Monday, 23 January 2017: 11:45 AM
Conference Center: Chelan 2 (Washington State Convention Center )
Stephanie S. Rushley, University of Washington, Seattle, WA; and
D. Kim
In this study, we examine changes in the characteristics of the Madden-Julian Oscillation (MJO) due to greenhouse gas induced warming. While previous studies have focused on individual models, we seek to expand understanding of the MJO’s behavior in a warmer climate by assessing the MJO response to warming in a multi-model ensemble. For this purpose, nine models are chosen based on their performance in representing the MJO from the Coupled Model Intercomparison Project Phase 5 models. Wavenumber-frequency filtering is used to isolate the MJO’s signal from the background variability for the preindustrial control and the Representative Concentration Pathways (RCP) 8.5 scenario simulations. The MJO-filtered precipitation data is considered during boreal winter (November to April) when the MJO is at its peak strength. Changes in MJO’s amplitude, horizontal size, phase speed, and north-south asymmetry are evaluated in the nine models.
Most models exhibit an increase in the MJO precipitation variance under the RCP 8.5 scenario relative to that in the preindustrial period. The changes in the MJO precipitation variance scale well with changes to the total precipitation variance. No robust relationship is found between the global mean surface temperature change and changes in the MJO precipitation variance. The MJO gets larger and faster under the RCP 8.5 warming scenario. The horizontal scale of the MJO increases in six among the nine models and the MJO consistently speeds up in all of the models. A majority of the models show an increase in the north-south asymmetry of the MJO variance, increasingly favoring the southern hemisphere with the warming. The north-south asymmetry in the MJO variance increases at a greater rate than that of the north-south asymmetry in the mean state.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner