The Impact of COVID-19 and the Russo-Ukraine War on Natural Gas Flow Through Time Series Forecasting

The natural gas market has experienced unprecedented volatility due to extreme events such as the COVID-19 pandemic and the Russia-Ukraine war. In Europe, natural gas is a major contributor to CO2 emissions, making its efficient transmission essential. This study examines the impact of these disruptions on natural gas distribution. We employ Long Short-Term Memory (LSTM) neural networks to analyze and forecast gas flow, assessing prediction errors. Our findings attribute the observed variation in prediction errors primarily to the COVID-19 pandemic and the Russia-Ukraine war. We also identify a positive correlation between maximum daily flow and prediction error. Furthermore, our analysis shows that municipal nodes were disproportionately affected by lockdown measures, while inflow nodes experienced the greatest impact from the invasion.