Jet formation and tracer cycles simulated by a new Jupiter GCM (JupiterMPAS)

Abstract:

We have converted a new atmospheric dynamical model – the National Center for Atmospheric Research (NCAR) Model for Prediction Across Scales (MPAS) – for application to planetary atmospheres (PlanetMPAS). PlanetMPAS has thus far been applied to the atmospheres of Jupiter, Mars and Titan. The purpose for developing this new planetary GCM is to: a) study atmospheric dynamics on multiple scales in a single GCM, e.g., from mesoscale to synoptic scale; b) build a platform where physics parameterizations developed in previous models (e.g., PlanetWRF and PlanetMITgcm) can be easily implemented; c) minimize various issues related to numerical stabilities, conservation properties and computational efficiencies.

JupiterMPAS currently focuses on weather-layer dynamics, but may extend to deep atmosphere by taking advantage of its non-hydrostatic dynamical core. It includes the following key physics parameterizations: a mass-flux based moist convection scheme for water cycle, a simple microphysics scheme for water based on Kessler scheme, a moist convective adjustment scheme for ammonia cycle based on Betts-Miller scheme, a deep convective mixing scheme that mixes momentum, heat and tracers below the water cloud deck, and a grey-radiative transfer model to represent the radiative processes in the atmosphere. These physics parameterizations are chosen because they are generic enough for applications to planetary atmospheres.

Preliminary results with two active tracers water and ammonia show that JupiterMPAS is able to produce banded jet pattern with equatorial superrotation exceeding 100m/s. Similar to the cloud-model results, ammonia cloud forms between ~0.4 and ~0.7 bars, and water cloud forms between ~3.5 and 7.5 bars assuming deep water and ammonia abundances of three times solar according to recent JUNO observations. Ammonia vapor abundance also shows apparent depletion at mid-to-high latitudes (+/- 15 degree poleward). Such depletion extends a few bars deep, below which the ammonia vapor is mostly well mixed. The modeled ammonia vapor distribution resembles some features seen in JUNO observations, though the observed ammonia vapor depletion extends far deeper (near 60 bars).