As the severe impacts of climate change become increasingly apparent, concerns about climate-related issues have grown in recent years. The news media plays an important role in disseminating information about climate change and its consequences to the wider public and thus can influence public climate concern. Here, we investigate how extreme weather affects issue attention to climate change in the European online news media and how extreme weather and news coverage jointly shape changes in climate change concern. For the analysis, we combine 12 harmonized Eurobarometer survey waves, measuring public concerns about climate issues, with meteorological data and indices of environmental news coverage based on publications from 2481 media outlets in 200 regions of 22 European countries. Using fixed effects panel models, we estimate effects of temperature anomalies on climate news and climate concern and explore the role of the news media in explaining changes in concerns in response to temperature anomalies. The results indicate that unusually high temperatures exhibit a robust positive effect on media attention, especially when they overlap with other events that draw attention to the climate topic, such as major climate change conferences. We furthermore find evidence that the climate news in national outlets increases public concern about climate change and show that reporting by such outlets is likely to partly explain the effects of temperature anomalies on concerns. We do not find any significant effects of climate reporting in regional news outlets on climate concern. Our results suggest that the national news media partly mediates the effects of extreme weather on public climate change concern. The findings also highlight that focusing events strongly influence issue attention of the media, providing windows of opportunity to raise awareness about climate issues, while pointing to challenges in sustaining attention to related topics beyond short-lived news cycles.
Accurate forecasting of air pollution is important for environmental monitoring and policy support, yet data-driven models often suffer from limited generalization in regions with sparse observations. This paper presents Meteorology-Driven GPT for Air Pollution (GPT4AP), a parameter-efficient multi-task forecasting framework based on a pre-trained GPT-2 backbone and Gaussian rank-stabilized low-rank adaptation (rsLoRA). The model freezes the self-attention and feed-forward layers and adapts lightweight positional and output modules, substantially reducing the number of trainable parameters. GPT4AP is evaluated on six real-world air quality monitoring datasets under few-shot, zero-shot, and long-term forecasting settings. In the few-shot regime using 10% of the training data, GPT4AP achieves an average MSE/MAE of 0.686/0.442, outperforming DLinear (0.728/0.530) and ETSformer (0.734/0.505). In zero-shot cross-station transfer, the proposed model attains an average MSE/MAE of 0.529/0.403, demonstrating improved generalization compared with existing baselines. In long-term forecasting with full training data, GPT4AP remains competitive, achieving an average MAE of 0.429, while specialized time-series models show slightly lower errors. These results indicate that GPT4AP provides a data-efficient forecasting approach that performs robustly under limited supervision and domain shift, while maintaining competitive accuracy in data-rich settings.
Tropical cyclones (TC) are among the most destructive natural disasters, causing catastrophic damage to coastal regions through extreme winds, heavy rainfall, and storm surges. Timely monitoring of tropical cyclones is crucial for reducing loss of life and property, yet it is hindered by the computational inefficiency and high parameter counts of existing methods on resource-constrained edge devices. Current physics-guided models suffer from linear feature interactions that fail to capture high-order polynomial relationships between TC attributes, leading to inflated model sizes and hardware incompatibility. To overcome these challenges, this study introduces the Kolmogorov-Arnold Network-based Feature Interaction Framework (KAN-FIF), a lightweight multimodal architecture that integrates MLP and CNN layers with spline-parameterized KAN layers. For Maximum Sustained Wind (MSW) prediction, experiments demonstrate that the KAN-FIF framework achieves a $94.8\%$ reduction in parameters (0.99MB vs 19MB) and $68.7\%$ faster inference per sample (2.3ms vs 7.35ms) compared to baseline model Phy-CoCo, while maintaining superior accuracy with $32.5\%$ lower MAE. The offline deployment experiment of the FY-4 series meteorological satellite processor on the Qingyun-1000 development board achieved a 14.41ms per-sample inference latency with the KAN-FIF framework, demonstrating promising feasibility for operational TC monitoring and extending deployability to edge-device AI applications. The code is released at https://github.com/Jinglin-Zhang/KAN-FIF.
Abstract Rain-on-snow events, characterized by rainfall occurring over existing snowpack, have the potential to trigger significant hydrological and environmental impacts including snowmelt, flooding, landslides, and other natural hazards. Despite prior evidence suggesting the Connecticut River watershed has endured socioeconomic and environmental setbacks due to rain-on-snow events in the past, limited prior research has been performed on the role of rain-on-snow events within the region’s hydroclimatology. A clearer understanding of the occurrence and impacts of these events is essential for risk and water resource managers. This study created a climatology of rain-on-snow events in the Connecticut River watershed from 1981 to 2022, focusing on their spatiotemporal variability, trends, and hydrological impacts. Using daily gridded observations of snow water equivalent, precipitation, and temperature, rain-on-snow events were identified in the region, followed by an analysis of their frequency, intensity, and temporal patterns. Results revealed that higher elevations experienced more frequent rain-on-snow events and a higher average magnitude of snow water equivalent loss, while precipitation decreased with increasing latitude. There were significant decreasing trends in annual rain-on-snow frequency for parts of Vermont and Massachusetts, significant increasing (decreasing) trends in event-based snow water equivalent loss for parts of the lower (upper) watershed, and significant increasing trends in event-based precipitation for many parts of the watershed. The findings from this research will play a crucial role in future mitigation strategies, water resource management, and resilience within the watershed.
The Antarctic ice sheet (AIS) is sensitive to short‐term extreme meteorological events that can leave long‐term impacts on the continent's surface mass balance (SMB). We investigate the impacts of atmospheric rivers (ARs) on the AIS precipitation budget using an AR detection algorithm and a regional climate model (Modèle Atmosphérique Régional) from 1980 to 2018. While ARs and their associated extreme vapor transport are relatively rare events over Antarctic coastal regions (∼3 days per year), they have a significant impact on the precipitation climatology. ARs are responsible for at least 10% of total accumulated snowfall across East Antarctica (localized areas reaching 20%) and a majority of extreme precipitation events. Trends in AR annual frequency since 1980 are observed across parts of AIS, most notably an increasing trend in Dronning Maud Land; however, interannual variability in AR frequency is much larger. This AR behavior appears to drive a significant portion of annual snowfall trends across East Antarctica, while controlling the interannual variability of precipitation across most of the AIS. AR landfalls are most likely when the circumpolar jet is highly amplified during blocking conditions in the Southern Ocean. There is a fingerprint of the Southern Annular Mode (SAM) on AR variability in West Antarctica with SAM+ (SAM−) favoring increased AR frequency in the Antarctic Peninsula (Amundsen‐Ross Sea coastline). Given the relatively large influence ARs have on precipitation across the continent, it is advantageous for future studies of moisture transport to Antarctica to consider an AR framework especially when considering future SMB changes.
A numerical simulation experiment is conducted to study the first-ever waterspout observed in Victoria Harbour, Hong Kong, in 2024, namely, a mesoscale meteorological model with a spatial resolution of 200 m coupled with a computational fluid dynamics model with a spatial resolution of 4 m. It is found that the simulation could reproduce the observed wind field near the surface reasonably well, as well as the location of the waterspout and showers, as shown in the weather image. By conducting simulations with and without buildings, it is found that the inclusion of buildings is essential for the successful reproduction of the flow fields near the surface and up to several hundred metres high. This may suggest that urbanization plays a role in the occurrence of this waterspout. The resultant horizontal vorticity is then stretched by strong vertical motion at around 850 hPa, resulting in the waterspout, though no closed circulation could be simulated at the location of the waterspout. Moreover, the cyclonic feature for the flow field near the surface has a time lag of about 30 min compared with the actual waterspout occurrence. Nonetheless, the simulation is considered to be generally satisfactory and provides useful insight into the occurrence of the waterspout.
Florian E. Roemer, Stefan A. Buehler, Kaah P. Menang
Abstract Earth’s climate feedback quantifies the response of Earth’s energy budget to temperature changes and thus determines climate sensitivity. The climate feedback is largely controlled by water vapor which absorbs both longwave radiation emitted by Earth and shortwave radiation from the Sun. For the clear-sky shortwave water vapor feedback λ SW, a gap remains between process understanding and estimates from comprehensive climate models. Therefore, we present a hierarchy of simple models for λ SW. We show that λ SW is proportional to the change with temperature in the square of atmospheric transmissivity that depends on the atmospheric concentration of water vapor and its ability to absorb shortwave radiation. The global mean λ SW is well captured by a simple analytical model that approximates the strong spectral variations in water vapor absorption, whereas its temperature dependence results from spectral details in water vapor absorption. With this study, we expand the conceptual understanding of an important but understudied feedback component.
The statistical retrieval of atmospheric parameters will be greatly affected by the accuracy of the simulated brightness temperatures (BTs) derived from the radiative transfer model. However, it is challenging to further improve a physical-based radiative transfer model (RTM) developed based on the physical mechanisms of wave transmission through the atmosphere. We develop a deep neural network-based RTM (DNN-based RTM) to calculate the simulated BTs for the Microwave Temperature Sounder-II onboard the Fengyun-3D satellite under different weather conditions. The DNN-based RTM is compared in detail with the physical-based RTM in retrieving the atmospheric temperature profiles by the statistical retrieval scheme. Compared to the physical-based RTM, the DNN-based RTM can obtain higher accuracy for simulated BTs and enables the statistical retrieval scheme to achieve higher accuracy in temperature profile retrieval in clear, cloudy, and rainy sky conditions. Due to its ability to simulate microwave observations more accurately, the DNN-based RTM is valuable for the theoretical study of microwave remote sensing and the application of passive microwave observations.
Monitoring air pollution is crucial for protecting human health from exposure to harmful substances. Traditional methods of air quality monitoring, such as ground-based sensors and satellite-based remote sensing, face limitations due to high deployment costs, sparse sensor coverage, and environmental interferences. To address these challenges, this paper proposes a framework for high-resolution spatiotemporal Air Quality Index (AQI) mapping using sparse sensor data, satellite imagery, and various spatiotemporal factors. By leveraging Graph Neural Networks (GNNs), we estimate AQI values at unmonitored locations based on both spatial and temporal dependencies. The framework incorporates a wide range of environmental features, including meteorological data, road networks, points of interest (PoIs), population density, and urban green spaces, which enhance prediction accuracy. We illustrate the use of our approach through a case study in Lahore, Pakistan, where multi-resolution data is used to generate the air quality index map at a fine spatiotemporal scale.
Numerical weather prediction requires initial estimates of the atmospheric state. Since the atmospheric density field is intricately woven into the atmosphere's governing equations, advancing atmospheric density estimation will improve numerical weather prediction. However, current meteorological instrumentation cannot directly measure the atmospheric density field over large volumes. Existing techniques rely on sparse point measurements, limiting our ability to accurately estimate the three-dimensional atmospheric density field. One potential solution is to employ measurements of the atmospheric muon flux. Atmospheric muons are particles produced when energetic atomic nuclei (cosmic rays) collide with nuclei in the upper atmosphere, producing a shower of secondary particles (muons) that propagates to the Earth's surface. The surface atmospheric muon flux is known to be proportional to the local atmospheric density field, implying that this technique can be used as a measurement of atmospheric density. This study examines the potential for using atmospheric muon flux measurements to improve atmospheric state estimation via a case study of simulated atmospheric muon observations in the path of tropical cyclone Freddy. We show that improvement in data assimilation performance can be achieved using data from a relatively small astroparticle detector, well within the capabilities of existing astroparticle technology. We additionally show that the improvements to atmospheric state estimates associated with muon flux assimilation are at least partially unique to muon flux measurements, as comparable surface pressure point measurements do not reproduce a similar effect.
Daniele Zambon, Michele Cattaneo, Ivan Marisca
et al.
Accurate weather forecasts are essential for supporting a wide range of activities and decision-making processes, as well as mitigating the impacts of adverse weather events. While traditional numerical weather prediction (NWP) remains the cornerstone of operational forecasting, machine learning is emerging as a powerful alternative for fast, flexible, and scalable predictions. We introduce PeakWeather, a high-quality dataset of surface weather observations collected every 10 minutes over more than 8 years from the ground stations of the Federal Office of Meteorology and Climatology MeteoSwiss's measurement network. The dataset includes a diverse set of meteorological variables from 302 station locations distributed across Switzerland's complex topography and is complemented with topographical indices derived from digital height models for context. Ensemble forecasts from the currently operational high-resolution NWP model are provided as a baseline forecast against which to evaluate new approaches. The dataset's richness supports a broad spectrum of spatiotemporal tasks, including time series forecasting at various scales, graph structure learning, imputation, and virtual sensing. As such, PeakWeather serves as a real-world benchmark to advance both foundational machine learning research, meteorology, and sensor-based applications.
To support the selection of large optical/infrared telescope sites in western China, long-term monitoring of atmospheric conditions and astronomical seeing has been conducted at the Muztagh-Ata site on the Pamir Plateau since 2017. With the monitoring focus gradually shifting northward, three stations were established: the South Point, North-1 point, and North-2 point. The North-1 point,selected as the site for the Muztagh-Ata 1.93 m Synergy Telescope (MOST), has recorded seeing and meteorological parameters since late 2018. In 2023,the North-2 point was established approximately 1.5 km northeast of North-1 point as a candidate location for a future large-aperture telescope. A 10m DIMM tower and a PC-4A environmental monitoring system were installed to evaluate site quality. This study presents a comparative analysis of data from the North-1 and North-2 points during 2018-2024.The median seeing is 0.89 arcsecs at North-1 and 0.78 arcsecs at North-2. Both points show clear seasonal and diurnal variations,with winter nights offering optimal observing conditions.On average, about 64% of the nighttime duration per year is suitable for astronomical observations. Nighttime temperature variation is low :2.03 at North-1 and 2.10 at North-2 .Median wind speeds are 5-6 m/s, with dominant directions between 210 and 300, contributing to stable airflow. Moderate wind suppresses turbulence, while strong shear and rapid fluctuations degrade image quality. These findings confirm that both the North-1 and North-2 points offer high-quality atmospheric conditions and serve as promising sites for future ground-based optical/infrared telescopes in western China.
Wind energy is the most mature renewable energy technology, however, its exploitation in cities is often met with skepticism. Thanks to their ability to operate effectively at low wind from any direction, vertical axis wind turbines (VAWTs) offer an attractive opportunity for wind energy harvesting in cities, but limited evidence exists on their potential in complex urban environments, and the role of different geographical settings, local meteorological conditions, and urban characteristics remains unclear. Here we use realistic Weather Research and Forecast model high-resolution wind speed simulations alongside representative VAWT power curves to quantify the range of micro-generation potentials at the annual, seasonal, and diurnal scale across two Swiss cities (Lausanne and Geneva) residing in complex terrain. Our results show that Lausanne generally experiences higher (+24%) wind speed than Geneva. Both cities present the greatest micro-generation potential during the summer months, although Lausanne shows non-negligible potential also during the wintertime. Wind speed is higher during the nighttime in Lausanne and during the daytime in Geneva, due to the different interaction between the local lake-breeze circulation and the synoptic flow. Simulated performance of case-study VAWTs is dominated by cut-in wind speed and power curve inflection point. On average, in 2022, an individual VAWT would have produced 2665 kWh of total annual electricity, equivalent to 16.5 square meters of photovoltaic panels. These results highlight the need for research on urban wind energy featuring detailed city-scale assessments that account for urban heterogeneities and regional circulation patters, to inform future planning investment and engineering development.
Erin Coughlan de Perez, Weston Anderson, Eunjin Han
et al.
People have known that El Niño events are associated with low rainfall in Southern Africa for a century, and seasonal rainfall forecasts are now available in agricultural advisories for farmers. While there is abundant theory as to how farmers might (or should) use seasonal rainfall information on their farms, little is known about whether this information has been widely used or has had widespread benefit. In this study, we use subnational data on cropping area and yield to see if we can detect any macro-level patterns in agricultural choices or outcomes that are related to knowledge of the El Niño Southern Oscillation or seasonal forecast information in Southern Africa. We find that in Lesotho and parts of South Africa, planted area of maize and sorghum is reduced when there is a dry start to the season and an El Niño event is apparent at the time of planting. Similarly, we find that in both Lesotho and most provinces of South Africa, drought years associated with El Niño have worse yields than drought years that are not associated with El Niño (controlling for rainfall). This association could indicate that people are discouraged during El Niño years by the potential for drought, and they might be reducing cropping area, reducing agricultural investments, or turning to other income-generating activities. We are unable to detect a relationship between yields and the accuracy of seasonal rainfall forecasts, therefore we are unable to observe any additional yield benefit when more accurate seasonal forecast information is available.
Meteorology. Climatology, Social sciences (General)
Numerous countries have built urban stations for monitoring the amount of PM2.5 in the atmosphere. In Iraq, there aren't enough stations to monitor PM2.5 pollution levels across all governorates. As a result, satellite remote sensing data is used in the majority of studies aimed at monitoring PM2.5 and the impact of other factors on it. The current study aimed to analyze the spatial and temporal distribution of (PM2.5) and its relationship with the meteorological parameters.(Air temperature, Relative humidity, Precipitation and wind speed) in Iraq during two periods (2001 and 2022). The dataset adopted in the study were downloaded from the Giovanni user interface which is based on satellite remote sensing data and reanalysis by MERRA-2model which produce by NASA. The output results shows that, the seasonal and annual PM2.5 concentration values increased from 2001 to 2022 due especially in the center and south of Iraq with the highest values of PM2.5 concentration recorded in the summers of 2001 and 2022 being 172.41 micro.g/m3 and 190.06 micro.g/m3 (increased 10.24%), respectively. Because of the low average temperature and the influence of northeasterly winds bringing continental air from Central Asia, PM2.5 values in northern and northeastern Iraq are lower than those in the center and southern regions. in 2001, they ranged from 8.41 to 12.6 micro.g/m3, whereas in 2022, they ranged from 9.02 to 15.98 micro.g/m3 throughout the year. Rainfall during the cold months in the north and northeast is an essential factor in cleaning the air of PM2.5. Also, study results indicate that the max. of PM 2.5 values have consistently exceeded the upper limits of PM2.5 quarterly standards set by both the US and Iraqi regulations, for the years 2001 and 2022, but the min. PM2.5 values are within both standards.
As climate change intensifies, the shift to cleaner energy sources becomes increasingly urgent. With wind energy production set to accelerate, reliable wind probabilistic forecasts are essential to ensure its efficient use. However, since numerical weather prediction models are computationally expensive, probabilistic forecasts are produced at resolutions too coarse to capture all mesoscale wind behaviors. Statistical downscaling, typically applied to enchance the resolution of climate model simulations, presents a viable solution with lower computational costs by learning a mapping from low-resolution (LR) variables to high-resolution (HR) meteorological variables. Leveraging deep learning, we evaluate a downscaling model based on a state-of-the-art U-Net architecture, applied to an ensemble member from a coarse-scale probabilistic forecast of wind velocity. The architecture is modified to incorporate (1) a learned grid alignment strategy to resolve LR-HR grid mismatches and (2) a processing module for multi-level atmospheric predictors. To extend the downscaling model's applicability from fixed spatial domains to the entire Canadian region, we assess a transfer learning approach. Our results show that the learned grid alignment strategy performs as well as conventional pre-processing interpolation steps and that LR wind speed at multiple levels is sufficient as a predictor, enabling a more compact architecture. Additionally, they suggest that extending to new spatial domains using transfer learning is promising, and that downscaled wind velocities demonstrate potential in improving the detection of wind power ramps, a critical phenomenon for wind energy.
Anthony Frion, Lucas Drumetz, Guillaume Tochon
et al.
In the context of an increasing popularity of data-driven models to represent dynamical systems, many machine learning-based implementations of the Koopman operator have recently been proposed. However, the vast majority of those works are limited to deterministic predictions, while the knowledge of uncertainty is critical in fields like meteorology and climatology. In this work, we investigate the training of ensembles of models to produce stochastic outputs. We show through experiments on real remote sensing image time series that ensembles of independently trained models are highly overconfident and that using a training criterion that explicitly encourages the members to produce predictions with high inter-model variances greatly improves the uncertainty quantification of the ensembles.
Myanmar’s climate is heavily influenced by its geographic location and relief. Located between the Indian summer monsoon (ISM) and the East Asian summer monsoon (EASM), Myanmar’s climate is distinguished by the alternation of seasons known as the monsoon. The north-south direction of peaks and valleys creates a pattern of alternate zones of heavy and scanty precipitation during both the northeast and southwest monsoons. The majority of the rainfall has come from Myanmar’s southwest monsoon (MSwM), which is Myanmar’s rainy season (summer in global terms, June–September). This study explained both threshold-based and nonthreshold-based objective definitions of the onset and withdrawal of large-scale MSwM. The seasonal transitions in MSwM circulation and precipitation are convincingly represented by the new index, which is based on change point detection of the atmospheric moisture flow converging in the MSwM region (10–28 N, 92–102 E). A transition in vertically integrated moisture transport (VIMT), the reversal of surface winds, and an increase in precipitation may also be considered when defining MSwM onset objectively. We also define a change point of the MSwM (CPI) index for MSwM onset and withdrawal dates. The climatological mean onset of MSwM is day 135 (May 14), withdrawal is day 278 (October 4), and the total season length is 144 days. We are investigating spatial patterns of rainfall progression at and after the start of the monsoon, rather than transitions within a single region of the MSwM. The local southwest monsoon duration is well correlated with the CPI duration on interannual timescales, particularly in the peak rainfall regions, with a delay (advance) in large-scale onset or withdrawal associated with a delay (advance) of onset or withdrawal by local index. Hence, the next phase of this research is to study the maintenance and break of the monsoon to understand the underlying physical processes governing the monsoon circulation. The results of this study provide a possibility to reconstruct Myanmar’s monsoon climate dynamics, and the findings of this study can help unravel many remaining questions regarding the greater Asian monsoon system’s variability.