The response of the Pacific Walker circulation (WC) to warming in both observations and simulations is uncertain. We diagnose contributions to the WC response in comprehensive and idealized general circulation model (GCM) simulations. We find that the spread in WC response is substantial across both the Coupled Model Intercomparison Project (CMIP6) and the Atmospheric Model Intercomparison Project (AMIP) models, implicating differences in atmospheric models in the spread in projected WC strength. Using a moist static energy (MSE) budget, we evaluate the contributions to changes in the WC strength related to changes in gross moist stability (GMS), horizontal MSE advection, radiation, and surface fluxes. We find that the multimodel mean WC weakening is mostly related to changes in GMS and radiation. Furthermore, different GMS responses can explain a substantial portion of the spread in WC responses. The GMS response is potentially sensitive to parameterized convective entrainment which can affect lapse rates and the depth of convection. We thus investigate the role of entrainment in setting the GMS response by varying the entrainment rate in an idealized GCM. The idealized GCM is run with a simplified Betts-Miller convection scheme, modified to represent entrainment. The WC weakening with warming in the idealized GCM is dampened when higher entrainment rates are used. However, the spread in GMS responses due to differing entrainment rates is much smaller than the spread in GMS responses across CMIP6 models. Therefore, further work is needed to understand the large spread in GMS responses across CMIP6 and AMIP models.

Climate models project a distinct seasonality to future changes in daily extreme precipitation. In particular, models project that over land in the extratropical Northern Hemisphere the summer response is substantially weaker than the winter response in percentage terms. Here we decompose the projected response into thermodynamic and dynamic contributions and show that the seasonal contrast arises due to a negative dynamic contribution in northern summer, and a positive dynamic contribution and an anomalously strong thermodynamic contribution in northern winter. The negative dynamic contribution in northern summer is due to weakened ascent and is strongly correlated with decreases in mean near-surface relative humidity which tend to inhibit convection. Finally, we show that the summer-winter contrast is also evident in observed trends of daily precipitation extremes in northern midlatitudes, which provides support for the contrast found in climate-model simulations.

Tropical convective organization and associated clusters of precipitation affect the global circulation and Earth’s energy budget, but many aspects such as the power-law distribution of precipitation clusters and the relation to convective self-aggregation remain poorly understood. Here, we present a physics-informed 2D conceptual model for tropical convective organization. The model is based on the budget equation of column moist static energy (CMSE) with terms that are parameterized based on diagnosing them in a high-resolution simulation with explicit convection. The conceptual model combines a reaction-diffusion equation, where the reaction term has memory, with a temporal red noise. We find that vertical advection has a strengthening effect on CMSE perturbations, whereas horizontal advection has a mitigating effect on CMSE perturbations. The CMSE tendencies from vertical and horizontal advection terms are larger in magnitude than those from radiation and surface fluxes. The strengthening effect of vertical advection corresponds to a negative gross moist stability (GMS) at short spatiotemporal scales, but the GMS becomes positive when the convection changes from shallow to deep, killing off the growth in CMSE. The conceptual model is faithful in simulating self-aggregation in that it shows a domain-size dependence of aggregation found in many prior works, and its CMSE power spectrum matches that of the high-resolution simulation except for a shallowing of the slope at high wavenumbers. Through analyzing the conceptual model equation, we show that 1) the domain-size dependence of self-aggregation is due to a competition between the strengthening effect of vertical advection and the smoothing effect of horizontal advection, and 2) the combination of temporal red noise and diffusive horizontal advection sets the shape of the CMSE power spectrum. Furthermore, the conceptual model also reproduces power-law distributions of precipitation clusters when precipitation clusters are viewed as thresholded islands on the CMSE topography. Thus, the simple conceptual model captures and helps to explain important aspects of convective self-aggregation and tropical convective organization.

Attempts to use machine learning to develop atmospheric parameterizations have mainly focused on subgrid effects on temperature and moisture, but subgrid momentum transport is also important in simulations of the atmospheric circulation. Here, we use neural networks to develop a parameterization of subgrid momentum transport that learns from coarse-grained data of a high-resolution atmospheric simulation in an idealized aquaplanet domain. We show that substantial subgrid momentum transport occurs due to convection and non-orographic gravity waves. The parameterization has a structure that ensures the conservation of momentum, and it has reasonable skill in predicting momentum fluxes associated with convection, although its skill is lower as compared to subgrid energy and moisture fluxes. The neural-network parameterization is implemented in the same atmospheric model at coarse resolution and leads to stable simulations. Overall, our results show that it is challenging to predict subgrid momentum fluxes and that machine-learning momentum parameterization gives promising results.