Professor Adrian Matthews School of Environmental Sciences and School of Mathematics, University of East Anglia, Norwich, UK
Home


Research
Publications
Abstracts
Research group
PhD projects

MJO introduction
Current MJO forecast
image
image
image
image
MJO forecast method
MJO forecast validation
MJO forecast archive
MJO EMD archive
Other MJO forecasts

Centre for Ocean and Atmospheric Sciences

MJO EMD archive

The empirical mode decomposition (EMD) technique that is used to isolate (filter) the MJO signal from the raw data has problems with end effects, as do all real-time filtering procedures. Even though care has been take to minimise these end effects, under certain circumstances they can be quite large. This can lead to an erroneous MJO state being fed into the forecast model, which then leads to a poor forecast.

Follow the links below to access archive validation data for the EMD process that isolates the MJO signal, that then goes into the forecast. Each link is a page with a grid of maps for that month. Each row corresponds to a particular day. Within each row, from left to right, is the analysis EMD map (i.e., the actual state of the MJO on that day) followed by the raw OLR anomaly maps for that day and the previous 10 days. The ability of the EMD process to filter the raw OLR data and its susceptibility to end effects can be qualitatively assessed here. Note, these pages may not be of great interest to the general reader!

Created: Fri Nov 24 02:03:26 2017