Phase variations of the summer and winter seasons in the Bohai Sea during the last four decades

Abstract

In most coastal oceans, the impacts of global warming on season duration and timing of seasonal transitions remain unknown. To mirror the reality of the ongoing climate change, the summer and winter seasons are redefined using the local water temperature thresholds in the Bohai Sea. Then the phase variations of these seasons are quantified using the duration and transition timing indices, including the duration (DUR), onset (ONS), and withdrawal (WIT) indices derived from the OSTIA SST dataset at a very high resolution (0.05°). During the last four decades (1982–2019), secular trends of summer indices extracted by the ensemble empirical mode decomposition (EEMD) method reveal that the summer DUR has an accumulated increase of about 17 days (4.5 days decade-1), which is primarily induced by the phase advance of the summer ONS by about 16 days (4.2 days decade-1). Spatial features of the duration and timing indices demonstrate that the lengthening of summer DUR and the phase advance of summer ONS have significantly enhanced in the shallow regions, due to the limited thermal inertia and the shorter period of the ocean’s memory. In contrast, the secular trend of winter DUR exhibits an accumulated shortening of about 18 days (4.8 days decade-1), which is induced by a moderately delayed winter ONS of 6 days (1.6 days decade-1) and a significantly advanced winter WIT of 12 days (3.2 days decade-1). The potential linkage between the phase variations in the oceanic seasonal cycle and those of the atmospheric forcing was investigated by analyzing both the interannual variability and the secular trend. Over the analysis period, the secular trend of an earlier summer ONS is related to a total reduction of cloud cover by 30% of its climatological mean and an increase of incoming solar radiation of 10 W m-2 month-1 in the late spring. Thus, our results highlight the influence of cloud cover in addition to wind speed on the temporal variations of season transition timing.