Environmental context. To assess the risk posed by environmental contaminants such as metals, one needs to be able to identify the key chemical species that prevail in natural waters. One of the recognised stumbling blocks is the need to quantify the influence of heterogeneous dissolved organic matter (DOM). Here we explore the possibility of using the optical signature of DOM to determine its quality, to alleviate the need to make assumptions about its metal-binding properties and to improve the prediction of trace metal species distributions in natural waters.Abstract. To calculate metal speciation in natural waters, modellers must choose the proportion of dissolved organic matter (DOM) that is actively involved in metal complexation, defined here as the percentage of active fulvic acid (FA); to be able to estimate this proportion spectroscopically would be very useful. In the present study, we determine the free Cd 2þ , Cu 2þ , Ni 2þ and Zn 2þ concentrations in eight Canadian Shield lakes and compare these measured concentrations to those predicted by the Windermere Humic Aqueous Model (WHAM VI). For seven of the eight lakes, the measured proportions of Cd 2þ and Zn 2þ fall within the range of values predicted by WHAM; the measured proportion of Cu 2þ falls within this range for only half of the lakes sampled, whereas for Ni, WHAM systematically overestimated the proportion of Ni 2þ . With the aim of ascribing the differences between measured and modelled metal speciation to variations in DOM quality, the percentage of active FA needed to fit modelled and measured free metal concentrations was compared with the lake-to-lake variation in the spectroscopic quality of the DOM, as determined by absorbance and fluorescence measurements. Relationships between the percentage of active FA and DOM quality were apparent for Cd, Cu, Ni and Zn, suggesting the possibility of estimating the percentage of active FA spectroscopically and then using this information to refine model predictions. The relationships for Ni differed markedly from those observed for the other metals, suggesting that the DOM binding sites active in Cd, Cu and Zn complexation are different from those involved in Ni complexation. To our knowledge, this is the first time that such a distinction has been resolved in natural water samples.

No abstract

In the present study, we explored the use of various optical parameters to detect differences in the composition of the dissolved organic matter (DOM) in a set of lakes that are all located on the Canadian Precambrian Shield, but within which Cu and Ni speciation predictions were previously shown to diverge from measured values in some lakes but not in others. Water samples were collected with in situ diffusion samplers in 2007 (N = 18 lakes) and 2008 (N = 8 lakes). Significant differences in DOM quality were identified between the sampling regions (Rouyn-Noranda, Québec and Sudbury, Ontario) and among lakes, based on dissolved organic carbon concentrations ([DOC]), specific UV absorbance (SUVA 254 ), fluorescence indices (FI), and excitation-emission matrix (EEM) fluorescence measurements. Parallel factor analysis (PARAFAC) of the EEM spectra revealed four components, two of which (C3, oxidized quinone fluorophore of allochthonous origin, and C4, tryptophan-like protein fluorescence of autochthonous origin) showed the greatest inter-regional variation. The inter-lake differences in DOM quality were consistent with the regional watershed characteristics as determined from satellite imagery (e.g., watershed-to-lake surface area ratios and relative percentages of surface water, rock outcrops vegetative cover and urban development). Source apportionment plots, built upon PARAFAC components ratios calculated for our lakes, were used to discriminate among DOM sources and to compare them to sources identified in the literature. These results have implications for other areas of research, such as quantifying lake-to-lake variations in the influence of organic matter on the speciation of trace elements in natural aquatic environments.

It is now well established that the bioavailability of metals toward aquatic organisms varies as a function of the free metal concentration. The ion-exchange technique (IET), which consists of equilibrating a calibrated cation-exchange resin with the water sample, is one of the few existing speciation methods that provide sufficient sensitivity and specificity to measure free metals in natural waters. In the present study, we developed an in situ IET (field-IET) in which the resin was directly equilibrated on site using dialysis cassettes. The field-IET was tested in six Canadian Shield lakes and in the Athabasca River (AB, Canada) for Cd 2? , Co 2? (only in the Athabasca River), Ni 2? , and Zn 2? . Measurements were compared with those obtained on samples collected from the same sites with in situ diffusion samplers and analyzed in the laboratory (lab-IET). IET results were also compared with predictions from the Windermere Humic Aqueous Model [WHAM VII; the Co and Ni carbonato complex formation constants were updated according to the US National Institute of Science and Technology (NIST compilation)]. Good agreement was obtained between the field-IET and the lab-IET for Co 2? and Ni 2? concentrations, but field-IET Cd 2? and Zn 2? concentrations were systematically higher than the lab-IET results (factors of *1.59 and *2.49, respectively). Uncertainties in the field-IET resin calibration and a significant Zn contamination could explain these results. WHAM VII predicted the free metal concentrations reasonably well, except for Ni 2? concentrations in the lakes, probably due to inappropriate formation constants for complexation of Ni with Electronic supplementary material The online version of this article (dissolved organic matter. This study showed that the field-IET, with its relative simplicity and its low detection limits, could be a useful method for the determination of free metal concentrations in acidic to neutral freshwaters.

No abstract