TECHNICAL REPORTS Seasonal Trends of Aluminum Chemistry in a Second-Order Massachusetts Stream DREW C. McAvoY* ABSTRACT An investigation of streamwater chemistry was conducted to evaluate seasonal and spatialvariations of AIin an acid-sensitive watershed. Monthly grab samples were collected from three stream sites along the West Wachusett Brook, MA, and analyzed for all major solutes from March 1985 to January 1986. Total monomeric AIconcentrations were greatest during high runoff periods with values as high as 22 anti11 pM in upland and wetland runoff, respectively. Survey results alsoil- lustrated theimportance of ionic solute concentrations to themobility and transport of AI. The mountain stream showed that basiccation dilution was an important factor contributing to greater A! transport during high flow periods. In contrast, organic acids in thewetland stream served as a mobile anion, and thus greatly affected AI transport by com- plexation and charge balance effects. Solubilitywith an aluminum hydroxide mineral phase didnotappear to be a controlling factor of streamwater AIconcentrations throughout the year. Results from a chemical equilibrium model showed the mountain stream to be poten- tially toxic to fish during high-flow periods, whereas thewetland stream, being dominated by organo-AI complexes, had a minimal potential for toxicological effects. Acidic deposition has had an impact on surface water quality in northern Europe and eastern North America by altering biogeochemical cycles in acid-sensitive water- sheds (Cronan and Schofield, 1979; van Breemen et al., 1982). One consequence has been an observed increase in surface water AI concentrations (Dickson, 1978; Johnsonet al., 1981; Driscoll et al., 1984). Elevated A1 concentrations in low ionic strength waters are of con- cern because the role of AI in pH buffering (Dickson, 1978) and the potential toxicological effects of A1 to aquatic organisms (Baker and Schofield, 1982). Major factors thought to affect A1 mobility in a watershed system are bedrock geology (Norton, 1980), soil chemistry (Reuss and Johnson, 1985), soil depth (Lawrenceet al., 1986), vegetation (Cronan and Reiners, 1983), hydrologic flowpaths (Chen et al., 1984). The speciation of AI is extremely important when assessing toxicity to aquatic organisms. Aquo AI is thought to be the most toxic form with inorganic and organic complexesbelieved to be less toxic (Driscoll et al., 1980; Baker and Schofield, 1982). Laboratory bioassay studies have shown A1 to be detrimental to fish at concentrations of aquo plus hydroxyl complex species Department of Civil Engineering, 220Hinds Hall, Syracuse Univ., Syracuse, NY 13244-1190. Received 10 July 1987. *Corresponding author. Published in J. Environ.Qual. 17:528-534 (1988). as low as 7.5 /~mol/L (Muniz and Leivestad, 1980; Schofield and Trojnar, 1980; Baker and Schofield, 1982). Although this value can be considered as a bench mark for assessing streamwater toxicity, other factors such as fish life history stage and water quality parameters(e.g., pH, calcium, and temperature) should also be considered. Despite the importance of A1 chemistry in surface water systems, a recent literature review (Driscoll, 1988) has shown few studies that focus soley on seasonal trends of AI chemistry in a stream ecosystem. Investigations of tem- poral variability have primarily been conducted during: fall rainfall and spring snowmelt episodic events. In ad- dition, few studies have investigated the role of a wetland with regard to streamwater acidification effects. Thus, the purpose of this study was to investigate seasonal and spatial variations of AI in an acid-sensitive Massachusetts stream that includes both upland and wetland runoff and also to evaluate the potential toxicological effect of A1. The West Wachusett Brook was chosen for study because of its history of elevated concentrations of AI, because’. it has contrasting mountain and wetland branches and. also becauseof a decline in fish populationreported since.. the middle 1960s (Arthur Screpetis, personal communica- tion). STUDY SITE AND METHODS This study was conducted in the West Wachusett Brook catch- mentlocated approximately 80 km from the Atlantic coast in central Massachusetts, USA (Fig. 1). West WachusettBrook originates on the westernslope of Mt. Wachusett (elev. 580m) and flows in a westerlydirection for approximately 3 km before entering the Bickford Pond (elev. 325 m). Sampling sites were located on two upstreambranches. Thesestream branches are considerably different, with one consisting of freely draining mountain runoff and the other consisting of wetlandrunoff. Athird site, located approximately 1 km downstream from the confluence of the two upstream sites, provides informationon changes in streamcomposition after mixing of the twoupstream branches. The catchment consists mostly of protected lands (i.e., Wachusett MountainState Reservation and Minns Wildlife Sanctuary), and encompasses a drainage area of 578 ha. The wetland streamdrains a larger area than the mountain stream, 220and 145 ha, respectively. The downstream site includes an additional 77 ha of drainage area. The dominant tree species are red oak (Quercus borealis), red maple (Acer rubrum), white pine (Pinus strobus), and eastern hemlock (Tsugacanadesis), with a scattering of American beech (Fagusgrandifolia) and yellow birch (Betual alleghaniensis) on the mountain slopes (Eshleman, 1982). The surficial geology consists primarily of’ glacial till (Stone, 1978)with a fragipan layer approximately 70 to 90 cmbelow the surface. Glacial outwash areas are also 528 J. Environ. Quai., Vol. 17, no. 4, 1988 Published October, 1988