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


     

Published in Agron J 99:1710-1718 (2007)
DOI: 10.2134/agronj2006.0314s
© 2007 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow An erratum has been published
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 Google Scholar
Google Scholar
Right arrow Articles by Walley, F. L.
Right arrow Articles by Lafond, G. P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Walley, F. L.
Right arrow Articles by Lafond, G. P.
Agricola
Right arrow Articles by Walley, F. L.
Right arrow Articles by Lafond, G. P.
Related Collections
Right arrow Other Legumes
Right arrow Crop Rotation Systems
Right arrow Dryland Cropping Systems
Right arrow Symbiosis
Right arrow Nitrogen
Right arrow Crop Systems
Right arrow Nutrient Management
Right arrow Production Agriculture
Right arrow Soil Fertility and Productivity

Symposium Papers

Nitrogen Economy of Pulse Crop Production in the Northern Great Plains

Fran L. Walleya,*, George W. Claytonb, Perry R. Millerc, Patrick M. Carrd and Guy P. Lafonde

a Dep. of Soil Science, Univ. of Saskatchewan, 51 Campus Drive, Saskatoon, SK, S7N 5A8, Canada
b Agriculture & Agri-Food Canada, P.O. Box 3000, Lethbridge, AB, Canada T1J 4B1
c Land Resources and Environmental Sciences, 334 Leon Johnson Hall, Montana State University, Bozeman, MT 59717-3120
d Dickinson Res. Ext. Ctr., North Dakota State University, 1041 State Avenue, Dickinson, ND 58601
e Agriculture and Agri-Food Canada, Indian Head Research Farm, Box 760, Indian Head, SK S0G2K0

* Corresponding author (fran.walley{at}usask.ca)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Previously published data were used to examine the N economy of pulse crops typically grown on the Northern Great Plains with the goal of assessing the potential contribution of field pea (Pisum sativum L.), lentil (Lens culinaris Medik.), chickpea (Cicer arietinum L.), common bean (Phaseolus vulgaris L.), and faba bean (Vicia faba L.) to soil N accretion. Incremental changes in soil N associated with the pulse crops (i.e., the nitrogen increment, Ninc), were strongly correlated to N2 fixation and were highly variable. Data suggest that crops that can achieve relatively high levels of N2 fixation, such as faba bean, field pea, and lentil are more likely to contribute positively to the overall N economy, particularly when a cropping system is evaluated over a long term. In contrast, pulse crops that typically achieve only modest levels of N2 fixation such as desi and kabuli chickpea and common bean are more likely to be either N neutral or contribute to a soil N deficit. Because of extreme variability in levels of N2 fixation achieved, presumably reflecting variability in soil productivity as well as variations in local climate and weather, the Ninc of pulse crops likewise is highly variable. Thus, the N contribution to a subsequent crop is difficult to predict with any certainty, particularly on a yearly or short-term basis.

Abbreviations: %Ndfa, percentage of nitrogen derived from the atmosphere • NHI, nitrogen harvest index • Ninc, nitrogen increment

Nitrogen Economy of Pulse Crop Production in the Northern Great Plains

Fran L. Walleya,*, George W. Claytonb, Perry R. Millerc, Patrick M. Carrd and Guy P. Lafonde

a Dep. of Soil Science, Univ. of Saskatchewan, 51 Campus Drive, Saskatoon, SK, S7N 5A8, Canada
b Agriculture & Agri-Food Canada, P.O. Box 3000, Lethbridge, AB, Canada T1J 4B1
c Land Resources and Environmental Sciences, 334 Leon Johnson Hall, Montana State University, Bozeman, MT 59717-3120
d Dickinson Res. Ext. Ctr., North Dakota State University, 1041 State Avenue, Dickinson, ND 58601
e Agriculture and Agri-Food Canada, Indian Head Research Farm, Box 760, Indian Head, SK S0G2K0

* Corresponding author (fran.walley{at}usask.ca)

Received for publication August 11, 2006.
Previously published data were used to examine the N economy of pulse crops typically grown on the Northern Great Plains with the goal of assessing the potential contribution of field pea (Pisum sativum L.), lentil (Lens culinaris Medik.), chickpea (Cicer arietinum L.), common bean (Phaseolus vulgaris L.), and faba bean (Vicia faba L.) to soil N accretion. Incremental changes in soil N associated with the pulse crops (i.e., the nitrogen increment, Ninc), were strongly correlated to N2 fixation and were highly variable. Data suggest that crops that can achieve relatively high levels of N2 fixation, such as faba bean, field pea, and lentil are more likely to contribute positively to the overall N economy, particularly when a cropping system is evaluated over a long term. In contrast, pulse crops that typically achieve only modest levels of N2 fixation such as desi and kabuli chickpea and common bean are more likely to be either N neutral or contribute to a soil N deficit. Because of extreme variability in levels of N2 fixation achieved, presumably reflecting variability in soil productivity as well as variations in local climate and weather, the Ninc of pulse crops likewise is highly variable. Thus, the N contribution to a subsequent crop is difficult to predict with any certainty, particularly on a yearly or short-term basis.

Abbreviations: %Ndfa, percentage of nitrogen derived from the atmosphere • NHI, nitrogen harvest index • Ninc, nitrogen increment


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
ONCE CONSIDERED A SPECIAL CROP in cereal-based cropping systems in the Northern Great Plains, pulse crops are now routinely grown for food, feed, and forage. Pulse production in the Northern Great Plains accounted for more than three million seeded hectares in 2005, and was dominated by field pea, lentil, common bean, and chickpea (National Agricultural Statistics Service, 2006; Statistics Canada, 2006). The area grown to pulse crops continues to increase.

One important benefit of pulse crop production is the ability of these crops to form symbiotic associations with N2-fixing Rhizobium bacteria, thereby reducing their dependence on inorganic N fertilizers. It has long been recognized that N fertilizer represents one of the main economic and energy costs of continuous cropping systems in the Northern Great Plains (Zentner et al., 1984, 2002a, 2002b, 2004). Given recent escalation in N fertilizer prices, crop rotations that include N2-fixing pulse crops offer an attractive alternative to cereal-based rotations that typically rely exclusively on N fertilizer. Generally it is assumed that a well-inoculated pulse crop can fix sufficient quantities of N to eliminate the need for N fertilizer inputs in the year of the pulse crop. Moreover, several reports have suggested that some pulse crops can enhance the N supplying power of the soil over time, presumably by increasing N inputs via N2 fixation relative to crop removal, thereby resulting in a net addition of N to the soil, or a positive Ninc (Soper and Grenier, 1987; Beckie and Brandt, 1997; Van Kessel and Hartley, 2000; Gan et al., 2003). However, pulse crops have high grain protein contents; thus, the net export of N in the grain frequently is high and often exceeds the total amount of fixed N in the biomass (Beck et al., 1991). A positive Ninc is achieved only if the percentage of nitrogen derived from the atmosphere (%Ndfa) is greater than the nitrogen harvest index (NHI). It follows that pulse crops are likely to contribute significantly to the soil N pool only when N2 fixation is relatively high and/or the NHI is relatively low, although low grain yield or quality is implied by the latter (Van Kessel and Hartley, 2000).

Published estimates of the contribution of pulse crops to a positive Ninc in long-term cereal-based cropping systems vary widely, reflecting in part the varied impact of soil and environmental conditions that influence N inputs via N2 fixation, and outputs via N harvested in the grain (Van Kessel and Hartley, 2000). Not surprisingly, studies conducted on the Northern Great Plains to quantify pulse crop N-benefits associated with pulse crop production have resulted in highly variable estimates (Stevenson and Van Kessel, 1996a, 1996b; Beckie and Brandt, 1997; Beckie et al., 1997). Consequently, on the basis of individual research studies, it is difficult to draw any conclusions regarding the contribution of pulse crops to the N economy of cropping systems in the Northern Great Plains. For example, Stevenson and Van Kessel (1996b) reported that grain yield of wheat grown on field pea stubble in the Black soil zone of Saskatchewan was 43% greater than wheat grown on wheat stubble, and accumulated an extra 27 kg N ha–1; however, only 6 to 14 kg ha–1 of the additional N was directly related to additional N contributed by field pea as compared with wheat residue. Using isotopic A-value estimates as an indicator of soil N supplying power, they estimated that only 8% of the rotation benefit associated with field pea stubble could be explained by direct N benefits, with the remaining benefit largely due to the reduction of wheat root diseases in the subsequent crop. Beckie and Brandt (1997), working in the same soil zone, observed that the residual effect of field pea stubble was approximately 37 kg N ha–1 in the first year of their study and 18 kg N ha–1 in the second, and suggested that the enhanced N acquisition by the nonlegume crop was due overwhelmingly to direct N contributions from field pea residue. Their results led them to suggest an appropriate N credit (i.e., N fertilizer replacement value) for field pea to a succeeding nonlegume in the moist Black soil zone of 15 kg N ha–1 for every 1000 kg pea grain produced.

Uncertainty regarding the N contribution is exacerbated by methodological constraints. Specifically, most N balance studies do not account for N in the root systems, and thus undoubtedly underestimate the true N contribution from N2 fixation. There are few studies on which to base estimates of root N contributions, largely because it is difficult and time-consuming to obtain reliable root measurements, particularly for field-grown crops. According to Soon and Arshad (2002), field pea returned 31 to 41 kg N ha–1 in straw, whereas only 2 to 3 kg N ha–1 were returned in roots, suggesting a relatively minor root N contribution; however, in their study roots were excavated only to 12.5 cm and thus it is likely that the total root N contribution was underestimated.

Bremer (1991) estimated that roots of Lens culinaris ‘Laird’ excavated to 90 cm contributed ~19 kg N ha–1, equivalent to ~14% of the total biomass N. This estimate is consistent with others for lupin (Unkovich et al., 1994) and field pea (Armstrong et al., 1994), in which root biomass accounted for 15 and 12% of peak biomass N (midflowering–early pod fill), respectively. Van Kessel (1994) used Bremer's estimate of 19 kg N ha–1 to determine that lentil provided at least 59 kg N ha–1 to the soil.

In addition to the contribution of root biomass to potential N accretion, Sawatsky and Soper (1991) suggested that a considerable amount of N is lost during the natural growth cycle of an annual crop in the form of root lysates and roots exudates. Using a split-root technique, they assessed the loss of N from field pea roots and concluded that rhizodeposited N accounted for between 22 and 46% of the total belowground N budget. Accordingly, they suggested that estimates of N2 fixation based on aboveground biomass may underestimate the quantity of fixed N by ~10% due to the loss of rhizodeposited N during plant growth. They acknowledged, however, that the extent to which rhizodeposited N might be recycled back to plants grown in the field is not known. Mayer et al. (2003) used a 15N stem labelling method to estimate rhizodeposition and reported that at maturity N rhizodeposition constituted ~13% of total plant N for faba bean and field pea. They concluded that N rhizodeposition of grain legumes represents a significant N pool and is likely to contribute to more positive N balances for grain legumes when taken into consideration.

The enhanced N availability following pulse crop production has led many commercial soil testing labs to reduce fertilizer N recommendations for crops grown on pulse stubble by assigning N credits, either on the basis of pulse crop yield or residue quality (i.e., cereal versus pulse residue). However, the manner in which N credits are calculated, and the ultimate magnitude of these credits varies widely between labs, in part reflecting the controversy that currently exists in the literature regarding the true nature of the N benefit.

Our review is intended to estimate the contribution of pulse crops typically grown in the Northern Great Plains to the soil N balance using previously published research in which N2 fixation was reported together with grain yield. Our approach was similar to that previously used by Evans et al. (2001), in which they collated information on the effects of cool-season legumes on soil N balance in Australia.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
The potential addition or removal of N from the cropping system, or Ninc was calculated by subtracting N harvested in the grain (Ngr) from the total N input derived from N2 fixation (Nfix) according to Evans et al. (2001):

Formula 1[1]

Published data used to calculate Ninc were limited to Northern Great Plains research in which both grain yield and N2 fixation values were reported as a minimum requirement (Table 1 ). In most instances, estimates of N2 fixation were determined using 15N enriched or natural abundance methods, although a limited number of studies used a difference method [i.e., a comparison of the N uptake in the fixing crop to a nonfixing reference crop, e.g., Soon and Arshad (2004)] to estimate N2 fixation.


View this table:
[in this window]
[in a new window]

  		Table 1. Published sources of data for estimating N increments of pulse crops grown on the Northern Great Plains.

 
Estimates of total N2 fixed included fixed N in grain, shoot (at harvest), root biomass, and root exudates. Although many publications included estimates of either grain N uptake or grain protein yield together with shoot biomass estimates or harvest index (from which shoot biomass was subsequently calculated), few publications included all of the required data to calculate total fixed N using measured values. Consequently, our data set was constructed, in part, using mean estimates calculated from available measured published data. For example, although all of the lentil publications either reported grain N or provided data from which grain N could be calculated directly (e.g., grain yield and percentage grain N or percentage grain protein), some publications did not include measurements of straw N concentration (e.g., Matus et al., 1997; Cowell et al., 1989). For these latter cases, the mean value for straw N concentration calculated using data provided in the other lentil publications (e.g., Rennie and Dubetz, 1986; Bremer, 1987, 1991; Van Kessel, 1994; Moolecki, 2000) were used to estimate total straw N for the missing values.

We assumed that root N of all pulse crops accounted for 14% of the total N contribution in the biomass, according to Bremer (1991). We viewed this as a relatively conservative estimate. In studies of lupin, Unkovich et al. (1994) reported that root biomass accounted for 15% of peak biomass N and as much as 25% of total N in recoverable plant N at grain harvest. Similarly, Armstrong et al. (1994) reported that root biomass N of field pea accounted for as much as 12% of total plant N at peak biomass and ~25% of recoverable plant N after grain harvest. Additionally, we assumed that rhizodeposited N accounted for an additional 10% of the total N contribution in the biomass (Sawatsky and Soper, 1991). Although it is not known how much of the rhizodeposited N is recycled, acknowledging the potential contribution of rhizodeposited N at worst contributes to an overestimation of the contribution of N2 fixation to a positive N balance.

Although the total N contributed from Nfix included N in grain, shoot residues (i.e., aboveground straw), roots, and root exudates, in most instances published N2 fixation estimates were limited to grain measurements only, and thus the fixed N in shoot residues, root, and root exudates was estimated. For calculation purposes, we assumed that %Ndfa did not differ between residue and grain. This assumption likely overestimates the contribution of N in the shoot residues from N2 fixation because legumes typically utilize soil- and fertilizer-N sources during early growth stages when vegetative material is being laid down (Van Kessel, 1994). Furthermore, legumes start relying on atmospheric N sources once the available soil N pool is depleted and an active N2-fixing association has been established. Consequently, %Ndfa in residue and grain is likely to differ. Van Kessel (1994) examined partitioning between leaves, stem, pods, and seed of lentil and reported that at maturity %Ndfa in lentil pods and seed was as high as 91%, whereas significantly less (i.e., 79 to 80%) N in the leaves and stems was derived from N2 fixation. Thus, our assumption that %Ndfa did not differ between grain and residue, like our earlier assumption regarding rhizodeposited N, at worst may have contributed to an overestimation of N from N2 fixation, and consequently, an inflated Ninc.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
The contribution of N2 fixation to pulse crop N nutrition was highly variable (Fig. 1 ), reflecting the variable conditions and experimental treatments imposed in the different studies. Variability in N2 fixation is not unexpected and it is well recognized that N2 fixation is sensitive to numerous environmental and edaphic factors. Moreover, high short-range spatial variability in N2 fixation has been demonstrated (Walley et al., 2001) and may have contributed to the overall variability in estimates of N2 fixation. Several reviews have examined the many factors controlling N2 fixation in grain legumes (e.g., Brockwell and Bottomley, 1995; Brockwell et al., 1995) including an excellent discussion of the impact of agricultural management factors on N2 fixation (Van Kessel and Hartley, 2000). Available inorganic N, in particular, often is cited as one of the most important factors controlling N2 fixation. Indeed, where inorganic N levels are sufficient or exceed crop N requirements, little N2 is fixed irrespective of other factors including the effectiveness of the rhizobia–host association (Van Kessel and Hartley, 2000). Unfortunately, we were unable to assess the impact of soil inorganic N on N2 fixation because of inconsistencies in the manner by which soil N sources were reported. For example, different publications reported different sampling depths, or reported soil N concentration but not bulk density. In some instances, organic N alone was reported. Given the importance of available inorganic N to N2 fixation, standardized sampling depths (e.g., 0- to 0.3-, 0.3- to 0.6-, and 0.6- to 1.2-m depths) for which inorganic (nitrate and ammonium) and organic N (i.e., potentially available N) are reported would be of considerable value in N2 fixation literature.


Figure 1
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 1. Boxplots representing the proportion of N derived from N2 fixation according to published estimates of %Ndfa. Numbers reported for each pulse crop indicate the total number of published means comprising the estimates. The thick dark line indicates the median value, the box represents 50% of all data and the whiskers contain the remaining 50% of the data.

 
In general, faba bean achieved the highest median levels of N2 fixation (88% Ndfa), whereas median values for both common bean and kabuli chickpea were <45 %Ndfa (Fig. 1). Desi chickpea and field pea shared similar median %Ndfa values (~55%), whereas the median %Ndfa value for lentil was ~60%. These estimates of %Ndfa are similar to other reports. For example, Evans et al. (2001) reported an Australian national average of 66 %Ndfa (± 2.0 SE) for field pea and 43 %Ndfa (± 4.1 SE) for chickpea.

When data from all pulses were combined, the %Ndfa significantly influenced the Ninc (Fig. 2 ) with the regression of the Ninc described by the linear relationship Ninc = –60.5 + 1.27(%Ndfa), r2 = 0.63, P < 0.01. Setting the Ninc to zero, positive increments were achieved when N2 fixation accounted for, on average, >48% of pulse crop N. Different crops, however, differed in the level of %Ndfa required to achieve a positive Ninc (Table 2 ). For example, field pea, lentil, and desi chickpea achieved levels of N2 fixation that favored a positive Ninc in ~50% or more of the cases, with required values equal to 46.7, 47.8, and 55.2 %Ndfa, respectively. Data suggest that lentil and field pea, in particular, have the greatest likelihood of contributing to a positive N balance as mean %Ndfa values for both crops typically exceeded levels required to achieve a positive Ninc, (i.e., 55.0 and 59.9 %Ndfa, respectively). In contrast, the mean values for %Ndfa typically achieved by common bean and kabuli chickpea were well below the values of 52.1 and 57.0 %Ndfa, respectively, required to achieve a positive Ninc. Faba bean consistently contributed to a positive Ninc and fixed a greater proportion of N than was required (>65.3 %Ndfa) to achieve a positive Ninc in all cases (n = 10) reported. Evans et al. (2001) similarly observed that the N balance was positively correlated to the proportion of N derived from N2 fixation and reported that positive N balances typically were achieved at %Ndfa values (termed "Pfix" in their publication) ranging from 41.9 to 44.2% for field pea, lupin, and chickpea.


Figure 2
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 2. Fig. 2. Relationship between N2 fixation (%Ndfa) and the N increment (kg N ha–1). The overall relationship is described by a linear equation as follows: Ninc = 1.27 x %Ndfa – 60.5, r2 = 0.63, P ≤ 0.01. For individual crops, linear relationships are described as follows: pea Ninc = 1.78 x %Ndfa – 83.3, r2 = 0.47, P ≤ 0.05; lentil Ninc = 1.18 x %Ndfa – 56.4, r2 = 0.47; P ≤ 0.05; desi chickpea Ninc = 0.82 x %Ndfa – 45.3, r2 = 0.55, P ≤ 0.05; kabuli chickpea Ninc = 0.66 x %Ndfa – 37.5, r2 = 0.57, P ≤ 0.01; bean Ninc = 0.66 x %Ndfa – 34.4, r2 = 0.52, P ≤ 0.01; faba bean Ninc = 3.67 x %Ndfa – 240.2, r2 = 0.65, P ≤ 0.01.

 

View this table:
[in this window]
[in a new window]

  		Table 2. Contribution of N from N2 fixation (Ndfa) required to achieve a positive N increment according to linear regression equations describing the relationship between N2 fixation, given as percentage N2 fixed (%Ndfa) and total N2 fixed (kgNdfa), and the N increment (kg N ha–1).

 
It is interesting that the Australian data indicated that levels of %Ndfa required to achieve a positive Ninc (i.e., 41.9 to 44.2%) were less than what we have predicted based on the data from the Northern Great Plains (i.e., ~48%). This observation implies that the NHI of Australian pulse crops likely is lower than that of pulses of the Northern Great Plains. A lower NHI indicates that N removed in the grain relative to N in the straw is low suggesting either grain quality is lower (i.e., lower grain protein content) or grain yield relative to straw yield is reduced. Data presented by Evans et al. (2001) indicate that mean Australian yields differ from those achieved on the Northern Great Plains. The median yield for field pea grain in the Northern Great Plains was 3.03 Mg ha–1, and ranged from 1.10 to 6.29 Mg ha–1. In contrast, estimates of grain yield from field pea crops grown at three locations in Australia ranged from 0.55 to 1.82 Mg ha–1. Thus, in the Northern Great Plains, more N is exported in field pea grain and a higher %Ndfa would be required to achieve a positive N balance.

Evans et al. (2001) indicated that chickpea may provide the least benefit to soil total N accumulation in Australia, and suggested this was due to relatively low average shoot dry matter yields and %Ndfa. In their calculations, Evans et al. (2001) used the Australian Bureau of Statistics average farm chickpea yield of 1060 kg ha–1 but indicated that average farm yields likely understate true yield by ~10% due to grain loss by shattering. The median Northern Great Plains desi chickpea yield was 1.51 Mg ha–1 with a range of 0.93 to 2.25 Mg ha–1, and the median kabuli chickpea yield was 1.19 Mg ha–1 with a range of 0.48 to 2.24 Mg ha–1. Thus, as with field pea, higher average chickpea yields in the Northern Great Plains as compared with the Australian estimates contributed to more N being exported in the grain.

The relationship between the Ninc and the quantity of N derived from N2 fixation (i.e., kg Ndfa) (Fig. 3 ) reflects not only the contribution of N from N2 fixation (%Ndfa) but also the total grain yield—both of which differed among crops. As with %Ndfa, the quantity of fixed N2 required to achieve a positive N balance differed between crops as did the strength of the relationship between the Ninc and the quantity of fixed N2 (Table 2). Specifically, a strong relationship (r2 = 0.91, P < 0.01) existed between kgNdfa and Ninc for faba bean; setting the linear relationship to zero indicated that a positive Ninc was achieved in all cases with the mean kgNdfa (188.6 kg N ha–1 ± 15.3 SE) far exceeding the value required to achieve a positive Ninc (85.7 kg N ha–1). The mean levels of N2 fixed for field pea and lentil also exceeded that required for a positive Ninc in greater than 50% of the cases although the relationships were weaker than for faba bean. The mean levels of N2 fixation achieved by desi chickpea was approximately equivalent to levels required to achieve a positive Ninc. In contrast, on average, both kabuli chickpea and bean typically fixed less N2 than was required to achieve a positive Ninc, suggesting that these crops are net exporters of soil N.


Figure 3
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 3. Relationship between the accumulation of fixed N (kg N ha–1) and total N increment.

 
It is important to note that there were fewer observations available for faba bean (n = 10) and all faba bean cases were irrigated. Moreover, data suggest that when grown under irrigation, lentil, field pea, and desi chickpea also consistently contributed to a positive Ninc, whereas there remained some cases of a negative Ninc for both irrigated kabuli chickpea and bean (data not shown). Presumably irrigation removed a significant moisture constraint that otherwise limits the potential for some pulse crops to contribute to a positive Ninc. Consequently, the apparent superiority of faba bean to achieve a positive Ninc should be viewed with caution until additional studies examine the faba bean Ninc under dryland conditions.

Miller et al. (2003) compared the apparent N margin, calculated as the difference between N harvested in the grain and N inputs of field pea, lentil, and chickpea, and used this accounting of N inputs and outputs to estimate N2 fixation. They observed that the apparent N margin for field pea averaged 30 and 32 kg N ha–1 greater than that for lentil and chickpea, and concluded that field pea was a superior N2 fixing crop in that environment. Their observations are consistent with our assertion that total fixed N2, and consequently the Ninc, differs between pulse crops.

Achieving a positive Ninc for all grain legume crops is desirable, particularly in light of increasing inorganic N fertilizer costs and a growing appreciation for the importance of N accretion in sustainable cropping systems. The data suggest that although some crops, like faba bean, are likely to consistently contribute to a positive Ninc and thus contribute to N accretion, it is clear that other pulse crops, such as chickpea and common bean, are unlikely to contribute to a positive N economy (Fig. 4 ). Moreover, although field pea and lentil, on average, are likely to contribute to an overall accretion of N over the long term, year-to-year variations in the magnitude of N accretion should be expected. For example, Beckie et al. (1997) reported that field pea contributed to a positive N benefit in both years of a 2-yr study; however, the N benefit was only 8 kg N ha–1 in the first year and 19 kg N ha–1 in second. They concluded that the greater N benefit observed in the second year resulted from a lower NHI due to greater residue production. Similarly, observations from long-term experiments indicated that inclusion of lentil in a cereal-based cropping system at Swift Current, SK, increased the potential mineralizable N in the soil and reduced fertilizer N requirements for subsequent wheat crops (Campbell et al., 1992; Zentner et al., 2001). However, Campbell et al. (1992) reported that the effect of the lentil was relatively small initially but increased over time, suggesting a cumulative effect. Van Kessel and Hartley (2000), using data compiled from Campbell et al. (2000), reported that lentil increased the total soil N pool in the Swift Current long-term rotation plots at an annual average rate of 23 kg N ha–1 compared with 8 kg N ha–1 for fertilized wheat. These observations are consistent with the notion that although the nature of the Ninc may vary from year to year (i.e., positive and negative), on average it is expected that both pea and lentil typically will contribute to a positive Ninc.


Figure 4
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 4. Boxplots representing the calculated N increment based on published estimates of grain yield and N2 fixation. Numbers reported for each pulse crop indicate the total number of calculated means comprising the estimates. The thick dark line indicates the median value, the box represents 50% of all data, and the whiskers contain the remaining 50% of the data. Outliers and extreme outliers (together with their numeric values) are indicated by open circles and asterisks, respectively.

 
Przednowek et al. (2004) examined the rotation benefit of four annual legumes in Manitoba and concluded that field pea contributed the largest and most consistent apparent N benefit to a succeeding wheat crop, whereas the benefits of both chickpea and bean were inconsistent, depending on location and local growing conditions. Soybean provided little N benefit. These results support our observation that the Ninc is pulse crop specific and highly variable within differing environmental contexts (Fig. 4). Moreover, the Ninc achieved at any given location reflects a myriad of interactions between crop characteristics and adaptability, edaphic factors, climate, and local weather.

The Ninc was not correlated with grain yield when all data were considered (Fig. 5 ) although when examined individually, weak correlations were detected for lentil, desi chickpea, and faba bean (Table 3 ). Similar correlations were observed between the Ninc and aboveground straw yields (Table 3). Miller et al. (2006) reported that there was no relationship between field pea shoot biomass and the amount of N taken up by a subsequent spring wheat crop, irrespective of whether the biomass was left on plots or removed (i.e., cut to 0.1 m and removed prior to threshing) in low-disturbance no-till systems under dry to normal growing conditions in Montana. One might expect a relationship between Ninc and straw yield as it generally has been assumed that the aerial biomass residues (i.e., straw and stubble) make the greatest contribution of N to positive N balances due to greater biomass and high N concentration relative to belowground residues (Evans et al. 2001). The fact that correlations were weak at best, likely reflects the variability in shoot dry matter yield (i.e., straw) relative to grain yield (Fig. 6 ). Although the data suggest a linear relationship between straw and grain yield when all data were considered, the strength of this relationship varied between crops with only weak relationships detected for lentil and field pea, in particular. For example, for lentil grain yields of ~1.8 Mg ha–1, aboveground straw yields ranged from 2.7 to 6.5 Mg ha–1. For field pea, grain yields of ~2.6 Mg ha–1 had aboveground straw yields ranging from 2.3 to 6.5 Mg ha–1, with straw N yields ranging from 29.4 to 44.2 kg N ha–1. Thus, both harvest index and NHI would be expected to vary presumably as a consequence of crop x environment interactions.


Figure 5
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 5. Relationship between pulse grain yield (kg ha–1) and total N increment (kg N ha–1).

 

View this table:
[in this window]
[in a new window]

  		Table 3. Pearson correlations (r) between the N increment and measured pulse crop growth parameters.

 

Figure 6
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 6. Relationship between pulse grain yield (kg ha–1) and shoot dry matter (kg ha–1). Only data for which either harvest index or both shoot and straw biomass yield were reported are included.

 
Using the data available to us, we could only estimate root N inputs and thus our estimates of N return are incomplete; nonetheless, the data suggest that many factors can influence N return to the soil and large grain yields do not necessarily correspond to greater N returns. This variability is important to recognize, particularly if the grain yield is used to develop fertilizer recommendations for subsequent crops, as is the practice in some soil testing labs. For example, some soil testing labs that apply N credits for pulse crops base the credits on the previous crop grain yield (e.g., Green, 2005), presumably because it is assumed that grain yields are a reflection of straw N yields and subsequent return of N to the soil. For example, chickpea is assigned an N credit of 0.28 kg N ha–1 (0.25 lb N ac–1) for every 112 kg ha–1 (100 lb ac–1) grain yield once a minimum yield of 2240 kg ha–1 (2000 lb ac–1) grain yield is achieved. Field pea and faba bean are assigned a credit of 0.56 kg N ha–1 (0.5 lb ac–1) for every 112 kg ha (100 lb ac–1) grain yield after yields of 1680 and 2240 kg ha–1 (1500 and 2000 lb ac–1), respectively, are achieved. Given our observations that grain yield is at best only weakly correlated to shoot dry matter, and grain yield is at best only weakly correlated to the Ninc, the practice of basing soil N credits on the previous pulse crop yield is not supportable.

Other soil testing labs do not assign credits based on grain yield but rather use flat rates. For example, South Dakota State University Fertilizer Recommendations (2005) assign a credit of 44.8 kg N ha–1 (40 lbs acre–1) for all pulse crops, irrespective of grain yield. Presumably, this recommendation is based on the assumption that the Ninc of a pulse crop exceeds 44.8 kg N ha–1 and that at least this amount of N is available to subsequent crops. Data suggest that total N accumulation in straw, residue, and roots of pulses can exceed 44.8 kg N ha–1 (Fig. 7 ). However, assuming that usually no more than 25 to 30% of the N in mature pulse residue is available via decomposition and mineralization in the year following a pulse (Müller and Sundman, 1988; Rees et al., 1993; Beckie et al., 1997; Broersma et al., 2000; Evans et al., 2001), it is unlikely that the N released from pulses grown on the Northern Great Plains have the potential to contribute as much as 44.8 kg N ha–1. Indeed, research suggests that N contributions from pulses are likely to be far more modest. For example, in the moist Black soil zone of Saskatchewan, the residual effect of field pea was estimated to be equivalent to 15 kg N ha–1 for every 1000 kg pea grain (Beckie and Brandt, 1997). In a similar study in the same soil zone, Stevenson and Van Kessel (1996b) estimated that field pea contributed only an extra 6 to 14 kg N ha–1 to wheat and they concluded that the N contribution probably represented only a small component of the overall rotation benefit of field pea. In a study conducted in the Brown soil zone of Saskatchewan (dry semiarid prairie), N removal in a wheat crop grown on various crop residues indicated that the difference in N removal between pulse and cereal stubble averaged 15 kg N ha–1 (Miller et al. 2002). The variability in these estimates is not surprising given the variability in field pea yields and calculated Ninc among studies.


Figure 7
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 7. Boxplots representing total residue N inputs (kg N ha–1) (straw, roots, and rhizosphere exudates) based on published estimates. Numbers reported for each pulse crop indicate the total number of calculated means comprising the estimates. The thick dark line indicates the median value, the box represents 50% of all data, and the whiskers contain the remaining 50% of the data. Outliers (together with their numeric values) are indicated by open circles.

 
Results of this examination of published data suggest that although crops such as faba bean, field pea, and lentil are likely to contribute positively to the N economy in the Northern Great Plains, particularly when a cropping system is evaluated over the long term, other pulse crops such as desi and kabuli chickpea and common bean are more likely to be either N neutral or contribute to a soil N deficit. Year-to-year variations in the N contribution of any pulse crop are highly variable, and thus the N contribution to a subsequent crop is difficult to predict and, in particular, cannot be deduced from grain yields. Thus, the accuracy of N fertilizer recommendations that use assigned N credits for pulse crop stubble are subject to the same vagaries of weather and local growing conditions that contribute to the dramatic variability in N2 fixation, HI and NHI. Developing fertilizer recommendations using N credits is further complicated by potential differences in mineralization rates of pulse residues. For example, a spring soil N test may already include N derived from the previous pulse crop if conditions have favored rapid mineralization; thus, assigning an additional N credit could overestimate the potential N supply and underestimate N fertilizer requirements. Irrespective of the variability associated with different environmental conditions, the strong relationship between Ninc and %Ndfa suggests that efforts to enhance the contribution of N2 fixation to the N nutrition of a pulse crop are likely to favor N accretion in rotations that include pulse crops.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow An erratum has been published
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 Google Scholar
Google Scholar
Right arrow Articles by Walley, F. L.
Right arrow Articles by Lafond, G. P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Walley, F. L.
Right arrow Articles by Lafond, G. P.
Agricola
Right arrow Articles by Walley, F. L.
Right arrow Articles by Lafond, G. P.
Related Collections
Right arrow Other Legumes
Right arrow Crop Rotation Systems
Right arrow Dryland Cropping Systems
Right arrow Symbiosis
Right arrow Nitrogen
Right arrow Crop Systems
Right arrow Nutrient Management
Right arrow Production Agriculture
Right arrow Soil Fertility and Productivity

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