https://arxiv.org/api/lWl7v0XGxX2hUpVIDjlmOOS5xAU 2026-06-18T13:04:43Z 8381 390 15 http://arxiv.org/abs/2603.19488v1 2026-03-19T21:33:21Z

Heat stress has harmful effects that impact communities across the Unitedt States, particularly when high temperatures are accompanied by high humidity. The combined impact of temperature and humidity can be summarized by the heat index (HI). Current state-of-the-art numerical weather prediction models are often biased when forecasting temperature and humidity even within a 24-hour forecast lead time. This study explores the ability of machine learning (ML) models to accurately predict the next-day heat index using Random Forest and single-layer Gated Recurrent Unit (GRU) models in four locations across the United States. We find that Random Forest and GRU models perform reasonably well at all four selected locations. Mean absolute HI error ranges from 4.5 to 6.6 °F. All model versions have an accuracy rate exceeding 80% in three of the four locations in terms of successfully forecasting an extreme heat day, as indicated by a high afternoon HI. The GRU model achieves over 95% accuracy in these three locations. Model performance details vary by location. In Minneapolis and Portland, which have relatively few days with high HI values, models' accuracy is high, but the recall and precision are generally very low. In contrast, Dallas, a location with many high HI days, shows moderately high accuracy, as well as extremely high recall and precision. These differences are likely due to distinct causes of heatwaves in different climatological regions of the United States, as reflected in the feature importance scores output by Random Forest models. The ML models designed in this study can be used to assist with local heat index forecasting and extreme heat warning issuance at minimal computational cost.

2026-03-19T21:33:21Z The two authors contributed equally to this work Yushan Han Calen Randall http://arxiv.org/abs/2603.19058v1 2026-03-19T15:50:38Z

Non-Gaussian statistics are a challenge for data assimilation. Linear methods oversimplify the problem, yet fully nonlinear methods are often too expensive to use in practice. The best solution usually lies between these extremes. Triangular measure transport offers a flexible framework for nonlinear data assimilation. Its success, however, depends on how the map is parametrized. Too much flexibility leads to overfitting; too little misses important structure. To address this balance, we develop an adaptation algorithm that selects a parsimonious parametrization automatically. Our method uses P-spline basis functions and an information criterion as a continuous measure of model complexity. This formulation enables gradient descent and allows efficient, fine-scale adaptation in high-dimensional settings. The resulting algorithm requires no hyperparameter tuning. It adjusts the transport map to the appropriate level of complexity based on the system statistics and ensemble size. We demonstrate its performance in nonlinear, non-Gaussian problems, including a high-dimensional distributed groundwater model.

2026-03-19T15:50:38Z 24 pages, 10 figures Berent Å. S. Lunde Maximilian Ramgraber http://arxiv.org/abs/2510.25444v2 2026-03-19T12:24:18Z

Convectively Coupled Equatorial Waves (CCEWs) dominate atmospheric variability on timescales of 2--30 days in the Tropics, bringing episodes of widespread heavy precipitation. This study compares the representation of CCEWs and their connection to upscale energy transfer in two Met Office Unified Model simulations of the full tropical channel with identical km-scale resolution for the DYAMOND Summer period. The principal difference between the simulations is that one parametrizes convection (GAL9), while the other (RAL3) is convection permitting. The GAL9 convection scheme acts to remove vertical instability without explicitly representing the resolved-scale circulation associated with convective plumes. We present the first quantitative diagnosis of interscale energy transfer and its relation to CCEWs. This diagnosis is important because upscale energy transfer between convection and large-scale waves may influence accurate simulation of tropical weather systems. The average upper-tropospheric upscale transfer simulated by RAL3 is approximately 50% higher than GAL9. CCEWs are more coherent in RAL3, with an average phase-speed variability 80% higher than observations, compared with 166% higher in GAL. RAL3 also simulates greater upscale energy transfer within waves than GAL9 with a stronger correlation between the interscale energy transfer and equatorial wave winds. Kelvin and Rossby waves are associated with upscale energy transfer from scales 4-8 times smaller than their wavelength, related to active deep convection within a particular sector of the wave phase. Our findings show that the explicit representation of convection has a significant impact on the simulation of upscale energy transfer, and is very likely to be a significant factor in the faithful simulation of convective coupling within CCEWs.

2025-10-29T12:10:24Z This article is in review at Journal of the Atmospheric Sciences. Copyright in this article may be transferred without further notice E. McKinnon-Gray D. Shipley J. Methven T. H. A. Frame C. Sanchez A. McCabe N. M. Roberts http://arxiv.org/abs/2502.16116v3 2026-03-19T10:37:36Z

Short-term precipitation nowcasting is essential for flood management, transportation, energy system operations, and emergency response. However, many existing models fail to fully exploit the extensive atmospheric information available, relying primarily on precipitation data alone. This study examines whether integrating multi variable weather-station measurements with radar can enhance nowcasting skill and introduces two complementary architectures that integrate multi variable station data with radar images. The SmaAt-fUsion model extends the SmaAt-UNet framework by incorporating weather station data through a convolutional layer, integrating it into the bottleneck of the network; The SmaAt-Krige-GNet model combines precipitation maps with weather station data processed using Kriging, a geo-statistical interpolation method, to generate variable-specific maps. These maps are then utilized in a dual-encoder architecture based on SmaAt-GNet, allowing multi-level data integration. Experimental evaluations were conducted using four years (2016--2019) of weather station and precipitation radar data from the Netherlands. Results demonstrate that SmaAt-Krige-GNet outperforms the standard SmaAt-UNet, which relies solely on precipitation radar data, in low precipitation scenarios, while SmaAt-fUsion surpasses SmaAt-UNet in both low and high precipitation scenarios. This highlights the potential of incorporating discrete weather station data to enhance the performance of deep learning-based weather nowcasting models.

2025-02-22T06:46:04Z 13 pages, 6 figures Jie Shi Aleksej Cornelissen Siamak Mehrkanoon http://arxiv.org/abs/2603.18319v1 2026-03-18T22:05:56Z

Marine heatwaves (MHWs) threaten marine ecosystems and significantly impact weather patterns. In the Arabian Sea, summer MHWs are of particular concern due to their potential impacts on the Indian summer monsoon, a lifeline for nearly a billion people. However, the drivers of these MHWs and their influence on atmospheric circulation and monsoon rainfall remain poorly understood. Using satellite observations, reanalysis datasets, and numerical model experiments, we investigate the key drivers of MHW events and assess their impacts. When SST warming trends are retained, the eastern and northern Arabian Sea emerge as MHW hotspots, showing rapid increases during 1982-2023, largely due to anthropogenic warming. On detrending the SSTs to remove the influence of anthropogenic warming on individual MHWs, we find that most MHWs are short-lived (lasting <= 20 days) and are initiated by enhanced surface shortwave radiation and reduced latent heat loss associated with the suppressed convection phase of the Boreal Summer Intraseasonal Oscillations (BSISOs). Interannual SST anomalies, including ENSO and Indian Ocean Dipole (IOD), further modulate the year-to-year MHW variability. Conversely, the warm SSTs during MHWs exert strong atmospheric feedbacks. MHWs in the eastern Arabian Sea drive cyclonic winds, intensify moisture convergence and increase the risk of extreme precipitation along the southwest coast of India. In the northern Arabian Sea, MHW-induced cyclones trigger intense rainfall over northwestern India and Pakistan, contributing to extreme events like the 2022 Pakistan floods. These findings improve our capacity to predict Arabian Sea MHWs and assess their risks, offering significant socio-economic and ecological benefits.

2026-03-18T22:05:56Z Under review at the Journal of Climate D. L. Suhas Weiqing Han Toshiaki Shinoda Rui Sun Aneesh Subramanian Mark Bourassa Michael Alexander http://arxiv.org/abs/2411.06629v3 2026-03-18T20:30:43Z

Robust and convergent high-order numerical methods for solving partial differential equations are highly attractive due to their efficiency on modern and next-generation hardware architectures. However, designing such methods for nonlinear hyperbolic conservation laws remains a significant challenge. In this work, we introduce a framework based on dual-pairing (DP) and upwind summation-by-parts (SBP) finite difference (FD) and discontinuous Galerkin (DG) finite element methods, aimed at achieving accurate and robust numerical approximations of nonlinear conservation laws. The framework ensures entropy consistency and features an intrinsic high-order accurate filter designed to detect and resolve regions where the solution is poorly captured or discontinuities are present. The DP SBP FD/DG operators form a dual pair of discrete derivative operators that collectively preserve the SBP property. Furthermore, these operators are constructed to be upwind, allowing them to incorporate dissipation within the elements themselves.This contrasts with traditional SBP and collocated DG spectral element methods, which typically induce dissipation solely through numerical fluxes at element interfaces. Our framework facilitates the systematic combination of DP SBP FD/DG operators with skew-symmetric and upwind flux splitting techniques. This integration enables the development of robust, high-order accurate schemes for nonlinear hyperbolic conservation laws.

2024-11-10T23:38:14Z Dougal Stewart Nathan Lee Kenneth Duru http://arxiv.org/abs/2603.17185v1 2026-03-17T22:28:27Z

In many geophysical and planetary environments, such as Earth's ocean and atmosphere as well as subsurface oceans of icy satellites, convection driven by bottom geothermal heating usually coexists with baroclinic eddies driven by lateral buoyancy/temperature gradients. These processes compete against each other, with convection destabilizing the stratification and baroclinic eddies re-stabilizing it, thereby controlling whether the bottom heat flux is significantly redistributed as it is transmitted to the upper surface. Using scaling analysis and numerical simulations, we show that a stratified layer persists near the upper surface up to ${\rm Ra}_{v}\sim {\rm Ra}_h^{5/2}$, where ${\rm Ra}_h\equiv Δb_0/(L_zf^2)$ measures the imposed upper-surface buoyancy contrast $Δb_0$ and ${\rm Ra}_v\equiv B_0/(L_z^2f^3)$ measures the strength of the bottom buoyancy flux $B_0$, $L_z$ is the domain depth and $f$ is the Coriolis parameter. For ${\rm Ra}_v<{\rm Ra}_h^{5/2}$, baroclinic eddies dominate over convection, maintain the upper stratified layer, and completely deflect the bottom buoyancy/heat input into meridional transport. In contrast, when ${\rm Ra}_v>{\rm Ra}_h^{5/2}$, convective plumes penetrate the stratification and transport buoyancy/heat vertically with negligible deflection. Building on these results, we further propose a scaling law for the meridional buoyancy/heat transport in this system. Applications to icy satellites are discussed.

2026-03-17T22:28:27Z Shuang Wang Wanying Kang Cheng Li http://arxiv.org/abs/2603.17116v1 2026-03-17T20:18:25Z

Traditional significant wave height (SWH) estima- tion from HF radar typically relies on spectral analysis of the received radar signals. This process was previously simplified by establishing a linear relationship between SWH and the standard deviation of received HF radar voltages under first- order scattering. Building on this approach, this paper presents a physics-informed regression model that incorporates second- order scattering effects through a quadratic formulation derived from a Neumann expansion. The proposed method is evaluated using HF radar data collected in July 2018 at Argentia, New- foundland, with collocated buoy measurements as ground truth. The model achieves a minimum root-mean-square error (RMSE) of approximately 19 cm.

2026-03-17T20:18:25Z 5 pages, 6 figures, IEEE conference format. Preprint Senal Chandrasekara Reza Shahidi http://arxiv.org/abs/2508.19826v3 2026-03-17T17:27:13Z

The Atlantic Meridional Overturning Circulation (AMOC) is a major tipping element in the present-day climate, and could potentially collapse under sufficient freshwater or CO2-forcing. While the effect of the Bering Strait on AMOC stability has been well studied, it is unknown whether a constructed closure of this Strait can prevent an AMOC collapse under climate change. Here, we show in an Earth system Model of Intermediate Complexity that an artificial closure of the Strait can extend the safe carbon budget of the AMOC, provided that the AMOC is strong enough at the closure time. Specifically, for this model, an equilibrium AMOC with a reduction below (6.1 +/- 0.5)% from pre-industrial has an additional budget up to 500PgC given a sufficiently early closure, while for a weaker AMOC a closure reduces this budget. This indicates that constructing this closure could be a feasible climate intervention strategy to prevent an AMOC collapse.

2025-08-27T12:25:05Z Jelle Soons Henk A. Dijkstra http://arxiv.org/abs/2603.16811v1 2026-03-17T17:15:00Z

We investigate analytically and numerically correlation functions of the phase of light waves that propagate through turbulent media. We examine the case of strong scintillations that occur at large values of the Rytov dispersion, $σ^2_R$. Then, it is possible to relate the pair correlation function of phase gradients to the known pair correlation function of the envelope dependent on the distance $r$ assuming Gaussianity of the envelope of the beam. Our direct numerical simulations show that the profile of the pair correlation function for phase gradients gradually approaches the theoretical expression as the value of $σ_R^2$ increases, if $r<r_0$ where $r_0$ is the Fried length. For larger $r$ the behavior of the computed correlation function is quite different because of destroying the Gaussianity.

2026-03-17T17:15:00Z arXiv admin note: substantial text overlap with arXiv:2601.08272 V. A. Bogachev I. V. Kolokolov V. V. Lebedev A. V. Nemtseva F. A. Starikov http://arxiv.org/abs/2603.26703v1 2026-03-17T15:38:19Z

Short-term background ensemble covariances (BEC) are crucial for ensemble-based data assimilation (DA). However, limited studies so far have examined the fidelity of the cost-effective data-driven model in producing the short-term BEC for hurricane data assimilation. In this study, we evaluate the background ensemble spread and correlations from GraphCast against those of GEFS for Hurricane Lee (2023) during both its intensification and non-intensification phases. Specifically, the BEC in the hurricane vortex, the hurricane environment, and the vortex-environment interactions are examined. Within the hurricane vortex, the background ensemble of GraphCast is less dispersive than GEFS. The two models agree well on the background ensemble correlations that are tied to the primary circulation but show a larger correlation difference associated with the secondary circulation, indicating the two models represent unbalanced and diabatic processes differently. In the hurricane environment, binned univariable correlations show linear relationships between the two models, with a weaker horizontal geopotential height correlation in GraphCast. GraphCast also shows a reduced spread and a flatter empirical orthogonal function spectrum of the 500 hPa geopotential height background ensemble, with more perturbation growth distributed to smaller-scale features such as shortwaves. For the vortex-environment interaction, the two models produce close background ensemble correlation patterns for Lee's track but differ more for intensity. Overall, GraphCast can produce broadly consistent short-term BEC for Hurricane Lee compared to those of GEFS. However, systematic difference exists in certain variables, scales, and processes, suggesting the need to further investigate its fidelity in a cycled hurricane DA context.

2026-03-17T15:38:19Z Zhihong Chen Xuguang Wang http://arxiv.org/abs/2512.06026v2 2026-03-17T11:14:03Z

Whereas there is growing interest in exploring longer-term climate with overshoot, including tipping elements, beyond 2100, most Integrated Assessment Models (IAMs) generate emissions scenarios only till 2100. Here we propose a framework to extend scenarios until 2150 using an emulator of IAMs. Our framework offers a potential interim solution for developing very long-term scenarios, such as the Scenario Model Intercomparison Project (ScenarioMIP), circumventing the challenges of fully simulating IAMs beyond 2100.

2025-12-04T14:18:38Z Weiwei Xiong Katsumasa Tanaka http://arxiv.org/abs/2603.16312v1 2026-03-17T09:44:28Z

The North Atlantic Oscillation (NAO) is the dominant mode of atmospheric variability over the North Atlantic sector, influencing temperature and precipitation across Europe. While the NAO's impact on North Atlantic sea surface temperatures (SSTs) is well understood, the NAO can also be driven by SST anomalies. However, this NAO response to SST anomalies is believed to be weak and nonlinear. Former studies highlight that during early winter (November-December), El Nino Southern Oscillation (ENSO) events modulate the NAO, with El Nino (La Nina) events being linked to positive (negative) NAO phases, and an opposite effect observed in late winter (January-February). Indian Ocean SSTs and the North Atlantic Horseshoe SST anomaly have also been suggested as contributors to early winter NAO variability. However, climate models often struggle to capture these SST-NAO teleconnections, particularly in early winter. To address this, a statistical framework based on convolutional neural networks (CNNs) is developed to predict the early winter NAO using observed SST fields one-, two-, and three-month before. A linear model serves as a benchmark, and both models are trained on ERA5 reanalysis data from 1940 to 2023. A sensitivity analysis is used to interpret the CNN's decision-making process, revealing that it focuses on regions such as the tropical Pacific and North Atlantic, confirming results from previous works. The CNN outperforms the linear model, highlighting the value of capturing nonlinear SST-NAO relationships. Prediction skill appears to be linked to ENSO, with strong ENSO events associated with greater skill in forecasting the NAO than neutral events. These findings underscore the potential of deep learning to build medium-range NAO prediction.

2026-03-17T09:44:28Z 8 pages, 5 figures Elena Provenzano Guillaume Gastineau Carlos Mejia Didier Swingedouw Sylvie Thiria http://arxiv.org/abs/2603.26700v1 2026-03-17T02:51:38Z

It has been hypothesized that the Earth may have experienced snowball events in the past, during which its surface became completely covered with ice. Previous studies used general circulation models to investigate the onset and climate of such snowball events. Using the MIROC4m coupled atmosphere--ocean climate model, this study examined the changes in the oceanic circulation during the onset of a modern snowball Earth and elucidated their evolution to steady states under the snowball climate. Abruptly changing the solar constant to 94% of its present-day value caused the modern Earth climate to turn into a snowball state after ~1300 years and initiated rapid increase in sea ice thickness. During onset of the snowball, extensive sea ice formation and melting of sea ice in the mid-latitudes caused substantial freshening of surface waters and salinity stratification. By contrast, such salinity stratification was absent if the duration between the change in the solar flux and the snowball onset was short. After snowball onset, the global sea ice cover and the buildup of salinity stratification caused drastic weakening in the deep ocean circulation. However, the meridional overturning circulation resumed within several hundred years after the snowball onset because the density flux by sea ice production weakens the salinity stratification. While the evolution of the oceanic circulation would depend on the continental distribution and the evolution of continental ice sheets, our results highlight the gradual growth of sea ice and associated brine rejection are essential factors for the transient evolution of the oceanic circulation in the snowball events.

2026-03-17T02:51:38Z 23 pages, 12 figures, accepted for publication in Climate of the Past (CP) Takashi Obase Takanori Kodama Takao Kawasaki Sam Sherriff-Tadano Daisuke Takasuka Ayako Abe-Ouchi Masakazu Fujii http://arxiv.org/abs/2603.16031v1 2026-03-17T00:36:18Z

We explore hurricane and ocean reanalysis data to understand how rapid intensification (RI) of tropical cyclones is impacted by the upper ocean density structure, with an emphasis on barrier layer (BL) thickness and thermocline depth in the eastern Caribbean Sea and adjacent western tropical North Atlantic. This analysis leverages transition path theory (TPT), supported by basic statistical methods. In TPT, Markov chains are constructed by discretizing data series related to weather system intensity, changes in intensity, translational speed, and BL thickness and thermocline depth. These series are viewed as trajectories in abstract state spaces, following a memoryless stochastic process. RI imminence is rigorously framed using a newly derived TPT statistic, which gives the time distribution to first reach a target -- the RI state -- from a source -- for instance, the state determined by a certain BL range and system intensity -- conditional on connecting paths exhibiting minimal detours. Increased RI frequency is observed in the eastern Caribbean and nearby Atlantic, influenced by river runoff, primarily in tropical storms and category 2 hurricanes. RI frequently correlates with a well-developed BL; however, increased translational speed is necessary for RI. TPT shows a stronger connection between RI and thermocline depth than BL presence, with RI likelihood rising for hurricanes with a thin BL, especially category 1. Across all strength categories, a deep thermocline consistently elevates RI probability, a factor missed by basic statistical analysis. Furthermore, translational speed is crucial, with faster, stronger hurricanes more susceptible to RI, while slower systems are less so.

2026-03-17T00:36:18Z Submitted to the Journal of Atmospheric Science F. J. Beron-Vera G. Bonner M. J. Olascoaga S. Dong H. Lopez