Coupled model sea ice loss experiments have been a primary tool for isolating and assessing the impact of sea ice loss on the global climate. However, recent work cautioned that the methods used to constrain sea ice introduce artificial heating, leading to spurious polar warming. In previous work, artificial heating in the widely used ghost flux, albedo reduction, and sea ice nudging methods was studied using idealized models, such as the energy balance model and slab ocean model. The extent to which the results from previous work can be applied to the comprehensive coupled models remains unclear. Here, we apply energetic frameworks to assess artificial heating in the coupled sea ice loss experiments. Artificial heating in the ghost flux and nudging method experiments manifests as a significant residual in the energy budget of the coupled climate system. The same applies to the ice coupling method experiment, which constrains sea ice by modifying surface fluxes. In the albedo reduction method experiments, artificial heating manifests as an excessive surface shortwave radiation response. Across different experiments, the global-mean artificial heating is about 0.04-0.18 W m22 per 106 km2 of sea ice loss. Its impact on the surface energy balance is comparable to the increased net surface shortwave radiation from sea ice loss, suggesting that artificial heating could lead to spurious warming in these experiments. Our results show how artificial heating can be quantified in the coupled sea ice loss experiments and how its impact on the global climate can be interpreted through energetic frameworks. SIGNIFICANCE STATEMENT: Coupled model sea ice loss experiments, widely used to study the climate impacts of polar sea ice loss, have recently been shown to involve artificial heating from the methods used to constrain sea ice. While previous studies examined its effects using idealized models, it remains unclear how these results hold in fully coupled models. Using energetic frameworks, we quantify artificial heating across multiple fully coupled sea ice loss experiments. We show artificial heating has impacts comparable to increased net shortwave radiation due to sea ice loss on the climate response in these experiments. Our results offer an assessment of coupled sea ice loss experiments thus far, providing insights for the interpretation of their climate responses and future modeling strategies.

AI models have emerged as potential complements to physics-based models, but their skill in capturing observed regional trends with important societal impacts remains unexplored. Here, we benchmark satellite-era regional thermodynamic trends, including extremes, in an AI emulator (ACE2) and a hybrid model (NeuralGCM), against physics-based models and ERA5. Both AI models capture regional temperature trends such as satellite-era Arctic warming. ACE2 outperforms other models in capturing midlatitude vertical temperature trends. However, the AI models do not capture trends in heat extremes over the US Southwest. Furthermore, they do not capture drying trends in arid regions, but generally outperform physics-based models. Our results show that a data-driven AI emulator can perform comparably to, or better than, hybrid and physics-based models in capturing regional thermodynamic trends. We also find that ACE2 learns much of the signal from .

The northward displacement of the Intertropical Convergence Zone is a defining feature of the tropical climate that global models struggle to reproduce. While often attributed to interhemispheric energy imbalance, other processes may also shape its annual-mean structure. Here we use a hierarchy of models-from a fully coupled model to simplified frameworks-to examine how seasonal insolation variations influence the symmetry of the tropical rain band. Using idealized obliquity experiments alongside observations from 1979 to 2024, we show that amplifying seasonal insolation, while preserving hemispheric symmetry in the annual mean, shifts the rain belt towards a more symmetric state. Enhanced austral-summer heating raises southern tropical sea surface temperatures above a convective threshold, triggering a transient southern rain band amplified by wind-evaporation-surface temperature feedback, while convection remains suppressed outside this season south of the Equator. This seasonally selective response thus affects the annual mean. The rain-belt position thus reflects the nonlinear integration of seasonal dynamics rather than a simple response to annual-mean forcing. Consistently, many models overestimate southeastern tropical Pacific sea surface temperature seasonality, contributing to the double rain-belt bias. Our findings highlight the need for improved representation of seasonal processes in models to better capture annual-mean rainfall patterns.

Reanalysis data reveal significant summertime stationary wave trends across Eurasia with implications for regional warming and extremes, yet the underlying physical mechanism remains unclear. Here we investigate the connection between these trends and European aerosol emissions using a thermal wind balance framework linking stationary waves to surface air temperature and lapse rate gradient trends. The reanalysis stationary wave trends are driven by surface air temperature gradients shaped by shortwave radiation and aerosol emission trends. In idealized European aerosol forcing experiments and historical aerosol single-forcing simulations, stationary wave trends resembling the reanalysis pattern emerge. However, attribution to aerosol forcing is limited because all-forcing simulations systematically underestimate the reanalysis trend magnitude. The thermal wind balance framework identifies that the underestimation is related to the surface air temperature gradient contribution being excessively offset by the lapse rate gradient contribution upstream and being too weak downstream, in climate models. This mechanistic decomposition offers a pathway to diagnose the robust reanalysis–model circulation discrepancy in the region, and suggests that aerosol forcing could play a more important role in regional climate than current models indicate.

Stratospheric sudden warmings (SSWs) predominantly occur in the Northern Hemisphere (NH) with only 1 major event recorded in the Southern Hemisphere in the satellite era. Investigating factors that contribute to this asymmetry can help to reveal the cause of SSWs and lead to improved forecasts. Here we use climate model simulations to investigate the impact of boundary conditions (topography and ocean circulation) on the hemispheric asymmetry. Flattening topography eliminates NH SSWs, while removing the ocean meridional overturning circulation reduces their frequency by half. The SSW response to boundary conditions is controlled by the hemispheric asymmetry of eddy heat flux. The reduction is driven by a decrease in amplitude of both eddy meridional wind and eddy temperature, as well as an increase in the cosine of the difference between their phases. The results suggest boundary conditions play an important role in shaping SSWs, especially topographic forcing, but that the boundary condition interactions are nonlinear.

As Earth warms, regional climate signals are accumulating. Some signals, for example, land warming more than the ocean and the Arctic warming the most, were expected and successfully predicted. Underlying this success was the application of physical laws under the assumption that large and small spatial scales are well separated. This established what we call the standard approach, climate science's dominant paradigm. With additional warming, however, discrepancies between real-world signals and expectations based on this standard approach are piling up, especially at regional scales. At the same time, disruptive computational approaches are advancing new paradigms. Philosophers of science characterize situations where accumulating discrepancies (anomalies) and disruptions lead to a loss of confidence in the dominant paradigm as a 'crisis'. Here we articulate what we consider to be the dominant paradigm, or standard approach, and the discrepancies and disruptions that have emerged in recent years. The policy implications of a purported crisis are discussed, as well as paths forward, crisis or no crisis. These paths include using signals to test assumptions and processes driving a warming Earth for the first time, developing testable hypotheses, and revitalizing conceptual thinking by filling gaps across climate-system components and spatial scales.

Evidence has emerged of a discrepancy in tropical Pacific sea surface temperature (SST) trends over the satellite era, where most coupled climate models struggle to simulate the observed La Niña-like SST trends. Here we highlight wider implications of the tropical Pacific SST trend discrepancy for global circulation trends during boreal winter, using two complementary methods to constrain coupled model SST trends: conditioning near-term climate prediction (hindcast) simulations, and pacemaking coupled climate simulations. The robust circulation trend response to constraining the tropical Pacific SST trend resembles the interannual La Niña response. Constraining tropical Pacific SST robustly reduces tropical tropospheric warming, improving agreement with reanalyses, and moderately shifts the zonal-mean jets poleward. It also improves surface air temperature and precipitation trends in ENSO-sensitive regions, such as the Americas, South Asia, and southern Africa. Our results underline the importance of tropical Pacific SST for achieving confidence in multidecadal model projections.

Emission of anthropogenic greenhouse gases has resulted in greater Arctic warming compared to global warming, known as Arctic amplification (AA). From an energy-balance perspective, the current Arctic climate is in radiative-advective equilibrium (RAE) regime, in which radiative cooling is balanced by advective heat flux convergence. Exploiting a suite of climate model simulations with varying carbon dioxide (CO2 ${\text{CO}}_{2}$) concentrations, we link the northern high-latitude regime variation and transition to AA. The dominance of RAE regime in northern high-latitudes under CO2 ${\text{CO}}_{2}$ reduction relates to stronger AA, whereas the RAE regime transition to non-RAE regime under CO2 ${\text{CO}}_{2}$ increase corresponds to a weaker AA. Examinations on the spatial and seasonal structures reveal that lapse-rate and sea-ice processes are crucial mechanisms. Our findings suggest that if CO2 ${\text{CO}}_{2}$ concentration continues to rise, the Arctic could transition into a non-RAE regime accompanied with a weaker AA.

Much has been learned about the response of the mean circulation under climate change. In particular, the subtropical jet will accelerate, the eddy-driven jet will shift poleward, storminess in the Southern Hemisphere will increase whereas storminess in the Northern Hemisphere will be impacted by a tug of war between different factors. However, very little is known about how circulation extremes will respond to climate change beyond blocking. This is in stark contrast to our understanding of the response of extreme temperatures, which follow the mean via an additive increase, and the response of extreme precipitation, which increase faster than the mean because of a multiplicative increase connected to the non-linear Clausius-Clapeyron relation. Here as a starting point, we investigate changes in upper-level circulation extremes defined using a daily distribution. We show fast upper-level jet stream (zonal) winds get faster under climate change. We also show extreme jet stream meandering or waviness (meridional wind) increases under climate change. These responses are geostrophic, robust across a climate model hierarchy (CMIP/AMIP/AQUA), and not connected to sea ice loss. The increase in upper-level circulation extremes is shown via moist thermal wind to be related to a multiplicative response connected to the non-linear Clausius-Clapeyron relation. Thus, upper-level circulation extremes exhibit a multiplicative increase similar to precipitation extremes. The results can be used to explain projected changes in commercial flight times, record-breaking winds, clear-air turbulence and a potential increase in severe weather occurrence under climate change.

The upper-level jet stream exhibits a robust increase in strength and shear under climate change. Previous work also noted a fast-get-faster response and connected it diagnostically to the Clausius-Clapeyron relation. Here we derive a moist adiabatic scaling that explains the upper-level jet stream wind response. Given the daily surface air temperature distribution and assuming a moist adiabatic atmosphere, the upper-level mean and fast jet stream wind increase by ∼2%/K and the jet stream shear increases by ∼4%/K across a climate model hierarchy. The scaling shows the increase of the surface moisture gradient following the Clausius-Clapeyron relation dominates the response. The scaling connects the increasing surface moisture gradient to the upper-level temperature gradient thereby reconciling dry and moist perspectives. The results show record-breaking upper-level jet stream wind and increased clear-air turbulence are tied to the Clausius-Clapeyron relation and are therefore robust and well-understood consequences of climate change.

Reanalysis data show the summertime circulation in the Northern Hemisphere midlatitudes has weakened significantly in the satellite era. Recent work shows the circulation weakening is not significantly affected by Arctic Amplification and Arctic Sea ice loss, but did not examine the role of other anthropogenic forcings such as aerosols. Here we use Detection and Attribution Model Intercomparison Project (DAMIP) simulations, which capture the weakening trend in reanalysis data, to quantify the impact of anthropogenic forcing due to aerosols and greenhouse gases. The DAMIP simulations show aerosol forcing dominates the weakening of the circulation across the Eurasia-Pacific sector, including the Pacific jet and storm track. Aerosol and greenhouse gases contribute equally to weakening the Atlantic jet and storm track. We use an energetic framework to understand the impact of aerosols on the storm track. In particular we show aerosol forcing leads to an increasing surface shortwave radiation trend over Western Europe and a decreasing surface shortwave trend over South and East Asia. These shortwave trends induce a weakening trend of the equator-to-pole energy gradient that leads to a weaker downstream storm track. Overall, our results show aerosol forcing is a dominant factor in regional circulation trends during Northern Hemisphere summertime in the satellite era. They have important implications for interpreting summertime heatwave trends in the Northern Hemisphere midlatitudes during summertime.

General Circulation Models (GCMs) are widely used to understand our climate and to simulate and predict the effects of global warming, revealing the dynamical convergence of storm tracks and jet streams at horizontal grid spacing of 50 km (e.g., Lu et al. 2015). Nevertheless, they have shown persistent biases in the large-scale features of the general circulation and basic climate statistics, which are attributed mainly to the parameterization, specifically, convection parameterization. To address this, Global storm-resolving models (GSRMs) provide an alternative approach to parameterization by explicitly resolving convection and its interaction with other processes, through the refinement of the horizontal grid, thus, offering new insights into the climate system. In a prior study, we showed the physical convergence of the tropical and general circulation structure at horizontal grid spacing of 2.5 km using aquaplanets. However, questions linger: Does the response under climate change of the storm tracks and jet streams converge at similar horizontal grid spacing, and what mechanism controls this convergence? We will present the effect of increasing horizontal grid spacing on the convergence of the storm tracks and jet stream location and intensity using the global storm-resolving model ICON. Control runs and idealised climate change experiments (increasing sea-surface temperature by 4 Kelvin) were conducted at horizontal grid spacing from 160 km to 2.5 km using an aqua-planet configuration. We adopt an aqua-planet configuration to focus on atmospheric phenomena, specifically convection and cloud feedback, meanwhile reducing the effect of complex interaction with land, topography, sea ice, and seasons. We will discuss the convergence rate of the eddy driven jet, subtropical jet, storm track, and large-scale circulation and their response to climate warming, characterised by the location, width, and intensity.

Anthropogenically forced climate change signals are emerging from the noise of internal variability in observations, and the impacts on society are growing. For decades, Climate or Earth System Models have been predicting how these climate change signals will unfold. While challenges remain, given the growing forced trends and the lengthening observational record, the climate science community is now in a position to confront the signals, as represented by historical trends, in models with observations. This review covers the state of the science on the ability of models to represent historical trends in the climate system. It also outlines robust procedures that should be used when comparing modeled and observed trends and how to move beyond quantification into understanding. Finally, this review discusses cutting-edge methods for identifying sources of discrepancies and the importance of future confrontations.

Trends in atmospheric circulation have begun to emerge in recent decades. Summertime mean circulation trends aloft have been attributed to human influence. For low-level extreme winds, the extent of human influence and climate model fidelity remains unclear. Here, we compare satellite-era trends in extratropical low-level mean and extreme (>90th percentile) winds defined using daily distribution in reanalyses and climate model simulations. In summer, Southern Hemisphere midlatitude winds have strengthened, driven by greenhouse gas and stratospheric ozone forcings. The summertime European wind stilling trend is dominated by aerosol and greenhouse gas forcings. In winter, models cannot capture the strengthening over the Southern Hemisphere and the weakening over Europe and the subtropical North Pacific. These discrepancies, particularly in the Pacific, are reduced but persist when observed sea surface temperatures are prescribed and affect the low-level baroclinicity. Our results highlight human influence on summertime low-level extreme wind trends and reveal regional wintertime discrepancies.

Earth’s upper-level jet streams influence the speed and direction of travel of weather systems and commercial aircraft, and are linked to severe weather occurrence. Climate change is projected to accelerate the average upper-level jet stream winds. However, little is known about how fast (>99th percentile) upper-level jet stream winds will change. Here we show that fast upper-level jet stream winds get faster under climate change using daily data from climate model projections across a hierarchy of physical complexity. Fast winds also increase ~2.5 times more than the average wind response. We show that the multiplicative increase underlying the fast-get-faster response follows from the nonlinear Clausius–Clapeyron relation (moist-get-moister response). The signal is projected to emerge in both hemispheres by 2050 when considering scenario uncertainty. The results can be used to explain projected changes in commercial flight times, record-breaking winds, clear-air turbulence and a potential increase in severe weather occurrence under climate change.

AbstractThe circulation response to climate change shapes regional climate and extremes. Over the last decade an increasing number of atmospheric circulation signals have been documented, with some attributed to human activities. The circulation signals represent an exciting opportunity for improving our understanding of dynamical mechanisms, testing our theories and reducing uncertainties. The signals have also presented puzzles that represent an opportunity for better understanding the circulation response to climate change, its contribution to climate extremes, interactions with moisture, and connection to thermodynamic discrepancies. The next decade is likely to be a golden age for dynamics with many advances possible.

Reanalysis data show a significant weakening of summertime circulation in the Northern Hemisphere (NH) midlatitudes in the satellite era with implications for surface weather extremes. Recent work showed the weakening is not significantly affected by changes in the Arctic, but did not examine the role of different anthropogenic forcings such as aerosols. Here we use the Detection and Attribution Model Intercomparison Project (DAMIP) simulations to quantify the impact of anthropogenic aerosol and greenhouse gas forcing. The DAMIP simulations show aerosols and greenhouse gases contribute equally to zonal-mean circulation weakening. Regionally, aerosol dominates the Pacific storm track weakening whereas greenhouse gas dominates in the Atlantic. Using a regional energetic framework, we show why the impact of aerosol is the largest in the Pacific. Reduced sulfate aerosol emissions over Eurasia and North America increase (clear-sky) surface shortwave radiation and turbulent fluxes. This enhances land-to-ocean energy contrast and energy transport via stationary circulations to the ocean. Consequently, energy converges poleward of oceanic storm tracks, demanding weaker poleward energy transport storm tracks, and the storm tracks weaken. The impact is larger over the Pacific following the larger emission decrease over Eurasia than North America. Similar yet opposite, increased aerosol emissions over South and East Asia decrease shortwave radiation and weaken land-to-ocean energy transport. This diverges energy equatorward of the Pacific storm track, further weakening it. Our results show aerosols are a dominant driver of regional circulation weakening during the NH summertime in the satellite era and a regional energetic framework explaining the underlying processes.

Earth's upper-level jet streams primarily flow in the eastward direction. They often exhibit a north-south component or waviness connected to extreme weather at the surface. Recently the upper-level eastward jet stream was found to exhibit a fast-get-faster response under climate change explained by the impact of the nonlinear Clausius-Clapeyron relation on the latitudinal density contrast. Here we show the fast-get-faster mechanism also applies to the upper-level north-south jet stream wind and the longitudinal density contrast, implying increased waviness under climate change. Arctic Sea ice loss, which has been proposed as a driver of increased waviness, cannot explain the response. It leads to a fast-get-slower waviness response at all vertical levels. We demonstrate the fast-get-faster waviness signal has emerged in reanalysis data in the Southern Hemisphere but not yet in the Northern Hemisphere. The results show the fast-get-faster mechanism explains upper-level waviness changes and highlights a tug of war between upper- and mid-level waviness under climate change. Climate change causes upper-level jet stream waviness to increase in both hemispheres, with extreme southward and northward excursions increasing faster than the mean, according to results from a fast-get-faster mechanism connecting waviness to density contrast changes.