| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Agron., DMI Tillage Business Unit, CNH Global, Rt. 150E., Box 65, Goodfield, IL 61742
b Dep. of Agron., Univ. of Wisconsin-Madison, 1575 Linden Dr., Madison, WI 53706
* Corresponding author (kaalbrec{at}facstaff.wisc.edu)
Received for publication March 2, 2001.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: ANR, apparent nitrogen recovery • DM, dry matter • NUE, nitrogen use efficiency
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Fertilizer N recovery by grass is affected by such factors as species, growth habit, N rate, precipitation, and soil type. George et al. (1973) found that, under favorable conditions in 1965 in Indiana, the apparent N recovery (ANR) after split applications of N to timothy (Phleum pratense L.), smooth bromegrass, and orchardgrass in a four-harvest system was as high as 0.48, 0.72, and 0.67 kg N (kg N applied)-1, respectively. These ANR values uniformly decreased as annual N rates exceeded 126 kg ha-1 for timothy and 168 kg ha-1 for smooth bromegrass and orchardgrass. During 1964, when dry conditions were present, the cumulative annual ANR for any of these species did not exceed 0.49 and was as low as 0.14 for timothy for N rates of 1008 kg ha-1. In Maryland, Wagner (1954) reported an ANR of 0.85 for orchardgrass when N was applied at either 90 or 179 kg ha-1, but this fell to 0.69 when N was applied at 269 kg ha-1. Work in Pennsylvania has shown that the cumulative annual ANR of orchardgrass ranged from 0.26 to 0.30, depending on soil type and associated temperature regime in a two-harvest system where N was split-applied (Stout and Jung, 1992). The researchers noted that the ANR was approximately 0.42 in the spring growth period and fell to 0.15 in the fall period when averaged across all soil types.
One of the tradeoffs of having excellent persistence in some temperate grass species is the somewhat poor seasonal distribution of forage DM yield. For example, smooth bromegrass yields substantially more dry matter (DM) in first-cut harvest than in subsequent harvests (Knowles et al., 1985; Van Esbroeck et al., 1995). Efforts to improve the mid- to late-summer forage production through multiple applications of fertilizer N during the growing season have met with varied results. Washko and Pennington (1956) showed that applying N at 1, 28, 56, 112, and 224 kg N ha-1 in either single or split applications to smooth bromegrass still resulted in the majority of forage yield coming in the early part of the growing season. Daly and Mackenzie (1983) demonstrated that delaying N applications on ryegrass (Lolium perenne L.) until later in the growing season may result in reduced annual yields because of less first-cut yield. Long et al. (1991) found that for first-cut ryegrass silage production, little precision was required in selecting the date to apply fertilizer N within a 5-wk period in the spring to achieve maximum DM yield.
In addition to forage yield improvement efforts related to fertilizer N applications, research has been conducted examining N use efficiency (NUE). Grass NUE is defined as the amount of forage DM produced for each unit of applied N. The NUE has been studied under a variety of field conditions for a wide array of grass species (Ramage et al., 1958; Drake et al., 1963; Donohue et al., 1965; George et al., 1973; Stout and Jung, 1992; Guillard et al., 1995) and cultivars (Sabata and Mason, 1992). Grass NUE may be affected by species, soil type, latitude, temperature, N fertilizer rate, and field moisture conditions (Wright and Davison, 1964). Donohue et al. (1965) reported that the NUE for orchardgrass was greatest at 247 kg N ha-1. Long et al. (1991) noted that split N applications on ryegrass significantly increased NUE only some of the time. In Connecticut, warm-season species, such as corn (Zea mays L.) grown for silage, had a NUE that ranged from 42 to 47 kg DM (kg N applied)-1 (hereafter, kg DM kg-1) at an N fertilizer rate of 112 kg N ha-1 and decreased to a low of 10 kg DM kg-1 at 448 kg N ha-1 (Guillard et al., 1995). In contrast, cool-season species such as orchardgrass had a NUE that ranged from 15 to 20 kg DM kg-1 at 112 kg N ha-1 and decreased to a low of 10 kg DM kg-1 at 448 kg N ha-1. The authors reported that for corn and oat (Avena sativa L.) forage systems followed by forage cover crops, NUE decreased with N fertilization rates >112 kg ha-1. They concluded that at fertilizer rates >112 kg N ha-1, there was no significant difference in NUE among these systems and monoculture orchardgrass and that prudent producers would benefit by considering NUE in addition to crop yield potential. This is consistent with the work of Ramage et al. (1958) in New Jersey and George et al. (1973) in Indiana, who reported decreased ANR with N fertilization rates >112 and 168 kg ha-1, respectively. Early one-time applications of N have been shown to increase N losses to the environment via leaching, denitrification, and volatilization (Ryden, 1984) and would therefore contribute to reduced grass NUE.
Results from these studies may help improve manure management within whole-farm nutrient plans because grass N demand can be met by manure N applications within livestock operations. Application techniques will of course be different between manure and inorganic N sources (i.e., injected rather than broadcast, based on plant-available N rather than total N), but baseline ANR and NUE information could aid in determining rate and timing of manure applications. For example, Kaffka and Kanneganti (1995) reported that in dairy production systems, orchardgrass recovered 40% of the liquid manure N and 26% of the solid manure N applied each year. Repeated applications throughout the growing season give perennial forages an advantage over annual crops such as corn because producers require smaller manure storage facilities. Continued dependence on sward renovation via repeated legume seeding, or on commercially available fertilizer N sources for high grass yields, coupled with a relatively short growing season reduces competitive margins for the North-Central region's livestock operations.
These factors increase the importance of ANR and NUE comparisons over several seasons, soil types, grass species, and a range of N rates. Limited information is available where these factors have been evaluated together over several seasons. This is particularly important in the North-Central USA where comparatively short growing seasons and high land values are frequently coupled with confined animal systems within environmentally sensitive watersheds. These factors dictate that forage swards are managed intensively to generate high-volume, high quality forage with as little N loss to the environment as possible. The purpose of this research is to (i) determine forage yield and gross N yield and recovery of Kentucky bluegrass, smooth bromegrass, and orchardgrass over a range of N fertilizer rates; (ii) assess the efficiency with which these species use N to produce DM; and (iii) determine whether these parameters are affected by different soil types within similar temperature and precipitation regimes typical for this region.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Spring seedbed preparation in 1993 at both locations included moldboard plowing to 15 cm followed by disking and cultimulching. Plots that were 6.1 by 0.9 m were sown on 30 Apr. 1993 at Arlington and 14 May 1993 at Lancaster. Using a six-row Carter planter (Carter Manufacturing, Brookston, IN), treatments were sown with ‘Park’ Kentucky bluegrass, ‘Badger’ smooth bromegrass, and ‘Orion’ orchardgrass at 17.9, 17.9, and 9.0 kg seed ha-1, respectively. Kentucky bluegrass was also sown at 17.9 kg ha-1 to establish a 2.0-m grass border surrounding all the experiments and a 0.9-m corridor between blocks. Swards were given no N fertilizer during the establishment year to recover available N from previous crop rotations. Beginning in 1994, broadcast N was split-applied as ammonium nitrate (NH4NO3) to six treatments in each experiment at annual rates of 0, 56, 112, 168, 224, and 336 kg N ha-1, respectively, in early April and after the first two harvests in a three-harvest system. Minor amounts of dead residue resulting from the previous late-autumn growth were clipped and discarded each year in early April.
Minor amounts of broadleaf weeds such as broadleaf plantain (Plantago major L.) and dandelion (Taraxacum officianale Weber in Wiggers) were controlled with dicamba (3,6-dichloro-2-methoxybenzoic acid) applied at 0.6 kg a.e. ha-1. Fertilizer P and K were applied annually at rates ranging from 0.0 to 50.1 kg ha-1 for P and from 0.0 to 141.9 kg ha-1 for K at both locations to maintain a very high status based on soil test recommendations for grass pasture (Kelling et al., 1991). These recommendations also account for potential subsoil P contributions by soil series as well as the P-buffering capacity of these soils.
Harvest and Sample Handling
All forage within each plot was harvested to a 7.0-cm stubble height three times during each growing season (approximately 5 June, 15 July, and 1 Sept.) using a 0.9-m-wide small-plot flail harvester. This harvest schedule coincided with the early heading to late-flower stage for grasses in the first harvest. The wet weight of the harvested vegetation was recorded, and a subsample of approximately 0.5 kg of fresh forage was oven-dried at 60°C for 72 h to determine forage DM. Remaining clippings were discarded and not returned to the plots. Plot yields were computed on a dry weight basis and summed across harvests in each year. Subsamples were ground using a Thomas-Wiley mill (A.H. Thomas Co., Philadelphia, PA) to pass a 1.0-mm screen.
Laboratory estimates of forage N content from each plot at each harvest were determined by analyzing the ground forage samples. Kjeldahl-N was determined using a semimicro-Kjeldahl procedure (Bremner and Breitenbeck, 1983) with a salicylic acid modification (Bremner, 1965) for the recovery of NO3. Means were computed and weighted to reflect the contribution of each harvest to annual plot N yields.
Statistical Analysis and Calculations
Analysis of variance (ANOVA) procedures were conducted using the GLM procedure of SAS (SAS Inst., 1990) at P = 0.05 to test the effects of year, location, N treatment, and all interactions in models for forage DM yield, ANR, and NUE. In all models, the effects of location were tested using year x location as the error term. Similarly, the treatment x year and treatment x location interactions were tested using treatment x year x location as the error term. Regression models were fitted for combined data (i.e., not averaged) in the absence of interactions using the REG procedure of SAS. In these models, DM yield, ANR, and NUE were the dependent variables while fertilizer N rate was the independent variable.
Forage N yield was calculated by multiplying plant DM yield by the concentration of N in the harvested forage. The ANR for each harvest and for the total season was calculated using the difference method where ANRx = [(kg of N recovered at Nx - kg of N recovered from N0)/kg N applied at Nx] x 100, where x = N rate > 0. Fertilizer NUE for each harvest was calculated as [(grass yield at Nx - grass yield at N0)/kg of N applied at Nx] x 100, where x = N rate > 0. Values of ANR and NUE for each harvest were mathematically weighted in proportion to their biomass and combined for each year.
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
|
|
|
|
Significant (P < 0.05) forage DM yield differences between locations for a given year, forage species, and N rate occurred in 22 out of 54 cases when individual harvests were considered (data not shown). Such cases were rarest for orchardgrass, intermediate for smooth bromegrass, and most common for Kentucky bluegrass and tended to occur at higher rates of N fertilizer and in 16 out of the 22 cases, favored Arlington. For all three species, the greatest DM production occurred in the first harvest and was followed by reduced DM production in subsequent harvests. An exception was in 1995 when second-harvest DM yield in all experiments for all N rates tended to be greater than first-harvest yields for Kentucky bluegrass and orchardgrass at both locations and across all N rates (data not shown). These results are explained by a dry period during mid- to late May (especially at Lancaster) in 1995 (Table 2) that reduced plant growth. Normal rainfall resumed in mid-June, which improved plant growth leading up to the second harvest.
Nitrogen Yield
No significant (P < 0.05) treatment x year or treatment x location interactions were present in the model comparing annual cumulative forage N yield. Data for forage N yield (Table 3) are therefore presented within years and combined over years for each of the three grass species. For each species, N yield increased with increasing fertilizer N rates. Across all N rates, smooth bromegrass generally had the greatest N yield, followed by orchardgrass, which was followed by Kentucky bluegrass. Forage N yield exceeded the N application rate in the smooth bromegrass experiment at both locations for the 0, 56, and 112 kg N ha-1 rates and in the orchardgrass experiment at Lancaster for these same N rates. However, crop N yield did not exceed the sum of the N applied and the N recovered by forage at the 0 kg N ha-1 rates. This indicates that a substantial portion of total forage N yield at the low-to-intermediate N rates was due to N mineralized from the soil.
Averaged over years, the forage N yield at the 0 kg N ha-1 rate was similar for both soil types and was 38 kg N ha-1 in Kentucky bluegrass, 78 in smooth bromegrass, and 65 in orchardgrass. As fertilizer N rates were increased, the forage N concentration also increased in each experiment (Table 1). However, as the fertilizer N rates were increased from 0 to 336 kg N ha-1, the harvest-weighted forage N concentration of the total season increased by a maximum of 8 g kg-1 DM for Kentucky bluegrass, 6 for smooth bromegrass, and 6 for orchardgrass. Meanwhile, forage DM yields in all experiments increased three- or fourfold over the same range of fertilizer N treatments. This suggests that the majority of the increase in forage N yield in any of the experiments comes from the increase in DM yield rather than an increase in N concentration. For example, based on aboveground harvests and tissue analysis, the lower N yield of Kentucky bluegrass across fertilizer N rates compared with smooth bromegrass or orchardgrass was associated with lower forage DM yields, which had a maximum of approximately 6 Mg ha-1 at both locations and tended to be less than the other species. Because resources such as precipitation and soil type are the same regardless of species, these data suggest that Kentucky bluegrass is either less efficient at sequestering soil N or allocates more sequestered N to roots. These results indicate that grass species affected the harvestable aboveground N pool more than soil type and may therefore influence potential N cycling in hay or pasture systems.
Apparent Nitrogen Recovery and Nitrogen Use Efficiency
Because there were no significant (P < 0.05) treatment x year interactions for either the ANR or NUE models in each experiment, these data were combined over years. For all three grass species, first-order models were fitted to predict forage ANR from N fertilizer rate for each location, except for smooth bromegrass at Arlington, which required a second-order model (Fig. 2). One other exception was the ANR model for Kentucky bluegrass at Lancaster where a linear-plus-plateau model best described the data. For all three grass species, as fertilizer N rates were increased, there was a divergence in ANR between locations, which suggested a significant treatment x location interaction though this was not significantly different and may have been related to less precipitation at Lancaster in August of all 3 yr (Table 2). Within the range of N rates in this study, the models predicted that ANR at Arlington increased as N fertilizer was increased. Meanwhile, models for ANR at Lancaster predicted that with increased N rates, ANR would plateau at a low N rate in Kentucky bluegrass, improve but to a lesser extent than at Arlington in smooth bromegrass, or even decrease in orchardgrass. Moreover, orchardgrass generally had greater ANR at Arlington than did Kentucky bluegrass or smooth bromegrass during the study (Fig. 2).
In contrast, there was generally little difference in ANR between species at Lancaster. Stout and Jung (1992) reported differences in ANR for orchardgrass on two soils with widely different thermal regimes. They suggested that greater ANR is associated with a higher rate of plant growth, which leads to a more rapid assimilation of fertilizer N. However, in the present experiment, environments at the Arlington and Lancaster sites have very similar temperature regimes. We attribute the greater ANR at Arlington to the superior yield potential of that soil for most crops. The Mollisol at Arlington is well aggregated, better drained, has a higher organic matter content, and was therefore more productive than the Alfisol at Lancaster under similar temperature and precipitation regimes (Table 2).
A more similar ANR between soil types for smooth bromegrass at N application rates 
112 kg N ha-1 may have resulted from the abundant DM production at first harvest in the spring when soil moisture was less limiting at either location. Shallow-rooted forages such as Kentucky bluegrass (Peterson et al., 1979) may have especially benefited from these conditions as a similar difference between soil types in ANR was not seen in the other experiments. In 1995, precipitation at Lancaster was 75 mm less than it was at Arlington. This resulted in ANR values for Kentucky bluegrass that were especially low in the third harvest at Lancaster and led to differences between locations for the year (data not shown). The influence of annual precipitation total and distribution were also reported by George et al. (1973) in Indiana where the ANR values were reported for smooth bromegrass, orchardgrass, and timothy across N rates that ranged from 0 to 1344 kg N ha-1. Under favorable conditions, maximum ANR for all N treatments and species in that study was achieved at 168 kg N ha-1, which corresponded with ANR values of 0.72 kg kg-1 for smooth bromegrass, 0.67 for orchardgrass, and 0.51 for timothy. In another year, when rainfall was 25% less, ANR was reduced and ranged from 0.14 to 0.49 kg kg-1 among all species.
In general, ANR values averaged over years ranged from 0.28 to 0.47 kg kg-1 for Kentucky bluegrass, 0.17 to 0.44 for smooth bromegrass, and 0.32 to 0.50 for orchardgrass at both Arlington and Lancaster. However, a few important exceptions showed the potential for ANR results for a given harvest to deviate greatly from these ranges depending on precipitation and species-associated growth habit. For example, a dry period during late summer of 1996 reduced the third-harvest DM yields for all species from 20 to 35% of total annual DM yield in 1994 or 1995 to 10 to 23% of total annual yield (data not shown). Yield reductions in the third harvest of 1996 were also related to reduced third-harvest ANR values where Kentucky bluegrass ranged from 0.06 to 0.23 kg kg-1, smooth bromegrass from -0.06 to 0.28, and orchardgrass from 0.20 to 0.39.
Reduced yield and ANR was least extreme for orchardgrass in August 1996 because of differences in rooting depth and seasonal yield distribution. Orchardgrass is more drought tolerant than either timothy or Kentucky bluegrass but not as drought tolerant as smooth bromegrass (Christie and McElroy, 1995). Orchardgrass is reported to have an extensive root system at depths of 0.4 to 0.6 m below the soil surface (Evans, 1978) and is able to access moisture from these depths during dry periods. Conversely, Kentucky bluegrass is notorious for being shallow rooted and prone to reduced production during dry periods (Sprague and Graber, 1938; Mortimer and Ahlgren, 1944). Relatively stable ANR values for orchardgrass during drought coupled with a more seasonally uniform growth habit suggest that it may be the most prodigious N user among the species studied and would provide the least risk to producers in terms of N losses to the environment.
Similar trends of divergence by location and species for NUE were generally consistent with previously described trends for yield and ANR though divergence was slightly less (Fig. 3). All NUE models had a negative slope, indicating reduced NUE with increased N fertilization rate regardless of soil type. In general, the NUE of orchardgrass was markedly greater than that of Kentucky bluegrass, which was greater than smooth bromegrass. Diminishing NUE was related to increased N fertilization rates and was particularly poignant for orchardgrass at both locations. This result was caused by very high NUE for 56112, and 168 kg N ha-1 fertilization rates followed by NUE that trends lower at 224 or 336 kg N ha-1. At higher rates of fertilizer N, NUE was similar among the three grasses.
In general, NUE values averaged over years ranged from 12 to 18 kg DM kg-1 for Kentucky bluegrass, 9 to 16 for smooth bromegrass, and 11 to 28 for orchardgrass at both locations. However, similar to ANR, a few important exceptions demonstrated how NUE for any given harvest can deviate greatly from these ranges depending on precipitation and species-associated growth habit. For example, the dry period during late summer of 1996 that reduced third-harvest DM yield was also related to reduced third-harvest NUE values where Kentucky bluegrass ranged from 3 to 9 kg DM kg-1, smooth bromegrass from 0 to 11, and orchardgrass from 8 to 14 (data not shown). As with ANR, the negative impact of dry conditions on yield and NUE was least for orchardgrass in August 1996.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
We conclude that producers in the North-Central region could benefit by using orchardgrass in forage systems where high N rates are combined with either soil type studied, provided the soil water status and prevailing climatic conditions are considered before applying N and that harvested forage could be utilized effectively in balanced animal rations. This would help to optimize DM yield, ANR, and NUE while managing risk in farm nutrient management plans. Though we cannot know for certain what the NUE for orchardgrass is beyond 336 kg ha-1, the slope of the fitted NUE model indicates that it would not be prudent to exceed this N rate because of the high likelihood of increased N losses to the environment and costs to producers. There may be other reasons for choosing Kentucky bluegrass (e.g., improved mixture compatibility with certain legumes; Zemenchik et al., 2001) or smooth bromegrass (e.g., soil conservation on sloping landscapes; Zemenchik et al., 1996) for forage production in the region.
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Q. M. Ketterings, J. H. Cherney, G. Godwin, T. F. Kilcer, P. Barney, and S. Beer Nitrogen Management of Brown Midrib Sorghum x Sudangrass in the Northeastern USA Agron. J., September 10, 2007; 99(5): 1345 - 1351. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S. Mangiafico and K. Guillard Nitrate Leaching from Kentucky Bluegrass Soil Columns Predicted with Anion Exchange Membranes Soil Sci. Soc. Am. J., January 1, 2007; 71(1): 219 - 224. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Zhang, R. Gavlak, A. Mitchell, and S. Sparrow Solid and Liquid Cattle Manure Application in a Subarctic Soil: Bromegrass and Oat Production and Soil Properties Agron. J., October 3, 2006; 98(6): 1551 - 1558. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Burner and C. T. MacKown Nitrogen Effects on Herbage Nitrogen Use and Nutritive Value in a Meadow and Loblolly Pine Alley Crop Sci., March 27, 2006; 46(3): 1149 - 1155. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. P. Beyaert and R. C. Roy Influence of Nitrogen Fertilization on Multi-Cut Forage Sorghum-Sudangrass Yield and Nitrogen Use Agron. J., October 19, 2005; 97(6): 1493 - 1501. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Burner and C. T. MacKown Herbage Nitrogen Recovery in a Meadow and Loblolly Pine Alley Crop Sci., August 1, 2005; 45(5): 1817 - 1825. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. M. Fitzpatrick and K. Guillard Kentucky Bluegrass Response to Potassium and Nitrogen Fertilization Crop Sci., September 1, 2004; 44(5): 1721 - 1728. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Hall, D. B. Beegle, R. S. Bowersox, and R. C. Stout Optimum Nitrogen Fertilization of Cool-Season Grasses in the Northeast USA Agron. J., July 1, 2003; 95(4): 1023 - 1027. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Singer and K. J. Moore Nitrogen Removal by Orchardgrass and Smooth Bromegrass and Residual Soil Nitrate Crop Sci., July 1, 2003; 43(4): 1420 - 1426. [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 | |||
