Explore the vast range of titles available.

The new generation of heterogeneous CPU/GPU computer systems offer much greater computational performance but are not yet widely used for climate modeling. One reason for this is that traditional climate models were written before GPUs were available and would require an extensive overhaul to run on these new machines. In addition, even conventional “high–resolution” simulations don't currently provide enough parallel work to keep GPUs busy, so the benefits of such overhaul would be limited for the types of simulations climate scientists are accustomed to. The vision of the Simple Cloud‐Resolving Energy Exascale Earth System (E3SM) Atmosphere Model (SCREAM) project is to create a global atmospheric model with the architecture to efficiently use GPUs and horizontal resolution sufficient to fully take advantage of GPU parallelism. After 5 years of model development, SCREAM is finally ready for use. In this paper, we describe the design of this new code, its performance on both CPU and heterogeneous machines, and its ability to simulate real‐world climate via a set of four 40 day simulations covering all 4 seasons of the year. Plain Language Summary This paper describes the design and development of a 3 km version of the Energy Exascale Earth System Model (E3SM) atmosphere model, which has been fully rewritten in C++ using the Kokkos library for performance portability. This newly rewritten model is able to take advantage of the state–of–the–science high performance computing systems which use graphical processor units (GPUs) to mitigate much of the computational expense which typically plagues high–resolution global modeling. Taking advantage of this high–performance we are able to run four seasons of simulations at 3 km global resolution. We discuss the biases, including the diurnal cycle, by comparing model results with satellite and Atmospheric Radiation Measurement ground‐based site data. Key Points Describes the C++/Kokkos implementation of the Simple Cloud–Resolving E3SM Atmosphere Model (SCREAMv1) SCREAMv1 leverages GPUs to surpass one simulated year per compute day at global 3 km resolution High resolution improves some meso‐scale features and the diurnal cycle but large‐scale biases require improvement across all four seasons

Cirrus dominate the longwave radiative budget of the tropics. For the first time, the variability in cirrus properties and longwave cloud radiative effects (CREs) that arises from using different microphysical schemes within nudged global storm‐resolving simulations from a single model, is quantified. Nudging allows us to compute radiative biases precisely using coincident satellite measurements and to fix the large‐scale dynamics across our set of simulations to isolate the influence of microphysics. We run 5‐day simulations with four commonly‐used microphysics schemes of varying complexity (SAM1MOM, Thompson, M2005 and P3) and find that the tropical average longwave CRE varies over 20 W m−2 between schemes. P3 best reproduces observed longwave CRE. M2005 and P3 simulate cirrus with realistic frozen water path but unrealistically high ice crystal number concentrations which commonly hit limiters and lack the variability and dependence on frozen water content seen in aircraft observations. Thompson and SAM1MOM have too little cirrus. Plain Language Summary Recently, advancements in computing have made it possible for atmospheric scientists to simulate Earth's global atmosphere with higher resolution than ever before. This new generation of models, called global‐storm resolving models, have a horizontal grid spacing of just a few kilometers, which permits the formation of thunderstorms. As a result, they simulate clouds more realistically than traditionally climate and weather models and are a great tool for diagnosing cloud biases in atmospheric models. Here, we run a single global storm‐resolving model with four different representations of cloud physics called M2005, P3, SAM1MOM and Thompson. We evaluate simulated tropical cirrus, which are stratiform ice clouds at the top of the troposphere that reduce the amount of infrared radiation emitted by the Earth, with satellite and aircraft data to see which representations have the best performance. SAM1MOM and Thompson make too little cirrus causing too much infrared radiation to be emitted, M2005 makes too much cirrus, causing too little infrared radiation to be emitted, and P3 makes about the right amount. Key Points Nudged global storm‐resolving simulations are valuable for microphysics sensitivity studies Mean tropical longwave cloud radiative effect varies over 20 W m−2 depending on microphysics scheme Two‐moment schemes outperform simpler one‐moment and partial double‐moment schemes, and P3 has the smallest longwave radiative bias

Accurately representing the precipitation diurnal cycle has long been a challenge for global climate models (GCMs). Here we evaluate the precipitation diurnal cycle in the DYAMOND global convection‐permitting models (CPMs) and CMIP6 HighResMIP models. Comparison of the high‐ (25–50 km) and low‐resolution (100–250 km) models with parameterized convection in HighResMIP shows that simply increasing model resolution does not noticeably improve the precipitation diurnal cycle. In contrast, CPMs can better capture the observed amplitude and timing of precipitation diurnal cycle. However, the simulated spatial variation of timing in CPMs is smaller than observed, leading to an exaggeration of the spatially averaged diurnal amplitude. The better‐simulated precipitation diurnal cycle in the CPMs is tied to mesoscale convective systems (MCSs), which contribute about half of the total precipitation. The observed life cycle of MCSs, including initiation and mature stages, is well captured in the CPMs, leading to a more realistic precipitation diurnal cycle. Plain Language Summary As a basic mode of climate variability, the diurnal cycle is a key metric that has been used to benchmark climate models. The current state‐of‐the‐art GCMs struggle to accurately represent the precipitation diurnal cycle, frequently peaking too early compared to the observations. Due to the coarse resolution and the use of convection parameterization, GCMs are also unable to simulate organized convective storms, which often exhibit a distinct diurnal cycle. With the emergence of global storm‐resolving models at kilometer‐scale resolution, this study evaluates the diurnal cycle of precipitation simulated by global storm‐resolving models and high‐resolution GCMs. We find that at resolutions between ∼25 and 250 km, increasing resolution has limited effects on the precipitation diurnal cycle but global CPMs can reproduce the observed precipitation diurnal cycle much better as they can better represent organized convective storms. Key Points Precipitation diurnal cycle is systematically evaluated in DYAMOND simulations from global convection‐permitting models (CPMs) and CMIP6 models Both the amplitude and phase of the precipitation diurnal cycle are better simulated by CPMs than CMIP6 high‐resolution models The global CPMs excel in reproducing the precipitation diurnal cycle owing to better simulated mesoscale convective systems

Future climate projections focusing on precipitation and water resource trends over South America (SA) are investigated using two ensembles. One of them is composed of three global climate models (GCMs), and the other of eight regional climate models (RCMs) from the Coordinated Regional Climate Downscaling Experiment (CORDEX). The present (1970–2005) and the future (2006–2100) climate trends are analyzed for representative pathway scenarios 4.5 (RCP4.5) and 8.5 (RCP8.5). For the most pessimistic scenario (RCP8.5), trends in water resources are assessed considering the terrestrial branch of the hydrologic cycle by analyzing the precipitation minus evapotranspiration (P-ET). For the present climate, RCMs added value to the GCMs in simulating more realistic precipitation fields in several regions. GCMs and RCMs project, in general, the same precipitation change signal for the end of the 21st century over SA, which is stronger in RCP8.5 than in RCP4.5. For RCP8.5 in most regions, GCMs and RCMs ensembles have the same precipitation trend signal, but a great spread between the ensemble members, which is greater in austral summer than winter, can be noted. In winter a negative trend in rainfall in most members and regions predominates. At the end of the 21st century, relative changes in rainfall in RCP8.5 are in the range of +14% (over northeastern Brazil in summer) to − 36% (over the Andes Mountains in winter). In RCP8.5, the ensembles project an increase in air temperature with a similar magnitude, while in RCP4.5 the trends are weaker. For air temperature, there is small spread between members, and the positive trend is statistically significant for all ensemble members in the RCP8.5 scenario. In terms of water resources, on an annual scale, for RCP8.5 the RCM ensemble projects a larger area with wetter conditions in the future than GCMs. Regionally, it is expected a decrease in water availability in the Amazon basin and an increase over northeast Brazil and southeast SA during the summer. In other regions (northern Amazon, the Andes Mountains and Patagonia) the ensembles indicate drier conditions in the future winter, except in southern Amazon. It is expected that such information could be useful for devising adaptation and mitigation policies due to climate change over the SA.

There is a lack in representation of biosphere–atmosphere interactions in current climate models. To fill this gap, one may introduce vegetation dynamics in surface transfer schemes or couple global climate models (GCMs) with vegetation dynamics models. As these vegetation dynamics models were not designed to be included in GCMs, how are the latest generation dynamic global vegetation models (DGVMs) suitable for use in global climate studies? This paper reviews the latest developments in DGVM modelling as well as the development of DGVM–GCM coupling in the framework of global climate studies. Limitations of DGVM and coupling are shown and the challenges of these methods are highlighted. During the last decade, DGVMs underwent major changes in the representation of physical and biogeochemical mechanisms such as photosynthesis and respiration processes as well as in the representation of regional properties of vegetation. However, several limitations such as carbon and nitrogen cycles, competition, land-use and land-use changes, and disturbances have been identified. In addition, recent advances in model coupling techniques allow the simulation of the vegetation–atmosphere interactions in GCMs with the help of DGVMs. Though DGVMs represent a good alternative to investigate vegetation–atmosphere interactions at a large scale, some weaknesses in evaluation methodology and model design need to be further investigated to improve the results.

Global Water Models (GWMs), which include Global Hydrological, Land Surface, and Dynamic Global Vegetation Models, present valuable tools for quantifying climate change impacts on hydrological processes in the data scarce high latitudes. Here we performed a systematic model performance evaluation in six major Pan-Arctic watersheds for different hydrological indicators (monthly and seasonal discharge, extremes, trends (or lack of), and snow water equivalent (SWE)) via a novel Aggregated Performance Index (API) that is based on commonly used statistical evaluation metrics. The machine learning Boruta feature selection algorithm was used to evaluate the explanatory power of the API attributes. Our results show that the majority of the nine GWMs included in the study exhibit considerable difficulties in realistically representing Pan-Arctic hydrological processes. Average APIdischarge (monthly and seasonal discharge) over nine GWMs is > 50% only in the Kolyma basin (55%), as low as 30% in the Yukon basin and averaged over all watersheds APIdischarge is 43%. WATERGAP2 and MATSIRO present the highest (APIdischarge > 55%) while ORCHIDEE and JULES-W1 the lowest (APIdischarge ≤ 25%) performing GWMs over all watersheds. For the high and low flows, average APIextreme is 35% and 26%, respectively, and over six GWMs APISWE is 57%. The Boruta algorithm suggests that using different observation-based climate data sets does not influence the total score of the APIs in all watersheds. Ultimately, only satisfactory to good performing GWMs that effectively represent cold-region hydrological processes (including snow-related processes, permafrost) should be included in multi-model climate change impact assessments in Pan-Arctic watersheds.

The extension of a cloud‐resolving model, the System for Atmospheric Modeling (SAM), to global domains is described. The resulting global model, gSAM, is formulated on a latitude‐longitude grid. It uses an anelastic dynamical core with a single reference profile (as in SAM), but its governing equations differ somewhat from other anelastic models. For quasihydrostatic flows, they are isomorphic to the primitive equations (PE) in pressure coordinates but with the globally uniform reference pressure playing the role of actual pressure. As a result, gSAM can exactly maintain steady zonally symmetric baroclinic flows that have been specified in pressure coordinates, produces accurate simulations when initialized or nudged with global reanalyses, and has a natural energy conservation equation despite the drawbacks of using the anelastic system to model global scales. gSAM employs a novel treatment of topography using a type of immersed boundary method, the Quasi‐Solid Body Method, where the instantaneous flow velocity is forced to stagnate in grid cells inside a prescribed terrain. The results of several standard tests designed to evaluate the accuracy of global models with and without topography as well as results from real Earth simulations are presented. Plain Language Summary The System for Atmospheric Modeling (SAM), a model widely used to study evolution of clouds and small‐scale atmospheric motions using computational grids with horizontal spacings of several meters to a few hundred kilometers, has been extended to simulations of the whole Earth with global grid spacings of 1–5 km that realistically represent intense and localized rain and snow storms. The resulting global atmosphere model, gSAM, is unique in its use of a computationally efficient approximation to the exact equations of air motion and its method of representing flow over mountains, which is well‐suited to steeply sloped terrain. The paper documents gSAM's updated equations, representation of physical processes, and computational methods. Despite its algorithmic simplicity, gSAM produces realistic global weather forecasts and cloud distributions and agrees well with other more complicated global models on standard benchmarking tests. Key Points The anelastic System for Atmospheric Modeling is extended to a global latitude‐longitude grid The model uses a novel treatment of topography forcing the flow to instantaneously stagnate in the cells inside a prescribed terrain The model performs well in several standard tests that evaluate the accuracy of global models and in real Earth simulations

The CALIOP (Cloud‐Aerosol Lidar with Orthogonal Polarization) layer product is used for a multimodel evaluation of the vertical distribution of aerosols. Annual and seasonal aerosol extinction profiles are analyzed over 13 sub‐continental regions representative of industrial, dust, and biomass burning pollution, from CALIOP 2007–2009 observations and from AeroCom (Aerosol Comparisons between Observations and Models) 2000 simulations. An extinction mean height diagnostic (Zα) is defined to quantitatively assess the models' performance. It is calculated over the 0–6 km and 0–10 km altitude ranges by weighting the altitude of each 100 m altitude layer by its aerosol extinction coefficient. The mean extinction profiles derived from CALIOP layer products provide consistent regional and seasonal specificities and a low inter‐annual variability. While the outputs from most models are significantly correlated with the observed Zα climatologies, some do better than others, and 2 of the 12 models perform particularly well in all seasons. Over industrial and maritime regions, most models show higher Zα than observed by CALIOP, whereas over the African and Chinese dust source regions, Zα is underestimated during Northern Hemisphere Spring and Summer. The positive model bias in Zα is mainly due to an overestimate of the extinction above 6 km. Potential CALIOP and model limitations, and methodological factors that might contribute to the differences are discussed. Key Points Mean regional tropospheric aerosol extinction profiles are calculated from CALIOP data. An extinction mean height diagnostic is defined. The performance of 12 global models in simulating the aerosol profiles is evaluated.

Realistically representing the present-day characteristics of extreme precipitation has been a challenge for global climate models, which is due in part to deficiencies in model resolution and physics, but is also due to a lack of consistency in gridded observations. In this study, we use three observation datasets, including gridded rain gauge and satellite data, to assess historical simulations from sixteen Coupled Model Intercomparison Project Phase 6 (CMIP6) models. We separately evaluate summer and winter precipitation over the United States (US) with a comprehensive set of extreme precipitation indices, including an assessment of precipitation frequency, intensity and spatial structure. The observations exhibit significant differences in their estimates of area-average intensity distributions and spatial patterns of the mean and extremes of precipitation over the US. In general, the CMIP6 multi-model mean performs better than most individual models at capturing daily precipitation distributions and extreme precipitation indices, particularly in comparison to gauge-based data. Also, the representation of the extreme precipitation indices by the CMIP6 models is better in the summer than winter. Although the 'standard' horizontal-resolution can vary significantly across CMIP6 models, from ∼0.7° to ∼2.8°, we find that resolution is not a good indicator of model performance. Overall, our results highlight common biases in CMIP6 models and demonstrate that no single model is consistently the most reliable across all indices.