In order to carry out the analysis of the rainfall
over the country as a whole has been expressed as
a linear combination of orthogonal functions. Let
P be a n  m matrix of monsoon rainfall over m
stations and a series of n years. In the present
study, n ˆ 41 years, and m ˆ 68 stations. Let P
represents the time and space variability of
rainfall. It has been defined that
P ˆ Q F;
(1)
where the matrix Q represents the time variation
and F represent the space variation of monsoon
rainfall. The element of the P matrix is given by
mX
qr;k …t† Â fks …x; y†;…(2)
Prs …x; y; t† ˆ
k ˆ1
where the element qrk …t† represents the time, and
fks …x; y† represents the space, respectively. The
matrix F is orthogonal matrix, hence the transpose
and product of this matrix should be represented
as a unique identity matrix. It has define that
F FH ˆ I;
(3)
where F H is the transpose of F and I is the identity
matrix. It has been further stated that F and Q
matrix drive from the matrix P after define
the matrix S, where
P H P ˆ S:
This matrix S is an square matrix and PH is the
transpose at the matrix P hence from the above
equation it has been concluded that
F H S F ˆ Q QH ˆ D;
(4)
where D is a diagonal matrix. The column of F are
the eigen vector of S, while the element, of D are
the eigen value of S. Every element of D is a
measure of the percentage variance explained by
the corresponding eigen vector. F and D were
calculated from S by Jacobi's method (Greenstadt,
1960). The study has been extended into normal,
flood and drought years over India.
The locations of all the 68 stations have been
shown in Fig. 1.
In this study a rainy day has been defined as
.25 cm or more as described by India Meteoro-
logical Department.
The probability of a rainy day in a normal,
flood and drought year has also been calculated
and shown in Table 2±4, respectively, along with
the number of rainy days. Table 5 shows the