APPLICATION OF TIME SERIES MODEL FOR GLOBAL MEAN SEA LEVEL PREDICTION

Abstract

Global warming refers to the ongoing and long-term increase in the Earth's average surface temperature. It is primarily caused by human activities such as burning fossil fuels, deforestation, and releasing harmful gases into the atmosphere, like carbon dioxide and methane. Global warming has profound effects on oceans and seas worldwide. The rising sea levels result in various consequences on Earth and require urgent mitigation measures. This paper investigates how various factors influence monthly sea level rise and its temporal fluctuations. These factors include temperature, glacier melting rates, sea density, salinity, and carbon dioxide levels. The first part of the paper focuses on data collection, data wrangling, and exploratory data analysis (EDA), all centered around the common element - time. The second part of the paper concentrates on applying the Time Series model, specifically using the XGB Regressor, to predict precise global mean sea level changes. Using cleaned and analyzed data, the model can incorporate various factors affecting sea levels and forecast its changes based on historical patterns with a very low error rate.