Environmental Engineering Reference
In-Depth Information
Table 11.4
Consumption and subsistence farming of plants and animal products in France
(calculated from INSEE
1991
)
Agricultural population
Non agricultural population
Subsistence
farming
[%]
Subsistence
farming
[%]
Consumption
[kg/person/year]
Consumption
[kg/person/year]
Vegetables
115.8
67.9
93.0
25.7
Fruit
∗
47.9
28.0
38.5
12.0
Beef
18.1
32.3
15.1
3.3
Pork
24.15
27.5
19.9
3.5
Poultry
17.5
68.3
13.2
15.3
Eggs
$
165.5
58.7
138.0
16.9
Dairy products
121.7
18.7
106.7
2.3
∗
Excluding tropical fruit
$
Consumption expressed in unit per year
Uncertainties and variability related to transfer parameters are large, as described
in Section
11.3.3.2
. If a regression relation is used instead of empirical data, the
uncertainties linked to the definition of the equation have to be added.
In case of a linear regression (
y
=
ax
+
b
), like for the relation of Travis and Arms
(
1988
) (where log
B
T
=
y
and log
Kow
=
x
), the (1-
α
) confidence interval of log
B
T
for a new contaminant is given by:
⎡
⎣
⎤
⎦
S
2
(
x
∗
−¯
x
)
2
×
(
n
+
1)
x
∗
+
a
×
b
±
t
α
×
+
(11.48)
i
=
1
n
n
x
)
2
(
x
i
−¯
in which
x
∗
=
log Kow for a new contaminant
x
¯
=
average of the x
i
observed in the dataset used to define the regression
relationship
t
α
=
value of t-distribution with
n
-2 degrees of freedom, at
α
level
n
=
number of (
y
i
,
x
i
) in the dataset used to define the regression relationship
[-]
s
=
residual standard deviation
There are also uncertainties linked to the assessment of animal exposure,
because:
•
Some phenomena are not
taken into account
in models,
like ingestion of
earthworms by hens which are kept (partly) outdoors.
Search WWH ::
Custom Search