ABSTRACT Information on extreme rainfall is needed for flood risk management and design and assessment of infrastructure. We review techniques used for estimating the characteristics of extreme rainfall events, covering both research and its application in practice. The review is motivated by a desire to update rainfall estimation methods for the UK, but we include examples from many other parts of the world. We cover estimation of rainfall depth‐duration‐frequency relationships, including regionalisation, development of extreme value models, and their incorporation of seasonality, climatic non‐stationarity and the areal extent of rainfall. Other aspects covered include methods for characterising the temporal and spatial variation of rainfall depths for design rainfall events. For each aspect we identify the main categories of method seen in the literature and then review their advantages, disadvantages and prevalence in practice. We highlight opportunities for incorporating meteorological understanding in the development and application of statistical models of extreme rainfall. We find increasing use of spatial statistics, machine learning and climate model outputs to complement traditional rainfall frequency analysis techniques. The design rainfall event concept, while the focus of this paper, is not necessarily capable of answering some more demanding questions about societal resilience. These may require information on the meteorological context of rainfall: its larger‐scale spatial and longer‐term temporal structure.
ABSTRACT Human cloud observations are by design to some extent subjective since they are based on the observer's visual perception of the sky and features therein. In this work, we assess the uncertainty of operational SYNOP cloud observations in three different settings: (i) seven observers at the same station but at different times, (ii) observations conducted at the same time but at different stations about 20 km apart, and (iii) five volunteer meteorologists classifying clouds from the same set of ground‐based camera images. Investigations indicate the presence of substantial systematic differences for different observers and also at different stations, which cannot be explained purely by meteorological circumstances. When evaluated on a complex cloud classification scheme, where for each instance up to three out of 30 cloud types have to be chosen, inter‐observer agreement on image‐based classifications reached MCC scores up to 0.33. However, combining observations from different participants to a single report and using a classification scheme of reduced complexity strongly increases MCC scores up to 0.67. Results suggest that human cloud observations have to be used with caution if they are used for example, to evaluate new automatized cloud classification methods.
ABSTRACT 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.
ABSTRACT 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.
ABSTRACT 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 Tavg, Tmax, and Tmin 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.
ABSTRACT 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 α = 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.
ABSTRACT 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.
ABSTRACT 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 R2 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 R2 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.
ABSTRACT The forecasting of clear‐air wind shear presents a significant challenge in aviation meteorology. To improve the accuracy of wind shear identification and forecasting using coherent Doppler wind lidar, this study proposes an advanced method for identifying and forecasting wind shear and shear line. The method first performs point‐by‐point sliding traversal of potential wind shear along the radial and tangential direction based on retrieved two‐dimensional wind vector. Subsequently, shear points are identified and extracted by the wind vector, and spatial clustering analysis is applied to reconstruct low‐level wind shear line. Finally, the movement of wind shear line is forecasted with the average wind vector in the guidance region of shear line segments. This study simulated two simplified wind shear scenarios that retain key wind field features: gust front passage and divergent airflow. It compared the identification and forecasting performance of wind shear and shear line based on radial speed and two‐dimensional wind vector. Results demonstrated that under controlled kinematic conditions, the two‐dimensional wind vector method provides more accurate identification and improved forecast precision. The proposed method effectively identified and forecasted shear lines induced by gust front and severe convection in the airport field experiments. This study provides technical support for airport low‐level wind shear early warning and flight safety assurance.
ABSTRACT This study provides a comprehensive evaluation for the prediction of wind power ramping events in the Belgian Offshore Zone. These rapid, large‐scale power fluctuations pose significant challenges to grid reliability. The research uses operational Numerical Weather Prediction (NWP) models from the Royal Meteorological Institute of Belgium, as well as its version enhanced with Wind Farm Parameterization (WFP). Power predictions are generated with both typical power curves and machine learning approaches. Standard verification metrics, such as Mean Absolute Error (MAE), often fail to capture the operational significance of ramp events. To address this, we develop a flexible verification framework designed to assess ramp forecast performance. This framework incorporates adjustable time and power buffers, which tolerate minor, operationally acceptable discrepancies in the timing and magnitude of predicted events. Application of this framework to both intraday and day‐ahead forecasts reveals that WFP‐enhanced models consistently improve ramp predictions over the operational baseline. Further analysis reveals that while the WFP model with power curves effectively reduced false alarms, it comes at the cost of more misses. In contrast, ML‐based approaches achieve slightly higher overall skill scores by striking a better balance between reducing these error types. Moreover, we introduce the Ramp Alignment Score (RAS), an event‐based metric that quantifies the temporal alignment between predicted and observed ramps, to supplement the model evaluation by lead time. RAS analysis demonstrates that WFP models achieve better temporal alignment and reveals a distinct diurnal cycle in ramping prediction errors. Finally, we investigate the impact of a specific meteorological driver, finding an association between severe precipitation and large, highly predictable ramp events. Conversely, moderate and light precipitation are linked to a higher incidence of missed events and false alarms. This work provides both an operationally relevant evaluation methodology and insights into ramp predictions under specific meteorological conditions.
ABSTRACT In the context of global warming, ensuring the comparability and traceability of air temperature measurements is essential for accurately characterising climate change. However, quantifying the environmental effects under real measurement conditions is still a key challenge for the meteorological community. This study addresses this issue by improving the understanding of air temperature observation uncertainties from ground‐based stations. Data from six identical thermometers at 1.60 m and 2.10 m, protected with the same ventilated solar shield and connected to the same datalogger, were analysed during summer and winter. Air temperature difference (ΔT) was calculated as the difference between the readings of each thermometer and the mean, while wind speed and relative humidity were considered quantities of influence. The analysis shows that ΔT depends on season, sensor height, wind speed and relative humidity. During winter nights, ΔT reflects the influence of fog and condensation, intensified by low wind speed. ΔT also exhibits a clear diurnal cycle, indicating the presence of a solar heating effect. The uncertainty of ΔT was then calculated across different ranges of air temperature, wind speed, and relative humidity, obtaining values between 0.03°C and 0.12°C for both thermometer heights. For instance, during a summer daytime case, the air temperature at 1.60 m reached 32.25°C with an uncertainty due to wind speed and relative humidity equal to 0.06°C. Overall, the results show that the uncertainty decreases as wind speed increases, independently of seasonal conditions and thermometer height. This work provides uncertainty values derived from real environmental conditions that can be incorporated in the environmental effects component of the overall measurement uncertainty, in line with the World Meteorological Organisation requirements. These results also support the evaluation of uncertainty budgets for climate reference stations, providing uncertainty values applicable to other stations with similar measurement setups.
ABSTRACT In cold regions across the globe, snowpack dynamics play a fundamental role in regulating hydrological regimes, influencing water resource availability, and triggering natural hazards such as floods, landslides, and debris flows. As climate warming accelerates, particularly in high‐latitude and high‐altitude areas, the need for accurate snow accumulation and snowmelt estimation has become increasingly critical for both environmental management and disaster risk reduction. To overcome the limitations of traditional constant‐density assumptions, this study presents an improved snow accumulation and snowmelt estimation approach incorporating temporal snow density variability and compaction effects. Subsequently, a comparative assessment of six widely used snow density estimation models is conducted based on the observed snow depth and density measurements. Results demonstrate that snow density estimation requires partitioning the entire snow cover duration into two distinct phases, that is, snowpack season and snowmelt season, and accurate snow density estimation must separately account for compaction effects induced by snow depth and duration of snow cover during the snowpack season and snowmelt season. That is, during the snowpack season, the increases in snow density are mainly caused by the gravitational compaction of overlying fresh snow layers, while during the snowmelt season, the snow density enhancement is predominantly governed by the duration of snow cover. Afterward, based on the proposed snow accumulation and snowmelt estimation approach, the impacts of temperature rise induced by climate change on snowpack dynamics were systematically investigated. Results indicate that the rising temperatures contribute to a reduction in snow depth during the snowpack season, primarily in the initial unstable phase of snowpack formation. Conversely, during the snowmelt season, temperature increases exert a more pronounced influence on snowmelt dynamics. Analysis of meteorological data from Nakayama Pass, Hokkaido, Japan, indicates that for every 1.0°C increase in temperature, the duration of snow cover at this location will be reduced by approximately 14.5 days, which may have implications for understanding snowpack responses to climate warming in other snow‐dominated regions. The findings of this study provide critical assessment benchmarks and modeling support for evaluating snowmelt‐induced disasters, including but not limited to floods, landslides, and debris flows.
ABSTRACT Verification for weather prediction is highly developed and has guided, motivated, and documented improvements in recent decades. In comparison, verification is far less developed for climate projections. Verification is an important part of the scientific process for testing and identifying ways to improve our understanding, and for establishing credibility and trust. In contrast to weather forecasts, climate projections are not initialized with an estimate of the state of the atmosphere or climate system as an initial condition; instead, the goal is to solve a boundary condition problem of estimating the climate state given a set of greenhouse gas and aerosol emissions and other forcings that are external to the model components. Because the primary use of climate projections is around future changes, which are often cast as anomalies relative to a baseline state, verification should include or even focus on the responses of the climate to external forcing, rather than just evaluation of the baseline climatology. Verification of climate projections is a challenge because (1) the relevant lead times are long, resulting in very small sample sizes, and (2) climate projections are conditional on the forcing pathway, which determines the boundary conditions, so they depend on additional information about the future external to the projection itself. Some existing activities contain elements of verification, such as the evaluation of climatological characteristics and the comparison of simulations of the past historical forcing with observed trends. Three concepts from weather forecast verification that present opportunities for advancing verification of climate projections are discussed: representativeness, forecast skill, and system design for verifiability.
ABSTRACT We used a new IREQ‐type (Required Insulation) model to estimate the heat excess caused by weather in the warm season of the temperate, lowland, and continental climates. The model we developed estimates heat excess by calculating both the “compensatory sweating”, λEswcomp, and the “clothing thermal resistance”, rcl,t. The model is very simple and can be used to estimate thermal load under any weather condition. It can be applied to individuals and all activity types (lying, standing, walking, running). Longitudinal data collection method was used, with 240 observations conducted for the period 2023–2025 in the Hungarian lowland, measuring sweating (referred to as “regulatory sweating”, λEsw) and comparing it to “compensatory sweating”. The λEswcomp and rcl,t values were also compared to the estimated UTCI (Universal Thermal Climate Index) values. The main results are as follows: (1) λEswcomp values registered when running changed between 80 and 650 W m−2 in the warm season in the Hungarian lowland, (2) observations revealed that λEsw < λEswcomp in general, (3) during extreme loads, the upper limit of physically interpretable rcl,t values is around 4 clo‐t. These high rcl,t values occur when λEswcomp is high (above 500 W m−2), the λEsw is low (around 100 W m−2) and (4) the model is very sensitive to changes in irradiance and regulatory sweating values in situations with high heat excess. One of our most important future tasks is to parameterize sweating as simply as possible, which would have acceptable accuracy for as many people as possible.
ABSTRACT Multi‐day periods featuring low wind and solar generation (‘Dunkelflaute’, DF) are an increasing concern for European electricity systems. Understanding DF and how they may change is critical for assessing the risk of electricity supply shortfalls in power systems containing high levels of variable renewable generation. This study assesses the suitability of data from two versions of the EU's flagship Copernicus climate service (C3S‐Energy and ECEM) for characterising DF events in the present‐day, and uses these datasets to develop plausible scenarios of change under 2°C global warming. After controlling for issues of dataset quality, a broadly consistent picture of DF behaviour over the historic period (1980–2010) emerges. As expected, DF predominantly occurs in winter and responds coherently to the North Atlantic Oscillation (NAO) as the dominant large‐scale mode of regional atmospheric variability. The continental‐scale patterns of behaviour seen in the observationally based components of the datasets are well‐replicated in the corresponding climate model‐based data, but individual simulations can differ substantially at the scale of individual countries (these differences are found both within and between each dataset). The multi‐model mean response to a 2°C global warming scenario in both ECEM and C3S‐Energy suggests an increase in DF events ~5%–25% over much of the European domain (particularly the north and west). However, individual model responses exhibit very different patterns with one model suggesting a widespread ~5%–25% decrease in DF (i.e., a change of similar magnitude but opposing direction). Five distinct storylines of a 2°C global warming scenario are therefore proposed, providing a compromise between representing diversity of the individual responses while retaining a tractable set of outcomes. The ability to robustly project future DF behaviour (and future renewable energy climate more broadly) is severely limited by the small sample of climate projections typically available. Future analysis should therefore seek to consider a more extensive and comprehensive ensemble of climate simulations to develop greater confidence and understanding.
ABSTRACT This study first applied the Petrel‐II unmanned aerial vehicle in China to analyze Northeast Cold Vortex weather processes, revealing land‐sea convective differences in coastal Liaoning during June 18–19, 2025. Observations from an S‐band weather radar indicate that convective cells with intensities reaching 50 dBz formed and gradually organized over land, while convective activity over the sea was weak. Vertical cloud radar measurements from the unmanned aerial vehicle further reveal that the circulation structure, with upward airflow of 2.52 m/s at 3–5 km altitude and downward airflow of −1.7 m/s near the surface, facilitated precipitation. In contrast, the atmospheric layer over the sea displayed consistent downward airflow, with −3.5 m/s at upper levels and −2.9 m/s near the surface, which suppressed convective development. Comparative analysis of environmental parameters shows that the convective available potential energy over land (768.58 J/kg) was significantly higher than that over the sea (30.70 J/kg), while the convective inhibition energy over land (−29.13 J/kg) was considerably greater than that over the sea (−182.30 J/kg). This indicates a higher degree of atmospheric instability over land, while stronger lifting forces are required to trigger convection over the sea. To further elucidate the mechanisms underlying land‐sea differences, from a dynamical perspective, divergence analysis at 925 hPa showed pronounced convergence over coastal land (−7.63 × 10−5 s−1) but divergence over the sea (6.65 × 10−5 s−1). From a thermodynamic perspective, by examining the overlay of 500 and 1000 hPa temperatures, it was found that a steeper vertical lapse rate develops over land. In summary, the land area exhibited more favorable dynamic and thermodynamic conditions for convective development, which served as the key factor driving the observed land‐sea differences during this event. This study aims to provide a basis for improving the accuracy of related weather forecasts and warnings.
ABSTRACT The study focuses on rainfall variability within Ghana's agroecological zones, employing varied techniques including spatial autocorrelation, GARCH family volatility models, vector autoregression, impulse response functions, and extreme rainfall analysis. The mean result of Moran's I (−0.317) indicates negative spatial autocorrelation. This means that extremes in rainfall are highly localised because only 1.04% of considered periods showed significant clustering. Volatility modelling reveals zones as separate regimes: highly persistent shocks in the Sudan and Guinea Savanna zones (β1 ≈0.97), asymmetric drought sensitivity in the Deciduous Forest under the E‐GARCH (α1 ≈ −0.206), and event‐driven volatility in the Transitional Zone. Weak correlations between the savannas and the forest zones are found within the VAR analysis, which suggests correlations lagged in time within defined zones. The forest zone leads rainfall over the Guinea and Sudan Savannas 2–3 days later. IRF results indicate systematic moisture‐carrying flows operating at 1–2‐day intervals, with sustained northward cross‐border zonal responses lasting over 20 days. Criteria for extreme rainfall show a strong north–south gradient, with the transitional zone containing the most intense events (95th percentile = 25.3 mm; 99th = 53.5 mm) and the deciduous forest enduring persistent droughts. The results of this study exemplify the fragmentation of rainfall regimes and their ecological volatility with a specific focus on the negative impacts of unadopted volatility. This study serves as a clarion call for zonal‐specific measures to safeguard agriculture, water, climate adaptability and resilience.
ABSTRACT Urban thermal environments have deteriorated due to global warming and urbanization in recent years. This is expected to become prominent in the near future, even in cities classified under the subarctic climate zone like Sapporo. This study aims to evaluate future thermal environments under progressing urban reconstruction and global warming through thermal environment surveys and numerical modeling in central Sapporo. First, air temperature, humidity, and globe temperature were measured in central Sapporo on two hot and clear days and a hot and partial cloudy day. Field observations were sequentially conducted at four locations in a high‐rise building district, a low‐rise residential area, a park, and a botanical garden, and compared with those at a point fixed during the day. The results showed a clear spatial variation in temperature and humidity within the city center. Moreover, a numerical simulation well reproduced this variation among the observation sites. Another hypothetical simulation was performed with a uniform temperature increase of 2 K on one of the observation dates and full redevelopment to be completed in 2030s. The simulation revealed not only a uniform temperature rising but mitigation of hot environment in areas shaded by newly constructed buildings. Downstream turbulence was found to be reinforced behind the new high‐rise buildings. The wet‐bulb globe temperature assessment suggested that the urban thermal environment in central Sapporo would deteriorate due to global warming and would change the spatial distribution by redevelopment.
ABSTRACT Spurred by technological advances, major operational centers and scientific research institutions have developed convection‐allowing ensemble prediction technologies and systems. To meet the demand for accurate localized severe weather forecasting, the convection‐allowing ensemble prediction system of North China (CMA‐BJ‐EN), developed by the Institute of Urban Meteorology, China Meteorological Administration (CMA), became officially operational in January 2023. This paper provides a comprehensive introduction to the system's research and development background and key technologies. The CMA‐BJ‐EN system has a convection permitting grid spacing of 3 km, covering 21 members, and can provide hourly probability prediction results of 48 h forecasts over North China. The system is coupled with technologies such as ensemble of data assimilations, Stochastically Perturbed Parameter Tendencies (SPPT), and has complete operational configuration. The qualitative and quantitative evaluation of the CMA‐BJ‐EN system and its comparison with the NCEP global ensemble forecast system show that the system can effectively forecast several major weather events in North China and can obtain local refined probability forecast results relative to global ensemble. The statistical results also show that compared with the NCEP global ensemble forecast, the CMA‐BJ‐EN system can effectively reduce the RMSE of 2 m temperature and 10 m wind speed and can improve the AROC score of precipitation forecast by 5%–10% for heavy rainfall. The operation of the system can provide users with effective refined probability forecast reference.
ABSTRACT With the growing use of kilometer‐scale atmospheric models in climate‐scale applications, remapping data from high‐resolution grids to coarser grids and performing statistical diagnostics have become increasingly routine. Precipitation data often exhibit strong spatial and temporal discontinuities, making the choice of remapping algorithm critical for maintaining certain properties after remapping. The choice among different remapping methods, in particular between conservative and nonconservative schemes, is closely tied to which aspect of precipitation one wishes to preserve most faithfully. This study first highlights the substantial differences between conservative and nonconservative remapping methods/tools in fine‐to‐coarse remapping. Because these methods are inherently designed to prioritize either area‐mean values or pointwise‐like values, they produce markedly different error characteristics in metrics such as mean precipitation amount, frequency, intensity, diurnal peak timing, and the probability distribution function (PDF) of precipitation frequency. These differences become progressively more pronounced as the target resolution decreases. Conservative remapping tends to increase the proportion of weak rainfall events, thereby substantially altering the PDF of the frequency–intensity relationship. Because of its strong smoothing effect, it may lead to apparently improved RMSE/PCC values, as compared with nonconservative remapping. Inverse‐distance‐weighted remapping, as a representative pointwise‐type nonconservative method, is not flux conserving but more effectively preserves location‐specific rainfall characteristics. This makes error metrics and frequency–intensity spectra less sensitive to spatial coarsening. It is recommended that kilometer‐scale model evaluations should explicitly account for the distinct characteristics of both remapping categories, in alignment with the specific research objectives.
