Hybrid receptor model

Han, Y.-J., Holsen, T.M., Hopke, P.K., Cheong, J.-P., Kim, H., Yi, S.-M., 2002. Identification of source locations for atmospheric dry deposition of heavy metals during yellow-sand events in Seoul, Korea in 1998 using hybrid receptor models. Atmos. Environ. 38, 5353-5361. 

Lewis and Stevens (This Volume and 12) have provided a useful framework for hybrid receptor modeling in which one calculates concentrations of various sulfur species relative to those of some tracer that is fairly unique to the sulfur source, e.g., Se as a tracer from coal-fired power plants. Equations are written for S02 conversion... 

One of the most important projects in progress in the field of hybrid receptor modeling is the Allegheny Mountain study by Pierson et al. of Ford Motor Co. (This Volume). Concentrations of many ions, major, minor and trace elements in airborne particles, rain, dew and fog and other parameters were measured at Allegheny Mt., PA and Laurel Hill, 35 km to the northwest, from 5 to 28 Aug 1983, approximately simultaneously with the Deep Creek Lake studies discussed above. These two huge data sets are now nearly complete and ready for detailed interpretations by the participants and other researchers in the field. In particular, Keeler is working with the Ford group to apply the Samson method to the data. 

In my view, hybrid receptor models are the most likely approach for provide reasonable answers to the sulfate deposition problem within a time that they might be of use in influencing controls that may be imposed on S02 and NO sources. This does not mean that there is no need for further field studies. The Allegheny Mt. and Deep Creek Lake data sets were taken so close together that one would feel much safer if similar data were available at several other sites, e.g., the three sites of the Ohio River Valley study (20) and one or two sites to the northeast of Allegheny Mt.,... 

Tracer Hybrid Receptor Model (Lewis). Lewis and Stevens (3) have derived a hybrid receptor model for describing the secondary sulfate from an SO2 point source. The resulting expression for secondary sulfate concentration M o at the receptor has the form... 

Source Finding Hybrid Receptor Model (Yamartino). The starting point of this model (5) is a iet of equations of the form... 

Diffusion Hybrid Receptor Model (Fay). This approach, beginning with the work of Fay and Rosenzweig (7), is perhaps the most interesting of all the hybrid models that have been proposed to date. Not only is it able to address the usual source apportionment problem of estimating source impacts (of SO2 and secondary sulfate) at a receptor site but it simultaneously generates estimates for the conversion and deposition rate constants and meteorological parameters that are influencing the pollutant transfer between source and receptor. Consequently, we choose to review this model in more detail than the others considered here. 

Tracer Hybrid Receptor Model Application. Even though the Deep Creek Lake data base is not yet complete, in terms of all the chemical analyses that are anticipated to be included, it may be useful to give a simple illustration of how a hybrid receptor model can provide some insight into the data. 

Under the assumption that the gaseous sulfur, fine particle selenium and secondary fine particle sulfur measured at an ambient site originates from a single point source the tracer hybrid receptor model can be expressed in terms of the two equations... 

If CMBs can be used to distinguish between primary and secondary emissions from coal-fired plants, the results can be used in hybrid receptor models to determine important parameters such as distance scales for transport and transformation of S species. 

The meaning of the the term "hybrid receptor model" is not consistent in the literature. Following the definition proposed at the Quail Roost Receptor Modeling Workshop (15), we take it to be a combination of some meteorological aspects of traditional source-based models with some tracer aspects of receptor models. An important feature of such models is that one often works with ratios of species so that some of the most uncertain absolute parameters of classical models cancel out. As noted below, for example, one can calculate the concentration ratio of gas-phase SO2 to gas-phase B as a function of distance from a common source more accurately than the absolute concentration of either species. 

An important step towards treatment of SO2 conversion to sulfate and deposition of both species that avoids absolute uncertainties of dispersion and deposition rates was taken by Lewis and Stevens, who investigated the mathematical basis of one form of hybrid receptor modeling (.16). Their model assumes that one measures concentrations of SO2 and SO4 relative to that of some species borne by particles from the plant. They assumed that (1) dispersion, deposition and transformation of the three species (SO2, SO4 and fine primary particles) are linear or pseudo first-order processes, but may have complex dependences on time (2) dispersion affects all three pollutants identically (3) dry deposition is the only type of deposition which occurs (4) deposition velocity is the same for all fine particles, but may be different for SO2 (5) secondary sulfate is produced only by homogeneous oxidation of SO2. 

In our more detailed calculations to test hybrid receptor modeling against these new data, we relaxed the assumption of a constant SO2 density vs. distance up the Valley, and put in estimates based on the SURE emissions inventory (21). Results from two calculations are shown in Figures 2 and 3. We started the model far to the SW, in Texas, to avoid boundary problems at the lower... 

Figure. 2. Hybrid receptor model predictions of concentrations of SO2, SO4, particulate Se and the S/Se ratio from Texas up through the Ohio River Valley. Experimental data from XRF studies of Shaw and Paur (1) for southwest back-trajectories are connected by dashed lines.

Application of observed concentrations of atmospheric particles and acidic gases are compared to the results predicted by a hybrid receptor model. 

In this study we have employed the simultaneous collection of atmospheric particles and gases followed by multielement analysis as an approach for the determination of source-receptor relationships. A number of particulate tracer elements have previously been linked to sources (e.g., V to identify oil-fired power plant emissions, Na for marine aerosols, and Pb for motor vehicle contribution). Receptor methods commonly used to assess the interregional impact of such emissions include chemical mass balances (CMBs) and factor analysis (FA), the latter often including wind trajectories. With CMBs, source-strengths are determined (1) from the relative concentrations of marker elements measured at emission sources. When enough sample analyses are available, correlation calculations from FA and knowledge of source-emission compositions may identify groups of species from a common source type and identify potential marker elements. The source composition patterns are not necessary as the elemental concentrations in each sample are normalized to the mean value of the element. Recently a hybrid receptor model was proposed by Lewis and Stevens (2) in which the dispersion, deposition, and conversion characteristics of sulfur species in power-plant emissions... 

Using both a hybrid receptor model, developed by Lewis and Stevens ( 2) and modified by Gordon and Olmez (3), and a simple model of emission from the Ohio River Valley, we compare the results of the College Park (CP) samples as well as those of another continuous set of samples taken from July 3-29, 1983 at Wallops Island, VA (WI), to predicted results. Single-source differential equations (2) are used to describe the time-varying concentrations of SO2, SO and a particulate element characteristic of coal-fired power plant emissions (chosen here as Se). An additional equation (3) can be added to describe the concentration variation of B(0H)3 The following rate constants apply to the concentrations of the four species in question ... 

Table II. Comparison of Hybrid Receptor Model with Observed...

---