The Eastern Gangetic Plains are a densely populated region of South Asia with comparatively low productivity yet a strong potential to intensify production to meet growing food demands. Conservation agriculture-based sustainable intensification (CASI) has gained academic and policy traction in the region, yet despite considerable promotional activities, uptake remains limited. Based on emerging evidence delving beyond a binary classification of adoption, this qualitative study seeks to explore the experiences and perspectives of smallholder farmers who express positive sentiments about CASI, yet have not progressed to (autonomous) adoption. After thematic coding of semi-structured interviews with 44 experimenting farmers and 38 interested non-users, ten common themes emerged that explain why farmers stagnate in their adoption process. Seven of the ten themes were non-specific to CASI and would constraint promotion and uptake of any agri-system change, highlighting the need for contextual clarity when promoting practice changes in smallholder systems. We summaries this to propose the ‘four T's’ that are required to be addressed to enable agricultural change in smallholder systems: Targeting; Training; Targeted incentives; and Time. Through this more nuanced evaluation approach, we argue the need for a stronger focus on enabling environments rather than technological performance evaluations generically, if promotional efforts are to be successful and emerging sustainable intensification technologies are to be adopted by smallholder farmers.

The North China Plain is an important summer maize/winter wheat rotation area. However, over the years, continued intensive tillage has destroyed the soil aggregate accelerating the mineralization and decomposition of soil organic carbon (SOC), which plays an important role in soil quality, as increased organic carbon storage improves soil fertility and crop yields. Thus, the objective of this study was to explore the comprehensive impact of tillage methods on soil aggregates, aggregate-associated SOC, and carbon sequestration capacity under a regime of straw return. In 2002, we started a 14-year long-term tillage experiment; then in 2016–2017, we tested the following tillage methods, zero tillage (ZT), rotary tillage (RT), subsoiling (SS), and conventional tillage (CT). The results showed that in the 0–10 cm soil layer, tillage methods significantly reduced the proportion of aggregates in the order of 2–0.25 > 5–2 > 0.25–0.053 mm. Additionally, conservation tillage (i.e., SS and ZT) significantly increased the percentage of macroaggregates (0–40 cm) and their SOC content, compared to CT. Additionally, the contribution rate of macroaggregates to SOC was 17.2% and 30.6% higher under SS and ZT than under CT, respectively. Conservation tillage methods improved the carbon sequestration capacity of soil aggregates. Our study provides a theoretical basis for the development of more suitable tillage methods. Furthermore, long-term conservation tillage seemingly protected large aggregates and, SOC, whereby carbon sequestration was enhanced and soil carbon emissions were effectively reduced.

Continuous-cropping conservation tillage systems may provide a viable alternative to the practice of summer fallow; however, concerns have been raised regarding potentially negative changes in weed communities in continuous cropping. Field experiments were established in Saskatchewan at three locations to determine the nature of weed community differences between a crop sequence with and without fallow in zero-, minimum-, and conventional-tillage systems from 1986 to 1990. Weed communities in continuous-cropping treatments tended to have greater total densities and were more similar in composition than crop-fallow treatments. Inclusion or exclusion of fallow within the rotation had a greater impact on weed community composition than did tillage system at Ituna and Waldron, but the reverse was true at Tadmore due to poor crop growth in all tillage systems. Differences in weed community composition were generally characterized by fluctuational changes in species associations. Volunteers of summer-annual crops, such as canola, flax, and barley, were associated with continuous cropping, but other species including perennial weeds, such as Canada thistle, perennial sowthistle, and quackgrass, were not strongly associated with the presence or absence of fallow. The practice of fallowing land to manage weeds may not be necessary.

The objective of this field research was to compare relative effectiveness and profitability of alternative herbicides applied to the same plots for four consecutive years for controlling and reducing dense Canada thistle infestations in no-till spring wheat. Chlorsulfuron at 30 g ai ha−1 plus nonionic surfactant or clopyralid plus 2,4-D at 70 plus 280 g ae ha−1, respectively, applied annually for 4 yr controlled Canada thistle and was more effective for gradually reducing Canada thistle stands than 2,4-D at 560 g ae ha−1, MCPA plus bromoxynil at 280 plus 280 g ae ha−1, or tribenuron at 20 g ai ha−1. Chlorsulfuron and clopyralid plus 2,4-D also controlled Canada thistle ≥ 90% earlier (by 2 yr) than other treatments. Stochastic dominance analysis, a form of economic analysis, predicted that either chlorsulfuron or clopyralid plus 2,4-D would be preferred by farmers to the untreated check, MCPA plus bromoxynil, or 2,4-D treatments. Chlorsulfuron also would be preferred to clopyralid plus 2,4-D by risk-neutral farmers, whereas clopyralid plus 2,4-D would be preferred to chlorsulfuron by highly risk-averse farmers, those who are most likely to pick only consistently effective herbicides.

Adverse changes in weed communities are a limiting factor for the adoption of conservation tillage practices. Predictions of an increased association of annual and perennial grasses, perennial dicot weeds, wind-disseminated species, and volunteer crops as weeds, and decreased association of annual dicot weeds in reduced-tillage systems were tested. Field experiments involving zero-, minimum-, and conventional-tillage systems were conducted in Saskatchewan from 1986 to 1990 at Ituna and Waldron, and from 1986 to 1988 at Tadmore. Weed community composition was analyzed for years 1988 to 1990 by canonical discriminant analysis. An increased association of perennial and annual grasses with zero tillage did not generally occur. Wind-dispersed species and volunteer crops were associated with reduced tillage and summer annual dicots with conventional tillage, but exceptions occurred. Species responded differently among sites or within a site over time. Within the time frame of this research, changes in weed communities were influenced more by location and year than by tillage systems, indicating fluctuational rather than directional or consistent changes in community composition.

Downy brome control in winter wheat is often inadequate. The effects of three crop rotations and two tillage intensities on downy brome populations and associated crop yields were determined in an experiment at Lethbridge, Alberta from 1987 to 1993. Downy brome densities in continuous winter wheat increased from 24 to 970 plants/m2 between 1988 and 1993; and were often higher with zero tillage. Inclusion of fallow or spring canola in rotation with winter wheat suppressed downy brome densities to less than 55 and 100 plants/m2, respectively, over the six years. In continuous winter wheat, yields decreased as downy brome densities increased progressively over years, indicating that monoculture winter wheat production will not be viable in regions where downy brome is prevalent unless effective herbicides are developed. In the more arid areas of the Canadian prairies, a winter wheat-fallow rotation may be most suitable but in higher precipitation areas, a winter wheat-canola rotation is a viable alternative. Crop rotation is a key component of an improved management system for control of downy brome.

The comparison of incorporation vs no incorporation on a mid-October application of the granular formulation of trifluralin to control infestations of wild oat and green foxtail was evaluated in spring barley, wheat, and canola in west central Saskatchewan over a 3 yr period. Incorporation treatments consisted of: no incorporation, one incorporation in fall, one incorporation in fall and a second in spring, one incorporation in fall and two additional incorporations in spring. All treatments eliminated green foxtail. In all three crops wild oat panicle counts were equivalent from incorporated and non-incorporated trifluralin. Wild oat fresh weights in crops grown on stubble were similar for incorporated and non-incorporated trifluralin. In fallow crops, wild oat fresh weight reductions were greater in three of nine site years with incorporation. There was little difference in crop yields from incorporated and non-incorporated trifluralin.

Field experiments were initiated at Alliance and Hairy Hill, Alberta, in 1989 to investigate the effects of conventional tillage, zero tillage, and four levels of nitrogen fertilizer on continuous barley production. In both tillage systems, the nitrogen was banded 6 to 8 cm deep between alternate barley rows. Herbicides were used for weed control each year. The influence of tillage and nitrogen on weed seed population dynamics was determined in 1991 and 1992. In the zero-tillage system, a large proportion of the weed seeds were present either at the soil surface or at the 5- to 10-cm depth. Green foxtail, the dominant species at Alliance, was also present at Hairy Hill where field pennycress was dominant. Green foxtail was consistently associated with low (residual) nitrogen and, in most cases, with conventional tillage. At both locations, green foxtail populations tended to decrease to very low levels as nitrogen rate increased, especially in zero tillage. At Hairy Hill, field pennycress populations in the soil seedbank were higher in zero tillage compared with conventional tillage, but plants that emerged from the soil seedbank in the field in spring were lower in zero tillage. Field pennycress populations were highest under low nitrogen. The results indicate that banding nitrogen has the potential to be an effective tool for green foxtail and field pennycress management in conventional- and zero-tillage systems, resulting in less dependence on herbicides for their control.

The robustness of competitive attributes of cereals such as rapid and uniform seedling emergence, tillering, early biomass accumulation and canopy closure, and height advantage over weeds have not yet been tested under environmental conditions typical of no-till (NT) cropping systems. Our objective was to evaluate the effects or NT practices on growth and productivity of Avena sativa, Triticum aestivum, Hordeum vulgare, and associated weeds. The experiment was conducted on a Kamouraska clay at La Pocatière, QC, in 1994, 1995, and 1996. Avena sativa, T. aestivum, and H. vulgare were grown under tilled and NT practices. Cereal growth parameters were measured six (1994) or seven (1995) times between planting and the 11th week after planting but only once in 1996. Grain yields and yield components were determined at crop maturity. Avena sativa and H. vulgare populations were little affected by tillage, whereas T. aestivum populations were reduced by 16 to 20% in NT systems. Growth in height in NT systems was either similar or greater than in tilled systems in all three cereals. Cereal leaf area index (LAI) and biomass accumulation was also comparable between tillage systems, except for T. aestivum LAI in 1994, which was greater in tilled plots on two sampling dates. Response of annual dicots to tillage was inconsistent in all crops. Annual monocots dominated in some but not all NT systems. Perennial dicots dominated in NT systems, whereas perennial monocots were more abundant in tilled systems in all three cereals. Avena sativa and T. aestivum yields in NT plots were comparable or greater than in tilled plots, in spite of having either lower test weights (A. sativa) or lower 1,000-grain weights (T. aestivum). NT T. aestivum productivity was maintained in spite of reduced plant establishment. Hordeum vulgare yields were also similar across tillage systems, except in 1995, when yields in tilled plots were greater than in NT plots. The height advantage observed for NT H. vulgare did not result in improved yields. All three cereals, and particularly A. sativa, appeared well suited to NT systems, despite the pressure provided by different weed groups, compared to tilled systems. However, results suggest that NT production of cereals could benefit from improved attention to perennial dicot control and crop seedling establishment, particularly for T. aestivum.

Field experiments were conducted at Lacombe and Ellerslie, Alberta, Canada, to study the interaction of tillage with glyphosate and sethoxydim on quackgrass shoot and rhizome biomass, rhizome bud viability, and crop yield. Glyphosate was most effective in the conventional tillage regime, whereas sethoxydim was effective only in conventional tillage. Tillage reduced viable rhizome buds more than shoot or rhizome biomass. The large tillage effect on quackgrass control with either herbicide would likely be diminished in zero tillage systems that employ preseeding burn-off or preharvest application of glyphosate. Under zero and conventional tillage, glyphosate at 220 g ha−1 often was as effective as glyphosate at 880 g ha−1.A split application of glyphosate at 220 g ha−1 in the fall and spring was as effective as a single 880 g ha−1 treatment in conventional tillage, and sometimes more effective than 880 g ha−1 in zero tillage. Glyphosate followed by sethoxydim was usually no better than glyphosate alone. Short-term split applications of sethoxydim offered no consistent advantage over single applications.

Five herbicides (cloproxydim, fluazifop, haloxyfop, quizalofop, and sethoxydim) were compared from 1984 to 1988 at 250 and 400 g ha−1 for controlling quackgrass within different crop competition and tillage conditions at the Lacombe Research Station. Crop competition usually augmented quackgrass control with the herbicides in conventional-tillage plots although direct statistical comparisons were not made. Without crop competition, haloxyfop and quizalofop at 250 g ha−1 were more effective than the other herbicides in conventional tillage 1 mo after treatment (MAT). In a conventional-tillage situation on a unit active ingredient basis, the herbicides ranked in order of decreasing activity as follows: quizalofop ≥ haloxyfop > fluazifop > cloproxydim > sethoxydim. In a zero-tillage situation, none of the herbicides reduced quackgrass shoot weights by 50% 3 MAT. However, haloxyfop and quizalofop were more effective in suppressing quackgrass shoot weight than the other herbicides in the zero-tillage experiments 3 MAT.

In the semiarid northern Great Plains, the adoption of zero tillage improves soil water conservation, allowing for increased crop intensification and diversification. Zero-tillage crop production relies heavily on herbicides for weed management, particularly the herbicide glyphosate, increasing selection pressure for herbicide-resistant weeds. Barley is well adapted to the northern Great Plains, and may be a suitable herbicide-free forage crop in zero-tillage systems. A 2-yr field study was conducted to determine if planting date influenced crop and weed biomass, water use (WU), and water-use efficiency (WUE) of barley and weed seed production in three preplant weed management systems: (1) conventional preplant tillage with a field cultivator (TILL); (2) zero tillage with preemergence glyphosate application (ZTPRE); and (3) zero tillage without preemergence glyphosate (ZT). None of the systems included an in-crop herbicide. Planting dates were mid-April (early), late May (mid), and mid-June (delayed). Early planting of ZT barley resulted in excellent forage yields (7,228 kg/ha), similar to those from TILL and ZTPRE. Early planting resulted in a small accumulation of weed biomass, averaging 76 kg/ha, and no weed seed production regardless of preplant weed management system. Early planting resulted in higher WU than delayed planting, averaging 289 and 221 mm, respectively, across management systems and years. The WUE of crop and total biomass did not differ among preplant weed management systems at harvest from the early planting date. Delayed planting resulted in decreased forage yield with high amounts of weed biomass and seed production, especially in ZT. A pre-emergence glyphosate application was not necessary for early-planted ZT forage barley. Early planting of herbicide-free barley for forage can be an excellent addition to northern Great Plains cropping systems as part of a multitactic approach for improved weed and water management.

Herbicides are very effective tools to control weeds but there has been an overreliance on their use at the expense of other useful methods of weed management. Farmers are interested in alternative methods of weed management but are concerned about the risk of adopting such practices with current small profit margins. Research on the Canadian Prairies has found that cropping systems that utilize zero tillage, diverse crop rotations, competitive cultivars, higher crop seed rates, specific fertilizer management, and cover crops can effectively manage weed populations, especially when used in conjunction with targeted but limited use of herbicides. Farmers are gaining confidence in the merits of such agronomic practices in terms of sustainable weed management and are gradually adopting these integrated cropping systems on their farms. Further research and extension efforts are required to ensure that these integrated weed management systems are biologically and economically robust to facilitate greater adoption at the farm level.

The influence of four tillage systems, varying from intensive to zero tillage, on weed populations and the vertical distribution of weed seeds in the soil was determined at Alliance, Hairy Hill, and Wainwright in northeastern Alberta. The soil was sampled at two depths (0 to 5 and 5 to 10 cm) in fall. Weed seedling emergence in the greenhouse over the winter was assumed to represent the type and amount of weed seeds present in the soil seedbank. Emerged weed seedlings were also identified and counted in the field in spring. In the zero-tillage system, most of the weed seeds were close to the soil surface (0 to 5 cm) at Alliance and Wainwright but were deeper (5 to 10 cm) at Hairy Hill. The winter annuals, field pennycress, shepherd's-purse, and flixweed, and the summer annuals, wild buckwheat and common lambsquarters, increased in the soil seedbank as tillage was reduced, but the higher populations in the soil seedbank did not always result in higher spring seedling populations under zero tillage. In contrast to the seedbank, spring seedling populations of common lambsquarters at Alliance and field pennycress and ball mustard at Hairy Hill were lowest in the zero-tillage system, suggesting that the requirement for herbicides for controlling these weeds in the crop may be least under zero tillage. Both soil seedbank and spring seedling populations of shepherd's-purse at Wainwright and Alliance and of flixweed at Alliance were highest in the zero-tillage system. At Alliance, wild buckwheat seedling emergence in the spring tended to be highest in the minimum-tillage system (one tillage operation prior to seeding). Both soil seedbank and spring seedling populations of green foxtail decreased as tillage was reduced, indicating that green foxtail should become less of a problem under reduced tillage.

Minimizing inputs such as fertilizers, herbicides, or tillage may be sought by producers to satisfy economic as well as environmental goals. One of the challenges in reducing inputs, whether synthetic fertilizers or herbicides, or substituting a synthetic nutrient with an organic source, is to identify practices that will provide optimum growing conditions for the crop while maintaining an adequate level of weed control. Our objective was to measure the cumulative effects of 12 yr of nitrogen (N) and phosphorus (P) fertilization treatments applied to two tillage systems [conventional tillage (CT) vs. no tillage (NT)] in a corn–soybean rotation on weed communities and crop yields. Residual (postherbicide treatment) weed species assembly was determined by multivariate analysis and was influenced mainly by tillage, with weeds more strongly associated with NT than with CT. Diversity of weed communities as measured by richness, evenness (E), and a diversity index (H′), and total weed biomass were greater for NT than for CT. Nutrient treatments had little or no effect on these parameters. Corn yields were reduced by 70% in the absence of N and by 25% in NT compared to CT treatments. Soybean yields were reduced in NT with increasing P rates compared to other treatments, but reductions never exceeded 10%. Overall, corn and soybean had different responses to treatments, with corn yields being far more affected by fertilization and tillage than soybean yields. Conversely, the absence of tillage had a much greater effect than the absence of nutrient input on weed community assembly and biomass, suggesting the importance of a weed management program specifically tailored for NT systems.

Development of conservation tillage practices for dry bean has lagged behind that of many other crops. A field study was conducted to determine the effects of various crop residues and herbicide treatments on weed management and dry bean yield within a zero-tillage system. Main plot treatments included wheat stubble, canola stubble, fall-seeded winter rye, fall-seeded spring rye, and a no-cover control. Subplot treatments included various preplant and POST herbicides. Wheat stubble, canola stubble, and winter rye residue provided sufficient ground cover to prevent soil erosion, and they effectively reduced weed density compared with the no-cover control in all years. Fall-seeded spring rye provided only partial soil-erosion protection and reduced weed density in only 1 of 3 yr. Dry bean emergence was 3 to 5 d slower in the crop residue treatments compared with the no-cover control, but crop density was not adversely affected. However, winter rye residue delayed dry bean maturity by 2 to 6 d. Fall-applied granular ethalfluralin followed by POST bentazon/imazethapyr or imazamox provided the most effective weed control. A sole POST imazamox application also provided good weed control when weed densities were reduced by winter rye residue or wheat stubble. Overall, results indicate that with suitable herbicide programs, similar yields were attained when dry bean was seeded directly into crop stubble or cover crop residues compared with the no-cover control. Information gained in this study will be used to encourage greater farmer adoption of conservation tillage practices for dry bean production on the Canadian prairies.

Crop rotations, particularly those that include legumes, often result in improved soil quality and crop yields. A study was conducted to confirm the presence and persistence of the residual effects of crop rotation and weed management on a test crop and weeds in three tillage systems (moldboard plow [MP]; chisel plow [CP]; no tillage [NT]). Rotation (spring barley monoculture and spring barley–red clover rotation) and weed management (intensive, moderate, minimum) treatments, initiated in 1987, were terminated, and a test crop of spring wheat was grown in 1995 and 1996. Tillage treatments were maintained throughout. Multivariate analysis showed that weed communities were more affected by treatment termination in the rotation NT treatment with minimum weed management than in all other treatments. The former treatment was dominated by perennial broadleaf weeds but sustained adequate wheat yields (3.3 Mg ha−1) compared with the monoculture (1.0 Mg ha−1) one year after termination. Weed communities in CP and MP plots were less affected by treatment termination. Yet, changes in herbicide use at termination caused the virtual elimination of quackgrass from tilled plots and allowed field pennycress to become ubiquitous across treatments. Residual effects from crop rotation were more important than those from weed management in increasing wheat yields in tilled systems. Differences in wheat yield in NT systems 2 yr after treatment termination were attributed to residual effects from previous weed management rather than from crop rotation. Beneficial effects of crop rotation and weed management may persist for 2 yr but will vary according to tillage system.

Glyphosate-resistant canola has been widely adopted in western Canada. This has prompted producer interest in the timing of glyphosate application, particularly under zero tillage, where glyphosate is often applied preseeding. Field experiments were conducted at Lacombe, Edmonton, and Beaverlodge in Alberta in 1997, 1998, and 1999 to determine the importance of preseeding glyphosate and the most effective growth stage to apply glyphosate to canola to optimize yield and weed management. Treatments consisted of zero-tillage systems, with and without preseeding glyphosate, and a conventional-tillage system involving preseeding tillage operations. Glyphosate was applied at the one- to two-, three- to four-, or five- to six-leaf stages of canola in each tillage system. Canola yield and weed dry weight were largely unaffected by the tillage system. In most instances, the highest canola yields occurred when glyphosate was applied early to the crop. The opposite occurred at Lacombe and Edmonton in 1999, however, where canola yield increased as glyphosate was applied at later crop growth stages. This yield benefit likely resulted from the control of late-emerging weeds that exerted competitive pressure on canola. Early glyphosate timing in glyphosate-resistant canola may eliminate the need for preseeding glyphosate in zero-tillage systems, and optimize yield and weed control.

In western Canada, little is known about the seedbank ecology of volunteer canola. Therefore, integrated recommendations for the management of this weed are limited. In this study, we investigated the seedbank persistence and seedling recruitment of two spring canola genotype groups with different secondary seed dormancy potentials under contrasting tillage systems. The study was conducted at two locations with different soils in the Mixed Moist Grassland ecoregion of Saskatchewan. A single cohort seedbank was established in 1999 and was followed for 3 yr in successive wheat crops. In a separate laboratory study, the six canola genotypes examined were classified as those with high and those with medium potentials for the development of secondary seed dormancy (HD and MD, respectively). After one, two, and three winters, maximum persistence of 44, 1.4, and 0.2% of the original seedbank was observed among the treatments, respectively. In 2001, HD canola genotypes tended to exhibit 6- to 12-fold greater persistence than MD canola genotypes, indicating lower seedbank mortality in HD canola. Seedling recruitment of HD canola also was higher than MD canola when differences were observed between these genotype groups. Therefore, long-term seedbank persistence of canola can be reduced by growing genotypes with low inherent potential for the development of secondary seed dormancy. The proportion of persisting seeds tended to be higher under conventional tillage than under zero tillage because of lower seedbank mortality, but no clear distinction in seedbank persistence in terms of absolute time could be made between these two tillage systems. Volunteer canola seedling recruitment followed the pattern of a typical summer-annual weed, where seedling emergence was observed only during May and June.

As a weed, wheat has recently gained greater profile. Determining wheat persistence in cropping systems will facilitate the development of effective volunteer wheat management strategies. In October of 2000, glyphosate-resistant (GR) spring wheat seeds were scattered on plots at eight western Canada sites. From 2001 to 2003, the plots were seeded to a canola–barley–field-pea rotation or a fallow–barley–fallow rotation, with five seeding systems involving seeding dates and soil disturbance levels, and monitored for wheat plant density. Herbicides and tillage (in fallow systems) were used to ensure that no wheat plants produced seed. Seeding systems with greater levels of soil disturbance usually had greater wheat densities. Volunteer wheat densities at 2 (2002) and 3 (2003) yr after seed dispersal were close to zero but still detectable at most locations. At the end of 2003, viable wheat seeds were not detected in the soil seed bank at any location. The majority of wheat seedlings were recruited in the year following seed dispersal (2001) at the in-crop, prespray (PRES) interval. At the PRES interval in 2001, across all locations and treatments, wheat density averaged 2.6 plants m−2. At the preplanting interval (PREP), overall wheat density averaged only 0.2 plants m−2. By restricting density data to include only continuous cropping, low-disturbance direct-seeding (LDS) systems, the latter mean dropped below 0.1 plants m−2. Only at one site were preplanting GR wheat densities sufficient (4.2 plants m−2) to justify a preseeding herbicide treatment in addition to glyphosate in LDS systems. Overall volunteer wheat recruitment at all spring and summer intervals in the continuous cropping rotation in 2001 was 1.7% (3.3 plants m−2). Despite the fact that volunteer wheat has become more common in the central and northern Great Plains, there is little evidence from this study to suggest that its persistence will be a major agronomic problem.