Revisiting the biogenic silica burial flux determinations: A case study for the East China seas

Abstract

The Coastal and Continental Margin Zones (CCMZs) contribute to 40% of the total burial flux of biogenic silica (bSi) of the world ocean. However, the accurate determination of the bSi content (bSiO2%) in marine sediments remains a challenge. The alkaline methods commonly used to quantitatively determine bSiO2% can completely digest the amorphous silica of diatoms but are less effective at digesting radiolarians and sponge spicules. In addition, the lithogenic silica (lSi) found in sediments is partly digested during these alkaline extractions, which can bias the accuracy of the determined bSiO2%. This is of importance in CCMZs where the lSi:bSi ratio is high. In this study, we examined sediments collected in the CCMZs of East China seas, an environment of peculiar interest given the large amount of lSi deposited by the Yellow River and the Yangtze River. The results show that alkaline digestions using stronger solutions and pretreatment steps resulted in an overestimate of the bSiO2% due to increased leaching of silica mainly from authigenic silicates and clays, whereas weak digestions underestimated the bSiO2% owing to incomplete digestion of sponge spicules. We found that the use of the Si/Al method accurately corrects for the lSi fraction in marine sediments, and thereby improves the determinations of bSiO2% in the sediments of East China seas CCMZs. Ensuring full digestion of all bSi remains challenging, in particular for sponge spicules, motivating both verifications via microscopy and longer extraction times. To emphasize the influence of these methodological differences, we revised the bSi burial flux in the East China seas and provide a new estimate of 253 (± 286) Gmol-SiO2 yr-1, which is one third of the previous estimates. We discuss the potential contribution from radiolarian and sponges and we propose a new general protocol for the determination of bSi in sediments that minimizes the methodological bias in bSi determination.