Project at a Glance
Title: Industrial surface coatings-wood furniture and fixtures emission inventory development.
Principal Investigator / Author(s): Anex, Lund, Chang
Contractor: UC Davis
Contract Number: 93-343
Research Program Area: Emissions Monitoring & Control
Topic Areas: Stationary Sources
For estimation purposes, the broad range of coatings used in coating wood furniture must be divided into subcategories. In any statistical estimation procedure, dividing a sample into smaller groups can increase the accuracy of the estimate by reducing the "averaging" effect of lumping disparate items together. However, too much of this division process can also decrease accuracy because it necessarily reduces the number of samples within each subcategory and can thereby increase the statistical variance. Thus subcategories should be chosen to minimize estimation error by balancing the error causing impacts of aggregation and disaggregation of coating categories. This study has involved three formal survey processes. Formulations of wood furniture and fixture coatings have been evaluated through a survey of coating manufacturers, and furniture manufacturing practice and coating use data have been gathered through two mail surveys of coating applicators. These surveys were complemented by extensive telephone conversations with those in the industry, site visits to industrial facilities, and discussions with local air quality management personnel. The number of survey responses to all surveys was surprisingly low (11.4% of questioned applicable firms). Applicators that responded to the use surveys indicated that they felt burdened by local regulation and reporting requirements and thus were often unwilling to answer what they saw as another governmental request for data. Manufacturers of furniture generally feel regulation imposes costs on them in terms of both purchase of required coatings and application equipment and in time spent responding to reporting requirements and frequent changes in regulation. The lower than expected response rate to repeated survey efforts has resulted in emission estimates that are based on limited data and higher than anticipated levels of estimate uncertainty. However, the resulting emission estimates are sufficiently accurate to be useful and reveal interesting emission patterns and trends. These statistical impressions appear to be confirmed by conversations with those in the industry.
Estimates of VOC emissions from the industrial coating of wood furniture and fixtures are tabulated in several ways in this report to facilitate different uses of the data. Emission estimates computed from the "bottom-up" approach described above are presented grouped by county, air district, and air basin, as well as by coating category and firm category. These appear in Tables 8-2, 8-3, and 8-5. Statewide, 1993 VOC emissions have an estimated mean of 8,330,000 lbs/year, and a 90% chance of falling between 7,791,000 and 9,907,000 lbs/year. Additional emission estimates are made using a "top-down" approach based on coating production. These appear, by county and air basin, in Chapter 9. Comparison of the results of the "bottom-up" and "top-down" estimates (Table 9-2) indicate that the bottom-up estimates are reasonable, though about a third smaller statewide. County estimates are re-produced and summarized in Table ES-1. Estimate uncertainty is large, particularly regarding the small number of large firms. Each estimation method has its own biases and it is difficult to conclude which method yields more accurate estimates. However, the independent estimates typically support each other, providing some confidence that actual emissions lie somewhere in the neighborhood of these estimates.
For questions regarding this research project, including available data and progress status, contact: Heather Choi at (916) 322-3893
Stay involved, sign up with ARB's Research Email Listserver