45 WS Electric Field Mill Lightning Prediction Threshold Analysis

Abstract

Electric field mills at Cape Canaveral continuously record data from 31 separate EFM sites 24 hours a day at a rate of 50 Hz. This produces 4,320,000 lines of recorded data daily for each EFM site, a total of more than 16 billion data points annually for the active thunderstorm season. This study seeks to determine a single electric field mill reading threshold for lightning onset and a separate single EFM reading threshold for lightning cessation. Statistical analysis of the EFM and Lightning Detection and Ranging (LDAR) parameters show there is no measurable correlation between EFM readings and lightning activity. Further, attempts to build models using threshold analysis, standard least squares regression fitting, nominal logistic regression fitting, and negative binomial regression fitting are unable to accurately predict any meaningful amount of lightning activity. The best of these models can only account for 16 percent of the variance in the dataset. Overall results show EFM readings do not correlate well with lightning activity and any attempts to predict lightning proved ineffective.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 2020
Accession Number
AD1101495

Entities

People

  • Charles A. Skrovan

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Sensors

DTIC Thesaurus Topics

  • Air Force
  • Computer Programming
  • Correlation Analysis
  • Data Mining
  • Data Science
  • Databases
  • Detection
  • Electric Fields
  • Information Science
  • Lightning
  • Measurement
  • Neural Networks
  • Predictive Modeling
  • Regression Analysis
  • Standards
  • Statistical Analysis
  • Warning Systems

Readers

  • Atmospheric Science/Meteorology
  • Computational Modeling and Simulation
  • Mathematics or Statistics