Statistical receptor models

Statistical Receptor Models Solved by Partial Least Squares... 

The two most widespread statistical receptor models in the literature are regression model of chemical mass balance (CMB) and target transformation factor analysis (TTFA) (. ) The questions to be answered by the receptor models are ... 

In this paper the PLS method was introduced as a new tool in calculating statistical receptor models. It was compared with the two most popular methods currently applied to aerosol data Chemical Mass Balance Model and Target Transformation Factor Analysis. The characteristics of the PLS solution were discussed and its advantages over the other methods were pointed out. PLS is especially useful, when both the predictor and response variables are measured with noise and there is high correlation in both blocks. It has been proved in several other chemical applications, that its performance is equal to or better than multiple, stepwise, principal component and ridge regression. Our goal was to create a basis for its environmental chemical application. 

The problem mentioned before has led to the development of statistical receptor models (Fig. 3) which nowadays are the most widely used tools for PM source apportionment. They can be applied even to a single site and need a time series of PM mass concentrations and corresponding chemical composition data. Depending on the method, these analyses are based only on the receptor data or additionally use information on the chemical composition from the relevant emission sources (emission profiles). 

Fig. 3 Statistical receptor model categories and specific models (italics/dashed arrows) (reprinted from [1] as modified from [12] with permission from Elsevier)...

Industrial emissions may contribute significantly to the PM10 burden at selected receptor sites as was also shown using the Lenschow approach. To a large part this assessment is due to the assignment of measured secondary aerosol components to industrial emissions of sulphur dioxide and nitrogen oxides which are not produced locally. Statistical receptor models like PMF, on the other hand, are able to identify local industrial impacts which can be seen in elevated levels of trace compounds, but hardly attribute secondary aerosol compounds to any specific industrial source. For example, a PMF study was carried out for a receptor site located a few... 

There are statistical methods to determine the verisimilitude of experimental data to models. One major procedure to do this is nonlinear curve fitting to dose-response curves predicted by receptor models. 

The two mostly used statistical methods for calculating receptor models are CMB and TTFA. 

A couple of years ago a workshop was organized to compare the performance of the various statistical methods applied for receptor model (12). To create an objective basis for the comparison of the different analyses, a synthetic data set was generated according to the following equation ... 

Among the multivariate statistical techniques that have been used as source-receptor models, factor analysis is the most widely employed. The basic objective of factor analysis is to allow the variation within a set of data to determine the number of independent causalities, i.e. sources of particles. It also permits the combination of the measured variables into new axes for the system that can be related to specific particle sources. The principles of factor analysis are reviewed and the principal components method is illustrated by the reanalysis of aerosol composition results from Charleston, West Virginia. An alternative approach to factor analysis. Target Transformation Factor Analysis, is introduced and its application to a subset of particle composition data from the Regional Air Pollution Study (RAPS) of St. Louis, Missouri is presented. 

The mathematical obfuscation of these models must not remove the requirement that every receptor model must be representative of and derivable from physical reality as represented by the source model. A statistical relationship between the variability of one observable and another is Insufficient to define cause and effect unless this physical significance can be established. 

A widely used approach for estimation of source contributions at receptor sites is receptor modelling [31, 32], In receptor models, source contributions are estimated based on the measurements of various chemical constituents in a sufficiently large number of ambient PM samples, often filter samples that are collected during 24 h. Depending on the available knowledge about the main sources, CMB or multivariate statistical models can be applied CMB requires a priori knowledge of the chemical profile of all relevant sources, i.e. the percentage of the chemical... 

Q)SAR may also be classified as (a) statistically based (relying on statistical, deterministic, or probabilistic association) or (b) mechanistically based (e.g., receptor modeling, electrophilicity-based), or they may be a combination of both. Ideally,... 

In the first part of this chapter a number of receptor modeling approaches will be discussed. These models are used for apportionment of the contributions of each source, identification of sources and their emission composition, and for determination of the spatial distribution of emission fluxes from a group of sources. In the second part, we will develop the tools needed to analyze the statistical character of air quality data. 

Lombardo F, Obach RS, Shalaeva MY and Gao F, Prediction of human volume of distribution values for neutral and basic drugs. 2. Extended data set and leave-class-out statistics, /. Med. Chem., 47, 1242-50 (2004). Ref. 299 = Hogberg T, Ramsby S, de Paulis T, Stensland B, Csorre I and Wagner A, Solid state conformations and antidopaminergic effects of remoxipride hydrochloride and a closely related salicylamide, FLA797, in relation to dopamine receptor models. Mol. Pharmacol., 30, 345-351 (1986). 

Figure 18 Binding of a divalent ligand AA to a divalent receptor model BB for the evaluation of the reference EM value. The statistical factor is obtained by considering that each species has cr = 2 for the presence of a single twofold symmetry axis.

The Noncooperative Model, (a = 7 = 1, c= 0). This model applies to assembhes that involve only intemiolecular interactions without any allosteric effect. The occupation of the various binding sites of the receptor is dictated only by statistics. This model is the reference for spotting the presence of allosteric effects in real systems. It also applies to the formation of a given ohgomer in isodesmic polymerizations. This process is exemplified by a monomer A—B that undergoes a reversible polymerization in which all of the stepwise association constants are identical and equal to K. The formation constant of each oligomer (A—B), is given by Eq. [51] in which a = y = 1, c = 0, = 1 and f) = 1 -... 

The Offshore and Coastal Dispersion (OCD) model (26) was developed to simulate plume dispersion and transport from offshore point sources to receptors on land or water. The model estimates the overwater dispersion by use of wind fluctuation statistics in the horizontal and the vertical measured at the overwater point of release. Lacking these measurements the model can make overwater estimates of dispersion using the temperature difference between water and air. Changes taking place in the dispersion are considered at the shoreline and at any points where elevated terrain is encountered. 

Receptor Exposure. Exposure modeling should produce a statistically representative profile of pollutant intake by a set of receptors. This is done by combining the space/time distribution of pollutant concentrations with that of receptor populations (whether they be people, fish, ducks or property made of some material that is vulnerable to pollutant damage). The accuracy and resolution of the exposure estimates are chosen to be consistent with the main purposes of decision making. These purposes include the following ... 

---