Prediction of Lightning Occurrence in South Africa Using Neural Network and Meteorological Data

12 Pages Posted: 23 Jan 2023 Last revised: 13 Apr 2023

See all articles by Sizwe Mzila

Sizwe Mzila

University of the Witwatersrand

Hugh Hunt

University of the Witwatersrand

Ritesh Ajoodha

University of the Witwatersrand

Date Written: November 23, 2022

Abstract

Many advances are being made in the prediction of lightning across the world. Researchers have investigated many different ways to predict lightning, from statistical techniques to the traditional machine learning techniques and most recently the use of deep learning techniques. The data used for these predictions range from the use of numerical weather prediction (NWP) models and parameterization schemes to using satellite images or the commonly available meteorological features which are easier to obtain and make for computationally less expensive models. South Africa is one of the leading African countries in lightning research and lightning detection, but very few papers have focused on using the available lightning detection networks to develop lightning prediction models to manage the risk of lightning in the country especially in rural areas which are prone to such incidents due to the lack of protection and efficient warning systems. In this paper, we use deep neural networks and meteorological data from the SAWS to predict the occurrence of lightning.

Keywords: Music genre classification, Q-learning, recommendation, Content-based Features, Machine Learning

Suggested Citation

Mzila, Sizwe and Hunt, Hugh and Ajoodha, Ritesh, Prediction of Lightning Occurrence in South Africa Using Neural Network and Meteorological Data (November 23, 2022). Available at SSRN: https://ssrn.com/abstract=4331675 or http://dx.doi.org/10.2139/ssrn.4331675

Sizwe Mzila (Contact Author)

University of the Witwatersrand

Hugh Hunt

University of the Witwatersrand

Ritesh Ajoodha

University of the Witwatersrand ( email )

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
66
Abstract Views
261
Rank
637,474
PlumX Metrics