Observing ENSO-modulated tides from space

Abstract

The understanding of sea level variability is fundamental for coastal communities with high population density. As important parts of sea levels, tides are often overlooked because they tend to be regarded as simple periodic oscillations by numerous studies. In fact, due to the interference of various non-astronomical factors, observed tides are non-stationary and display multi-time scale variability. Here, we focus on inter-annual tidal variations induced by El Niño-Southern Oscillation (ENSO) in the western Pacific. Although satellite altimeters have been widely used in tidal research, to date, inter-annual tidal changes from satellite altimeters are never reported. Classical harmonic analysis (CHA) model with moving windows is inappropriate for satellite altimeter records due to tidal aliasing originated from long-period sampling intervals. In this paper, we successfully quantify the influences of ENSO on tides from multi-satellite altimeters through a revised harmonic analysis (RHA) model which directly builds ENSO forcing into the basic functions of CHA. To eliminate mathematical artifacts caused by over-fitting, Lasso regularization is applied in the RHA model to replace widely-used ordinary least squares. The performances of RHA are generally better than those of CHA as a result of considering the impacts of ENSO. The combination of RHA and satellite altimeters extends the studies on tidal evolution using spatially limited tide gauges. Compared to diurnal tides, it is found that semi-diurnal tides are more vulnerable to ENSO in the western Pacific. Of note, both linear and non-linear tidal responses to ENSO show significant spatial heterogeneity. Moreover, as a universal method, ENSO index used in this study can be replaced by other factors such as river flow and sea waves to explore tidal evolution driven by other physical processes.