATION, AND POST- APPLICATION IRRIGATION VOLUME INFLUENCE ON CONTROL OF COMMON LANDSCAPE WEED SPECIES ...... 28

Introduction ...... 28 Materials and Methods...... 31 Results and Discussion...... 36 Herbicide-only Treatments ...... 36 Mulch-only Treatments ...... 39 Herbicide vs. Mulch-only Treatments ...... 40 Herbicide + Mulch Treatments...... 41

3 RESPONSE OF WEED SEED EMERGENCE AND GROWTH TO DIFFERENT PHYSICAL PROPERTIES OF COMMON LANDSCAPE MULCH IN NURSERY CONTAINER PRODUCTION ...... 66

Introduction ...... 66 Materials and Methods...... 69 Outdoor Container Experiment ...... 69 Greenhouse Experiment...... 71 Moisture Retention by Mulch Materials...... 72 Particle Size Analysis of Mulch Materials ...... 73

Results and Discussion...... 74 Outdoor Container Experiment ...... 74 Greenhouse Experiment...... 76

4 ASSESSING HERBICIDE MOVEMENT THROUGH MULCH MATERIALS TO IMPROVE WEED CONTROL ...... 87

Introduction ...... 87 Materials and Methods...... 90 Bioassay ...... 90 Chemical Assay ...... 92 Results and Discussion...... 93 Bioassay ...... 93 Chemical Assay ...... 93

5 ALLELOPATHIC EFFECTS OF COMMON LANDSCAPE AND NURSERY MULCH MATERIALS ON WEED CONTROL ...... 99

Introduction ...... 99 Materials and Methods...... 101 Results and Discussion...... 103

6 CONCLUSIONS ...... 110

LIST OF REFERENCES ...... 115

BIOGRAPHICAL SKETCH ...... 125

LIST OF TABLES

Table page

2-1 Effect of herbicide formulation and post-treatment irrigation volume on control of three landscape/container nursery weed species...... 54

2-2 Effect of mulch type and depth on control of three landscape/container nursery weed species...... 55

2-3 Control of three common landscape/nursery weed species with use of herbicide or mulch at two different depths...... 56

2-4 Response of eclipta, large crabgrass, and garden spurge to the main effects of herbicide formulation, mulch type and depth, and irrigation volume and all three way interactions...... 57

2-5 Effect of herbicide formulation, mulch type and mulch depth on control of three landscape/container nursery weed species...... 59

2-6 Effects of mulch type and irrigation volume on eclipta controlz with indaziflam + mulch combinations...... 61

2-7 Influence of indaziflam formulation, mulch depth, and irrigation volume on mulch effectz in eclipta...... 62

2-8 Influence of herbicide formulation and irrigation volume on mulch effect in garden spurge and large crabgrass...... 63

3-1 Mulch type, depth, and seed placement effects on emergence of large crabgrass (Digitaria sanguinalis) and garden spurge (Euphorbia hirta) in outdoor container experiments...... 80

3-2 Influence of mulch type and depth on percent reduction in light (PAR, µ mol m-2 s-1) in outdoor container experiments over 12 weeks...... 82

3-3 Mulch type, depth, and seed placement effects on emergence of large crabgrass (Digitaria sanguinalis) and garden spurge (Euphorbia hirta) in greenhouse experiments...... 83

3-4 Percent of water retention by three different mulch materials...... 85

3-5 Percentages of particle sizes present in three different mulch materials...... 86

4-1 Percent control of three herbicides applied to different mulch materials...... 96

4-2 Average amount of herbicide detected in soil samples following application to pots mulched with pinebark and those containing no mulch...... 97

4-3 Percent of preemergence herbicides retained by pinebark...... 98

5-1 Response of large crabgrass (Digitaria sanguinalis) and garden spurge (Euphorbia hirta) seeds to extracts of three different common landscape mulch materials...... 109

LIST OF FIGURES

Figure page

2-1 Different particle sizes present in hardwood, pinebark and pinestraw mulch materials after screening through soil sieves. 1 mm = 0.0394 inch...... 64

2-2 Influence of mulch type and mulch depth on the herbicide effect observed in large crabgrass (treated with prodiamine) and garden spurge (treated with dimethenamid-P + pendimethalin)...... 65

LIST OF ABBREVIATIONS

ANOVA Analysis of Variance

DAT Days After Treatment

FW Fresh Weight

HW Hardwood

LSD Least Significant Difference

PB Pinebark

POST Postemergence

PPFD Photosynthetic Photon Flux Density

PRE Preemergence

PS Pinestraw

SC Shredded Cypress

WAS Weeks After Seeding

WAT Weeks After Treatment

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

INFLUENCE OF MULCH CHEMICAL AND PHYSICAL PROPERTIES ON PREEMERGENCE HERBICIDE INTERACTIONS AND WEED CONTROL

Debalina Saha

May 2019

Chair: Stephen Christopher Marble Cochair: Héctor Eduardo Pérez Major: Horticultural Sciences

Weed control is an important aspect in ornamental crop production as weeds reduce crop growth and market value. Currently, growers rely on preemergence herbicides with supplemental hand weeding for weed control. However, an alternative strategies for long- term weed control could combine preemergence herbicides and organic landscape mulch materials. Research conducted at Mid-Florida Research and

Education Center, Apopka, FL showed that 2 inches depth of mulch materials including pinebark, pinestraw, and hardwood, alone or in combination with preemergence herbicides including prodiamine, indaziflam, and dimethenamid-P + pendimethalin were effective in controlling large crabgrass (Digitaria sanguinalis), eclipta (Eclipta prostrata), and garden spurge (Euphorbia hirta). Weed seeds placed above the mulch surface showed greater emergence compared to seeds placed below the mulch layer.

Hardwood mulch provided less weed control due to higher moisture content (60%) and smaller particl size than the pinebark or pinestraw mulch. There was no difference in light intensity below mulch depth of 1 inch or greater. However, in certain cases, mulch depths of > 2 inches showed better weed control probably attributed to the mulch’s

physical barrier and limited moisture availability. The mulch layer in nursery containers can retain significant amount (80% to 90%) of preemergence herbicides but was determined to be pinestraw the appropriate mulch type for these preemergence herbicides. Mulch materials evaluated in this research will have little to no effect on control of large crabgrass or garden spurge from an allelopathic standpoint. However, while shoot and root length of garden spurge was not adversely affected by mulch extracts, germination decreased significantly when garden spurge seed were treated with pinestraw mulch extract. Overall, the findings from this research can be applied for weed control in landscapes and container nurseries for about 9 weeks as mulch in combination with herbicides can be an effective and environment-friendly weed control strategy.

CHAPTER 1 LITERATURE REVIEW

Overview of Landscape Weed Management

Weed control is important for horticultural and agronomic crops because weeds compete for light, nutrients, water, and space cause reductions in crop growth. In addition, weeds can harbor insects, pests, diseases, and pathogens resulting in further reduction of market value.

Weed control in landscape planting beds, or non-turf areas, continues to be a challenge for pest control applicators, partially due to a lack of research (and funding for research) in this area. Government agencies and private funding sources unfamiliar with the landscaping industry often dismiss this area as a small economically insignificant sector despite being a multibillion-dollar industry. In 2007, there were over 100,000 landscaping service companies in the United States, which generated $54 billion in sales and employed more than 1 million people (Hodges et al., 2011). Although financial impact represents all landscaping services, a predominate portion of these sales comes from landscape maintenance contracts. One of the most costly aspects of landscape maintenance is weed control in perennial or woody ornamental planting beds and other nonturf areas; In many cases, companies do not offer this service due to labor required

(for hand-weeding) and fear of causing damage to ornamental plants. Many preemergence (PRE) and postemergence (POST) herbicides are labeled for use in turf, which makes weed control in lawns more manageable and cost effective. In contrast, few herbicide exist for weed control around desirable ornamentals, and the common planting of many different ornamental species in a planting bed make finding safe and effective herbicides difficult. In many cases, landscape contractors must resort to hand

weeding in planting beds, thus paying high labor costs that continue to increase (Martin and Calvin, 2010). Small improvements in this area, such as finding the most effective mulch and herbicide combinations, could potentially represent significant labor savings for these companies and consequently higher profit margins. As there are no recommendations on herbicide labels for making applications to mulched beds, research is also needed to determine proper placement (above or below the mulch layer) (Wilen and Elmore, 2007), and irrigation required for activation based on mulch depth. This could also be a service offered with more regularity once companies determine that various mulch-herbicide combinations can be used with little to no non- target damage.

Overview of Nursery Weed Management

In a restricted growth environment, such as container plant production, weeds have been shown to reduce marketability and crop growth by up to 60% (Fretz, 1972).

Other than a select group of graminicides, which can be applied to certain ornamentals, there are virtually no POST weed control options in container nursery production other than handweeding. Thus, weed control is typically achieved through use of PRE herbicides in combination with supplemental handweeding. Weed control in container nursery production is often the highest production cost encountered by nursery growers, often exceeding $4,000 per acre (Case et al., 2005; Mathers, 2003). In Florida, the costs can be significantly more due to the ornamental species grown in the state, which are not grown in other areas, and thus are not listed on herbicide labels. Crops such as jungle flame (Ixora coccinea), golden dewdrop (Duranta erecta), garden croton

(Codiaeum variegatum), blue plumbago (Plumbago auriculata), and ti plant (Cordyline fruticose) are economically important in Florida but are listed on no or very few

herbicide labels, which makes weed management a challenge for growers. As limited funding is available to screen ornamentals, which are considered a niche from a national perspective for herbicide tolerance, research is needed to determine effective and economical weed control methods for these species. One possible way in which phytotoxicity could be reduced in certain crops is the application of herbicide treated mulches to containers at the time of potting (Somireddy, 2012). Preliminary research has shown promising results in this area. Mathers et al. (2004) found that acetochlor treated pine nuggets or hardwood was not phytotoxic to neither shrub rose (Rosa ‘Care

Free Beauty’) nor boxwood (Buxus ‘Green Gem’) and but was only slightly phytotoxic to

Japanese spirea (Spirea ‘Little Princess’) while direct herbicide applications were significantly injurious or caused plant death. Applying these principles to weed control in tropical ornamentals or bedding plants in landscapes has the potential to significantly reduce phytotoxicity issues and overall weed control costs. However, the correct mulch- herbicide combinations must first be determined as well as weed species response to various mulch materials applied at different depths.

Another issue with weed control in nursery production is the number of herbicide applications needed per year and associated costs with these applications. Nursery growers typically make anywhere from 4 to 8 PRE herbicide applications per year, primarily using granular herbicide formulations (Samtani et al., 2007) due to phytotoxicity issues associated with some liquid formulations (Adams, 1990). Although costs will vary depending on active ingredient, an average price for a 50 lb. bag of preemergence herbicide is approximately $80. As most products such as indaziflam

(Marengo G), dimethenamid-P + pendimethalin (Freehand) are labeled for rates of 200

lbs. per acre, this would represent an approximate cost of $240 per acre, per application

(information obtained from a personal communication with Dr. Chris Marble). On a yearly basis, for a 50-acre nursery making six applications per year, herbicide chemical costs alone would be approximately $72,000, not including labor costs needed to make the application.

An issue with broadcasting granular herbicides is non-target losses associated with applying these herbicides to spaced containers. When applying herbicides to spaced containers, over 80% may be deposited into the spaces between the containers

(Gilliam et al., 1992) and become unavailable for weed control. In addition to an added cost, this promotes leaching of herbicides into irrigation ponds and offsite movement with runoff water can also occur. Certain herbicide active ingredients have been shown to cause phytotoxicity to non-target plants when residues are found in irrigation water

(Samtani et al., 2007). Uniform application of herbicides to the container substrate may also not be possible in some cases because of the plant canopy that unevenly covers the container (Gorski, 1993). Using organic mulches as a herbicide carrier have been shown to reduce leaching (Knight et al., 2001), the number of herbicide applications needed (Derr, 1994), and the herbicide rates needed for commercially acceptable weed control (Chen et al., 2013). However, much of the previous research was conducted using active ingredients which are no longer registered such as EPTC (or Eptam 5G,

Gowan USA Trf & ornamental Co., Yuma, AZ) or widely used in container plant production or landscape management such as simazine. Therefore, future work should focus on the newer herbicides, which are currently being widely used by practitioners.

Mulching as a Weed Control Method

Any material that grows upon or is applied to the soil surface can be defined as mulch (Chalker-Scott, 2007). The common materials that can be used as mulch include yard trimmings, straw, crop or plant residues, leaves, paper, plastic films or geotextiles, bark, gravel and many other materials (Marble, 2015). Waste or byproducts which include pinebarks, coconut coir, nut hulls, wood shavings, newspaper pellets and sea shells can also be used as (Sibley et al., 2004; Somireddy, 2012) but availability and popularity are typically restricted to particular regions. For instance, pinestraw, barks, shredded hardwood, cedar chips, gravel or stones are the popular materials in many regions of the United States (Marble, 2015). Although usually applied for appearance, organic mulch can increase the organic content, reduce erosion, reduce soil compaction, moderate soil temperature, and increase water availability (Chalker-Scott,

2007).

Although mulch is multi-functional, it is primarily applied for aesthetic purposes and weed control. Multiple types of mulch have been shown to reduce labor costs associated with handweeding and reduce the need for POST herbicides (Wilen and

Elmore, 2007). While the mechanism of weed control for all mulch types is not well understood, for most species, control can be predominately attributed to light exclusion

(Teasdale and Mohler, 2000). Most of the annual weed species have small seeds that require light to germinate (Wesson and Wareing, 1967). These seeds mostly germinate at a very shallow depth or at the soil surface (Popay and Roberts, 1970). So, when mulch is applied uniformly at a depth of 5 cm or more, there is a reduction of weed seed germination by the exclusion of light and promotion of prolonged dormancy (Fitter and

Hay, 1987). A thin layer of mulch can also reduce the photosynthetic ability of those

weeds, which germinate near the soil surface (Crutchfield et al., 1986; Facelli and

Pickett, 1991). Several mulch materials like the pinebark nuggets possess hydrophobic properties, and as a result, dry out very quickly immediately after the irrigation or rainfall, thereby minimizing the availability of water to the germinating weeds

(Richardson et al., 2008). Thus, application of mulch at a depth of 5 cm or more can effectively control the weeds by acting as a physical barrier to weed seed germination and growth.

Another way in which certain types of mulch can reduce weed growth is through allelopathy. Weeds can be controlled by using organic mulch, which inhibit germination and suppress growth due to allelochemicals released after application (Duryea et al.,

1999; Skroch et al., 1992). When mulches are applied fresh, the control effect is greater due to higher levels of allelochemicals (Duryea et al., 1999). Research by Duryea et al.

(1999) showed that almost all fresh mulch materials had some amount of allelopathic property and ranked their effects on weed species as greatest by pinestraw followed by pinebark and shredded cypress. These fresh mulch materials contained hydroxylated aromatic compounds, which exhibited allelopathic effects on lettuce (Lactuca sativa) seed germination. However, when fresh mulch is applied there can also be negative effects on ornamental plant health due to these same allelochemicals (Mary et al.,

1996) or potentially due to high carbon: nitrogen ratio, which causes nitrogen immobilization by soil microorganisms (Pickering and Shepherd, 2000). Again, more research is needed to determine optimum herbicide-mulch combinations not only from a weed control perspective but also from a phytotoxicity perspective accounting for both injury caused by herbicides and also by mulch allelochemicals.

Herbicide and Mulch Interaction

Herbicide placement in regards to the mulch layer (i.e. above or below the mulch) is an important factor to be considered because different mulch materials interact differently with various types of herbicides (Marble, 2015). With the increased level of organic matter on the soil surface, there can be a decrease in herbicide efficacy with certain herbicides, such as dinitroanaline herbicides, which bind tightly to organic materials (Buhler, 1992). This reduction in herbicide efficacy occurs due to mulch materials that intercept and bind the herbicide, preventing from reaching the soil surface where it is needed for weed control (Banks and Robinson, 1986; Chauhan and Abugho,

2012). An additional reduction could be due to an increased rate of degradation caused by increased microbial activity there can be reduction in herbicide efficacy (Locke and

Bryson, 1997). In most cases a herbicide can only be applied below a mulch layer at one time and subsequent applications would need to be made on top. Therefore, determining which products are most effective either above or below a mulch layer could help develop timing recommendations for these products in the landscape and nursery industries.

Most of the previous research in this area has focused on agronomic cropping systems and conservation tillage where crop residues are left on the soil after harvest

(Locke and Bryson, 1997). However, some studies have focused on interactions between common landscape herbicides and mulch materials. In a study by Chen et al.

(2013), EPTC was applied either above or below pinestraw (PS), pinebark (PB), or shredded cypress (SC) mulch to evaluate yellow nutsedge (Cyperus esculentus) control. Results indicated that better yellow nutsedge control was achieved when EPTC was applied under mulch, with the greatest effect noticed in SC mulch. This result may

be due to the volatile nature of EPTC (4,530 mPa) (Abu-Qare and Duncan, 2002; Baker et al., 1996), which could have degraded more quickly when applied on top of mulch materials. Based on their data, Chen et al. (2013) recommended that EPTC be applied under mulch or before replenishing existing landscape beds with fresh mulch. Chauhan and Abugho (2012) investigated the use of rice (Oryza sativa) residue mulch with pendimethalin and oxadiazon, both of which are common landscape PRE herbicides. In this study, rice residue was applied to the soil surface of pots filled with field soil. Pots were then treated with either oxadiazon or pendimethalin using spray formulations on top of mulch. Based on the study by Chauhan and Abugho (2012), overall data suggested that some weed seedlings may be able to survive herbicide treatment in the presence of residue (i.e., mulch), which acts to intercept herbicide. Banks and Robinson

(1984) investigated the effects of oryzalin applied to straw-covered and nonmulched soils. Presence of straw at the time of application reduced the amount of oryzalin reaching the soil surface, and the concentration of oryzalin in the soil declined as the amount of straw increased. A study by Crutchfield et al. (1986) investigating the effects of metolachlor (a common PRE herbicide labeled for landscape use) when applied to mulch showed that although soil concentrations of metolachlor were lower when applied in the presence of wheat (Triticum aestivum) straw mulch, weed control increased along with increasing mulch level, likely due to the weed suppression ability of the mulch and because metolachlor is highly mobile (Sanchez-Martin et al., 1995). In studies by

Richardson et al. (2008), pinebark nuggets were applied to nursery containers at depths of 0, 3.8, and 7.6 cm and either treated with flumioxazin or left untreated (mulch only). In general, weed control increased with increasing mulch depth, regardless of herbicide

treatment, indicating that in deeper mulch depths, weed control may be attributable more to mulch than to herbicides and that herbicides will have a more significant effect on efficacy at lower mulch depths (Somireddy, 2012). Another innovative approach to investigate herbicide-mulch interactions would be the application of mulches already treated with herbicides as described by Mathers (2003) and Mathers et al. (2004). This practice involves pre-treating mulches with PRE herbicides and then applying the mulch in and around the landscape plantings or to a container (Somireddy, 2012). While some work has been conducted in this area showing improved weed control over time and less phytotoxicity effects (Mathers, 2003; Mathers et al., 2004), only a few active ingredients (isoxaben, trifluralin, oryzalin, oxadiazon, etc.) have been investigated and limited work has been conducted on the newest or most widely used herbicides such as indaziflam, dimethenamid-p, or flumioxazin.

Physical Properties of Mulch

Mulch materials such as pinebark and hardwood chips can vary significantly in their physical properties depending on how they were manufactured. Physical properties of mulch includes, mulch particle size, water holding capacity, light transmittance, and mulch depth. Teasdale and Mohler (2000) introduced the theoretical relationships among mulch mass, area index, height, cover, light extinction, and weed emergence. They found, successful weed emergence through mulches was related to the capacity of seedlings to grow under limiting light conditions around mulch material as the obstructing element. Previous studies have shown that organic mulches can retain more water than inorganic (Balvinder et al.,1988; Iles and Dosmann, 1999).

Water holding capacity of mulch depends on particle size and interception of water varies with mass and water holding capacity of organic matter (Naeth et al., 1991). Most

of the studies conducted with mulch for weed control efficacy are based only on different mulch types (pinebark, wood chips etc.) or evaluation of mulch depths. The results reported are inconsistent as none (or rarely) of the authors have studied the physical properties of the organic mulch materials.

Allelopathic Properties of Mulch

Molisch (1937) first used the term allelopathy, defined as any direct or indirect harmful effect by one plant on another through production of chemicals that are released into the environment (Rice, 1984). Allelochemicals are diverse in chemical structure and produced by plants as secondary metabolites (Hadacek, 2002; Paiva,

2000). They are released by root exudation, volatilization, and death and decay of plants, through leaching from living or decaying residues (Anaya, 1999; Rice, 1984).

Toxicity of these allelochemicals are determined by several factors including concentration, flux rate, age and metabolic state of the plant, and prevailing climate and environmental conditions (Gallet and Pellissier, 1997; Kohli et al., 1993; Nilsson et al.,

1998; Wardle et al., 1993; Weidenhamer, 1996).

Allelochemicals can also be used for weed management as they act as natural herbicides. Many investigations have been completed or are ongoing evaluating use of cover crops and their residues for weed suppression. Some results are positive showing enhanced weed suppression and thereby reducing needed herbicide applications, and others with mixed results. Cover crop residue provides a weed-suppressive “mulch” effect due in part by providing a physical barrier, but also due to phytotoxins being released from decomposing residues which impacts weed control selectivity (Burgos and Talbert, 2000; Nagabhushana et al., 2001; Putnam, 1988; Weston, 1996).

Duryea et al. (1999) compared chemical, allelopathic, and decomposition properties of six common landscape mulch materials including cypress (Taxodium distichum and T. distichum var. nutans Sweet), eucalyptus (Eucalyptus grandis ex

Maiden), pinebark {splash pine (Pinus elliottii) and loblolly pine (P. taeda)}, pine needle

(Pinus elliottii), melaleuca (Melaleuca quinquenervia), and a utility-trimming mulch

(GRU) composed of multiple species [oaks (Quercus laurifolia and Q. rubra, Q. virginiana), and cherry (Prunus serotina)], with a small amount of southern redcedar

[Juniperus virginiana var. silicicola (Small) Silba] and southern pines (Pinus spp.).

Bioassays were conducted by extracting water-soluble chemicals from the mulches followed by application to germinating lettuce seeds and germinants for each mulch extract were recorded. Results showed hydroxylated aromatic compounds were highest in GRU and lowest in melaleuca, pinebark, and pinestraw, but all showed levels of significant activity in bioassay (Duryea et al., 1999). The authors hypothesized allelopathic properties of these mulches could potentially reduce germination of landscape weed species, even though this was not evaluated.

Hardwood chips from eucalyptus and cypress, both common in southeast United

States, contain more phenolic compounds (tannins) than pinebark and pinestraw

(Duryea et al., 1999) and may inhibit germination of weed seeds and seedling growth.

Some of the phenolic compounds identified in eucalyptus hardwood are quinic acid, gallic acid, protocatechuic acid, catechin, and chlorogenic acid (Santos et al., 2013).

Maimoona et al. (2011) demonstrated that bark of chir pine (Pinus roxburghii) and Bhutan pine (Pinus wallichiana) contained catechin and gallocatechin derivative, quercetin, kaempferol, secoisolariresinol, 3, 4-dihydroxybenzoic acid, and rhamnetin.

Previous studies have showed that allelopathy in conifers is due to the presence of phenolic compounds. The above-mentioned phenolic compounds present in the pinebark may be responsible for allelopathy. Terpenoids, such as β-pinene, myrcene, camphor, and cineole, a group of naturally occurring chemicals have allelopathic effects as toxic, inhibitory or deterrent compounds (Langenheim, 1994). Harman-Ware et al.

(2016) reported presence of monoterpenes, α- and β-pinene, camphene, and δ-carene in the terpenoids extract of loblolly pine saplings and pine lighter wood. The β-pinene and camphene are two important potential allelopathic compounds that may be present in pinebark mulch materials and responsible for inhibition of weed seed germination and growth. Further studies are needed to identify the specific allelochemicals present in pinebark mulch, as not much information is available.

Although research has focused on allelopathic properties of various agronomic crop residues and cover crops and their effect on weed suppression or potential as natural herbicides/herbicide templates (Weston, 2005), these materials would not be suitable in landscapes due to rapid decomposition, availability, and appearance

(Marble, 2015). There remains a significant knowledge gap concerning identification and quantification of potential allelochemicals present in the common landscape mulch materials.

Knowledge Gaps

While some studies have investigated herbicide–mulch combinations in both field and container studies, there is still a significant knowledge gap in this area pertaining to interactions between the most common landscape herbicide and mulch combinations.

For example, there are essentially no previous reports examining various landscape herbicides and pinestraw mulch, one of the most common mulch materials in the southeastern United States, in terms of how herbicide efficacy is influenced by its use as a mulch in the landscape. Several new herbicides have become available to the landscape sector in recent years, including dimethenamid-P, indaziflam, and new formulations of older materials such as isoxaben. Another aspect to consider is to determine if there is a need to alter irrigation procedures after application to mulch in the landscape. Herbicides will need to be watered after application to be incorporated

(Altland et al., 2003), but few studies have examined whether more irrigation is needed for improved efficacy in mulched landscape beds. Herbicide efficacy in soilless container substrates can be influenced by irrigation volume after application (Yang et al., 2013), and it would seem that more irrigation may be needed to move herbicides through organic mulch layers to reach the soil surface to control germinating weeds. It is also important to note that most of the previous studies evaluated weed germination from underneath mulch layers. Richardson et al. (2008) showed that at shallow mulch depths, weed seed entering mulched areas are more likely to germinate than seeds on the soil surface. Therefore, it is not only critical to examine different mulch–herbicide combinations, but also how weed seed placement (either existing on the soil or blown into the bed after installation) affects weed germination through various mulch–herbicide combinations.

Research is also needed to determine how different mulch materials affect herbicide leaching and runoff after application. In a study by Wilson et al. (1995), maximum cumulative herbicide losses of isoxaben, oryzalin, and trifluralin occurred within the first 6 d after application, as similarly reported by Wauchope (1987) with atrazine. Indaziflam leaching can increase with the increase in rainfall as it can leach up to 12.2 + 0.8 cm (values are expressed + SD) and 27.2 + 2.6 cm at 5 and 15 cm ha-1 rainfall, respectively (Jhala and Singh, 2012). Precipitation levels 0-17 mm within 7 d after dimethenamid-P application provided unacceptable weed control and cumulative precipitation during the 14 d after pendimethalin application reduced green foxtail

(Setaria viridis) control (Stewart et al., 2012). It is known that many different landscape mulch materials reduce soil erosion and slow water runoff (Chalker-Scott, 2007), and different mulch materials have been shown to reduce herbicide movement and off-target contamination (Fawcett et al., 1994). Organic mulch materials used as herbicide carriers can reduce herbicide leaching by up to 74% in the landscape (Knight et al., 2001). More research is needed with common landscape herbicides to determine how different mulch materials affect leaching and runoff potential and whether environmental concerns resulting from herbicide runoff and leaching can be mitigated through the use of certain mulch materials.

(Marble, 2015). There remains a significant knowledge gap concerning identification

and quantification of potential allelochemicals present in common landscape mulch materials. In addition to characterizing allelopathic properties of current and potential mulch materials, further investigation of the mechanism of action of allelochemicals is also needed (Weston, 2005). Another aspect to consider is the extraction process of newly identified allelochemicals from landscape mulches and if their persistence in the environment in sufficient concentration to affect weed species (Ferguson et al., 2003).

From a commercial perspective, further work is also needed to identify how aging and handling these materials after harvest impacts allelochemical composition, as age has been shown to have a significant effect on allelochemicals composition (Achakzai et al.,

2009). Mulch availability typically varies by region and what materials are available locally (Marble, 2015). A better understanding of potential allelopathic effects of these mulch materials could be used by mulch manufacturers for promotion and to aid the horticulture industry in selecting mulch for different applications. Due to a high degree of variability in both allelopathic potential and weed species response reported here, it is important that researchers identify key characteristics of mulch materials used including plant species, age, plant parts used, and harvesting and handling procedures prior to experimentation. Identifying activity of these compounds on economically important weed species would also be beneficial from a weed management perspective.

CHAPTER 2 MULCH TYPE AND DEPTH, HERBICIDE FORMULATION, AND POST-APPLICATION IRRIGATION VOLUME INFLUENCE ON CONTROL OF COMMON LANDSCAPE WEED SPECIES

Introduction

Mulch provides many benefits to landscape ornamentals including soil temperature regulation (Fraedrich and Ham, 1982; Montague and Kjelgren, 2004), increased soil moisture (Fraedrich and Ham, 1982; Iles and Dosmann, 1999; Kraus,

1998; Litzow and Pellett, 1993; Watson, 1988; Watson and Kupkowski, 1991), and improved overall plant growth and survival (Green and Watson, 1989; Greenly and

Rakow, 1995; Litzow and Pellett, 1993). Similarly, mulch increases growth of container- grown ornamentals by providing the same benefits in a production (i.e., nursery) environment (Amoroso et al., 2010; Lohr, 2001).

Mulch is most often applied in landscape planting beds for aesthetic purposes and for weed management (Chalker-Scott, 2007). Mulch is less commonly used in container nursery production but may be employed as a non-chemical weed management option for sensitive plant species (Case et al., 2005). Many different mulch materials have been evaluated for weed control in container plants. Richardson et al.

(2008) reported up to 150 d of yellow woodsorrel (Oxalis stricta) and hairy bittercress

(Cardamine hirsuta) control in large (3 to 7 gal) container-grown ornamentals with 3 inches of pinebark mini-nuggets. Similarly, Cochran et al. (2009) showed that 1-inch pinebark mulch reduced garden spurge and eclipta fresh weights and weed counts by over 80% compared with a non-mulched control. Reviews of different mulch materials as a sole means for weed control have been summarized for landscape (Chalker-Scott,

2007) and nursery production (Case et al., 2005).

A review of earlier research focusing on use of mulch in combination with, or in comparison to, preemergence (PRE) herbicides was recently published by Marble

(2015). In many cases, research focused on evaluating different mulch or herbicide + mulch combinations to determine the most effective on target weed species, or evaluated use of herbicide treated mulches (Case and Mathers, 2006a, 2006b). For example, Bartley et al., (2017) evaluated three different mulch types applied at three depths (1 inch, 2 inches, 4 inches) with and without addition of dimethenamid-P. The authors reported that 168 d after treatment, herbicide was no longer a significant factor as dimethenamid-P had lost all efficacy, and mulch depth was the only significant factor, with depths of 1 to 4 inches providing 90% to 100% garden spurge control for up to 90 d after seeding. All of these previous reports establish that different herbicide + mulch combinations can potentially provide a high level of control of many different weed species.

Previous studies have shown that mulch can provide substantial weed control when applied alone at adequate depths (Cochran et al., 2009; Richardson et al., 2008;

Wilen et al. 1999); however, it is unclear whether it is mulch or herbicide contributed most to observed weed control. Furthermore, a decrease in herbicide efficacy with certain herbicides such as dinitroanilines, which bind tightly to mulch, can occur with increased levels of organic matter on the soil surface. Consequently, the herbicide becomes unavailable for weed control (Buhler, 1992). Mulch depth is an important factor to consider (Banks and Robinson, 1986; Chauhan and Abugho, 2012) because application of thin mulch layers reduces PRE efficacy in several agronomic studies due to rapid degradation caused by increased microbial activity (Locke and Bryson, 1997).

Alternatively, herbicide placement (i.e. application of herbicides under mulch, or making the application prior to mulch addition) improves weed control compared with PRE application on top of mulch (Chen et al., 2013). However, in most landscape situations, this application could only be made initially, and subsequent applications would have to be applied on top of existing mulch layers.

Increasing post-treatment irrigation levels to mulched areas could be a means of increasing herbicide concentrations in the soil as PRE herbicides must be incorporated into the soil by irrigation following application. In most cases, PRE herbicide labels recommend irrigation volumes of 0.2 to 0.5 inch soon after application. Banks and

Robinson (1986) reported that reduced amounts of either acetochlor, alachlor, or metolachlor were received on the soil surface as wheat (Triticum aestivum) straw mulch depth increased resulting in the need for higher irrigation volumes for thicker mulch layers. However, wheat straw has very different physical properties compared with common landscape mulch materials and the herbicides that were evaluated are not commonly applied in landscape situations. Additional research is required to determine the extent to which activation irrigation can improve efficacy of different herbicide + mulch combinations commonly used in landscapes and container nurseries. This research was conducted to accomplish three primary objectives. First, we wanted to determine efficacy of multiple herbicide + mulch combinations and determine which factors significantly affected control, specifically focusing on herbicide formulation and post-treatment irrigation volumes. Second, we wanted to determine efficacy derived from mulch or herbicides used alone under the same conditions as the herbicide +

mulch combinations. Third, our goal was to identify differences in the additive effects of the herbicide + mulch combinations compared with use of only herbicides or only mulch.

Materials and Methods

Greenhouse experiments were conducted at the Mid-Florida Research and

Education Center in Apopka, FL in 2016 and 2017. Square black plastic nursery containers [4 inches (width) × 4 inches (length) × 5 inches (height) dimensions] were filled with pinebark:peat mix (Fafard®52 growing mix, SunGro Horticulture, Agawam,

MA) amended with 8 lb/yard3 controlled release fertilizer 15N-3.9P-9.9K (Osmocote®

Plus, Everris, Geldermalsen, Netherlands) based on the manufacturer’s recommended medium rate for incorporation. A volume containing approximately 35 seeds of either large crabgrass, garden spurge or eclipta were then surface-sown to the filled containers. All emerged weeds were allowed to grow for the duration of the experiment.

Large crabgrass seeds were supplied by Azlin Seed Services (Leland, MS) whereas eclipta and garden spurge seed were collected from the natural population present at the Mid-Florida Research and Education Center, Apopka, FL. Following seeding, three different mulch types including pinestraw [PS (Pine Straw of Central Florida, Winter

Garden, FL)], pinebark mini-nuggets [PB (Timberline, Old Castle Lawn & Garden Inc.,

Atlanta, GA)], or hardwood chips [HW (Florida Select™ Natural Eucalyptus Mulch,

Scotts®, Marysville, OH)] were applied to containers at depths of 0 inch, 1 inch or 2 inches. Particle size was determined for each mulch material using established methodology (Bilderback et al., 2005). Four samples of mulch material were collected randomly from different bags (HW and PB) or bales (PS) and air-dried in the laboratory for 7 d at 22°C to achieve a constant weight was achieved. From each sample, four 100 g samples were measured out for each mulch type and hand-shaken through a series of

soil sieves (W.S. Tyler, Mentor, OH) ranging from 1 to 25-mm for 3 min. Mulch that was retained in each sieve was weighed and the percent of the total mulch mass was calculated for each screen size and the pan (<1 mm) (Fig. 2-1). Over 50% of PB and PS particles were retained in the 25-mm sieve. Only 11% of HW particles were retained in the largest screen indicating that it had a higher percentage of smaller particles, which would be expected from a shredded wood mulch. Shallow mulch depths of 1 and 2 inches were chosen as deeper mulch depths (e.g. 4 inches or greater) have been associated with increased exposure to pathogens and/or decreased ornamental plant growth (Bartley et al., 2017; Chalker-Scott, 2007; Koski and Jacobi, 2004). In addition to negative impacts associated with high mulch depths, lower mulch depths were chosen to more closely evaluate herbicide + mulch combinations. Mulch applied alone (without the addition of a herbicide) at depths of 3 to 4 inches is often sufficient for season long weed control (Bartley et al., 2017; Richardson et al., 2008), which could potentially reduce the ability to determine what effects herbicides may have when used in combination with mulch. Mulch applied at 2 inches is also often used as standard mulch depth in the landscape industry. In the nursery industry, mulch may be applied at depths up to 2 inches, but is rarely applied at a higher depth because greater mulch depths would require more space in a container, and thus limit the potential root area in a pot.

Depths of 1 to 2 inches were chosen in order to closely evaluate herbicide and mulch effects and also chose depths that would be applicable to both the landscape and nursery and landscape industries.

Liquid or granular formulations of indaziflam (Marengo® 0.622 SC and Marengo®

0.0224 G, Bayer Crop Science, Research Triangle Park, NC), prodiamine ( Barricade® 4

FL, Syngenta Crop Protection, Greensboro, NC and RegalKade® 0.5 G, Regal

Chemical Co., Alpharetta, GA), and dimethenamid-P + pendimethalin (Tower® 6 EC +

Pendulum® 3.3 EC and Freehand® 1.75 G, BASF Corp., Research Triangle Park, NC) were applied on August 17, 2016 (first experimental run) (application time 10:30 AM, temperature 28 °C, humidity 69%, wind speed 6 mph, sunny condition) and April 2,

2017 (second experimental run) (application time 10:40 AM, temperature 27.7 °C, humidity 65%, wind speed 5 mph, sunny condition) at labeled rates to pots seeded with eclipta, large crabgrass, or garden spurge respectively. These herbicides/weed combinations were selected because these are the common weed species that are found in landscapes and container nurseries. The herbicides used are the most widely and commonly applied in ornamental crop production. Liquid formulations were applied with a carbon dioxide (CO2) backpack sprayer (Bellspray R&D sprayer Inc., Opelousas,

LA) calibrated to deliver 20 gal/acre using an 8004 flat-fan nozzle (TeeJet Technologies,

Wheaton, IL) at a pressure of 30 psi. Granular formulations were applied to each container separately using a hand-shaker. Following treatment, pots were placed inside a greenhouse which was fully lighted by natural light. Each container received either 0.5 inch, 1 inch, or 2 inches of irrigation by hand using an irrigation wand (Melnor® Titanium

Series Front-Trigger 7-Pattern Aqua Gun®, Melnor, Inc., Winchester, VA) that was previously calibrated to determine flow rate. The three irrigation volumes were applied 1 d after herbicides application. Following initial hand watering, used as a one-time herbicide activation irrigation, pots were kept dry for 3 d. After 3 d, all containers were irrigated using overhead sprinklers and received 0.2 inch total per day via two separate

irrigation cycles throughout the remainder of the trial. Air temperature was maintained between 21 to 35 oC inside the greenhouse throughout the trial.

The experiment consisted of a factorial treatment arrangement of the two herbicide formulations (granular or spray-applied), three mulch types (HW, PB, PS), two mulch depths (1 and 2 inches), and three levels of one-time, post-treatment irrigation volumes (0.5 inch, 1 inch, and 2 inches). Three sets of controls were included in the study to accomplish the objectives noted previously. The first set of controls included only the three mulch types applied at the two depths and received only the 0.5 inch irrigation volume. The second set of controls included only the two herbicide formulations and the three one-time irrigation volumes, while the last set of control pots received no treatment (no herbicide or mulch) and received only the 0.5 inch irrigation volume. This design yielded 49 individual treatments per weed species (36 formulation

× mulch depth × mulch type × irrigation combinations, six formulation × irrigation combinations, six mulch type × mulch depth combinations, and one no herbicide, no mulch control) with eight single pot replications per treatment, per weed species. Pots were grouped by weed species in a completely randomized block design and each weed species was treated as a separate experiment.

Weed shoot fresh weights (FW) were recorded at 60 DAT. Shoot FW data were collected by clipping weeds at the soil line and weighing on a portable scale. Data collected in non-treated pots (no mulch, no herbicide) and pots receiving either a herbicide or a mulch treatment was used to calculate percent control and the percent increase in control achieved with the herbicide + mulch combination compared with either herbicide or mulch applied alone. First, percent control (shoot FW reduction) of all

treated pots (herbicide + mulch, herbicide-only, or mulch-only) was calculated based on

FW of the non-treated pots (no mulch or herbicide) using the formula [(FW non-treated

– FW treated)/FW non-treated)] × 100. The percentage increase in control was then calculated for the herbicide + mulch combination based on 1) mulch treatments receiving no herbicide (herbicide caused increase in control or “herbicide effect”)

[(percent control combination - percent control mulch)/percent control mulch] × 100, and

2) herbicide treatments receiving no mulch (mulch caused increase in control or “mulch effect”) [(percent control combination - percent control herbicide)/ percent control herbicide] × 100. This resulted in three different data points for each weed species on each evaluation date: 1) percent control; 2) herbicide caused increase (herbicide effect); and 3) mulch caused increase (mulch effect). This methodology allowed us to determine which herbicide + mulch combination was most efficacious (Bartley et al., 2017) and additionally the significance of experimental variables (herbicide formulation, mulch type, depth, irrigation vol.) when used in combination at different levels.

All data were subjected to mixed model analysis of variance (ANOVA) using SAS

(Ver. 9.4, SAS Institute, Cary, NC) reflecting the factorial treatment arrangement and pooled over trial runs as there were no treatment × trial run interactions. Herbicide only treatments were analyzed to determine significant effects of formulation and irrigation volume when no mulch was included. Replication was considered random while all other experimental variables and interactions between these variables were considered fixed factors. Herbicide-only treatments were analyzed to determine significant effects of formulation and irrigation volume when no mulch was included. Mulch-only treatments were analyzed to determine significant effects of mulch type and depth when no

herbicide was included and to determine the overall efficacy of the mulch. Data from mulch-only treatments was also compared to herbicide-only treatments after pooling data over all irrigation volumes to compare efficacy. The data was pooled to compare the overall efficacy of all mulch treatments to overall efficacy derived from herbicide treatments without regard to irrigation affects. Data from herbicide + mulch combinations (percent control, herbicide effect, mulch effect) were then analyzed as 3

(mulch types) × 2 (mulch depths) × 3 (irrigation levels) × 2 (herbicide formulations) × 3

(weed species) factorial. In all cases, linear or quadratic trends over irrigation rate were determined with orthogonal contrasts and least significant difference (LSD) values were calculated when main effects or interactions for other parameters were significant (P <

0.05). Visual percent control data and seedling counts collected at 30 and 60 DAT are reflected in data derived from shoot FW, therefore only percent control data derived from shoot FW are discussed for brevity. Data with no variance were not analyzed but are discussed and treatment least squares means are presented.

Results and Discussion

Herbicide-only Treatments

Percent control data from herbicide treated pots (no mulch) showed that formulation was significant for all three herbicides as dimethenamid-P + pendimethalin, indaziflam, and prodiamine were all more effective when applied as a spray-applied formulation compared with the granular formulation (Table 2-1). Use of the spray- applied formulations of indaziflam, prodiamine, or dimethenamid-P + pendimethalin resulted in an increase in control of 42%, 17%, and 13% for eclipta, large crabgrass, and garden spurge, respectively, in comparison with the granular formulation. These findings are consistent with earlier reports evaluating flumioxazin in which greater

control was consistently achieved with spray-applied formulations compared with a granular formulation (Wehtje et al., 2015).

Irrigation volume was not significant for large crabgrass, nor were there any formulation × irrigation interactions. Prodiamine is widely recommended for PRE control of large crabgrass and other grassy weeds (Neal et al., 2017). As prodiamine is highly efficacious for large crabgrass, irrigation likely plays little role in efficacy as long as a sufficient amount is applied to properly incorporate the herbicide. In pots seeded with garden spurge, irrigation volume was significant for dimethenamid-P + pendimethalin as there was a linear decrease in garden spurge control as irrigation volume increased from 0.5 to 2 inches. While the decrease was significant, commercially acceptable control (~80%) was achieved at all three irrigation levels averaged over both formulations. Similar to results with large crabgrass, there were no formulation × irrigation interactions with gardens spurge, again likely due to the high efficacy of this combination on Euphorbia species (Neal et al., 2017).

Irrigation was not a significant main effect for eclipta, but there was a significant formulation × irrigation interaction. For the granular formulation of indaziflam, control decreased as irrigation volume increased, ranging from 59% control at 0.5 inches to

41% control at 2 inches. The spray formulation was more effective than the granular formulation and no irrigation effect was observed. It is unclear why these differences occurred. However, higher variability in eclipta control could be due to the fact that although indaziflam is effective for eclipta control, is typically slightly comparatively less efficacious on eclipta than prodiamine is on large crabgrass or dimethenamid-P + pendimethalin is on garden spurge (Neal et al., 2017) and this variability may have led

to greater variability in results. Granular herbicides are often less effective than spray- applied herbicides because granules need to be uniformly dispersed on the soil surface and in adequate quantities so that the herbicidal active ingredient is in close proximity to germinating seedlings (Wehtje et al., 2015). While it was not recorded, it is possible that the higher irrigation volume of 2 inches may have slightly dispersed granules inside pots resulting in less weed control. This is not likely, as all pots were irrigated carefully and uniformly. The influence of post treatment irrigation on granular herbicide formulations likely depends upon the weed species and herbicide. Wehtje et al. (2015) reported that higher post treatment irrigation levels had no influence on efficacy of a granular flumioxazin application for spotted spurge (Euphorbia maculata), but did for bittercress

(Cardamine hirsuta) (Yang et al., 2013). While post treatment irrigation may play a role in granular efficacy, it appears that any effects would be minimal if the required amount of irrigation that is needed to activate the herbicide were applied.

Based on these results across all three species, herbicide formulation will have a greater influence on weed control than irrigation, similar to previous reports (Saha et al.,

2016; Wehtje et al., 2015; Yang et al., 2013). While some irrigation effects were noted in garden spurge treated with dimethenamid-P + pendimethalin, all irrigation levels resulted in commercially acceptable weed control and would have little practical importance. Irrigation and formulation interactions were significant for eclipta, as difference results were observed over the three irrigation levels within the granular formulation but not the more effective spray formulation. While indaziflam is effective for eclipta, eclipta is often difficult to control with PRE herbicides (Marble, unpublished

data). Irrigation levels and other environmental variables could have a greater influence on weed control when troublesome weeds or less efficacious herbicides are evaluated.

Mulch-only Treatments

Mulch type was not significant for large crabgrass or garden spurge, but was for eclipta, as HW provided greater control of eclipta (64%) compared with PB (44%) or PS

(46%) (Table 2-2). Eclipta seeds have been shown to be strongly photoblastic with 0% germination observed in darkness (Altom and Murray, 1996). Albeit less than in high light levels, large crabgrass can germinate in darkness (Chauhan and Johnson, 2008).

Similar to eclipta, garden spurge is also photoblastic with only 1% germination occurring in darkness (Rooden et al., 1970). Although light levels underneath mulch were not evaluated, greater eclipta control with HW may be related to the higher percentage of smaller particle sizes in HW (Fig. 2-1). The HW mulch may have reduced light more effectively than PB or PS applied at similar depths and reduced eclipta emergence, but did not block out enough light to affect garden spurge or large crabgrass. It is unknown how the physical properties of mulch influence light levels on the soil surface and additional research is needed in this area. It is also important to note that while the small particle size of HW mulch may have reduced light levels and prevented growth and/or germination of eclipta, in this study all seeds were placed on the soil surface.

When seeds are placed on the surface of a mulch, as would be expected to occur over time in a production or landscape environment, smaller particle size materials can be prone to encourage greater weed growth (Chalker-Scott, 2007) due to greater water holding capacity.

In addition to the light requirement for seed germination, the seed burial depth (or mulch depth) is an important factor to consider. Mulch depth was significant for all three

weed species, with the 2 inches depth resulting in a 71, 147, and 57% increase in control compared to the 1 inch depth for eclipta, large crabgrass, and garden spurge, respectively. Seedling emergence through 2 inches of mulch depth requires more resources than emergence through lower mulch depths (0 or 1 inch). Greater depths will require more time for emergence and generally uses more of the seed energy reserve

(Black, 1956). Smaller seeds, such as those used these experiments, often cannot emerge from deeper depths as they do not have the ability to store the required resources to emerge from greater depths. Mulch type × depth interactions were not observed for any weed species, indicating that for most weed species, mulch depth will be a more critical factor for control than will mulch type as has been previously reported

(Marble et al., 2015).

Herbicide vs. Mulch-only Treatments

The best eclipta control was observed in pots mulched with HW at 2 inches

(83%) and pots treated with the spray formulation of indaziflam (71%) which were similar (Table 2-3). The spray formulation of indaziflam provided efficacy similar to that of PB and PS applied at the 2 inches depth and outperformed all three mulch types applied at the 1 inch depth. The spray formulation of prodiamine provided the highest large crabgrass control of any treatment (94%) followed by the granular formulation

(78%) (Table 2-3). Both formulations provided greater control than any mulch material, which ranged from 15 to 45% control. As large crabgrass has the ability to germinate in darkness (Chauhan and Johnson, 2008; Holm et al., 1977), mulch was less effective in reducing emergence of this species. Dimethenamid-P + pendimethalin provided 97% control of garden spurge when applied as the spray formulation, and this treatment provided better control than any of the mulch treatments applied at the 1 inch depth.

HW (87%) and PS (87%) at the 2 inches depth provided control similar to that observed with the spray formulation. For all three weed species, the spray formulation provided greater or similar control to that observed with any of the mulch types applied at the 2 inches depth and greater control than the 1 inch depth. The granular formulation was generally less effective than the spray as discussed previously, but still provided greater control than a 1 inch mulch layer in all cases except eclipta mulched with HW at 1 inch which was similar. This suggests that over the short-term (~ 3 months) herbicides will generally outperform mulch materials. However, mulch materials will be slow to degrade

(Duryea et al., 1999) while herbicides will require reapplication every 2 to 3 months

(Neal et al., 2017). As proposed and discussed by Bartley et al. (2017), mulching for weed control will become more beneficial and cost effective over the long-term, as mulch has been shown to provide weed control for several seasons without need for reapplication.

Herbicide + Mulch Treatments

Eclipta control with indaziflam was influenced by formulation, mulch type, mulch depth, and the interactions of formulation × mulch type, formulation × mulch depth, and mulch type × irrigation volume (Table 2-4). Within the granular formulation, HW (84%) provided greater control than PB (66%) or PS (67%), but there was no difference in mulch type within the spray-applied treatments as the spray formulation generally provided better eclipta control throughout the experiment (Table 2-5). In all cases, a higher level of control was achieved with the 2 inches mulch depth compared with 1 inch depth. Similarly, the spray formulation outperformed the granular formulation at all mulch depths. Irrigation volume had no influence on eclipta control in HW or PS; however, greater eclipta control was observed in PB at the 0.5 inch irrigation volume

compared with the 2 inch volume (Table 2-6) as there was a linear decrease in control with increasing irrigation volume. In herbicide only pots, we also observed a decrease in eclipta control at 2 inch irrigation volume compared with the 0.5 inch volume (Table 2-

1). In this case, herbicides were applied to the pots mulched with PB. As we speculated previously, the higher irrigation volume may have dispersed the herbicide, especially the granular formulation, and the herbicide was not uniform on the soil surface and control decreased. However, there was no formulation × mulch type × irrigation interaction, so this does not seem likely. It is unknown why this difference occurred but the difference was of little practical importance as ~70 to 80% control was achieved at all irrigation levels.

The herbicide effect, calculated by determining the percent increase in control achieved with the herbicide + mulch combination in comparison with mulch-only pots, showed that formulation and mulch depth were significant main effects but were influenced by interactions of formulation × mulch type, formulation × depth, and mulch type × mulch depth. The granular formulation of indaziflam contributed to a higher herbicide effect in pots mulched with HW (74%) compared with pots mulched with PS

(34%), while the spray formulation resulted in an increase of 108 to 130% control across all three mulch types and there was no difference in herbicide effect across the mulch types (Table 2-5). In all but one case, the herbicide effect from both formulations of indaziflam was less at the 2 inches mulch depth, indicating that application of indaziflam would have a higher influence on weed control when mulch was at the lower 1 inch depth. The one exception was the granular formulation applied to PB, which caused a

59% increase in control at the 1 inch depth and a 51% increase in control at the 2

inches depth and were similar. Similar to results observed in non-mulched pots, addition of the spray formulation resulted in higher control (118%) compared with the granular formulation (54%) (Table 2-5). As the spray formulation was more efficacious in all instances, it would be expected that the spray formulation provide a greater herbicide effect. Additionally, as mulch depth was found to be significant for all three weed species and all mulch types when evaluating mulch alone (Table 2-2), a greater herbicide effect would be expected at the lower mulch depths.

Mulch effects, calculated by comparing control of the herbicide + mulch combination with pots treated with herbicide only, were influenced by all main effects but significant interactions occurred including formulation × mulch type, formulation × mulch depth, mulch type × mulch depth, formulation × irrigation volume, and the three- way interactions of formulation × mulch type × mulch depth and formulation × mulch depth × irrigation volume. Hardwood resulted in the highest increase in control (121%) within the granular formulation of indaziflam, followed by PB (63% increase) and PS

(16% increase) (Table 2-5). In all cases, a greater mulch effect was observed at the 2 inch depth compared with 1 inch within the granular formulation (Table 2-5). A decrease in control (-105%) was observed in PS applied at the 1 inch depth indicating that the granular formulation was less effective when used in combination with PS compared with its use alone (Table 2-5). There was no difference in mulch types or mulch depth within the spray formulation, indicating that mulch type and depth would not be as significant of a factor in eclipta control when using the more effective spray formulation.

This suggests that at least in the short-term, control would be adequate with the use of the spray formulation alone.

The granular indaziflam + mulch combination also resulted in less control than the granular formulation applied alone when examining the formulation × depth × irrigation interaction evidenced by negative values at the 0.5 and 1 inch irrigation volume (Table 2-7). Across all three mulch types, addition of mulch at the 1 inch depth resulted in 17 and 31% decrease in control when receiving a 0.5 or 1 inch irrigation volume, respectively, compared with the granular formulation applied alone and irrigated similarly. This suggests that the herbicide + mulch combination was less effective than the herbicide applied alone at similar irrigation volumes, indicating that binding occurred tying up the herbicide and making it unavailable for weed control.

There were however, no irrigation effects within the granular formulation applied to the 1 inch mulch depths. Mulch applied at the 2 inches depth provided a 62 to 226% increase when used in combination with the granular formulation compared to the granular herbicide applied alone, and increased linearly from the lowest irrigation volume to the highest irrigation volume. This linear increase was a result of the granular herbicide providing less control at the higher irrigation volume as discussed in Table 2-1. When the spray formulation of indaziflam was applied, there was no difference in the mulch additive effect between the different irrigation volumes when mulch was applied at the 1 inch depth. At the 2 inch depth, the mulch caused a greater percentage increase in control at the 1 inch irrigation volume compared with the 2 inches volume, again coinciding with the control observed in the herbicide-only treatments as shown in Table

2-1. Mulch effect had a quadratic response at both mulch depths in the spray formulation. The mulch effect for the spray formulation ranged from 6% to 40% at the 1 inch mulch depth and 36% to 78% at the 2 inch mulch depth. The higher mulch level

contributed more significantly to overall eclipta control compared with the granular formulation and the mulch effect increased at the higher mulch depth as would be expected.

The spray formulation of prodiamine resulted in 100% large crabgrass control when combined with any mulch treatment, while the granular formulation resulted in

100% control when combined with HW; therefore, these data were excluded from the analysis and only data from the granular formulation used in combination with PB and

PS were analyzed (Tables 2-4, 2-5). The spray formulation resulted in a herbicide effect of 331% to 763% increase in large crabgrass control compared with mulch alone applied at the 1 inch depth and over a 100% increase in control compared to mulch applied at the 2 inches depth (Table 2-5, Fig. 2-2). The mulch additive effect was 8% across all mulch types and depths, indicating use of spray-applied prodiamine will likely contribute to large crabgrass control to a much greater degree than would mulch.

Mulch depth was the only significant effect for large crabgrass control within the granular prodiamine, with the 2 inches depth resulting in greater control than the 1 inch depth. The herbicide effect was influenced by mulch type, mulch depth, and the interaction of these terms (Table 2-4). In all three mulch types, the granular prodiamine had a higher herbicide effect at the lower mulch depth, similar to results observed with eclipta (Table 2-5, Fig. 2-2). Again, this indicates that similar to results with indaziflam, prodiamine will have more of an influence on weed control when lower mulch depths are used. Addition of granular prodiamine resulted in increases of 763, 388% and 442% in control for HW, PB, or PS, respectively, compare to these mulch materials applied alone at the 1 inch depth (Table 2-5). There was a greater mulch effect in PB (427%)

compared with PS (359%). Although not included in the analysis, granular prodiamine resulted in a 763% increase in control compared with the HW used alone (Fig. 2-2). At the 2 inch depth, there was no difference in herbicide effect among mulch types, with prodiamine resulting in a 110% to 140% increase in control for all three mulch materials.

Again, these results indicate that similar to results with indaziflam, prodiamine will have more of an influence on weed control when lower mulch depths are used and as mulch depth increases, the use of herbicides becomes less significant.

The mulch effect for prodiamine treated pots was influenced by the main effects of mulch type and irrigation volume. Pinebark had a greater mulch effect (26%) than PS

(18%) while HW resulted in a 32% increase in control compared with granular prodiamine applied alone (Table 2-5). A greater mulch effect was observed following the

2 inches irrigation volume (39%) compared with the 1 (22%) or 0.5 inch (18%) irrigation volumes as there was a linear increase in mulch effect with increasing irrigation volume

(Table 2-8). As control with herbicide only treatments generally was reduced with higher irrigation volumes, the mulch effect would consequently increase and account for a greater percent of the weed control observed.

Similar to large crabgrass data, there was a high level of spurge control in all pots that were mulched and treated with dimethenamid-P + pendimethalin and all combinations resulted in > 96% control (Table 2-5). Mulch depth was the only significant effect for spurge control with the 2 inches depth (100%) providing 2% greater control than the 1 inch depth (98%) mulch depth (Tables 2-4 and 2-5) across both formulations and all three mulch types.

Herbicide effect was influenced by mulch type, mulch depth, and the interaction of these terms (Table 2-4). At the 1 inch mulch depth, a greater herbicide effect was observed in HW (134% increase), followed by PB (109% increase) and PS (73% increase). At the 2 inch depth, there was a greater herbicide effect for PB (39% increase) compared with either HW or PS (both 17% increase) (Fig. 2-2). Less herbicide effect would be expected at the 2 inch depth as spurge control was greater at 2 inches, thus the herbicide would have less influence. The increase in herbicide effect was related to the ability of the mulch materials to control spurge on their own. That is, there was a greater herbicide effect in less efficacious mulch materials and greater herbicide effect at lower mulch depths.

Mulch effect was influenced by formulation, irrigation, and the interaction of formulation and irrigation (Table 2-4). In pots treated with the granular formulation of dimethenamid-P + pendimethalin, the mulch effect increased with increasing irrigation volume (11, 22, and 41% mulch effect at the 0.5, 1, and 2 inch irrigation volume, respectively). In the spray formulation, there was again a linear increase in mulch effect with increasing irrigation volume (Table 2-8). As the spray formulation was more effective, the mulch effect was lower (4%) compared with the mulch effect of the less effective granular formulation (25%). This shows that similar to eclipta and large crabgrass, mulch contributes more to overall weed control in cases where herbicides are not as effective. Regardless of any herbicide or mulch effect, all mulch and herbicide combinations resulted in commercially acceptable weed control of spotted spurge.

Overall, these results suggest that for the weed species evaluated, PRE herbicides will likely provide similar or greater control than common mulch materials applied at a 1 inch or 2 inches depth, especially when considering spray formulations that consistently outperformed granular formulations. Mulch applied at adequate depths, generally 1 inch or greater, has provided better control of common nursery and landscape weed species when compared to PRE herbicides (Bartley et al., 2017;

Burrows, 2017; Marble et al., 2017). However, short-term studies often suggest PRE herbicides are as, or more, effective than mulch (Ferguson et al., 2008; Smith et al.,

1998), similar to our results over a 60 d evaluation period.

High levels of large crabgrass and garden spurge control (88 to 100%) were observed with all herbicide + mulch combinations evaluated. With these species, herbicides provided more effective control than mulch in most cases, but control increased with the combination. The highest eclipta control was achieved with the spray formulation and with higher mulch depths. While mulch type was not a significant factor in control when the more effective spray formulation was applied, the best eclipta control was observed with the HW mulch when the granular formulation was used. In almost all cases, herbicide additive effects were greater at the lower mulch depths, and mulch additive effects were greater at the higher mulch depth. Similar to findings by

Somireddy (2012), use of herbicides in combination with mulch become more meaningful in terms of weed control when lower mulch depths are used. By examining weed control achieved with use of herbicides alone (Tables 2-1, 2-3), mulch depth also becomes a more important factor when troublesome weed species are evaluated, as was the case with indaziflam applied for eclipta control.

Irrigation volume was not a significant factor for any of the herbicide + mulch combinations when evaluating overall weed control, but was significant when accounting for the mulch additive effect, that is, comparing the herbicide + mulch combination to herbicide-only treatments. However, interactions that occurred are more likely related to the lower control observed in herbicide-only treatments at higher irrigation levels (Table 2-1) and not specifically to any effect irrigation volume had on control.

For eclipta, the indaziflam + mulch combination provided less control than when granular indaziflam was applied alone. Similarly, PS mulch applied at 1 inch + granular indaziflam resulted in a 105% reduction in eclipta control compared with use of granular indaziflam alone. As all weed seeds were placed on the soil surface, underneath mulch, this reduction in control indicates that low PS mulch levels may impede indaziflam movement to the soil surface and reduce weed control. Indaziflam is a relatively new active ingredient for use in landscapes and nurseries and little research has focused on its use in combination with common mulch materials. Burrows (2017) evaluated indaziflam when used in combination with 2 inches of PB mini-nuggets but reported a high level of control and no difference when indaziflam was applied either above or below the mulch layer. However, a high level of control was also reported with mulch- only treatments in that experiment; therefore, it is not clear whether mulch caused any antagonistic effects.

Overall, results from these experiments show that spray formulations of prodiamine and dimethenamid-P + pendimethalin are more effective than granular formulations when applied alone, while indaziflam was more effective as a spray

formulation when used both alone and in combination with mulch. Increasing irrigation volume lead to statistically lower levels of weed control in the case of indaziflam and dimethenamid-P + pendimethalin in some cases, but considering the overall level of weed control observed, these differences would likely have little practical significance.

Further, as control was generally greatest at the lower irrigation volumes, increasing irrigation volume to “move herbicide through the mulch” would likely provide little improvement in weed control, given that label directions are followed. Addition of PRE herbicides becomes more important to overall weed control when lower mulch depths are used, and similarly, addition of mulch becomes more critical when the PRE herbicide is not as effective or troublesome weed species are present. Due to some negative effects observed with the combination, specifically with application of indaziflam to a 1 inch layer of PS, additional research is needed to determine herbicide sorption to various mulch materials. If mulch were applied at an adequate depth, this potential sorption would become less important as the mulch would be able to provide adequate weed control in many cases. Additional studies should also investigate efficacy of these and other combinations when seeds are placed on top of the mulch. If some weed species can germinate and grow within a mulch layer, herbicide binding to these materials could potentially offer greater control of weed seeds that are introduced over the course of a growing season.

In the landscape industry, mulch is usually used in landscape beds for both aesthetic purposes and weed control, and typically applied at 2 to 3 inches (Marble et al., 2015). Mulch is less commonly used in the nursery industry due to the increased cost and difficulty of application, but may be used in longer-term crops or for crops that

are sensitive to PRE herbicide applications (Case et al., 2005). In larger pots or pot-in- pot production systems, herbicide and mulch combinations may be used to extend the length of weed control (Mathers, 2003). Results from these studies indicate that in both the landscape and nursery industries, use of PRE herbicides, specifically spray applied formulations, will outperform common mulch materials over the short-term (~9 weeks).

While long-term weed control was not evaluated in these studies, mulch usually provides effective weed control for a greater period of time compared with a single herbicide application, thus, the benefits of using mulch from both a weed management and cost perspective increase over time (Bartley et al., 2017). As seeds were placed on the soil surface prior to mulching in our study, additional research is needed to determine if there are any synergistic or antagonistic effects of herbicide + mulch combinations over time. Over time, weed seeds would be continually introduced on-top of the mulch layer. If certain weed species had the ability to germinate within a layer of mulch, use of herbicides could provide additional benefits not captured in these experiments. Our data also suggests that weed control will likely not improve with additional overhead irrigation when herbicides and mulch are used concurrently. There does not seem to be any benefit to additional irrigation once the recommended irrigation volume is applied for each herbicide. However, more research is needed on additional herbicides with differing solubility characteristics and more mulch materials.

As would be expected, the use of herbicides became more advantageous when lower mulch levels were used as the lower mulch depth consistently provided less control than the higher depth. Similarly, the use of higher mulch depths became more important when granular herbicides were used (which were less effective), or when

eclipta was evaluated as it is difficult to control with preemergence herbicides alone. For eclipta and large crabgrass, acceptable weed control was rarely achieved with the use of mulch alone, but with garden spurge, over 85% control was achieved with HW and

PS at 2 inches. For the herbicide + mulch combinations, acceptable weed control was achieved for all three-weed species at the 2 inch depth when herbicides were used.

Based on these findings, there is an advantage for using PRE herbicides when mulch is applied at depths of 2 inches or less. As mulch depths of 3 inches or even 4 inches are sometimes used in the landscape and provide long-term weed control when used alone

(Chalker-Scott, 2007), it is unlikely that PRE herbicides would provide a meaningful benefit until the mulch began to decompose. In “real-world” scenarios, it is also important to consider that it is difficult to uniformly apply mulch at a certain level throughout a landscape bed or within a nursery container. Mulch materials will settle, move off target during rain or wind events, or may be displaced by countless other means. Use of PRE herbicides either soon before or after mulch application could provide additional assurance that a high level of weed control would be achieved in cases where mulch was applied non-uniformly. A topic that has not been studied adequately is the effects of mulch particle size on weed growth and determining what weed species have the ability to germinate and grow within mulch layers. For weed species that can grow within mulch layers, a preemergence herbicide application would be beneficial regardless of mulch depth.

Overall, it is likely that PRE herbicides will provide a significant improvement in weed control when mulch depths of 2 inches or less are used. In landscapes where mulch is applied at depths greater than 2 inches, PRE herbicide applications could

possibly be delayed until mulch materials began to degrade or weed control began to fail. In the nursery industry, mulch depths of 2 inches or less or most often used as it would be difficult and expensive to apply significantly higher depths inside a nursery container. At these lower mulch depths, growers could use PRE herbicides in combination with mulch in tolerant crops to increase weed control. In crops where herbicides could not be used, mulch would still provide a benefit and would likely be much more cost effective than handweeding (Case et al., 2005). In all cases, spray- applied herbicides will likely provide additional benefits over granular herbicides and additional irrigation is likely not warranted when using mulch in combination with herbicides. Additional research is needed to determine how mulch physical properties affect weed germination and growth and to determine how various herbicide + mulch combinations interact so that herbicides could be selected for use with specific mulch materials.

Table 2-1. Effect of herbicide formulation and post-treatment irrigation volume on control of three landscape/container nursery weed species. Formulationz Irrigation vol Control (%)x y (inches) Eclipta Large crabgrass Gargen spurge

Granular 0.5 59 82 89 1.0 50 78 84 2.0 41 71 77 L*w NSw NS Mean granular 50 bv 78 b 84 b Spray 0.5 71 92 99 1.0 65 98 98 2.0 77 91 94 NS NS NS Mean spray 71 a 94 a 97 a Irrigationu NS NS L* ANOVAt Formulation (F) 0.001 0.001 0.001 Irrigation vol (I) 0.152 0.211 0.034 F × I 0.004 0.344 0.590 zGranular and spray-applied formulations of indaziflam, prodiamine, and dimethenamid-P + pendimethalin were evaluated with eclipta, large crabgrass, and garden spurge, respectively. y1 inch = 2.54 cm. xCalculated as percent decrease based upon shoot fresh weights (FW) in non-treated, non-mulched control pots by using the formula: [(FW non- treated – FW treated) ÷ FW non-treated)] × 100; means are shown for pots treated with herbicide only. wL* represent a significant linear response (P ≤ 0.05) with respect to irrigation volume based on orthogonal contrasts; NS = not significant linear or quadratic response with respect to irrigation volume based on orthogonal contrasts. vMeans within species followed by the same letter are not significant different (P ≤ 0.05) based on ANOVA. uL* and NS = linear (P ≤ 0.05) or no significant response with respect to irrigation volume over both formulations. L* or NS show linear or not significant responses within each formulation where interactions are significant. tANOVA was performed using a mixed model analysis in SAS (version 9.4; SAS Institute, Cary, NC); effects are considered significant at P ≤ 0.05.

Table 2-2. Effect of mulch type and depth on control of three landscape/container nursery weed species. Control (%)z Mulch type Eclipta Large Garden spurge crabgrass Hardwood 64 ay 29 69 Pinebark 44 b 33 62 Pinestraw 46 b 36 73 Mulch depth (inches)x 1.0 38 b 19 b 53 b 2.0 65 a 47 a 83 a ANOVAw Mulch type (T) 0.002 0.545 0.267 Mulch depth (D) 0.001 0.001 0.001 T × D 0.301 0.879 0.782 zCalculated as percent decrease based upon shoot fresh weights (FW) in non-treated, non-mulched control pots by using the formula: [(FW non- treated – FW treated) ÷ FW non-treated)] × 100; means are shown for pots treated with mulch only. yMeans within a column and factor (type or depth) followed by the same letter or no letters are not significantly different based on Fisher's protected LSD (P < 0.05). x1 inch = 2.54 cm. wANOVA was performed using mixed model analysis in SAS (Ver. 9.4, SAS Institute, Cary, NC); effects are considered significant at P < 0.05.

Table 2-3. Control of three common landscape/nursery weed species with use of herbicide or mulch at two different depths. Control (%)z Treatment Eclipta Large crabgrass Garden spurge Herbicide Granular 50 cy 78 b 84 b Spray 71 ab 94 a 97 a Mulch Hardwood 1 inchx 45 cd 15 d 51 d 2 inches 83 a 43 c 87 ab Pinebark 1 inch 34 d 21 d 49 d 2 inches 59 bc 45 c 75 bc Pinestraw 1 inch 34 d 21 d 59 cd 2 inches 59 bc 45 c 87 ab zCalculated as percent decrease based upon shoot fresh weights (FW) in non-treated, non-mulched control pots by using the formula: [(FW non- treated – FW treated) ÷ FW non-treated)] × 100; indaziflam, prodiamine and dimethenamid-P + pendimethalin were evaluated as granular and spray applied formulations for eclipta, large crabgrass, and garden spurge, respectively. yMeans within a column followed by the same letter are not significantly different based on Fisher's protected LSD (P < 0.05). x1 inch = 2.54 cm.

Table 2-4. Response of eclipta, large crabgrass, and garden spurge to the main effects of herbicide formulation, mulch type and depth, and irrigation volume and all three way interactions. Eclipta Large crabgrass Garden spurge Main effects Control (%)y Herbicide Mulch Control (%)y Herbicide Mulch Control (%)y Herbicide Mulch effectx effectw effect effect effect effect

Formulation (F) <0.001 <0.001 <0.001 NAv NA NA 0.081 0.646 <0.001 Mulch type (T) <0.001 0.660 <0.001 0.655 0.002 0.017 0.137 <0.001 0.781 Mulch depth (D) <0.001 <0.001 <0.001 0.001 <0.001 0.257 0.043 <0.001 0.483 Irrigation vol. (I) 0.731 0.480 <0.001 0.606 0.943 <0.001 0.591 0.956 <0.001

Interactions F × T 0.002 0.008 <0.001 NA NA NA 0.090 0.848 0.749 F × D <0.001 <0.001 <0.001 NA NA NA 0.099 0.657 0.555 T × D 0.203 <0.001 0.001 0.099 0.048 0.843 0.128 <0.001 0.773 F × I 0.888 0.813 <0.001 NA NA NA 0.576 0.955 <0.001 T × I 0.007 0.281 0.382 0.762 0.698 0.313 1.656 0.955 0.989 D × I 0.487 0.241 0.003 0.704 0.923 0.409 0.604 0.958 0.939 F × T× D 0.144 0.077 0.001 NA NA NA 0.087 0.847 0.743 F × T× I 0.451 0.594 0.299 NA NA NA 0.645 0.955 0.988 F × D× I 0.280 0.818 0.007 NA NA NA 0.607 0.957 0.939 T × D× I 0.183 0.543 0.710 0.280 0.578 0.800 0.658 0.995 0.989 zFormulation refers to granular or spray-applied formulations of indaziflam, prodiamine, or dimethenamid-P + pendimethalin which were evaluated with eclipta, large crabgrass, and garden spurge, respectively. Mulch types evaluated included hardwood, pinebark, and pinestraw. Mulch depths included 1inch or 2 inches. Irrigation volumes included 0.5 inch, 1 inch, and 2 inches; 1 inch = 2.54 cm. ANOVA was performed using a mixed model in SAS (Ver. 9.4, SAS Institute, Cary, NC). Main effects and interactions are significant at P < 0.05. yCalculated as percent decrease in shoot fresh weights (FW) in pots that received the herbicide + mulch combination in comparison to pots receiving no herbicide or mulch treatment by using the formula: [(FW non-treated – FW treated) ÷ non-treated)] × 100. xCalculated as percent increase in control with the herbicide + mulch combination in comparison with mulch only treatments by using the formula: [(percent control combination - percent control mulch) ÷ percent control mulch] × 100.

wCalculated as percent increase in control with the herbicide + mulch combination in comparison with herbicide only treatments by using the formula: [(percent control combination - percent control herbicide) ÷ percent control herbicide] × 100. vNA = not analyzed. Data from spray-applied treatments in large crabgrass were not analyzed due to zero variance (all 100% control). Data relating to herbicide formulation, mulch type, and mulch depth and interactions are included in Table 2-5 and Fig. 2-2. Irrigation and interactions involving irrigation are included in Tables 2-6 to 2-8.

Table 2-5. Effect of herbicide formulation, mulch type and mulch depth on control of three landscape/container nursery weed species. Eclipta Large crabgrass Garden spurge

Formulationz Mulch type Depth Control Herbicide Mulch Control Herbicide Mulch Control Herbicide Mulch (inches)y (%)x effectw effectv (%) effect effect (%) effect effect

Granular Hardwood 1 70 bu 128 a 67 b 100*t 763* 32* 99 134 25

2 98 a 20 b 175 a 100* 140* 31* 99 16 25 Mean 84 A 74 A 121 A 100* 452* 32* 99 75 25

Pinebark 1 53 b 59 a 7 b 93 442 a 24 96 105 21

2 79 a 51 a 118 a 97 131 b 27 99 39 26 Mean 66 B 55 AB 63 B 95 269 A 26 A 98 72 23

Pinestraw 1 46 b 4 b -105 b 88 388 a 17 99 73 25 2 88 a 64 a 136 a 99 110 b 19 99 17 26

Mean 67 B 34 B 16 C 94 219 B 18 B 99 45 25

Mean granular 73 B 54 B 67 A 96 313 27 99 64 25 A

Spray Hardwood 1 79 b 196 a 20 a 100* 763* 8* 99 134 3 2 98 a 20 b 57 a 100* 140* 8* 98 16 4 Mean 88 A 108 A 39 A 100* 452* 8* 99 76 3

Pinebark 1 76 b 159 a 18 a 100* 412* 8* 98 113 4

2 92 a 78 b 46 a 100* 125* 8* 100 39 4

Mean 84 A 118 A 32 A 100* 286* 8* 99 76 4 Pinestraw 1 79 b 174 a 25 a 100* 331* 8* 99 74 4

Table 2-5. Continued Eclipta Large crabgrass Garden spurge

Formulationz Mulch type Depth Control Herbicide Mulch Control Herbicide Mulch Control Herbicide Mulch (inches)y (%)x effectw effectv (%) effect effect (%) effect effect

2 96 a 86 b 54 a 100* 106* 8* 100 16 4 Mean 87 A 130 A 40 A 100* 249* 8* 100 45 4 Mean spray 86 A 118 A 37 B 100* 329* 8* 100 66 4 B zFormulation refers to granular or spray-applied formulations of indaziflam, prodiamine, or dimethenamid-P + pendimethalin which were evaluated with eclipta, large crabgrass, and garden spurge, respectively. y1 inch = 2.54 cm. xCalculated as percent decrease in shoot fresh weights (FW) in pots that received the herbicide + mulch combination in comparison to pots receiving no herbicide or mulch treatment by using the formula: [(FW non-treated – FW treated) ÷ non-treated)] × 100. wCalculated as percent increase in control with the herbicide + mulch combination in comparison with mulch only treatments by using the formula: [(percent control combination - percent control mulch) ÷ percent control mulch] × 100. vCalculated as percent increase in control with the herbicide + mulch combination in comparison with herbicide only treatments by using the formula: [(percent control combination - percent control herbicide) ÷ percent control herbicide] × 100. uMeans within columns followed by different lowercase letter show differences in depth within each mulch type and formulation while different uppercase letters show differences in formulation or mulch type based on Fisher's protected LSD (P < 0.05). Means followed by no letters are not significantly different. Significant mulch type × mulch depth interactions across both formulations in large crabgrass and garden spurge are shown in Fig. 2-2. t*Shows data that were not analyzed due to zero variance.

Table 2-6. Effects of mulch type and irrigation volume on eclipta controlz with indaziflam + mulch combinations. Mulch type Irrigation vol (inches)y Hardwood Pinebark Pinestraw 0.5 84 82 75 1.0 86 75 75 2.0 89 69 81 NSx L**x NS zCalculated as percent decrease in shoot fresh weights (FW) in pots that received the indaziflam + mulch treatment in comparison to pots receiving no herbicide or mulch treatment by using the formula: [(FW non-treated – FW treated) ÷ non-treated)] × 100. y1 inch = 2.54 cm. xNS represents no significant response while L** indicates a significant (P < 0.01) linear response based on orthogonal contrast statements.

Table 2-7. Influence of indaziflam formulation, mulch depth, and irrigation volume on mulch effectz in eclipta. Mulch depth Formulation Irrigation vol (inches)y 1 inch 2 inches Granular 0.5 -17 62 1.0 -31 141 2.0 18 226 NSx L***x Spray 0.5 16 43 1.0 40 78 2.0 6 36 Q**x Q*** zCalculated as percent increase in control with the indaziflam + mulch combination in comparison with herbicide only treatments by using the formula: [(percent control combination - percent control herbicide) ÷ percent control herbicide] × 100. y1 inch = 2.54 cm. xNS, L, and Q represent no, linear, or quadratic response in mulch effect with respect to irrigation volume, respectively based on orthogonal contrasts; **, *** represent significant effects when P < 0.01 and 0.001, respectively.

Table 2-8. Influence of herbicide formulation and irrigation volume on mulch effect in garden spurge and large crabgrass. Mulch effectz Formulation Irrigation vol (inches)y Large crabgrass Garden spurge Granular 0.5 18 11 1.0 22 22 2.0 39 41 L***x L*** Spray 0.5 12 1 1.0 2 3 2.0 1 7 NAx L*** zCalculated as percent increase in control with the herbicide + mulch combination in comparison with herbicide only treatment by using the formula: [(percent control combination - percent control herbicide) ÷ percent control herbicide] × 100. Prodiamine was evaluated with large crabgrass and dimethenamid-P + pendimethalin was evaluated with garden spurge. y1 inch = 2.54 cm. xL*** indicates a significant (P < 0.001) linear response based on orthogonal contrast statements; NA = not analyzed due to zero variance.

Figure 2-1. Different particle sizes present in hardwood, pinebark and pinestraw mulch materials after screening through soil sieves. 1 mm = 0.0394 inch.

Figure 2-2. Influence of mulch type and mulch depth on the herbicide effect observed in large crabgrass (treated with prodiamine) and garden spurge (treated with dimethenamid-P + pendimethalin). Herbicide effect is the percent increase in control with the herbicide + mulch combination in comparison with control achieved with mulch only treatments and is calculated by using the formula: [(percent control combination - percent control mulch) ÷ percent control mulch] × 100. Means followed by the same letter within each species and mulch depth are not significantly different based on Fisher’s protected LSD (P < 0.05). NA* = not analyzed due to zero variance. 1 inch = 2.54 cm.

CHAPTER 3 RESPONSE OF WEED SEED EMERGENCE AND GROWTH TO DIFFERENT PHYSICAL PROPERTIES OF COMMON LANDSCAPE MULCH IN NURSERY CONTAINER PRODUCTION

Introduction

Weed control is an important aspect for container crop production as weeds can reduce crop growth and development by competing for nutrients, light, and water.

According to Bartley et al. (2017), the effect of weed competition is highly variable and depends on both weed and ornamental species. Many researchers have documented the detrimental effects of weeds on container-grown ornamentals such as eclipta

(Eclipta alba) on the container-grown Gumpo White Sport’azalea (Rhododendron eriocarpum) and large crabgrass (Digitaria sanguinalis) can interfere with the growth and development of container-grown ‘Gold Drop’bush cinquefoil (Potentilla fruticosa)

(Berchielli-Robertson et al., 1990; Fretz, 1972; Walker and Williams, 1989).

The most common method of weed control in containerized ornamental production is hand weeding and application of PRE herbicides. Hand weeding can be labor intensive, time consuming and expensive for growers. With the push for new immigration law reform, it is now more difficult to find long-term labor for weed control and other production tasks (Taylor et al., 2012). In 2003, Mathers reported that nurseries spend an estimated $500 - $4000/acre annually for hand weeding. Therefore, to avoid hand weeding, growers often greatly rely on application of PRE herbicides for an effective weed control. Direct application of herbicides can cause phytotoxic injuries to ornamentals leading to plant death and reduction in their market and aesthetic values. On an average, a 50 lbs. bag of granular preemergence herbicide costs approximately $80 and mostly preemergence herbicides are applied at their labeled

rates of 200 lb/acre such as granular formulations of indaziflam (Marengo G) or dimethenamid-P + pendimethalin (Freehand). So, for one application per acre, it would cost around $240 excluding labor cost (from a personal communication with Dr. Chris

Marble). For six applications per year in a 50 acre nursery would cost $72,000 only for the herbicide chemical (excluding the labor cost). Moreover, when broadcasting granular herbicides, there is almost 80% target loss as the herbicides may be deposited in the spaces between the containers (Gilliam et al., 1992). Out of $72,000, only approximately $14,000 is actually utilized within the container for the weed control and

$57,000 is wasted, or deposited on the soil surface. This off-target herbicide loss reduces profitability and consequently may lead to pesticide runoff and possibly groundwater or irrigation pond contamination (Pereira and Hostettler, 1993). Certain active ingredients have also been shown to cause phytotoxicity when residues are found in irrigation water (Samtani et al., 2007). Repeated use of the same PRE herbicides can lead to herbicide resistance in some weed species (Bartley et al., 2017).

Organic mulch material have been used as an alternative non-chemical weed control method in landscape and nursery container industries. Tree derived mulch materials have provided effective long-term weed control in nursery production

(Richardson et al., 2008; Wilen et al., 1999). In landscape studies, long-term weed control has been reported by using tree-derived mulch materials compared to non- treated plots (Billeaud and Zajicek, 1989; Broschat, 2007; Greenly and Rakow, 1995).

Bartley et al. (2017) reported that mulch from red cedar (Juniperus virginiana), ground whole loblolly pine (Pinus taeda), chinese privet (Ligustrum sinense), and sweetgum

(Liquidambar styraciflua) at a depth of at least 2 inches provided long-term weed control

in nursery containers. In other container studies, organic mulch in combination with herbicides have proved to be effective for weed control. Douglas fir (Pseudotsuga menziesii) and mini pinebark nugget mulch materials in combinations with either acetochlor (Harness®; Monsanto, St. Louis, MO), flumioxazin (SureGuard®; Nufarm,

Alsip, IL) or oryzalin (Surflan®; United Phosphorous, King of Prussia, PA) provided acceptable long-term weed control in containers when compared with these herbicides alone (Case and Mathers, 2003). Mulches when applied at depths of 2 to 3 inches are most effective for weed control (Greenly and Rakow, 1995; Richardson et al., 2008).

Most of the studies conducted with mulch for weed control efficacy are based only on different mulch types (pinebark, wood chips) or evaluation of mulch depths. The results reported are inconsistent as none (or rarely) of the authors have studied the physical properties (particle size, water holding capacity, light transmittance) of the organic mulch materials. Teasdale and Mohler (2000) introduced the theoretical relationships among mulch mass, area index, height, cover, light extinction, and weed emergence. According to them, successful weed emergence through mulches was related to the capacity of seedlings to grow under limiting light conditions and around mulch materials as the obstructing elements. Mulch materials such as pinebark and hardwood chips can vary significantly in their physical properties depending on how they were manufactured. While mulch is an effective method for weed control and can outperform herbicides (Marble et al., 2017), data is needed to identify and quantify their physical characteristics that are needed for optimum weed control and how to use different types of mulch materials in the most effective manner. Developing guidelines in terms of physical properties (particle size, water holding capacity, light transmittance),

depths needed for different materials, and the most effective mulch types would provide information to the nursery and landscape industry. In addition to that, such guidelines can also provide information to bark/mulch manufacturers on how these materials should be processed for weed control. Generating data on what physical properties are needed for weed control would also allow growers the ability to choose the most cost- effective option in terms of material based on regional availability when they know the properties needed for weed control. Hence, the objective of this research was to determine the response of large crabgrass and garden spurge emergence and growth to different physical properties of common organic landscape mulch materials for nursery container production and for landscape use.

Materials and Methods

Outdoor Container Experiment

Research was conducted at the Mid-Florida Research and Education Center,

University of Florida, in Apopka, FL in Summer 2017 and repeated in Fall 2017. Round black plastic nursery containers [11 inches (diameter) × 9 inches (height) dimensions] were filled with pinebark: peat mix (Fafard®52 growing mix, SunGro Horticulture,

Agawam, MA) amended with 8 lb/yard3 controlled release fertilizer 15N-3.9P-9.9K

(Osmocote® Plus, Everris, Geldermalsen, Netherlands) based on the manufacturer’s recommended medium rate for incorporation. Square transparent plastic tubes 12 inches × 1.5 inches × 1.5 inches (Sinclair & Rush Inc., Arnold, MO) were inserted in the center of each container dividing it into two equal halves. Thirty five seeds of large crabgrass or garden spurge were surface sown to one-half of each container

(representing seeds below the mulch layer). Large crabgrass seeds were supplied by

Azlin Seed Services (Leland, MS) whereas garden spurge seed was collected from

natural populations present at the Mid-Florida Research and Education Center, Apopka,

FL. Following seeding, three different types of mulch materials including pinestraw [PS

(Pine Straw of Central Florida, Winter Garden, FL)], pinebark [PB (Timberline, Old

Castle Lawn & Garden Inc., Atlanta, GA)] or hardwood chips [HW (Florida Select™

Natural Eucalyptus Mulch, Scotts®, Marysville, OH)] were added on top of each container at depths of 0, 0.5, 1, 2, and 4 inches. Containers without mulch (control) were also included for comparison. Another 35 seeds of large crabgrass or garden spurge were sown to the surface of mulch layer on the other half of each container

(representing seeds above the mulch layer). All the containers were kept under full sun condition (natural outside light) and received 0.5 inch of irrigation per day via overhead impact sprinklers.

Data collection included biweekly weed counts and light intensity measurements of photosynthetic photon flux density (PPFD in µmol m-2 s-1) under the mulch layers at different depths using a light measuring sensor (LI-191R Line Quantum Sensor,

LICOR®, Inc. Environmental, Lincoln, NB). Light intensity was recorded bi-weekly by inserting the sensor into the transparent plastic tubes for 12 weeks after seeding (WAS).

Prior to insertion of the quantam sensor, ambiet light levels were first recorded for each block in order to calculate the percent decrease in light under the mulch in comparison with ambient (full sun) light levels. All data was collected for 12 weeks after trial iniation.

In this experiment, weeds were not removed after counting.The experiment was a randomized complete block design with four replications per mulch and depth combination, with factorial arrangement of three different mulch types, five different mulch depths and two weed seed placements (above and below the mulch).

All data were converted to percent reduction in either light or weed count in relation to the non-mulch controls prior to analysis and inspected to ensure the assumptions of ANOVA were met. Data were arcsine transformed when needed to satisfy the assumptions of ANOVA but back-transformed data is presented in all cases for interpretation. Data were then subjected to a mixed model analysis of variance

(ANOVA) using SAS® (SAS 9.4, SAS Institute, Inc., Cary, NC) with trial run and replication nested within trial run as random effects and all other variables (mulch type, depth, and seed placement) and interactions between these terms as fixed factors.

Fisher’s Protected Least Significant Difference (LSD) test was used to compare between individual means of experimental variables at the 0.05 probability level.

Orthogonal contrasts were used to test linear and quadratic response trends over mulch depth and interactions with mulch depth when significance was detected. Data for each weed species was analyzed separately.

Greenhouse Experiment

In addition to the above experiment, another greenhouse study was conducted in

2017 and 2018 at the Mid-Florida Research and Education Center, University of Florida, in Apopka, FL. Round black plastic nursery containers [7 inches (diameter) × 7 inches

(height) dimensions] were filled with substrate similar to the previous experiment. Plastic corrugated sheets (Highway Traffic Supply, Pearl River, NY) 5 inches × 6 inches were placed at the middle of each container. Twenty seeds of large crabgrass or garden spurge were sown on one-half of each container representing seeds below the mulch layer. Large crabgrass seeds were supplied by Azlin Seed Services (Leland, MS) whereas garden spurge seed was collected from the natural population present at the

Mid-Florida Research and Education Center, Apopka, FL. Then mulch materials

including PS (Pine Straw of Central Florida, Winter Garden, FL)], PB (Timberline, Old

Castle Lawn & Garden Inc., Atlanta, GA)] or HW (Florida Select™ Natural Eucalyptus

Mulch, Scotts®, Marysville, OH), were added on top of the soil in each container at depths of 0 inch, 0.5, 1, 2, and 4 inches. Another 20 seeds of either large crabgrass or garden spurge were sown above the mulch layers and these represents seeds above the mulch layers. Control sets with weed seeds but without any mulch layers were also included in the study for comparison. All the containers were placed inside a greenhouse with natural light where the temperature was maintained between 21 to 35 oC and received irrigation by overhead sprinklers of 0.2 inch total per day via two separate irrigation cycles, throughout the trial.

Weed counts were conducted bi-weekly. Immediately after weed counts, emerged weeds were pulled out once their species were identified correctly so as to remove weeds before seed was produced. Data was collected for 12 weeks and data were analyzed similar to that of the previous experiment.

Moisture Retention by Mulch Materials

In Spring 2018, round black plastic nursery containers [7 inches (diameter) × 7 inches (height) dimensions] were filled up with either PS or PB or HW. Soil moisture sensors (S-SMD-M005 soil moisture smart sensor, Onset®, Hobo®, Bourne, MA) were inserted one in each mulch-containing container and the data were recorded for 12 weeks. After 12 weeks, all data analyzed using SAS® (SAS 9.4, SAS Institute, Inc.,

Cary, NC).

Another experiment was conducted to measure the percent moisture retention by three different mulch materials. Two-piece plastic Buchner funnels (13.1 cm inner diameter, 6.6 cm tall) (Thermo Scientific, Waltham, MA) were filled with 2.5 inches of

either PB or PS or HW mulch materials. It was established that 2.5 inches layer of PB,

PS and HW weighed 157 g, 41 g and 75 g respectively. These weights were used to uniformly apply the same exact mass of each mulch material to replicate funnels. Each mulch-filled funnel was weighed (Wi) and placed over a 900 ml glass jar (Golden harvest mason jar, Ball Crop®, Broomfield, CO) and 171 ml of water (equivalent to 0.5 inch of irrigation) was applied uniformly to each of them. The water passing through the mulch layer were collected in the glass jar beneath. The funnels were weighed after the irrigation ceased and funnels stopped dripping (W0). The funnels were weighed again 1 h, 4 h, and 24 h after the irrigation event (W1, W4, and W24, respectively). The volume of water passing through the mulch and into the jar beneath was measured (V). The percent of water passing through the mulch layer in each funnel was calculated as [V ÷

(W0 – Wi + V)] × 100. The mass of water retained in the mulch layer at 1 h, 4 h, and 24 h was calculated as W1 - Wi, W4 - Wi, and W24 - Wi, respectively. Percent of water retained my the mulch at 1 h, 4 h, and 24 h was calculated by the formula: [(W1to 24 – Wi) ÷ Wi] ×

100. There were five replications per mulch material consisting of a complete randomize design. The data were pooled because there were no experimental run by treatment interactions. The experiment was repeated and the combined data were analyzed by

ANOVA using SAS (Ver. 9.4, SAS Institute, Cary, NC) and Fisher’s LSD test was performed to separate out the means.

Particle Size Analysis of Mulch Materials

One hundred g each of PS, PB, and HW mulch materials were subjected to rotating USA standard sieves (WS Tyler®, Mentor, Ohio) of sizes 25 mm, 19 mm, 12.5 mm, 9.5 mm, 6.3 mm, 3.35 mm, and 1 mm. The mulch residues at each sieve was collected, weighed and recorded separately. There were three replications per

treatment and the experiment was repeated two times.All weights were converted to percentages and were analyzed using ANOVA by SAS (Ver. 9.4, SAS Institute, Cary,

NC) and Fisher’s LSD test was performed to separate out the means.

Results and Discussion

Outdoor Container Experiment

The results from data captured with the quatum sensor showed that WAS was not a significant factor as light reductions underneath the mulch materials did not vary over the 12 weeks of the study, therefore, data were pooled together across all data collection periods. Previous research has shown that mulch materials such as HW, PB, and PS do degrade and decompose over time, but significant decomposition is typically not observed for 4 to 6 months (Duryea et al., 1999). Over the short duration of our study (12 weeks), there was not enough time to observe a significant rate of mulch decomposition and thus the ability of the mulch materials to block out light remained constant throughout. Mulch type was also found to be a non-significant factor (P =

0.0659). This suggests that at least over the short-term, these materials did not differ in their ability to reduce light levels when applied at similar depths. Mulch depth was significant in this study over the 12 weeks of monitoring (P < 0.001). Analysis showed that once a 1 inch mulch was added, over 99% of ambient light was eliminated (Table 3-

2) and no additional benefit was realized in terms of light reduction on the soil surface.

In the moisture retention study, HW retained the highest amount of applied irrigation for 24 hours (29%), followed by PS (22%) and then PB (14%) (Table 3-4). This indicates that HW would be the slowest mulch material to dry out after a rainfall or irrigation event while PB would dry the fastest. The results from the moisture retention study can be explained by examining the particle size analysis (Table 3-5). Differences

were detected at each sieve size, but by examining the percent of mulch particles present in the largest and smallest sieves, data shows that HW had the highest percentage of smaller particles while PB and PS, to a lesser degree, had a higher percentage of larger particle sizes. Porometer procedures which identify water holding capacity of different growth substrates shows that smaller particle materials will hold a higher percentage of water compared with substrates comprised of larger particle materials (Buamscha et al., 2007). Pinebark and PS drained more rapidly following the simulated irrigation due to a higher percentage of large particle materials. As higher moisture levels in soils often result in a greater incidence of weed seed germination, and the fact that critical moisture levels must be reached for some weed species to break dormancy (Pérez-Fernández et al., 2000), mulch materials that dry quickly would be expected to provide superior weed control.

In large crabgrass, the main effects of mulch type and mulch depth were significant but significant interactions were detected in, mulch × depth, depth × placement, and mulch × depth × placement (Table 3-1). Linear or quadratic trends were observed in all three mulch types when seeds were placed both below and above the mulch. With the exception of HW when seeds were placed above the mulch (a linear trend), quadratic trends were observed in all other cases, indicating a non-linear relationship between observed reduction in weed counts (control) and mulch depth. By examining the percent reduction in weed counts across the differement mulch materials, it is clear that once 1 inch of mulch depth was applied, increased weed control was generally not consistently observed, especially in the case where seeds were placed below the mulch (Table 3-1).

In garden spurge, the main effects of mulch type, depth, and seed placement were significant but no interactions were detected. The highest reduction in weed counts were observed in PB and PS (82 and 81%, respectively) while HW only reduced weed counts by 54% (Table 3-1). Overall three mulch types, a quadratic trend was detected for mulch depth, similar to the results observed in large crabgrass. Higher weed control (i.e. weed count reduction) was consistently observed when seeds were placed below mulch (73%) compared to seeds placed above (69%) (P < 0.001).

Greenhouse Experiment

In both large crabgrass and garden spurge, mulch type, depth, seed placement, and the three-way interaction between these terms was significant (Table 3-3). In large crabgrass, linear or quadratic trends were observed in all cases when seed were placed below the mulch. When seeds were placed above the mulch, no relationship was observed in HW or PS. In HW, an increase in weed counts was detected at the 0.5 and

4.0 inch depth while PS reduced weed counts at all depths between 67% and 92% but trends were still not detected. In the case of garden spurge, linear or quadratic trends were detected in all cases with the exception of HW when seeds were placed above the mulch, similar to the results observed with large crabgrass. Also similar to large crabgrass, an increase in weed counts was detected in HW mulch when seeds were placed above the mulch layer.

In cases where weed seeds have the ability to germinate within the mulch layer, mulch depth will likely be a non-significant factor, at least in cases when depths of only a few inches are applied. This has been observed previously with eclipta and spotted spurge in containers using depths ranging from 0.5 to 4 inches (Cochran et al., 2009).

The poor control resulting from HW, and to a lesser degree PS mulch when seeds were

placed above the mulch layer is due to the greater capacity of these materials to hold water compared with PB. Weed seeds will have different levels of soil moisture requirements in order to germinate, and when mulch provides a suitable environment, the seeds will likely germinate. While high rates of germination were dected in the HW mulch in these studies when seeds were placed above the mulch layer, studies with bittercress (Cardamine flexuosa) and rice hull mulch have shown that seed germination was reduced when seeds were placed above the mulch and germination increased in seeds placed below the mulch (Altland et al., 2016). However, in this study, Altland et al. (2016) also reported that greater moisture content was found within the soil compared to the mulch, and only lower depths were evaluated (only up to 1 inch).

In both outdoor and greenhouse experiments, weed counts tended to consistently respond either linearly or quadratically with mulch depth. It is important to note that in outdoor container experiments, emerged seedlings were not removed following data collection, thus both species reached the reproductive stage during the

12 weeks these studies were conducted. Both large crabgrass and garden spurge produce copius amounts of seeds in a relatively short timeframe, and seeds of these species have no dormancy requirement (Neal and Derr, 2005). After 8 weeks, seedlings resulting from mature, reproductive emerged plants initially seeded had begun to emerge, and so even when seeds were placed below mulch initially, seeds were introduced on top of the mulch at later evaluation dates. The ability of mulch to reduce weed counts when seeds were placed above mulch is likely related to the water holding capacity of the mulch materials. Both PB and PS consistently reduced weed counts from seeds placed both above mulch in both experiments, and both of these materials

also dried more rapidly than HW. In contrast, the depth of HW mulch had no effect on seedling emergence in the greenhouse trials when seeds were placed above the mulch because the small particle material of the HW resulted in an environment suitable for seed germination.

Based upon results of these experiments, particle size is likely the most important factor in determination of the weed suppressive ability of mulch materials, if equal depths are applied. In many previous studies, authors only report weed control achieved with different mulch materials (e.g. wood chips, pine straw, pine bark nuggets, etc.) but omit the particle size analysis from the discussion. All of these organic mulch materials can be processed in many different ways, and this processing and the resulting particle size distribution present in the mulch will likely have a profound influence on the ability of the mulch to control weed growth.

Weed control was consistently highest when weed seeds were placed below the mulch layer, similar to findings with mulberry weed (Fatoua villosa)

(Penny and Neal, 2003), bittercress (Cardamine spp.) (Richardson et al., 2008), and eclipta (Cochran et al., 2009). When seeds are placed below the mulch, the depth of the mulch will become a more important factor as higher depths, or at least depths above

1.0 inches, will both reduce light levels and provide a physical barrier to emergence. For both species, weed counts tended to overwhelmingly respond in a non-linear, quadratic manner. This indicates that once a certain level of mulch is applied, likely approximately

1.0 inch, greater weed suppressive ability will not be achieved through reduction of more light or through the addition of a greater physical barrier. However, many weed species have the ability to germinate and grow through mulch applied at much higher

depths, especially in the case of perennial species such yellow nutsedge (Cyperus esculentus) (Chen et al., 2013). Therefore, higher depths may be needed for other weed species, or may be used in addition to preemergence herbicides for more thorough control (Marble et al., 2015).

Based on the results of these experiments, nursery growers and landscape contractors should chose mulch materials that contain a high percentage of large (> 25 mm) particles so that moisture inside the mulch layer can be lost as soon as possible after rainfall or irrigation events. Mulch type tended to not be a significant factor in weed control as control was more closely correlated to the mulch particle size and subsequent water holding capacity. While greater weed control was not consistently achieved at depths above 1.0 inches, it should be noted that these experiments were short-term in nature and mulch could be applied very carefully and thoroughly in all cases to ensure that the materials were applied at precisely the level we indicated. In a nursery or landscape scenario, mulch will typically need to provide weed control for a longer period of time, and it is not feasible to apply it as accurately as could be done in a small plot experiment. Therefore, depths of at least 2 inches could be applied to ensure adequate coverage once mulch settles and is partially displaced by foot traffic, rainfall, or other environmental factors. Applying mulch at depths of at least 2 inches would also allow more time to pass in between subsequent mulch applications to maintain mulch at needed levels once the mulch begins to decompose over time (Duryea et al., 1999).

Table 3-1. Mulch type, depth, and seed placement effects on emergence of large crabgrass (Digitaria sanguinalis) and garden spurge (Euphorbia hirta) in outdoor container experiments. Large crabgrass Mulch type Hardwood Pinebark Pinestraw Mean depth

Depth (inches) Abovez Belowy Above Below Above Below

Decrease (%)x 0.5 39 16 58 70 44 26 42 1.0 62 55 88 97 83 81 77 2.0 69 98 99 100 100 100 94 4.0 73 99 93 100 100 99 94 L*w Q*** Q*** Q** Q*** Q*** NA Mean placement 61 67 85 92 82 76 Mean type 64 88 79 Garden spurge 0.5 31 30 52 68 57 63 50 1.0 40 62 84 88 76 79 71 2.0 57 76 88 91 90 93 83 4.0 61 72 92 93 96 98 85 NA NA NA NA NA NA Q** Mean placement 47 60 79 85 80 83 Mean type 54bv 82a 81a

ANOVAu Large crabgrass Garden spurge Mulch type (T) <0.001 <0.001 Mulch depth (D) <0.001 <0.001

Table 3-1. Continued. ANOVAu Large crabgrass Garden spurge Seed placement 0.1785 <0.001 (P) M × D 0.0316 0.4339 M × P 0.1253 0.1553 D × P 0.0327 0.7662 M × D × P 0.0345 0.8499 zAbove indicates data where seeds of each species were placed above the mulch layer. yBelow indicates data where seeds of each species were placed below the mulch layer. xDecrease (%) was calculated based upon percentage decrease in cumulative weed counts observed in comparison with a non-mulched control over 12 weeks. wNS, L, and Q represent no, linear, or quadratic response in mulch depth, respectively, based on orthogonal constrasts; *, ** and *** represent significant effects with P < 0.05, 0.01, and 0.001, respectively. NA = not analyzed due to significant interactions. vMeans followed by the same letter within a row are not significantly different according to Fisher's protected LSD (P < 0.05). uAnalysis of variance was performed using a mixed model to test for significance of main effects and interactions. Effects are considered significant at P < 0.05.

Table 3-2. Influence of mulch type and depth on percent reduction in light (PAR, µ mol m-2 s-1) in outdoor container experiments over 12 weeks. Mulch type Reduction (%)z Hardwood 98.5 Pinebark 97.1 Pinestraw 97.5

Depth (inches) 0.5 91.2 by 1.0 99.7 a 2.0 99.9 a 4.0 99.9 a zCalculated as percent decrease in light in pots that received the mulch materials in comparison to pots receiving no mulch treatment by using the formula: [(non-treated – treated) ÷ non-treated)] × 100. xMeans within a column followed by the same letter are not significantly different based upon Fisher’s Protected LSD (P < 0.05).

Table 3-3. Mulch type, depth, and seed placement effects on emergence of large crabgrass (Digitaria sanguinalis) and garden spurge (Euphorbia hirta) in greenhouse experiments. Large crabgrass Mulch type Hardwood Pinebark Pinestraw Mean depth

Depth (inches) Abovez Belowy Above Below Above Below

Decrease (%)x 0.5 -6 56 27 81 67 82 51 1.0 22 96 74 94 84 95 78 2.0 16 95 88 97 84 100 79 4.0 -7 99 78 98 92 100 76 NS*w Q** Q*** L* NS Q** NA Mean placement 6 87 67 93 82 95 Mean type 46 80 88 Garden spurge 0.5 9 55 40 72 54 64 49 1.0 -15 92 60 95 90 97 70 2.0 -12 99 83 99 99 100 78 4.0 6 100 91 100 99 100 83 NS Q*** L*** Q*** Q*** Q*** NA Mean placement -3 87 68 91 85 90 Mean type 42 80 88

ANOVAv Large crabgrass Garden spurge Mulch type (T) <0.001 <0.001 Mulch depth (D) <0.001 <0.001

Table 3-3. Continued. ANOVAv Large crabgrass Garden spurge Seed placement <0.001 <0.001 (P) M × D 0.3937 0.0716 M × P <0.001 <0.001 D × P 0.6261 0.0986 M × D × P 0.0289 0.0013 zAbove indicates data where seeds of each species were placed above the mulch layer. yBelow indicates data where seeds of each species were placed below the mulch layer. xDecrease (%) was calculated based upon percentage decrease in cumulative weed counts observed in comparison with a non-mulched control over 12 weeks. Negative values indicate a percent increase. wNS, L, and Q represent no, linear, or quadratic response in mulch depth, respectively, based on orthogonal constrasts; *, ** and *** represent significant effects with P < 0.05, 0.01, and 0.001, respectively. NA = not analyzed due to significant interactions. vAnalysis of variance was performed using a mixed model to test for significance of main effects and interactions. Effects are considered significant at P < 0.05.

Table 3-4. Percent of water retention by three different mulch materials. Percent water retentionz Mulch type 0 h 1 h 4 h 24 h Hardwood 32.3 ay 32.1 a 31.3 a 29.1 a Pinestraw 25.5 b 25.2 b 24.0 b 21.9 b Pinebark 16.5 c 16.1 c 15.5 c 14.0 c zCalculated as percentage of water retained in mulch after 0 h, 1 h, 4 h, and 24 h by using the formula [(W0 to 24 – Wi) ÷ Wi] × 100. Where W0, W1, W4, and W24 are the mulch weights after application of irrigation at 0 h, 1 h, 4 h, and 24 h respectively and Wi is the initial weight of the dry mulch materials before application of irrigation. y Means within a column followed by the same letter are not significantly different based upon Fisher’s Protected LSD (p < 0.05).

Table 3-5. Percentages of particle sizes present in three different mulch materials. Sieve size Mulch Initial total 25 mm 19 mm 12.5 mm 9.5 mm 6.3 mm 3.35 mm 1 mm Pan types weight (g) Hardwood 100 11.0 bz 6.9 b 13.9 a 12.3 a 15.2 a 17.1 a 12.9 a 8.5 a Pinebark 100 51.6 a 18.0 a 16.3 a 6.4 b 5.8 b 1.4 c 0.2 b 0.2 c Pinestraw 100 56.6 a 1.4 b 3.3 b 2.9 c 6.4 b 9.9 b 12.5 a 1.6 b zMeans within a column followed by the same letter are not significantly different based upon Fisher’s Protected LSD (p < 0.05).

CHAPTER 4 ASSESSING HERBICIDE MOVEMENT THROUGH MULCH MATERIALS TO IMPROVE WEED CONTROL

Introduction

Mulching is one of the most effective methods of weed control in landscape planting beds. Alternatively, use of preemergence (PRE) herbicides in combination with mulch may be used to provide longer-term weed control and suppression of different weed species that may not be controlled with PRE or mulch alone. Use of herbicide and mulch combinations is also often recommended to reduce labor costs associated with hand weeding, repeated application of POST-applied herbicides such as glyphosate, or both (Wilen and Elmore, 2007). There are many different PRE herbicide active ingredients labeled for use in residential and commercial landscapes, and most are available in different formulations or are commonly sold as combination products or premixes containing two or more active ingredients. Many different types of mulch are also available and vary widely in chemical and physical properties.

As there are many choices available for PRE herbicides and mulch types, it is not clear which herbicide is best suited for different types of mulch. Another important factor to consider is how management practices and application procedures could potentially be altered depending on mulch type and herbicide to achieve the best weed control. It has been well established that soil type and organic matter content can have a dramatic effect on herbicide behavior (Carter, 2000; Weber, 1990) and efficacy (Blumhorst et al.,

1990). Therefore, it is important to determine which group(s) of herbicides or formulations are best suited for different mulches to achieve the best weed control.

Different mulch materials will generally interact differently with various herbicides

(Marble, 2015). The efficacy and duration of soil-applied herbicides is dependent on its

soil-activity, that is, how the herbicide binds or interacts with soil particles based on the herbicides solubility and other chemical properties (Banks and Robinson, 1984). Knight et al. (2001) evaluated movement of isoxaben, metolachlor, and pendimethalin when applied to mulches composed of pinestraw, pinebark, or recycled newspaper pellets.

Results showed that newspaper pellets absorbed more herbicide (57% to 82% retention) than did pinestraw (34% to 88%) or pinebark (37% to 83% retention) with metolachlor being absorbed less than any other herbicide in all mulch materials.

Increasing levels of plant residue on the soil surface in conservation tillage systems have shown reduction in herbicide efficacy (Buhler, 1992). These mulch, or residue materials, usually decrease efficacy by binding and intercepting the herbicide and inhibiting it from reaching the soil surface (Banks and Robinson, 1986; Chauhan and Abugho, 2012) or by increasing microbial activity and thereby increase the speed of degradation (Locke and Bryson, 1997). Banks and Robinson (1984) studied the effects of oryzalin applied to straw-covered and nonmulched soils. The amount of oryzalin reaching the soil surface was reduced in presence of straw at the time of herbicide application, and the soil oryzalin concentration declined as the amount of straw increased. Crutchfield et al. (1986) investigated the effects of metolachlor (a common

PRE herbicide labeled for landscape use) when applied to mulch showed that although soil concentrations of metolachlor were lower when applied in the presence of wheat straw mulch, weed control increased with the increasing mulch level. This could be due to the weed suppression ability of the mulch material and because metolachlor is highly mobile (Sanchez-Martin et al., 1995).

Chauhan and Abugho (2012) investigated the use of rice residue mulch with pendimethalin and oxadiazon, which are common landscape PRE herbicides. In their study, rice residue was applied to the soil surface of pots which were filled with field soil at rates of 0, 5.3, and 10.6 g/pot. Pots were then treated at 0, 0.5, and 1.0 Kg ai/ha of oxadiazon or 0, 1.0, and 2.0 Kg ai/ha of pendimethalin using spray formulations (making applications on top of mulch). Overall data suggested that some weed seedlings may be able to survive herbicide treatment in the presence of mulch, which acts to intercept herbicide. However, Chauhan and Abugho (2012) noted that additional studies were required with more weed species.

Herbicide leaching and runoff can cause off-target movement of herbicides which can be a major concern for nursery growers as run-off contaminants are regulated and monitored in many cases. Location of container nurseries in relation to surface water, possible ground water contamination, and the potential plant injury from recycled runoff water are the three main concerns regarding herbicides in nursery runoff water (Gillam et al., 1992). While application of mulch can be used as an effective weed management tool alone (Richardson et al., 2008), application of mulch can also reduce runoff and leaching of PRE herbicides both in nursery environments and in the landscape. In landscape studies, herbicide leaching was reduced by 35 to 74% when herbicides were applied along with mulch in comparison to herbicides applied alone (Knight et al., 2001).

It is important to determine how different mulch materials affect herbicide leaching and runoff after application (Marble, 2015). It is known that many different landscape mulch materials reduce soil erosion and slow water runoff (Chalker-Scott,

2007), and different mulch materials have been shown to reduce herbicide movement

and off-target contamination (Fawcett et al., 1994). According to Mathers (2001), combining herbicides with mulch materials or use of PRE herbicides in areas that are already mulched would allow the herbicides to be adsorbed to the mulch, causing the herbicide to be released slowly into the soil or media. This would not only decrease the amount of herbicide needed but also the amount leached from each container.

However, it is unknown if the efficacy of PRE herbicides is reduced when they are applied to mulch. The objective of this research was to assess herbicide movement through organic mulch to determine the most appropriate mulch type for use with PRE herbicides.

Materials and Methods

Bioassay

Research was conducted at the Mid-Florida Research and Education Center,

Apopka, FL in summer 2018. Square black plastic nursery containers [4 inches (width) ×

4 inches (length) × 5 inches (height) dimensions] were filled with a pinebark: peat

(Fafard®52 growing mix, SunGro Horticulture, Agawam, MA) amended with 8 lb/yard3 controlled release fertilizer 15N-3.9P-9.9K (Osmocote® Plus, Everris, Geldermalsen,

Netherlands) based on the manufacturer’s recommended medium rate for incorporation.

Pinestraw [PS (Pine Straw of Central Florida, Winter Garden, FL)], pinebark mini- nuggets [PB (Timberline, Old Castle Lawn & Garden Inc., Atlanta, GA)], or hardwood chips [HW (Florida Select™ Natural Eucalyptus Mulch, Scotts®, Marysville, OH)] were then applied at a depth of 2 inches on top of each container. Liquid formulations of indaziflam (Marengo® 0.622 SC, Bayer Crop Science, Research Triangle Park, NC), prodiamine (Barricade® 4 FL, Syngenta Crop Protection, Greensboro, NC), and dimethenamid-P + pendimethalin (Tower® 6 EC + Pendulum® 3.3 EC, BASF Corp.,

Research Triangle Park, NC) were then applied to the mulched containers with a CO2 backpack sprayer (Bellspray R&D sprayer Inc., Opelousas, LA) calibrated to deliver 20 gal/acre using an 8004 flat-fan nozzle (TeeJet Technologies, Wheaton, IL) at 30 psi. on

June 06, 2018 (run 1) and July 12, 2018 (run 2) at their labeled rates. A separate group of nontreated pots were maintained for each herbicide and mulch combination. All containers were placed on a full sun container nursery pad and received 1.4 inches of irrigation via two irrigation cycles through overhead sprinklers over 3 days (0.46 inches per day). Following the irrigation and three days after treatment, mulch was carefully removed from each pot so that only the herbicide reaching the soil surface was available for weed control and the presence of mulch did not confound results.

Following mulch removal, 20 seeds of either large crabgrass, garden spurge or eclipta were surface sown to the previously mulched pots that were treated with prodiamine, dimethenamid-P + pendimethalin, or indaziflam, respectively. Large crabgrass seeds were supplied by Azlin Seed Services (Leland, MS) whereas eclipta and garden spurge seed were collected from populations present at the Mid-Florida Research and

Education Center, Apopka, FL. The experiment was a completely randomized design with six replicates per herbicide and mulch combination.

Data collection included weed counts at 2 and 4 weeks after treatment (WAT). At

4 WAT, all weed species were cut at the soil line and shoot fresh weights were determined for each weed species. Shoot fresh weights (FW) were converted to percent control by using the formula [((FW Nontreated control – FW treated) / FW nontreated control) × 100]. All percent control data were subjected to a mixed model analysis of variance (ANOVA) after meeting the assumptions of normalility using SAS (Ver. 9.4,

SAS Institute, Cary, NC). Mulch types were the fixed effects and the different experimental runs were the random effects. Fisher’s Least Significance Difference Test was used to compare between individual means of experimental variables. All differences were considered significant at P < 0.05 and each weed species was analyzed separately.

Chemical Assay

In addition to the bioassay described previously, chemical assays were performed to quantify herbicide movement through pinebark mulch. Pine bark mulch was chosen for these experiments, as it is the most commonly used mulch material in both container nurseries and in landscapes in Florida. Square black plastic nursery containers [4 inches (width) × 4 inches (length) × 5 inches (height) dimensions] were filled with substrate and amendments as previously described. Pine bark mulch was then applied at a depth of 2 inches on top of each container. Liquid formulations of indaziflam, prodiamine, and dimethenamid-P + pendimethalin were applied as described previously, and pots were irrigated in the same manner as the bioassay experiment. Following treatment and 1.4 inches of irrigation, mulch was removed carefully without disturbing the underlying substrate. The substrate was then sampled to a depth of 1 inch out from each container. Quantification of each herbicide in the soil layer underneath mulch was determined using previously described methods (EPA,

1996; EPA, 2007; EPA, 2018). All herbicide quantification data were converted to percent retention by PB mulch by using the formula [((No mulched control – herbicide treated) / no mulch control) × 100] to determine which herbicide was more tightly bound by the PB mulch. There were two runs of experiments each consisting of a completely

randomize design with three replications for each treatment. Data were analyzed as described previously.

Results and Discussion

Bioassay

When indaziflam was applied to pots initially mulched with PB (89% control) and

PS (99% control), eclipta control was similar to that of nonmulched pots (100% control), indicating that these two mulch types had no detrimental effect on indaziflam efficacy of eclipta (Table 4-1). Only 67% eclipta control was observed in pots originally mulched with HW, which indicates that indaziflam was more tightly bound to this mulch, or was intercepted to a greater degree with HW compared with PB or PS. Large crabgrass data showed that PB (65% control) was the only mulch type that caused a significant reduction in prodiamine efficacy. Prodiamine provided similar large crabgrass control when pots were originally mulched with HW (80% control), PS (91%), and when no mulch was present. The combination of dimethenamid-P + pendimethalin provided similar control of garden spurge when it was applied to bare soil and pots originally mulched with PS, which both treatments resulting in 100% control. Pots that were originally mulched with either HW or PB (94 to 95% control) provided commercially acceptable control, but to a less degree than was observed in no mulched pots or pots originally mulched with PS.

Chemical Assay

Averaged over all three herbicide treatments (four active ingredients) to determine the over all effect, PB reduced the amount of herbicide reaching the soil surface by 85% (Table 4-2). Approximately 20% of pendimethalin, prodiamine, and indaziflam that was applied reached the soil surface and was detected using chemical

assay (Table 4-3). This indicates that 10 to 20% of the herbicide that was applied was available for weed control on the soil surface. More dimethenamid-P reached the soil surface (30%) than any other herbicide, with only 69% being retained by the PB mulch.

Similar to previous findings, herbicides evaluated in these experiments provided a high level of control of each bioassay weed species when applied to the soil surface when no mulch was present at the time of application (Marble et al., 2011; Johnson,

1997). When mulch was present during the application, results differed by herbicide as some herbicides will bind tighter to organic matter than others. Pinestraw was the only mulch material in which control of the target weed species was similar to control achieved when no mulch was present across all herbicide treatments and weed species evaluated. This indicates that the herbicides evaluated in this study may move more effectively through PS compared with HW or PB mulch.

Results from this study show that many of the most commonly used PRE herbicides in container nurseries are bound in organic mulch materials. Only 10% to

30% of the herbicide that was applied reached the soil surface, at least following only approximately 1.5 inches of irrigation over a short time period. As most herbicide labels indicate that 0.25 inch to 0.5 inch of irrigation is needed to water in herbicides following application, more irrigation may be needed with PRE herbicides applied to mulched nursery containers or landscape beds. While only a small portion of the total herbicide applied reached the soil surface, in most instances commercially acceptable weed control resulted over a 4-week evaluation period. It is unknown, however, how efficacy would be affected over a longer period and or if seeds were sown on multiple dates.

While this data shows a high degree of herbicide binding to mulch, use of PRE herbicides to mulched containers or landscape beds would still offer significant advantages. In many cases, weed germination and growth significantly increase when seeds are placed on top of mulch compared to seeds below mulch (Richardson et al.,

2008). Therefore, herbicide that is retained in the mulch layer would be available to prevent growth from weed seeds introduced on top of the mulch layer that may germinate/grow within that mulch layer. Weed control from these different herbicide + mulch combinations (when mulch was left on the soil surface) was not determined in the current study, but these combinations have been shown to provide season long weed control in previous studies (Bartley et al., 2017; Somireddy, 2012). While this data shows that PS may be the most compatible mulch for use with the PRE herbicides evaluated here, more data is needed to evaluate long-term control with this combination in a variety of environments. Cost, aesthetics, availability, and consumer acceptance should also be considered when evaluating mulch either with or without the use of PRE herbicides.

Table 4-1. Percent control of three herbicides applied to different mulch materials. Weed species Herbicide Mulch typesz Percent controly Eclipta Indaziflam Hardwood 67 bx Pinebark 89 a Pinestraw 99 a No-mulch 100 a

Large crabgrass Prodiamine Hardwood 80 ab Pinebark 65 b Pinestraw 91 ab No-mulch 100 a

Garden spurge Dimethenamid-P + Hardwood 94 b pendimethalin Pinebark 95 b Pinestraw 100 a No-mulch 100 a zLiquid formulations of each herbicide (or combination) were to applied to pots mulched with hardwood, pinebark, pinestraw, or contained no mulch. Two days after application, mulch was removed prior to bioassay. yCalculated as a percent decrease in shoot fresh weights (FW) in pots receiving no herbicide or mulch treatment by using the formula: [((FW nontreated control – FW treated)/ FW nontreated control) × 100)]. zMeans within a column followed by the same letter are not significantly different based upon Fisher’s Protected LSD (p < 0.05).

Table 4-2. Average amount of herbicide detected in soil samples following application to pots mulched with pinebark and those containing no mulch. Mulch type Herbicide detected (mg/Kg)z Pinebark 49.4 by

No-mulch 321.7 a zHerbicides applied were indaziflam, prodiamine, and dimethenamid-P + pendimethalin. The amount of herbicide presented in this table is the average of all four herbicides. yMeans within a column followed by the same letter are not significantly different based upon Students t- test (p < 0.05).

Table 4-3. Percent of preemergence herbicides retained by pinebark. Mulch type Herbicides Retention in mulch (%)z Pinebark Pendimethalin 88 ay Prodiamine 84 a Indaziflam 80 a Dimethenamid-P 69 b zCalculated as a percent decrease in herbicide amount in soil samples receiving no mulch treatment by using the formula: [((no mulch control - treated)/ no mulch control) × 100)]. yMeans within a column followed by the same letter are not significantly different based upon Fisher’s Protected LSD (p < 0.05).

CHAPTER 5 ALLELOPATHIC EFFECTS OF COMMON LANDSCAPE AND NURSERY MULCH MATERIALS ON WEED CONTROL

Introduction

The landscape industry represents a diverse network of service companies contributing over $54 billion in sales in the U.S. (Hodges et al., 2011). Weed management in non-turf areas of residential and commercial landscapes is primarily achieved through application of organic mulch materials that serve as both a weed management tool and provide aesthetic value (Marble, 2015). Materials such as pinebark, pinestraw, hardwood chips from various plant species and other, sometimes inorganic mulches (i.e. gravel or stone) are commonly used due to their low cost and/or availability, and consumer preferences (Chalker-Scott, 2007). Mulch is known to suppress weed growth through light exclusion, by creating a physical barrier, or by reducing available moisture to seeds within mulch layers (Chalker-Scott, 2007).

Potential mechanisms of control that have been under-investigated for landscape mulch are allelopathic compounds.

Research has shown that mulch primarily inhibits weed growth through light exclusion (Wesson and Wareing, 1967; Popay and Roberts, 1970; Fitter and Hay, 1987;

Richardson et al., 2008), creation of a physical barrier (Crutchfield et al., 1986; Facelli and Pickett, 1991; Marble, 2015), and reducing available water within mulch layers

(Jordan et al., 2010). While physical characteristics and depth of mulch often explain efficacy in regards to weed control (Chalker Scott, 2007), allelopathic properties present in some mulch materials may also inhibit weed growth in certain instances.

Allelochemicals are diverse in chemical structure and produced by plants as secondary metabolites (Paiva, 2000; Hadacek, 2002). They are released by root

exudation, volatilization, and death and decay of plants, through leaching from living or decaying residues (Rice, 1984; Anaya, 1999). For example, catmint (Nepeta × faasennii) is an ornamental groundcover which has several secretory glands on the abaxial surfaces of its leaves. This plant releases volatile mixture over the time which contains numerous secondary products, three of them are related to high concentrations of nepetalactones which are potent inhibitors of seedling growth and germination in enclosed bioassays (Inderjit at al., 2005). Toxicity of these allelochemicals are determined by several factors including concentration, flux rate, age and metabolic state of the plant, and prevailing climate and environmental conditions

(Kohli et al., 1993; Wardle et al., 1993; Weidenhamer, 1996; Gallet and Pellissier, 1997;

Nilsson et al., 1998).

Allelopathic production and quantity varies within plant species, cultivar, age, plant organ, and time of the year (Argandona et al., 1981; Hanson et al., 1981; Wyman-

Simpson et al., 1991; Devi et al., 1997; Burgos et al., 1999; Cambier et al., 2000;).

Allelochemicals produced by some species may be toxic enough to lead to death of others (Bewick et al., 1994). Rietveld et al. (1983) reported decline and death of

European black alder trees (Alnus glutinosa) due to black walnut (Juglans nigra) allelopathy. This resulted from a combination of high walnut biomass causing significant release of juglone to the environment and wet soil that restricted aerobic metabolism by soil microorganisms, allowing juglone to build up to toxic levels.

100

Maiden), pinebark {splash pine (Pinus elliottii) and loblolly pine (P. taeda)}, pine needle

(Pinus elliottii), melaleuca (Melaleuca quinquenervia), and a utility-trimming mulch

(GRU) composed of multiple species [oaks, Quercus laurifolia and Q. rubra, Q. virginiana, and cherry (Prunus serotina)], with a small amount of southern redcedar

[Juniperus virginiana var. silicicola (Small) Silba] and southern pines (Pinus spp.)].

Bioassays were conducted by extracting water-soluble chemicals from the mulches followed by application to germinating lettuce (Lactuca sativa) seeds and germinants for each mulch extract were recorded. Results showed hydroxylated aromatic compounds were highest in GRU and lowest in melaleuca, pinebark, and pinestraw, but all showed levels of significant activity in bioassay (Duryea et al., 1999). The authors hypothesized allelopathic properties of these mulches could potentially reduce germination of landscape weed species, but were not evaluated.

There remains a significant knowledge gap concerning quantification of potential allelochemicals present in the common landscape mulch materials. Although research has focused on allelopathic properties of various agronomic crop residues and cover crops and their effect on weed suppression or potential as natural herbicides/herbicide templates (Weston, 2005), these materials would not be suitable in landscapes due to rapid decomposition, availability, and appearance (Marble, 2015). Therefore, the objective of this research was to assess the effects of allelopathic properties of common landscape and nursery mulch materials on weed control efficacy.

Materials and Methods

Experiments were conducted two times in Spring 2019 at Mid-Florida Research and Education Center, Apopka, FL. The mulch extraction was prepared by following the modified EPA 1312 synthetic precipitation procedure. Mulch materials including

101

pinestraw [PS (Pine Straw of Central Florida, Winter Garden, FL)], pinebark mini- nuggets [PB (Timberline, Old Castle Lawn & Garden Inc., Atlanta, GA)], or hardwood chips [HW (Florida Select™ Natural Eucalyptus Mulch, Scotts®, Marysville, OH)] were crushed and ground to < 9.5 mm in size. Extraction fluid was prepared adding the 60/40 weight percent mixture of concentrated sulfuric (Fisher Scientific International, Inc.,

Hampton, NH) and nitric (Fisher Scientific International, Inc., Hampton, NH) acids to deionized ASTM type II water (LabChem Inc., Zelienople, PA) until the pH was 4.2 +

0.05. Following the extraction fluid preparation, 1 g of each crushed mulch material was transferred to the borosilicate extraction bottle [1 inch diameter, 3.5 inches length

(dimensions)]. Twenty ml of the extraction fluid was added to each of the extraction bottle containing 1 g of mulch material. The extraction bottle was closed tightly and was secured in rotary extractor device (custom built), and was rotated at 30 + 2 rotation per minute (rpm) for 18 h. After 18 h extraction, materials in extractor bottles were separated out into their liquid and solid phases by filtering through the glass fiber filter of pore size 0.7 µm (Whatman™, GE Healthcare, Life Sciences, Chicago, IL). The liquid extract was collected separately for PS, PB, and HW mulch materials.

Each of these extracts (2 mL) was added to the germination paper in the 47 mm diameter perti dish (Fisher Scientific International, Inc., Hampton, NH). Twenty seeds of either large crabgrass or garden spurge were placed in each 47 mm petri dish and were sealed with paraflim (Bemis Company Inc., Neenah, WI). Petri dishes were then placed in constant temperature of 24 oC for 9 d. Large crabgrass seeds were supplied by Azlin

Seed Services (Leland, MS) and garden spurge seed were collected from naturally occurring populations present at the Mid-Florida Research and Education Center,

102

Apopka, FL. A control was also included where the weed seeds were sown on the germination paper wetted by double distilled water instead of mulch extract.

Data collection included number of germinants, root and shoot lengths of each germinated weed species for each mulch extract at 9 d after treatment (DAT). The experiment was a completely randomize design with 4 replications per treatment in each trial repetition. As no trial by treatment interactions were detected with an initial analysis of variance (ANOVA), results from both experimental runs were pooled together for anlaysis. Data were subjected to ANOVA and Dunnett’s test at the 0.05% significance level was used to calculate least significant differences for each mulch extraction in relation to the water control using SAS (Ver. 9.4, SAS Institute, Cary, NC).

Each weed species was analyzed separately.

Results and Discussion

No differences were observed in large crabgrass germination among any of the mulch treatments in relation to the control. Average percent germination for large crabgrass was 20, 21, 22, and 26% in the PB, PS, HW, and control treatments, respectively. Average shoot and root length of large crabgrass was similar to or greater than the non-treated control for all three mulch types (Table 5-1). Hardwood mulch resulted in an increase in shoot and root length in comparison to the control while seed treatment with PB or PS extract had no effect.

In garden spurge, only PS mulch resulted in a decrease in germination in relation to the control. Germination rates of 49, 54, and 55% were observed for garden spurge seed treated with extracts from PB, HW, and the control, respectively, while only 38% germination was observed for seed treated with the PS extract. Average shoot and root

103

length of seed treated with any of three mulch types was similar to or greater than the non-treated control.

Based upon these results, the mulch materials evaluated in this trial will have little to no effect on control of large crabgrass or garden spurge from an allelopathic standpoint. However, while shoot and root length of garden spurge was not adversely affected by mulch extracts, germination did decrease significantly when garden spurge seed were treated with PS mulch extract (Table 5-1). In contrast to our results, several studies have noted allelopathic properties in mulch materials, albiet these results were observed with materials different from the mulch materials used in our studies.

Rathinasabapathi et al. (2005) demonstrated that water eluates from wood chips of southern redcedar, red maple (Acer rubrum), swamp chestnut oak (Quercus michauxii), neem (Azadirachta indica), and magnolia (Magnolia grandiflora) inhibited radicle growth in germinating lettuce. It was also observed that eluates from wood chip mulch of neem, swamp chestnut oak, and red cedar inhibited the hypocotyl growth. While alleopathic properties have been noted in some other studies, in our experiment there was little or no allelopathic effects of HW, PB, or PS mulch materials on large crabgrass and garden spurge. Only PS extract showed significant decrease in germination of garden spurge.

Reasons that few or no alleopathic effects were noted in our study is likely related to both the method of extract that was used and also the weed species that were evaluated. In most of these previous studies, only water or methanol extracts from plant materials has been used for identifying allelopathic properties. Vrchotová et al. (2011) reported allelopathic properties of Impatiens species by comparing water, methanol, and dichloromethane extracts. Results showed that the methanol extract was most

104

inhibitory and phytotoxic to the rapeseeds (Brassica napus) in comparison to the water and dichloromethane extracts. This clearly shows that toxicity of the allelochemicals also depend on the nature of the solvent used. Methanol is an organic solvent which can have significant reactions with the natural allelochemicals present within the mulch materials. However, in our research, mulch extracts were prepared by following the modified EPA 1312 synthetic precipitation procedure which mimics leaching under acidic rainfall condition and more applicable to a natural setting of landscape/nursery conditions compared with water eluates or methonal extracts of mulch materials.

Quality of irrigation water, chemical and physical properties of soil/soilless substrates used in landscape/nursery container production, precipitation or wind movement, and the microbial activity in the substrates can interfere or interact with the allelochemicals released mulch. In addition, allelopathic production and quantity will vary within plant species, cultivars, plant age, plant organ, and time of the year that the mulch (or other biomass material) is harvested (Argandona et al., 1981; Hanson et al.,

1981; Wyman-Simpson et al., 1991; Devi et al., 1997; Burgos et al., 1999; Cambier et al., 2000;). Precipitation or wind may cause leaching of foliage and associated allelochemicals into the surrounding soil/substrate (Inderjit et al., 2005).

When identifying allelochemicals, sensitive plant species such as lettuce or radish or rape seeds have been used for previous allelopathic studies (Duryea et al.,

1999). Little research has been done by using weed seeds. Kato-Noguchi et al. (2009) reported allelopathic properties of red pine (Pinus densiflora) needles that inhibited root growth of weed species such as cress (Lepidium sativum), ryegrass (Lolium multiflorum), timothygrass (Phleum pretense), large crabgrass and barnyardgrass

105

(Echinochloa crus-galli) by 9% to 18% and shoot growth of these species by 20% to

65%. Increasing the concentration of the pine needle extract concentration increased the inhibitory effects on these weed species.The aqueous methanol extract of pine needles was purified, and the main inhibitory substance was isolated and determined by spectral data as 9α,13β-epidioxyabeit-8 (14)en-18-oic acid (Kato-Noguchi et al., 2009).

In our experiment, two of the most common landscape/nursery weeds species i.e. large crabgrass and garden spurge were used instead of lettuce or radish. The mulch extracts used in this trial could possibly have had a negative inhibitory effect on other sensitive plant seeds, but this was not evaluated. So, plant species is also an important factor to consider because allelopathic effects will vary depending on species that is evaluated.

It is also important to note that in many experiments evaluation alleopathic properties, authors may extract and isolate pure forms of a particular alleochemical for testing (Vrchotová et al., 2011) in order to determine if the particular chemical has alleopathic effects. However, the amounts of allelochemicals that seeds are exposed to may be an order of magnitude higher than would be encountered in a landscape or nursery environment where these alleochemicals would be released slowly at minute quantities over time. Maimoona et al. (2001) demonstrated that bark of chir pine (Pinus roxburghii Sarg.) and Bhutan pine (Pinus wallichiana A.B. Jacks) contained catechin and gallocatechin derivative, quercetin, kaempferol, secoisolariresinol, 3, 4- dihydroxybenzoic acid, and rhamnetin. Harman-Ware et al. (2016) reported presence of monoterpenes, α- and β-pinene, camphene, and δ-carene in the terpenoids extract of loblolly pine saplings and pine lighter wood. In most of these previous allelopathic studies, authors have tried to identify and isolate specific chemical compounds from

106

plant extracts. However, in natural settings, the joint action of chemicals in a mixture is potentially more important in explaining allelopathic interference (Inderjit et al., 2005).

The concentration of each chemical in a mixture might be significantly reduced than the concentration of an individual chemical required to induce an inhibitory effect on higher plant growth and seed germination (Blum, 1996). Designing allelopathic studies for common landscape mulch materials would be challenging because the results obtained must be (ideally) transferable to the natural landscape/nursery settings. According to

Inderjit et al. (2005) there are two main issues related to both laboratory and field research conducted on allelopathy. First, utilization of inadequate methodology with respect to chemical extraction and identification of compounds, and secondly, difficulties in relating results from bioassay studies to the vegetation patterning in the natural field.

Another major problem encountered in allelochemicals research is related to the absence of an appropriate methodology that can separate the effects of allelopathic compounds from other factors, because allelochemicals are prone to chemical oxidation and microbial degradation in the environment, allelochemicals do not proliferate over time, and allelopathic activity is often due to the joint action of chemicals in a mixture

(Williamson, 1990).

These experiments utilizing bioassay were preliminary studies conducted to identify the allelopathic properties of mulch materials on weed seeds. There remains a significant knowledge gap concerning identification and quantification of particular/mixture of allelochemicals present in these landscape mulch material extracts and hence more research is required in this area. A better understanding of potential allelopathic effects of these mulch materials could be used by mulch manufacturers for

107

promotion and to aid the horticulture industry in selecting mulch for different applications

(Saha et al., 2018).

108

Table 5-1. Response of large crabgrass (Digitaria sanguinalis) and garden spurge (Euphorbia hirta) seeds to extracts of three different common landscape mulch materials. Large crabgrass Mulch type Germinants Shoot length (mm) Root length (mm) Pinebark 3.9z 4.3 8.0 Pinestraw 4.1 4.5 7.0 Hardwood 4.4 5.8* 13.3* Control 5.1 4.2 6.8 Garden Spurge Pinebark 9.8 3.8* 10.1* Pinestraw 7.6* 4.0* 11.1* Hardwood 10.8 3.7* 10.8* Control 10.9 3.2 7.5 zDunnett’s test for least significant difference: means within a column and weed species followed by an asterisk differ from the mean of the control (p < 0.05).

109

CHAPTER 6 CONCLUSIONS

Mulch is often applied in landscape planting beds for weed control, but little research has focused specifically on mulch and PRE herbicide combinations. The first experiment was performed to determine efficacy of herbicide + mulch combinations and which factors significantly affected weed control including herbicide formulation and post-treatment irrigation volumes. Additional objectives were to determine efficacy derived from mulch or herbicides used alone under herbicide + mulch combinations; and to identify differences in the additive (herbicide + mulch combinations) or singular

(herbicide or mulch) effects compared with use of herbicides or mulch only. Weed species including large crabgrass (Digitaria sanguinalis), garden spurge (Euphorbia hirta), and eclipta (Eclipta prostrata) were used as bioassay species for prodiamine, dimethenamid-P + pendimethalin and indaziflam efficacy, respectively. The experiment consisted of a factorial treatment arrangement of two herbicide formulations (granular or spray-applied), three mulch types [hardwood chips (HW), pinebark (PB), and pinestraw

(PS)], two mulch depths (1 inch and 2 inches), and three levels of one-time, post- treatment irrigation volumes (0.5 inch, 1 inch, and 2 inches). Three sets of controls were used: the first set included three mulch types applied at two depths receiving only 0.5 inch irrigation volume, the second set included only two herbicide formulations and three one-time irrigation volumes, while the last set received no treatment (no herbicide or mulch) and only 0.5 inch irrigation volume. High levels of large crabgrass and garden spurge control (88% to 100%) were observed with all herbicide + mulch combinations evaluated at mulch depths of 1 inch or greater. When comparing mulch types, the best eclipta control was achieved with hardwood at 2 inches depth. The spray formulation of

110

indaziflam outperformed the granular in most cases when used alone or in combination with mulch. Overall, results showed that spray formulations of prodiamine and dimethenamid-P + pendimethalin were more effective than granular when applied alone, while indaziflam was more effective as a spray formulation when used both alone and in combination with mulch. Increasing irrigation volume was not a significant factor for any of the herbicide + mulch combinations when evaluating overall weed control.

The second group of experiments included an outdoor container and a greenhouse experiment which were performed to determine the response of large crabgrass and garden spurge emergence and growth to different physical properties of common organic landscape mulch materials for nursery container production and for landscape use. The mulch material studied were PS, PB, and HW. Results showed that once a 1 inch mulch was added, over 99% of ambient light was eliminated and no additional benefit was realized in terms of light reduction on the soil surface. Hardwood retained the highest amount of applied irrigation for 24 hours (29%), followed by PS

(22%) and then PB (14%). Data showed that HW had the highest percentage of smaller particles while PB and PS, to a lesser degree, had a higher percentage of larger particle sizes. In the case of outdoor container experiment, with the exception of HW when large crabgrass seeds were placed above the mulch there was a non-linear relationship between observed reduction in weed counts (control) and mulch depth. Higher garden spurge control was consistently observed when seeds were placed below mulch (73%) compared to seeds placed above (69%). In the greenhouse experiment, for both large crabgrass and garden spurge, mulch type, depth, seed placement, and interactions between these terms was significant. In large crabgrass, linear or quadratic trends were

111

observed in all cases when seed were placed below the mulch. When seeds were placed above the mulch, no relationship was observed in HW or PS. In HW, an increase in weed counts was detected at the 0.5 and 4.0 inches depth while PS reduced weed counts at all depths between 67% and 92% but trends were still not detected. In the case of garden spurge, linear or quadratic trends were detected in all cases with the exception of HW when seeds were placed above the mulch. An increase in weed counts was detected in HW mulch when seeds were placed above the mulch layer.

The third group of experiments were conducted to assess herbicide movement through organic mulch materials including PB, PS, and HW. Weed species evaluated were large crabgrass, garden spurge, and eclipta. Liquid formulations of prodiamine, dimethenamid-P + pendimethalin, and indaziflam were evaluated in combination with mulch materials applied at a depth of 2 inches. Quantification of these herbicides was performed using biological and chemical assays from the soil samples collected from below the mulch layers. Results showed that only 67% eclipta control was observed in pots originally mulched with HW, which indicates that indaziflam was more tightly bound to this mulch. Large crabgrass data showed that PB (65% control) was the only mulch type that caused a significant reduction in prodiamine efficacy. Dimethenamid-P + pendimethalin efficacy on garden spurge was reduced in pots originally mulched with

HW or PB, but all treatments provided >94% control. Chemical assays showed that approximately 20% of pendimethalin, prodiamine, and indaziflam that was applied reached the soil surface when mulch was present during the application. More dimethenamid-P reached the soil surface than any other herbicide, with 69% being retained by the PB mulch.

112

The last experiment focused on assessing the effects of allelopathic properties of common landscape and nursery mulch materials on weed control efficacy. Large crabgrass and garden spurge weed seeds were subjected to germination in presence of mulch extracts of PB, PS, and HW. Number of germinants, root and shoot lengths of the seedlings were recorded. Results showed that the mulch materials evaluated in this trial will have little to no effect on control of large crabgrass or garden spurge from an allelopathic standpoint. However, while shoot and root length of garden spurge was not adversely affected by mulch extracts, germination did decrease significantly when garden spurge seed were treated with PS mulch extract. Designing allelopathic experiments are challenging because the results obtained must be suitable to the natural settings. Researchers most of the time face difficulties in utilization of inadequate methodology with respect to chemical extraction and identification of compounds, and difficulties in relating results from bioassay studies to the vegetation patterning in the natural field.

From this research it can be concluded that using common landscape mulch at a depth of 2 inches or greater in combination with spray-applied PRE herbicides can provide a high level of weed control both in landscapes and nursery container production. Mulch depth and particle size are more important considerations than mulch type, at least in the short term (3 to 4 months). Smaller particles mulch materials, such as shredded wood products similar to the HW used in our studies, may not be a suitable mulch material for all situations as it held a greater degree of moisture than PB or PS and the weed species evaluated here were able to germinate within that layer due to this moisture retention. Overall, as mulch depth increased, PRE herbicides contributed

113

less to the observed weed control, and conversely, as the mulch depth decreased, use of PRE herbicides became more critical when lower mulch depths were applied.

Activation irrigation does not need to be increased or decreased in order to move the mulch through the mulch layer to enhance control. However, this research showed the mulch layer may retain up to 80% of the applied herbicide. In our studies, this did not negatively impact weed control as the herbicide retained in the mulch layer often reduced emergence of the incoming weed seeds (i.e. seeds placed on top of the mulch layer). This is important, as these mulch materials generally applied a high level of control of seeds below the mulch layer. If it was known that the soil beneath the mulch layer contained a significant seed bank, use of pinestraw could be used as it was the most compatible mulch material with all PRE herbicides that were evaluated. Mulch materials evalued previously have shown that allelopathic chemicals are present, however, we found that in normal nursery or landscape scenarios, the allelopathic affect of these materials likely contributes little to none of the observed weed control.

114

LIST OF REFERENCES

Abu-Qare, A.W., and H.J. Duncan. 2002. Photodegradation of the herbicide EPTC and the safener dichlormid, alone and in combination. Chemosphere 46:1183-1189.

Achakzai, A.K.H., P. Achakzai, A. Masood, S.A. Kayani, and R.B. Tareen. 2009. Response of plant parts and age on the distribution of secondary metabolites on plants found in Quetta. Pak. J. Bot. 41:2129-2135.

Adams, D. 1990. Chemical weed control in containerized hardy ornamental nursery stock. Professional Hort. 4:70-75.

Altland, J.E., C.H. Gilliam, and G.W. Wehtje. 2003. Weed control in field nurseries. HortTechnology 13:9-14.

Altland, J.E., J.K. Boldt, and C.C. Krause. 2016. Rice hull mulch affects germination of bittercress and creeping woodsorrel in container plant culture. American J. Plant Sci. 7:2359-2375.

Altom, J.V. and D.S. Murray. 1996. Factors affecting eclipta (Eclipta prostrata) seed germination. Weed Technol. 10:727-731.

Amoroso, G., P. Frangi, R. Piatti, A. Fini, and F. Ferrini. 2010. Effect of mulching on plant and weed growth, substrate water content, and temperature in container- grown giant arborvitae. HortTechnology. 20:957-962.

Anaya, A.L. 1999. Allelopathy as a tool in the management of biotic resource in agroecosystems. Crit. Rev. Plant Sci. 18:697-739.

Argandona, V.H., H.M Niemeyer, and L.J. Corcuera.1981. Effects of content and distribution of hydroxamic acids in wheat on infestation by the aphid Schizaphis graminum. Phytochemistry 20:673-676.

Baker, J.M., W.C. Koskinen, and R.H. Dowdy. 1996. Volatilization of EPTC: Simulation and measurement. J. Environ. Qual. 25:169-177.

Balvinder, S., G.N. Gupta, and K.G. Prasad. 1988. Use of mulches in establishment and growth of tree species on dry lands. Indian For. 114:307-316.

Banks, P.A. and E.L. Robinson. 1984. The fate of oryzalin applied to straw-mulched and nonmulched soils. Weed Sci. 32:269-272.

Banks, P.A. and E.L. Robinson. 1986. Soil reception and activity of acetochlor, alachlor, and metolachlor as affected by wheat (Triticum aestivum) straw and irrigation. Weed Sci. 34:607-611.

115

Bartley, III., P.C., G.R. Wehtje, A.M. Murphy, W.G. Foshee, III., and C.H. Gilliam. 2017. Mulch type and depth influences control of three major weed species in nursery container production. HortTechnology 27:465-471.

Berchielli-Robertson, D.L., C.H. Gilliam, and D.C. Fare. 1990. Competitive effects of weeds on the growth of container grown plants. HortScience 25:77–79.

Bewick, T.A., D.G. Shilling, J.A. Dusky, and D. Williams. 1994. Effects of celery (Apium graveolens) root residues on growth of various crops and weeds. Weed Technol. 8:625-629.

Bilderback, T.E., S.L. Warren, J.S. Owen, and J.P. Albano. 2005. Healthy substrates need physicals too! HortTechnology 15:747-751.

Billeaud, L.A. and J.M. Zajicek. 1989. Influence of mulches on weed control, soil pH, soil nitrogen content, and growth of Ligustrum japonicum. J. Environ. Hort. 7:155- 157.

Black, J.N. 1956. The influence of seed size and depth of sowing on pre-emergence and early vegetative growth of subterranean clover (Trifolium subterraneum L.). Aust. J. Agr Res. 7:98-109.

Blum, U. 1996. Allelopathic interactions involving phenolic acids. J Nematol. 28:259- 267.

Blumhorst, M.R., J.B. Weber, and L.R. Swain. 1990. Efficacy of selected herbicides as influenced by soil properties. Weed Technol. 4:279-283.

Broschat, T.K. 2007. Effects of mulch type and fertilizer placement on weed growth and soil pH and nutrient content. HortTechnology 17:174-177.

Buamscha, M.G., J.E. Altland, D.M. Sullivan, D.A. Horneck, and J. Cassidy. Chemical and physical properties of douglas fir bark relevant to the production of container plants. HortScience 42:1281-1286.

Buhler, D.D. 1992. Population dynamics and control of annual weeds in corn (Zea mays) as influenced by tillage systems. Weed Sci. 40:241-248.

Burgos, N.R., R.E. Talbert, and J.D. Mattice. 1999. Cultivar and age differences in the production of allelochemicals by Secale cereale. Weed Sci. 47:481-485.

Burrows, M. 2017. Evaluation of pine bark mulch-herbicide combinations for weed control in nursery containers. Masters Thesis. Auburn, AL: Auburn University. https://etd.auburn.edu/handle/10415/5933. Accessed 9 July 2018.

Cambier, V., T. Hance, and E. Hoffman. 2000. Variation of DIMBOA and related compounds content in relation to the age and plant organ in maize. Phytochemistry 53:223-229.

116

Carter, A.D. 2000. Herbicide movement in soils: principles, pathways and processes. Weed Res. 40:113-122.

Case, L.T. and H.M. Mathers. 2003. Long term effects of herbicide treated mulches for ornamental weed control. Proc. Northeast. Weed Sci. Soc. 57:118-121.

Case, L.T., H.M. Mathers, and A.F. Senesac. 2005. A review of weed control practices in container nurseries. Weed Technol. 15:535-545.

Case, L.T. and H.M. Mathers. 2006a. Field evaluation of herbicide treated mulches. Proc. Southern. Nursery. Assn. Res. Conf. 51:402.

Case, L.T. and H.M. Mathers. 2006b. Herbicide-treated mulches for weed control in nursery container crops. J. Environ. Hort. 24:84-90.

Chalker-Scott, L. 2007. Impact of mulches on landscape plants and the environment – A review. J. Environ. Hort. 25:239-249.

Chauhan, B.S. and D.E. Johnson. 2008. Germination ecology of southern crabgrass (Digitaria ciliaris) and Indian crabgrass (Digitaria longiflora): Two important weeds of rice in the tropics. Weed Sci. 56:722-728.

Chauhan, B.S. and S.B. Abugho. 2012. Interaction of rice residue and pre herbicides on emergence and biomass of four weed species. Weed Technol. 26:627-632.

Chen, Y., R.E. Strahan, and R.P. Bracy. 2013. Effects of mulching and preemergence herbicide placement on yellow nutsedge control and ornamental plant quality in landscape beds. HortTechnology 23:651-658.

Cochran, D.R., C.H. Gilliam, D.J. Eakes, G.R. Wehtje, P.R. Knight, and J. Olive. 2009. Mulch depth affects weed seed germination. J. Environ. Hort. 2:85– 90.

Crutchfield, D.A., G.A. Wicks, and O.C. Burnside. 1986. Effect of winter wheat (Triticum aestivum) straw mulch level on weed control. Weed Sci. 34:110-114.

Derr, J.F. 1994. Weed control in container-grown herbaceous perennials. HortScience 29:25-27.

Devi, S.R., F. Pellissier, and M.N.V. Prasad. 1997. Allelochemicals, p. 253-303. In: Plant Ecophysiology, (ed. M.N.V. Prasad). John Wiley & Sons, New York.

Duryea, M.L., R.J. English, and L.A. Hermansen. 1999. A comparison of landscape mulches. J. Arboricult. 25:88–97.

EPA. 1996. Method 3540C. Soxhlet extraction. https://www.epa.gov/sites/production/files/2015-12/documents/3540c.pdf. Accessed 22 August 2018.

117

EPA. 2007. Method 8321B. Solvent-extractable nonvolatile compounds by high- performance liquid chromatography/ thermospray /mass spectrometry (HPLC/TC/MS) or ultraviolet (UV) detection. https://www.epa.gov/sites/production/files/2015-12/documents/8321b.pdf. Accessed 22 August 2018.

EPA. 2018. Method 8270E. Semivolatile organic compounds by gas chromatography/ mass spectrometry. https://www.epa.gov/sites/production/files/2017 04/documents/method_8260d_update_vi_final_03-13-2017_0.pdf. Accessed 22 August 2018.

Facelli, J.M., and S.T.A. Pickett. 1991. Plant litter: Its dynamics and effects on plant community structure. Bot. Rev. 57:1-32.

Fawcett, R.S., B.R. Christensen, and D.P. Tierney. 1994. The impact of conservation tillage on pesticide runoff into surface water: A review and analysis. J. Soil Water Conserv. 49:126-135.

Ferguson, J.F., B. Rathinasabapathi, and C.A. Chase. 2003. Allelopathy: How plants suppress other plants. Gainesville: University of Florida, Institute of Food and Agricultural Sciences HS944. Available online: http://edis.ifas.ufl.edu/hs186 Accessed 12 October 2017.

Ferguson, J., B. Rathinasabapathi, and C. Warren. 2008. Southern redcedar and southern magnolia wood chip mulches for weed suppression in containerized woody ornamentals. HortTechnology 18:266-270.

Fitter, A.H., and R.K.M. Hay. 1987. Environmental physiology of plants. 2nd ed. London: Academic Press.

Fraedrich, S.W. and D.L. Ham. 1982. Wood chip mulching around maples: Effect on tree growth and soil characteristics. J. Arboricult. 8:85-89.

Fretz, T.A. 1972. Weed competition in container grown Japanese holly. Hort Science 7:485-486.

Gallet, C., and F. Pellissier. 1997. Phenolic compounds in natural solutions of coniferous forests. J. Chem. Ecol. 23:2401-2412.

Gilliam, C.H., D.C. Fare, and A. Beasley. 1992. Nontarget herbicide losses from application of granular Ronstar to container nurseries. J. Environ. Hort. 10:175- 176.

Gorski, S.F. 1993. Slow-release delivery system for herbicides in container-grown stock. Weed Technol. 7:894-899.

118

Green, T.L. and G.W. Watson. 1989. Effects of turfgrass and mulch on establishment and growth of bare root sugar maples. J. Arboricult. 15:268-272.

Greenly, K. and D. Rakow. 1995. The effects of mulch type and depth of weed and tree growth. J. Arboricult. 21:225-232.

Hadacek, F. 2002. Secondary metabolites as plant traits: Current assessment and future perspectives. Crit. Rev. Plant Sci. 21:273-322.

Hanson, A.D., P.L.Traynor, K.M. Ditz, and D.A. Reicosky.1981. Gramine in barley forage-effects of genotype and environment. Crop Sci. 21:726-730.

Harman-Ware, A.E., R. Sykes, G.F. Peter, and M. Davis. 2016. Determination of terpenoids content in pine by organic solvent extraction and fast-GC analysis. Frontiers in Energy Res. doi:10.3389/fenrg.2016.00002.

Hodges, A.W., C.R. Hall, and M.A. Palma. 2011. Economic contributions of the green industry in the United States in 2007-08. HortTechnology 21:628-638.

Holm, L.G., D.L. Plucknett, J.V. Pancho, and J.P. Herberger. 1977. The world’s worst weeds. Distribution and biology. University Press of Hawaii, Honolulu, H.I.

Iles, J.K. and M.S. Dosmann. 1999. Effect of organic and mineral mulches on soil properties and growth of ‘Fairview Flame R’ red maple trees. J. Arboricult. 25:163-167.

Inderjit, A., L.A. Weston, and S.O. Duke. 2005. Challenges, achievements, and opportunities in allelopathy research. J. Plant Interact. 1:69-81.

Jhala, A.J. and M. Singh. 2012. Leaching of indaziflam compared with residual herbicides commonly used in Florida citrus. Weed Technol. 26:602-607.

Johnson, B.J. 1997. Sequential applications of preemergence and postemergence herbicides for large crabgrass (Digitaria sanguinalis) control in tall fescue (Festuca arundinacea) turf. Weed Technol. 11: 693-697.

Jordan, A., L.M. Zavala, and J. Gil. 2010. Effects of mulching on soil physical properties and runoff under semi-arid conditions in southern Spain. Catena 81:77-85.

Kato-Noguchi, H., Y. Fushimi, and H. Shigemori. 2009. An allelopathic substance in red pine needles (Pinus densiflora). J. Plant Physiol. 166:442-446.

Knight, P.R., C.H. Gilliam, S.L. File, and D. Reynold. 2001. Mulches reduce herbicide loss in the landscape. Proc. Southern Nursery Assn. Res. Conf. 46:461-463.

Kohli, R.K., D. Rani, and R.C. Verma. 1993. A mathematical model to predict tissue response to parthenin- an allelochemic. Biol. Plantarum 35:567-576.

119

Koski, R. and W.R. Jacobi. 2004. Tree pathogen survival in chipped wood mulch. J. Arboricult. 30:165-171.

Kraus, H.T. 1998. Effects of mulch on soil moisture and growth of desert willow. HortTechnology 8:588-590.

Langenheim, J.H. 1994. Higher plant terpenoids: A phytocentric overview of their ecological roles. J. Chem. Ecol. 20:1223-1280.

Litzow, M. and H. Pellett. 1993. Influence of mulch materials on growth of green ash. J. Arboricult. 9:7-11.

Locke, M.A. and C.T. Bryson. 1997. Herbicide-soil interactions in reduce tillage and plant residue management systems. Weed Sci. 45:307-320.

Lohr, V.I. 2001. Mulching reduces water use of containerized plants. HortTechnology 11:277-278.

Maimoona, A., I. Naeem, Z. Saddiqe, N. Ali, G. Ahmed, and I. Shah. 2011. Analysis of total flavonoids and phenolics in different fractions of bark and needle extracts of Pinus roxburghii and Pinus wallichiana. J. Med. Plants Res. 5:2724-2728.

Marble, S.C., C.H. Gilliam, G.R. Wehtje, A.J. Van Hoogmoed, and C. Palmer. 2011. Early postemergence control of spotted spurge in container production. J. Environ. Hort. 29:29-34.

Marble, S.C. 2015. Herbicide and mulch interactions: A review of the literature and implications for the landscape maintenance industry. Weed Technol. 29:341-349.

Marble, S.C., A.K. Koeser, and G. Hasing. 2015. A review of weed control practices in landscape planting beds: Part I – non-chemical weed control methods. HortScience 50:851-856.

Marble, S.C., A.K. Koeser, G. Hasing, D. McClean, and A. Chandler. 2017. Efficacy and estimated annual cost of common weed control methods in landscape planting beds. HortTechnology 27:199-211.

Martin, P., and L. Calvin. 2010. Immigration reform: what does it mean for agriculture and rural America? Appl. Econ. Perspect Policy 32:232-253.

Mary, B., S. Recous, D. Darwis, and D. Robin. 1996. Interactions between decomposition of plant residues and nitrogen cycling in soil. Plant Soil 181:71-82.

Mathers, H., E. Ozkan. 2001. Herbicide treated mulches. Nursery Mgt. Production. 17:61-64.

Mathers, H.M. 2003. Novel methods of weed control in containers. HortTechnology 3:28-34.

120

Mathers, H.M., J. Pope, and L.T. Case. 2004. Evaluation of micro-encapsulated herbicide alone and applied to bark mulches. Proc. Northeastern Weed Sci. Soc. 58:47.

Molisch, H. 1937. Origin and meaning of allelopathy, p. 1-7. In: Allelopathy 2nd ed. Academic Press, Orlando, FL.

Montague, T. and R. Kjelgren. 2004. Energy balance of six common landscape surfaces and the influence of surface properties on gas exchange of four containerized tree species. Scientia Hort. 100:229-249.

Naeth, M.A., A.W. Bailey, D.S. Chanasyk, and D.J. Pluth. 1991. Water holding capacity of litter and soil organic matter in mixed prairie and fescue grassland ecosystems of Alberta. J. Range. Mgt. 44:13-17.

Nagabhushana, G.G., A.D. Worsham, and J.P. Yenish. 2001. Allelopathic cover crops to reduce herbicide use in sustainable agricultural systems. Allelopathy J. 8:133- 146.

Neal, J.C., and J.F. Derr. 2005. Weeds of container nurseries in the United States. Raleigh, NC: North Carolina Assoc. of Nurserymen, Inc.

Neal, J.C., J.C. Chong, and J. Williams-Woodward. 2017. Southeastern US pest control guide for nursery crops and landscape plantings. NC State Extension Publication.

Nilsson, M.C., C. Gallet, and A. Wallstedt. 1998. Temporal variability of phenolics and batatastin III in Empetrum hermaphroditum leaves over an eight-year period: Interpretations of ecological functions. Oikos 81:6-16.

Paiva, N.L. 2000. An introduction to the biosynthesis of the chemicals used in plant- microbe communication. J. Plant Growth Regul. 19:131-143.

Penny, G.M., and J.C. Neal. 2003. Light, temperature, seed burial, and mulch effects on mulberry weed (Fatoua villosa) seed germination. Weed Technol. 17:213-218.

Pereira, W.E. and F.D. Hostettler. 1993. Nonpoint source contamination of the Mississippi river and its tributaries by herbicides. Environ. Sci. Technol. 27:1542- 1552.

Pérez-Fernández, M.A., B.B. Lamont, A.L. Marwick, W.G. Lamont. 2000. Germination of seven exotic weeds and seven native species in south-western Australia under steady and fluctuating water supply. Acta Oecologica 21:323-336.

Pickering, J.S., and A. Shepherd. 2000. Evaluation of organic landscape mulches: Composition and nutrient release characteristics. J. Arboricult. 24:175-187.

121

Popay, A.I., E.H. Roberts. 1970. Ecology of Capsells bursa-pastoris (L.) Medik and Senecio vulgaris L. J. Ecol. 58:123-139.

Putnam, A.R. 1988. Allelochemicals from plants as herbicides. Weed Technol. 2:510- 518.

Rathinasabapathi, B., J. Furguson, and M. Gal. 2005. Evaluation of allelopathic potential of wood chips for weed suppression in horticultural production systems. HortScience. 40:711-713.

Rice, E.L. 1984. Allelopathy. 2nd ed. Academic Press, Orlando, FL.

Richardson, B., C.H. Gilliam, G.B. Fain, G.R. Wehtje. 2008. Container nursery weed control with pine bark mini-nuggets. J. Environ. Hort. 26:144-148.

Rietveld, W.J. 1983. Allelopathic effects of juglone on germination and growth of several herbaceous and woody species. J. Chem. Ecol. 9:295-308.

Rooden, J.V., L.M.A. Akkermans, and R. Van Der Veen. 1970. A study on photoblastism in seeds of some tropical weeds. Acta Botanica Neerlandica 19:257-264.

Saha, D., C. Marble, N.S. Boyd, and S. Steed. 2016. Impacts of preemergence herbicide formulation on cost and weed control efficacy for container nursery crop producers. Proc. Intl. Plant Prop. Soc. 66:319-324.

Saha, D., S.C. Marble, and B.J. Pearson. 2018. Allelopathic effects of common landscape and nursery mulch materials on weed control. Front. Plant Sci. 9:733.

Samtani, J.B., G.J. Kling, H.M. Mathers, and L. Case. 2007. Rice hulls, leaf-waste pellets, and pine bark as herbicide carriers for container-grown woody ornamental. HortTechnology 17:289-295.

Sanchez-Martin, M.J., T. Crisanto, L.F. Lorenzo, M. Arienzo, M. Sanchez-Camazano. 1995. Influence of leaching rates on 14C metolachlor mobility. Bull. Environ. Contam. Toxicol. 54:562-569.

Santos, S.A.O., C. Vilela, C.S.R. Freire, C.P. Neto, and A.J.D. Silvestre. 2013. Ultra- high performance liquid chromatography coupled to mass spectrometry applied to the identification of valuable phenolic compounds from Eucalyptus wood. J. Chromatogr. B 938:65-74.

Sibley, J.L., D.M. Cole, and W. Lu. 2004. Waste is a terrible thing to mind. Proc. Intl. Plant Prop. Soc. 54:596-603.

Skroch, W.A., M.A. Powell, T.E. Bilderback, and P.H. Henry. 1992. Mulches: Durability, aesthetic value, weed control, and temperature. J. Environ. Hort. 10:43-45.

122

Somireddy, U. 2012. Effect of herbicide-organic mulch combinations on weed control and herbicide persistence. Ph.D. dissertation. Columbus, OH: Ohio State University https://etd.ohiolink.edu/. Accessed 9 July 2018.

Smith, D.R., C.H. Gilliam, J.H. Edwards, J.W. Olive, D.J. Eakes, and J.D. Williams. 1998. Recycled waste paper as a non-chemical alternative for weed control in container production. J. Environ. Hort. 16:69-75.

Stewart, C.L., N. Soltani, R.E. Nurse, A.S. Hamill, and P.H. Sikkema. 2012. Precipitation influences pre and post-emergence herbicide efficacy in corn. Scientific Research 3:1193-1204.

Taylor, J.E., D. Charlton, and A. Yunez-Nuade. 2012. The end of farm labor abundance. Appl. Econ. Perspect. Pol. 34:587-598.

Teasdale, J.R., and C.L. Mohler. 2000. The quantitative relationship between weed emergence and the physical properties of mulches. Weed Sci. 48:385-392.

Vrchotová, N., B. Será, and J. Krejčová. 2011. Allelopathic activity of extracts from Impatiens species. Plant Soil Environ. 57:57-60.

Walker, K.L. and D.J. Williams. 1989. Annual grass interference in container grown bush cinquefoil (Potentilla fruiticosa).Weed Sci. 37:73–75.

Wardle, D.A., K.S. Nicholson, and A. Rahman. 1993. Influence of plant age on the allelopathic potential of nodding thistle (Carduus nutans L.) against pasture grasses and legumes. Weed Res. 16:223-227.

Watson, G.W. 1988. Organic mulch and grass competition influence tree root development. J. Arboricult. 14:200-203.

Watson, G.W. and G. Kupkowski. 1991. Effects of a deep layer of mulch on the soil environment and tree root growth. J. Arboricult. 17:242-245.

Wauchope, R.D. 1987. Tilted-bed simulation of erosion and chemical runoff from agricultural fields: II. Effects of formulation on atrazine runoff. J. Environ. Quality 16:212–216.

Weber, J.B. 1990. Behavior of dinitroanaline herbicides in soils. Weed Technol. 4:394- 406.

Wesson, G., and P.F. Wareing. 1967. Light requirements of buried seeds. Nature (London) 21:600-601.

Wehtje, G.R., C.H. Gilliam, A.M. Murphy, and J. Fausey. 2015. Preemergence control of spotted spurge (Chamaesyce maculata) with flumioxazin as influenced by formulation and activation moisture. Weed Technol. 29:108-114.

123

Weidenhamer, J. 1996. Distinguishing resource competition and chemical interference: overcoming the methodological impasse. Agron. J. 88:866-875.

Weston, L.A. 1996. Utilization of allelopathy for weed management in agroecosystems. Agron. J. 88:860-866.

Weston, L.A. 2005. History and current trends in the use of allelopathy for weed management. HortTechnology. 15:529-534.

Wilen, C.H., U.K. Schuch, and C.L. Elmore. 1999. Mulches and subirrigation control weeds in container production. J. Environ. Hort. 17:174.

Wilen, C.A., and C.L. Elmore. 2007. Weed management in landscapes. Davis, CA: University of California Pest Notes No. 7441.7p.

Williamson, G.B. 1990. Allelopathy, Koch’s postulates and the neck riddles. In: Grace JB, Tilman D, editors. Prespectives on plant competition. New York: Academic Press. P 143-162.

Wilson, P.C., M.B. Riley, and T. Whitwell. 1995. Effects of ground cover and formulation on herbicides in runoff water from miniature nursery sites. Weed Sci. 43:671-677.

Wyman-Simpson, C.L., G.R. Waller, M. Jurzysta, J.K. McPherson, and C.C. Young. 1991. Biological activity and chemical isolation of root saponins of six cultivars of alfalfa. (Medicago sativa L.). Plant Soil 135:83-94.

Yang, Q. C.H. Gilliam, G.R. Wehtje, J.S. McElroy, J.L. Sibley, and J. Chamberlin. 2013. Effect of pre and post moisture level on preemergence control of hairy bittercress (Cardamine hirsuta L.) with flumioxazin. J. Environ. Hort. 31:49-53.

124

BIOGRAPHICAL SKETCH

Debalina Saha was born in Kolkata, West Bengal, India. She graduated in 2006 from Salt Lake School (English Medium) and immediately began her undergraduate career at the Presidency College, University of Calcutta, India. In 2009, Debalina Saha graduated with a Bachelor of Science degree with Honors in botany from the University of Calcutta, India.

Upon completion of her undergraduate studies, Debalina started her graduate studies and obtained Master of Science in botany from the University of Calcutta, India in 2011 with a concentration in plant physiology, biochemistry, and molecular biology.

Her master’s thesis was on “Establishment of Agrobacterium mediated transformation in

Solanum tuberosum”. Debalina got married to Somnath Saha in 2012 after she completed her master’s degree and moved to Bangalore, India.

In 2013, she earned her Bachelor of Education from Bangalore University, India.

Immediately she started working as a Subject Matter Expert in biology for the educational company, Fitkids Education and Training Pvt. LTD in 2014. While working at the company, Debalina realized that, she wanted to pursue her doctoral degree and started her Ph.D. program in horticultural sciences at the University of Florida, United

States, under the supervision of Dr. Chris Marble in August 2015.

In May 2019, Debalina graduated with a doctoral degree from the University of

Florida, United States and began working as an Assistant Professor in weed science for ornamental crop production at the Department of Horticulture, Michigan State

University, United States.

125