Papers

Seasonal precipitation anomalies are largely regulated by atmospheric circulation, which dynamical models predict with greater reliability than precipitation itself. Here, we employ a deep learning model that translates dynamical circulation predictions into precipitation estimates. Predictions initialized from March to May consistently indicate a dry anomaly over central China in summer 2026. Retrospective evaluations revealed higher predictive skill in the analogue years, which also tended to feature central equatorial Pacific warming persisting from the preceding winter into summer. This warming favors an anomalous cyclonic circulation over the western North Pacific-South China Sea-South China region, which induces northerly winds and moisture divergence that jointly suppress rainfall over central China. Supporting this mechanism, layer-wise relevance propagation (LRP) independently identifies these northerly winds as the dominant driver of the prediction among all model inputs. Perturbation tests supported this attribution: removing LRP-identified features effectively eliminates the dry anomaly. Our framework thus provides physically interpretable explanations for AI-derived regional climate projections, facilitating evidence-based assessment before observational data become available.

Scenario planning for rare, high-impact events often requires massive ensembles to stochastically sample relevant trajectories. Although autoregressive weather emulators can efficiently generate such ensembles, isolating trajectories of interest requires sifting through petabytes of data, a challenge that grows exponentially with lead time and rarity. In contrast, a non-autoregressive foundation model like Climate in a Bottle video (cBottle-video) can directly sample trajectories terminating in extremes, avoiding large-ensemble search. We use cBottle-video to generate 1000-member ensembles with start- and/or end-conditioning across three extreme events—the 2021 Pacific Northwest (PNW) heatwave, Superstorm Sandy, and Hurricane Ian. Antecedent 500 hPa geopotential height ($z_{500}$) spread at the free end of end-conditioned ensembles reaches 84–89% of the final-state spread of start-conditioned ensembles, revealing substantial diversity consistent with each extreme event. For the 2021 PNW heatwave, end-conditioned ensemble members begin uniformly warmer than reanalysis and stay warm, replacing the observed rapid intensification with persistent antecedent heat. For Superstorm Sandy, the leading modes of $z_{500}$ at the antecedent end of the end-conditioned ensemble explain 44% of the variance in track latitude, and roughly 10% of ensemble members begin as stronger hurricanes than Sandy. For Hurricane Ian, variation in the first landfall location among end-conditioned trajectories underscores the importance of accounting for intermediate hazard exposure in risk planning.

Detection and attribution studies typically rely on free running climate models with modified anthropogenic forcings, yet they fail to reproduce key observed decadal trends. Despite reduced biases, higher predictive skill, large ensembles, and improved trends through frequent reinitialization, seasonal forecast models have not been used for this purpose. We introduce a new set of counterfactual seasonal hindcasts using the ECMWF coupled seasonal forecasting system based on alternative forcing scenarios by modifying atmospheric forcing and ocean and sea ice initial conditions. An observation based estimate of the forced ocean temperature signal is derived to amplify or remove this signal from the ocean initial conditions. Retrospective forecasts include a control configuration together with enhanced and reduced forcing experiments in which either CO2 increases together with doubled anthropogenic ocean warming or CO2 is fixed and the anthropogenic ocean-warming signal is removed. Additional experiments isolate aerosol forcing. The counterfactual hindcasts substantially alter long term temperature trends while largely preserving seasonal prediction skill, interannual variability, and model drift. Enhanced forcing strengthens several observed climate trends underestimated by the control, including top of atmosphere radiative fluxes and aspects of tropical Pacific atmospheric circulation. However, the coupled atmosphere ocean system fails to sustain the strengthened tropical Pacific temperature gradient through Bjerknes feedbacks, suggesting a fundamental model limitation. Aerosol forcing has little impact, likely because indirect aerosol effects are omitted. These results establish counterfactual seasonal hindcasts as a powerful tool for dynamic attribution and diagnosing model deficiencies in responses to anthropogenic climate forcing.

Climate model simulations indicate that surface ultraviolet (UV) radiation would change after soot injection into the stratosphere during a nuclear war, due to the competing effects of ozone depletion and aerosol attenuation. Using climate model simulations, we evaluate UV impacts under two scenarios: a regional India-Pakistan conflict producing 5 Tg of soot and a global U.S.-Russia war producing 150 Tg. UV enhancements due to ozone depletion substantially shorten safe outdoor exposure time, particularly for individuals with lighter skin types. Applying UV dose-response relationships to the year 2000 population data, without accounting for direct conflict mortality or famine-related population loss, the 5 Tg scenarios result in approximately 5,300-9,800 additional skin cancer deaths within 10-15 years and up to 75,000 cumulative excess deaths over the following century. In contrast, strong aerosol attenuation under the 150 Tg scenario initially suppresses surface UV, resulting in about 8,500 fewer skin cancer deaths within 15 years and a maximum cumulative reduction of approximately 17,000 deaths over the following century. These findings demonstrate that nuclear war-induced changes in surface UV radiation represent a persistent but previously understudied health impact of nuclear war. While skin cancer would not dominate overall mortality following a nuclear war, enhanced cumulative surface UV radiation represents an additional, long-lasting threat to human health that compounds other global impacts such as climate disruption and food insecurity. The excess UV may also pose negative impacts on animals and plants, including those used for agriculture, which remain to be quantified.

Regional climate zones are expected to shift under global warming. Whether AI climate models have learned to generalize climate-zone distributions under warming in a physically meaningful way affects their suitability for climate projection. We address this question by applying a Köppen-Geiger climate-zone decomposition to AIMIP Phase 1 models under prescribed +4K SST forcing and comparing their responses to physics-based AMIP models. Using this diagnostic, we compare baseline classification skill, per-zone responses in temperature, precipitation, and near-surface specific humidity, and the spatial structure of departures from physics-based models. All AI models considered reproduce the 1979-2014 ERA5 climatology within the physics-based models' range, but only the hybrid physics-AI model NeuralGCM-HRD reorganizes zones in agreement with established thermodynamic and hydrological scaling relations. The remaining emulators have distinct failure modes traceable to their architectural treatment of land cells. A physically consistent climate-zone response is therefore necessary for AI models intended for climate projection.

Anthropogenic heat flux, the heat released to the environment from human activities such as building energy use, transportation, industrial processes, and human metabolism, is a defining feature of the urban climate system. It is an important contributor to the urban heat island (UHI) effect and influences a wide range of urban meteorological processes. Its significance extends beyond urban climatology because of its close connections to energy consumption, greenhouse gas emissions, and climate adaptation. Despite decades of research, anthropogenic heat flux remains one of the least well-constrained components of the urban energy balance. Moreover, its climatic significance has often been assessed from an applied perspective, with less emphasis on developing transferable physical understanding of how anthropogenic heat flux acts as a forcing, how urban temperatures respond, and how feedbacks modify that response. This review develops a forcing-response-feedback framework for synthesizing current understanding of the role of anthropogenic heat flux in the urban climate system and identifies priorities for future research.

Optical wavelengths have received significant attention in free-space channels and are vital for applications in communications, imaging and sensing. Their widespread implementation is motivated by a variety of factors including improved bandwidths, increased security and higher energy efficiency. However, these advantages cannot be fully brought to bear in real-world scenarios due to the deleterious effects of atmospheric turbulence. Induced by small temperature and pressure fluctuations in the environment that vary rapidly in space and time, these effects cause significant power losses which decrease SNR, induce severe crosstalk in communication links, and greatly limit resolution of long-range imaging systems. To overcome this, we numerically and experimentally investigate the reconstruction of the transmission matrix of a time-evolving atmospheric channel with a real-time recursive optimization routine. We demonstrate that this estimation technique is able to keep up with the evolution of the channel and enables a significant improvement of communication-relevant quantities such as the coupling of the received light into a single-mode fiber while notably reducing the probability and duration of power outages, even in strong turbulence. Our results have immediate applications in free-space optical communication in both the classical and quantum regimes.

The generalised Lagrangian mean (GLM) theory of Andrews & McIntyre provides a powerful framework to study the interactions between waves and flows. A drawback of this theory is that the Lagrangian mean velocity is divergent even for incompressible fluids because the mean flow map, which sends the Lagrangian labels of fluid parcels to their mean positions, does not preserve volume. This results, for instance, in vortices shrinking under Lagrangian averaging. We overcome this drawback by revising the definition of the mean flow map, choosing it as the volume-preserving map closest to the "bare" GLM mean map. A standard result of optimal-transport theory then shows that the new mean map is the volume-preserving factor in the polar factorization of the GLM mean map. We develop and implement a numerical method for the computation of the corresponding Lagrangian mean fields from simulation data. The implementation builds on recently developed algorithms for the on-the-fly computation of Lagrangian means using the exponential and Butterworth filters. We demonstrate the value of volume-preserving Lagrangian averaging in simulations of the two-dimensional incompressible and shallow-water models. We compare the Lagrangian-mean fields obtained with and without the volume-preservation constraint.

Accurate simulation and prediction of ocean waves are essential for coastal risk management and climate studies. Deep learning has shown promising results for wave modeling, but most approaches still operate on regular grids and on forecasting time scales, and do not generalize to unstructured discretization or to long time horizons. Here we present WaveGraph, a model based on Graph Neural Networks (GNNs) that emulates basin-scale wave dynamics directly on unstructured meshes with high resolution along the coasts (up to 2-3 km). Trained on bias-corrected simulation data over the Mediterranean Sea, WaveGraph uses a multiscale architecture combining the unstructured model mesh with a uniform graph, allowing simultaneous representation of local coastal interactions and large-scale wave dynamics. The model reconstructs the evolution of significant wave height, mean period, and mean direction, and is applied autoregressively for a continuous 17-year period without reinitialization or drift. Validation against buoy and satellite observations shows skill comparable to the input data set, and ablation experiments indicate that wind forcing drives most of the long-term stability while wave history improves swell-driven and basin-scale dynamics. These results show that GNNs can provide stable and efficient emulators of spectral wave models on unstructured domains, enabling decadal wave reconstructions.

Air-quality and climate monitoring routinely convert native-resolution measurements into exceedance counts before testing whether high values cluster in time. A surrogate null can be imposed on the derived index or on the native series and then propagated through the same thresholding and aggregation. We show that the two placements can encode different scientific hypotheses and lead to opposite conclusions. In 31 Chilean PM2.5 monitoring stations (2018–2025), weekly clustering of days above the 50 ug m^-3 national 24-hour standard was significant at all stations under an index-level IAAFT null but at only two under a daily native-resolution null; a native generative null with seasonal marginals and short-memory Gaussian-copula dependence agreed with the native surrogate at 30 of 31 stations. Across thresholds spanning the WHO guideline (15 ug m^-3) to the Chilean episode-alert level (80 ug m^-3), index-only rejections increased from 8 of 31 to all eligible stations as monthly event probabilities became more seasonally concentrated, and equalizing monthly event probabilities at the 50 ug m^-3 level reduced index-only decisions from 29 to five. We formalize the issue as null transport through a lossy observation map: an exact aggregate-level equivalent of a native null exists if and only if the randomization kernel is lumpable with respect to that map. A within-fiber diagnostic, which perturbs native amplitudes while keeping the observed index exactly fixed, shows that this condition fails in the monitoring data: the native null distribution moved by a median of 18 times its Monte Carlo variability. Weekly clustering of PM2.5 exceedances in Chile is therefore fully compatible with seasonality plus short-memory daily persistence, and index-level surrogate rejections should not be read as evidence about daily concentration dynamics.

Mesoscale eddies are fundamental to the ocean circulation, yet the extent to which submesoscale motions, a few kilometers across, influence mesoscale eddy energetics through a kinetic energy cascade remains uncertain. High-resolution simulations predict that submesoscale fronts are key regulators of the cascade, transferring energy both downscale towards dissipation and upscale to sustain and shape the seasonality of mesoscale eddies. Testing these predictions has remained difficult because existing observations and state estimates cannot resolve submesoscale currents over sufficiently broad domains. Here we map the ocean's submesoscale energy cascade by combining multi-source satellite observations with a generative deep learning framework, reconstructing gap-free, kilometer-scale surface currents with physically plausible dynamics learned from simulations. Applying this to the eddy-rich Agulhas Current system, we find that submesoscales energize the mesoscale through an upscale energy cascade above 10 km, contributing to the seasonality of mesoscale eddies. Below 10 km, convergence at submesoscale fronts drives a downscale cascade towards dissipation. Both upscale and downscale pathways concentrate within fronts, where cross-scale transfer is up to an order of magnitude more efficient. Despite their limited extent, fronts account for a substantial fraction of the domain-integrated cascade, establishing them as key regulators of the cascade and targets for next-generation eddy parameterizations.

Turbulent fluxes between the surface and the atmosphere are typically parameterized using empirically fit relationships. Here we test machine learning techniques for fitting the relationship for the offshore environment. To do that, data from three offshore sites are used: the Martha's Vineyard Coastal Observatory (MVCO) air-sea interaction tower, the FINO1 research platform, and the CASPER-West FLIP research vessel deployed off the coast of California. Two machine learning methods were employed: Neural Networks (NN) and Random Forests (RF). Because the observational sites had towers with measurements at different levels, the vertical differences were input as gradients. Models were built for both momentum flux and heat flux. ML models trained at the individual sites were competitive with and in some cases, better than the physically-based COARE-3 model tailored to offshore fluxes. The heat flux ML models generally outperformed the physics-based parameterizations for most metrics, but the results were mixed for momentum flux, with only the site with the most training data (MVCO) producing results better than COARE-3. When the ML models from that site were applied to the other sites, results were degraded from using data from the site being tested. ML models built from data combined from the three sites generally showed improvements for the sites with less available training data. When assessing which variables were most important, the wind speed was most important for momentum flux and temperature gradient for heat flux.

Particulate matter with aerodynamic diameter less than 2.5 {\mu}m (PM2.5) is of great concern for human health. Here, for the first time, we examine the impact of sulfate aerosol geoengineering on PM2.5, using the output from the Geoengineering Large Ensemble (GLENS) project. GLENS is an ensemble of climate model simulations injecting SO2 into the stratosphere to balance RCP8.5 forcing using the Community Earth System Model, version 1. GLENS geoengineering reduces global averaged surface PM2.5 mass concentrations compared with RCP8.5 and also changes PM2.5 composition with more percentages of organic carbon and sulfate. The total reduction of PM2.5 is a result of less dust and sea salt concentrations. Dust emission is declined under GLENS geoengineering because of increased soil moisture and leaf area index over desert regions, and less emission of sea salt is due to slower wind speeds when geoengineering is applied. Excluding dust and sea salt, there is more global averaged PM2.5 under GLENS geoengineering relative to RCP8.5, predominantly due to more aerosol phase secondary organic aerosol (SOA). Since gas precursors of SOA are prescribed in the simulations, a cooler environment with geoengineering tends to transfer more gas phase SOA to aerosol phase. Changes in PM2.5 concentration and composition with applied geoengineering may have potential human health impact. Another new finding of this study is that the large amount of injected SO2 does not increase surface sulfate aerosol as a component of PM2.5, as the majority of sulfate aerosol reaching the boundary layer is in the coarse mode, which represents a very small fraction of the PM2.5. But the difference of deposition spatial distribution between geoengineering and RCP8.5 may have potential impacts on ecosystem.

Theoretical arguments have for decades anticipated substantial weakening of tropical convection as the surface warms, yet convection-resolving models show no robust change in column-integrated convective mass transport and an increase in transported volume. Here we reconcile this apparent discrepancy by demonstrating that the two main theoretical arguments identify a redistribution, rather than a straightforward reduction, of convective motions. A commonly invoked hydrological argument accordingly captures a weakening at a single altitude that is offset by strengthening aloft as the troposphere expands upward. Further, reduced convective mass flux following atmospheric structures that rise with warming simply reflects decreasing air density aloft. Convective motions and transported water are shown to be highly invariant to surface warming along isotherms, and the ability to predict convective responses at individual altitudes is a manifestation of this invariance. Together, these findings recast and simplify long-standing expectations of how the tropical atmosphere will respond to warming.

Quantifying forced and internal variability in climate data is fundamental to detecting the climate change signal and assessing uncertainty in climate projections. We propose a metric that quantifies the relative magnitude of internal variability in single-model initial-condition large ensembles (SMILEs). We measure forced and internal variability by two 1-Wasserstein distances, a form of optimal transport cost. The proposed metric is the distance for internal variability divided by the sum of the two. Computing this ratio requires only sorting the samples and differencing the resulting quantiles, with no additional parameters. The metric reflects the entire distribution shape and applies to non-Gaussian variables, because the 1-Wasserstein distance quantifies the difference between any two probability distributions. We validate the metric with synthetic climate data from Gaussian, uniform, and lognormal distributions. Unlike the two existing metrics, the proposed metric gives stable estimates irrespective of the distribution shape and the presence of outliers, provided that the ensemble has about 40 members or more. We then apply the proposed and existing metrics to the 2 m air temperature and total precipitation of the Community Earth System Model Large Ensemble (CESM-LE) under two different forcing scenarios. All the metrics indicate that the relative contribution of internal variability decreases as the forcing increases, but the proposed metric shows this response most clearly. The 1-Wasserstein distance thus provides a simple and useful tool for analyzing large-ensemble datasets.

Current understandings of the tropical cyclone (TC) warm core primarily relies on statistical averages from soundings and satellite products, which provide a climatological thermal state but cannot adequately characterize the continuous three-dimensional evolution of the warm core throughout the TC life cycle. In this study, Empirical Orthogonal Function (EOF) analysis was applied to three-dimensional temperature anomalies of Western North Pacific (WNP) TCs derived from the ERA5 reanalysis dataset for the period 1979-2024. We found that the three leading EOF modes effectively capture the primary characteristics of the TC thermal structure: the first mode represents the typical warm core structure; the second mode characterizes vertical baroclinicity; and the third pattern captures the horizontal asymmetry of the temperature anomalies. Using the first three principal components (PCs), we established a three-dimensional Cartesian coordinate system. Within this framework, two phase parameters define a thermal phase space to visualize the TC's three-dimensional thermal structure: one diagnoses the vertical barotropic or baroclinic structure, and the other quantifies the horizontal asymmetry of the thermal field. The results demonstrate that the trajectory in the phase diagram effectively captures the observed intensity changes and thermal structural evolution. The EOF-based phase diagrams offer promising insights and provide a novel, objective tool for analyzing and diagnosing structural thermodynamic characteristics and intensity evolution by quantifying key thermal features and their dynamic relationships with TC intensity. This capability thereby holds substantial potential for advancing both theoretical understanding and operational forecasting of TCs.

Which meteorological processes control exposure to fugitive gases downwind of a source, and on what timescales, have largely been inferred from dispersion theory and partial field evidence. Here we show that the meteorological drivers of elevated hydrogen sulphide (H$_2$S) exposure at a long-monitored European landfill, and the timescales over which each acts, can be identified directly from monitoring data. Wind direction, wind speed and atmospheric pressure form the causal core, with the share of directed information carried by pressure increasing with aggregation scale. The recovered timescales are consistent with those expected from the underlying atmospheric processes. We use these driver timescales to initialise CAIRN (Causal-Anchored Inference for Receptor Nowcasting), a machine-learning nowcaster with fast and slow memory components. Trained on past exceedances of WHO guideline levels, CAIRN nowcasts them from surface weather measurements and the calendar alone, without hand-engineered features. Combining four such nowcasters produces a site-level, tiered alert that agrees substantially with that generated by a direct sensor network and tracks an independent record of community odour reports. Meteorological variables can therefore serve as an inference-time proxy for exposure relative to WHO guideline levels, and they link atmospheric dynamics to community impact as an episode unfolds.

Transient weather systems are often associated with alternating warm and cold advections of air masses and changing wind directions, which drive coastal ocean and estuarine waters to oscillate quasi-periodically. Quantifying water transport under these meteorological oscillations between inland waterways and the coastal ocean helps interpret land-ocean interactions and the impact of migrating weather systems. The challenge lies in the difficulty of obtaining continuous, long-term direct measurements of transport due to logistical constraints. Here, we apply a method to determine the meteorological tide-induced volume transport of water through an intensive survey, correlating transport values measured by a boat-mounted ADCP with vertically averaged velocities from a bottom-mounted ADCP, which recorded a much longer time series. The correlation is then used to compute transport over the period of the bottom-mounted ADCP deployment. Observations were conducted at Belle Pass, Port Fourchon. The transport data revealed the impact of weather systems, including four cold fronts. A model of volume transport, accounting for rotary cold front wind variations, was applied, where both along-channel and along-coastline wind components contribute to the remote wind effect, leading to a more complex response to passing weather systems. The local wind effect is much smaller than the remote wind effect, and transport is primarily controlled by water level fluctuations resulting from open boundary input. Finally, the channel orientation relative to the coastline is found to be critical in determining both the magnitude and phase of the transport.

Infrasound signal period is widely used as a proxy for source energy in bolide studies, but the reliability of period-yield relations for small, high-altitude regional meteors has not been systematically evaluated. We analyze 90 infrasound detections from well-constrained regional meteoroid events (source altitudes 20-111 km, ranges 47-268 km) and demonstrate a strong, monotonic decrease in receiver dominant frequency with increasing source altitude (Spearman r_{s} = -0.629, p = 3.3 x 10^11). No detections above 80 km retain dominant frequencies exceeding 3 Hz, and 75% of detections above 100 km are dominated by sub-1 Hz content. Partial correlation analysis indicates this is primarily an altitude effect, not a propagation distance artifact. Near-source dissipation modeling using the generalized Burgers equation supports a physical mechanism: the exponential increase in kinematic viscosity with altitude drives the acoustic Reynolds number downward, imposing frequency-dependent molecular absorption that selectively attenuates high-frequency content within the first 5 km below the source. Our results suggest that this atmospheric low-pass filter systematically modulates the observed period and can bias period-based energy estimates upward by one to two orders of magnitude for sources above 80-90 km when uncorrected period-yield relations are applied. These findings are relevant to any high-altitude infrasound source, including space debris and controlled reentries.

Frontogenesis is the process by which ocean fronts–regions of strong horizontal density gradients in the upper ocean–strengthen. It is a highly non-linear phenomenon arising from many competing effects, such as background strain, instabilities, and surface forcing. Vertical mixing plays a complicated role in governing both frontogenesis and the long-term dynamics of ocean fronts. For weak fronts, it has been shown to drive shear dispersion which further weakens the front over long timescales. However, vertical mixing also mixes the vertically-sheared thermal wind velocity, driving the front out of balance and causing a secondary circulation to develop that may strengthen the front. Here, it is shown that vertical mixing can directly act to oppose the frontogenetic secondary circulation, provided the frontal width is larger than a critical value. A non-linear iteration method is described for finding balanced states in which vertical mixing arrests frontogenesis and the parameter dependence of this regime is discussed. Further, existing asymptotic solutions for the buoyancy and velocity fields are extended and found to closely match these balanced states, even outside their region of formal validity.

AI weather emulators benefit from clear objectives and metrics, which have led to the rapid development of models that outperform traditional benchmarks. In contrast, long-term climate emulators must reliably reproduce forced responses over months to centuries, while relying on training objectives that span a small number of model time steps. We assess autoregressive, full-depth ocean emulators using data from the midHolocene experiment of a numerical climate model to examine their skill in responding to surface forcings from an in-distribution, out-of-sample climate. We demonstrate that these emulators generalize to new orbital forcings, reproducing the spatial structure of the large-scale response as well as changes in seasonal patterns and in the spatial structure of ocean variability, while underestimating their amplitude. Baselines that infer the ocean state directly from the boundary forcings also recover much of the large-scale pattern, but only near the surface, and capture neither the seasonal nor the variability changes, indicating that these require some representation of dynamics. Despite these successes, the emulators fail to reproduce the slow, internally driven evolution of the ocean interior. We then show that the emulators' total forced response is well reconstructed by linearly composing their independent responses to each forcing component. Tracking response across training epochs, we find that convergence on mean state metrics in the training climate does not guarantee that the emulators capture the dynamics necessary for a skillful response. Together, these experiments establish the midHolocene as a controlled, ground-truthed setting for diagnosing forced-response failures before emulators are pushed to out-of-distribution climates.

Earth's feedback parameter measures how the Earth system responds to forcing and is inversely proportional to climate sensitivity. Sea-surface temperature (SST) patterns can modulate the value of the feedback parameter. Differences between observed and simulated SSTs have raised the question how the observed SSTs evolution impacts the global feedback. The standard method for estimating this effect uses observed SSTs prescribed to an atmospheric model with fixed pre-industrial atmospheric forcing. This method makes two assumptions: first, that the observed SSTs capture all relevant effects from the forcing, so that prescribing a time-varying forcing is unnecessary; second, that the temporal variations in the feedback parameter are driven by the evolving SST pattern and can be estimated via moving-window regressions. We test these assumptions by running controlled experiments in which SSTs from fully-coupled historical simulations are prescribed to an atmospheric model. We find that the prescribed-SST experiments fail to capture the coupled feedback evolution. This is explained by two effects: First, the absence of prescribed atmospheric forcing, and second, statistical noise arising from the computation of moving-window regressions. We find no evidence of any significant relationship between evolving SST patterns and changes in the feedback time series in a 4000-year pre-industrial control simulation. Any trends in the feedback parameter detected on timescales shorter than ~100 years are indistinguishable from statistical noise, making their attribution to the evolving SST pattern extremely difficult. Our results imply that prescribed SST simulations offer limited potential for inferring temporal changes in Earth's parameter over the observational period.

Satellite precipitation products such as IMERG exhibit biases that vary with terrain, season, and precipitation regime, leaving the applicability boundaries of machine learning correction unclear. This study proposes the Terrain-Moisture-Intensity (TMI) framework, centered on mechanism purity, extending the correction problem from purely algorithmic optimization to physical consistency diagnosis. A proof-of-concept study in Hunan Province employs IMERG V07, SRTM DEM, and ERA5 variables (tcwv, u10, v10). Ablation results indicate that, under the conditions of this study, terrain-moisture relationships are predominantly additive: RF-Full yields merely +0.001 R^2 gain over LR-Full, while bias rises to 1.282 mm d^-1; MAE decreases by approximately 14%, reflecting a trade-off between tail-fitting improvement and mean shift. SHAP diagnostics identify three categories of boundaries. Spatially, Central Hunan exhibits significant degradation (R^2=0.133) despite strong variable activation, consistent with mechanism fragmentation induced by mixed terrain. Temporally, u10 undergoes directional reversal between summer and spring (+0.096 to -0.156), presenting "silent failure." Extreme precipitation (>=50 mm d^-1) approximates a mechanism saturation frontier rather than isolated out-of-distribution samples, with DEM showing the largest relative amplification in SHAP disorder (+150%). The results demonstrate that machine learning correction performance is primarily constrained by mechanism purity. A pre-registered cross-regional test (Hunan, Guangxi, Guangdong) confirms this screening capability out of sample: a priori coherence proxies predict correction efficiency with a mean absolute error of 2.6 percentage points, while the transfer-versus-retraining contrast separates mechanism mismatch (coastal Guangdong) from portability (Guangxi), establishing the framework as a validated applicability screen.

An “Attention Residual U-Net” method is described for probabilistic quantitative precipitation forecasting (PQPF) that predicts the hourly probability of no precipitation plus the distribution of positive precipitation from a weighted mixture of two Gamma distributions. The neural network is trained on patches of numerical weather prediction (NWP) hourly precipitation from The Weather Company's convection-permitting GRAF (Global high-Resolution Atmospheric Forecasting) model along with terrain information and column-average relative humidity from the National Oceanic and Atmospheric Administration's (NOAA's) Global Forecast System (GFS). The target data are NOAA's Multi-Radar, Multi-Sensor (MRMS) gauge-corrected, quality controlled radar data sampled to the same grid as the GRAF data. The network outputs distributional parameters for each model grid point. Training uses negative log-likelihood as a proper scoring rule, with climatological initialization for stable convergence. Inference is performed as a single forward pass over the contiguous United States (CONUS) domain, with edge-replication padding to satisfy the network's spatial-divisibility requirement. The subsequent forecasts are spatially detailed, highly reliable, and skillful with respect to climatology and a simpler reference forecast method. The method is particularly useful for estimating probabilities in regions with large terrain variation.

Gridded-dataset suitability varies by support, season, variable, and application. This study evaluated E-OBS, AgERA5, MARS-STAT/JRC Agri4Cast, ERA5, and ERA5-Land against observations from 624 European Climate Assessment & Dataset (ECA&D) and Hellenic National Meteorological Service (HNMS/EMY) stations, using 18.18 million source-station-day records containing minimum (TN) and maximum (TX) temperatures. Derived variables included mean temperature (Tmean), diurnal temperature range (DTR), and growing degree days (GDD). Identical-support rankings and station bootstraps assessed daily and GDD performance; Fisher's z transformation summarized seasonal correlations; temporal and network tests screened trends. E-OBS minimized daily and monthly-climatology-removed anomaly root mean square error (RMSE) for TN, TX, Tmean, and DTR in all 16 country-variable comparisons. With at least 10 matched years, E-OBS led 42 of 64 seasonal RMSE comparisons and 59 of 64 correlation comparisons; MARS-STAT/JRC Agri4Cast led the remaining 22 and five, respectively. E-OBS minimized GDD RMSE in Spain and Italy, whereas MARS-STAT/JRC Agri4Cast led in Greece and narrowly led in France. GDD RMSE equaled 3.3-7.2% of observed mean GDD. Robustness tests supported increases in Spanish annual DTR, French annual TX, and French growing-season GDD. The lapse-rate correction improved 22 of 40 comparisons but usually worsened TN. Dataset choice should reflect support, variable, season, terrain, and application.

The Hadley circulation is Earth's dominant tropical overturning circulation, regulating atmospheric energy transport, tropical rainfall, and subtropical aridity. Although its variability has been extensively studied on seasonal to decadal climate-change timescales, its subseasonal behaviour remains poorly understood. Here we show that the Madden-Julian Oscillation (MJO), the leading mode of tropical intraseasonal variability, systematically modulates the Hadley circulation and influences global hydroclimate variability. Using reanalysis data and dynamical diagnostics, we identify a robust hemispherically asymmetric Hadley circulation anomaly associated with active MJO events, with amplitudes comparable to the climatological intraseasonal variability of the Hadley circulation. Breaking down the dynamical components reveals that the response is primarily maintained by latent heating associated with moist convection, while the overturning strength is driven by interhemispheric moisture gradients. Lead-lag analyses further show that the Hadley circulation lags the MJO by 4-15 days, indicating that MJO convection may drive overturning adjustments on subseasonal timescales. This coupled MJO-Hadley circulation state reveals a previously overlooked pathway linking tropical intraseasonal oscillations, meridional circulation and the hydrological cycle.

Global weather reanalyses and forecasts resolve the evolving atmospheric state on coarse grids, but site-specific applications require predictions at arbitrary locations where near-surface conditions also depend on unresolved terrain and land-surface properties. Existing probabilistic downscalers address this gap using hand-crafted topographic and surface descriptors. We ask instead whether Earth observation foundation models can provide transferable subgrid surface representations for probabilistic weather downscaling. We augment a convolutional conditional neural process (ConvCNP) that downscales coarse ERA5 reanalysis fields at ~25 km resolution with a learned local surface descriptor, obtained by compressing a patch of TESSERA embeddings at 10 m resolution. Although these embeddings summarize annual surface conditions, they improve downscaling by encoding persistent surface properties that capture a location's departure from the coarse-grid atmospheric state. Across five climatically diverse regions, the embedding improves point and probabilistic skill at stations held out in both space and time, overall improving CRPS skill by 11.5% for 2 m temperature and 6.2% for 10 m wind speed relative to a topography-only ConvCNP baseline. A hand-crafted descriptor incorporating richer surface information than topography alone captures comparable persistent subgrid signal but yields far smaller predictive gains than the learned embedding representation. These improvements persist when forecasts from the Aurora AI model replace ERA5 reanalysis fields and when predicting at newly deployed weather station networks. To our knowledge, this is the first evidence that long-timescale Earth observation embeddings can support short-timescale weather downscaling where subgrid departures are systematically structured by persistent surface properties.

While AI has shown remarkable promise in atmospheric and meteorological forecasting, accurately simulating other components of the Earth system with AI remains an active frontier. We present DLESyM-Ocean, a Deep Learning Earth System Model that simulates global present-day sea ice and upper ocean conditions. Unlike conventional probabilistic models optimized via diffusion objectives or losses such as continuous-ranked probability score, DLESyM-Ocean is trained using a patch energy score loss. When driven by atmospheric forcing, DLESyM-Ocean produces a well-calibrated, spatially coherent, and skillful ensemble of sea ice and upper ocean conditions with minimal bias relative to reanalysis products. DLESyM-Ocean is stable when autoregressively run for multi-year simulations and produces a climatology and variability with minimal bias compared with reanalysis. We evaluate case studies including a recent sea ice extreme, a severe marine heatwave, the 2023 El Niño transition, and the 2023 spike in global mean temperature. In all of these case studies, DLESyM-Ocean produces realistic surface and subsurface trajectories and ample ensemble diversity in response to common atmospheric forcing, suggestive of learned autoregressive ocean dynamics. When coupled with other Earth system components, such as the atmosphere, the computational efficiency of DLESyM-Ocean makes it a promising tool for subseasonal to seasonal forecasting.

Atmospheric reanalyses combine observations with model forecasts using complex data assimilation systems. We test whether a differentiable weather model permits a simpler and more accurate method based on a long-window four-dimensional variational data assimilation (4D-Var) formulation that omits the conventional background-error term. The method uses automatic differentiation to find optimal NeuralGCM initial conditions that minimize the misfit to real surface-pressure observations distributed across overlapping windows of two to seven days, assuming no model error. Cycling at 6-hour intervals for three months beginning 1 January 2015 yields a stable reanalysis with smaller error relative to ERA5 in 500-hPa geopotential height than the Twentieth Century Reanalysis version 3 (20CRv3), which uses an ensemble Kalman filter to assimilate the same observations. Every window produces smaller errors than 20CRv3, with analysis error for the four-day window approximately 55% smaller than for 20CRv3. At the end of the four-day window, which does not benefit from future observations, error remains approximately 38% smaller than 20CRv3. Analyses degrade slightly beyond four days, which we attribute to the increasing importance of model error.

Many air quality models indicate that rapid reductions in nitrogen oxides (NOx), without comparable controls on volatile organic compounds, have worsened summertime ozone pollution in urban China, producing a short-term strong ozone penalty. Other models, however, simulate the opposite response, suggesting that cutting down NOx has already helped mitigate ozone pollution. This contradiction obscures understanding of atmospheric chemistry and weakens guidance on control policy design. Here, we reconcile this disagreement and reveal the underestimated benefits of NOx emission reductions using a machine learning framework integrated with an observational constraint. We first constrain ozone responses under a 30% NOx reduction, comparable to the magnitude of NOx emission declines across major Chinese city clusters between 2015 and 2023. The constrained results indicate that ozone decreases prevail across urban China, with only small increases mainly in July 2015. This challenges the widespread ozone worsening that many models predict. We then extend the constraint across 10-60% NOx reductions, establishing its use for rapid ozone sensitivity diagnosis without exhaustive scenario modeling. This diagnosis shows that sustained NOx control increasingly favored ozone mitigation during 2015-2023, benefiting a growing share of China's population. These results underscore that continued NOx reductions can deliver larger ozone mitigation benefits than many models suggest.