4.2	METHODOLOGICAL OPTIONS FOR ANALYSIS OF THE ATMOSPHERIC DEPOSITION PATHWAY

The mechanism and influences governing the atmospheric deposition pathway are complex. In contrast, for example, to the direct loading pathway - in which liquid effluent is discharged directly into the lake - air emissions are subject to advection, dispersion, transformation, deposition and possible re-emission processes between the source and the lake. The characteristics of weather patterns between the source and the lake influence this pathway significantly, and, different pollutants can exhibit quite different behavior in the atmosphere. In general, pollutant-specific atmospheric modeling is necessary to understand the linkages between sources and the lakes. As modeling is an inherently uncertain process, evaluation of the validity of the simulation results -- generally by comparison against ambient monitoring data -- is crucial.

There are several distinct methodological approaches available to develop information on the atmospheric deposition pathway. Three major approaches are summarized in Table 3: semi-empirical loading estimates, back-trajectory calculations, and comprehensive modeling.

In the semi-empirical loading approach, ambient measurements are used to estimate the deposition of a given pollutant to a given lake. Analyses of this type include those by Hoff et al. (1996) and Eisenreich and Strachan (1992). In this approach, the extent to which the available ambient measurements are also representative of the concentrations over the entire lake - information which is sometimes not known - will strongly influence the accuracy of the estimates. In addition, it is difficult to develop any information about source-receptor relationships using only this method. ⁽¹⁾ That is, an estimate of the total deposition to a given lake can be made, but the sources of this deposition will remain unidentified.

In receptor-oriented techniques, information about sources is inferred from measurements at the receptor. In one common receptor-based methodology - the back-trajectory approach - the paths of air masses that arrive at a particular location are estimated based on a meteorological model. Generally, locations with ambient monitoring data are selected for analysis. A common technique is to investigate where the air masses, associated with high (or low) ambient measurements at the study site, originated. This yields information on source regions expected to be large (or small) contributors to the observed concentrations. If the study site is sufficiently representative of the concentrations of the specific Great Lake as a whole, then inferences can be made regarding the likely source regions of importance to the lake. In addition, when the back-trajectory analysis results are compared to emissions inventory information, inferences regarding the likely source categories contributing significantly to the lake can sometimes be made. Examples of back-trajectory analyses applied to Binational Toxics Strategy (Canada and United States 1997) pollutants in the Great Lakes region include those by Hoff et al. (1992a, b), Gao et al. (1996), and Blanchard et al. (1997).⁽²⁾ Biegalski et al. (1998) applied several alternative receptor-based methodologies (factor analysis, elemental ratios, and enrichment factor analysis) to investigate source-receptor relationships in the Great Lakes basin.

In the comprehensive modeling approach, an emissions inventory of the pollutant of interest is used as an input to an atmospheric fate and transport model. The model attempts to simulate the transport, dispersion, transformation, and deposition of the pollutant emitted from each source in the inventory. Examples of this approach for one or more BTS pollutants in the Great Lakes basin include the analyses of Voldner and Schroeder (1989), Clark (1992ab), Shannon and Voldner (1995), and Cohen et al. (1995, 1997b).

In the work described here, the comprehensive modeling approach has been taken, using a particular fate and transport model (HYSPLIT). In the following sections, the general methodology used will be briefly described. Following these general methodological considerations, preliminary results for dioxin will be presented.

1. In this chapter, the term "receptor" is used to signify a lake or lake basin.

2.Interpretation of back-trajectory analytical results can be complicated by a number of factors. The distance between a known or hypothesized source location and the measurement site obviously influences the extent of its measured concentrations; dispersion, deposition, and chemical transformation en route will all generally serve to lower the atmospheric concentration of the pollutant with increasing distance from any given source. These processes are dependent on the physical-chemical properties of the particular pollutant and the many features of the atmospheric environment encountered by the polluted air parcel in between the source and the measurement site. These factors can make it difficult to determine the location - on a given back trajectory - of the significant source(s).