Unveiling Indonesian Sea Surface Hidden Dynamics Using the FSLE Lagrangian Method

Due to their inherent turbulent nature, ocean motions and dynamics exhibit a great complexity, from the main patterns of the general circulation at large scale to circulation features at mesoscale and lower scales that makes the assessment of ocean evolution quite complicated. These difficulties are especially relevant when one tries to study problems of Lagrangian character, such as mixing, dispersion or stirring and transport of oceanic properties, or ecosystems dynamics. The broad context is based on the fact that fluid advection can effectively be studied by considering special material surfaces, which are also referred to as Lagrangian Coherent Structures (LCS). Notably, LCS are often locally the most strongly attracting or repelling material surfaces in the flow, and as such have a strong influence on the flow dynamics. In this study the dynamics of Indonesian seas is considered under the prism of one Lagrangian approach. Thus, it allows us to reconstruct the structure of transport dynamics that cannot be captured by the standard Eulerian approach. In this work, we present a study of the horizontal mixing properties in the Indonesian Seas using Finite Size Lyapunov Exponents (FSLE) analysis generated using numerical ocean model that simulated surface velocity fields. FSLE is a local measure of particle dispersion, determined at each grid point, which serves to characterize Lagrangian structures. Such method is widely used todays in physical oceanography but remains rarely applied in a closed area such as Indonesian seas. As the part of Maritime Continent, Indonesian seas have a complex topography. This complexity is due to the fact that Indonesia has thousands of islands with narrow straits and shallow sea regions. Roughly, Indonesia has shallow water region in its western part, and deeper water region in its eastern part. The regions such as Karimata Strait and Java Sea have around 200-meter depth. In the eastern region such as Banda Sea, the depth of the ocean can reach more than 4000-meter. In addition, Indonesian Seas is one unique choke point of the global ocean circulation that provides the only low-latitude linkage between the tropical Pacific and Indian Ocean. The water from the Pacific Ocean enters Indonesian regions through Makassar Strait and Banda Sea. This saline water, then, meets the fresh water (lower salinity) over the Java Sea and southern part of the Makassar Strait. A quite complex resulting mixing occurs and provides favorable conditions for LCS and stirring. Sensitivity of FSLE fields with respect to the final Relative Dispersion (RD) distance suggest that the FSLE features are optimally resolved when the distance is set at <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$0.3^{\circ}$</tex> value, which is determined for a time window sets to 180 days (instead of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$0.6^{\circ}$</tex> and 200 days used in the literature for ocean mesoscale dynamics). Comparisons are also made with classical FSLE estimated from geostrophic currents based on satellite altimetry (source from AVISO+), and with direct observations of trajectories of surface drifters from the GDP center. Sensitivity tests how that reducing the final separation distance with slight reduction of prescribed time has impact on increasing the magnitude of FSLE but not the structure and its time variability. In a second part, the role of barrier to material transport in fluids is specifically investigated with the FSLE in relation with the riverine discharge over the Java Sea, and more generally with the water cycle at regional scales. It is shown that the FSLE manifolds are able to outline the borders of the fresh water discharge, in particular in the Java Sea. Finally, we study the impact of different physics parametrization applied to the ocean circulation model to the distribution of passive particle advected by the flows. The standard physics parameter used fields such as river discharges, tide forcing and wave impacts on the currents, whereas the second velocity dataset used is based on the outputs of the model without tide and wave effects. At the first glance, the quantity of particles which are stranded (defined as the particles drifted within 9 meters from the borders of the model ocean-land mask) at the end of simulation is higher for the simulation using all physics parameters. This is confirmed by the fact that 81 % particles are stranded on shore for the simulation using the velocity field with complete physics, while if we use the velocity field without tide and without wave the percentage of stranded particles only reach 50 %. The particles are mostly stranded in regions which consists of smaller islands such as the Philippines region or the Nusa Tenggara archipelago. In the open ocean, the particles are driven toward the convergent regions due to eddies or strait dynamics that are depicted by the convergent FSLE lines. As the FSLE diagnostics can be used in detecting LCS, we are interested to use the FSLE to understand more about marine debris and sea pollution within the Indonesian Seas.