[A-OS09] Climate variability and predictability on subseasonal to multidecadal timescales

[AOS09-01] Beyond day 2: Variable-resolution convective-scale prediction on medium-range, subseasonal, and longer timescales.

★Invited Papers

*Lucas Harris¹ (1.NOAA/Geophysical Fluid Dynamics Laboratory)

Keywords:Subseasonal-to-seasonal, Variable-resolution Modeling, Cloud-resolving model, Severe Weather, Madden-Julian Oscillation

Current medium-range and longer (beyond 48–60 hour) prediction is limited to global models with resolutions too coarse to explicitly represent deep convection, and are limited by the quality of existing deep convective and gravity-wave drag parameterizations. Regional convective-scale models are often found to be inadequate for prediction on these timescales. In this talk I will discuss the prospects for using variable-resolution global-to-regional models for extending convective-scale prediction into these extended lead times. Results from the GFDL SHiELD model, which heavily relies upon the variable-resolution capabilities in the FV3 dynamical core, demonstrates medium-range predictability of severe thunderstorm outbreaks and significantly enhanced predictability of the Madden-Julian Oscillation. I also discuss the role of explicit deep convection in these predictions and challenges that arise in the absence of a deep convective parameterization. I also discuss how these models fit into a seamless modeling suite with traditional-resolution weather and climate models, and how variable-resolution models can be a “stepping stone” to global cloud-resolving models.