Storm-Based Probabilistic Hail Forecasting with Machine Learning Applied to Convection-Allowing Ensembles

AbstractForecasting severe hail accurately requires predicting how well atmospheric conditions support the development of thunderstorms, the growth of large hail, and the minimal loss of hail mass to melting before reaching the surface. Existing hail forecasting techniques incorporate information about these processes from proximity soundings and numerical weather prediction models, but they make many simplifying assumptions, are sensitive to differences in numerical model configuration, and are often not calibrated to observations. In this paper a storm-based probabilistic machine learning hail forecasting method is developed to overcome the deficiencies of existing methods. An object identification and tracking algorithm locates potential hailstorms in convection-allowing model output and gridded radar data. Forecast storms are matched with observed storms to determine hail occurrence and the parameters of the radar-estimated hail size distribution. The database of forecast storms contains information a...

[1]  G. Brier.  VERIFICATION OF FORECASTS EXPRESSED IN TERMS OF PROBABILITY , 1950 .

[2]  R. C. Miller,et al.  A Method for Forecasting Hailstone Size at the Earth's Surface , 1953 .

[3]  J. Munkres.  ALGORITHMS FOR THE ASSIGNMENT AND TRANSIORTATION tROBLEMS* , 1957 .

[4]  H. Glahn,et al.  The Use of Model Output Statistics (MOS) in Objective Weather Forecasting , 1972 .

[5]  A. H. Murphy.  A New Vector Partition of the Probability Score , 1973 .

[6]  G. Mellor,et al.  Development of a turbulence closure model for geophysical fluid problems , 1982 .

[7]  S. Nelson,et al.  The Influence of Storm Flow Structure on Hail Growth. , 1983 .

[8]  G. Foote,et al.  A Study of Hail Growth Utilizing Observed Storm Conditions. , 1984 .

[9]  A. H. Murphy,et al.  The attributes diagram A geometrical framework for assessing the quality of probability forecasts , 1986 .

[10]  Roy Rasmussen,et al.  Melting and Shedding of Graupel and Hail. Part I: Model Physics , 1987 .

[11]  R. Rasmussen,et al.  Melting and Shedding of Graupel and Hail. Part II: Sensitivity Study , 1987 .

[12]  John P. Pino,et al.  An Interactive Method for Estimating Maximum Hailstone Size from Forecast Soundings , 1990 .

[13]  R. Johns,et al.  Severe Local Storms Forecasting , 1992 .

[14]  M. Dixon,et al.  TITAN: Thunderstorm Identification, Tracking, Analysis, and Nowcasting—A Radar-based Methodology , 1993 .

[15]  Rich Caruana,et al.  Multitask Learning , 1998, Encyclopedia of Machine Learning and Data Mining.

[16]  A. Witt,et al.  An Enhanced Hail Detection Algorithm for the WSR-88D , 1998 .

[17]  Caren Marzban,et al.  A Bayesian Neural Network for Severe-Hail Size Prediction , 2001 .

[18]  E. Ebert.  Ability of a Poor Man's Ensemble to Predict the Probability and Distribution of Precipitation , 2001 .

[19]  G. Reuter,et al.  Modeling Maximum Hail Size in Alberta Thunderstorms , 2002 .

[20]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[21]  Rich Caruana,et al.  Multitask Learning , 1997, Machine Learning.

[22]  Kevin W. Manning,et al.  Explicit Forecasts of Winter Precipitation Using an Improved Bulk Microphysics Scheme. Part I: Description and Sensitivity Analysis , 2004 .

[23]  H. Niino,et al.  An Improved Mellor–Yamada Level-3 Model with Condensation Physics: Its Design and Verification , 2004 .

[24]  Ryan E. Jewell,et al.  Evaluation of Alberta Hail Growth Model Using Severe Hail Proximity Soundings from the United States , 2004 .

[25]  Russell S. Schneider,et al.  The May 2003 extended tornado outbreak , 2005 .

[26]  M. Yau,et al.  A Multimoment Bulk Microphysics Parameterization. Part I: Analysis of the Role of the Spectral Shape Parameter , 2005 .

[27]  B. Galperin,et al.  ‘Application of a New Spectral Theory of Stably Stratified Turbulence to the Atmospheric Boundary Layer over Sea Ice’ , 2005 .

[28]  H. Zou,et al.  Regularization and variable selection via the elastic net , 2005 .

[29]  J. Dudhia,et al.  A New Vertical Diffusion Package with an Explicit Treatment of Entrainment Processes , 2006 .

[30]  W. Briggs.  Statistical Methods in the Atmospheric Sciences , 2007 .

[31]  Kevin W. Manning,et al.  Experiences with 0–36-h Explicit Convective Forecasts with the WRF-ARW Model , 2008 .

[32]  S. J. Weiss,et al.  Some Practical Considerations Regarding Horizontal Resolution in the First Generation of Operational Convection-Allowing NWP , 2008 .

[33]  G. Thompson,et al.  Impact of Cloud Microphysics on the Development of Trailing Stratiform Precipitation in a Simulated Squall Line: Comparison of One- and Two-Moment Schemes , 2009 .

[34]  Jeffrey L. Anderson,et al.  The Data Assimilation Research Testbed: A Community Facility , 2009 .

[35]  Stanley A. Changnon,et al.  Increasing major hail losses in the U.S. , 2009 .

[36]  Amy McGovern,et al.  Classification of Convective Areas Using Decision Trees , 2009 .

[37]  Paul J. Roebber,et al.  Visualizing Multiple Measures of Forecast Quality , 2009 .

[38]  Valliappa Lakshmanan,et al.  An Efficient , General-Purpose Technique to Identify Storm Cells in Geospatial Images , 2010 .

[39]  Travis M. Smith,et al.  An Objective Method of Evaluating and Devising Storm-Tracking Algorithms , 2010 .

[40]  John S. Kain,et al.  Extracting Unique Information from High-Resolution Forecast Models: Monitoring Selected Fields and Phenomena Every Time Step , 2010 .

[41]  K. Elmore,et al.  Reaching Scientific Consensus Through a Competition , 2010 .

[42]  Song-You Hong,et al.  Development of an Effective Double-Moment Cloud Microphysics Scheme with Prognostic Cloud Condensation Nuclei (CCN) for Weather and Climate Models , 2010 .

[43]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[44]  Jian Zhang,et al.  National mosaic and multi-sensor QPE (NMQ) system description, results, and future plans , 2011 .

[45]  S. J. Weiss,et al.  Probabilistic Forecast Guidance for Severe Thunderstorms Based on the Identification of Extreme Phenomena in Convection-Allowing Model Forecasts , 2011 .

[46]  Richard L. Thompson,et al.  Convective Modes for Significant Severe Thunderstorms in the Contiguous United States. Part I: Storm Classification and Climatology , 2012 .

[47]  S. J. Weiss,et al.  An Overview of the 2010 Hazardous Weather Testbed Experimental Forecast Program Spring Experiment , 2012 .

[48]  Richard L. Thompson,et al.  Convective Modes for Significant Severe Thunderstorms in the Contiguous United States. Part II: Supercell and QLCS Tornado Environments , 2012 .

[49]  Harold E. Brooks,et al.  An Objective High-Resolution Hail Climatology of the Contiguous United States , 2012 .

[50]  James Correia,et al.  Tornado Pathlength Forecasts from 2010 to 2011 Using Ensemble Updraft Helicity , 2013 .

[51]  Agostino Manzato,et al.  Hail in Northeast Italy: A Neural Network Ensemble Forecast Using Sounding-Derived Indices , 2013 .

[52]  John K. Williams,et al.  Using random forests to diagnose aviation turbulence , 2013, Machine Learning.

[53]  Roshanak Nateghi,et al.  Power Outage Estimation for Tropical Cyclones: Improved Accuracy with Simpler Models , 2014, Risk analysis : an official publication of the Society for Risk Analysis.

[54]  I. Giammanco,et al.  Evaluating Hail Damage Using Property Insurance Claims Data , 2015 .

[55]  Walker S. Ashley,et al.  Spatiotemporal analysis of tornado exposure in five US metropolitan areas , 2015, Natural Hazards.

[56]  Gregory Thompson,et al.  Parameterization of Cloud Microphysics Based on the Prediction of Bulk Ice Particle Properties. Part II: Case Study Comparisons with Observations and Other Schemes , 2015 .

[57]  Kathryn R. Fossell,et al.  NCAR’s Experimental Real-Time Convection-Allowing Ensemble Prediction System , 2015 .

[58]  James Correia,et al.  Day-Ahead Hail Prediction Integrating Machine Learning with Storm-Scale Numerical Weather Models , 2015, AAAI.

[59]  H. Morrison,et al.  Parameterization of Cloud Microphysics Based on the Prediction of Bulk Ice Particle Properties. Part I: Scheme Description and Idealized Tests , 2015 .

[60]  David John Gagne,et al.  Coupling Data Science Techniques and Numerical Weather Prediction Models for High-Impact Weather Prediction , 2016 .

[61]  Rebecca D. Adams-Selin,et al.  Forecasting Hail Using a One-Dimensional Hail Growth Model within WRF , 2016 .

[62]  Matthias Steiner,et al.  Probabilistic forecasts of mesoscale convective system initiation using the random forest data mining technique , 2016 .

[63]  Craig S. Schwartz,et al.  Severe Weather Prediction Using Storm Surrogates from an Ensemble Forecasting System , 2016 .

[64]  Matthew R. Kumjian,et al.  The Impact of Vertical Wind Shear on Hail Growth in Simulated Supercells , 2017 .

[65]  S. E. Haupt,et al.  Using Artificial Intelligence to Improve Real-Time Decision-Making for High-Impact Weather , 2017 .