Characterization of an organic acid analog model in Adirondack, New York, surface waters

Natural waters include a variety of organic matter that differs in composition and functional groups. Dissolved organic matter is important but difficult to characterize acidic and metal binding (e.g., Al) functional groups in chemical equilibrium models. In this study data from Adirondack Lake Survey were used to calibrate an organic acid analog model in order to quantify the influence of organic acids on surface water chemistry. The study sites in the Adirondack region of New York have diverse levels of dissolved organic carbon (DOC), used as a surrogate for organic acids. DOC in 55 Adirondack surface waters varies from 180 μmol C/l (in Little Echo Pond) to 1263 μmol C/l (in Sunday Pond). To reduce the variability inherited in the large raw data set, suite of mean observations was constructed by grouping and averaging measured data into pH intervals of 0.05 pH units from pH 4.15 to 7.3. A chemical equilibrium model, which includes major solutes in natural waters, was linked to an optimization algorithm (genetic algorithm) to calibrate a triprotic organic analog model which includes proton and aluminum binding by adjusting the dissociation constants and site density of DOC. The object of fitting procedure was to simultaneously minimize the discrepancy between observed and simulated pH, acid neutralizing capacity (ANC), organic monomeric aluminum and inorganic monomeric aluminum. A sensitivity analysis on calibrated values indicate that the speciation of the modeled solutes are most responsive to the dissociation constant of AlOrg= Al3+ + Org3- reaction (Org3- represents organic anion), the site density of DOC and the second H+ dissociation constant of the triprotic organic analog (i.e. H2Org- = 2H+ + Org3- reaction).