Using Empirical Mode Decomposition to Identify Multi-Decadal Climate Variability and its Contribution to Sea Level Acceleration
College of Sciences
Climate variability has an effect on sea level on a global scale. There are decadal impacts of sea level with the climate events El Nino Southern Oscillation (ENSO) and Pacific Decadal Oscillation (PDO). The Pacific is effected by ESNO and PDO and they have been shown to impact climate and sea level. There also appears to be multi decadal periods seen within tide gauges. With a 113-year reconstruction of sea level, we examine decadal to multi-decadal variability seen in the sea level fluctuations in the Global Oceans. We use empirical mode decomposition (EMD) as a low pass filter. We break down regional sea level into a series of intrinsic mode functions (IMFs) and attempt attribution of these IMFs to specific climate modes of variability. We applied Empirical Orthogonal Function analysis (EOF) to the low filtered time series. On a basic level, the resulting EOF Loading Vectors (LVs; i.e. spatial maps) exhibit large-scale patterns, focused largely in the Pacific Ocean. In the first and second EOF, there is a multi-decadal signal that bearing similarity to ENSO. The first two principal component time series (PCTS) exhibits a signal with a period of ~60-106 years, while the third and fourth EOF displays a shorter period near ~53 years and exhibits a regional pattern similar to PDO. In each of the four EOFs, the variability appears to be dominated by signals centered in the Pacific Ocean. By identifying and separating these periodic climate signals, we can gain a better understanding of how the sea level variability associated with these modes can impact sea level on shorter timescales and serve to exacerbate the effects of long-term sea level change.
Sommers, Lauren; Hamlington, Ben; Ezer, Tal; and Clayton, Sophie, "Using Empirical Mode Decomposition to Identify Multi-Decadal Climate Variability and its Contribution to Sea Level Acceleration" (2019). College of Sciences Posters. 26.