Hasil untuk "Meteorology. Climatology"

Jing Wu, Yang Liu, Lin Zhang et al.

Accurate weather forecasting is more than grid-wise regression: it must preserve coherent synoptic structures and physical consistency of meteorological fields, especially under autoregressive rollouts where small one-step errors can amplify into structural bias. Existing physics-priors approaches typically impose global, once-for-all constraints via architectures, regularization, or NWP coupling, offering limited state-adaptive and sample-specific controllability at deployment. To bridge this gap, we propose Agent-Guided Cross-modal Decoding (AGCD), a plug-and-play decoding-time prior-injection paradigm that derives state-conditioned physics-priors from the current multivariate atmosphere and injects them into forecasters in a controllable and reusable way. Specifically, We design a multi-agent meteorological narration pipeline to generate state-conditioned physics-priors, utilizing MLLMs to extract various meteorological elements effectively. To effectively apply the priors, AGCD further introduce cross-modal region interaction decoding that performs region-aware multi-scale tokenization and efficient physics-priors injection to refine visual features without changing the backbone interface. Experiments on WeatherBench demonstrate consistent gains for 6-hour forecasting across two resolutions (5.625 degree and 1.40625 degree) and diverse backbones (generic and weather-specialized), including strictly causal 48-hour autoregressive rollouts that reduce early-stage error accumulation and improve long-horizon stability.

Irene Iele, Giulia Romoli, Daniele Molino et al.

Accurate short-term forecasting of vegetation dynamics is a key enabler for data-driven decision support in precision agriculture. Normalized Difference Vegetation Index (NDVI) forecasting from satellite observations, however, remains challenging due to sparse and irregular sampling caused by cloud coverage, as well as the heterogeneous climatic conditions under which crops evolve. In this work, we propose a probabilistic forecasting framework specifically designed for field-level NDVI prediction under clear-sky acquisition constraints. The method leverages a transformer-based architecture that explicitly separates the modeling of historical vegetation dynamics from future exogenous information, integrating historical NDVI observations with both historical and future meteorological covariates. To address irregular revisit patterns and horizon-dependent uncertainty, we introduce a temporal-distance weighted quantile loss that aligns the training objective with the effective forecasting horizon. In addition, we incorporate cumulative and extreme-weather feature engineering to better capture delayed meteorological effects relevant to vegetation response. Extensive experiments on European satellite data demonstrate that the proposed approach consistently outperforms a diverse set of statistical, deep learning, and recent time series baselines across both point-wise and probabilistic evaluation metrics. Ablation studies further highlight the central role of target history, while showing that meteorological covariates provide complementary gains when jointly exploited. The code is available at https://github.com/arco-group/ndvi-forecasting.

Peter Bespalov, Olga Savina, Polina Shkareva

The models of excitation of quasiperiodic ELF/VLF emissions with spectral shape repetition periods from 10 to 300 s are discussed. The primary cause of quasiperiodic (QP) emissions is cyclotron instability of electron radiation belts. Relatively slow processes of cyclotron instability evolution are well described within the framework of the plasma magnetospheric maser (PMM) theory based on the averaged self-consistent system of quasilinear equations for particles and waves. The presence of an eigen-frequency of oscillations of PMM parameters allows explaining many properties of QP 1 emissions, in which not very clear spectral bursts are hiss with resonant modulation mainly near the upper spectral boundary by geomagnetic pulsations of the Pc 3–4 range. The analysis of the general problem of equilibrium of radiation belts shows the possibility of its instability, which is caused by the difference in the pitch-angle dependences of the particle source power and the steady state distribution function. In the nonlinear mode of the specified instability, QP 2 emissions are formed, often with an increase in frequencies in individual spectral bursts. This paper mainly focuses on the study of QP 2 emissions with both a normal and an atypical time structure, as well as with large and fast dynamics of the frequency spectrum. Periodic large-scale atmospheric disturbances with a suitable frequency on the ionosphere can significantly affect the operating modes of the PMM and, as a consequence, the quasiperiodic VLF emissions in the magnetosphere. Infrasonic waves at the altitudes of the E region of the ionosphere can provide excitation of atypical quasiperiodic emissions due to a change in the reflection coefficient of whistler waves from the ionosphere from above. The obtained results are important for interpreting observational data on emissions associated with large-scale processes in the atmosphere. To analyze the magnetosphere response to earthquakes, observation data from the Van Allen Probe spacecraft were used. Also, specific examples of quasiperiodic emissions, probably associated with large-scale atmospheric processes, were obtained during the analysis of observational data.

Ho Yeon Shin, Jaemin Kim, Daegeun Shin et al.

Atmospheric greenhouse gases (GHGs) affect Earth’s radiation balance and are the primary drivers of climate change. This study analyzed the spatiotemporal variability of carbon dioxide (CO₂), methane (CH₄), and nitrous oxide (N₂O) at domestic World Meteorological Organization/Global Atmosphere Watch (WMO/GAW) sites located at Anmyeondo (AMY), Jeju Gosan, and Ulleungdo, examining the local influences on GHG variations and comparing the relationships between gases. Long-term records from the AMY site were used to investigate temporal changes in CO₂–CH₄. The results showed that short-term variations were influenced by local emissions, sink processes, and anthropogenic signals, whereas medium-to-long-term variations displayed clear seasonality driven by vegetation and meteorological changes, with continuously increasing trends. The sites strongly reflected the effects of nearby point sources, and the ∆CO₂–∆CH₄ relationships revealed site-specific spatiotemporal differences. At AMY, 46–49% of top-quartile CO₂, CH₄, and N₂O enhancements occurred under easterly winds from nearby industrial and agricultural sources, whereas 14–16% under northwesterly flow indicated episodic transport from eastern China, highlighting the site’s combined exposure to domestic and foreign emissions. The observed strengthening long-term ∆CO₂–∆CH₄ correlation may be related to continuously increasing emissions in East Asia. However, uncertainties remain, owing to changes resulting from the 2012 instrument replacement and calibration scale. Overall, this study provides baseline insights into domestic GHG variability and offers fundamental information for understanding East Asian emissions and supporting climate policy.

Juliette Lavoie, Louis-Philippe Caron, Travis Logan et al.

Bias-adjusted climate simulations are increasingly disseminated through online platforms to support adaptation actions. However, there is no consensus on an operational framework to choose what to include in these “decision-ready” ensembles and for communicating the related uncertainty. In this paper, we use a systematic approach to assess the uncertainty related to bias-adjusted climate simulations across five dimensions: internal variability, greenhouse gases scenario, global climate model, observational reference and bias-adjustment method. We calculate the fraction of uncertainty associated with each dimension for precipitation-based, temperature-based and multivariate indicators over eastern Canada and focus particularly on three locations: Montréal, Gaspé and Kawawachikamach. The results show that the uncertainty associated with the reference dataset can be very large and in some instances can become the first or second largest source of uncertainty. Using simple examples, we show that the resulting differences could lead to different conclusions with respect to some adaptation solutions or possibly create confusion with users. These results raise questions on the robustness of climate projections distributed through these web platforms and the ethical responsibility of data providers to adequately evaluate and communicate the underlying uncertainty.

Hedibert F. Lopes, Filippo Ascolani

Mike West is currently the Arts & Sciences Distinguished Professor Emeritus of Statistics and Decision Sciences at Duke University. Mike's research in Bayesian analysis spans multiple interlinked areas: theory and methods of dynamic models in time series analysis, foundations of inference and decision analysis, multivariate and latent structure analysis, stochastic computation and optimisation, among others. Inter-disciplinary R&D has ranged across applications in commercial forecasting, dynamic networks, finance, econometrics, signal processing, climatology, systems biology, genomics and neuroscience, among other areas. Among Mike's currently active research areas are forecasting, causal prediction and decision analysis in business, economic policy and finance, as well as in personal decision making. Mike led the development of academic statistics at Duke University from 1990-2002, and has been broadly engaged in professional leadership elsewhere. He is past president of the International Society for Bayesian Analysis (ISBA), and has served in founding roles and as board member for several professional societies, national and international centres and institutes. Recipient of numerous awards, Mike has been active in research with various companies, banks, government agencies and academic centres, co-founder of a successful biotechnology company, and board member for several financial and IT companies. He has published 4 books, several edited volumes and over 200 papers. Mike has worked with many undergraduate and Master's research students, and as of 2025 has mentored around 65 primary PhD students and postdoctoral associates who moved to academic, industrial or governmental positions involving advanced statistical and data science research.

Christopher J. Paciorek, Daniel Cooley

Weather extremes produce major impacts on society and ecosystems and are likely to change in likelihood and magnitude with climate change. However, very low probability events are hard to characterize statistically using observations or climate model output because of short records/runs. For precipitation, consideration of such events arises in quantifying Probable Maximum Precipitation (PMP), namely estimating extreme precipitation magnitudes for designing and assessing critical infrastructure. A recent National Academies report on modernizing PMP estimation proposed using huge climate model-based ensembles to estimate extreme quantiles, possibly through machine learning-based ensemble boosting. Here we assess such an approach for the contiguous United States using a huge ensemble (10560 years) from a state-of-the-art emulator (ACE2) trained on ERA5 reanalysis. The results indicate that one can practically estimate very extreme precipitation and temperature quantiles using appropriate statistical extreme value techniques. More specifically, the results provide evidence for (1) the use of threshold-exceedance methods with a sufficiently high threshold for reliable estimation (necessary for precipitation), (2) the robustness of results to variations in extremes by season and storm type, and (3) well-constrained statistical uncertainty. Our results also show that the emulator produces extremes outside the range of the ERA5 training data. While this suggests that such emulators have potential for quantifying the climatology of extremes, we do not extensively investigate if this particular emulator is fit for purpose. Our focus is on how to use huge ensembles to estimate very extreme statistics, and we expect the results to be relevant for future improved emulators.

Eamonn Organ, Maeve Upton, Denis Allard et al.

Accurate high-resolution spatial and temporal wind speed data is critical for estimating the wind energy potential of a location. For real-time wind speed prediction, statistical models typically depend on high-quality (near) real-time data from official meteorological stations to improve forecasting accuracy. Personal weather stations (PWS) offer an additional source of real-time data and broader spatial coverage than official stations. However, they are not subject to rigorous quality control and may exhibit bias or measurement errors. This paper presents a framework for incorporating PWS data into statistical models for validated official meteorological station data via a two-stage approach. First, bias correction is performed on PWS wind speed data using reanalysis data. Second, we implement a Bayesian hierarchical spatio-temporal model that accounts for varying measurement error in the PWS data. This enables wind speed prediction across a target area, and is particularly beneficial for improving predictions in regions sparse in official monitoring stations. Our results show that including bias-corrected PWS data improves prediction accuracy compared to using meteorological station data alone, with a 5% reduction in prediction error on average across all sites. The results are comparable with popular reanalysis products, but unlike these numerical weather models our approach is available in real-time and offers improved uncertainty quantification. are comparable with popular reanalysis products, but unlike these numerical weather models our approach is available in real-time and offers improved uncertainty quantification.

Md Sultanul Arifin, Abu Nowshed Sakib, Yeasir Rayhan et al.

Lightning, a common feature of severe meteorological conditions, poses significant risks, from direct human injuries to substantial economic losses. These risks are further exacerbated by climate change. Early and accurate prediction of lightning would enable preventive measures to safeguard people, protect property, and minimize economic losses. In this paper, we present DeepLight, a novel deep learning architecture for predicting lightning occurrences. Existing prediction models face several critical limitations: i) they often struggle to capture the dynamic spatial context and the inherent randomness of lightning events, including whether lightning occurs and its variability in location and timing even under similar meteorological conditions; ii) they underutilize key observational data, such as radar reflectivity and cloud properties; and iii) they rely heavily on Numerical Weather Prediction (NWP) systems, which are both computationally expensive and highly sensitive to parameter settings. To overcome these challenges, DeepLight leverages multi-source meteorological data, including radar reflectivity, cloud properties, and historical lightning occurrences through a dual-encoder architecture. By employing multi-branch convolution techniques, it dynamically captures spatial correlations across varying extents. Furthermore, its novel Hazy Loss function explicitly addresses the spatio-temporal uncertainty of lightning by penalizing deviations based on proximity to true events, enabling the model to better learn patterns amidst randomness. Extensive experiments show that DeepLight improves the Equitable Threat Score (ETS) by 18\%--30\% over state-of-the-art methods, establishing it as a robust solution for lightning prediction.

Scott A. Martin, Noah Brenowitz, Dale Durran et al.

Accurate long-range weather forecasting remains a major challenge for AI models, both because errors accumulate over autoregressive rollouts and because reanalysis datasets used for training offer a limited sample of the slow modes of climate variability underpinning predictability. Most AI weather models are autoregressive, producing short lead forecasts that must be repeatedly applied to reach subseasonal-to-seasonal (S2S) or seasonal lead times, often resulting in instability and calibration issues. Long-timestep probabilistic models that generate long-range forecasts in a single step offer an attractive alternative, but training on the 40-year reanalysis record leads to overfitting, suggesting orders of magnitude more training data are required. We introduce long-range distillation, a method that trains a long-timestep probabilistic "student" model to forecast directly at long-range using a huge synthetic training dataset generated by a short-timestep autoregressive "teacher" model. Using the Deep Learning Earth System Model (DLESyM) as the teacher, we generate over 10,000 years of simulated climate to train distilled student models for forecasting across a range of timescales. In perfect-model experiments, the distilled models outperform climatology and approach the skill of their autoregressive teacher while replacing hundreds of autoregressive steps with a single timestep. In the real world, they achieve S2S forecast skill comparable to the ECMWF ensemble forecast after ERA5 fine-tuning. The skill of our distilled models scales with increasing synthetic training data, even when that data is orders of magnitude larger than ERA5. This represents the first demonstration that AI-generated synthetic training data can be used to scale long-range forecast skill.

Corina Dunn, Xing Meng, Olga P. Verkhoglyadova

Abstract The Equatorial Ionization Anomaly (EIA) crest location is known to vary over a variety of temporal scales. For the first time we perform a statistical survey of the temporal variation of the EIA crest location viewed globally and spanning 20 years. We extract the crest location for double‐peaked EIAs from a data set of total electron content intensifications identified on global ionospheric maps from 2003 to 2022. We show that the dominant temporal variations of the crest latitude are annual and semi‐diurnal for the northern crest, and annual and diurnal for the southern crest. For the annual variation, we find that both crests move poleward in local summer and equatorward in local winter, which is more pronounced for the southern crest than the northern crest, and more pronounced at solar minimum than solar maximum. For the diurnal and semi‐diurnal variations in universal time, both crests dip southward around 15UT and the northern crest additionally dips southward around 2.5UT. We consider apparent universal time dependence to be a proxy for the longitudinal distribution of the crest geomagnetic latitude, which exhibits the known wave‐number‐four longitudinal structure of EIA crests. In local time, the EIA crests form earlier than 10LT and move poleward to their maximum distance at 14LT, and remain at constant latitude until 18LT. Solar cycle modulation on the diurnal/semi‐diurnal variations and the local time evolution of the crest latitude is minimal.

Yanfang Hou, Wenliang Liu, Litao Wang et al.

The regional differences in carbon dioxide (CO₂) variations from the Orbiting Carbon Observatory-2 (OCO-2) over the Beijing-Tianjin-Hebei (Jing-Jin-Ji) region from 2015 to 2021 are analyzed in this study. This study shows an annual increase and a seasonal cycle; the CO₂ annual growth rate was about 2.63 ppm year⁻¹, with the highest value being in spring and the lowest in summer. The spatial distribution is unbalanced, regional differences are prominent, and the CO₂ concentration is lower in the north of the Jing-Jin-Ji region (like Zhangjiakou, Chengde, and Qinhuangdao). Land-type structures and population economy distributions are the key factors affecting CO₂ concentration. By analyzing the land-type structures over Jing-Jin-Ji in 2020, we find that cropland, woodland, and grassland (CWG) are the main land cover types in Jing-Jin-Ji; the proportion of these three types is about 83.3%. The woodland areas in Zhangjiakou, Chengde, and Qinhuangdao account for about 65% of the total woodland areas in Jing-Jin-Ji; meanwhile, the grassland areas in these three regions account for 62% of the total grassland areas in Jing-Jin-Ji. CO₂ concentration variation shows a high negative correlation with CWG land areas (coefficient of determination (R²) > 0.76). The regions with lower population and GDP secondary industry (SI) density also have lower CO₂ concentration (like Zhangjiakou, Chengde, and Qinhuangdao), and the regions with higher population and GDP SI density also have higher CO₂ concentration (like the southeast of Jing-Jin-Jin).

Xinyu Li, Riyu Lu, Guixing Chen et al.

Abstract It is generally believed that the Maritime Continent (MC) is rarely affected by tropical cyclones (TCs) due to its equatorial location. However, this study reveals that TCs in the tropical western North Pacific can significantly suppress rainfall over the MC and its surrounding seas, based on the composite analysis. This suppression effect of TCs exists across all phases of the Madden–Julian Oscillation (MJO). TCs greatly alleviate rainfall enhancement during the convective phases of the MJO and aggravate rainfall suppression during the suppressive phases. Particularly, TCs reduce the likelihood of extremely high rainfall in convective MJO phases from 9% to 5% and increase the likelihood of extremely low rainfall in suppressive MJO phases from 10% to 16%. The rainfall suppression is attributed to the lower-tropospheric southwesterly anomalies to the south of TCs, which result in moisture divergence over the MC. Additionally, the upper-tropospheric equatorward outflows of TCs also promote subsidence and suppress rainfall. This study introduces a new factor influencing the rainfall over the MC from a synoptic climatology perspective.

Arnab Kumar Mallik

The deposition of micron particles finds importance in meteorology and several engineering applications such as deposition of dust in gas lines, carbon deposition in engine exhaust, designing effective air-cleaning systems and estimating deposition of inhaled drug or atmospheric pollutants to determine its consequences on human health. Although the existing literature on deposition in straight tubes is quite mature, an experimental study on deposition in micro capillaries with a wide ranges of Re that models particle dynamics in lungs, is missing. The deposition of atmospheric pollutants and nebulized drugs in the lung depends on various biological factors such as flow properties, lung morphology, breathing patterns, particle properties, deposition mechanism, etc. To complicate matters, each breath manifests flows spanning a wide range of Reynolds numbers in various regions of the lung. In this study, the deposition of nebulized aerosol was experimentally investigated in phantom bronchioles of diameters relevant to the 7th to the 23rd branching generations and over the entire range of Re manifest during one breathing cycle. The aerosol fluid was loaded with boron doped carbon quantum dots as a fluorophore. An aerosol was generated of this mixture fluid using an ultrasonic nebulizer, producing droplets of 6.5$μ$m as the mean diameter. The amount of aerosol deposited on the bronchiole walls was measured using a spectrofluorometer. Finally, a universal bronchiole scale deposition model is proposed which can form the building block for lung-scale aerosol deposition prediction.

Forkan Uddin Ahmed, Annesha Das, Md Zubair

The development of an intelligent agricultural decision-supporting system for crop selection and disease forecasting in Bangladesh is the main objective of this work. The economy of the nation depends heavily on agriculture. However, choosing crops with better production rates and efficiently controlling crop disease are obstacles that farmers have to face. These issues are addressed in this research by utilizing machine learning methods and real-world datasets. The recommended approach uses a variety of datasets on the production of crops, soil conditions, agro-meteorological regions, crop disease, and meteorological factors. These datasets offer insightful information on disease trends, soil nutrition demand of crops, and agricultural production history. By incorporating this knowledge, the model first recommends the list of primarily selected crops based on the soil nutrition of a particular user location. Then the predictions of meteorological variables like temperature, rainfall, and humidity are made using SARIMAX models. These weather predictions are then used to forecast the possibilities of diseases for the primary crops list by utilizing the support vector classifier. Finally, the developed model makes use of the decision tree regression model to forecast crop yield and provides a final crop list along with associated possible disease forecast. Utilizing the outcome of the model, farmers may choose the best productive crops as well as prevent crop diseases and reduce output losses by taking preventive actions. Consequently, planning and decision-making processes are supported and farmers can predict possible crop yields. Overall, by offering a detailed decision support system for crop selection and disease prediction, this work can play a vital role in advancing agricultural practices in Bangladesh.

Nikolaos A. Bakas

Stergios Misios, Andreas Chrysanthou, Kostas Tsigaridis et al.

Chengyan Liu, Zhaomin Wang, Xi Liang et al.

Tides play a key role in regulating the circulation and water properties around Antarctica, yet tidal currents and the corresponding influences in Prydz Bay have not been quantified with observational datasets. This study focuses on the observed characteristics of tidal currents and quantifies the tidal contribution to the basal melting of the Amery Ice Shelf (AIS). Long-term hydrography observations are provided by 10 moorings over the continental shelf and 6 borehole sites drilled through the AIS. Based on the mooring observations, we analyze the observed tidal currents, and evident seasonality in the mixed diurnal–semidiurnal tidal currents is identified. Barotropic tides dominate the tidal currents at the AIS front, except at the western corner where the seabed is abruptly deeper. The spatially and temporally averaged magnitude of tidal currents for all the current meter records is only ~3 cm s−1. However, an observed maximum tidal velocity of ~11 cm s−1 occurs at the AIS front, and the maximal time-averaged tidal kinetic energy reaches ~31% of the total kinetic energy over the outer continental shelf. Based on the borehole observations, tide-like pulsing is identified in the ocean layer adjacent to the AIS basal surface. The maximal tidal contribution of the AIS basal melting is estimated at 69% based on a simple model. Due to the paucity of long-term velocity observations in the sub-ice-shelf cavity, the uncertainties of the estimated tide-induced melting in this study could be attributed to the simulated velocity.

Chuloh Jung, Naglaa Sami Abdelaziz Mahmoud

This study investigates the effectiveness of different ventilation methods in reducing indoor air pollutants in newly constructed residential buildings, focusing on indoor air quality (IAQ) in Dubai. The paper highlights the growing concern for IAQ in response to residents’ increasing awareness of their well-being and environmental sustainability. The study examines the concentrations of formaldehyde (HCHO), volatile organic compounds (VOCs), and total volatile organic compounds (TVOC) in bedrooms and living rooms before and after implementing various ventilation methods during the construction phase. The findings indicate that mechanical exhaust ventilation, mainly through bathroom and kitchen exhaust fans, was highly effective in reducing HCHO levels. Combining kitchen and bathroom exhaust fans demonstrated the most significant reduction in HCHO concentrations. Similarly, reductions in VOCs, such as ethylbenzene, toluene, and xylene, were observed with different ventilation methods. Natural ventilation also proved effective in reducing pollutant concentrations. The results emphasize the importance of implementing appropriate ventilation strategies to improve IAQ in residential buildings. However, the study acknowledges the limitations of a single-location measurement and recommends further research to validate the findings across different building types and locations. Additionally, long-term studies are necessary to assess the sustained effects of ventilation methods on IAQ. The study highlights the significance of addressing IAQ concerns in residential buildings and suggests potential research directions to explore other ventilation strategies and their energy efficiency implications. Ultimately, this research contributes to developing healthier and sustainable living environments by promoting effective ventilation strategies to mitigate indoor air pollutants.

A. S. Pepler, I. Rudeva

<p>Cold fronts make a significant contribution to cool season rainfall in the extratropics and subtropics. In many regions of the Southern Hemisphere the amount of frontal rainfall has declined in recent decades, but there has been no change in frontal frequency. We show that for southeast Australia this contradiction cannot be explained by changes in frontal intensity or moisture at the latitudes of interest. Rather, declining frontal rainfall in southeast Australia is associated with weakening of the subtropical westerlies in the mid-troposphere, which is part of a hemispheric pattern of wind anomalies that modify the extratropical zonal wave 3. Fronts that generate rainfall are associated with strong westerlies that penetrate well into the subtropics, and the observed decrease in frontal rainfall in southern Australia can be linked to a decrease in the frequency of fronts with strong westerlies at 25<span class="inline-formula">^∘</span> S.</p>