Page 88 - Modelisation du devenir des pesticides...

Chapitre 3 Evaluation du modèle STICS version pesticide sur différents sites
expérimentaux

h i
¯ 2
(O i − O) −
P P (P i − O i ) 2
EF = (3.2.11)

2
P ¯
O i − O
equation 3.2.10 et 3.2.11 are extracted from (Loague et al., 1991) 70 where P i are
¯
the predicted values ; O i are the observed values ; n is the number of samples ; O
is the mean of observed data The bias is expressed in % and normalized by the
mean of observed data.

site Model Ref.
Vredepeel Pestla 62
PLM 28
PELMO 26
PRZM 63
GLEAMS 65
PESTRAS 66
WAVE 34
SIMULAT 67
LEACHP 33
MACRO 68
a
10models 69
Kerlavic AGRIFLUX 56
Thiverval- PRZM
Grignon VARLEACH 58
LEACHP

Table 3.4 – Pesticide fate model application at Vredepeel, Kerlavic and
Thiverval-Grignon and associated references. a) is a compilation of
the results obtained with 10 models at Vredepeel

The STICS model predictions were first compared to observed data and then
compared to previous application of several pesticide fate models on the three
datasets. The different model application and associates references are summarized
in Table 3.4.

3.3 Results and discussion

As one specificity of the model is to be able to simulate the plant growth and
its impact on the water budget, the results are first analyzed on the basic of the

76