Page 44 - Contributed Paper Session (CPS) - Volume 5

CPS795 Nestor B.
To analyze the evolution of the daily precipitation, we use series of five years data,
ten years data, and twenty year data. From the fitting of the model to these data with
each stations we have the following results:
 This impact is the cause of the reduction of the amount of precipitation
proved by the decrease of the shape parameters, and the poor quality of the
rainy season proved by the fact that the shape parameters (which decrease)
are less than the parameters of scales (which increase).
 This poor rainy season quality is due to the very small amount of rain on som
rainy days and the high amount of rainfall observed for a low number of rainy
days. This is the cause of the floods and the destruction of crops and certain
infrastructures.
Station Breaking Good rainy season Bad rainy season
Cotonou 1972 1952-1972 1973-2007
Bohicon - 1940-1960, 1980-2007 1960-1980
Save` 1973 1921-1973 1974-2007
Parakou 1963 1921-1963 1964-2007
Kandi 1963 1921-1963 1964-2007
Natitingou 1974 1921-1974 1975-1921

Table 2: Results of estimations

Périodes de dix ans Périodes de 10 ans
Figure 1: Shape and scale evolution in Kandi and Parakou
Graphique de détection de rupture à Kandi Graphique de détection de rupture à Parakou

rapports de vraisemblance rapports de vraisemblance

Années Années

Figure 2: Breaking detection in Kandi and Parakou

33 | I S I W S C 2 0 1 9