The influence of dissolved organic carbon, suspended particulates, and hydrology on the concentration, partitioning and variability of trace metals in two contrasting Wisconsin watersheds (USA)

Factors controlling the levels and forms of trace metals in streams, and transport from watersheds into streams, are poorly understood. We determined levels of filtrable (<0.4 mu m) and particulate Al, Cd, Cu, Pb, and Zn in rivers draining two contrasting watersheds in eastern Wisconsin. Both are developed on glaciated terrain, but the North Branch Milwaukee watershed (agricultural) is characterized by gently sloping uplands with thick silt loam soils, while the Wolf River (forested) has peat loam soils on hilly terrain. Watershed differences resulted in higher mean dissolved organic carbon (DOC) concentrations under baseflow conditions in the Wolf River (7 mg l(-1)) than in the Milwaukee River (4 mg l(-1)), but higher suspended particulate matter (SPM) levels in the Milwaukee (18 vs. 2 mg l(-1)). Mean filtrable trace metal concentrations (ng l(-1)) under baseflow conditions were similar in the Milwaukee and Wolf rivers (Al = 4.0 and 5.8; Cd = 4.2 and 3.2; Cu = 660 and 200; Pb = 56 and 53; Zn = 340 and 240), but particulate metal levels (ng l(-1)) were higher in the Milwaukee (Al = 230 vs. 16; Cd = 14 vs. 0.8; Cu = 736 vs. 90; Pb = 750 vs. 30; Zn = 2290 vs. 180). Filtrable metal concentrations exhibited low variability (relative standard deviation) under baseflow (Milwaukee: 5-20% Cu, Pb, Zn; 20-40% Al, Cd-Wolf: 10-35% Cu, Pb, Zn; 30-60% Al, Cd), while variability in particulate metal concentrations was somewhat higher (18-73%). During events (spring melt, storms), concentrations increased 1- to 5-fold for filtrable metals and 1- to 17-fold for particulate metals. Differences between watersheds and within-watershed responses to events were related in part to SPM (particulate metals) and DOC (filtrable metals) levels in the rivers and their associated influences on metal partitioning. In the Milwaukee River (higher SPM and lower DOG), particulate metal phases were more dominant (range among the five metals in mean % particulate = 47-98% vs. 24-74% for the Wolf. Partitioning (K-d) to SPM followed the order Pb > Zn > Cd > Cu, whereas relationships to DOC indicated essentially the reverse order for metal binding to DOG. SPM was a good predictor of particulate metal levels, especially in the Milwaukee River (> 94% of variance in particulate Cd, Pb, and Zn was accounted for, vs. 44-69% in Wolf River). While DOC was able to account for a significant portion of the variation in certain filtrable metal levels (r(2) = 0.52-0.65), other metals showed poorer correlations, probably due to variability in the composition of DOC and SPM and to variability in the forms of metal (aqueous complexes, colloid-bound) in the filtrable fraction.