Record Display for the EPA National Library Catalog

RECORD NUMBER: 6 OF 31

OLS Field Name OLS Field Data
Main Title Determination of an Empirically Derived IP/TSP Relationship.
Author Nelson, Jr., A. Carl ; Wijnberg, Luke ;
CORP Author PEDCo-Environmental, Inc., Durham, NC.;Environmental Monitoring Systems Lab., Research Triangle Park, NC.
Year Published 1982
Report Number EPA-68-02-3173; EPA-600/4-82-034;
Stock Number PB82-190554
Additional Subjects Particles ; Air pollution ; Mathematical models ; Concentration(Composition) ; Monitoring ; Comparison ; Lognormal distribution functions ; Outliers
Holdings
Library Call Number Additional Info Location Last
Modified
Checkout
Status
NTIS  PB82-190554 Most EPA libraries have a fiche copy filed under the call number shown. Check with individual libraries about paper copy. 06/23/1988
Collation 70p
Abstract
The primary objective of this study was to provide researchers with statistical methodology for comparing data on inhalable particulate (IP) and on the IP/TSP ratios from various sites, predicting IP concentration as a function of total suspended particulate (TSP) concentration, and detecting potential outliers. Eleven sites were selected for study based on the completeness of both IP and TSP data. The frequency distributions of IP and TSP data indicate that the lognormal distribution provides a better approximation than the normal distribution. No attempt is made to derive numerical results for all the available IP/TSP data because the 11 sites are not a random selection from a population of sites. The data from these 11 sites were compared graphically by means of box plots showing the dispersion of the date, the median value, and specified percentiles. One of the major results was the study of relationships between the measurements of IP and TSP. Two forms of this relationship were compared, a simple linear one and a first order approximation to a model relating the logarithms of IP and TSP. The linear relationship was considered to be satisfactory for most applications; however, the second model form indicates that the simple ratio prediction (i.e., using a constant ratio of IP/TSP over all values of TSP) may not be appropriate for all TSP. Simple statistical test procedures are given for comparing the average ratios at two or more sites (or for two or more years at a single test site), detecting outlying observations, and for selecting IP monitoring sites based on available TSP data and a proposed standard for IP.