Signal Processing and Analysis Methodology Applied to Climatological Anomalies

This paper addresses the processing and anomaly analysis of climatological data using filters and in time and frequency domain correlation. Nowadays, various organizations need to know the future behavior of certain phenomena in order to plan to prevent disasters. The E1 Nino-Southern Oscillation (ENSO) is a climatic phenomenon that has relationship with precipitation and temperature variability worldwide and is linked to droughts and flooding in some regions. In this work, Sea Surface Temperature (SST) in Equatorial Pacific Ocean and precipitation in the Santa River Basin are signals of climate origin which are processed and analyzed. The methodology includes Oceanic Nino Index (ONI) and precipitation data from Tropical Rainfall Measurement Missions (TRMM) satellite. The precipitation is filtered and denoised using Fourier Transform and Savitzky-Golay filter providing stationarity and anomaly of precipitation. The results demonstrate that the cutoff frequencies ω =0 and 1 represent the annual behavior of precipitation in Santa River Basin. In addition, Wavelet Analysis shows similar power patterns between Sea Surface Anomalies and precipitation with high correlation of around 3 and 6 years.