PDF version
for printing

Impact
Statement

© 2009 Plant Management Network.
Accepted for publication 2 June 2009. Published 12 June 2009.

Nitrogen Needs of Teff Managed as Forage Crop in New York

Michael Hunter, Cornell Cooperative Extension of Jefferson County, Watertown, NY 13601; Quirine M. Ketterings, Nutrient Management Spear Program, Department of Animal Science, Cornell University, Ithaca, NY 14853; Jerome H. Cherney, Department of Crop and Soil Sciences, Cornell University, Ithaca, NY 14853; Peter Barney, Cornell Cooperative Extension of St Lawrence, Canton, NY 13617 (currently Barney Agronomic Services, Potsdam, NY 13676); Tom Kilcer, Cornell Cooperative Extension of Rensselaer County, Troy, NY 12180; and Greg Godwin, Nutrient Management Spear Program, Department of Animal Science, Cornell University, Ithaca, NY 14853

Corresponding author: Quirine M. Ketterings. qmk2@cornell.edu

Hunter, M., Ketterings, Q. M., Cherney, J. H., Barney, P., Kilcer, T., and Godwin, G. 2009. Nitrogen needs of teff managed as forage crop in New York. Online. Forage and Grazinglands doi:10.1094/FG-2009-0612-01-RS.

Abstract

Teff [Eragrostis tef (Zucc.) Trotter] is an annual C₄ grass native to Ethiopia. It has potential as an emergency hay, pasture, or silage crop in the United States. Our objective was to determine the optimum N rate for single and multiple cut systems evaluating both dry matter (DM) yield and forage quality. Eight N-rate studies (0, 50, 75, and 100 lb/acre) were conducted over 2 years (four as single-cut systems and two each as 2- or 3-cut systems). Yields ranged from 0.8 to 2.4, 3.3 to 3.6, and 4.8 to 4.9 ton/acre for the 1-, 2- and 3-cut systems, respectively. Nitrogen application increased yield at all but two sites which had manure or rotation nutrient inputs. Only at one site was there a yield benefit from applying > 50 lb N per acre. Crude protein levels of 1st cutting increased 3.3% units when 50 lb per acre was added with similar trends for 2nd and 3rd cuttings. For multiple-cut systems, an application of 50 lb N per acre per cut reduced forage neutral detergent fiber (NDF) while NDF digestibility was not impacted. We conclude teff grown under New York growing conditions needs 50 lb N per acre per cut for maximum DM yield.

Introduction

Teff [Eragrostis tef (Zucc.) Trotter] is an annual C₄ grass native to Ethiopia (9). It is adapted to environments ranging from drought-stressed to waterlogged soil conditions (17). Teff is a major cereal crop in Ethiopia but it is also grown as a hay crop (19). In its native habitat maximum production occurs with a growing season rainfall of 17 to 22 inches (9) and a temperature range of 50 to 85°F (17).

Challenging spring weather in the humid Northeast, where growing seasons are already short, has resulted in a search for fast-growing high-quality forage alternatives to corn that can still be planted in June and July (2,12). Earlier work in New York explored the potential of brown midrib sorghum × sudangrass hybrids (BMR S×S) as "emergency forage" (10,11,12). These studies showed 3.5 to 6.2 ton/acre of high quality BMR S×S forage could be produced in 2-cut harvest systems but that 110 to 130 lb N per acre per cut was needed for fields without a manure history (10). Although these studies showed BMR S×S to be a promising alternative to corn, the high N fertilizer need is not conducive to growing BMR S×S in non-manured fields. Furthermore, the high moisture content of the forage can cause harvest challenges (8,11).

Because teff is well-adapted to heat and drought, as well as excess water, it has potential as an emergency hay, pasture, or silage crop for dairy or beef cattle or as an annual hay crop for the horse market. Like BMR S×S, it establishes quickly resulting in ground coverage that could reduce erosion and N leaching (Fig. 1). Teff has been grown successfully in South Dakota (19), Nevada (4,5,6), Oregon and Washington (16,17), and the Mid-Atlantic region (1,8). Yields in Oregon and Washington ranged from 5.1 to 6.1 ton/acre (2-cut systems) (16). In Nevada, teff yielded 3.9 to 4.6 ton/acre (6), whereas in Pennsylvania, yield ranged from 4.6 to 6.9 ton/acre (8). Although these yields imply great potential for teff as emergency forage, little is known about its yield potential, forage quality, and N needs when grown in New York environments.

Fig. 1. Teff, Eragrostis tef (Zucc.) Trotter, is an annual C4 grass native to Ethiopia with potential as emergency forage crop in New York.

Our objective was to determine the N needs for teff grown in New York as a forage crop harvested in single- and multiple-cut systems, evaluating both dry matter yield and feed quality.

Experimental Details

"Corvallis" Teff was established in Jefferson, St. Lawrence, and Columbia counties in New York in 2006 (Table 1). At each site four N rate treatments (0, 50, 75, and 100 lb/acre) were established at planting in a randomized complete block design with four replications. Plots were 8 by 40 ft. Phosphorus and potassium fertilizer were broadcast before planting at three of four locations to ensure no P or K limitations (Table 2). For all sites forage was tested for K to verify that forage K levels were 1.2% or higher (3). Site 3 had received an estimated 10 ton/acre rate of bedded pack cattle manure in the fall of 2005.

Table 1. Soil series, cropping history and planting and harvest dates of eight teff trials conducted in 2006 and 2007 in New York.

 ^w BUK = buckwheat, GRT = forage grasses, OAT = oats, RYE = ryegrass cover crop, SSG = sorghum × sudangrass, STC = spring triticale, SWC = sweet corn.

 ^x Site 3 received an estimated 10 ton/acre rate of bedded pack cattle manure in fall 2005.

 ^y Originally seeded 7 June 2007 but reseeded 19 June 2007 to replace poor stand.

 ^z Sites 2 and 6 were 950 ft apart. Sites 1 and 5 were neighboring fields.

Table 2. Initial soil fertility status (0 to 20 cm depth) and phosphorus and potassium fertilizer additions at planting for eight teff trials conducted in 2006 and 2007 in New York.

 ^w Morgan extraction (14); L = low; M = medium; H = high; VH = very high.

 ^x Soil organic matter (OM) was determined by loss-on-ignition at 500°C (18).

 ^y Soil fertility samples were accidently disposed of after soil nitrate analyses and before general soil fertility analyses could be done.

 ^z Site 3 had received an estimated 10 ton/acre rate of bedded pack cattle manure in the fall of 2005.

Climate data were obtained from weather stations located within 2 miles of the plots (Northeast Regional Climate Center CLIMOD database, 2006, 2007). The average monthly temperature was derived from calculated daily averages ([maximum - minimum]/2). All locations were planted after mid-June in 2006 when average air temperatures exceeded 64°F (Table 3). Each of these sites was harvested once (single-cut system) at the end of August (Table 1). Harvest areas were 3 by 20 ft. Forage fresh weights were determined and subsamples were analyzed for moisture content, crude protein (CP), acid detergent fiber (ADF), neutral detergent fiber (NDF), and digestibility of NDF (dNDF at 48 h).

Table 3. Monthly precipitation and temperature for eight sites where teff trials were conducted in New York in 2006 and 2007.

The CP concentration of the forage samples was determined by dry combustion (7) using a Leco FP-528 C/N analyzer (Leco Instruments Inc., St. Joseph, MI), assuming a CP:N ratio of 6.25. Neutral detergent fiber was determined according to Van Soest et al. (20) using the ANKOM system (ANKOM Technology, Fairport, NY). In vitro true digestibility was determined according to ANKOM Application Note 11/00 using the Daisy II 200/220 in vitro incubator and an ANKOM200/220 fiber analyzer. The 48-h NDF digestibility (dNDF) was calculated as (NDF – IVTD residue at 48 h)/NDF × 100.

Soil samples (0 to 8 inches; 5 per plot) were taken before N application in June. Samples were dried at 65°C, ground and passed through a 2-mm sieve before analyses at the Cornell Nutrient Analysis Laboratory, Ithaca, NY. Soil organic matter was determined by loss-on-ignition upon exposure for 2 h to 500°C (18). Available P, K, Ca, and Mg were determined by shaking dried samples in a 1:5 (v/v) ratio for 15 min in Morgan solution (1 N sodium acetate buffered at pH 4.8) and filtering the extract through a Whatman No. 2 equivalent filter paper (14) before analysis using an Inductively Coupled Plasma Atomic Emission Spectrometer (JY70 Type II, Jobin Yvon, Edison, NJ) for K, Ca, and Mg. Morgan extractable PO₄-P was measured colorimetrically using an Alpkem automated rapid flow analyzer (RFA/2-320) (OI Corporation, College Station, TX). Soil pH was measured in a 1:1 (w/v) water extract.

In 2007, four additional trials were established with the same treatments (0, 50, 75, and 100 lb N per acre per cut; Sites 5 through 8), design and number of replications. These trials, located in Jefferson, St. Lawrence, Columbia, and Tompkins counties in New York (Table 1), were harvested in a 2-cut system (2 sites) or 3-cut system (2 sites), depending on the growing conditions for each individual location. The warmer spring than in 2006 allowed for planting in the last week of May or first week of June although one site (Site 5) had to be reseeded on June 19 to replace a poor stand. Harvest took place between mid July and early August (first cutting), followed by second cutting mid August with the exception of the site in Jefferson Co., which was harvested on October 4 due to drought-induced lack of growth in August and September. At two locations (Sites 7 and 8) a third cut was harvested between September 20 and 9 October (Table 1). Harvested samples were processed as in 2006. Soils were sampled for Morgan extractable NO₃-N prior to N application in May or June and after each cutting. Samples were processed as described before. Morgan extractable NO₃-N was measured colorimetrically using an Alpkem automated rapid flow analyzer (RFA/2-320) (OI Corporation, College Station, TX).

The experimental design was a randomized complete block in four replications. A mixed model analysis of the data was conducted with PROC MIXED in SAS version 7.0 (13). For yield and quality data for first cutting, if the Site × N rate interaction was significant (P < 0.05; PROC MIXED with N rate, location and their interaction term as fixed effects and replicate within site as random effect), each site was analyzed individually with N rate as a fixed effect and replicates as random effects. For the sites established in 2007, yield and quality data for each cutting were analyzed by site. The PDIFF option of the LSMEANS statement was used to generate differences among least square means, with comparisons at P ≤ 0.05.

Yield

Teff K levels at harvest ranged from 1.2 to 2.8% of DM (results not shown), suggesting no K limitations for any of the sites or cuts (3). First cut yields ranged from 0.8 to 2.4 ton/acre DM with maximum yields at the various sites ranging from 1.6 to 2.5 ton/acre DM (Table 4). These ranges are similar to the 1.0 to 3.2 ton/acre teff yields obtained in first cutting under medium or high irrigation treatments in experiments at the Klamath Experiment Station in Oregon (17), and the 2.3 to 2.9 ton/acre yields for first cuts reported for Oregon and Washington (16).

Table 4. Effect of N application rate on first harvest yield of teff grown at eight location/years in New York State^x.

 ^x Yield values within a row followed by the same letter are not significantly different at P = 0.05.

Nitrogen application increased first cut DM yield at all but two single-cut locations (Sites 3 and 4) and one 3-cut location (Site 8) (Table 4). Sites 3 and 4 were the highest yielding (1.7 and 2.4 ton/acre) single-cut sites. The bedded pack cattle manure applied at Site 3 the autumn before teff seeding might explain the lodging observed at this location upon N fertilizer addition. At Site 4, the teff crop followed a grass sod in the rotation (oats was seeded in 2005 but the seeding failed resulting in no crop, or nutrient, removal in 2005) and likely benefitted from N released from decomposition following plow-down of the grass/weeds biomass. At Site 8, average yield of the zero-N plots was lower than for all other plots at each cutting but for 1st cutting, the differences were not statistically significant. At this location, a 9-year sod was rotated to rye and buckwheat before seeding of teff, which possibly explains the lack of a significant N response for first cutting.

At Site 6, first-cut yield increased linearly with N application rate and optimum N rate could not be determined. For all other trials, N application increased yield but no yield benefit was obtained for N application rates exceeding 50 lb/acre. The average yield response to 50 lb/acre was 0.7 ton/acre (ranging from 0.4 to 0.9 ton/acre across the N-responsive sites). These results are very consistent with Roseberg et al. (17) who reported the addition of 40 to 45 lb N per acre (across three sites) resulted in a first cut yield response of 0.7 ton/acre with no further gain in yield upon doubling of the N application.

For three of the four multiple-cut trials, 50 lb N per acre per cut resulted in maximum yield (Table 5). The fourth site (Site 6) showed a yield increase with N addition up to 100 lb/acre per cut. Total DM yield of multiple cut systems ranged from 3.3 to 3.6 ton/acre for 2-cut systems and from 4.8 to 4.9 ton/acre for the 3-cut systems (Table 6). Such DM yields are comparable to 3.5 to 4.3 ton/acre DM for BMR S×S at sites without prior organic N inputs and 4.6 to 6.2 ton/acre at sites with a manure or sod history (10). Furthermore, for the 3-cut system, yield was only 1.1 to 1.2 ton/acre DM less than average corn silage yield in New York in 2007 (6.0 ton/acre) (15). Thus, based on DM yield, teff could be a competitive alternative to BMR S×S as emergency forage.

Table 5. Effect of N application rate on dry matter (DM) yield of teff grown in 2- or 3-cut harvest systems in New York State^x.

 ^x Yield values within a row followed by the same letter are not significantly different at P = 0.05.

Table 6. Effect of N application rate on crude protein (CP), neutral detergent fiber (NDF), fiber digestibility (dNDF), and acid digestible fiber (ADF) of first cutting of teff (average of eight different location/years in New York State^x).

 ^x Yield values within a row followed by the same letter are not significantly different at P = 0.05.

Quality

Crude protein levels of 1st cut teff increased with N application from an average of 12.2% without N application to 15.5% and 18.6% when 50 and 100 lb N per acre were applied, respectively (Table 6). These results are consistent with average CP levels of 13.1% across three locations in Oregon and Washington (16) and the CP increases upon N addition in studies by Roseberg et al. (17). Similar CP responses were reported for BMR S×S trials in New York (10).

Across all sites, N addition decreased NDF of first cutting forage from 65.4% (no N applied) to 62.2% (100 lb N per acre), and ADF from 36.3% (no N applied) to 33.4% (100 lb N per acre) (Table 6). However, when 50 lb N per acre were added, only NDF decreased significantly (from 65.4 to 63.6%). Nitrogen addition did not affect NDF digestibility (average of 70.6%). Similar NDF and ADF concentrations were obtained in studies in Medford, OR, where in a low irrigation treatment N addition of 40 to 45 lb/acre to an N deficient site reduced teff NDF concentration from 65.8 to 56.6% and ADF from 35.4 to 29.4% (17). However, research at other locations in Oregon show no change in NDF or ADF with N addition (17). Our New York data add to the existing literature that suggest that although a small decline in NDF and ADF can occur, N addition to obtain maximum yield (50 lb N per acre in this study) has very little effect on forage digestibility. In comparison to the teff studies, studies with BMR S×S in central New York showed NDF digestibility to be slightly higher than for teff. However, side by side comparisons are needed to determine if quality differences are significant.

Crude protein levels increased with similar trends for 2nd and 3rd cuttings as noted for 1st cutting (Table 7). Nitrogen application reduced forage NDF although for 3rd cutting more than 50 lb N per acre was needed for a significant reduction (Table 7). Nitrogen addition at 50 lb/acre or higher reduced ADF only for 1st cutting (Table 7), whereas digestibility of NDF was not affected by N addition at any of the sites for any of the cuts (results not shown). Thus, N addition increases CP levels without affecting forage digestibility as long as N application rates did not exceed 50 lb/acre.

Table 7. Effect of N application rate (lb N per acre per cut) on crude protein (CP), neutral detergent fiber (NDF), and acid detergent fiber (ADF) of teff grown in two 2- and two 3-cut harvest systems in New York^x.

 ^x Yield values within a row followed by the same letter are not significantly different at P = 0.05.

Nitrogen Uptake

Nitrogen uptake increased with N application due to increases in both yield and CP content (Table 8). The N use efficiency (NUE) at the 50 lb of N per acre application rate ranged from 12.1 to 24.4 lb DM per lb of N applied (Table 8). For 2nd and 3rd cutting, the NUE was considerably higher than the 14.5 lb DM per lb of N average NUE for BMR S×S grown on field without prior organic N history (10). The average apparent N recovery (ANR) ranged from 61% for 1st cutting to 83% for 3rd cutting (Table 8), 1.5 to 2 times higher than the average ANR for BMR S×S described in Ketterings et al. (10). Although side by side comparisons are needed, these results support the hypothesis that teff is a more N-efficient option than BMR S×S in N limiting situations (i.e., fields without a manure or sod history). Carryover of soil nitrate from applications to first cut was limited with N application rates of 50 lb/acre (results not shown) and the significant yield response to 50 lb/acre applied after 1st and 2nd cutting in the multiple-cut systems support a recommendation of split-application of 50 lb N per acre per cut for optimum yield and CP levels.

Table 8. Effect of N application rate (lb N per acre per cut) on N uptake, nitrogen use efficiency (NUE), and apparent nitrogen recovery (ANR) of teff grown in two 2- and two 3-cut harvest systems in New York^x.

 ^x Yield values within a row followed by the same letter are not significantly different at P = 0.05.

Practical Implications

High-quality teff forage can be grown in New York in 2- or 3-cut systems with DM yields ranging from 3.3 to 4.9 ton/acre, depending mostly on rainfall in the latter part of the summer. Yield and forage quality results of the eight trials support a recommendation of no more than 50 lb N per acre per cut (Fig. 2), whereas on soils with recent organic N additions (sod rotation or manure application), no additional N is recommended for 1st cutting. In the latter case N addition is not recommended as the probability of a yield response is small and lodging might occur.

Fig. 2. Teff responded to nitrogen application but an application of 50 lb N per acre per cut was enough for maximum yield at most locations.

Acknowledgments

This project was funded by grants from the Northeast Sustainable Agriculture Research and Extension Program (NESARE) and the Northern New York Agricultural Development Program (NNYADP). We thank James Van Leeuwen in Halsey, OR, for donating the teff seed and Ken Paddock, Wayne Berry, Kevin Dietzel, Chie Miyamoto, and Charles Hitchman for help with soil sampling and harvest of the field trials.

Literature Cited

1. Abaye, O., Teutsch, C., Claham, W., and Phillips, T. 2006. The influence of nitrogen rates, planting dates, and seeding rates on the herbage biomass yield and nutritive values of teff. ASA-CSSA-SSSA 2006 Intl. Meetings Poster, November 2006, Indianapolis, IN.

2. Braunwart, K., Putnam, D., and Fohner, G. 2001. Alternative annual forages, now and in the future. Proc. of the 31st California Alfalfa and Forage Symp. Univ. of California, Davis, CA.

3. Cherney, J. H., Cherney, D. J. R., and Casler, M. D. 2003. Low intensity harvest management of reed canarygrass. Agron. J. 95:627-634.

4. Davison, J., and McKnight, C. 2004. Tef demonstration plantings results for 2003. Coop. Ext. Publ. No. FS 04-51. Univ. of Nevada, Reno, NV.

5. Davison, J., and Peraldo, M. 2005. Tef demonstration planting results for 2004. Coop. Ext. Publ. No. FS 05-28. Univ. of Nevada, Reno, NV.

6. Davison, J. 2006. Tef demonstration planting results for 2005. Coop. Ext. Publ. No. FS 06-58. Univ. of Nevada, Reno, NV.

7. Greweling, T. 1976. Dry ashing. Agri. Exp. St. Research Bull. 6. Cornell Univ., Ithaca, NY.

8. Hall, M. 2007. Teff evaluation in Pennsylvania. Penn State Field Crops News 7:27.

9. Ketema, S. 1997. Tef Eragrostis tef (Zucc.) Trotter. Promoting the conservation and use of underutilized and neglected crops. Institute of Plant Genetics and Crop Plant Research, Gatersleben/International Plant Genetic Resources Institute, Rome, Italy.

10. Ketterings, Q. M., Cherney, J. H., Godwin, G., Kilcer, T. F., Barney, P., and Beer, S. 2007. Nitrogen management of brown midrib sorghum × sudangrass in the Northeastern USA. Agron. J. 99:1345-1351.

11. Ketterings, Q. M., Godwin, G., Cherney, J. H., and Kilcer, T. F. 2005. Potassium management for brown midrib sorghum × sudangrass in the Northeast. J. Agron. Crop Sci. 191:41-46.

12. Kilcer, T. F., Ketterings, Q. M., Cherney, J. H., Cerosaletti, P., and Barney, P.  2005. Optimum stand height for forage brown midrib sorghum × sudangrass in Northeastern USA. J. Agron. Crop Sci. 191:35-40.

13. Littell, R. C., Milliken, G. A., Stroup, W. W., and Russell, R. D.  1996. SAS System for Mixed Models. SAS Institute Inc., Cary, NC.

14. Morgan, M. F. 1941. Chemical soil diagnosis by the universal soil testing system. Agri. Exp. St. Bull. 450. Univ. of Connecticut, West Hartford, CT

15. NASS, 2009. National Agricultural Statistics, New York. USDA-NASS, Washington, DC.

16. Norberg, S., Roseberg, R. J., Charlton, B., and Shock, C.  2008. Teff: a new warm-season annual grass for Oregon. Coop. Ext. Publ. No. EM8970. Oregon State Univ., Corvallis, OR.

17. Roseberg, R. J., Norberg, S., Smith, J., Charlton, B., Rykbost, K., and Shock, C. 2005. Yield and quality of teff forage as a function of varying rates of applied irrigation and nitrogen. Research in the Klamath Basin 2005 Annual Report. Agric. Exp. Stn. Spec. Rep. 1069:119-136. Oregon State Univ., Corvallis, OR.

18. Storer, D. A. 1984. A simple high sample volume ashing procedure for determination of soil organic matter. Commun. Soil Sci. Plant Anal. 15:759-772.

19. Twidwell, E. K., Boe, A., and Casper, D. P. 2002. Teff: A new annual forage grass for South Dakota? Coop. Ext. Serv. Bull. Ex8071. South Dakota State Univ,. Brookings, SD.

20. 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