Agriculture Reference
In-Depth Information
strongly dependent on
Ved
, thus on depos-
ition of protein and lipid. The work of Han-
cock
et al
. (1995) followed by that of Gous
et al
. (1999) and Sakomura
et al
. (2011)
showed that body composition of broilers
for a given body weight is comparable from
one genotype to another because of a very
high pressure of genetic selection. Based on
the data of Gous
et al
. (1999), we used Eqn
9.3 to estimate the value of
Ved
. The result
of linear regression (
R
² = 0.998) of
Ved
ac-
cording to the body weight (kilogrammes)
power 0.6 is described by Eqn 9.6:
Ved
= 1.56 + 0.63
BW
0.6
represent the evolution of
EPA
and modu-
late it according to the situations.
EPA
can
be strongly modified by nutritional or envir-
onmental conditions of production. For ex-
ample, physical characteristics of feed (Savory,
1974) and temperature strongly influence
the physical activity of chickens. INAVI has
the objective of representing the largest pos-
sible number of production systems and
conditions. It seems relevant to separate the
part of
MEI
used for
PA
from the part used
for maintenance to improve the adaptability
of the model.
(9.6)
Modelling of the level and the energy
cost of the physical activity (
EPA
)
Nevertheless, as fattening, feathering and
protein deposition potential are subject to
genetic variations, a factor (
Feg
) was intro-
duced into this equation to allow a modifi-
cation of
Ved
by the user for a genotype
with a very different body composition:
Ved
=
Feg
× (1.56 + 0.63
BW
0.6
)
In INAVI, the activity of the animal is repre-
sented as the percentage of time during
which the animal is standing up. This meas-
ure is made by scan-sampling (Picard
et al
.,
1999) at the beginning of a period. It defines
the initial activity (Initial
PA
) of animals. In
INAVI, we consider that the physical activ-
ity level (
PAL
in %, i.e. the percentage of
time of activity) decreases linearly with
time from initial
PA
, using a constant, the
activity factor (
AF
, %) (Eqn 9.8). The obser-
vation of fast-growing broilers in experi-
mental pens from
4
to
6
weeks of age sup-
ports this view
(Fig. 9.2)
.
(9.7)
Importance of physical activity
in energy partitioning
Physical activity (
PA
) can represent from
7% to 15% of the
MEI
in broilers (Wenk and
Van Es, 1980). Other behaviours besides
movement can be costly from an energetic
point of view, such as engaging in social
behaviour, eating or perching. Notably, the
slow-growing chickens are known to be more
active than fast-growing chickens (Bizeray
et al
., 2000; Bokkers and Koene, 2003).
PA
represents a cost for the animal but
it can also have positive effects. An increase
in
PA
of broilers induced by high ventila-
tion rate from
6
to 41 days improves growth
without changing feed conversion. Increases
in breast meat yield and feed efficiency, and
a decrease in fattening were also measured
in active animals compared to less active
ones (Lei and Van Beek, 1997). A positive
correlation between feed conversion and
the 'standing' behaviour was clearly dem-
onstrated by Skinner-Noble
et al
. (2003).
In most models, energy related to the
physical activity (
EPA
) is included in main-
tenance requirements, so it is impossible to
PAL
= Initial
PAL
−
AF
(9.8)
The main challenge in the modelling of
PA
is to transform the level of activity (
PAL
)
into energy. Baker and Gleeson (1999) con-
sidered that the heavier the animal is, the
higher the energy cost of
PA
. Therefore, to
model this approach in a simple way, we con-
sider the energy cost of
PA
(
EPA
, kcal) as a func-
tion of
PAL
and
BW
, using an activity unit (
AU
,
kcal/%
PAL
/g
BW
) that represents the energy
cost of 1% of
PAL
per gramme of body weight.
EPA
=
PAL
×
AU
×
BW
(9.9)
To estimate
AU
, we used the data from an ex-
periment carried out in the Poultry Research
Unit (INRA Nouzilly, France) and the data
used for the study of
Ed
and
Ved
(see above).
The initial
PA
was 30% for standard broilers
of 28 days with an
AF
of 0.8%/day. According
