Evaluating the MJO's Impact on North American Subseasonal Forecasts in a Real-time Linear Inverse Model

Development of reliable subseasonal-to-seasonal (S2S) forecasts is driven by the needs of a wide range of stakeholders from energy, water management, to agriculture sectors as such long-lead time predictions can provide actionable information for resource allocation. Teleconnections originating in the tropics, including those related to the MJO and ENSO, provide potential major sources of predictability for global S2S forecasts, and in particular, S2S forecasts over North America. However, current operational models differ in their ability to accurately simulate tropical variability, as well as the subsequent downstream teleconnections. As a result, official forecast guidance is typically based on information from a suite of dynamical and statistical models and tools. In support of the NOAA Climate Prediction Center Weeks 3-4 temperature outlooks over the United States, the NOAA Physical Sciences Laboratory recently developed an empirical-dynamical linear inverse model (LIM) that has 2m temperature skill comparable to that of the real-time ECMWF IFS. The LIM provides real-time forecasts of the coupled atmosphere-ocean system, including representations of the Northern Hemisphere extratropical circulation, tropical sea surface temperature and heating, and North American 2m temperature.

In this talk, we will evaluate the LIM’s overall skill in predicting tropical variability and subsequent downstream teleconnections. In particular, we will focus on its performance when forecasts are conditioned on various MJO indices including two widely used indices, the Real-time Multivariate MJO (RMM) and the OLR-based MJO Index (OMI). The results are expected to identify the ability of the LIM to skillfully simulate key large-scale teleconnections important to S2S forecasts and improve our understanding of physical processes influencing MJO-modulated forecasts of opportunity. This framework can also be adapted operationally to provide insights into the reasons, timing, and locations of expected skillful forecasts, fostering confidence and enhancing the applicability to use the S2S forecasts.