NOAA Climate Process Team: Understanding Processes Affecting Madden‐Julian Oscillation Initiation and Propagation
Improvement of MJO simulation in NCEP Coupled Forecast System: Upper ocean and air- sea coupled processes
Toshiaki Shinoda (Department of Physical and Environmental Sciences, Texas A&M University, Corpus Christi)
Joshua Fu (IPRC, SOEST, University of Hawaii at Manoa)
Ren-Chieh Lien (APL, University of Washington)
Hyodae Seo (Woods Hole Oceanographic Institution)
Alexander Soloviev (Oceanographic Center, Nova Southeastern University)
Wanqiu Wang (NOAA/NCEP/CPC)
Award: $180,000 for WHOI
Accurate simulation and prediction of the Madden-Julian Oscillation (MJO) is one of the major challenges for climate modeling and operational weather forecasts. The MJO in the NCEP Coupled Forecast System (CFS) is too weak and propagates too slowly, particularly during its initiation and evolution over the Indian Ocean. With the objective to advance our understanding of the MJO initiation processes and improve MJO prediction, DYNAMO international field campaign provides a substantial amount of oceanic and atmospheric in-situ data. In the last few years, the DYNAMO data have been used to identify important oceanic, atmospheric, and air-sea coupled processes in the MJO initiation and propagation. A primary goal of this proposed study is to advance MJO simulation and prediction in NOAA CFS by improving the representation of the air-sea flux and upper-ocean vertical mixing. The DYNAMO data and the outcome from our previous DYNAMO projects will be maximally utilized for the improvement of MJO simulations. To accomplish this goal, we propose to:
(1) Improve the one-dimensional General Ocean Turbulence Model (GOTM) by including a new mixing scheme developed by Soloviev et al. (2001) that has realistic performance in the tropics and extra-tropics in capturing large diurnal warming and responding to strong westerly wind bursts. The improved GOTM will be tested at DYNAMO field observation sites where accurate surface fluxes as well as high quality upper ocean data are available. These schemes will be further tested in the uncoupled ocean component of CFS with an enhanced vertical resolution.
(2) Develop computationally efficient surface flux algorithm using the most updated version of TOGA COARE bulk flux algorithm, in-situ flux observations, and the method used by Kara et al. (2000, 2005). The algorithms will be carefully validated against DYNAMO observations and tested in the atmospheric component of CFS.
(3) Implement the improved ocean mixing parameterization and air-sea flux algorithm in coupled CFS, and evaluate the MJO simulation and prediction skill based on the comparison with a variety of in-situ and satellite observations, and regional coupled model experiments.
We anticipate that the proposed research with the improved CFS will result in a significant improvement in the forecast of subseasonal variability including the MJO and associated variability such as tropical storms and North America weather. The schemes developed, tested, and implemented in the project also provide guidance for improving the next generation CFS and other coupled models in the climate community, which generally have poor representation of the upper ocean processes and deficient surface fluxes critical to the simulation of the MJO.