Agronomy Journal Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Miller, P. R.
Right arrow Articles by Holmes, J. A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Miller, P. R.
Right arrow Articles by Holmes, J. A.
Agricola
Right arrow Articles by Miller, P. R.
Right arrow Articles by Holmes, J. A.
Related Collections
Right arrow Crop Rotation Systems
Right arrow Dryland Cropping Systems
Published in Agron. J. 97:189-200 (2005).
© American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA

Production Papers

Cropping Sequence Effects of Four Broadleaf Crops on Four Cereal Crops in the Northern Great Plains

P. R. Miller* and J. A. Holmes

Dep. of Land Resour. and Environ. Sci., P.O. Box 173120, Montana State Univ., Bozeman, MT 59717-3120

* Corresponding author (pmiller{at}montana.edu)

Received for publication January 26, 2004.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSION
 REFERENCES
 
The productivity and cropping sequence effects of broadleaf crops are not well known within no-till systems in the semiarid northern Great Plains. We compared productivity of cool- and warm-season broadleaf crops with spring wheat (Triticum aestivum L.), measured cropping sequence effects of broadleaf crops on subsequent cereal crops, and tested if different cereal crops interacted with previous crops. During 1999–2001 we conducted 2-yr cropping sequence experiments at five sites in Montana under climatic conditions ranging from severe drought to average rainfall. Grain yield and quality were measured for chickpea (Cicer arietinum L.), flax (Linum usitatissimum L.), pea (Pisum sativum L.), proso millet (Panicum miliaceum L.), sunflower (Helianthus annuus L.), and wheat in Year 1. In Year 2, four cereal test crops [barley (Hordeum vulgare L.), durum (Triticum durum L.), spring wheat, and winter wheat] were grown following the Year 1 crops and a fallow control to measure cropping sequence effects. Comparative productivity varied among crops by site, affirming that crop diversity mitigates production risk. Under average rainfall, the cereal test crop yields following flax, pea, and chickpea ranged from 84 to 101% of the fallow control and were greater than that following wheat in seven of nine cases. However, yields following sunflower were equal or less than those following wheat. Under severe drought, cereal test crop yields following broadleaf crops ranged from 21 to 41% of the fallow control and were equal or less than those following wheat. Previous crops affected four cereal test crop species similarly.

Abbreviations: AER, agroecoregion • WUE, water use efficiency


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSION
 REFERENCES
 
AGRONOMICALLY, three unsustainable aspects of farming systems in the semiarid northern Great Plains are summer fallow, tillage, and cereal monoculture. In 2002, 34% of the 4.2 million dryland ha in Montana were fallowed (2002 Certified Crop Acreage, USDA Farm Service Agency, Bozeman, MT), and no grain crop other than some type of wheat or barley occupied >40000 ha. Larney et al. (1994) referred to summer fallow as the single most devastating cause of soil degradation through soil nutrient and organic matter depletion, soil erosion, salinization, and inefficient use of precipitation. Tillage is detrimental to fragile soils in semiarid regions, and no-till systems are being advanced even beyond the agronomic community as part of the solution for increasing agricultural production within a sustainable framework (Huang et al., 2002; Trewavas, 2002). Crop diversity is a key concept in managing the risk of unpredictable rainfall and market patterns and is essential to successful management of no-till systems (Peterson et al., 1996; Zentner et al., 2002). Montana farmers recognize their reliance on the cereal grain market sector and the growing need for nutrient and pesticide inputs required to maintain productivity in cereal monoculture systems. However, with scant research information, undeveloped market infrastructure for broadleaf crops, and low returns from their traditional commodities, alternative cropping ventures are especially risky in Montana.

Diversified no-till cropping systems have been promoted to increase environmental and economic sustainability (Tanaka et al., 2002; Zentner et al., 2002). Oilseed and pulse crops increase market diversification since these crop prices respond somewhat independently of cereal grain markets (Zentner et al., 2002). Oilseed and pulse crops also increase production diversification due to differential responses to growing season rainfall and temperature patterns (Johnston et al., 2002; Miller et al., 2002a). Crop diversity may also add value to cropping systems by increasing the efficiency of cereal crop production (Johnston et al., 2002; Miller et al., 2002a). Rotational benefits to wheat and barley from broadleaf crops may include enhanced soil fertility (pulse crops), increased water use efficiency (WUE), as well as decreased yield and quality losses from weeds and soil-borne disease (Derksen et al., 2002; Krupinsky et al., 2002). Diversification and intensification of wheat-based cropping systems have been key to sustaining farm profitability for producers in the driest parts of the Canadian prairies (Zentner et al., 2002). In fact, an unprecedented acreage shift into oilseed and pulse crops has occurred in a dryland cropping region where historically the distribution of crops was similar to Montana's current situation. In 1991, oilseed and pulse crops accounted for less than 4% of the seeded area in the "Brown" soil zone (the region of southwestern Saskatchewan adjacent to northern Montana). In 2002, oilseed and pulse crops accounted for 24% of seeded area in the Brown soil zone of Saskatchewan (Statistics Canada). Importantly, the broadleaf crop stubbles for this same land area can benefit cereal crops.

In tilled systems, pulse crops have proven better adapted to driest parts of the northern Great Plains due to high WUE and greater rotational benefits to cereals (soil N and water) than oilseed crops (Miller et al., 2001, 2002b, 2003a, 2003b). Miller et al. (2003b) reduced N fertilizer amounts for crops following three pulse crops to account for greater predicted soil N mineralization. After this N "equalization," the average wheat yield following pulse crops did not differ from that following mustard (Brassica juncea L.) (P < 0.10). However, for both the loam and clay soil sites, the average yield of wheat following wheat was 12 and 26% less, respectively, than wheat following the average of four broadleaf crops. It was uncertain if the additional yield of wheat following broadleaf crops was due to a failure to completely account for mineralized N or some other effect(s) such as conserved soil water or disruption of cereal disease cycles. Cropping sequence effects from broadleaf crops managed in no-till systems have not been previously reported within Agroecoregions (AER) 9 and 12, the most drought-prone area in the northern Great Plains (Padbury et al., 2002).

Previous cropping sequence studies have relied on a single cereal crop to measure the test crop response. Since key growth periods for different cereal crop species vary within a growing season, it is possible that broadleaf crops will affect cereal crop productivity differently. The objectives of this research were to compare productivity of early seeded cool-season and late-seeded warm-season broadleaf crops with spring wheat to determine if broadleaf crops differed from spring wheat in their effect on a subsequent cereal crop and to test if four cereal crop species interacted with four broadleaf crop stubbles. These questions were posed within the context of no-till management within dryland cropping systems in Montana.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSION
 REFERENCES
 
A field study was conducted for consecutive years at five locations in Montana: Amsterdam, Bozeman, Denton, Dutton, and Havre. Experimental site characteristics are summarized in Table 1. Amsterdam, Bozeman, and Denton are in AER 9, and Dutton and Havre are in AER 12 of the northern Great Plains (Padbury et al., 2002). All trials were established on commercial farms in wheat stubble fields with a minimum 3-yr history of no-till management, except at Bozeman (A.H. Post Research Farm) where the trial was established in a wheat stubble field that had been previously managed with intensive tillage. Growing season precipitation was measured on-site at all locations either at existing meteorological stations (Bozeman, Havre) or with a combination of portable electronic and manual rain gauges (Amsterdam, Denton, and Dutton), and the monthly distributions are presented in Table 2.


View this table:
[in this window]
[in a new window]
  		Table 1. Location, soil type, cropping history, seeding date, and fertilizer level at each site.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Overwinter and monthly growing season precipitation (mm) during consecutive growing seasons at five locations in Montana.

 
Treatments in this study consisted of a combination of five (1999) or six (2000) different crop species [pea, flax, spring wheat, chickpea, sunflower, and proso millet (2000 only)] planted as the "previous crop" with four "test crop" species (barley, durum wheat, hard red spring wheat, and hard red winter wheat) following the various previous crops. Treatments were arranged as split plots, with previous crops as main plots and test crops as subplots using four replications in a randomized complete block design. Main plots were 14.7 by 7.3 m, and subplots were 7.3 by 3.7 m. Pea, flax, and spring wheat were designated as "cool-season" crops in this experiment and planted 32 to 35 d earlier than the remaining crops at all sites except Bozeman and Denton where a combination of wet soils and logistical difficulties delayed seeding of the cool-season crops until the warm-season crops were planted (Table 3). Winter wheat was sown in mid- to late September, and the remaining cereal test crops were sown in early to late April (Table 3). All crops were direct-seeded into standing wheat stubble at rates described in Table 4, anticipating 80% of live seeds to emerge, except at Bozeman where plots were direct-seeded into wheat stubble that had been tilled intensively the previous fall. Seeding rates for flax, pea, and chickpea were based on recommendations from western Canada since research-based recommendations did not exist for these alternative crops in Montana (Hnatowich, 2000; Anonymous, 2002). The seeding rate for sunflower was based on recommendations from western North Dakota (D. Tanaka, personal communication, 1999) and the seeding rate for millet was based on recommendations from South Dakota (Wietgrefe, 1990). However, for flax, pea, chickpea, and spring wheat, recommended plant densities typically exceed threshold plant densities to attain optimal yield in the semiarid northern Great Plains (Pelton, 1969; Lafond, 1993; Johnston et al., 2002; Gan et al., 2003). Seedling densities were determined 3 to 4 wk after seeding by counting three or four 0.5-m-length rows in two randomly chosen locations in each plot (0.91–1.02 m2). Crop densities at all sites met or exceeded reported threshold densities.


View this table:
[in this window]
[in a new window]
  		Table 3. Agronomic factors for cropping sequence study at five Montana locations, 1999–2001.

 

View this table:
[in this window]
[in a new window]
  		Table 4. Cultivars, seed rates, fungicidal, and inoculant seed treatments for cropping sequence study at five Montana locations, 1999–2001.

 
Plots were sampled in the spring and fall for Year 1 crops and the following spring before seeding the cereal test crops by taking cores to a depth of 1.2 m; dividing the cores into 0- to 0.3-, 0.3- to 0.6-, 0.6- to 0.9-, and 0.9- to 1.2-m segments; and bulking by depth. The core diameter was approximately 30 mm and was measured precisely with calipers at each sampling time. For Year 1 crops, spring soil N sampling occurred 1 d before 35 d after seeding. When soil sampling occurred after seeding, it was conducted in a grid pattern in the alleyways around the plots to prevent interference from fertilizer N applied during seeding. For Year 1, spring soil N and water values were averaged by rep and considered uniform within a rep. Eight soil cores per rep were taken at Bozeman and Havre and 14 cores per rep at Amsterdam, Denton, and Dutton. After that, postharvest and before seeding, the Year 2 cereal test crop soil samples were taken within the main plots. Soil samples were analyzed for water (gravimetrically) and NO3–N (Hamm et al., 1970). Site values for soil bulk densities were obtained by measuring the volume of the soil increments and averaging the oven-dried soil mass values, by depth increment, for each site. It is possible that the small-diameter soil cores inflated bulk density values due to greater sample compaction often noted with small-diameter soil cores, but since bulk density values were held constant by site, they did not bias treatment comparisons. Bulk density values were used to express N and water content on a volumetric basis.

Year 1 crops at all sites were managed so that N fertility was equal among the nonlegume crops to permit a fair comparison of crop productivity (Table 3). Pea and chickpea were inoculated with appropriate strains of rhizobia to prevent N fertility as a yield limitation. Monitoring of root nodulation confirmed successful inoculation of pulse crops at all sites. Year 1 crops were managed to supply sufficient N fertility for a wheat yield target of 3.0 t ha–1 using 30 kg of N per tonne of targeted grain yield. This corresponded well with targeted oilseed crop yields of 1.5 t ha–1 using 60 kg of N per tonne of targeted seed yield (Jackson, 1999). Since soil available nitrate N was not known at the time of Year 1 crop seeding, we relied on farmer soil tests from the same field (Amsterdam, Denton, and Dutton) or farmer intuition in regard to residual soil nitrate N under wheat stubble (Bozeman and Havre). The targeted N supply was achieved for all sites except Amsterdam where available soil N was less than expected. However, precipitation strongly limited wheat yields at Amsterdam (as well as Denton and Dutton) such that N was not limiting. Previous soil tests at all sites showed P–K–S levels were generally not yield limiting for wheat, but these nutrients were applied to ensure they were not limiting and to maintain soil nutrient status.

Plant-available soil water before seeding was determined to depths of 0.6 m at Denton, 0.9 m at Dutton and Havre, and 1.2 m at Amsterdam and Bozeman and estimated by subtracting lower limits (Ritchie, 1981) of 109 mm at Bozeman, 86 mm at Denton, 123 mm at Havre, 113 mm at Amsterdam, and 118 mm at Dutton (Table 3). The soil depths at Denton and Havre were limited by obstructive coarse fragments (i.e., rocks), and at Dutton, soil wetting below 0.9 m did not occur during the 2-yr cycle at that site, even in the fallow control plots. At Denton and Havre, it is likely that additional water was accessed by plants below the depths of our soil measurements, but the water-holding capacity in those gravelly layers would be expected to be minimal. It is also likely that sunflower was able to access water at depths greater than 1.2 m at Amsterdam and Bozeman (Johnston et al., 2002).

Cultivars, fungicides, and inoculants are described in Table 4. Different seeding implements were used at each site, according to availability. At Denton and Havre, a 3.6-m-wide Conserva Pak no-till air drill (Conserva Pak Seeding Systems, Indian Head, SK, Canada) was used with fertilizer banding tips set 5 cm below and 2.5 cm to the side of the seed furrow, with 30-cm spacing between seed rows. In 1999 at Bozeman, a 1.8-m-wide Hege disc seeder (Wintersteiger USA, Salt Lake City, UT) was used with 26-cm row spacing, and fertilizer was broadcast independently with a 3.6-m-wide Valmar pneumatic granular applicator (Valmar Airflo, Elie, MB, Canada). In 2000 at Bozeman, a 1.8-m-wide custom-built no-till seeder was used with hoe shank single-shoot openers at 26-cm row spacing and with granular applicators (Gandy Co., Owatonna, MN) mounted on the seeder to dribble a band of dry fertilizer on top of the seed row. In 2000 at Amsterdam, a customized 3.5-m-wide Flexi-Coil no-till air drill (Flexi-Coil Ltd., Saskatoon, SK, Canada) was used with rear-mounted double-shoot (seed + fertilizer) Barton angle disc openers on 23-cm row spacing. In 2001 at Amsterdam and at Dutton, a 1.8-m-wide Fabro no-till drill (Fabro Enterprises Ltd., Swift Current, SK, Canada) was used with disc openers on 26-cm row spacing and with disc coulters banding fertilizer 2.5 cm below and 2.5 cm to the side of the seed furrow.

Weeds were managed through use of recommended herbicides for all crops. Herbicides were applied in 94 L ha–1 of water with flat fan nozzles at a pressure of 207 to 276 kPa using a custom-made Gregoire 7.3-m-wide shrouded sprayer (Brad Gregoire, Havre, MT) at all sites. Minimal supplemental hand weeding was conducted to ensure weeds did not affect crop yield, with one exception. In 2001 at Amsterdam, infestation of downy brome (Bromus tectorum L.) caused yield loss in the winter wheat plots, but there was no interaction (P = 0.40) between Year 1 and Year 2 crops. In 1999, pea was injured by MCPA-amine (2-Methyl-4-chlorophenoxyacetic acid) applied at a high rate (830 g a.i. ha–1, the high end of the range for recommended application in Montana), which delayed crop growth for 2 to 3 wk and likely reduced yield.

Shoot biomass yield (seed and straw) was determined at maturity from two randomly chosen locations in each plot from three or four 0.5-m length rows (0.91–1.02 m2). Biomass samples were bagged, oven-dried at 40 to 50°C, and weighed. All crops were harvested directly with a plot harvester. Grain samples were oven-dried at 40 to 50°C for 72 h, cleaned, weighed, and reported on a dry matter basis. A subsample was used to determine seed weight. Grain N concentrations were determined using a LECO CNS combustion analyzer (LECO Corp., St. Joseph, MI) for Year 1 crops and whole-grain NIR with an Infratec 1225 instrument (FOSS Analytical A/S, Hillerød, Denmark) for cereal test crops. Grain N concentrations were converted to protein concentration using a factor of 5.7 for the cereal crops and 6.25 for the broadleaf crops (Jones, 1941). Grain N yield was reported because it integrates yield and protein response to N in cereals and was calculated by multiplying the grain N concentration by the grain yield. Seed density was determined by measuring the dry weight of grain in a 0.95-L container.

Water use efficiency was determined by measuring the dry grain yield by water used. Crop water use was determined as the difference in soil water between spring and fall soil sampling plus the sum of all rainfall between soil sampling dates (Table 3). Nitrogen margin was used to approximate biological N fixation and was calculated as N output minus N input: grain N yield minus the difference in soil N between spring and fall soil sampling to a 0.6-m depth and minus fertilizer N. The estimate of N margin is biased to the extent that soil N use was not accounted below 0.6 m, which may increase N margin values for crops with extensive rooting below 0.6 m, such as wheat, millet, and sunflower.

Data were analyzed with the PROC GLM procedure of SAS (SAS Inst., 1988, p. 549–640). Sites were analyzed independently due to heterogeneity of error variances. Year 1 previous crops and Year 2 cereal test crops were considered "fixed" effects. The variance estimates for each site for Year 1 crop x cereal test crop were used to determine if cereal test crops responded differentially to Year 1 crop stubbles. Effects were declared significant at P < 0.10 unless otherwise specified.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
SUMMARY AND CONCLUSION
 REFERENCES
 
Year 1 Crop Productivity
In Year 1, Amsterdam, Denton, and Dutton experienced more severe drought than Bozeman or Havre (Fig. 1). Because crop development patterns interact with variable weather, it is difficult to predict which crop will yield best in any year at a particular location. The data presented here support this contention (Table 5). For statistical yield rankings, pea varied from first to last, flax varied from second to last, wheat was first or second, chickpea was third or last, and sunflower ranked from first to last among these five sites. Thus, multiple crops provided yield diversification within the climatic context of these five sites.



View larger version (50K):
[in this window]
[in a new window]
  		Fig. 1. Cumulative plant-available water and potential evapotranspiration (Thornthwaite) during consecutive growing seasons at five locations in Montana. Plant-available soil water at the start of the growing season was estimated by using the "lower limit" method in wheat stubble (Ritchie, 1981).

 

View this table:
[in this window]
[in a new window]
  		Table 5. Crop productivity and quality parameters for Year 1 crops grown at five locations in Montana.

 
In this study, it was hypothesized that early seeded cool-season crops (pea, flax, and wheat) would outyield their late-seeded warm-season counterparts (chickpea, sunflower, and millet) because delayed seeding and/or development associated with the warm-season crops typically increases drought stress in the semiarid northern Plains. Pea yielded greater than chickpea at three of four sites (Havre excluded), but flax yielded greater than sunflower at one of five sites and less than sunflower at three of five sites. Deeper soil water extraction by sunflower explains the superior yield performance of sunflower at Amsterdam and Bozeman (Fig. 2). At Denton, a hail event (22 June) injured sunflower relatively early in its growth cycle, allowing greater yield recovery than other crops at that site (Schneiter et al., 1987). At the driest site, Dutton, the late-seeded warm-season crops yielded zero or very nearly so, providing flax a yield advantage over sunflower. Spring wheat yielded greater than proso millet at both locations where it was grown. Thus, late-seeded warm-season pulse and cereal crops were less productive than early seeded cool-season counterparts, but a deeper rooting system and different growth pattern permitted sunflower to be more productive than flax.



View larger version (61K):
[in this window]
[in a new window]
  		Fig. 2. Cumulative soil water depletion by Year 1 crops measured in 0.3-m-depth increments by comparing soil water status with the chem fallow control at four sites in Montana. Bars marked with the same letter do not differ (P < 0.10). Denton is not included because the soil depth was only 0.6 m and soil water differences among crops did not occur at that site due to the shallow soil.

 
In the semiarid Canadian prairies, spring wheat and pea yields were comparable when both were grown on wheat stubble (Miller et al., 2001), but that occurred in this study only at the 2000 sites. In 1999, a high rate of MCPA-amine (within recommended range) severely injured pea at all sites, stalling growth for 2 to 3 wk. Thus, it is likely that those yields would have been much greater with a lower rate of herbicide. It is also important to consider grain quality for its direct effect on gross returns. Although seed size of pea varied from 160 to 231 mg seed–1, the grain density, referred to in the grain industry as "test weight," remained consistently high (Table 5). However, the grain density (test weight) for wheat was sufficiently low (680–686 kg m–3 corresponds to a test weight of 53 lb bushel–1 and industry standard is 60) at three of the five sites to incur price discounts. Statistical rankings for WUE paralleled the grain yield response at all sites, except at Bozeman and Denton where pea yields were equal to flax but the WUE of pea was greater than flax.

One potential economic advantage to growing N-fixing pulse crops such as pea and chickpea is the positive effect on soil N margin. The N margin values for pea and chickpea averaged 84 and 47 kg ha–1 greater than wheat, reflecting the value of biological N fixation for these legumes (Table 5). We cannot explain the anomalous response for N margin of sunflower at Dutton. The fall residual soil N values were consistently high across all four replicates. The landowner observed persistent deer (Odocoileus virginianus) presence after harvest, and they may have preferred to feed on the aborted heads of the sunflower plants, with resultant N cycling from animal wastes. At all five sites, the N margin for pea was greater than that for chickpea, consistent with earlier reports by Miller et al. (2001)(2003a). This was primarily a reflection of greater pea than chickpea grain yields, but there was also a trend of greater residual soil nitrate N following pea (Fig. 3).



View larger version (67K):
[in this window]
[in a new window]
  		Fig. 3. Spring soil nitrate N to 0.6 m and crop available water to 0.6 m at Denton, 0.9 m at Havre and Dutton, and 1.2 m at Bozeman and Amsterdam, MT, at seeding of the Year 2 cereal test crops. > or < denote differences from spring wheat within each site (P = 0.10).

 
Pea used less soil water than wheat at three locations; otherwise, there were no consistent differences in soil water use among crops (Fig. 2). However, by the following spring, soil water status following the Year 1 crops was equal except at Amsterdam, with only fallow having greater soil water than the wheat stubble at all sites (Fig. 3). This diminishing of postharvest soil water differences was reported in a previous study in the semiarid northern Great Plains (Miller et al., 2003a). Thus, the Year 1 crops and their stubbles reasonably reflected the growing conditions at each site and appeared to differ in soil N status, but not soil water status, at the time of seeding the cereal test crops.

Year 2 Cereal Test Crops
The previous crop had a large effect (P < 0.01) on cereal test crop yield, grain protein, and grain N yield at all locations except one (grain protein at Amsterdam; Table 6). However, the effect of previous crop varied strongly. Among sites, the previous crop explained 8 to 86% of the variance for grain yield, 1 to 62% of the variance for grain protein, and 13 to 85% of the variance for grain N yield (Table 6). The cereal test crop itself also had a large and variable effect (P < 0.01) on cereal test crop yield, grain protein, and grain N yield at all locations. Among sites, the test crop explained 8 to 90% of the variance for grain yield, 30 to 96% of the variance for grain protein, and 8 to 82% of the variance for grain N yield. It was hypothesized that since these four cereal test crops differed in primary growth period, plant development rates, and physiological determination of grain yield, test crop would interact with the previous crop. Previous crop x test crop interactions were significant in only 4 of 15 cases and, where they were significant, explained ≤1% of the variance (Table 6). This occurred despite using cereal test crops with contrasting growth and development patterns grown at sites with contrasting water availability (Fig. 1). Importantly, this result validates previous cropping sequence research that has typically used only a single cereal crop to test previous crop effects.


View this table:
[in this window]
[in a new window]
  		Table 6. Analysis of variance (ANOVA) mean squares for cereal test crop yield, grain protein, and grain N yield at five locations in Montana, 2000–2001.

 
Previous Crop Effect
Sites were distinctly classified into two different climatic conditions during the cereal test crop year: (i) near-average crop water availability (Bozeman, Amsterdam, and Denton), and (ii) severe drought (Havre and Dutton) (Table 2). The Dutton site was unique in this study because it experienced severe drought in both crop years. The severity of drought at Havre in 2000 was less obvious. There, rainfall was only 59 mm less than the May–August long-term average. However, no effective rainfall events (>5 mm) were received between 1 June and 6 July. This 34-d drought period caused severe stress on all cereal test crops grown on stubble while those on fallow used stored soil water to bridge the gap until timely July rain occurred (6 July—7 mm, 10 July—10 mm, and 21 July—24 mm) to complete grain fill. Previous economic research in the driest region of Saskatchewan (Brown soil zone) has shown the break-even yield for recropped spring wheat to be within the range of 67 to 84% of the yield attained on tilled fallow at the same location (Zentner et al., 1986). Using that yield guideline, it was clearly unprofitable to grow cereals following any crop under severe drought at Havre or Dutton where recropped cereal yields averaged only 33 and 30% of that attained on the chem fallow control, respectively (Table 7). Conversely, under near-average rainfall conditions at the remaining three sites, recropped cereal yields averaged 82 to 89% of that attained on the chem fallow control, indicating potential profitability. Meaningful rotational benefits to the cereal test crops from previous alternative crops to wheat were apparent at the three sites receiving near-average moisture but not at the two drought sites.


View this table:
[in this window]
[in a new window]
  		Table 7. Mean effect of previous crop on five cereal test crop response parameters grown at five locations in Montana (columns left to right = wettest to driest).

 
The previous crop did not affect plant densities at any of the sites, except at Dutton where final shoot densities on fallow were 24% greater than all crop stubbles (data not shown). The shoot biomass response corresponded closely with grain yield (Table 7) and so is not discussed separately here. At the three average-rainfall sites, relative rankings of crop stubbles were consistent, with the chem fallow yield in the highest statistical grouping, and cereals following sunflower in the lowest statistical grouping. Sunflower used greater soil water to a 1.2-m depth (data not shown), and in a related study at Bozeman (2001–2003), sunflower used greater soil water to a 1.8-m depth than all other crops in the study, reducing subsequent pea yields by 25% compared with pea following wheat stubble (P < 0.10, P. Miller, unpublished data, 2001–2003). Similarly, Nielsen et al. (1999) reported that winter wheat yields following sunflower were reduced 30% compared with winter wheat grown on summer fallow. With average rainfall, cereal yields following flax ranged from 15 to 19% greater than wheat in all 3 yr while those following pea were 19 and 24% greater than wheat in 2 of 3 yr and those following chickpea were 8 and 11% greater than wheat in 2 of 3 yr (Table 7). Thus, with the notable exception of sunflower, diversified crop sequences increased wheat yields in 7 of 9 cases under average rainfall. Conversely, at the two drought sites, cereal yields following spring wheat were greatest but were only 36 and 44% of the fallow yield at Dutton and Havre, respectively. Under severe drought, it is likely that the superior stubble microclimate of spring wheat resulted in increased WUE as has been reported previously for pea grown in wheat stubble (Miller et al., 2003b). Thus, cereal grain yield benefitted from rotational broadleaf crops under average-rainfall conditions but not during severe drought.

At the average-rainfall sites, the grain protein concentrations of the cereal test crops were lowest following wheat and chickpea (Table 7). Thus, in the case of wheat stubble, both yield and grain protein were reduced. This may indicate greater immobilization of soil nitrate N due to wide C/N ratio wheat straw interacting with synchrony of cereal nutrient demand. However, only at Amsterdam were grain protein concentrations possibly below critical threshold values (123 to 147 g kg–1, dry basis) for optimal yield (Engel et al., 1999; Selles and Zentner, 2001). Reduced grain protein concentration on chickpea stubble is difficult to explain but may have been related to the low chickpea yields at these three locations, which limited biological N fixation and subsequent N cycling. Since N clearly did not limit yield at the two drought sites, there were no important differences among the very high grain protein concentrations.

Crop yield or grain protein may be affected differentially, depending on the timing of soil N release from crop stubbles (Miller et al., 2002a). In fact, the statistical ranking of crop stubble effects on grain yield and grain protein differed at each site, and so grain N yield was considered an important parameter because it integrated both effects. Cereals following pea (range = 118 to 127%) and flax (range = 115 to 127%) had greater grain N yield than wheat stubble at all three average-rainfall sites and were lower than the chem fallow control only at Amsterdam (Table 7). Cereals following chickpea had greater grain N yield only at Denton while cereal grain N yield following sunflower did not differ from wheat at the average-rainfall sites. At the drought sites, a similar ranking of previous crop effects on average cereal grain N yield was observed at Havre but not at Dutton where two consecutive years of severe drought made water the sole limiting factor. Under those very low yield conditions, cereals following wheat, sunflower, and millet had the highest grain N yields. Presumably, the spring wheat stubble caused the most favorable microclimate for promotion of WUE while sunflower and millet died from drought the previous year before reaching anthesis, limiting soil water extraction.

At the average-rainfall sites, there were few differences among previous crops for effects on cereal test crop grain density (i.e., test weight). At Denton, grain densities were lowest on wheat and sunflower stubbles but likely were not so low as to incur price discounts. At Havre, grain densities were lowest following chickpea and sunflower but again were likely not so low as to incur price discounts. At Dutton, grain densities were lowest on flax, chickpea, sunflower, and millet stubbles, risking price discounts, making very low grain yields even less valuable.

Cereal Test Crops
Shoot biomass ranged widely among cereal test crops and sites, from less than 2 t ha–1 for durum wheat at Dutton to nearly 13 t ha–1 for barley at Bozeman (Table 8). Shoot biomass at maturity paralleled the grain yield response among cereal test crops except at Dutton where winter wheat had a very low harvest index (0.20) compared with spring wheat (0.34). Grain yields ranged even more widely from 0.5 t ha–1 for durum wheat at Dutton to 4.2 t ha–1 for barley at Bozeman. Shoot biomass and grain yields for cereal test crops varied among sites, with barley ranking first or second, durum second to fourth, spring wheat first to third, and winter wheat first to fourth. The spring cereals received equal fertilizer N (67 to 101 kg N ha–1) while winter wheat received 5 to 23 kg N ha–1 greater than the spring cereals depending on site (Table 3). Despite higher fertilizer rates for winter wheat, grain yield was greater than spring wheat at two of five sites and less at two of five sites. The only consistent yield relationship among cereal crops was that durum wheat yielded less than spring wheat at all locations.


View this table:
[in this window]
[in a new window]
  		Table 8. Mean cereal test crop response for five parameters grown at five locations in Montana (columns left to right = wettest to driest).

 
Grain protein differences among cereal test crops also varied among sites, with barley ranking second to fourth, durum first, spring wheat first or second, and winter wheat first to third. Barley grain protein concentrations likely were too high for malt grade at three of five locations while wheat grain protein concentrations were lower than expected at Amsterdam, below the critical threshold value of 147 mg g–1 (dry matter basis) for N-sufficient yield in spring wheat (Engel et al., 1999). The grain N yield response was similarly variable as the grain yield response, with parallel crop rankings by site with the exceptions of barley at Amsterdam (less) and durum wheat at Bozeman (greater). Grain density was generally acceptable, except spring wheat at Denton, Havre, and Dutton had unacceptably low values, which likely would have resulted in price discounts. Thus, yield formation in spring wheat was more injured by drought than the other cereal test crops.


   SUMMARY AND CONCLUSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSION
REFERENCES
 
This research concluded that comparative productivity among alternative crops varied by site, indicating that diversification can mitigate production risk. There was an inconsistent yield advantage for early seeded cool-season crops over late-seeded warm-season crops. Additionally, cropping sequence benefits to cereal crops from broadleaf crops were observed only at sites with near-average growing season rainfall and not at sites experiencing severe drought. Cropping sequence differences between wheat and flax or pea as the previous crop were not explained by soil water but were related to differences in soil N despite the use of high N fertilizer rates for the cereal test crops. Finally, the four cereal test crops were affected similarly by previous crop stubbles, indicating that a single cereal test crop is sufficient for measuring cropping sequence response for cereal crops.


   ACKNOWLEDGMENTS
 
This research was supported by the Montana Wheat and Barley Committee and the Montana Agricultural Experiment Station. Gregg Carlson, Dave Wichman, Jody McConnell, Brad Gregoire, and Doug Ryerson supplied valuable technical assistance. We appreciate the farm field access provided by Rich Barber (Denton), Alec McIntosh (Havre), Matt Flikkema (Amsterdam), and Dale Schuler (Dutton). A special thanks goes to Doug Ryerson of Monsanto for providing seeding, spraying, and harvesting equipment for the Dutton site.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND CONCLUSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 Agron. J.Home page
P. R. Miller, D. E. Buschena, C. A. Jones, and J. A. Holmes
Transition from Intensive Tillage to No-Tillage and Organic Diversified Annual Cropping Systems
Agron. J., May 7, 2008; 100(3): 591 - 599.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
H. W. Cutforth, S. M. McGinn, K. E. McPhee, and P. R. Miller
Adaptation of Pulse Crops to the Changing Climate of the Northern Great Plains
Agron. J., November 6, 2007; 99(6): 1684 - 1699.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
R. L. Lemke, Z. Zhong, C. A. Campbell, and R. Zentner
Can Pulse Crops Play a Role in Mitigating Greenhouse Gases from North American Agriculture?
Agron. J., November 6, 2007; 99(6): 1719 - 1725.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
M. A. Liebig, D. L. Tanaka, J. M. Krupinsky, S. D. Merrill, and J. D. Hanson
Dynamic Cropping Systems: Contributions to Improve Agroecosystem Sustainability
Agron. J., June 5, 2007; 99(4): 899 - 903.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
D. L. Tanaka, J. M. Krupinsky, S. D. Merrill, M. A. Liebig, and J. D. Hanson
Dynamic Cropping Systems for Sustainable Crop Production in the Northern Great Plains
Agron. J., June 5, 2007; 99(4): 904 - 911.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
P. R. Miller, R. E. Engel, and J. A. Holmes
Cropping Sequence Effect of Pea and Pea Management on Spring Wheat in the Northern Great Plains
Agron. J., October 31, 2006; 98(6): 1610 - 1619.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
C. Chen, P. Miller, F. Muehlbauer, K. Neill, D. Wichman, and K. McPhee
Winter Pea and Lentil Response to Seeding Date and Micro- and Macro-Environments
Agron. J., October 31, 2006; 98(6): 1655 - 1663.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Miller, P. R.
Right arrow Articles by Holmes, J. A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Miller, P. R.
Right arrow Articles by Holmes, J. A.
Agricola
Right arrow Articles by Miller, P. R.
Right arrow Articles by Holmes, J. A.
Related Collections
Right arrow Crop Rotation Systems
Right arrow Dryland Cropping Systems

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