Spreading of water masses and regeneration of silica and ²²⁶Ra in the Indian Ocean

Abstract

The magnitudes of silica and ²²⁶Ra inputs to water (through particle regeneration, in situ, and from sediments) and the validity of observed Si and ²²⁶Ra as tracers of water masses and advective processes were examined in the Indian Ocean using the GEOSECS data. The regenerated quantities of these two parameters were calculated as the difference between the observed and the expected concentrations; the latter were estimated from a three end-member mixing model employing potential temperature and salinity as conservative tracers. Here we present results on the quantitative spreading of the Antarctic Bottom Water (AABW); the Modified North Atlantic Deep Water (MNADW, also known as the Circumpolar Water) and the North Indian Deep Water (NIDW)-both these were represented together as High Salinity Deep Waters (HSDW); the Antarctic Intermediate Water (AAIW); the North Indian Intermediate Water (NIIW) and the Central Indian Water (CIW). Our results concur with recent results in the literature. Briefly, the northward flow of the AABW is uneven; the MNADW core layer is found to be closer to the Antarctic that spreads to the north, and AAIW is largely restricted to the Indian Ocean south of 10°S. Our results also reveal that: roughly 10% more AABW enters the Bay of Bengal than the Arabian Sea; there is greater possibility for deep waters to enter the Central Indian Basin from the Bay of Bengal; CIW occupies a larger part of the Bay of Bengal than of the Arabian Sea; and 10% of the NIIW reaches 30°S in the western Indian Ocean. The regenerations of Si and ²²⁶Ra are mainly from the underlying sediments rather than through the dissolution of particles in the water column. The sediments in the northern parts seem to supply ²²⁶Ra and Si to the rest of the Indian Ocean. At 10δS there is a subsurface ( ~600 m) maximum in regenerated Si, which is possibly connected to the advection of particles by Indonesian waters. The maxima in regenerated ²²⁶Ra and Si contribute about 50% and 30%, respectively, to the observed abundances, suggesting that the observed Si is a more useful tracer of water masses and mixing processes than ²²⁶Ra. Linear relationships were found between regenerated Si and ²²⁶Ra, but departures noticed for ²²⁶Ra in the eastern Indian Ocean may be attributed to its release from particles transported by the Indian rivers. Diverse regimes with respect to the extent of sources and dissolution of opal were noted in the Indian Ocean: high diatom abundance but low Si regeneration in the Antarctic, high diatom abundance and high regenerated Si in the Arabian Sea, and low diatom abundance but high regenerated Si in the Bay of Bengal.