Expectation-maximization analysis of spatial time series
Citable URI
https://hdl.handle.net/1912/1571As published
https://doi.org/10.5194/npg-14-73-2007Abstract
Expectation maximization (EM) is used to estimate the parameters of a Gaussian Mixture Model for spatial time series data. The method is presented as an alternative
and complement to Empirical Orthogonal Function (EOF) analysis. The resulting weights, associating time
points with component distributions, are used to distinguish
physical regimes. The method is applied to equatorial Pacific
sea surface temperature data from the TAO/TRITON mooring
time series. Effectively, the EM algorithm partitions the
time series into El Nino, La Nina and normal conditions. The
EM method leads to a clearer interpretation of the variability
associated with each regime than the basic EOF analysis.
Description
© Author(s) 2007. This work is licensed
under a Creative Commons License. The definitive version was published in Nonlinear Processes in Geophysics 14 (2007): 73-77, doi: 10.5194/npg-14-73-2007
Collections
Suggested Citation
Nonlinear Processes in Geophysics 14 (2007): 73-77The following license files are associated with this item:
Except where otherwise noted, this item's license is described as Attribution-NonCommercial-ShareAlike 2.5 Generic
Related items
Showing items related by title, author, creator and subject.
-
Trajectory encounter volume as a diagnostic of mixing potential in fluid flows
Rypina, Irina I.; Pratt, Lawrence J. (Copernicus Publications on behalf of the European Geosciences Union&the American Geophysical Union, 2017-05-03)Fluid parcels can exchange water properties when coming into contact with each other, leading to mixing. The trajectory encounter mass and a related simplified quantity, the encounter volume, are introduced as a measure ... -
Controllability, not chaos, key criterion for ocean state estimation
Gebbie, Geoffrey A.; Hsieh, Tsung-Lin (Copernicus Publications on behalf of the European Geosciences Union & the American Geophysical Union, 2017-07-19)The Lagrange multiplier method for combining observations and models (i.e., the adjoint method or "4D-VAR") has been avoided or approximated when the numerical model is highly nonlinear or chaotic. This approach has been ... -
Strongly nonlinear, simple internal waves in continuously-stratified, shallow fluids
Ostrovsky, Lev A.; Helfrich, Karl R. (Copernicus Publications on behalf of the European Geosciences Union and the American Geophysical Union, 2011-02-14)Strongly nonlinear internal waves in a layer with arbitrary stratification are considered in the hydrostatic approximation. It is shown that "simple waves" having a variable vertical structure can emerge from a wide class ...