| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
a Dep. of Agricultural, Food and Nutritional Science, Univ. of Alberta, Edmonton, AB, T6G 2P5
b Dep. of Plant Sciences, Univ. of Saskatchewan, 51 Campus Dr., Saskatoon, SK, S7N 5A8
c Agriculture and Agri-Food Canada, Lacombe Research Centre, 6000 C & E Trail, Lacombe, AB, T4L 1W1
* Corresponding author (dean.spaner{at}ualberta.ca)
Received for publication September 15, 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: ERS, Edmonton Research Station • GMOs, genetically modified organisms
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
In Canada, competition with weeds on conventional land has reduced crop yields significantly; with documented losses from 16 to 29% in barley (Didon and Bostrom, 2003; Harker, 2001; Scursoni and Satorre, 2005), and 8 to 63% in wheat (Hucl, 1998; Kirkland and Hunter, 1991). The three most abundant weed species in conventional spring wheat production fields in Alberta are wild buckwheat (Polygonum convolvulus L.), wild oat (Avena fatua L.), and chickweed (Stellaria media L.) (Leeson et al., 2002). In organic cereal production fields, greater weed species diversity and higher weed populations have been reported (Leeson et al., 2000; Samuel and Guest, 1990), with wild mustard (Sinapis arvensis L.) and Canada thistle (Cirsium arvense L.) the most problematic weed species (Entz et al., 2001).
Organic production systems must have reliable nonchemical weed control methods to maximize returns (Jordan, 1993; Lemerle et al., 1996). Weed control may be accomplished by using various tillage regimes (Barberi et al., 2000), crop rotations and intercrops (Hartl, 1989), changes to crop seeding density (Korres and Froud-Williams, 2002), and the use of competitive cultivars (Huel and Hucl, 1996; Lemerle et al., 1996).
Generally, barley has been found to be more competitive than wheat (Cousens, 1996; Fischer et al., 2000; O'Donovan et al., 1985; Pavlychenko and Harrington, 1934). In addition, there are differences in the competitive ability of genotypes or cultivars of both wheat and barley crops (Huel and Hucl, 1996; Lemerle et al., 1996; O'Donovan et al., 2000; Wicks et al., 1986). The competitive ability of a crop or a plant cultivar may be due to (i) tolerance to weed pressure by maintaining grain yield (crop competitive response), and/or (ii) the ability to suppress weed growth (crop competitive effect) (Coleman et al., 2001; Goldberg and Landa, 1991). Both are important since yield stability and the prevention of weed seed production (and subsequent seed bank build-up) are desirable in crops growing in association with weeds (Jordan, 1993). When considering crop competitive ability, weed tolerance and weed suppression need to be considered separately, because they may or may not occur together (Jordan, 1993).
Morphological, physiological, and biochemical traits are thought to control plant competitiveness (Lemerle et al., 2001). Plant height, tillering capacity, canopy structure, light interception, early biomass accumulation, ground cover, flag leaf length, and timing of spike emergence have been found to contribute to competitiveness (Champion et al., 1998; Hucl, 1998; Huel and Hucl, 1996; Korres and Froud-Williams, 2002; Lemerle et al., 1996; Wicks et al., 1986). From an agronomic perspective, increases in seeding density have resulted in higher levels of weed suppression and increased yields in wheat (Champion et al., 1998; Lemerle et al., 1996; Weiner et al., 2001) and barley (O'Donovan et al., 1999). Korres and Froud-Williams (2002) concluded that altering crop density was a more reliable tool than cultivar selection to reduce weed–crop competition.
The identification of plant traits that improve competitive ability may help crop breeders develop competitive crop cultivars. Choosing cultivars and management techniques that increase competitive ability will help growers to maximize production. The objectives of the present study were to determine the effect of cultivar and seeding rate on the competitive ability and agronomic performance of Canadian spring wheat and barley cultivars grown under organic management.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
330 mm, with 185 mm of precipitation at ERS and 80 mm at New Norway. Rainfall during the 2004 growing season totaled 367 mm at ERS and 281 mm at New Norway. All trials were planted in late May and harvested in early to mid-September.
Trials did not receive any applications of chemical fertilizer or herbicide, and were managed in accordance with the Organic Crop Improvement Association International Certification Standards (Organic Crop Improvement Association, 2000). Edmonton sites were situated on a section of land at the ERS that was first designated to be organically managed in the spring of 2001. The 2003 organically managed trial followed a cereal plowdown and an application of composted dairy manure at a rate of 60 Mg ha–1. The 2004 organically managed trial followed a cereal–legume rotation and an application of composted dairy manure at a rate of 60 Mg ha–1. Composted dairy manure was estimated to be 50% dry matter content, with 1.3% total N. At the certified organic farm, experimental trials followed cereal–legume plowdowns without crop removal in the year before planting.
Data Collection
After stem elongation was complete, plant height (representing the distance from the soil surface to the tip of the spike, excluding awns in awned cultivars) was recorded on a per-plot basis. Maturity was recorded as the day when 75% of the spikes and peduncles in the plot were tan brown, estimated to be approximately 30% seed moisture content. At maturity, all spikes in either a 1-m2 (2003) or 1-m row (2004) section of each plot were counted and used to calculate spikes m–2.
In 2003, dry weed biomass in each plot was determined from the aboveground portion of weeds from the harvested 1 m2. In 2004, dry weed biomass in each plot was determined by harvesting the aboveground portion of weeds from within two randomly placed 0.0625 m2 quadrats (25 by 25 cm) at crop maturity. In both years, samples were dried at 60°C for 24 h and then weighed. Weeds present in both 2003 and 2004 at the ERS included field pennycress (Thlaspi arvense L.), common lambsquarters (Chenopodium album L.), wild buckwheat, shepherd's-purse (Capsella bursa-pastoris L.) and Canada thistle whereas weeds in both years at the New Norway site were mainly wild oat and common lambsquarters.
In both years, tame oat samples were harvested from plots just before crop harvest. Tame oat was harvested from a 1-m2 area in 2003 and from within two randomly placed 0.0625-m2 quadrats in 2004. The samples were dried at 60°C for 24 h and weighed to obtain oat biomass and threshed to calculate grain weight.
Before harvest, 10 randomly chosen wheat/barley spikes in each plot were collected and used to determine kernels spike–1 and thousand kernel weight. At crop maturity, grain was harvested from the entire plot using a Wintersteiger plot combine. Grain yield was recorded on a dry-weight basis. Harvested grain samples were dried at 60°C for 24 h and weed seeds were removed using a 2-mm mesh sieve (Canadian Standard Sieve Series no.10). The weed seed–free grain samples were weighed and plot yields were recorded for those plots without tame oat. For plots with tame oat, the weed seed–free grain sample was weighed, a 100-g sample of grain was removed, and tame oat and wheat were separated and weighed. Grain yields for plots with tame oat were based on multiplication of the weed seed–free plot grain yield by the grain/oat ratio from the 100-g sample. For the sake of varietal comparison, grain yield and kernel weight of the hulled cultivar Seebe were adjusted downward by 15% to account for the weight of the hull (Bhatty et al., 1993). Percentage yield loss was calculated as the difference between grain yield in plots without and with tame oat, divided by the grain yield in plots without.
Data Analysis
A variance-stabilizing square root transformation [(Y + 0.5)1/2] was used for all natural weed, tame oat, and total weed biomass measures (Gomez and Gomez, 1984). A preliminary analysis of variance was performed to detect significant location x treatment differences using the MIXED procedure of SAS (SAS Institute, 2003), where location was considered to be a fixed effect while year was considered random. The effects of seeding rate, location, and cultivar were most important, respectively, while few significant location x treatment interactions occurred (data not shown). Seven of a possible 51 location x treatment interactions were significant (P < 0.10) (data not shown). Subsequent analyses of variance were therefore performed using data combined across environments (year x location), again using the MIXED procedure of SAS (Littell et al., 2006). Environment (year x location) was considered to be a random effect, whereas competition from tame oat, cultivar, and seeding rate were considered fixed effects. Preliminary analysis of variance showed a high degree of variation existed in the naturally occurring weed biomass; thus, an analysis of covariance was attempted to control error, and increase the precision of treatment effect estimation (Steel et al., 1996). Analysis of covariance was conducted (where appropriate) using the previous model with natural weed biomass as a covariate. Due to the conservative nature of probability estimation in the horizontal- and vertical-strip plot factors of the strip plot design (Gomez and Gomez, 1984), effects were considered significant at P < 0.10. Single degree of freedom contrasts were performed to detect significant differences in weed and oat biomass between Seebe, a competitive barley cultivar (O'Donovan et al., 2000), and the other 10 cultivars tested.
A simple economic analysis was conducted to determine the net return associated with doubling the seeding rate, based on the seeding rates used, and yield gains and kernel weights observed in this trial. Crop prices were obtained from documents detailing Alberta purchase prices for organically grown crops in 2005 (Organic Agriculture Centre of Canada, 2006). Seed costs were calculated as the crop price per bushel plus an additional $0.65 per bushel to account for seed cleaning and transportation (Bernie Ehnes, Ehnes Organic Seed Cleaning, personal communication, 2006). The net return was calculated as:
 | [1] |
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
|
Tame oat grain weight and tame oat total plant biomass (dry weight) were correlated (r = 0.95, P < 0.01) and were similarly reduced by cultivar and seeding rate (data not shown). These data suggest that any decreases observed in overall oat biomass would result in decreased oat grain production, ultimately leading to a reduced soil weed seed bank.
Effect of Cultivar on Agronomic Performance and Competitive Ability
Cultivars differed for all measured traits (Tables 2 and 3). Seebe barley and the semidwarf wheat CDC Go yielded more grain than all other cultivars, averaging 3.46 Mg ha–1 and 3.27 Mg ha–1, respectively (Table 2). Marquis wheat was among the lowest yielding cultivars, at 2.24 Mg ha–1.
Average natural weed biomass was highest in plots with Kohika, Peregrine, and CDC Go (Table 2). Seebe barley was identified in a previous trial as a weed-suppressive cultivar, capable of consistently reducing wild oat seed production (O'Donovan et al., 2000). Similarly, Seebe barley was the most weed-suppressive cultivar in the current study. Single degree of freedom contrasts between Seebe and other cultivars highlight some of the major trends among cultivars (Table 3). Without competition from tame oat, total weed biomass accumulation in plots with Seebe was lower than for all other cultivars except Hard Red Calcutta, Marquis, and McKenzie. With competition from tame oat, total weed biomass (natural + tame oat) in Seebe plots was lower than all cultivars except for Hard Red Calcutta, Katepwa, Marquis, and McKenzie (Table 3).
Competition x cultivar interaction effects were detected for grain yield and kernels spike–1 (Table 2; Fig. 1A and 1B). Although oat competition decreased grain yield for all cultivars, the magnitude of the losses differed, where only the wheat cultivars CDC Go and Sapphire experienced statistically significant yield losses (Fig. 1A). Five of the 11 cultivars experienced reduced kernels spike–1 as a result of tame oat competition (Fig. 1B). These results are indicative of genotypic differences in the response of wheat and barley cultivars to competition from weeds on organically managed land, as other studies done on conventionally managed land have suggested (Huel and Hucl, 1996; Lemerle et al., 1996; O'Donovan et al., 2000; Wicks et al., 1986).
Competitive Traits: Height, Tillering Capacity, and Maturity
Previous research has determined that height plays a role in competitive ability (Champion et al., 1998; Cosser et al., 1997; Hucl, 1998; Huel and Hucl, 1996; Korres and Froud-Williams, 2002; Lemerle et al., 1996; Wicks et al., 1986). Two of the three semidwarf cultivars (CDC Go and Sapphire) in the current study incurred significant yield reductions as a result of weed competition, but the other cultivars did not (Fig. 1A and 1B). Percentage yield loss was weakly correlated with height at (r = –0.12, P < 0.05). Overall, height and natural weed biomass were correlated (r = –0.34, P < 0.01), with some of the tallest cultivars (i.e., Hard Red Calcutta, Marquis) suppressing the most natural weeds (Table 2). Similar trends were observed for weed biomass in plots without tame oat competition and for natural weed, tame oat, and total weed biomass in plots with tame oat competition (Table 3). Total weed biomass was found to be correlated with height in both tame oat (r = –0.32, P < 0.01) and non-tame oat plots (r = –0.34, P < 0.01). While there appears to be a relationship between height and weed suppression, other plant traits, in association with height, likely contribute to the ability of a cultivar to suppress and/or tolerate weeds.
Tillering capacity (i.e., spikes m–2) may affect competitive ability, but weed competition is known to affect tillering capacity. The absence of weed-free controls in the present report complicates analysis of tillering effects on competitive ability. Tillering capacity alone does not appear to be a consistent predictor of crop response to weeds, as the least weed-tolerant cultivars (CDC Go and Sapphire) differed in their tillering ability (Table 2). The relationship between crop competitive effect and tillering is similarly unclear. Overall, natural weed biomass and tame oat were highest in plots with two of the lower tillering cultivars (Kohika and Peregrine), but those cultivars were also the shortest cultivars in the trial. Further, relatively high total weed biomass accumulation was observed in plots of the high tillering CDC Go. Champion et al. (1998) similarly reported a high degree of variability in the potential for high tillering wheat cultivars to suppress weeds.
While time to maturity as a competitive trait has been less studied, the relationship between the timing of phenological events and competitive ability has been investigated, with results linking early biomass accumulation to increased competitive ability in wheat (Cousens et al., 2003; Lemerle et al., 1996). Cultivars with differing growth rates may respond differently to environmental stresses (e.g., moisture, light), and those that develop quickly could avoid some of those stresses. Yield loss was positively correlated (r = 0.75, P < 0.01) with days to maturity. This was most apparent with the cultivar Sapphire, which was very late maturing and incurred the highest yield loss. Marquis, however, was also late maturing and did not incur yield loss from competition with tame oat. Furthermore, CDC Go was relatively early maturing. These data suggest that time to maturity alone is not a reliable predictor of crop tolerance to weeds.
Overall natural weed biomass and total weed biomass in plots without tame oat competition were positively associated (r = 0.40, P < 0.01 and r = 0.42, P < 0.01, respectively) with days to maturity, as were tame oat biomass and total weed biomass in plots with tame oat competition (r = 0.47, P < 0.01 and r = 0.50, P < 0.01, respectively). The positive associations between all measures of weed biomass and days to maturity suggests that early maturing cultivars allow less weed growth than later maturing cultivars. As with the other competitive traits in this study, these associations are somewhat inconsistent, yet it remains that time to maturity may play a role in the ability of wheat and barley cultivars to tolerate competition from weeds and to suppress their growth.
Grain Yield and Weed Suppressive Nature of Cultivars
Many studies (Huel and Hucl, 1996; Lemerle et al., 1996; Wicks et al., 1986) have focused on competitive effect (i.e., weed suppression) and response (i.e., yield loss) of cultivars. For the organic producer, more important measures may be overall grain yielding ability under competition combined with weed suppressive ability. In the present study, cultivars differed in their ability to achieve high grain yield and suppress weeds (Fig. 2A
and 2B). A cluster of wheat cultivars, including Park, 9207-DB3*D, and Katepwa were found to be relatively high yielding and good at suppressing weeds, a desirable combination for organic grain production. Despite their favorable performance, these wheat cultivars exhibited varying levels of each of the four "competitive traits" discussed in this article, again suggesting that competitive ability in general cannot be controlled by one trait alone and that it results from a number of traits working together.
|
Competitive Effect and Response
Some cultivars maintain yield primarily by suppressing weeds (competitive effect) whereas others maintain yield without suppressing weeds, that is, by tolerating weeds (competitive response). Some studies have demonstrated a positive association between competitive response and effect (Goldberg and Fleetwood, 1987), but others have reported no relationship between the two (Goldberg and Landa, 1991; Keddy et al., 1994). Goldberg and Fleetwood (1987) suggest that the association may depend on the nature of competition (i.e., size symmetry) and/or plant traits that are important for competitive ability in a particular system (i.e., growth rate).
Huel and Hucl (1996) and Lemerle et al. (1996) reported correlations between percentage yield loss and weed suppression in wheat, suggesting that competitive response and effect in wheat cropping systems may be related. In the current study, tame oat biomass and total weed biomass were correlated with percentage yield loss (r = 0.41, P < 0.01 and r = 0.38, P < 0.01, respectively). Despite this association between weed biomass and yield loss, the cultivars tested appear to differ in their ability to both suppress weeds and maintain yields (Table 2). These results seem to support the theory that the competitive effect and response of cultivars should be considered separately when trying to identify cultivars with superior competitive ability. The use of weed suppressive cultivars, for example, could supply organic crop producers with an approach for managing weeds that could be combined with other management tools (e.g., seeding density) to both increase yield and suppress weeds.
The Effect of Doubling the Seeding Rate
On average, doubling the seeding rate increased grain yield by 10%, from 2.62 Mg ha–1 at the single seeding rate to 2.85 Mg ha–1 at the double seeding rate (Table 2). Similar results have been reported for wheat (Champion et al., 1998; Weiner et al., 2001) and barley (O'Donovan et al., 2000; Scursoni and Satorre, 2005) grown in the presence of weeds. In terms of yield components, average number of spikes m–2 increased by almost 19% at the higher seeding rate, although kernels spike–1 did not change. Champion et al. (1998) reported increases in spikes m–2 and decreases in kernels spike–1 and kernel weight with increased seeding rate, a reflection of increased intraspecific competition. For kernel weight, a cultivar x seeding rate interaction was detected (Table 2), where Marquis and Sapphire wheat were the only cultivars to exhibit kernel weight reductions in response to increased seeding rates (data not shown), indicating the potential for cultivars to respond differently to higher seeding rates. No cultivar x seeding rate interaction effects were detected for grain yield (Table 2), suggesting that both barley and wheat cultivar grain yield is similarly affected by a doubling of the seeding rate.
Overall natural weed biomass m–2 decreased from 98 to 71 g m–2 at the higher seeding rate, a reduction of 28% (Table 2). A competition x seeding rate interaction was detected for natural weed biomass, where doubling the seeding rate more greatly reduced natural weed biomass in plots without competition from tame oat than in plots with competition from tame oat (Table 3). In plots without tame oat, total weed biomass decreased by 33% at the doubled seeding rate. Plots with tame oat experienced a total weed biomass reduction of 27%, with a 16% decrease in natural weed biomass and a 33% reduction in tame oat biomass at the higher seeding rate. Tame oat and crop competition may have suppressed the natural weed population at the single seeding rate so that the double seeding rate appeared less suppressive to natural weed growth. However, total weed biomass (natural + tame oat) was reduced by the double seeding rate treatment, indicating that doubling the seeding rate remains an effective strategy in reducing weeds. Moreover, the mixture of tame oat (a grassy weed) and natural weed populations (mostly broadleaf) in this study reflects the more diverse population of weeds common to organic wheat cropping systems when to compared with conventional systems (Frick, 1993; Leeson et al., 2000). Previous researchers have reported increased weed suppression with increased seeding density (Champion et al., 1998; Lemerle et al., 1996; Weiner et al., 2001), although Korres and Froud-Williams (2002) reported that weed suppression was cultivar specific. The lack of cultivar x seeding rate interactions for any of the weed-related parameters in this trial suggest that the relationship between weed suppression and seeding density in spring wheat and barley is not cultivar specific. Additionally, the presence of only one significant cultivar x seeding rate interaction (kernel weight) in this study substantiates the idea that the relationship between competitive response and seeding rate is not cultivar specific (Table 2).
Economic Analysis: Net Return Associated with Doubling the Seeding Rate
Increasing the seeding rate may increase overall wheat and barley yields; however, net returns may not be high enough to justify such a practice. An economic analysis was conducted to determine the net return associated with the extra yield. Typical organic management practices were considered, factoring in the cost of seed, obtained on farm, but cleaned off farm.
Based on the average yield of Canadian Western Red Spring wheat grown at single and doubled rates in the current study (2.55 and 2.81 Mg ha–1, respectively), net economic gains associated with doubling the seeding rate ranged from $34.62 ha–1 to $53.72 ha–1, depending on grade. Seebe and Peregrine barley yield differed enough to warrant separate analyses. Based on 2006 organic feed barley prices, Peregrine barley produced an additional net return of $15.74 ha–1 when seeded at the doubled rate. Net returns for Seebe barley seeded at the doubled rate were negative, however, with a net loss of $7.99 ha–1. This was likely due to the comparatively smaller increase in grain yield of Seebe as a result of doubling the seeding rate.
Although a number of factors were not considered in this analysis (e.g., extra labor and fuel costs, yield and market fluctuations, climatic factors), it appears to be generally advantageous for producers to consider an increased seeding rate to improve weed suppression and grain yield. Net returns could increase further if the value of weed suppression to the farmer was factored into the analysis. Reduced weed seed bank build-up has many potential benefits in an organic production system, including less fuel and labor costs, a possibility to diversify crop rotations, reduced yield loss, and more yield stability over time.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Although the competitive effect and response of cultivars were correlated, not all wheat cultivars that accumulated high weed biomass experienced the same degree of yield loss as a result of weed competition. The traits that control these two measures of competitive ability may differ, which suggests that wheat cultivars could be bred specifically for yield maintenance and/or weed suppression. This may be especially significant when considering breeding for organic wheat production, where increased weed populations may render weed suppression more important than yield maintenance.
Choice of cultivar can have an impact on the ability of the crop to achieve high grain yield and suppress weeds, which could be a potential benefit for organic wheat producers, giving them another tool for overcoming weed problems. Doubling the seeding rate was effective for suppressing weeds and increasing grain yield under our growing conditions; however, these results may not hold true in different soil types, or under different weed pressure or moisture regimes. In areas of low rainfall, for example, the practice of doubling the crop seeding rate may present a greater risk to crop quality than weed competition. McKenzie et al. (2005) reported reduced benefits of doubling the barley seeding rate under low moisture conditions (<300 mm per season) compared with irrigated conditions. Increases in grain yield were less prominent, whereas reductions in kernel number and weight were more pronounced under low soil moisture.
Because the overall benefits of doubling the seeding rate do not appear to be cultivar specific, increasing the seeding rate may be a more viable management option for organic producers who are aiming to ameliorate problems associated with weed competition. Increasing seeding rates may currently be less complex and more effective than choosing a cultivar that is both weed suppressive and high yielding under elevated weed conditions.
The presence of naturally occurring weeds, which can be nonuniform in distribution, is one of the problems associated with conducting field trials on organically managed land, and one that we attempted to address by using covariate analysis. Though the increased variation in the data can pose certain analytical difficulties, studies conducted on organically managed land aim to better represent the realities of such production systems.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
on plots grown without tame oat competition and on natural weed, tame oat, total weed biomass in plots grown with tame oat competition at four organically managed sites in 2003 and 2004 at Edmonton, AB, and New Norway, AB.
