Pan-regional characterization of the variability in the Indonesian Seas

Abstract

The Indonesian Seas feature a wide spectrum of variations in hydrography and circulation. This study applies a simple frequency-based time series decomposition method on the 20-year (2000–2019) model simulation, revealing the spatial distribution of the variations in intra-seasonal, semi-annual, annual, and inter-annual bands, respectively. K-means clustering based on the decomposed variability is used to explore dynamical similarities and differences, highlighting the competing impacts of annual and inter-annual variations. The Indonesian Seas lie at a “crossroad” where the inter-annual variation dominates in a NE-SW-oriented deep-ocean regime, while the annual variation prevails in a NW-SE-oriented marginal sea regime. The latter is primarily driven by the local monsoonal winds. Moreover, annual current profiles reveal significant differences between the main and east branches of the Indonesian Throughflow (ITF), with annual variations being more prominent above (below) the thermocline in the main (east) branch. Overall, ITF variability is influenced by remote inter-annual factors, regional monsoonal forces, and local flow instabilities, with the semi-annual variations being minor but non-trivial.