Topological Insights Into Weather Dynamics in the Indian Context

Abstract

Analysis of day-to-day weather patterns is critical and essential in daily life. Although traditional methods exist, in modern times, we have developed realistic and reliable methods to provide better insights and understanding of complex weather patterns for various surges, especially in these times of global warming. Implementing clustering and topological data analysis in this analysis has looked into a vast understanding of how regions with similar characteristics behave when weather changes occur due to heat, pressure, or wind-related phenomena. The classification model developed using Mapper analysis has produced 95.8% accuracy, concluding that weather follows a transient weather pattern due to various resources and how stagnant conditions affect transient weather patterns, causing rise in sub-clusters. Thus, fitting and interpreting newer models helps us understand weather analysis and classification.