Due to the major role played by diatoms in the biological pump of CO2, and to the presence of silica-rich sediments in areas that play a major role in air-sea CO2 exchange (e.g. the Southern Ocean and the Equatorial Pacific), opal has a strong potential as a proxy for paleoproductivity reconstructions. However, because of spatial variations in the biogenic silica preservation, and in the degree of coupling between the marine Si and C biogeochemical cycles, paleoreconstructions are not straitghtforward. A better calibration of this proxy in the modern ocean is required, which needs a good understanding of the mechanisms that control the Si cycle, in close relation to the carbon cycle.This review of the Si cycle in the modern ocean starts with the mechanisms that control the uptake of silicic acid (Si(OH)4) by diatoms and the subsequent silicification processes, the regulatory mechanisms of which are uncoupled. This has strong implications for the direct measurement in the field of the kinetics of Si(OH)4 uptake and diatom growth. It also strongly influences the Si:C ratio within diatoms, clearly linked to environmental conditions. Diatoms tend to dominate new production at marine ergoclines. At depth, they also succeed to form mats, which sedimentation is at the origin of laminated sediments and marine sapropels. The concentration of Si(OH)4 with respect to other macronutrients exerts a major influence on diatom dominance and on the rain ratio between siliceous and calcareous material, which severely impacts surface waters pCO2. A compilation of biogenic fluxes collected at about 40 sites by means of sediment traps also shows a remarkable pattern of increasing BSi:Corg ratio along the path of the "conveyor belt", accompanying the relative enrichment of waters in Si compared to N and P. This observation suggests an extension of the Si pump model described by Dugdale and Wilkerson (1989, doi:10.1038/34630), giving to Si(OH)4 a major role in the control of the rain ratio, which is of major importance in the global carbon cycle.The fate of the BSi produced in surface waters is then described, in relation to Corg, in terms of both dissolution and preservation mechanisms. Difficulties in quantifying the dissolution of biogenic silica in the water column as well as the sinking rates and forms of BSi to the deep, provide evidence for a major gap in our understanding of the mechanisms controlling the competition between retention in and export from surface waters. The relative influences of environmental conditions, seasonality, food web structure or aggregation are however explored. Quantitatively, assuming steady state, the measurements of the opal rain rate by means of sediment traps matches reasonably well those obtained by adding the recycling and burial fluxes in the underlying abyssal sediments, for most of the sites where such a comparison is possible. The major exception is the Southern Ocean where sediment focusing precludes the closing of mass balances. Focusing in fact is also an important aspect of the downward revision of the importance of Southern Ocean sediments in the global biogenic silica accumulation. Qualitatively, little is known about the duration of the transfer through the deep and the quality of the material that reaches the seabed, which is suggested to represent a major gap in our understanding of the processes governing the early diagenesis of BSi in sediments. The sediment composition (special emphasis on Al availability), the sedimentation rate or bioturbation are shown to exert an important control on the competition between dissolution and preservation of BSi in sediments. It is suggested that a primary control on the kinetic and thermodynamic properties of BSi dissolution, both in coastal and abyssal sediments, is exerted by water column processes, either occuring in surface waters during the formation of the frustules, or linked to the transfer of the particles through the water column, which duration may influence the quality of the biogenic rain. This highlights the importance of studying the factors controlling the degree of coupling between pelagic and benthic processes in various regions of the world ocean, and its consequences, not only in terms of benthic biology but also for the constitution of the sediment archive.The last section, first calls for the end of the "NPZD" models, and for the introduction of processes linked to the Si cycle, into models describing the phytoplankton cycles in surface waters and the early diagenesis of BSi in sediments. It also calls for the creation of an integrated 1-D diagnostic model of the Si:C coupling, for a better understanding of the interactions between surface waters, deep waters and the upper sedimentary column. The importance of Si(OH)4 in the control of the rain ratio and the improved parametrization of the Si cycle in the 1-D diagnostic models should lead to a reasonable incorporation of the Si cycle into 3-D regional circulation models and OGCMs, with important implications for climate change studies and paleoreconstructions at regional and global scale.

Supplement to: Ragueneau, Olivier; Tréguer, Paul; Leynaert, Aude; Anderson, Robert F; Brzezinski, Mark A; DeMaster, David J; Dugdale, Richard; Dymond, Jack R; Fischer, Gerhard; Francois, Roger; Heinze, Christoph; Maier-Reimer, Ernst; Martin-Jézéquel, Véronique; Nelson, David M; Quéguiner, Bernard (2000): A review of the Si cycle in the modern ocean: recent progress and missing gaps in the application of biogenic opal as a paleoproductivity proxy. Global and Planetary Change, 26(4), 317-365