Previous PageTable Of ContentsNext Page

Stability of competitive ability in wheat genotypes across different weed species

Michael Zerner and Gurjeet Gill

School of Agriculture, Food and Wine, The University of Adelaide, Waite Campus, Glen Osmond SA 5064. Email michael.zerner@adelaide.edu.au

Abstract

In the field, weeds are invariably present as a mixture of different species. Therefore, in developing crops with improved competitive ability (CA) with weeds, it is critical to determine the stability of CA against different weed species. During 2009, 14 wheat genotypes varying in CA were grown in weed-free and weedy plots to assess their tolerance to and suppression of three weed species, annual ryegrass (Lolium rigidum), oat (Avena sativa) and mustard (Brassica juncea). The suppression of weed seed production across the three weed species was strongly correlated (r= 0.86-0.90), indicating consistent weed species response to wheat genotypes. As compared to the cultivar Wyalkatchem, the most suppressive wheat genotype reduced weed seed production by up to 92% (mustard), 65% (ryegrass) and 53% (oat). The degree of weed tolerance of wheat genotypes varied with weed species, with grain yield losses ranging between 3-21% in mustard, 18-49% in ryegrass and 39-63% in oat. However, the ranking of wheat genotypes against the three weed species was maintained, as indicated by the significant correlations between wheat grain yield losses caused by each of the weeds (r = 0.58-0.77). There was a positive correlation between weed tolerance and suppression for each of the three weeds, ryegrass (0.79), oat (0.72) and mustard (0.59), suggesting that a genotype with high tolerance to weeds was also likely to have high suppressive ability.

Key Words

Integrated weed management, weed suppression,

Introduction

Weed management in many cropping systems in Australia has been heavily reliant on the use of selective herbicides. However, rapid evolution of herbicide resistance across multiple modes of actions and weed species has forced many growers in southern Australia to consider adoption of more integrated strategies for weed population management. In some districts, weeds such as annual ryegrass can no longer be controlled with selective post-emergence herbicides, thus further increasing reliance on pre-emergent herbicides. In such cropping systems, weeds that escape pre-emergent herbicide application can produce large amounts of seed due to poor competitive ability of current wheat cultivars. Integration of weed competitive crop cultivars with other control techniques such as crop-topping and wind-row burning has the potential to significantly improve effectiveness of weed management programs.

Wheat is known as a poor competitor enabling weeds surviving various control tactics including herbicides to set a large amount of seed (Lemerle et al. 1995). Therefore improvements in the competitive ability of wheat could reduce weed seed set and the dependence on herbicides in the cropping rotation. As farmer fields are often infested with mixed weed flora, such competitive cultivars would need to provide suppression of a range of weed species. In order to address this research question, a field study was undertaken to investigate the competitive ability of high vigour wheat lines against ryegrass, oats and mustard.

Methods

A field experiment was undertaken at Roseworthy, South Australia during 2009 to investigate the stability of competitive ability in wheat across different weed species. Fourteen wheat lines varying in competitive ability were used including nine breeding lines that have increased early vigour, an important plant trait in competitiveness (Seavers and Wright 1999). Ryegrass (Lolium rigidum), cultivated oat (Avena sativa) and oriental mustard (Brassica juncea) were used as the weed species in this study. The weeds were spread on the soil surface prior to the wheat being sown. All weeds were sown in monocultures, ryegrass was sown at 250 plants m-2, oat (weed mimic for wild oat) at 60 plants m-2 and mustard (weed mimic for wild radish) at 30 plants m-2. These densities were chosen to target 50% yield reduction in commercial wheat cultivars (Poole and Gill 1987; Streibig et al. 1989). Wheat was sown using knife points on 25cm row spacings. Plots were 3 by 1.5m and wheat seeding rates were adjusted according to the seed size of each line, targeting a density of 200 plants m-2. All wheat lines were also sown without any weeds and with each of the three weed species in separate plots in a strip-plot design.

Plant density was recorded for both wheat and all weed species. Just prior to grain harvest within the plots containing weeds, spike density was recorded for ryegrass, panicle density for oats and plant density for mustard. Ryegrass seed set was calculated by randomly sampling plants and measuring the length and seed number of each spike. Seed number was then plotted against spike length and used as a calibration equation for ryegrass seed set, y=7.7356x, where y= seed number and x= average spike length (r2= 0.93). Ryegrass spikes were then randomly sampled from each plot to determine the average length, and using the equation in combination with spike density, ryegrass seed set was calculated. Mustard seed set was measured by collecting all plants in each plot at maturity and after cleaning the seed it was weighed and grain size recorded to estimate seed production. Oats were harvested with the wheat using a plot harvester and later separated from the wheat using a seed grading machine. Once separated oats were weighed and grain weight recorded for an estimation of seed production.

The experiment was blocked in three replications and within each block plots were randomly allocated to each wheat line. Then each wheat plot was further randomly sub-divided into four strips within each block for each weed and a weed-free strip. All statistical analyses were completed using the GENSTAT statistical analysis software.

Results

Wheat grain yields varied significantly between the lines. In the plots containing no weeds yields ranged from 4.7t/ha to 2.7t/ha, averaging 3.5t/ha across all wheat lines (Table 1).

Table 1. Effect of wheat genotypes varying in competitive ability on grain yield and weed seed-set in the presence of annual ryegrass (ARG), mustard (MUS) and oat (OAT). (WF denotes crop in weed-free environment)

Line ID

Wheat grain yield

(kg/ha)

Grain yield loss

(%)

Weed seed production (seed/m2)

ARG

MUS

OAT

WF

ARG

MUS

OAT

ARG

MUS

OAT

Espada

2999

3978

2270

4497

30.7

10.1

47.4

19201

2379

1088

Frame

2218

2921

1894

3031

27.0

4.3

39.0

13715

1986

1079

Halberd

2167

3043

1479

3433

36.4

11.7

57.0

14202

1319

1392

Janz

2196

3230

1487

3653

37.6

10.3

58.3

23824

2473

1737

WCD1-0375

2779

3383

2014

3945

27.7

13.0

48.0

11898

1916

1423

WCD1-1462

2601

3285

1918

3542

28.2

7.8

46.8

13780

781

1088

WCD1-2204

2271

2629

1684

2968

23.5

11.5

43.8

14043

1194

1033

WCD2-110615

2557

3303

1787

4057

34.6

16.6

54.6

16012

1661

854

WCD2-280504

2826

3096

1742

3467

18.0

10.1

48.9

9961

819

828

WCD2-390403

2390

3012

1868

3438

30.5

12.1

45.9

9145

233

820

WCD2-400107

1826

2603

1276

2722

31.3

2.7

51.0

12148

1136

852

WCD2-400203

2325

2674

1720

3065

25.3

14.0

44.7

10030

1517

838

WCD2-420403

2085

2837

1329

3603

41.9

20.9

62.9

32262

5884

2245

Wyalkatchem

2419

3750

2054

4676

48.6

20.6

56.9

25797

2975

1708

                     

Mean

2404

3125

1752

3578

32

12

50

16144

1877

1213

                     

P values

                   

Line ID

*** (555)

*** (8.25)

*** (2041)

Weed

*** (335)

*** (3.72)

*** (920)

Line ID.Weed interaction

*** (647)

ns

*** (3473)

*** P<0.001, values in parenthesis represent LSD

Most of the commercial varieties out-yielded the experimental high vigour lines. In the presence of weeds yields were significantly reduced in all lines except in the presence of mustard, where yields were not significantly lower than the weed-free treatment for many lines. Weed competition pressure was greatest under oats, with yield penalties, ranging from 39-63%. Yield losses from ryegrass ranged between 18-49% and between 3-21% for mustard. Despite actual grain yield losses varying significantly between weed species, wheat lines ranked similarly between weeds. This is indicated by significant positive correlations between yield losses caused by each of the three weeds, with r values ranging between 0.58 and 0.77, shown in Figure 1. This relationship indicates that despite changing the weed species, a wheat line showing high tolerance to ryegrass will also have high tolerance to oats and mustard. Wyalkatchem suffered consistently severe yield losses across all weed species. It is difficult to identify any wheat line with consistently high weed tolerance due to considerable variation in yield losses. The other component of competitive ability, weed suppression provided a much more consistent pattern of performance of wheat lines across the three weed species.

a.(i)

a.(ii)

b.(i)

b.(ii)

c.(i)

c.(ii)

Figure 1. Relationships for (i) weed seed production and (ii) wheat yield losses between three weed species used in this study, (a) ryegrass and mustard, (b) ryegrass and oat, and (c) oat and mustard. Significant correlations also were found between weed seed production and wheat yield loss in each weed species with r values of 0.79 in ryegrass, 0.72 in oat and 0.59 in mustard.

Weed seed production varied between weeds due to inherent differences between them in potential seed set. On average ryegrass produced 16000 seeds/m2, compared to 1900 seeds/m2 and 1200 seeds/m2 for mustard and oat respectively. Rankings of lines based on weed suppression was much more consistent between weeds than weed tolerance, which is supported by a significant interaction between wheat line and weed species as well as highly significant correlations of seed production between each of the weeds (Figure 1). (r values: ARG/MUS=0.90, ARG/OAT=0.88, OAT/MUS=0.86). These results clearly demonstrate that the CA of a specific wheat line is consistent across weed species, whether it’s a grass or broad-leaf species. Although the high vigour breeding lines did not show considerable advantages in terms of crop tolerance, they showed superior weed suppression as compared to the commercial cultivars used in this study. These results indicate that wheat lines with improved CA used in this study may not provide greater economic returns in the presence of weeds through achieving lower yield losses. However, these weed competitive wheat lines will assist in reducing the weed burden in future crops and could reduce dependence on herbicides. Wyalkatchem allowed for the highest weed seed set, with the exception of the extreme dwarf line, WCD2-420403 (Table 1). WCD2-390403 was consistently the most competitive breeding line with lowest weed seed production in each of the weed species. This high vigour lines reduced weed seed production by 92%, 65% and 53% in mustard, ryegrass and oat when compared to seed production of these species grown in competition with Wyalkatchem.

The crop yield loss (weed tolerance) and weed seed production (weed suppression) have been shown to be positively correlated in some studies (Lemerle et al. 2006), but they may also have no or a very weak relationship (Vandeleur and Gill 2004). This study showed strong positive correlations between weed tolerance and suppression for each of the three weeds with r values ranging from 0.79 in ryegrass to 0.72 and 0.59 in oats and mustard (Figure 1). These results indicate that the wheat lines expressing high weed suppression will also provide greater levels of weed tolerance (yield maintenance), which is highly desirable for efforts to breed crop cultivars with improved competitive ability.

Conclusion

The CA of wheat measured in terms of weed seed production was highly stable across three weed species. This trend was reflected in significant (positive) correlations between each of the weeds. There was also a strong positive correlation between weed suppression and weed tolerance which is desirable for breeding efforts to develop cultivars with increased competitive ability as they must be able to compete with multiple weed species as weeds exist as mixed populations in field situations.

References

Lemerle D, Verbeek B and Coombes NE (1995). Losses in grain yield of winter crops from Lolium rigidum competition depend on cultivar and season. Weed Research 35, 503-509.

Lemerle D, Smith A, Verbeek B, Koetz E, Lockley P and Martin P (2006). Incremental crop tolerance to weeds: A measure for selecting competitive ability in Australian wheats. Euphytica 149, 85-95.

Poole ML and Gill GS (1987). Competition between crops and weeds in southern Australia. Plant Protection Quarterly 2, 86-96.

Seavers GP and Wright KJ (1999). Crop canopy development and structure influence weed suppression. Weed Research 39, 319-328.

Streibig JC, Combellack JH, Pritchard GH and Richardson RG (1989). Estimation of thresholds for weed control in Australian cereals. Weed Research 29, 117-126.

Vandeleur RK and Gill GS (2004). The impact of plant breeding on the grain yield and competitive ability of wheat in Australia. Australian Journal of Agricultural Research 55, 855-861.

Previous PageTop Of PageNext Page