Premium
A comparison of synoptic classification schemes based on ‘objective’ procedures
Author(s) -
Key Jeff,
Crane Robert G.
Publication year - 1986
Publication title -
journal of climatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.58
H-Index - 166
eISSN - 1097-0088
pISSN - 0196-1748
DOI - 10.1002/joc.3370060404
Subject(s) - principal component analysis , statistics , subjectivity , mathematics , circulation (fluid dynamics) , atmospheric circulation , pattern recognition (psychology) , artificial intelligence , computer science , geography , meteorology , engineering , philosophy , epistemology , aerospace engineering
Various ‘objective’ methods of classifying the atmospheric circulation are compared in order to determine whether the different methods result in similar groupings of circulation types, and also to examine the question of how much subjectivity is present in each of the typing schemes. The results indicate a high degree of subjectivity within each approach and that the different classification schemes do give rise to the production of different synoptic types. A method which groups days based on the sum of squares difference between cases (Kirchhofer method) gives the best overall classification, but only classifies 80 per cent of the days. A technique that clusters days based on principal component scores gives a larger mean sum of squares difference between the type patterns and the individual cases, but classifies all of the days.