A machine‐learning approach to thunderstorm forecasting through post‐processing of simulation data

Thunderstorms pose a major hazard to society and the economy, which calls for reliable thunderstorm forecasts. In this work, we introduce SALAMA, a feedforward neural network model for identifying thunderstorm occurrence in numerical weather prediction (NWP) data. The model is trained on convection‐resolving ensemble forecasts over central Europe and lightning observations. Given only a set of pixel‐wise input parameters that are extracted from NWP data and related to thunderstorm development, SALAMA infers the probability of thunderstorm occurrence in a reliably calibrated manner. For lead times up to 11 h, we find a forecast skill superior to classification based only on NWP reflectivity. Varying the spatiotemporal criteria by which we associate lightning observations with NWP data, we show that the time‐scale for skillful thunderstorm predictions increases linearly with the spatial scale of the forecast.

• Publication year

  2023

• Venue

  Quarterly Journal of the Royal Meteorological Society

• Publication date

  2023-03-15

• Fields of study

  Physics, Computer Science, Environmental Science

• Identifiers
  DOI 10.1002/qj.4777arXiv 2303.08736
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Quantitative Precipitation Forecast Experiment Based on Basic NWP Variables Using Deep Learning

  2022cited by this paper
• A deep learning framework for lightning forecasting with multi‐source spatiotemporal data

  2021influential reference
• Cb‑Fusion – forecasting thunderstorm cells up to 6 hours

  2021cited by this paper
• Convection indicator for pre-tactical air traffic flow management using neural networks

  2021cited by this paper
• Nowcasting thunderstorm hazards using machine learning: the impact of data sources on performance

  2021cited by this paper
• Calibration of Machine Learning–Based Probabilistic Hail Predictions for Operational Forecasting

  2020cited by this paper
• Characteristics of deep moist convection over Germany in multi-source data

  2020cited by this paper
• A Comparison of Neural-Network and Surrogate-Severe Probabilistic Convective Hazard Guidance Derived from a Convection-Allowing Model

  2020cited by this paper
• Generating Probabilistic Next-Day Severe Weather Forecasts from Convection-Allowing Ensembles Using Random Forests

  2020cited by this paper
• A deep‐learning model to predict thunderstorms within 400 km2 South Texas domains

  2020cited by this paper
• Frequency of severe thunderstorms across Europe expected to increase in the 21st century due to rising instability

  2019cited by this paper
• Bad weather and flight delays: The impact of sudden and slow onset weather events

  2019cited by this paper
• Evaluation of Machine Learning Classifiers for Predicting Deep Convection

  2019influential reference
• Money Doesn't Grow on Trees, but Forecasts Do: Forecasting Extreme Precipitation with Random Forests

  2018cited by this paper
• Scientific challenges of convective-scale numerical weather prediction

  2017cited by this paper
• Storm-Based Probabilistic Hail Forecasting with Machine Learning Applied to Convection-Allowing Ensembles

  2017cited by this paper
• Lightning as a major driver of recent large fire years in North American boreal forests

  2017cited by this paper
• A Summary of Recent National-Scale Lightning Fatality Studies

  2016cited by this paper
• Upscale Error Growth in a High-Resolution Simulation of a Summertime Weather Event over Europe*

  2015cited by this paper
• The quiet revolution of numerical weather prediction

  2015cited by this paper
• Improved Nowcasts by Blending Extrapolation and Model Forecasts

  2015cited by this paper
• The ICON (ICOsahedral Non‐hydrostatic) modelling framework of DWD and MPI‐M: Description of the non‐hydrostatic dynamical core

  2015cited by this paper
• Use of NWP for Nowcasting Convective Precipitation: Recent Progress and Challenges

  2014cited by this paper
• Adam: A Method for Stochastic Optimization

  2014cited by this paper
• Some aspects of global lightning impacts

  2014cited by this paper
• Robust increases in severe thunderstorm environments in response to greenhouse forcing

  2013cited by this paper
• Classification of lightning damage to wind turbine blades

  2012cited by this paper
• Mitigating the Impact of Adverse Weather on Aviation

  2012cited by this paper
• Objective Prediction of Warm Season Afternoon Thunderstorms in Northern Taiwan Using a Fuzzy Logic Approach

  2012cited by this paper
• Blending a probabilistic nowcasting method with a high‐resolution numerical weather prediction ensemble for convective precipitation forecasts

  2012cited by this paper
• Mesoscale Meteorology in Midlatitudes

  2010cited by this paper
• LINET—An international lightning detection network in Europe

  2009influential reference
• Classification of Imbalanced Data: a Review

  2009cited by this paper
• Assessing the spatial and temporal variation in the skill of precipitation forecasts from an NWP model

  2008cited by this paper
• Increasing the Reliability of Reliability Diagrams

  2007cited by this paper
• Statistical Methods in the Atmospheric Sciences

  2007cited by this paper
• Scoring Probabilistic Forecasts: The Importance of Being Proper

  2007cited by this paper
• Predicting good probabilities with supervised learning

  2005cited by this paper
• Predictability of Precipitation from Continental Radar Images. Part III: Operational Nowcasting Implementation (MAPLE)

  2004cited by this paper
• NCAR Auto-Nowcast System

  2003cited by this paper
• Nowcasting Thunderstorms: A Status Report

  1998cited by this paper
• TITAN: Thunderstorm Identification, Tracking, Analysis, and Nowcasting—A Radar-based Methodology

  1993influential reference
• A New Vector Partition of the Probability Score

  1973influential reference
• The predictability of a flow which possesses many scales of motion

  1969cited by this paper

• An AI Training Dataset for Thunderstorm Monitoring and Forecasting over China

  2026cites this paper
• Thunderstorm Forecasting by Using Machine Learning Techniques: A Comparative Model Analysis Leveraging Historic Climatic Records of Bangladesh

  2025cites this paper
• Exploring Deep Learning and Machine Learning Models for HailStorm Detection

  2025cites this paper
• Application of AI in Lightning and Thunderstorm Forecasting: A Vision for the Future

  2025cites this paper
• Machine Learning-Based Identification of Key Predictors for Lightning Events in the Third Pole Region

  2025influential citation