Search for: All records

We design a new strategy to load‐balance high‐intensity sub‐grid atmospheric physics calculations restricted to a small fraction of a global climate simulation's domain. We show why the current parallel load balancing infrastructure of Community Earth System Model (CESM) and Energy Exascale Earth Model (E3SM) cannot efficiently handle this scenario at large core counts. As an example, we study an unusual configuration of the E3SM Multiscale Modeling Framework (MMF) that embeds a binary mixture of two separate cloud‐resolving model grid structures that is attractive for low cloud feedback studies. Less than a third of the planet uses high‐resolution (MMF‐HR; sub‐km horizontal grid spacing) relative to standard low‐resolution (MMF‐LR) cloud superparameterization elsewhere. To enable MMF runs with Multi‐Domain cloud resolving models (CRMs), our load balancing theory predicts the most efficient computational scale as a function of the high‐intensity work's relative overhead and its fractional coverage. The scheme successfully maximizes model throughput and minimizes model cost relative to precursor infrastructure, effectively by devoting the vast majority of the processor pool to operate on the few high‐intensity (and rate‐limiting) high‐resolution (HR) grid columns. Two examples prove the concept, showing that minor artifacts can be introduced near the HR/low‐resolution CRM grid transition boundary on idealized aquaplanets, but are minimal in operationally relevant real‐geography settings. As intended, within the high (low) resolution area, our Multi‐Domain CRM simulations exhibit cloud fraction and shortwave reflection convergent to standard baseline tests that use globally homogenous MMF‐LR and MMF‐HR. We suggest this approach can open up a range of creative multi‐resolution climate experiments without requiring unduly large allocations of computational resources.

 

Subkilometer processes are critical to the physics of aerosol‐cloud interaction (ACI) but have been dependent on parameterizations in global model simulations. We thus report the strength of ACI in the Ultra‐Parameterized Community Atmosphere Model (UPCAM), a multiscale climate model that uses coarse exterior resolution to embed explicit cloud‐resolving models with enough resolution (250 m horizontal, 20 m vertical) to quasi‐resolve subkilometer eddies. To investigate the impact on ACIs, UPCAM's simulations are compared to a coarser multiscale model with 4 km horizontal resolution. UPCAM produces cloud droplet number concentrations (N_d) and cloud liquid water path (LWP) values that are higher than the coarser model but equally plausible compared to observations. Our analysis focuses on the Northern Hemisphere (20–50°N) oceans, where historical aerosol increases have been largest. We find similarities in the overall radiative forcing from ACIs in the two models, but this belies fundamental underlying differences. The radiative forcing from increases in LWP is weaker in UPCAM, whereas the forcing from increases inN_dis larger. Surprisingly, the weaker LWP increase is not due to a weaker increase in LWP in raining clouds, but a combination of weaker increase in LWP in nonraining clouds and a smaller fraction of raining clouds in UPCAM. The implication is that as global modeling moves toward finer than storm‐resolving grids, nuanced model validation of ACI statistics conditioned on the existence of precipitation and good observational constraints on the baseline probability of precipitation will become key for tighter constraints and better conceptual understanding.

 

Abstract For the Community Atmosphere Model version 6 (CAM6), an adjustment is needed to conserve dry air mass. This adjustment exposes an inconsistency in how CAM6’s energy budget incorporates water—in CAM6 water in the vapor phase has energy, but condensed phases of water do not. When water vapor condenses, only its latent energy is retained in the model, while its remaining internal, potential, and kinetic energy are lost. A global fixer is used in the default CAM6 model to maintain global energy conservation, but locally the energy tendency associated with water changing phase violates the divergence theorem. This error in energy tendency is intrinsically tied to the water vapor tendency, and reaches its highest values in regions of heavy rainfall, where the error can be as high as 40 W m −2 annually averaged. Several possible changes are outlined within this manuscript that would allow CAM6 to satisfy the divergence theorem locally. These fall into one of two categories: 1) modifying the surface flux to balance the local atmospheric energy tendency and 2) modifying the local atmospheric tendency to balance the surface plus top-of-atmosphere energy fluxes. To gauge which aspects of the simulated climate are most sensitive to this error, the simplest possible change—where condensed water still does not carry energy and a local energy fixer is used in place of the global one—is implemented within CAM6. Comparing this experiment with the default configuration of CAM6 reveals precipitation, particularly its variability, to be highly sensitive to the energy budget formulation. Significance Statement This study examines and explains spurious regional sources and sinks of energy in a widely used climate model. These energy errors result from not tracking energy associated with water after it transitions from the vapor phase to either liquid or ice. Instead, the model used a global fixer to offset the energy tendency related to the energy sources and sinks associated with condensed water species. We replace this global fixer with a local one to examine the model sensitivity to the regional energy error and find a large sensitivity in the simulated hydrologic cycle. This work suggests that the underlying thermodynamic assumptions in the model should be revisited to build confidence in the model-simulated regional-scale water and energy cycles.