Abstract In this paper, we propose a new capability for an oceanic hazards outlook containing delineations of where winds and waves are expected to have the potential of posing a hazard to either life or property at sea. This new product is based on NOAA’s Global Ensemble Forecast System, focused on developing a probabilistic forecast of 10-m wind speed (U10) and significant wave height (Hs) for week 2. The methodology involves defining temporal and spatial windows to postprocess the ensemble members and analyzing the tail of the probability density function generated by the pooled dataset. Specific thresholds for U10 and Hs as well as probability levels are defined in the algorithm to generate probability maps. The performance is assessed through statistical validation using NDBC buoy data and reanalysis, followed by case studies. Results indicate that the probabilistic forecast is skillful in predicting waves up to 9 m and winds up to 48 kt (1 kt ≈ 0.51 m s−1), particularly for extratropical systems. In addition, with weaker signals and lower probabilities forecasted, extreme extratropical events associated with hurricane-force winds could be in many cases successfully detected in the week 2 forecast. Therefore, the probability maps provide valuable guidance for practical applications, helping in decision-making for maritime operations.

Abstract As the deployment of solar power utilizing devices increases, accurate forecasts of solar energy become increasingly important for managing and stabilizing the electricity grid. This study investigates the validity of 7-day ahead global horizontal irradiance (GHI) forecasts for South Africa provided by the Global Forecast System (GFS). The clear-sky index was used to interpolate GFS mixed 3-hourly and 6-hourly averages to generate hourly forecasts. The model’s performance was validated against radiometric GHI data from 15 locations spread across South Africa’s six climatic zones. The results showed that the GFS model can satisfactorily forecast GHI in South Africa. Overall, the model performed best in interior climatic zones and relatively poorly in coastal climatic zones. The accuracy of the GFS model performance decreases as the forecast lead time increases. The model overestimates GHI in overcast conditions while it underestimates GHI in clear and partly cloudy conditions. On average, the model underestimates GHI in all seasons. Forecasting errors are highest in summer and lowest in winter and moderate in autumn and spring. The inadequacy of the GFS model to precisely determine the cloud quantities is the major source of error; errors increase as the cloud fraction increases. Despite this, the study’s findings suggest that GFS forecasts may be used for forecasting solar photovoltaic output for energy planning in South Africa. Significance Statement This paper provides the first comprehensive assessment of Global Forecast System (GFS) forecasts of global horizontal irradiance (GHI) in six climatic regions of South Africa, exploring how cloud cover, site location, climatic zone, season, forecast horizon, and hour of the day impact the model’s performance. This research was carried out in a country where GFS GHI models have never been validated before and that presents a wide range of climatic conditions. The results allow the accuracy of other global and mesoscale GHI forecasting models to be compared against the current results in the region. The outcomes of this study indicate the possibility of using GFS GHI forecasts for forecasting solar photovoltaic output for energy planning. South Africa has an electricity supply deficit, and being able to accurately forecast the output of the increasing capacity of solar PV is one of the things needed to stabilize the grid.

Abstract Global Navigation Satellite System (GNSS) radio occultation (RO) profiles provide atmospheric temperature and water vapor information, complementing microwave and infrared satellite radiances by enhancing coverage in data-sparse oceanic regions due to their high vertical resolution and all-sky observability. This study uses data denial experiments to evaluate the impacts of assimilating commercial RO profiles from Spire Global, Inc. (Spire) on Hurricane Weather Research and Forecasting Model (HWRF) analyses and forecasts of four 2022 Atlantic hurricanes. HWRF currently assimilates RO data from government-funded missions such as the Constellation Observing System for Meteorology, Ionosphere, and Climate-2 (COSMIC-2). Compared to COSMIC-2 observations, Spire data exhibit similar retention statistics under HWRF’s quality control screens and similar diagnosed observation uncertainties, except around the extratropical boundary layer where Spire’s data retention is improved, and its observation errors are smaller. Compared to a Control experiment that withholds COSMIC-2 and Spire data, a C2 experiment assimilating COSMIC-2 observations yields a ∼0%–10% reduction in mean absolute minimum central sea level pressure (PMIN) intensity errors for most forecast times. A C2Spire experiment assimilating Spire RO data on top of COSMIC-2 generates mean PMIN absolute errors similar to C2 except for the t = 24–36-h period when they exhibited a statistically significant 10%–15% improvement over Control. The 3–6-hPa mean PMIN overintensification bias of the Control showed a statistically significant 17%–38% reduction in t = 66–108-h C2Spire forecasts. Additionally, short-range C2Spire forecast temperature and specific humidity root-mean-square deviations (RMSDs) against dropsondes below 700 hPa are ∼5%–10% smaller than those of Control; these differences are statistically significant for some layers.

Abstract To enhance the utilization and assimilation of the observations from the Geostationary Interferometric Infrared Sounder (GIIRS) into the China Meteorological Administration (CMA) Global Forecast System (CMA-GFS), a novel three-dimensional cloud detection (3DCD) method for GIIRS data has been recently developed. This study is an extension of the 3DCD method. In the original 3DCD (3DCD-OLD) method, the ideal atmosphere profile and observational error are used to construct a look-up table. In this study, the actual profiles and stricter threshold are used to improve the 3DCD method (3DCD-NEW). We aim to investigate the benefits of using the 3DCD-NEW and compare its impact on the analysis and forecasting through one-month cycling and forecast experiments. Preliminary assessment results indicate that the number of assimilated GIIRS observations using the 3DCD-NEW is approximately double that of using completely clear-sky detection. The analysis fields of geopotential height, temperature, specific humidity, zonal wind, and meridional wind are neutral for the 3DCD-NEW when compared to the 3DCD-OLD. A neutral to positive impact is observed in the forecast fields of geopotential height, temperature, and vector winds over the Northern Hemisphere, Tropics, and East Asia during the initial 7-day forecast period when compared to 3DCD-OLD.

Abstract Lake-effect snowstorms can produce large snowfall accumulations that are challenging to simulate and forecast. One source of forecast uncertainty for these events is the uncertain parameterization of sub-grid processes, such as planetary boundary layer and surface layer turbulence (PBL/SL) and cloud and precipitation microphysics (MP), in numerical weather prediction models. One way to quantify this uncertainty is to design ensembles that use stochastic parameter perturbations (SPP) to vary individual uncertain parameters within physics schemes. This research aims to evaluate and improve the utility of SPP for convection-permitting ensemble forecasts of lake-effect snow, with a focus on PBL/SL and MP parameterizations. We focus on a snowfall event observed during the Ontario Winter Lake-effect Systems (OWLeS) field campaign, which is simulated with 1-km horizontal grid spacing using the Weather Research and Forecasting model. A suite of 20-member ensemble simulations are run, including ensembles where SPP is applied only to PBL/SL or MP, where SPP applied to multiple schemes concurrently, where perturbations to initial and boundary conditions (ICs/BCs) are applied instead of SPP, and where SPP and IC/BC perturbations are applied together. SPP perturbations produce substantial spread in simulated precipitation, despite having only modest impacts on the synoptic-scale flow. They accomplish this by modulating lake-atmosphere fluxes, boundary layer characteristics, precipitation growth processes, and hydrometeor terminal fall speeds. The spread and skill of simulated precipitation from an ensemble using SPP alone is comparable to that from ensemble that uses IC/BC perturbations alone. The physical pathways whereby SPP perturbations generate spread are examined and discussed.

Abstract California is vulnerable to large, rapidly growing wildfires that can threaten human lives and damage physical infrastructure. This study analyzes the daily growth of all satellite-observed California fires from 2003 through 2020 and relates that growth to local meteorological and environmental conditions. Fire growth is defined using both absolute and relative metrics (absolute growth—the daily burned area; relative growth—the daily percent increase in burned area). Near-surface environmental conditions are evaluated before, during, and after periods of fire growth for individual fires of varying sizes and spread rates across events with both high and low relative growth. Regional spatial patterns and distributions of near-surface dryness and winds during large absolute-growth events are examined and compared to climatological conditions. Overall, large absolute-growth events generally occur when dead fuels are driest, trailing short periods of increased atmospheric dryness. For large absolute-growth events, high relative growth is usually associated with strong winds, while low relative growth is generally associated with quiescent climatological winds. Winds during high-relative-growth events are greatest in autumn, surpassing the 95th percentile on average. For these autumnal events, atmospheric dryness is less than during other seasons, but atmospheric and fuel dryness is still sufficient for fire. Irrespective of the season, winds during particularly rapid-growth events (high absolute and relative growth) are frequently downslope and influenced by local terrain regardless of the climatological conditions. Significance Statement Wildfires that undergo rapid growth are particularly dangerous to human life and infrastructure. This research finds that, given sufficient dryness, wind is the most important environmental factor in differentiating fire growth of varying sizes and spread rates and for enabling explosive wildfire growth.

Abstract Numerical model simulations are sensitive to the choice of physical parameterizations, particularly those related to convection processes, especially over complex terrains like the western Himalayan region. This study evaluates the sensitivity of the Weather Research and Forecasting (WRF) Model to three convective parameterization schemes [Betts–Miller–Janjić (BMJ), Kain–Fritsch (KF), and Grell–Freitas (GF)] and their multiphysics ensemble mean (ENSM) through long-term (2001–16) winter (December–March) simulations. Model outputs are validated against multisource datasets [India Meteorological Department (IMD), IMERG, ERA5, and Indian Monsoon Data Assimilation and Analysis (IMDAA)], with extreme precipitation events (EPEs) identified as those exceeding the 95th-percentile threshold. Climatology analysis reveals systematic precipitation biases across all single schemes, significantly reduced in the ENSM. Improvements in skill scores observed for ENSM highlight the reliability of the ensemble approach as an alternative to account for model errors. Underestimations of low-level clouds and hydrometeors improve in ENSM. BMJ and KF (GF) schemes exhibit overestimated (underestimated) magnitudes and location-wise discrepancies for baroclinic instability and moisture transport. These issues are mitigated in ENSM, resulting in better alignment with reference datasets. Additionally, ENSM and KF’s storm tracks align well with reanalyses, while individual schemes show a minor northward placement. For EPEs, ENSM highlights Jammu and Kashmir, Himachal, and Uttarakhand as zones of maximum heavy precipitation and frequency. ENSM realistically represents the western disturbance (WD)-induced circulation characteristics, southward-displaced subtropical jet, shifts in WD-associated vorticity maxima, and potential vorticity intrusions during extremes. Consistent with reference datasets, ENSM demonstrates the crucial role of vertical advection and dynamical controls in moisture distribution during EPEs. Our analysis indicates that a suitable WRF cumulus configuration includes either the multiphysics ENSM or the KF scheme, while the GF scheme yields the poorest results. The findings underscore the importance of evaluating both precipitation characteristics and underlying physical mechanisms to identify sources of systematic biases in model simulations.

Abstract Reanalysis proximity vertical profile attributes associated with long-track tornadoes [LTTs; pathlength ≥ 48 km (30 mi)] and short-track tornadoes [STTs; pathlengths < 48 km (30 mi)] for a total of 48 212 tornadoes with pathlengths ≥ 0.016 km (0.01 mi) from 1979 to 2022 in the United States were examined. Both longer- and shorter-track tornadoes were associated with vast ranges of mixed-layer convective available potential energy, together with relatively low mixed-layer lifted condensation level heights and minimal convective inhibition. A large range of 500–9000-m wind speeds and bulk wind differences, 500–3000-m streamwise vorticities, storm-relative helicities, and storm-relative wind speeds were found for STTs. In stark contrast, LTTs only occurred when these kinematic attributes were larger in amplitude through the troposphere, supporting previously documented associations between observed longer-track tornado pathlengths and faster-propagating parent storms. A novel parameter, heavily weighted by kinematic parameters and lightly weighted by thermodynamic parameters, outperformed the significant tornado parameter in differentiating environments that were more supportive of both LTTs and tornadoes rated <EF5. The high correlation values R2 = 0.79 between tornado pathlength and Bunkers’ approximate tornado duration (pathlength/VBunkers) call for improved understanding of mesocyclone periodicities, which impact tornado longevity, to improve tornado pathlength diagnoses and forecasts. Pragmatically, diagnosing LTT environments using vertical profile attributes, perhaps, is not so much a problem of determining when there might be higher expectations for LTTs, but rather a problem of when there might be lower expectations for LTTs, e.g., weaker kinematic attributes in the lower troposphere. Significance Statement The majority of tornadoes have pathlengths less than a few kilometers. As tornado pathlengths increase, their probability of causing impacts to society also increases. We study >40 years of modeled atmospheric vertical profiles to better understand the environmental conditions that support long-track tornadoes (pathlength ≥ 48 km or ≥30 mi). Consistent with previous studies, long-track tornadoes occurred with substantially stronger vertical wind shear profiles and low-level winds compared to short-track tornadoes; however, most tornadoes did not form in environments with exceedingly large vertical instability, regardless of pathlength or intensity. A proposed composite long-track tornado parameter (LTTP) provided better discrimination between longer and shorter pathlength events compared to preexisting parameters.

Abstract Though discrete supercells are usually emphasized in severe weather forecasting, hazard production is often preceded by their interaction with external features. Past studies have examined the impacts of cell mergers, boundaries, other supercells, convective systems, etc. but usually in isolation. Here, we investigate 230 significant tornadoes, 246 significant hail events, and 191 null cases across the United States using WSR-88D data. We find that in over 90% of cases, supercells that produced significant hazards were accompanied by external features. These features varied between hazards; for example, hailstorms were more frequently near boundaries than tornadic storms. That said, the positions of these features with respect to the storm (and storm-relative inflow) distinguished between hazard potential and type. For example, tornadic storms were predominantly on the more unstable side of a boundary, while nontornadic storms and hailstorms were on the less unstable side. Similarly, tornadic storms had more cells in their rear flanks than forward flanks, while hailstorms had more cells in their forward flanks than rear flanks. Although these conditions were observed regardless of the background environment, they were affected by certain variables in the vertical profile, especially in tornadic cases. Namely, when storm-relative inflow was stronger and lifting condensation level (LCL) was lower, tornadic storms were accompanied by more rear-flank cells that were closer to the storm, more directly opposite the storm-relative inflow, for a longer period of time. We propose that these interactions likely modulate hazard potential, in ways that are not accounted for in traditional environmental parameter-based forecasting.

Abstract Synoptic-scale vortices known as monsoon low pressure systems (LPSs) frequently produce intense precipitation and hydrological disasters in South Asia, so accurately forecasting LPS genesis is crucial for improving disaster preparedness and response. However, the accuracy of LPS genesis forecasts by numerical weather prediction models has remained unknown. Here, we evaluate the performance of two global ensemble models—the U.S. Global Ensemble Forecast System (GEFS) and the Ensemble Prediction System of the European Centre for Medium-Range Weather Forecasts (ECMWF)—in predicting LPS genesis during the years 2021–22. The GEFS successfully predicted about half the observed LPS genesis events 1–2 days in advance; the ECMWF model captured an additional 10% of observed genesis events. Both models had a false alarm ratio (FAR) of around 50% for 1–2-day lead times. In both ensembles, the control run typically exhibited a higher probability of detection (POD) of observed events and a lower FAR compared to the perturbed ensemble members. However, a consensus forecast, in which genesis is predicted when at least 20% of ensemble members forecast LPS formation, had POD values surpassing those of the control run for all lead times. Moreover, probabilistic predictions of genesis over the Bay of Bengal, where most LPSs form, were skillful, with the fraction of ensemble members predicting LPS formation over a 5-day lead time approximating the observed frequency of genesis, without any adjustment or bias correction.