517 Journal of Oceanography, Vol. 56, pp. 517 to 525, 2000 Keywords: ⋅ Dissolved aluminum, ⋅ estuarine mixing, ⋅ flocculation, ⋅ adsorption, ⋅ authigenic mineral formation, ⋅ St. Lawrence Estuary. * Corresponding author. E-mail: kazufumi@nria.affrc.go.jp * Present address: National Research Institute of Aquaculture, 422-1 Nakatsuhamaura, Nansei, Mie 516-0193, Japan. Copyright © The Oceanographic Society of Japan. Dissolved Aluminum in the Upper St. Lawrence Estuary KAZUFUMI TAKAYANAGI* and CHARLES GOBEIL Institut Maurice-Lamontagne, Ministère des Pêches et des Océans, C.P. 1000, Mont-Joli, Québec G5H 3Z4, Canada (Received 5 October 1999; in revised form 26 February 2000; accepted 29 March 2000) The concentrations of dissolved aluminum (Al) in the upper St. Lawrence Estuary were determined during periods of high and intermediate river-discharge. Labora- tory experiments simulating estuarine processes were also conducted in order to ex- amine possible mechanisms controlling the Al distribution. During the high river- discharge, the Al concentration at river end-member was 1.63 μM and decreased exponentially with increasing salinity. An almost complete removal of dissolved Al was observed in the low salinity area up to 10 with an intensive removal in the turbid- ity maximum zone. Principal mechanisms responsible for the Al removal inferred from the laboratory experiments were flocculation and adsorption onto suspended particulate matter (SPM). During the intermediate river-discharge, the Al concen- tration was 0.72 μM at the river end-member and again decreased with increasing salinity. However, the removal was less pronounced, being only about 25%. Good fits with model predictions and laboratory experiments suggest that principal removal mechanisms were authigenic aluminosilicate formation and adsorption onto SPM. In the upper St. Lawrence Estuary, Al distribution is controlled by a combination of three removal mechanisms: flocculation, authigenic aluminosilicate formation, and adsorption. Each mechanism can become a dominant factor depending on the con- centration level and speciation of dissolved Al in the river water. et al., 1970; Hydes and Liss, 1977; Van Bennekom and Jager, 1978; Mackin and Aller, 1984a, b; Morris et al ., 1986; Hydes, 1989). One of these processes is salt-in- duced flocculation of riverine colloidal Al as demonstrated by Sholkovitz (1976, 1978) and Eckert and Sholkovitz (1976). In this case, riverine dissolved Al, which exists as colloid, is removed in an estuary together with other dissolved constituents during the mixing of river water and seawater. Colloidal Al has indeed been observed in the Conway River (Hydes and Liss, 1977) and in the Zaire River (Van Bennekom and Jager, 1978). In estuar- ies with low dissolved iron (Fe) and organic matter, Mackin and Aller (1984a) and Mackin (1986, 1989) have demonstrated that authigenic aluminosilicate formation is the major mechanism controlling the concentration of dissolved Al. Their authigenic aluminosilicate equilibrium model can explain a removal of dissolved Al at low salinities as well as an addition of dissolved Al at high salinities. In contrast, Morris et al. (1986) proposed that suspended sediment dynamics control the distribution of Al in an estuary based on their results from the Tamar Estuary. In this case, riverine dissolved Al is removed in a well-developed turbidity maximum zone and some Al flux is expected from mid-estuarine sediments. 1. Introduction Aluminum (Al) is a geochemically active element in the marine environment. It has a relatively short oceanic residence time of 100–200 years (Orians and Bruland, 1985) reflecting its affinity for particles. The distribution of Al in open ocean seawaters can be explained by an enrichment in the surface due to atmospheric inputs fol- lowed by its adsorption onto suspended particulate mat- ter, resulting in a depletion at mid-depths and an enrich- ment at depths due to redissolution from the solid phase (Mackenzie et al., 1978; Hydes, 1979, 1983; Moore, 1981; Olafsson, 1983; Moore and Millward, 1984; Measures et al., 1984, 1986; Orians and Bruland, 1985, 1986; Maring and Duce, 1987; Moran and Moore, 1989; Upadhyay and Sen Gupta, 1994). The reactive nature of Al has also been observed in estuarine environments and the riverine input is known to be modified by several estuarine processes (Hosokawa