| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Agric. Food and Nutritional Sci., 410 Agric.–Forestry Cent., Univ. of Alberta, Edmonton, AB, T6G 2P5 Canada
b Battelle Pacific Natl. Lab., Washington, DC 20024-2115
c Northern Agric. Res. Cent., Agric. and Agri-Food Canada, Beaverlodge, AB, T0H 0C0 Canada
* Corresponding author (jking{at}afns.ualberta.ca)
Received for publication May 19, 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
Abbreviations: WAP, weeks after planting
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
The crop–weed competitive relationship in an agroecosystem is determined by climatic, soil, biological, and cultural factors (Altieri and Liebman, 1988). The characteristics of the Parkland and low Boreal regions of Alberta include a cool, subhumid climate; a short, but long-day growing season; fertile black soils that promote both crop and weed growth; and low-fertility gray soils that require a long-term strategy to build up soil organic matter. Spring cereals, oilseeds, and forages are the predominant crops in north-central Alberta.
Clover species are considered poor competitors because of small seed size, lack of seedling vigor, and slow establishment (Lee, 1985). Grown as spring-seeded cover crops in Oklahoma, nine grassy species were more effective than five legume species (including three clovers) at providing ground cover and suppressing weeds (Nelson et al., 1991). Under dryland prairie conditions, large-seeded legumes and sweetclover (Melilotus officinalis Lam.) grown as green fallow provided better weed suppression than Trifolium spp. (Schlegel and Havlin, 1997; Jensen, 1992). Grown as winter cover in Iowa, alfalfa (Medicago sativa L.) and sweetclover usually produced better ground cover than red or alsike clover (Exner and Cruse, 1993).
Research suggests that competitive abilities vary among clover species and cultivars within a species and with intent of use. Three clover species failed to establish under conditions of severe weed interference and when interseeded as a cover crop with corn (Zea mays L.), but crimson clover established well (Abdin et al., 1998). After 14 mo as ground cover for red birch (Betula pubescens Ehrh. f. rubra Ulvinen) seedlings, three perennial clover species reduced birch stem growth while three replanted annual clover species had no significant effect on stem growth (Hanninen, 1998). In a study of seven berseem clover cultivars intercropped with oat (Avena sativa L.) cultivars, Holland and Brummer (1999) observed considerable variability in important agronomic traits (forage stand, plant health, maturity, yield, height, and weeds) due to berseem cultivar effects. They suggest that the variable performance of berseem cultivars in intercrops may not be predicted from monoculture evaluations.
Of the 240 Trifolium spp., approximately two-thirds are annuals and one-third perennials (Zohary, 1972). Annual clovers may have some advantages over perennial clovers as cover crops. In a study of 32 cover crops in California, annual clover species were taller than perennial clovers (Bugg et al., 1996). At 47 to 66 d after planting, Nelson et al. (1991) observed less weed biomass with an annual clover (crimson) than with two perennial clovers (red and white) used as spring cover crops. In a study with three annual and two perennial clover species grown as green manure crops, an annual clover (Persian) had the most aboveground biomass while perennials (red and white clover) had the most root biomass (Kirchmann, 1988). Clover species also differ in germination response to temperature (Evers, 1980), seedling growth (Evers, 1999), N fixation (Trytsman and Smith, 1998), and morphology (Brink and Fairbrother, 1992).
Many studies of cover crops have used the native weed population as part of the experimental design (Nelson et al., 1991; Bugg et al., 1996; Schlegel and Havlin, 1997; Abdin et al., 1998). Given that the differences between clover species may be small, the use of a fixed weed population may provide greater precision in assessing weed suppression. Experimental control of the density, proportion, and spatial arrangements of interacting species often enhances the utility and predictability of interference studies (Radosevich, 1988).
Few studies of cover crops have assessed the effect of mowing on weed suppression. Mowing is recommended as a method to control weeds in cover crops and during establishment of clovers (Teasdale, 1996; Lee, 1985). However, if the weeds are too small when mowed, new growth from the lateral buds of weeds may compete more effectively than if they had not been mowed (Lee, 1985). Brandsaeter and Netland (1999) advocate springtime mowing to control winter annual weeds in clovers planted as winter cover crops, but they caution that reduction of legume biomass may decrease weed suppression. Mowed subterranean clover (T. subterraneum L.) sometimes had more weed biomass than nonmowed treatments when used as mulch with vegetable crops (Ilnicki and Enache, 1992).
The objective of this study was to assess weed suppression by three species of perennial clover (alsike, red, and white) during the establishment year and by four species of annual clover (balansa, berseem, crimson, and Persian). The experiments were designed to determine the effects of mowing on clover and weed biomass and to measure weed suppression on the two main soil orders cultivated in Alberta.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
Fields were disked and harrowed before seeding. Experiments at Breton in 1996 and 1997 followed oat and barley (Hordeum vulgare L.), respectively. Edmonton experiments followed tilled fallow. The 1996 Edmonton site (pH of 6.5 and 5.7 g kg-1 total organic N at 0 to 15 cm) had been limed in 1995 to raise the pH. The 1997 Edmonton site was more acidic with pH of 5.4 and 5.9 g kg-1 total organic N at a 0- to 15-cm depth. No fertilizer was added at Edmonton. The 1996 and 1997 Breton sites (pH of 6.1, 1.4 g kg-1 total organic N at 0 to 15 cm) were on adjacent areas that had been fertilized with 50 and 60 kg ha-1 P2O5 and K2O, respectively, in May 1996.
The experimental design was a split-plot randomized complete block replicated four times. Whole plots were two mowing treatments (mowed or not mowed), and subplots were nine cover crop treatments (seven clover species, rye, and without cover crop). Subplots were 2 by 6 m in 1996 and 2 by 5 m in 1997. Plots were seeded 3 and 4 June at Edmonton and 7 June at Breton in 1996 and 30 May at Edmonton and 9 June at Breton in 1997. Clover species seeded were ‘Aurora’ alsike, ‘Paradana’ balansa, and common white Dutch clover at 8 kg ha-1; ‘Felix’ Persian and ‘Altaswede’ red clover at 12 kg ha-1; and ‘Bigbee’ berseem and ‘Au Robin’ crimson clover at 15 kg ha-1. Clover seeds were inoculated with appropriate strains of Rhizobium leguminosarum biovar trifolii and then broadcast onto the soil surface by hand and incorporated by hand raking. ‘Kodiak’ fall rye was seeded with a four-row Fabro drill at Edmonton and with a single-row cone seeder at Breton at a rate of 70 kg ha-1, a depth of approximately 2 cm, and in rows 18 to 20 cm apart. Brown mustard was added to all plots, including mustard-only control plots, by broadcasting 15 seeds m-2 onto the soil surface and incorporating them by hand raking. Brown mustard, a minor crop in Alberta, was chosen to represent an annual broadleaf weed. Wild mustard (Sinapsis arvensis L.) and others from the Cruciferae family are common, aggressive weeds in annual crops in Alberta. Cultivated mustard has more synchronous germination than wild mustard, making it preferred for some weed experiments (Liebman, 1989).
After emergence, a 1-m2 quadrat with a 0.25-m2 subquadrat was permanently marked in each subplot. Quadrats were placed away from plot margins and in areas with fairly uniform crop and mustard growth. In 1997, quadrats were placed in a manner so that they would contain 12 evenly spaced mustard plants within the 1-m2 area. Where necessary, the mustard population was thinned by hand to 12 plants m-2. All plants other than the respective cover crop and mustard were removed from plots by hand, with the exception of volunteer oat at Breton in 1996. Oat was removed from the subquadrats but left in the remaining areas of quadrats. Clover density within subquadrats was determined at 6 to 7 wk after planting (WAP) in 1996 and at 4 to 6 WAP in 1997.
Half of the plots were mowed at 10 WAP in 1996 and at 6 to 7 WAP in 1997. Plant growth was cut to a stubble height of 7 to 10 cm using a sickle bar side mower in 1996 and a flail type small-plot harvester in 1997. Cut material was removed from plots. Before mowing, aboveground biomass was harvested from each quadrat by clipping plants at a stubble height of 5 to 7.5 cm. Mustard and oat plants (Breton in 1996) were counted for each quadrat harvested. Harvested vegetation was separated by species, dried for 72 h at 52°C, and weighed. In 1996, mustard and crop canopy heights were measured before harvest. Where clover growth was prostrate or lodged, maximum stem length of clovers was also measured. Towards the end of the growing season, harvest procedures were repeated for nonmowed plots and the regrowth of mowed plots. In 1996, at 14 WAP at Breton and 15 to 16 WAP at Edmonton, nonmowed plots were harvested concurrently with 4 wk of regrowth for Breton mowed plots and 5 to 6 wk of regrowth for Edmonton mowed plots. In 1997, at 15 WAP at Breton and 14 WAP at Edmonton, nonmowed plots were harvested concurrently with the 8 wk of regrowth for Breton mowed plots and 7 to 8 wk of regrowth for Edmonton mowed plots.
Data were subjected to analysis of variance to determine significant treatment effects and interactions (P < 0.05) using SAS (SAS Inst., 1995). Data for each year and location are presented separately because error variances were not homogeneous across locations or years. Before F-test and mean separation analysis, results were transformed to square root values if the coefficient of variation exceeded 25. Nontransformed means are presented in tables. When significant treatment effects occurred, means were separated using Fisher's protected LSD test at P = 0.05. Simple linear regression was used to measure response of mustard biomass to crop biomass.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
|
|
Access to light was probably a major factor in clover–mustard competition at Edmonton. Competition for light will be strongest under high-productivity conditions (Goldberg, 1990). Positive relationships between height and competitive ability have been found in many studies of crop varieties and weeds (Berkowitz, 1988). In 1996 at 10 WAP, the mean heights of mustard and clover were 117 and 53 cm, respectively (Table 2). The shortest species (red and white clover) became almost completely overgrown by the mustard canopy. Alsike clover was taller than the other perennials (red and white clover) and was more successful at accumulating biomass and suppressing mustard. As the mustard plants formed a closed canopy, the longer, erect stems of alsike, berseem, crimson, and Persian clover would have had an advantage over the shorter clovers in accessing light. Balansa clover's prostrate growth habit probably reduced its ability to compete for light. Crimson clover had moderate height but, like balansa clover, it suppressed mustard biomass in 1997 but not in 1996. Persian clover had long stems, but its biomass yields were relatively low and it failed to reduce mustard biomass. Persian clover lacked vigor, and it had the lowest rates of emergence. Characteristics of rapid germination, vigorous growth, large leaf area development, greater plant height, profuse branching, and rapid canopy closure have been shown to enhance the ability of a crop to compete with weeds (Pester et al., 1999).
Clover flowering date probably affected competitive ability. Biomass production declines with the switch from vegetative to reproductive growth. Balansa clover flowered much earlier than the other clovers and was in early bloom by 6 WAP. Crimson clover also flowered relatively early, blooming by 10 WAP. It has been suggested that crimson clover is poorly suited for use as a forage at northern latitudes because daylengths longer than 12 h stimulate flowering and because the flowering response is accelerated when seeds germinate at low temperatures (Panciera and Sparrow, 1995; Knight, 1985). Early flowering of balansa clover has been identified as a problem with using it in mixtures for hay or silage (Snowball, 1993). Pester et al. (1999) suggest that the traits that need to be considered in choosing cultivars for mixtures (crop maturity, photoperiod sensitivity, temperature sensitivity, morphology, root system, seedling growth rate, and density response) are also important traits to consider in crop–weed competition.
The mean density of mustard plants in quadrats was 11.5 ± 4.0 plants m-2 in 1996 and 12.0 ± 1.1 plants m-2 in 1997. In 1996, sources of variation (Table 1) and suppression by alsike clover (Table 2) differed between mustard data sets (Mg ha-1 and g plant-1). In 1997, results were consistent between the two mustard data sets. Greater uniformity of mustard density and proximity may account for greater consistency in 1997.
With Mowing
In 1996, when mowing was applied at late flowering of the mustard plants (10 WAP), the mustard did not regrow. In 1997, mowing was applied when the mustard plants were in early flowering (6 to 7 WAP), and the majority of mustard plants regrew. Mowing had a significant effect on mustard in both years, reducing total mustard biomass (Mg ha-1) by 51% in 1996 and by 70% in 1997 (Table 1). Early stage mowing (1997) resulted in a greater benefit to clover/mustard proportion than late-stage mowing (1996). Mowing increased clover biomass by 47% (2.0 Mg ha-1) in 1997 vs. no effect in 1996.
In 1997, all seven clover species reduced the total mustard biomass (Mg ha-1) in mowed treatments (Table 3). The average mustard biomass reduction by clover species was 68%. All clovers except berseem had higher clover biomass yields in mowed treatments than in nonmowed treatments. All clover species, including berseem, produced greater percent reduction of mustard biomass in mowed than in nonmowed treatments. Berseem clover reduced mustard biomass by 83% in mowed treatments and by 69% in nonmowed treatments. In 1996, berseem, crimson, and white clover reduced total mustard biomass per hectare in mowed treatments, but the effects per plant were not significant (data not shown).
|
The mechanisms of mustard regrowth suppression warrant further research. Interference from the clovers may have inhibited mustard regrowth from lateral buds through effects on light transmittance.
Low-Fertility Site—Breton
Without Mowing
All seven clover species and rye decreased the weed biomass (Mg ha-1) in both years (Table 4). In 1996, suppression of weed biomass (Mg ha-1) by alsike and white clover was greater than that by Persian and red clover. In 1997, the data for mustard biomass per plant provided greater separation among clover species than the data for mustard biomass per hectare, with greater mustard suppression by alsike, berseem, crimson, and white clover than by red clover. Mean clover reduction of weed biomass (Mg ha-1) was 51% in 1996 and 57% in 1997.
|
The annual clovers (balansa, berseem, crimson, and Persian) had greater clover biomass than the perennial clovers (alsike, red, and white clover), with the exception of balansa in 1997 (Table 4). The greater biomass of the annual clovers was not associated with greater weed suppression. The relationship between clover biomass and mustard biomass was not significant (r2 = 0.38 in 1996 and r2 = 0.39 in 1997). Two of the perennials (alsike and white clover) had relatively little biomass, but they had some of the lowest weed yields. Belowground interference by alsike and white clover may have been aided by greater root biomass. Kirchmann (1988) found that white and red clover had greater root dry weights than three annual clovers at 19 WAP. The relatively poor performance by red clover at Breton shows that weed suppression was not consistent among the perennial species.
Initial growth at Breton was slow and sparse, probably limited by low soil N. Mustard plants were probably stressed by low availability of N, and the competition from the clover for soil resources would have placed additional stress on the mustard. Mustard plants would have experienced ongoing N deficits while the N status of the clover plants gradually improved after the onset of N2 fixation. Associated research found that N fixation occurred with all clover species at Breton (Ross, 1999).
There were more cases of significant suppression of weed biomass at Breton than at Edmonton (Tables 2 and 4). Clovers reduced weed biomass (Mg ha-1) by an average of 54% at Breton vs. 21% at Edmonton. Liebman and Robichaux (1990) observed a similar trend where competition from a barley–pea mixture reduced white mustard biomass by 87% under low soil-N availability vs. 49% under high soil-N availability. Shading of mustard by barley–pea mixtures was greatest when N fertilizer was not applied. Deficits of N or light had major negative effects on mustard's photosynthetic performance and yield.
Allelopathy was probably a factor in weed suppression by rye, given that weed suppression was greatly disproportionate to rye biomass. Weed biomass suppression and mustard height were most greatly reduced in rye treatments even though biomass was less for rye than all other cover crop treatments. Greater allelopathic interference by rye at Breton than at Edmonton would be consistent with the finding of Mwaja et al. (1995) that the phytotoxicity of rye is higher under low or moderate fertility than under high fertility. Allelopathy may have also played a role in clover–mustard interference. Allelochemicals have been identified in clover and mustard species (Weston, 1996; Krishnan et al., 1998). Allelochemicals derived from crimson clover inhibited germination and seedling growth of wild mustard (White et al., 1989). The dynamics of plant interference on infertile soils tend to be more difficult to interpret and predict than on fertile soils (Goldberg, 1990). Goldberg suggests that under conditions of low resource levels, the species that dominate may be good stress tolerators rather than good competitors.
With Mowing
At Breton, mowing had little effect on clover/weed proportion. Although reductions in weed biomass were not significant, mowing reduced the weed biomass (Mg ha-1) by 27% in 1996 and by 32% in 1997 (Table 1). Mowing did reduce clover biomass, with reductions of 38% in 1996 and 14% in 1997. With early mowing in 1997, 72% of the mustard plants regrew (data not shown). In theory, the optimal point for mowing mustard plants would be after the plants were mature enough not to regrow from lateral buds but before any seed set. Further research, with additional mowing dates between early flowering and late flowering, might determine a more optimal point for mowing mustard plants. Weed ecologists could make major contributions to weed management by discerning the conditions and times under which weeds would be most vulnerable to management tactics (Altieri and Liebman, 1988).
|
SUMMARY AND CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
The annual clover species (balansa, berseem, crimson, and Persian) had no consistent weed suppression advantages over the perennial clovers (alsike, red, and white clover). The seven clover species differed in growth rate, morphology, aboveground biomass, and flowering date. At Breton, the annual clovers had greater biomass than the perennial clovers. At Edmonton, berseem clover produced the largest amount of biomass. Aboveground biomass of cover crop was a significant factor in suppressing the mustard biomass at Edmonton but not at Breton. Berseem clover was the most suppressive species in nonmowed 1997 treatments at Edmonton. Berseem's competitive ability was aided by an upright growth habit, long stems, high biomass production, and late flowering. Similarly, the upright growth and long stems of alsike clover made it more competitive than the other perennials (red and white clover) at Edmonton. In a study of the primary spring growth of two annual and three perennial clovers, Brink and Fairbrother (1992) concluded that differences in stem and forage yield were linked to growth habit rather than life cycle.
The effects of mowing varied with location, timing, and species. Mowing was beneficial to crop/weed proportion on the high-productivity site (Edmonton) but not on the low-productivity site (Breton). Without mowing, the clover proportion of total biomass was much lower at Edmonton (31%) than at Breton (77%). At Edmonton, there was greater benefit when mowing at early flowering than at late flowering of mustard. With early mowing at Edmonton, all seven clovers reduced mustard regrowth (g plant-1); and balansa, berseem, Persian, and red clover also reduced the number of mustard plants producing regrowth.
Clovers may be perceived as poor cover crops because of slow establishment, small leaves, and low height. Yet compared with rye, which competes well with weeds, clovers provided much better ground cover at Breton, and berseem clover suppressed weeds as well as or better than rye at Edmonton. Clovers are adapted to grazing, and their characteristics may be advantageous under mowing.
Berseem clover has potential as a cover crop or annual forage in the Parkland and Boreal regions of Alberta. Further research is being conducted to test berseem clover in mixtures with cereals for forage. Persian clover shared some characteristics with berseem clover but lacked vigor. The early flowering of crimson clover, which occurred at this northern location, was a disadvantage to biomass production and regrowth after mowing. Balansa clover may have potential for use as a smother crop with row crops. It produced abundant, prostrate, early growth for rapid ground cover; tolerated flooding; and then became less competitive after initiation of flowering at 6 WAP.
The effects of the clover species on mustard may be used as an indicator of the ability of clovers to suppress a highly competitive annual weed, but the results cannot be generalized for all weeds. Research with a range of weeds (e.g., perennials, short species, weeds with low plasticity, and large-seeded weeds) would provide a wider picture of weed interference by clover. Testing a number of cultivars for each clover species would enhance understanding of the relative suppressive abilities of clover species.
Further research is needed to (i) test clover suppression of other types of weeds, (ii) measure clover root and belowground interference factors, and (iii) gain a better understanding of the effects of mowing on clover–weed interference. Greater understanding of the growth characteristics of clover species will aid the selection of species for a particular crop use and aid decisions about crop–weed management.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. M. Cherr, J. M. S. Scholberg, and R. McSorley Green Manure Approaches to Crop Production: A Synthesis Agron. J., February 7, 2006; 98(2): 302 - 319. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Ross, J. R. King, J. T. O'Donovan, and D. Spaner Forage Potential of Intercropping Berseem Clover with Barley, Oat, or Triticale Agron. J., July 1, 2004; 96(4): 1013 - 1020. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
