Melvin R. Willis, PhD
Donald Bren School of Environmental Science and Management
University of California, Santa Barbara
2003
Despite the relationship between the spatial pattern of regional land use and air toxics exposure, review of the literature found little evidence that community health risk assessment is a significant component of long-range land use policy planning. The principal question preventing development of a risk assessment methodology for regional land use policy analysis involves how to estimate air toxics mass flux for the general categories or zones of land use used in policy planning, given the wide range and inherent uncertainty about potential uses of individual properties.
LUAIRTOX is an interactive spreadsheet model capable of generating individual air toxic emission factors for broad categories of industrial and commercial land uses, normalized by area. It enables a user to generate air toxics emission factors unique to the specific land use policies of a community or region, as embodied in the set of permitted uses for each land use district generally specified in a zoning ordinance. This paper is a brief introduction to the model and instructions for its use.
LUAIRTOX models the relationship:
Q'pz = f (Ψz)
where,
Q'pz = normalized mass flux for pollutant p (g·s-1·m-2)
Ψz = vector of binary variables with 1 indicating use ui is permitted in zone z and 0 indicating the use is not permitted.
f = probability mass function that aggregates the geometric mean emission rate of pollutant p of all facilities of land use type ui after weighting the rate by the probability of occurrence in zone z.
A detailed description of the functional relationship f used in LUAIRTOX, and the census and emission inventory data used to derive the probability mass function, will be published in a paper currently in review.
After obtaining values for Q'p using LUAIRTOX, they can be used to derive mass flux rates Qp for land areas zoned z, assumed to be area sources of pollutant p using the relationship:
Qp = Q'pz
Az
where,
Qp = mass flux for pollutant p (g·s-1)
Q'pz = normalized mass flux for pollutant p from zone j (g·s-1·m-2)
Az = area of unit of zone j (m2)
Once derived, these area source air toxic emission rates can then be input to an atmospheric fate and transport model to simulate the spatial pattern of air toxics exposure within a region.
The fundamental assumption of the model is that the functional relationship f is spatially stationary and that all variation from mean values is random and independent of location. Because California emissions and census data were used to derive the probability mass function, this assumption may not hold true outside the state. The user must be aware that application of the model to its region implies an assumption that the probability of occurrence and pollutant emission rates of the set of uses in the region is the same as in California, and that any differences from mean values are random. It is possible, however, to modify the model to fit another region of the country using a different set of emissions and census data specific to that region.
LUAIRTOX is an Excel workbook containing eight worksheets used to estimate Q'pz for individual zoning districts, given a unique set of permitted uses. All worksheets are protected with cells locked, except those cells for parameter entry.
The following are step-by step instructions for obtaining results.
At this time, permitted uses–the input variables to LUAIRTOX–are only classified at the 2-digit SIC level, which limits the model’s versatility for land use planners, who often require a more detailed breakdown of land uses for groupings into permitted uses. Expansion of the model to the 3-digit SIC level would involve using smaller sample sizes for individual land uses, thus resulting in wider confidence intervals (i.e., a tradeoff between accuracy and precision). We hope to eventually expand our database so that LUAIRTOX can be extended to the SIC 3-digit level and the newer North American Industrial Classification System (NAICS).