Numerical Weather Prediction Models Research Articles

Relationships between Terrain Features and Forecasting Errors of Surface Wind Speeds in a Mesoscale Numerical Weather Prediction Model

Relationships between Terrain Features and Forecasting Errors of Surface Wind Speeds in a Mesoscale Numerical Weather Prediction Model

Improving forecasts of precipitation extremes over northern and central Italy using machine learning

AbstractThe accurate prediction of intense precipitation events is one of the main objectives of operational weather services. This task is even more relevant nowadays, with the rapid progression of global warming which intensifies these events. Numerical weather prediction models have improved continuously over time, providing uncertainty estimation with dynamical ensembles. However, direct precipitation forecasting is still challenging. Greater availability of machine‐learning tools paves the way to a hybrid forecasting approach, with the optimal combination of physical models, event statistics, and user‐oriented postprocessing. Here we describe a specific chain, based on a random‐forest (RF) pipeline, specialised in recognising favourable synoptic conditions leading to precipitation extremes and subsequently classifying extremes into predefined types. The application focuses on northern and central Italy, taken as a testbed region, but is seamlessly extensible to other regions and time‐scales. The system is called MaLCoX (Machine Learning model predicting Conditions for eXtreme precipitation) and is running daily at the Italian regional weather service of ARPAE Emilia‐Romagna. MalCoX has been trained with the ARCIS gridded high‐resolution precipitation dataset as the target truth, using the last 20 years of the European Centre for Medium‐Range Weather Forecasts (ECMWF) reforecast dataset as input predictors. We show that, with a long enough training period, the optimal blend of larger‐scale information with direct model output improves the probabilistic forecast accuracy of extremes in the medium range. In addition, with specific methods, we provide a useful diagnostic to convey to forecasters the underlying physical storyline which makes a meteorological event extreme.

PlanetiQ Radio Occultation: Preliminary Comparative Analysis of Neutral Profiles vs. COSMIC and NWP Models

Radio Occultation (RO) is pivotal for profiling the neutral and ionized atmosphere, with the PlanetiQ mission, via its GNOMES satellites, striving to establish an advanced atmospheric observing system. However, an assessment of the spatiotemporal distributions of PlanetiQ observations and comparisons with reliable datasets are lacking. This study addresses this gap by examining the temporal and spatial distribution of RO observations from PlanetiQ during its initial 198 operational days in 2023, alongside comparisons with COSMIC and Numerical Weather Prediction (NWP) models. Data from GN02, GN03, and GN04 satellites, yielding 1099, 1313, and 1843 RO events per day, respectively, were analyzed. The satellite constellation’s observations demonstrate a generally well-distributed pattern, albeit minor deficiencies in equatorial and polar regions. Single-profile comparisons with COSMIC data reveal strong correlations for pressure, temperature, Water Vapor Pressure (WVP), and refractivity profiles, with temperature exhibiting larger variations (RMSE = 1.24 °C). Statistical analyses confirm statistically insignificant differences between the PlanetiQ and COSMIC profiles at the same spatio-temporal coordinates. Comparisons with NWP models show slight differences with GFS, with overall RMSE values of 0.23 mb (WVP), 0.6 mb (pressure), 1.3 (refractivity), and 1.5 °C (temperature). However, assessments against GFS/ECMWF models indicate overall compatibility, with insignificant differences between PlanetiQ profiles and model observations.

Investigating the impact of coupling HARMONIE-WINS50 (cy43) meteorology to LOTOS-EUROS (v2.2.002) on a simulation of NO2 concentrations over the Netherlands

Abstract. Meteorological fields calculated by numerical weather prediction (NWP) models drive offline chemical transport models (CTMs) to solve the transport, chemical reactions, and atmospheric interaction over the geographical domain of interest. HARMONIE (HIRLAM ALADIN Research on Mesoscale Operational NWP in Euromed) is a state-of-the-art non-hydrostatic NWP community model used at several European weather agencies to forecast weather at the local and/or regional scale. In this work, the HARMONIE WINS50 (cycle 43 cy43) reanalysis dataset at a resolution of 0.025° × 0.025° covering an area surrounding the North Sea for the years 2019–2021 was coupled offline to the LOTOS-EUROS (LOng-Term Ozone Simulation-EURopean Operational Smog model, v2.2.002) CTM. The impact of using either meteorological fields from HARMONIE or from ECMWF on LOTOS-EUROS simulations of NO2 has been evaluated against ground-level observations and TROPOMI tropospheric NO2 vertical columns. Furthermore, the difference between crucial meteorological input parameters such as the boundary layer height and the vertical diffusion coefficient between the hydrostatic ECMWF and non-hydrostatic HARMONIE data has been studied, and the vertical profiles of temperature, humidity, and wind are evaluated against meteorological observations at Cabauw in The Netherlands. The results of these first evaluations of the LOTOS-EUROS model performance in both configurations are used to investigate current uncertainties in air quality forecasting in relation to driving meteorological parameters and to assess the potential for improvements in forecasting pollution episodes at high resolutions based on the HARMONIE NWP model.

Impacts of Fengyun-4A and Ground-Based Observation Data Assimilation on the Forecast of Kaifeng’s Heavy Rainfall (2022) and Mechanism Analysis of the Event

The advancement of Numerical Weather Prediction (NWP) is pivotal for enhancing high-impact weather forecasting and warning systems. However, due to the high spatial and temporal inhomogeneity, the moisture field is difficult to describe by initial conditions in NWP models, which is the essential thermodynamic variable in the simulation of various physical processes. Data Assimilation techniques are central to addressing these challenges, integrating observational data with background fields to refine initial conditions and improve forecasting accuracy. This study evaluates the effectiveness of integrating observations from the Fengyun-4A (FY-4A) Advanced Geosynchronous Radiation Imager (AGRI) and ground-based microwave radiometer (MWR) in forecasts and mechanism analysis of a heavy rainfall event in the Kaifeng region of central China. Our findings reveal that jointly assimilating AGRI radiance and MWR data significantly enhances the model’s humidity profile accuracy across all atmospheric layers, resulting in improved heavy rainfall predictions. Analysis of the moisture sources indicates that the storm’s water vapor predominantly originates from westward air movement ahead of a high-altitude trough, with sustained channeling towards the rainfall zone, ensuring a continuous supply of moisture. The storm’s development is further facilitated by a series of atmospheric processes, including the interplay of high and low-level vorticity and divergence, vertical updrafts, the formation of a low-level jet, and the generation of unstable atmospheric energy. Additionally, this study examines the influence of Tai-hang Mountain’s terrain on precipitation patterns in the Kaifeng area. Our experiments, comparing a control setup (CTL) with varied terrain heights, demonstrate that reducing terrain height by 50–60% significantly decreases precipitation coverage and intensity. In contrast, increasing terrain height enhances precipitation, although this effect plateaus when the elevation increase exceeds 100%, closely mirroring the precipitation changes observed with a 75% terrain height increment.

Simulating the seeder–feeder impacts on cloud ice and precipitation over the Alps

Abstract. The ice phase impacts many cloud properties as well as cloud lifetime. Ice particles that sediment into a lower cloud from an upper cloud (external seeder–feeder process) or into the mixed-phase region of a deep cloud from cirrus levels (internal seeder–feeder process) can influence the ice phase of the lower cloud, amplify cloud glaciation and enhance surface precipitation. Recently, numerical weather prediction modeling studies have aimed at representing the ice crystal number concentration in mixed-phase clouds more accurately by including secondary ice formation processes. The increase in the ice crystal number concentration can impact the number of ice particles that sediment into the lower cloud and alter its composition and precipitation formation. In the Swiss Alps, the orography permits the formation of orographic clouds, making it ideal for studying the occurrence of multi-layered clouds and the seeder–feeder process. We present results from a case study on 18 May 2016, showing the occurrence frequency of multi-layered clouds and the seeder–feeder process. About half of all observed clouds were categorized as multi-layered, and the external seeder–feeder process occurred in 10 % of these clouds. Between cloud layers, ≈60 % of the ice particle mass was lost due to sublimation or melting. The external seeder–feeder process was found to be more important than the internal seeder–feeder process with regard to the impact on precipitation. In the case where the external seeder–feeder process was inhibited, the average surface precipitation and riming rate over the domain were both reduced by 8.5 % and 3.9 %, respectively. When ice–graupel collisions were allowed, further large reductions were seen in the liquid water fraction and riming rate. Inhibiting the internal seeder–feeder process enhanced the liquid water fraction by 6 % compared to a reduction of 5.8 % in the cloud condensate, therefore pointing towards the de-amplification in cloud glaciation and a reduction in surface precipitation. Adding to the observational evidence of frequent seeder–feeder situations, at least over Switzerland, our study highlights the extensive influence of sedimenting ice particles on the properties of feeder clouds as well as on precipitation formation.

Assessment for operational assimilation of horizontal line of sight winds from the European Space Agency's Aeolus at the Met Office

AbstractThe European Space Agency's Aeolus satellite was launched in August 2018 and began delivering horizontal line‐of‐sight (HLOS) wind observations in early September 2018. In early 2019, the Met Office began assessing the suitability of the HLOS winds for operational assimilation into its global numerical weather prediction system. We performed a number of assimilation experiments to assess the impact of HLOS wind observations on our global forecasts. We have found that assimilating HLOS winds changes the zonal winds in the analysis fields predominantly in the Tropics and Southern Hemisphere, with the largest changes being in the upper troposphere and lower stratosphere. This has a positive impact on the accuracy of the global weather forecasts, with improvements in the root‐mean‐square error seen throughout the troposphere. Assimilation of Aeolus HLOS winds improves the standard deviation of the observation minus background (a 6 hr forecast) of almost all other observation types, suggesting that the numerical weather prediction model analysis is improved, which consequently improves the 6 hr forecast. In a set of short‐period observation denial experiments, we found that assimilating Aeolus has an impact similar in magnitude to assimilating surface winds from scatterometers. Assimilating winds from the Rayleigh channel has approximately three times the impact that assimilating HLOS winds from the Mie channel does. Both channels contribute a measureable improvement to the global forecast, and we therefore started operational assimilation of winds from the Mie channel in December 2020 and the Rayleigh channel operationally in May 2022.

How frequent and which variables of automatic weather station data should be assimilated into WRF-3DVar model? A case study of a squall line event in Beijing

Accurate precipitation forecasting is one of the most challenging problems in mesoscale Numerical Weather Prediction (NWP) models. The utilization of 3-Dimensional Variational (3DVar) Data Assimilation technique based on automatic weather station (AWS) data can significantly enhance the accuracy of precipitation simulations and forecasts using the Weather Research and Forecasting (WRF) model. However, the impacts of different assimilation frequencies and meteorological variables on the accuracy of precipitation forecasts remain unclear. This study comprehensively evaluated the impact of different assimilation frequencies and meteorological variables on precipitation forecasts, as illustrated by a case study of a squall line event on August 2, 2017, in Beijing. Eight experiments were conducted by assimilating different combinations of meteorological variables (air pressure, temperature, relative humidity, wind speed, and wind direction) at various time intervals (1 h, 3 h, and 6 h). The results indicated that the WRF model roughly simulated the evolution of this event but overestimated the precipitation amount, accompanied by a large deviation in precipitation areas. Assimilating detailed AWS data significantly improves the model performance in precipitation forecasts. The experiment assimilating all variables at a 3 h interval yielded the most accurate forecasts, with the maximum threat score (TS) increasing from 0.02 to 0.56. Higher assimilation frequencies do not guarantee a better performance. In practice, a 3 h assimilation frequency emerges as an optimal choice for AWS data assimilation. The assimilation of various variables enhanced precipitation forecasts to varying degrees, with the optimum results achieved when all variables were assimilated. Wind speed and direction were the most significant factors in dynamically enhancing the simulation of precipitation areas. Relative humidity and temperature influenced the precipitation intensity by affecting the evolution of convective precipitation thermodynamically. The findings of this study can contribute to the development of AWS data assimilation strategies for the WRF-3DVar model, thereby enhancing the precipitation forecast accuracy.

Machine learning parameterization of the multi-scale Kain–Fritsch (MSKF) convection scheme and stable simulation coupled in the Weather Research and Forecasting (WRF) model using WRF–ML v1.0

Abstract. Warm-sector heavy rainfall along the south China coast poses significant forecasting challenges due to its localized nature and prolonged duration. To improve the prediction of such high-impact weather events, high-resolution numerical weather prediction (NWP) models are increasingly used to more accurately represent topographic effects. However, as these models' grid spacing approaches the scale of convective processes, they enter a “gray zone”, where the models struggle to fully resolve the turbulent eddies within the atmospheric boundary layer, necessitating partial parameterization. The appropriateness of applying convection parameterization (CP) schemes within this gray zone remains controversial. To address this, scale-aware CP schemes have been developed to improve the representation of convective transport. Among these, the multi-scale Kain–Fritsch (MSKF) scheme enhances the traditional Kain–Fritsch (KF) scheme, incorporating modifications that facilitate its effective application at spatial resolutions as high as 2 km. In recent years, there has been an increase in the application of machine learning (ML) models across various domains of atmospheric sciences, including efforts to replace conventional physical parameterizations with ML models. This work introduces a multi-output bidirectional long short-term memory (Bi-LSTM) model intended to replace the scale-aware MSKF CP scheme. This multi-output Bi-LSTM model is capable of simultaneously predicting the convection trigger while also modeling the associated convective tendencies and precipitation rates with a high performance. Data for training and testing the model are generated using the Weather Research and Forecast (WRF) model over south China at a horizontal resolution of 5 km. Furthermore, this work evaluates the performance of the WRF model coupled with the ML-based CP scheme against simulations with the traditional MSKF scheme. The results demonstrate that the Bi-LSTM model can achieve high accuracy, indicating the promising potential of ML models to substitute the MSKF scheme in the gray zone.

Study of boundary layer parameterization simulation uncertainties of sand-dust storm windfield using high-resolution three-dimensional Doppler wind lidar data

The processes of dust emission, the development, and dissipation of sand and dust storms all exhibit dynamic and rapidly changing characteristics, posing significant challenges for us to accurately represent their complex behaviors in numerical simulations. In the numerical weather prediction model, the selection and optimization of planetary boundary layer (PBL) parameterization schemes have a direct impact on the simulation results of PBL wind fields, primarily due to the significant uncertainty in the fine-grained simulation of PBL wind field structure. To gain a deeper understanding of the sources of uncertainty in simulating sand-dust storms through vertical wind field structure simulations of the PBL, this study utilizes high-resolution 3D Doppler wind lidar (DWL) data. Focusing on sand-dust storm wind fields with distinctive regional characteristics in Inner Mongolia, China, it examines the turbulent momentum equation and assumptions of the PBL. A comprehensive and systematic evaluation was conducted on the characteristics of horizontal and vertical wind fields, wind shear, as well as the dynamic and thermodynamic differences that arise when simulating typical sand-dust storm wind fields using the Weather Research and Forecasting (WRF) Model with various PBL schemes. This evaluation focused on simulating the typical wind field of sand-dust storms during various stages, including dust generation, occurrence, development, and dissipation. The findings indicate that non-adiabatic heating plays a significant role in the sustainability and evolution of sand-dust storms. The wind shear index holds particular importance as it embodies the dynamic factor in the model. The PBL schemes such as Mellor-Yamada Nakanishi and Niino Level 2.5 (MYNN2), A new version of asymmetric convective model (ACM2), and Grenier-Bretherton-McCaa (GBM), which factor in buoyancy effects, demonstrate more accurate wind shear simulations in sand-dust storms. Furthermore, the ACM2 scheme effectively replicates the fundamental variation characteristics of meteorological elements throughout the sand-dust storm lifecycle.

Improved freezing rain forecast using machine learning

Freezing rain is one of the most damaging weather phenomena in winter or early spring in many parts of the world, affecting traffic, power lines and agriculture. Thus, reliable and computationally efficient prediction of its occurrence is urgently needed in weather forecast operations. However, there are different thermodynamic processes that can lead to freezing rain, resulting in unsatisfactory forecasting performance of the state-of-the-art Numerical Weather Prediction (NWP) models. Here a data-driven forecasting method for freezing rain using machine learning technologies is proposed. Observations of weather phenomenon collected from 2515 national weather stations of China for winter of 2016 to 2019 and the corresponding atmospheric predictors derived from ERA5 reanalysis are used. The prediction function is constructed based on the classification and regression tree, and the predicting variables include temporal and vertical profiles of fundamental thermodynamic and kinematic parameters from 500hPa to 1000hPa, with a total dimension of 2304. The LightGBM (Light Gradient Boosting Machine) framework is adopted to train our prediction model and an algorithm-level approach of modifying the loss function is used to address the imbalance of classes to improve forecasting skill. Results show that the data-driven prediction model, namely DDFR (data driven forecast of freezing rain), out-performs the benchmark NWP, i.e., ECMWF IFS product. It’s improvements in terms of TS score range from 120% to 258% depending on different forecast leading times, which range from 0-12 hours. In addition, DDFR is applied in an operational NWP model of China. The problem of domain adaptation is tackled and transfer learning method is employed to adapt the original DDFR to this NWP model. The effectiveness of such adaptation has been demonstrated by its performance on both training and testing datasets.

A Comprehensive Northern Hemisphere Particle Microphysics Data Set From the Precipitation Imaging Package

AbstractMicrophysical observations of precipitating particles are critical data sources for numerical weather prediction models and remote sensing retrieval algorithms. However, obtaining coherent data sets of particle microphysics is challenging as they are often unindexed, distributed across disparate institutions, and have not undergone a uniform quality control process. This work introduces a unified, comprehensive Northern Hemisphere particle microphysical data set from the National Aeronautics and Space Administration precipitation imaging package (PIP), accessible in a standardized data format and stored in a centralized, public repository. Data is collected from 10 measurement sites spanning 34° latitude (37°N–71°N) over 10 years (2014–2023), which comprise a set of 1,070,000 precipitating minutes. The provided data set includes measurements of a suite of microphysical attributes for both rain and snow, including distributions of particle size, vertical velocity, and effective density, along with higher‐order products including an approximation of volume‐weighted equivalent particle densities, liquid equivalent snowfall, and rainfall rate estimates. The data underwent a rigorous standardization and quality assurance process to filter out erroneous observations to produce a self‐describing, scalable, and achievable data set. Case study analyses demonstrate the capabilities of the data set in identifying physical processes like precipitation phase‐changes at high temporal resolution. Bulk precipitation characteristics from a multi‐site intercomparison also highlight distinct microphysical properties unique to each location. This curated PIP data set is a robust database of high‐quality particle microphysical observations for constraining future precipitation retrieval algorithms, and offers new insights toward better understanding regional and seasonal differences in bulk precipitation characteristics.

A dynamic approach to three-dimensional radiative transfer in subkilometer-scale numerical weather prediction models: the dynamic TenStream solver v1.0

Abstract. The increasing resolution of numerical weather prediction models makes inter-column three-dimensional (3D) radiative transport more and more important. However, 3D radiative-transfer solvers are still computationally expensive, largely preventing their use in operational weather forecasting. To address this issue, Jakub and Mayer (2015) developed the TenStream solver. It extends the well-established two-stream method to three dimensions by using 10 instead of 2 streams to describe the transport of radiative energy through Earth's atmosphere. Building upon this method, this paper presents the dynamic TenStream solver, which provides a further acceleration of the original TenStream model. Compared to traditional solvers, this speedup is achieved by utilizing two main concepts. First, radiation is not calculated from scratch every time the model is called. Instead, a time-stepping scheme is introduced to update the radiation field, based on the result from the previous radiation time step. Secondly, the model is based on incomplete solves, with just the first few steps of an iterative scheme towards convergence performed every time it is called. Essentially, the model thereby just uses the ingoing fluxes of a grid box to update its outgoing fluxes. Combined, these two approaches move radiative transfer much closer to the way advection is handled in the dynamical core of a numerical weather prediction (NWP) model, as both use previously calculated results to update their variables and thereby just require access to the neighboring values of an individual grid box, facilitating model parallelization. To demonstrate the feasibility of this new solver, we apply it to a precomputed shallow-cumulus-cloud time series and test its performance in terms of both speed and accuracy. In terms of speed, the dynamic TenStream solver is shown to be about 3 times slower than a traditional 1D δ-Eddington approximation but noticeably faster than currently available 3D solvers. To evaluate the accuracy of the dynamic TenStream solver, we compare its results as well as calculations carried out using a 1D δ-Eddington approximation and the original TenStream solver, to benchmark calculations performed with the 3D Monte Carlo solver MYSTIC. We demonstrate that at the grid box level, dynamic TenStream is able to calculate heating rates and net irradiances at domain boundaries that are very close to those obtained by the original TenStream solver, thus offering a much better representation of the MYSTIC benchmark than the 1D δ-Eddington results. By calling the dynamic TenStream solver less frequently than the δ-Eddington approximation, we furthermore show that our new solver produces significantly better results than a 1D δ-Eddington approximation carried out with a similar computational demand. At these lower calling frequencies, however, the incomplete solves in the dynamic TenStream solver also lead to a buildup of bias with time, which becomes larger the lower the calling frequency is.

Life cycle dynamics of Greenland blocking from a potential vorticity perspective

Abstract. Blocking over Greenland stands out in comparison to blocking in other regions, as it favors accelerated Greenland Ice Sheet melting and has substantial impacts on surface weather in adjacent regions, particularly in Europe and North America. Climate models notoriously underestimate the frequency of blocking over Greenland in historical periods, but the reasons for this are not entirely clear, as we are still lacking a full dynamical understanding of Greenland blocking from formation through maintenance to decay. This study investigates the dynamics of blocking life cycles over Greenland based on ERA5 reanalysis data from 1979–2021. A year-round weather regime definition allows us to identify Greenland blocking as consistent life cycles with an objective onset, maximum, and decay stage. By applying a new quasi-Lagrangian potential vorticity (PV) perspective, following the negative, upper-tropospheric PV anomalies (PVAs−) associated with the block, we examine and quantify the contribution from different physical processes, including dry and moist dynamics, to the evolution of the PVA− amplitude. We find that PVAs− linked to blocking do not form locally over Greenland but propagate into the region along two distinct pathways (termed “upstream” and “retrogression”) during the days before the onset. The development of PVAs− differs more between the pathways than between seasons. Moist processes play a key role in the amplification of PVAs− before the onset and are linked to midlatitude warm conveyor belts. Interestingly, we find moist processes supporting the westward propagation of retrograding PVAs− from Europe, too, previously thought to be a process dominated by dry-barotropic Rossby wave propagation. After onset, moist processes remain the main contribution to PVA− amplification and maintenance. However, moist processes weaken markedly after the maximum stage, and dry processes, i.e., barotropic, nonlinear wave dynamics, dominate the decay of the PVAs− accompanied by a general decrease in blocking area. Our results corroborate the importance of moist processes in the formation and maintenance of Greenland blocking and suggest that a correct representation of moist processes might help reduce forecast errors linked to blocking in numerical weather prediction models and blocking biases in climate models.

Use of CAMS near Real-Time Aerosols in the HARMONIE-AROME NWP Model

Near real-time aerosol fields from the Copernicus Atmospheric Monitoring Services (CAMS), operated by the European Centre for Medium-Range Weather Forecasts (ECMWF), are configured for use in the HARMONIE-AROME Numerical Weather Prediction model. Aerosol mass mixing ratios from CAMS are introduced in the model through the first guess and lateral boundary conditions and are advected by the model dynamics. The cloud droplet number concentration is obtained from the aerosol fields and used by the microphysics and radiation schemes in the model. The results show an improvement in radiation, especially during desert dust events (differences of nearly 100 W/m2 are obtained). There is also a change in precipitation patterns, with an increase in precipitation, mainly during heavy precipitation events. A reduction in spurious fog is also found. In addition, the use of the CAMS near real-time aerosols results in an improvement in global shortwave radiation forecasts when the clouds are thick due to an improved estimation of the cloud droplet number concentration.

Development of an Emulator for Radiation Physics Simulator Using Stochastic Variational Gaussian Process Model

This paper describes an emulator that uses a Stochastic Variational Gaussian process (SVGP) regression model to parameterize radiation in a numerical weather prediction (NWP) model that meteorologically models the Earth's weather system. The computation of radiative processes is very large, accounting for most of the total NWP model computation. Statistical emulators are surrogate models that represent simulators and can overcome the computational limitations of very complex simulators such as radiative processes. Recently, artificial neural network-based radiative transfer emulators have been developed, and in this study, a statistical model, GP, is used to develop a radiative transfer emulator. The GP model has the advantage of calculating the uncertainty of the prediction along with the prediction, so the uncertainty of the prediction can be utilized appropriately. However, the computational complexity of the conventional GP model is very high, making it difficult to apply to large data. To solve this problem, an approximate approach, the SVGP model, was utilized. To further reduce the dimensionality of the input variables, we used a combined neural network and SVGP model. As a result, the SVGP-based radiative physics emulator improved its accuracy by about 20% compared to the artificial neural network emulator. However, the computation speed was about 3 to 9 times slower than the neural network emulator, but it was faster than the computation speed of the NWP model. This suggests that statistical emulators can be used to replace NWP model simulators.

Do AI models produce better weather forecasts than physics-based models? A quantitative evaluation case study of Storm Ciarán

There has been huge recent interest in the potential of making operational weather forecasts using machine learning techniques. As they become a part of the weather forecasting toolbox, there is a pressing need to understand how well current machine learning models can simulate high-impact weather events. We compare short to medium-range forecasts of Storm Ciarán, a European windstorm that caused sixteen deaths and extensive damage in Northern Europe, made by machine learning and numerical weather prediction models. The four machine learning models considered (FourCastNet, Pangu-Weather, GraphCast and FourCastNet-v2) produce forecasts that accurately capture the synoptic-scale structure of the cyclone including the position of the cloud head, shape of the warm sector and location of the warm conveyor belt jet, and the large-scale dynamical drivers important for the rapid storm development such as the position of the storm relative to the upper-level jet exit. However, their ability to resolve the more detailed structures important for issuing weather warnings is more mixed. All of the machine learning models underestimate the peak amplitude of winds associated with the storm, only some machine learning models resolve the warm core seclusion and none of the machine learning models capture the sharp bent-back warm frontal gradient. Our study shows there is a great deal about the performance and properties of machine learning weather forecasts that can be derived from case studies of high-impact weather events such as Storm Ciarán.

Uncertainties in cloud-radiative heating within an idealized extratropical cyclone

Abstract. Cloud-radiative heating (CRH) within the atmosphere affects the dynamics and predictability of extratropical cyclones. However, CRH is uncertain in numerical weather prediction and climate models, and this could affect model predictions of extratropical cyclones. In this paper, we present a systematic quantification of CRH uncertainties. To this end, we study an idealized extratropical cyclone simulated at a convection-permitting resolution of 2.5 km and combine large-eddy-model simulations at a 300 m resolution with offline radiative transfer calculations. We quantify four factors contributing to the CRH uncertainty in different regions of the cyclone: 3D cloud-radiative effects, parameterization of ice optical properties, cloud horizontal heterogeneity, and cloud vertical overlap. The last two factors can be considered essentially resolved at 300 m but need to be parameterized at a 2.5 km resolution. Our results indicate that parameterization of ice optical properties and cloud horizontal heterogeneity are the two factors contributing most to the mean uncertainty in CRH at larger spatial scales and can be more relevant for the large-scale dynamics of the cyclone. On the other hand, 3D cloud-radiative effects are much smaller on average, especially for stratiform clouds within the warm conveyor belt of the cyclone. Our analysis in particular highlights the potential to improve the simulation of CRH by better representing ice optical properties. Future work should, in particular, address how uncertainty in ice optical properties affects the dynamics and predictability of extratropical cyclones.

Analytical and adaptable initial conditions for dry and moist baroclinic waves in the global hydrostatic model OpenIFS (CY43R3)

Abstract. This article presents a description of an analytical, stable, and flexible initial background state for both dry and moist baroclinic wave simulation on an aquaplanet in order to test the dynamical core of numerical weather prediction models and study the dynamics and evolution of extratropical cyclones. The initial background state is derived from an analytical zonal wind speed field, or jet structure, and the hydrostatic primitive equations for moist adiabatic and frictionless flow in spherical coordinates. A baroclinic wave can develop if a perturbation is added to the zonal wind speed field. This new baroclinic wave configuration has been implemented in the Open Integrated Forecasting System (OpenIFS) CY43R3, a global numerical weather prediction model developed by the European Centre for Medium-Range Weather Forecasts. In total, seven parameters can be used to control the generation of the initial background state and hence the development of the baroclinic waves in the OpenIFS configuration file: the jet's width, the jet's height, the maximum zonal mean wind speed of the jet, the horizontal mean of the surface virtual temperature, the surface relative humidity, the lapse rate, and the surface roughness. Nine dry and nine moist initial background states have been generated to test their stability without perturbations. The meteorological stability of the initial states is investigated by examining the spatial distributions of the equivalent potential temperature, the absolute vorticity, and the Brunt–Väisälä frequency. Moreover, the root mean square error (RMSE) of the zonal wind speed has been computed to assess their numerical stability. Finally, six dry and six moist initial background state have been used with an unbalanced perturbation to ensure that the baroclinic life cycles that develop are physically realistic. The resulting baroclinic wave is shown to be sensitive to the jet's width. This configuration for baroclinic wave simulations will be used to create a large ensemble of baroclinic life cycles to study how extratropical cyclones may evolve in the future.

Weather‐Induced Satellite Orbit Perturbations

AbstractSatellites in Earth's orbit are exposed to Earth radiation, consisting of reflected solar and emitted thermal radiation, thereby exerting a non‐conservative force that causes acceleration and affects the orbits. Gravity Recovery and Climate Experiment Follow‐On (GRACE‐FO) mission aiming to retrieve the Earth's gravity potential is critically dependent on accounting for this force and all other non‐gravitational forces. There are both diurnal and seasonal variations in the Earth's radiation pressure, of which the seasonal variability can be represented by climatology. Nevertheless, the daily variations in the Earth's radiation pressure, due to the transient changes in the weather; for example, clouds and their properties, are not accounted for in the orbit perturbations studies. We show here that the top‐of‐atmosphere radiation fluxes computed with a numerical weather prediction (NWP) model explain most of the measured short‐term variations in the radial acceleration of the GRACE‐FO satellite. Our physics‐based modeling corrects a hitherto unexplained lack of power spectral density in the measured accelerations. For example, we can accurately model the accelerations associated with a tropical storm in the Indian Ocean in December 2020, which would not be possible when using climatological data. Our results demonstrate that using a global numerical weather prediction model significantly improves the simulation of non‐gravitational effects in the satellites' orbits. In the 7‐day data set, OpenIFS‐simulated acceleration exhibited higher accuracy than climatological‐data‐simulated acceleration (2.5 compared to 2.6 nms−2) and an improved precision (2.6 compared to 3.0 nms−2). This advancement contributes to a more precise orbit determination across various applications in Earth sciences.