|
|
||||||||
a Dep. of Land Resour. and Environ. Sci., Montana State Univ., P.O. Box 173120, Bozeman, MT 59717-3120
b Semiarid Prairie Agric. Res. Cent., Agric. and Agri-Food Can., Swift Current, SK, Canada S9H 3X2
* Corresponding author (pmiller{at}montana.edu)
Received for publication May 6, 2002.
| ABSTRACT |
|---|
|
|
|---|
Abbreviations: WUE, water use efficiency
| INTRODUCTION |
|---|
|
|
|---|
Farmers frequently ask if soil texture (clay vs. loam) influences the magnitude of the effect of pulse crops on other broadleaf crops. A literature review did not reveal any published papers on this topic. Townley-Smith (1994) reported that in east-central Saskatchewan, Canada (subhumid), yields were reduced for pea, canola, and spring wheat when grown on their own stubbles. The objective of this study was to compare the effects of chickpea, lentil, and pea stubbles on yield and quality of wheat, mustard or canola, and lentil or pea when grown on soils with clay and loam textures.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Experimental Design and Field Operations
All 3-yr crop sequences were initiated on tilled fallow at independent sites each year. A three-replicate split-plot randomized complete block design was used, with five Year-1 crops (wheat, pea, lentil, desi chickpea, and Oriental mustard) as the 12- by 16-m main plots, three Year-2 test crops (wheat, Oriental mustard or canola, and lentil or pea) occurring in 4- by 16-m subplots, and durum wheat (Triticum durum L.) grown on all plots in Year 3 (Gan et al., 2003). This paper reports the test crop responses from the second year of the 3-yr crop sequence. The experimental design included a 4- by 16-m fallow control plot (2 yr of consecutive fallow) in the center of each block (systematic arrangement). Only the wheat test crop was grown on the fallow control to provide a reference for soil water use and water use efficiency (WUE). Cultivars, fungicides, and inoculants used were described previously (Miller et al., 2003). Imazethapyr-{(±)-2-[4,5-dihydro-4-methyl-4-(1-methylethyl)-5-oxo-1H-imadazol-2-yl]-5-ethyl-3-pyridine-carboxylic acid} and tolerant-treated (disease and insect) canola (45A71) was planted at a rate of 240 seeds m-2 in 1998 and 1999.
All crops were seeded using a modified hoe drill with a cone and spinner assembly for precise seed placement. Seeding dates ranged from late April to mid-May (Table 1) when soil moisture conditions permitted. Wheat and canola or mustard fertilizer N was banded midway between pairs of seeded rows (20-cm row spacing) through double-disc openers running 1 m ahead of the hoe openers while fertilizer P was placed in the seed row. Pea or lentil received no fertilizer N aside from the small amount accompanying P2O5 in a fertilizer formulation of 11510. Less fertilizer N was applied to the Year-1 pulse crop stubbles to account for measured postharvest (061 cm depth) differences in mean soil NO3N and expected net soil N mineralization (Table 2). Nitrogen credits associated with pulse crop stubbles were calculated according to the Saskatchewan Soil Testing Lab, Saskatoon, SK, Canada (B. Green, personal communication, 1995), as follows:
![]() |
|
|
,
,
-trifluoro-2,6-dinitro-N, N-dipropyl-p-toluidine] (1.11.4 kg a.i. ha-1) was applied in the spring for broad-spectrum weed control in all crops but wheat. Trifluralin was incorporated with two passes of a field cultivator in 1997 and 1998 and was managed as a surface broadcast application in 1999 to protect seedbed moisture. Additionally, commercially available postemergent herbicides were used to effect weed control in all test crops.
Soil and Crop Data
Year-2 subplots were sampled before seeding and after harvest (Table 1) by taking two cores per plot to a depth of 122 cm; dividing into 0- to 30-, 30- to 61-, 61- to 91-, and 91- to 122-cm segments down the profile; and bulking segments by depth. Soil samples were analyzed for water (gravimetrically) and NO3N (Hamm et al., 1970). Soil bulk densities were estimated from a previously conducted study (Campbell et al., 1983). Bulk densities at Stewart Valley were measured by grid sampling in 1997 and 1999. The 1997 and 1998 sites at Stewart Valley were 100 m apart. Bulk densities were used to convert gravimetric values to a volumetric basis. Runoff was considered insignificant because the field sites are level, and deep percolation was considered negligible.
A plot combine was used to harvest a 1.22- by 16-m area for all crops except canola. For canola, a 1.22- by 16-m area was swathed before harvesting. Seed yield and seed N concentration were reported on a dry matter basis. Seed N concentration was determined using the standard micro-Kjeldahl method (AOAC, 2000). Seed N concentration was multiplied by 5.7 for wheat and by 6.25 for oilseed and pulse crops to determine seed protein concentration. Water use efficiency was determined as the dry matter seed yield divided by total water used, which was defined as the difference between spring and fall soil water measurements plus all rain received between spring and fall sampling dates. The difference between N harvested in the seed and N inputs was defined as the apparent N margin. Nitrogen inputs were defined as fertilizer N minus spring NO3N + fall NO3N. The N margin was used to compare N contributions among crop stubbles.
Statistical Analyses
Each test crop was analyzed independently. Sites were considered fixed and years random effects. The data were analyzed with the GLM procedure of SAS (SAS Inst., 1988, 549640). Analyses of variance were run for the full model (Table 3). Consistent with the objective of comparing pulse cropping sequence effects in different soil textures, mean comparisons were made within sites where statistical differences occurred (wheat only). Where crop x year was significant, it was used as the error term for testing the crop effect and for calculating an appropriate standard error. Due to the large proportion of variance due to year, and its interaction with site and crop stubble, mean comparisons were mainly made within years and sites. A P value of 0.10 was used for testing the significance of interaction terms (F test) and for testing mean differences with the protected LSD procedure. Cropping sequence data for the canola test crop at Stewart Valley in 1998 were omitted from analyses due to severe grasshopper (Malanoplus bivittatus Say) predation. Cropping sequence data for the canola and pea test crops at Stewart Valley in 1999 were also omitted from analyses due to severe deer (Odocoileus virginianus) grazing and treated as missing data. The systematic fallowwheat sequence was not included in any statistical analysis, but data were reported to assist with cropping sequence interpretation for wheat.
|
| RESULTS AND DISCUSSION |
|---|
|
|
|---|
Wheat Test Crop
The previous crop influenced wheat yields on both the clay (Stewart Valley) and silt loam (Swift Current) soils (Tables 4 and 5). In both soils, wheat yields were equal on all broadleaf crop stubbles and were lowest on wheat stubble. In the clay soil at Stewart Valley, broadleaf crop stubble increased wheat yield by 35%, and in the loam soil at Swift Current, broadleaf crop stubble increased yield 14%. This apparent difference between soil types is not readily explained by preseeding soil water or N content, reported by Miller et al. (2003), especially since fertilizer N was adjusted by crop stubble to account for estimated differences in soil available N (Table 2). The low yields in the wheat-following-wheat rotation may partially result from increased soil-borne plant pathogens as has been reported in other cropping sequence studies (Stevenson and van Kessel, 1996; Beckie and Brandt, 1997). Yield reductions were most on the clay soil at Stewart Valley in 1997 and 1999 and on the silt loam soil at Swift Current in 1999. At these sites, precipitation during May was high, which resulted in cool, wet soil conditions ideal for many known cereal root and crown pathogens. In the drier year of 1998, water conservation differences among stubbles were more important. For example, in 1998 at Stewart Valley, wheat yields were greatest when grown on pea stubble; intermediate on lentil, mustard, and wheat stubbles; and least when grown on desi chickpea stubble. This response was positively correlated with soil water use by the wheat test crop [r = 0.70 (P < 0.01), data not shown].
|
|
The results from this study agree partially with those of Townley-Smith (1994) where wheat yielded greatest on pea stubble, least on wheat stubble, and intermediate on canola stubble though the low soil N status maintained in their study likely magnified the cropping sequence effects from the pea stubble. The adjustment of fertilizer N by crop stubble likely prevented a difference occurring between the pulse and mustard stubbles in this study. This study also agrees partially with the preliminary results from a cropping sequence study in a semiarid location in North Dakota (USDA-ARS Northern Great Plains Res. Lab., 2002). There, the rotational effect on spring wheat was reported to be similar for pea and canola stubbles (a uniform rate of fertilizer N was applied under all crop stubbles) and 10 to 14% greater than when grown on wheat stubble.
The economic break-even yield for wheat grown on wheat stubble vs. that grown on fallow has been reported to be in the range of 67 to 84% (Zentner et al., 1986). In this study, wheat yields when grown on wheat stubble averaged 67%, and ranged from 43 to 83%, of that grown on the fallow control, with no consistent difference between sites. This wheat stubblefallow relationship was slightly lower, on average, than the comparative value of 71% reported by Zentner et al. (2001) from a 19-yr rotation study conducted at Swift Current on the same silt loam soil type. However, wheat yields on broadleaf crop stubbles averaged 81% of fallow, indicating greater profitability. This 3-yr study was conducted under generally wetter conditions than the longer-term study, and the fallow control in this study was a double fallow (32 mo), which created a 2-yr break from cereal plant pathogens and might have had greater soil N and water availability compared with normal 20-mo fallow.
Wheat protein differences among crop stubbles were observed only on the clay soil. In 1997, wheat grown on pulse crop stubbles had 16% greater grain protein than wheat grown on wheat stubble. In 1998, wheat grown on broadleaf crop stubbles had 15% greater protein than wheat grown on wheat stubble. Based on minimum grain protein concentration of 123 to 147 g kg-1 as an indicator of N sufficiency (Engel et al., 1999; Selles and Zentner, 2001), N sufficiency to attain optimal yield was achieved only in 1998 when grain protein varied from 125 to 158 g kg-1 and 149 to 175 g kg-1 on the clay and loam soils, respectively. On the loam soil, grain protein averaged 110 and 116 g kg-1 in 1997 and 1999, respectively, while on the clay soil, grain protein averaged 114 and 112 g kg-1 in 1997 and 1999, respectively. These results indicate that wheat would have been most responsive to N in 1997 and 1999. As expected in a N-deficient scenario, wheat protein differences among crop stubbles generally were not observed, with one exception. On the clay soil in 1997, seed protein concentrations of wheat grown on wheat stubble were less than those for wheat grown on the pulse crop stubbles. This difference may have resulted from denitrification resulting from saturated soil conditions that occurred in late May. Mineral N released from decomposition of pulse crop residues after these wet soil conditions abated would not have been subject to loss due to denitrification.
Differences in the N margin for the wheat test crop reflected N contributions from the pulse crop stubbles that became apparent after spring soil sampling. On the clay soil, the pulse crop stubbles averaged 11 to 15 kg ha-1 greater N margin compared with mustard stubble. At Swift Current, a similar trend was evident where the N margin for pulse crop stubbles averaged 6 to 12 kg ha-1 greater than the mustard stubble although a large crop x year interaction prevented significant differences. These values were consistent with the magnitude of estimated N credits for pulse crops used to apply different fertilizer N rates in this study.
On the clay soil, wheat grown on wheat stubble had lower WUE values than wheat grown on all broadleaf stubbles in 1997 and 1999. On the silt loam, a similar nonsignificant trend (P = 0.11) occurred in 1999. These results suggest that reduced wheat productivity in the wheat stubble was not related to water availability and that some unknown factor compromised the productivity per unit of available water for wheat grown on wheat stubble. This would be consistent with plant disease interference, as discussed for the wheat yield response above. The average WUE values were greater on the clay soil (average = 9.5 kg ha-1 mm-1; range = 6.014.1 kg ha-1 mm-1) than on the loam soil (average = 7.7 kg ha-1 mm-1; range = 5.39.7 kg ha-1 mm-1), reflecting greater soil water storage under the clay soil (Miller et al., 2003). The range of WUE values on the loam soil was consistent with those reported previously for wheat in a previous study at the same site (5.78.1 kg ha-1 mm-1; Miller et al., 2001).
Brassica sp. Oilseed Test Crop (Oriental mustard in 1997 and Argentine canola in 19981999)
The mean seed yield for the Brassica sp. oilseed test crop differed for the only year on the clay soil and for 1 of 3 yr on the silt loam soil (Table 4). In 1997, on the clay soil, mustard yield averaged 79% more when grown on pea or lentil stubbles and averaged 36% more on chickpea or mustard stubbles compared with wheat stubble (Table 6)
. In 1998, on the silt loam soil, canola yield averaged 29% more when grown on pea or lentil stubbles and 32% less when grown on mustard stubble compared with wheat stubble. The Brassica sp. oilseed test crop response differed from wheat in that mustard or canola generally did not suffer yield losses when grown on mustard stubble. Townley-Smith (1994) had different results and reported that canola yielded least when grown on its own stubble in a subhumid region of east central Saskatchewan where canola had a long production history. Differences between the studies may result from differential amounts of plant diseases (Petrie, 1995; Thompson et al., 1995).
|
Although seed protein of the Brassica sp. oilseed test crop differed among crop stubbles at only one site, the apparent N margin was highest when grown on pea stubble in three of the four site-years. On the clay soil, the apparent N margin for mustard was 32 kg ha-1 greater when grown on pea stubble compared with the average of all other crop stubbles, which did not differ from each other. On the loam soil in 1998, the N margin for canola was 19 kg ha-1 greater when grown on pea stubble compared with the average of all other crop stubbles. In 1999, the apparent N margin for canola averaged 16 kg ha-1 greater on pea or lentil stubble compared with that grown on wheat stubble. This might reflect greater soil N availability under pea stubble than was estimated for the fertilizer N adjustment, or greater soil water availability, consistent with the observations for seed yield (Miller et al., 2003).
Pulse Test Crop
The pulse test crop was expected to respond to residual effects of soil resident rhizobia from inoculation of the previous pea and lentil crops, in addition to crop stubble effects on soil N and water. On the silt loam soil at Swift Current in 1997, lentil grown on pea stubble yielded 26% greater than the average of all other crop stubbles (Table 7). In 1998, pea grown on lentil and wheat stubbles yielded 43 and 28% greater, respectively, than the average of the other crop stubbles. However, the greatest variability among crop stubbles occurred on the clay soil in 1998. There, pea yielded most when grown on wheat stubble, nearly double that grown on chickpea stubble, on which the least pea yield occurred. Two main crop stubble effects were evident. First, because lentil and pea share a common species of rhizobia, each yielded most when grown on the other's stubble but not when grown on its own stubble (i.e., lentil on lentil stubble and pea on pea stubble). Despite application of recommended rates of peat granular inoculant for the pulse test crop, the vegetation of pea or lentil that was grown on pea or lentil stubbles had a visibly richer green hue before flowering compared with that grown on the other stubbles (personal observation). This suggested that resident soil rhizobia from the previous pea or lentil crop accelerated early N2 fixation. Supporting this observation, seed protein for the lentil and pea test crops was 9 and 23% greater when grown on lentil or pea stubble compared with the average of other crop stubbles on the silt loam soil in 1997 and 1998, respectively. A similar nonsignificant trend (P = 0.12) was observed for the clay soil at Stewart Valley in 1998 where pea grown on lentil or pea stubble had 6% more seed protein than pea grown in other crop stubbles. McConnell et al. (2002) showed that visibly greener pea vegetation before flowering was associated with increased seed yield and seed protein content in Montana. Similarly, McKenzie et al. (2001) showed increased grain protein content in pea where a positive response to rhizobial inoculation occurred in Alberta, Canada. However, obvious growth advantages due to timely rhizobial nodulation did not translate into increased yield where lentil or pea was grown on its own stubble. Foliar diseases in the pea or lentil were not evident in any site-year; root diseases were not monitored. It is expected that lentillentil or peapea crop sequences would lead to disease problems (Krupinsky et al., 2002).
|
Water use efficiency for pulse test crops differed among crop stubbles for 1 of 2 yr on the clay soil and for all 3 yr on the silt loam soil. Similar to that observed for wheat and mustard, the WUE of pulse crops was greater in the clay (6.7 kg ha-1 mm-1) than the loam soil (5.0 kg ha-1 mm-1). The range of measured WUE values was wider than that reported by Miller et al. (2001). Miller et al. (2001) reported that WUE for lentil ranged from 2.7 to 5.3 kg ha-1 mm-1, and for pea, WUE values ranged from 5.9 to 10.8 kg ha-1 mm-1.
| CONCLUSIONS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
| NOTES |
|---|
|
|
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. W. Gesch, B. S. Sharratt, and K.-I. Kim Yield and Water Use Response of Cuphea to Irrigation in the Northern Corn Belt Crop Sci., August 7, 2009; 49(5): 1867 - 1875. [Abstract] [Full Text] [PDF] |
||||
![]() |
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] |
||||
![]() |
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] |
||||
![]() |
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] |
||||
![]() |
J. M. Krupinsky, D. L. Tanaka, S. D. Merrill, M. A. Liebig, M. T. Lares, and J. D. Hanson Crop Sequence Effects on Leaf Spot Diseases of No-Till Spring Wheat Agron. J., June 5, 2007; 99(4): 912 - 920. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. D. Hanson, M. A. Liebig, S. D. Merrill, D. L. Tanaka, J. M. Krupinsky, and D. E. Stott Dynamic Cropping Systems: Increasing Adaptability Amid an Uncertain Future Agron. J., June 5, 2007; 99(4): 939 - 943. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. C. Nielsen, P. W. Unger, and P. R. Miller Efficient Water Use in Dryland Cropping Systems in the Great Plains Agron. J., March 1, 2005; 97(2): 364 - 372. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. R. Miller and J. A. Holmes Cropping Sequence Effects of Four Broadleaf Crops on Four Cereal Crops in the Northern Great Plains Agron. J., January 1, 2005; 97(1): 189 - 200. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. S. Sharratt and R. W. Gesch Water Use and Root Length Density of Cuphea spp. Influenced by Row Spacing and Sowing Date Agron. J., September 1, 2004; 96(5): 1475 - 1480. [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 | |||