Agriculture Reference
In-Depth Information
ME
id
AA
Maintenance
activity
Uterus
Milk
Body
lipid
Body
protein
Backfat thickness
Body weight
Fig. 4.1.
Description of nutrient utilization in the InraPorc sow model. (From Dourmad
et al
., 2008.)
of ME according to the NE system proposed
by Noblet
et al
. (1994). The supply of amino
acids is considered as standardized ileal di-
gestible (SID) amino acid (INRA-AFZ, 2004),
implying that the indigestible fraction and
specific endogenous amino acid losses are
combined together. The supply of phosphorus
is considered as apparent digestible phos-
phorus (INRA-AFZ, 2004), but the supply of
calcium is considered in terms of total calcium.
The different equations describing the
utilization of nutrients and energy by gestat-
ing and lactating sows were derived from
the literature and used to build a computer-
ized simulator (Dourmad
et al
., 2008). This
simulator determines on a daily basis the
flow of nutrients and energy from the feed to
storage in the body, excretion or dissipation.
Other functionalities were added to the
simulator so that it can be used as a decision
('sow profile') is used to describe the animal's
characteristics and adjust some model param-
eters to account for variation in genotype and
performance. Three other modules are used
to describe the types of feeds used ('feed se-
quence plan'), the quantity of feed consumed
('feed rationing plan') and the housing condi-
tions ('housing plan'). The sow module is
connected to the 'feed' modules, which can
be used to calculate dietary nutrients from
feed ingredients using the INRA-AFZ (2004)
database. When defining the sow profile, a
calibration procedure is used to adjust some
model parameters for each specific sow
genotype/phenotype in relation to observed
traits in a reference situation. This calibration
is based on an automated optimization pro-
cedure that minimizes the difference be-
tween observed and predicted performances.
The model can then be used to deter-
mine the nutritional requirements according
to a classical factorial approach, or to predict
performance and analyse nutrient utiliza-
tion, including nutrient excretion, through
simulations. In the current version of the
software, reproductive performance data (i.e.
litter size, piglet weight, milk production)
are considered as user inputs and are there-
fore not sensitive to nutrient supply.
Factorial Calculation of
Sow Requirements
As an example of the use of InraPorc, the
energy, amino acid and phosphorus require-
ments of sows from a herd weaning
25
pig-
lets per sow per year, with, respectively,
12.5 and 10.8 piglets born alive and weaned
per litter have been calculated (
Table 4.1)
.
The daily energy requirement during gesta-
tion increases from parity
1
to parity
3
and
remains constant thereafter. Conversely, the
amino acid requirement (lysine), expressed
per day or per kilogramme of feed, decreases
