Analysis of MJO Wind-Flux Feedbacks in the Indian Ocean Using RAMA Buoy Observations

Abstract

Observations spanning 2004-2012 from two Research Moored Array for African-Asian-Australian Monsoon Analysis and Prediction (RAMA) buoys along the equator in the Indian Ocean are used in conjunction with Tropical Rainfall Measuring Mission (TRMM) to assess the relative importance of surface latent heat fluxes to intraseasonal convection. This work is motivated by previous observational and modeling studies that have suggested the importance of wind-induced surface fluxes to the dynamics of the Madden-Julian Oscillation (MJO).Intraseasonal variability is isolated in two ways: 1) 20-100-day bandpass filtering and 2) using the global real time multivariate MJO index. Linear regression shows latent heat flux anomalies to be between 4 % and 8 % of precipitation anomalies when the two variables are compared using similar Wm −2 energy units. From a moist static energy budget viewpoint, these results confirm the potential of wind-induced latent heat fluxes to aid destabilization of MJO convection. Results derived from using both simple intraseasonal filtering and global MJO indices indicate that precipitation leads latent heat flux on the order of a few days, indicating surface fluxes may be more important for maintenance of deep MJO convection in addition to helping set the MJO's propagation speed. Sensitivity tests using smoothed wind speed or thermodynamics (i.e., air temperature, relative humidity, and sea surface temperature) to compute latent heat flux show wind speed variability explains most of the latent heat flux variability on intraseasonal timescales. A similar conclusion is found via linearization of the latent heat flux formula. Additional analysis shows mesoscale and synoptic scale wind variability have negligible impact on intraseasonal latent heat flux anomalies.

Atmosphere‐ocean coupled processes in the Madden‐Julian oscillation

“…The SST feedback is modulated by the atmosphere, since both ocean and atmosphere determine Δq and ΔT. This is demonstrated in Figure 17 using 2 years of latent heat flux data from the 0 ∘ N, 90 ∘ E RAMA buoy and methodologies of Hendon and Glick [1997], , Maloney and Sobel [2004], Araligidad and Maloney [2008], Andersen and Kuang [2012], Riley Dellaripa and Maloney [2015], and others. Flux-related variables and rainfall are lag regressed onto SST anomalies.…”

“…Overbars represent background mean quantities, and primed quantities (hereafter not retained) are time‐varying departures from the background means. The nonlinear terms and are omitted, as they are negligibly small [e.g., Riley Dellaripa and Maloney , ]. Secondary dependencies on stability and ocean wave state (section 3.6) are not included.…”

“…(These values are based on regressing rainfall onto LH. However, regressing LH onto rainfall is perhaps a calculation more directly relevant to moisture mode theory and yields a 5∼10% contribution of LH to rainfall [e.g., Riley Dellaripa and Maloney , ].) Fluxes computed using long‐term means of SST and near‐surface humidity were about 20% greater than those using observed SST and humidity variations, generally consistent with Shinoda et al [].…”

“…Rather, they occur during the westerly wind (active and transition to suppressed conditions) phase and are partially offset by surface flux enhancements during the suppressed and developing stages [ Shinoda et al , ]. Riley Dellaripa and Maloney [] demonstrate this result using data from an Indian Ocean RAMA buoy (Figure ). This finding suggests two possible feedbacks to the MJO by SST‐modulated latent heat fluxes: first, the SST‐enhanced flux during MJO developing and active phases increases available boundary layer moisture and amplifies and maintains the developing convective anomalies, respectively, via a so‐called “modified WISHE” effect [ Maloney and Sobel , ]; second, the westerly wind during the transition‐to‐suppressed phase impacts the ocean mixed layer energy budget and influences the SST recovery time.…”

“…Here latent heat flux enhancement by warm SSTs is 10∼15 W m −2 , or about 10% of the total flux. This modest increase in the latent heat flux may nevertheless be important for maintaining the MJO column moist static energy anomaly, as the anomaly is depleted by convection at a rate of 10∼20% d −1 [e.g., Riley Dellaripa and Maloney , , and references therein]. By maintaining MJO convection, the ocean feedbacks may indirectly supplement longwave heating and moistening associated with stratiform cloud feedbacks [ Chikira , ].…”