PDF version
for printing

Impact
Statement

© 2006 Plant Management Network.
Accepted for publication 7 September 2006. Published 11 December 2006.

Split Application of Nitrogen on Temperate Perennial Grasses in the Northeast USA

Debbie J. R. Cherney, Department of Animal Science, Cornell University, Ithaca, NY 14853; and Jerome H. Cherney, Department of Crop & Soil Sciences, Cornell University, Ithaca, NY 14853

Corresponding author: Debbie J. R. Cherney. djc6@cornell.edu

Cherney, D. J. R., and Cherney, J. H. 2006. Split application of nitrogen on temperate perennial grasses in the northeast USA. Online. Forage and Grazinglands doi:10.1094/FG-2006-1211-01-RS.

Abstract

Timing of N applications on grasses can impact agronomic forage traits as well as the environment. We studied the impact of split N applications on yield, quality, and N recovery in orchardgrass, reed canarygrass, and tall fescue. Nitrogen fertilizer was applied at 200 lb N per acre at spring greenup, or 100 lb N fertilizer per acre at spring greenup followed by 100 lb N per acre after first cut, or 100 lb N fertilizer/acre at spring greenup followed by 50 lb N per acre after first cut and 50 lb N per acre after second cut. Plots were harvested three times per season for three years at two locations in central New York. There was about a 12% yield increase associated with split application of N in the spring and after first cut, but no yield advantage to using a three-way split of N over the season. Nitrogen yield and apparent N recovery were not influenced by timing of N application. Highest N fertilizer rates prior to the first and second cuts resulted in increased crude protein and reduced neutral detergent fiber, but fertilization after second cut had no impact on third cut quality. Fertilization with a split application of N will maximize yield, most of which is suitable forage for lactating dairy cows.

Introduction

Wise use of perennial grasses on dairy farms can reduce nutrient management problems (15). Grass-based diets have been shown to produce as much milk as alfalfa-based diets for dairy cattle (3). While it is possible to get three, four, or five cuts of high quality grass in a single season in the northeastern USA, it is often more practical to take two cuts of lactating dairy quality forage in spring and early summer, followed by a third cut in the fall suitable for non-lactating animals (7).

Dry matter yield responds primarily to N fertilizer in grasses (1). The increasing cost of commercial N fertilizer makes it imperative to optimize N use efficiency. While most states recommend split application of N fertilizer, a study using reed canarygrass (Phalaris arundinacea) in southern Minnesota concluded that optimum dry matter (DM) yield and crude protein (CP) content occurred with a single spring application of N fertilizer (14). Timing of N fertilization applications on perennial grasses potentially impacts total yield, the proportion of total yield suitable for lactating cows, quality of the forage, and N recovery. The objective of this study was to determine the impact of N application timing on forage yield and quality, and on N recovery in perennial grasses.

Experimental Details

Four field replicates were established at two sites in central New York. ‘Rival’ reed canarygrass, ‘Okay’ orchardgrass (Dactylis glomerata), and ‘Select’ tall fescue (Festuca arundinaceae) were sown in the spring of 2000 on a Niagara silt loam (fine-silty, mixed, active, mesic Aeric Endoaqualfs) soil with 0 to 2% slope in Ithaca, NY and on a Mardin silt loam (coarse-loamy, mixed, active, mesic Typic Fragiudepts) soil with 0 to 6% slope in Dryden, NY. Each replicate consisted of three species blocks (main plots) that were each split into subplots (3.1 by 6.1 m) that received 0, 100, or 200 lb N per acre. Zero-N plots were included to allow estimation of N recovery. All three N treatments totaled 200 lb N per acre annually, with 200 lb applied in the spring (200-0-0), or split-applied prior to spring growth and second cut regrowth (100-100-0), or split-applied prior to spring growth and prior to both regrowth periods (100-50-50). Both sites were fertilized with P and K according to soil test recommendations.

Broadleaf weeds were controlled by annual applications of either 2,4-D amine (2,4-dichlorophenoxy acetic acid dimethylamine) at 0.12 lbs/acre in the spring or Banvel (0.28 lb/gal dicamba) at 0.10 lbs/acre in the fall. Sites were harvested three times per season in 2001, 2002, and 2003. The goal for the first two cuts of the season was 55% neutral detergent fiber (NDF), which is suitable quality for lactating dairy cows. These harvests were followed by a late fall cut. One exception was the Dryden site in 2002, which was droughty in late summer and did not produce any harvestable yield for a third cut. Three- × 20-ft harvest strips at a 4-inch stubble height were taken for yield with a Carter Harvester (Carter Mfg. Co., Inc., Brookston, IN).

Forage samples were oven-dried at 60°C for dry matter (DM) determination and ground in a cyclone mill (Udy Corp., Ft. Collins, CO) to pass a 1-mm screen. Nitrogen concentration of samples was determined using a Leco N analyzer (LECO Corp., St. Joseph, MI) with Dumas combustion (11,16). Samples (0.5 g) were analyzed for NDF using the procedure described by Van Soest et al. (13), except that the ANKOM fiber analyzer (ANKOM Technology, Maceon, NY) was used. In vitro digestibility was determined by incubating ground samples in buffered rumen fluid with urea for 48 h (9) using the ANKOM incubator (5). Digested residues were subject to NDF analysis to determine in vitro true digestibility (IVTD). Digestibility of NDF (NDFD) was calculated as the proportion of the NDF digested after 48 h incubation. Milk2000 (11) was used to estimate milk/acrere and relative forage quality (RFQ). In Milk2000, forage energy intake is calculated for a 1350-lb milking cow consuming a 30% NDF diet. Energy for maintenance of this cow is subtracted, and the remainder is an estimate of the energy available for conversion to milk.

Statistical design was a split-plot with two locations and four replications per site for three years. A split-plot analysis of variance was used to test for statistical significance of treatment effects and interactions using PROC MIXED in SAS, version 7.0 software (SAS Institute Inc., Cary, NC). Location, species, and fertilization timing were considered fixed effects, while year and rep were considered random effects. The PDIFF option of the LSMEANS statement was used to generate differences among least square means, with comparisons at P ≤ 0.05. The zero-N plots were included in the experiment to allow estimation of N recovery, but were not included in the statistical analysis of split N application treatments.

Environment

Precipitation at both sites was at or above normal during the April to September growing season for all three years (range of 21 to 32 inches of rainfall), although the Dryden site received little rainfall during the second regrowth period of 2002, resulting in no harvestable growth that fall. Stands remained in excellent condition throughout the experimental period. At the conclusion of the experiment there were no differences in soil pH, organic matter, K, P, Mg, or Ca among the N treatments. Soil test K at the conclusion of the experiment was significantly higher (P < 0.05) in reed canarygrass plots compared to the other grasses. There were no other differences in soil parameters among species. At the conclusion of a six-year experiment (6) soil test K for reed canarygrass stands also was significantly higher than either orchardgrass or tall fescue. Since annual K uptake for reed canarygrass was not different from the other species (6), the higher soil test K in reed canarygrass plots may be caused in part by the vigorous root system of reed canarygrass.

Impact on Yield and Yield Distribution

Spring cut occurred in late May or early June, and second cut in late June to early July. A few pre-harvest samples were analyzed for NDF to help with harvest date estimation. All fall cuts occurred in September. Although DM yields were not statistically different among grass species at P ≤ 0.05, tall fescue did produce significantly greater milk yields (Table 1). Dairy cows fed fescue silage did produce more milk than cows fed orchardgrass silage, both fed in balanced rations (3). The Ithaca location produced significantly more DM than Dryden, but was not significantly greater in milk yield. There was no difference in either DM or milk yield between a two-way split application of N fertilizer versus a three-way split of N, averaged over three years, two locations, and three species (Table 1). Both split N applications resulted in greater DM yields than a single spring application of N.

Table 1. Yield, yield components, and milk yield of total forage as influenced by species, split application of N, and location (four replicates and three years).

 ^v Milk per acre was estimated using Milk2000 (10).

 ^w For each parameter, means followed by a different letter within a column differ (P ≤ 0.05).

 ^x Standard error of the difference.

 ^y Split application of N fertilizer in the spring, after first harvest, and after second harvest.

 ^z Probability level.

The amount of DM produced of lactating dairy quality (approximately 55% NDF) from the first two cuts was not influenced by N treatment (Table 1). This was primarily due to a significant (P ≤ 0.05) linear decrease in the proportion of DM yield in the first two cuts with increased number of split applications of N (1, 2, or 3 N applications). A single, early-season application of N will skew yield distribution towards the first cutting (14). The Dryden location had greater DM yield and a greater proportion of yield in the first two cuts, compared to Ithaca. As there was no third cut at Dryden in 2002, no yield proportion data was included in the analysis for that site-year combination. A significant location × N split interaction for yield proportion (Table 1) was due to changes in magnitude of response, with the Dryden yield proportion greater than Ithaca for all three species.

Impact on N Fertilizer Utilization

Timing of N fertilizer application had minimal effect on N utilization (Table 2). Neither N yield nor N recovery were influenced by timing of N application. Apparent N recovery averaged 53% of applied N fertilizer, and was not influenced by grass species or location. Hall et al. (8) applied 240 lb N per acre in Pennsylvania and calculated an apparent N recovery of 58% for orchardgrass, 64% for tall fescue, and 52% for timothy (Phleum pratense). In 1994 in Minnesota, Vetsch et al. (14) optimized reed canarygrass yield with approximately 200 lb N per acre, with an apparent N recovery of approximately 51%. Location × N split interactions were significant for both N yield and N recovery. The N yield interaction was due to changes in magnitude of response, as N yield was greatest for all N treatments at the Ithaca site. The 200-0-0 treatment resulted in greater N recovery at Dryden compared to Ithaca (50% vs 56%) (P ≤ 0.05), while N recovery for the other N treatments was similar at both locations.

Table 2. Nitrogen yield, apparent N recovery, and crude protein for each harvest as influenced by species, split application of N, and location (four replicates and three years).

 ^v Quantity harvested minus the quantity in the zero-N check, divided by the amount of N applied.

 ^w For each parameter, means followed by a different letter within a column differ (P ≤ 0.05).

 ^x Standard error of the difference.

 ^y Split application of N fertilizer in the spring, after first harvest, and after second harvest.

 ^z Probability level.

As expected, crude protein concentration reflected the quantity of N applied at spring greenup and after first cut (Table 2). The 200-0-0 treatment resulted in the greatest CP concentration in the spring, while the 100-100-0 treatment resulted in greatest CP concentration at second cut. Concentration of CP at cut 3 was similar for all N treatments. Reed canarygrass was significantly greater in CP than the other species at cut 1, and tall fescue was significantly lower in CP than the other species at cut 3. Fall-harvested tall fescue was significantly lower in CP than orchardgrass in a northern New York study (4). Four two-way interactions were significant for cut 2 and cut 3 CP. Three of these interactions were the result of reed canarygrass CP not responding to either location or N treatment, compared to the other species. The fourth significant two-way interaction (cut 2, location × N split) was due to differences in magnitude of response.

Impact on Fiber, Digestibility, and RFQ

In general, forage quality was not greatly impacted by N treatment, location, or species (Tables 3 and 4). High N fertilization of grasses tended to lower NDF slightly in both cut 1 and cut 2 (Table 3) as observed in the past (2). Neutral detergent fiber was in the targeted range (low to mid-50s) for both locations. Reed canarygrass was lower in NDF and NDFD at cut 3, compared to the other species, similar to that observed in another study that included these three species (6). In vitro true digestibility and fiber digestibility were lowest at cut 3, resulting in correspondingly lower RFQ values.

Table 3. Neutral detergent fiber (NDF) and in vitro true digestibility (IVTD) for each harvest as influenced by species, split application of N, and location (four replicates and three years).

 ^w For each parameter, means followed by a different letter within a column differ (P ≤ 0.05).

 ^x Standard error of the difference.

 ^y Split application of N fertilizer in the spring, after first harvest, and after second harvest.

 ^z Probability level.

Table 4. Neutral detergent fiber digestibility (NDFD) and relative forage quality (RFQ) for each harvest as influenced by species, split application of N, and location (four replicates and three years).

 ^v Relative forage quality was estimated using Milk2000 (10). RFQ is based on the concept of digestible dry matter intake relative to a standard forage (12).

 ^w For each parameter, means followed by a different letter within a column differ (P ≤ 0.05).

 ^x Standard error of the difference.

 ^y Split application of N fertilizer in the spring, after first harvest, and after second harvest.

 ^z Probability level.

A number of two-factor interactions were significant for quality components, most were due to changes in magnitude of response. For several species × location interactions, reed canarygrass values were similar between locations while the other two species had higher values at the Dryden location, resulting in significant interactions. Two-factor interaction responses were similar among IVTD, NDFD, and RFQ, as they are related parameters. Correlations between IVTD and NDFD ranged from r = 0.93 to 0.96 among the three cuts, and correlations were very high between IVTD and RFQ (r = 0.97 to 0.98 among cuts). Neutral detergent fiber was correlated with IVTD for cut 1 (r = -0.85), because cut 1 occurred before NDF concentration reached a plateau. In late summer regrowth, NDF concentration reaches a plateau, while lignin continues to accumulate, eliminating the relationship between NDF and IVTD for cut 3 (r = -0.03).

Impact on Economics

Since timing of N fertilizer application had essentially no effect from an environmental standpoint, the practice can be evaluated from a strictly economic standpoint. A partial economic budget for N application treatments can be calculated by assigning monitary value to the forage and subtracting the cost of commercial application of N ($7/acre/application). All other costs were assumed to be similar across N application treatments. Forage from cuts 1 and 2 was assumed to be of equal value ($100/ton of hay equivalent). Third cut forage was valued at $80/ton. Since the yield and yield distribution with a two-way split application of N are not significantly different from a three-way split application of N, the economic difference between these treatments is the cost of an additional N application ($7/acre). Averaged over years and sites, a two-way split application of N would result in an additional $27/acre return over a single application of N in the spring, using the assumptions above.

Summary and Conclusions

This study focused on N utilization under a three-cut system, with the production of lactating dairy quality forage for the first two cuts and a fall cut of non-lactating quality forage. The first two cuts for all three grass species and all three N fertilization rates were near optimum NDF and relatively high in IVTD and NDFD, suitable for lactating dairy cattle. The final cut in the fall was higher in NDF and considerably lower in NDFD and CP than earlier cuts, but suitable for non-lactating animals. A 12% yield increase resulted from split application of N in the spring and after first cut, but no yield advantage to using a three-way split of N over the season. Nitrogen yield and apparent N recovery were not influenced by timing of N application. Fertilization with a split application of 200 lbs N per acre with two high quality cuts early in the season will maximize yield, most of which is suitable forage for lactating dairy cows.

Acknowledgments

This research was supported in part by the Cornell University Agricultural Experiment Station federal formula funds, Project No. NYC-1277455 received from Cooperative State Research, Education, and Extension Service, U.S. Department of Agriculture. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the U.S. Department of Agriculture. The assistance of Samuel Beer and Leon Hatch is especially appreciated.

Literature Cited

1. Allinson, D. W., Guillard, K., Rafey, M. M., Grabber, J. H., and Dest W. M. 1992. Response of reed canarygrass to nitrogen and potassium fertilization. J. Prod. Agric. 5:595-601.

2. Cherney, D. J. R., and Cherney, J. H. 1997. Grass forage quality and digestion kinetics as influenced by N fertilization and maturity. J. Applied Anim. Res. 11:105-120.

3. Cherney, D. J. R., Cherney, J. H., and Chase, L. E. 2004. Lactation performance of Holstein cows fed fescue, orchardgrass or alfalfa silage. J. Dairy Sci. 87:2268-2276.

4. Cherney, D. J. R., Cherney, J. H., and Mikhailova, E. A. 2002. Nitrogen utilization by orchardgrass and tall fescue from dairy manure or commercial fertilizer nitrogen. Agron. J. 94:405-412.

5. Cherney, D. J. R., Traxler, M. J., and Robertson, J. B. 1997. Use of Ankom fiber determination systems to determine digestibility. NIRS Forage and Feed Testing Consortium Ann. Conf., 19-20 Feb., 1997, Madison, WI. NIRS Forage and Feed Testing Consortium, St. Paul, MN.

6. Cherney, J. H., and Cherney, D. J. R. 2005. Agronomic responses of cool-season grasses to low-intensity harvest management and low potassium fertility. Agron. J. 97:1216-1221.

7. Cherney, J. H., Cherney, D. J. R., and Parsons, D. 2006. Grass Silage Management Issues. Pages 37-49 in: Proc. from "Silage for Dairy Farms: Growing, Harvesting, Storing, and Feeding". NRAES-181. 23-25 Jan., 2006. Harrisburg, PA. Nat. Resource, Agric., and Engineering Serv., Ithaca, NY.

8. Hall, M. H., Beegle, D. B., Bowersox, R. S., and Stout, R. C. 2003. Optimum nitrogen fertilization of cool-season grasses in the Northeast USA. Agron. J. 95:1023-1027.

9. Marten, G. C., and Barnes, R. F. 1980. Prediction of energy digestibility of forages with in vitro rumen fermentation and fungal enzyme systems. Pages 61-71 in: Standardization of Analytical Methodology for Feeds. W. J. Pigden, C. C. Balch, and M. Graham, eds. Int. Devel. Res. Ctr., Ottawa, ON, Can.

10. Shaver, R., Undersander, D., Schwab, E., Hoffman, P., Lauer, J., Combs, D., and Coors, J. 2004. Milk2000: Combining yield and quality into a single term. Univ. of Wis. Coop. Ext. Publ., Madison, WI.

11. Tate, D. F. 1994. Determination of nitrogen in fertilizer by combustion: Collaborative study. J. AOAC Int. 77:829-839.

12. Undersander, D., and Moore, J. E. 2002. Relative forage quality. Focus on Forage 4:1-2.

13. Van Soest, P. J., Robertson, J. B., and Lewis, B. A. 1991. Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J. Dairy Sci. 74:3583-3597.

14. Vetsch, J. A., and Russelle, M. P. 1999. Reed canarygrass yield, crude protein, and nitrate N response to fertilizer N. J. Prod. Ag. 12:465-471.

15. Wang, S. J., Fox, D. G., Cherney, D. J. R., and Chase, L. E. 2000. Whole herd optimization with the Cornell Net Carbohydrate and Protein System. III. Application of an optimization model to evaluate alternatives to reduce nitrogen and phosphorus mass balance. J. Dairy Sci. 83:2160-2169.

16. Wiles, P. G., Gray, I. K., and Kissling, R. C. 1998. Routine analysis of proteins by Kjeldahl and Dumas methods: review and interlaboratory study using dairy products. J. AOAC Int. 81:620-632.