The CPC MJO index is an index developed at CPC that uses Extended Empirical Orthogonal Function (EEOF) analysis of 200-hPa velocity potential (a measure of upper tropospheric divergence) from the NCEP/NCAR Reanalysis. Currently, the index is applicable to describe the temporal evolution of the MJO for both the Northern Hemisphere cold season (defined as November -April) and warm season (defined as May – October) during ENSO-neutral and weak ENSO years. Non-overlapping pentad data were used for the period 1979-2004 to obtain 10 patterns (corresponding to the first EEOF mode) that describe the convectively active phase of the MJO at different longitudes around the global tropics. The projection of daily analyses of 200-hPa velocity potential onto these ten patterns defines the index at ten locations or phases. The CPC MJO index has been run operationally for several years and has proven beneficial to a number of activities at CPC and to the larger scientific and general public communities as well.

The CPC MJO index has been extended to forecast mode by applying a Markov model trained on the 200-hPa velocity potential for the period 1979-2004. The Markov model is the first order regression of the future pentad data onto the current pentad data in a reduced EOF space. The forecasts of the CPC MJO index are then obtained by projecting the forecast field onto the first EEOF of 200-hPa velocity potential. The predicted CPC MJO index catches features of eastward propagation and phase change represented for MJO. The forecast skill is relative high over the Indian Ocean and the western Pacific. The CPC MJO index operating in forecast mode will further aid CPC monitoring activities as well as provide useful predictive input to CPC's MJO weekly update and experimental global tropics benefits/hazards assessment currently under development.