Journal of Advances in Modeling Earth Systems最新文献

Aerosol-cloud interactions are a persistent source of uncertainty in climate research. This study presents findings from a model intercomparison project examining the impact of aerosols on clouds and climate in convection-permitting radiative-convective equilibrium (RCE) simulations. Specifically, 11 different modeling teams conducted RCE simulations under varying aerosol concentrations, domain configurations, and sea surface temperatures (SSTs). We analyze the response of domain-mean cloud and radiative properties to imposed aerosol concentrations across different SSTs. Additionally, we explore the potential impact of aerosols on convective aggregation and large-scale circulation in large-domain simulations. The results reveal that the cloud and radiative responses to aerosols vary substantially across models. However, a common trend across models, SSTs, and domain configurations is that increased aerosol loading tends to suppress warm rain formation, enhance cloud water content in the mid-troposphere, and consequently increase mid-tropospheric humidity and upper-tropospheric temperature, thereby impacting static stability. The warming of the upper troposphere can be attributed to reduced lateral entrainment effects due to the higher environmental humidity in the mid-troposphere. However, models do not agree on aerosol impacts on convective updraft velocity based on the preliminary examination of high-percentiles of vertical velocity at a single mid-troposheric layer (500 hPa). In large-domain simulations, where convection tends to self-organize, aerosol loading does not consistently influence self-organization but tends to reduce the intensity of large-scale circulation forming between convective clusters and dry regions. This reduction in circulation intensity can be explained by the increase in static stability due to the upper tropospheric warming.

Climate models initialized near the observed state typically drift toward their own climatology as the forecast evolves. This drift is commonly corrected through a lead-time and start-month dependent bias adjustment, derived from a hindcast data set. While widely used, this traditional correction has well-known limitations: it is statistically inefficient, prone to introducing artificial discontinuities, and offers little insight into the underlying causes of forecast error. This paper presents an alternative framework that addresses these limitations and provides a more process-oriented diagnostic. The proposed method fits separate autoregressive models with exogenous input (ARX models) to both forecasts and observations. Forecast errors are then predicted and removed using the difference between the two ARX models. The method is demonstrated and compared to traditional methods using seasonal forecasts of global mean temperature from the SPEAR model, a contributor to the North American Multi-Model Ensemble (NMME). The ARX approach outperforms traditional methods in independent data even when traditional approaches include a linear trend correction. The analysis reveals that SPEAR exhibits an exaggerated response to radiative forcing, leading to significant trend errors. Notably, these errors are already present in the first month. These initial trend errors can be reproduced by a one-dimensional data assimilation system, indicating that they originate from SPEAR's exaggerated response to radiative forcing, which is carried forward into the first-guess fields used in the data assimilation system.

Land-atmosphere coupling involves multiple processes occurring across different temporal and spatial scales. These complex processes are not yet fully represented in climate or numerical weather prediction models. To evaluate these coupled models and improve or develop new parameterizations, four different single-column model setups are proposed. These setups vary in the type of spatially homogeneous land cover: grassland, wheat field, corn field, and pine forest. They are based on the single-column version of the CNRM-CM6-1 model, which couples the ARPEGE-Climat atmospheric component with the SURFEX surface modeling platform. First, the simulation results are compared with observational data, focusing on the radiative balance, surface fluxes, soil temperature and moisture, as well as air temperature, specific humidity, and wind speed near the surface. The evaluation shows that the ARPEGE–SURFEX single-column model can accurately replicate both surface and atmospheric conditions. Next, the same modeling framework is applied to the Meso-NH–SURFEX large-eddy simulation model. These simulations provide a consistent reference for evaluating the boundary-layer in CNRM-CM6-1, particularly in terms of potential temperature and specific humidity throughout the day. Finally, a sensitivity study on the impact of surface parameters on land-atmosphere coupling reveals that the hydric stress parameterization, by modifying canopy conductance and transpiration, is the most influential driver of temperature and moisture in the mixed layer. This work highlights the advantages of using the ARPEGE–SURFEX single-column configuration as a reliable platform for experimentation and parameterization improvement, and local climate modeling. The study cases developed are made available for further evaluation of land-atmosphere coupling.

Organic aerosol (OA) is an important constituent of the Earth's atmosphere, yet the extent of its destruction by photolysis remains an active research question. Recent laboratory studies reveal evidence for rapid short-term photolysis for secondary OA, but the rate declines to negligible levels over time. Here we use the stratosphere to investigate long-term OA photolysis because of the relatively simple sources and sinks of OA in this region. Airborne campaign observations show that the organic content in organic-sulfate aerosols remains stable with altitude and time in the stratosphere, indicating no significant photolysis. Satellite observations of the 2020 Australian wildfires reveal OA persists over a year in the stratosphere, consistent with model simulations excluding long-term photolysis. These findings suggest long-term OA photolysis is negligible in the real atmosphere. The current Community Earth System Model (CESM) significantly underestimates the abundance of stratospheric OA due to assumed rapid photolysis. We add this well-validated mechanism into CESM by turning off secondary OA photolysis after it is 50 days old, effectively simulating stratospheric OA consistent with observations. In summary, multiple lines of evidence confirm that the long-term photolysis of OA is negligible or extremely slow. Incorporating this mechanism into CESM addresses a key model deficiency, improving simulation of stratospheric OA.

The differential cycling of marine macromolecules, such as dissolved carbohydrates, proteins, and lipids play a fundamental role in marine microbial metabolisms, ultimately regulating the ocean's capacity to sequester carbon downwards via the biological pump or move carbon up the food chain, supporting marine food webs. To date, their representation in global scale models of marine biogeochemistry and ecosystems is lacking. Here we add explicit representation of marine macromolecular cycling for dissolved polysaccharides, lipids, amino polysaccharides, and proteins, within the dissolved organic matter pools of the Marine Biogeochemistry Library (MARBL) ecosystem model, implemented within the ocean circulation model of the Community Earth System Model. The resulting dissolved macromolecule distributions identify polysaccharide and amino polysaccharide accumulation within the upper ocean of the subtropics owing to longer lifetimes than dissolved lipids and proteins which are diagnosed with shorter lifetimes and accumulate in more biologically productive regions. Representation of marine macromolecules is found to better match observed constraints of dissolved organic carbon to nitrogen stoichiometry patterns with depth than its MARBL predecessor which considers only bulk pools. This implementation of macromolecular cycling within the marine dissolved organic matter pool represents an important next step in better characterizing the complexity of natural organic matter within the marine ecosystem.

Global climate models parameterize a range of atmospheric-oceanic processes, including gravity waves (GWs), clouds, moist convection, and turbulence, that cannot be sufficiently resolved. These subgrid-scale closures for unresolved processes are a substantial source of model uncertainty. Here, we present a new approach to developing machine learning (ML) parameterizations of small-scale climate processes by fine-tuning a pre-trained AI foundation model (FM). FMs are largely unexplored in climate research. A pre-trained encoder-decoder from a 2.3 billion parameter FM (NASA and IBM Research's Prithvi WxC)—which contains a latent probabilistic representation of atmospheric evolution—is fine-tuned (or reused) to create a deep learning parameterization for atmospheric gravity waves (GWs); a process unseen during pre-training. The parameterization captures GW effects for a coarse-resolution climate model by learning the fluxes from an atmospheric reanalysis with 10 times finer resolution. A comparison of monthly averages and instantaneous evolution with a machine learning model baseline (an Attention U-Net) reveals superior predictive performance of the FM parameterization throughout the atmosphere, even in regions excluded during pre-training. This performance boost is quantified using the Hellinger distance, which is 0.11 for the baseline and 0.06 for the fine-tuned model. Our findings emphasize the versatility and reusability of FMs, which could be used to accomplish a range of atmosphere- and climate-related applications, leading the way for the creation of observations-driven and physically accurate parameterizations for more earth system processes.

Accurately simulating clouds remains a key challenge in global climate models, primarily because cloud formation involves sub-grid processes that are parameterized and crudely represented in models. This study examines the performance of DOE's Simple Cloud-Resolving Energy Exascale Earth System (E3SM) Atmosphere Model (SCREAM) in simulating cloud properties and their spatio-temporal distribution by comparing against satellite observations. Two horizontal resolutions of SCREAM (3 and 12 km) are examined, and both depict a realistic spatial structure of mean-state cloud cover but underestimate its global mean magnitude. SCREAM 3 km reasonably reproduces the distribution of mean-state cloud properties across various cloud optical thickness and cloud-top pressure regimes, with performance comparable to CMIP5 and CMIP6 ensemble and marginally outperforming SCREAM 12 km. Still, SCREAM 3 km tends to underpredict low clouds and optically thin clouds, highlighting the need for continued improvement in representing unresolved processes. This study provides a basis for confidence in the representation of clouds in SCREAM, as simulating mean-state clouds is a necessary prerequisite for trusting its cloud responses to changes in aerosols and greenhouse gases.

Urban areas are increasingly vulnerable to the impacts of climate change, necessitating accurate simulations of urban climates in Earth system models (ESMs) in support of large-scale urban climate adaptation efforts. ESMs underrepresent urban areas due to their small spatial extent and the lack of detailed urban landscape data. To enhance the accuracy of urban representation, this study integrated the local climate zones (LCZs) scheme within the Community Earth System Model (CESM) to better represent urban heterogeneity. We adopted a modular approach to incorporate the 10 built LCZ classes into CESM as a new option in addition to the default urban three-class scheme (i.e., tall building district, high density, and medium density). CESM simulations using the LCZ-based urban characteristics were validated globally at 20 flux tower sites, showing site-averaged improvement in modeling upward longwave radiation ($��L�W_{\text{up}}�$) and anthropogenic heat flux ($��Q_{\text{ahf}}�$), but increased uncertainties in modeling sensible heat flux ($��Q_h�$). The root-mean-square error between the observed and simulated $��Q_{\text{ahf}}�$ using the LCZ decreased by 4% compared to using the default. Model sensitivity experiments revealed that $��L�W_{\text{up}}�$ and $��{}_{}$

The rapid loss of Arctic sea ice is a striking consequence of anthropogenic global warming. Its remote impacts on mid-latitude weather and climate have attracted scientific and media attention. In this study, we use a hybrid (dynamical plus machine-learning) atmospheric model—Google's NeuralGCM—to investigate the mid-latitude atmospheric circulation responses to Arctic sea-ice loss for the first time. We conduct experiments in which NeuralGCM is forced with pre-industrial and future sea-ice concentrations following the protocol of the Polar Amplification Model Intercomparisom Project. To assess the performance of NeuralGCM, we compare the results with those simulated by two physics-based climate models. NeuralGCM produces a comparable response of near-surface warming to sea-ice loss and the subsequent weakened zonal wind in mid-latitudes. However, there is a substantial discrepancy between the two models' stratospheric responses, where different temperature responses in these models are associated with different zonal wind and geopotential height responses. Further investigation of North Atlantic blocking shows that NeuralGCM produces stronger, more frequent, and more realistic blocking events. Our results demonstrate the capability of NeuralGCM in simulating the tropospheric responses to Arctic sea-ice loss, but improvements may be needed for the stratospheric representation.