Sediment processes and mercury transport in a frozen freshwater fluvial lake (Lake St. Louis, QC, Canada)

An open-bottom and a closed-bottom mesocosm were developed to investigate the release of mercury from sediments to the water column in a frozen freshwater lake. The mesoscosms were deployed in a hole in the ice and particulate mercury (Hg_P) and total dissolved mercury (TDHg) were measured in sediments and in water column vertical profiles. In addition, dissolved gaseous mercury (DGM) in water and mercury water/airflux were quantified. Concentrations of TDHg, DGM, and mercury flux were all higher in the open-bottom mesocosm than in the closed-bottom mesocosm. In this paper we focus on the molecular diffusion of mercury from the sediment in comparison with the TDHg accumulation in the water column. We conclude that the molecular diffusion and sediment resuspension play a minor role in mercury release from sediments suggesting that solute release during ebullition is an important transport process for mercury in the lake.     

Estimation of Emissions of γ-Hexachlorcyclohexane from the Great Lakes–St. Lawrence Ecosystem Using a Coupled Efficient Kalman Filter–Atmospheric Transport–Soil Model

An efficient linear Kalman filter has been combined with a coupled atmospheric transport and soil–air exchange model to determine organochlorine pesticides emissions on the regional scale. In this study, results of -HCH emissions from the Great Lakes–St. Lawrence ecosystem, estimated from the coupled model, are presented and discussed. A source receptor technique is used to identify a priori the locations of emission sources of -HCH, the emissions are then updated through a Kalman filtering procedure which minimizes the weighted difference between the predicted mixing ratios from the coupled model and the measured concentrations over the Great Lakes–St. Lawrence river region. Two experiments using the inverse algorithm are carried out. In the first experiment, the coupled atmospheric transport and soil–air exchange model is implemented to predict -HCH air and soil concentrations. Emissions are then updated every 12 days using the updated soil concentrations and emission factors. However, the updated emissions are not input into the coupled atmospheric transport and soil–air exchange model. On the other hand, in the second experiment the updated emissions are fed back to the coupled model, so that the model is reinitialized in each 12 days. The results from the inverse technique for the year 1995 have been compared with grided -HCH emission inventory in Canada, generated by emission factors. It is shown that the estimated emissions of -HCH are consistent with the measured emissions. It is found that the St. Lawrence valley has larger emissions of -HCH than the Great Lakes region, indicating an opposite distribution to the emission usage inventory, but in agreement with the measured -HCH concentration.     

Mercury in sediments and vegetation in a moderately contaminated salt marsh (Tagus Estuary, Portugal)

Depth variations of total mercury (Hg) and methylmercury (MeHg) concentrations were studied in cores from non-colonized sediments, sediments colonized by Halimione portulacoides, Sarcocornia fruticosa and Spartina maritima and belowground biomass, in a moderately contaminated salt marsh (Tagus Estuary, Portugal). Concentrations in belowground biomass exceeded up to 3 (Hg) and 15 (MeHg) times the levels in sediments, and up to 198 (Hg) and 308 (MeHg) times those found in aboveground parts. Methylmercury in colonized sediments reached 3% of the total Hg, 50 times above the maximum values found in non-colonized sediments. The absence of correlations between total Hg concentrations in sediments and the corresponding MeHg levels suggested that methylation was only dependent on the environmental and microbiological factors. The analysis of belowground biomass at high-depth resolution (2 cm) provided evidence that Hg and MeHg were actively absorbed from sediments, with higher enrichment factors at layers where higher microbial activity was probably occurring. The results obtained in this study indicated that the biotransformation of Hg to the toxic MeHg could increase the toxicity of plant-colonized sediments.     

Mercury transport between sediments and the overlying water of the St. Lawrence River area of concern near Cornwall, Ontario

Contaminated sediments in the St. Lawrence River remain a difficult problem despite decreases in emissions. Here, sediment and pore water phases were analyzed for total mercury (THg) and methyl mercury (MeHg) and diffusion from the sediment to the overlying water was 17.5 ± 10.6 SE ng cm⁻² yr⁻¹ for THg and 3.8 ± 1.7 SE ng cm⁻² yr⁻¹ for MeHg. These fluxes were very small when compared to the particle-bound mercury flux accumulating in the sediment (183 ± 30 SE ng cm⁻² yr⁻¹). Studies have reported that fish from the westernmost site have higher Hg concentrations than fish collected from the other two sites of the Cornwall Area of Concern, which could not be explained by differences in the Hg flux or THg concentrations in sediments, but the highest concentrations of sediment MeHg, and the greatest proportions of MeHg to THg in both sediment and pore water were observed where fish had highest MeHg concentrations.     

Misspecification of within-area exposure distribution in ecological Poisson models

Ecological studies enable investigation of geographic variations in exposure to environmental variables, across groups, in relation to health outcomes measured on a geographic scale. Such studies are subject to ecological biases, including pure specification bias which arises when a nonlinear individual exposure-risk model is assumed to apply at the area level. Introduction of the within-area variance of exposure should induce a marked reduction in this source of ecological bias. Assuming several measurements per area of exposure and no confounding risk factors, we study the model including the within-area exposure variability when Gaussian within-area exposure distribution is assumed. The robustness is assessed when the within-area exposure distribution is misspecified. Two underlying exposure distributions are studied: the Gamma distribution and an unimodal mixture of two Gaussian distributions. In case of strong ecological association, this model can reduce the bias and improve the precision of the individual parameter estimates when the within-area exposure means and variances are correlated. These different models are applied to analyze the ecological association between radon concentration and childhood acute leukemia in France.