¹ Post-depositional mobility of trace elements in northern forest ecosystems

Hale, B², W. Hendershot³, D. MacDonald³, and D. Johnson²

² University of Guelph, Guelph, ON
³ McGill University, Ste. Anne de Bellvue, QU

Current knowledge of post-deposition metal behavior in boreal forests is incomplete for the purpose of predicting the fate, and thus eventual concentrations, of trace metals added as a result of human activity. Understanding trace metal fate and using that knowledge to predict the concentrations of metals in soils and vegetation, as well as losses of metals from soils, are essential for ecological risk assessment. The goal of this project is to better understand the processes that control the cycling of metals once deposited within boreal forest ecosystems. The processes under study include: the binding of metals to soils, processes which are responsible for the solubility, mobility and bioavailability; and, the sequestration of metals in vegetation and their release, processes which move and concentrate metals in environmental compartments different from initial deposition.

Soil solutions were collected from the lysimeters installed at Rouyn-Noranda and Sudbury sites from June to November 2001. An ion exchange technique was used to measure the concentrations of Cd2+ and Pb2+. Other solution properties such as pH, DOC, anions and dissolved metal cations were also measured. A total of 34 soils (sampled in 1999 and 2000) were used to investigate the surface charge and metal adsorption on the soil surface materials. Back-titration demonstrated that the surface charge was strongly affected by soil organic matter and oxides. Batch equilibrium experiments demonstrated that more metals were bound to the soil surface when the soil contained higher organic matter. Speciation data of Cd, Cu and Pb, which were obtained by the ion exchange technique, were used in the NICA-Donnan model to determine the binding constants of Cd, Cu and Pb to DOC in lysimeter solutions. The results of the back-titration and batch adsorption experiments, along with a computer program, FITEQL, are being used to determine the equilibrium constants of surface reactions (KH for protonation and KM for metal sorption). Water soluble metal concentrations in soils have been determined for comparison to data on root metal accumulation.

The relative inputs of metals (Cu, Ni, Pb and Zn) from the decomposition of leaves and fine roots to soils at study sites located along transects following soil metal gradients near smelters in the regions of Rouyn-Noranda, QC and Sudbury, ON. Fine roots played the dominant role in Cu and Pb transfer to soils, consistent with the strong binding properties of these elements to root tissues. Fine roots also were important in Ni transfer to soils. Zinc flux to soils was proportional to biomass turnover, and therefore transfer to soils was equally attributable to decomposition of fine roots and leaves. Litterbag experiments demonstrated that decomposing litter actully is a sink for metals, as the metal content of the litter in the bag increased despite a loss of biomass. These gains in metals were compared to atmospheric input data, and in most cases, atmospheric deposition was sufficient to account for the increases in metal content. The role of litter as a sink was metal-specific, and was related to the generalized affinity of the metals for organic matter.

Top of Page

Development of rational models for relating metal accumulation by aquatic organisms to metal concentrations in their environment: relative importance of ambient metal sources

Hare, L. (PI), L. Croisetière, M.-N. Croteau, C. Gallon, A. Gosselin, J. Orvoine & A. Tessier.

Institut National de la Recherche Scientifique - Eau, Terre et Environnement (INRS-ETE), Université du Québec, Sainte-Foy, QC.

The overall goal of our research is to develop and test in the field theoretically based models that relate metal concentrations in aquatic organisms to those in their surroundings. To develop such models we need to better understand the chemical and biological processes involved in the bioaccumulation of metals. First, to understand the exposure of burrowing animals to metal in the sediment compartment versus that in the overlying water-column compartment, we are studying the manner in which insects construct and irrigate their burrows. We have shown (Gallon, Gosselin) that burrowing insects of various types differ in their potential exposure to metals in anoxic interstitial waters; the phantom midge spends its day in anoxic sediments whereas mayflies and alderflies irrigate burrows thereby maintaining their surroundings oxygenated. Overall, our results suggest that a ‘one size fits all' approach to ERA for sediments would ignore important behavioural differences among taxa that could be a major factor in their accumulation of sedimentary metals. Experiments are now underway to determine if burrowing behaviour is influenced by metal bioaccumulation. Second, we are working to determine the relative importance of food and water as metal sources (Croisetière). To this end, we are determining if our laboratory observation that alderfly larvae take up most of their Cd from prey also holds true for this and other trace metals in nature. Chironomus larvae reared in the laboratory were placed in the sediments of contaminated Lake Dufault to accumulate metals whereas others were not exposed to high metal concentrations. Results suggest that water is a negligible metal source compared to prey in nature. Because our earlier studies suggest that prey are also the major metal source for Chaoborus, we compared Cd concentrations in this predator to those of its planktonic prey (Croteau). We found that Cd in predator and prey can be related but only if measurements in prey are made on well-defined taxonomic units rather than on vague assemblages (that is, "copepods" versus "bulk plankton"). Furthermore, Cd concentrations in prey were generally more than those in the predator suggesting that Cd concentrations decline along food chains (biodilution). In contrast, Hg concentrations in prey were similar to those of an invasive predator (Bythotrephes) suggesting an absence of Hg biomagnification (Croteau). This complementary series of projects is aiding us to build mechanistic models that we are testing in lakes and streams by measuring metal concentrations in water, a plant and several invertebrates. Such models can serve as the basis for using organisms as metal biomonitors. In addition to using organisms to compare metal concentration among sites, we have also shown that declines in Cd over the last decade can lead to counter-intuitive increases in the Cd present in biota (Croteau). Our explanation of this trend springs from the observation that Cd and H ions appear to compete for biological uptake sites; we now hope to determine at what level in the food chain this competition occurs (Orvoine). Because metal accumulation precedes toxic effects, such models are an important step towards predicting toxicity. Environmental regulations springing from ecological risk assessments resting on a sound understanding of metals and their interactions with living organisms are more likely to be defensible scientifically than are other more empirical approaches.

Top of Page

Quantification and Modelling of Metal Mobility in Lakes

Diamond, M.L1^,3., L.J. Evans², S. Bhavsar³, P. Cypas², K. Rudnitski^2
1 Department of Geography, University of Toronto, Toronto, Ontario M5S 3G3
² Department of Land Resource Science, University of Guelph, Guelph, Ontario N1G 2W1
³ Department of Chemical Engineering and Applied Chemistry, University of Toronto, Toronto, Ontario

The aim of our research is to develop a general model of metal chemistry and fate in an aquatic system. The research derives from the need to link metal emissions and loadings to lakes with resultant concentrations, including the bioavailable fraction of metal, and metal distribution within a lake. This linkage provides the pathways analysis component necessary to estimate metal concentrations and hence exposure within an ecological risk assessment. For metals, achieving this goal demands consideration of metal speciation as a function of ambient chemistry.

We have developed a general chemistry-fate model named TRANSPEC (Transport and Speciation Model). The model is applicable to most metals and most aquatic systems. The chemistry component of the model is an adaptation of the equilibrium speciation model MICROQL while the fate model is based on the multi-species version of the aquivalence-based QWASI model (Quantitative Water Air Sediment Interaction) of Mackay (1991) and Diamond et al. (1992) and co-workers. We have developed steady- and unsteady-state versions of TRANSPEC that estimates metal fate in dissolved, colloidal (DOC-bound) and particulate phases within a stratified water column and two vertical sediment layers.

To calibrate and evaluate model performance, we used the model to estimate the fate of zinc and copper in Ross Lake, Manitoba, in which these metals have accumulated to high concentrations in the organic-rich sediments. This system has provided a rigorous test of various aspects of the model due to its complexity of fluctuating redox conditions in surface sediments and the remobilization of sedimentary metals. The model is also applicable to simpler systems such as those with minimal historically-accumulated metal and with organic-poor sediments.

For the application of TRANSPEC to Ross Lake we sampled and analysed the following: weekly zinc and copper concentrations in the lake's inflow and water column, year-round estimates of sediment deposition and resuspension using sediment traps, pore water chemistry using peepers, sediment accumulation rate by means of Pb-210 dating of a vertical sediment profiles, mass transfer coefficient for the diffusive release of metal between pore water and the water column by means of a diffusive release experiment, and finally depth profiles of acid volatile sulphides and simultaneously extracted metals.

The results of the model indicate that zinc mobility is controlled by redox-sensitive diffusive fluxes from sediment-to-water and the resuspension of metal-rich sediments. The main route of zinc mobilization changes seasonally in response to changing sediment redox conditions. The competing ligand exchange method with electrochemical techniques for determination of speciation parameters in lake waters.

Top of Page

Are speciation parameters significant in terms of the needs of MITE-RN?

Chakrabarti, C.L., J. Murimboh, F. Raoufi, J.W. Guthrie, M.S.A. Salam, N.M. Hassan and A. Jamaluddin

Ottawa-Carleton Chemistry Institute, Department of Chemistry, Carleton University, 1125 Colonel By Drive, Ottawa, ON, K1S 5B6

The Free Ion Activity Model is generally regarded as a good indicator of metal bioavailability. However, the determination of very low free-metal ion concentrations is a difficult challenge. Here, we present the results of our investigations of the speciation of copper and cadmium in lake water samples from Lake Opasatica, Lake Vaudray, Lake Osisko, and Lake Dufault using the Competing Ligand Exchange Method. The speciation parameters determined include: free metal ion concentration, the concentration of naturally-occurring complexants, and stability constants of the metal complexes. Both the choice of the competing ligand and the electrochemical technique required (e.g. Anodic Stripping Voltammetry or Adsorptive Cathodic Stripping Voltammetry) depend on the specific metal under investigation. Copper speciation was studied by Square Wave Adsorptive Cathodic Stripping Voltammetry using catechol as the competing ligand. Cadmium speciation was investigated by Square Wave Anodic Stripping Voltammetry using ethylenediamine as the competing ligand. Both methods are based on competition for the complexation of the metal between the naturally-occurring ligands in the samples and the added competing ligand. Free metal ion concentrations were determined indirectly from equilibrium calculations after titration of the samples with the metal of interest. Free cupric ion and cadmium ion concentrations in the lake water samples were pCu 14-17 and pCd 9-11. Conditional stability constants of the Cu(II) and Cd(II) complexes, and the concentration of naturally-occurring ligands in the lake water samples were calculated by fitting the titration data to a one-ligand model using FITEQL. The results suggest the presence of very strong ligands. The main advantages of the method are that freshwater samples can be studied at their native total metal concentrations (» 10-7 mol/L) and that very low free metal ion concentrations can be determined.

The link between metal speciation and bioavailability is now widely accepted. This work has sought to extend this linkage further to include the lability of metal complexes. The linkage between the lability of metal complexes and bioavailability will be discussed. Since MITE-RN needs information on metal speciation and their bioavailability, determination of speciation parameters of metals serves the need well.

The speciation parameters provide speciation data needed for assessing the risk to ecosystems from these metals. This Project is directly applicable to the other Projects (B1, B2, B3, C1, C3, C9) for which it provides speciation data; for other Projects, the application is indirect. This Project is also directly applicable to PSL-2 Risk Assessment (SOURCES domain) because the speciation data provided by this Project will build on and improve the PSL-2 Risk Assessment.

Top of Page

Importance of reduced sulfur species in controlling metal speciation in the surface waters on the Canadian Shield: Methodological development and preliminary results

Wang, F.¹, A. Tessier²

¹. Environmental Science Program and Department of Chemistry, University of Manitoba, Winnipeg, MB.
². INRS-Eau, Université du Quebec, Sainte-Foy, QC.

Most metal ions of environmental concern (e.g., Cd, Cu, Pb, Zn, Hg) are Class B metals (or "soft Lewis acids") and tend to form strong complexes with reduced sulfur species (RSS) which are soft Lewis bases. Although RSS are not stable in oxic surface waters, their complexes with these metals may be more resistant to oxidation and thus affect the metal speciation in oxic surface waters. We have developed several methods to unambiguously determine the significance of RSS in controlling metal speciation in the Canadian Shield waters.

We have determined the solubility of rhombic sulfur by square wave cathodic striping voltametry (SWCSV). The solubility data allowed us to express the thermodynamic constants for polysulfides and polysulfide-metal complexes as a function of dissolved sulfur; this information was then incorporated into the speciation models HYDRAQL and WHAM to enable the calculations of metal speciation in the presence of sulfide and polysulfides.

Methods for direct identifications and quantifications of metal-RSS complexes were also investigated. Aqueous Mn-, Pb-, and Zn-sulfide complexes were synthesized in the laboratory by titrating the metal solution with sulfide under controlled conditions. Although there was measurable sulfide in the solution immediately after the synthesis, the sulfide was not measurable after the solution was exposed to air for 3-5 days, indicating limited stability of these complexes in oxic waters. Mass spectra were obtained, for the first time, for aqueous Pb-sulfide solution by directly electrospraying the aqueous samples into a mass spectrometer without any pretreatments.

Field measurements of total dissolved sulfide (along with a variety of other geochemical variables) were conducted in 4 Canadian Shield lakes. Sulfide concentrations were very low (1-17 nM) in the oxic waters of Lakes Tantaré, St-Augustin, and Despériers. Sub-micromolar levels (43 to 480 nM) of sulfide were found in the oxic waters of Lake Rouyn, but cross contamination by organic matter during the sampling could not be ruled out. Since free sulfide is not stable in oxic waters, the dissolved sulfide measured in these oxic waters is likely in the forms of metal-sulfide complexes.

The results suggest that the current practice of ecological risk assessment (ERA) of Class B metals in oxic surface waters needs to be revisited as it does not take into account the roles of RSS when estimating metal speciation. This can be readily improved with our revised speciation models. The results also question the usefulness of porewater toxicity testing, during which process both the sulfide concentrations and metal speciation can be significantly changed.

Top of Page

BIOGEOCHEMICAL CYCLING IN THE BOREAL FOREST – METAL CONCENTRATIONS IN WOODY TISSUES – PHASE II

Bégin, C., M. Savard, M. Parent, J. Marion, and A. Smirnov

Geological Survey of Canada, GSC-Q, Québec, Qué.

Dendrogeochemical investigations of a series of six forest sites in of Rouyn-Noranda region have shown that part of metals emitted from the Horne smelter accumulate in woody tissues. As an example, concentrations of Cd range between 0.02 to 1.05 ppm for contaminated sites near Rouyn (GSC MITE 1998-2001) while they vary from 0.01 and 0.03 ppm for our control site in the Hudson Bay region. What does this represent at the scale of the forest? How will these metals be incorporated and cycled in the forest system? The quantitative assessment of the total metal pool in woody vegetation is the first step to calculate the long term return of these metals to forest soils. Such investigations will directly contribute to the global understanding of metal biogeochemical cycles in the boreal ecosystem.

As a first phase in addressing this long term objective we have focused on two boreal forest sites exposed to contrasting levels of airborne pollutants in the Rouyn region and calculate the total metal content in black spruce trees using a method that takes into account metal variations along tree stem (MITE-RN 2000-2001). In fact, for this species, it was shown that lead concentration gradually decrease from 0.27 ppm at ground level to 0.01 ppm at the apex in response to distinct assimilation processes (MITE-RN 1999-2000). Our results indicate that for spruce trees growing at the polluted site, the total Cd load of standing wood can reach 1.2 kg/km2, which is about 10 time higher then our estimation for the unpolluted site. For zinc, the total load is much higher, reaching 80 kg/km2 at the contaminated site. However, in terms of wood volume, black spruce trees count for 30% to 50% of a typical boreal forest stand. To provide a quantitative estimation for a fuller spectrum of forest constituents at the two selected sites, co-dominant tree species were investigated this year.

We have adapted the field sampling strategy and the biomass evaluation methodology developed last year (MITE-RN 2000-2001) to the study of the balsam fir, paper birch and aspen. At each site, four trees per species, each representing a different growth stage, were selected and sampled at one meter intervals along the main stem and an average of 6 branches were sampled on each tree using a stratified sampling protocol. Detailed measurements were made to calculate the biomass of every single tree. At our tree-ring laboratory, branches were resampled (3 samples/branch) and all wood samples were prepared for metal analysis. Over 600 samples were submitted to the geochemistry laboratory for the analysis of metal concentrations by ICP-MS. Metal concentrations, along with individual tree biomass will provide an accurate estimation of the total metal content in spruce trees. Demographic data for boreal spruce populations provided by provincial forest services will serve to extend the total metal content to a given surface unit. Analyses are currently finalized and our results will be summarized at the symposium in a Process Domain Team presentation.

Top of Page

The availability of trace metals in the rhizosphere of contaminated soils from the Sudbury area

Courchesne, François, Université de Montréal, Montréal, QC

Context. The rhizosphere is the narrow soil volume (the first few mm) surrounding plant roots and affected by their activity. Because of its proximity to the site of elemental uptake by plants, the rhizosphere is a key component of terrestrial ecosystems. Clearly, several lines of evidence suggest that the extent of the functional role of the rhizosphere on the biogeochemical cycling of elements is much larger than the volume fraction it occupies in the field. Our previous work on the rhizosphere supports this statement and shows that the rhizosphere is enriched in organic C (20 to 600%) and more acidic (0,1 to 0,5 pH unit) than the surrounding bulk soil. We also found the rhizosphere of soils growing under Norway spruce (Picea Abies) to be depleted in weatherable minerals such as amphiboles, chlorite and plagioclases. Recently, water soluble, salt exchangeable and total trace metal contents (Cd, Cu, Pb, Zn) were shown to be higher, by up to a factor of five, in the rhizosphere than in the bulk soil. In this context, we believe that an approach that integrates the soil rhizosphere is needed to gain insights into the processes controlling the storage, cycling and availability of trace metals in forest ecosystems, particularly in contaminated environments.

The approach. The objective of this project is to contrast the fractionation of metals (Al, Cd, Cu, Fe, Ni, Pb, Zn) between the rhizosphere and bulk components of forest soils exposed to different levels of contamination and forming under distinct canopies. Soils were collected in October 2001 at three locations along a transect in the Sudbury area. These sites are already studied in Project B1 (Hale & Hendershot). At each location, the soil material adhering to the roots was collected in the B horizon under at least six different trees (Betula papyrifera). The bulk soil was collected 1 to 5 cm away from individual roots. In the laboratory, the soil still adhering to the roots after they had been shaken was considered as ‘inner' rhizosphere. The material that detached from the roots was regarded as the ‘outer' rhizosphere. All soil components were freed from root fragments using plastic tweezers. The mineralogical analyses are presently under way together with pH and organic C determinations. The fractionation and analysis of metals by ICP-AES or ICP-MS will start in March 2002

The impact of the data. The advances in the area of rhizosphere research resulting from this work will: 1) challenge our current concept of the mechanisms that the plants implement when exposed to trace metals, 2) help understand the role of roots and organic substances on the biogeochemistry of trace metals in soils and, 3) allow the identification of indicators of the response of soils to environmental stresses such as metal contamination. The knowledge generated by this project will also prove useful for the development of remediation technologies for the cleanup of contaminated soils. Moreover, despite the existence of a substantial body of literature, the published work on the rhizosphere has not been systematically integrated to the study of the associated bulk soil and vegetation components. Our research will help bridge this gap in knowledge by linking the work on trace metal activity and speciation in the bulk soil solution (Hendershot) to that on metal accumulation in the vegetation (Hale). It thus represents a unique study where the soil, the rhizosphere and the vegetation components are integrated both at the plot and the regional scale. This original approach will contribute to ecological risk assessment of anthropogenic activities on soil and forest health, productivity and sustainability.

Top of Page

Trace metal mass balances for Plastic Lake and its catchment

Dillon, P., S. Watmough, Y. Shi and K. Epova

Environmental and Resource Studies and Chemistry Departments, Trent University, Peterborough, ON

The objective of this project is to measure the size of the compartments or pools of metals and the fluxes between compartments in a lake and catchment that are not affected by any local point sources. These results should be useful for comparison with other MITE study sites that are affected by substantial inputs of metals. Plastic Lake is a small, oligotrophic, low ionic strength lake on the Precambrian Shield in south-central Ontario that has been under study since 1979. Both the lake and its catchment have been impacted by the long-range transport of strong acids. The pH of the lake is about 5.6, but is gradually increasing as sulphur deposition declines. On the other hand, the soil continues to deteriorate through continuous net loss of base cations, particularly calcium. The hydrology of the system is monitored, with continuous measurement of the lake outflow and the biggest inflow, both above and below the major feature of this sub-catchment, a Sphagnum-conifer swamp, being carried out. We are measuring (since September 2001) mass fluxes of metals, a well as of DOC, ions and nutrients, from the upland portion of the catchment into the wetland, out of the wetland and into and out of the lake itself. We are also measuring changes in chemistry as water passes through the soil horizons (using lysimeters placed below each horizon in the upland portion of the catchment), within the wetland and in the lake. Beginning this spring, we will measure input to the system from the atmosphere (wet-only deposition and bulk deposition), and changes in flux resulting from the precipitation passing through the canopy (which is mainly coniferous) using throughfall collectors. At this point we are measuring only total metals in aqueous samples rather than specific chemical fractions. We have also measured the forest biomass in the upland catchment as part of another study, and are currently analyzing vegetation (including wood) samples for metal content. In addition, we are measuring the soil and peat metal content to estimate the size of these reservoirs. All analyses are carried out using a collision cell quadrupole ICP-MS and/or a time-of-flight ICP-MS. Preliminary estimates of the size of the metal compartments will be available this spring.