Journal of Advances in Modeling Earth Systems最新文献

{"title":"Prediction Beyond the Medium Range With an Atmosphere-Ocean Model That Combines Physics-Based Modeling and Machine Learning","authors":"Dhruvit Patel,&nbsp;Troy Arcomano,&nbsp;Brian Hunt,&nbsp;Istvan Szunyogh,&nbsp;Edward Ott","doi":"10.1029/2024MS004480","DOIUrl":"https://doi.org/10.1029/2024MS004480","url":null,"abstract":"<p>This paper explores the potential of a hybrid modeling approach that combines machine learning (ML) with conventional physics-based modeling for weather prediction beyond the medium range. It extends the work of Arcomano et al. (2022, https://doi.org/10.1029/2021ms002712), which tested the approach for short- and medium-range weather prediction, and the work of Arcomano et al. (2023, https://doi.org/10.1029/2022gl102649), which investigated its potential for climate modeling. The hybrid model used for the forecast experiments of the paper is based on the low-resolution, simplified parameterization atmospheric general circulation model SPEEDY. In addition to the hybridized prognostic variables of SPEEDY, the model has three purely ML-based prognostic variables: the 6 hr cumulative precipitation, the sea surface temperature, and the heat content of the top 300 m deep layer of the ocean (a new addition compared to the model used in Arcomano et al., 2023, https://doi.org/10.1029/2022gl102649). The model has skill in predicting the El Niño cycle and its global teleconnections with precipitation for 3–7 months depending on the season. The model captures equatorial variability of the precipitation associated with Kelvin and Rossby waves and MJO. Predictions of the precipitation in the equatorial region have skill for 15 days in the East Pacific and 11.5 days in the West Pacific. Though the model has low spatial resolution, for these tasks it has prediction skill comparable to what has been published for high-resolution, purely physics-based, conventional, operational forecast models.</p>","PeriodicalId":14881,"journal":{"name":"Journal of Advances in Modeling Earth Systems","volume":"17 4","pages":""},"PeriodicalIF":4.4,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2024MS004480","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143836186","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Modeling Antarctic Sea Ice Variability Using a Brittle Rheology","authors":"Rafael Santana,&nbsp;Guillaume Boutin,&nbsp;Christopher Horvat,&nbsp;Einar Ólason,&nbsp;Timothy Williams,&nbsp;Pierre Rampal","doi":"10.1029/2024MS004584","DOIUrl":"https://doi.org/10.1029/2024MS004584","url":null,"abstract":"<p>Sea ice is a composite solid material that sustains large fracture features at scales from meters to kilometres. These fractures can play an important role in coupled atmosphere-ocean processes. To model these features, brittle sea ice physics, via the Brittle-Bingham-Maxwell (BBM) rheology, has been implemented in the Lagrangian neXt generation Sea Ice Model (neXtSIM). In Arctic-only simulations, the BBM rheology has shown a capacity to represent observationally consistent sea ice fracture patterns and breakup across a wide range of time and length scales. Still, it has not been tested whether this approach is suitable for the modeling of Antarctic sea ice, which is thinner and more seasonal compared to Arctic sea ice, and whether the ability to reproduce sea ice fractures has an impact on simulating Antarctic sea ice properties. Here, we introduce a new 50-km grid-spacing Antarctic configuration of neXtSIM, neXtSIM-Ant, using the BBM rheology. We evaluate this simulation against observations of sea ice extent, drift, and thickness and compare it with identically-forced neXtSIM simulations that use the standard modified Elastic-Visco-Plastic (mEVP) rheology. In general, using BBM results in thicker sea ice and an improved correlation of sea ice drift with observations than mEVP. We suggest that this is related to short-duration breakup events caused by Antarctic storms that are not well-simulated in the viscous-plastic model.</p>","PeriodicalId":14881,"journal":{"name":"Journal of Advances in Modeling Earth Systems","volume":"17 4","pages":""},"PeriodicalIF":4.4,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2024MS004584","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143836074","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Simulating Atmospheric Processes in Earth System Models and Quantifying Uncertainties With Deep Learning Multi-Member and Stochastic Parameterizations","authors":"Gunnar Behrens,&nbsp;Tom Beucler,&nbsp;Fernando Iglesias-Suarez,&nbsp;Sungduk Yu,&nbsp;Pierre Gentine,&nbsp;Michael Pritchard,&nbsp;Mierk Schwabe,&nbsp;Veronika Eyring","doi":"10.1029/2024MS004272","DOIUrl":"https://doi.org/10.1029/2024MS004272","url":null,"abstract":"<p>Deep learning is a powerful tool to represent subgrid processes in climate models, but many application cases have so far used idealized settings and deterministic approaches. Here, we develop stochastic parameterizations with calibrated uncertainty quantification to learn subgrid convective and turbulent processes and surface radiative fluxes of a superparameterization embedded in an Earth System Model (ESM). We explore three methods to construct stochastic parameterizations: (a) a single Deep Neural Network (DNN) with Monte Carlo Dropout; (b) a multi-member parameterization; and (c) a Variational Encoder Decoder with latent space perturbation. We show that the multi-member parameterization improves the representation of convective processes, especially in the planetary boundary layer, compared to individual DNNs. The respective uncertainty quantification illustrates that methods (b) and (c) are advantageous compared to a dropout-based DNN parameterization regarding the spread of convective processes. Hybrid simulations with our best-performing multi-member parameterizations remained challenging and crash within the first days. Therefore, we develop a pragmatic partial coupling strategy relying on the superparameterization for condensate emulation. Partial coupling reduces the computational efficiency of hybrid Earth-like simulations but enables model stability over 5 months with our multi-member parameterizations. However, our hybrid simulations exhibit biases in thermodynamic fields and differences in precipitation patterns. Despite this, the multi-member parameterizations enable improvements in reproducing tropical extreme precipitation compared to a traditional convection parameterization. Despite these challenges, our results indicate the potential of a new generation of multi-member machine learning parameterizations leveraging uncertainty quantification to improve the representation of stochasticity of subgrid effects.</p>","PeriodicalId":14881,"journal":{"name":"Journal of Advances in Modeling Earth Systems","volume":"17 4","pages":""},"PeriodicalIF":4.4,"publicationDate":"2025-04-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2024MS004272","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143826732","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Navigating the Noise: Bringing Clarity to ML Parameterization Design With \u0000 \u0000 \u0000 O\u0000 \u0000 $boldsymbol{mathcal{O}}$\u0000 (100) Ensembles","authors":"Jerry Lin,&nbsp;Sungduk Yu,&nbsp;Liran Peng,&nbsp;Tom Beucler,&nbsp;Eliot Wong-Toi,&nbsp;Zeyuan Hu,&nbsp;Pierre Gentine,&nbsp;Margarita Geleta,&nbsp;Mike Pritchard","doi":"10.1029/2024MS004551","DOIUrl":"https://doi.org/10.1029/2024MS004551","url":null,"abstract":"<p>Machine-learning (ML) parameterizations of subgrid processes (here of turbulence, convection, and radiation) may one day replace conventional parameterizations by emulating high-resolution physics without the cost of explicit simulation. However, uncertainty about the relationship between offline and online performance (i.e., when integrated with a large-scale general circulation model) hinders their development. Much of this uncertainty stems from limited sampling of the noisy, emergent effects of upstream ML design decisions on downstream online hybrid simulation. Our work rectifies the sampling issue via the construction of a semi-automated, end-to-end pipeline for <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mi>O</mi>\u0000 <mrow>\u0000 <mo>(</mo>\u0000 <mn>100</mn>\u0000 <mo>)</mo>\u0000 </mrow>\u0000 </mrow>\u0000 <annotation> $mathcal{O}(100)$</annotation>\u0000 </semantics></math> size ensembles of hybrid simulations, revealing important nuances in how systematic reductions in offline error manifest in changes to online error and online stability. For example, removing dropout and switching from a Mean Squared Error to a Mean Absolute Error loss both reduce offline error, but they have opposite effects on online error and online stability. Other design decisions, like incorporating memory, converting moisture input from specific humidity to relative humidity, using batch normalization, and training on multiple climates do not come with any such compromises. Finally, we show that ensemble sizes of <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mi>O</mi>\u0000 <mrow>\u0000 <mo>(</mo>\u0000 <mn>100</mn>\u0000 <mo>)</mo>\u0000 </mrow>\u0000 </mrow>\u0000 <annotation> $mathcal{O}(100)$</annotation>\u0000 </semantics></math> may be necessary to reliably detect causally relevant differences online. By enabling rapid online experimentation at scale, we can empirically settle debates regarding subgrid ML parameterization design that would have otherwise remained unresolved in the noise.</p>","PeriodicalId":14881,"journal":{"name":"Journal of Advances in Modeling Earth Systems","volume":"17 4","pages":""},"PeriodicalIF":4.4,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2024MS004551","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143824633","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Implementation of Dynamic Fire Injection Height in GFDL's Atmospheric Model (AM4.0): Impacts on Aerosol Profiles and Radiation","authors":"Arman Pouyaei,&nbsp;Paul Ginoux,&nbsp;Daniel S. Ward,&nbsp;Yan Yu,&nbsp;Larry W. Horowitz","doi":"10.1029/2024MS004407","DOIUrl":"https://doi.org/10.1029/2024MS004407","url":null,"abstract":"<p>Wildfires inject aerosols into the atmosphere at varying altitudes, modifying long-range transport, which impacts Earth's climate system and air quality. Most global climate models use prescribed fixed-height injections, not accounting for the dynamic variability of wildfires. In this study, we enhance the injection method of biomass burning aerosols implemented in the Geophysical Fluid Dynamic Laboratory's Atmospheric Model version 4.0, shifting to a more mechanistic approach. We test several injection height schemes to assess their impact on the Earth's radiation budget by performing 18-year global simulations. Comparison of modeled injection height from the mechanistic scheme with observations indicates error within instrumental uncertainty (less than 500 m). Aerosol Optical Depth is systematically underestimated due to biases in the emission data set, but the mechanistic scheme significantly reduces this bias by up to 0.5 optical depth units during extreme wildfire seasons over boreal forests. In term of the vertical profile of the aerosol extinction coefficient, a comparison with satellite observations indicates significant improvement below 4 km altitude. Dynamic injection of biomass burning emissions changed the net clear-sky radiative flux at top of the atmosphere regionally (±1.5 Wm⁻²) and reduced it by −0.38 Wm⁻² at the surface globally, relative to a baseline with no fire emissions. The temperature gradient anomaly associated with the dynamic injection of absorbing aerosols affects the atmospheric stability and circulation patterns. This study highlights the need to implement dynamic injection of fire emissions to simulate more accurately the atmospheric distribution of aerosols and their interactions with Earth's climate system.</p>","PeriodicalId":14881,"journal":{"name":"Journal of Advances in Modeling Earth Systems","volume":"17 4","pages":""},"PeriodicalIF":4.4,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2024MS004407","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143824632","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Immersion Freezing in Particle-Based Aerosol-Cloud Microphysics: A Probabilistic Perspective on Singular and Time-Dependent Models","authors":"Sylwester Arabas,&nbsp;Jeffrey H. Curtis,&nbsp;Israel Silber,&nbsp;Ann M. Fridlind,&nbsp;Daniel A. Knopf,&nbsp;Matthew West,&nbsp;Nicole Riemer","doi":"10.1029/2024MS004770","DOIUrl":"https://doi.org/10.1029/2024MS004770","url":null,"abstract":"<p>Cloud droplets containing immersed ice-nucleating particles (INPs) may freeze at temperatures above the homogeneous freezing threshold temperature in a process referred to as immersion freezing. In modeling studies, immersion freezing is often described using either so-called “singular” or “time-dependent” parameterizations. Here, we compare both approaches and discuss them in the context of probabilistic particle-based (super-droplet) cloud microphysics modeling. First, using a box model, we contrast how both parameterizations respond to idealized ambient cooling rate profiles and quantify the impact of the polydispersity of the immersed surface spectrum on the frozen fraction evolution. Presented simulations highlight that the singular approach, constituting a time-integrated form of a more general time-dependent approach, is only accurate under a limited range of ambient cooling rates. The time-dependent approach is free from this limitation. Second, using a prescribed-flow two-dimensional cloud model, we illustrate the macroscopic differences in the evolution in time of ice particle concentrations in simulations with flow regimes relevant to ambient cloud conditions. The flow-coupled aerosol-budget-resolving simulations highlight the benefits and challenges of modeling cloud condensation nuclei activation and immersion freezing on insoluble ice nuclei with super-particle methods. The challenges stem, on the one hand, from heterogeneous ice nucleation being contingent on the presence of relatively sparse immersed INPs, and on the other hand, from the need to represent a vast population of particles with relatively few so-called super particles (each representing a multiplicity of real particles). We discuss the critical role of the sampling strategy for particle attributes, including the INP size, the freezing temperature (for singular scheme) and the multiplicity.</p>","PeriodicalId":14881,"journal":{"name":"Journal of Advances in Modeling Earth Systems","volume":"17 4","pages":""},"PeriodicalIF":4.4,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2024MS004770","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143824634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Forward Modeling of Bending Angles With a Two-Dimensional Operator for GNSS Airborne Radio Occultations in Atmospheric Rivers","authors":"P. Hordyniec,&nbsp;J. S. Haase,&nbsp;M. J. Murphy Jr.,&nbsp;B. Cao,&nbsp;A. M. Wilson,&nbsp;I. H. Banos","doi":"10.1029/2024MS004324","DOIUrl":"https://doi.org/10.1029/2024MS004324","url":null,"abstract":"<p>The Global Navigation Satellite System (GNSS) airborne radio occultation (ARO) technique is used to retrieve profiles of the atmosphere during reconnaissance missions for atmospheric rivers (ARs) on the west coast of the United States. The measurements of refractive bending angle integrate the effects of variations in refractive index over long near-horizontal ray-paths from a spaceborne transmitter to a receiver onboard an aircraft. A forward operator is required to assimilate ARO observations, which are sensitive to pressure, temperature, and humidity, into numerical weather prediction models to support forecasting of ARs. A two-dimensional (2D) bending angle operator is proposed to enable capturing key atmospheric features associated with strong ARs. Comparison to a one-dimensional (1D) forward model supports the evidence of large bending angle departures within 3–7 km impact heights for observations collected in a region characterized by the integrated water vapor transport (IVT) magnitude above 500 kg <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <msup>\u0000 <mi>m</mi>\u0000 <mrow>\u0000 <mo>−</mo>\u0000 <mn>1</mn>\u0000 </mrow>\u0000 </msup>\u0000 <msup>\u0000 <mi>s</mi>\u0000 <mrow>\u0000 <mo>−</mo>\u0000 <mn>1</mn>\u0000 </mrow>\u0000 </msup>\u0000 </mrow>\u0000 <annotation> ${mathrm{m}}^{-1}{mathrm{s}}^{-1}$</annotation>\u0000 </semantics></math>. The assessment of the 2D forward model for ARO retrievals is based on a sequence of six flights leading up to a significant AR precipitation event in January 2021. Since the observations often sample regions outside the AR where moisture is low, the significance of horizontal variations is obscured in the average bending angle statistics. Examples from individual flights sampling the cross-section of an AR support the need for the 2D forward model. Additional simulation experiments are performed to quantify forward modeling errors due to tangent point drift and horizontal gradients suggesting contributions on the order of 5% and 20%, respectively.</p>","PeriodicalId":14881,"journal":{"name":"Journal of Advances in Modeling Earth Systems","volume":"17 4","pages":""},"PeriodicalIF":4.4,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2024MS004324","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143824717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using Machine Learning to Generate a GISS ModelE Calibrated Physics Ensemble (CPE)","authors":"Gregory S. Elsaesser,&nbsp;Marcus van Lier-Walqui,&nbsp;Qingyuan Yang,&nbsp;Maxwell Kelley,&nbsp;Andrew S. Ackerman,&nbsp;Ann M. Fridlind,&nbsp;Gregory V. Cesana,&nbsp;Gavin A. Schmidt,&nbsp;Jingbo Wu,&nbsp;Ali Behrangi,&nbsp;Suzana J. Camargo,&nbsp;Bithi De,&nbsp;Kuniaki Inoue,&nbsp;Nicolas M. Leitmann-Niimi,&nbsp;Jeffrey D. O. Strong","doi":"10.1029/2024MS004713","DOIUrl":"https://doi.org/10.1029/2024MS004713","url":null,"abstract":"<p>A neural network (NN) surrogate of the NASA GISS ModelE atmosphere (version E3) is trained on a perturbed parameter ensemble (PPE) spanning 45 physics parameters and 36 outputs. The NN is leveraged in a Markov Chain Monte Carlo (MCMC) Bayesian parameter inference framework to generate a second <i>posterior</i> constrained ensemble coined a “calibrated physics ensemble,” or CPE. The CPE members are characterized by diverse parameter combinations and are, by definition, close to top-of-atmosphere radiative balance, and must broadly agree with numerous hydrologic, energy cycle and radiative forcing metrics simultaneously. Global observations of numerous cloud, environment, and radiation properties (provided by global satellite products) are crucial for CPE generation. The inference framework explicitly accounts for discrepancies (or biases) in satellite products during CPE generation. We demonstrate that product discrepancies strongly impact calibration of important model parameter settings (e.g., convective plume entrainment rates; fall speed for cloud ice). Structural improvements new to E3 are retained across CPE members (e.g., stratocumulus simulation). Notably, the framework improved the simulation of shallow cumulus and Amazon rainfall while not degrading radiation fields, an upgrade that neither default parameters nor Latin Hypercube parameter searching achieved. Analyses of the initial PPE suggested several parameters were unimportant for output variation. However, many “unimportant” parameters were needed for CPE generation, a result that brings to the forefront how parameter importance should be determined in PPEs. From the CPE, two diverse 45-dimensional parameter configurations are retained to generate radiatively-balanced, auto-tuned atmospheres that were used in two E3 submissions to CMIP6.</p>","PeriodicalId":14881,"journal":{"name":"Journal of Advances in Modeling Earth Systems","volume":"17 4","pages":""},"PeriodicalIF":4.4,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2024MS004713","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821986","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Characteristics and Representation of Subgrid Convective Flux in a Tropical Cyclone Convection System at Convection-Permitting Resolution","authors":"Xu Zhang","doi":"10.1029/2024MS004558","DOIUrl":"https://doi.org/10.1029/2024MS004558","url":null,"abstract":"<p>A large-eddy simulation (LES) of an idealized tropical cyclone convection system was conducted as a benchmark to provide the statistical characteristics of subgrid convective flux at convection-permitting resolution and to evaluate three existing scale-adaptive convection parameterization schemes in the Weather Research and Forecasting model and a nonlinear horizontal gradient (<i>H</i>-gradient) term. Coarse-grained results showed that the vertical profiles of subgrid convective flux presented various modes and each mode exhibited a thin vertical extent, in contrast to the ensemble-mean profile with a bottom-heavy, deep structure. A good spatial correspondence between subgrid convective flux and grid-scale vertical velocity was found in the LES benchmark. The existing convection parameterization schemes were unable to reasonably represent such vertical profiles and the spatial distributions of subgrid convective flux, particularly at high levels. The <i>H</i>-gradient term was able to represent these characteristics of subgrid convective flux, both in terms of vertical profile and spatial distribution. The limitations of traditional mass-flux convection parameterization schemes at convection-permitting resolution and related underlying assumptions are discussed. The physical backgrounds and significances of mass-flux convection parameterization schemes and the <i>H</i>-gradient term are clarified.</p>","PeriodicalId":14881,"journal":{"name":"Journal of Advances in Modeling Earth Systems","volume":"17 4","pages":""},"PeriodicalIF":4.4,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2024MS004558","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143809676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Toward Calibrated Ensembles of Neural Weather Model Forecasts","authors":"J. Baño-Medina,&nbsp;A. Sengupta,&nbsp;D. Watson-Parris,&nbsp;W. Hu,&nbsp;L. Delle Monache","doi":"10.1029/2024MS004734","DOIUrl":"https://doi.org/10.1029/2024MS004734","url":null,"abstract":"<p>Neural Weather Models (NWM) are novel data-driven weather forecasting tools based on neural networks, that have recently achieved comparable deterministic forecast skill to current operational approaches using significantly less real-time computational resources. They require short inference times, which can potentially improve the characterization of forecast uncertainty by designing very large ensembles, which is of paramount importance for for example, extreme events, and critical for various socio-economic sectors. Here we propose a novel ensemble design for NWMs spanning two main sources of uncertainty: model uncertainty and initial condition uncertainty. For the model uncertainty, we propose an effective strategy for creating a diverse ensemble of NWMs that captures uncertainty in key model parameters. For the initial condition uncertainty, we explore the “breeding of growing modes” for the first time on NWMs, a technique traditionally used for operational numerical weather predictions to estimate the initial condition uncertainty. The combination of these two types of uncertainty produces an ensemble of NWM-based forecasts that is shown to improve upon benchmark probabilistic NWM and is competitive with the 50-member ensemble of the European Centre for Medium-Range Weather Forecasts based on the Integrated Forecasting System (IFS), in terms of both error and calibration. Results are particularly promising over land for three key variables: total column water vapor, surface wind and surface air temperature. The proposed strategy is scalable, enabling the generation of very large ensembles (<span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 <mo>&gt;</mo>\u0000 </mrow>\u0000 <annotation> ${ &gt;} $</annotation>\u0000 </semantics></math>100) with potential applications for extreme events.</p>","PeriodicalId":14881,"journal":{"name":"Journal of Advances in Modeling Earth Systems","volume":"17 4","pages":""},"PeriodicalIF":4.4,"publicationDate":"2025-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2024MS004734","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143786980","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}