- Applied Animal Behaviour Science

Applied Animal Behaviour Science 139 (2012) 50–57
Contents lists available at SciVerse ScienceDirect
Applied Animal Behaviour Science
journal homepage: www.elsevier.com/locate/applanim
Comparison of methods to quantify the number of bites in calves
grazing winter oats with different sward heights
Laura B. Nadin a,b,∗ , Federico Sánchez Chopa a,b , Malcolm J. Gibb c , Júlio Kuhn da Trindade d ,
Glaucia Azevedo do Amaral d , Paulo C. de Faccio Carvalho d , Horacio L. Gonda a
a
Departamento de Producción Animal, Facultad de Ciencias Veterinarias, Universidad Nacional del Centro de la Provincia de Buenos Aires, Paraje Arroyo
Seco s/n. 7000, Tandil, Argentina
b
Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Argentina
c
Formerly at the Institute of Grassland and Environmental Research, North Wyke, Okehampton, Devon EX20 2SB, UK
d
Departamento de Plantas Forrageiras e Agrometeorologia, Faculdade de Agronomia, Universidade Federal do Rio Grande do Sul. Av. Bento Gonçalves 7712,
Agronomia, CEP 91501-970 - Caixa-Postal: 776, Porto Alegre, RS, Brazil
a r t i c l e
i n f o
Article history:
Accepted 7 March 2012
Available online 27 April 2012
Keywords:
Calves
Winter oats
Sward height
Number of bites
Acoustic recorder
IGER behaviour recorder
a b s t r a c t
The requirement to measure the constituents of ingestive behaviour in grazing ruminants,
such as the number and type of jaw movements, is essential for understanding the herbage
intake process. In this experiment, three methods of recording grazing behaviour were
compared: visual observation (VO; four trained observers), solid-state behaviour recorder
(SSBR) and acoustic recorder (AR), in order to quantify the number of bites taken by
Holstein–Friesian calves (138 ± 11 kg LW) grazing winter oat (Avena sativa) pastures with
different sward surface heights (SSH): tall (T; 52.4 ± 9.9 cm), medium (M; 26.5 ± 4.9 cm)
and short (S; 14.6 ± 3.4 cm). The number of bites was recorded on nine calves with VO,
SSBR and AR during 5-min grazing sessions on each sward height. In addition, the total
grazing jaw movements (GJM) and bites per GJM (PB) were calculated from the SSBR and
the AR methods. The experiment was conducted on 3 days, and on each day three animals
grazed on each of the three sward heights in a randomised sequence. Sound files from AR
were analysed visually and aurally (Sound Forge 9.0) and recordings by SSBR were analysed
using the dedicated software ‘Graze 8.0’. Data were analysed by three-way ANOVA. There
were no statistical differences (P > 0.05) in the number of bites counted by the observers.
The number of bites did not differ significantly between AR and VO, but were lower for
SSBR (P < 0.001). Mean values of number of bites were: 191, 153 and 192 for VO, SSBR
and AR, respectively. Comparison of recording analyses from AR and SSBR showed a small
difference (<5%; P < 0.05) in total GJM (388 and 407 for SSBR and AR, respectively) and a
larger difference (19%; P < 0.01) in PB (0.395 and 0.472 for SSBR and AR, respectively). The
main problem appeared to be in the identification of bites using SSBR, which led to an
underestimation of the number of bites and bites per GJM using that method under the
present grazing conditions. The AR method proved to be accurate for identifying biting jaw
movements.
© 2012 Elsevier B.V. All rights reserved.
Abbreviations: SSBR, solid state behaviour recorder; VO, visual observation; AR, acoustic recorder; SSH, sward surface height; GJM, total
grazing jaw movements; PB, bites per GJM.
∗ Corresponding author at: Departamento de Producción Animal, Facultad de Ciencias Veterinarias, Universidad Nacional del Centro de la
Provincia de Buenos Aires, Paraje Arroyo Seco s/n., 7000, Tandil, Argentina.
Tel.: +54 0249 4439850; fax: +54 0249 4439850.
E-mail address: [email protected] (L.B. Nadin).
0168-1591/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.applanim.2012.03.001
1. Introduction
The requirement to measure the constituents of ingestive behaviour by grazing ruminants, to either estimate
intake or explain the observed intake, has led to the
development of various automated devices (Beauchemin
L.B. Nadin et al. / Applied Animal Behaviour Science 139 (2012) 50–57
et al., 1989; Boland, 2005; Delagarde et al., 1998, 1999;
Desnoyers et al., 2009; Galli et al., 2006; Laca et al.,
1992; Rutter et al., 1997; Ungar and Rutter, 2006). Among
these automated devices, the IGER solid state behaviour
recorder (SSBR) has been widely used in many studies
(Gibb et al., 1998, 1999, 2002; Orr et al., 2005; Rutter
et al., 2002). This method allows the continuous recording of jaw movements over a 24-h period. Its associated
software ‘Graze 8.0’ (Rutter, 2000), utilises algorithms
formulated to analyse and discriminate different jaw
movements, namely bites and non-bites during grazing,
and rumination. However, these algorithms were developed and validated for interpreting recordings made with
cattle and sheep grazing temperate pastures (perennial
ryegrass and/or white clover) in which sward surface height (SSH) varied between approximately 3 and
25 cm.
In the Province of Buenos Aires, Argentina, swards
of Italian ryegrass (Lolium multiflorum) or oats (Avena
sativa) are frequently used to provide grazing for beef
and dairy cattle during the late autumn, winter and early
spring. In many cases, due to the high rate of growth,
by the time these pastures are grazed, sward height can
achieve 30 or 50 cm, resulting in a very different plant
morphology compared with the swards used in the development of ‘Graze’. As a consequence, a wider variety of
jaw movements may be performed in these taller swards.
In addition, these swards are generally utilised under
a rotational stocking system, with animals occupying a
paddock for a period of 1–7 days, resulting in a rapid
modification of the sward height and lamina:pseudostem
ratio by repeated defoliations over a short period and,
as a consequence, modification of grazing behaviour. In
previous studies with calves grazing on tall swards, we
have faced considerable difficulty trying to identify biting
jaw movements using the SSBR and its associated ‘Graze’
software.
More recently, methods of acoustic recording have
shown great potential for the identification of and differentiation between different grazing jaw movements (Galli
et al., 2006; Laca and Wallis De Vries, 2000) and provide an
alternative to the SSBR.
The present experiment was conducted with the objective of comparing methods (visual observation, SSBR and
acoustic recording) to quantify the number of bites in Holstein Friesian calves grazing winter oat (A. sativa) pastures
with three different sward surface heights (SSH). In addition, the effect of SSH on number of bites per grazing session
was also analysed.
2. Materials and methods
The experiment was conducted between 2 and 16 July
2008, at the campus of the Universidad Nacional del Centro
de la Provincia de Buenos Aires (37◦ 19 S, 59◦ 07 W), Tandil,
Argentina.
All procedures regarding the use of animals complied
with the requirements of the ethical guidelines published
by the International Society for Applied Ethology.
51
2.1. Animals
Nine,
6-month-old
Holstein–Friesian
calves
(138 ± 11 kg LW) were used. Two months prior to the
experiment the calves were provided with experience of
grazing oat swards, the close proximity of the observers
and the wearing of head collars. When not being used
for measurement, the calves were stocked on a common
paddock of winter oats with a daily forage allowance (kg
DM animal−1 ) equivalent to 0.05 of LW.
2.2. Sward preparation
The winter oats (A. sativa cv. Calén) were sown on 16
March 2008 at a seed rate of 120 kg ha−1 in an area treated
with glyphosate herbicide 4 weeks previously. The area of
the sward intended for use for the present study was kept
ungrazed until three different sward surface heights (SSH)
were created by cutting the sward. The SSHs were named
as: tall (T), medium (M) and short (S). The T sward was
left uncut and the M and S swards were cut to approximately 0.50 and 0.25 of SSH of the T sward, respectively,
using a reciprocating-cutter-bar mower, the day prior to
each recording day. After cutting, the cut plant material
was immediately removed. The required height of the cutter bar above ground level was maintained by using a taut
wire adjusted to the required height. The area of each
treatment sward for each day of observation, measured
approximately 6 m × 0.75 m. Before cutting, the SSH of the
whole area intended for use the next day was assessed by
making 60 height measurements using a sward stick similar to the design of the HFRO sward stick (Hill Farming
Research Organization, 1986), but in which the ‘contact’
window measured 15 mm × 35 mm.
Cutting the swards established a significant difference
in SSH between the three sward heights (F2,147 = 1187.6,
P < 0.001), and significantly affected the total herbage
masses (F2,18 = 155.1, P < 0.001), lamina (F2,18 = 202.4,
P < 0.001), pseudostem (F2,18 = 22.6, P < 0.001), dead
(F2,18 = 25.2, P < 0.001) and the L:P mass ratio (F2,18 = 112.0,
P < 0.001) (Table 1).
Following preparation of the three sward heights, the
plots corresponding to each SSH were subdivided into
three individual sub-plots with electric fencing, to provide
grazing for one calf. Each sub-plot measured 2 m × 2.75 m
Table 1
Mean sward surface height (SSH, cm), total, lamina (L), pseudostem (P)
and dead herbage mass (g DM m−2 ) and square root of L:P mass ratio data
for an Avena sativa sward with three different sward surface heights (SSH:
T, tall; M, medium; S, short) prior to grazing.
SSH
T
M
S
SEMa
P
Height
52.4 a
26.5 b
14.6 c
0.56
<0.001
Herbage DM mass
L:P
Total
Lamina
P
Dead
600.9 a
303.5 b
143.8 c
18.62
<0.001
358.8 a
132.3 b
24.8 c
11.98
<0.001
179.7 a
134.7 b
96.0 c
8.80
<0.001
62.4 a
36.6 b
22.9 c
3.99
<0.001
1.41 a
0.99 b
0.47 c
0.071
<0.001
Means not sharing common letters (a, b) within a column differ by P < 0.05.
a
SEM: standard error of the mean.
52
L.B. Nadin et al. / Applied Animal Behaviour Science 139 (2012) 50–57
Fig. 1. Two synchronous recordings representing a total of 15 jaw movements by their amplitudes (vertical axis) from signals recorded by the acoustic
recorder (AR, A) and by the solid state behaviour recorder (SSBR, B), over a period of 10 s (horizontal axis) from a calf grazing on Avena sativa with a sward
surface height of 32 ± 2.3 cm.
encompassing a 2 m × 0.75 m area of sward and a 2 m × 2 m
area of bare soil.
2.3. Experimental procedure
The experiment comprised 3 days of recordings in
which the number of bites taken by calves while grazing
on winter oats with different SSHs was registered during
5-min grazing sessions. Three methodologies were used to
measure the number of bites taken by the calves, as follows:
visual and aural recognition of bites counted by four trained
observers (visual observation, VO), recording of jaw movements using the IGER behaviour recorder (SSBR; Rutter
et al., 1997) and acoustic recording (AR; Galli et al., 2006;
Laca and Wallis De Vries, 2000).
Each recording day, three of the nine calves to be used
for study were removed from the common grazing paddock
to a handling area at 11:00 h. Beginning at 15:00 h, one calf
at a time was fitted with a head collar to which the SSBR
and the AR devices were attached. The AR apparatus incorporated a microphone held in place on the forehead of the
calf and connected to a digital recorder. The calf was then
walked to the experimental subplots and allowed to graze
for 5 min on each of the three sward SHHs in turn, following
a random sequence. During the grazing sessions, the number of bites was counted simultaneously by the observers
(VO) and recorded by the SSBR and AR methods. After completing the observations and recordings on the three sward
heights, the calf was returned to the handling area, where
both devices were removed. The same procedure was then
carried out with the other two calves in turn.
The allocation of the animals to the recording days, the
order in which the animals were studied within the recording days, as well as the order in which each animal grazed
the different sward heights were randomised.
To ensure that the time windows over which the results
of the three methodologies were compared were the same,
the following procedure was adopted. When a calf entered
an individual recording plot as soon as it started to graze,
the precise time was noted according to a digital watch, an
audible signal (a shout) was made and the visual observations (VO) commenced immediately. After 5 min the
procedure was terminated with a second audible signal.
Recordings were subsequently downloaded to a computer.
The audible signal picked up and recorded by the
microphone-digital recorder (AR) on the calf was used
L.B. Nadin et al. / Applied Animal Behaviour Science 139 (2012) 50–57
to identify the 5-min recording window on the acoustic
recording. The internal clock of the SSBR had previously
been synchronized with the digital watch used by the visual
observer who made the audible signal. However, to ensure
that the 5-min time windows, over which jaw movements
recorded by the SSBR and AR were precisely the same, the
wave patterns recorded by the two methods were aligned
on a screen and the time window identified from the AR
recording was then marked on the SSBR recording using
the facility in Graze. An example of the plots of AR and SSBR
is shown in Fig. 1.
2.4. Swards measurements
Before being grazed, 20 SSH measurements were made
across each of the T, M and S sub-plots. In addition, three
samples of herbage were cut to ground level within three
quadrats, measuring of 20 cm × 30 cm on each sub-plot,
using scissors. The herbage cut from each quadrat was
separated manually into lamina, pseudostem and dead
fractions, before being dried at 65 ◦ C until reaching a constant weight. The weights were used to calculate the total
herbage mass (g m−2 ), as well as the masses of the different
morphological components, lamina, pseudostem and dead
material.
2.5. Measurement of number of bites
2.5.1. Visual observations
Four trained observers, standing in close proximity to
the animals (within 3 m), counted the numbers of bites,
based on visual and aural cues, during all the grazing sessions. In addition, a video recording was made (SP mode,
real time, 25 frames s−1 ) by a fifth operator using a handheld digital camera (Sony, digital video camera recorder,
Digital 8, DCR-TRV530, Japan), to provide a record, if
required, for the interpretation of any of the methods used
to determine the number of bites.
2.5.2. Analysis of SSBR
The recordings from the three calves used on each day
on each of the three different SSHs were made using a single SSBR. After fitting the recorder to a calf, three recordings
were made, one on each of the three sward structures. The
recorder was then fitted to the next calf and the procedure repeated. After the final period of measurement was
completed with the third calf, the recorder was removed
and switched off. The memory card was then removed
and the single file containing the jaw movement data from
the three animals was analysed using ‘Graze 8.0’. The nine
individual, 5-min recordings corresponding to the precise
periods of visual observation were marked using the facility
in ‘Graze 8.0’, as described previously. The total grazing jaw
movements (GJM) and bites were identified by an operator
experienced in using the ‘Graze 8.0’ software. This software allowed the operator to set the parameters in order to
obtain what were considered to be the most reliable results
regarding identification of individual jaw movements and
discrimination between biting and non-biting grazing jaw
movements. This examination and analysis was conducted
53
without the operator knowing the number of bites counted
by the observers.
2.5.3. Analysis of AR
The sounds produced by the range of different jaw
movements were recorded using a microphone (Omni
directional microphone, ECM-F8, Sony, Japan) connected to
a digital recorder (IC recorder, ICD P320, Sony, Japan). After
completing the observations, the sound files were downloaded to a computer in a MP3 format. This allowed the
sound recordings to be interpreted by an operator examining both the digitised sound-wave patterns and aurally
by listening to the recordings (Milone et al., 2009; Sound
Forge 9.0, Sony Creative Software Inc., USA). The numbers
of bites and the total GJM were both registered. As for the
SSBR analysis, this examination was conducted without
the operator knowing the number of bites counted by the
observers.
2.6. Statistical analyses
Variables of the sward and the number of bites counted
by the observers were analysed by ANOVA using the
mixed model procedure (SAS Institute Inc., 2009). Classes
included into the model were sward height (fixed effect)
and day of measurement (random effect), and sward
height, day and observer (fixed effect); for sward variables
and bites counted by the observers, respectively. Because
the variable lamina:pseudostem mass ratio was not normally distributed, data were transformed to root square
before statistical analysis.
Before analysis, normal distribution of number of bites
(P = 0.38) and total grazing jaw movements (GJM; P = 0.31)
was tested by the Shapiro–Wilks test. The numbers of bites,
total GJM and bites per GJM data were analysed according
to a three-way ANOVA. Classes included into the model
were method and sward height (fixed effects) and the day
of measurement (random effect) (mixed model procedure,
SAS Institute Inc., 2009). Animals (n = 9) were treated as
independent replicates. Comparison of individual means
was performed by the lsmeans/DIFF statement of SAS (SAS
Institute Inc., 2009).
3. Results
3.1. Number of bites
Prior to the start of the experiment, care was taken to
accustom the animals to the close proximity of operators
in order to avoid disruption of their grazing activity. Thus,
during the 5-min recordings, despite the subject calf being
enclosed within a small sub-plot of 5.5 m2 and closely surrounded by four observers and a fifth operator with a video
recording camera, there were no signs of disturbance and
animals grazed continuously throughout observation sessions.
There were no statistical differences among the
numbers of bites counted by the four observers neither
across (F3, 72 = 0.59, P = 0.51) nor within (F3, 24 = 0.07,
P = 0.97; F3, 24 = 0.40, P = 0.75; F3, 24 = 0.30, P = 0.82,
for T, M and S, respectively) the SSHs (Table 2). The
54
L.B. Nadin et al. / Applied Animal Behaviour Science 139 (2012) 50–57
Table 2
Mean number of bites counted by four observers (L, G, J and H) in calves
grazing an Avena sativa sward with three different sward surface heights
(SSH: T, tall; M, medium; S, short) for periods of 5 min.
SSH
Observer
L
T
177
212
M
S
171
Mean
187
SEMa
Significance of effects (P)
Observer
SHH
Observer × SSH
Mean
G
J
H
179
216
172
189
181
221
176
193
184
224
181
196
180 a
218 b
175 a
7.73
0.510
<0.001
0.999
Means not sharing common letters (a, b) within a column differ by P < 0.05.
a
SEM: standard error of the mean.
interactions observer × SSH, observer × day, SSH × day and
SSH × observer × day, were not significant (F6, 72 = 0.02,
P > 0.99; F6, 72 = 0.26, P = 0.95; F4, 72 = 0.63, P = 0.64;
F12, 72 = 0.09, P > 0.99, respectively).
While there were no differences between the number
of bites registered by the VO and the AR methods, bites
counted with the SSBR method were lower than those from
VO and AR (F2, 64 = 23.1, P < 0.001) (Table 3). The interaction
SSH x method was not significant (F4, 64 = 1.64, P = 0.176,
Table 3).
Wave patterns for the jaw movements recorded by the
SSBR method showed an evident lack of clear sub-peaks
associated with the main peaks of biting jaw movements
(Fig. 2A).
3.2. Total grazing jaw movements and number of bites
per GJM
Total GJM and number of bites per GJM (PB) from
the two automatic methods, SSBR and AR, are presented
in Table 4. Analysis of variance showed a significant
effect due to recording methods on the total number of
GJM (F1, 40 = 4.2, P = 0.048), as well as on PB (F1, 40 = 12.9,
P = 0.001). For total GJM and PB, the interactions between
Table 3
Number of bites determined using visual observations (VO), solid-state
behaviour recorder (SSBR) and acoustic recorder (AR) in calves grazing
an Avena sativa sward with three different sward surface heights (SSH: T,
tall; M, medium; S, short) for periods of 5 min.
SSH
Method
VO
T
180
218
M
S
175
Mean
191 a
SEMa
Significance of effects (P)
Method
SSH
Method × SSH
Mean
SSBR
AR
134
165
159
153 b
7.11
183
213
180
192 a
166 y
199 z
171 y
<0.001
<0.001
0.176
Means not sharing common letters (a, b) within a row differ by P < 0.05.
Means not sharing common letters (z, y) within a column differ by
P < 0.05.
a
SEM: standard error of the mean.
method and SSH were not significant (Table 4; F2, 40 = 1.45,
P = 0.246 and F2, 40 = 3.12, P = 0.055; for total GJM and PB,
respectively).
The total GJM were higher on M than on T, and on T
than on S swards (F2, 40 = 8.7, P = 0.001). The PB ratios did
not differ among SSHs (F2, 40 = 2.9, P = 0.067, Table 4).
4. Discussion
Periods of fasting have been shown to significantly
increase bite rate (Gregorini et al., 2007; Patterson et al.,
1998). Nevertheless, in the present study the decision was
taken to fast the animals prior to undertaking the measurements in order to ensure a 15-min period of intensive
grazing for comparison of the different methodologies on
different sward heights.
The VO method has been useful in demonstrating the
effect of sward height on bite rate (Chilibroste et al., 2000;
Erlinger et al., 1990; McGilloway et al., 1999; Mezzalira,
2009; Realini et al., 1999), as well as a method of reference
in the development of automatic devices (Delagarde et al.,
1999; Rutter et al., 1997) and software for data analysis
(Clapham et al., 2011; Rutter, 2000). Although it could not
be assumed that VO gave an exact count of the number of
bites, in the conditions under which the VO method was
used in the present study, we are confident that it represented a reliable estimation of the actual number of bites
against which the automated methods could be compared.
The close agreement between the number of bites registered by VO and AR methods (Table 3), showed that the AR
method can be a reliable technique for counting the number of bites. The advantage of both the VO and AR methods
over the SSBR method is the ability of the operator to utilize
the diagnostic aural cue for a bite as well as a simply visual
one.
In contrast, the number of jaw movements identified as
bites from the SSBR was significantly lower compared with
the VO method. Also, the waveforms of the SSBR recordings were found to differ from those previously recorded
with dairy cows and heifers grazing perennial ryegrass
swards. The SSBR and its associated software ‘Graze’ were
developed and validated for interpreting recordings made
with cattle and sheep grazing temperate pastures (perennial ryegrass and/or white clover) in which sward heights
varied between approximately 3 and 25 cm. As shown in
Fig. 2, the wave patterns for the jaw movements recorded
on the temperate perennial ryegrass swards (Fig. 2A; from
a study described by Forbes et al., 2007, M.J. Gibb, pers.
comm.) differ slightly in shape from those of the present
study (Fig. 2B), but particularly in the presence of clear secondary sub-peaks associated with a main peak (one of the
criteria used by the algorithms of the ‘Graze’ software to
identify a jaw movement as a bite). The lack of such clear
sub-peaks in recordings from the present experiment made
distinction of bites from amongst the various jaw movements extremely difficult. The differences in waveforms
could have been due to the sward height and structure, pasture species morphology, the age and grazing experience
of the calves, and therefore, the way the animals behaved
when confronted with these swards.
L.B. Nadin et al. / Applied Animal Behaviour Science 139 (2012) 50–57
55
Fig. 2. Examples of SSBR recordings representing bites and non bites (manipulative and ingestive mastications during grazing) in a calf grazing winter oats
(Avena sativa) between 9 and 14 cm sward surface height (A) and in a dairy heifer grazing perennial ryegrass between 9 and 17 cm sward surface height
(B; from an experiment described by Forbes et al., 2007). In this example, the black box shows the main peak associated with a secondary peak (bite) being
more clearly discernable in (B) than in (A).
Comparing the two automated recording systems, in
contrast with previous studies by Champion et al. (1998)
and Ungar and Rutter (2006), numbers of bites determined
by SSBR were consistently lower than those determined by
AR. However, as stated previously, those studies were conducted with dairy cows grazing relatively short pastures.
Some explanation for the relatively large discrepancies
between the number of bites determined by SSBR and AR
methods may be found by examination of the total GJM
and PB values (Table 4). While a small difference (<5%) in
total GJM was observed between SSBR and AR, the difference in PB values between the automatic methods was 19%.
The difficulty, therefore, appears not to lie in the ability
of the operator using ‘Graze’ to identify GJM, but to distinguish between biting and non-biting GJM (movements
that are not bites and therefore include movements which
are masticatory or manipulative in function (Gibb et al.,
1999)).
Though well proven under temperate pasture conditions, the SSBR was clearly inferior to the AR system for
determining the number of bites taken by calves grazing
the winter oat swards offered in this experiment. Further
potential advantages of the AR over SSBR are the accurate
discrimination between compound jaw movements associated with the gathering or biting the herbage and chewing
the material already in the mouth (Laca et al., 1994); manipulative jaw movements and bites (Laca and Wallis De Vries,
2000); the determination of plant parts or species being
eaten; as well as the possible direct determination of dry
matter intake (Galli et al., 2006).
Table 4
Total number of grazing jaw movements (GJM) and the number of bites per GJM (PB) identified from recordings made using solid-state behaviour recorders
(SSBR) or acoustic recorder (AR) in calves grazing an Avena sativa sward with three different sward surface heights (SSH: T, tall; M, medium; S, short) for
periods of 5 min.
SSH
T
M
S
Mean
SEMa
Significance of effects (P)
Method
SSH
Method × SSH
Total GJM
PB
SSBR
AR
Mean
SSBR
AR
Mean
392
418
354
388
402
424
395
407
397 a
421 c
374 b
0.343
0.394
0.448
0.395
0.456
0.502
0.457
0.472
0.376
0.448
0.453
9.05
0.016
0.048
0.001
0.067
0.001
0.246
Means not sharing common letters (a, b, c) within a column differ by P < 0.05.
a
SEM: standard error of the mean.
0.055
56
L.B. Nadin et al. / Applied Animal Behaviour Science 139 (2012) 50–57
With regard to the effect of SSH, number of bites
(Tables 2 and 3) and total GJM and PB ratio (AR method,
Table 4) were higher in M than in T and S swards. This
may have been due to the occurrence of a higher proportion of more manipulative, non-biting jaw movements
during grazing in T and S than M, as suggested by the
PB values from the AR method (Table 4). In the present
experiment, where calves were offered a tall grass pasture, evidence from visual observation would suggest that
animals in the T swards performed more head movements while trying to sever a bite. This could have led to
an increase in handling time (Laca et al., 1994). Alternatively, in the S swards characterised by having the lowest
lamina:pseudostam ratio (Table 1), evidence from visual
observation would suggest that pseudostems could have
represented a physical barrier making the act of severance of a bite more difficult to perform (Benvenutti et al.,
2009).
5. Conclusions
Under the experimental conditions of the present study,
the close agreement found between the number of bites
estimated by VO and AR methods supports for the use
of the AR method as an accurate method for identifying
the biting jaw movements in grazing animals. A recent
development of software for identifying and quantifying
ingestive events – biting and non-biting jaw movements
– from sound files (Clapham et al., 2011) appears to
have overruled the limitation of the AR method. When
used in young calves offered tall grass swards, as presented in this study, the SSBR failed to allow reliable
discrimination between biting and non-biting grazing jaw
movements.
Acknowledgements
We thank Claudia Marinelli and Rosana Cepeda for
statistical consultations; and Guillermo Milano and Laura
Moreno for field assistance.
This research was financially supported by the Universidad Nacional del Centro de la Provincia de Buenos Aires,
Argentina, and the Consejo Nacional de Investigaciones
Científicas y Tecnológicas (CONICET, Argentina).
References
Beauchemin, K.A., Zelin, S., Genner, D., Buchanan-Smith, J.G., 1989. An
automatic system for quantification of eating and ruminating activities of dairy cattle. J. Dairy Sci. 72, 2746–2759.
Benvenutti, M.A., Gordon, I.J., Poppi, D.P., Crowther, R., Spinks, W., Moreno,
F.C., 2009. The horizontal barrier effect of stems on the foraging
behaviour of cattle grazing five tropical grasses. Livest. Sci. 126,
229–238.
Boland, H.T., 2005. Grazing behaviour of beef steers grazing endophyteinfected, endophyte-free and novel endophyte infected tall fescue,
and lakota prairie grass. Master of Science Thesis, Blacksburg, Virginia,
USA.
Champion, R.A., Rutter, S.M., Delagarde, R., 1998. Recent developments
with the IGER behaviour recorder. In: Gibb, M.J. (Ed.), Techniques for
Investigating Intake and Ingestive Behaviour by Farm Animals. Proceedings of the IXth European Intake Workshop. IGER North Wyke,
UK, pp. 31–34.
Chilibroste, P., Tamminga, S., Boer, H., Gibb, M.J., den Dikken, G., 2000.
Duration of regrowth of ryegrass (Lolium perenne). Effects on grazing
behavior, intake, rumen fill, and fermentation of lactating dairy cows.
J. Dairy Sci. 83, 984–995.
Clapham, W.M., Fedders, J.M., Beeman, K., Neel, J.P.S., 2011. Acoustic monitoring system to quantify ingestive behavior of free-grazing cattle.
Comput. Electron. Agric. 76, 96–104.
Delagarde, R., Caudal, J.P., Peyraud, J.L., 1998. Development of an automatic
bitemeter for grazing cattle. Ann. Zootech. 48, 329–339.
Delagarde, R., Peyraud, J.L., Delaby, L., 1999. Influence of carbohydrate or
protein supplementation on intake, behaviour and digestion in dairy
cows strip-grazing low-nitrogen fertilized perennial ryegrass. Ann.
Zootech. 48, 81–96.
Desnoyers, M., Béchet, G., Duvaux-Ponter, C., Morand-Fehr., P., GigerReverdin., S., 2009. Comparison of video recording and a portable
electronic device for measuring the feeding behaviour of individually
housed dairy goats. Small Rum. Res. 83, 58–63.
Erlinger, L.L., Tolleson, D.R., Brown, C.J., 1990. Comparison of bite size,
biting rate and grazing time of beef heifers from herds distinguished by mature size and rate of maturity. J. Anim. Sci. 68,
3578–3587.
Forbes, A.B., Huckle, C.A., Gibb, M.J., 2007. Evaluation of the effect of
eprinomectin in young dairy heifers sub-clinically infected with gastrointestinal nematodes on grazing behaviour and diet selection. Vet.
Parasitol. 150, 321–332.
Galli, J.R., Cangiano, C.A., Demment, M.W., Laca, E.A., 2006. Acoustic monitoring of chewing and intake of fresh and dry forages in steers. Anim.
Feed Sci. Technol. 128, 14–30.
Gibb, M.J., Huckle, C.A., Nuthall, R., 1998. Effect of time of day on
grazing behaviour by lactating dairy cows. Grass Forage Sci. 53,
41–46.
Gibb, M.J., Huckle, C.A., Nuthall, R., 2002. Effects of level of concentrate
supplementation on grazing behaviour and performance by lactating
dairy cows grazing continuously stocked grass swards. Anim. Sci. 74,
319–335.
Gibb, M.J., Huckle, C.A., Nuthall, R., Rook, A.J., 1999. The effect of physiological state lactating or dry/and sward surface height on grazing
behaviour and intake by dairy cows. Appl. Anim. Behav. Sci. 63,
269–287.
Gregorini, P., Eirin, M., Wade, M.H., Refi, R., Ursino, M., Ansin, O.E., Masino,
C., Agnelli, L., Wakita, K., Gunter, S., 2007. The effects of a morning
fasting on the evening grazing behavior and performance of stripgrazed beef heifers. Prof. Anim. Sci. 2, 642–648.
Hill Farming Research Organisation, 1986. Biennial report 1984–1985.
HFRO, Edinburgh, UK, pp. 29–30.
Society
for
Applied
Ethology,
2002.
Ethical
International
Guidelines,
Retrieved
from
http://www.appliedethology.org/ethicalguidelines.htm.
Laca, E.A., Ungar, E.D., Demment, M.W., 1994. Mechanisms of handling
time and intake rate of a large mammalian grazer. Appl. Anim. Behav.
Sci. 39, 3–19.
Laca, E.A., Ungar, E.D., Seligman, N.G., Ramey, M.R.N., Demment, M.W.,
1992. An integrated methodology for studying short-term grazing
behaviour of cattle. Grass Forage Sci. 47, 81–90.
Laca, E.A., Wallis De Vries, M.F., 2000. Acoustic measurement of intake and
grazing behaviour of cattle. Grass Forage Sci. 55, 97–104.
McGilloway, D.A., Cushnahan, A., Laidlaw, A.S., Mayne, C.S., Kilpatrick, D.J.,
1999. The relationship between level of sward height reduction in a
rotationally grazed sward and short-term intake rates of dairy cows.
Grass Forage Sci. 54, 116–126.
Mezzalira, J.C., 2009. O manejo do pastejo em ambientes pastoris heterogêneos: comportamento ingestivo e produção animal em distintas
ofertas de forragem [dissertação]. Universidade Federal do Rio Grande
do Sul, Porto Alegre, RS, Brazil, 159 pp.
Milone, D.H., Rufiner, H.L., Galli, J.R., Laca, E.A., Cangiano, C.A.,
2009. Computational method for segmentation and classification of ingestive sounds in sheep. Comput. Electron. Agric. 65,
228–237.
Orr, R.J., Cook, J.E., Young, K.L., Champion, R.A., Rutter, S.M., 2005. Intake
characteristics of perennial ryegrass varieties when grazed by yearling
beef cattle under rotational grazing management. Grass Forage Sci. 60,
157–167.
Patterson, D.M., McGilloway, A., Cushnahan, A., Mayne, C.S., Laidlaw, A.S.,
1998. Effect of duration of fasting period on short term intake rates of
lactating dairy cows. Anim. Sci. 66, 299–305.
Realini, C.E., Hodgson, J., Morris, S.T., Purchas, R.W., 1999. Effect of sward
surface height on herbage intake and performance of finishing beef
cattle. N. Z. J. Agric. Res. 42, 155–164.
Rutter, S.M., 2000. Graze: a program to analyse recordings of the
jaw movements of ruminants. Behav. Res. Meth. Inst. Comp. 32,
86–92.
L.B. Nadin et al. / Applied Animal Behaviour Science 139 (2012) 50–57
Rutter, S.M., Champion, R.A., Penning, P.D., 1997. An automatic system
to record foraging behaviour in free-ranging ruminants. Appl. Anim.
Behav. Sci. 54, 185–195.
Rutter, S.M., Orr, R.J., Penning, P.D., Yarrow, N.H., Champion, R.A., 2002.
Ingestive behaviour of heifers grazing monocultures of ryegrass or
white clover. Appl. Anim. Behav. Sci. 76, 1–9.
57
SAS Institute Inc., 2009. SAS/STAT® 9.2 User’s Guide, second edition. SAS
Institute Inc., Cary, NC.
Ungar, E.D., Rutter, S.M., 2006. Classifying cattle jaw movements: comparing IGER behaviour recorder and acoustic techniques. Appl. Anim.
Behav. Sci. 98, 11–27.