This study investigates the impact of wind velocity averaging on Dynamic Thermal Rating (DTR) calculations. It is based on a high-temporal-resolution (1 second) wind measurements obtained from a transmission line in Slovenia, Europe. Wind speed and direction variability are analysed, and two averaging methods, namely vector averaging, where velocity is averaged as vector, and hybrid averaging, where speed is averaged as scalar, are employed. DTR calculations are performed on both high-resolution data and averaged data (5 minute averaging window). It is demonstrated that averaging has a significant effect on both Nusselt number and ampacity, and the effect exhibits a strong angular dependency on the relative angle of the wind to the line. Therefore, two limit cases are studied: in the case of parallel wind, averaged data underestimates the ampacity, and there is a significant amount of cases where the underestimation is larger than 10 %. In the case of perpendicular wind, the two averaging methods affect the results in different ways, but both result in a substantial amount of cases where ampacity is overestimated, potentially leading to unsafe operation. The main takeaway of the study is that averaging wind velocity has a significant impact on DTR results, and special emphasis should be given to the averaging method, as different methods affect the results in different ways.

Machine learning-based precipitation nowcasting relies on high-fidelity radar reflectivity sequences to model the short-term evolution of convective storms. However, the development of models capable of predicting extreme weather has been constrained by the coarse resolution (1-2 km) of existing public radar datasets, such as SEVIR, HKO-7, and GridRad-Severe, which smooth the fine-scale structures essential for accurate forecasting. To address this gap, we introduce Storm250-L2, a storm-centric radar dataset derived from NEXRAD Level-II and GridRad-Severe data. We algorithmically crop a fixed, high-resolution (250 m) window around GridRad-Severe storm tracks, preserve the native polar geometry, and provide temporally consistent sequences of both per-tilt sweeps and a pseudo-composite reflectivity product. The dataset comprises thousands of storm events across the continental United States, packaged in HDF5 tensors with rich context metadata and reproducible manifests.

There is a growing need to evaluate Large Language Models (LLMs) on complex, high-impact, real-world tasks to assess their true readiness as reasoning agents. To address this gap, we introduce AgentCaster, a contamination-free framework employing multimodal LLMs end-to-end for the challenging, long-horizon task of tornado forecasting. Within AgentCaster, models interpret heterogeneous spatiotemporal data from a high-resolution convection-allowing forecast archive. We assess model performance over a 40-day period featuring diverse historical data, spanning several major tornado outbreaks and including over 500 tornado reports. Each day, models query interactively from a pool of 3,625 forecast maps and 40,125 forecast soundings for a forecast horizon of 12-36 hours. Probabilistic tornado-risk polygon predictions are verified against ground truths derived from geometric comparisons across disjoint risk bands in projected coordinate space. To quantify accuracy, we propose domain-specific TornadoBench and TornadoHallucination metrics, with TornadoBench highly challenging for both LLMs and domain expert human forecasters. Notably, human experts significantly outperform state-of-the-art models, which demonstrate a strong tendency to hallucinate and overpredict risk intensity, struggle with precise geographic placement, and exhibit poor spatiotemporal reasoning in complex, dynamically evolving systems. AgentCaster aims to advance research on improving LLM agents for challenging reasoning tasks in critical domains.

Global Station Weather Forecasting (GSWF) is a key meteorological research area, critical to energy, aviation, and agriculture. Existing time series forecasting methods often ignore or unidirectionally model spatial correlation when conducting large-scale global station forecasting. This contradicts the intrinsic nature underlying observations of the global weather system, limiting forecast performance. To address this, we propose a novel Spatial Structured Attention Block in this paper. It partitions the spatial graph into a set of subgraphs and instantiates Intra-subgraph Attention to learn local spatial correlation within each subgraph, and aggregates nodes into subgraph representations for message passing among the subgraphs via Inter-subgraph Attention -- considering both spatial proximity and global correlation. Building on this block, we develop a multiscale spatiotemporal forecasting model S$^2$Transformer by progressively expanding subgraph scales. The resulting model is both scalable and able to produce structured spatial correlation, and meanwhile, it is easy to implement. The experimental results show that it can achieve performance improvements up to 16.8% over time series forecasting baselines at low running costs.

We introduce LUCIE-3D, a lightweight three-dimensional climate emulator designed to capture the vertical structure of the atmosphere, respond to climate change forcings, and maintain computational efficiency with long-term stability. Building on the original LUCIE-2D framework, LUCIE-3D employs a Spherical Fourier Neural Operator (SFNO) backbone and is trained on 30 years of ERA5 reanalysis data spanning eight vertical {\sigma}-levels. The model incorporates atmospheric CO2 as a forcing variable and optionally integrates prescribed sea surface temperature (SST) to simulate coupled ocean--atmosphere dynamics. Results demonstrate that LUCIE-3D successfully reproduces climatological means, variability, and long-term climate change signals, including surface warming and stratospheric cooling under increasing CO2 concentrations. The model further captures key dynamical processes such as equatorial Kelvin waves, the Madden--Julian Oscillation, and annular modes, while showing credible behavior in the statistics of extreme events. Despite requiring longer training than its 2D predecessor, LUCIE-3D remains efficient, training in under five hours on four GPUs. Its combination of stability, physical consistency, and accessibility makes it a valuable tool for rapid experimentation, ablation studies, and the exploration of coupled climate dynamics, with potential applications extending to paleoclimate research and future Earth system emulation.

Small-scale gravity waves play a crucial role in atmospheric tracer transport, yet their effects remain unresolved in climate models and must be parameterized. This study investigates how gravity waves influence large-scale tracer distributions, utilizing a multiple-scale analysis to systematically identify the governing terms of gravity wave-induced tracer fluxes. The analysis reveals both leading-order and next-order impacts: the former being the inertia-gravity wave-induced tracer Stokes drift, which acts perpendicular to both the large-scale tracer gradient and the wave number, while the latter becomes significant at lower latitudes where Coriolis effects diminish. A numerical framework is developed to incorporate these fluxes into a gravity wave parameterization model, potentially enhancing climate model accuracy without requiring explicit resolution of small-scale wave dynamics. Model validation against high-resolution wave-resolving simulations confirms the effectiveness of this approach. By improving the representation of gravity wave-induced tracer transport, this research advances the accuracy of climate simulations, particularly in their depiction of microphysics and radiative processes.

Convection is a ubiquitous process driving geophysical/astrophysical fluid flows, which are typically strongly constrained by planetary rotation on large scales. A celebrated model of such flows, rapidly rotating Rayleigh-Bénard convection, has been extensively studied in direct numerical simulations (DNS) and laboratory experiments, but the parameter values attainable by state-of-the-art methods are limited to moderately rapid rotation (Ekman numbers $Ek\gtrsim10^{-8}$), while realistic geophysical/astrophysical $Ek$ are significantly smaller. Asymptotically reduced equations of motion, the nonhydrostatic quasi-geostrophic equations (NHQGE), describing the flow evolution in the limit $Ek\to 0$, do not apply at finite rotation rates. The geophysical/astrophysical regime of small but finite $Ek$ therefore remains currently inaccessible. Here, we introduce a new, numerically advantageous formulation of the Navier-Stokes-Boussinesq equations informed by the scalings valid for $Ek\to0$, the \textit{Rescaled Rapidly Rotating incompressible Navier-Stokes Equations} (RRRiNSE). We solve the RRRiNSE using a spectral quasi-inverse method resulting in a sparse, fast algorithm to perform efficient DNS in this previously unattainable parameter regime. We validate our results against the literature across a range of $Ek$ and demonstrate that the algorithmic approaches taken remain accurate and numerically stable at $Ek$ as low as $10^{-15}$. Like the NHQGE, the RRRiNSE derive their efficiency from adequate conditioning, eliminating spurious growing modes that otherwise induce numerical instabilities at small $Ek$. We show that the time derivative of the mean temperature is inconsequential for accurately determining the Nusselt number in the stationary state, significantly reducing the required simulation time, and demonstrate that full DNS using RRRiNSE agree with the NHQGE at very small $Ek$.

We present a lightweight, easy-to-train, low-resolution, fully data-driven climate emulator, LUCIE, that can be trained on as low as $2$ years of $6$-hourly ERA5 data. Unlike most state-of-the-art AI weather models, LUCIE remains stable and physically consistent for $100$ years of autoregressive simulation with $100$ ensemble members. Long-term mean climatology from LUCIE's simulation of temperature, wind, precipitation, and humidity matches that of ERA5 data, along with the variability. We further demonstrate how well extreme weather events and their return periods can be estimated from a large ensemble of long-term simulations. We further discuss an improved training strategy with a hard-constrained first-order integrator to suppress autoregressive error growth, a novel spectral regularization strategy to better capture fine-scale dynamics, and finally an optimization algorithm that enables data-limited (as low as $2$ years of $6$-hourly data) training of the emulator without losing stability and physical consistency. Finally, we provide a scaling experiment to compare the long-term bias of LUCIE with respect to the number of training samples. Importantly, LUCIE is an easy to use model that can be trained in just $2.4$h on a single A-100 GPU, allowing for multiple experiments that can explore important scientific questions that could be answered with large ensembles of long-term simulations, e.g., the impact of different variables on the simulation, dynamic response to external forcing, and estimation of extreme weather events, amongst others.

In recent years, artificial intelligence (AI) has deeply impacted various fields, including Earth system sciences. Here, AI improved weather forecasting, model emulation, parameter estimation, and the prediction of extreme events. However, the latter comes with specific challenges, such as developing accurate predictors from noisy, heterogeneous and limited annotated data. This paper reviews how AI is being used to analyze extreme events (like floods, droughts, wildfires and heatwaves), highlighting the importance of creating accurate, transparent, and reliable AI models. We discuss the hurdles of dealing with limited data, integrating information in real-time, deploying models, and making them understandable, all crucial for gaining the trust of stakeholders and meeting regulatory needs. We provide an overview of how AI can help identify and explain extreme events more effectively, improving disaster response and communication. We emphasize the need for collaboration across different fields to create AI solutions that are practical, understandable, and trustworthy for analyzing and predicting extreme events. Such collaborative efforts aim to enhance disaster readiness and disaster risk reduction.

Atmospheric interaction with light has been an area of fascination for many researchers over the last century. Environmental conditions, such as temperature and wind speed, heavily influence the complex and rapidly varying optical distortions propagating optical fields experience. The continuous random phase fluctuations commonly make deciphering the exact origins of specific optical aberrations challenging. The generation of eddies is a major contributor to atmospheric turbulence, similar in geometric structure to optical vortices that sit at the centre of OAM beams. Decomposing the received optical fields into OAM provides a unique spatial similarity that can be used to analyse turbulent channels. In this work, we present a novel mode decomposition assisted machine learning approach that reveals trainable features in the distortions of vortex beams that allow for effective environmental monitoring. This novel technique can be used reliably with Support Vector Machine regression models to measure temperature variations of 0.49C and wind speed variations of 0.029 m/s over a 36m experimental turbulent free-space channels with controllable and verifiable temperature and wind speed with short 3s measurement. The predictable nature of these findings could indicate the presence of an underlying physical relationship between environmental conditions that lead to specific eddy formation and the OAM spiral spectra.

To ensure efficient and reliable operation of a concentrating solar-thermal power (CSP) plant, its solar collector field needs to accurately focus sunlight. The optical efficiency and structural integrity of the solar collectors is significantly influenced by wind conditions in the field. In this study, we present insights into dynamic wind loading on parabolic trough CSP collectors. We derive novel conclusions by analyzing a first-of-a-kind measurement campaign of wind and structural loads, performed at an operational CSP plant. Previous research primarily relied on wind tunnel tests and simulations, leaving uncertainty about wind loading effects in operational settings. We demonstrate that the parabolic trough field significantly alters the turbulent wind field within the collector field, especially under winds perpendicular to the trough rows. Our measurements within the trough field show reduced wind speeds, changes in wind direction and turbulence properties, and vortex shedding from the trough assemblies. These modifications to the wind field directly impact both static and dynamic support structure loads. Our measurements reveal higher wind loads on trough assemblies compared to those observed previously in wind tunnel tests. The insights from this study offer a novel perspective on our understanding of wind-driven loads on CSP collectors. By informing the development of next-generation design tools and models, this research paves the way for enhanced structural integrity and improved optical performance in future parabolic trough systems.

Over the past few years, due to the rapid development of machine learning (ML) models for weather forecasting, state-of-the-art ML models have shown superior performance compared to the European Centre for Medium-Range Weather Forecasts (ECMWF)'s high-resolution forecast (HRES) in 10-day forecasts at a spatial resolution of 0.25 degree. However, the challenge remains to perform comparably to the ECMWF ensemble mean (EM) in 15-day forecasts. Previous studies have demonstrated the importance of mitigating the accumulation of forecast errors for effective long-term forecasts. Despite numerous efforts to reduce accumulation errors, including autoregressive multi-time step loss, using a single model is found to be insufficient to achieve optimal performance in both short and long lead times. Therefore, we present FuXi, a cascaded ML weather forecasting system that provides 15-day global forecasts with a temporal resolution of 6 hours and a spatial resolution of 0.25 degree. FuXi is developed using 39 years of the ECMWF ERA5 reanalysis dataset. The performance evaluation, based on latitude-weighted root mean square error (RMSE) and anomaly correlation coefficient (ACC), demonstrates that FuXi has comparable forecast performance to ECMWF EM in 15-day forecasts, making FuXi the first ML-based weather forecasting system to accomplish this achievement.

We introduce a new methodology to study marine heat waves, extreme events in the sea surface temperature (SST) of the global ocean. Motivated by previously large and impactful marine heat waves and by theoretical expectation that the dominant heating processes coherently affect large regions of the ocean, we introduce a methodology from computer vision to construct marine heat wave systems (MWHSs) -- the collation of SST extrema in dimensions of area and time. We identify 649,475 MHWSs in the 37 year period (1983-2019) of daily SST records and find that the duration t_dur (days), maximum area A_max (km$^2$), and total ``volume'' N_vox (days km$^2$) for the majority of MHWSs are well-described by power-law distributions: t_dur^(-3), A_max^(-2) and N_vox^(-2). These characteristics confirm SST extrema exhibit strong spatial coherence that define the formation and evolution of marine heat waves. Furthermore, the most severe MHWSs deviate from these power-laws and are the dominant manifestation of marine heat waves: extrema in ocean heating are driven by the ~200 systems with largest area and duration. We further demonstrate that the previously purported rise in the incidence of marine heat wave events over the past decade is only significant in these severe systems. A change point analysis reveals a rapid increase in days under a severe MHW in most regions of the global ocean over the period of 2000-2005. Understanding the origin and impacts of marine heat waves in the current and future ocean, therefore, should focus on the production and evolution of the largest-scale and longest-duration heating phenomena.

The warming of the Arctic, also known as Arctic amplification, is led by several atmospheric and oceanic drivers. However, the details of its underlying thermodynamic causes are still unknown. Inferring the causal effects of atmospheric processes on sea ice melt using fixed treatment effect strategies leads to unrealistic counterfactual estimations. Such models are also prone to bias due to time-varying confoundedness. Further, the complex non-linearity in Earth science data makes it infeasible to perform causal inference using existing marginal structural techniques. In order to tackle these challenges, we propose TCINet - time-series causal inference model to infer causation under continuous treatment using recurrent neural networks and a novel probabilistic balancing technique. Through experiments on synthetic and observational data, we show how our research can substantially improve the ability to quantify leading causes of Arctic sea ice melt, further paving paths for causal inference in observational Earth science.

The Planetary Boundary for Climate Change has been surpassed, and humanity must therefore decide on a pathway back to the safe operating space below the Planetary boundaries to minimise the risk of deleterious or even catastrophic environmental change at continental to global scales. However, the control variables used in the concept do not link to current indicators quantified in life-cycle assessments. Translating the Planetary Boundaries into maximum scores for life-cycle assessment indicators is therefore necessary to enable the assessment of whether life-cycle environmental impacts of e.g. products and services respect the Planetary Boundaries. This paper investigates two different approaches for translating the Planetary Boundary for Climate Change into estimations of maximum emission of greenhouse gasses expressed as annual emissions of CO2-equivalents, a commonly used Climate Change indicator in life-cycle assessments. The estimations may serve as an end goal benchmark for the life-cycle environmental impact of products and services in proposed pathways and roadmaps for the regeneration back to the Planetary Boundary for Climate Change.

Recently, the use of machine learning in meteorology has increased greatly. While many machine learning methods are not new, university classes on machine learning are largely unavailable to meteorology students and are not required to become a meteorologist. The lack of formal instruction has contributed to perception that machine learning methods are 'black boxes' and thus end-users are hesitant to apply the machine learning methods in their every day workflow. To reduce the opaqueness of machine learning methods and lower hesitancy towards machine learning in meteorology, this paper provides a survey of some of the most common machine learning methods. A familiar meteorological example is used to contextualize the machine learning methods while also discussing machine learning topics using plain language. The following machine learning methods are demonstrated: linear regression; logistic regression; decision trees; random forest; gradient boosted decision trees; naive Bayes; and support vector machines. Beyond discussing the different methods, the paper also contains discussions on the general machine learning process as well as best practices to enable readers to apply machine learning to their own datasets. Furthermore, all code (in the form of Jupyter notebooks and Google Colaboratory notebooks) used to make the examples in the paper is provided in an effort to catalyse the use of machine learning in meteorology.

FourCastNet, short for Fourier Forecasting Neural Network, is a global data-driven weather forecasting model that provides accurate short to medium-range global predictions at $0.25^{\circ}$ resolution. FourCastNet accurately forecasts high-resolution, fast-timescale variables such as the surface wind speed, precipitation, and atmospheric water vapor. It has important implications for planning wind energy resources, predicting extreme weather events such as tropical cyclones, extra-tropical cyclones, and atmospheric rivers. FourCastNet matches the forecasting accuracy of the ECMWF Integrated Forecasting System (IFS), a state-of-the-art Numerical Weather Prediction (NWP) model, at short lead times for large-scale variables, while outperforming IFS for variables with complex fine-scale structure, including precipitation. FourCastNet generates a week-long forecast in less than 2 seconds, orders of magnitude faster than IFS. The speed of FourCastNet enables the creation of rapid and inexpensive large-ensemble forecasts with thousands of ensemble-members for improving probabilistic forecasting. We discuss how data-driven deep learning models such as FourCastNet are a valuable addition to the meteorology toolkit to aid and augment NWP models.

Ensemble data from Earth system models has to be calibrated and post-processed. I propose a novel member-by-member post-processing approach with neural networks. I bridge ideas from ensemble data assimilation with self-attention, resulting into the self-attentive ensemble transformer. Here, interactions between ensemble members are represented as additive and dynamic self-attentive part. As proof-of-concept, I regress global ECMWF ensemble forecasts to 2-metre-temperature fields from the ERA5 reanalysis. I demonstrate that the ensemble transformer can calibrate the ensemble spread and extract additional information from the ensemble. As it is a member-by-member approach, the ensemble transformer directly outputs multivariate and spatially-coherent ensemble members. Therefore, self-attention and the transformer technique can be a missing piece for a non-parametric post-processing of ensemble data with neural networks.