Variability in surface meteorological variables over the equatorial Indian Ocean and its relationship with the Madden-Julian Oscillation

To examine statistical behavior of variability in surface meteorological variables with sub-daily time scales is expected to provide fundamental information for understanding air-sea interaction, surface flux variability, and interaction between cumulus convection and a boundary layer. Here we study relationship between the behavior and Madden-Julian Oscillation (MJO), through analysis of observation data obtained over the equatorial eastern Indian Ocean during CINDY/DYNAMO intensive observation period of October-December, 2011. The data analyzed include observations by Research Vessels Roger Revelle and Mirai, and by a moored buoy installed near the equator at 1.5S, 90E as a part of the RAMA project. The sub-daily variability is defined as an anomaly of hourly-mean variable from its daily running mean. The buoy data show that horizontal wind and temperature fluctuations had larger variance during convective active phases of the MJO than pre- and post-active phases. The sub-daily fluctuation contributed to approximately 30 % of surface latent heat flux before intrusion of the MJO-scale westerly wind, while after that, its contribution reduced to less than 10 % of the flux. In the latter period, not only the MJO-scale westerly but also vigorous synoptic-scale meridional wind fluctuation plays a role on the flux. Furthermore, it seems that a major axis of the horizontal wind fluctuation shifted from southwest-northeast to northwest-southeast directions when the westerly wind started to flow. Such characteristics were generally consistent with observations by Roger Revelle staying near the equator, while data obtained by Mirai staying 8S exhibited different characteristics, with smaller amplitude of the sub-daily fluctuations. In addition to the above-mentioned case study, we use RAMA buoy data of other periods to perform statistical analysis, intending to compare results with those of the case study. Furthermore, we plan to compare the observed behavior with one simulated by a global cloud-system resolving model, NICAM, to discuss its reproducibility of the boundary layer variability.