Share Email Print

Proceedings Paper

Multiscale analysis of climate data in Changchun, China
Author(s): Hong Shu; Kai Guo
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

To some extent, the multi-level dynamics of an atmosphere system implies temporal structures in time-varying climate data. Here the multi-period issue of temperature data is studied. Since the scale parameter of wavelets is not easily understandable, the relationship between time period and time scale is formulated in Morelet wavelet. Unlike overall multi-period and -frequency information in Fourier analysis, wavelets analysis provides us with local multi-period information. At Changchun meteorological station, our experimental dataset are daily temperature measurements from 1951 to 2005. After a wavelet transform of climate dataset, modes and real parts of wavelets coefficients are drawn for visually exploring local multi-period information. In particular, it is seen that the time periods of 1 year, 3 to 4 years are globally apparent, and the time period of 8 to 12 years is locally apparent. For the temperature, there is an overall trend of colder and warmer interchanging time periods, i.e., a colder period before the middle 1980s and a warmer period from the middle 1980s to present. These two time periods are further divided into four cold periods and three warm periods respectively. In a large time period, 1987 is the year of abrupt temperature change. In a middle time period, 1970 is the year of abrupt temperature change. In the time period of year, there exist specific years of abrupt temperature change. In our framework of spatiotemporal data mining, these local multiple periods are used for creating multi-level spatiotemporal meteorological association rules. It is proved that the Morelet wavelet is feasible for exploring temporal structures in climate data.

Paper Details

Date Published: 30 October 2009
PDF: 6 pages
Proc. SPIE 7498, MIPPR 2009: Remote Sensing and GIS Data Processing and Other Applications, 74982J (30 October 2009); doi: 10.1117/12.833693
Show Author Affiliations
Hong Shu, Wuhan Univ. (China)
Kai Guo, Wuhan Univ. (China)

Published in SPIE Proceedings Vol. 7498:
MIPPR 2009: Remote Sensing and GIS Data Processing and Other Applications
Faxiong Zhang; Faxiong Zhang, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?