| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a 9782 Merrill Rd., Whitmore Lake, MI 48189
b S.A. Clay, Plant Sci. Dep., South Dakota State Univ., Brookings, SD 57007
* Corresponding author (david_clay{at}sdstate.edu)
Received for publication June 21, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
) in corn grain was measured at Aurora in 1998 and Beresford in 1998. Medic biomass at Aurora in 1997, Aurora in 1998, and Beresford in 1998 was 1290, 215, and 913 kg ha-1, respectively. At Aurora, on 11 July 1997, medic increased water infiltration from 0.0156 to 0.035 cm s-1. Similar results were measured at Aurora in 1998. Medic reduced corn growth as early as 17 July at Aurora in 1997 and 19 June at Beresford in 1998. Associated with reduced corn biomass production in the fertilized treatments were lower N concentration and fertilizer efficiency. Interseeding medic into corn did not increase in corn at Beresford in 1998 or Aurora in 1998. These results suggest that the negative effects of medic on corn primarily resulted from N stress and not increased water stress. Given these results, management strategies that reduce the competition between medic and corn for N should be developed.
Abbreviations: ET, evapotranspiration • , 13C discrimination • 
N stress, the change in 13C discrimination for each percentage of yield lost due to N stress • water stress, the change in 13C discrimination for each percentage of yield lost to water stress
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
Annual snail medic may be suited for interseeding into corn in the Great Plains because it has a short life cycle (12–14 wk), is a cool-season plant, and has a determinate growth habit (DeHann et al., 1997; Vos, 1999). Medic is also capable of fixing atmospheric N and may suppress weed growth (Buhler et al., 1998; DeHaan et al., 1997; Vos, 1999). In the north-central USA, annual medics have been evaluated for their potential use as a forage and smother crop (Moynihan et al., 1996; DeHaan et al., 1997; Buhler et al., 1998). This research has shown that medic can reduce weed and corn biomass production at seeding rates as low as 14 kg ha-1 seed. The planting date of the intercrop in the Great Plains is critical because if they are planted too late, then they may have a limited impact on reducing wind or water erosion and if planted too early, they may reduce primary crop yield (Jeranyama et al., 1998).
Interseeded crops can reduce primary crop yields through allelopathy and competition for nutrients, water, and light. In semiarid environments such as the Great Plains, water used by the interseeded crops may reduce plant-available water, which can reduce yields. Yield loss due to the interseeded crop can be reduced by delaying the planting date of the interseeded crops (Jeranyama et al., 1998). However, delaying the planting date also reduces the potential for the interseeded crop to reduce wind and water erosion. To resolve this dilemma, the mechanisms causing yield reductions in the primary crop must be known. Once mechanisms are known, solutions can be developed.
Interseeded crops have the potential to compete with or provide nutrients to the primary crop. The impact of the interseeded crop on N utilization by the primary crop can be evaluated by using nonisotopic as well as 15N natural abundance (Clay, 1997; Zhu et al., 1998), 15N enrichment, and 15N depleted methods. A common feature of all isotopic methods is that different N sources have different 15N/14N ratios.
If the interseeded crop is a legume, then symbiotic N fixation may add N to the cropping system (Heichel and Barnes, 1984; Frye et al., 1988; Loi et al., 1993). The amount of N fixed by medic is variable and has been reported to be as high as 200 kg N ha-1 (Zhu et al., 1998). The amount of N transferred from leguminous to nonleguminous plants depends on the legume, cropping system, climatic conditions, and N uptake kinetics of the nonleguminous crop (Tomm et al., 1994; Walley et al., 1996). Benefits from N fixation may occur during the year that the legume is planted or in subsequent years. Hargrove (1986) reported that the average fertilizer replacements of four winter legume cover crops was 72 kg N ha-1. Jeranyama et al. (1998) reported that medic intercropped into corn had a fertilizer replacement value for the subsequent crop of 13 to 18 kg N ha-1. Amado et al. (1998) reported that 61 kg N ha-1 inorganic N was needed for corn following fallow to equal the corn yield following common vetch (Vicia sativa L.) in southern Brazil. The transfer mechanism has been attributed to the mineralization of plant residues followed by plant uptake.
The impact of water stress on crops grown under field conditions is difficult to assess because water can be lost from the system by a variety of mechanisms, including evaporation, transpiration, runoff, and leaching. Techniques that measure the impact of management on water stress may combine evaporation and transpiration into a single measurement of evapotranspiration (ET). Separating ET into transpiration and evaporation may be required to assess the impact of interseeded crops on water availability to the primary crop because if the interseeded crop obtains water from the soil surface that typically is lost to evaporation and the primary crop obtains water from deeper soil depths, then the two crops may not directly compete for water.
Many of the commonly used techniques for measuring water use do not separate evaporation and transpiration. For example, in the mass balance approach, ET has been estimated using the following equation: ET = precipitation - soil water at the end of the season + soil water at the beginning of the season. This equation assumes that runoff, runon, and leaching are insignificant. Clearly, this assumption is not valid for fields with topographic relief. This assumption may not even be valid in relatively flat fields with course-textured soils. Delin et al. (2000) reported that water redistribution occurs in coarse-textured soils with relatively small slopes (<2%). In sloping land, runoff or runon can be measured, modeled, or estimated from topographic information. For example, Halvorson and Doll (1991) addressed the runoff and runon problem using topography to calculate a topographic factor. If the factor was positive, then a net gain in runon was expected and if negative, then a net loss from runoff was expected. They used the topographic factor to predict areas where runoff was expected and to move an appropriate amount of water to areas where runon was expected. This approach worked better in wet than dry years.
An alternative approach to measuring transpiration is to measure a time-integrated value that is related to water stress. The amount of 13C discrimination () occurring in C3 and C4 plants provides an time-integrated index of stomatal closure (Farquar and Lloyd, 1993). In order to understand why provides an index of water stress, background information is needed. First, is influenced by plant type, stomatal conductance, photosynthesis capacity, and the plant water demand (Clay et al., 2001a, 2001b). Water stress influences because the amount of 13CO2 and 12CO2 fixed during photosynthesis is influenced by the relative amounts 13CO2 and 12CO2 in the leaf. In C4 plants, the relative amount of 12CO2 fixed during photosynthesis increases with increasing water stress (stomatal closure). Photosynthesis-induced C isotope fractionation in C4 (c4) plants is described by the equation:
 | [1] |
) ) 
= ratio of the rate of CO2 leakage from the bundle sheath to the rate of PEP carboxylation (leakiness)
When s = 0 and = 0.21, Eq. [1] predicts that as Ci/Ca approaches 0 (stomata closed), c4 approaches 4.4 and as Ci/Ca approaches 1 (stomata open), c4 approaches 0.6. An advantage of using as a water stress index is that it provides a spatially and temporally integrated value.
Clay et al. (2001a) tested if could be used to evaluate water stress–induced yield reductions in corn. In this study, corn located at different landscape positions was watered and not watered. In the nonwatered areas, summit soils were dryer than footslope soils for the entire growing season. In the footslope area, watered (154 g grain plant-1) and unwatered (164 g grain plant-1) plants had similar yields and values. However, in summit soils, yields were lower in nonwatered (87 g plant-1) than watered plants (145 g plant-1), and nonwatered plants (3.35) had higher than watered plants (3.11). In the same study, the authors reported that contained spatial structure, was highest in summit soils and lowest in footslope soils, and that could be used to estimate yield losses due to water stress. Data from this study were used to calculate that for every 1% decrease in relative corn yield due to water stress, increased 0.0117.
Clay et al. (2001b) also used C isotopic discrimination to calculate yield losses due to N and water stress in wheat (Triticum aestivum L.). They showed that by defining the impact of N on , yield losses due to water and N stress could be calculated. In this study, under non–N limiting conditions, the impact of water stress on yield was defined by the equation: yield (kg ha-1) = -11000 + 884 [r = 0.92 (significant at the 0.01 level)].
Once the causes for yield reduction are understood, management practices designed to make best use of the available resources can be developed. The objective of this study was to determine the influence of an annual medic (variety Sava) interseeded into corn at planting on N and water stress in corn.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
The experimental design for each site was a randomized complete block with four replications. Factorial treatments were two snail medic (cultivar Sava) seeding rates (0 and 33 kg ha-1) and two N fertilizer rates (0 and the N recommendation based on the NO3–N test) (Gerwing and Gelderman, 1996). The row spacing at all sites was 76 cm. Based on spring soil tests, all plots were fertilized with P and K (Gerwing and Gelderman, 1996).
At Aurora in 1997, corn was planted on 5 May at a rate of 69000 seeds ha-1, urea was broadcast-applied following planting at rates of 0 or 134 kg N ha-1, and inoculated snail medic was planted with an alfalfa seeder on 9 May. To stimulate medic germination and growth and to move the urea into the soil, 1 cm of irrigation water was applied in May. To control weeds, an interrow cultivation was conducted on 2 July. At Aurora in 1998, corn was planted at a rate of 69000 seeds ha-1 on 5 May, urea was broadcast following planting at rates of 0 or 146 kg N ha-1, inoculated snail medic was broadcast, and urea and medic were raked into the soil. Irrigation water was not applied at Aurora in 1998. At Beresford in 1998, corn was planted at the rate of 69000 seeds ha-1 on 24 April, urea was broadcast-applied following planting at rates of 0 and 143 kg ha-1 urea N, inoculated medic was broadcast, and urea and medic were raked into the soil on 24 April. Irrigation water was not applied at Beresford 1998.
Soil samples, a composite of eight 2-cm cores, were collected from each plot in 15-cm increments to a depth of 60 cm before fertilizing and after corn harvest. Nitrate N and NH3–N were extracted from 10 g of air-dried soil with 100 mL of 1 M KCl and analyzed using the Cd reduction and phenate methods, respectively (Maynard and Kalra, 1993).
Water Infiltration
Water infiltration at Aurora was measured at one site in each plot on 11 June, 19 June, 11 July, and 19 July 1997 and on 20 August and 28 August 1998. The double-ring infiltrometer used to measure infiltration consisted of an outside ring with a diameter of 34 cm and an inside ring with a diameter of 15 cm (Lowery et al., 1996). The rings were driven 15 cm into the soil in the center of the interrow. The amount of water needed to cover the ring area to a depth of 2.5 cm was added to each ring. Twenty-four hours later, rings received an additional 2.5 cm of water, and the length of time for water to completely infiltrate into inner ring soil was measured.
Nitrogen Budget Calculations
Net N mineralization in the unfertilized plots was calculated using the N balance approach. Net N mineralization was calculated with the equation: net N mineralization = N contained in grain and stover at harvest - final inorganic soil N at harvest + initial inorganic soil N at planting. In this equation, plant N at harvest was the total amount of N contained in the aboveground biomass, and inorganic soil N at planting and harvest was the amount inorganic N contained in the 0- to 60-cm depth of soil. This method assumed that N lost through leaching and denitrification was insignificant.
Medic biomass was harvested, when maximum biomass was expected, from two 0.35-m2 areas in each plot at Aurora 1997 on 2 July, Aurora 1998 on 27 July, and Beresford 1998 on 19 July. Corn biomass was harvested from a 1.52-m2 area at Aurora 1997 (18 June, 2 July, and 17 July), Aurora 1998 (11 June and 27 July), and Beresford 1998 (19 June and 5 August). On 1 October, 8 October, and 23 September, corn grain and stover were harvested from 9.27-m2 areas at Aurora 1997, Aurora 1998, and Beresford 1998, respectively. Harvested plants were dried, weighed, and analyzed for total N and 15N on an Europa 20-20 ratio mass spectrometer (Europa Sci., Westchester, UK). The 15N values were calculated with the following equation:
To determine the impact of the N and medic treatments on N uptake by the subsequent crop, oat (Avena sativa L.) was planted in 1998 at Aurora 1997, 1999 at Aurora 1998, and 1999 at Beresford 1998. Oat biomass samples were harvested at the point of maximum biomass accumulation (July) in 1998 (Aurora 1997) and 1999 (Aurora 1998 and Beresford 1998) from a 0.3-m2 area in each plot. Oat biomass was dried, weighted, ground, and analyzed for total N and 15N on an Europa 20-20 ratio mass spectrometer.
The contribution of medic toward meeting corn N requirement in the unfertilized plots was calculated using the following equation:
15Na is the 15N of corn in the medic–zero N plots, 15Nb is the 15N of medic in the zero-N plots, and 15Nc is the 15N of corn in the no medic–zero N plots (Shearer and Kohl, 1992; Clay, 1997).
Fertilizer efficiency (%FE) was calculated using the equa-tion:
 | [4] |
Carbon-13 Discrimination and Yield Losses Due to Water and Nitrogen Stress
Grain and stover samples from Aurora 1998 and Beresford 1998 were analyzed for 13C and . The 13C was calculated with the following equation:
13C values for air, C3, and C4 plants are -8, -27, and -13, respectively. A negative sign indicates that the sample has a lower 13C/12C ratio than PDB. The 13C values were used to calculate using the following equation:
 | [6] |
13Ca is the 13C value of air (-8) and 13Cp is the 13C value of the plant.
The data from Clay et al. (2001a), combined with findings from this study, and the approach described by Clay et al. (2001b) were used to calculate yield losses due to water and N stress. Clay et al. (2001b) suggested that the total yield losses due to water and N stress could be separated into two components. Yield loss due to N stress was defined as the difference between measured yield and the expected yield if N was applied in excess of the plant requirement. Yield loss due to water stress was defined as the difference between the maximum yield and the yield obtained under non–N limiting conditions. Using this approach, the equations were
 | [7] |
 | [8] |
= the change in in corn due to adding medic water stress = the change in for each percentage of yield lost to water stress (Clay et al., 2001a) N stress = the change in for each percentage of yield lost due to N stress (derived below)
Clay et al. (2001a) showed that to produce a 30% yield loss to water stress, increased 0.35. Using this data, water stress was estimated to be 0.0117 (percentage loss in relative yield)-1.
Analysis of variance was used to evaluate treatment differences. For main effects, a F-test at the 0.05 level was used to determine treatment differences, and for interactions, a Fisher LSD at the 0.05 level was used to determine treatment differences.
Climatic Calculations and Conditions
Daily maximum and minimum temperatures and rainfall for each day were measured at weather stations located at each site. Growing degree days (GDD) were calculated using the following equation:
 | [9] |
Rainfall plus irrigation between May and September at Aurora 1997, Aurora 1998, and Beresford 1998 was 30.4, 34.5, and 24.9 cm, respectively. At Aurora and Beresford, the rainfall plus irrigation totals were less than the long-term precipitation averages between May and September of 38.0 and 40.6 cm, respectively.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
|
|
contained in the surface 60 cm before planting corn at Aurora 1997, Aurora 1998, and Beresford 1998 was 9, 43, and 47 kg N ha-1, respectively. The small amount of inorganic N at Aurora 1997 was attributed to fertilizer not being applied to the site for the previous 7 yr. Following harvest, inorganic N contained in the surface 60 cm of soil at Aurora 1997, Aurora 1998, and Beresford 1998 was not influenced by fertilizer or medic and averaged 39, 52, and 34 kg N ha-1, respectively.
Corn Production
Corn biomass reduction in medic-interseeded areas was detected on 17 July and 19 June at Aurora 1997 and Beresford 1998, respectively (Table 3). Associated with biomass reductions, in the fertilized treatments, a reduction in corn N concentration from 26.7 to 21.5 g kg-1 was detected on 17 July 1997 at Aurora. Medic also reduced the N concentration in the fertilized plots at Beresford 1998 on 19 June from 34.8 to 26.7 g kg-1. Medic-induced reductions in corn biomass and N concentration suggest that medic reduced yields either through allelopathy or N stress. Allelopathic effects of medic on corn was discounted because Vos (1999) reported that leaf extracts from medic did not inhibit corn seed germination or seedling growth. Yield reductions due to medic were still evident at harvest (Table 3).
|
|
Transfer to Medic Nitrogen to Corn
At Aurora 1997, the average 15N values of medic in the medic–unfertilized treatment (+M/0N) and corn in the no medic–unfertilized treatment (0M/0N) and the medic–unfertilized treatment (+M/0N) were 1.41, 2.42, and 1.89, respectively (Tables 4 and 5). The 15N values of medic suggest that medic N was derived N from several sources, including N fixation, residual N, or some of the 87 kg N ha-1 mineralized in the unfertilized treatments. Based on 15N values, 63 kg N ha-1 contained in the corn at harvest in the unfertilized medic treatment was transferred from medic (Eq. [2]) (Shearer and Kohl, 1992). Nitrogen transfer from medic to corn may have resulted from the following events: (i) Medic died in July; and (ii) N contained in the medic was mineralized and taken up by corn. The transfer of N from a legume plant to a nonlegume plant in a pasture has been previously reported by Tomm et al. (1994) and Walley et al. (1996).
|
15N value in the grain and stover. These results suggest that corn in the unfertilized plots had a large reliance on the 103 kg N ha-1 that was mineralized in the unfertilized plots.
At Beresford 1998, the average 15N values of medic in the medic–unfertilized treatment (+M/0N), corn in the no medic–unfertilized treatment (0M/0N), and corn in the medic–unfertilized treatment (+M/0N) were 0.262, 2.06, and 1.04, respectively. The relatively low 15N value of the medic (0.262) suggests that medic had a relatively small reliance on the 71 kg of N that was mineralized in the unfertilized treatments. Based on 15N values, 32.5 kg of N contained in corn in the unfertilized treatment was transferred from medic. The N transferred from medic to corn most likely was derived from both above- and belowground portions of the plant.
Medic and Nitrogen Fertilizer Impact on Oat Nitrogen
The impact of N and medic on N cycling was not just limited to the year that the treatments were applied. At Aurora 1997, medic did not influence N contained in oat harvested from the plots in 1998. However, oat harvested from fertilized plots contained 23 kg ha-1 more N than oat harvested from unfertilized plots (Table 6). The impact of N fertilizer on N contained in oat was attributed to the remineralization of immobilized N fertilizer contained in soil microbial biomass, corn residue, or medic residues. These findings are conceptually in agreement with those of Clay et al. (1990). Oat grown at Aurora 1998 in the medic and N treatments had similar results.
|
Medic and Nitrogen Impacts on Carbon-13 Discrimination in Corn
At Aurora 1998, medic did not influence , and fertilizing with N increased in corn grain from 2.62 to 2.81. The lack of differences in due to medic was expected because medic did not impact yields or fertilizer efficiency. Based on the yield and values in the fertilized plots (Table 3), N stress (Eq. [8]) was estimated to be -0.0091 (percentage loss in yield due to N stress)-1.
Similar results were observed at Beresford 1998 where N increased in corn grain from 2.14 to 2.36 while interseeding corn with medic did not influence (2.25). Care must be used in evaluating because both water and N interact to influence . At Beresford 1998, similar values in medic and nonmedic corn most likely resulted from medic increasing both water and N stress in corn. If this occurred, then Eq. [7] and [8] could be used to separate the water and N effects of medic on corn from each other. However, these equations require that water stress and N stress be known. Based on the only data available (Clay et al., 2001a), water stress in Eq. [8] was estimated to be 0.0117 (percentage loss in relative yield)-1. Grain yields from Table 3 and presented above were used to estimate that N stress was -0.0042 (% yield loss from N stress)-1 [= (2.14 - 2.36)/(51.8% grain yield loss from N stress)]. By simultaneously solving Eq. [7] and [8] using the total percentage yield loss (TYL) and (change in in corn due to adding medic) values of 31.6% [yield loss from medic in fertilized treatments = 100(7210 - 4930 kg ha-1 grain)/7210 kg ha-1 grain)] and zero, respectively, the yield losses due to medic-induced water and N stresses were estimated to be 8.34 (601 kg ha-1 grain) and 23.3% (1680 kg ha-1 grain), respectively. This analysis suggests that the primary factor causing the yield reduction at Beresford 1998 was a medic-induced N stress in corn. The decrease in fertilizer efficiency and N concentrations in biomass samples, due to medic, at Aurora 1997 and Beresford 1998 support this hypothesis.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
Based on calculated fertilizer efficiencies, 15N, and , the corn yield losses at Aurora 1997 and Beresford 1998, due to inseeding with medic, were primarily due to N stress and not water stress. The relatively small impact of medic on increased water stress in corn may have resulted from medic increasing water infiltration rates and/or medic utilizing water that otherwise would have been lost to evaporation. Given these findings, additional research is needed to assess if alternative N fertilizer management strategies can be used to reduce the competition between medic and corn for N.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. N. Baributsa, E. F. Foster, K. D. Thelen, A. N. Kravchenko, D. R. Mutch, and M. Ngouajio Corn and Cover Crop Response to Corn Density in an Interseeding System Agron. J., June 16, 2008; 100(4): 981 - 987. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Clay, K.-I. Kim, J. Chang, S. A. Clay, and K. Dalsted Characterizing Water and Nitrogen Stress in Corn Using Remote Sensing Agron. J., April 11, 2006; 98(3): 579 - 587. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. S. Snapp, S. M. Swinton, R. Labarta, D. Mutch, J. R. Black, R. Leep, J. Nyiraneza, and K. O'Neil Evaluating Cover Crops for Benefits, Costs and Performance within Cropping System Niches Agron. J., January 1, 2005; 97(1): 322 - 332. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Clay, S. A. Clay, J. Jackson, K. Dalsted, C. Reese, Z. Liu, D. D. Malo, and C. G. Carlson Carbon-13 Discrimination Can be Used to Evaluate Soybean Yield Variability Agron. J., March 1, 2003; 95(2): 430 - 435. [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 | |||
