Identification, extraction and interpretation of multi-period variations of coastal suspended sediment concentration based on unevenly spaced observations

Abstract

Suspended sediment concentration (SSC) is an important indicator for coastal morphodynamics and health, exhibiting significant periodic variations at multiple timescales in response to various cyclic forcing factors. Identification of key periods and the extraction of corresponding amplitudes and phases can greatly aid effort to characterize the representative portrayal of SSC, which contributes to further physical interpretation. Remote sensing missions are intended to satisfy the retrieval of long-term SSC variations. However, the unevenly spaced characteristic and low effective sampling rates of the retrieved time series would hamper the extraction and interpretation of the feature changes. Here, we retrieved a 9-year SSC time series in a coastal tidal basin of Jiangsu coast, China, based on hourly GOCI satellite images from 2011 to 2019, and applied the Lomb-Scargle periodogram with the phase-folded method to extract their periodic information. On the basis of the unevenly spaced data with effective rate of 6.9%, the semidiurnal, diurnal, fortnightly, and seasonal cycles were successfully identified as the main periods, with amplitudes of 33.1, 23.4, 54.1, and 71.4 mg/L, respectively. In accordance with investigations into phase lags between SSC and forcing factors, we found that the unimodal SSC pattern at the semidiurnal timescale was induced by flood dominance, while the diurnal inequality of water level and wave intensity accounts for the SSC variation at the diurnal timescale. Tidal ranges vary consistently with the SSC in the fortnightly scale while background wave dominates the SSC variation in the seasonal timescale, the latter resulting in the seasonal difference in semidiurnal and fortnightly SSC cycles. The annual semidiurnal SSC variations experience a regime shift from 2014 to 2016, which is in consistency with the sequential reclamation. The research presents a typical application of long-term unevenly spaced observations into coastal SSC, and the proposed methodology can be extended to similar studies aiming at uncovering the mechanism of SSC variations at multiple timescales.