Management Strategies for Weed Suppression During Transition to Organic Agriculture

Management Strategies for Weed Suppression during Transition to Organic Agriculture

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Stephanie Lynne Wedryk

Graduate Program in Horticulture and Crop Science

The Ohio State University

2011

Dissertation Committee:

John Cardina, Advisor

Doug Doohan

Allison Snow

Nicholas Basta

Copyrighted by

Stephanie Lynne Wedryk

2011

Abstract

Concerns about public health and environmental quality due to the use of pesticides in conventional agriculture have driven increased demand for organic products.

Although growers have obtained higher prices and demand with organic products, many farmers are reluctant to transition to organic agriculture. Farmers view the challenge of weed management and risk of lower output as barriers to converting to organic production. The mandated three years before organic certification can be used to suppress weeds and improve soil fertility for enhanced yields in the first year of organic production. Smother cropping is an alternative strategy of weed management that uses living plants in monoculture or mixture to control weeds with the potential to improve soil fertility. Potential smother crops and smother crop mixtures, their effectiveness without chemical or mechanical management, mechanisms of suppression, and impacts on productivity under organic management are not fully understood. In this study, we investigated the use of smother cropping and associated transition strategies for weed suppression and productivity through 1) evaluation of smother crop species and mechanisms of weed suppression through a literature review; 2) determining the potential of using tef [Eragrostis tef (Zucc.) Trotter] and warm-season annual crop mixtures; 4) assessing smother crop planting dates; and 5) comparing mechanical and cropping-based organic transition strategies. ii

The results of this research indicate that crop growth and ancillary management practices are most important in determining the effectiveness of smother crops.

Exploitation of ecological niches in designing smother crop systems and targeting specific weeds can improve weed suppression. Tef can be used to suppress annual weeds under organic management, but is a weak competitor against Canada thistle [Cirsium arvense (L.) Scop]. In designing smother crop mixtures, the choice of grass species in mixture can affect biomass production. The effect of grass species in crop mixture dynamics may be related to height, morphology, spread, and aggressivity. Multi-species mixtures can increase ground cover by smother crops and reduce the cover of weeds, but are not more effective than monocultures in suppressing weed biomass.

Canada thistle is a particularly problematic weed for organic growers and planting smother crop mixtures when root carbohydrate reserves are at a seasonal nadir can improve suppression. Crop mixtures of warm-season, highly competitive crops were most effective at suppressing Canada thistle while a mixture of cool-temperature adapted species suppressed annual weed biomass. Smother cropping and the use of high-diversity prairie species as organic transition strategies were most suppressive of weed density and biomass after three years of transition. Compost application improved vegetable yields in the first organic year, while plant available nutrients had the greatest influence on potato yield and organic matter strongly affected tomato yields in comparison to other soil variables. Transition strategies before conversion to organic agriculture can influence productivity and weed populations. Smother cropping is a viable strategy for organic

iii transition, but the choice of crops and management must be carefully considered in order to realize optimal benefits.

Acknowledgments

I would like to thank my advisor, Dr. John Cardina, for his guidance with experimental design, data collection and analysis, and writing during my dissertation research. I would further like to thank my advisory committee, Dr. Doug Doohan, Dr.

Allison Snow, and Dr. Nick Basta for their guidance. I acknowledge Jennifer Moyseenko and Catherine Herms for their assistance with data collection and experimental management. I would further like to acknowledge Lynn Ault and K. Gregory Smith for their help with field operations. I am thankful for the assistance of Bert Bishop with statistical analysis. I am appreciative to the undergraduate students and visiting scholars at the OARDC who helped with data collection and entry. I would also like to thank my friends and family for their support and love throughout this journey. Last, I am grateful to the Department of Horticulture and Crop Science at The Ohio State University for allowing me to flourish, learn, and develop during my doctoral education. Thank you.

Vita

2007...... B.A. Biology, The College of Wooster

2007 to present ...... Graduate Research Associate, Department

of Horticulture and Crop Science, The Ohio

State University

Fields of Study

Major Field: Horticulture and Crop Science

Table of Contents

Abstract ...... ii

Acknowledgments...... v

Vita ...... vi

Fields of Study ...... vi

Table of Contents ...... vii

List of Tables…………………………………………………………………………..…ix

List of Figures ...... xiv

Chapter 1: Introduction ...... 1

Chapter 2: A Review of Smother Cropping and Associated Factors of Weed Suppression

...... 12

Chapter 3: Evaluation of Tef as a Smother Crop during Transition to Organic

Management ...... 61

Chapter 4: Summer Annual Crop Mixtures for Biomass Production and Weed

Suppression ...... 88

Chapter 5: Smother Crop Mixtures for Canada Thistle Suppression in Organic Transition

...... 118 vii

Chapter 6: Strategies for weed suppression and improvement of soil fertility during transition from conventional to organic vegetable production ...... 148

Conclusions ...... 182

Literature Cited ...... 185

viii

List of Tables

Table 2.1. Rating system used to evaluate publication quality…………………………..42

Table 2.2. Factors of weed suppression………………………………………………….43

Table 2.3. Description of smother cropping research and journals reviewed…………....45

Table 2.4. Locations of smother cropping research……………………………………...46

Table 2.5. Most commonly studied smother crops……………………………………....47

Table 2.6. Weed groups most commonly studied in research on smother crops………..49

Table 2.7. Effectiveness of categories of the factors of weed suppression………………50

Table 2.8. Rating and frequency of occurrence of most and least effective weed

suppression factors……………………………………………………………….51

Table 2.9. Highest and lowest rated smother cropping systems by species……………..53

Table 2.10. Most and least effective smother cropping systems that target specific

weeds……………………………………………………………………………..55

Table 2.11. Interaction of weed types and highest and lowest rated suppression

factors…………………………………………………………………………….59

Table 3.1. Seed color and source of smother crop varieties of tef and sorghum-

sudangrass………………………………………………………………………..80

Table 3.2. Final height of Canada thistle and smother crops and biomass of smother

crops in greenhouse trial. Canada thistle root pieces were planted at depths of 7.6

and 15 cm. Data are means of six replications and two runs of the

experiment………………………………………………………………………..83

Table 3.3. Monthly and 20-yr mean total precipitation and mean maximum and minimum

temperatures during the growing season…………………………………………85

Table 3.4. Biomass of tef varieties, Canada thistle, and other weeds harvested in field

trial, 2008 to 2009. Data are means of six replications each year……………….86

Table 3.5. Rate of vertical growth (cm d-1) and ground cover spread (% d-1) of tef,

Canada thistle, and annual weeds in field trial 2008 and 2009…………………..87

Table 4.1. Composition of summer annual crop mixtures and seeding rates…………..107

Table 4.2. Mean daily and 20 yr-mean maximum and minimum temperature during the

growing season………………………………………………………………….108

Table 4.3. Monthly and 20-yr mean total precipitation during the growing season……109

Table 4.4. Crop biomass and land equivalent ratio (LER) of summer annual crop

mixtures in 2008 and 2009……………………………………………………...110

Table 4.5. Percent cover of crops at harvest in summer annual crop mixtures in 2008 and

2009……………………………………………………………………………..113

Table 4.6. Equation parameters, adjusted R2, and P values for spread of weed cover over

growing degree days in 2008 and 2009. Data fit a sigmoidal logistic three

parameter curve (Equation 5)…………………………………………………...116

Table 4.7. Pearson correlation coefficients for response variables for grass, legume, and

forb crops in 2008 and 2009……………………………………………………117

Table 5.1. Varieties, seeding rates and depths of smother crop mixtures in 2009 and

2010……………………………………………………………………………..134

Table 5.2. Dates of planting and harvesting of smother crop mixtures in 2009 and

2010……………………………………………………………………………..135

Table 5.3. Monthly total precipitation in 2009 and 2010 and 20-yr mean……………..137

Table 5.4. Analysis-of-variance results for biomass of Canada thistle (CIRAR), annual

weeds, and crops and percent cover of crops in 2009 and 2010………………..138

Table 5.5. Canada thistle (CIRAR) and crop biomass in 2009 and 2010 at the early,

middle, and late planting dates and in cropping mixtures……………………...139

Table 5.6. Biomass of annual weeds in 2009 and 2010 and in each cropping treatment at

the early, middle, and late planting dates……………………………………….140

Table 5.7. Analysis-of-covariance results for Canada thistle (CIRAR) shoot density after

fall harvest and the following spring and percent cover at harvest in 2009 and

2010……………………………………………………………………………..141

Table 5.8. Canada thistle shoot density as affected by planting date and smother crop

mixture after biomass harvest in the fall of 2009 and the following spring in

2010……………………………………………………………………………..142

Table 5.9. Canada thistle shoot density as affected by the interaction of planting date and

smother crop mixture after biomass harvest in fall 2010 and the following spring

2011……………………………………………………………………………..143

Table 5.10. Percent cover of Canada thistle (CIRAR) and crops in the early, middle, and

late planting dates and in the cropping mixtures in 2009 and 2010…………….144

Table 5.11. Pearson correlation coefficients for measured response variables of Canada

thistle (CIRAR), annual weed, and crop biomass populations in 2009………...145

Table 5.12. Pearson correlation coefficients for measured response variables of Canada

thistle (CIRAR), annual weed, and crop biomass populations in 2010………...147

Table 6.1. Analysis of variance results of weed density and biomass in organic transition

strategies and compost applications in potato and tomato crops……………….170

Table 6.2. Density of CIRAR, monocot and broadleaf weeds in organic transition

strategies and compost applications in potato and tomato crops and total weed

biomass in tomato crop at harvest………………………………………………171

Table 6.3. Total biomass of weeds at harvest of potato in organic transition strategies and

compost applications……………………………………………………………173

Table 6.4. Analysis of covariance results of soil nutrients, OM and pH after four years of

organic transition strategies and compost application in 2010…………………174

Table 6.5. Soil pH and Bray-1 exchangeable P as affected by four years of organic

transition strategies within a compost application in 2010……………………..175

Table 6.6. Exchangeable K, Ca, and Mg, and OM as affected by four years of organic

transition strategies and compost application in 2010………………………….176

Table 6.7. Canonical correlation analysis for each set of canonical variables describing

marketable yield of potato or tomato and soil chemistry in 2010………………180

xii

Table 6.8. Standardized canonical coefficients for original variables of marketable potato

yield and soil chemistry for significant canonical correlations (P < 0.05) described

by canonical variates……………………………………………………………181

xiii

List of Figures

Figure 3.1. Biomass ± SE (g pot-1) of Canada thistle shoots and rhizomes harvested

from the greenhouse trial with eight varieties of tef, one variety of sorghum-

sudangrass, and a no smother crop control. Bars without the same letter differ

according to Fisher‟s Protected Least Significant Difference at P < 0.05……….81

Figure 3.2. Biomass ± SE (g pot-1) of smother crop varieties growing in competitioin with

Canada thistle planted at 7.6 and 15 cm depths. Bars followed by the same letter

do not differ according to Kruskal-Wallis Test at P < 0.05……………………...82

Figure 3.3. Height ± SE (cm) of Canada thistle and smother crops in pots where Canada

thistle rhizomes were planted at 7.6 and 15 cm depths. Bars within a group

followed by the same letter do not differ according to Kruskal-Wallis Test at P <

0.05……………………………………………………………………………….84

Figure 4.1. Linear regression of total land equivalent ratio (LERt) and aggressivity (AGX)

of grass (G) crops to legume or forb (X) crops in two-species mixtures and

aggressivity (AGLGF) of grass (G) crops to legume (L) and forb (F) crops in three-

species mixtures. Data presented is from 2008 and 2009………………………112

xiv

Figure 4.2. Weed biomass in the non-treated control and crop mixtures in 2008 and

2009. Bars within a year followed by the same letter do not differ according to

single degree-of-freedom contrasts (P < 0.05)…………………………………114

Figure 4.3. Weed cover (%) over growing degree days (base 10 C) in 2008 and 2009.

Adjusted means ± s.e. at each sampling time are presented. Equation parameters

are listed in Table 4.5…………………………………………………………...115

Figure 5.1. Mean monthly maximum and minimum temperatures (C) in 2009, 2010,

and 20-yr mean…………………………………………………………………136

Figure 6.1. The total and marketable number and yield of tomato fruits as affected by

compost treatment with plant density of 2.5x104 plants ha-1. Bars within a group

with the same letter do not differ (P < 0.05)……………………………………177

Figure 6.2. The number and yield of U.S. No. 1 and No. 2 potato tubers as affected by

compost treatment with plant density of 1.4x103 plants ha-1. Bars within a group

with the same letter do not differ (P < 0.05)……………………………………178

Figure 6.3. The number and weight of cull potato tubers as affected by organic transition

strategy with plant density of 1.4x103 plants ha-1. Bars within a group (cull count;

cull weight) with the same letter do not differ (P < 0.05). Abbreviations: FA,

fallow; NT, non-treated; PR, prairie; SC, smother crops; VG, vegetables……..179

Chapter 1: Introduction

Public concerns about environmental and human health consequences of the use of pesticides in conventional agriculture have driven a large increase in demand for organic food in the United States (Dorais 2007). Growth in retail sales of organic products has increased by 20% or more each year since 1990 (Dmitri and Greene 2002).

Between 1995 and 2008, there was an 81% increase in the quantity of certified organic farmland in the United States (USDA 2010). Although the organic market and total acreage of organic cropland and pasture has increased, the total organic farm acreage in

2008 was only 1.2% of agricultural land in production in the United States (USDA 2010).

Despite the increased demand for organic products and higher price premiums realized by growers, many farmers are reluctant to convert to organic agriculture. Barriers to adoption include higher managerial costs and risks of adopting new farming practices

(USDA 2010). Many farmers are reluctant to transition to organic production due to perceived challenges of lower yields and pest management (Beveridge and Naylor 1999;

Hanson et al. 2004; Oberholtzer et al. 2005).

In order to obtain organic certification, farmland must be managed without the use of pesticides or chemical fertilizers for three years without realizing the premium

1 organic prices (Greene and Kremen 2003). During the mandated transition period from conventional to organic agriculture in the United States, many farmers adopt biological, cultural, and mechanical techniques that aid in building soil fertility and suppressing weeds in order to prepare for certified organic crop production (Hanson et al. 2004).

Despite three years of transition to prepare land for organic farming, many growers view weed management as too great a challenge to undergo transition to organic agriculture

(Beveridge and Naylor 1999; Hanson et al. 2004; Oberholtzer et al. 2005). In Ohio,

74.3% of organic growers named weed control as their primary concern (Rzewnicki

2000). Management of perennial weeds, such as Canada thistle, [Cirsium arvense (L.)

Scop] poses a particular problem to growers without herbicides at their disposal

(Beveridge and Naylor 1999). In order to encourage growers to transition to organic agriculture to meet current market demands, research is needed on organic transition strategies that can suppress weeds before initiation of certified organic production

(Bàrberi 2002; Turner et al. 2007).

Current strategies in organic weed management. Organic farmers employ a variety of mechanical and cultural techniques for weed suppression. Mowing is often used for control of weeds, but may not be compatible with annual crop production systems.

Cultivation and tillage are commonly used to control weeds before planting and for inter- row control during early stages of crop growth (Bond and Grundy 2001). However, tillage and cultivation cannot provide season-long weed control in crops and can result in loss of soil nutrients and soil erosion (Havlin et al. 1990). Additionally, research has

2 demonstrated that tillage and cultivation alone may serve to increase populations of perennial or grass weeds (Buhler 1995). For example, using tillage to control Canada thistle may serve to cut up and spread underground roots capable of vegetative reproduction (Evans 1984). Perennial weeds are often cited as one of the most daunting pest management challenges for organic growers (Turner et al. 2007). Additionally, cultivation and tillage can encourage weed emergence, which may be increasingly difficult to manage with crop maturity (Parish 1990). Mechanical management may not be effective at short- or long-term weed control in organic agriculture (Bàrberi 2002). To effectively suppress weeds and enhance soil fertility in order to improve yields in the first year of organic production, alternatives to mechanical management may need to be used.

Farmers often use crop species in rotation or as cover crops to enhance weed suppression in organic agriculture. The use of different crop species in rotation provides varying patterns of resource utilization, soil disturbance, and mechanical impacts that provide an unstable environment for the emergence of a particular weed (Bond and

Grudy 2001; Liebman and Dyck 1993). Cover crops are also frequently used by organic farmers to benefit soils and compete with weeds for resources before and after the growing season (Bàrberi 2002). Specifically, cover crops planted with the intention of suppressing weed emergence and growth are usually termed smother crops. Smother cropping involves the use of a living plant to reduce the growth, development, or reproduction of weeds predominantly through competition for resources (Teasdale 1998).

Competition for light is often considered one of the mechanisms that improves the suppressive effect of smother crops (Holt 1995). Smother cropping is an alternative weed

3 management technique that is well adapted to the three-year transition required for organic production (Bàrberi 2002). Smother crops may be used alone or in mixture during the growing season or in conjunction with a cash crop to provide weed control

(Teasdale 1998). Unlike mechanical weed management, smother crops may improve soil fertility and discourage further germination of weed seeds (Teasdale 1998).

Why smother cropping? Smother crops can effectively suppress a diverse group of weeds through a variety of competition mechanisms. Monoculture plantings of leguminous and grass smother crop species have been shown to reduce aboveground biomass of annual broadleaf and grass weed species (Creamer and Baldwin 2000).

Selection of smother crop species based on ecological and morphological characteristics can enhance weed control. The growth habit of the crop and weed can interact to improve weed suppression or reduce crop growth (Wang et al. 2006). Research has demonstrated that annual clover species seeded for suppression of brown mustard [Brassica juncea (L.)

Czern.] were more competitive than perennial clover species (Ross et al. 2001).

Allelopathic potential of smother crops may also enhance weed suppression (Barnes and

Putnam 1983). Management of smother cropping systems can also affect weed control.

Suppression of smooth pigweed (Amaranthus hybridus L.) was found to be a function of smother crop planting density and the subsequent effect of light attenuation through the canopy (Collins et al. 2008). Plant density of intercropped wheat (Triticum aestivum L.) and field beans (Vicia faba L.) has been demonstrated to affect the suppression of weed populations (Bulson et al. 1997). The selection of appropriate smother crop species and

4 management regimes to compete with target weed species may enhance weed control.

However, selection of crop species is often not based on standardized criteria or knowledge of mechanisms that may enhance weed suppression.

Smother cropping systems may also be effective at suppressing weeds when planted in mixture. The use of multi-species cropping systems is known to increase system productivity and improve pest management (Mohler and Liebman 1987; Tilman et al. 2002). The advantage to polyculture crop plantings is often attributed to the occupation of multiple ecological niches (Linares et al. 2008; Willey 1979). The use of crop species of different functional groups such as a grass and legume in mixture has been shown to reduce weed populations more effectively than monoculture smother crops

(Creamer and Baldwin 2000; Linares et al. 2008). Increasing the number of functional groups represented by crops in an intercropping system resulted in greater weed suppression (Banik et al. 2006; Zuofa et al. 1992). Occupation of different spatial niches by crops can aid in weed control through competition for space and light (Unamma et al.

1986; Wang et al. 2006). However, competition between crops in mixture can reduce the degree of weed suppression observed (Valverde et al. 1995; Zuofa et al. 1992).The use of smother crop mixtures may enhance weed suppression in transition to organic agriculture, but how to design mixtures while reducing competition between smother crops is not clearly understood.

Smother cropping has been shown to be effective at suppressing annual and perennial weeds without the use of chemical management. Velvetbean [Mucuna pruriens

(L.) DC. var. utilis] used as a smother crop in tropical agriculture has been successful at

5 suppressing emergence of speargrass [Imperata cylindrica (L.) Beauv.], which reproduces through underground rhizomes (Udensi et al. 1999). According to Collins et al. (2007), smother crops have also reduced the quantity and weight of yellow nutsedge

(Cyperus esculentus L.) tubers. For suppression of Canada thistle, smother cropping with mechanical control has been effective. Mowing in combination with grass-clover smother crops or sudangrass [Sorghum sudanese (Piper) Stapf.] with and without cowpea [Vigna unguiculata (L.) Walp.] reduced density of Canada thistle after more than one year of management (Bicksler and Masiunas 2009; Graglia et al. 2006). Growth and development of Canada thistle is dependent on seasonally-controlled cycles of carbohydrate storage in underground roots (McAllister and Haderlie 1985). Taking advantage of Canada thistle biology may be an effective management strategy through appropriately timed and designed smother crop mixtures (Bicksler and Masiunas 2009).

However, previous studies have not separated the effects of smother crops and mowing on Canada thistle. Research needs to address the ability of smother crops and other organic transition strategies to suppress Canada thistle and other weeds to encourage farmers to meet market demands for organic products.

The central hypothesis of my Ph.D. research is that the use of smother crops and smother crop mixtures would provide non-chemical suppression of Canada thistle and other weeds appropriate for use in organic production. The long-term goal of this research is to increase our understanding of smother crop species and mechanisms of smother cropping to ease organic transition of cropland through enhanced weed suppression. Ultimately, we want to improve transition strategies to provide farmers with

6 more options when considering organic agriculture to implement their goals. The objectives of this research, corresponding to each of the subsequent chapters of this dissertation, were to

1) identify features of smother crops and associated mechanisms of weed suppression through an extensive literature review;

2) evaluate the use of available tef [Eragrostis tef (Zucc.) Trotter] varieties as smother crops for annual weed and Canada thistle control;

3) assess warm-season smother crop mixtures for biomass production and weed suppression;

4) determine planting dates and smother crop mixtures for effective Canada thistle management; and

5) compare the efficacy transition strategies to suppress weeds, enhance soil fertility, and improve yields and quality of organic tomato (Solanum lycopersicum L.) and potato

(Solanum tuberosum L.).

Literature Cited

Banik, P., A.Midya, B.K. Sarkar, and S.S. Ghose. 2006. Wheat and chickpea intercropping systems in an additive series experiment: advantages and weed smothering. Eur. J. Agron. 24:325-332.

Bàrberi, P. 2002. Weed management in organic agriculture: are we addressing the right issues? Weed Res. 41:177-193.

Barnes, J.P. and A.R. Putnam. 1983. Rye residues contribute weed suppression in no- tillage cropping systems. J. Chem. Ecol. 9:1045-1057.

Beveridge, L. E. and R.E.L. Naylor. 1999. Options for organic weed control – what farmers do. Pages 939-944 in The 1999 Brighton Crop Protection Conference– Weeds. Brighton, U.K.: British Crop Protection Council.

Bicksler, A. J. and J. B. Masiunas. 2009. Canada thistle (Cirsium arvense) suppression with buckwheat or sudangrass cover crops and mowing. Weed Technol. 23:556- 563.

Bond W. and A. C. Grundy. 2001. Non-chemical weed management in organic farming systems. Weed Research 41:383-405.

Buhler, D.D. 1995. Influence of tillage systems on weed population dynamics and management in corn and soybean in the central USA. Crop Sci. 35:1247-1258.

Bulson, H.A.J., R.W. Snaydon, and C.E. Stopes. 1997. Effects of plant density on intercropped wheat and field beans in an organic farming system. J. Agric. Sci. 128:59-71.

Collins, A. S., C. A. Chase, W. M. Stall, and C. M. Hutchinson. 2007. Competitiveness of three leguminous cover crops with yellow nutsedge (Cyperus esculentus) and smooth pigweed (Amaranthus hybridus). Weed Sci. 55:613-618.

Collins, A.S., C.A. Chase, W.M. Stall, and C.M. Hutchinson. 2008. Optimum densities of three leguminous cover crops for suppression of smooth pigweed (Amaranthus hybridus). Weed Sci. 56:753-761.

Creamer, N.G. and K.R. Baldwin. 2000. An evaluation of summer cover crops for use in vegetable production systems in North Carolina. HortScience 35:600-603.

Dimitri, C. and C. Greene. 2002. Recent growth patterns in the U.S. organic foods market. Washington, D.C.: U.S. Department of Agriculture, Economic Research Service, Resources Economics Division, Agriculture Information Bulletin 777.

Dorais, M. 2007. Organic production of vegetables: state of the art and challenges. Can. J. Plant Sci. 87:1055-1066.

Evans, J. E. 1984. Canada thistle (Cirsium arvense): a literature review of management practices. Natural Areas Journal 4:11-21.

Graglia, E., B. Melander, and R.K. Jensen. 2006. Mechanical and cultural strategies to control Cirsium arvense in organic arable cropping systems. Weed Res. 46:304- 312.

Greene, C. and A. Kremen. 2003. U.S. Organic Farming in 2000-2001: Adoption of Certified Systems. Washington, D.C.: U.S. Department of Agriculture, Economic Research Service, Resource Economics Division, Agriculture Information Bulletin No. 780. 55 p.

Hanson, J., R. Dismukes, W. Chambers, C. Greene, and A. Kremen. 2004. Risk and risk management in organic agriculture: views of organic farmers. Renew. Agr. Food Syst. 19:218-227.

Havlin, J.L., D.E. Kissel, L.D. Maddux, M.M. Claassen, and J.H. Long. 1990. Crop rotation and tillage effects on soil organic carbon and nitrogen. Soil Sci. Soc. Am. J. 54:448-452.

Holt, J.S. 1995. Plant responses to light: a potential tool for weed management. Weed Sci. 43:474-482.

Liebman, M. and E. Dyck. 1993. Crop rotation and intercropping strategies for weed management. Ecol. Appl. 3:92-122.

Linares, J., J. Scholberg, K. Boote, C.A. Chase, J.J. Ferguson, and R. McSorley. 2008. Use of the cover crop weed index to evaluate weed suppression by cover crops in organic citrus orchards. HortScience 43:27-34.

McAllister, R. S. and L. C. Haderlie. 1985. Seasonal variations in Canada thistle (Cirsium arvense) root bud growth and root carbohydrate reserves. Weed Sci. 33:44-49.

Mohler, C.L. and M. Liebman. 1987. Weed productivity and composition of sole crops and intercrops of barley and field pea. J. Appl. Ecol. 24:685-699.

Oberholtzer, L., C. Dimitri, and C. Greene. 2005. Price premiums hold on as U.S. organic produce market expands. Washington, D.C.: U.S. Department of Agriculture, Economic Research Service, Resources Economics Division, VGS-308-01.

Parish, S. A review of non-chemical weed control techniques. Biol. Agric. Hortic. 7:117- 137.

Ross, S.M., J.R. King, R.C. Izarrualde, and J.T. O'Donovan. 2001. Weed suppression by seven clover species. Agron. J. 93:820-827.

Rzewnicki P. E. 2000. Ohio Organic Producers: Final Survey Results. Available at http://ohioline.osu.edu/sci174_11.html. Accessed 18 June 2008.

Teasdale, J. R. 1998. Cover crops, smother plants, and weed management. Pages 247-270 in J. L. Hatfield, D. D. Buhler, and B. A. Stewart, eds. Integrated Weed and Soil Management. Chelsea, MI: Sleeping Bear Press.

Tilman, D., K.G. Cassman, P.A. Matson, R. Naylor, and S. Polasky. 2002. Agricultural sustainability and intensive production practices. Nature 418:671-677.

Turner, R. J., G. Davies, H. Moore, A. C. Grundy, and A. Mead. 2007. Organic weed management: a review of the current UK farmer perspective. Crop Prot. 26:377- 382.

Udensi, U.E., I.O. Akobundu, A.O. Ayeni, and D. Chikoye. 1999. Management of cogongrass (Imperata cylindrica) with velvetbean (Mucuna pruriens var. utilis) and herbicides. Weed Technol. 13:201-208.

Unamma, R.P.A., L.S.O. Ene, S.O. Odurukwe, and T. Enyinnia. 1986. Integrated weed management for cassava intercropped with maize. Weed Res. 26:9-17.

United States Department of Agriculture. 2010. Organic production. Available at http://www.ers.usda.gov/Data/Organic/. Accessed 16 August 2011.

Valverde, B.E., A. Merayo, C.E. Rojas, and T. Alvarez. 1995. Interaction between a cover crop (Mucuna sp.), a weed (Rottboellia cochinchinensis), and a crop (maize). Pages 197-200 in Proceedings of the Brighton Crop Protection Conference – Weeds, Brighton, UK.: British Crop Protection Council.

Wang, G., M.E. McGiffen Jr., J.D. Ehlers, and E.C.S. Marchi. 2006. Competitive ability of cowpea genotypes with different growth habit. Weed Sci. 54:775-782.

Willey, R.W. 1979. Intercropping – its importance and research needs. Part 1. Competition and yield advantages. Field Crop Abst. 32:1-11.

Zuofa, K., N.M. Tariah, and N.O. Isirimah. 1992. Effects of groundnut, cowpea and melon on weed control and yields of intercropped cassava and maize. Field Crop Res. 28:309-314.

Chapter 2: A Review of Smother Cropping and Associated Factors of Weed

Suppression

Stephanie Wedryk and John Cardina*

Smother crops are frequently mentioned in the literature as a technique for suppressing weeds through competition from living plants. Studies on smother crops have been conducted in a variety of contexts and with associated management inputs that make it difficult to determine the effectiveness of smother crops at suppressing weeds and what accounts for their success or failure. This review presents an analysis of smother cropping and related mechanisms of weed suppression. We examined hundreds of studies reported since 1929 and identified those that tested the ability of living plants to suppress weeds without confounding inputs. The degree of weed suppression in each smother cropping scenario was rated and analyzed to determine the most and least effective species and associated factors for weed suppression. Factors related to crop growth and development were most important in determining the effectiveness of cover crops. Ancillary management practices such as fertilization and irrigation contributed to smother crop

* Graduate Research Associate and Professor, Department of Horticulture and Crop Science, The Ohio State University, Wooster, OH 44691. Corresponding author‟s E-mail: [email protected].

12 effectiveness and may serve to support crop growth and development. Crop interaction factors were least important, due to competitive interactions of component smother crops.

Development of a successful smother crop system depends on the competitive ability of the smother crop and its physiological and chemical attributes, as well as crop complementarity within the cropping system and against target weed species.

Keywords: Organic weed management, intercropping, cover crops, weed suppression.

The use of living plants to compete with weeds in conventional or organic agriculture has been of interest for many years as a way to reduce herbicide inputs and the harmful effects of tillage on soil erosion and fertility (Akobundu 1978; Liebman and

Dyck 1993; Teasdale 1998). These plants may be grown before, during, or after a cash crop, for the purpose of providing weed suppression, improved soil nutrition, and habitat for beneficial pollinators and other insects (Akobundu 1978; Bàrberi 2002; Liebman and

Staver 2001). In this review, we focus on smother crops, which we define as a living plant species or mix of species growing alone or in combination with a main (cash) crop to reduce the growth, development, and reproduction of undesirable plants. A smother crop can be grown without a main crop, in a crop rotation, or simultaneously with a main crop. This definition distinguishes a smother crop from residue of a killed crop. This description of the use of smother crops for weed suppression has been given many names in the literature. What we define as a smother crop has also been referred to as a cover crop, catch crop, companion crop, competitive crop, intercrop, mixed crop, living mulch and green manure (Akobundu 1984; Crafts 1975; Liebman and Staver 2001; Paine and

Harrison 1993; Regnier and Janke 1990; Robbins et al. 1942). Determining smother crop effectiveness and function in the literature is confounded by the use of multiple terms that often imply other economic or ecological benefits in addition to weed suppression.

Smother crops are often integrated with other weed control practices in research.

Smother crops have been investigated in tandem with herbicides, mowing, or tillage

(Ateh and Doll 1996; Bicksler and Masiunas 2009; DeHaan et al. 1994; Teasdale 1993).

Combining different management techniques confounds isolating and detecting the

14 possible beneficial effects of the smother crop. This makes it difficult to assess the source of the observed reduction in weed density or growth.

The idea of using a crop plant to suppress weedy plants, while obtaining other environmental benefits, has great appeal. The underlying biology that would allow a smother crop to exhibit selectivity in suppressing weed growth without reducing main crop yield or economic return has not been described. Yet there are numerous reports that smother crops are used successfully in some cropping systems. The key to their utility is likely a combination of plant traits and management practices that have not been fully described. However, given the lack of a cohesive, reinforced theory of smother crop function in competition with weedy and non-weedy plants and employment in the literature, the evidence is not clear whether smother crops are an effective approach for reducing the growth, development, or reproduction of weeds in cropping systems. Not surprising then, there are no clear recommendations guiding farmers in the use of smother crops, explaining what makes them work, or what causes them to fail. Here we present a review of pertinent literature that examines the utility of smother crops for organic and alternative weed management. Our goal was to identify the probable physiological and ecological mechanisms governing this method of weed control in order to determine factors underlying the success or failure of smother crops, with the hope of identifying the useful features that will help growers in developing effective smother cropping systems for weed management.

Materials and Methods

To investigate the use of smother crops, we searched BIOSIS, BioOne, CAB

Direct, WorldWideScience.org, ISI Web of Knowledge, Garden Landscape and

Horticulture Index, Agricola (EBSCO Host), Agricola (NAL Host), AGRIS, CRIS,

Environmental Complete (EBSCO Host), Environmental Sciences and Pollution

Management, and Google Scholar using the keywords and Boolean operatives “smother crop,” for all time periods from 1920 until January 1, 2010. Additionally, theses and dissertations were searched for in the same time period with keywords “smother crop” using the databases Australasian Digital Theses Program, Darenet, Electronic Theses and

Dissertations (OhioLink Host), Ethos, Networked Digital Library of Theses and

Dissertations (Virginia Technical University Host), OAIster (University of Michigan

Host), ProQuest, Theses Canada Portal and WorldCat. The literature search was then extended to references cited by these studies. Other search terms were excluded from the literature search as related terms often include studies that do not focus on weed management or include management practices that confound the competitive ability of the living smother crop.

Literature included in the following review was chosen based on several criteria.

First, the study had to include data on any parameter of weed biology while the smother crop was still living. Second, if herbicides were applied directly preceding the collection of weed data in the smother crop, the study was excluded to eliminate confounding effects. If herbicides were used after the collection of pertinent weed data or were applied

16 to control non-target weed species, the study was included as the herbicide would not interfere with the measurement of smother crop effectiveness. Third, studies were excluded in which the smother crop was tilled under or mowed before weed data were collected, because it was impossible to distinguish tillage and mowing effects from smother crop effects on weed populations. However, studies in which tillage occurred before seeding of the smother crop or mowing occurred with ample time for regrowth of the smother crop before weed data were collected were included in the review. Studies in which weed data were collected after the senescence of a smother crop or after rotation of several smother crops were not included, as our definition of a smother crop involves suppression by living plants. Studies that used a smother crop system for management of invasive species in natural areas were excluded. In total, 89 publications were considered for this review.

Each publication was evaluated for quality on a scale of 0-10 with 10 being superior quality. The rating system is explained in Table 2.1. Briefly, publications were awarded points based on journal impact factor, use of statistics, experimental design, descriptions of species, varieties and sites, treatments and data collected. Points were assigned for presence of listed criteria within each category (Table 2.1).

Following evaluation of publication quality, individual publications were scrutinized to delineate individual smother cropping scenarios characterized by individual crop and weed species, year and site of study, and management practices. A single smother cropping scenario would include a single smother crop species or mixture and variety in combination with a single main crop species and variety present during

17 smother crop life (where applicable), one year of the study, one site of the study (where applicable), a single target weed species or type (where applicable), and one set of management practices (irrigation, fertilizer, tillage regime, etc…). We used this approach in order to exclude sources of variation that may affect weed suppression. For example, in Creamer and Baldwin (2000), multiple smother crop species were compared for weed suppression. Each species would be classified under a different smother cropping scenario to account for the effect of an individual species on weed growth. Combination of multiple species in one scenario may not lead to an accurate depiction of why smother cropping was or was not effective. When different varieties or genotypes of a single smother crop species were evaluated as in Wortmann (1993), each genotype was classified in a different scenario. The same considerations applied for publications that used different main crops or main crop varieties or targeted different weed species or weed types. Each unique combination of smother crop species, smother crop variety, main crop species, main crop variety, and weed species or type constituted a different scenario. In the scenario classification spreadsheet, descriptions of the type of crop

(grass, legume, or forb) or weed (grass, broadleaf) and the life cycle (annual, biennial, perennial) of the crop or weed were added.

Classification of smother cropping scenarios continued with consideration of year, site, and management. Each year or site of a study was separated into different scenarios.

In Ekeleme et al. (2003), smother crops were planted in four different agroecological zones of Nigeria, constituting four different sites, which were classified into separate scenarios in addition to any aforementioned crop-weed classifications. Linares et al.

(2008) conducted experiments from 2002-2005. Each year would be separated into a scenario in addition to any aforementioned crop-weed classifications. Finally, different management regimes were put into separate scenarios. For example, three different mechanical control strategies used to manage speargrass [Imperata cylindrica (L.) P.

Beauv] before smother crop seeding, were each classified into different scenarios in combination with aforementioned criteria (Aflakpui and Bolfrey-Arku 2007). For descriptive purposes only, studies were also classified by journal of publication, cropping system, geography, smother crops tested, main crops tested (where applicable), and weeds of interest (where applicable).

Each smother cropping scenario was rated on a scale of 0-10 where 0 indicated no weed suppression and 10 indicated complete control. Descriptions of the studies based on journal, cropping system, geography, and crops used were not given weed suppression ratings nor analyzed statistically. Rating values were assigned by determining the amount of reduction in weed growth (biomass, density, height, or other measured value) between the smother crop scenario and the control treatment in the publication. When multiple parameters of weed growth were measured, the average weed reduction among weed growth measures for each smother crop scenario was used. Weed suppression factors, the mechanisms by which smother crop species were deemed to contribute to weed suppression, were determined for each scenario based solely on the data collected in each publication. For example, Lotz et al. (1991) compared the suppressive effect of four smother crops on yellow nutsedge (Cyperus esculentus L.). Some of the crops in this publication were more effective at suppressing the number of yellow nutsedge tubers than

19 others. The scenarios with the more suppressive smother crops would be labeled with the weed suppression factor “smother crop species,” as the choice of species was determined to affect suppression. There were 49 weed suppression factors that we organized into 8 broader categories. The eight broad categories were environment and climate, soil inputs and properties, crop and soil management, crop characteristics, crop growth and development, photosynthesis and leaf development, weed competition, and crop interactions as listed in Table 2.2.

Data of weed suppression ratings were analyzed using analysis of variance, where species of smother crop or weed or weed suppression factor or category were regarded as independent variables and weed suppression rating was the dependent variable. Each scenario was weighted by publication quality rating. Scenarios that were rated as zero were excluded from analysis of variance for analysis of weed suppression factors only.

Data that did not meet the assumptions of analysis of variance were subjected to non- parametric methods using Kruskal-Wallis test statistics (P < 0.05). Descriptions of smother crop research included in the review such as journal of publication, location of research, named research focus, and types of smother crops, main crops, and weeds studies were analyzed for frequency of occurrence and were not subjected to analysis of variance. All statistical analyses were computed using SAS v.9.21.

Results

Falling within our definition of smother cropping we found publications using the terms smother crops, intercrops, and cover crops to describe the phenomenon of weed suppression by living plant species (Table 2.3). From this point forward, use of the term smother crop will refer to the definition previously given (see Introduction) and any publications included in this review. Grain crops were the most frequent cash crop of interest in intercropping studies (Table 2.3). Sixty-five percent of the research on smother crops has been conducted in North America or Africa (Table 2.4). Nigeria was the most common location for smother crop research (Table 2.4). Fifty-five percent of smother crop species that have been evaluated were annual legumes and 16 % were annual grasses

(Table 2.5). The most commonly studied smother crop species were cowpea [Vigna unguiculata (L.) Walp.], in 7% of studies, and velvetbean [Mucuna pruriens (L.) DC.] in

4% of studies (Table 2.5). Thirty-six percent of the reviewed publications reported measurements assessing all weed species present without distinction of type or species

(Table 2.6). The effect of smother crops on broadleaf weeds or grass weeds accounted for

36% and 28% of all weed types measured, respectively. Specifically, target weeds measured included common lambsquarters (Chenopodium album L.) and speargrass

[Imperata cylindrica (L.) Beauv.] (Table 2.6).

Among the 8 broad categories of weed suppression factors determined from the data collected in each reviewed publication, smother crop growth and development was the underlying factor category giving the most effective weed suppression (Table 2.7).

The category of crop interaction factors was least effective (a score of 5.2 out of 10) in

21 contributing to weed suppression, and there was no difference among factor categories related to weed competition, photosynthesis, environment, or soils.

Among the 49 factors within each weed suppression factor category evaluated that contributes to weed suppression by smother crops, we identified 5 with the greatest contribution to weed suppression, and the 5 with the least contribution (Table 2.8).

Irrigation and fertilization treatments had the highest scores (7.6 and 7.5, respectively), indicating they helped smother crops to more effectively suppress weeds in comparison to other factors. Soil properties and main crop species scored only 3.7 to 4.0, indicating they contributed little.

We examined specific smother crops and main crops where weed suppression was most and least effective, to help identify trends in combinations that allow smother crops to succeed or to fail (Table 2.9). Weed suppression was 7.7 to 8.0 for smother crops

Italian ryegrass (Lolium multiflorum Lam) and subterranean clover (Trifolium subterreaneum L.) intercropped in maize (Zea mays L.). In contrast, weed suppression was 0.58 and 1.1 with smother crops of Asian pigeonwings (Clitoria ternatea L.) and cowpea [Vignua unguiculata (L.) Walp] intercropped in maize.

Since the success of smother crops depends on which weeds are being suppressed as well as which smother and main crop are used, we examined the data for studies with specific weeds (Table 2.10). The most effective combination of smother cropping system and weed was intercropping of subterranean clover in maize for management of ivyleaf morningglory [Ipomoea hederacea (L.) Jacq.] (WS Rating = 9.8). Use of smother crop species rape (Brassica napus L.), winter vetch (Vicia villosa Roth) and triticale (x

Triticosecale rimpaui Wittm.) without a main crop generally resulted in no suppression of target weeds (Table 2.10).

In order to determine whether target weed types such as annual broadleaves, annual grasses, or perennial grasses are affected differently by weed suppression factors in unique smother cropping systems, we examined the data for weed suppression factors that were most effective at suppressing different weed types (Table 2.11). Smother crop variety was highly rated in suppressing annual and perennial grasses. Niche occupation was identified as the underlying mechanism for suppression annual grass (WS Rating =

6.3) and annual broadleaf weeds (WS Rating = 9.3). Smother crop life cycle contributed to suppression of annual broadleaf (WS Rating = 7.9) and perennial grass weeds (WS

Rating = 6.2). Some crop species have traits that allow them to suppress some weed species more effectively than others due to specific traits of the weed such as growth habit, time of emergence, or life cycle. This is referred to as weed-crop complimentarity and was a factor contributing to suppression of annual grasses (WS Rating = 6.2).

Discussion

Measurement of weed suppression factors biomass or ground cover can help determine competitive ability of a smother crop (Gaudet and Keddy 1988). Weed suppression factors associated with smother crop growth and development, such as biomass and ground cover, had the highest mean weed suppression rating of all

23 mechanism categories (Table 2.7). Greater biomass production by assessed smother crops was related to greater suppression of weeds. Rye (Secale cereale L.) planted as a smother crop produced 20% more crop biomass and provided a 10% or greater increase in weed biomass suppression compared to smother crops of alfalfa (Medicago sativa L.) (Schoofs and Entz 2000). Similarly, a smother crop of sunnhemp (Crotalaria juncea L.) produced

50% or more crop biomass than hairy indigo (Indigofera hirsuta L.), alyceclover

(Alysicarpus vaginalis L.), or velvetbean while increasing weed suppression by 14% or more compared to these same smother crops (Linares et al. 2008). Biomass accumulation reflects competitive ability for below- and above-ground resources that may also contribute to the ability of a smother crop to occupy above-ground space as measured in ground cover. The ability of a smother crop to compete for above-ground space as measured by percent cover was related to effective weed suppression. Barley (Hordeum vulgare L.) was rated as one of the more weed-suppressive smother crops (WS Rating =

6.6) (Table 2.9). Percentage ground cover measurements of barley as a smother crop, which suppressed weeds by 98% compared to the weedy control, were 40% higher than smother crops that provided no weed suppression (Nelson et al. 1991). Competition for space is related to resources that can be sequestered from such above-ground space, such as light captured by smother crops that would otherwise reach weedy plants or the soil surface. Factors related to photosynthesis and leaf development including reduction of light were among the most effective mechanisms of weed suppression (Table 2.7).

Factors that contribute to optimum crop growth and development are likely to play an important role in weed suppression. The importance of irrigation, fertilization and

24 land preparation illustrate how management decisions influence weed suppression outcomes (Table 2.8). Soil inputs can provide smother crops resources to accumulate biomass or height and occupy more space, contributing to weed suppression. The addition of fertilizer to smother crops of Mucuna spp. resulted in an approximately 17 to

30% increase in crop biomass production, depending on species, and a 10 to 30% increase in weed suppression (Akobundu et al. 2000). Decisions concerning planting time and land preparation can maximize weed suppression during the best time of year or when environmental conditions are ideal (Olasantan 2007). Biomass of weeds was suppressed when a smother crop of pumpkin (Cucurbita maxima Duchesne) and a yam

(Dioscorea spp.) main crop were planted in March (21% reduction) or April (15% reduction) instead of May (Olasantan 2007). In Aflakpui and Bolfrey-Arku (2007), plowing fields reduced density of speargrass shoots by 20 to 99%, depending on site, compared to slashing speargrass shoots prior to smother crop seeding. However, smother crop suppression of weeds is not always enhanced through management decisions. Weed biomass increased while smother crop biomass decreased with increasing rates of nitrogen fertilization in a Medicago polymorpha (L.) cv. Santiago smother crop seeded with a maize main crop (DeHaan et al. 1997). Careful consideration of the smother cropping system and further testing of management strategies in smother cropping systems can illustrate how management practices influence weed suppression.

Decisions about cropping systems and species choices can impact the effectiveness of smother crops to suppress weeds. Factors related to crop interactions were least effective in suppressing weeds (Table 2.7). It is commonly agreed that

25 intercropping, included in crop interactions, can reduce weed pressure (Liebman and

Dyck 1993). Smother crops used for economic return in an intercropping system were not as effective at suppressing weeds as monoculture smother crops seeded without the purpose of economic return or without a main crop. Smother crops chosen in intercropping systems have a dual role of producing food and suppressing weeds. Since resources must be allocated to production of seeds or fruits, this reduces allocation toward plant components that may contribute to weed suppression such as leaf expansion, height, or root system growth (Jordan 1993). Interspecific competition between smother crops and main crops may also reduce effectiveness of smother crops in suppressing weeds. Using a smother crop without the need for seed or fruit production in monoculture eliminates interspecific competition in intercropping scenarios, which might allow the smother crop to be more competitive with target weeds.

Spatial arrangement in monoculture and intercropped smother crops also contributed little to weed suppression (Table 2.8). If spatial arrangement of smother crops or smother crops and main crops are not conducive to weed suppression either by increased competition between component crops or between weeds and crops, smother cropping is not effective. In wheat (Triticum aestivum L.) and chickpea (Cicer arietinum

L.) intercropping, weed biomass was reduced by 33% when row spacing between intercrops was reduced, likely due to competitive effects of the crops (Banik et al. 2006).

In pigeonpea [Cajanus cajan (L.) Millsp.] systems with intercropped smother species, row arrangements of two pigeonpea rows to one smother crop row resulted in 46%

26 reduction in weed suppression compared to 1:1 row arrangements, but the change in row arrangement did not affect weed suppression in other tested species (Ali 1988).

The niche occupation and species choice of smother crops in mixture or in intercropping situations can also affect weed suppression. Creamer and Baldwin (2000) used complementary crops without the purpose of food production in mixture of different functional groups (grass or legume) and spatial niches that effectively suppressed weeds.

Niche occupation was an effective factor in suppressing annual broadleaves (WS Rating

= 9.3) and annual grasses (WS Rating = 6.3) (Table 2.12). Using smother crops that occupy different niches through different utilization or resources, occupation of space, or temporal patterns of growth and resource acquisition can compete more effectively with weeds. Monocultures of saffron (Crocus sativus L.), a monocot, were less effective at suppressing grass weed density than monocultures of black zira (Bunium persicum

Boiss), a member of the Apiaceae (Mesgaran et al. 2008). Crops from different phylogenetic clades than target weed species can improve weed suppression.

The characteristics of the smother crop(s) in the smothering system are important in determining the level of weed suppression. Specific characteristics, such as smother crop life cycle, are for weed suppression despite the low rating for the crop characteristics category (Table 2.7). Weed suppression was improved by 32% in an annual smother crop of crimson clover (Trifolium incarnatum L.) compared to perennial red clover (Trifolium pretense L.) (Linares et al. 2008). A few of the most effective smother crops, such as subterranean clover (WS Rating = 7.7), M. cochinchinensis (WS Rating 7.3), and rye

(WS Rating = 6.1), are considered to be allelopathic, and this may have added to the high

27 level of weed suppression against annual grasses (WS Rating = 6.9) and perennial grasses

(WS Rating = 6.4). (Table 2.9; Table 2.11) (Barnes and Putnam 1983; Chikoye and

Ekeleme 2001; Enache and Ilnicki 1990; Ilnicki and Enache 1992; Ross et al. 2001;

Schoofs and Entz 2000). Allelopathy may be an important trait to select for in a smother crop (Table 2.11).

Crop growth habit and main crop species were two of the least important contributors to weed suppression (Table 2.8). Smother crops planted as an intercrop in a main cash crop are understood to assist in weed suppression, otherwise the planting would be unnecessary. The low weed suppression rating associated with crop growth habit may be the result of using smother crops that could not effectively compete with highly problematic weeds. Growth habit aided in suppression of the perennial grass I. cylindrica by sunnhemp (Crotalaria juncea L.), tropical kudzu (Pueraria phaseoloides

L.), stylo (Stylosanthes guianensis L.), and M. pruriens even though overall suppression was low (Bolfrey-Arku et al. 2002; Guritno et al. 2002). Using varieties of cowpea with different growth habits aided in the suppression of Helianthus annuus L. (sunflower), but the target weed was more competitive than the smother crop (Wang et al. 2006). Growth habit may aid in weed suppression, but is not necessarily the most important crop characteristic to select for.

The results of this review necessarily reflect the data that has been collected by researchers testing smother crop systems since 1929 (Arny 1929). Our assessment of smother crop species and effective factors of weed suppression is limited by the data that have been collected on studied smother crop species. We only presented smother

28 cropping scenarios in our results that were replicated by two or more publications so that conclusions drawn were based on rigorously tested data and not outlying data. However, there may be highly successful smother cropping scenarios that were not included in presented results. One idea attributed to the success of smother cropping is to limit the availability of light to weeds (Holt 1995). Photosynthesis and leaf development was one of the highest rated categories of weed suppression (WS Rating = 6.2) (Table 2.7).

However, these mechanisms were used in only 5.6% of the smother cropping scenarios presented in this review (Table 2.7). Further research on smother cropping should focus on light interception and competition for light to determine how they function in reducing weed pressure. Competition for light was important for the suppression of C. esculentus by a smother crop of hemp (Lotz et al. 1991). LAI contributed to high weed suppression by M. cochinchinensis, but did not contribute to effective weed suppression by peanut or sweet potato [Ipomoea batatas (L.) Lam] (Table 2.10) (Chikoye and Ekeleme 2001;

Zuofa and Tariah 1992). The importance of weed suppression factors may differ between smother crop species.

Further smother crop research should explore the ability of crops to suppress weeds through more extensive measurements within mechanism categories listed here.

Choice of smother crop species should consider its complementarity within the cropping system and its ability to grow and outcompete target weeds. Although all mechanisms analyzed helped to suppress weeds, those with higher ratings may be considered more important when trying to develop an ideal smother cropping system.

Conclusions

The use of smother crops to suppress weeds without chemical inputs is a viable weed management alternative that is dependent on the following criteria a) ability of smother crop to grow and compete effectively for resources and space, b) complementarity of component crops through occupation of different functional, spatial and temporal niches and ability to suppress weeds through occupation of said niches and c) management decisions that can enhance smother crop growth. Crop characteristics such as allelopathy should also be carefully considered when choosing a candidate smother crop species. Development of a successful smother cropping system can be enhanced by further research that looks more deeply at mechanisms of suppression for a large number of species and utilization of genetic research.

Sources of Materials

1 SAS 9.2 Statistical Software, Statistical Analysis Systems, SAS Institute, Inc., 100 SAS

Campus Drive, Cary NC 27513-2414.

Acknowledgements

We would like to thank Bert Bishop, Senior Statistician, for assistance in data analysis. This review was supported, in part, by the USDA-NRI-ORG project entitled

“Transition Strategies that Control Perennial Weeds and Build Soil.” Salaries and research support were provided by State and Federal Funds appropriated to the Ohio

Agriculture Research and Development Center, The Ohio State University. Manuscript

No. HCS-11-00.

Literature Cited

Abdin, O.A., X.M. Zhou, D. Cloutier, D.D. Coulman, M.A. Faris, and D.L. Smith. 2000. Cover crops and interrow tillage for weed control in short season maize (Zea mays). Eur. J. Agron. 12:93-102.

Aflakpui, G.K.S. and G.E.K. Bolfrey-Arku. 2007. Suppression of Imperata cylindrica (Speargrass) and changes in weed flora in yam and cassava fields. J.Plant Sci. 2:25-34.

Agegnehu, G., A. Ghizaw, and W. Sinebo. 2008. Yield potential and land-use efficiency of wheat and faba bean mixed intercropping. Agron. Sustain. Dev. 28:257-263.

Agegnehu, G., A. Ghizaw, and W. Sinebo. 2006a. Crop productivity and land-use efficiency of a teff/faba bean mixed cropping system in a tropical highland environment. Exp. Agr. 42:495-504.

Agegnehu, G., A. Ghizaw, and W. Sinebo. 2006b. Yield performance and land-use efficiency of barley and faba bean mixed cropping in Ethiopian highlands. Eur. J. Agron. 25:202-207.

Akobundu, I.O. 1978. Weed control strategies for multiple cropping systems of the humid and subhumid tropics. Pages 80-100 in Proceedings of the Conference Weeds and their Control in the Humid and Subhumid Tropcis, Ibadan, Nigeria.

Akobundu, I.O. 1984. Advances in live mulch crop production in the tropics. Pages 51-57 in Proceedings of the Western Society of Weed Science: Western Weed Science Society.

Akobundu, I.O., U.E. Udensi, and D. Chikoye. 2000. Velvetbean (Mucuna spp.) suppresses speargrass (Imperata cylindrica (L.) Raeushcel) and increases maize yield. Int. J. Pest Manage. 46:103-108.

Ali, M. 1988. Weed suppressing ability and productivity of short duration legumes intercropped with pigeonpea under rainfed conditions. Trop. Pest Manage. 34: 384-387.

Arny, A.C. 1929. Flax cropping in mixture with wheat, oats, and barley. USDA Technical Bulletin No. 133:1-47.

Ateh, C.M. and J.D. Doll. 1996. Spring-planted winter rye (Secale cereale) as a living mulch to control weeds in soybean (Glycine max). Weed Technol. 10:347-353.

Banik, P., A.Midya, B.K. Sarkar, and S.S. Ghose. 2006. Wheat and chickpea intercropping systems in an additive series experiment: advantages and weed smothering. Eur. J. Agron. 24:325-332.

Bantilan, R.T., M.C. Palada, and R.R. Hardwood. 1974. Integrated weed management: I. key factors effecting crop-weed balance. Philippine Weed Science Bulletin 1:14- 36.

Bàrberi, P. 2002. Weed management in organic agriculture: are we addressing the right issues? Weed Res. 41:177-193.

Barnes, J.P. and A.R. Putnam. 1983. Rye residues contribute weed suppression in no- tillage cropping systems. J. Chem. Ecol. 9:1045-1057.

Baumann, D.T., M.J. Kropff, and L. Bastiaans. 2000. Intercropping leeks to suppress weeds. Weed Res. 40:359-374.

Baumann, D.T., L. Bastiaans, and M.J. Kropff. 2001. Effects of intercropping on growth and reproductive capacity of late-emerging Senecio vulgaris L., with special reference to competition for light. Ann. Bot. 87:209-217.

Becker, M. and D.E. Johnson. 1999. The role of legume fallows in intensified upland rice-based systems of West Africa. Nutr. Cycl. Agroecosys. 53:71-81.

Beveridge, L.E. and R.E.L. Naylor. 1999. Options for organic weed control - what farmers do. Pages 939-944 in Proceedings 1999 Brighton Conference – Weeds, Brighton, U.K.: Brighton Crop Protection Council.

Bicksler, A.J. and J.B. Masiunas. 2009. Canada thistle (Cirsium arvense) suppression with buckwheat or sudangrass cover crops and mowing. Weed Technol. 23:556- 563.

Bolfrey-Arku, G.E.K., O.U. Onokpise, C.C. Coultas, and D. Shilling. 2002. Land preparation and legume cover crops for the biological control of cogongrass. Pages 4-9 in Proceedings of the Soil and Crop Science Society of Florida, Clearwater, FL: Soil and Crop Science Society of Florida.

Brainard, D.C. and R.R. Bellinder. 2004. Weed suppression in a broccoli-winter rye intercropping system. Weed Sci. 52:281-290.

Brainard, D.C., R.R. Bellinder, and A.J. Miller. 2004. Cultivation and interseeding for weed control in transplanted cabbage. Weed Technol. 18:704-710.

Brandsaeter, L.O. and J. Netland. 1999. Winter annual legumes for use as cover crops in row crops in northern regions: I. field experiments. Crop Sci. 39:1369-1379.

Buhler, D.D., M. Liebman, and J.J. Obrycki. 2000. Theoretical and practical challenges to an IPM approach to weed management. Weed Sci. 48:274-280.

Bulson, H.A.J., R.W. Snaydon, and C.E. Stopes. 1997. Effects of plant density on intercropped wheat and field beans in an organic farming system. J. Agr. Sci. 128:59-71.

Caamal-Maldonado, J.A., J.J. Jimenez-Osornio, A. Torres-Barragan, and A.L. Anaya. 2001. The use of allelopathic legume cover and mulch species for weed control in cropping systems. Agron. J. 93:27-36.

Carr, P.M., J.C. Gardener, B.G. Schatz, S.W. Zwinger, and S.J. Guldan. 1995. Grain yield and weed biomass of a wheat-lentil intercrop. Agron. J. 87:574-579.

Chikoye, D. and F. Ekeleme. 2001. Growth characteristics of ten Mucuna accessions and their effects on the dry matter of Imperata cylindrica (L.) Rauesch. Biol. Agric. Hortic. 18:191-201.

Chikoye, D., F. Ekeleme, and U.E. Udensi. 2001. Cogongrass suppression by intercropping cover crops in corn/cassava systems. Weed Sci. 49:658-667.

Collins, A.S., C.A. Chase, W.M. Stall, and C.M. Hutchinson. 2007. Competitiveness of three leguminous cover crops with yellow nutsedge (Cyperus esculentus) and smooth pigweed (Amaranthus hybridus). Weed Sci. 6:613-618.

Collins, A.S., C.A. Chase, W.M. Stall, and C.M. Hutchinson. 2008. Optimum densities of three leguminous cover crops for suppression of smooth pigweed (Amaranthus hybridus). Weed Sci. 56:753-761.

Coultas, C.L., T.J. Post, J.B. Jones, Jr, and Y.P. Hsieh. 1996. Use of velvetbean to improve soil fertilty and weed control in corn production in northern Belize. Commun. Soil Sci. Plan. 27:2171-2196.

Crafts, A.S. 1975. Ecology of Weeds. Pages 67-100 in Modern Weed Control. Berkeley, CA: University of California Press.

Creamer, N.G. and K.R. Baldwin. 2000. An evaluation of summer cover crops for use in vegetable production systems in North Carolina. HortScience 35:600-603.

Daugovish, O., D.C. Thill, and B. Shafii. 2002. Competition between wild oat (Avena fatua) and yellow mustard (Sinapis alba) or canola (Brassica napus). Weed Sci. 50:587-594.

DeHaan, R.L., D.L. Wyse, N.J. Ehlke, B.D. Maxwell, and D.H. Putnam. 1994. Simulation of spring-seeded smother plants for weed control in corn (Zea mays). Weed Sci. 42:35-43.

DeHaan, R.L., C.C. Sheaffer, and D.K. Barnes. 1997. Effect of annual medic smother plants on weed control and yield in corn. Agron. J. 89:813-821.

Dyke, G.V. and A.J. Barnard. 1976. Suppression of couch grass by Italian ryegrass and broad red clover undersown in barley and field beans. J. Agr. Sci. 87:123-126.

Ekeleme, F., I.O. Akobundu, R.O. Fadayomi, D. Chikoye, and Y.A. Abayomi. 2003. Characterization of legume cover crops for weed suppression in the moist savanna of Nigeria. Weed Technol. 17:1-13.

El Saeed, E.A.K. 1967. The smothering effect of Dolichos labalab L. Weed Res. 7:68-69.

Ellis, D.R., K. Guillard, and R.G. Adams. 2000. Purslane as a living mulch in broccoli production. Am. J. Alternative Agr. 15:50-59.

Enache, A.J. and R.D. Ilnicki. 1990. Weed control by subterranean clover (Trifolium subterraneum) used as a living mulch. Weed Technol. 4:534-538.

Gaudet, C.L. and P.A. Keddy. 1988. A comparative approach to predicting competitive ability from plant traits. Nature 334:242-243.

Guritno. B., S.M. Stiompul, and J. van der Heide. 1992. Reclamation of alang-alang land using cover crops on a ultisol in Lampung. Agrivita 15:87-89.

Hartl, W. 1989. Influence of undersown clovers on weeds and on the yield of winter wheat in organic farming. Agr. Ecosyst. Environ. 27:389-396.

Hauggaard-Nielsen. H., P. Ambus, and E.S. Jensen. 2001. Interspecific competition, N use and interference with weeds in pea-barley intercropping. Field Crop Res. 70:101-109.

Haymes, R. and H.C. Lee. 1999. Competition between autumn and spring planted grain intercrops of wheat (Triticum aestivum) and field bean (Vicia faba). Field Crop Res. 62:167-176.

Holt, J.S. 1995. Plant responses to light: a potential tool for weed management. Weed Sci. 43:474-482.

Ilnicki, R.D. and A.J. Enache. 1992. Subterranean clover living mulch: an alternative method of weed control. Agr. Ecosyst. Environ. 40:249-264.

Jordan, N. 1993. Prospects for weed control through crop interference. Ecol. Appl. 3:84- 91.

Kropff, M.J. and H. Walter. 2000. EWRS and the challenges for weed research at the start of the new millennium. Weed Res. 40:7-10.

Lanini, W.T., S.B. Orloff, R.N. Vargas, J.P. Orr, V.L. Marble, and S.R. Grattan. 1991 Oat companion seeding rate effect on alfalfa establishment, yield, and weed control. Agron. J. 83:330-333.

Liebman, M. 1989. Effects of nitrogen fertilizer, irrigation, and crop genotype on canopy relations and yields of an intercrop/weed mixture. Field Crop Res. 22:83-100.

Liebman, M. and E. Dyck. 1993. Crop rotation and intercropping strategies for weed management. Ecol. Appl. 3:92-122.

Liebman, M. and R.H. Robichaux. 1990. Competition by barley and pea against mustard: effects on resource acquisition, photosynthesis and yield. Agr. Ecosyst. Environ. 31:155-172.

Liebman, M. and C.P. Staver. 2001. Crop diversification for weed management. Pages 322-374 in M. Liebman, C.L. Mohler, and C.P. Staver, eds. Ecological Management of Agricultural Weeds. Cambridge, U.K.: Cambridge University Press.

Linares, J., J. Scholberg, K. Boote, C.A. Chase, J.J. Ferguson, and R. McSorley. 2008. Use of cover crop weed index to evaluate weed suppression by cover crops in organic citrus orchards. HortScience 43:27-34.

Lotz, L.A.P., R.M.W. Groeneveld, B. Habekotte, and H. Van Oene. 1991. Reduction of growth and reproduction of Cyperus esculentus by specific crops. Weed Res. 31:153-160.

Lowe, H.J. and K.P. Buchholtz. 1952. Cultural methods for control of quackgrass. Weeds 4:346- 351.

McFadyen, R.E.C. 1998. Biological control of weeds. Annu. Rev. Entomol. 43:369-393.

Medina, J.L. 1999. Interaction of rapeseed (Brassica naphus L.) residue management and smother crops for weed control in corn (Zea mays L.) and soybean (Glycine max (L.) Merr.). Ph.D dissertation. Ames, IA: Iowa State University.

Melander, B., I.A. Rasmussen, and P. Bàrberi. 2005. Integrating physical and cultural methods of weed control - examples from European research. Weed Sci. 53:369- 81.

Mesgaran, M.B., H.R. Mashhadi, M. Khosravi, E. Zand, and H. Mohammad-Alizadeh. 2008. Weed community response to saffron-black zira intercropping. Weed Sci. 56:400-407.

Midya, A., K. Bhattacharjee, S.S. Ghose, and P. Banik. 2005. Deferred seeding of blackgram (Phaseolus mungo L.) in rice (Oryza sativa L.) field on yield advantages and smothering of weeds. J. Agron. Crop Sci. 191:195-201.

Mohler, C.L. and M. Liebman. 1987. Weed productivity and composition in sole crops and intercrops of barley and field pea. J. Appl. Ecol. 24:685-699.

Moynihan, J.M., S.R. Simmons, and C.C. Sheaffer. 1996. Intercropping annual medic with conventional height and semidwarf barley grown for grain. Agron. J. 88:823-828.

Mugabe, N.R., M.E. Sinje, and K.P. Sibuga. 1980. A study of crop/weed competition in intercropping. Pages 96-101 in Proceedings of the Second Symposium on Intercropping in Semi-Arid Areas. Morogoro, Tanzania.

Nelson, W.A., B.A. Kahn, and B.W. Roberts. 1991. Screening cover crops for use in conservation tillage systems for vegetables following spring plowing. HortScience 26:860-862.

Obiefuna, J.C. 1989. Biological weed control in plantains (Musa AAB) with egusi melon (Colocynthis citrullus L.). Biol. Agric. Hortic. 6:221-227.

Oladokun, M.A.O. 1978. An assessment of cultural weed control methods in Coffea canephora (var. Quillou). Pages 362-365 in Weeds and Their Control in the Humid and Sub-Humid Tropics Ibadan, Nigeria: IITA.

Olasantan, F.O. 2007. Effect of population density and sowing date of pumpkin on soil hydrothermal regime, weed control and crop growth in a yam-pumpkin intercrop. Exp. Agr. 43:365-380.

Paine, L.K. and H. Harrison. 1993. The historical roots of living mulch and related practices. HortTechnology 3:137-143.

Palada, M.C., S. Ganser, R. Hofstetter, B. Volak, and M. Culik. 1983. Association of interseeded legume cover crops and annual row crops in year-round cropping systems. Pages 193-213 in W. Lockeretz, ed. Environmentally Sound Agriculture. New York: Praeger Publishers.

Peters, R.A. 1961. Legume establishment as related to the presence or absence of an oat companion crop. Agron. J. 53:195-198.

Phatak, S.C., R.L. Bugg, D.R. Sumner, J.D. Gay, K.E. Brunson, and R.B. Chalfant. 1991. Cover crop effects on weeds, diseases, and insects of vegetables. Pages 153-154 in W.L. Hargrove, ed. Cover Crops for Clean Water. Ankeny, IA: Soil and Water Conservation Society.

Pimentel, D., C. Harvey, P. Resosudarmo, K. Sinclair, D. Kurz, M. McNair, S. Crist, L. Shpritz, L. Fitton, R. Saffouri, and R. Blair. 1995. Environmental and economic costs of soil erosion and conservation benefits. Science 267:1117-1123.

Poggio, S.L. 2005. Structure of weed communities occurring in monoculture and intercropping of field pea and barley. Agr. Ecosyst. Environ. 109:48-58.

Poku, J.A. and I.O. Akobundu. 1984. Control of Imperata cylindrica. International Institute of Tropical Agriculture (IITA) Annual Report. Ibadan, Nigeria: IITA. p. 174.

Rao, M.R. and S.V.R. Shetty. 1976. Some biological aspects of intercropping systems on crop-weed balance. Indian J. Weed Sci. 8:32-43.

Regnier, E.E. and R.R. Janke. 1990. Evolving Strategies for Managing Weeds. Pages 174-202 in C.A. Edwards, R. Lal, P. Madden, R.H. Miller, and G. House, eds. Sustainable Agricultural Systems. Boca Raton, FL: St. Lucie Press.

Robbins, W.W., A.S. Crafts, and R.N. Raynor. 1942. Weed Control, 1st ed. New York: McGraw-Hill Book Company, Inc.

Roder, W., S. Maniphone, and B. Keoboulapha. 1998. Pigeon pea for fallow improvement in slash-and-burn systems in the hills of Laos? Agroforest. Syst. 39:45-57.

Ross, S.M., J.R. King, R.C. Izarrualde, and J.T. O'Donovan. 2001. Weed suppression by seven clover species. Agron. J. 93:820-827.

Samson, R.A. 1991. The weed suppressing effects of cover crops. Pages 11-22 in Proceedings of the Fifth Annual REAP Conference. Ste-Anne-de-Bellevue, Quebec, Canada: Macdonald College.

Sangakkara, U.R., P.R.S.D. Bandaranayake, D.N.K. Weerasekera, and P. Stamp. 2006. Interseasonal cropping - its potential for managing weeds in the Asian tropics. J. Plant Dis. Protect. 20:921-927.

Sarma, M.K., S.S. Narwal, and J.S. Yadava. 1992. Smothering effect of Brassica species on weeds. Pages 51-55 in Proceedings of the First National Symposium on Allelopathy in Agroecosystems. Hisar, India: Haryana Agricultural University.

Saucke, H. and K. Ackermann. 2006. Weed suppression in mixed cropped grain peas and false flax (Camelina sativa). Weed Res. 46:453-461.

Schonbeck, M., J. Browne, G. Deziel, and R. DeGregorio. 1991. Comparison of weed biomass and flora in four cover crops and a subsequent lettuce crop on three New England organic farms. Biol. Agric. Hortic. 8:123-143.

Schoofs, A. and M.H. Entz. 2000. Influence of annual forages on weed dynamics in a cropping system. Can. J. Plant Sci. 80:187-198.

Sengupta, K., K.K. Bhattacharyya, and B.N. Chatterjee. 1985. Intercropping of upland rice with blackgram [Vigna mungo (L.)]. J. Agr. Sci. 104:217-221.

Severino, F.J. and P.J. Christoffoleti. 2004. Weed suppression by smother crops and selective herbicides. Sci. Agric. 61:21-26.

Sheaffer, C.C., J.L. Gunsolus, J.G. Jewett, and S.H. Lee. 2002. Annual Medicago as a smother crop in soybean. J. Agron. Crop Sci. 188:408-416.

Shetty, S.V.R. and A.N. Rao. 1979. Weed-management studies in sorghum/pigeonpea and pearl millet/groundnut intercrop systems - some observations. Pages 238-248 in Proceedings of the International Workshop on Intercropping. Hyderabad, India: ICRISAT.

Sinha, S.P., S.M. Prasad, and S.J. Singh. 2005. Nutrient utilization by winter maize (Zea mays) and weeds as influenced by integrated weed management. Indian J. Agron. 50:303-304.

Sivakumar, C. and P. Subbian. 2002. Growth and yield of irrigated cotton (Gossypium hirsutum) as influenced by different chemical and non-chemical weed- management practices. Indian J. Agron. 47:123-129.

Tamado, T. and P. Milberg. 2004. Control of parthenium (Parthenium hysterophorus) in grain sorghum (Sorghum bicolor) in the smallholder farming system in eastern Ethiopia. Weed Technol. 18:100-105.

Teasdale, J.R. 1993. Reduced-herbicide weed management systems for no-tillage corn (Zea mays) in a hairy vetch (Vicia villosa) cover crop. Weed Technol. 7:879-883.

Teasdale, J.R. 1998. Cover crops, smother plants, and weed management. Pages 247- 270 in J.L. Hatfield, D.D. Buhler, and B.A. Stewart, eds. Integrated Weed and Soil Management. Chelsea, MI: Sleeping Bear Press.

Teasdale, J.R. and C.S. Daughtry. 1993. Weed suppression by live and desiccated hairy vetch (Vicia villosa). Weed Sci. 41:207-212.

Turner, R.J., G. Davies, H. Moore, A.C. Grundy, and A. Mead. 2007. Organic weed management: a review of the current UK farmer perspective. Crop Prot. 26:377- 382.

Udensi, U.E., I.O. Akobundu, A.O. Ayeni, and D. Chikoye. 1999. Management of cogongrass (Imperata cylindrica) with velvetbean (Mucuna pruriens var. utilis) and herbicides. Weed Technol. 13:201-208.

Unamma, R.P.A., L.S.O. Ene, S.O. Odurukwe, and T. Enyinnia. 1986. Integrated weed management for cassava intercropped with maize. Weed Res. 26, 9-17.

Valverde, B.E., A. Merayo, C.E. Rojas, and T. Alvarez. 1995. Interaction between a cover crop (Mucuna sp.), a weed (Rottboellia cochinchinensis) and a crop (maize). Pages 197-200 in Proceedings of the Brighton Crop Protection Conference Weeds 1995. Brighton, U.K.: Brighton Crop Protection Council.

Wang, G., J.D. Ehlers, E.J. Ogbuchiekwe, S. Yang, and M.E. McGiffen, Jr. 2004. Competitiveness of erect, semierect, and prostrate cowpea genotypes with sunflower (Helianthus annuus) and purslane (Portulaca oleracea). Weed Sci. 52:815-820.

Wang, G., M.E. McGiffen, Jr, J.D. Ehlers, and E.C.S. Marchi. 2006. Competitive ability of cowpea genotypes with different growth habit. Weed Sci. 54:775-782.

Wang, G., M.E. McGiffen, Jr, and E.J. Ogbuchiekwe. 2008. Crop rotation effects on Cyperus rotundus and C. esculentus population dynamics in southern California vegetable production. Weed Res. 48:420-428.

Weiling, K.Y. 1982. Weeds, intercropping, and mulch in the temperate zones and the tropics - some ecological implications for low-technology agriculture. Ph.D dissertation. Ann Arbor, MI: University of Michigan.

Wortmann, C.S. 1993. Contribution of bean morphological characteristics to weed suppression. Agron. J. 85:840-843.

Zuofa, K. and N.M. Tariah. 1992. Effects of weed control on maize and intercrop yields and net income of small-holder farms, Nigeria. Trop. Agr. 69:167-170.

Zuofa, K., N.M. Tariah, and N.O. Isirimah. 1992. Effects of groundnut, cowpea and melon on weed control and yields of intercropped cassava and maize. Field Crop Res. 28:309-314.

Table 2.1. Rating system used to evaluate publication quality.

Category Maximum points Points per criteria Criteria

Journal impact factor 1 0.25 For each percent quartile of impact factor, i.e. a

journal with an impact factor in the second quartile

would receive 0.5 points

Statistics 2 1 Presence of statistics and appropriate analyses

42 Experimental design 2 0.66 Replication within a trial, repetition of trial, sound

methods

Descriptions 1 0.33 Species names, variety names, site descriptions

Treatments 2 0.5 Species characteristics, comparison of cropping

systems, soil management and planting management

Data 2 0.5 Measures of crop growth, photosynthesis/light,

competition, abiotic factors

Table 2.2. Factors of weed suppression.

Weed suppression factor category Weed suppression factor

Environment and climate Climatic conditions, land geography, pest infestation, planting conditions, planting

season, site characteristics, year

Soil inputs and properties Fertilization (N-P-K), irrigation, liming, soil nitrogen, soil properties, soil temperature

Crop and soil management Land preparation, planting density, planting time, spatial arrangement

Crop characteristics Allelopathy, smother crop life cycle, environmental adaptation, smother crop growth

43 habit, main crop species, niche occupation, smother crop species, smother crop variety

Smother crop growth and Smother crop biomass, smother crop emergence, smother crop ground cover, smother

development crop height, smother crop nitrogen accumulation, smother crop nodulation, smother crop

survival

Photosynthesis and leaf development Canopy structure, green surface area, LAI, leaf area duration, leaf size, leaf weight,

photosynthetic rate, reduction in light

Continued

Table 2.2 continued

Weed suppression factor category Weed suppression factor

Weed competition Weed-crop complementarity, weed density, weed emergence

Crop interactions Component crop complementarity, intercropping, nitrogen competition, nutrient

competition, time of competition

Table 2.3. Description of smother cropping research and journals reviewed.

Frequency

Descriptor Name N %

Journal Weed Science 9 10.1

Agronomy Journal 8 8.99

Weed Technology 7 7.87

Weed Research 6 6.74

Research focus Intercropping 31 34.8

Cover cropping 27 30.3

Smother cropping 11 12.4

Types of main crops a Grain 41 57.8

Vegetable 7 9.86

Grain legume 7 9.86

Fruit 6 8.45

Edible roots 4 5.63

Forages 3 4.23 a Main crop species is considered the cash crop or crop of focus for weed control in intercropping situations only.

Table 2.4. Locations of smother cropping research.

Frequency

Descriptor Name N %

Continent North America 43 45.3

Africa 22 23.2

Asia 14 14.7

Europe 10 10.5

Latin America 6 6.32

African location Nigeria 12 12.6

Ethiopia 4 4.21

Asian location India 9 9.47

North American California, U.S.A. 6 6.32 location

Minnesota, U.S.A. 4 4.21

European location England 3 3.16

Switzerland 2 2.11

Latin American Mexico 2 2.11 location

Table 2.5. Most commonly studied smother crops.

Frequency

Descriptor Name N %a

Smother crop type Annual legume 179 54.7

Annual grass 53 16.2

Perennial legume 51 15.6

Annual forb 32 9.79

Annual legumes Cowpea [Vigna unguiculata (L.) Walp.] 22 6.67

Velvetbean [Mucuna pruriens (L.) DC.] 13 3.94

Crimson clover (Trifolium incarnatum L.) 12 3.64

Winter vetch (Vicia villosa Roth) 10 3.03

Annual grasses Rye (Secale cereale L.) 13 3.94

Sorghum-sudangrass [Sorghum bicolor (L.) 5 1.52

Moench x Sorghum sudanense (P.) Stapf.]

Oat (Avena sativa L.) 5 1.52

Lopsided oat (Avena strigosa Schreb.) 5 1.52

Perennial legumes Red clover (Trifolium pratense L.) 8 2.42

Pigeonpea [Cajanus cajan (L.) Millsp.] 6 1.82

Tropical kudzu (Pueraria phaseoloides Roxb. 6 1.82

Benth.)

White clover (Trifolium repens L.) 6 1.82

Continued

Table 2.5 continued

Frequency

Descriptor Name N %a

Annual forbs Rape (Brassica napus L.) 4 1.21

Cultivated radish (Raphanus sativus L.) 4 1.21

Buckwheat (Fagopyrum esculentum Moench) 3 0.91

Colocynth [Citrullus colocynthis (L.) Schrad.] 3 0.91 a Percent of total smother crops used in all reviewed publications.

Table 2.6. Weed groups most commonly studied in research on smother crops.

Frequency

Weeds N %a

Total weeds 63 36.4

Broadleaves 62 36.1

Grasses 47 27.3

Target weed species

Common lambsquarters (Chenopodium album L.) 10 5.81

Common purslane (Portulaca oleracea L.) 3 1.74

Speargrass [Imperata cylindrica (L.) Beauv.] 7 4.07

Yellow nutsedge (Cyperus esculentus L.) 4 2.33 a Percent of all weed groups studied.

Table 2.7. Effectiveness of categories of the factors of weed suppression

Rating Frequency

Weed suppression factor category WSa N %

Crop growth and development 6.6 a 499 21.0

Weed competition 6.3 b 123 5.2

Photosynthesis and leaf development 6.2 b 132 5.6

Environment and climate 6.2 b 204 8.6

Soil inputs and properties 6.1 b 95 4.0

Crop characteristics 5.8 c 753 32.0

Crop and soil management 5.6 d 205 8.7

Crop interactions 5.2 e 350 15.0

Abbreivations: WS, average weed suppression calculated for studies that identify the specific weed suppression category as the underlying factor. a Adjusted least squares means followed by the same letter within a column are not significantly different (P < 0.05).

Table 2.8. Rating and frequency of occurrence of most and least effective weed suppression factors.a

Factors of weed suppression Rating Frequency

Factor Category WSb PQ N %

Irrigation Soil inputs and properties 7.6 a 7.1 9 0.30

Fertilization (N-P-K) Soil inputs and properties 7.5 a 6.2 6 0.20

Soil nitrogen Soil inputs and properties 7.3 a 6.9 35 1.2

Smother crop life Crop characteristics 7.1 a 6.2 113 3.8 cycle

Land preparation Crop and soil management 7.0 a 6.7 54 1.8

Smother crop growth Crop characteristics 4.7 b 6.2 55 1.9 habit

Smother crop Crop growth and 4.5 b 6.5 49 1.7 emergence development

Spatial arrangement Crop and soil management 4.4 b 6.5 14 0.47

Main crop species Crop characteristics 4.0 bc 6.0 10 0.34

Soil properties Soil inputs and properties 3.7 c 7.2 27 0.91

Abbreviations: WS, average weed suppression calculated for factors of weed suppression; PQ, average rating for quality of publication. a Only factors contributing to weed suppression in two or more publications are described.

b Adjusted least squares means followed by the same letter within a column are not significantly different (P < 0.05).

Table 2.9. Highest and lowest rated smother cropping systems by species.a

Cropping System Rating Frequency

Smother Crop Main Crop WSb PQ N %

Italian ryegrass (Lolium multiflorum Lam) None 8.0 a 5.7 4 0.28

Subterranean clover (Trifolium subterraneum L.) Maize (Zea mays L.) 7.7 a 6.8 40 2.8

Velvetbean [Mucuna cochinchinensis Lour. A. None 7.3 a 6.5 9 0.63

Chev.]

Barley (Hordeum vulgare L.) None 6.6 ab 6.4 4 0.28

Rye (Secale cereale L.) None 6.1 b 6.5 20 1.4

Mung bean [Vigna radiata (L.) Wilczek] Sorghum [Sorghum bicolor (L.) Moench.] 1.4 c 3.2 5 0.35

Cowpea [Vigna unguiculata (L.) Walp.] Sorghum [Sorghum bicolor (L.) Moench] 1.3 c 5.5 5 0.35

Sweet potato [Ipomoea batatas (L.) Lam.] Maize (Zea mays L.) 1.1 c 5.2 5 0.35

Cowpea [Vigna unguiculata (L.) Walp.] Maize (Zea mays L.) 1.1 c 5.6 4 0.28

Asian pigeonwings (Clitoria ternatea L.) None 0.58 c 6.5 5 0.35

Abbreviations: WS, average weed suppression calculated for each combination of smother crop and main crop; PQ, average

rating for quality of publication.

a Species listed suppress weeds in two or more publications.

b Adjusted least squares means followed by the same letter within a column are not significantly different (P < 0.05).

Table 2.10. Most and least effective smother cropping systems that target specific weeds.a

Cropping System Rating Frequency

Weed Smother Crop Main Crop WSb PQ N %

Ivyleaf morningglory Subterranean clover (Trifolium Maize (Zea mays L.) 9.8 a 6.1 12 0.84

(Ipomoea hederacea subterraneum L.)

Jacq.)

Speargrass [Imperata Velvetbean (Mucuna None 8.0 b 7.2 5 0.35

cylindrica (L.) Beauv.] cochinchinensis Lour. A.

Chev.)

Speargrass [Imperata Velvetbean [Mucuna pruriens Maize (Zea mays L.) 7.7 bc 6.4 2 0.14

cylindrica (L.) Beauv.] (L.) DC. var. utilis (Wright)

Burck.]

Fall panicum (Panicum Subterranean clover (Trifolium Maize (Zea mays L.) 6.7 cd 6.1 10 0.70

dichotomiflorum subterraneum L.)

Michx.)

Continued

Table 2.10 continued

Cropping System Rating Frequency

Weed Smother Crop Main Crop WSb PQ N %

Speargrass [Imperata Cowpea [Vigna Cassava (Manihot esculenta 6.6 cd 7.5 5 0.35

cylindrica (L.) Beauv.] unguiculata (L.) Walp.] Crantz.) + Maize (Zea mays L.)

Smooth pigweed (Amaranthus Sunn hemp (Crotalaria None 6.4 d 7.8 10 0.70

hybridus L.) juncea L.)

Amaranthus spp., wild oat Rape (Brassica napus None 0 5.7 16 1.1

(Avena fatua L), field L.)

bindweed (Convolvulus

arvensis L.), Pennsylvania

smartweed (Polygonum

pensylvanicum L.), giant

foxtail (Setaria faberi Herrm.)

Continued

Table 2.10 continued

Cropping System Rating Frequency

Weed Smother Crop Main Crop WSb PQ N %

Quackgrass [Agropyron repens (L.) P. Garden pea (Pisum Barley (Hordeum 0 7.2 2 0.14

Beauv.] sativum L.) vulgare L.)

Pineappleweed [Chamomilla suaveolens Winter vetch (Vicia None 0 7.2 6 0.42

(Pursh.) Rydb.], purple nutsedge villosa Roth)

(Cyperus rotundus L.), barnyardgrass

[Echinochloa crus-galli (L.) Beauv.]

Common lambsquarters (Chenopodium Triticale (x Triticosecale None 0 6.7 24 1.7

album L.), wild buckwheat (Polygonum rimpaui Wittm.)

convolvulus L.), green foxtail [Setaria

viridis (L.) P. Beauv.]

Abbreviations: WS, average weed suppression calculated for each combination of weed, smother crop, and main crop; PQ,

average rating for quality of publications.

a Cropping system and weed combinations listed occur in two or more experimental situations.

b Adjusted least squares means followed by the same letter within a column are not significantly different (P < 0.05).

Table 2.11. Interaction of weed types and highest and lowest rated suppression factors.a,b

Rating Frequency

Weed Type Weed suppression factor WS PQ Nc %

Annual Broadleaf Niche occupation 9.3 a 6.8 12 0.84

Land preparation 8.5 a 6.8 11 0.77

Smother crop life cycle 7.9 ab 6.3 2 0.14

Irrigation 7.6 b 6.7 9 0.62

Smother crop biomass 7.3 b 6.6 59 4.1

Smother crop species 7.2 b 5.0 34 2.3

Annual Grass Allelopathy 6.9 a 6.1 15 1.0

Smother crop variety 6.5 ab 5.0 5 0.35

Niche occupation 6.3 ab 6.8 4 0.28

Weed-crop complementarity 6.2 b 4.0 16 1.1

Smother crop biomass 6.1 b 6.3 28 1.9

Smother crop species 5.8 b 4.9 18 1.3

Perennial Grass Smother crop variety 6.9 a 6.9 20 1.4

Smother crop emergence 6.8 ab 6.9 6 0.42

Smother crop ground cover 6.7 ab 6.9 45 3.1

Allelopathy 6.4 abc 7.1 12 0.84

Land preparation 6.3 bc 6.0 10 0.69

Smother crop life cycle 6.2 c 5.9 12 0.84

Abbreviations: WS, average weed suppression calculated for each combination of weed and weed suppression mechanism; PQ, average rating for quality of publications. a Weed and mechanism interactions occur in two or more publications. b Perennial broadleaves are excluded as cropping scenarios did not suppress it. c Number of times scenario occurs out of all analyzed experimental situations.

Chapter 3: Evaluation of Tef as a Smother Crop during Transition to Organic

Management

Stephanie Wedryk and John Cardina*

Management of weeds is often a barrier to conversion from conventional to organic agriculture. Tef [Eragrostis tef (Zucc.) Trotter ] is a C-4 annual cereal that is valued for its small seeds, rapid establishment, and wide adaptation. The objective of this study was to evaluate tef as a smother crop for management of weeds during transition to organic production. Greenhouse and field trials were conducted in 2008 and 2009 to evaluate the growth of eight tef varieties and their effect on Canada thistle and annual weeds. In greenhouse studies, tef decreased the biomass of Canada thistle shoots and rhizomes 44 to 74%, depending on variety. Tef varieties VA-T1, Corvalis, and Excalibur suppressed Canada thistle more than other varieties. In field studies, tef varieties suppressed annual weeds by 35% to 54%, but there were no differences among varieties.

Canada thistle growth was suppressed 73% by tef in 2008 and 37% in 2009, a year of cooler temperatures and unseasonal rainfall. All tef varieties except „Pharaoh‟ were competitive with Canada thistle in the field experiment.

* Graduate Research Associate and Professor, Department of Horticulture and Crop Science, The Ohio State University, Wooster, OH 44691. Corresponding author‟s E-mail: [email protected]. 61

Nomenclature: Tef, Eragrostis tef (Zucc.) Trotter; Canada thistle, Cirsium arvense (L.)

Scop.

Keywords: smother crop, organic weed management

Organic producers use the required three-year transition from conventional to organic management to improve soil fertility while suppressing weeds and other pests in preparation for growing certified organic crops (Hanson et al. 2004). During transition, farmers must adopt organic certification rules, which prohibit the use of synthetic herbicides and stipulate the use of biological, cultural, and mechanical controls (Greene and Kremen 2003). Since premium organic prices cannot be realized during this transition period, growers face contradictory goals of minimizing inputs while reducing potential weed populations in subsequent crops.

Weed management during transition is especially important in fields infested with species that are very difficult to control, such as rapidly growing annuals and deeply rooted perennials. For example, Canada thistle (Cirsium arvense (L.) Scop.) is regarded by farmers as one of the most troublesome perennial weeds in organic agriculture

(Beveridge and Naylor 1999; Turner et al. 2007; Verschwele and Häusler 2004). Canada thistle can reproduce by seeds or underground propagative roots (hereafter roots) that have been found up to 6.75 m below the soil surface (Donald 1994; Evans 1984). Organic farmers rely mostly on mechanical management practices, which are often not effective against all Canada thistle roots. The action of machinery cuts the roots into segments from which Canada thistle can regenerate (Evans 1984). Growth of Canada thistle in early spring comes at the expense of carbohydrates stored in rhizomes (Gustavsson

1997). Root reserves and bud numbers are at their lowest between May 15 and July 15 when Canada thistle is flowering (McAllister and Haderlie 1985).

Smother cropping is an alternative weed management technique that is well adapted to the three-year transition required for organic production (Bàrberi 2002).

Smother cropping involves the use of a living plant to reduce the growth, development, or reproduction of weeds predominantly through competition for resources (Teasdale 1998).

Many successful horticultural weeds, including Canada thistle, are shade intolerant, producing less robust growth when shaded by neighboring plants (Donald 1994).

Therefore, a fast growing smother crop that competes successfully for light, when annuals are in their logarithmic growth phase or when root carbohydrate reserves in perennials are low, might suppress emerging annual seedlings or perennial shoots. For example, a smother crop of live hairy vetch (Vicia villosa Roth.) reduced annual and perennial weed biomass and density more effectively than desiccated hairy vetch

(Teasdale and Daughtry 1993). Smother crops have been used to suppress perennial weeds and prevent vegetative reproduction (Regnier and Janke 1990). In the tropics,

Udensi et al. (1999) found that a smother crop of velvetbean (Mucuna pruriens (L.) DC. var. utilis) reduced speargrass (Imperata cylindrica (L.) Beauv.) shoot density.

Leguminous smother crops have also reduced the number and weight of yellow nutsedge

(Cyperus esculentus L.) tubers (Collins et al. 2007). Spring planted grass smother crops such as rye (Secale cereale L.) and triticale (x Tritosecale rimpaui Wittm.) have reduced weed density without additional inputs (Barnes and Putnam 1983; Schoofs and Entz

2000). Sudangrass (Sorghum sudanese (Piper) Stapf) smother crops reduced Canada thistle shoot density and biomass (Bicksler and Masiunas 2009).

We have been evaluating tef (Eragrostis tef (Zucc.) Trotter), a C4 annual cereal commonly grown in Ethiopia, as a potential smother crop for use during transition from conventional to organic production. Characteristics of tef that makes it a viable candidate smother crop are rapid establishment, drought tolerance, and lack of significant disease issues (DeHaan et al. 1994; Ketema 1997). Germination rates of tef seeds are greater than

90% within 24 h of planting when daytime temperatures are 25C or greater (Debelo

1992). For use in a cropping system where perennials like Canada thistle are present, we expected tef to grow rapidly during the time when Canada thistle roots are low in carbohydrate reserves during late spring and early summer in the midwestern U.S. Tef is commonly planted at high seeding rates (up to 55 kg/ha) to provide densities of 30,000 seedlings m-2, which are expected to compete effectively with annual weeds and eventually aid in shading of Canada thistle (Ketema 1997; Yu et al. 2007). Studies of quantitative traits have demonstrated significant genetic variation among germplasm accessions and potential for improvement as a grain crop; however, few varieties are recognized and none have been evaluated as smother crops (Adnew et al. 2005). The objective of this study was to evaluate available varieties of tef as smother crops for annual weed and Canada thistle suppression in greenhouse and field experiments.

Materials and Methods

Greenhouse Experiment. A greenhouse experiment was conducted at the Ohio State

University, Wooster, Ohio, during the spring and fall of 2009 to determine the ability of smother crops to suppress Canada thistle shoot growth. Canada thistle roots were collected in April 2009 and October 2009 at the Schaffter Farm near Wooster, Ohio.

Roots were stored under moist conditions at 5C until used in this experiment. Root pieces (7.6-10.2 cm) were weighed and visible bud number was measured before planting.

The experiment was arranged as a 2x10 factorial with two Canada thistle root planting depths (7.6 and 15.2 cm) and ten smother cropping treatments with three replications. Smother cropping treatments included eight tef varieties, one sorghum- sudangrass, and a non-treated control.. The tef varieties used were available commercially in the U.S. for potential use by organic producers, and represent tef seed production from different regions in the U.S. (Table 3.1). Sorghum-sudangrass was included because it has recently been shown to be effective as a smother crop against

Canada thistle (Bicksler and Masiunas 2009).

Plastic pots of 38 cm depth and 23 cm diameter were filled with a 1:1 mix of commercial potting media1 and Wooster silt loam soil. Three Canada thistle root pieces were planted at the respective planting depth and then covered with the soil mix and watered. Tef and sorghum-sudangrass were seeded at an equivalent to the recommended field seeding rate of 30 kg ha-1 (Assefa et al. 2001; Kefyalew et al. 2000). A small amount of soil was mixed with 0.83 g of tef seeds and pressed into the top of the soil mix to ensure uniform seeding. The average number of tef seeds in 0.83 g was 2530. Sorghum-

66 sudangrass seeds (0.83 g=37 seeds) were planted into the designated pots at a depth of

2.5 cm and then covered with a thin layer of soil. Media in non-treated pots were also covered with a thin layer of soil. The smother crops and Canada thistle roots were allowed to grow under greenhouse conditions for 4 weeks with 14 h/10 h light/dark and

24C/18C day/night cycles. Pots were irrigated with water for 1 min. twice daily and re- randomized within blocks weekly. The experiment was repeated under the same conditions.

Emergence of Canada thistle shoots was measured daily until the completion of the experiment. Height of Canada thistle shoots and smother crop varieties was measured weekly. After four weeks, aboveground biomass of smother crops and Canada thistle were weighed, dried at 55C for 72 hours, and re-weighed. Belowground roots of Canada thistle were removed from soil, rinsed in H2O, weighed, and the number of buds counted.

Roots were dried at 55C for 72 hours and re-weighed.

Field Experiment. A field experiment was conducted at the Ohio State University

Schaffter Farm near Wooster, Ohio in 2008 and 2009 to evaluate commercial varieties of tef as smother crops. The soil type at the site is a Wooster silt loam with pH of 6.5 and

2.9% organic matter. Percent cover of Canada thistle was visually assessed, before treatments were imposed, on June 9, 2008, and June 5, 2009. The field was disked and prepared for planting June 10, 2008, and June 11, 2009. Treatments were arranged in a randomized complete block design with six replications. Treatments consisted of eight tef varieties and a non-treated control. In 2009, plots were located in a different part of the

67 same field as the previous year. Each plot was 2.3 m2 and in 2008, plots with low Canada thistle density were augmented with Canada thistle roots. The roots had been propagated in the greenhouse from shoots collected during April 2008. Three roots that were 10 to 15 cm in length (average weight 3.88 g), with at least one adventitious bud per root piece, were planted at a depth of 15 cm. Varieties of tef used were the same as in the greenhouse study. Tef seeds were broadcast seeded at a rate of 30 kg ha-1 (approximately

16,000 seeds m-2) by hand on June 11, 2008, and June 15, 2009. After seeding, a roller was passed over each plot to help insure seed-soil contact.

Canada thistle shoots were counted on June 26, 2008 and 2009. The percent cover and height of tef and Canada thistle were measured weekly for 8 weeks, starting 10 days after planting. Percent cover of annual weeds was also recorded weekly for 8 weeks beginning 10 days after planting. Total plant biomass (tef plus annual weeds and Canada thistle) was harvested from 0.093 m2 quadrat per plot on August 22, 2008, and between

August 27 and October 6, 2009. Plots were harvested when tef seeds were 80% filled.

Due to differences in days to maturity for individual plots, plots in 2009 were harvested on different days. Harvested materials were separated into tef, Canada thistle, and annual weeds, weighed and dried at 55C for 72 hours, and weighed again. The most common annual weeds present were common lambsquarters (Chenopodium album L.), common ragweed (Ambrosia artemisiifolia L.), yellow foxtail [Setaria glauca (L.) Beauv.], and green foxtail [Setaria viridis (L.) Beauv.].

Data Analysis. Greenhouse trial data were combined for the two repetitions of the study.

Data were first analyzed for effect of root planting depth on Canada thistle shoot responses using ANOVA in SAS2. When the interaction of root planting depth and smother crop treatments was significant (P < 0.05), data from the two planting depths were analyzed separately. Field experiment data were analyzed separately by year.

Harvest data from both trials were subjected to ANOVA in SAS and means separated by

Fisher‟s Protected Least Significant Difference (LSD p < 0.05). Data that did not meet the assumptions of ANOVA were log transformed. Transformed data that did not meet the assumptions of ANOVA were analyzed using Kruskal-Wallis analysis of variance.

Pearson correlation coefficients were calculated for biomass measures. Percent cover and height data were subjected to repeated measures analysis using PROC MIXED in SAS.

Least squares means of percent cover and height were divided by the number of days between the initial and final sampling times to determine the rate of ground cover spread and vertical growth. Differences between means were determined using the PDIFF optionin PROC MIXED.

Crop harvest data, final percent cover, and final height data from the field trial were plotted in SigmaPlot3 as dependent variables with initial Canada thistle rating and shoot count as the independent variables. Crop harvest data and final height data from the greenhouse trial were plotted in SigmaPlot as dependent variables with initial Canada thistle root weight, bud number and emergence as the independent variables. Final

Canada thistle measurements in the field trial were subjected to ANCOVA with initial

Canada thistle rating and shoot count as the covariates. Final Canada thistle

69 measurements in the greenhouse trial were similarly subjected to ANCOVA with initial

Canada thistle root weight, bud number and emergence as the covariates.

Results and Discussion

Greenhouse Experiment. Dry matter biomass of Canada thistle was not different between root planting depths; therefore, data for root planting depths are combined. Total biomass of Canada thistle decreased 44 to 74% in pots where tef was grown compared with the non-treated control (Figure 3.1). Sorghum-sudangrass did not suppress Canada thistle growth.Three tef varieties (Corvalis, Excalibur, and VA-T1) reduced Canada thistle biomass more than sorghum-sudangrass. Biomass of Canada thistle was lowest (0.4 g pot-

1) when tef variety VA-T1 was grown and highest in the non-treated control (1.5 g pot-1) and sorghum-sudangrass variety Special Effort (0.93 g pot-1). Tef seed weight is about

1.5% that of sorghum-sudangrass, and since the same weight of seeds was planted for both species, the number of individual plants in tef treatments was much greater than in the sorghum-sudangrass pots. Results suggest that the higher density of tef smother crops

(~2500 seedlings per pot) compared with sorghum-sudangrass (37seedlings per pot) may have had a greater effect on the ability of Canada thistle to compete with smother crops for light. Canada thistle is known to be allelopathic and the exudation of allelochemicals from Canada thistle may have affected the growth of tef and sorghum-sudangrass

(Donald 1994).

The biomass of smother crops was 58% lower, averaged over smother crop variety, where Canada thistle roots were planted 7.6 cm deep compared with 15 cm deep

(Figure 3.2). The tef varieties Corvalis, VA-T1, and Pharaoh had the greatest biomass

(9.2 to 9.6 g pot-1) and the sorghum-sudangrass produced the least amount of biomass

(2.8 g pot-1) at the 7.6 cm planting depth (Table 3.2). Biomass of plants competing with weeds is highly correlated with reduction of weed biomass and a proxy for competitive ability (Gaudet and Keddy 1988). There were no differences in biomass of smother crops at the 15 cm planting depth. At the 15 cm planting depth, the competition from emerging

Canada thistle shoots was delayed compared to the 7.6 cm planting depth. This delay allowed smother crop populations to be established before Canada thistle could exert competitive pressure. Similarly, root growth of smother crops would not have been affected by more deeply buried Canada thistle roots during early stages of establishment.

The differences in biomass and height at the two different planting depths demonstrate the effect that competing Canada thistle can have on smother crop varieties. Results suggest that establishment of smother crop populations in the field before emergence of

Canada thistle shoots from underground roots is important for the success of smother crops to suppress Canada thistle.

The heights of Canada thistle shoots and smother crops were affected by planting depth of Canada thistle roots (Figure 3.3). Canada thistle shoots were 3.5 cm shorter and smother crops averaged 4.5 cm taller where roots were planted 15 cm deep compared to those 7.6 cm deep (Figure 3.3). Canada thistle shoots from roots buried at 7.6 cm reached the soil surface more quickly and competed with smother crops earlier than shoots from

71 roots planted at 15 cm depth. Canada thistle shoots were shortest (6.3 to 8.6 cm) in the non-treated control and tallest (9.8 to 15.9 cm) in pots with the tef variety Dessie (Table

3.2). Competition from Canada thistle shoots from roots planted at 7.6 cm apparently reduced the growth of smother crops. Canada thistle shoots grew taller in competitive smother crop treatments, presumably to reach light above the canopy of the smother crops. Canada thistle shoots in tef smother crops appeared to be lighter green with less pronounced thistles and weaker stems, suggesting competition for light and reduced photosynthesis in competition with the smother crops (Moore 1975).

The height of smother crops can be an effective measure of the ability of a variety to compete effectively with Canada thistle shoots (Gaudet and Keddy 1988). Weed growth during early developmental stages can decrease the relative growth rate (Wang et al. 2006). The Special Effort variety of sorghum-sudangrass had the highest final height

(33.0 and 45.8 cm for 7.6 and 15 cm root planting depths, respectively) (Table 3.2). Tef varieties Pharaoh and Ivory had the highest and lowest final height, respectively, a difference of 4 cm for treatments with roots planted 7.6 cm deep. There were no differences in the heights of Canada thistle shoots and smother crop varieties at the 15 cm root planting depth. The onset of later competition by Canada thistle at the 15 cm planting depth may have allowed less competitive smother crop varieities to grow similarly to more competitive varieties. Regressions between the growth of smother crop varieties and initial Canada thistle root mass and bud number and emergence were not significant (data not shown). The differences among smother crops are not the result of

72 the existing Canada thistle population and biology of roots, but rather the differences between varieties and Canada thistle rhizome depth.

Field Experiment. Monthly maximum temperatures in July and August were about 11% and 5% lower than the 20-year mean in 2009 (Table 3.3). August maximum temperatures in 2008 were 5% lower than the 20-year mean (Table 3.3). Monthly precipitation was

43% higher in June and July 2008 compared to the 20-year mean with a subsequent decrease in August rainfall of 62% (Table 3.3). In 2009, there was an 18% decrease in

July precipitation and an increase in August precipitation of 75% (Table 3.3).

There were no differences in biomass among tef varieties in 2008 and 2009

(Table 3.4). Tef varieties Emerald, Ivory, Excalibur and Tiffany suppressed Canada thistle biomass 79% to 83% in 2008 (Table 3.4). In 2009, all varieties of tef except

Emerald and VA-T1 suppressed Canada thistle biomass 47% to 62% (Table 3.4). There was no significant correlation between tef biomass and Canada thistle biomass (data not shown). Tef varieties suppressed the growth of annual weeds both years (Table 3.4). In

2008, varieties Excalibur and Ivory had annual weed biomass of 256 and 397 g m-2, compared to 1690 g m-2 for the non-treated control. In 2009, annual weed biomass was

64 to 80% lower in plots with tef than in the non-treated control (Table 3.4). Tef biomass and annual weed biomass were negatively correlated in 2008 (Pearson correlation coefficient=-0.68; P < 0.0001).

Tef varieties differed in rate of vertical growth over the course of the growing seasons in 2008 and 2009 (Table 3.5). Dessie had the highest rate of vertical growth with

1.62 cm d-1 in 2008 and 1.58 cm d-1 in 2009 (Table 3.5). Emerald (1.40 cm d-1) and

Pharaoh (1.29 cm d-1) had the lowest rates in 2008 and 2009, respectively (Table 3.5).

Tef varieties also affected the growth of Canada thistle shoots (Table 3.5). Canada thistle shoots had lower vertical growth rates than the non-treated control in the Ivory (0.31 cm d-1) and Tiffany (0.39 cm d-1) varieties in 2008 (Table 3.5). In 2009, all tef varieties suppressed Canada thistle vertical growth (average 22%), with the most suppression in the Dessie (33%) and Emerald (24%) varieties (Table 3.5). The colder temperatures in

2009 may have favored growth of Canada thistle as its growth rate was 54% greater than in 2008 (Table 3.5).

Ground cover of tef, Canada thistle, and annual weeds differed between tef varieties in 2008 and 2009 (Table 3.5). In 2008, the Tiffany treatment had the lowest rate of Canada thistle spread (0.055 % d-1) and highest rate of tef ground cover (1.76 % d-1)

(Table 3.5). In 2009, the Dessie treatment had the lowest rate of Canada thistle spread

(0.70 % d-1) and highest rate of tef ground cover (1.91 % d-1) (Table 3.5). The inconsistency of variety performance between years may be due to cooler temperatures in

2009 and change in rainfall pattern (Table 3.3). The ability of smother crops to compete for space can decrease the space available for weeds to occupy. All tef varieties in 2008 and 2009 suppressed the rate of ground cover of annual weeds by 35 to 54% (Table 3.5).

The rate of ground cover of annual weeds in Corvalis (1.31 % d-1) was highest in both years and lowest in Excalibur (0.93 % d-1) in both years (Table 3.5).

In summary, tef was a weak potential smother crop for management of Canada thistle, but was effective for controlling annual weeds. In the greenhouse trial, the

74 varieties VA-T1, Corvalis, and Excalibur reduced Canada thistle biomass and demonstrated higher competitive ability through height and biomass production.

However, differences in emergence of Canada thistle shoots affected by the planting depth of roots affected the competitive ability of smother crops. In the field trial, there was little consistency between years for differences between tef varieties. Tef varieties suppressed growth of annual weeds and usually outcompeted Canada thistle. No specific variety suppressed Canada thistle more effectively in both years than others, although the variety Pharaoh was less effective as a smother crop than other tef varieties. Based on the data from the greenhouse and field trials, tef varieties VA-T1 and Corvalis had higher growth and more suppression of Canada thistle. These varieties need more testing before use as a smother crop. However, the results of the study demonstrate the possibility of using a grass smother crop for weed suppression during the transition to organic production.

Sources of Materials

1Pro-Mix, Premier Tech Horticulture, 127 South Fifth Street #300, Quakertown, PA

18951.

2 SAS 9.2 Statistical Software, Statistical Analysis Systems, SAS Institute, Inc., 100 SAS

Campus Drive, Cary NC 27513-2414.

3 SigmaPlot 11 Software, Systat Software, Inc., 1735 Technology Drive, Suite 430, San

Jose, CA 95110.

Acknowledgements

Field assistance was provided by Lynn Ault and K. Gregory Smith. Critical knowledge of tef management was provided by James VanLeeuwen. Thank you to undergraduate students who assisted in data collection and field work. Salaries and research support were provided by State and Federal Funds appropriated to the Ohio

Agriculture Research and Development Center, The Ohio State University. Manuscript

No. HCS-09-00

Literature Cited

Adnew, T., S. Ketema, H. Tefera, and H. Sridhara. 2005. Genetic diversity in tef [Eragrostis tef (Zucc.) Trotter] germplasm. Genet. Resour. Crop Ev. 52:891-902.

Assefa, K., H. Tefera, A. Merker, T. Kefyalew, and F. Hundera. 2001. Quantitative trait diversity in tef [Eragrostis tef (Zucc.) Trotter] germplasm from Central and Northern Ethiopia. Genet. Resour. Crop Ev. 48:53-61.

Bicksler, A. J. and J. B. Masiunas. 2009. Canada thistle (Cirsium arvense) suppression with buckwheat or sudangrass cover crops and mowing. Weed Technol. 23:556- 563.

Debelo, A. 1992. Germination, Yield and Yield Components of tef (Eragrostis tef) as affected byEnvironment, Tillage and Weed Control Practices. Ph.D dissertation. Stillwater, OK: Oklahoma State University. 162p.

DeHaan, R. L., D. L. Wyse, N. J. Ehlke, B. D. Maxwell, and D. H. Putnam. 1994. Simulation of spring-seeded smother plants for weed control in corn (Zea mays). Weed Sci. 42:35-43.

Donald, W.W. 1994. The biology of Canada thistle (Cirsium arvense). Rev. Weed Sci. 6:77-101.

Evans, J. E. 1984. Canada thistle (Cirsium arvense): a literature review of management practices. Natural Areas Journal 4:11-21.

Gaudet, C. L. and P. A. Keddy. 1988. A comparative approach to predicting competitive ability from plant traits. Nature 334:242-243.

Gustavsson, A.M.D. 1997. Growth and regenerative capacity of plants of Cirsium arvense. Weed Res. 37:229-236.

Hanson, J., R. Dismukes, W. Chambers, C. Greene, and A. Kremen. 2004. Risk and risk management in organic agriculture: views of organic farmers. Renew. Agr. Food Syst. 19:218-227.

Kefyalew, T., H. Tefera, K. Assefa, and M. Ayele. 2000. Phenotypic diversity for qualitative and phonologic characters in germplasm collections of tef (Eragrostis tef). Genet. Resour. Crop Ev. 47:73-80.

Ketema, S. 1997. Tef (Eragrostis tef (Zucc.) Trotter). 50p. in J. Heller, J. Engels, and K. Hammer, eds. Promoting the Conservation and Use of Underutilized and Neglected Crops. 12. Rome: Institute of Plant Genetics and Crop Plant Research.

McAllister, R. S. and L. C. Haderlie. 1985. Seasonal variations in Canada thistle (Cirsium arvense) root bud growth and root carbohydrate reserves. Weed Sci. 33:44-49.

Moore, R. J. 1975. The biology of Canadian weeds 13. Cirsium arvense (L.) Scop. Can. J. Plant Sci. 55:1033-1048.

Regnier, E. E. and R. R. Janke. 1990. Evolving strategies for managing weeds. Pages 174-202 in C. A. Edwards, R. Lal, P. Madden, R. H. Miller and G. House, eds. Sustainable Agricultural Systems. Boca Raton, FL: CRC Press.

Schoofs, A. and M. H. Entz. 2000. Influence of annual forages on weed dynamics in a cropping system. Can. J. Plant Sci. 80:187-198.

Teasdale, J. R. and C.S.T. Daughtry. 1993. Weed suppression by live and desiccated hairy vetch (Vicia villosa). Weed Sci. 41:207-212.

Turner, R. J., G. Davies, H. Moore, A. C. Grundy, and A. Mead. 2007. Organic weed management: a review of the current UK farmer perspective. Crop Prot. 26:377- 382.

Udensi, U. E., I. O. Akobundu, A. O. Ayeni, and D. Chikoye. 1999. Management of cogongrass (Imperata cylindrica) with velvetbean (Mucuna pruriens var. utilis) and herbicides. Weed Technol. 13:201-208.

Verschwele, A. and A. Häusler. 2004. Effect of crop rotation and tillage on infestation of Cirsium arvense in organic farming systems. Pages 187-194 in 6th EWRS Workshop on Physical and Cultural Weed Control. Lillehammer, Norway: European Weed Research Society.

Wang, G., McGiffen, M. E. Jr., Ehlers, J. D., and E.C.S. Marchi. 2006. Competitive ability of cowpea genotypes with different growth habit. Weed Sci. 54:775-782.

Yu, J. K., E. Graznak, F. Breseghello, H. Tefera, and M. E. Sorrells. 2007. QTL mapping of agronomic traits in tef (Eragrostis tef (Zucc) Trotter). BMC Plant Biol. 7:30.

Table 3.1. Seed color and source of smother crop varieties of tef and sorghum-

sudangrass.

Species Variety Seed color Seed source

Tef Corvalis Brown and King‟s Agriseeds, Ronks, PA

white

Tef Dessie Brown and red The Teff Company, Nampa, ID

Tef Emerald White The Teff Company, Nampa, ID

Tef Excalibur White United Seed, DeGraff, MN

Tef Ivorya White Byron Seeds, LLC., Marshall, IN

Tef Pharaoh Brown United Seed, DeGraff, MN

Tef Tiffanyb White Target Seed Company, Parma, ID

Tef VA-T1 Brown and James VanLeeuwen, Halsey, OR

white

Sorghum- Special Effort Tan Production Plus Quality Seed, sudangrass Plainview, TX a Certified organic. b Seed coated with Pinnacle Nutrient Coating™.

2.5

)

1 A - 2

1.5

AB BC BC 1 BCD BCD BCD CD CD D

Biomass Biomass Canada of thistle (g pot 0.5

Smother crop

Figure 3.1. Biomass ± SE (g pot-1) of Canada thistle shoots and rhizomes harvested from the greenhouse trial with eight varieties of tef, one variety of sorghum-sudangrass, and a no smother crop control. Bars without the same letter differ according to Fisher‟s

Protected Least Significant Difference at P < 0.05.

)

1 - 20 A

10 B

5 Biomass of smother smother Biomass of crops pot(g 0 7.6 cm 15 cm Canada thistle root planting depth

Figure 3.2. Biomass ± SE (g pot-1) of smother crop varieties growing in competitioin with Canada thistle planted at 7.6 and 15 cm depths. Bars followed by the same letter do not differ according to Kruskal-Wallis Test at P < 0.05.

Table 3.2. Final height of Canada thistle and smother crops and biomass of smother crops in greenhouse trial. Canada thistle root pieces were planted at depths of 7.6 and 15 cm. Data are means of six replications and two runs of the experiment.

Height Biomass

Canada thistle Smother Crop Smother Crop

Treatment 7.6a 15b 7.6a 15b 7.6a 15b

cm g pot-1

Non-treated 6.3 c 8.6 N/A N/A N/A N/A

Corvalis 12 ab 3.8 23 bc 25 9.6 a 19

Dessie 16 a 9.8 21 c 25 7.6 c 19

Emerald 11 bc 8.1 21 c 27 6.5 c 16

Excalibur 11 bc 6.0 21 c 24 7.8 bc 19

Ivory 14 ab 11 20 c 26 9.1 ab 18

Pharaoh 13 ab 12 24 b 26 9.2 a 19

Special Effort 12 ab 6.6 33 a 46 2.8 d 13

Tiffany 12 ab 11 21 bc 25 6.7 c 17

VA-T1 9.7 bc 8.6 23 bc 27 9.5 a 22 a Least squares means within a column followed by the same letter are not significantly different at P < 0.05. b Least squares means presented. Data in column are not significantly different according to Kruskal-Wallis Test at P < 0.05.

30 7.6 cm A 25 15 cm B

A Height Height (cm) 10 B

0 Canada Thistle Smother Crop

Figure 3.3. Height ± SE (cm) of Canada thistle and smother crops in pots where Canada thistle rhizomes were planted at 7.6 and 15 cm depths. Bars within a group followed by the same letter do not differ according to Kruskal-Wallis Test at P < 0.05.

Table 3.3. Monthly and 20-yr mean total precipitation and mean maximum and minimum temperatures during the growing season.

Monthly total precipitation

Year Jun Jul Aug Sep Oct Total

2008 14.7 11.3 3.24 7.26 3.89 40.4

2009 9.52 7.37 14.9 6.65 8.55 47.0

20-yr mean 9.13 8.98 8.52 7.32 6.75 40.7

Monthly maximum temperature

Year Jun Jul Aug Sep Oct Mean

 C

2008 26.8 28.3 26.7 25.4 16.7 24.8

2009 25.9 25.6 26.7 22.9 14.9 23.2

20-yr mean 26.8 28.7 28.0 24.1 17.5 25.0

Monthly minimum temperature

Year Jun Jul Aug Sep Oct Mean

 C

2008 15.0 15.8 13.6 11.5 3.70 11.9

2009 13.2 13.6 15.5 11.5 4.06 11.6

20-yr mean 13.9 16.0 15.2 10.9 5.06 12.2

Table 3.4. Biomass of tef varieties, Canada thistle, and other weeds harvested in field trial, 2008 to 2009. Data are means of six replications each year.

Tefa Canada thistleb Other weedsb

Treatment 2008 2009 2008 2009 2008 2009

g m-2

Non-treated N/A N/A 11 ab 149 a 1690 a 507 a

Corvalis 577 520 4 ab 79 bc 1180 ab 155 b

Dessie 833 432 6 ab 66 c 1030 abc 147 b

Emerald 955 492 2 b 136 ab 495 cd 132 b

Excalibur 1060 375 2 b 76 bc 256 d 129 b

Ivory 815 397 2 b 57 c 397 d 98 b

Pharaoh 1010 358 21 ab 72 bc 424 cd 105 b

Tiffany 974 420 2 b 69 c 563 bcd 150 b

VA-T1 789 446 26 a 104 abc 884 abc 128 b aMeans presented. Data in column not significant according to ANOVA at P < 0.05. bMeans within a column followed by the same letter are not significantly different according to Fisher‟s Protected Least Significant Difference at P < 0.05.

Table 3.5. Rate of vertical growth (cm d-1) and ground cover spread (% d-1) of tef, Canada thistle, and annual weeds in field

trial 2008 and 2009.

Tefa Canada Thistlea Tefa Canada Thistlea Annual Weedsa Treatment 2008 2009 2008 2009 2008 2009 2008 2009 2008 2009 -1 -1 cm d % d Non-treated N/A N/A 0.96 a 2.1 a N/A N/A 0.75 ab 1.3 a 1.9 a 2.1 a Corvalis 1.5 ab 1.6 a 0.78 ab 1.7 b 1.4 cd 1.8 ab 0.36 bcd 0.96 b 1.2 b 1.4 b Dessie 1.6 a 1.6 a 0.99 a 1.6 bc 1.5 bcd 1.9 a 0.76 a 0.70 c 1.1 bc 1.2 bc Emerald 1.4 c 1.4 abc 0.85 a 1.4 c 1.6 abc 1.7 ab 0.49 abc 0.94 bc 1.0 bc 1.2 bc

87 Excalibur 1.5 ab 1.4 bc 0.56 abc 1.7 b 1.5 bcd 1.5 cd 0.28 cd 1.2 a 0.88 c 0.98 c Ivory 1.6 ab 1.4 abc 0.31 c 1.5 bc 1.5 bcd 1.7 ab 0.040 d 1.0 ab 1.2 b 1.4 b Pharaoh 1.5 bc 1.3 c 0.75 ab 1.7 b 1.4 d 1.4 d 0.36 bcd 1.2 a 1.1 bc 1.2 bc Tiffany 1.5 ab 1.5 abc 0.39 bc 1.6 b 1.8 a 1.7 bc 0.060 d 0.94 bc 0.96 c 1.1 c VA-T1 1.6 ab 1.5 ab 0.60 abc 1.7 b 1.6 ab 1.7 ab 0.055d 0.87 bc 1.2 b 1.3 b a Least squares means within a column followed by the same letter are not significantly different at P < 0.05.

Chapter 4: Summer Annual Crop Mixtures for Biomass Production and Weed

Suppression

S. Wedryk and J. Cardina*

Summer-annual crop mixtures can be used to increase system productivity, reduce chemical inputs, and manage weeds. The objective of this study was to determine optimal tef and sorghum-sudangrass mixtures for biomass production and weed suppression. Tef and sorghum-sudangrass were planted in monoculture and in mixture with soybean and sunflower in 2008 and 2009. The percentage cover of crops and weeds and height of crops were measured weekly. Final biomass was measured for component crops and weeds. Biomass was used to calculate the land equivalent ratio (LER) and aggressivity indices of crops in mixture. Biomass production was affected by the species of grass in monoculture or mixture, but LERt was greater in three-species mixtures than monoculture. Differences in crop height between tef and sorghum-sudangrass may have influenced biomass production. Aggressivity of grass crops was linearly related to total

LER in two- and three-species mixtures. Percent cover of crops was greater in multi- species mixtures and weed cover was suppressed 70 and 45 percent in 2008 and 2009 by

* Graduate Research Associate and Professor, Department of Horticulture and Crop Science, The Ohio State University, Wooster, OH 44691. Corresponding author‟s E-mail: [email protected]. 88 multi-species mixtures. The biomass and percent cover of the grass crop in mixture or monoculture was related to total biomass production and percent cover. Management and choice of grass crops for biomass production may affect productivity.

Nomenclature: Tef, Eragrostis tef (Zucc.) Trotter; sorghum-sudangrass Sorghum bicolor (L.) Moench. x Sorghum sudanese (P.) Stapf.; soybean Glycine max (L.) Merr.; sunflower Helianthus annuus L.

Keywords: smother crop, land equivalent ratio (LER), aggressivity, cover crop

Demand for increased global agricultural output has shifted the emphasis of agriculture to optimize cropping systems for productivity while reducing environmental impacts. Multi-species cropping systems can contribute to management of pests and greater yields compared with monoculture (Tilman et al. 2002). Intercropping with two or more species has been used for minimizing risk of crop failure, increasing yield and yield stability, improving the use of resources by crops, and suppression of weeds (Agegnehu et al. 2006, 2008; Ghosh 2004). Additionally, research on intercropping systems is often conducted to enhance productivity on smaller spaces due to shrinking farm size in the tropics and to maximize economic return from cropping systems managed with no fertilizers and pesticides (Agegnehu et al. 2008; Rao and Shetty 1976). The use of multi- species cropping mixtures in temperate locations can offer similar benefits to intercropping systems in tropical agriculture to increase productivity, manage pests and reduce chemical inputs (Maléziuex et al. 2009).

Previous studies have shown that multi-species mixtures had greater yields, biomass production, and system productivity than monoculture plantings (Linares et al.

2008; Odo 1991). The advantage to polyculture was attributed to the occupation of multiple ecological niches by crops in mixture. Using species of different functional groups, such as a grass and a legume, increased yields compared with monoculture due to resource use efficiency and differences in nitrogen use (Agegnehu et al. 2006; Willey

1979). Combinations of crops that occupy different spatial niches in canopy architecture and belowground structure had increased biomass production over monoculture due to reduced interspecies competition (Creamer and Baldwin 2000; Willey 1979). Mixtures of

90 component crops that occupy different temporal niches of resource use and growth patterns have also been reported to increase biomass production through decreased competition for resources (Willey 1979). However, other studies have shown reduced biomass production and yields due to competition between component crops (Creamer and Baldwin 2000; Mandal et al. 1990; Rezende and Ramalho 1994).

Growing crops in mixtures often has an additional benefit of greater weed suppression than monoculture (Creamer et al. 1997). Warm-season grass and legume smother crop mixtures have been shown to reduce weed populations more effectively than monoculture smother crops (Creamer and Baldwin 2000; Linares et al. 2008).

Increasing the number of functional groups represented by crops in intercropping resulted in greater weed suppression than one or two species cropping systems (Banik et al. 2006;

Zuofa et al. 1992). Occupation of different spatial niches by crops suppressed weeds through competition for space and light (Unamma et al. 1986; Wang et al. 2006).

However, competition between crops in mixture can reduce the degree of weed suppression observed if crops are competing for the same resources at the same time or are not sufficiently divergent ecologically (Valverde et al. 1995; Zuofa et al. 1992).

This research was conducted to develop annual crop species mixtures for enhanced productivity in temperate climate with the additional benefit of weed suppression for organic and reduced input production. Crop mixtures grown during the summer months could be used in rotation with fall-planted cereals such as winter wheat

(Triticum aestivum L.) or rye (Secale cereale L.) or short-season vegetables like lettuce

(Lactuca sativa L.) or collard greens (Brassica oleracea L.) (Akemo et al. 2000). The use

91 of cover crop mixtures during the summer in rotation can provide benefits to soil through carbon deposition and soil aggregation or reduce chemical inputs during cash crop production (Forcella and Reicosky 1998). The use of annual crop mixtures may be attractive for farmers wanting to diversify cropping rotations without investment in perennial crops for bioenergy that can take three to four years before economic benefits are realized (Tilman et al. 2006).

A critical question in developing species mixtures is how to maximize the contribution of each species for total biomass production while reducing competition between component crops. The approach included grass, forb, and legume species in mixtures, with the assumption that these three plant types and functional groups occupy different spatial and ecological niches (Unamma et al. 1986). Here results are reported for mixtures containing one of two warm season grasses of interest: tef [Eragrostis tef

(Zucc.) Trotter] and sorghum-sudangrass [Sorghum bicolor (L.) Moench x Sorghum sudanese (P.) Stapf.]. Tef is a rapidly emerging, fine-leaved cereal crop that can be seeded at high rates for biomass production and may be effective at suppressing annual weeds (Ketema 1997). Sorghum-sudangrass is a robust forage crop that has been used as a smother crop, suppressing weeds by 90 percent and producing more than 10,000 kg ha-1 of biomass in mixture (Creamer and Baldwin 2000). Soybean [Glycine max (L.) Merr.] was included to aid in reducing competition for nitrogen between component crops in order to improve nutrient use efficiency (Ghosh 2004). We included sunflower

(Helianthus annuus L.) in the crop mixture for its biomass and allelopathic potential

(Leather 1983). The objective of this study was to evaluate warm-season grasses with a

92 legume and/or forb in mixture for biomass production, and secondly, weed control under organic conditions. It was hypothesized that crop mixtures would be more effective at biomass production and weed suppression than monoculture.

Materials and Methods

Experimental Design. Field experiments were conducted at The Ohio State University

Schaffter Farm near Wooster, Ohio, in 2008 and 2009 to evaluate summer annual cover crop mixes for biomass production. The soil type at the site was a Wooster silt loam with pH of 7.3, 2.9% organic matter, and available P, K of 21.3 and 80.6 mg kg-1 soil, respectively. The field was disked and the seedbed prepared for planting on June 10,

2008, and June 11, 2009. The experimental design was a randomized complete block design with six replications of eleven treatments: sole crops of two grasses (tef and sorghum-sudangrass), a legume (soybean) and a forb (sunflower); two-species mixtures

(tef+soybean, tef+sunflower, sorghum-sudangrass+soybean, and sorghum- sudangrass+sunflower); three-species mixtures (tef+soybean+sunflower and sorghum- sudangrass+soybean+sunflower); and a non-treated control. Details of the varieties, seeding rates and densities are given in Table 4.1. The seeding rate for each component crop was approximately 80 and 60 percent of monoculture seeding rates for two- and three-species mixtures, respectively. Seeding rates were altered in two- and three-species mixtures to reduce competition between component crops. Each plot was 2.3 m2.

Mixtures were seeded by hand on June 11, 2008, and June 15, 2009. Sorghum-

93 sudangrass, soybean, and sunflower seeds were planted in the same rows at a depth of 3 cm with row spacing of 23 cm. Soybean seeds were inoculated with Rhizobium bacteria at rate of 2.5 g TerraMax Dry per kg seed immediately prior to planting.1 Tef was broadcast seeded and incorporated into the soil surface using a hand-held roller. Crop species were seeded as a crop mixture in the same rows and not species-specific rows of crops as may be more common in intercropping.

Percent cover of individual component crops and weeds was visually estimated every week for 8 wk starting 10 d after planting and at final harvest. Height of crop mixtures was measured in each plot weekly for 8 wk starting 10 d after planting. Height was measured at the two tallest points within each plot. Total plant biomass (crop mixtures plus weeds) was harvested when seeds from all crop species were at least 80 percent filled. Plants were harvested at this time to capture maximum crop biomass before senescence. Total plant biomass was harvested from one 0.093 m2 quadrat per plot on August 22, 2008, and August 27 and September 11, 2009. Harvested materials were separated into grass, legume, and forb crop species, and weeds, weighed, dried at 55C for 72 hours, and weighed again.

Total dry matter biomass of crops was used to calculate the total Land Equivalent

Ratio (LERt) using the following equation for two-species mixtures

LERt = (Mab/Maa) + (Mba/Mbb) [1] where Mab was the biomass of crop a in mixture with crop b and Maa was the biomass of crop a in monoculture. The following equation was used to calculate total LER in three- species mixtures

LERt = (Mabc/Maaa) + (Mbac/Mbbb) + (Mcab/Mccc) [2] where Mabc was the biomass of crop a in mixture with crops b and c in three-species mixtures and Maaa was the biomass of crop a in monoculture. The LERt was used to calculate the yield advantage to polyculture in tef and sorghum-sudangrass mixtures

(Ghosh 2004). The LER is frequently used in intercropping studies to give an estimate of the land area needed for monoculture plantings in order to produce equivalent intercropping yields (Wiley 1979). The LER can be used to give measures of yield advantage in mixtures as well (Odo 1991). In order to relate the change in biomass of component crops in mixture to monoculture, aggressivity (A) of crops in two- and three- species mixtures was calculated (Willey and Rao 1980). It was calculated as

Aab = (Mab/MaaXab) – (Mba/MbbXba) [3] where Xab was the proportion of the sown density of crop a in mixture to crop a in monoculture. If Aab was greater than zero, crop a was the dominant crop in mixture; if Aab was less than zero, then crop b was dominant in the mixture.

Statistical Analysis. Data for crop and weed biomass and percent cover were analyzed separately by year due to a significant treatment x year interaction. Crop and weed biomass, LERt, and crop final percent cover and height data were subjected to ANOVA in SAS v9.2.2 Outliers (less than 5 percent of data points) were removed from data that did not meet the assumptions of ANOVA before square root transformation. Single degree-of-freedom orthogonal contrasts were used to test the effect of species of grass in

95 the mixture and the number of crop species in mixture for crop biomass, percent cover, height, and LERt.

Data for regression analyses were combined for both years when appropriate and

3 analyzed using SigmaPlot v11. Data for LERt and aggressivity of grasses were plotted for two- and three-species crop mixtures and fitted to a linear regression. Weed percent cover (WC) data were plotted as a function of growing degree days (GDD) with a base temperature of 10 C and a start date of January 1. Results were fit to a three parameter logistic curve as follows

b WC = a/(1+(GDD/GDD0) ) [6] where GDD is growing degree days and a and b are equation parameters.

Pearson correlation coefficients were calculated separately by year for total crop, grass, legume, forb biomass and percent cover, LERt, crop height, and seeding rate.

Significant correlations (P < 0.05) between response variables with r < 0.4 were considered weak; 0.4 < r < 0.8 were considered moderate; r < 0.8 were considered strong.

Results and Discussion

Climatic Conditions. In July and August 2009, monthly maximum temperatures were

11% and 5% lower than the 20-yr mean (Table 4.2). Total precipitation in June and July

2008 was 43% greater than the 20-yr mean (Table 4.3). There was an 18% decrease in

96 precipitation during July 2009 and a 75% increase in precipitation during August 2009 compared to the 20-yr mean.

Crop Production. Crop biomass was affected by grass species in 2008 and 2009 (Table

4.4). Sorghum-sudangrass mixtures yielded 44 and 40 percent more biomass on average than tef mixtures in 2008 and 2009, respectively. The difference in biomass between tef and sorghum-sudangrass mixtures may be attributable to greater height in sorghum- sudangrass. Sorghum-sudangrass mixtures (2008=270 cm; 2009=220 cm) were, on average, 58 and 65 percent taller than tef mixtures (2008=110 cm; 2009=77 cm) in 2008 and 2009, respectively (P < 0.05). Although the seeding rate and density of tef mixtures was greater than sorghum-sudangrass mixtures, presumably allowing for more plants to germinate and grow, this did not contribute to greater biomass in tef mixtures (Table 1).

The greater leaf area of Sorghum spp. in comparison to tef may also have contributed to greater biomass production (Hammer et al. 1987; Ketema 1997). Crop biomass was different in one- and two- or three-species mixtures in 2009 only (Table 4.4). The magnitude of crop biomass production was greater in 2008 than 2009 (Table 4.4). Cooler temperatures and changes in precipitation patterns may have affected growth as increased temperatures can lead to increased yields in warm-season grasses (Table 4.2; Table 4.3)

(Henderson and Robinson 1982).

The number of crop species in mixture affected LERt in 2008 and 2009, but only the difference between one- and three-species mixtures was consistent in both years

(Table 4.4). An advantage to using crops in polyculture is evidenced when LERt is

97 greater than one (Willey 1979). The biomass advantage to using three-species mixtures compared with monoculture may be the result of greater seeding rate and density (Table

4.1). Planting multi-species mixtures is often considered advantageous due to the occupation of different ecological niches by component crops (Mandal et al. 1990;

Willey 1979). This may be a reason why LERt in three-species mixtures in both years and two-species mixtures in 2009 is greater than one. However, differences between the LERt of two- and three-species would be expected if occupation of additional ecological niches contributed to biomass advantage. Although niche occupation may play a role in the advantage of three-species mixtures to monoculture, more evidence about crop resource use is needed for this conclusion.

Aggressivity was used to determine which component crop in mixture was more dominant in relation to other crop species using biomass measures. Total LER and aggressivity of grasses in two- and three-species mixtures fit a linear relationship (Figure

4.1). In two-species mixtures, LERt declined with increasing aggressivity of grass component crops regardless of the presence of legume or forb (slope = -0.62; Adj. R2 =

0.77; P < 0.0001). In two-species mixtures, higher aggressivity values of one component crop in mixture may suppress growth of the other crop, reducing its contribution to polyculture biomass (Agegnehu et al. 2006). The mixture may be more reflective of the dynamics of monoculture if other component crops are unable to compete effectively for resources. Multi-species mixtures in this context would not result in greater biomass production or LERt in comparison to monoculture if one crop does not sufficiently contribute to biomass.

In three-species mixtures, LERt increased with increasing aggressivity of grass component crops compared to either the legume or forb species (slope = 0.41; Adj. R2 =

0.19; P = 0.009) (Figure 4.1). The addition of a third crop in mixture may affect competition between component crops. In maize (Zea mays L.)-cassava (Manihot esculenta Crantz.) intercrops, the yield of maize and total LER increased when smother crops were included in the mixture (Zuofa et al. 1992). Grass crops in three-species mixtures may enhance production with increasing competitiveness.

Percent cover of crop mixtures was affected by the number of species in mixture

(Table 4.5). In 2008 and 2009, percent cover was different in one- and two- or three- species mixtures. The differences may be the result of including percent cover of soybean and sunflower monocultures in calculating contrasts that were 18 to 67 percent and 14 to

45 percent lower in 2008 and 2009, respectively, than tef or sorghum-sudangrass monocultures (Table 4.5). The greater seeding rate in two- and three-species mixtures within a finite space may also have contributed to increased percent cover (Table 4.1).

However, percent cover between tef and sorghum-sudangrass mixtures and two- and three-species mixtures would be expected to be different if seeding rate and density were the primary cause of increased ground cover (Table 4.1; Table 4.5). Increased percent cover of crop mixtures in two- and three-species mixtures may be the result of occupation of complementary spatial niches by component crops. Differences in above-ground growth habit and height and below-ground root systems between component crops may allow for reduced competition in acquisition of resources dependent on space such as light or non-mobile soil nutrients (Davis et al. 1984; Lynch 2011). The change in

99 magnitude of percent cover of crops between 2008 and 2009 may be related to lower temperatures and timing of precipitation in 2009 as warm-season crops may not have been able to attain maximum growth (Table 4.2; Table 4.3).

Weed Suppression. The presence of crops reduced weed biomass compared to the non- treated control in 2008 and 2009 (Figure 4.2). However, there were no differences in suppression of weed biomass among crop mixtures. Weed cover was reduced 51 and 31 percent or more by the monocultures in 2008 and 2009, respectively (Figure 4.3). Further reduction in weed cover occurred with two-species mixtures in 2008 (70 percent) and

2009 (45 percent) and three-species mixtures in 2008 (78 percent) and 2009 (50 percent).

The logistic curve did not fit for three-species mixtures in 2008 due to a decrease in weed cover at the end of the growing season (Adj. R2 = 0, P = 0.73; Table 4.6). Although suppression of weed biomass did not differ among cropping treatments, two- and three- species mixtures gave greater suppression of weed cover than monocultures. Previous research has demonstrated that weed biomass is related to changes in crop biomass

(Beckie et al. 2008; Gaudet and Keddy 1988). Since there were no consistent differences in crop biomass related to the number of species in mixture, it was expected that weed biomass may not differ may not differ among cropping treatments (Table 4.4). Similarly, differences in percent cover between crop mixtures may be reflected in differences in weed cover since the potential space to be occupied is finite. The differences in weed suppression reflected in measures of percent cover are limited to what is visible through a crop canopy. Weeds may emerge after closure of the crop canopy and not be visible in

100 percent cover estimates, but can be harvested and included in weed biomass measurements. Growth and reproduction of weeds that emerge in mature crop stands can be affected by light attenuation (Baumann et al. 2001). Measurements of light and development of weed species in different crop mixtures would be needed to more accurately determine the effects of crop cover on weed populations.

Correlation Analysis. Pearson correlation coefficients were calculated between measured response variables for 2008 and 2009 to determine the relationship between parameters of crop growth and components of crop mixtures to biomass production and LERt (Table

4.7). Grass crop biomass was directly correlated with total crop biomass in 2008 (r =

0.81, P < 0.001) and 2009 (r = 0.90, P < 0.001) (Table 4.7). Grass crop biomass was weakly correlated with LERt in both years also. Similarly, total and grass crop percent cover were moderately, directly correlated to total crop biomass and LERt in 2008 and

2009 (Table 4.7). Crop height was directly correlated with total crop biomass in 2008 (r =

0.66, P < 0.001) and 2009 (r = 0.83, P < 0.001) (Table 4.7). The lack of consistent relationships between legume and forb crops and biomass production and previous observations of grass performance in monoculture and mixtures might indicate that the choice of grass species could be crucial in determining biomass production (Table 4.4;

Table 4.5; Figure 4.1) (Rao and Willey 1983). Crop height can be an indicator for crop competitiveness and therefore its ability to acquiesce and utilize resources (Gaudet and

Keddy 1988). Further, seeding rate was correlated with LERt in 2008 (r = 0.51, P <

0.001) and 2009 (r = 0.27, P < 0.05), even though the relationship was not strong in 2009

101

(Table 4.7). Seeding rate may enhance biomass production as reflected in LERt and give greater biomass advantage to crop mixtures (Table 4.4).

Conclusions

In low-input production systems, biomass production may be enhanced through the choice and management of grass species used in monoculture or mixture. Crop height and seeding rates were related to biomass output. The aggressivity of grasses used in mixtures can affect whether there is an advantage to using polyculture. The use of crop mixtures may enhance suppression of weed cover through greater seeding rates and greater cover of finite space. However, further measurements of light and the competitive effects of crops on weeds may demonstrate why crop mixtures were more suppressive of weed cover than monocultures. There was a biomass advantage to using three-species mixtures in comparison to monoculture, but the role of ecological niches in conferring that advantage is unclear. The number of species in mixture affected percent cover of crops. Seeding rate may have allowed more potential plants to occupy space, or different occupation of space may have improved visual estimates. Measurements of crop species morphology such as leaf area, internode length, or root architecture may demonstrate whether spatial occupation of crops can affect biomass output of crop mixtures.

102

Sources of Materials

1 TerraMax Dry, TerraMax, Inc., 7769 95th Street South, Cottage Grove, MN 55106.

2 SAS 9.2 Statistical Software, Statistical Analysis Systems, SAS Institute, Inc., 100 SAS

Campus Drive, Cary NC 27513-2414.

3 SigmaPlot Version 11.0, Systat Software, Inc., 225 W. Washington St., Chicago, IL

60606.

Acknowledgements

We would like to thank Lynn Ault and K. Gregory Smith for field assistance.

Knowledge of tef management was graciously provided by James Van Leeuwen. Thank you to the undergraduate students who helped with data collection and field work.

Salaries and research support were provided by State and Federal Funds appropriated to the Ohio Agriculture Research and Development Center, The Ohio State University.

Manuscript No. HCS-09-00.

103

Literature Cited

Agegnehu, G., A. Ghizaw, and W. Sinebo. 2006. Yield performance and land-use efficiency of barley and faba bean mixed cropping in Ethiopian highlands. Eur. J. Agron. 25:202-207.

Agegnehu, G., A. Ghizaw, and W. Sinebo. 2008. Yield potential and land-use efficiency of wheat and faba bean mixed intercropping. Agron. Sustain. Dev. 28:257-263.

Akemo, M.C., E.E. Regnier, and M.A. Bennett. 2000. Weed suppression in spring-sown rye (Secale cereale)-pea (Pisum sativum) cover crop mixes. Weed Technol. 14:545-549.

Banik, P., A. Midya, B.K. Sarkar, S.S. Ghose. 2006. Wheat and chickpea intercropping systems in an additive series experiment: advantages and weed smothering. Europ. J. Agron. 24:325-332.

Beckie, H.J., E.N. Johnson, R.E. Blackshaw, and Y. Gan. 2008. Weed suppression by canola and mustard cultivars. Weed Technol. 22:182-185.

Bulson, H.A.J., R.W. Snaydon, and C.E. Stopes. 1997. Effects of plant density on intercropped wheat and field beans in an organic farming system. J. Agric. Sci., Camb. 128:59-71.

Creamer, N.G., M.A. Bennett, and B.R. Stinner. 1997. Evaluation of cover crop mixtures for use in vegetable production systems. HortScience 32:866-870.

Creamer, N.G. and K.R. Baldwin. 2000. An evaluation of summer cover crops for use in vegetable production systems in North Carolina. HortScience 35:600-603.

Davis, J.H.C., L. van Beuningen, M.V. Ortiz, and C. Pino. 1984. Effect of growth habit of beans on tolerance to competition from maize when intercropped. Crop Sci. 24:751-755.

Forcella, F. and D.C. Reicosky. 1998. Cover crop and soil quality interactions in agroecosystems. J. Soil Water Conserv. 53:224-229.

104

Gaudet, C.L. and P.A. Keddy. 1988. A comparative approach to predicting competitive ability from plant traits. Nature 334:242-243. Ghosh, P.K. 2004. Growth, yield, competition and economics of groundnut/cereal fodder intercropping systems in the semi-arid tropics of India. Field Crops Res. 88:227- 237.

Hammer, G.L., K. Hill, and G.N. Schrodter. 1987. Leaf area production and senescence of diverse grain sorghum hybrids. Field Crop Res. 17:305-317.

Henderson, M.S. and D.L. Robinson. 1982. Environmental influences on yield and in vitro true digestibility of warm-season perennial grasses and the relationships to fiber components. Agron. J. 74:943-946.

Ketema, S. 1997. Tef (Eragrostis tef (Zucc.) Trotter). In J. Heller, J. Engels, and K. Hammer, eds. Promoting the Conservations and Use of Underutilized and Neglected Crops. Institute of Plant Genetics and Crop Plant Research, Rome. 50 p.

Leather, G.R. 1983. Sunflowers (Helianthus annuus) are allelopathic to weeds. Weed Sci. 31:37-42.

Lynch, J.P. 2011. Root phenes for enhanced soil exploration and phosphorus acquisition: tools for future crops. Plant Physiol. 156:1041-1049.

Malézieux, E., Y. Crozat, C. Dupraz, M. Laurans, D. Makowski, H. Ozier-Lafontaine, B. Rapidel, S. de Tourdonnet, and M. Valantin-Morison. 2009. Mixing plant species in cropping systems: concepts, tools and models. A review. Agron. Sust. Dev. 29:43-62.

Mandal, B.K., M.C. Dhara, B.B. Mandal, S.K. Das, and R. Nandy. 1990. Rice, mungbean, soybean, peanut, ricebean, and blackgram yields under different intercropping systems. Agron. J. 82:1063-1066.

Odo, P.E. 1991. Evaluation of short and tall sorghum varieties in mixtures with cowpea in the Sudan savanna of Nigeria: land equivalent ratio, grain yield and system productivity index. Expl. Agric. 27:435-441.

Rao, M.R. and S.V.R. Shetty. 1976. Some biological aspects of intercropping systems on crop-weed balance. Indian J. Weed Sci. 8:32-43.

105

Rao, M.R. and R.W. Willey. 1983. Effects of genotype in cereal/pigeonpea intercropping on the alfisols of the semi-arid tropics of India. Expl. Agric. 19:67-78.

Rezende, G.D.S.P. and M.A.P. Ramalho. 1994. Competitive ability of maize and common bean (Phaseolus vulgaris) cultivars intercropped in different environments. J. Agric. Sci., Camb. 123:185-190.

Tilman, D., K.G. Cassman, P.A. Matson, R. Naylor, and S. Polasky. 2002. Agricultural sustainability and intensive production practices. Nature 418:671-677.

Tilman, D., J. Hill, and C. Lehman. 2006. Carbon-negative biofuels from low-input high- diversity grassland biomass. Science 314:1598-1600.

Unamma, R.P.A., L.S.O. Ene, S.O. Odurukwe, and T. Enyinnia. 1986. Integrated weed management for cassava intercropped with maize. Weed Res. 26:9-17.

Wang, G., M.E. McGiffen Jr., J.D. Ehlers, and E.C.S. Marchi. 2006. Competitive ability of cowpea genotypes with different growth habit. Weed Sci. 54:775-782.

Willey, R.W. 1979. Intercropping – its importance and research needs. Part 1. Competition and yield advantages. Field Crop Abst. 32:1-11.

Willey, R.W. and M.R. Rao. 1980. A competitive ratio for quantifying competition between intercrops. Expl. Agric. 16:117-125.

Zuofa, K., N.M. Tariah, and N.O. Isirimah. 1992. Effects of groundnut, cowpea and melon on weed control and yields of intercropped cassava and maize. Field Crops Res. 28:309-314.

106

Table 4.1. Composition of summer annual crop mixtures and seeding rates.

Crop mix Seeding rate Seeding density

kg ha-1 seeds ha-1

Tef „VA-T1‟ 34 9.7 x 107

Sorghum-sudangrass „Special Effort‟ 28 1.2 x 106

Soybean „Stonewall‟ 61 3.7 x 105

Sunflower „620CL‟ 2.6 2.9 x 104

Tef + soybean 27 + 48 7.7 x 107 + 2.9 x 105

Tef + sunflower 27 + 2.0 7.7 x 107 + 2.3 x 104

Tef + soybean + sunflower 20 + 36 + 1.5 5.7 x 107 + 2.2 x 105 + 1.7 x 104

Sorghum-sudangrass + soybean 22 + 48 9.8 x 105 + 2.9 x 105

Sorghum-sudangrass + sunflower 22 + 2.0 9.8 x 105 + 2.3 x 104

Sorghum-sudangrass + soybean + 17 + 36 + 1.5 7.5 x 105 + 2.2 x 105 + 1.7 x 104 sunflower

107

Table 4.2. Mean daily and 20 yr-mean maximum and minimum temperature during the growing season.

Mean daily maximum temperature

Year June July Aug Sep Oct

C

2008 26.8 28.3 26.7 25.4 16.7

2009 25.9 25.6 26.7 22.9 14.9

20-yr mean 26.8 28.7 28.0 24.1 17.5

Mean daily minimum temperature

Year June July Aug Sep Oct

C

2008 15.0 15.8 13.6 11.5 3.70

2009 13.2 13.6 15.5 11.5 4.06

20-yr mean 13.9 16.0 15.2 10.9 5.06

108

Table 4.3. Monthly and 20-yr mean total precipitation during the growing season.

Monthly total precipitation

Year June July Aug Sep Oct Total

2008 14.7 11.3 3.24 7.26 3.89 40.4

2009 9.52 7.37 14.9 6.65 8.55 47.0

20-yr mean 9.13 8.98 8.52 7.32 6.75 40.7

109

Table 4.4. Crop biomass and land equivalent ratio (LERt) of summer annual crop mixtures in 2008 and 2009.

Crop biomass LERt

Crop treatment 2008 2009 2008 2009

g m-2

Tef 1200 430 1.0 1.0

Sorghum-sudangrass 2800 1200 1.0 1.0

Soybean 160 410 1.0 1.0

Sunflower 2500 170 1.0 1.0

110

Tef + Soybean 1200 680 1.2 1.6

Tef + Sunflower 1900 380 2.2 1.8

Sorghum-sudangrass + soybean 3500 1000 1.5 1.5

Sorghum-sudangrass + sunflower 1600 1000 0.71 3.6

Tef + soybean + sunflower 1800 690 2.2 1.8

Continued

110

Table 4.4 continued

Crop biomass LERt

Crop treatment 2008 2009 2008 2009

g m-2

Sorghum-sudangrass + soybean + sunflower 3000 1100 1.4 2.0

P>F 0.0095 0.0001 0.023 0.024

Contrasts P values

One- vs. two-species mixtures 0.339 0.024 0.081 0.003

111

One- vs. three-species mixtures 0.149 0.003 0.007 0.045

Two- vs. three-species mixtures 0.497 0.210 0.167 0.617

Sorghum-sudangrass vs. tef mixtures 0.006 0.0001 0.027 0.206

111

10 Two-species mixes

LER 4 LERt = 0.41*AGX + 1.2

Adj. R2 = 0.19 2 P < 0.009

0 -10 -8 -6 -4 -2 0 2 4

AggressivityGX

5 Three species mixes

LERt = -0.62*AGX + 1.69 4 Adj. R2 = 0.77

P < 0.0001 3

LER 2

0 -1 0 1 2 3 Aggressivity GLGF

Figure 4.1. Linear regression of total land equivalent ratio (LERt) and aggressivity (AGX) of grass (G) crops to legume or forb (X) crops in two-species mixtures and aggressivity

(AGLGF) of grass (G) crops to legume (L) and forb (F) crops in three-species mixtures.

Data presented is from 2008 and 2009. 112

Table 4.5. Percent cover of crops at harvest in summer annual crop mixtures in 2008 and

2009.

Final percent cover

Crop treatment 2008 2009

Tef 84.3 55.8

Sorghum-sudangrass 82.5 52.5

Soybean 27.5 46.7

Sunflower 61.3 24.5

Tef + Soybean 83.5 56.2

Tef + Sunflower 89.0 50.2

Sorghum-sudangrass + soybean 88.8 60.8

Sorghum-sudangrass + sunflower 91.8 62.5

Tef + soybean + sunflower 93.3 60.0

Sorghum-sudangrass + soybean + 90.0 65.0 sunflower

P>F 0.0001 0.0001

Contrasts P values

One- vs. two-species mixtures 0.0001 0.0002

One- vs. three-species mixtures 0.0001 0.0001

Two- vs. three-species mixtures 0.108 0.192

Sorghum-sudangrass vs. tef mixtures 0.655 0.143

113

1800 Crop mixtures Non-treated 1600 A

1400

)

2 - 1200 1000

800 A 600 B Weed biomass biomass Weed m (g 400 B 200 0 2008 2009

Figure 4.2. Weed biomass in the non-treated control and crop mixtures in 2008 and 2009.

Bars within a year followed by the same letter do not differ according to single degree-of- freedom contrasts (P < 0.05).

114

120 No crops 2008 Monocultures 100 Two crops Three crops

Weed cover (%) cover Weed 40

0 800 1000 1200 1400 1600 1800 Growing degree days

100 2009 No crops Monocultures 90 Two crops Three crops 80

Weed cover (%) cover Weed

20 800 1000 1200 1400 1600 1800 Growing degree days

Figure 4.3. Weed cover (%) over growing degree days (base 10 C) in 2008 and 2009.

Adjusted means ± s.e. at each sampling time are presented. Equation parameters are listed in Table 4.5.

115

Table 4.6. Equation parameters, adjusted R2, and P values for spread of weed cover over growing degree days in 2008 and 2009.

Data fit a sigmoidal logistic three parameter curve (Equation 5).

2008 2009

Species number a b GDD Adj. R2 P a b GDD Adj. R2 P

Zero 101 -7.72 1060 0.97 0.02 94.5 -7.91 1060 0.99 0.001

One 49.5 -7.26 967 0.94 0.03 65.8 -16.0 1020 0.98 0.001

Two 28.9 -18.8 892 0.64 0.18 54.7 -14.4 1000 0.92 0.01

Three 26.3 13.8 1900 0 0.73 52.7 -13.0 951 0.51 0.16

116

Abbreviations: GDD, growing degree days

116

Table 4.7. Pearson correlation coefficients for response variables for grass, legume, and forb crops in 2008 and 2009.

Total crop biomass LERt

2008 2009 2008 2009

Biomass

Grass 0.81*** 0.90*** 0.35** 0.28*

Legume -0.021 0.38** 0.15 0.18

Forb 0.40** 0.21 0.14 0.45***

Percent cover

Total 0.63*** 0.70*** 0.59*** 0.53***

Grass 0.41** 0.46*** 0.41** 0.32**

Legume -0.035 0.24 0.14 0.081

Forb 0.35** 0.075 0.21 0.39**

Crop height 0.66*** 0.83*** 0.29* 0.45***

Seeding rate 0.25 0.48*** 0.51*** 0.27*

*Significant at P<0.05

**Significant at P<0.01

***Significant at P<0.001

117

Chapter 5: Smother Crop Mixtures for Canada Thistle Suppression in Organic

Transition

Stephanie Wedryk and John Cardina*

Canada thistle is a noxious weed in temperature agriculture that poses a particular threat to organic producers. The life cycle, growth, and development of Canada thistle are seasonally affected and exploiting this biology may be useful for weed management. The objective of this study was to evaluate smother crop mixtures seeded, at different times, for Canada thistle control. Field trials were established in 2009 and 2010 to evaluate the ability of smother crop mixtures to suppress Canada thistle growth and development.

Canada thistle biomass was suppressed 50% in 2009 and 87% in 2010 by the sorghum- sudangrass mixture, averaged over planting times. The oat mixture suppressed annual weed biomass more than 58% in 2009 and 67% in 2010 in all planting dates. Canada thistle shoot density and percent cover were affected by crop mixture in 2009 and 2010, with sorghum-sudangrass being the most suppressive. Planting date affected smother crop suppression of Canada thistle growth, but the effect was not consistent between

2009 and 2010 due to differences in weather conditions.

Nomenclature: Canada thistle, Cirsium arvense (L.) Scop.

Keywords: smother crop, organic weed management

* Graduate Research Associate and Professor, Department of Horticulture and Crop Science, The Ohio State University, Wooster, OH 44691. Corresponding author‟s E-mail: [email protected]. 118

Canada thistle [Cirsium arvense (L.) Scop] is a noxious weed throughout temperate agriculture that causes extensive yield losses, particularly in organic production where the use of herbicides is banned (Holm et al. 1977; Turner et al. 2007).

Canada thistle can infest new fields by seeds or vegetative reproduction through deep and extensive root systems that have been found up to 6.75 m below the soil surface (Donald

1994a; Evans 1984). Through these propagative roots, Canada thistle produces new shoots that can occupy above-ground space in a field. Previous research to address

Canada thistle management has closely followed farmers‟ practices in examining mechanical control methods (Riemens et al. 2010). However, cultivation and tillage machinery cut the roots into smaller pieces that can be spread more widely (Evans 1984).

Mowing can be effective for Canada thistle, but is not appropriate for annual crop production (Hodgson 1958).

Canada thistle shoot emergence from underground buds begins in spring and peaks in June or July (Donald 1994a). Root carbohydrate reserves are depleted during the

“bud to bloom” stage that starts during early summer (Donald 1994a; McAllister and

Haderlie 1985b). Carbohydrate reserves are replenished by photosynthesis during the fall months before cool temperatures limit growth (Donald 1994a). Strategies that take advantage of this seasonal cycle in Canada thistle biology have been used for management. In pastures, for example, appropriately timed mowing of established perennial species such as white clover (Trifolium repens L.) with grass or red clover

(Trifolium pretense L.) has been effective at reducing Canada thistle populations after 3 years (Graglia et al. 2006). The perennial species present in early spring may compete

119 with emerging Canada thistle shoots and decrease replenishment of root carbohydrate reserves in the fall. Summer crops of buckwheat (Fagopyrum esculentum Moench.) or a mixture of sudangrass [Sorghum sudanese (P.) Stapf.] and cowpea [Vigna unguiculata

(L.) Walp.], in combination with mowing, have also been shown to reduce Canada thistle biomass and shoot density (Bicksler and Masiunas 2009). The use of crops that have maximum growth during summer when Canada thistle root reserves are diminished may be an effective strategy for suppression. However, mowing is an integral part of these strategies and the effect of competitive crops on Canada thistle growth is confounded by mechanical management.

Smother crops are living plant species or mixes of species growing alone or in combination with a main crop to reduce the growth, development, and reproduction of undesirable plants through resource competition (Teasdale 1998). If smother crops could effectively suppress Canada thistle growth and reproduction, the need for mechanical or chemical inputs could be reduced. This is especially important during the transition from conventional to organic agriculture, when the use of herbicides is prohibited and growers need methods to reduce weeds in preparation for organic production. Annual smother crops with the potential for rapid biomass production may be more effective in suppressing Canada thistle than perennial species. Additionally, the use of multi-species smother crop mixtures may be more effective at suppression than monocultures due to occupation of different above- and below-ground niches within the stand (Creamer and

Baldwin 2000; Linares et al. 2008). Mixtures to suppress Canada thistle may also diverge ecologically from each other in order to compete most effectively at different stages in

120 the life cycle of Canada thistle. However, previous research has not addressed the functionality of smother crop mixtures adapted to different stages of Canada thistle growth.

This research was conducted to compare differently adapted smother crop mixtures for suppression of Canada thistle. Smother crop mixtures containing species adapted to different growing conditions might exert different competitive effects on

Canada thistle and the competitive effects would depend on time of planting. Mixtures of smother crops with species adapted to cool temperatures, species capable of rapid spread and growth, and species adapted to warm temperatures with potential for high biomass output were developed. Each mixture may be more effective at competing with Canada thistle at different temperatures and seasonal stages of Canada thistle growth. Our hypothesis is that Canada thistle suppression will be a function of planting date of crops and the ecological adaptation of crop mixtures. Therefore, we conducted studies with the objective to determine the suppressive ability of differently adapted smother crop mixtures on Canada thistle seeded at different stages of development.

Materials and Methods

Experimental Design. Field experiments were conducted at The Ohio State University

Schaffter Farm near Wooster, Ohio, in 2009 and 2010 to evaluate time of planting and species composition of smother crop mixtures for Canada thistle management. The soil

121 type at the site was classified as a fine, mixed, Typic Fragiaqualf (Luvisols) of the

Wooster series, a moderately well-drained silt loam with pH of 7.3, 2.9% organic matter, and available P, K of 21.3 and 80.6 mg kg-1 soil. The experiment was managed without the use of pesticides or fertilizers to represent the transition period from conventional to organic production. The experimental design was a 3x4 factorial with three planting dates and three smother crop mixtures and a non-treated control with four replications.

Individual plots were 3- x 6-m. Smother crop mixture composition, seeding rates and depths in 2009 and 2010 are listed in Table 5.1. Planting dates are described in Table 5.2.

The three planting dates will be referred to as „early,‟ „middle,‟ and „late‟ and the cropping treatments as „oat‟ (Avena sativa L.), „sorghum-sudangrass‟ [Sorghum bicolor

(L.) Moench. x Sorghum sudanese (Piper) Stapf.], and „tef‟ [Eragrostis tef (Zucc.)

Trotter] hereafter. The oat mixture was designed to represent an early season-adapted mix that may be able to effectively compete under cooler temperatures when Canada thistle shoots are emerging and beginning to form rosettes. The tef mixture contains tef that can be planted densely and is composed of species that are low-growing and mature rapidly.

The quick-growing species in the tef mixture may be able to outcompete Canada thistle for resources and reduce light attenuation. Crop species in the sorghum-sudangrass mixture are capable of high-biomass production and are adapted to warmer temperatures when Canada thistle carbohydrate reserves are depleted. Planting dates were chosen to target competitive effects at different stages of Canada thistle biology and underground storage capacity of carbohydrate reserves (McAllister and Haderlie 1985). The early planting date was the first available time for planting the smother crop mixtures with

122 decreasing concentrations of root carbohydrates at subsequent planting dates. The late planting date was chosen to represent the time when the lowest amount of root carbohydrates were available for Canada thistle growth.

The field was disked and seedbed prepared before planting. Soybean seeds were inoculated with Rhizobium bacteria at rate of 2.5 g TerraMax Dry per kg seed immediately prior to planting.1 Treatments were drill-seeded using a Great Plains no-till drill with row spacing of 18 cm, with two offset passes of the drill per plot to obtain more uniform planting pattern. Tef was broadcast seeded on appropriate plots after drill- seeding and incorporated into the soil using a roller.

Data Collection. Pre-experiment visual assessments of percent cover and shoot density of

Canada thistle were conducted on May 4, 2009, and April 16, 2010. Percent cover was estimated visually, and shoot density was determined by counting emerged Canada thistle shoots in four 0.09-m2 quadrats per plot. Before each planting date, density of Canada thistle shoots was counted in four 0.09 m2 quadrats per plot. Shoot density Canada thistle was measured every wk starting 1 wk after planting for 4 wk and again at destructive biomass harvests in four permanent 0.09 m2 quadrats per plot. Density of grass, legume, and forb components of the smother crop mixtures were measured every wk starting 1 wk after planting for 4 wk in two permanent 0.09 m2 quadrats per plot in 2009. Percent cover of grass, legume, and forb crops, Canada thistle, and annual weeds was visually assessed using two permanent 0.09 m2 quadrats per plot every wk starting 2 wk after planting and continuing until biomass harvest. Height of the grass component of the smother crops

123 and Canada thistle was measured every week starting 2 wk after planting and continuing until biomass harvest in 2009. Biomass was harvested from two 0.09 m2 quadrats per plot when all component crops of the smother crop mixture reached maturity. Harvest dates for each planting date and cropping treatment are listed in Table 5.2. Non-treated control plots were harvested at the mid-point between harvest of the first and last cropping treatments within a planting date to represent growth of Canada thistle and annual weeds without smother crop for the entire planting time. Harvested biomass was separated into grass, legume, and forb crops, Canada thistle, and annual weeds, weighed, dried at 55C for 72 h, and weighed again.

Statistical Analysis. Data were analyzed separately by year due to differences in planting and harvest dates. Data for biomass of crops, Canada thistle, and annual weeds and final percent cover of crops were subjected to ANOVA in SAS v9.2 to test for the effects of planting date, smother crop mixture, and the interaction of planting date and smother crop mixture.2 When the interaction of planting date and smother crop mixture was significant

(P < 0.05), data were analyzed separately for each planting date x smother crop treatment. Data that did not meet the assumptions of ANOVA were square root transformed prior to analysis, and means separated by Fisher‟s Protected LSD (P < 0.05).

Data for final percent cover and shoot density of Canada thistle were subjected to analysis of covariance to account for differences in the Canada thistle populations at experiment initiation. Visual estimates of initial percent cover before experiment initiation and Canada thistle shoot count before seeding at each planting date were used

124 as covariates in a General Linear Model in SAS v9.2. The effects of planting date, smother crop mixture, and the interaction of planting date and smother crop mixture on final Canada thistle percent cover and shoot density were tested for in the General Linear

Model and data were analyzed separately when the interaction of planting date and smother crop mixture was significant (P < 0.05). Differences between means adjusted for the covariate were determined using the PDIFF option in PROC GLM. Pearson correlation coefficients were calculated for relationships between biomass measures, percent cover, crop emergence, Canada thistle shoot density, and Canada thistle and crop height. Significant correlations (P < 0.05) between response variables with r < 0.4 were considered weak; 0.4 < r < 0.8 were considered moderate; r < 0.8 were considered strong.

Results and Discussion

Monthly total precipitation and maximum and minimum temperatures in 2009 and

2010 did not follow the climatic patterns of the 20-yr mean (Figure 5.1). Precipitation was 88% and 38% higher than the 20-yr mean in June and July of 2010 and 32% lower in

August 2010. In 2009, total precipitation was 75% higher than the 20-yr mean (Table

5.3). The mean monthly maximum temperature was lower than the 20-yr mean in June,

July, August, and September 2009. The mean monthly minimum temperature was 6% to

32% higher than the 20-yr mean in 2010 (Figure 5.1). The change in precipitation and temperature in 2009 and 2010 may have contributed to differences in the magnitude of

125 crop, Canada thistle, and annual weed biomass production between 2009 and 2010 (Table

5.5; Table 5.6).

Due to the differences in climatic conditions between 2009 and 2010, results were analyzed separately by year. In 2009 and 2010, planting date and crop mixture affected

Canada thistle biomass in 2009 and 2010 (Table 5.4). Biomass of Canada thistle in the early planting date was more than 52% and 61% greater than the middle or late planting dates in 2009 and 2010 (Table 5.5). Canada thistle biology is strongly affected by seasonal fluctuations in root carbohydrate reserves (McAllister and Haderlie 1985b).

Field operations followed by subsequent planting of smother crops when carbohydrate reserves are low in early summer could have affected the ability of Canada thistle to compete with crops for resources. In 2009, all crop mixtures suppressed Canada thistle biomass by greater than 50% compared to the non-treated control (Table 5.5). However, only the sorghum-sudangrass mixture suppressed Canada thistle biomass in 2010 (Table

5.5). The warmer temperatures in 2010 could have favored the growth of warm-season adapted crops in the sorghum-sudangrass mixture (Figure 5.1). The oat mixture adapted to cooler temperatures may have been able to compete more effectively with Canada thistle in 2009 (Figure 5.1).

Crop biomass was affected by planting date in 2009 and crop mixture in 2009 and

2010. The interaction of planting date x crop mixture was not significant for Canada thistle or crop biomass (P > 0.05) (Table 5.4). The sorghum-sudangrass mixture produced 44% and 71% more biomass than the oat or tef mixtures in 2009 and 2010

(Table 5.5). The interaction between planting date and crop mixture was significant for

126 annual weeds in 2009 and 2010 (Table 5.4). The oat mixture at each planting date in 2009 and 2010 effectively suppressed the biomass of annual weeds compared to the non- treated control (Table 5.6). The higher seeding rate (114 kg ha-1) of crops in the oat mixture may have contributed to greater weed suppression through greater occupation of space by crops (Table 5.1). The oat mixture was harvested earlier than other mixtures within each planting date and the amount of time between seeding and harvest could have affected biomass production by competing annual weeds (Table 5.2). In 2010, all crop mixtures effectively suppressed annual weeds in comparison to the non-treated control

(Table 5.6). The presence of crop species in mixture was able to effectively compete with annual weeds for resources and suppress their growth.

The ability of Canada thistle to occupy space through shoot emergence was affected by the planting date and crop mixture in 2009 and 2010 after fall harvest and the following spring (Table 5.7). The interaction between planting date and crop mixture was significant in 2010 for Canada thistle shoot density in the fall after harvest and the following spring. The quantity of emerged Canada thistle shoots in 2009 was 25% and

28% greater in the late planting date than the middle and early planting dates in the fall

(Table 5.8). In the following spring, the density of Canada thistle shoots was reduced by

28% in the middle planting date compared to the late planting date. However, in fall

2010, shoot density was greatest in the early planting date (oat mixture = 33 shoots m-2) and least in the late planting date (sorghum-sudangrass mixture = 0.07 shoots m-2) (Table

5.9). In spring 2011 after harvest the previous fall, Canada thistle shoot density was greatest in the early planting date (oat mixture = 55 shoots m-2) and lowest in the late

127 planting date (tef mixture = 0.09 shoots m-2). The elongation of root buds and emergence from the soil in Canada thistle is regulated by temperature (McAllister and Haderlie

1985a). The cooler temperatures in 2009 may have caused a shift in the flowering time and deposition of carbohydrates in root reserves (Figure 5.1). In fall 2009 and at the early and middle planting dates in fall 2010, the shoot density of Canada thistle was reduced in the sorghum-sudangrass crop mixture compared to the non-treated control (Table 5.8;

Table 5.9). Previous studies have shown that biomass of crops competing with weeds can function as a proxy for competitive ability (Gaudet and Keddy 1988). The sorghum- sudangrass crop mixture may be more effective at competing with Canada thistle for resources as evidenced by higher biomass output (Table 5.5). In spring measurements of

Canada thistle shoot density in both years, the suppressive effect of smother crop mixtures in comparison to the non-treated control was not apparent (Table 5.8; Table

5.9). Since there was no cropping cover during the late fall, winter, and early spring,

Canada thistle would be able to photosynthesize and vegetatively spread. Continuous cover by smother cropping treatments has the potential to further suppress Canada thistle emergence before spring-planted smother crops can be seeded.

The cover of Canada thistle as estimated visually by percentage cover was affected by crop mixture in both years (Table 5.7). The sorghum-sudangrass mixture suppressed spread of Canada thistle by 74% in 2009 and 92% in 2010 (Table 5.10).

Percent cover of Canada thistle was not affected by the oat or tef crop mixtures in 2009 and 2010 (Table 5.10). In 2009 and 2010, percent cover of crops was affected by crop mixture (Table 5.4). Sorghum-sudangrass mixtures covered more ground than the oat or

128 tef mixtures in both years (Table 5.10). The results of visual estimates of percent cover suggest that sorghum-sudangrass was able to grow into two-dimensional space at the expense of lateral propagation of Canada thistle. Canada thistle spreads through underground roots that can occupy unused space, although how Canada thistle roots can sense unoccupied space is not clear (Donald 1994b). In the early planting date in 2010, percent cover of Canada thistle was more than 51% greater than the middle and late planting dates (Table 5.10). Similarly, the percent cover of crops in 2010 was affected by planting date with cover being reduced by greater than 34% (Table 5.4; Table 5.10). The early planting date could have favored growth of Canada thistle at the expense of smother crops due to the ability to regrow after field operations before depletion of carbohydrate reserves. The ability of crops to occupy space can affect its competitive ability with a weed.

Pearson correlation coefficients were calculated between measured response variables for 2009 and 2010 to determine what parameters of crop growth may contribute to suppression of Canada thistle (Table 5.11; Table 5.12). The height of crops were moderately inversely correlated with percent cover of Canada thistle (r = -0.49; P <

0.001) and Canada thistle shoot density (r = -0.40; P < 0.001) in 2009 (Table 5.11). The height potential of crops in mixture can reflect the ability of the crops to outcompete

Canada thistle for resources, particularly light. After canopy closure, light may not reach emerged Canada thistle shoots, preventing growth. Sorghum-sudangrass crop mixtures grew taller than the oat or tef mixtures and also suppressed Canada thistle growth more than other mixtures (Table 5.5; Table 5.8; Table 5.9; height data not shown). Other

129 relationships tested were not significant, were weak relationships, or do not contribute to understanding how smother crops can suppress Canada thistle.

In summary, suppression of Canada thistle with smother crops depended on the planting date and crop mixture. The sorghum-sudangrass mixture suppressed Canada thistle more effectively than the oat and tef mixtures regardless of planting date, despite different ecological adaptation of the mixtures to planting conditions and temperatures.

The oat mixture suppressed annual weeds more effectively than the other mixtures at each planting date. The success of the sorghum-sudangrass and oat mixtures at suppressing Canada thistle and annual weeds, respectively, may also be attributed to the allelopathic potential of component species in each mix. Canada thistle suppression tended to be more effective when smother crop mixtures were planted at a nadir in

Canada thistle root carbohydrate reserves. However, changes in annual climatic conditions can affect Canada thistle biology and planting date should reflect this for better suppression. Further research is needed to test additional species in mixture for

Canada thistle suppression or additional planting dates in late summer or winter smother crops.

Sources of Materials

1 Sprint Royal oats and Packer peas. La Crosse Forage and Turf Seed Crop., 2541

Commerce St., La Crosse, WI 54603.

130

2 SAS 9.2 Statistical Software, Statistical Analysis Systems, SAS Institute, Inc., 100 SAS

Campus Drive, Cary NC 27513-2414.

Acknowledgements

Agriculture Research and Development Center, The Ohio State University. Manuscript

No. HCS-09-00

131

Literature Cited

Bicksler, A. J. and J. B. Masiunas. 2009. Canada thistle (Cirsium arvense) suppression with buckwheat or sudangrass cover crops and mowing. Weed Technol. 23:556- 563.

Creamer, N.G. and K.R. Baldwin. 2000. An evaluation of summer cover crops for use in vegetable production systems in North Carolina. HortScience 35:600-603.

Donald, W.W. 1994a. The biology of Canada thistle (Cirsium arvense). Rev. Weed Sci. 6:77-101.

Donald, W.W. 1994b. Geostatistics for mapping weeds, with a Canada thistle (Cirsium arvense) patch as a case study. Weed Sci. 42:648-657.

Evans, J. E. 1984. Canada thistle (Cirsium arvense): a literature review of management practices. Natural Areas Journal 4:11-21.

Gaudet, C. L. and P. A. Keddy. 1988. A comparative approach to predicting competitive ability from plant traits. Nature 334:242-243.

Graglia, E., B. Melander, and R.K. Jensen. 2006. Mechanical and cultural strategies to control Cirsium arvense in organic arable cropping systems. Weed Res. 46:304- 312.

Hodgson, J.M. 1958. Canada thistle (Cirsium arvense Scop.) control with cultivation, cropping, and chemical sprays. Weeds 6:1-11.

Holm, L.G., D.L. Plucknett, J.V. Pancho, and J.P. Herberger. 1977. The world‟s worst weeds, distribution and biology. Honolulu: University Press of Hawaii. 607 p.

McAllister, R. S. and L. C. Haderlie. 1985a. Effects of photoperiod and temperature on root bud development and assimilate translocation in Canada thistle (Cirsium arvense). Weed Sci. 33:148-152.

McAllister, R. S. and L. C. Haderlie. 1985b. Seasonal variations in Canada thistle (Cirsium arvense) root bud growth and root carbohydrate reserves. Weed Sci. 33:44-49.

132

Riemens, M. M., R. M. W. Groeneveld, M. J. J. Kropff, L. A. P. Lotz, R. J. Renes, W. Sukkel, and R. Y. van der Weide. 2010. Linking farmer weed management behavior with weed pressure: more than just technology. Weed Sci. 58:490-496.

Turner, R. J., G. Davies, H. Moore, A. C. Grundy, and A. Mead. 2007. Organic weed management: a review of the current UK farmer perspective. Crop Prot. 26:377- 382.

133

Table 5.1. Varieties, seeding rates and depths of smother crop mixtures in 2009 and 2010.

Smother crop mixture Variety Seeding rate Seeding depth

kg ha-1 cm

„Oat‟

Oat Royal 108a 2.5

Field pea (Pisum sativum L.) Packer - 2.5

India mustard (Brassica juncea L.) Florida 6 1.2

broadleaf

„Sorghum-sudangrass‟

Sorghum-sudangrass Special Effort 25 1.2

Soybean [Glycine max (L.) Merr.] Stonewall 20 2.5

Sunflower (Helianthus annuus L.) 620CL 3 2.5

„Tef‟

Tef VA-T1 27 Surface

Burr medic (Medicago Santiago 8 1.2 polymorpha L.)

Buckwheat Common 25 2.5 a Oat and field pea seeded as commercially available Sprint oat and pea mix.1

134

Table 5.2. Dates of planting and harvesting of smother crop mixtures in 2009 and 2010.

Planting date Harvest date

Early Middle Late Early Middle Late

Smother crop mixture 2009 2010 2009 2010 2009 2010 2009 2010 2009 2010 2009 2010

Non-treated 5/13 5/7 5/28 5/28 6/9 6/16 8/7 8/17 8/24 8/30 9/9 9/23

Oat-field pea-mustard 5/13 5/7 5/28 5/28 6/9 6/16 7/23 7/9 8/14 7/22 8/24 8/10

Sorghum-sudangrass- 5/13 5/7 5/28 5/28 6/9 6/16 8/26 8/30 9/9 9/23 9/22 9/23

13 sunflower-soybean

5 Tef-burr medic-buckwheat 5/13 5/7 5/28 5/28 6/9 6/16 8/7 8/17 8/21 8/30 9/22 9/23

135

50 Mean maximum temperature 2009 Mean maximum temperature 2010 20-yr mean maximum temperature 40 Mean minimum temperature 2009 Mean minimum temperature 2010 20-yr mean minimum temperature

Temperature (°C) Temperature

0 May June July Aug Sept Month

Figure 5.1: Mean monthly maximum and minimum temperatures (C) in 2009, 2010, and 20-yr mean.

136

Table 5.3. Monthly total precipitation in 2009 and 2010 and 20-yr mean.

Monthly total precipitation

Year May June July Aug Sep

2009 8.1 9.5 7.4 14.9 6.7

2010 10.7 17.2 12.4 5.8 7.1

20-yr mean 9.47 9.13 8.98 8.52 7.32

137

Table 5.4. Analysis-of-variance results for biomass of Canada thistle (CIRAR), annual weeds, and crops and percent cover of

crops in 2009 and 2010.

Probability values

Biomass Percent cover

CIRAR Annual weeds Crops Crops

Main Effects 2009 2010 2009 2010 2009 2010 2009 2010

PD 0.0001 0.0001 0.46 0.001 0.03 0.36 0.07 0.004

138 CM 0.0002 0.05 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001

PD x CM 0.11 0.11 0.0006 0.0001 0.09 0.08 0.72 0.33

Abbreviations: PD, planting date; CM, cropping mixture.

138

Table 5.5. Canada thistle (CIRAR) and crop biomass in 2009 and 2010 at the early, middle, and late planting dates and in cropping mixtures.

Biomass

CIRAR Crop

Planting date 2009a 2010a 2009a 2010a

g m-2

Early 199 a 28 a 512 b 434 a

Middle 96 b 11 b 639 a 544 a

Late 81 b 0.84 c 402 b 582 a

Smother crop mixture

Non-treated 210 a 14 a N/A N/A

Oat-field pea-mustard 81 b 19 a 349 b 109 c

Sorghum-sudangrass- 106 b 1.8 b 771 a 1130 a sunflower-soybean

Tef-burr medic-buckwheat 102 b 18 a 433 b 325 b a Means within a column followed by the same letter do not differ according to Fisher‟s

Protected LSD (P < 0.05).

139

Table 5.6. Biomass of annual weeds in 2009 and 2010 in each cropping treatment at the early, middle, and late planting dates.

Biomass

Annual weeds

Planting date Smother crop mixture 2009a 2010a

g m-2

Early Non-treated 276 ab 645 b

Oat-field pea-mustard 85 e 189 efg

Sorghum-sudangrass-sunflower-soybean 223 abc 452 c

Tef-burr medic-buckwheat 185 bcd 279 def

Middle Non-treated 285 ab 524 c

Oat-field pea-mustard 121 de 171 fg

Sorghum-sudangrass-sunflower-soybean 191 bcd 84 g

Tef-burr medic-buckwheat 138 cde 330 d

Late Non-treated 308 a 844 a

Oat-field pea-mustard 14 f 110 g

Sorghum-sudangrass-sunflower-soybean 251 ab 69 g

Tef-burr medic-buckwheat 198 bcd 309 de a Means within a column followed by the same letter do not differ according to Fisher‟s

Protected LSD (P<0.05).

140

Table 5.7. Analysis-of-covariance results for Canada thistle (CIRAR) shoot density after fall harvest and the following spring and percent cover at harvest in 2009 and 2010.

Probability values

CIRAR shoot density CIRAR percent cover

Fall post-harvest Spring

Main effects 2009 2010 2010 2011 2009 2010

PD 0.004 0.0001 0.015 0.0001 0.08 0.0001

CM 0.0001 0.0001 0.003 0.001 0.0001 0.003

PD x CM 0.20 0.007 0.60 0.0002 0.66 0.11

Covariate

Initial density 0.0001 0.68 0.0001 0.0001 0.94 1

Abbreviations: PD, planting date; CM, cropping mixture.

141

Table 5.8. Canada thistle shoot density as affected by planting date and smother crop mixture after biomass harvest in the fall of 2009 and the following spring in 2010.

Canada thistle shoot density

Planting date Fall 2009a Spring 2010a

shoots m-2

Early 44 b 48 ab

Middle 46 b 39 b

Late 61 a 54 a

Smother crop mixture

Non-treated 57 ab 45 b

Oat-field pea-mustard 49 b 59 a

Sorghum-sudangrass- 33 c 35 b sunflower-soybean

Tef-burr medic-buckwheat 62 a 50 a a Means within a column followed by the same letter do not differ according to Fisher‟s

Protected LSD (P < 0.05).

142

Table 5.9. Canada thistle shoot density as affected by the interaction of planting date and smother crop mixture after biomass harvest in fall 2010 and the following spring 2011.

Canada thistle shoot density

Planting date Smother crop mixture Fall 2010a Spring 2011a

shoots m-2

Early Non-treated 22 b 29 c

Oat-field pea-mustard 33 a 55 a

Sorghum-sudangrass-sunflower-soybean 3.3 cd 45 ab

Tef-burr medic-buckwheat 24 ab 42 b

Middle Non-treated 11 c 20 cd

Oat-field pea-mustard 25 ab 21 cd

Sorghum-sudangrass-sunflower-soybean 0.14 d 2.5 e

Tef-burr medic-buckwheat 10 c 9.8 d

Late Non-treated 2.0 d 5.1 e

Oat-field pea-mustard 4.1 cd 4.5 e

Sorghum-sudangrass-sunflower-soybean 0.07 d 0.57 e

Tef-burr medic-buckwheat 0.71 d 0.09 e a Means within a column followed by the same letter do not differ according to Fisher‟s

Protected LSD (P<0.05).

143

Table 5.10. Percent cover of Canada thistle (CIRAR) and crops in the early, middle, and late planting dates and in the cropping mixtures in 2009 and 2010.

Percent cover

CIRAR Crops

Planting date 2009a 2010a 2009a 2010a

Early 27 ab 7.8 a 57 a 33 b

Middle 24 b 3.8 b 61 a 51 a

Late 36 a 0.34 b 50 a 50 a

Smother crop mixture

Non-treated 47 a 5.2 a N/A N/A

Oat-field pea-mustard 23 ab 6.2 a 52 b 14 c

Sorghum-sudangrass-sunflower-soybean 12 b 0.42 b 72 a 77 a

Tef-burr medic-buckwheat 34 a 4.1 a 44 b 42 b a Means within a column followed by the same letter do not differ according to Fisher‟s

Protected LSD (P < 0.05).

144

Table 5.11. Pearson correlation coefficients for measured response variables of Canada thistle (CIRAR), annual weed, and crop

biomass populations in 2009.

Biomass Percent cover Emergence Shoot density Height

Annual Crops CIRAR Crops Crops CIRAR CIRAR Crops

weeds

Biomass

CIRAR 0.03 -0.23* 0.44*** -0.28** -0.11 0.30** 0.45*** -0.16

Annual weeds -0.16 -0.09 -0.31** -0.29** -0.27** 0.07 -0.20*

145

Crops -0.29** 0.61*** 0.06 -0.18 0.19 0.69***

Percent cover

CIRAR -0.03 0.64*** 0.09 -0.49***

Crops 0.21* -0.37*** 0.04 0.83***

Crop emergence 0.07 -0.02 -0.06

Continued 145

Table 5.11 continued

Biomass Percent cover Emergence Shoot density Height

Annual Crops CIRAR Crops Crops CIRAR CIRAR Crops

weeds

Percent cover

CIRAR shoot -0.03 -0.40***

density

CIRAR height 0.16

146

*Significant at P < 0.05

**Significant at P < 0.01

***Significant at P < 0.001

146

Table 5.12. Pearson correlation coefficients for measured response variables of Canada thistle (CIRAR), annual weed, and crop biomass populations in 2010.

Biomass Percent cover Shoot density

Biomass Annual weeds Crops CIRAR Crops CIRAR

CIRAR -0.07 -0.20* 0.41*** -0.19 0.53***

Annual weeds -0.48*** 0.02 -0.51*** -0.05

Crops -0.31** 0.84*** -0.41***

Percent cover

CIRAR -0.34*** 0.65***

Crops -0.40***

*Significant at P < 0.05

**Significant at P < 0.01

***Significant at P < 0.001

147

Chapter 6: Strategies for weed suppression and improvement of soil fertility during

transition from conventional to organic vegetable production

Stephanie Wedryk, Joel Felix, Doug Doohan, and John Cardina*

Farmers view weed management and the risk of lower yields as barriers to transition from conventional to organic agriculture. The three transition years before organic certification can be used to implement strategies to suppress weeds and improve soil fertility before organic production. The objective of this research was to evaluate the effects of five organic transition strategies on soil quality, weed suppression, and yield of tomato and potato in the first year of organic production. The transition strategies included a tilled fallow, non-treated weedy, high diversity prairie mixture, smother crops, and vegetable rotation. Subplots with and without compost application were also included. Transition strategies affected weed density and biomass in the first organic year with the prairie strategy being the most suppressive. Compost application increased plant available nutrients and soil organic matter. The fallow transition strategy improved the quantity of plant available P and K by 33% and 15%, respectively, while the prairie strategy improved soil organic matter by 10% or more. Compost application increased

* First, third, and fourth authors: Graduate Research Associate, Professor, and Professor, Department of Horticulture and Crop Science, The Ohio State University, 1680 Madison Avenue, Wooster, OH 44691; second author: Assistant Professor, Crop and Soil Science Department, Oregon State University, Malheur Experiment Station, 595 Onion Avenue, Ontario, OR 97914. 148

yields of potato by 50% and tomato by 17% with transition strategy affecting the number and weight of cull potato tubers. Canonical correlation analysis showed that plant available nutrients strongly influenced potato yield while organic matter influenced tomato yield. The selection of transition strategies before conversion to organic agriculture may affect weed pressure, soil quality, and crop production in the first year.

Nomenclature: Tomato, Solanum lycopersicum L., potato, Solanum tuberosum L.

Keywords: Organic weed management, smother crop, prairie, canonical correlation, soil quality.

149

Environmental quality and public health concerns about the use of chemicals in conventional agriculture have driven a large increase in demand for organic food (Dorais

2007). Acreage of certified organic farm land in the United States has increased by 81% between 1995 and 2008 (USDA 2010). The number of acres in certified organic vegetable production has increased by 62% between 2000 and 2008 (USDA 2010).

Despite the increased demand for organic products and higher price premiums realized by growers, many farmers are reluctant to convert to organic agriculture due to perceived risks of lower yields and challenges in managing pests (Beveridge and Naylor 1999;

Hanson et al. 2004; Oberholtzer et al. 2005). Weed control, especially management of perennial weeds, is often cited by farmers as a barrier to transition or organic production

(Turner et al. 2007; Verschwele and Häusler 2004). During the mandated 3-yr transition period from conventional to organic agriculture in the United States, many farmers adopt biological, cultural and mechanical techniques that aid in building soil fertility and suppressing weeds with the potential to enhance yields in the first year of certified organic production (Hanson et al. 2004). Improvement of existing transition strategies and implementation of novel transition strategies may encourage farmers to convert to organic practices if the risk of lower yields and weed competition is minimized.

Organic farmers have identified the improvement of soil fertility and pest management as the most crucial research needs for vegetable production (Delate et al.

2003). Research conducted on transition strategies for organic vegetable production has included several approaches to build soil fertility and improve yields. Compost applications during transition to organic vegetable production have helped to improve

150 soil organic matter and microbial biomass, alter nematode communities, and increase the concentration of plant available macro- and micronutrients (Briar et al. 2011; Martini et al. 2004; Smukler et al. 2008). Improvements in soil quality as a result of organic material addition are often associated with improvements in vegetable yield (Briar et al.

2011; Russo and Taylor 2006). However, the cropping system used during transition can interact with compost application to affect increases in yield and soil fertility. Intensive vegetable production in high-tunnels resulted in higher tomato (Solanum lycopersicum

L.) yields in the first organic year than other transition strategies including tilled fallow, hay, and field vegetable production (Briar et al. 2011). Transition strategies for organic grain production have used different cover crops, cropping rotations, and tillage regimes to both suppress weeds and improve soil fertility. Weed suppression and grain yields were affected by the species of cover crop used during 3 years of transition (Smith et al.

2011). Transitional strategies of different agronomic crop rotations affected weed seedbank density and soil aggregate size in the first year of certified organic production

(Corbin et al. 2010). Weed competition and soil fertility during organic transition are affected by compost application, transition strategy approach, cover crops, and crop species. However, previous research has not compared potential differences in transition strategies to simultaneously suppress weeds, improve soil fertility, and increase yields in organic vegetable production.

This research was conducted to assess cropping and management strategies for weed suppression, enhancing soil fertility and organic vegetable yields during transition to organic agriculture in Ohio. The 3-yr transition period was considered a time to

151 improve agricultural conditions for optimum yields after organic certification and not as a period itself for organic production. Five transition strategies were implemented to mirror current farmer and researcher practices such as growing vegetables, tilled fallow, and mowing, with novel strategies such as prairie biomass production and summer annual smother crops. The objective of this study was to evaluate soil quality, response of weed populations, and compare yields in tomato and potato (Solanum tuberosum L.) production in the first certifiable organic year after 3 years of transition strategies and compost application. We hypothesize that weed competition and soil quality will be affected by 3 years of transition strategy and compost application. Further, we hypothesize that transition strategy and compost application as well as resulting weed populations and soil fertility will affect tomato and potato yields.

Materials and Methods

Experimental Design. Field experiments were conducted at The Ohio State University

Schaffter Farm near Wooster, Ohio, from 2007 to 2010 to evaluate the effect of five organic transition strategies on weed suppression, soil fertility, and production of organic tomato and potato. The soil type at the site was classified as a fine, mixed, Typic

Fragiaqualf (Luvisols) of the Wooster series, a moderately well-drained silt loam with pH of 6.8, 2.2% organic matter, and available P, K of 26.8 and 116.7 mg kg-1 soil. The

152 experiment was managed without pesticides or chemical fertilizers to represent the transition from conventional to organic agriculture.

Organic transition strategies. Five organic transition strategies were established in 2007 and continued through 2009 to represent the three years of organic transition. The experimental setup was a split-plot design with six replications. Main plot factors were organic transition strategy and sub-plot factors were compost application. Main plot dimensions were 12 m x 15 m. Experimental plots were maintained with the same treatment each year. The five organic transition strategies included 1) tilled fallow; 2) non-treated; 3) tallgrass prairie; 4) vegetable; and 5) smother crops. The tilled fallow treatment was cultivated 3-4 times each growing season when Canada thistle [Cirsium arvense (L.) Scop] reached the green bud stage. In the non-treated strategy, weeds were mowed four times per growing season in order to prevent annual weeds from setting seed. The tallgrass prairie was broadcast seeded on June 12, 2007, and cultimulched to ensure seed and soil contact. Straw was subsequently laid on the plots to encourage prairie seed germination. Species included in the mix were one C3 grass [Canada wildrye

(Elymus canadensis L.)], two C4 grasses [big bluestem (Andropogon gerardii Vitman) and indian-grass (Sorghastrum nutans (L.) Nash)], one spiderwort [Ohio spiderwort

(Tradescantia ohioensis Raf.)], two legumes [wild senna (Senna hebecarpa (Fernald)

Irwin & Barnaby) and round bushclover (Lespedeza capitata Michx.)], and ten forbs

[gray-headed coneflower (Ratibida pinnata (Vent.) Barnhart), biennial gaura (Gaura biennis L.), stiff goldenrod (Solidago rigida L.), tall coneflower (Rudbeckia laciniata L.), smooth penstemon (Penstemon digitalis Nutt ex. Sims), false sunflower (Helianthus

153 helianthoides L.), brown-eyed susan (Rudbeckia triloba L.), tall coreopsis (Coreopsis tripteris L.), Lowrie‟s aster (Aster lowrieanus Porter), and New England aster (Aster novae-anglia (L.) G.L. Nesom)]. There was no additional management in the prairie transition strategy in 2008 or 2009.

The vegetable transition strategy followed an edamame soybean [Glycine max

(L.) Merr.]-lettuce (Lactuca sativa L.) and Brassicaceae species-edamame soybean rotation between 2007 and 2009. In 2007, edamame soybean cv. Gardensoy was seeded on June 7 at a rate of 296,000 seeds ha-1 in 76 cm rows. Cultivation was used to control inter-row weeds when weeds were approximately 4 cm tall. Edamame soybean was harvested September 15, 2007. In 2008, lettuce cvs. North Star, Green Towers, New Red

Fire, and Tropicana seeds were germinated under greenhouse conditions on April 14 and

April 28 and hand transplanted on May 22 and June 12 into black landscape cloth.1 Inter- row and intra-row spacing was 30 cm, respectively. Brassica crops, red cabbage

(Brassica oleracea var. capitata f. rubra) cv. Super Red 80, broccoli (Brassica oleracea var. botrytis) cv. Diplomat, Brussels sprouts (Brassica oleracea var. gemmifera) cv.

Oliver, and kale (Brassica oleracea var. viridis) cv. Winterbor were seeded in the greenhouse on May 13, May 27, and June 12, 2008.2 Seedlings were hand transplanted on

July 1, July 21, and August 4 into black landscape cloth after lettuce harvest. Inter-row spacing was 61 cm for all species and intra-row spacing for cabbage, kale, and Brussels sprouts was 30 cm and 61 cm for broccoli. Black landscape cloth was used for weed control and vegetables were irrigated throughout the season as necessary. In 2009,

154 edamame soybeans cv. Gardensoy were seeded on June 5 and followed the same management protocol as 2007.

The smother crop strategy followed a rotation of oat (Avena sativa L.) cv. Royal and field pea (Pisum sativum L.) cv. Packer mixture in 2007, tef [Eragrostis tef (Zucc.)

Trotter] cv. VA-T1 and sorghum-sudangrass [Sorghum bicolor (L.) Moench. x Sorghum sudanese (Piper) Stapf.] cv. Special Effort in 2008, and sorghum-sudangrass cv. Special

Effort in 2009.3,4,5 The oat-pea mixture was seeded on May 25, 2007, at 108 kg ha-1 and harvested July 12, 2007. In 2008, a split-split plot design was used in the smother crop transition strategy. Main plot factor was organic transition strategy, sub-plot factor was compost application, and sub-sub-plot factor was smother crop species. Tef was broadcast seeded at 27 kg ha-1 and cultipacked to ensure soil and seed contact on May 29,

2008. Sorghum-sudangrass was drill-seeded at 25 kg ha-1 on May 29, 2008. Tef and sorghum-sudangrass were mowed in October 2008 and residues were tilled under the following spring. In 2009, sorghum-sudangrass was planted May 24 with the same management as the previous year.

Compost was applied as a sub-plot factor each year of organic transition. The treatment levels were no compost and composted dairy manure applied at a rate of

3.9x104 kg crude compost ha-1. Compost was applied to the same sub-plots each year on

April 30, 2007, April 17, 2008, and April 15, 2009.

First-year organic production. In the first year of production after three years of organic transition strategies, tomato and potato production were evaluated. The experimental setup was a split-plot design with six replications. The main plot factor was organic

155 transition strategy and the sub-plot factor was compost application. Plots were divided for production of tomato and potato with weedy and weed-free sub-plots, which were considered separate experiments superimposed on the transition strategy and compost application study. The experiments for the respective crops constituted a split-split-split- plot design. On April 15, 2010, compost was applied to the appropriate sub-plots at the same rate as transition years. The field was disked and seedbed prepared before planting vegetables. Potato cv. Rio Grande (russet type) was planted on April 29, 2010, with seed tubers of approximately 35 g into 1.5 m rows with 30 cm intra-row spacing at a depth of

10 cm. Tomato seedlings cv. Heinz 3402 (processing variety) were grown in the greenhouse and mechanically transplanted on May 27, 2010 into 1.5 m rows with 30 cm intra-row spacing to give a final population of 24,000 plants ha-1. Weeds were controlled in the first organic year with inter-row cultivation when weeds were 5 cm and hand hoeing for intra-row control in non-weedy sub-sub-sub-plots. Weeds were not controlled in the weed sub-sub-sub-plots.

Data Collection. Soil samples were collected on April 15, 2007, before experiment initiation and on April 14, 2010, after three years of organic transition. Sixteen soil cores per subplot were collected with a diameter of 3 cm to a depth of 10 cm. Soil samples were amassed for each subplot and dried before analysis. Samples were tested for pH, available P using a Bray-1 extractant, exchangeable K, Ca, and Mg using an ammonium acetate extractant, and organic matter using the loss on ignition method.

156

Weed emergence was counted after planting of potato on May 24, 2010, and tomato on June 16, 2010 from 2-0.25 m2 quadrats per subplot to determine the effect of transition strategies on weed suppression. Weed biomass was harvested from tomato on

August 25, 2010 and potato plots on September 15, 2010 from 2-0.5 m2 quadrats per subplot to determine the effect of transition strategies and one year of organic production on weed suppression. All weed data were collected from the weedy sub-plots. Weed biomass was separated into Canada thistle, monocot and broadleaf weeds; these were weighed, dried at 55C for 72 h, and weighed again.

Tomatoes were harvested by hand from 1.5 m sections of two rows in each non- weedy sub-plot between August 26 and September 26, 2010. Harvest times varied by replication and transition strategy due to differences in time to maturity. Tomatoes were separated into marketable composed of ripe tomatoes meeting standards for categories A and B and non-marketable composed of tomatoes that were immature or culls (USDA

1983). The number of fresh weight of tomatoes in each category was measured. Six marketable tomato fruits from each sub-plot were ground for pH and BRIX analysis of fruit quality. A 500 mL sample from each sub-plot was stored at -20C until analysis using an American Optical 10480 Mark II Digital Refractometer for BRIX analysis and a

Fisher Scientific Accument AB15 Basic Meter for pH analysis.6,7 Potatoes were mechanically harvested from 6.1 m sections of two rows in each non-weedy sub-plot on

October 11, 2010. Potato tubers were graded and classified as U.S. No.1 with diameter of greater than 4.8 cm, U.S. No. 2 with diameter of greater than 3.8 cm and cull tubers

(USDA 2011). The number and fresh weight of tubers in each category was measured.

157

Statistical Analysis. Data for each vegetable crop were handled separately and analyzed in SAS v.9.2.8 The effect of transition strategy, compost application and the interaction of transition strategy and compost application on soil pH, P, K, Ca, Mg, and percent OM were tested using PROC MIXED with data from 2007 used as a covariate. Fixed effects were replication, transition strategy, and compost application; random effects were replication by transition strategy interactions. When the interaction of transition strategy and compost application was significant (P < 0.05), data were analyzed separately for each transition strategy x compost application. Means were adjusted for the covariate using the LSMEANS statement in PROC MIXED and means separated using the PDIFF option.

The effect of transition strategy, compost application, and the interaction of transition strategy and compost application on weed density and biomass and tomato and potato yield were tested using PROC MIXED. Fixed effects were replication, transition strategy, and compost application; random effects were replication by transition strategy interactions. When the interaction of transition strategy and compost application was significant (P < 0.05), data were analyzed separately for each transition strategy x compost application. Means and mean separation were determined using the LSMEANS statement and PDIFF option. Canonical correlation analysis was performed to assess the association of soil properties with tomato and potato yield, respectively, using PROC

CANCORR.

158

Results and Discussion

Weed Suppression. Data were analyzed separately by vegetable crop due to differences in dates of planting and data collection. Transition strategies affected density of Canada thistle, monocot, and broadleaf weeds in both potato and tomato crop production (Table

6.1). In potato, the density of Canada thistle, monocot, and broadleaf weeds was 74% lower in the prairie than the fallow or non-treated strategies (Table 6.2). Similarly, the density of monocot and broadleaf weeds was 51% lower or more in the prairie transition strategy compared to the other treatments (Table 6.2). Restoration of prairie species has been shown to reduce weed invasions previously (Blumenthal et al. 2005). The prairie transition strategy included a highly diverse mix of perennial species that occupy multiple ecological niches and give year-round ground cover. The use of species of multiple functional groups may utilize resources in a complementary manner that facilitates competition with weed species (Liebman and Staver 2001). Ground cover provided by prairie species may prevent growth and emergence of Canada thistle shoots in early spring. Without shoots to produce photosynthetic product, underground storage of root carbohydrates is affected, leading to reduced shoot production (Gustavsson 1997). In tomato production, Canada thistle shoot emergence was reduced 85% in the smother crop transition strategy compared to the fallow or non-treated (Table 6.2). Smother crops in

2008 and 2009 were planted when Canada thistle root carbohydrate reserves were at a seasonal minimum (McAllister and Haderlie 1985). Canada thistle shoots emerged after

159 smother crop planting may not have been able to effectively compete for light and produce carbohydrates for storage, thus affecting the ability of roots to produce shoots.

Weed biomass in potato was affected by the interaction of transition strategy and compost application (Table 6.1). In tomato, compost application affected weed biomass

(Table 6.1). The application of plant nutrients through compost likely encouraged growth of weeds. Weed biomass in tomato was not affected by transition strategy (Table 6.1).

The weed biomass was harvested after three months of tomato production and the effect of the crop may have reduced any differences between transition strategies. In potato, weed biomass was 29% greater or more in the fallow + compost treatment and 30% greater or more in the non-treated + no compost treatment compared to other transition strategies + compost application treatments (Table 6.3). In the fallow + compost treatment, plants were not allowed to grow and develop and utilize soil nutrients. The soil nutrients may have accumulated during the transition years, providing greater benefit to weeds in the first organic year. The minimal weed management executed during the transition years may not have effectively controlled weeds in the long-term after organic transition.

The fallow, non-treated, and vegetable transition strategies were less effective at suppressing weeds than the prairie or smother crops in both tomato and potato crops

(Table 6.2; Table 6.3). Although landscape cloth suppressed weeds in the lettuce-Brassica crops in 2008, poorly timed rainfall in 2007 and 2009 made timely cultivation in edamame soybeans impossible, resulting in weed seed return and survival of Canada thistle roots. The density of Canada thistle shoots was highest in the non-treated strategy

160 compared to other treatments in both tomato and potato crops (Table 6.2). Canada thistle was likely able to assimilate and store carbon in underground roots without mechanical management or competition from competitive smother crops or prairie species. The carbohydrate storage may allow Canada thistle to produce shoots in subsequent years of organic production, but little is known about the longevity of Canada thistle roots

(Donald 1994). In both tomato and potato, the fallow strategy was not suppressive of monocot or broadleaf weed density in comparison to other treatments (Table 6.2).

Although seed set was prevented in the treatment and the mechanical management would encourage depletion of the seed bank, the strategy did not appear to have long-term effects for weed management once crop production was initiated.

Soil Fertility. Soil pH and available P were affected by the interaction of transition strategy and compost application (Table 6.4). Available K, Mg, and percent OM were affected by transition strategy (Table 6.4). Compost application affected available Ca,

Mg, and percent OM (Table 6.4). The addition of soil nutrients and organic matter in the compost would be expected to increase soil fertility during the transition years without depletion by vegetation or soil disturbance. Soil OM was generally lowest in the vegetable transition strategy (Table 6.6). Harvesting aboveground plant biomass removed a source of organic C for soil and annual root systems may contribute less to soil organic carbon pools in comparison to perennial root systems (Kögel-Knabner 2002).

Additionally, soil disturbance through tillage and cultivation in vegetable crops may decrease the rate of soil organic matter build-up (Jackson et al. 2004).

161

Tomato and Potato Yield. Tomato and potato yield were both affected by compost application (Figure 6.1; Figure 6.2). Tomato fruit quality as measured by pH and BRIX were not affected by transition strategy or compost application (data not shown). The number of marketable tomato fruits was 17% higher and the weight of marketable tomato fruits was 21% higher with compost application compared to the no compost treatment

(Figure 6.1). Compost application increased the number of marketable potato tubers by

79% and the weight of marketable tubers by 50% (Figure 6.2). The application of compost increased yield through the supply of plant available and potentially mineralizable nutrients. Transition strategy did not affect the marketable yield of tomato or potato (data not shown). Since cultivation and hoeing were used for weed control during the first organic year in the weed-free sub-plots, the effects of transition strategy on weeds may not be observed in vegetable yield data (Table 6.1). However, the effect of transition strategy on number (P = 0.0006) and weight (P = 0.0005) of cull potato tubers was significant. There were 73% more cull potato tubers and the weight of cull tubers was 90% greater in the non-treated strategy compared to the smother crop or vegetable strategy (Figure 6.3). The non-treated and prairie transition strategies had the least amount of soil disturbance over the three years of transition compared to the other treatments. Soil physical properties as affected by tillage can influence potato yield

(Pierce and Burpee 1995). Soil disturbance in the fallow, smother crops, and vegetable strategies may have affected the quantity of cull potato tubers.

162

Canonical correlation analysis was used to assess the relationship between marketable yield of tomato or potato and soil properties. One of the canonical correlations in potato was significant (P = 0.007) whereas two of the canonical correlations in tomato were significant (P = 0.009; P = 0.02) (Table 6.7). The first canonical correlation between potato and soil explained 83% of the cumulative variance

(Table 6.7). The first and second canonical correlations between tomato and soil explained 51 and 49% of the cumulative variance, respectively (Table 6.7). Canonical correlation analysis was not used to describe associations between yield and weed variables since marketable yield data was collected from weed-free plots.

The first canonical variate of soil properties in the potato crop had positive standardized coefficients for available K, Ca, and percent OM and negative standardized coefficients for pH, exchangeable P, and available Mg (Table 6.8). The first potato yield canonical variate had positive standardized coefficients for number of U.S. No. 1 grade tubers and weight of U.S. No. 2 grade tubers and negative standardized coefficients for weight of U.S. No.1 grade tubers and number of U.S. No.2 grade tubers (Table 6.8).

These results suggest that plant available nutrients K and Ca and P and Mg had strong, yet opposing, influences on the yield and grade of potato tubers. Previous research has demonstrated that addition of P, K, Ca, and Mg through soil amendments can affect potato tuber quality, size, and yield (Porter et al. 1999). Although compost additions in this study affected pH and percent OM, they contributed weakly to effects on potato yield

(Table 6.8).

163

In the first canonical variate of soil properties in tomato production, available Ca and percent OM had positive standardized coefficients and pH, available P, K, and Mg had negative standardized coefficients (Table 6.8). The weight of marketable tomato fruits in the first canonical variate had a positive standardized coefficient and fruit number had a negative standardized coefficient (Table 6.8). In the second canonical variate, only available Ca and Mg had negative standardized coefficients where the sign of the other variables remained unchanged (Table 6.8). Percent OM strongly influenced the yield of tomato in the first and second canonical variates. The accumulation of organic matter in the soil can be affected by compost application during transition to organic agriculture which can subsequently affect tomato yields (Briar et al. 2011).

Conclusions

The transition strategy implemented during the three years prior to organic farming can affect the weed density and biomass in the first year of organic production.

The use of native, perennial prairie species can provide suppression of Canada thistle, monocot, and broadleaf weeds. Prairie species also represent a potential source of biofuel feedstocks for burgeoning ligno-cellulosic fuel markets that may provide economic return during transition (Jefferson et al. 2004; Tilman et al. 2006). Smother crops are also effective at reducing populations of Canada thistle and weed biomass.

164

Compost applications and transition strategies during organic transition affected soil chemical properties. However, during the first year of organic production, compost application strongly affected potato and tomato yield and quality. In potato production, plant available nutrients had the strongest influence on yield whereas SOM had the greatest impact on tomato yield. Transition strategies that aid in building OM through perennial root systems and minimal soil disturbance may be most effective before initiating organic tomato production.

Before conversion to organic agriculture, consideration of first-year production may allow farmers to choose transition strategies that impact soil quality and weed populations to favor the choice of organic crop species. Further research may consider the effect of novel cropping strategies on a wider variety of organic crops and other measures of soil quality such as microbial communities and physical properties.

Sources of Materials

1 K.W. Zellers & Son, Inc., 13494 Duquette Avenue NE, Hartville, OH 44632.

2 Johnny‟s Selected Seeds, 955 Benton Avenue, Winslow, ME 04901.

3 Sprint Royal oats and Packer peas. La Crosse Forage and Turf Seed Crop., 2541

Commerce St., La Crosse, WI 54603.

4 James VanLeeuwen, 27666 Peoria Road, Halsey, OR 97328.

5 Production Plus Quality Seed, 800 E. 6th, Plainview, TX 79073.

165

6 Reichert, Inc., 3362 Walden Avenue, Depew, NY 14043.

7 Fisher Scientific, 200 Park Lane Drive, Pittsburgh, PA 15275.

8 SAS 9.2 Statistical Software, Statistical Analysis Systems, SAS Institute, Inc., 100 SAS

Campus Drive, Cary NC 27513-2414.

Acknowledgements

Field assistance was provided by Lynn Ault and K. Gregory Smith. Critical knowledge of tef management was provided by James VanLeeuwen. Knowledge of vegetable production was provided by Sonia Walker. Management of the transition strategies and first-year organic production was provided by Jennifer Moyseenko and

Catherine P. Herms. Thank you to all OARDC visiting scholars and undergraduate students who assisted in data collection and field work. Salaries and research support were provided by State and Federal Funds appropriated to the Ohio Agriculture Research and Development Center, The Ohio State University. Manuscript No. HCS-09-00

166

Literature Cited

Blumenthal, D.M., N.R. Jordan, and E.L. Svenson. 2005. Effects of prairie restoration on weed invasions. Agr. Ecosyst. Environ. 107:221-230.

Briar, S.S., S.A. Miller, D. Stinner, M.D. Kleinhenz, and P.S. Grewal. 2011. Effects of organic transition strategies for peri-urban vegetable production on soil properties, nematode community, and tomato yield. Appl. Soil Ecol. 47:84-91.

Corbin, A.T., K.D. Thelen, G.P. Robertson, and R.H. Leep. 2010. Influence of cropping systems on soil aggregate and weed seedbank dynamics during the organic transition period. Agron. J. 102:1632-1640.

Creamer, N.G. and K.R. Baldwin. An evaluation of summer cover crops for use in vegetable production systems in North Carolina. HortScience 35:600-603.

Delate, K., H. Friedrich, and V. Lawson. 2003. Organic pepper production systems using compost and cover crops. Biol. Agric. Hortic. 21:131-150.

Donald, W.W. 1994. The biology of Canada thistle (Cirsium arvense). Rev. Weed Sci. 6:77-101.

Dorais, M. 2007. Organic production of vegetables: state of the art and challenges. Can. J. Plant Sci. 87:1055-1066.

Gutastavsson, A.M.D. 1997. Growth and regenerative capacity of plants of Cirsium arvense. Weed Res. 37: 229-236.

Hanson, J., R. Dismukes, W. Chambers, C. Greene, and A. Kremen. 2004. Risk and risk management in organic agriculture: views of organic farmers. Renew. Agr. Food Syst. 19:218-227.

Jackson, L.E., I. Ramirez, R. Yokota, S.A. Fennimore, S.T. Koike, D.M. Henderson, W.E. Chaney, F.J. Calderón, and K. Klonsky. 2004. On-farm assessment of organic matter and tillage management on vegetable yield, soil, weeds, pests, and economics in California. Agr. Ecosyst. Environ. 103:443-463.

167

Jefferson, P.G., W.P. McCaughey, K. May, J. Woosaree, and L. McFarlane. 2004. Potential utilization of native prairie grasses from western Canada as ethanol feedstock. Can. J. Plant Sci. 84:1067-1075.

Kögel-Knabner, I. 2002. The macromolecular organic composition of plant and microbial residues as inputs to soil organic matter. Soil Biol. Biochem. 34:139-162.

Martini, E.A., J.S. Buyer, D.C. Bryant, T.K. Hartz, and R.F. Denison. 2004. Yield increases during the organic transition: improving soil quality or increasing experience? Field Crop Res. 86:255-266.

McAllister, R. S. and L. C. Haderlie. 1985b. Seasonal variations in Canada thistle (Cirsium arvense) root bud growth and root carbohydrate reserves. Weed Sci. 33:44-49.

Nelson, W.A., B.A. Kahn, and B.W. Roberts. Screening cover crops for use in conservation tillage systems for vegetables following spring plowing. HortScience 26:860-862.

Pierce, F.J. and C.G. Burpee. 1995. Zone tillage effects on soil properties and yield and quality of potatoes (Solanum tuberosum L.). Soil Till. Res. 35:135-146.

Porter, G.A., G.B. Opena, W.B. Bradbury, J.C. McBurnie, and J.A. Sisson. 1999. Soil management and supplemental irrigation effects on potato: I. soil properties, tuber yield, and quality. Agron. J. 91:416-425.

Purakayastha, T.J., D.R. Huggins, and J.L. Smith. 2008. Carbon sequestration in native prairie, perennial grass, no-till, and cultivated Palouse silt loam. Soil Sci. Soc. Am. J. 72:534-540.

Russo, V.M. and M. Taylor. 2006. Soil amendments in transition to organic vegetable production with comparison to conventional methods: yields and economics. HortScience 41:1576-1583.

168

Smith, R.G., M.E. Barbercheck, D.A. Mortensen, J. Hyde, and A.G. Hulting. 2011. Yield and net returns during the transition to organic feed grain production. Agron. J. 103:51-59.

Smukler, S.M., L.E. Jackson, L. Murphree, R. Yokota, S.T. Koike, and R.F. Smith. 2008. Transition to large-scale organic vegetable production in the Salinas Valley, California. Agr. Ecosyst. Environ. 126:168-188.

Tilman, D., J. Hill, and C. Lehman. 2006. Carbon-negative biofuels from low-input high- diversity grassland biomass. Science 314:1598-1600.

Turner, R. J., G. Davies, H. Moore, A. C. Grundy, and A. Mead. 2007. Organic weed management: a review of the current UK farmer perspective. Crop Prot. 26:377- 382.

United States Department of Agriculture. 1983. United States standards for grades of tomatoes for processing. United States Department of Agriculture, Agricultural Marketing Service, Fruit and Vegetable Programs, Fresh Products Branch.

United States Department of Agriculture. 2010. Organic production. Available at http://www.ers.usda.gov/Data/Organic/. Accessed 16 August 2011.

United States Department of Agriculture. 2011. United States standards for grades of potatoes. United States Department of Agriculture, Agricultural Marketing Service, Fruit and Vegetable Programs, Fresh Products Branch.

169

Table 6.1. Analysis of variance results of weed density and biomass in organic transition strategies and compost applications in

potato and tomato crops.

Probability values

Potato Tomato

Density Biomass Density Biomass

Effect CIRAR Mono BL Total CIRAR Mono BL Total

TS 0.03 0.0001 0.008 0.33 0.04 0.001 0.0001 0.37

17 Compost 0.08 0.52 0.06 0.003 0.97 0.22 0.92 0.0001

0 TS x compost 0.06 0.82 0.64 0.01 0.93 0.14 0.96 0.38

Abbreviations: TS, transition strategy; Mono, monocots; BL, broadleaves.

170

Table 6.2. Density of CIRAR, monocot and broadleaf weeds in organic transition strategies and compost applications in potato

and tomato crops and total weed biomass in tomato crop at harvest.

Potatoa Tomatoa

Weed density Weed density Biomass

Transition strategy CIRAR Mono BL CIRAR Mono BL Total

weeds m-2 weeds m-2 g m-2

Fallow 38 ab 570 b 350 a 28 a 880 a 640 a 230

Non-treated 50 a 410 b 76 b 29 a 360 b 74 b 230

1 Prairie 10 c 200 c 31 b 16 ab 190 b 45 b 200

Smother crops 17 bc 470 b 76 b 4.3 b 980 a 64 b 210

Vegetables 22 bc 880 a 350 a 18 ab 980 a 150 b 210

Compost application

Compost 33 480 230 19 720 190 240 y

No compost 22 530 130 19 630 200 190 z

Abbreviations: Mono, monocots; BL, broadleaves. 171

a Means within a column and main effect followed by the same letter do not differ (P < 0.05).

172

Table 6.3. Total biomass of weeds at harvest of potato in organic transition strategies and compost applications.

Transition strategy Compost Weed biomassa

g m-2

Fallow Yes 420 a

Non-treated Yes 300 b

Prairie Yes 330 ab

Smother crops Yes 280 b

Vegetables Yes 310 b

Fallow No 230 y

Non-treated No 370 z

Prairie No 230 y

Smother crops No 200 y

Vegetables No 260 y a Means within a compost treatment followed by the same letter do not differ (P < 0.05).

173

Table 6.4. Analysis of covariance results of soil nutrients, OM, and pH after four years of organic transition strategies and compost application in 2010.

Probability values

Effect pH Soil nutrients % OM

P K Ca Mg

Transition strategy 0.002 0.10 0.004 0.39 0.003 0.0007

Compost 0.02 0.0001 0.0001 0.0001 0.0001 0.0001

Transition strategy x compost 0.03 0.05 0.19 0.84 0.21 0.54

Covariate

2007 0.0001 0.005 0.34 0.01 0.008 0.007

174

Table 6.5. Soil pH and Bray-1 exchangeable P as affected by four years of organic transition strategies within a compost application in 2010.

Transition strategy Compost pH P

mg kg soil-1

Fallow Yes 7.1 55

Non-treated Yes 6.8 40

Prairie Yes 7.0 42

Smother crops Yes 7.1 47

Vegetables Yes 7.2 43

S.E. 0.062 2.8

Fallow No 6.9 26

Non-treated No 6.8 26

Prairie No 7.0 28

Smother crops No 7.1 25

Vegetables No 6.9 24

S.E. 0.062 2.8

175

Table 6.6. Exchangeable K, Ca, and Mg, and OM as affected by four years of organic transition strategies and compost application in 2010.

Soil nutrients OM

Transition strategy K Ca Mg

mg kg soil-1 %

Fallow 280 1300 180 2.8

Non-treated 250 1300 190 3.1

Prairie 260 1300 200 3.2

Smother crops 260 1400 190 2.9

Vegetables 190 1300 180 2.7

S.E. 14 29 3.3 0.068

Compost application

Compost 360 1400 200 3.4

No compost 140 1200 170 2.5

S.E. 8.9 20 2.1 0.043

176

) Compost No compost 5 A

10 A

x 10x

1 - 8 A 6 B

2 Number Number fruit (Noof fruit ha 0 Marketable Total Tomato fruit yield

50 45 Compost No compost A 40

) 1 - 35 A

30 B 25 20

15 Fruit yield yield (Mg Fruit ha 10 5 0 Marketable Total Tomato fruit yield

Figure 6.1. The total and marketable number and yield of tomato fruits as affected by compost treatment with plant density of 2.5x104 plants ha-1. Bars within a group with the same letter do not differ (P < 0.05).

177

160

) A 3 Compost No compost

140

x x 10

1 - 120 B 100 A 80 60 B 40 20 0

Number tubers of Number tubers (No ha U.S. No. 1 U.S. No. 2 U.S. potato grade

12 A Compost No compost

)

1 - 8

6 B A

4 B Tuber yield (Mg ha 2

0 U.S. No. 1 U.S. No. 2 U.S. potato grade

Figure 6.2. The number and yield of U.S. No. 1 and No. 2 potato tubers as affected by compost treatment with plant density of 1.4x103 plants ha-1. Bars within a group with the same letter do not differ (P < 0.05).

178

) 3

80 A 7

x x 10

70 6 )

A 1 - 60 AB 5 50 B AB 4 40 BC 3 30 C C 2 20 C C 10 1 of cull Weight tubers (Mg ha

Number Number of cull tubers (No. tubers ha 0 0 FA NT PR SC VG FA NT PR SC VG

Cull count Cull weight

Figure 6.3. The number and weight of cull potato tubers as affected by organic transition strategy with plant density of 1.4x103 plants ha-1. Bars within a group (cull count; cull weight) with the same letter do not differ (P < 0.05). Abbreviations: FA, fallow; NT, non-treated; PR, prairie; SC, smother crops; VG, vegetables.

179

Table 6.7. Canonical correlation analysis for each set of canonical variables describing marketable yield of potato or tomato and soil chemistry in 2010.

Crop Canonical correlation Cumulative variance P < F

CC1 CC2 CC1 CC2 CC1 CC2

Potato 0.79 0.41 0.83 0.93 0.007 0.75

Tomato 0.51 0.51 0.51 1.0 0.009 0.02

180

Table 6.8. Standardized canonical coefficients for original variables of marketable potato yield and soil chemistry for significant canonical correlations (P < 0.05) described by canonical variates.

Canonical variate

Soil Potato Tomato

Soil 1 Soil 1 Soil 2 pH -0.32 -0.52 0.14

P -1.2 -0.52 0.27

K 2.0 -0.23 0.34

Ca 1.4 1.5 -0.28

Mg -1.8 -1.2 -0.44

%OM 0.40 0.90 0.98

Yield Yield 1 Yield 1 Yield 2

No. U.S. No.1 1.3 -3.1a -1.4a

Weight U.S. No. 1 -0.59 2.6b 2.3b

No. U.S. No. 2 -0.16 - -

Weight U.S. No. 2 0.55 - - a Number of marketable tomato fruits b Weight of marketable tomato fruits

181

Conclusions

The research presented in this dissertation supports the hypothesis that smother cropping is a viable non-chemical weed management strategy appropriate for organic production. A literature review of previous smother cropping research demonstrates that the success of smother crops for weed management depends on the ability of a smother crop to compete effectively with a target weed while not reducing the growth and development of other crop species through occupation of different ecological niches.

Also, soil and crop management can impact the effectiveness of smother crops to suppress weeds. The choice of smother crop species can affect its ability to smother target weeds. This literature review is limited by what data were collected in previous publications and by what smother cropping research has been conducted. Evaluation of tef as a smother crop showed that it is effective in suppressing annual weeds, but is weakly competitive with Canada thistle, depending on variety and depth of Canada thistle roots. Depending on the weed population present at onset of organic transition, seeding tef is a possible strategy for reducing weed pressure. The mechanisms of how tef suppresses annual weeds were not fully elucidated with the data presented.

Planting multi-species crop mixtures during transition to organic agriculture has been shown to be a feasible strategy to suppress the cover of weeds. However, biomass production is more affected by the identity of grass species in mixture than the number of 182 species present. Using multi-species mixtures to suppress Canada thistle at seasonally affected growth stages may also be an effective strategy of weed management. Planting warm-season crops when Canada thistle root carbohydrate reserves are depleted reduced

Canada thistle population more than cool-season or rapid-growth smother crop mixtures.

Measurements of Canada thistle root biology and light attenuation in the crop canopy may be helpful in further explaining the suppressive effect of smother crop mixtures.

Smother cropping is not the only effective strategy for weed suppression during transition to organic agriculture. The use of high-diversity perennial prairie species can suppress Canada thistle, monocot and broadleaf weed populations while building soil organic matter during organic conversion. Compost applications during transition can influence vegetable yields in the first certified organic year, but transition strategy does not affect marketable yields. Plant available nutrients in the soil can influence organic potato yields in the first year of organic production while organic matter may affect tomato yields. Economic return during the organic transition years may be an important motivator for farmers wanting to convert to organic practices, but were not addressed in this study.

Development of transition strategies and smother crops for weed suppression needs further research. Evaluation of more species and their interactions with other problematic weeds may demonstrate further possible crops for weed suppression.

Assessing the competitive ability of smother crops through investigating light attenuation, occupation of space, and below-ground competition for water and nutrients may reveal more information about the mechanisms underlying the ability of crops to

183 suppress weeds. Temporal competition as a mechanism for weed suppression can be further explored through evaluation of different weed species or crops in mixture that do not develop or utilize resources at the same time. Further attention to the effect of cropping strategies on soil properties is merited as the relationship between plant effects on soil and subsequent crops in organic production is not fully understood.

184