Edward R. Sholkovitz

Flocculation of dissolved organic and inorganic matter during the mixing of river water and seawater

Abstract A new approach to estuarine studies (termed the product approach) is presented which establishes 1. (1) the extent and salinity dependence of non-conservative behaviour; 2. (2) composition and chemical form of removal products; and 3. (3) abiological removal mechanisms. In the product approach the composition of removal products (i.e. flocculants), which result from mixing of river waters (of Scotland) and seawater at varying salinities, are determined using laboratory experiments designed with the minimum of constraints. Rapid flocculation of Fe, Mn, Al, P, organic carbon and humic substances occurs when filtered river waters are mixed with filtered seawater. The amounts of flocculated constituents increase as salinity increases from 0 to 15–20%., above which little additional removal occurs. The extent of flocculation is very salinity-dependent indicating the destabilization of river-introduced colloidal humic substances during the mixing with sea water. The close association of Fe, Mn, Al and P with both river-dissolved humic substances and the sea water-flocculated humates demonstrates the important role of dissolved organic matter in controlling the non-conservative behaviour of inorganic constituents. The extent of removal, relative to the river water, ranges between 75 and 115% for Fe, 90 and 100% for Mn, 50 and 200% for P and 10 and 70% for Al. Although only 3–11% of the river DOM is flocculated, the organic component (i.e. humates) outweigh the sum total of the inorganic trace elements. Silica shows a lack of reactivity (3–6% removal). Removal processes, described in this paper, are applicable to the major rivers of the world where they may play an important role in the chemical mass-balance between the rivers and the ocean.

The mechanism of iron removal in estuaries

Abstract A survey of U.S. east coast estuaries confirms that large-scale rapid removal of iron from river water is a general phenomenon during estuarine mixing. The river-borne ‘dissolved’ iron consists almost entirely of mixed iron oxide-organic matter colloids, of diameter less than 0.45 μm, stabilized by the dissolved organic matter. Precipitation occurs on mixing because the seawater cations neutralize the negatively charged iron-bearing colloids allowing flocculation. The process has been duplicated in laboratory experiments using both natural filtered and unfiltered river water and a synthetic colloidal goethite in 0.05 μm filtered water. The colloidal nature of the iron has been further confirmed by ultracentrifugation and ultrafiltration. A major consequence of the precipitation phenomena is to reduce the effective input of ‘dissolved’ iron to the ocean by about 90% of the primary river value, equivalent to a concentration of less than 1 μmol per liter of river water.

The removal of dissolved humic acids and iron during estuarine mixing

Abstract The estuarine chemistry of dissolved humic acids was determined by carrying out both field and laboratory studies. These approaches were combined in an investigation of the Amazon estuary while laboratory mixing experiments were performed using filtered (0.45−0.001 μm) river water fractions of the Water of Luce (Scotland). The results demonstrate that a small fraction of river dissolved organic matter is preferentially and rapidly flocculated during estuarine mixing. This fraction is the high molecular weight component of dissolved humic acids (0.45−0.1 μm filtered). Approximately 60–80% of the dissolved humic acid in these rivers flocculates during estuarine mixing. This represents a removal of only 3–6% of river dissolved organic matter and is responsible for the non-conservative behaviour of dissolved humic acid in the Amazon estuary even though total dissolved organic carbon appears conservative. The salinity dependence with which humic acid flocculates in estuaries is similar to that of iron. This implies that both constituents may be removed from river water by a common mechanism of colloid flocculation.

The flocculation of dissolved Fe, Mn, Al, Cu, Ni, Co and Cd during estuarine mixing

Abstract A laboratory experiment was carried out in which the flocculation products, formed from the mixing of filtered (0.4 μm) river water and seawater, were analysed. This study established that Fe, Mn, Al, Cu, Ni, Cd and Co have resolvable and well-defined estuarine chemistries. Copper, Ni, Mn and Co have salinity dependences of removal which are similar to those of dissolved Fe and humic acids. The amount of removal of the above trace metals increases between 0 and 15–18‰, after which little additional removal occurs. The extents of removal from river water are very different: Fe, 95%; Al, 20%; Cu, Ni, 40%; Co, 10%; Cd, 5% and Mn, 25–45%. The basic removal mechanism appears to be the estuarine flocculation of trace metals which exist, in part, in river water as colloids in association with colloidal humic acids and hydrous iron oxides. A qualitative model, based on this mechanism, supports the observations of this flocculation study. The results of this study give the most complete and consistent set of data presently available, from which to postulate the most important processes controlling the estuarine chemistry of trace metals. The generality of their behaviours still needs to be determined by future investigations.

The coagulation, solubility and adsorption properties of Fe, Mn, Cu, Ni, Cd, Co and humic acids in a river water

Abstract River water (Water of Luce, Scotland) is used in laboratory experiments designed to investigate physical and chemical properties of Fe. Mn, Cu, Ni, Co, Cd and humic acids in riverine and estuarine systems. Using NaCl, MgCl 2 and CaCl 2 as coagulating agents, coagulation of dissolved (0.4 μm filtered) Fe, Cu, Ni, Cd and humic acids increases in a similar matter with increasing salt molarily: Ca 2+ is the most dominant coagulating agent. Removal by coagulation with Ca 2+ at seawater concentrations ranges from large (Fe-80%. HA-60%, Cu-40%) to small (Ni, Cd-15%) to essentially nothing (Cd, Mn-3%). Destabilization of colloids is the indicated mechanism. Solubility-pH measurements show that between a pH of 3 and 9, Fe, Cu, Ni, Mn, Co and Cd are being held in the dissolved phase by naturally occurring organic substances. Between pH of 2.2 and 1.2 a large proportion of dissolved Fe, Cu. Ni and Cd (72, 35,44 and 36% respectively) is precipitated along with the humic acids; in contrast, Mn and Co show little precipitation (3%). Adsorption-pH experiments, using unfiltered river water spiked with Cu, indicate that adsorption of Cu onto suspended particles is inhibited to a large extent by the formation of dissolved Cu-organic complexes. The experimental results demonstrate that solubilities and adsorption properties of certain trace metals in freshwaters can be opposite to those observed with artificial solutions or predicted with chemical models. Interaction with organic substances is a critical factor.

The flocculation of iron, aluminium and humates from river water by electrolytes

This paper reports on the flocculation of iron, aluminium and humates from Scottish river water by the three major salts in sea water. The molarities of the salts required to precipitate the maximum amounts of Fe, Al and humates decrease in the order: NaCl, MgCl2, CaCl2. The maximum amounts of a species precipitated by the salts are shown to increase in the same order. Chemical and electrostatic interactions are both involved in the flocculation process.

The chemistry of suspended matter in Esthwaite Water, a biologically productive lake with seasonally anoxic hypolimnion

Ten detailed vertical water column profiles were taken between April and November, 1979, in Esthwaite Water (English Lake district), a lake with high biological productivity and a seasonally anoxic hypolimnion. Measurements of the major-element particle composition (organic C, P, S, Si, Al, Ti, K, Mg, Ca, Fe, Mn, and Ba) and hydrochemical constituents (temperature, pH, dissolved oxygen, total suspended load, dissolved Fe, Mn, P, and Ba) were carried out. These have revealed new information about the mechanisms and kinetics of biogeochemical cycles in a lake. Pronounced seasonal cycles exist in which large excess concentrations (those unsupported by detrital components) of particulate organic C, Fe, Mn, P, S, Mg, K, Ba, and Ca are being generated and lost in situ in the water column (15m deep). In the epilimnion these elements (excepting Fe and Mn) are incorporated into the organic components of growing phytoplankton during the spring and summer. Simultaneously, in the hypolimnion there is a build-up and then a decrease in the excess concentrations of particulate C, P, S, Mg, K, Ba and Ca; this cycle is due to the indirect involvement of these elements with the iron redox cycle. As the hypolimnion becomes anoxic, dissolved ferrous Fe is released from the sediments and large concentrations of excess particulate iron (III) oxides accumulate; these oxides act as adsorbing substrates for the above mentioned elements. As conditions become more reducing, these same elements are solubilized as the iron (III) oxide particles are reduced to dissolved ferrous iron. Adsorption equations are derived from the field data which relate the concentration of excess particulate Fe to those of POC, P, S, Ca, Mg, Ba, and K. At the last stages of anoxia (before the lake overturns) large populations of bacteria and the formation of iron sulfide particles control the concentrations of excess particulate C, S, P, Mg, K, and Ca.

On the association of iron and manganese with organic matter in anoxic marine pore waters

Abstract Experiments were performed to establish directly the extent of association of Fe and Mn with dissolved organic matter in anoxic pore waters from the sediments of Loch Duich, Scotland. Of the dissolved Fe, 74–84% was retained by an ultrafiltration membrane (MW ⩾ 1000), in the high molecular weight (HMW) phase, compared with 3–17% of the dissolved Mn. After u.v. oxidation the measured total concentration of Fe increased by 43–70% while that in the HMW phase was reduced to 27–44%. Simultaneously, the concentration of dissolved Mn increased by only 1–12%. It was concluded that most if not all the Fe was organically bound while Mn was present mainly as labile inorganic complexes.

Chemical and physical processes controlling the chemical composition of suspended material in the River Tay Estuary

Abstract The chemical composition (Al, Si, Ti, K, Ca, Mg, P, Org. C, Fe and Mn) of suspended material in the Tay Estuary and River Tay have been measured to determine the relation between river and estuary material and chemical reactions which may be occurring during estuarine mixing. Variations in the ratios of Fe Al , Mn Al , Ti >Al , etc., with salinity and suspended load during a tidal station suggest that sedimentological and hydrological processes, rather than chemical ones, are responsible for the observed compositional changes. This interpretation is confirmed by laboratory mixing experiments which also contradict published reports of Fe and Mn desorption in estuaries. Measurements of suspended matter composition will not easily determine whether desorption or adsorption of element occurs when river-borne suspended material enters the saline environment. A tentative conclusion on the River Tay study is that the suspended matter in the Tay Estuary results from the input of material at times of high suspended loads and of high river water discharge.

Non-uniform vertical distribution of fine sediment in the Amazon River

IN studies of sediment transport in rivers, it is frequently assumed that suspended particles finer than a certain size (usually of the order of 60 µm) are uniformly concentrated from river bed to water surface1,2. The assumption seems to be mostly a matter of convenience; it is well established that the vertical concentration gradient for sediment particles suspended in rivers is directly proportional to fall velocity of the particles and inversely proportional to shear velocity of the flow3. For very fine particles suspended in highly turbulent flows, concentration gradient may be small; the assumption has been used, therefore, to justify collection of near-surface samples in streams where velocities are high and turbulence is intense2. Our recent studies of sediment in suspension in the Amazon River show that this assumption is not valid for deep rivers and that appreciable errors result from using concentrations of surface samples to represent concentration of suspended sediment and its associated adsorbed constituents through the entire stream depth.