Abstract

Dissolved and colloidal concentrations of Al, Fe, Mn, Cu, Zn, Cd, and Pb were determined in the Mullica River, New Jersey, and its tributaries on nine occasions between August 1994 and September 1995. Cross‐flow filtration (CFF) at a nominal 10 kDa cutoff was used to separate total dissolved (>0.45 μm) riverine metals into colloidal and truly dissolved fractions. Reproducibility tests on full procedural replicates demonstrated CFF's precision and indicated that temporal or spatial differences exceeding 10–15% in colloidal fraction can be resolved using these techniques. For the whole data set, mean colloidal fractions of 0.45‐ μm filtrates were 65% for Al, 82% for Fe, 29% for Mn, 66% for Cu, 31% for Zn, 46% for Cd, and 88% for Pb. The magnitude of the colloidal fraction followed an elemental order that was largely consistent from site to site and among different sampling dates, despite large variations in total dissolved metal concentrations. The order was very similar to that observed previously in rivers of varying geochemical properties, suggesting that the degree of colloidal association is driven by a metal's affinity for humic materials and is relatively insensitive to other riverine properties such as pH, dissolved organic carbon, and suspended mass. The constancy of this elemental pattern in Pinelands streams suggests that the relative fates of metals during estuarine colloid flocculation may be predictable and temporally stable. Iron behavior was distinct from that of the other elements: colloidal Fe fraction was positively correlated with pH (range, 4.3–7.1) and the colloidal Fe/Al ratio showed clear seasonal variation. The other metals did not exhibit these behaviors, which suggests that their colloidal association was distinct from that of Fe. Colloidal metal/Al ratio for Fe, Pb, and Zn was clearly different from that of >0.45 μm particles; that finding is consistent with the distinct compositions and sources for colloidal and particulate fractions. The distribution coefficient (Kd) for Pb was positively correlated with suspended particulate mass (<10‐fold variation) over time at one representative site. This inverse particle concentration effect runs contrary to recent observations in other freshwater and marine systems and suggests that colloidal and particulate size fractions vary independently over the course of the year.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call