As part of a long-term project examining short-period tropical rainfall characteristics, a preliminary classification of stations for a wide range of localities based on such characteristics is presented. The approach adopted, using principal component analysis and a clustering procedure, functioned well and the resulting classification would not have been anticipated simply by using general annual and seasonal variables alone. Interpretation of the components allowed the important characteristics of each group of stations to be described. Although the stations used did not represent every region of the tropics, the range of groups with a wide variety of characteristics suggests that they are indicative of conditions applicable to many areas. Although the data base was for varying periods and record lengths, it appeared to be adequate for the purpose of the analysis.
[1]
H. Weinand,et al.
Regional Weil‐Being in Australia Revisited
,
1991
.
[2]
Laurence S. Kalkstein,et al.
Development of an automated spatial synoptic climatological classification
,
1990
.
[3]
R. Stone.
Weather types at Brisbane, Queensland: An example of the use of principal components and cluster analysis
,
1989
.
[4]
I. J. Jackson.
Daily rainfall over Northern Australia: Deviations from the world pattern
,
1988
.
[5]
L. Kalkstein,et al.
An Evaluation of Three Clustering Procedures for Use in Synoptic Climatological Classification
,
1987
.
[6]
I. Jolliffe.
Principal Component Analysis
,
2005
.
[7]
I. G. Jackson.
Relationships between raindays, mean daily intensity and monthly rainfall in the tropics
,
1986
.