2 o ‘o * LIBRARY
Michigan State
U: liversity

This is to certify that the
thesis entitled

EVALUATION OF TEFF [Eragrostis tef (Zucc.) Trotter] AS AN
ANNUAL FORAGE CROP IN MICHIGAN

presented by
Stephanie Ilene Peck

has been accepted towards fulfillment
of the requirements for the

MASTER OF degree in Crop and Soil Science
SCIENCE

 

 

Major Professor’s Signature

q//(; /0

f I

 

MSU is an Affinnative Action/Equal Opportunity Employer

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 KzlProleccauProleIRCIDateDue.indd

EVALUATION OF TEFF [Eragrostis tef(Zucc.) Trotter] AS AN
ANNUAL F ORAGE CROP IN MICHIGAN

By

Stephanie Ilene Peck

A THESIS

Submitted to
Michigan State University
In partial fulfillment of the requirements
For the Degree of

MASTER OF SCIENCE
Crop and Soil Science

2010

ABSTRACT

EVALUATION OF TEFF [Eragrostis tef(Zucc.) Trotter] AS AN ANNUAL FORAGE
CROP IN MICHIGAN

By

Stephanie Ilene Peck

Forage producers in Michigan could benefit with a forage crop that can produce high
yields during the summer slump where environmental conditions are warm and dry. Teff
[Eragrostis tef(Zucc.) Trotter] is a warm season, annual grass from Ethiopia that fills this
need. Teff is an aggressive crop with good seedling vigor that continues to grow well
under high temperatures, is drought tolerant, and grows rapidly. This study compared
forage yield and quality (crude protein [CP], acid detergent fiber [ADF] and neutral
detergent fiber [NDF]) of teff to four warm season grasses; Brown midrib sudangrass
[Sorghum bicolor (L.) Moench], hybrid pearl millet [Pennisetum glaucum (L.)], and the
foxtail [Setaria italica (L.)] cultivars, Siberian millet and golden German millet under
multiple planting dates over two years (2008-9) and two locations (East Lansing [EL] and
Lake City [LC], MI). Teff forage yield and quality responses were also under five
nitrogen rates (0, 56, 84, 112, and 168 kg N ha'l). The millets and the sudangrass resulted
in greater dry matter yield than teff; however, teff exhibited greater forage quality (crude
protein and fiber). Post harvest re-growth of teff was faster (about 40 days to harvest)
than for millet and sudangrass. Nitrogen rate of 84 kg N ha'1 was optimal for higher
yields and desirable forage quality. The increase of nitrogen had a positive effect on teff
yields and forage quality. Teff appears to be a desirable alternative forage crop for forage
producers in need of an emergency forage crop that grows well in warm air temperatures

with adequate moisture.

ACKNOWLEDGMENTS

I would like to give a special thanks and great appreciation to Dr. Richard Leep
for this opportunity of research and study. I would like to give sincere gratidude to my
committee member Dr. Donald Penner for his encouragement and assisting me with my

manuscript. I would like to thank my committee member Dr. Roy Black for all his help.

I would like to give special thanks to Timothy Dietz for helping out with field work, his
encouragement, and evaluation and assistance in preparing my manuscript. I would like

to thank my parents for their encouragement and helping me through graduate school.

My deepest thank to my husband, Ryley Peck, for his encouragement, assistance, and

support through my time at Michigan State University.

iii

TABLE OF CONTENTS

 

 

 

 

 

LIST OF TABLES ........................................................................................................... vi
LIST OF FIGURES- - -- - - - -- - -- - - - - --.vii
INTRODUCTION ..... - - - - -- - - - .......... . -- -- 1
CHAPTER l-INTRODUCTION -- - - ......... - - 6
Materials and Methods
Study sites and experimental design ............................................................................... 8
Planting Methods ................................................................................................ I ............. 9
Forage dry matter and yields ......................................................................................... 11
Forage Quality ............................................................................................................... 12
Statistical analysis ......................................................................................................... 12
Weather Conditions ....................................................................................................... 13
Results and Discussion---- - -- - -- - - - - - -- l6
Forage Yields ................................................................................................................ 16
East Lansing 2008 ..................................................................................................... 18
East Lansing 2009 ..................................................................................................... 18
Lake City 2008 .......................................................................................................... 21
Lake City 2009 .......................................................................................................... 21
Acid Detergent Fiber ..................................................................................................... 23
East Lansing 2008 ..................................................................................................... 24
East Lansing 2009 ..................................................................................................... 25
Lake City 2008 ................................................ ‘. ......................................................... 27
Lake City 2009 .......................................................................................................... 27
Neutral Detergent Fiber ................................................................................................. 27
East Lansing 2008 ..................................................................................................... 29
East Lansing 2009 ..................................................................................................... 30
Lake City 2008 .......................................................................................................... 33
Lake City 2009 .......................................................................................................... 33
Crude Protein ................................................................................................................. 34
East Lansing 2008 ..................................................................................................... 36

iv

East Lansing 2009 ..................................................................................................... 36

 

Lake City 2008 .......................................................................................................... 38
Lake City 2009 .......................................................................................................... 38
Teff Forage Quality and Yield Summary ...................................................................... 39
CONCLUSION - -- - - - -- ------------- - - - ------ 4]
LITERATURE CITED - - - - - - - 42

 

 

Chapter 2- INTRODUCTION -- - - - - - - - - - 45

 

 

 

Materials and Methods - - - - - - -- -- - - 46
Study sites and experimental design ............................................................................. 46
Teff Nitrogen Rate Response ........................................................................................ 47
Dry Matter Yields .......................................................................................................... 48
Forage Quality ............................................................................................................... 48
Statistical Analysis ........................................................................................................ 49
Weather Conditions ....................................................................................................... 49

Results and Discussion .................................................................................................... 52
Teff Yields ..................................................................................................................... 52
Acid Detergent Fiber ..................................................................................................... 54
Neutral Detergent Fiber ....................................... . ......................................................... 55

East Lansing, MI ........................................................................................................ 55
Lake City, MI ............................................................................................................ 56
Crude Protein ................................................................................................................. 57
East Lansing, MI ........................................................................................................ 57
Lake City, MI ............................................................................................................ 59

CONCLUSION ---------------- - --------------------------- ----- - 60

LITERATURE CITED ------- - - - - - -- -- - - 61

APPENDICES.--“ -- - - -- -- - - - - . - - -- - 62

 

LIST OF TABLES

Table 1. Variety of the species used in this experiment along with the companies that
provide the seed ...................................................................................... 9

Table 2. Planting and harvest dates for 5 species of WSG at East Lansing, MI in 2008 and
2009. Seeding rates are in kg ha'l ................................................................ 10

Table 3. Planting and harvest dates for 5 species of WSG at Lake City, MI in 2008 and
2009. Seeding rates are in kg ha.1 ................................................................ 11

Table 4. East Lansing, MI planting and harvest dates for teff with five different fertility
rates in 2008 and 2009 ............................................................................... 47

Table 5. Lake City, Michigan planting and harvest dates for teff with five fertility rates in
2008 and 2009 ........................................................................................... 48

Table 6. Near infrared reflectance spectroscopy calibration and validation statistics for
crude protein, acid detergent fiber, and neutral detergent fiber for stands harvested over
two years (2008-9) in Lake City and East Lansing, MI ........................................ 62

vi

LIST OF FIGURES

Figure 1. Monthly average precipitation (cm) received in East Lansing, M1 for the
growing season (April-October) in 2008 and 2009 ........................................... 13

Figure2. Monthly averages for temperatures (°C) received in East Lansing, M1 for the
growing season (April-October) in 2008 and 2009 ........................................... 14

Figure 3. Monthly averages for precipitation (cm) received in Lake City, MI for the
growing season (April-October) in 2008 and 2009 ........................................... 15

Figure4. Monthly averages for temperatures (°C) received in Lake City, MI for the
growing season (April-October) in 2008 and 2009 ........................................... 15

Figure 5. Accumulative forage yield from BMRS, GGM, HPM, SM, and teff located at
East Lansing, MI in 2008, per species per planting date in dry matter Mg ha'l. LSD-=1 .07
(n=270) Standard Error=0.8292 ................................................................... 17

Figure 6. Accumulative forage yield from BMRS, GGM, HPM, SM, and teff located at
East Lansing, MI in 2009, per species per planting date in dry matter Mg ha". LSD=0.95
(n=270) Standard error=0.7672 ................................................................... 17

Figure 7. Accumulative forage yield from BMRS, GGM, HPM, SM, and teff located at

Lake City, MI in 2008, per species per planting date in dry matter Mg ha]. LSD=0.82
(n=270) Standard error=0.63 86 ................................................................... 20

Figure 8. Accumulative forage yield from BMRS, GGM, HPM, SM, and teff located at

Lake City, MI in 2009, per species per planting date in dry matter Mg ha]. LSD=1.14
(n=270) Standard error=1 .09 ....................................................................... 20

Figure 9. Total weighted acid detergent fiber (g kg'l) from BMRS, GGM, HPM, SM, and

teff located at East Lansing in 2008. Teff had the lowest ADF value 354 g kg".
LSD=0.87 (n=270) Standard error=3.386 ....................................................... 23

Figure 10. Total weighted acid detergent fiber (g kg'l) from BMRS, GGM, HPM, SM,
and teff located at East Lansing, Mi in 2009. Teff showed significantly lower ADF than
the other 4 species at 353 g kg]. LSD=0.87 (n=270) Standard error=3.1432 ............. 24

Figure 11. Total weighted acid detergent fiber (g kg'l) from BMRS, GGM, HPM, SM,
and teff located at Lake City, Mi in 2008. Teff showed significantly lower ADF than the
other 4 species at 320 g kg]. LSD=1.52 (n=270) Standard error=5.656 .................... 26

vii

Figure 12. Total weighted acid detergent fiber (g kg'l) from BMRS, GGM, HPM, SM,
and teff located at Lake City, Mi in 2009. HPM showed significantly lower ADF than the
other 4 species at 346 g kg'l. LSD=1.01 (n=270) Standard error=3.64l ................. 26

Figure 13. Total weighted neutral detergent fiber (g kg']) from BMRS, GGM, HPM, SM,
and teff located at East Lansing 2008 per species per planting date. LSD=2 (n=270)
Standard error=7.255. ................................................................................. 28

Figure 14. Total weighted neutral detergent fiber (g kg'l) from BMRS, GGM, HPM, SM,
and teff located at East Lansing 2009 (n=270) per species per planting date. LSD=2.I
(n=270) Standard Error=7.287. ................................................................... 29

Figure 15. Total weighted neutral detergent fiber (g kg!) from BMRS, GGM, HPM, SM,
and teff located at Lake City 2008 (n=270) per species per planting date. LSD=3.1
(n=270) Standard Error=11.449. .................................................................. 32

Figure 16. Total weighted neutral detergent fiber (g kg'l) from BMRS, GGM, HPM, SM,

and teff located at Lake City 2009 per species per planting date. LSD=2.8 (n=270)
Standard Error=10.063. ............................................................................ 32

Figure 17. Total weighted crude protein (g kg'l) from BMRS, GGM, HPM, SM, and teff
located at East Lansing, MI in 2008 per species per planting date (n=270). LSD=1.04
(n=270) Standard error=7.38. ..................................................................... 35

Figure 18. Total weighted crude protein (g kg!) from BMRS, GGM, HPM, SM, and teff
located in East Lansing, MI in 2009 per species per planting date. LSD=1.1 (n=270)
Standard Error=7.73. ............................................................................... 35

Figure 19. Total weighted crude protein (g kg") from BMRS, GGM, HPM, SM, and teff
located at Lake City, MI in 2008 per species per planting date (n=270). LSD=1.82
(n=270) Standard EI’I'OFI 1.44. ................................................................... 37

Figure 20. Total weighted crude protein (g kg'l) from BMRS, GGM, HPM, SM, and teff
located in Lake City, MI in 2009 per species per planting date. LSD=O.92 (n=270).
Standard Error=5 .705. . ............................................................................ 38

Figure 21. The dry matter (Mg ha") and forage quality as NDF, ADF, and CP (g kg!) of
teff located at East Lansing in 2008 and 2009 with four planting dates. Error bars for any
given line are based on LSD (0.05). Each line was analyzed separately ..................... 39

viii

Figure 22. The dry matter (Mg ha") and forage quality as NDF, ADF, and CP (g kg") for
teff located at Lake City in 2008 and 2009 with three planting dates. Error bars for any
given line are based on LSD (0.05). Each line was analyzed separately ..................... 40

Figure 23. Monthly average precipitation (cm) received in East Lansing, M1 for the
growing season (April-October) in 2008 and 2009 ............................................ 50

Figure 24. Monthly means for temperatures (°C) received in East Lansing, M1 for the
growing season (April-October) in 2008 and 2009 ............................................ 50

Figure 25. Monthly average precipitation (cm) received in Lake City, MI for the growing
season (April-October) in 2008 and 2009 ........................................................ 51

Figure 26. Monthly average temperatures (°C) received in Lake City, M1 for the growing
season (April-October) in 2008 and 2009 ........................................................ 52

Figure 27. Accumulative forage yield per nitrogen treatment across two years (2008-9)
located at two locations (East Lansing and Lake City, MI) in dry matter Mg ha" (N =80).
LSD=8.5 Standard error=6.050 ................................................................... 53

Figure 28. Total weighted acid detergent fiber (g kg") with five nitrogen treatments
across two years (2008-9) located at East Lansing and Lake City, MI (n=80). LSD=5.9
Standard error=2.104 ............................................................................... 55

Figure 29. Total weighted acid detergent fiber (g kg") with five nitrogen treatments
across two years (2008-9) located at East Lansing, MI (n=80). LSD=11.3 Standard _
error=4.346 .......................................................................................... 56

Figure 30. Total weighted acid detergent fiber (g kg") with five nitrogen treatments
across two years (2008-9) located at Lake City, MI (n=80). LSD=12.6 Standard
Error=4.933 .......................................................................................... 57

Figure 31. Total weighted crude protein (g kg") with five nitrogen treatments across two
years (2008-9) located at East Lansing, MI (n=80). LSD=11.4 Standard Error=3.969... 58

Figure 32. Total weighted crude protein (g kg") with five nitrogen treatments across two
years (2008-9) located at Lake City, MI (n=80). The 56, 84, 112, and 169 kg ha'l N

treatments were significantly higher thanthe 0 kg ha.1 N. LSD=9.5 Standard
Error=4.28l .......................................................................................... 59

Figure 33. Teff forage quality (CP, ADF, NDF) from East Lansing in 2009. Samples
were collected every two weeks from planting date of June 4. Collected samples when

teff was ~10 cm tall until the teff lodged ........................................................ 63

INTRODUCTION

Michigan experiences a wide range of temperatures throughout the state. In
northern Michigan, temperatures tend to be cooler, whereas the mid to southern parts of
Michigan tend to be warmer. Michigan producers have an advantage of being able to
grow both warm-season (C4) and cool-season (C3) grasses. Warm season grasses (WSG)
primarily grow during the warm summer months (Hall, 1994) and perform well in
Michigan where there tends to be drought periods and warm air temperatures. Cool-
season grasses (CSG) do well in the spring and fall when there is adequate moisture and
cooler air temperatures. Several WSGS do well in the summer and are suitable for
pastures and hay production when CSGs become less productive.

Michigan often experiences a summer slump in both pasture and hay crop growth
in July and August due to lack of moisture and high temperatures. It has often been stated
in extension meetings that Michigan has the least amount of rainfall of all states east of
the Mississippi River. Teff, a warm-season grass, grows well when soil temperatures are
15°C or warmer (Norberg, 2008). Cool season grasses and legume crops grow optimally
at air temperatures of 21°C, while warm season grasses have optimum temperature of
32°C. The relatively high summer temperatures and precipitation deficit in Michigan
summer creates a summer forage deficit problem for some livestock producers who grow
cool season grasses and legumes. Producers are looking for a crop that can be planted and
produce forage with rapid and vigorous growth from which they can get hay or graze
with livestock during the summer months. A potential crop which is a fast growing and

drought tolerant is teff [Eragrostis tef(Zucc.) Trotter].

Teff is a warm-season, annual forage grass known for high dry matter yields,
vigorous growth, and high forage quality (Hunter, 2007). It can be used as either a
pasture grass or a hay crop.

Other studies have shown additional uses for teff such as a green manure (Hunter,
2008) to till into the soil in the late fall before frost. Green manure improves soil
structure. When green manure is incorporated by tillage it improves soil structure and
increases oxygen availability to soil organisms which aid in the breakdown of nutrients.
This green manure can provide readily available nutrients for crops the following year,
when tilled into the soil 2 weeks before planting of a crop (Card, 2006).

Teff is a grass native to Ethiopia where it is grown as a grain. In Ethiopia, teff is
an important grain crop with one million hectares of land cultivated with teff each year
(Ketema, 1996). Teff grain is used in the United States to produce gluten free flour for
people who have gluten allergies. Teff is also grown for grain in the United States.
particularly in Idaho and South Dakota, to produce gluten free flour. In several other
states, New York, Virginia, Kentucky, Wyoming, and Minnesota, teff is being researched
and successfully used as a forage grass.

Teff is adapted to environments that range from waterlogged conditions, to
environments that are dry and drought stressed (Hunter et al., 2008). A study by Seyfu
Ketema in 1983 found that teff is a long day plant and needs a certain amount of daylight
to flower (Ketema, 1996). Michigan summers have day lengths suitable for teff growth.
Teff has a fibrous root system. After first cutting, teff roots will establish sufficiently to

allow grazing (after 45 days). The leaves are smooth and erect. The inflorescence is a

panicle that starts out compact and then becomes loose. The size of the teff seed is small
(1.23 million seeds per pound) (Hunter et al., 2007).

In Michigan, teff is a newly introduced annual forage grass by several seed
marketers. There are currently other WSGs that are being sold by marketers as
emergency forage crops in Michigan. Brown midrib sudangrass (BMRS) is a crop which
was introduced by several seed companies as forage. BMRS is identified by the reddish-
brown midrib on the leaf margins. BMRS results from a genetic mutation in which the
total lignin content is reduced in different parts of the plant such as the leaves (Miller,
2003). Low lignin increases digestibility to the livestock, whereas, sudangrass has a
higher lignin content which has less digestible content (Miller, 2003). The BMRS has a
greater fiber digestibility that contributes to increased milk production (Kilcer et al. 2005.
Like teff, BMRS is a warm-season annual grass that is known for growing well in high
temperatures and drought conditions (Kilcer et al., 2005).

Pearl millet is another newly introduced forage crop in Michigan. Originated in
Africa and India, pearl millet is used as a grain crop. Pearl millet is an annual, warm
season grass that can adapt to higher temperatures and low rainfall (Maman 1999). The
pearl millet is known to have vigorous growth when the right conditions are present
(Maiti 1981) such as temperature and moisture. Siberian millet is an annual, warm season
grass known to grow rapidly in warm weather and low moisture, yet has a disadvantage
of slow to no re-growth after one cutting or grazing event (Koch, 1992). Other millets
that are less commonly grown for emergency summer annuals include Siberian and
golden German millet. Golden German millet is known to produce high yields with

adequate moisture and warm air temperatures (May, 2007). Teff, along with Brown mid-

rib sudangrass (Sorghum bicolor L. Moench), hybrid pearl millet (Pennisetum glaucu),
and the foxtails of golden German millet (Echinochloafrumentaceae) and Siberian millet
(Setaria italica L.) will be evaluated in Michigan. There has been no research done in
Michigan that compares teff to other WSGs. In addition, there is no optimal planting date
for teff compared to other warm season annual grasses. There has been no research done

in Michigan to evaluate the optimal nitrogen rate of teff.

ABSTRACT

EVALUATION OF TEFF [Eragrostis tef(Zucc.) Trotter] AS AN ANNUAL FORAGE
CROP IN MICHIGAN

By
Stephanie Ilene Peck

Producers in the North Central region could benefit from a forage species which will
maintain high yields though the summer. Teff [Eragrostis tef(Zucc.) Trotter] is a warm
season, annual grass from Ethiopia that has potential in Michigan during the warm, dry
summers (summer slump). Teff is an aggressive crop with good seedling vigor that
continues to grow well under high temperatures but has not yet been tested in Michigan
for forage production. Teff possesses several desirable characteristics including drought
tolerance and rapid growth. This study compared teff yields and forage quality to other
newly introduced warm season grasses; Brown midrib sudangrass [Sorghum bicolor (L.)
Moench], hybrid pearl millet [Pennisetum glaucum (L.)], and the foxtail [Setaria italica
(L.)] cultivars, Siberian millet and golden German millet. Trials were established as a
split-plot design with planting dates, in two week intervals, as the whole plot factor
beginning late-May and ending mid-July with sub plots being species. The millets and the
sudangrass resulted in greater dry matter yield than teff; however, teff exhibited greater
forage quality (crude protein and fiber). Post harvest re-growth of teff was faster (about
40 days to harvest) than for millet and sudangrass. Teff appears to be a desirable
alternative forage crop for producers in need of an emergency forage crop that grows well

in warm air temperatures with adequate moisture.

Chapter 1- INTRODUCTION

Michigan experiences a summer slump in forage growth of cool season grasses
from dryer and warmer weather, which creates a forage deficit for producers. These
producers have very few crop choices if they desire a high quality, annual, warm-season
grass (WSG) to plant as emergency or supplementary forage to provide to their livestock.
Corn (Zea mays L.) is a widely grown WSG but has limitation of forage and grazing
qualities. Several cool-season (CSG) perennial grasses including Timothy (Phleum
pretense) have traditionally been grown in pastures for summer forage. These cool-
season perennial grasses have limitations when the air temperatures are hot and soil
moisture declines. Millets (Panicum spp.) and sudangrass (Sorghum bicolor) grow well
under these conditions but also become fibrous early, re-growth also can be slow. A hi gh-

quality, annual WSG would increase options for producers and increase profitability.

Teff [Eragrostis tef(Zucc.) Trotter], an annual WSG, has the potential to be

grown in mid season as a replacement to CSG that are affected by hot and dry summer
conditions. Teff is known to have comparable forage quality to timothy a cool season
grass grown in Michigan for hay and pasture (Miller and Stroup, 2008). Teff needs a soil

temperature of 15.5 degrees C and adequate soil moisture for good growth (Norberg et

al., 2008). Crude protein at boot stage in teff ranged from 80-140 g kg" compared to 90-
140 g kg'1 in timothy while neutral detergent fiber (NDF) was 530-590 g kg" and 530-

650 g kg", for teff and timothy, respectively (Miller and Stroup, 2008). Teft’s has an

advantage over for timothy, in that timothy tends to have low production in the summer

months when teff thrives (Miller and Stroup, 2008). This is due to the shallow root
system of timothy resulting in low production during the hot, dry periods of summer.
Producers desire a forage species that has vigorous growth and produces high-quality hay
in a short time to supplement their normal grazing program with cool-season perennial
grasses. A species with these characteristics may work well as an emergency forage crop

that can be seeded in late spring or early summer.

Golden German millet (GGM) requires a warm climate and is known to produce
forage with little rainfall and fertility (Moser et al., 2004). Nitrogen and phosphorus are
the most limiting nutrients in GGM (National Research Council, 1996). In the USA,
GGM is used as high quality forage (Moser et al., 2004) which has the advantage of rapid

growth and drought resistance. Crude protein of GGM is comparable to teff, and ranges

from 100 to 140 g kg" when out at boot stage (Government of Alberta, 2007).

Siberian millet (SM) grows rapidly in warm, dry weather producing ample forage
(Koch, 1992). At 110 g kg", the crude protein (CP) content is not as high as that of teff.

A disadvantage of SM is the limited grth capacity, which limits SM to one cutting

(Koch, 1992).

Hybrid pearl millet (HPM) favors high temperatures and adequate moisture and

can tolerate poor soils (Sedivec and Schatz, 1991). HPM is known for producing high CP
at boot stage 150g kg", comparable to teff (Sedivec and Schatz, 1991). HPM can be used

for hay but is difficult to dry. When out early enough, it has the potential to re-grow.

Brown mid—rib sudangrass (BMRS) is adapted to higher temperatures, drought,

and lower soil pH (Kilcer et al., 2004). BMRS is known for reduced lignin content

7

throughout the plant and is named for the reddish-brown color on the midrib and stem.
Crude protein of BMRS at boot stage is much lower than in teff, at 92 g kg" (Miller,

2003). These warm season grasses have traits that result in potential forage for use in

Michigan.

The objectives of this study were as follows:

1) Determine yields of teff compared to German millet, Siberian millet, hybrid pearl
millet, and brown mid-rib sudangrass.

2) Compare teff forage quality with German millet, Siberian millet, hybrid pearl
millet, and brown mid-rib sudangrass.

3) Determine the optimal planting date for teff compared to German millet, Siberian
millet, and hybrid pearl millet in Michigan.

Materials and Methods

Study sites and experimental design

Teff research trials were established in 2008 and 2009 at two locations in
Michigan using a split-plot design with four replications with a plot size of 0.9 x 3.7 m.
The planting date was the whole-plot and species was the subplot which was arranged in
a randomized complete block design. Four planting dates were used at the Michigan State
University Crop and Soil Science Farm in East Lansing (EL), Ingham County, MI
(Latitude: 420 72 ’N, Longitude: 84°49 ’W) and three planting dates were used at the
Michigan State University Lake City Experiment Station in Lake City (LC), Missaukee
County, MI (Latitude: 44°35’ N, Longitude: 85°18’W). The planned dates for
establishment of each planting date was l-Jun, 15-Jun, l-Jul, and 15-Jul for EL and 6-

Jun, 16-Jun, and 6-Ju1 for LC. The actual planting date for each location and year are

presented in table 2 and 3. The soils in East Lansing consist of a Colwood-Brookston
loam (fine-loamy, mixed, mesic Typic Haplaquolls, and fine-loamy, mixed, mesic Typic
Argiaqoulls) with a pH of 6.5. The soil at Lake City is a Nester Sandy loam (fine, mixed,

semiactive, frigid Oxyauic Glossudalfs) with a pH of 6.9.

Cultivars

Table 1. Variety of the species used in this experiment along with the companies that
provide the seed.

Teff Land O’ Lakes
Siberian Millet Windfield Solutions Common

brid-Pearl Millet Land O’ Lakes 23
German Millet Brier ' Golden
Brown mid-rib Cisco Seed Pro Max

 

Planting Methods

The seedbed was prepared by conventional tillage using a moldboard plow and
disking twice. The ground was then prepared using a mat drag and/or chain drag to
smooth and compact the seedbed. Teff was planted 0.635 cm deep, due to the small seed
size and millets and sudangrass were planted 1.27 cm deep using an Almaco 1.52-m wide
drill (double-disk openers with press wheels) according to the seeding rate in table 1.

Granular ammonium sulfate was applied by hand broadcasting immediately after seeding
at 56 kg N ha'l in 2008 and increased to 112 kg N ha" in 2009 based upon the results of a

teff fertility trial conducted in 2008 (Peck, unpublished data, 2009).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

303 82-8 3-2 mom-8 3-2
303 meal; 33 30-2 828 3-8 8 200
weal-N 3-2-. 3-2 30-2 30,-: 3-2
83-: 3-8 34. 30-2 mom-8 3-8 33.8
303 22-8 3-2 mom-8 3-2
303 23.2 3-0 30-2 828 3-8 8 3::
228 3-8 3-2 30-2 w:<-: 3-2
w:<-8 3-8 3-: 30-2 .328 3-8 32.8
303 “-3-8 3-2 mom-8 3-2
303 w:<-2 3-0 30-2 was--8 3-8
23.-8 3-8 3-2 30-2 weer: 3-2 am $30
w:<-8 3-8 53 30-2 mam-8 3-8 e2.8
8.28 3-2 was--8 3-2
3..-: 3-0 mg: 2 3-8 8 2.3.
3-2-. :32 3E 3-2
3-8 3-4 3-: 32.8
303 “03-8 3-2 38-8 3.:
30A 30:: 3-3 30-2 mam-8 3-8 a 23.:
30A m:<-8 3-8 3-2 30-2 gem-m 2.: 3-2
:3? we} 3-8 3-3 30-2 828 3-: 22.8
2:0 ~30 2:0 2232 2:0 2:0 2:0 83: 9.3 000 3325
S8 28 3:2 2538

._-a._ w: E 0.3 83-. 955% $ch .25 ”SN 5 :2 $33.55 unam— ua Om?» .3 8.00% m be v.32. “mo-Ca.— .Ea wit—«E .N 030,—-

 

10

Table 3. Planting and harvest dates for 5 species of WSG at Lake City, MI in 2008 and
2009. Seeding rates are in kg ha".

Lake ('in

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Seeding Rate 2008 2009
Species (Kg ha'l) Planting Cut 1 Cut 2 Planting Cut 1 Cut 2

12-Jun 15-Aug 9-Oct 4—Jun 12-Aug 12-Oct

Teff 9 24-Jun 26-Aug 1 8-Sep 16-Jun 12-Aug 12-Oct
9-Jul 18-Sep 6-Jul 3 l-Aug
12-Jun 5-Aug 4-Jun 12-Aug
SM 22 24-Jun lS-Aug 16-Jun 12-Aug
9-Jul 26-Aug 6-Jul 3 l-Aug

12-Jun 15-Aug 9-Oct 4-Jun 12-Aug 12-Oct

BMRS 34 24-Jun 26-Aug 9-Oct l6-Jun l2-Aug l2-Oct
9-Jul 18-Sep 6-Jul 3 1 -Aug

12-Jun 15-Aug 9-Oct 4-Jun 12-Aug 12-Oct

HPM 28 24-Jun 26-Aug 9-Oct l6-Jun 12-Aug 12-Oct
9—Jul 1 8-Sep 6-Jul 3 1 -Aug

l2-Jun 15-Aug 9-Oct 4-Jun 12-Aug 12-Oct

GGM 22 24-Jun 26-Aug 9-Oct 16-Jun 12-Aug 12-Oct
9-Jul 18-Sep 6-Jul 31-Aug

 

Forage dry matter and yields

 

The experiment was harvested for forage using a rotary flail harvester (Carter
Manufacturing Co. Inc., Brookston, IN) set to harvest at a height of 10.2 cm from plots
0.9 x 3.7 m. A ~200 g subsample was obtained at the time of harvest from each plot and
dried for 96 h at 60°C, and weighed again for dry matter determination and forage quality
analysis. Moisture content was determined by: Dry weight (g)/ Wet weight (g) * 100= %

dry matter content.

11

Forage Quality

Samples were ground with a laboratory hammer mill (Christy Turner, LTD.,
Chelmsford, UK.) to pass through a 1 mm screen. Then the sample was scanned using a
near-infrared spectrophotometer (NIRS) Foss model 6500 (FOSS NIRSystems, Inc.,
Eden Prairie, MN). The wavelengths (between 700-2500 nm) reflected off the samples
were recorded and used to estimate crude protein (CP), the neutral detergent fiber (N DF)
and the acid detergent fiber (ADF) from equations created by the NIRS Forage and Feed
Testing Consortium (Madison, WI). Subsets of samples were chemically analyzed for
CP, NDF, ADF, and dry matter content to validate NIRS values. The Goering-VanSoest
method (1970) was used to determine NDF and ADF, the standard protocol was modified
with the addition of 1 ml of alpha amylase. Total nitrogen (N) was determined by the
Hach modified Kjeldahl Nitrogen method (Hach, et al., 1985) and crude protein was
estimated by multiplying total N by 6.25. Dry matter content was determined 3 by drying

0.50 g of sample in a crucible at 90°C for 12 h.

Statistical analysis

PROC UNIVARIATE in SAS (SAS Institute, v9.1) was used to check normality,
residuals, and unequal variances. Analysis of variance (ANOVA) was performed on
yield, forage quality, CP content, NDF, and ADF using PROC MIXED. Interactions were
considered significant at P>0.1 and main effects considered significant at P>0.05.
Contrasts of the means were used to determine mean separation with a confidence
interval of 95%. The replication by planting date and the replication were both considered

random effects. The cultivar was considered a fixed effect.

12

Weather Conditions

East Lansing, MI

The total amount of precipitation East Lansing received between April and
October was higher than the 30 year average in both years. East Lansing received an
average of 15.1 cm more precipitation than the 30 yr average of 50.6 cm, for the 2008
and 2009 growing season. The total precipitation in 2008 and 2009 growing season of
April through October was 57.3 cm and 74.1 cm, respectively. In 2008, there was 21.4
cm in September, higher than the 30 yr average. In 2009, August had high rain amounts

Of 16.8 cm (Figure 1).

Precipitation in East Lansing

 

25 ~ - W _ g - W _-_- W_LL ‘
20 4
15 J- . T H" " ‘ """“ " "'--~-—~-—-e-~---7—...“__,_;.€;__“___ ___ _.-.,u..--._

10 I 2009

a: 30 year avg '

      

Apr May June July Aug

Figure 1. Monthly average precipitation (cm) received in East Lansing, M1 for the growing
season (April-October) in 2008 and 2009.

Temperatures during the growing season (April to October) in both years
averaged 0.23°C warmer than the 30 year average. Temperatures were warmer than the

normal for 2008 and slightly cooler in 2009. Seasonal averages for 2008 and 2009 were

13

164°C and 15.0°C, respectively. The 30 year average temperature for East Lansing was

15.5°C (Figure 2).

1 Temperature in East Lansing

I 25 - , , -, , 7w - W _--- ~ 73’ 7732222723 ~77- I
i I
| 20 ' 1
I u
I g 15 . I2008
E? 10 ‘ 2009
lNormal ‘
5 -
0

 

I
I
$ APR MAY JUN JUL AUG SEP OCT

Figure 2. Monthly means for temperatures (°C) received in East Lansing, MI for the
growing season (April-October) in 2008 and 2009.

Lake City, MI

The total amount of precipitation during the growing season in 2008 and 2009
was higher than the 30 yr average of 52.9 cm. In 2008, June had more than normal
precipitation resulting in a higher than normal average of61.4 cm. The average
precipitation in 2009 (53.5 cm) was just slightly higher than the 30 year average. In 2009,
October had 12.3 cm, higher than the 30 year average. The total precipitation for both

years combined was above the normal by 4.6 cm (Figure 3).

14

Precipitation in Lake City 1

20 n— 3 i - - 3» ——'———--— -———'—v —

15
l E 10 - W ‘ .2008 l
I u .
E - 2009

0

Apr May June July Aug Sept Oct

 

Figure 3. Monthly average precipitation (cm) received in Lake City, MI for the growing
season (April-October) in 2008 and 2009.

The average temperature in Lake City was close to the 30 year average of
133°C. In 2008, the average temperature (137°C) during the growing season was
slightly higher than the 30 yr normal. The 2009 growing season temperatures had a lower
average than the 30 yr normal, with an average of 126°C. Lake City averaged -0.1°C less

in both years combined during the growing season (Figure 4).

Lake City Temperature
25

20'

I I I
' 15 '* * ' _ ' * * .2008 E
I 10 g ' 2009 g
l 5 ' I I I I lNormal
i 0 I '

APR MAY JUN JUL AUG SEP OCT

Degress C

 

Figure 4. Monthly average temperatures (°C) received in Lake City, M1 for the growing
season (April-October) in 2008 and 2009.

Results and Discussion
Forage Yields

The harvest-stage target for both locations and both years was boot stage with no
more than 10% of the plants headed. In 2008, LC received 15 cm of precipitation in a few
hours in June; washing away some of the seed in the first planting date (Figure 3). These
stands were thin and seeds between plots mixed. After the first cut, the stands improved,
grew thicker, and outgrew weeds. In Lake City in 2009, temperatures were below the
normal during the growing season (Figure 4) and plots did not germinate quickly nor
grow well until late July which delayed harvest. When temperatures increased, the plant
growth accelerated and plants matured quickly. The plants from the first planting date in
LC in 2009 were more mature than desired at harvest, with about 40% of the plants
headed. In LC, the first frost in 2009 was I-Oct, and the last harvest took place 12-Oct,
therefore plant growth had ceased. In 2008 and 2009, LC had lower temperatures than

EL; therefore LC only has three planting dates whereas EL had four.

There was a significant interaction with the main effect, species and planting date.
A significant interaction between planting date, species, and year and between planting
date and species, therefore data will be presented within a year and planting date. Since
there are four planting dates in EL versus the three planting dates in LC, data will be

presented by location.

16

Yield In East Lansing In 2008

 

 

 

 

 

l 16 I
l “I? 14 I 7"? I
I .2 12 E ,
g” 10 l - 7
l a 8 -f -
l 4 , -
, E‘
l a 2 I "
E O +4 --
I e 2 a .- 3...: a
I g 8 I F' ‘ g 8 I
E 29-May - 16-Jun 30—Jun 14—Jul

L-Lu ___. - .7 ,L L L, ,7 - ,, LL LL,- ,,--,-- L, , L, _ L ,. ,

Figure 5. Accumulative forage yield from BMRS, GGM, HPM, SM, and teff located at East
Lansing, MI in 2008, per species per planting date in dry matter Mg ha". LSD (0.05) =1.07
(n=270) Standard Error=0. 8292.

Yield' In East Lansing' In 2009

Dry Matter (Mg ha'1)
m
o

Teff H

 

I I
I L 4-Jun 16—Jun ? 6 Jul ‘ 15- Jul

Figure 6. Accumulative forage yield from BMRS, GGM, HPM, SM, and1 teff located at East
Lansing, M] In 2009, per species per planting date' In dry matter Mg ha'1 . LSD (04,5) =0.95
(n=270) Standard error=0.7672

17

East Lansing 2008

In 2008, EL had three cuts of teff that were made from the first and second
planting date, two cuts from the third planting date and one cut from the last planting
date. The SM was out once per planting date due to lack of re-growth. The BMRS, HPM,
and GGM were out three times from the first planting date, two cuts from the second and
third planting date, and one cut from the last planting date (Table 1). The BMRS resulted
in the highest yield at in all four planting dates. This supports the findings of Twidwell et
al., (2002) who compared sudangrass to teff and millets is South Dakota. The highest

yield (12.7 Mg ha") of BMRS resulted from the first plating date. Teff was the lowest

yielding species in three of the four planting dates. Teff resulted in the second highest

yield on the first planting date with 7.2 Mg ha". Teff, GGM, and HPM yield from the

first planting date were not significantly different, with yields at 7.2, 7.7, and 6 Mg ha",

respectively. Teff was not significantly different from SM on the second, third and fourth

planting dates (Figure 5).

East Lansing 2009

In 2009, three cuts of teff were taken from the first and second planting date and
two cuts from the third and fourth planting date. One cut per planting date was taken
from the SM due to lack of re-growth. Two cuts were taken with the BMRS, HPM, and
GGM from the first, second and third planting date (Table 1). Teff resulted in higher
yields on the first and third planting date and was the second highest yielding species on

the second and fourth planting dates. The highest yield of teff was from the first planting

date with 9.3 Mg ha". The SM had the lowest yield in all four planting dates and the

18

lowest yield from the fourth planting date (3.1 Mg ha"). The BMRS resulted in the

highest yielding on the second planting date with 10.8 Mg ha'1 (Figure 6). The BMRS

from the fourth planting date had poor emergence, therefore there no data was Obtained

for that treatment.

The precipitation in 2008 was above the 30-yr average (36.5 cm) during the
growing season of J un-Oct with 48.4 cm. The temperatures were slightly above the 30-yr
(173°C) normal during the growing season of Jun-Oct with an average of 183°C. In
2009, the precipitation was above the 30—yr average (36.5 cm) during the growing season
of J un-Oct receiving a total of 46.6 cm. The temperature was slightly below the 30-yr
average (17.3°C) during the growing season of Jun—Oct with an average of 168°C. July
was cooler than normal with a temperature mean of 192°C when the average for July is

214°C and July and August were slightly below the 30-yr average.

BMRS, HPM, and GGM resulted in higher yields than teff from all planting dates
while teff was higher in yield tha SM was lower or similar to teff in 2008. Teff resulted in
higher yields compared to all species and planting dates in 2009. The BMRS resulted in
higher yields in 2008 when air temperatures were above the 30-yr normal and rainfall
was above the 30-yr average. The teff resulted in higher yields in 2009 when

temperatures were below the 30-yr normal and rainfall was above the 30-yr average. In
2008, the species received 56 kg N ha", less than the 112 kg N ha" split application that

was received in 2009. Both teff and BMRS responded positively to the increased N with

higher dry matter yield in 2009.

19

1. ._LL_- LL7-

14.o
12.0
,T 10.0 +
E 8.0
M ‘.
g 6.0 I
3 4.0
g 20
E ' E
r 0.0 I"
D

L L- L. L __LL.. ._ -L _.L L. L._L. L LL. - L - LL L-E

Yield in Lake City in 2008

 

     

 

12-Jun 24-Jun E 9-Jul I

i..._ L LLI

 

Figure 7. Accumulative forage yield from BMRS, GGM, HPM, SM, and teff located at Lake
City, MI in 2008, per species per planting date in dry matter Mg ha' . LSD (0,05) =0.82
(n=270) Standard error=0.6386

I.

12.0 '-

Dry Matter (Mg ha'l)

10.0 -

6.0 --

4.0 5
2.0

 

Yield in Lake City in 2009

 

SM
Teff

12-Jun

24-Jun 9-Jul

Figure 8. Accumulative forage yield from BMRS, GGM, HPM, SM, and teff located at Lake

City, MI in 2008, per species per planting date in dry matter Mg ha". LSD (0.05) =1.l4
(n=270) Standard error=l.09

20

Lake City 2008

In 2008, two cuts were taken with teff, BMRS, HPM, and GGM from the first and
second planting dates and one from the third planting date. One cut per planting date
from the SM. A heavy rain event in June (18.2 cm) washed away some of the seed along

with the fertilizer from the first planting date.

Teff was the lowest yielding in the second and third planting date. Teff yield was
between the highest yielding species (BMRS) and the lowest yielding species (SM) from
the first planting date. The BMRS was the highest yielding species in all three planting

dates. BMRS resulted in the highest overall yield. from the second planting date, with

11.5 Mg ha". The SM from the first planting date resulted in the lowest yield with 0.41

Mg ha".

Teff and HPM were not significantly different in the first planting date and GGM
was the second highest yielding species. The GGM and HPM were higher yielding than
teff on the second planting date. The GGM and HPM resulted in yields comparable to

BMRS from the third planting date (Figure 7).
Lake City 2009

In 2009, two cuts were taken with teff, BMRS, HPM, and GGM from the first and
second planting dates and one cut from the third planting date. One cutting was taken

from SM per planting date. Teff resulted in the highest yield from the first and second
planting dates with yields of 8.6 Mg ha" and 8 Mg ha", respectively. The SM resulted in

high yields from all three planting dates; however the SM yields were not significantly

21

different than teff from the first and second planting dates. Although the SM was not
significantly different than teff from the second planting date, BMRS and GGM were not
significantly different than the SM. The HPM was the lowest yielding species from the
second planting date. SM was not significantly different in yield from the third planting
date than the species teff, BMRS, and GGM. The HPM was the lowest yielding species in

the third planting date but not different from teff (Figure 8).

The temperatures at LC in 2008 (156°C) was slightly above the 30-yr normal
(15.1°C) during the growing season of Jun-Oct (Figure 4). The precipitation was well
above the 30-yr average (38.7 cm), with 47.9 cm received during the growing season of
J un-Oct with the first planting date receiving the most precipitation with 18.9 cm (Figure
3). The SM resulted in low yields due to some seed washing away. The BMRS grew well
when the temperatures were above the 30-yr normal and the rainfall was adequate. In
2009, the temperatures (143°C) were below the 30-yr normal (15.1°C) during the
growing season of Jun-Oct (Figure 3). The rainfall amounts (39.8 cm) were slightly

above the 30—yr average of 38.7 cm.

Higher yields were obtained from the majority of species in 2008. Although the N
rate in 2008 was 56 kg N ha", the temperatures were above normal and the average

precipitation was above normal which is why all species responded well to the 2008
growing season environment. Similar to EL in 2008, BMRS resulted in higher yields
when temperatures and precipitation were above normal. Teff resulted in higher yields in
LC in 2009 when temperatures were below normal yet there was adequate precipitation
similar to E1 in 2009. Teff also responded to the higher rate of N applied in a split

application in 2009.

22

Acid Detergent Fiber

The part of the grass that is made up of lignin and cellulose is what determines the
acid detergent fiber (ADF) (Schroeder, 1994). Acid detergent fiber affects how animals
digest forage. Digestibility is greater with lower ADF levels since there is less lignin
present (Russell and Johnson, 1993). Plant maturity can play a key role in increasing the
ADF levels (Hunter et al., 2008) which will decrease the digestibility (Schroeder, 1994).
All ADF values are weighted averages. The weighted ADF is formulated by taking an
average of the yield and the proportional ADF component and then adding the results.

There was an interaction between location and year but not planting dates.

ADF in East Lansing in 2008

440 _-. ..__. _._L_.-L L L
420 E-
400
380 '

 

 

 

 

340 .»
320 .

   

Acid Detergent Fiber (g kg‘l)
C”
Ch
0

GGM HPM Teff SM

Species

 

Figure 9. Total weighted acid detergent fiber (g kg") from BMRS, GGM, HPM, SM, and

teff located at East Lansing in 2008. Teff had the lowest ADF value 354 g kg". LSD (0.05)
=8.7 (n=270) Standard error=3.386.

23

ADF in East Lansing in 2009
400 , - f. 2. L ---
390
380 .-
370 E 4-
360 i
350 E— ,
340
330

 

 

 

 

Acid Detergent Fiber (g kg'1)

BMRS GGM HPM Teff SM

Species

Figure 10. Total weighted acid detergent fiber (g kg") from BMRS, GGM, HPM, SM, and
teff located at East Lansing, Mi in 2009. Teff showed significantly lower ADF than the other

four species at 353 g kg". LSD (0,05, =8.7 (n=270) Standard error=3.1432.

East Lansing 2008
In EL in 2008, teff had significantly lower ADF content in EL with 345 g kg"
compared to the other four species on the four planting dates. The BMRS had

significantly higher ADF content of 411 g kg". The HPM and SM were not different

from each other. The GGM ADF content of 389 g kg" was higher than teff (Figure 9).

These results indicate that teff was not as lignified, compared to the other four species
which makes teff more digestible compared to the other species in this study. Although
BMRS did have a higher yield than teff in all four planting dates, it resulted in a

significantly higher ADF content.

24

East Lansing 2009

In EL in 2009, teff again had significantly lower ADF content than the other
species with 353 g kg". The other four species, BMRS, GGM, HPM, and SM, were not
significantly different with values ranging from 374-3 82 g kg" (Figure 10). In 2009, teff
was consistently lower in ADF content compared to the other species in this trial from all
planting dates. The increased N application in 2008 (56 kg N ha") compared to the split
N application in 2009 (1 12 kg N ha") did not result in significantly different ADF values

for teff.

In 2008-9, teff in EL had the lowest ADF content compared to the other four
species with levels of 354 g kg" and 353 g kg" in 2009. This supports the findings of
Miller (2008) who found that teff ADF content ranged from 320-3 80 g kg". The SM was

harvested at a slightly later maturity which resulted in increased ADF content. In 2008
the BRMS was more mature on the first planting date resulting in an increase in ADF

content.

25

ADF in Lake City in 2008

 

 

 

Acid Detergent Fiber (g kg’l)

BMRS GGM HPM Teff SM

Species

 

L’qu- _

Figure 11. Total weighted acid detergent fiber (g kg") from BMRS, GGM, HPM, SM, and
teff located at Lake City, Mi in 2008. Teff showed significantly lower ADF than the other

four species at 320 g kg". LSD (0,05) =15.2 (n=270) Standard error=5.656.
ADF in Lake City in 2009
380
370 E——
360
350

u
I
1

340 E

330 E

320 Em-

 

 

  
 

 

 

Acid Detergent Fiber (g kg‘1)

BMRS GGM HPM Teff SM

Species

Figure 12. Total weighted acid detergent fiber (g kg") from BMRS, GGM, HPM, SM, and
teff located at Lake City, Mi in 2009. HPM showed significantly lower ADF than the other

four species at 346 g kg". LSD ((105) =10.l (n=270) Standard error=3.64l.

26

Lake City 2008

In LC in 2008, teff resulted in significantly lower ADF content among the five
species, with ADF content of 320 g kg" from the average of the three planting dates.
There was not a significant difference among the other species of BMRS, GGM, SM, and

HPM which had higher ADF values than teff, ranging from 338-352 g kg" (Figure 11).

Lake City 2009

SM resulted in significantly higher ADF content compared to the other species in 2009
at 375 g kg". BMRS, GGM, and teff, were not significantly different in ADF content

(Figure 12). The first fall frost date affected the ADF content of all species as

observations showed that the grasses had a brown color when harvested in mid-October.

In both years at LC, SM had the highest ADF content overall. This is because SM
was the fastest maturing plant compared to the other species. Although teff was not the
lowest both years, the early fall frost may have played a role in the higher ADF values in
2009. Teff was headed out on the last planting date in 2009, which could likely affect the
overall ADF. The BMRS was bred for low lignin but teff resulted in lower ADF than
BMRS. This would indicate teff as a good candidate for livestock and dairy animals

requiring a lower ADF content.

Neutral Detergent Fiber
Neutral detergent fiber measures the cellulose, hemicelluloses, and lignin in a
forage plant (Schroeder 1990). This is important in animal feed analysis, the lower the

NDF content the more a cow (Bos taurus) or other livestock can eat. A larger NDF value

27

will cause a decrease in dry matter intake. The planting date and year had significant

effects on NDF at both sites.

NDF in East Lansing in 2008
720 - a- .2- __ L -LL
700
680 E
660
640
620
600
580 ‘
560
540

 

 

 

Neutral Detergent Fiber (g kg’l)

 

 

I
L—__”_.._.._— - —- __.\_-.. .4

Figure 13. Neutral Detergent Fiber (g kg") from BMRS, GGM, HPM, SM, and teff located

at East Lansing 2008 per species per planting date. LSD (0,05) =20 (n=270) Standard
error=7.255

28

NDF in East Lansing in 2009
800 ~~wwm~——-u_- LLLLL-LLLLL-L
700 E
600 s
500 E
400 f
300 .
200 -
100 a

 

Neutral Detergent Fiber (g kg'l)

 

4Jun 16Jun . 640i 154u|

 

Figure 14. Neutral Detergent Fiber (g kg") from BMRS, GGM, HPM, SM, and teff located

at East Lansing 2009 (n=270) per species per planting date. LSD (0.05) =21 (n=270) Standard
Error=7.287.

East Lansing 2008

In 2008, the SM had significantly higher NDF content on the first, second, and
third planting dates which may be attributable to the earlier maturity. Observations of the
SM plots showed that the SM matured faster than the other WSGs which had a negative
effect on NDF. The BMRS had significantly higher NDF content on the second, third,
and fourth planting dates but not significantly lower than SM on the second and third
planting dates. The treatments from the first planting date did not show a statistical
difference between BMRS, GGM, HPM, and teff. There was not a difference between

BMRS, HPM, and SM from the second planting date. The lowest NDF content was from

GGM and teff from the second planting date (631-639 g kg") and the third planting date

(602-610 g ka" ). The GGM, HPM, and SM had the lowest NDF from the fourth planting

29

date (Figure 13). The millets from the third and fourth planting date were not as mature
when harvested, unlike the first and second planting date. The cooler temperatures in July
and August and the high precipitation in August resulted in slower maturity with the

millets and subsequent lower NDF.

East Lansing 2009

In 2009, SM had the highest NDF content from the first, second, and fourth
planting dates. The highest NDF content from SM (660 g kg") resulted from the first
planting date. Teff had the lowest NDF content on the first and fourth planting dates, with
the lowest NDF content (588 g kg") on the fourth planting date. Teff NDF content was

lower only on the first planting date. On the second and third planting dates, teff resulted

in high NDF content but not significantly different from SM (Figure 14).

The BMRS and HPM had the lowest NDF content from the second planting date.
The BMRS, GGM, and HPM had the lowest NDF content from the third planting date.
The lowest NDF content from the fourth planting date were GGM, HPM, and teff (Figure
14). Overall the BMRS resulted in the lowest NDF content from three of the four planting
dates, with the lowest NDF coming from the second planting date (565 g kg"). The
BMRS did not germinate in the fourth planting date; therefore NDF values are not given.
These species mature quickly within a few days of sunny weather or after a rainfall event.
Different species maturity may have caused the differences in NDF among the planting
dates with the same species. Overall, GGM and HPM were not significantly different

from the lowest NDF content or they were in the middle of the highest and lowest NDF

30

values. Observations of GGM and HPM indicate they do not mature as fast as SM and

BMRS.

In both years (2008-9) in EL, SM resulted in higher NDF content on almost all
planting dates. The BMRS was higher in NDF in 2008 from the third and fourth planting
dates yet was low in 2009 on the third planting date. The teff resulted in the lowest in
NDF in 2008 whereas in 2009 teff had higher NDF content or was not significantly
different from the highest NDF value. Maturity plays an important role in determining
NDF values. Different planting dates and different years resulted in some species
maturity more quickly than others. Weather may have also played an important role for in
2008 temperatures were warmer than the 30-year normal and in 2009 when temperatures

were cooler than the 30 year normal. The different nitrogen rates could have made an
impact on the NDF values. In 2008, 56 kg N ha" was applied whereas 112 kg N ha" was

applied in 2009 with a split application. There were a few species in 2009 that were
significantly different. The NDF values were lower overall in 2009 with the extra N input
which supports the findings by Hunter et al. (2008) where a fertility study was done
showing that the increase in N decreased the NDF content from an N fertility response

study.

31

660 ~—* "-W W“ -,.,-L.. ~ A ‘-
640 .-
620 -I
600 -‘
580 .
560 -'
540
520
500
480 --

 

Neutral Detergent Fiber (g kg‘1)

was H:

4-Jun I 24-Jun . 9-Jul

 

Figure 15. Neutral Detergent Fiber (g kg") from BMRS, GGM, HPM, SM, and teff located

at Lake City 2008 (n=270) per species per planting date. LSD (035) =31 (n=270) Standard
Error=ll.449

NDF in Lake City in 2009

700
600
500 E
400 4:
300 E
200
100-e-

 

BMRS

Neutral Detergent Fiber (g kg'1)
BMRS

 

4—Jun 16-Jun 6-Jul
Figure 16. Neutral Detergent Fiber (g kg") from BMRS, GGM, HPM, SM, and teff located

at Lake City 2008 per species per planting date. LSD (0.05) =28 (n=270) Standard
Err0F10.063

32

Lake City 2008
In 2008, the teff NDF content was not significantly different from BMRS or SM

from the three planting dates with the highest overall NDF content from the third planting
date with 614 g kg". Teff was not significantly different from GGM on the third planting
date. Overall, HPM resulted in the lowest NDF content from all three planting dates with

the lowest NDF being 547 g kg'l from the first planting date (Figure 15). Treatments

received 56 kg N ha"1 directly following first seeding. In 2008, air temperatures were

slightly above the 30-yr normal and with adequate rainfall with the first planting date
receiving the most rainfall. In 2008, the treatments received a lower N rate than 2009
which may have had an impact in the NDF values for the addition of N would likely see a

higher leaf to steam ratio for the grasses.

Lake City 2009

In 2009, teff had the highest NDF content from the first and second planting dates
with the highest NDF value of 665 g kg" from the second planting date. The HPM
resulted in the lowest NDF in the second and third planting dates. The lowest NDF
content was from the third planting date of HPM with 539 g kg". Teff and SM from the
first planting date had higher NDF content than GGM, HPM and BMRS which is likely
due to their earlier maturity compared to the other species. The BMRS, GGM, and SM
from the second planting date were significantly lOwer in NDF content compared to teff.
There was no significant difference in NDF content between any of the species in the last

planting date which was likely due to more immature plants at the time of frost (Figure

16). The air temperatures were below the 30-yr normal and the precipitation was above

33

. Crude Protein in East Lansing in 2008
E 130 E —-— i=7 —~v—e— W , ~

 

Crude Protein (g kg'1)
m
C

 

 

16-Jun I 30-Jun 14-Jul

I_.LL_._ - .L i- -__~___L _L_L-LLLLLLL .- L-L LL LL .L. g , , _ - fl L LL-

Figure 17. Total weighted Crude Protein (g kg") from BMRS, GGM, HPM, SM, and teff
located at East Lansing, MI in 2008 per species per planting date (n=270). LSD (035) =10.4
(n=270) Standard error-=7 .38

Crude Protein in East Lansing in 2009

 

Crude Protein (g kg'l)

 

I
I
I

E 4-Jun 5 16-Jun

I; L _

15-Jul

Figure 18. Total weighted Crude Protein (g kg") from BMRS, GGM, HPM, SM, and teff

located in East Lansing, MI in 2009 per species per planting date. LSD (0.05) =11 (n=270)
Standard Error=7.73

35

East Lansing 2008
In 2008, teff resulted in significantly higher CP than other species from all four

planting dates; the highest CP found in teff resulted from the third planting date with 147

g kg]. The BMRS resulted in the overall lowest CP from all four planting dates, with the

lowest content from the fourth planting date with 57 g kg]. This may be due to lack of

nitrogen used in the study to meet the demands for BMRS growth.

The GGM CP was not significantly different from the teff from the first, third and
fourth planting dates and was between the highest and lowest CP content from the second
planting date. The HPM resulted in the highest CP from the first planting date and
average CP from the second, third, and fourth planting dates. SM was not significantly
different than BMRS in the first planting but was significantly higher than BMRS in the
other planting dates (Figure 17). GGM resulted in the second highest CP from the second,

third, and forth planting dates.

East Lansing 2009

In 2009, the average CP content of all species were higher than in 2008 which is
likely due to increased nitrogen application in 2009 to the species compared to 2008. Teff
resulted in the highest CP from the first, third, and fourth planting dates; the highest teff
CP was from the fourth planting date with 163 g kg]. The higher C P content of all
species in the forth planting date was likely due to the immaturity of the species resulting
in a higher concentration of CP in the plants. The SM resulted in the lowest CP from the

first, third, and forth planting dates with the lowest CP from the first planting plant date

36

with 86 g kg]. The BMRS from the fourth planting date resulted in poor germination;

therefore there is no data (Figure 18).

In EL, both years resulted in teff being higher in CF content than the other
species. In a study by Twidwell et al. (2002) comparing sudangrass, teff, SM, and GGM,
teff resulted in the highest CP but was not significantly different to the other four species
at one location. At another location, Twidwell at al. (2002) found teff CF to be

significantly higher than the other four species.

Crude Protein in Lake City in 2008
250 . H . - -- a..- __- ----_-.._ “1--.... . ,1

200
150 .-~

100 --

Crude Protein (g kg‘1)
U1
C

 

i
12-Jun ’ 24-Jun I 9-Jul

Figure 19. Total weighted Crude Protein (g kg") from BMRS, GGM, HPM, SM, and teff

located at Lake City, MI in 2008 per species per planting date (n=270). LSD (0,05) =1.82
(n=270) Standard Error=l 1.44

37

Crude Protein in Lake City in 2009

 

 

   

‘ 250 WWW ---..----_--.- -_ . _
'2'
3‘2 200 E 1
£3 ; E
.s 150 7*- - I
3 E
g 100 — :
a .
8 so -
3
h 1
U o --. —:— --
I "’ V’ 2 2 2 «‘- V’ 2 2 2 a:
E o: a: o. m '5” a: CL :1) a)
3 a2: 8 I '- 2 8 I '— E
E 16dun E 6dul ‘

 

Figure 20. Total weighted Crude Protein (g kg!) from BMRS, GGM, HPM, SM, and teff

located in Lake City, MI in 2009 per species per planting date. LSD (0,05) =9.2 (n=270).
Standard Error=5.705

Lake City 2008

In 2008, CP in teff was highest in only the third planting date with 187 g kg".

However, teff CP was not significantly different than BMRS, GGM, or HPM in the
second and third planting dates. SM resulted in the lowest CP content from the second
and third planting dates. HPM resulted in significantly higer CP in the first planting date

(Figure 19).

Lake City 2009

In 2009, HPM resulted in the highest CP compared to other species from all three
planting dates. The highest CP content of teff was 187 g kg'l from the third planting date.
The SM resulted in the lowest CP among all three planting date with CP content of 101 g
kg'1 from the first planting date. Teff was not significantly different from BMRS, GGM,

or HPM in the first or second planting dates. Teff and HPM were not significantly

38

different in the third planting date. The third planting date resulted in the highest CP
content whereas the first planting date averaged the lowest (Figure 20). This is likely due

to less mature plants in the third planting date when the first fall frost occurred.

At LC, teff was not significantly different than the other species among the
planting dates which supports the finding by Twidwell (2002) where the WSG species
from one location did not result in a significant difference across CP. Overall, the first
planting date at both locations and years resulted in the lowest averaged CP among the
species. This may be due to the plants running out of available soil nitrogen by the time

the second harvest occurred from the first planting date.

Teff Forage Quality and Yield Summary

East Lansing Teff

 

 

 

750 — _- . 12
675 , , - . .. ‘ ., . o
600 2
525 g
450 2
a. 375 A CP
:3 300 g
no 225 :- +ADF
' m
E 150 r - , i» t. —x—NDF
- 75 7‘“ _, .,,._.__,u,.L,. W___--._._ _____-_....-____.._... ...--,,.__..- V .
g 2 .3 ‘2: :5. :5. 2 :2;
a. :2 a’ a . é 2 a a;
N _ .
2008 3 2009

 

Figure 21. The dry matter (Mg ha!) and forage quality as NDF, ADF, and CP (g kgq) of
teff located at East Lansing in 2008 and 2009 with four planting dates. Error bars for any
given line are based on LSD (0.05). Each line was analyzed separately.

39

Lake City Teff

 

 

U
W 2
2
~— 3
_ _ 2 CP
— g '~~-.e;..~=-ADF
on
g -+(-NDF
J5 -o—Yie|d

 

 

112-Jun 24-Jun 9-Jul

 

2008

Figure 22. The dry matter (Mg ha!) and forage quality as NDF, ADF, and CP (g kg'l) for
teff located at Lake City in 2008 and 2009 with three planting dates. Error bars for any
given line are based on LSD (0.05). Each line was analyzed separately.

Figures 21 and 22 contain a summary of the responses of teff dry matter (Mg ha'l)

and forage quality as CP, ADF and NDF (g kg!) to various planting dates at which the

teff was seeded. In East Lansing, the most notable response evident in both 2008 and

2009 was the loss of yield productivity with the later planting dates (Figure 21). In Lake
City, the loss of yield productivity was also evident with the later planting dates in 2008
and 2009 (Figure 22). The CP in Lake City appeared to increase in the teff planted at the

later dates in 2008 and 2009.

40

CONCLUSION

The treatments had interactions by year, by location, and by planting date for
yield, NDF, and CP. The ADF had a year by location interaction. There was an
establishment problem in LC in 2008 with the first planting resulting in some seeds
washed away with a major rainfall event in June. The BMRS in EL in 2008 on the fourth

planting date failed to germinate which may have been due to lack of soil moisture.

Planting and harvest timing is crucial. With adequate precipitation and warm air
temperatures, all annual grasses mature fast. When the air temperatures were cool, all
species resulted in delayed growth. This is to be expected as all species in this study are
warm season grasses and need optimum temperature to grow to their maximum. Teff
yielded better than other species when the air temperatures were cooler with adequate
rainfall. Thus, teff may perform better in summers where temperatures are slightly below
normal. The BMRS had higher yields when the air temperatures were warmer with
adequate rainfall. The SM matured earlier than all Other species with little to no re-
growth after the first harvest. The SM overall resulted in the lowest forage quality among

the millets. Therefore, SM would not be a good warm season annual grass for Michigan.

Teff would fit as an annual warm season grass that can be planted in early
summer to provide a forage crop with relatively high quality during a summer slump in
Michigan. Teff had sufficient yields, high CP, and low ADF but was higher than
expected in NDF content. With proper management and crucial harvest timing, teff

would fit as an emergency forage crop in Michigan.

41

LITERATURE CITED

Anonymous. 23 February 2007. Millet for Forage Use. Government of Alberta.
Agricultural and Rural Development.

Goering, H. K., and P. J. Van Soest. 1970. Forage fiber analyses (apparatus, reagents,
procedures, and some applications). Agric. Handbook 379. ARS, USDA,
Washington, DC.

Hach, C.C., S.V. Brayton and AB. Kopelove. 1985. A powerful Kjeldahl nitrogen
method using peroxymonosulfuric acid. J. Agric. Food Chem. 33:1117-1123.

Hunter, M, K. J. Moore, M. Wiedenhoeft and Thomas, Loynachan. 2008. Evaluation of
teff Eragrostis tef(Zucc.) Trotter, as a forage crop in New York. Iowa State
University. Ames,Iowa.

Ketema, S. 1997. Tef Eragrostis tef (Zucc.) Trotter. Promoting the Conservation and Use
of Underutilized and Neglected Cops. 12. Institute of Plant Genetics and Crop Plant
Research, Gatersleben/Intemational Plant Genetic Resources Insistute. Rome, Italy.

Kilcer, T.F., Q.M. Ketterings, J.H. Chemey, P. Cerosaletti and P. Barney. 2005.
“Optimum stand height for forage brown midrib sorghum x sudangrass in North-
eastern USA.” Journal of Agronomy and Crop Science. 191 p 35—40.

Koch, D. W. 1992. Foxtail Millet-Management for Supplemental and Emergency Forage.
University of Wyoming. SMRR Info Source.

Miller, D. 2008. Teff as an alternative summer forage. Proceedings, 2008 California
Alfalfa & Forage Symposium and Western Seed Conference. San Diego, CA.
University of California Cooperative Extension Plant Sciences Department.

Miller, F.R. and J .A. Stroup. “Brown midrib forage sorghum, sudangrass, and corn: What
is the potential?” California alfalfa and forage symposium. (2003) pp 143-150.

Moser, L.E., Burson, BL. and Sollenberger, LE. 2004. P.222-223, 1036-1039. Warm
season (C4) grasses. Agronomy number 45. American Society of Agronomy,
Madison, WI.

Norberg, S., R. Roseberg, B. Charlton and C. Shock. 2008. Teff: A new warm season
annual grass in Oregon. Oregon State University Extension Extention Services
Sheet EM 8970.

Rayburn, E. 2002. Use Forage Test To Diagnose Management Problems. West Virginia
University Extension http://www.wvu.edu/~agexten/forglvst/foragtst.htm

42

 

Russell, M and K. Johnson. 1993. Selecting Quality Hay for Horses. Cooperative
Extension Service Purdue University. ID-l90.

Schroeder, J .W. 1994. Interpreting Forage Analysis. North Dakota State University. AS-
1080.

Sedivec, K and B. Schatz. 1991. Pearl millet: Forage Production in North Dakota R-1016.
North Dakota State University.

Tranel, L. 2000. Nitrogen fertilizer on grass pastures. Iowa State University. ISU fact
sheet LT-107.

Twidwell, E.K., A. Boe and DP. Casper. 2002. Teff: A new annual forage grass for
South Dakota? Extension Service Bulletin ExEx8071. South Dakota State
University. Srookings, SD.

43

ABSTRACT

EFFECTS OF NITROGEN RATES ON TEFF (Eragrostis Iequcc. Trotter) IN
MICHIGAN

By
Stephanie Ilene Peck

Forage producers in the North Central region could benefit from a forage species which
will maintain high yields and forage quality though the summer. Teff [Eragrostis tef
(Zucc.) Trotter] is a warm season, annual grass from Ethiopia that has potential in
Michigan during the warm, dry summers (summer slump). Teff is an aggressive crop
with good seedling vigor that continues to grow well under high temperatures but has not
yet been tested in Michigan for forage production. Teff possesses several desirable
characteristics including drought tolerance and rapid growth. The objectives of this study
were to determine the optimal nitrogen rate for teff. Ammonium sulfate (AMS) was used
as the carrier of nitrogen since it does not volatilize when applied to the soil surface. The
effect of increasing nitrogen application rates was evaluated for forage quality (ADF,

NDF, CP). Five nitrogen treatments were used with rates of (0, 56, 84, 112, and 168 kg N
ha!) This study was conducted over two years (2008-9) at two locations in Michigan
(East Lansing and Lake City). Nitrogen rate of 84 kg N ha"l was optimal for higher yields

and desirable forage quality. Increasing nitrogen rates significantly affected yield and
forage quality of teff. Determining optimal nitrogen use is important to producers who

want to get high yields and forage quality but do not want to waste nitrogen inputs.

44

 

Chapter 2 - INTRODUCTION

Nitrogen (N), an essential element for plant growth, is one of the most limiting
growth factors in grasses (Cuomo, 2000). Nitrogen increases plant leaf color (Tranel,
2000) crude protein, increase leaf tiller, leaf weight and dry matter yield (Moser et al.,
2004). Teff is a newly introduced warm-season, annual forage grass to Michigan by
various seed markerters. Known for its wide range of adaptability, teff can adapt to both
drought stressed and waterlogged conditions (Hunter et al., 2008). Since no prior work on
nitrogen response has been done in Michigan, determining the optimal nitrogen rate will
provide growers with sound environmental and economical information on which to base
their decisions on both the rate and timing of N fertilizer applications to teff. Proper N
fertilization and management strategies are important to determine both yield and forage
quality. In 2008, the average price for ammonium sulfate (AMS) was $391/ton (USDA-
NASS). When too much N is applied, lodging may occur (Norberg et al., 2008), which
can make harvesting difficult. When too little N is available, the plants will become light
green to yellow in color and eventually stunting occurs. Using optimal N is important in

cost efficiency, maximum yield potential, and forage quality.

Nitrogen has some disadvantages when not managed properly. Leaching is a
major concern with N; not only is it a loss of N but can be costly and causes health
concerns but it can also cause ground water contamination. If N03' is over the regulation
limit of 10 mg NO3-N L-l (Moser et al., 2004) in groundwater, a potential of
methemoglobinemia (blue baby syndrome) may occur in babies. There are ways to

reduce the loss of nitrogen into the environment and reduce environmental waste.

45

Accurate fertilization will improve management along with improvement of the
environment. Finding an optimal rate based upon economic return for yield and forage
quality will decrease the chance of leaching and minimize the uptake of luxury nitrogen
to the plant (Satter el al. and Whitehead 2000; Moser 2004). Since farmers do not have
extensive experience with teff, it is important to determine the nitrogen rate that would
give the best economic return for teff forage while protecting the environment.

The objectives of this study were as follows:

1) Determine the optimal nitrogen rate in the form of ammonium sulfate (AMS)
for dry matter yield of teff.

2) Determine the effect of nitrogen rate upon the forage quality of teff.

Materials and Methods

Study sites and experimental design

This study was conducted in 2008 and 2009 at two locations in Michigan. The
first site was at Michigan State University Crop and Soil Science Farm in East Lansing,
MI (Latitude: 42.7551 Longitude: 84.48499) and the second at Michigan State University
Lake City Experiment Station in Lake City, MI (Latitude: 44.3352872, Longitude:
85.3352872). The soils in East Lansing consisted of a Colwood-Brookston loam (fine-
loamy, mixed, mesic Typic Haplaquolls, and fine-loamy, mixed, mesic Typic
Argiaqoulls) with a pH of 6.5. The soil at Lake City was Nester Sandy loam (fine, mixed,
semiactive, frigid Oxyauic Glossudalfs) with a pH of 6.9. In 2009, the Lake City trial was
planted in an area previously occupied by Kura clover (Trifolium ambiguum). Kura

clover is known for its persistence because of its massive rhizomes, which is an

46

underground stem. Trimec (2, 4-D, dicamba) was sprayed to control the Kura clover

(Trifolium ambiguum).
Teff Nitrogen Rate Response

The response trial to determine the effect of increasing nitrogen rates upon teff

dry matter yield was arranged in a randomized complete block design with four

replications with five nitrogen treatments. The nitrogen rates include: 0 kg ha], 56 kg ha"
1, 84 kg ha], 112 kg ha'], and 168 kg ha'l. Ammonium sulfate (AMS) (21-0-0 Sulfur
24%) was surfaced applied at time of seeding with the 0 kg N kg'I treatment receiving 40
kg S ha"l added to adjust for the sulfur in the content of the 168 kg N ha.1 treatment. The
56 and 84 kg N ha'1 treatments were applied at time of planting. The 112 and 168 kg N

ha'1 treatments were applied in a split application with half of the total rate being applied

at planting and the other half after the first harvest (see tables 4 and 5 for seeding and

harvest dates).

Table 4. East Lansing, MI planting and harvest dates for teff with five different fertility
rates in 2008 and 2009.

East Lansing, Michigan

N Rate 2008 2009

 

 

(kg ha']) Seeding Cut 1 Cut 2 Cut3 Seeding Cut 1 Cut 2 Cut 3

 

0
56
84

112
168

 

 

16-Jun l-Aug 3-Sept lO-Oct 16-Jun 30-Jul 25-Aug 5—Oct

 

 

 

 

 

47

Table 5. Lake City, Michigan planting and harvest dates for teff with five fertility rates in
2008 and 2009.

N rate 2008 2009
ha- Seeding Cut 1 Cut 2 Cut 3 Seeding Cut 1 Cut 2

0

56
84 24-Jun 26-Aug 18-Sept 9-Oct 16-Jun 12-Aug 12-Oct

112
168

 

Dry Matter Yields

The teff harvest timing goal was to take first cutting at late boot stage. Yield was
collected using a rotary flail harvester (Carter Manufacturing Co. Inc., Brookston, IN) set
to harvest at a 10.2 cm height and a subsample was obtained at time of harvest for dry
matter determination and forage quality analysis. The samples were weighed and placed

in a dryer for 96 h at 70°C and then weighed again to determine the moisture content.
Forage Quality

After determining dry matter, samples were ground through a 1 mm screen in a
Christy laboratory hammer mill (Christy Turner, Chelmsford, U.K.). Then the sample
was scanned using a near-infrared spectrophotometer (NIRS) Foss model 6500 (FOSS
NIRSystems, Inc., Eden Prairie, MN). The wavelengths (between 700-2500 nm) are
reflected off the samples and were recorded. The predicted wavelengths were used to
estimate crude protein (CP), neutral detergent fiber (NDF) and acid detergent fiber (ADF)
from equations created by the NIRS Forage and Feed Testing Consortium (Madison, WI).

Subsets of samples are chemically analyzed in the Michigan State University Forage

48

Physiology Laboratory for CP, NDF, ADF, and DM to validate NIRS prediction. The
Goering-VanSoest method (1960) was used to determine NDF and ADF with the addition
of 1 ml of alpha amylase. Total nitrogen (N) was determined by the Hach modified
Kjeldahl Nitrogen method (Hach et al., 1985) and crude protein was estimated by 6.25 x
total N. Dry matter content was determined, for the lab samples, drying 0.50 g of sample

in a crucible at 90°C for 12 h.

Statistical Analysis

Data was tested for normality and unequal variance using PROC UNIVARIATE.
Analysis of variance (ANOVA) was performed on yield, forage quality, crude protein
content, NDF, and ADF using PROC MIXED procedure (SAS Institute, v9.1).
Interactions were considered significant at P>0.1 and main effects considered significant
at P>0.05. Contrasts of the means were used to determine mean separation with a
confidence interval of 95%. The replication by plant date and the replication were both

considered random effects. The nitrogen rate is considered a fixed effect.

Weather Conditions
East Lansing, MI

The total amount of precipitation East Lansing received between April and
October was more than the 30 year average in both years. East Lansing received an
average of 15.1 cm more precipitation than the 30 yr average of 50.6 cm, for the 2008
and 2009 growing season. The total precipitation in 2008 and 2009 growing season of
April through October was 57.3 cm and 74.1 cm, respectively. In 2008, there was 21.4
cm in September, higher than the 30 yr average. In 2009, August had high rain amounts

of 16.8 cm (Figure 23).

49

Precipitation in East Lansing

 

- 2008
2009 1

g”
,, , I Ii 30 year avg

Apr May June July Aug Sept Oct

 

 

Figure 23. Monthly average precipitation (cm) received in East Lansing, M1 for the growing
season (April-October) in 2008 and 2009.

Temperatures during the growing season (April to October) in both years
averaged 0.23°C warmer than the 30 year average. Temperatures were warmer than the
normal for 2008 and slightly cooler in 2009. Seasonal averages for 2008 and 2009 were
164°C and 15.0°C, respectively. The 30 year average temperature for East Lansing was

15.5°C (Figure 24).

Temperature in East Lansing

   

25 :
20 - l
1 U if }
1 § 15 , .2008 g
h .
E? 10 . , 2009
. a .
, .— 7 INormal
1 APR MAY JUN JUL AUG SEP OCT

 

Figure 24. Monthly means for temperatures (°C) received in East Lansing, MI for the
growing season (April-October) in 2008 and 2009.

50

Lake City, MI
The total amount of precipitation during the growing season in 2008 and 2009
was higher than the 30 yr average of 52.9 cm. In 2008, June had more than normal
precipitation resulting in a higher than normal average of 61 .4 cm. The average
precipitation in 2009 (53.5 cm) was just slightly higher than the 30 year average. In 2009,
October had 12.3 cm, higher than the 30 year average. The total precipitation for both

years combined was above the normal by 4.6 cm (Figure 25).

Precipitation in Lake City

   

20 , . L.LL.LLL L L L L L L L L ,
15
I 2008
. E 10 7 .
y _ 2009 j
z 5 l i I i I 5 I I 30 year avg ‘I
o I 7 g

‘ Apr May June July Aug Sept Oct

Figure 25. Monthly average precipitation (cm) received in Lake City, M1 for the growing
season (April-October) in 2008 and 2009.

The average temperature in Lake City was close to the 30 year average of 133°C.
In 2008, the average temperature (137°C) during the growing season was slightly higher
than the 30 yr normal. The 2009 growing season temperatures had a lower average than
the 30 yr normal, with an average of 126°C. Lake City averaged -0.1°C less in both

years combined during the growing season (Figure 26).

51

Lake City Temperature

25 ~ 7 7 ~ ~ A w v —
20 . L
U
3 15 .2003
L
g” 10 ~ — —rv77 2009

 

U'l
—
I

O L.

u a Normal V

APR MAY JUN JUL AUG SEP OCT

Figure 26. Monthly average temperatures (°C) received in Lake City, M1 for the growing
season (April-October) in 2008 and 2009.

Results and Discussion

Teff Yields

Harvest times took place at boot stage with the plot having about 10% heading. In
2009, Lake City had a cool, dry July (Figures 25 and 26), which may have caused teff to
delay growth. The last harvest in 2009 had severe frost damage. The first from was 1-

October and the last harvest took place 12-October.

52

TefiYhfld

 

‘ 14 » 4~4774- 7 7 7»
i :r 12
1 'N
1 g 10
1 M
t E 8 i
1 a 1
1 3:: 6 , 1
N
2 4 .
E
. D 2 . 1
i 0 j
i
l 0 55 84 112 158
Nitrogen kg ha‘1 é
|

Figure 27. Accumulative forage yield per nitrogen treatment across two years (2008-9)
located at two locations (East Lansing and Lake City, MI) in dry matter Mg ha'l (N=80).
LSD=0.85 Standard error=6.050.

Three cuttings were taken in 2008, at East Lansing and Lake City while in 2009,
East Lansing had three cuts and Lake City had two cuts. There was not a significant
interaction between years by N rate and location by N rate interaction for yield, therefore,
data presented as an average over years and locations. The greatest teff yields were
obtained with 1 12 and 168 kg N ha", both resulting in 12 Mg ha" per year but not
significantly different from each other. The lowest yielding was from the 56 kg ha'] of N
yielding at 9.5 Mg ha". There was obviously soil nitrogen present in the 0 N treatments
as the yield was not different than the 56 kg ha"! treatment (Figure 27). Nitrogen rates had
a significant effect on yields per treatment. Nitrogen rates of 168 and 112 were not

different from one another. The 0 and 56 kg ha'1 treatments were not significantly

53

different (Figure 27). Treatments of 56 and 84 kg ha'1 are not significantly different from

one another on yield. The 112 and 168 kg ha.1 N were split applications.

Acid Detergent Fiber

The part of the grass that is made up of lignin and cellulose is what determines the
acid detergent fiber (Schroeder, 1994). Acid detergent fiber is important to the animals’
digest of the grass. The digestibility is greater with lower ADF levels (Russell and
Johnson, 1993). Plant maturity can play key roles in increasing the ADF levels (Hunter et
al., 2008) which then decrease the digestibility (Schroeder, 1994). There was not an
interaction between N rate and locations or year. All ADF values are weighted average.
The weighted ADF is formulated by: ((yield1*ADF1) + (yield2*ADF2)) /

(yieldl +yield2). There was a significant effect of N rate upon ADF content. The 168 kg

N ha'1 treatment resulted in the lowest ADF content at 334 g kg], but not significantly
different from 1 12 kg N ha.I at 337 g kg.l (Figure 28). The 0 kg N ha°l resulted in the

highest weighted ADF at 343 g kg'1 (Figure 28). The addition of nitrogen had a positive

effect on decreasing the ADF. The plant becomes more fibrous with less nitrogen which
will increase the percent ADF (DePeters, 1990). This is likely due to a higher leaf to stem

ratio with the stem containing more lignin than the leaves.

54

 

TeffADF
355 LL- gene—— -- ..--._-L_L__L-LL_.L
350 LLLL .. ___L
345 1L-
340 i-
335 j
330
325 --

 

 

 

 

 

   

O 56 84 112 168

Acid Detergent Fiber (3 kg’1)

Nitrogen kg ha'1

 

Figure 28. Total weighted acid detergent fiber (g kg!) with five nitrogen treatments across
two years (2008-9) located at East Lansing and Lake City, MI (n=80). LSD=5.9 Standard
error=2.104.

Neutral Detergent Fiber
East Lansing, MI

Neutral detergent fiber measures the cellulose, hemicelluloses, and lignin in a
forage plant (Schroeder, 1990). This is important in animal feed analysis, the lower the
NDF content the more the cow (Bos taurus) or other livestock can eat. A larger NDF
value will cause a decrease in the dry matter intake. There was a significant difference

between locations by treatment, therefore, the locations will be provided separately, and

treatments had a significant difference. In East Lansing, 168 kg ha'I N resulted in the
lowest weighted NDF value at 617 g kg], lower than 0, 56, 84, and 112 kg ha'1 N. The 0,
56, 84, and 112 kg ha'I N were not significantly different and had weighted NDF values

of 632, 639, 634, and 630 g kg", respectively (Figure 29).

55

 

East Lansing NDF

660 ‘- -LL .- LL. __
650 «i- we,» - — __L_L___ __
640

530 iLLL
620 . __
610
600 -

 

 

 

0 56 84 112 168

Neutral Detergent Fiber (3 kg'1)

Nitrogen kg ha'1

Figure 29. Total weighted acid detergent fiber (g kg'l) with five nitrogen treatments across
two years (2008-9) located at East Lansing, MI (n=80). LSD=11.3 Standard error=4.346.

Lake City, MI

In Lake City, there was a significant effect of N fertility level in both years of the

trial. In 2009, the teff at Lake City had 2 cuts whereas, in 2008, 3 cuts were taken,

therefore NDF comparisons are made with within a year. The 168 kg N ha'I resulted in
the lowest weighted NDF at 612 g kg"1 (Figure 30). All teff treatments receiving N had

significantly lower NDF contents than the 0 kg H ha.l treatment. Adding nitrogen had a

positive effect on the NDF values in Lake City which is likely due to a higher leaf to stem

ratio as previously mentioned.

56

 

 

 

 

650 ~ - -— - -~.—-——— ___.LLLLLLLLL
540 L T __L_LLLLLLLL

 

630
620
610
600 .
590

 

 

   

Neutral Detergent Fiber (g kg'1)

Nitrogen kg ha'1

 

Figure 30. Total weighted acid detergent fiber (g ng) with five nitrogen treatments across
two years (2008-9) located at Lake City, MI (n=80). LSD=12.6 Standard Error=4.933.

Crude Protein

East Lansing, MI

Crude protein (CP) is an estimate of the total protein content that is calculated

from the amount of nitrogen in the plant tissue (Hunter et al., 2008). The recommended

CP for a lactating cow requires 150-160 g kg'1 to maintain a healthy diet (Stallings et al.,
2009)

There was a significant interaction between location by treatment, and treatments

had a significant difference. In East Lansing, teff receiving rates from 84, 112, and 168

kg N ha"1 met the requirement for a lactating dairy cow however; teff receiving N rates of
0 and 56 kg ha.1 did not meet the diet needs of a lactating cow. The teff receiving 168 kg

N ha'l treatment had the highest CP content at 167 g N kg. The lowest CP content in

57

teff was 131 g kg'l from the 56 kg N ha'l treatment. There was a difference of 36 g kg-1
CP between the 56 and 168 kg N ha'] treatments. There was a difference of 29 g kg'l CP
between the 0 and the 168 kg ha'1 N treatments. The CP from 168 kg ha'1 N treatment
was significantly different than the 0, 56, 84, 112, and 168 kg N ha'l. Treatments
receiving 0, 84, and 112 kg N ha'1 were not significantly different from one another and
0, 56, and 84 kg N ha'1 were not significantly different (Figure 31). Teff was planted on
16 June in 2008 and 2009 in Lake City. The 0, 56, and 84 kg N ha'] were not split

applications. In August 2009, there was 16.8 cm of precipitation (Figure 23). The AMS
had potential to leach, which could have affected the non-split application treatments. In

2008 there were slightly warmer temperatures than the 30 year average in June-August

 

(Figure 24).
East Lansing Crude Protein ;
200 ~ «A» r _w ,_,_,,, M-.- I
1 ET .
T 3’:
g 3 150
‘ .E
U
i O.
z a so
i a ‘
U 1
1 o 56 84 112 168
“ I
Nitrogen kg ha’1 .

LMW ,__#_ _--_-,_._ M 7

Figure 31. Total weighted crude protein (g kg'l) with five nitrogen treatments across two
years (2008-9) located at East Lansing, MI (n=80). LSD=11.4 Standard Error=3.969.

58

Lake City, MI

In general, the increasing rates of nitrogen resulted in higher CP values in
teff. The teff from all treatments at Lake City met the CP requirements for a lactating
dairy cow. The range of teff CP from Lake City for both years was between 152 g kg—1
and 187 g kg". There was however, a significant difference between 0 kg ha'1 and 56, 84,
112, and 168 kg ha'1 N (Figure 32). The temperatures during the planting and last harvest

(2008: Jun-Sept, 2009: Jun-Oct) were slightly above normal (Figure 26). Precipitation in
2008 on average was below normal between the planting and last harvest whereas in
2009 the precipitation was above normal in most cases. The last harvest in 2009 occurred

after the first frost.

Lake City Crude Protein
200 . ~~ "fiwwg*~w LL, 7M 7" —~ L, . .

1

U1
C

1

Crude Protein (g kg‘)
0
O

     

0 56 84 112 168

Nitrogen kg ha'1

 

Figure 32. Total weighted crude protein (g kg-l) with five nitrogen treatments across two
years (2008-9) located at Lake City, MI (n=80). The 56, 84, 112, and 169 kg ha'I N
treatments were significantly higher than the 0 kg ha'l N. LSD=9.5 Standard Error=4.281

59

CONCLUSION

Good stand of teff were established in both locations (EL and LC) and years

(2008 and 2009). There was not a significant difference in yield between years or

locations. An application of 84 kg N hal (11 Mg ha'l) at the time of planting was optimal
for teff yield. Adding a split application of N totaling 168 kg N ha.1 was comparable in

yield (12 Mg ha'l).

The results of the study showed that with the addition of nitrogen and proper

management, teff forage quality improved. In most cases a split application of 168 kg ha-
I (half at planting and after first harvest) had a positive influence on teff yield, ADF, CP,
and lower NDF in EL. In LC, 112 kg N ha'1 was optimal for high yields and CP with

lower ADF and NDF. Although teff receiving 168 “kg N ha'1 resulted in the highest yield,

one application of 84 kg N ha'1 will result in the greatest economic yield rather the

highest yield resulting in savings on fertilizer cost. Teff has a potential market in
Michigan for lactating dairy cows since it provided a good nutritional profile for this

class of animals.

60

LITERATURE CITED

Cuomo, G. 2000. Nitrogen management for grass pastures.
Retrieved Noverber 5, 2009, website:

<http://www.uwex.edu/ces/forage/wfc/proceedings2000/cuomo.htm>

DePeters, E., Medrano, J ., Bath, D. and Harper, D. 1990. Cereal forage for dairy cattle.
University of California. California agriculture 44(6):21-23.

Goering, H. K., and P. J. Van Soest. 1970. Forage fiber analyses (apparatus, reagents,
procedures, and some applications). Agric. Handbook 379. ARS, USDA,
Washington, DC.

Hach, C.C., S.V. Brayton and AB. Kopelove. 1985. A powerful Kjeldahl nitrogen
method using peroxymonosulfuric acid. J. Agric. Food Chem. 33:1117-1123.

Hunter, M, K.J. Moore, M. Wiedenhoeft, and T. Loynachan. 2008. Evaluation of teff
Eragrostis tef(Zucc.) Trotter, as a forage crop in New York. Iowa State University.
Ames,Iowa.

Moser, L.E., B.L. Burson, and LE. Sollenberger. 2004. P.222-223, 1036-1039. Warm
season (C4) grasses. Agronomy number 45. American Society of Agronomy,
Madison, WI.

Norberg, S. R. Roseberg, B.Charlton and C. Shock. 2008. Teff: A new warm season
annual grass in Oregon. Oregon State University Extension. Extention Services
Sheet EM 8970.

Russell, M and Johnson, K. 1993. Selecting Quality Hay for Horses. Cooperative
Extension Service Purdue University. ID-l90.

Satter, L.D., Z. Wu, and J. M. Powell. 2000. Reduction of nitrogen and phosphorus in
dairy manure by diet manipulation. P. 31-52. In Proc. California Animal Nutrition
Conf., Fresno, CA.

Schroeder, J.W. 1994. Interpreting Forage Analysis. North Dakota State University. AS-
1080.

Stallings, C., MD. Hanigan and RE. James. 2002. Feeding protein to meet dairy cow
requirements can result in cheaper, environmentally friendly rations. Virginia
Cooperative Extension (404-354).

Whitehead, DC. 2000. Nutrient elements in grassland: Soil-plant-animal relationships.
CAB Int., Wallingford, Oxon, UK.

61

APPENDICES

Table 6. Near infrared reflectance spectroscopy calibration and validation statistics for
crude protein, acid detergent fiber, neutral detergent fiber for stands harvested over two
years (2008-9) in Lake City and East Lansing, MI. .

 

 

 

 

Constituent n Mean SEC'l' SECVI r2 § l-VRfl
-l
g kg
Crude protein 71 124 6.6 8.9 0.98 0.92
Acid detergent fiber 169 407 14.8 15.4 0.95 0.94
Neutral detergent fiber 152 601 20.5 22.5 0.97 0.88

 

TSEC, standard error of calibration.
iSECV, standard error of cross validation
§r2, coefficient of determination for calibration

1] l-VR, 1 minus the variance ratio in cross validation during modified partial least
squares regression

62

2009 Teff Forage Quality

 

 

 

 

 

 

 

 

 

 

 

800.0 — 1 — ——--—
600.0
“to rib"—
; 4000
ID
200.0
, 0-0 " ;—_i::_:';“‘” "T‘TTTLE :_'*' .. ' ‘
i 1-Jul 1 9-Jul 16-Jul 23-Jul
‘ -o—Protein: 246.9 . 194.2 . 146.9 T 110.9 -
5 -ADF a 232.2 l 251.7 313.2 348.9
ear-NDF 472.4 ‘ 500.0 589.2 § 637.8

 

304u|

"114.9

w2L 853w

 

' __ 687.2 "1

Figure 33. Teff forage quality (CP, ADF, NDF) from East Lansing in 2009. Samples were
collected every two weeks from planting date of June 4. Collected samples when teff was

~10 cm tall until the teff lodged.

63