Search Results

You are looking at 1 - 3 of 3 items for

  • Author or Editor: Noah D. Brenowitz x
  • Refine by Access: All Content x
Clear All Modify Search
Noah D. Brenowitz, Yevgeniy Frenkel, and Andrew J. Majda


Recent observational and theoretical studies show a systematic relationship between tropical moist convection and measures related to large-scale convergence. It has been suggested that cloud fields in the column stochastic multicloud model compare better with observations when using predictors related to convergence rather than moist energetics (e.g., CAPE) as per Peters et al. Here, this work is extended to a fully prognostic multicloud model. A nonlocal convergence-coupled formulation of the stochastic multicloud model is implemented without wind-dependent surface heat fluxes. In a series of idealized Walker cell simulations, this convergence coupling enhances the persistence of Kelvin wave analogs in dry regions of the domain while leaving the dynamics in moist regions largely unaltered. This effect is robust for changes in the amplitude of the imposed sea surface temperature (SST) gradient. In essence, this method provides a soft convergence coupling that allows for increased interaction between cumulus convection and the large-scale circulation but does not suffer from the deleterious wave–conditional instability of the second kind (CISK) behavior of the Kuo-type moisture-convergence closures.

Full access
Qiu Yang, Andrew J. Majda, and Noah D. Brenowitz


Atmospheric convection exhibits distinct spatiotemporal variability at different latitudes. A good understanding of the effects of rotation on the multiscale organization of convection from the mesoscale to synoptic scale to planetary scale is still lacking. Here cloud-resolving simulations with fixed surface fluxes and radiative cooling are implemented with constant rotation in a two-dimensional (2D) planetary domain to simulate multiscale organization of convection from the tropics to midlatitudes. All scenarios are divided into three rotation regimes (weak, order-one, and strong) to represent the idealized ITCZ region (0°–6°N), the Indian monsoon region (6°–20°N), and the midlatitude region (20°–45°N), respectively. In each rotation regime, a multiscale asymptotic model is derived systematically and used as a diagnostic framework for energy budget analysis. The results show that planetary-scale organization of convection only arises in the weak rotation regime, while synoptic-scale organization dominates (vanishes) in the order-one (strong) rotation regime. The depletion of planetary-scale organization of convection as the magnitude of rotation increases is attributed to the reduced planetary kinetic energy of zonal winds, mainly due to the decreasing acceleration effect by eddy zonal momentum transfer from mesoscale convective systems (MCSs) and the increasing deceleration effect by the Coriolis force. Similarly, the maintenance of synoptic-scale organization is related to the acceleration effect by MCSs. Such decreasing acceleration effects by MCSs on both planetary and synoptic scales are further attributed to less favorable conditions for convection provided by weaker background vertical shear of the zonal winds, resulting from the increasing magnitude of rotation.

Free access
Noah D. Brenowitz, Tom Beucler, Michael Pritchard, and Christopher S. Bretherton


Neural networks are a promising technique for parameterizing subgrid-scale physics (e.g., moist atmospheric convection) in coarse-resolution climate models, but their lack of interpretability and reliability prevents widespread adoption. For instance, it is not fully understood why neural network parameterizations often cause dramatic instability when coupled to atmospheric fluid dynamics. This paper introduces tools for interpreting their behavior that are customized to the parameterization task. First, we assess the nonlinear sensitivity of a neural network to lower-tropospheric stability and the midtropospheric moisture, two widely studied controls of moist convection. Second, we couple the linearized response functions of these neural networks to simplified gravity wave dynamics, and analytically diagnose the corresponding phase speeds, growth rates, wavelengths, and spatial structures. To demonstrate their versatility, these techniques are tested on two sets of neural networks, one trained with a superparameterized version of the Community Atmosphere Model (SPCAM) and the second with a near-global cloud-resolving model (GCRM). Even though the SPCAM simulation has a warmer climate than the cloud-resolving model, both neural networks predict stronger heating/drying in moist and unstable environments, which is consistent with observations. Moreover, the spectral analysis can predict that instability occurs when GCMs are coupled to networks that support gravity waves that are unstable and have phase speeds larger than 5 m s−1. In contrast, standing unstable modes do not cause catastrophic instability. Using these tools, differences between the SPCAM-trained versus GCRM-trained neural networks are analyzed, and strategies to incrementally improve both of their coupled online performance unveiled.

Full access