Daryl T. Kleist
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Daryl T. Kleist has a total of 81 co-authors. Here are the top co-authors by number of shared publications:
Shun Liu
Environmental Modeling Center National Centers for Environmental Prediction College Park MD USA
4 shared publications- Jacob R. Carley
NOAA/NCEP/EMC
3 shared publications - Sho Yokota
Numerical Prediction Development Center, Japan Meteorological Agency, Tsukuba, Ibaraki 305-0052, Japan
3 shared publications - Xuguang Wang
School of Meteorology University of Oklahoma Norman OK USA
3 shared publications - Quanhua Liu
Center for Satellite Applications and Research, National Environmental Satellite Data, and Information Service (NESDIS), National Oceanic and Atmospheric Administration, College Park, USA
2 shared publications Ting Lei
NOAA/NWS/NCEP/Environmental Modeling Center College Park MD USA
2 shared publications
Latest Publications
- Open Access
- 24 January 2026
- Published by American Geophysical Union (AGU) in Geophysical Research Letters
- Vol. 53 (2)
Evaluating Machine Learning Weather Models for Data Assimilation: Fundamental Limitations in Tangent Linear and Adjoint Properties
Machine learning (ML) weather models like GraphCast and NeuralGCM show forecasting promise but face fundamental limitations for data assimilation (DA) integration. This study reveals critical problems in error covariance representation and adjoint sensitivity patterns challenging their operational viability. We evaluate tangent linear and adjoint models of GraphCast and NeuralGCM by comparing perturbation responses with MPAS‐A, a well‐established numerical weather prediction model. ML models exhibit unphysical adjoint sensitivities, including persistent localized responses and excessive noise at various atmospheric levels, contrasting sharply with physically consistent MPAS‐A patterns and indicating fundamental error covariance representation issues. Implications extend across DA methodologies. Unrealistic sensitivity patterns would generate distorted error covariances in ensemble systems and unphysical analysis increments in variational approaches like 4DVar. Assimilating single observations could create spurious corrections far from observation locations, degrading forecast skill. Despite forecasting capabilities, current ML weather models require significant improvements in linearization properties before reliable operational DA integration.
- 24 January 2026
- Published by American Geophysical Union (AGU) in Geophysical Research Letters
- Vol. 53 (2)
Open AccessEvaluating Machine Learning Weather Models for Data Assimilation: Fundamental Limitations in Tangent Linear and Adjoint Properties
Machine learning (ML) weather models like GraphCast and NeuralGCM show forecasting promise but face fundamental limitations for data assimilation (DA) integration. This study reveals critical problems in error covariance representation and adjoint sensitivity patterns challenging their operational viability. We evaluate tangent linear and adjoint models of GraphCast and NeuralGCM by comparing perturbation responses with MPAS‐A, a well‐established numerical weather prediction model. ML models exhibit unphysical adjoint sensitivities, including persistent localized responses and excessive noise at various atmospheric levels, contrasting sharply with physically consistent MPAS‐A patterns and indicating fundamental error covariance representation issues. Implications extend across DA methodologies. Unrealistic sensitivity patterns would generate distorted error covariances in ensemble systems and unphysical analysis increments in variational approaches like 4DVar. Assimilating single observations could create spurious corrections far from observation locations, degrading forecast skill. Despite forecasting capabilities, current ML weather models require significant improvements in linearization properties before reliable operational DA integration.
- Open Access
- 10 October 2025
- Published by American Geophysical Union (AGU) in Earth and Space Science
- Vol. 12 (10)
Implementation and Evaluation of Storm‐Following 3DIAU for Hurricane Intensity Prediction Improvements in Operational HAFS
Accurate hurricane intensity prediction remains a critical challenge in numerical weather prediction (NWP). This study implements and evaluates a newly developed Storm‐Following Three‐Dimensional Incremental Analysis Update (3DIAU) methodology for high‐resolution regional hurricane models with storm‐following nest capabilities. Built upon the feature‐relative 4DIAU approach proposed by Lu and Wang (2021), https://doi.org/10.1175/MWR‐D‐21‐0068.1 , the method gradually introduces Data Assimilation (DA) increments relative to the storm's position, reducing spin‐up imbalances and improving intensity predictions. Retrospective experiments were conducted over three Atlantic hurricane seasons (2021–2023) using the 2024 operational Hurricane Analysis and Forecast System (HAFS) version 2.0A configuration. Sensitivity experiments suggest that increment weighting should depend on storm strength. The storm‐strength‐dependent configuration yields an average improvement of 3% in intensity prediction skill, with modest gains in long‐term track predictions. A case study further demonstrates that gradual, storm‐relative adjustments mitigate disruptions caused by intermittent DA and enhance forecast performance. The Storm‐Following 3DIAU will be incorporated into the 2025 operational HAFS V2.1 upgrade.
- 10 October 2025
- Published by American Geophysical Union (AGU) in Earth and Space Science
- Vol. 12 (10)
Open AccessImplementation and Evaluation of Storm‐Following 3DIAU for Hurricane Intensity Prediction Improvements in Operational HAFS
Accurate hurricane intensity prediction remains a critical challenge in numerical weather prediction (NWP). This study implements and evaluates a newly developed Storm‐Following Three‐Dimensional Incremental Analysis Update (3DIAU) methodology for high‐resolution regional hurricane models with storm‐following nest capabilities. Built upon the feature‐relative 4DIAU approach proposed by Lu and Wang (2021), https://doi.org/10.1175/MWR‐D‐21‐0068.1 , the method gradually introduces Data Assimilation (DA) increments relative to the storm's position, reducing spin‐up imbalances and improving intensity predictions. Retrospective experiments were conducted over three Atlantic hurricane seasons (2021–2023) using the 2024 operational Hurricane Analysis and Forecast System (HAFS) version 2.0A configuration. Sensitivity experiments suggest that increment weighting should depend on storm strength. The storm‐strength‐dependent configuration yields an average improvement of 3% in intensity prediction skill, with modest gains in long‐term track predictions. A case study further demonstrates that gradual, storm‐relative adjustments mitigate disruptions caused by intermittent DA and enhance forecast performance. The Storm‐Following 3DIAU will be incorporated into the 2025 operational HAFS V2.1 upgrade.
- Open Access
- 9 November 2024
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 16 (11)
Scale‐ and Variable‐Dependent Localization for 3DEnVar Data Assimilation in the Rapid Refresh Forecast System
This study demonstrates the advantages of scale‐ and variable‐dependent localization (SDL and VDL) on three‐dimensional ensemble variational data assimilation of the hourly‐updated high‐resolution regional forecast system, the Rapid Refresh Forecast System (RRFS). SDL and VDL apply different localization radii for each spatial scale and variable, respectively, by extended control vectors. Single‐observation assimilation tests and cycling experiments with RRFS indicated that SDL can enlarge the localization radius without increasing the sampling error caused by the small ensemble size and decreased associated imbalance of the analysis field, which was effective at decreasing the bias of temperature and humidity forecasts. Moreover, simultaneous assimilation of conventional and radar reflectivity data with VDL, where a smaller localization radius was applied only for hydrometeors and vertical wind, improved precipitation forecasts without introducing noisy analysis increments. Statistical verification showed that these impacts contributed to forecast error reduction, especially for low‐level temperature and heavy precipitation.
- 9 November 2024
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 16 (11)
Open AccessScale‐ and Variable‐Dependent Localization for 3DEnVar Data Assimilation in the Rapid Refresh Forecast System
This study demonstrates the advantages of scale‐ and variable‐dependent localization (SDL and VDL) on three‐dimensional ensemble variational data assimilation of the hourly‐updated high‐resolution regional forecast system, the Rapid Refresh Forecast System (RRFS). SDL and VDL apply different localization radii for each spatial scale and variable, respectively, by extended control vectors. Single‐observation assimilation tests and cycling experiments with RRFS indicated that SDL can enlarge the localization radius without increasing the sampling error caused by the small ensemble size and decreased associated imbalance of the analysis field, which was effective at decreasing the bias of temperature and humidity forecasts. Moreover, simultaneous assimilation of conventional and radar reflectivity data with VDL, where a smaller localization radius was applied only for hydrometeors and vertical wind, improved precipitation forecasts without introducing noisy analysis increments. Statistical verification showed that these impacts contributed to forecast error reduction, especially for low‐level temperature and heavy precipitation.
- Open Access
- 24 July 2024
- Published by Wiley
Scale- and Variable-Dependent Localization for 3DEnVar Data Assimilation in the Rapid Refresh Forecast System
This study demonstrates the advantages of scale- and variable-dependent localization (SDL and VDL) on three-dimensional ensemble variational data assimilation of the hourly-updated high-resolution regional forecast system, the Rapid Refresh Forecast System (RRFS). SDL and VDL apply different localization radii for each spatial scale and variable, respectively, by extended control vectors. Single-observation assimilation tests and cycling experiments with RRFS indicated that SDL can enlarge the localization radius without increasing the sampling error caused by the small ensemble size and decreased associated imbalance of the analysis field, which was effective at decreasing the bias of temperature and humidity forecasts. Moreover, simultaneous assimilation of conventional and radar reflectivity data with VDL, where a smaller localization radius was applied only for hydrometeors and vertical wind, improved precipitation forecasts without introducing noisy analysis increments. Statistical verification showed that these impacts contributed to forecast error reduction, especially for low-level temperature and heavy precipitation.
- 24 July 2024
- Published by Wiley
Open AccessScale- and Variable-Dependent Localization for 3DEnVar Data Assimilation in the Rapid Refresh Forecast System
This study demonstrates the advantages of scale- and variable-dependent localization (SDL and VDL) on three-dimensional ensemble variational data assimilation of the hourly-updated high-resolution regional forecast system, the Rapid Refresh Forecast System (RRFS). SDL and VDL apply different localization radii for each spatial scale and variable, respectively, by extended control vectors. Single-observation assimilation tests and cycling experiments with RRFS indicated that SDL can enlarge the localization radius without increasing the sampling error caused by the small ensemble size and decreased associated imbalance of the analysis field, which was effective at decreasing the bias of temperature and humidity forecasts. Moreover, simultaneous assimilation of conventional and radar reflectivity data with VDL, where a smaller localization radius was applied only for hydrometeors and vertical wind, improved precipitation forecasts without introducing noisy analysis increments. Statistical verification showed that these impacts contributed to forecast error reduction, especially for low-level temperature and heavy precipitation.
- Open Access
- 15 May 2024
- Published by Wiley
Filtered Ensemble-based Tangent Linear Model for the Operational EnVar Data Assimilation System
The ensemble-based tangent linear model (ETLM), which evolves the static background error covariance (BEC) in time in the data assimilation window, is beneficial for improving the analysis and avoids the need for developing the tangent linear forecast model and its adjoint as in four-dimensional variational data assimilation (4DVar). This study proposes to apply the filtered ETLM (FETLM), which is composed of ensemble perturbations at limited time slots and localized with a quasi-Gaussian filter in the way that avoids calculation of the inverse matrix. Tests with the Lorenz (1996) model showed that FETLM evolved the static BEC as in 4DVar and improved the analysis. In experiments with the FETLM implemented in the hourly-updated regional forecast system, the Rapid Refresh Forecast System, the imbalance of the analysis was mitigated due to the flow-dependent BEC. Moreover, the cross-variable covariance in the FETLM made it possible to assimilate radar reflectivity effectively even without hydrometeors as control variables.
- 15 May 2024
- Published by Wiley
Open AccessFiltered Ensemble-based Tangent Linear Model for the Operational EnVar Data Assimilation System
The ensemble-based tangent linear model (ETLM), which evolves the static background error covariance (BEC) in time in the data assimilation window, is beneficial for improving the analysis and avoids the need for developing the tangent linear forecast model and its adjoint as in four-dimensional variational data assimilation (4DVar). This study proposes to apply the filtered ETLM (FETLM), which is composed of ensemble perturbations at limited time slots and localized with a quasi-Gaussian filter in the way that avoids calculation of the inverse matrix. Tests with the Lorenz (1996) model showed that FETLM evolved the static BEC as in 4DVar and improved the analysis. In experiments with the FETLM implemented in the hourly-updated regional forecast system, the Rapid Refresh Forecast System, the imbalance of the analysis was mitigated due to the flow-dependent BEC. Moreover, the cross-variable covariance in the FETLM made it possible to assimilate radar reflectivity effectively even without hydrometeors as control variables.
- Open Access
- 13 April 2023
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 15 (4)
JEDI‐Based Three‐Dimensional Ensemble‐Variational Data Assimilation System for Global Aerosol Forecasting at NCEP
A three‐dimensional ensemble‐variational global aerosol data assimilation system based on the Joint Effort for Data assimilation Integration (JEDI) was developed for the Global Ensemble Forecast System‐Aerosols (GEFS‐Aerosols) coupled with the Goddard Chemistry Aerosol Radiation and Transport (GOCART) model at the National Centers for Environmental Prediction. Aerosol mass mixing ratios in GEFS‐Aerosols were selected as control or analysis variables and were adjusted by assimilating 550 nm Aerosol Optical Depth (AOD) retrievals from the Visible Infrared Imaging Radiometer Suite (VIIRS) instruments onboard the Suomi National Polar‐orbiting Partnership (S‐NPP) satellite produced by the National Environmental Satellite, Data, and Information Service (NESDIS) at National Oceanic and Atmospheric Administration (NOAA). The original NOAA/NESDIS S‐NPP VIIRS Level 2.0 550 nm AOD retrievals were converted to JEDI Interface for Observation Data Access format. AOD forward operator and its tangent‐linear and adjoint models were implemented based on GOCART in JEDI Unified Forward Operator. A stochastically perturbed emission (SPE) approach was developed in the Common Community Physics Package‐based GEFS‐Aerosols to account for aerosol emission uncertainty. One‐month retrospective and three‐month near‐real‐time experiments consistently showed improved GEFS‐Aerosols analyses and forecasts from assimilating VIIRS 550 nm AOD retrievals against independent NASA Aqua and Terra Moderate Resolution Imaging Spectroradiometer AOD retrievals, Aerosol Robotic Network AOD, and independent AOD and aerosol analyses from NASA and European Centre for Medium‐Range Weather Forecasts. Through scaling and perturbing aerosol emissions, SPE enhanced ensemble error‐spread consistency and further improved AOD assimilation. The valid‐time‐shifting ensemble approach in a cost‐effective manner of populating background ensembles showed positive impacts on AOD assimilation.
- 13 April 2023
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 15 (4)
Open AccessJEDI‐Based Three‐Dimensional Ensemble‐Variational Data Assimilation System for Global Aerosol Forecasting at NCEP
A three‐dimensional ensemble‐variational global aerosol data assimilation system based on the Joint Effort for Data assimilation Integration (JEDI) was developed for the Global Ensemble Forecast System‐Aerosols (GEFS‐Aerosols) coupled with the Goddard Chemistry Aerosol Radiation and Transport (GOCART) model at the National Centers for Environmental Prediction. Aerosol mass mixing ratios in GEFS‐Aerosols were selected as control or analysis variables and were adjusted by assimilating 550 nm Aerosol Optical Depth (AOD) retrievals from the Visible Infrared Imaging Radiometer Suite (VIIRS) instruments onboard the Suomi National Polar‐orbiting Partnership (S‐NPP) satellite produced by the National Environmental Satellite, Data, and Information Service (NESDIS) at National Oceanic and Atmospheric Administration (NOAA). The original NOAA/NESDIS S‐NPP VIIRS Level 2.0 550 nm AOD retrievals were converted to JEDI Interface for Observation Data Access format. AOD forward operator and its tangent‐linear and adjoint models were implemented based on GOCART in JEDI Unified Forward Operator. A stochastically perturbed emission (SPE) approach was developed in the Common Community Physics Package‐based GEFS‐Aerosols to account for aerosol emission uncertainty. One‐month retrospective and three‐month near‐real‐time experiments consistently showed improved GEFS‐Aerosols analyses and forecasts from assimilating VIIRS 550 nm AOD retrievals against independent NASA Aqua and Terra Moderate Resolution Imaging Spectroradiometer AOD retrievals, Aerosol Robotic Network AOD, and independent AOD and aerosol analyses from NASA and European Centre for Medium‐Range Weather Forecasts. Through scaling and perturbing aerosol emissions, SPE enhanced ensemble error‐spread consistency and further improved AOD assimilation. The valid‐time‐shifting ensemble approach in a cost‐effective manner of populating background ensembles showed positive impacts on AOD assimilation.
- Open Access
- 13 December 2022
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Atmospheres
- Vol. 127 (24)
Implementation of a Discrete Dipole Approximation Scattering Database Into Community Radiative Transfer Model
The Community Radiative Transfer Model (CRTM) is a fast model that requires bulk optical properties of hydrometeors in the form of lookup tables to simulate all‐sky satellite radiances. Current cloud scattering lookup tables of CRTM were generated using the Mie‐Lorenz theory thus assuming spherical shapes for all frozen habits, while actual clouds contain frozen hydrometeors with different shapes. The Discrete Dipole Approximation (DDA) technique is an effective technique for simulating the optical properties of non‐spherical hydrometeors in the microwave region. This paper discusses the implementation and validation of a comprehensive DDA cloud scattering database into CRTM for the microwave frequencies. The original DDA database assumes total random orientation in the calculation of single scattering properties. The mass scattering parameters required by CRTM were then computed from single scattering properties and water content dependent particle size distributions. The new lookup tables eliminate the requirement for providing the effective radius as input to CRTM by using the cloud water content for the mass dimension. A collocated dataset of short‐term forecasts from Integrated Forecast System of the European Center for Medium‐Range Weather Forecasts and satellite microwave data was used for the evaluation of results. The results overall showed that the DDA lookup tables, in comparison with the Mie tables, greatly reduce the differences among simulated and observed values. The Mie lookup tables especially introduce excessive scattering for the channels operating below 90 GHz and low scattering for the channels above 90 GHz.
- 13 December 2022
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Atmospheres
- Vol. 127 (24)
Open AccessImplementation of a Discrete Dipole Approximation Scattering Database Into Community Radiative Transfer Model
The Community Radiative Transfer Model (CRTM) is a fast model that requires bulk optical properties of hydrometeors in the form of lookup tables to simulate all‐sky satellite radiances. Current cloud scattering lookup tables of CRTM were generated using the Mie‐Lorenz theory thus assuming spherical shapes for all frozen habits, while actual clouds contain frozen hydrometeors with different shapes. The Discrete Dipole Approximation (DDA) technique is an effective technique for simulating the optical properties of non‐spherical hydrometeors in the microwave region. This paper discusses the implementation and validation of a comprehensive DDA cloud scattering database into CRTM for the microwave frequencies. The original DDA database assumes total random orientation in the calculation of single scattering properties. The mass scattering parameters required by CRTM were then computed from single scattering properties and water content dependent particle size distributions. The new lookup tables eliminate the requirement for providing the effective radius as input to CRTM by using the cloud water content for the mass dimension. A collocated dataset of short‐term forecasts from Integrated Forecast System of the European Center for Medium‐Range Weather Forecasts and satellite microwave data was used for the evaluation of results. The results overall showed that the DDA lookup tables, in comparison with the Mie tables, greatly reduce the differences among simulated and observed values. The Mie lookup tables especially introduce excessive scattering for the channels operating below 90 GHz and low scattering for the channels above 90 GHz.