ISO 9001:2015

INTERNATIONAL JOURNAL OF GLOBAL RESEARCH INNOVATIONS & TECHNOLOGY (IJGRIT) [ Vol. 4 | No. 2 | April - June, 2026 ]

Study the Occurrence of Lightning and Thunderstorm during February to May, with All Environmental Influencing Factors by Neural Network and Machine Learning Technique

Sumana Chatterjee

This paper is based on the study of the occurrence of lightning and or thunderstorm, the weather phenomena, during February to May, analysis on google collaborator platform. The environmental parameters used were ‘convective available potential energy(CAPE)’, ‘convective inhibition (CIN)’, ‘convective precipitation (CP)’, ‘2 meter dew point temperature, (d2m), ‘mean sea level pressure (MSL)’, ‘relative humidity at 500 HPA pressure level ’, ‘relative humidity at 850 HPA pressure level’, ‘earth’s surface skin layer temperature (skin)’, ‘earth surface pressure (sp)’, ‘sea surface temperature (sst)’, ‘air temperature (t at 500 HPA)’, ‘air temperature (t at 850 HPA)’, ‘air temperature measured at 2 meters above the surface (t2m)’, ‘ total cloud cover (tcc)’, ‘ total precipitation (tp)’, ‘u, horizontal wind at 500 HPA pressure level’, ‘u, horizontal wind at 800 HPA pressure level’, ‘v, vertical wind at 500 HPA pressure level’, ‘v, vertical wind at 800 HPA pressure level’, ‘w, speed of air movement at 500 HPA pressure level’, ‘w, speed of air movement at 800 HPA pressure level’, data source was ‘Copernicus Climate Data Store (ERA5 dataset )’ with sub region of data adequate for Alipore. Date wise ‘Copernicus Hub’ daily data merged with date wise daily surface data of Alipore to have insights of occurrence of lightning or thunderstorm during February to May as well as trend for near future. The multivariate analysis, with historical large data from 1969 to 2026, with associated variables, factors influencing lightning and thunderstorm, i.e. predictors from ERA5 data set, along with target data, i,e. surface data set of Alipore with actual data of lightning or thunderstorm occurrence, was subjected to parallel study with LSTM neural network and machine learning to study the occurrence of such phenomena, during the scheduled months.

  1. J. M., 2024. An explainable machine learning technique to forecast lightning density over North-Eastern India. Journal of Atmospheric and Solar Terrestrial Physics, 259(June),p.106255.
  2. Raheem, B. D., 2023. Techniques for lightning prediction: A review. Ukrainian Journal of Educational Studies and Information Technology, 11(4), pp. 227-241.
  3. Saha, K., 2025. Prediction of lightning events over Bangladesh: A machine learning perspective. Journal of Atmospheric and Solar Terrestrial Physics, 268(March), p. 106448.
  4. Song, G., 2023. Nowcasting lightning occurrence from commonly available. Nature, August, 6(126), pp. 1-10.
  5. Wang, X., 2023. A Survey of Deep Learning-Based Lightning Prediction. MDPI, 14(1698), pp. 1-17.

DOI:

Article DOI:

DOI URL:


Download Full Paper:

Download