Implications of space-time orientation for Principal Components Analysis of Earth observation image time series

A time series of geographic images can be viewed from two perspectives: as a set of images, each image representing a slice of time, or as a grid of temporal profiles (one at each pixel location). In the context of Principal Components Analysis (PCA), these different orientations are known as T-mode and S-mode analysis respectively. In the sparse literature on these modes it is recognized that they produce different results, but the reasons have not been fully explored. In this paper we investigate the interactions between space-time orientation and standardization and centering in PCA. Standardization refers to the eigenanalysis of the inter-variable correlation matrix rather than the variance-covariance matrix while centering refers to the subtraction of the mean in the development of either matrix. Using time series of monthly anomalies in lower tropospheric temperature from the Microwave Sounding Unit (MSU) as well as in CO2 in the middle troposphere from the Atmospheric Infrared Sounder (AIRS), we show that with T-mode PCA, standardization has the effect of giving equal weight to each time step while centering has the effect of detrending over time. In contrast, with S-mode PCA, standardization has the effect of giving equal weight to each location in space while centering detrends over space. Further, in the formation of components, S-mode PCA preferences patterns that are prevalent over space while T-mode PCA preferences patterns that are prevalent over time. The two orientations thus provide complementary insights into the nature of variability within the series.

[1]  I. Noy-Meir,et al.  Data Transformations in Ecological Ordination: I. Some Advantages of Non-Centering , 1973 .

[2]  W. T. Williams,et al.  Data Transformations in Ecological Ordination: II. On the Meaning of Data Standardization , 1975 .

[3]  R. Compagnucci,et al.  Surface Pressure Patterns During The Year Over Southern South America , 1997 .

[4]  M. Richman,et al.  Can principal component analysis provide atmospheric circulation or teleconnection patterns? , 2008 .

[5]  Thomas S. Pagano,et al.  Satellite remote sounding of mid‐tropospheric CO2 , 2008 .

[6]  Brian J. L. Berry,et al.  APPROACHES TO REGIONAL ANALYSIS: A SYNTHESIS , 1964 .

[7]  I. Jolliffe.  Principal Component Analysis , 2002 .

[8]  Raymond B. Cattell,et al.  Factor Analysis. An Introduction and Manual for the Psychologist and Social Scientist. , 1953 .

[9]  C. Bretherton,et al.  The ENSO Signal in Tropical Tropospheric Temperature , 2002 .

[10]  A. Barnston,et al.  Classification, seasonality and persistence of low-frequency atmospheric circulation patterns , 1987 .

[11]  R. Compagnucci,et al.  The spatial and temporal behaviour of the lower stratospheric temperature over the Southern Hemisphere: the MSU view. Part I: data, methodology and temporal behaviour , 2001 .

[12]  M. Richman,et al.  Rotation of principal components , 1986 .

[13]  B. Weare,et al.  Empirical Orthogonal Analysis of Pacific Sea Surface Temperatures , 1976 .

[14]  N. Lau,et al.  The Role of the “Atmospheric Bridge” in Linking Tropical Pacific ENSO Events to Extratropical SST Anomalies , 1996 .

[15]  Ngar-Cheung Lau,et al.  Remote Sea Surface Temperature Variations during ENSO: Evidence for a Tropical Atmospheric Bridge , 1999 .

[16]  H. V. D. Dool,et al.  Empirical Methods in Short-Term Climate Prediction , 2006 .

[17]  R. Compagnucci,et al.  The spatial and temporal behaviour of the lower stratospheric temperature over the Southern Hemisphere: the MSU view. Part II: spatial behaviour , 2001 .

[18]  W. Briggs.  Statistical Methods in the Atmospheric Sciences , 2007 .

[19]  Ron Johnston,et al.  Multivariate Statistical Analysis in Geography: A Primer on the General Linear Model , 1978 .

[20]  J. E. Jackson.  A User's Guide to Principal Components , 1991 .

[21]  J. Eastman,et al.  Long sequence time series evaluation using standardized principal components , 1993 .

[22]  Marc P. Armstrong,et al.  Temporality in Spatial Databases , 1988 .

[23]  David B. Stephenson,et al.  Arctic Oscillation or North Atlantic Oscillation , 2001 .

[24]  John M. Wallace,et al.  The Signature of ENSO in Global Temperature and Precipitation Fields Derived from the Microwave Sounding Unit , 1994 .

[25]  R. Cattell,et al.  Factor Analysis: An Introduction and Manual for the Psychologist and Social Scientist , 1953 .

[26]  M. Ehrendorfer.  A regionalization of Austria's precipitation climate using principal component analysis , 1987 .

[27]  John R. Christy,et al.  MSU Tropospheric Temperatures: Dataset Construction and Radiosonde Comparisons , 2000 .