Modeling mercury fluxes and concentrations in a Georgia watershed receiving atmospheric deposition load from direct and indirect sources

This paper presents a modeling analysis of airborne mercury (Hg) deposited on the Ochlockonee River watershed located in Georgia. Atmospheric deposition monitoring and source attribution data were used along with simulation models to calculate Hg buildup in the subwatershed soils, its subsequent runoff loading and delivery through the tributaries, and its ultimate fate in the mainstem river. The terrestrial model calculated annual watershed yields for total Hg ranging from 0.7 to 1.1 microg/m2. Results suggest that approximately two-thirds of the atmospherically deposited Hg to the watershed is returned to the atmosphere, 10% is delivered to the river, and the rest is retained in the watershed. A check of the aquatic model results against survey data showed a reasonable agreement. Comparing observed and simulated total and methylmercury concentrations gave root mean square error values of 0.26 and 0.10 ng/L, respectively, in the water column, and 5.9 and 1 ng/g, respectively, in the upper sediment layer. Sensitivity analysis results imply that mercury in the Ochlockonee River is dominated by watershed runoff inputs and not by direct atmospheric deposition, and that methylmercury concentrations in the river are determined mainly by net methylation rates in the watershed, presumably in wetted soils and in the wetlands feeding the river.     

An application of parametric and nonparametric models to the assessment of fluoride levels in vegetation exposed to stack emissions of an aluminum reduction plant in Greece

Various statistical models were developed for assessing airborne fluoride (F) levels in natural vegetation near an aluminum reduction plant using as predictor variables the distance from the emission source, the predominating wind, and characteristic topography-geomorphology parameters. Results revealed that F concentrations in vegetation showed a predictable response to both wind conditions and landscape features. The linear model was found to give good estimations, taking advantage of the relatively strong linear correlation between concentration and distance. A nonlinear relationship between the F concentration in vegetation and the other variables was also found, while interactions between the variables were found to be non-first-order. The nonlinear relationship hypothesis was supported by the improved results of various nonlinear models that also indicated the importance of the area's topography-geomorphology and meteorology in model predictions. The application of an artificial neural network (ANN) model showed the closest agreement between predicted and observed values with a correlation coefficient of 0.92. The improved reliability of the ANN and a regression tree model (RTM) also were indicated by the normal distribution of their residuals. The RTM and the ANN were further investigated and found to be capable of identifying the importance of the variables in vegetation exposure to air emissions.     

Modeling assessment of air emission flux of mercury from soils in terrestrial landscape components: model tests and sensitivities

The abilities of a screening-level model to predict variations in elemental mercury (Hg0) air emissions from soils in terrestrial landscapes are examined by comparing simulation results to published observational data and by performing sensitivity analyses. Despite uncertainties and simplifications, the model results obtained offer some degree of confidence in the model's joint ability to relate readily available environmental parameters to airborne emissions of Hg predicted by coupling simple atmospheric and soil parameters with Hg cycling and transport algorithms. The model reasonably predicted the observational data in the considered data sets except for one site for which significant uncertainty was associated with model input data. Predictions are consistent with many trends observed in the field studies; better predictions were obtained for nonvegetated systems (relative errors between 0.4 and 9.7%) than for shaded-soil landscapes (relative errors between 2.3 and 27%). The model reflected field data showing that daily average emission rates of Hg0, formed by the reduction of Hg(II), are primarily controlled by changes in solar radiation, soil moisture, temperature, and, to a lesser extent, wind conditions. The model may have potential use in several preliminary studies to characterize trends of airborne Hg emitted from terrestrial sources to the atmosphere.     

Combining neural network models to predict spatial patterns of airborne pollutant accumulation in soils around an industrial point emission source

Neural networks (NNs) have the ability to model a wide range of complex nonlinearities. A major disadvantage of NNs, however, is their instability, especially under conditions of sparse, noisy, and limited data sets. In this paper, different combining network methods are used to benefit from the existence of local minima and from the instabilities of NNs. A nonlinear k-fold cross-validation method is used to test the performance of the various networks and also to develop and select a set of networks that exhibits a low correlation of errors. The various NN models are applied to estimate the spatial patterns of atmospherically transported and deposited lead (Pb) in soils around an historical industrial air emission point source. It is shown that the resulting ensemble networks consistently give superior predictions compared with the individual networks because, for the ensemble networks, R2 values were found to be higher than 0.9 while, for the contributing individual networks, values for R2 ranged between 0.35 and 0.85. It is concluded that combining networks can be adopted as an important component in the application of artificial NN techniques in applied air quality studies.     

A screening model-based study of transport fluxes and fate of airborne mercury deposited onto catchment areas.

Dynamics of airborne mercury deposited onto catchment areas is investigated within the framework of a simulation model. Model results show that, for a particular atmospheric deposition rate, significant interannual variability in mercury transport flux in catchments is caused by climatology and corresponding differences in catchment soil loss rates; in comparison to the normal year, runoff flux increased by a factor of 2-3 for the wet year (rainfall 35% above normal) while for the dry year (rainfall 18% below normal) runoff flux decreased by factors of 5-7. The interaction of parameters describing soil type, topography and vegetation cover causes variability in both transport and emission fluxes among catchments; as soil loss rate increases by a factor of 5 due to variations in these parameters among the examined catchments, annual average transport flux increases by a factor of 3; and annual average emission flux of mercury (as Hg0) from soil to the atmosphere decreases by a factor of 2 due to the decreased levels of soil mercury associated with catchment soil loss increases. Seasonal variability of transport flux is associated with seasonal changes in precipitation and soil loss rates while seasonal changes of emission flux are primarily due to changes in soil moisture regime and temperature. Although modeled results are consistent with observational data from previous studies, they must be interpreted in a relative sense due to the screening-level character of this study.