Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Machine Learning for Real-time Prediction of Damaging Straight-line Convective Wind

Machine Learning for Real-time Prediction of Damaging Straight-line Convective Wind AbstractThunderstorms in the U.S. cause over 100 deaths and $10 billion of damage per year, much of which is attributable to straight-line (non-tornadic) wind. This paper describes a machine-learning system that forecasts the probability of damaging straight-line wind (≥ 50 kt or 25.7 m s−1) for each storm cell in the continental U.S., at distances up to 10 km outside the storm cell and lead times up to 90 min. Predictors are based on radar scans of the storm cell, storm motion, storm shape, and soundings of the near-storm environment. Verification data come from weather stations and quality-controlled storm reports. The system performs very well on independent testing data. The area under the ROC curve ranges from 0.88 to 0.95; critical success index (CSI) ranges from 0.27 to 0.91; and Brier skill score (BSS) ranges from 0.19 to 0.65 (> 0 is better than climatology). For all three scores, the best value occurs for the smallest distance (inside storm cell) and/or lead time (0-15 min), while the worst value occurs for the greatest distance (5-10 km outside storm cell) and/or lead time (60-90 min). The system was deployed in the 2017 Hazardous Weather Testbed. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Weather and Forecasting American Meteorological Society

Machine Learning for Real-time Prediction of Damaging Straight-line Convective Wind

Lagerquist, Ryan; McGovern, Amy; Smith, Travis
Weather and Forecasting , Volume preprint (2017): 1 – Nov 21, 2017
Download PDF
53 pages

Loading next page...
1
 
/lp/ams/machine-learning-for-real-time-prediction-of-damaging-straight-line-stLTVZcVa1

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
American Meteorological Society
Copyright
Copyright © American Meteorological Society
ISSN
1520-0434
DOI
10.1175/WAF-D-17-0038.1
Publisher site
See Article on Publisher Site

Abstract

AbstractThunderstorms in the U.S. cause over 100 deaths and $10 billion of damage per year, much of which is attributable to straight-line (non-tornadic) wind. This paper describes a machine-learning system that forecasts the probability of damaging straight-line wind (≥ 50 kt or 25.7 m s−1) for each storm cell in the continental U.S., at distances up to 10 km outside the storm cell and lead times up to 90 min. Predictors are based on radar scans of the storm cell, storm motion, storm shape, and soundings of the near-storm environment. Verification data come from weather stations and quality-controlled storm reports. The system performs very well on independent testing data. The area under the ROC curve ranges from 0.88 to 0.95; critical success index (CSI) ranges from 0.27 to 0.91; and Brier skill score (BSS) ranges from 0.19 to 0.65 (> 0 is better than climatology). For all three scores, the best value occurs for the smallest distance (inside storm cell) and/or lead time (0-15 min), while the worst value occurs for the greatest distance (5-10 km outside storm cell) and/or lead time (60-90 min). The system was deployed in the 2017 Hazardous Weather Testbed.

Journal

Weather and Forecasting – American Meteorological Society

Published: Nov 21, 2017

References

  • Fatalities associated with nonconvective high-wind events in the United States
    Ashley, W.; Black, A.
  • An hourly assimilation–forecast cycle: The RUC
    Benjamin, S.
  • SHARPpy: An open source sounding analysis toolkit for the atmospheric sciences
    Blumberg, W.; Halbert, K.; Supinie, T.; Marsh, P.; Thompson, R.; Hart, J.
  • Random forests
    Breiman, L.
  • An empirical model for assessing the severe weather potential of developing convection
    Cintineo, J.; Pavolonis, M.; Sieglaff, J.; Lindsey, D.
  • An overview of the 2010 Hazardous Weather Testbed Experimental Forecast Program Spring Experiment
    Clark, A.
  • An observational study of derecho-producing convective systems
    Coniglio, M.; Stensrud, D.; Richman, M.
  • A proposed revision to the definition of “derecho.”
    Corfidi, S.; Coniglio, M.; Cohen, A.; Mead, C.
  • Machine intelligent gust front detection
    Delanoy, R.; Troxel, S.
  • Climatological estimates of daily local nontornadic severe thunderstorm probability for the United States
    Doswell, C. A.; Brooks, H.; Kay, M.
  • Examination of derecho environments using proximity soundings
    Evans, J.; Doswell, C. A.
  • Greedy function approximation: A gradient boosting machine
    Friedman, J.
  • Downbursts: Meteorological features and wind field characteristics
    Fujita, T. T.
  • The attributes diagram: A geometrical framework for assessing the quality of probability forecasts
    Hsu, W.; Murphy, A.
  • The WSR-88D severe weather potential algorithm
    Kitzmiller, D.; McGovern, W.; Saffle, R.
  • A real-time, three-dimensional, rapidly updating, heterogeneous radar merger technique for reflectivity, velocity, and derived products
    Lakshmanan, V.; Smith, T.; Hondl, K.; Stumpf, G.; Witt, A.
  • The Warning Decision Support System–Integrated Information
    Lakshmanan, V.; Smith, T.; Stumpf, G.; Hondl, K.
  • An efficient, general-purpose technique for identifying storm cells in geospatial images
    Lakshmanan, V.; Hondl, K.; Rabin, R.
  • A method for extracting postevent storm tracks
    Lakshmanan, V.; Herzog, B.; Kingfield, D.
  • Deep learning
    LeCun, Y.; Bengio, Y.; Hinton, G.
  • Use of receiver operating characteristic curves to assess the performance of a microdilution assay for determination of drug susceptibility of clinical isolates of Mycobacterium tuberculosis
    Luna-Herrera, J.; Martinez-Cabrera, G.; Parra-Maldonado, R.; Enciso-Moreno, J.; Torres-Lopez, J.; Quesada-Pascual, F.; Delgadillo-Polanco, R.; Franzblau, S.
  • Challenges for data assimilation—From convective-scale to climate
    Marriott, R.
  • A neural network for damaging wind prediction
    Marzban, C.; Stumpf, G.
  • Solar energy prediction: An international contest to initiate interdisciplinary research on compelling meteorological problems
    McGovern, A.; Gagne, D.; Basara, J.; Hamill, T.; Margolin, D.
  • Using artificial intelligence to improve real-time decision making for high-impact weather
    McGovern, A.; Elmore, K.; Gagne, D.; Haupt, S.; Karstens, C.; Lagerquist, R.; Smith, T.; Williams, J.
  • Statewide monitoring of the mesoscale environment: A technical update on the Oklahoma Mesonet
    McPherson, R.
  • North American Regional Reanalysis
    Mesinger, F.
  • Basic principles of ROC analysis
    Metz, C.
  • Can routine laboratory tests discriminate between severe acute respiratory syndrome and other causes of community-acquired pneumonia?
    Muller, M.; Tomlinson, G.; Marrie, T.; Tang, P.; McGeer, A.; Low, D.; Detsky, A.; Gold, W.
  • Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods
    Platt, J.
  • Visualizing multiple measures of forecast quality
    Roebber, P.
  • Downbursts
    Rose, M.
  • Multi-Radar Multi-Sensor (MRMS) severe weather and aviation products: Initial operating capabilities
    Smith, T.
  • Use of NWP for nowcasting convective precipitation: Recent progress and challenges
    Sun, J.
  • Regression shrinkage and selection via the lasso
    Tibshirani, R.
  • Estimation of the probability of an event as a function of several independent variables
    Walker, S.; Duncan, D.
  • Experiences with 0–36-h explicit convective forecasts with the WRF-ARW model
    Weisman, M.; David, C.; Wang, W.; Manning, K.; Klemp, J.
  • Regularization and variable selection via the elastic net
    Zou, H.; Hastie, T.