COMPETENCE CENTERRESISTANCEWEED

Integrated Weed Management

Your guide to the way forward in weed control and yield protection

How to fight weeds and maximize yields in rice systems

Contents

1.0 INTRODUCTION 4 1.1 What is Integrated Weed Management (IWM)? 8 1.2 Bayer commitment to IWM 8 1.3 Best Weed Management Practices – Diversity is the Future 9

2.0 WEED CONTROL BASICS 10 2.1 What is a weed? 10 2.2 Why do we need to control weeds? 11 2.3 How can weeds be controlled? 12 2.4 Weed identification – why it is important to know your weeds 14 2.5 Weed biology and ecology 16 2.5.1 Germination characteristics and period 16 2.5.2 Growth habit & optimum conditions 16 2.5.3 Growth rate 17 2.5.4 Competitiveness with crops 17 2.5.5 Weed-free period for a crop 17 2.5.6 Weed size for optimal control 18 2.5.7 Seed production and dispersal 18 2.5.8 Vegetative propagation 18 2.6 How herbicides work 19 2.6.1 Basic biochemical mechanisms 19 2.6.2 Why is a mode of action important? 20

3.0 SYSTEMS RICE CULTURE AND WEED CONTROL 21 3.1 Transplanted rice 22 3.2 Direct-seeded rice 22 3.3 Lowland rice 22 3.4 Upland rice 22 3.5 Main impediment for adoption of direct-seeded rice 22 4.0 THE DEVELOPMENT OF HERBICIDE-RESISTANT WEEDS 23 4.1 Definitions of resistance 23 4.2 Effective dose 23 4.3 Global resistance trends 24 4.3.1 Global species count 24 4.3.2 Resistant species by mode of action 25 4.4 Resistance evolution – a Darwinian process 25 4.4.1 Basic principles 25 4.4.2 Resistance evolution dynamics 25 4.4.3 Resistance and fitness penalties 26

2

4.5 Resistance selection pressure and risk assessment 27 4.5.1 Use of below-labeled rates 27 4.5.2 Help from non-chemical supplemental measures (soil tillage, cover crop, seeding density, competitive varieties, etc.) 28 4.5.3 Lack of weed control innovations 28 4.6 Resistance mechanisms 29 4.7 Multiple resistance and cross resistance 31 4.8 Resistance dynamics - spread vs. independent evolution 32

5.0 RESISTANCE CONFIRMATION TESTING AND DIAGNOSIS 33 5.1 Validation testing 33 5.2 Types of testing 34 5.2.1 Greenhouse tests 34 5.2.2 “Quick” tests 35 5.2.3 Field tests 35 5.3 Interpreting results and making recommendations 35

6.0 INTEGRATED WEED MANAGEMENT (IWM) 36 6.1 Bayer IWM Core Guidelines 37 6.1.1 Know the weed spectrum 38 6.1.2 Develop a weed management strategy 38 6.1.3 Demonstrate in practice IWM techniques 40 6.2 The need for proactive weed control within a field and at community level 40 6.3 Other non-chemical measures 41 6.4 How to measure success 41 6.5 Other sources of information 41

7.0 THE BAYER WEED RESISTANCE COMPETENCE CENTER (WRCC) 42 7.1 Introduction 42 7.2 Mission & objectives 43 7.3 Key operational tasks 43

8.0 REFERENCES 44

3

China

Japan

Korea, South

Rest of East Asia

Indonesia

Vietnam

Thailand

Burma

Philippines

Cambodia

Rest of Southeast Asia/Pacific

India

Bangladesh

Pakistan

Rest of South, West,Central Asia/Russia

Egypt

Rest of Africa

Brazil

United States

Rest of Latin America

European Union/Eatern Europe

Others

■

■

■

■

■

■

■

■

■

■

■

■

■

■

■

■

■

■

■

■

■

■

6,6

9,1 4,6

13,57,8

7,18,8

2,13,8

144,9

%#&

4,22,736,6

27,8

18,612,5

11,5

4,7

4,4

106,5

34,6

7,8Fi

gur

e 1

Estimated global milled rice production in million metric tons for 2017.

Since the first cultivation of rice, weeds have always been a major impediment in its production. The need to control weeds has partly led to the evolution of production systems using transplanted rice in paddy (or puddled) fields (De Datta, 1981). Flooding fields helps to control and suppress weeds and gives rice seedlings a head start that helps them compete better. Other benefits are greater ease of transplanting into loosened soil, less effort required for tillage operations, improved soil fertility and nutrient availability through anaerobic soil conditions, and less water loss through reduced percolation (De Datta et al., 1978). However, this system is highly reliant on a reliable source of water and manual labor, partly for transplanting rice seedlings for establishment and partly to control weeds. With the increased difficulty of finding such labor in rural areas and the pressure from population growth on agriculture’s use of water, it is increasingly being replaced by direct-seeding production systems in which labor can be replaced by mechanized operations, but they are reliant on the proper use of herbicides to control weeds (Ismail et al., 2012; Joshi et al., 2013; Mahajan et al., 2014). Another factor in the future will be climate change and its effect on natural precipitation (von Braun and Sol, 2005).

The title of this brochure “How to fight weeds and maximize yields in rice systems” reflects the need to control weeds in rice crops and how important it is in order to protect its yield potential. Rice is the single largest global staple food grain consumed directly by humans (USDA, 2017) and is grown in a variety of production systems around the world (Kraehmer et al., 2015). China produces the most milled rice in the world. They are followed closely by India, Indonesia, Bangladesh, and Vietnam (Figure 1), which account for 73% of global production. The region (UN definition) with the largest production of milled rice (Figure 1) is East Asia (159.6 Mmt – million metric tons), followed by South, West and Central Asia (156.8 Mmt), and Southeast Asia (116.1 Mmt), together accounting for 90% of the global total (USDA, 2017). Production shares in Africa, the Americas, and the European Union as well as others are relatively low, making up the remaining 10%; however rice’s share of total calories in these regions is growing, particularly in Sub-Saharan Africa, while in Asia it is declining (von Braun and Sol, 2005). The importance of rice in this region is beyond just being considered a staple food. In Asia, “rice is not just a cereal; it is the root of civilization” (Gomez, 2001). In some cultures it has even achieved mystic significance (De Datta, 1981).

1.0 Introduction

Data Source: USDA, 2017

4

Fig

ure

2Fi

gur

e 3

Number of reported unique resistance cases in rice for single MoAs and cross-resistant populations.

Global area grown to rice and average yields 1960/61 – 2016/17.

0

20

40

60

80

100

120

140

Num

ber

of

Rep

ort

ed U

niq

ue C

ases

1980 1985 1990 1995 2000 2005 2010 2015 2020

Cross Resistance

Single Resistance

1,5

2,0

2,5

3,0

3,5

4,0

4,5

60

80

100

120

140

160

180

1960 1980 20001970 1990 2010 2020

Yie

ld (t

/ha)

Are

a (m

illio

n ha

)

Area (m ha)

Yield (t/ha)

Data Source: Heap, 2017

Redrawn from USDA, 2017

The use of properly applied herbicides at labelled rates has become the most consistent, effective, and economical method to control weeds in not just rice, but in many cropping systems. However, problems resulting from the resistance of weeds to many herbicides in various crops, including rice, are increasing globally. The number of unique cases of resistance to a single mode of action (MoA) first reported in rice is rising steadily (Figure 2). To date, there are 51 weeds that have developed resistance to herbicides in rice in 29 countries (Table 1). Leading are Japan (20), the USA (15), and South Korea (14). Conspicuously absent from the list is India, although it is known that some rice weeds there are herbicide-resistant, none have been reported to the global database (Heap, 2017). Although increasing at a slower rate, the cases of cross resistance to multiple MoAs are also rising steadily (Figure 2). Species within the Echinochloa genus seem to be the most adept at developing resistances, with a total of seven species. E. crus-galli var. crus-galli leads the group with 14 unique single- or cross-resistances (and nine total MoAs), followed by E. colona with nine and E. phyllopogon (=E. oryzicola) with five (Table 2). One population of E. galli var. crus-galli has developed resistance to four MoAs, and some populations of E. colona and Ischaemum rugosum are resistant to three MoAs. It is interesting to note that of the 51 resistant species, 41 are reported to be resistant to Group B (ALS inhibitors). This does not mean that this MoA is completely lost for weed control in rice. In fact, ALS-inhibiting herbicides are still working well in many rice fields. However, it does suggest that in order to maintain the sustainable use of this MoA, it is very important to give it some help. That’s where IWM comes in.

The twelve most troublesome weeds in rice in Asia are: Cyperus difformis, Cyperus iria, Echinochloa colona, Echinochloa crus-galli, Eclipta prostrata, Fimbristylis miliacea, Ischaemum rugosum, Leptochloa chinensis, Ludwigia hyssopifolia, Oryza sativa (weedy rice), Schoenoplectus juncoides, and Sphenoclea zeylanica, and include four sedges, five grass, and three broadleaf weeds (IRRI, 2017). Most weeds found in rice are C4 plants, whereas rice is a C3 plant and has difficulty competing with the weeds, especially under sunny and hot conditions (Mahajan et al., 2014). More details about the weeds plaguing rice can be found in Kraehmer et al., 2015.

Help is clearly needed, and not just with resistances to single MoAs. Resistances to more than one MoA or herbicide group greatly complicate the management of weeds and limit herbicide options. It has been recognized for over 50 years that “combinations of cultural and chemical practices are more effective in controlling weeds in rice than either practice used alone” (Smith & Shaw, 1966). Cultural weed control measures help to reduce weed populations and put crops into a better competitive position against weeds; in limited cases they can actually control weeds (i.e., tillage). However, if used exclusively in the absence of any other control measure, weeds can develop resistance, even to the oldest weed control method known to man, hand-weeding. How? By having weeds become indistinguishable from crops in early growth stages through mimicry, as has occurred with

51

Rice 29AS OF APRIL 2017

in

5

Rural and urban population of the world.

Fig

ure

4

certain populations of Echinochloa crus-galli in rice (Barrett, 1983). Historically, the rice paddy has been the ultimate proving ground for developing strategies for Integrated Weed Management (IWM). Today, we need to do even more on the farm, in industry, in distribution, as advisors – to emphasize the benefits of IWM and to promote practices that include the integration of chemical and non-chemical management practices in order to protect yields and help maximize farm income now and in the future.

As the world population has increased from 3 billion people in 1960 by a factor of 2.5 to nearly 7.5 billion today (US Census Bureau, 2017), rice production has kept pace mainly through an increase in yields by a similar factor of 2.4 (Figure 3). During the same period, the area used to grow rice increased by only 30%. This level of efficiency gain must be maintained as the overall population continues to grow, the demand for water becomes greater, and the population in rural areas declines (Figure 4), making it difficult to find manual labor. This trend is especially acute in Asia where most rice is grown and consumed. The changeover to direct seeding production systems will continue and will require further adaptation to local areas. Protecting the continued utility of herbicides will contribute greatly to the goal of feeding a growing global population. This brochure seeks to provide a guide on how to build integrated strategies and practices that lead to sustainable weed control in rice crops and why it is important to do so.

Number of resistant weed species reported in rice.

RuralUrban

0

1

2

3

4

5

6

7

1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

2000

2005

2010

2015

2020

2025

2030

2035

2040

2045

2050

Po

pul

atio

n (b

illio

ns)

Location No. of Resistant Species

Australia 4

Bolivia 2

Brazil 7

Chile 2

China 6

Colombia 2

Costa Rica 2

Egypt 2

El Salvador 1

France 2

Greece 4

Guatemala 1

Honduras 1

Indonesia 1

Italy 6

Japan 20

Malaysia 7

Nicaragua 1

Panama 1

Philippines 2

Portugal 1

South Korea 14

Spain 2

Sri Lanka 1

Thailand 3

Turkey 4

United States 15

Uruguay 1

Venezuela 6

Tab

le 1

Redrawn from UN, 2014

Data Source: Heap, 2017

6

List of resistant weed species reported in rice and resistance mode of action.

Tab

le 2

Data Source: Heap, 2017

Species Location of First Report First Year Resistance Mode of Action (HRAC Group)

Alisma canaliculatum Japan 2007 B

Alisma plantago-aquatica Portugal 1995 B

Ammannia auriculata United States (California) 1997 B

Ammannia coccinea United States (California) 2000 B

Bacopa rotundifolia Malaysia 2000 B

Blyxa aubertii South Korea 2006 B

Conyza canadensis United States (Mississippi) 2003 G

Cyperus difformis United States (California) 1993 B, C2

Cyperus esculentus United States (Arkansas) 2013 B

Cyperus iria United States (Arkansas) 2010 B

Cyperus odoratus Venezuela 2014 B

Damasonium minus Australia (New South Wales) 1994 B

Digitaria ischaemum United States (California) 2002 O

Dopatrium junceum Japan 2007 B

Echinochloa colona Costa Rica 1987 A, AB, ABC2, AG, B, C2, BC2L, O, G

Echinochloa crus-galli var. crus-galli Greece 1986 A, AB, ABL, ABC2L, AN, B, BO, C2, C2K3, C2O, F4, K3, N, O

Echinochloa crus-galli var. formosensis Japan 2009 A, AB

Echinochloa crus-galli var. zelayensis China 2013 O

Echinochloa crus-pavonis Brazil 1999 O

Echinochloa erecta Italy 2004 C2, C2L

Echinochloa oryzoides United States (California) 2000 AB, N

Echinochloa phyllopogon (=E. oryzicola) United States (California) 1998 A, B, AB, AN, N

Elatine triandra var. pedicellata Japan 1997 B

Eleocharis acicularis South Korea 2006 B

Eleusine indica Bolivia 2005 A

Fimbristylis miliacea Malaysia 1989 B, O

Ischaemum rugosum Colombia 2000 A, ABC2

Leptochloa chinensis Thailand 2002 A, C2

Leptochloa panicoides United States (Louisiana) 2009 A

Leptochloa scabra Venezuela 2013 C2

Limnocharis flava Indonesia 1995 BO, O

Limnophila erecta Malaysia 2002 BO

Limnophila sessiliflora Japan 1996 B

Lindernia dubia (=Lindernia dubia var. major) Japan 1995 B

Lindernia micrantha Japan 1996 B

Lindernia procumbens Japan 1996 B

Ludwigia prostrata South Korea 2011 B

Monochoria korsakowii Japan 1994 B

Monochoria vaginalis Japan 1998 B

Oryza sativa var. sylvatica United States (Arkansas) 2002 B

Rotala indica var. uliginosa Japan 1997 B

Rotala pusilla Japan 2007 B

Sagittaria guyanensis Malaysia 2000 B

Sagittaria montevidensis United States (California) 1993 B, BC3

Sagittaria pygmaea South Korea 2004 B

Sagittaria trifolia Japan 1997 B

Schoenoplectus fluviatilis South Korea 2004 B

Schoenoplectus juncoides Japan 1996 B

Schoenoplectus mucronatus (=Scirpus mucronatus) Italy 1995 B, C2

Schoenoplectus wallichii Japan 2000 B

Sphenoclea zeylanica Philippines 1983 O

Data Source: Heap, 2017; this database contains a collection of the first unique resistance cases for a specific weed species to a particular MoA (or unique MoA combination) within a particular geographical unit (state, province, or country).

7

1.2 Bayer commitment to IWM Bayer invests heavily in research to develop new methods and possibilities of combating weeds. The herbicide research activities are concentrated at the research facility in Frankfurt, Germany where Bayer has laboratories, production facilities, and its Weed Resistance Competence Center (WRCC), inaugurated in 2014. As our global reference center, the WRCC is developing and implementing pro-active programs to promote the sustainability of weed control. The growing resistance situation worldwide, along with its increasing complexity, is making weed control more arduous. The implementation of solutions requires a fundamental shift in thinking and acting – by farmers, advisors, the product distribution chain, and us. For more information, please see Section 7.0.

1.1 What is Integrated Weed Management (IWM)?

Integrated Weed Management (IWM) is a fundamental program in the production systems of farmers that enables the sustainable control and management of weeds in fields using methods designed to complement each other. It involves the use of a range of control techniques embracing physical, chemical, and biological methods in an integrated fashion without excessive reliance on any one method. Because of the increase in weed resistance over the past decade, it has been adopted as a tool for managing herbicide-resistant weeds.The purpose of IWM is to reduce weed pressure and keep weeds at low levels. The desired outcome is to put weeds off balance and thus make it easier for an herbicide to do its job — which is to protect the yield potential of a crop. The goals of an IWM plan can be simply stated as below.

1. Suppress weed growth and biomass accumulation to limit their ability to decrease yield2. Minimize weed seed production to limit return of seeds into the soil seed bank 3. Deplete weed seed reserves in the soil to minimize germination the following years4. Prevent or reduce the spread of weeds to keep problems from non-problem areas

The careful use of such methods should result in no negative environmental issues and, in fact, can contribute positive environmental results by helping to reduce soil erosion and increasing soil organic matter levels. It can also pay for itself by delivering economical weed control and reducing future weed problems. It may require doing some things differently than what was done before, but IWM is a dynamic program than can be fine-tuned to almost any agricultural production system.

8

1.3 Best Weed Management Practices – Diversity is the Future Our global initiative is called “Diversity is the Future." Diversity is the key to success in many aspects of crop protection, including diversity in herbicides, diversity in crops, and diversity in supplemental methods designed to disrupt the life cycle of weeds. In this way, Bayer is contributing to sustainable agriculture.

9

Weed control basics begin with understanding what a weed is and how it can affect you, even if you are not a farmer or work in agriculture.

2.0 Weed control basics

2.1 What is a weed?

A weed can be defined in many ways. The Webster-Miriam dictionary (Anon., 2015) defines a weed as “a plant that is not valued where it is growing and is usually of vigorous growth; especially: one that tends to overgrow or choke out more desirable plants.” This certainly refers to several attributes of weeds: not valued where it is growing (e.g., in a field of crops), vigorous growth, and interference with the growth of desirable plants (e.g., crop plants). Those attributes contribute to losses in potential yield through weeds – losses that result in less food, feed, and fiber being harvested and therefore less being made available to nourish and cloth our growing global population.

10

2.2 Why do we need to control weeds?

Weeds can be pretty, as in the picture of this monochoria (Monochoria vaginalis) plant to the left. However, in this situation they are not desired plants when they grow in a rice paddy. Although not known for being very competitive, a heavy infestation can significantly reduce yield. They can reproduce through stolons, plant fragments as well as plantlets developing from seeds, so care must be taken to limit their spread to avoid such infestations. All weeds compete with crops for water, nutrients, space, and – if allowed to grow taller than the crop canopy – can block out light (Figure 5). This competition can add up to significant yield losses over the course of a growing season. In some cases, weeds interfere with the harvest and cause additional yield loss. And if weeds are allowed to grow unchecked, they can reproduce and recharge the soil seed bank, causing protracted problems.

Figure 6 shows that yield losses to weeds in rice can vary anywhere from 12% to 90%, with the lowest losses in transplanted rice and the highest in DSR (Majahan et al., 2014). For example, barnyard grass (Echinochloa crus-galli) is a C4 plant that can outgrow rice at higher temperatures, giving it a built-in advantage (Figure 7). Weeds can contaminate the harvest with undesired seeds that result in dockage. In severe cases, a total loss of yield can be the result of allowing weeds to get the upper hand on a crop. The fact is that weeds need to be controlled – every weed, in every field, every year.In addition to affecting yield, weeds can cause many other problems (Zimdahl, 2007). For example, weeds can lead to human health issues such as hay fever that are caused by the allergic reaction to pollen. In addition, poisonous weeds can cause allergenic reactions to exposed skin or even result in severe injury or death if ingested. Dried weeds can also be a significant fire hazard during certain periods of the year. Certain weeds can harbor specific harmful pests (insects, diseases, and nematodes). Aquatic weeds can interfere with water management. And weeds can interfere with vision on on highways or railways and decrease the safety of these modes of transportation. One rarely thinks about the effects of weeds on wildlife, but poisonous invasive species can also be detrimental to game animal populations in addition to being damaging to communities of native flora. If they remain unchecked, weeds can turn into a large ecological problem.

Barnyard grass (Echinochloa crus-galli) can outgrow rice.

Fig

ure

7

WaterNutrients

Space

Ligh

t

Fig

ure

5

Picture source: http://www.bayercropscience.co.uk/your-crop/crop-diseases,-weeds-and-pests/grass-weeds/black-grass/

Effect of weeds on crop plants

Redrawn from Majahan et al., 2014

Potential yield loss caused by weeds in different rice cultivation systems in India

Fig

ure

6

Patna

Wet seeded rice

Dry seeded rice

Transplanted rice

Patnagar

0

10

20

30

40

55

60

70

80

90

100Y

ield

Lo

ss [

%]

11

Weeds can be controlled in broad-acre agriculture by two main means: through mechanical cultivation or with the use of an herbicide. Other agronomic, non-chemical means of weed control can significantly reduce weed populations but rarely result in the level of weed control achieved using the first two methods. Hand-weeding can be used for small-scale areas or as a complement to mechanical cultivation or herbicide usage on larger fields, but is not an economically sustainable method of weed control for larger farms because it is so labor intensive Mahajan et al., 2014.

2.3 How can weeds be controlled?

Pa

st

Source: Bayer

Pre

se

nt

12

Weed control can be divided into preventive weed management, with the intent to decrease the emergence or growth of weeds, and corrective weed management, with the intent to eliminate weeds once they have emerged. Preventative weed management techniques include crop rotation, alternation of spring and autumn crops and agronomic measures such as tillage, seeding time, starting with clean seed and the use of cover crops. Corrective weed management techniques include hand-weeding, mechanical cultivation, post-harvest weed seed control, biological or biotech weed management, and chemical weed control (herbicides). These techniques will be discussed in greater detail in Section 6.0 on Integrated Weed Management.

How do you define the level of weed control needed? Figure 8 shows commonly accepted definitions of weed control and suppression. Minimum acceptable levels of weed control are reached upon the elimination of weeds with more than 80% efficacy. Suppression is defined as the reduction of weed competition with 30-70% efficacy. Efficacy is usually measured by assessing the reduction in above-ground biomass. However, reduction in the below-ground roots and other organs (i.e., rhizomes) and the reduction of viable seeds to return to the soil are also important. Generally speaking, the higher the level of efficacy, the better the yield potential of the crop. In the past, emphasis was put on using economic thresholds to determine the level of intervention in weed growth. However, for some particularly noxious weeds that have a high degree of competitiveness and an ability to produce large numbers of seed, such as Palmer amaranth (Amaranthus palmeri), a “zero-tolerance” policy has

been adopted as a management strategy to prevent it from overrunning fields (Norsworthy et al., 2014a).

Sun Tzu, an ancient Chinese military general, strategist, and philosopher has been credited with underscoring the need to “understand one’s enemy.” The more you understand about him, his capabilities, strengths, weaknesses, and those of yourself, the better able you are able to defeat him. The same goes for weeds. They are year after year the major enemy for crops. The objective is to disrupt the weed’s life cycle and eliminate its threat to obtaining the maximum yield possible in a given field with the given inputs subjected to the conditions over the crop growing season. In order to do so, you must understand the weed, beginning with its identity and its biological characteristics. Some weeds have a much greater ability to compete (e.g., pigweeds, Amaranthus spp., and giant ragweed, Ambrosia trifida) than others (e.g., large crabgrass, Digitaria sanguinalis) and need to be treated differently. The same can be said for the crop. You must understand its ability to compete with the weeds present in the field and how long it takes for the crop canopy to close. Once the crop canopy is closed, the ability of weeds to affect the yield potential is severely reduced (Singh et al., 2014). In short, weeds are highly adaptable to many environments and strive for survival while crops are adapted to specific environments and bred for yield and uniformity.

Weeds are highly adaptable to many environments and strive for survival, while crops are adapted to specific environments and are bred for yield and uniformity.

Weed controlElimination of weed competition withmore than 80% efficacy

SuppressionReduction of weed competition with 30-70% efficacy

Annual weeds• Elimination of roots and shoots• Weed development stopped• No viable seeds

Perennial weeds• Elimination of above-ground plant material • Impact on below-ground organs

• Partial inhibition of growth and development • Reduction in number of viable seeds

Fig

ure

8

Source: Bayer, 2014

Definition of weed control and suppression

13

2.4 Weed identification – why it is important to know your weeds

Know your weeds. Differences in susceptibility to herbicides and growth characteristics by weeds that look similar but are, in fact, different species, can result in selection of improper measures. The following list (adapted from Shrestha, 2015) states several of the reasons.

• Knowing exactly which weeds you have in your field and their biological characteristics helps you choose the correct herbicide and other complementary control measures that will work

• Different weed species can respond to management measures very differently• Some weed species are more competitive than others and need to be treated differently• Seed production and shattering characteristics (return to the soil bank) may differ greatly

between weed species• Weed emergence characteristics can differ greatly between weed species• It is important to know whether a weed has an annual or perennial growth habit to help chose

a management strategy – perennial weeds may require control of plant organs in the soil (e.g. rhizomes, bulbs)

Differences in sensitivity between plant species can be significant, even if the phenotypes (weed appearance) are very similar. Let’s take two Echinochloa species currently causing problems in rice crops, barnyard grass (Echinochloa crus-galli) and jungle rice (Echinochloa colona), as an example. Figure 9 shows photos of seedlings and the inflorescence of barnyard grass and Figure 10 shows the same for jungle rice. In the early stages of growth, it is very difficult to tell them apart. Later the differences in the appearance of the inflorescences and the awnless seeds of jungle rice and awned seeds of barnyard grass enable differentiation (Galinato et al., 1999) and will help to correctly identify the species, but by then it may be too late if the incorrect dosage has been applied. The response of each species to treatments of post-emergent applied herbicides is shown in Figure 11 (Chauhan and Abugho, 2012). Although the activity to both species was similar for the fenoxaprop-p-ethyl + ethoxysulfuron treatment, both bispyribac-sodium and penoxsulam + cyhalofop-butyl treatments were clearly less active to barnyard grass than to jungle rice. This difference in activity could potentially have serious consequences for weed control if you misidentify the weed. Because the Echinochloa complex is so large and posseses a high degree of genetic and phenotypic variability, there could be significant regional differences in relative response to herbicides between the two species.

Fig

ure

9

Barnyard grass (Echinochloa crus-galli)

14

Fig

ure

11

2.4.1 Phenotypic variabilityNot only can different weed species look similar, but different populations of the same species can look different. This ability to have multiple growth forms is a desirable property for a weed, helping it adapt to novel forms of stress (Hantsch et al., 2013). Phenotypic variability in weeds can be extremely large and appears to be a predictor of the invasive ability of weeds to acclimate to new environments (Mal and Lovett-Doust, 2005). Sometimes the variability is so large that not enough is known about all the phenotypic variants and thus genetic tests must be implemented to truly distinguish whether an accession of a weed is merely another phenotypic variant or belongs to a completely different species. We must sometimes conduct these test ourselves at the WRCC (see Section 7.0) to distinguish individual weeds that are in mixed populations of Amaranthus (A. palmeri and A. tuberculatus).

Tabular data from Chauhan and Abugho, 2012 were transformed from % survival to % mortality and are drawn as a figure (bars represent LSD0.05). Herbicides included bispyribac-sodium at 30 g a.i. ha-1; fenoxaprop-p-ethyl + ethoxysulfuron at 10 + 35 g a.i. ha-1, penoxsulam + cyhalofop-butyl at 60 + 12 g a.i. ha-1.

0

10

20

30

40

55

60

70

80

90

100

See

dlin

g m

ort

alit

y [%

]

4-leaf

6-leaf

8-leaf

Bispyribac-sodium

Fenoxaprop-p-ethyl+ ethoxysulfuron

Penoxsulam +cyhalofop-butyl

Bispyribac-sodium

Fenoxaprop-p-ethyl+ ethoxysulfuron

Penoxsulam +cyhalofop-butyl

Echinochloa crus-galli Echinochloa colona

Treatment stage

Jungle rice (Echinochloa colona)

Response of Echinochloa crus-galli and Echinochloa colona to selected herbicide treatments

Fig

ure

10

15

2.4.2 Genetic variability (dioecious vs. monoecious, self-fertilization vs. outcrossing, etc.)High phenotypic variability seems to go hand-in-hand with high genetic variability. This leads to greater chances of a weed population surviving a new threat, such as the application of a new herbicide it has previously not experienced. Only one plant has to survive and produce viable seed, so that the trait or traits conferring resistance are passed on to the next generation. It is important to know the reproductive strategy of a species, whether it is self-fertilized, apomictic, or outcrossing in order to help you define the best control strategy to contain the weed to a particular area. It can also help you to zero in on particular reproductive structures (e.g., seeds, rhizomes, bulbs) to use in order to measure the success of your weed control management program.

2.4.3 Where to get information on weed identificationDue to the regional differences in weed phenotypes, the best place to get information on weed identification is from local sources – local agricultural extension agents, company representatives, agricultural universities, or from some of the increasing number of apps for smartphones and tablets. Some excellent general guides are available as downloads (Galinato et al., 1999; Caton et al., 2010).

2.5 Weed biology and ecology

Once a weed has been correctly identified, it is then important to understand the biological and ecological attributes of that species and how it interacts with the crop as the first step to developing a management plan for that weed (Labrada, 2006). An incomplete understanding of the biology of a weed in a particular area is one of the continuing impediments to its management. Attributes including germination characteristics and period, growth habit and optimum growth conditions, growth rate, competitiveness, flowering and seed dispersal, and seed production are important factors that can give insights on how best to approach managing that weed. The intent of an IWM program is to disrupt the life cycle of the target weed(s) with as many approaches as is economically feasible in order to make the job of the herbicide easier and to decrease the selection pressure for resistance.

The intent of an IWM program is to disrupt the life cycle of the target weed(s) with as many approaches as is economically feasible in order to make the job of the herbicide easier and to decrease the selection pressure for resistance.

2.5.1 Germination characteristics and periodDoes the target weed germinate in the winter, spring, or both? Does it germinate before, with, or after the crop? Does it germinate during the entire growing season? These are important considerations when designing a weed management program. For example, the germination of crowfoot grass (Dactyloctenium aegyptium), one of the main weeds in dry-seeded rice in Asia, was found to be stimulated by light and mulching with rice straw helped to reduce germination (Chauhan, 2011). It was also found that the seeds of D. aegyptium required moisture to germinate optimally, therefore making the weed a good candidate for control with a false seedbed strategy in no-till systems like dry-seeded rice. Employing this strategy, an initial shallow irrigation stimulates germination and emerged seedlings are then controlled with a non-selective herbicide. Additionally, seeds at or near the soil surface emerged at a much higher percentage than when buried at 4 cm (1.5 inches) deeper, suggesting that reduced tillage systems would favor this weed and that periodic seed burial with inversion tillage would also be an effective non-chemical control strategy.

2.5.2 Growth habit & optimum conditionsIt is very helpful to know if the weed is an annual or perennial, whether it grows in the wintertime or the summertime. It is harder to control a winter annual in a winter crop

16

CWFP

0 20 40 60 80 100 120 140 160

100

80

60

40

20

00 20 40 60 80 100 120 140 160

Days after sowing

weed-free

no weed control

CWFP

PR-115PR-114

Gra

in y

ield

(% o

f w

eed

-fre

e)

Fig

ure

12

and likewise a summer annual in a spring or summer crop. Rotations can also affect yield losses from weeds, with the rice-wheat rotation (in India) resulting in the highest losses (Singh et al., 2005). For example, if facing an annual grass in a winter cereal crop, it is easier to combat this in a spring crop, given the opportunity to use a cover crop during the winter to reduce the opportunity for the weed to grow freely or to use a non-selective burndown treatment to significantly reduce weeds prior to planting. Does the weed grow prostrate or vine up the crop structure? Does the weed form rhizomes or other underground reproductive structures the might impede complete control? Is it a C4 plant like Echinochloa crus-galli or Echinochloa colona, which are far more able to handle higher temperatures and thus able to outcompete a C3 crop like rice under these conditions (Galinato et al., 1999)? It helps to understand the biology of the target weeds to find its weaknesses and exploit them and avoid its strengths with other tactics.

2.5.3 Growth rateThe growth rate of weeds can differ greatly between species and thus play a large role particularly when planning post-emergent applications. Timely applications that are made to weeds at optimal growth stages are sometimes difficult to achieve due to the weather. Knowing how much time one might have before weeds grow out of the optimal treatment stage can be valuable in deciding the urgency of completing an application.

2.5.4 Competitiveness with cropsWeeds are naturally competitive with crops for water, nutrients, space and light. Certain weeds are much more competitive than others and need to be managed more closely. Emerged weeds should be removed before the end of Period 1 (maximum weed-infested period) to prevent significant reductions to crop yield. Period 2 (the critical weed-free period as in Figure 12) needs to be maintained weed-free. Weeds emerging later normally do not significantly affect crop yield, with the exception of vining weeds or others that may interfere with harvest operations. After closure of the crop canopy, the crop generally can effectively suppress major weed growth. However, it is important to keep populations from growing at the very end of the season in order to prevent additional contribution to the soil seed bank. This is particularly relevant for resistant populations.

2.5.5 Weed-free period for a cropDefining critical weed-free periods for different crops and cultivars helps to improve the timing of weed control and the choice of more competitive cultivars for particular rice cropping systems and weed, such as DSR systems. For example, the variety ‘PR 115’ had a much shorter critical weed-free period (shown between the dotted vertical lines) than the variety ‘PR 114’ (Figure 12), and thus it was shown to be much more competitive under the weed pressure in the growing system at this site.

Figure redrawn from Singh et al., 2014. Data only shown for 2013. CWFP: critical weed-free period. Gompertz and logistic equations were fit to grain yield (% of weed-free) and the critical weed-free periods to achieve 95% of maximum yield are shown between the dotted vertical lines.

Periods of weed growth and competition in rice for two rice varieties

17

2.5.6 Weed size for optimal controlThe weed size for optimal control by any post-emergent herbicide is generally small, at 15 cm (6 inches) or less. Younger weeds tend to be more susceptible to herbicides than older weeds. In the case of a contact herbicide or one that does not translocate to a great degree, they are generally even less effective on larger weeds because the developed foliage can cover the growing points, which must be coated by the herbicide application in order to achieve maximum efficacy.

2.5.7 Seed production and dispersalProlific seed production capacity appears to be coupled with rice emergence and a high degree of competitiveness by some of the Echinochloa species, i.e., E. crus-galli, which suggest that practices that delay germination of the weed would benefit IWM programs (Bagavathiannan et al., 2012). Seed production has an indirect influence on resistance evolution by changing the mathematics of resistance evolution. Non-chemical measures help tremendously by reducing the numbers of weeds (or their growth and development) and thus reducing the seed production, which for E. crus-galli can approach 1000 kg/ha-1 (approx. 900 pounds per acre (lbm/ac) (Galinato et al., 1999). The higher the seed production, the greater the chance of finding an individual plant with the mutation(s) conferring resistance. It is also helpful to understand how seeds are dispersed in order to control their spread within a field and between fields. The distribution of weeds in a field is generally found to be non-uniform and has been described as “an important source of inefficiency in weed management.” (Cardina et al., 1997). Keeping an eye on the “bad patches” within a field is one way of determining whether a potential problem is emerging and a good way to keep resistance from evolving faster.

2.5.8 Vegetative propagationSedges like Cyperus spp. can be vegetatively propagated through the formation of tubers (Stoller and Sweet, 1987). Rice varieties are often selected for traits that help them establish in flooded fields, including the ability to germinate and elongate faster under hypoxia, mobilize stored starch reserves, and generate energy through fermentation pathways. Some weeds can do exactly the same. In fact, some biotypes of weeds, like Cyperus rotundus, have developed larger storage tubers that enlarge further in deeper flooded soils, allowing them to survive puddling (Ismail et al., 2012).

18

Herbicides are chemicals that kill plants or arrest their growth through various mechanisms. They can block the function of a target protein leading to the accumulation of a a toxic product, as in the case for glufosinate ammonium which inhibits glutamine synthetase and leads to an accumulation of ammonia (Devine et al., 1993). This leads to further changes which ultimately release reactive oxygen species that destroy lipids and membrane integrity. Consequently, glufosinate acts very quickly. ALS-inhibiting herbicides, on the other hand, work slowly because they block the production of essential product. By inhibiting the acetolactate synthase enzyme, production of the branched-chain amino acids leucine, isoleucine and valine ceases, and is the mechanism of action (Ray, 1984). The lack of these essential amino acids stops plant growth and eventually leads to its demise.

2.6.1 Basic biochemical mechanismsThe mode of action, or more precisely site of action, of herbicides can loosely be classified into those affecting light processes, cell metabolism or growth and/or cell division (HRAC, 2010). The site of action (sometimes also referred to as mechanism of action) is defined as where physically the herbicide must bind to inhibit an enzymatic pathway and the mode of action is the process by which it does so as a consequence of the herbicide action (WSSA, 2011). For the purposes of this brochure and to limit confusion, “mode of action” is used to refer to all. The progression of phases of herbicide activity have been summarized by Délye et al. (2013): penetration into the plant (shoot or root), translocation to the site of action, accumulation at the site of action (specific concentration of intact herbicide molecules is required for activity), binding to the target protein, and the ensuing damage, cell and plant death. Further information can be found in the following sources (Devine et al., 1993; Powles & Yu, 2010; WSSA, 2014). The general classification of herbicides into categories and groups according to the site of action as per HRAC (Menne & Köcher, 2007, 2012; HRAC, 2010) is presented in Table 3. There are 21 different known sites of action, with one broad group containing individual compounds whose site of action is as yet unknown.

2.6 How herbicides work

19

Category Group Herbicide Site of Action

Inhibition of light processes

C1, C2 & C3 Inhibition of photosynthesis at photosystem II (3 subgroups)

D Photosystem-I-electron diversion

E Inhibition of protoporphyrinogen oxidase (PPO)

F1

Bleaching: Inhibition of carotenoid biosynthesis at the phytoene desaturase step (PDS)

F2

Bleaching: Inhibition of 4-hydroxyphenyl-pyruvate-dioxygenase (4-HPPD)

F3 Bleaching: Inhibition of carotenoid biosynthesis (unknown target)

F4 Bleaching: Inhibition of DOXP synthase

Inhibition of cell metabolism

A Inhibition of acetyl CoA carboxylase (ACCase)

BInhibition of acetolactate synthase (ALS) or alternatively called

acetohydroxyacid synthase (AHAS)

G Inhibition of EPSP synthase

H Inhibition of glutamine synthetase

I Inhibition of DHP (dihydropteroate) synthase

M Uncoupling (membrane disruption)

N Inhibition of lipid synthesis – not ACCase inhibition

Inhibition of growth and/or cell division

K1 Microtubule assembly inhibition

K2 Inhibition of mitosis/microtubule organization

K3

Inhibition of VLCFAs – very long-chain fatty acids (inhibition of cell division)

L Inhibition of cell wall (cellulose) synthesis

O Action like indole acetic acid (synthetic auxins)

P Inhibition of auxin transport

Unknown ZUnknown – Note: while the site of action of herbicides in Group Z

is unknown, it is likely that they differ in site of action between themselves and from other groups.

HRAC Classification of Herbicides According to Site of Action.

From Menne and Köcher, 2007, 2012 and HRAC, 2010

2.6.2 Why is a mode of action important?With the increase of resistant populations, it has become much more important to understand the need for rotating modes of action in the herbicide program for a particular field. Using the same mode of action over and over in the same field will lead to the evolution of herbicide resistance. Another consideration is the use of mixtures of products with different modes of action in order to reduce the selection pressure for resistance on either of the mixture partners.

Tab

le 3

20

Summary of planting methods, seeding and planting conditions, and soil moisture and water management of rice production systems.

From Kraehmer et al., 2017

Two species of cultivated rice are grown throughout the world. Oryza glaberrima is grown primarily in Western Africa and is considered to be a minor species, and the major species Oryza sativa is grown everywhere else, with the bulk of it grown in Asia (Sweeny and McCouch, 2006). The latter species can be classified into two major eco-geographical races: indica rice and japonica rice. These can be differentiated morphologically, but when crossed generally lead to sterile progeny (Sweeny and McCouch, 2006). They can be further subdivided into a number of distinct groupings that are still not agreed upon, even when using modern molecular biology tools to determine the genetic makeup (Sweeny and McCouch, 2006). Several species of wild rice are known, generally from the genus Oryza and Zizania.

Presented in Table 4 is a summary of global rice production systems presented in Kraehmer et al., (2017). It shows the variety of topographic situations, planting methods and seedling, seeding and planting conditions, soil moisture conditions, and water management system after planting. More details can be found in this reference.

Tab

le 4

3.0 Systems Rice Culture and Weed Control

Planting Method Habitat Special Seeding and Planting Characteristics

Soil Moisture Conditions at Planting

Water Management After Planting

Transplanted Lowland 1-2 week old seedlings Flooded, puddled soil Irrigated

>3 week old seedlings Flooded, puddled soil Rainfed

Upland >3 week old seedlings Flooded, puddled soil Irrigated

Lowland Hand broadcast > 3 week old seedlings

Flooded, puddled soil Irrigated

Seeded Lowland Pre-germinated, broadcast Flooded, puddled soil Irrigated

Flooded, onto water surface Irrigated

Dry seeded broadcast incorporated Dry soil Irrigated

Dry seeded drilled

Dry seeded broadcast incorporated Rainfed

Dry seeded drilled

Upland Dry seeded broadcast incorporated Dry soil Rainfed

Dry seeded drilled

21

3.1 Transplanted riceThe oldest practiced form of rice cultivation involves transplanting 1-3 week-old seedlings into puddled soil in continuously flooded paddies. It is still the favored method in Japan and the Republic of Korea. The practice of puddling the soil results in physical and chemical changes that include the filling of interstitial spaces of macropores, breakup of soil aggregates, the formation of hardpans below the field to prevent water drainage, an anaerobic layer at and slightly below the water-soil interface, and soil pH values that approach neutrality, whether starting from acidic or basic conditions (Sanchez, 1976; Kirchhof et al., 2000). The anaerobic conditions in the upper layer of the soil inhibit weed seed germination because aerobic conditions are required for germination. Deepwater rice planting is a specialized version that is restricted mainly to periodically flooded coastal areas. It requires specially adapted varieties and practices.

3.2 Direct-seeded riceThere are many variants of the rice production systems that use direct-seeding. One variation called wet-seeded rice, which can be in upland or lowland habitats, involves sowing pre-germinated seeds onto soil. Water seeded or flooded rice, found in lowland habitats, involves sowing pre-germinated seeds onto flooded fields. It is being used in cases where weedy rice is being encountered in dry direct-seeded systems because it retains some of the advantages of transplanted rice systems (Chauhan, 2013). One of the main impediments to greater adoption of this system is the lack of local rice varieties that are adapted to germinating under more anaerobic conditions. Aerobic rice is a relatively new alternative to production systems that utilize large amounts of water. It is dry- seeded but dependent upon locally adapted rice varieties that can grow under more challenging conditions (Kraehmer et al., 2017). Aerobic rice is kept unflooded throughout the growing season.

3.3 Lowland riceLowland rice is another term used for rainfed or irrigated rice grown in lowlands and relies on direct seeding or transplanting seedlings. After planting, the crop may be irrigated or just flushed with water and never flooded. In some cases, herbicides are applied to soil by tractors in some rice growing areas, and the field is then floodeda few days later.

3.4 Upland riceSystems that use dry direct-seeded rice sown directly onto the soil and relying on rainfall for water are generally referred to as upland rice.

3.5 Main impediment for adoption of direct-seeded riceA survey of rice farmers in the Punjab region of India (Mahajan et al., 2013) revealed that the yield gap for direct-seeded rice growing systems compared with more traditional systems was the result of not following recommended fertilizer rates and weed control practices, e.g., using inappropriate herbicides, not applying them at the optimum time, using improper water volumes, as well as inappropriate nozzles. These issues could be solved through training and education programs tailored to local conditions. Additionally, more locally adapted cultivars would help increase yields.

22

4.0 The development of herbicide-resistant weeds

The correct application of herbicides following label recommendations is the most reliable and economically viable weed control practice in broad-acre agriculture. The development of herbicide-resistant weeds is threatening the sustainability of agriculture in some areas. What, why, and how this happens is described in this section.

4.1 Definitions of resistanceAccording to the WSSA (1998) herbicide resistance is defined as “… the inherited ability of a plant to survive and reproduce following exposure to a dose of herbicide normally lethal to the wild type.” HRAC (2017) defines resistance as the “… naturally-occurring inheritable ability of some weed biotypes within a given weed population to survive a herbicide treatment that should, under normal use conditions, effectively control that weed population.” Cross-resistance is defined by HRAC as resistance to two or more herbicides through only a single resistance mechanism, and multiple resistance is defined as resistance to several herbicides resulting from two or more distinct resistance mechanisms in the same plant or population.

4.2 Effective doseEvery weed has the ability to survive an herbicide application at a particular dose. It may be much lower or much higher than the applied dose. The dose that a weed can survive will vary between individual weeds and populations as shown in Figure 13. Environmental conditions can affect herbicide activity and potential herbicide injury to rice (Awan et al., 2016), thus they can bring different levels of control or potential for injury in different years. A labelled rate of an herbicide is chosen to provide excellent weed control (generally >95%) under a wide variety of environmental conditions, in a wide variety of soils over the lifetime of a product, often over 30 years. Thus, the labelled rate must take into account the year-to-year variability and should provide excellent control in most conditions.

Fig

ure

13

Use rate

Num

ber

of

wee

ds

100% efficacyWeed population

Sufficient activity(dependent upon rate, environmental conditions, application, growth stage)

Efficacy range in a population of weeds

23

81

37

7

51

1

11

11

1

4 9

285

212

12

80

113 1 2

27

47

13

15

2

3315

1

1

3

2

2

1

1

8

1 23 3

1

4.3 Global resistance trends

4.3.1 Global species countHerbicide resistance is a global problem and is being investigated and pursued globally by Bayer. Our group works with regional and local experts to help determine the extent of the weed resistance problem and helps find solutions to them. The resistance cases reported in The International Survey of Herbicide-Resistant Weeds (Heap, 2017) are accepted only if they respect certain restrictive criteria and represent only unique cases, meaning populations within a particular geography, for a particular species, for a particular resistance (by mode of action) or combination thereof. It takes time to follow the correct validation procedure. Cases are often reported years after they are discovered. Therefore, the current extent of herbicide resistance is not limited to the cases within the database and must be followed in the field to properly understand its distribution and intensity. In many situations of a reported resistance, the resistance potentially exists but it has not been validated and reported as such. However, the database serves to provide an account of key trends. The numbers of unique confirmed resistant weed species keep increasing across the globe. The count in 2016 showed the USA leading with 81, Europe (collectively) next with 80, followed by Australia with 47 (Figure 14).

Data Source: Heap, 2017. European countries including Russia and Turkey represented together.

Fig

ure

14

Count of confirmed resistant weed species by country in 2016

251

91 69AS OF JULY 2017

24

160

140

120

100

80

60

40

20

0

Num

ber

of

resi

stan

t sp

ecie

s

1950 1960 1970 1980 1990 2000 2010 2020

ACCase-A

ALS-B

PSII-C1

Glycines-G

Auxins-O

VLCFAs-K3

PPO-E

4.3.2 Resistant species by mode of actionThe results presented in Figure 15 show the increase in resistant species for selected modes of action. It shows that the number of species is rising most rapidly for ALS (Group B) and least rapidly for the auxins and very long-chain fatty acids (Groups O and K3). For PSII (C1) and ACCase inhibitors, the increase in number of resistant species lies somewhere in the middle, which is also the case for glyphosate. However, another important point is that for all chemistries (even older classes) it is rising.

4.4 Resistance evolution – a Darwinian process

4.4.1 Basic principlesResistance is a global problem and it is growing. The trend of increasing resistance is a constant. The number of resistant weed species, of resistant weed populations and those with cross- and multiple-resistances is constantly increasing. Let us be clear about one thing: It is the result of a natural evolutionary process as first described by Charles Darwin, that nature favors individuals that are more tolerant of stresses and pass this tolerance on to their offspring. It is a natural process that arises in naturally occurring mutations caused by cosmic and solar radiation and through DNA repair processes that sometimes go awry. An herbicide application in a field is certainly not a natural stress, but the process governing resistance development is one.

4.4.2 Resistance evolution dynamicsHow herbicide resistance develops can really be quite simple. If you always use the same method to control a weed, eventually it will be selected by individual weeds that can survive it. The control method applies to everything: from using only one herbicide, to alternating use of herbicides from the same mode of action group, or even hand-weeding (Barrett, 1983). An ingenious strategy, right? We mentioned naturally occurring mutations as the driver of genetic changes leading to survival of a particular stress factor. The range in these genetic changes can be extensive in populations. It is estimated that such mutations, for example a point mutation involving a single amino acid exchange conferring target-site resistance, occur in many weeds at a frequency of one out of every million. It can vary widely of course, but for the purpose of this theoretical example, let's use this estimate. Let's assume that we always spray the same herbicide over and over every year in the same crop and use no other measures to control the weed. If we assume that each surviving individual passes this on to

Number of resistant species for selected modes of action

Fig

ure

15

Data Source: Heap, 2017

25

4.4.3 Resistance and fitness penaltiesThere is evidence that resistance can result in reduced fitness, but the criteria for determining fitness has not been applied properly in most studies (Vila-Aiub et al., 2009). In some cases, resistance can result in no measurable fitness cost (Giacomini et al., 2014) or even in a higher fitness of the resistant biotype (Wang et al., 2010). Reviews state that the effects on fitness of the transfer of herbicide-tolerant traits from rice to weedy relatives are inconclusive (Kumar et al., 2008; Lu and Yang, 2009). Thus, the common assumption that resistance is always associated with a fitness cost is not correct.

Theoretical case of how resistance frequency increases in a weed population

10 offspring that germinate the following year, we see enrichment in the number of resistant individuals that appear as shown in Figure 16. It shows that resistant individuals can increase to become 10% of the population in 5 years and 100% in 6 years. This reflects what farmers can see in their fields. It shows that resistance can realistically be observed in the field only in the last phases, as in the last year or two, before it explodes in their fields. That is why farmers believe that resistance only takes a few years to develop, because that is what they see. Consequently, they believe that it takes only a few years to get rid of the resistance once it has evolved in a field. This is not correct. A population in a field "remembers" a resistance (Powles SD, pers. comm. 2014). For how long will a population keep this resistance? We don't know exactly, but it we believe in many cases it will be for decades. We have sampled fields that have not seen selection pressure from ACCase herbicides for at least 15 years and found that 20-30% or more of the population still contains resistance to this class of herbicides (Bayer, 2013). It certainly takes much longer to devolve resistance than it does to evolve resistance.

Fig

ure

16

Source: Bayer, 2015

What our data show happens

What would happen if it were true

What farmers believe happens

Re

sist

an

t w

ee

ds

(% o

f to

tal p

op

ula

tio

n)

Years

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

100

90

80

70

60

50

40

30

20

10

0

26

The risk of developing resistance is increased by a combination of factors that increase the selection pressure for resistance to an herbicide. The table below (Table 5) gives general guidelines to help classify management options that, when combined, help to decrease or increase the risk of developing resistance in a particular field. The evolution of resistance is a “numbers game.” Higher weed infestations, along with lower numbers of herbicides (and modes of action), crops, and control methods within a cropping system lead to higher risk of resistance.

4.5 Resistance selection pressure and risk assessment

4.5.1 Use of below-labeled ratesIncreasing evidence indicates that the use of low (below-labeled) rates of herbicides can lead to the evolution of resistance, particularly through enhanced metabolism, which is evident in both controlled studies (Neve et al., 2005; Yu et al., 2013) and in the field (Manalil et al., 2011). Increases in tolerance (below resistance level at below-labeled use rates) appear to be incremental with each generation resulting in a relatively slow evolution of resistance. This is attributed to the assumed multi-genetic characteristic of this type of resistance mechanism. Soil-active herbicides are not exempt from this behavior and resistance can rapidly evolve (Busi et al., 2012). Further work (Busi et al., 2014) showed that even when resistance can develop rapidly, if management options are chosen properly, they can help to reduce the selection pressure for resistance.

Management Option

Resistance Risk

LOW MEDIUM HIGH

Herbicide mix or rotation in cropping system

>2 modes of action 2 modes of action 1 mode of action

Weed control in cropping system

Cultural, mechanical and chemical Cultural and chemical Chemical only

Use of same mode of action in cropping season

Once More than once Many times

Cropping system Full rotation Limited rotation No rotation

Resistance status to mode of action(s)

Unknown Limited Common

Weed infestation Low Moderate High

Control in last three years Good Declining Poor

From HRAC (2017), accessed 21 Mar 2017, available online at http://www.hracglobal.com/.

Assessment of the risk of resistance development per target species

Tab

le 5

27

4.5.2 Help from non-chemical supplementary measures (soil tillage, cover crop, seeding density, competitive varieties, etc.)The reality of the current situation (presented in general terms) is that we are relying mainly on herbicides to perform weed control in many areas of broad-acre agriculture and not relying much on mechanical or other cultural, non-chemical weed control practices, even in supplementary roles. In order to maintain the effectiveness of herbicides, which do the heavy work in controlling weeds, we need to include more non-chemical weed control practices to take the selection pressure off of the herbicides. And certainly we need to do more to protect individual modes of action and individual herbicides, by using full rates, mixtures, and sequences. See Section 6.1.2.2 for more information on how diversifying weed management strategies can reduce weed populations and help reduce selection pressure for herbicide resistance.

4.5.3 Lack of weed control innovationsMany classes of herbicides, as determined by their mode of action, have been discovered during the period of the late 1940s to the early 2000s (Figure 17). Since the mid-1980s no new mode of action of major market significance has been discovered. Herbicides representing more than the number of classes currently on the market have been discovered, but have not been registered. Most have failed due to toxicological, environmental, or production cost issues. The increasing regulatory hurdles, along with other reasons, have made it harder to find good herbicides and have made them more expensive to register and develop.

One of the most startling trends is shown in Figure 18, a graph depicting new active ingredient market launches of all modes of action starting in 1950 and shown in 5-year intervals. It appears that we have come full circle and returned to the beginning of the herbicide era. The heyday was reached in the period between 1990 and 2004, showing an abrupt drop in the four-year period of 2011-2015. This truly shows the lack of herbicide innovation in the industry today. A key recommendation by the WSSA to address herbicide resistance is to point out the rarity of the discovery and bringing to market of new modes of action and that existing herbicide resources are “exhaustible” (Norsworthy et al., 2012). It is sobering to think that each herbicide or mode of action lost to resistance may be “lost” for up to decades in a particular field. Additionally, each herbicide lost due to a regulatory issue will completely disappear from the arsenal of available tools. We are not discovering new herbicides fast enough to replace the ones we have lost. We must do all we can to protect each and every remaining herbicide. We do not know if an older herbicide will be the ideal complement to a brand-new herbicide with a new yet undiscovered mode of action. As Stephen Powles, a renowned professor of Weed Science from Australia puts it, “Each herbicide is a treasure that must be preserved.”

Commercially available and experimental modes of action never registered and marketed

1950 1960 1970 1980 1990 2000 2010

Com

mer

cial

Auxin

PS II

PS I

Uncoupling

Cell Division ACCase PPO

HPPD

ALS

PDS

DXS

EPSP

GS

Cytokinin Ox

(Cinmethylin)

(Oxaziclomefone)

Microtubuli

Cellulose

DHPS

VLCFA

Expe

rimen

tal

LyCyc

CGS

CBL ZDSGibB ObtDM AMPDA

DXR

TrpS

GPAT CLObtl

C24SMT

OSCyc

Δ7StC5Des

Δ24StRed

PyrDeh

KARI βIMDH

FPPS

ADSS

IGPDGSAT

Breakthrough towardsdiscovery of new

development candidates Fi

gur

e 17

Source: Bayer, 2015

28

It will not be easy to convince farmers that this is a necessary step. Actually, it will be hard (Norsworthy et al.,2013). But there are signs that working to relieve the selection pressure on resistance will pay for itself. But as we have seen, mere facts are not enough. We need to be more persuasive. And this means we need to communicate better.

4.6 Resistance mechanismsAn herbicide works by binding to a pocket in a particular enzyme and blocking a biochemical pathway. This results in the accumulation of a product that is deleterious to the plant or a decrease in a product that is essential for growth. Ultimately, the plant dies. Mutations occur naturally through cosmic rays, sunlight (heat), natural DNA repair mechanisms, or chromosomal hybrids gone awry. A target-site mutation can cause a single amino acid substitution and cause a structural change in the binding pocket. Studies have shown that in many cases, this is the result of a single gene. The result is that the herbicide only partially fits into the binding pocket and does not work properly or only partially.

We are not discovering new herbicides fast enough to replace the ones we have lost.

Herbicide active ingredient launches 1950-2014

Data from Phillips McDougall (2016) and Bayer (2016).

Fig

ure

18

1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

13

19

5

25

18

23

16

24

31 31

21

16

7

0

29

Every plant has the capability to degrade an herbicide at a particular rate and usually it is degraded to biologically non-active products. Thus, it is the speed of degradation that ultimately matters and affects whether the herbicide is structurally altered quickly enough so that it does not reach the target site intact. Figure 19 illustrates a few of the more common resistance mechanisms. The figure shows conceptually how herbicides work by fitting into the binding pocket at the target site. An altered target site results in a physical change to the binding pocket that prevents the herbicide from fitting properly, such as through a target-site mutation. Resistance through enhanced metabolism (metabolic resistance) results when naturally occurring degradation processes evolve to become more efficient or active to a particular chemistry, and alter the structure of the herbicide before it reaches the target site. Sometimes a weed evolves a mechanism in which it produces a large increase in the enzyme target so that the normally effective herbicide is in effect “diluted” to a lower level of activity. There are other resistance mechanisms. Herbicide resistance mechanisms can be broadly classified into target-site and non target-site resistance (Table 6). The list includes diverse mechanisms that utilize changes in biochemical processes within plants, changes to exterior structures or interior redirection away from the target site, or changes in germination to avoid peak soil herbicide concentrations or application windows. By far, target-site mutations and enhanced metabolism are two of the most common mechanisms encountered. It is currently believed that non target-

site resistance mechanisms involve multiple genes. For a more thorough and detailed discussion of this aspect and resistance mechanisms, please refer to Powles & Yu (2010), Délye et al. (2013) and Yu & Powles (2014). Weeds constantly evolve novel mechanisms of resistance as they are confronted with new management techniques, representing a new type of stress on the population. It is thus important, in order to preserve the techniques that are currently effective on a particular weed, to change them through pre-planned management strategy.

*Presented in Figure 19

Tab

le 6

Targetenzyme

Targetenzyme

Targetenzyme

HerbicideHerbicide HerbicideHigher

enzy

me

level

Altered site

of actionAlteredstructure

Herbicide fits intotarget, inhibits enzyme

and controls weed

Herbicide does not fitinto target and does

not control weed

Herbicide structure alteredby degradation process

and does not control weed

Herbicide fits into site of action, but too much

enzyme present

Fig

ure

19

Source: Bayer, 2015

Mechanism Resistance Class

Target-site mutation*

Target-siteIncreased gene copy number*

Enzyme overexpression

Enhanced metabolism*

Non target-site

Differential uptake

Differential redistribution

Sequestration

Delayed germination

Rapid necrosis/defoliation

Selected mechanisms of herbicide resistance

Diversity of selected herbicide resistance mechanisms

30

4.7 Multiple resistanceHRAC (2017) defines multiple resistance as occurring “… when resistance to several herbicides results from two or more distinct resistance mechanisms in the same plant.” This definition has also been routinely applied to weed populations within a field since control of weeds is usually performed at this scale. The appearance of resistance to multiple herbicide groups (MoAs) in a population of weeds complicates its management. An example can be found in rice in the US with barnyard grass (Echinochloa crus-galli). The tested barnyard grass population was resistant to three herbicide classes representing different MoAs (HRAC Group B, C2 and O), and to several representative herbicides within a MoA (Table 7). What is interesting to note is that for some compounds that can be applied either pre- or post-emergent, for example quinclorac, the activity on the resistant population was much stronger when applied pre-emergent. Although not resistant to pendimethalin, cyhalofop or thiobencarb, the resistant population showed less control than in the susceptible population. Perhaps this is indicative of previous selection pressure, and thus heavier use of any of these compounds alone in the future would tip the population to resistance. The lack of efficacy of ALS-inhibiting, auxinic and PSII-inhibiting (C2) herbicides as well as the reduced efficacy of cyhalofop and pendimethalin (HRAC Groups A and K3, respectively) are decreasing barnyard grass control options for this field. The increase in multiple- and cross-resistance has complicated weed management in other species and geographies. In Japan Iwakami et al. (2015) describe a biotype of Echinochloa crus-galli var. formosensis showing resistance to cyhalofop-butyl and several ALS inhibitors with mechanisms suspected to be non target-site based. Multiple- and cross-resistance are threatening the future of broad-acre agriculture and have resulted in a call for the discovery of novel herbicides with new modes of action (Tranel et al., 2011). This makes it even more important to implement IWM programs to safeguard the efficacy of herbicides still working.

Tab

le 7

Barnyard grass Control

Herbicide MoA (HRAC)

Chemical Family

Timing Dose (g ai ha-1)

Susceptible Resistant

Clethodim A DIM post 280 E E

Cyhalofop A FOP post 314 G M

Fenoxaprop A FOP post 120 E E

Bispyribac-sodium B PTB post 36 E P

Imazethapyr B IMI pre 106 G P

Imazethapyr B IMI post 106 E P

Imazamox B IMI post 45 E N

Penoxsulam B TAP post 40 E N

Pendimethalin K1 DNA pre 1120 G M

Quinclorac O QCA pre 560 E G

Quinclorac O QCA post 560 E P

Atrazine C¹ TAZ post 2240 E E

Propanil C2 AMI post 4480 E N

Thiobencarb N THC pre 4480 E G

Glyphosate G GLY post 870 E E

Glufosinate H PHA post 590 E E

Clomazone F4 - pre 36 E E

Paraquat D BIP post 701 E E

Barnyardgrass control scale is based on % visual control E: 95-100, G: 94-85, M: 84-70, P: 69-20, N: 19-0Chemical family abbreviations - DIM: cyclohexanediones, FOP: aryloxyphenoxy-propionates, PTB: pyri-midinyl(thio)benzoate, IMI: imidazolinones, TAP: triazolopyrimidines, DAN: dinitroanilines, QCA: quinoline carboxylic acids, TAZ: triazines, AMI: amides, THC: thiocarbamates, GLY: glycines, PHA: phosphinic acids, -: none given, BIP: bipyridiliums. The resistant barnyard grass biotype is not sufficiently controlled by the herbi-cides with the light gray shading. Clethodim and atrazine (shown in italics) are not selective in rice and were tested as products to be used in rotated crops. Tabular data reformulated from Norsworthy et al., 2014b.

Control of ALS-resistant barnyard grass from Arkansas by a variety of herbicides.

31

4.8 Resistance dynamics – spread vs. independent evolutionMany farmers believe that resistance comes from their neighbors. They believe it starts in a neighbor’s field and eventually spreads to their fields, as in Figure 20 (left). Although this does indeed occur, they do not understand that resistance can also develop independently in their fields. One reason is that many farmers use weed control products correctly, and despite following all product recommendations for application rate and optimum conditions, they obtain excellent weed control for a number of years yet weed resistance still occurs. They may not be paying attention to the overall picture over time. What farmers perhaps do not realize is that using one product that works well over and over, without any other complementary weed control measures, results in resistance evolution. Certainly it is known that resistance can spread. How quickly it will spread depends on the particular species and on agronomic factors such as the use of herbicide mixtures and sequences, crop rotations, tillage, and other non- chemical weed control measures. However, resistance evolution also depends on what farmers do in each field. A distinction in flooded rice is the potential for dispersal of resistant seeds through water movement, as in through normal irrigation water disbursement (which may be under the control of a local or regional water authority) or through shared machinery bringing in resistant seeds (Vongsaroj and Tjitosemito, 1993).

At Bayer, we are conducting in-depth resistance evolution studies in small landscapes with Alopecurus myosuroides in Germany that show that resistance evolution cannot be explained exclusively through movement of pollen or seed (Hess et al., 2012; Hermann et al., 2014). The distribution of resistance can take on a checkerboard pattern (Figure 20, right), with highly resistant fields, like those in red, side-by-side with fields that have sensitive weed populations, like those in dark green. The intensity and profile for individual fields can take on intermediate character, as shown with the mixture of different colors. Even all fields from one farmer can look different, or similar. We increasingly believe that a significant component of resistance evolution is attributable to independent evolution, even if the initial resistant plant occurs through selection or introduction, based upon the weed control history in an individual field. That means that the farmer has an opportunity to prevent, or at least significantly delay, resistance in his field based on what he does in each field. However, he must include diversity in his weed control measures. Diversity in herbicides and modes of action. Diversity of non-chemical methods, including crop rotations (where possible), tillage, and other measures. We must make a stronger effort to convince farmers of the need for this.

Figu

re 2

0

Source: Bayer, 2015

Farmer CFarmer A Farmer B

Farmer D

Farmer CFarmer A Farmer B

Farmer D

Resistance spread vs. independent evolution.

32

5.0 Resistanceconfirmation testing and diagnosis

The testing and confirmation of resistance is not a trivial exercise. Great care must be taken first in the sampling of the seed (or plant material) and consideration of the type of testing that should be done. Suspicion that resistance is responsible for the lack of expected efficacy is often associated with agronomic reasons having nothing to do with resistance (e.g., application error, environmental or soil conditions). Great care should be taken to ensure uniform conditions in the greenhouse, the workhorse for resistance testing. The selection of the susceptible population is another critical aspect of validation of resistance, with the use of populations that represent the middle of the range of susceptibility (see Figure 13).

Several publications describe the correct steps and protocols for determination of the resistance status of a population of weeds (Beckie et al., 2000; Heap, 2015; HRAC, 2015). For a weed biotype to be listed in the International Survey of Herbicide-Resistant Weeds (www.weedscience.org), it must meet all of the criteria summarized in the following list (Heap, 2015).

5.1 Validation testing

1. Fulfillment of the WSSA definition of resistance and the survey’s definition of an herbicide-resistant weed

2. Data confirmation using acceptable scientific protocols

3. The resistance must be heritable

4. Demonstration of practical field impact

5. Identification as a problem weed to species level, not the result of deliberate/artificial selection

33

5.2 Types of testingIn our studies on resistance, we use proven methods like whole-pot greenhouse bioassays and innovative technologies for metabolic resistance analyses as well as target-site resistance analyses. In the greenhouse, we use a number of compounds, some with multiple rates, to help determine what other compounds can still be used effectively, in addition to confirming resistance to a particular herbicide or herbicide class. To study the metabolism in weeds, currently we incubate them in solutions containing radiolabeled herbicides, extract them, and measure them on an HPLC with in-line detection. To analyze target-site mutations, we run them on a PCR and quantify results using a pyrosequencer. No other company routinely runs as many samples using the full breadth studies as does Bayer. We are constantly evaluating new technologies to use in studying resistance.

5.2.1 Greenhouse testsGreenhouse bioassays form the backbone of most diagnostics testing for suspected cases of resistance. Disadvantages include the fact that they are relatively slow, need a significant area of greenhouse space, but they are relatively robust regarding climate conditions. They generally require a fair amount of manual labor to conduct, particularly for seed cleaning operations, pre-germination, sowing, visual evaluation, etc. However, in terms of being closer to field or practical application and growth conditions, this gives them a considerable advantage over other assay techniques. They are more independent of weed species and have higher versatility. All compounds, MoA classes, and even mixtures, sequences and dose rates can be tested. The symptoms can give indications of resistance mechanisms. They can detect cross-resistance for compounds with the same or different MoAs.

34

5.3 Interpreting results and making recommendations

The purpose of testing for resistance is not only to determine the resistance status to a particular herbicide or class of herbicides. In the case of a positive confirmation of such resistance, it is also to determine the resistance status of other available control options (Beckie et al., 2000). To do so, it means that the potential options should have been considered beforehand. Ultimately, the objective is to understand the resistance status as well as possible in order to make the best recommendation to the farmer for their particular field.

5.2.2 “Quick” testsSo called “quick tests” have the advantage that they can generally be completed more quickly, cheaply, and with less space than traditional greenhouse testing. However, they do come with some disadvantages. For some tests, more manual work is actually required than for greenhouse studies. Some methods can give “artificial” application conditions not encountered in an actual field situation, such as a leaf-active herbicide being applied to germinating seedlings where root uptake is the primary mechanism of plant entry. These methods tend to be more weed and compound specific. They generally use a more discriminating (single) dose rate, which often allows only yes/no decisions. It is hard to assess developing resistance (not yet established) using this kind of system, particularly when the development of metabolic resistance is a possibility. The determination of potential resistance mechanisms is much more difficult because often such tests run over a shorter time interval and symptoms may not be fully developed. Despite the stated disadvantages, it is clear that there are several advantages of quick-tests over greenhouse tests. They are faster to run than full bioassays and thus provide answers more quickly, although perhaps the answers are not as complete as those received through greenhouse tests. They generally require less space through the use of smaller items (e.g., Petri plates vs. pots), although sometimes due to the need for more controlled climate conditions, the use of specialized equipment like climate chambers is necessary.

5.2.3 Field testsThe use of field studies is generally discouraged for validation of resistance except as a screening tool. The difficulty of comparing a field site location suspected of resistance with control of a sensitive population, unless it was already growing naturally at the site, make it less precise in determinations of resistance.

35

6.0 Integrated Weed Management

Integrated Weed Management can be defined as “a holistic approach to weed management that integrates different methods of weed control to provide the crop with an advantage over weeds” (Harker & O’Donovan, 2013). Or, as seen from another perspective, as sustainable weed management involving “… combined use of preventive, cultural, mechanical, chemical, and biological weed control techniques in an effective and economical way” (Majahan et al., 2014). It involves the integration of chemical, cultural, and biological methods of weed control and suppression using knowledge of the weed’s weaknesses to help keep weed populations manageable, reduce selection pressure for weed resistance, maintain sustainability of cropping systems while reducing the environmental impact of weed management practices. This section starts with some core IWM guidelines and describes in general terms how to set up an Integrated Weed Management strategy. All weed management is on a field-by-field basis and thus all generalized recommendations must be put into the context of local practices. In the past, chemical weed control was perceived as competing against manual labor and thus depriving some individuals of their livelihood. However, studies showed that it did not displace labor, but served to increase farm productivity and profitability. Now with the increasing difficulty of finding labor, more farmers are relying on herbicides (De Datta, 1981). However tempting it is for farmers to rely on herbicides only, recent modeling results indicate that integrating chemical and non-chemical measures leads not only to barnyard grass control that is equivalent to hand-weeding, but also maximizes profits over the long-term (Beltran et al., 2012).

Fig

ure

21

Biological life cycle of a weed (above) and potential control measures (below) for particular key decision points (numerals).

1

2

3

4

5

6

Seed survival

Seedling emergence

Seedling survival

Seed bank

Seedlings

Adults

Viability

Migration

Seed losses

Migration

Seed rain

Flowers / inflorescences

7

8

1. Inversion tillage, no-till, planting of certain cover crops to increase seed predation, soil solarization

2. Stale/false seedbed, planting date, water manage-ment, pre-emergence herbicide, cover crops to decrease germination, mulching, crop rotation counter to weed cycle, polyfilm covers

3. Early post-emergent herbicide, competitive crop variety, narrow row spacing, increased seeding rates, intercropping, fertility management, water management, introduction of exotic natural enemies

4. Late post-emergent herbicide, hand-weeding prior to flowering

5. Salvage herbicide, gametocide, harvest weed seed control, hand-weeding, and removal from field

6. Field border or water canal management, attracting foraging wildlife

7. Crop residue burning, water management, fallowing, cover crops to increase seed predation, silaging

8. Land preparation, soil erosion, contamination from crop seed lots or machinery

Redrawn from Labrada, 2006.

36

Preventivemeasures Stale

seedbedtechnique

Herbicideapplicationand dose

Cropsowing

time

Croprotation

Cropresidue

Narrowrow

spacing

Highseeding

rates

Weedcompetitive

cultivars

IntegratedWeed

ManagementStrategies

Potential components of an integrated weed management strategy in rice.

Fig

ure

22

Chauhan and Mahajan (2012) succinctly list eight non-chemical components that can be combined with optimized herbicide application in an integrated fashion to manage weeds more sustainably (Figure 22). Their perspective is that integrated approaches are necessary to be able to continue to benefit from the many positive aspects which conservation tillage practices have brought to rice crop, using integrated strategies is essential. They also categorically state they “… will also ensure that herbicide use remains profitable and environmentally sound over a long period of time.”

1. Know the weed spectrum

2. Develop a weed management strategy

2.1. Develop a plan

2.2. Diversify weed management measures

2.3. Enhance crop competitiveness

2.4. Start clean

2.5. Aggressively manage weeds

2.6. Maximize herbicide activity

3. Demonstrate in practice IWM techniques

6.1 Bayer CropScienceIWM Core Guidelines

Tab

le 4

Much has been written about IWM guidelines. The Bayer IWM Core Guidelines are an attempt to offer simple guidelines that are understandable to a large audience and applicable globally and that can be adapted to local conditions. However, the intent is to ensure that a multi-faceted and well thought-out approach be undertaken to manage weeds. Included is the importance of the need to convince others to adopt them using one of the most effective means available – demonstrating them in practice – in the field. The core guidelines consist of three main points: know the weed spectrum, develop a weed management strategy, and demonstrate in practice IWM techniques. The weed management strategy addresses several aspects that should be included to maximize effectiveness: the need to plan ahead and to stick to the plan, the emphasis on adopting diversified weed management measures, maximizing the help that crops themselves can give to suppressing weed growth, the need to be aggressive in managing weeds and not to let up, the advantage of “starting clean” to help give crop establishment additional help, and the need to maximize herbicide activity to keep the surviving numbers of weeds to a minimum.

It is extremely important to understand the biological cycle of a key weed within the context of a cropping cycle. Doing so helps to highlight key decision points and suggest key control measures that may not be currently implemented in the region for a particular crop, as shown in Figure 21. For any IWM program to be effective at the farmer level, training is required and must address the ecology of key weeds and the crop and how they interact (Valverde and Itoh, 2001; Labrada, 2006).

Redrawn from Chauhan and Mahajan (2012).

37

6.1.1 Know the weed spectrumIt is important to know driver weed is, or weeds are, in a field. It is important to identify them correctly (see Section 2.4) and to understand to understand their biology (see Section 2.5) in order to to understand their weaknesses and to guide the development of an effective management strategy. Scouting a field before application is important to identify the weeds present and to identify areas of a field that may have higher weed populations that may need special attention.

6.1.2 Develop a weed management strategyThe goal of a weed management strategy is to disrupt the life cycle of the driver weeds to make the management thereof easier through the chosen combination of measures. Another goal of the management strategy is that it should also be sustainable and contribute to the preservation of currently working herbicides and non-chemical weed management measures. Individual components are detailed below.

6.1.2.1 Develop a planThe plan needs to be made for more than just one year. It takes a multi-year approach to combat weeds, particularly resistant weeds. Start with the goal of the plan. For example, a 99% reduction in the population of resistant weeds in field X within three years. The goal of the plan should be realistic. Once the goal of the plan has been achieved, there should be no letup in effort. The buildup of weeds can happen very quickly, so that attention needs to be paid to continuing a weed management strategy. Thus the goal of the program should then be changed to avoiding the return of resistance and of higher weed populations. A new plan that emphasizes diversity must be implemented or you will just be facing the return of a new kind of resistance.

6.1.2.2 Diversify weed management strategiesDiversity in weed management using strategies working at several levels in an integrated manner is one of the most effective ways to keep weeds off balance and make it easier for herbicides to do their job. The stale seedbed technique is a good example of an integrated measure that employs knowledge of the biology (germination) with various chemical (herbicides) or non-chemical (inversion tillage) weed destruction technologies to give the rice crop a head start over weeds in direct-seeded systems (Rao et al., 2007). Another trial from Nepal investigated the use of stale seedbeds, mulch, hand-weeding, and a co-culture of Sesbania with the use of herbicides in various programs on weed control, yield, and net profit (Figure 23). The treatment with pendimethalin

followed by 2,4-D and a subsequent hand-weeding resulted resulted in a rice yield that was slightly lower than the weed-free plot in

2010 and comparable to the 2011 yield. The difference between the two years in the production of rice was

significant. The year 2010 appeared to be a particularly strong year for the weeds and a difficult year for

the rice, which was illustrated by the fact that in the weedy plot the relative net profit fell to

minus 70% because the yield was very low. Some of the treatments barely

made a net profit. The other two best treatments in both years

were the pendimethalin followed by two

hand-weedings and the stale seedbed followed by pendimethalin

38

Diversity includes the management of chemical resources. As a basic principle, one should avoid repeated and continued use of the same herbicide or herbicides having the same mode of action in the same field, in the same growing season, and in the following year when only these are being applied. Mixtures or sequential treatments of herbicides that are active on the driver weed but have different modes of action are an effective strategy to reduce selection pressure on a particular herbicide or herbicide group. From experience, we can conclude that rotation of herbicides alone is not enough to prevent the eventual development of resistance. To retain these valuable tools over the long-term, the chemical rotation explained must be employed in association with at least some of the other weed control measures outlined. In cases where metabolic resistance is already present, the mode of action of the herbicide is not always the key criterion for the selection of products. In these cases, the mechanism of degradation can be very important across site of action groups and chemistries. No classification of herbicides relating to metabolic resistance is available yet and such examples need to be handled on a case-by-case basis.

6.1.2.3 Enhance crop competitivenessCrop competitiveness of rice can be enhanced with the selection of cultivars that have higher seedling vigor, early and faster seedling establishment, and exhibit allelopathy (Dass et al., 2017). Increasing seeding rates in direct-seeded rice and planting using narrower rows in both direct-seeded and transplanted rice resulted in lower weed biomass. Farmers generally choose cultivars that are perceived to bring the highest yield that are bred and normally grown in a weed-free environment (Andrew et al., 2015). However, if weed resistance leads to greater competition by weeds within the crop, yield losses can outpace any potential gains due to superior germplasm or breeding (Brennan et al., 2001). Thus, if faced with a situation of resistant weeds in dense populations, greater attention to the competitive ability of the cultivar can bring dividends. Other factors that can help the competitiveness of rice if used properly include: tillage systems, the use of crop residue, appropriate

Figu

re 2

3

Tabular data from Bhurer et al., 2013 reformatted as graph. Weed control program key: W weedy, WF weed-free (by continuous hand-weeding), P pendimethalin, B bispyribac-sodium, / followed by, HW hand-weeding, SS stale seedbed with glyphosate, M mulch, C Sesbania co-culture, D 2,4-D Na salt. The WF was used as basis for calculation of relative yield and net profit. For more details see reference.

Rice yield and net profit as influenced by different weed management treatments.

120

100

80

60

40

20

0Rel

ativ

e Y

ield

Rel

ativ

e N

et P

rofit

(%)

2010 2011

P/D

/HW

P/C

/D/H

W

SS

/M/B

SS

/P/B

SS

/B

P/H

W/H

W

M/B

/HW

P/BWFW

P/D

/HW

P/C

/D/H

W

SS

/M/B

SS

/P/B

SS

/B

P/H

W/H

W

M/B

/HW

P/BWFW

120

100

80

60

40

20

0

-20

-40

-60

-80

followed by bispyribac-sodium weed control programs. None of the four previously mentioned programs were statistically different from one another. However, when net profit was calculated factoring inputs and yield by price as the output, the pendimethalin followed by 2,4-D and a subsequent hand-weeding clearly resulted in the highest net profit, while the weed control programs of pendimethalin and two subsequent hand-weedings and the stale seedbed followed by pendimethalin followed by bispyribac-sodium were not far behind. Interestingly, the stale seedbed followed by bispyribac-sodium only gave similar results for net profit. The net profit for the treatment with pendimethalin and Sesbania co-culture followed by the 2,4-D Na salt and a subsequent hand-weeding was comparable to the net profit from the weed-free program (through hand-weeding). The use of mulch added little in this experiment. The results show that using solely the herbicide treatment (without a stale seedbed) was not enough to curtail early weed competition, and that the mixed practices, especially those incorporating hand-weeding and the stale seedbed, gave the highest yield and net profit.

39

water depth and duration, appropriate agronomic practices such as row spacing and seeding rates, manual or mechanical weeding, and appropriate herbicide timing, rotation, and mixtures (Chauhan, 2013). The economic value to farmers of addressing crop competitiveness was discussed in the previous section (6.1.2.2).

6.1.2.4 Start cleanPlanting into weed-free fields is one of the best ways to ensure early and strong crop establishment (IRRI, 2017). Controlling early weed flushes with tillage or use of non-selective herbicides before crop emergence is one way to achieve this. Using pre-emergent herbicides after tillage or in combinations with non-selective herbicides in burndown applications (sometimes with overlapping pre-emergent herbicides) is a good strategy (IRRI, 2017). Using seed saved from the previous season may bring false economic savings if resistance is developing in a field and a farmer is merely spreading over more fields on his farm.

6.1.2.5 Aggressively manage weedsAggressive weeds need to be managed aggressively. In reality, all weeds need to be managed aggressively without letup. The best strategy is to keep weed densities low so that the contribution to the soil seed bank remains low. Studies have shown that for some species, low densities, even after successful management and reduction over several years, can quickly lead to reestablishment of high densities in a short time frame if abandoning successful management practices (Norsworthy et al., 2014a; Buhler et al., 2001; Yamada et al., 2007).

6.1.2.6 Maximize herbicide activityThe key justifications for paying attention to maximizing herbicide activity are to reduce the potential number of weed escapes, thus reducing the opportunities to increase the weed population and to derive the maximum economic benefit from the treatment. Knowing which weeds infest a field or border area is essential to guide the correct choice of herbicides. Following label use instructions carefully, particularly those concerning recommended use rates and application timing (weed size) and optimum application conditions (environment, adjuvants, nozzles, etc.) is essential. The scouting of weeds after herbicide applications is encouraged to determine the effectiveness of the treatment(s) and to allow the detection of subsequent weed flushes. Maintaining accurate and detailed records of field history ensures that associations with certain practices become more evident and that effective adjustments can be made to management strategies in subsequent years.

6.1.3 Demonstrate in practice IWM techniquesOne of the most effective ways to convince farmers to adopt IWM techniques is to show them how it can work in practical field situations. It helps farmers who have not yet adopted IWM become aware about strategies alternative to the ones they are following and the potential value of integrated weed management to them. After all, “seeing is believing” as one farmer at a Respect the Rotation event in the USA put it.

A successful program for the management of a very competitive weed affecting an entire community comes from Arkansas. The program provided the recognition that the simultaneous management of several aspects of infestations of Amaranthus palmeri (including in-field, field borders, and roadsides) need to be done not by individual farmers, but by a community working together simultaneously (Barber et al., 2015). Some of the recommendations for how to institute a successful program include starting with an appropriate local leader within the agricultural community, developing a common goal and creating an identity or branding for the program, using science-based information to guide the development of practices to be adopted, and publicizing results. The adage that together growers are stronger than when working alone certainly applies in this case.

6.2 The need for proactive weed control within a field and at a community level

40

6.3 Other non-chemical measuresCover crops can be a useful non-chemical tool to disrupt the life cycle of a weed and can lead to decreases in biomass and seed production, but they need to be integrated into cropping systems and adapted to the biological and germination characteristics of the target weed (Price et al., 2011; Korres & Norsworthy, 2015). In rice, there have been some attempts at controlling weeds with plant pathogens, but this approach generally suffers because of specificity of the pathogen is generally limited to one particular weed species (Charudattan, 2001). Using novel approaches of treatment with multiple pathogens is achieving some success. Managing weed seeds at harvest and after harvest through controlled burning (where permitted), chaff collection, and mechanical destruction(Harrington Weed Seed Destructor Figure 24)help to prevent a buildup of the weed seed bank (Walsh et al., 2013), and are currently being evaluated in rice in Australia. However, it has been acknowledged that they are not suitable for all weed types and that many technical issues relating to production systems have to be resolved (Broster, 2017). Others are using narrow windrow burning of rice residues in Arkansas to reduce weed seed return to the soil (Schwartz et al., 2016). Using plastic film to cover soil and raise temperatures to a maximum of between 50-70 °C to keep weeds in check, has also been successful. The differences observed were between film thicknesses, color (clear or black), and duration of soil coverage (Khan et al., 2003). Other strategies are reviewed in Norsworthy et al., 2012.

Fig

ure

24

Source: http://www.producer.com/wp-content/uploads/2013/08/Harrington_Seed_Destructor.jpg

A Harrington Seed Destructor – post-harvest weed seed control

6.4 How to measure successIt is important to measure the success of IWM programs. This helps when setting goals for IWM programs by permitting the assessment of progress. A previous version of this brochure stated that the purpose of IWM is to “… reduce weed pressure and keep weeds below their economic thresholds.” With the increase of weed resistance, particularly multiple resistance, and with the escalation of species like Palmer amaranth that can rapidly take over fields, the economic threshold concept is being realistically superseded by less acceptance of weeds, such as in the zero-tolerance concept. Thus the previous emphasis on population density (number of plants per square meter or acre) or weed biomass reduction as a measure of program effectiveness (and accepting a certain level of weeds) is becoming less relevant. A more appropriate measure for the developing situation is weed seed set. The emphasis on decreasing the numbers of seeds returned to the soil seed bank is more versatile and can be used, for example, to measure post-harvest weed seed control methods. It is also a measure, like weed density, that farmers can easily understand and to which they can relate.

6.5 Other sources of information

Please refer to our references for detailed scientific information and to our website http://www.iwm.bayer.com/ for current sources of information.

41

COMPETENCE CENTERRESISTANCEWEED

Weed management is being complicated through the increased resistance in more and more species, in more and more fields, to more and more herbicides, and to an ever increasing number of products and modes of action. Farmers love simplicity. In fact, they value it. However, the future means that we will need to offer solutions that address this complexity while being as simple as possible. That is one of our biggest challenges.

7.1 Introduction

7.0 The Weed ResistanceCompetence Center (WRCC)

Just what is the Weed Resistance Competence Center (WRCC)? It is the Bayer global reference center for weed resistance management, understanding resistance mechanisms and evolution in the field by weeds, testing and developing new concepts, strategies, and tools to diagnose and manage resistant weeds, and communicating and sharing Bayer knowledge and solutions with farmers, advisors, distributors, and officials. It has been operating functionally in Frankfurt for almost two decades, with the official launch of the WRCC occurring in November of 2014. We work closely with our partners in the Marketing Division, whose focus is on not only providing compelling products for weed management, but also in increasing farmers’ profitability, preserving the farm environment, and improving farmers’ livelihoods. The WRCC provides the science and partnerships, as well as contributes to developing best management practices in support that form the basis for locally adapted versions.

IWMIm

provin

g farmers’

livel

ihood

Increasing farm

ers’profitab

ility

Preserving farm

environment

42

We have three objectives for the WRCC. Firstly, we strive to be leaders in the competence of weed resistance, understanding resistance to herbicides in depth, how resistance mechanisms work, how resistance evolves, and how it can be managed in each field, no matter the current status. Secondly, we want to take this knowledge and use it to develop and offer the best strategies and specific solutions for resistance management. Through our partners in the respective countries, we work to tailor these strategies to the individual fields of each farmer. And thirdly, we want to effectively communicate our knowledge and solutions. We are aware that leadership will bring the responsibility of supplying effective weed control products and programs, and we are ready to meet the challenge.

7.2 Mission and objectives

Just what do we do at the WRCC? We have a multifunctional research group with an overwhelming external focus engaging in projects in different countries and regions. We support our Weed Control Discovery group in the search for finding new weed control compounds with novel modes of action. We conduct the tried-and-true resistance diagnostics bioassays in the greenhouse with seeds harvested from plants in the field, which is still the best way to get a clear profile of the level of resistance and learn what products still work and which ones do not. We conduct a full range of advanced diagnostics tests in our Weed Resistance Research Laboratory using the latest biochemistry and molecular biology techniques. We develop strategies, diagnostic technologies, and management tools for IWM programs that support our outreach activities. This includes diverse things such as models that predict resistance evolution, training modules, integrated weed management strategies, and platforms for delivering recommendations to farmers. We supplement our internal capabilities through external collaborations with top researchers and institutes across the globe.

7.3 Key operational tasks

Please refer to our website http://www.iwm.bayer.com/competencies/wrcc for more details about the Bayer Weed Resistance Competence Center.

Our objectives at the WRCC

1. Understand resistance mechanisms and evolution

2. Develop and offer the best specific solutions and management strategies

3. Communicate its knowledge and solutions

43

8.0 References Ahmed S, Awan TH, Salim M, Chauhan BS 2015 The Journal of Animal and Plant Sciences, 25(6):1675-1684.

Andrew IKS, Storkey J, Sparkes DL 2015 A review of the potential for competitive cereal cultivars as a tool in integrated weed management. Weed Research 55:239–248.

Anon. 2015 Meriam-Webster Dictionary, online version, available online at http://www.merriam-webster.com/inter?dest=/dictionary/weed, accessed 31 March, 2015.

Awan TH, Cruz PCS, Chauhan BC 2016 Effect of pre-emergence herbicides and timing of soil saturation on the control of six major rice weeds and their phytotoxic effects on rice seedlings. Crop Protection, 83:37-47.

Bagavathiannan, M V, Norsworthy, J K, Smith, K L and Neve, P 2012 Seed production of barnyard grass (Echinochloa crus-galli) in response to time of emergence in cotton and rice. The Journal of Agricultural Science, 150:717-724.

Barber LT, Smith KL, Scott RC, Norsworthy JK, Vangilder AM 2015 Zero Tolerance: a community-based program for glyphosate-resistant Palmer amaranth management. University of Arkansas Cooperative Extension Service, Bulletin FSA2177-PD-3-2015N, 6 pp.

Barrett SH 1983 Crop mimicry in weeds. Economic Botany 37: 255-282.

Bayer, 2013 Observations from fields in Germany where studies on resistance evolution were being conducted. See Herrmann et al., 2014 and Hess et al., 2012.

Beckie HJ, Heap IM, Smeda RJ, Hall LM 2000 Screening for herbicide resistance in weeds. Weed Technology 14; 428-445.

Bhurer KP, Yadav DN, Ladha JK, Thapa RB, Pandey K 2013 Effect of integrated weed management practices on performance of dry direct seeded rice (Oryza sativa L.). Agronomy Journal of Nepal, 3:53-63.

Beltran JC, Pannell DJ, Doole GJ, White B 2012 A bioeconomic model for analysis of integrated weed management strategies for annual barnyard grass (Echinochloa crus-galli complex) in Philippine rice farming systems.

Brennan JP, Lemerle D, Martin P 2001 Economics of increasing wheat competitiveness as a weed control weapon. Contributed paper presented to the 45th annual conference of the Australian Agricultural and Resource Economics Society, available online http://ageconsearch.umn.edu/bitstream/125543/2/BrennanJ.pdf, 15 pp.

Broster J 2017 Are there opportunities for HWSC in irrigated cropping systems? Graham Centre for Agricultural Innovation, presentation, 22 pp.

Buhler DD, Kohler KA, Thompson RL 2001 Weed seed bank dynamics during a five-year crop rotation. Weed Technology, 15:170–176.

Busi R, Gaines TA, Walsh MJ, Powles SB 2012 Understanding the potential for resistance evolution to the new herbicide pyroxasulfone: field selection at high doses versus recurrent selection at low doses. Weed Research 52: 489–499.

Busi R, Gaines TA, Vila-Aiub MM, Powles SB 2014 Inheritance of evolved resistance to a novel herbicide (pyroxasulfone). Plant Science 217– 218 (2014) 127– 134.

Cardina J, Johnson GA, Sparrow DH 1997 The nature and consequence of weed spatial distribution. Weed Science 45:364-373.

Caton BP, Mortimer M, Hill JE, Johnson DE 2010 A practical field guide to weeds of rice in Asia, 2nd ed. Los Baños (Philippines): International Rice Research Institute, 118 p. Available as a download (http://books.irri.org/9789712202568_content.pdf).

Charudattan R 2001 Biological control of weeds by means of plant pathogens: Significance for integrated weed management in modern agro-ecology. BioControl 46: 229–260.

Chauhan BS, Abugho, SB 2012 Effect of growth stage on the efficacy of postemergence herbicides on four weed species of direct-seeded rice. The ScientificWorld Journal, Article ID 123071, 7 pp.

Chauhan BS, Mahajan G 2012 Role of integrated weed management strategies in sustaining conservation agriculture systems. Current Science, 103:135-136.

Chauhan BS 2011 Crowfootgrass (Dactyloctenium aegyptium) germination and response to herbicides in the Philippines. Weed Science 59:512–516.

44

Chauhan BS 2013 Management strategies for weedy rice in Asia. Crop Protection, 48:51-56.

Dass A, Shekhawat K, Choudhary AK, Sepat S 2017 Weed management in rice using crop competition-a review. Crop Protection, 95:45-52.

De Datta SK, Morris RA, and Barker R 1978 Land preparation and crop establishment for rainfed lowland rice. IRRI Research Paper Series 22, 24 pp.

De Datta SK 1981 Principles and practices of rice production. John Wiley & Sons, New York, Chap. 1 Rice in perspective, p. 1-8, Chap. 12 Weeds and their control, p. 460-512.

Délye C 2013 Unravelling the genetic bases of non-target-site-based resistance (NTSR) to herbicides: a major challenge for weed science in the forthcoming decade. Pest Management Science 69:176-187.

Délye C, Jasieniuk M, Le Corre V 2013 Deciphering the evolution of herbicide resistance in weeds. Trends in Genetics 29:649-58.

Devine M, Duke SO, Fedtke C 1993 Physiology of herbicide action. PTR Prentice-Hall, New Jersey, 441 pp.

Galinato MI, Moody K, Piggin CM 1999 Upland rice weeds of South and Southeast Asia. Makati City (Philippines): International Rice Research Institute. 156 p. Available as a download (http://books.irri.org/9712201309_content.pdf).

Giacomini D, Westra P, Ward SM 2014 Impact of genetic background in fitness cost studies: an example from glyphosate-resistant Palmer amaranth. Weed Science 62:29-37.

Gomez KA 2001 Rice, the grain of culture. Thai Rice Foundation under Royal Patronage, 10 pp.

Hantsch L, Bruelheide H, Erfmeier A 2013 High phenotypic variation of seed traits, germination characteristics and genetic diversity of an invasive annual weed. Seed Science Research 23:27–40.

Harker KN, O’Donovan JT 2013 Recent weed control, weed management, and integrated weed management. Weed Technology 2013 27:1–11.

Heap I 2017 The International Survey of Herbicide Resistant Weeds. Accessed 21 July 2017, available: www.weedscience.org

Herrmann J, Hess M, Schubel T, Strek H, Richter O, Beffa R 2014 Spatial and temporal development of ACCase and ALS resistant Black-grass (Alopecurus myosuroides Huds.) populations in neighboring fields in Germany. Julius-Kühn-Archiv 443:273–279.

Hess M, Beffa R, Kaiser J, Laber B, Menne H, Strek H 2012 Status and development of ACCase and ALS resistant blackgrass (Alopecurus myosuroides Huds.) in neighboring fields in Germany. Julius-Kühn-Archiv 434:163–170.

HRAC 2010 Herbicide Resistance Action Committee - The world of herbicides – according to HRAC classification on mode of action 2010, poster, available online at: http://www.hracglobal.com/pages/world%20of%20herbicides%20map.aspx

HRAC 2017 Herbicide Resistance Action Committee, available online at: http://hracglobal.com/, accessed on 24 Jul 2017.

IRRI 2017 Rice knowledge bank: Your information source for rice farming, Herbicides, available online at http://www.knowledgebank.irri.org, accessed on 24 Jul 2017.

Ismail AM, David E. Johnson DE, Ella ES, Vergara GV, Baltazar AM 2012 Adaptation to flooding during emergence and seedling growth in rice and weeds, and implications for crop establishment. AoB PLANTS, document pls019, 18 pp.

Iwakami S, Uchino A, Watanabe H, Yamasue Y, Inamura T 2012 Isolation and expression of genes for acetolactate synthase and acetyl-CoA carboxylase in Echinochloa phyllopogon, a polyploid weed species. Pest Management Science 68: 1098–1106.

Joshi E, Kumar Lal B, Nepalia V, Gautam P, Vyas AK 2013 Management of direct seeded rice for enhanced resource – use efficiency. Plant Knowledge Journal, 2:119-134.

Khan AR, Srivastava RC, Ghorai AK, Singh SR 2003 Efficient soil solarization for weed control in the rain-fed upland rice ecosystem. International Agrophysics, 17:99–103.

Kirchhof G, Priyono S, Utomo WH, Adisarwanto T, Dacanay EV, So HB 2000 The effect of soil puddling on the soil, physical properties and the growth of rice and post-rice crops. Soil and Tillage Research, 56:37-50.

Korres NE, Norsworthy JK 2015 Influence of a rye cover crop on the critical period for weed control in cotton. Weed Science 63:346–352.

Kraehmer H, Jabran K, Mennan H, Chauhan BS 2015 Global distribution of rice weeds - A review. Crop Protection 80:73-86.

45

Kraehmer H, Echle J, Schreiber B, Ueno C, Sato Y, Park M 2017 The role of industry in managing rice weeds in the Asian Pacific Region during the past, present and in future. In press in: Rao AN, Matsumoto H (eds.) 2017 Weed management in rice in the Asian-Pacific Region., 30 pp. Asian-Pacific Weed Science Society (APWSS), Indian Society of Weed Science (ISWS), India and The Weed Science Society of Japan (WSSJ), Japan.

Kumar V, Bellinder RR, Brainard DC, Malik RK, Gupta RK 2008 Risks of herbicide-resistant rice in India: A review. Crop Protection 27:320–329.

Labrada R 2006 Recommendations for improved weed management. FAO, Plant Production and Protection Division, Rome, Italy, 50 pp.

Lu B-R, Yang C 2009 Gene flow from genetically modified rice to its wild relatives: Assessing potential ecological consequences. Biotechnology Advances, 27:1083–1091.

Mahajan G, Chauhan BS, Gill MS 2013 Dry-seeded rice culture in Punjab State of India: Lessons learned from farmers. Field Crops Research, 144:89–99.

Mahajan G, Chauhan BS, Kumar V 2014 Integrated Weed Management in Rice. Chapter 6 in: Chauhan, B.S., Mahajan, G. (eds.), Recent Advances in Weed Management. Springer Science+Business Media, New York, USA, pp. 125-153.

Mal TK, Lovett-Doust J 2005 Phenotypic plasticity in vegetative and reproductive traits in an invasive weed, Lythrum salicaria (Lythraceae), in response to soil moisture. American Journal of Botany 92:819-25.

Manalil S, Busi R, Renton M, Powles SB 2011 Rapid evolution of herbicide resistance by low herbicide dosages. Weed Science 2011 59:210–217.

Menne H, Köcher H 2007 HRAC Classification of herbicides and resistance development. In: Krämer W, Shirmer U, eds. Modern crop protection compounds. Wiley-Vch, pp 5-27.

Menne H, Köcher H 2012 HRAC Classification of herbicides and resistance development. In: Krämer W, Shirmer U, Jeschke P, Witschel M, eds. Modern crop protection compounds, 2nd ed. Wiley-VCH Verlag GmbH & Co. KGaA, pp 5-28.

Neve P, Powles SB 2005 Recurrent selection with reduced herbicide rates results in the rapid evolution of herbicide resistance in Lolium rigidum. Theoretic and Applied Genetics 110:1154-1166.

Norsworthy JK, Ward SM, Shaw DR, Llewellyn RS, Nichols RL, Webster TM, Bradley KW, Frisvold G, Powles SB, Burgos NR, Witt WW, Barrett M 2012 Reducing the Risks of herbicide resistance: best management practices and recommendations. Weed Science 2012 Special Issue:31–62.

Norsworthy JK, Bond J, Scott RC 2013 Weed management practices and needs in Arkansas and Mississippi rice. Weed Technology 27:623–630.Norsworthy JK, Griffith G, Griffin T, Bagavathiannan M, Gbur EE 2014a In-field movement of glyphosate-resistant Palmer amaranth (Amaranthus palmeri) and its impact on cotton lint yield: Evidence supporting a zero-threshold strategy. Weed Science, 62:237-249.

Norsworthy JK, Wilson MJ, Scott RC, Gbur EE 2014b Herbicidal activity on acetolactate synthase-resistant barnyard grass (Echinochloa crus-galli) in Arkansas, USA. Weed Biology and Management, 14:50–58.

Phillips McDougall 2016 The cost of new agrochemical product discovery, development and registration. Midlothian, Scotland, UK, CLA / ECPA / CropLife R&D Survey, 13 pp.

Powles SB, Yu Q 2010 Evolution in action: plants resistant to herbicides. Annual Review of Plant Biology 61: 317-347.

Powles SD 2014 Personal communication.

Rao AN, Johnson DE, Sivaprasad B, Ladha JK; Mortimer AM 2007 weed management in direct-seeded rice. Advances in Agronomy, 93:153-255.

Ray TB 1984 Site of action of chlorsulfuron inhibition of valine and isoleucine biosynthesis in plants. Plant Physiology, 75:827-831.

Sanchez PA 1976 Properties and Management of Soils in the Tropics. Wiley, New York, 618 pp.

Schwartz LM, Norsworthy JK, Barber LT, Scott RC 2016 Harvest weed seed control – An alternative method for measuring the soil seedbank. University of Arkansas System, Division of Agriculture, Research & Extension, Agriculture and Natural Resources Extension Bulletin FSA2180PD12016N, 4 pp.

Shrestha A 2015 Know your weeds. Division of Agriculture and Natural Resources, University of California, UC Cooperative Extension, Fresno County, online version, accessed 01 April, 2015, available: http://ucanr.edu/sites/Weed_Management/files/138789.pdf

Singh M, Bhullar MS, Chauhan BS 2014 The critical period for weed control in dry-seeded rice. Crop Protection 66:80-85

46

Singh S, Singh G, Singh VP, Singh AP 2005 Effect of establishment methods and weed management practices on weeds and rice in rice-wheat cropping system. Indian Journal of Weed Science, 37:51-57.

Singh Y, Singh G 2008 Cropping systems and weed flora of rice and wheat in the Indo-Gangetic Plains. p. 33-43 in: Direct seeding of rice and weed management in the irrigated rice-wheat cropping system of the Indo-Gangetic Plains, eds., Singh Y, Singh VP, Chauhan B, Orr A, Mortimer AM, Los Baños (Philippines): International Rice Research Institute, and Pantnagar (India): Directorate of Experiment Station, G.B. Pant University of Agriculture and Technology. 272 p

Smith RJ, Shaw WC 1966 Weeds and their control in rice production. Agricultural Research Service, United States Department of Agriculture, Washington, DC, Agriculture Handbook No. 292, 64 pp.

Stoller EW, Sweet RD 1987 Biology and life cycle of purple and yellow nutsedges (Cyperus rotundus and C. esculentus). Weed Technology, 1:66-73.

Sweeny M, McCouch S 2006 The complex history of the domestication of rice. Annals of Botany, 100:951–957.

Tranel PJ, Riggins CW, Bell MS, Hager AG 2011 Herbicide resistances in Amaranthus tuberculatus: a call for new options. Journal of Agricultural & Food Chemistry 59:5808-5812.

UN 2014 World Urbanization Prospects: The 2014 Revision, Highlights (ST/ESA/SER.A/352). United Nations, Department of Economic and Social Affairs, Population Division, 32 pp.

US Census Bureau 2017 International Data Base, updated Aug 2016, available online at: https://www.census.gov/population/international/data/idb/informationGateway.php, accessed online 12 Apr 2017.

USDA 2017 Foreign Agricultural Service, Production, Supply, & Distribution Online Data Base, projected market year production on a milled basis updated 13 Feb 2017, accessed online on 09 Mar 2017 at http://www.fas.usda.gov/psdonline/psdHome.aspx.

Valverde BE, Itoh K 2001 World rice and herbicide resistance. In Powles SB, Shaner DL, eds., Herbicide resistance and world grains, CRC Press, New York, p. 195-249.

Vila-Aiub MM, Neve P, Powles SB 2009 Fitness costs associated with evolved herbicide resistance alleles in plants. New Phytologist 184: 751–767.

von Braun J, Bos MS 2005 The changing economics and politics of rice: implications for food security, globalization, and environmental sustainability. In Rice is Life – p. 7-20.

Vongsaroj P, Tjitosemito S 1993 Integrated management of paddy weeds in Thailand. ASPAC Food & Fertilizer Technology Center. 13 pp., available online at, http://www.fftc.agnet.org/htmlarea_file/library/20110721175022/eb367a.pdf, accessed 27 Apr 2017.

Walsh M, Newman P, Powles S 2013 Targeting Weed seeds in-crop: a new weed control paradigm for global agriculture. Weed Technology 27:431–436.

Wang T, Picard JC, Tian X, Darmency H 2010 A herbicide-resistant ACCase 1781 Setaria mutant shows higher fitness than wild type. Heredity 105:394–400.

WSSA 1998 "Herbicide resistance" and "herbicide tolerance" defined. Weed Technology, Notes, 12:789.

WSSA 2011 Training module “How Herbicides Work, Lesson 2”, available online at http://wssa.net/wp-content/uploads/resistancemodules/two/index.htm, accessed 26 Jul 2017.

WSSA 2014 Herbicide Handbook, 10th ed. Weed Science Society of America, 500 pp.

Yamada S, Okubo S, Kitagawa Y, Takeuchi K 2007 Restoration of weed communities in abandoned rice paddy fields in the Tama Hills, central Japan. Agriculture, Ecosystems and Environment, 119:88-102.

Yu Q, Han H, Cawthray GR, Wang SF, Powles SB 2013 Enhanced rates of herbicide metabolism in low herbicide-dose selected resistant Lolium rigidum. Plant, Cell and Environment 36:818–827.

Yu Q, Powles SB 2014 Resistance to AHAS inhibitor herbicides: current understanding. Pest Management Science 70:1340–1350.

Zimdahl RL 2007 Fundamentals of weed science. Academic Press, 666 pp.

* All tables, figures, and photos credited to Bayer, 2017 unless otherwise noted.

47

For more information about Bayer’s Integrated Weed Management activities please visit us online:www.iwm.bayer.com

Bayer AG Crop Science Division Alfred-Nobel-Str. 50 D - 40789 Monheim am Rhein Germany

www.cropscience.bayer.com

© Bayer · August 2017