Previous PageNext Page


Deirdre Lemerle1, Robert Blackshaw2, Trent Potter3, Stephen Marcroft4 and Robert Barrett-Lennard5

1 NSW Agriculture and CRC for Weed Management Systems, Agricultural Institute, PMB, Wagga Wagga, NSW 2650, Australia
Agriculture and Agri-Food Canada, Box 3000, Lethbridge, AB T1J4B1, Canada
South Australian Research and DeveIopment Institute, PO Box 618 Naracoorte, SA 5271
Department of Natural Resources, Victorian Institute of Dryland Agriculture, PMB 260, Horsham Vic 3401, Australia
University of Western Australia and CRC for Weed Management Systems, PMB, Nedlands WA 6009, Australia


A random survey of the incidence and abundance of weeds in 62 canola crops was carried out in Spring 1998 across the wheat-belt of southern Australia, mainly in eastern Australia. The frequency of weeds in triazine-tolerant (TT) canola compared to conventional cultivars was determined. Canola maturity ranged from six leaves to early flowering, when most weed management practices were already completed by farmers. Thirty-seven percent of the crops were TT cultivars. A total of 73 weed species were identified, some at densities of up to several hundred plants per m2. The most widespread grass species were Lolium rigidum (annual ryegrass), Avena spp. (wild oats), and Vulpia spp. (silver grass), while the main broadleaf weeds were Arctotheca calendula (capeweed), Polygonum aviculare (wireweed), and Fumaria spp. (fumitory). In TT canola, the average number of weeds was 49 plants per m2 (range of 0-235 plants per m2), compared with 72 plants per m2 (range of 1-358 plants per m2) in the conventional cultivars. The implications of weeds surviving current management practices are discussed, especially in relation to herbicide resistance and the long-term reduction of weed seedbanks in the soil.

KEYWORDS: Weed density, Lolium rigidum, Arctotheca calendula, Polygonum aviculare, Raphanus raphanistrum, Triazine-tolerant.


Canola is a rapidly expanding crop in the southern wheat-belt of Australia. A total of 1.19 million ha was grown in 1998, and this area is conservatively expected to rise to 1.65 million ha in 1999 (Canola Association of Australia). Canola is more profitable than cereals and provides growers with the opportunity to use more diverse cropping rotations.

Herbicide resistance in weeds is a major problem for crop production in Australia, especially in Lolium rigidum (annual ryegrass). Resistance to herbicides has developed to a lesser extent in other species, such as Avena spp. (wild oats), Raphanus raphanistrum (wild radish) Sisymbrium orientale (Indian hedge mustard), and Arctotheca calendula (capeweed), but is expected to become more widespread in the future unless radical changes in weed management practices are adopted.

Triazine-tolerant (TT) canola cultivars provide greater cost-effective opportunities for growers to manage more weeds, particularly Raphanus raphanistrum, Fumaria spp. (fumitory) and Capsella bursa-pastoris (shepherd’s purse), than conventional cultivars. As a result, there has been rapid adoption of TT canola in Australia, especially where Raphanus raphanistrum occurs. In 1998, it is estimated that the areas of TT canola grown were 90% in Western Australia (WA) because Raphanus raphanistrum is a major problem, and 25-30% in South Australia (SA), Victoria (VIC) and New South Wales (NSW) (N. Wratten, personal communication). The area of production of TT canola is expected to continue to rise in Australia, despite the associated penalties of lower grain yield and oil content, lower resistance to blackleg, and the persistence of triazine herbicides in the soil. TT cultivars are also more flexible for direct-drilled crops.

The survey reported here aimed to determine the incidence and abundance of weeds species in canola crops across the southern wheat-belt of Australia. This included a range of conventional and TT canola crops in NSW, VIC, SA and WA. The aim also was to compare the numbers of plants that survived the weed management practices in TT canola and conventional canola.


The survey was conducted in August 1998. Sixty-three canola crops were selected at random, thirty-two sites in NSW, fourteen in VIC, nine in SA and eight in WA. One of the sites (a TT crop in WA) had over 743 Lolium rigidum plants per m2 and was clearly atypical due to a high level of triazine resistance or missed herbicide application. The only other weeds at this site were Bromus spp. and Raphanus raphanistrum. This site was therefore not included in the survey. Soils ranged from red-brown earths in NSW and VIC, to sand on clays in WA, while the annual average rainfall of a site ranged from 375 mm in WA to 600mm in parts of eastern Australia.

At each crop, weed species and plant density were recorded in twenty 1 m2 quadrats. Five quadrats were spaced 50 m apart along each arm of a W-transect of the field. The canola growth stage and cultivar, and the soil type were recorded at each site. The percentage incidence of a weed species was calculated using the mean of the 20 positions on the transect for each species over the 62 sites. The average weed plant density of each species was calculated for each site.


Canola maturity at the time of assessment ranged from the six-leaf stage to flowering. Twenty-three of the sixty-two crops or 37% were TT cultivars, 31% of these crops were in NSW, 36% in VIC, 11% in SA and 100% in WA. The high adoption of TT canola in WA reflects the need to control widespread and dense infestations of Raphanus raphanistrum. In WA, Raphanus raphanistrum was found at 57% of all sites surveyed, compared with being at 12% of the NSW sites, 22% of SA sites, 7% of VIC sites.

A total of 73 weed species were identified in the survey. Total weed numbers of all species at a site, ranged from 0-358 plants per m2. The three most widespread species, Lolium rigidum, Arctotheca calendula and Polygonum aviculare (wireweed), were present at more than 50% of the sites (Table 1). Other widespread species that were present in more than 15% of fields are shown in Table 1. The incidence of grass weeds found in canola was less than recorded in a previous survey of cereals in south-eastern Australia (Lemerle et al. 1996), however, the distribution of broadleaf weeds in canola was greater. There are less selective herbicide options available for broadleaf weeds in canola.

In TT canola, the average number of weeds was 49 per plants per m2 (range of 0-235 plants per m2), compared with 72 plants per m2 (range of 1-358 plants per m2) in the conventional cultivars.

Table 1 The percentage incidence of the most widespread weed species recorded at 62 sites, and the average weed density (plants per m2) in TT and conventional canola

Weed species

Incidence at field sites (%)

Proportion of TT canola (as a % of total sites where the weed occurred)

Average weed density in TT canola (plants per m2)

Average weed density in conventional canola (plants per m2)

Lolium rigidum (annual ryegrass)





Arctotheca calendula (capeweed)





Polygonum aviculare (wireweed)





Avena spp. (wild oats)





Fumaria spp.(fumitory)





Vulpia spp. (silver grass)





Hordeum spp. (barley grass)





Sisymbrium orientale (Indian hedge mustard





Capsella bursa-pastoris (shepherd’s purse)





Raphanus raphanistrum (wild radish)





Some weeds such as Lolium rigidum, Hordeum spp. and Raphanus raphanistrum were present in a large proportion of the TT crops (Table 1). In contrast, other species (eg Capsella bursa-pastoris and Fumaria spp) were less prevalent in the TT crops. It is surprising that densities of some weed species were similar or even greater (eg Sisymbrium orientale and Polygonum aviculare) in the TT crops, compared with the conventional cultivars. In contrast, Arctotheca calendula, Vulpia and Capsella bursa-pastoris had lower in the TT crops.

There is a perception that TT-canola will provide a ”magic bullet” weed control solution for growers. However, this evidence shows that weeds survive even the TT canola weed control practices, thus producing seed and replenishing the seedbank in the soil, and exacerbating subsequent weed infestations. The survival of such large numbers of weeds will facilitate the development of herbicide resistance in these weed species. This is of particular concern given the suspected discovery of a population of Raphanus raphanistrum in WA resistant to triazine herbicides in 1998, with a further 53 populations with confirmed resistance to Group B herbicides (A Cheam, personal communication.).


A surprisingly large number of weeds were found in the survey. The results confirmed the distribution of problem weeds across the southern wheat-belt and showed that TT canolas, while reducing the problem in some species, are not removing weed problems in canola.


We thank Steve Sutherland and Chris Watson for assisting with the survey. Financial support was provided by the Grains Research and Development Corporation of Australia.


1. Lemerle, D., Yuan, T.H, Murray, G.M. and Morris, S. (1996). Survey of weeds and diseases in cereal crops in the southern wheat-belt of New South Wales. Australian Journal of Experimental Agriculture 36,545-54.

Previous PageTop Of PageNext Page