Geoscience Reference
In-Depth Information
-
P is the average of the square roots of the monthly rainfall (the square of
P gives the monthly rainfall median, Pmed);
-
σ is the standard deviation of the square roots of the monthly rainfall.
The adjustments were applied to every rain gauge that formed part of the rain
gauge network. This network held data that provided information on observations
that had been measured for a period of at least 20 years during the period from
September 1965 to August 1995. Depending on the month being studied, the
number of stations used to take rainfall measurements varied from 499 to 535.
Distribution of rainfall and average rainfall are dependent on one another. These
two parameters are closely linked so that any differences that exist between the two
can be explained by uncertainties that are associated with the estimations of the
values of
σ . Figure 7.7 shows that it is possible to obtain
σ from the
following equation:
P
σ
0.2188
P
+
1.5852 (1
e
0.5489
)
P
The variable U(P) can, therefore, be considered as a random reduced centered
Gaussian variable:
PP
U(P)
=
P 0.5489
0.2188
P
+
1.5852 (1
e
)
7.2.3.2. Mapping the averages
For any given month the square root of rainfall is a random Gaussian variable
with an average of P . Only one value is known for each rain gauge and some
doubt is associated with this value. The doubt arises because the values of rain
gauges vary depending on how long they have been in active use to measure
rainfall. This parameter is non-stationary, increases with altitude and decreases with
distance from the sea. As the square of P represents the median values of rainfall,
these values will be the first to be represented on a map.
The median values of monthly rainfall are associated with the following factors:
the altitude at which the rain gauge is located (referred to as Z , measured in meters);
the longitude, which is measured by the Lambert x in kilometers; and distance from
the sea (referred to as d and measured in kilometers). Distance from the sea has an
influential role on the median values of monthly rainfall for the first few kilometers
and for this reason we have decided to use distance from the sea as an exponential
decline, with the parameter k being a regional optimum for a particular month. The
parameter k refers to the speed at which the effect of distance from the sea
decreases. The regression used to represent the median values is as follows:
 
Search WWH ::




Custom Search