Meteorological Applications最新文献

{"title":"Harnessing Subseasonal-to-Seasonal Predictability for Wind Energy in Kenya","authors":"Gibbon Innocent Tirivanhu Masukwedza,&nbsp;Linda C. Hirons,&nbsp;Steven J. Woolnough,&nbsp;Cohen Ang'u,&nbsp;Hannah C. Bloomfield,&nbsp;Caroline M. Wainwright,&nbsp;David James Brayshaw,&nbsp;Eve de Vincelles,&nbsp;Wilson Gitau,&nbsp;Joseph Mutemi,&nbsp;Willis Owino Ochieng,&nbsp;Daniel Olago,&nbsp;Christopher Oludhe","doi":"10.1002/met.70221","DOIUrl":"https://doi.org/10.1002/met.70221","url":null,"abstract":"<p>Understanding how climate variability influences wind-power resources is critical for energy security, particularly in countries such as Kenya where reliance on weather-sensitive generation is increasing. Using ERA5 reanalysis, climate-mode indices, site observations where available and modelled wind-power potential, we quantify how three major tropical climate modes relevant to subseasonal-to-seasonal (S2S) wind-power variability, the Madden–Julian Oscillation (MJO), El Niño–Southern Oscillation (ENSO) and Indian Ocean Dipole (IOD), influence Kenya's wind-energy system. Results show that MJO Phases 2–4 generally suppress wind-energy potential across Kenya, while Phases 6–8 enhance it. ENSO influences are weaker and more seasonally selective, with El Niño favouring suppressed wind-energy potential during October–November–December and La Niña favouring enhancement during April–May–June. The IOD imprint is strongest in October–November–December, when positive IOD conditions weaken wind-energy potential across key inland regions and negative IOD conditions strengthen it. Using a perfect-forecast framework that assumes known MJO, ENSO and IOD states, with forecast skill evaluated through leave-one-out cross-validation, we quantify the potential value of climate-mode information for wind-power prediction. We show that conditioning on the MJO provides the largest single-mode improvements in deterministic and probabilistic wind-power forecast skill relative to monthly climatology. Conditioning on ENSO and the IOD yields smaller, season-dependent skill gains, with the strongest improvements occurring in October–November–December. Combined MJO–ENSO–IOD conditioning gives the largest overall skill improvement. Taken together, these results identify a clear hierarchy of predictability: MJO information provides the strongest subseasonal contribution to wind-power prediction, while ENSO and IOD provide broader seasonal context and smaller refinements. This hierarchy supports the development of climate-mode-informed “windows of risk and opportunity” that translate S2S climate information into practical services for anticipatory, risk-informed wind-energy operations and planning in Kenya, with potential relevance for wider East Africa.</p>","PeriodicalId":49825,"journal":{"name":"Meteorological Applications","volume":"33 4","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/met.70221","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148849098","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Transferability of Machine Learning-Based Calibration Models for Low-Cost PM2.5 Sensors Across Heterogeneous African Cities","authors":"Priscilla Adong,&nbsp;Engineer Bainomugisha","doi":"10.1002/met.70228","DOIUrl":"https://doi.org/10.1002/met.70228","url":null,"abstract":"<p>The expansion of low-cost particulate matter (PM_2.5) sensor networks offers an opportunity to strengthen air quality monitoring and management across Africa, where reference-grade instrumentation remains sparse. However, their effectiveness depends on robust calibration against reference-grade monitors. Machine learning (ML) calibration methods have proven effective in correcting low-cost sensor biases under local collocation conditions, but their scalability and transferability across heterogeneous environments have not been systematically evaluated. As sensor networks expand beyond well-instrumented cities, calibration is no longer solely a modelling challenge but a deployment and scalability problem. This study assesses the transferability of ML-based calibration models across five monitoring sites in four African cities representing diverse meteorological patterns and PM_2.5 concentration profiles. We compare four calibration strategies: (1) localised calibration using on-site reference monitors; (2) direct transfer without local reference data; (3) multicity pooled models evaluated using leave-one-out cross-validation (LOOCV) and (4) environmental clustering, in which cities were grouped using K-means according to similarity in meteorological conditions and PM_2.5 distributions prior to LOOCV evaluation (cluster-based LOOCV). Results show that localised calibration consistently provides the highest predictive performance, confirming the value of site-specific reference data. Direct transfer performs poorly when applied to cities not from the same cluster, highlighting the risks of naive model transfer. In contrast, both LOOCV and cluster-based LOOCV improve ML model generalisation, with cluster-based LOOCV achieving performance comparable to LOOCV despite requiring less training data. These findings indicate that environmental similarity is a key determinant of calibration transferability. Our results demonstrate that strategically placed reference-grade monitors in representative environmental clusters can support scalable calibration of distributed low-cost networks across multiple cities. This approach reduces reliance on extensive collocation campaigns while maintaining acceptable performance. The findings provide a practical framework for designing scalable, data-driven air quality networks in resource-constrained settings.</p>","PeriodicalId":49825,"journal":{"name":"Meteorological Applications","volume":"33 4","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/met.70228","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148848537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multidecadal Projection of Annual Precipitation in Iran Using Spatially Bias-Adjusted ARIMA Models","authors":"Hossein Asakereh,&nbsp;Soma Zandkarimi,&nbsp;Koohzad Raispour","doi":"10.1002/met.70236","DOIUrl":"https://doi.org/10.1002/met.70236","url":null,"abstract":"<p>Precipitation forecasting is critical for water resource management, particularly in arid and semi-arid regions like Iran, where climate change is expected to exacerbate water scarcity. This study uses a spatially adjusted autoregressive integrated moving average (ARIMA) model for projecting annual precipitation over the next three decades. Using data from 113 meteorological stations across Iran, ARIMA models were fitted to precipitation records (1979–2023), with bias adjustment applied via spatial regression incorporating coordination variables (latitude and longitude) and elevation. The adjusted ARIMA models demonstrated high accuracy, with a correlation coefficient (CC) of 0.98 between observed and modeled precipitation and a root mean square error (RMSE) of 35.78 mm, significantly outperforming raw ARIMA outputs (CC = 0.96, RMSE = 53.50 mm). Nevertheless, projections indicate a widespread decline in precipitation across Iran, with reductions of 50%–60% anticipated in the water-critical Zagros and Alborz Mountains and milder decreases (≤ 30%) in central and eastern arid regions. The findings underscore the utility of spatially calibrated ARIMA models for mid-term precipitation forecasting in topographically diverse regions, offering actionable insights for climate adaptation.</p>","PeriodicalId":49825,"journal":{"name":"Meteorological Applications","volume":"33 4","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/met.70236","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148784560","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Recent Climate Change in Benin Since the Mid-20th Century: Spatiotemporal Trends, Regime Shifts, and Extremes in Precipitation and Temperature","authors":"René Bodjrènou,&nbsp;Luc Ollivier Sintondji,&nbsp;Amza Labouda,&nbsp;Patrick Koyo,&nbsp;Alfred Godui,&nbsp;Oyedé Inès,&nbsp;Amou Martial,&nbsp;Françoise Comandan","doi":"10.1002/met.70230","DOIUrl":"https://doi.org/10.1002/met.70230","url":null,"abstract":"<p>This study investigates long-term climate change in Benin over the period 1950–2024, focusing on precipitation and temperature variability, trends, breakpoints, and extreme events. The results reveal that annual precipitation ranges from 1034 mm in Kandi to 1320 mm in Cotonou, with overall declining trends (−0.4 to −4.5 mm/year). A recovery in rainfall is observed in Kandi after 2000 (trend = +2.73 mm/year). Annual Precipitation Concentration Index (PCI) values indicate generally moderate rainfall concentration, except in northern Benin, where irregular temporal distribution is observed (PCI≈20 in Kandi), accompanied by a slight reduction in drought occurrence, from 16.5% during 1950–1999 to 14% during 2000–2024. Extreme precipitation indices reveal increasing trends in the number of consecutive dry days (0.013–0.149 days/year) and decreasing trends in the number of consecutive wet days (−0.007 to −0.039 days/year). Seasonal analyses comparing the 1950–1999 and 2000–2024 periods highlight delays in the onset of the rainy season in central regions (about 10 days in Bohicon and 2 days in Savè), while the length of the rainy season remains relatively stable. In contrast, an extension of the rainy season is observed in Cotonou (+16 days) and Kandi (+7 days), whereas a shortening is detected in Parakou (−13 days) and Natitingou (−17 days). Most precipitation breakpoints were detected during the 1960s and 1970s. Mean annual temperature averages approximately 27.5°C nationwide, with higher maximum temperatures in the north (up to 33.6°C) and higher minimum temperatures in the south (around 24.4°C in Cotonou). Over the past half-century, mean, maximum, and minimum temperatures have increased by approximately 0.88°C, 1.03°C, and 1.04°C, respectively. Breakpoints vary depending on the station and variable. Significant warming trends are observed across all stations, reaching up to +0.029°C/year. Seasonal analyses also indicate a significant increase in heat wave frequency (e.g., from 0.134 to 0.333 days/year). Overall, these findings emphasize the urgent need for adaptation measures to mitigate the growing impacts of climate change on agriculture and water resources in Benin.</p>","PeriodicalId":49825,"journal":{"name":"Meteorological Applications","volume":"33 4","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/met.70230","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148783660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rainfall Onset Timing and Hydroclimatic Variability in a Tropical Catchment: Insights From Observations and Climate Projections","authors":"Albert Elikplim Agbenorhevi,&nbsp;Julian Klaus,&nbsp;Valery Bessely Stanislas Kouassi,&nbsp;Ernest Biney,&nbsp;Nelly Carine Kelome,&nbsp;Leonard Kofitse Amekudzi","doi":"10.1002/met.70231","DOIUrl":"https://doi.org/10.1002/met.70231","url":null,"abstract":"<p>Variability in the seasonal onset of rainfall remains a challenge in tropical regions, yet its spatial and temporal patterns remain poorly characterised at the catchment scale in West Africa. Here, we perform a station-level analysis of rainfall onset, centre of mass dynamics and multi-day accumulation across the Pra River Catchment. We used the Mann–Kendall trend test and linear regression to detect systematic shifts in onset over the observational record, and Kernel density estimation to characterise scale-dependent rainfall distribution over 1-, 5-, 15- and 20-day accumulation windows. The method is evaluated against historical simulations and future projections of the CMIP6 multi-model ensemble under SSP2-4.5 and SSP5-8.5 scenarios. Results show spatial heterogeneity in rainfall onset trends, with statistically significant earlier shifts at several southern, eastern and western stations, while most stations show no significant trend. Furthermore, ENSO shows only a weak probabilistic influence on onset timing, with a significant phase response detected at Mfensi. CMIP6 simulations consistently underestimate rainfall variability and upper-tail extremes relative to observations across all accumulation windows. Future projections indicate that climate risk in the catchment will be driven primarily by the intensification of extreme events rather than shifts in mean rainfall. Our results point toward a dominance of interannual variability over systematic regional change, and a stronger proximal control on Guinea Coast onset dynamics than the remote Pacific teleconnection by the Atlantic equatorial mode. These findings provide a quantitative basis for improving seasonal forecasting, adaptive agricultural planning and long-term water resources governance across tropical hydrologically significant catchments.</p>","PeriodicalId":49825,"journal":{"name":"Meteorological Applications","volume":"33 4","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/met.70231","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148783635","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-Source Data-Driven Downscaling and Bias Correction of ERA5-Land Near-Surface Air Temperature Over Complex Terrain","authors":"Jiawei Huang,&nbsp;Xiaojuan Wu,&nbsp;Yongren Chen,&nbsp;Xiaoxiao Du,&nbsp;Hao Niu","doi":"10.1002/met.70226","DOIUrl":"https://doi.org/10.1002/met.70226","url":null,"abstract":"<p>Near-surface air temperature exhibits local spatial differences in complex terrain, and the native spatial resolution of ERA5-Land near-surface air temperature cannot directly meet application requirements. Taking Sichuan Province as the study area, this study developed a statistical downscaling and bias-correction workflow for daily ERA5-Land near-surface air temperature based on observations from 6045 stations during 2020–2024, ERA5-Land 2 m air temperature, high-resolution land-surface predictors, and ERA5/ERA5-Land Reanalysis-Derived Auxiliary Predictors (RDAPs). The performance of Elastic Net (EN), Bayesian-Optimized Random Forest (BO-RF), and Random-Search-based Long Short-Term Memory network (RS-LSTM) under different combinations of model and predictor set was compared across multiple evaluation dimensions. The results showed that all three models performed better after RDAPs were introduced. BO-RF with RDAPs and RS-LSTM with RDAPs were the two best-performing combinations, with the lowest RMSE values for <i>T</i>_avg, <i>T</i>_max, and <i>T</i>_min on the independent final validation set of 0.94°C, 1.40°C, and 1.12°C, respectively. The two best-performing combinations also showed relatively stable monthly RMSE reductions, with no evident month-to-month fluctuations in accuracy; for all three temperature variables, the proportions of validation stations with RMSE improvement exceeded 97%. Grouped Permutation Feature Importance (GPFI) further indicated that the near-surface humidity and surface thermal state group and the pressure–wind–circulation background group were the RDAP predictor groups with relatively high contributions, whereas the terrain group was the main contributing group among the high-resolution land-surface predictors. However, the error reduction of the best-performing combinations remained limited for sporadic extreme samples.</p>","PeriodicalId":49825,"journal":{"name":"Meteorological Applications","volume":"33 4","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/met.70226","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148753559","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Novel Multi-Model Climate Ensemble Approach to Assess Hydrological Drought Under Climate Change With Reduced Uncertainty","authors":"Saira Akram,&nbsp;Hussnain Abbas,&nbsp;Aamina Batool,&nbsp;Zulfiqar Ali,&nbsp;Miklas Scholz,&nbsp;Amna Nazeer","doi":"10.1002/met.70205","DOIUrl":"https://doi.org/10.1002/met.70205","url":null,"abstract":"<p>Drought is a natural outcome of global warming which brings serious effects in various fields. To avert its consequences and facilitate long-term planning, proper forecasting of droughts is critical. Global climate models (GCMs) are useful tools to forecast hazards related to climate. But GCMs are biased, uncertain, and interdependent. Multi-model ensembles (MMEs) are useful to overcome these deficiencies. However, even the classical ensemble methods like equal model averaging (EMA) and mutual information weighting still have the weaknesses of being sensitive to outliers and less robust in highly variable areas. This study intended to propose a new machine learning-based solution, the K-Nearest-Neighbors Adaptive Weighting Scheme (KNNAWS-Ensemble) that resolves the discrepancies in traditional weighting approaches and enhances the accuracy of the drought forecasting. KNNAWS-Ensemble uses the K-Nearest Neighbors (KNN) algorithm to assign adaptive weights to 18 CMIP6 GCMs that are based on historical performance. As a reference data set, precipitation data of the Tibetan Plateau is utilized from 1961 to 2014. Additionally, K-Component Gaussian Mixture Model (K-CGMM) is used for the statistical distribution fitting and standardization. KNNAWS-Ensemble is validated across conventional ensemble methods using quality measures such as the normalized root mean square error (NRMSE), mean absolute error (MAE), and the correlation coefficient. The findings prove that the KNNAWS-Ensemble is more effective than the conventional ensemble methods. Meanwhile, it has the lowest NRMSE (7.2560) and MAE (1.4127) and largest correlation coefficient (0.9074). Conversely, there are greater values of NRMSE in EMA and the Mutual Information Ensemble (MIE) (8.5516 and 8.4122), and lower correlation values (0.8958 and 0.8960). These results verify high predictive validity and less uncertainty of the proposed scheme. Furthermore, long-term drought projections underscore the elevated risk of severe drought (SD) events. Under SSP1-2.6, the average SD probability increased from 0.533 at the 1-month scale to 0.705 at the 48-month scale. Similarly, under SSP2-4.5, average SD exhibits persistently high probabilities, ranging from 0.668 (1 month) to 0.681 (48 months), while under the high-emission SSP5-8.5 scenario, the average probability SD is 0.406 at the 48-month scale. These findings reveal the higher probabilities of severe drought than other categories under future climate scenarios. The KNNAWS-Ensemble offers an accurate framework on prediction of drought and is useful to address uncertainties and outlier sensitivity in MMEs. Such development serves as a good aid to managing and mitigating droughts especially in case of rising climatic variability.</p>","PeriodicalId":49825,"journal":{"name":"Meteorological Applications","volume":"33 4","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/met.70205","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148753246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Warming-Level-Dependent Restructuring of Precipitation Extremes in Shandong Province, China: Asymmetric Dry-Wet Anomaly Responses and Compound Severity Escalation","authors":"Ali Salem Al-Sakkaf,&nbsp;Jiahua Zhang,&nbsp;Abdulhaleem H. Labban,&nbsp;Adeel Ahmad,&nbsp;Fengmei Yao","doi":"10.1002/met.70227","DOIUrl":"https://doi.org/10.1002/met.70227","url":null,"abstract":"<p>Understanding regional precipitation extreme responses to global warming is critical for climate adaptation planning, yet threshold behaviors and dry-wet anomaly asymmetries remain poorly characterized at provincial scales. This study quantifies drought and wet-anomaly frequency evolution across four warming levels (1.5°C, 2.0°C, 3.0°C, 4.0°C) in Shandong Province, China, over the period 1979–2100 (relative to a 1979–2008 baseline), using high-resolution (0.1°) downscaled CMIP6 data from 19 climate models under multiple emission scenarios (SSP2-4.5 and SSP5-8.5). We apply the Standardized Precipitation Index and the Intensity-Area-Duration framework to characterize compound event severity. Results reveal systematic dry-wet asymmetry: dry-anomaly frequency (SPI-12 ≤ −1.0) declines 43.5%–51.8% from its 6.5% baseline, while wet-anomaly frequency (SPI-12 ≥ +1.0) increases from its 6.4% baseline by 82.3% at 1.5°C to 232.7% at 4.0°C. The historical 0.81:1 dry-wet anomaly ratio collapses to 0.50:1 at 4.0°C, creating unprecedented wet-anomaly-dominated conditions. Notably, the largest single-step increment in wet-anomaly frequency (+4.1 percentage points from 3.0°C to 4.0°C) is approximately double the preceding 2.0 to 3.0°C increment (+2.1 pp), a pattern consistent with published process studies linking high-warming regimes to reorganized convective systems, soil moisture saturation thresholds, and positive wetness-evaporation feedbacks—though these mechanisms are not directly tested and formal statistical tests do not confirm significant nonlinear acceleration at <i>α</i> = 0.05. Spatial analysis reveals remarkable uniformity in wet-anomaly amplification, with increases of +22–37 percentage points in northwestern and central plains. The Intensity-Area-Duration framework demonstrates that compound event severity escalates dramatically, with the severity index reaching 43.9 ± 35.1 at 4.0°C—a + 307% increase from 1.5°C—driven by event durations extending to 38.3 months. Scenario comparison reveals pathway divergence at 2040–2060; limiting warming to approximately 3.0°C avoids progression to the highest warming level examined and substantially constrains wet-anomaly frequency escalation in the latter half of the century. These findings underscore the importance of near-term mitigation and highlight strong warming-level dependence in regional wet-anomaly responses.</p>","PeriodicalId":49825,"journal":{"name":"Meteorological Applications","volume":"33 4","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/met.70227","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148753247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DustCast: An Ensemble Machine Learning Model for Monthly Atmospheric Dust Aerosol Forecasting Across the Arabian Peninsula","authors":"Christopher Ramos,&nbsp;Brandi Gaertner,&nbsp;Steven J. Greybush","doi":"10.1002/met.70215","DOIUrl":"https://doi.org/10.1002/met.70215","url":null,"abstract":"<p>This study presents DustCast, an ensemble machine learning model developed to forecast monthly atmospheric dust concentrations across the Arabian Peninsula. Motivated by the increasing frequency and intensity of dust storms in the region and their associated adverse impacts on health, agriculture, and the environment, the model integrates multiple meteorological and aerosol datasets, including ERA5 reanalysis, MERRA-2 aerosol diagnostics, and the Indian Ocean Dipole index. The methodology employs a heterogeneous parallel ensemble framework that combines four machine learning techniques: multiple linear regression, K-nearest neighbors, decision tree, and random forest, with weights assigned based on each model's performance as evaluated by root mean squared error (RMSE). Spatial aggregation facilitates efficient and precise data binning and analysis. Results indicate that multiple linear regression and random forest exhibit superior predictive capabilities among the individual models on the surface, with the aggregated ensemble prediction achieving an RMSE of 0.00972 μg/m³ and an <i>R</i>² of 0.887, outperforming each base learner. When applied to the atmospheric column, DustCast derives the majority of the predictive contributions from decision tree and random forest, with the ensemble prediction achieving an RMSE of 0.00550 mg/m² and an <i>R</i>² of 0.984. The DustCast ensemble model captures seasonal patterns of dust mobilization across the Arabian Peninsula for both near-surface and atmospheric column dust concentrations. The model performs particularly well during the summer months (June July and August) when the Shamal wind strongly influences sand and dust storms throughout the region. It also captures seasonal dust events associated with frontal systems and dynamic pressure gradients throughout the year.</p>","PeriodicalId":49825,"journal":{"name":"Meteorological Applications","volume":"33 4","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/met.70215","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148752662","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ensemble Weather Forecasting Scenarios Tailored to Users' Needs: Application to Wind Energy","authors":"Flore Roubelat,&nbsp;Arnaud Mounier,&nbsp;Bruno Joly,&nbsp;Laure Raynaud","doi":"10.1002/met.70225","DOIUrl":"https://doi.org/10.1002/met.70225","url":null,"abstract":"<p>Ensemble Prediction Systems (EPS) are commonly used in weather forecasting to account for uncertainties, but their exploitation for operational prediction remains a challenge and often lacks a user-oriented perspective. To address this issue, this article examines how a scenario-based processing of an EPS, which summarizes ensemble information in a few meaningful scenarios, can be adapted to an end-user. The proposed methodology follows a 2-step procedure where ensemble information is first projected in a reduced space with a convolutional autoencoder, and a clustering algorithm is then applied in this reduced space to define climatological patterns that will guide the final ensemble members classification. Our main contribution is to integrate a user impact variable in the dimension reduction step, in order to structure the latent space and subsequent clusters according to the user's need. The approach is illustrated for wind power prediction, with 100 m wind fields as predictors and the capacity factor as the impact variable. Compared to a pure wind-based clustering, leveraging an impact variable provides more coherent scenarios that better represent the correlation between wind and energy production. An application to a case of ramp event shows that our clustering provides a relevant decision support tool. Finally, our design is generic enough to be applied easily to other case studies with different impact variables.</p>","PeriodicalId":49825,"journal":{"name":"Meteorological Applications","volume":"33 4","pages":""},"PeriodicalIF":3.1,"publicationDate":"2026-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/met.70225","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148752661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}