Sites and Spills. The development of a chemical signature profiling techniques for toxic release sites based on outcomes for all chemical reports filed for a given site in relation to the Standard Industrial Classification (SIC) value(s) assigned to each site.
The development of a SIC-reclassification techniques used to correlate toxic chemical release histories for different release sites.
Several methods have been attempted to classify and pull together industries with similar chemical histories and related health risk histories. These reclassification were produced relying upon the 450 CRI sites, and then used to fill in the few CRI sites with no chemical reports filed, and other TRI sites considered important due to their SIC, but not yet fully documented in the state database.
To analyze site spills or releases in relation to SIC, we have to take several types of site and chemical data into consideration. First, each site has a unique spill/release history in terms of numbers of spills, types of spills, length of spills time-wise, and chemistry of spills. In particular, the variation in chemicals from one site to the next can be significantly different, even if the sites bear very similar industrial types and histories. In spite of these differences between seemingly similar sites, there are also certain features which sites with a common background share. Most often sites with similar industrial or land use backgrounds do have some commonalities in regard to chemical history. A site that is related to farming will more often than not have issues pertaining to fertilizer, herbicide, pesticide, and gas/diesel-powered vehicles use, a site related to dry cleaning typically has just two chemicals commonly assessed as part of its chemical release field testing process, a paper industry has a unique set of chemicals related to paper manufacture which are significantly different from chemicals related to wood processing (i.e. plywood manufacture) or lumber (log-cutting) and tree-cutting industries.
Applications of the data to Evaluating Claims related to Wood Products Industries
Second, it is important to remember that these differences that exist from one industry to the next are fairly distinctive, but if analyzed and mapped properly, they can be used to demonstrate how certain kinds of industries are responsible for specific risks whereas others are not, or at least less likely to be a source of concern. In general, the chemicals they do release make them distinctly different from other kinds of industries with the same SIC and which the general public may feel is responsible for the “leak” responsible for a recent “outbreak” of a specific form of cancer.
We can partially distinguish the different industries and their chemical histories from each other by the use of the Standardized Industrial Classification (SIC) identifier assigned to the site. The SIC identification process, developed decades ago by agencies overseeing the overall economic importance of the various industries researched and monitored by these agencies, relates to the level at which the industry resides in a comparative industrial evolutionary sense, and in turn relates to how the site creates its products and may or may not be a reason for serious chemical exposure.
In general, SIC classes (the 4-digit version), begin with its lowest values (<<1000) indicating natural resource industries such as mines, gravel pits, quarries, natural water sources, very crude petroleum industry materials such as tar, etc. The next level up refers to a certain amount of production of goods by the use of these resources, such as primitive stone products useful for sculpting, etc. The next levels up (about 3000 to 5000) refer to various aspects of the industries that make use of these natural resource products, for example the various stages involved with the production and preparation of various wood products (from cut log to pulp to paper, or shaved log to veneer to plywood product), or the levels of production of treated metal alloy-based steel bars and rods, which in turn become machined objects such as screws and bolts, which in turn become frames, edging, trim, and the like. In the case of electronics, this level of industry might also refer to other levels of natural resource use at the factory level, such as the conversion of metal through some sort of electrolysis treatment required for chrome plating, or the conversion of sheet metal into some form of container meant to house or store other products that are kept. At the 5000 level, industries are pretty much involved in some sort of factory level trade. The product they produce is very much modified from its original resources forms, and typically not only includes one type of resource such as metal, but also other numerous products formed based on other material types, such as a computer, toaster or metal garbage can. The 6000 and 7000 levels represent sellers, distributors, and storage companies for the product. The 8000 level refers to upper level distribution features (smaller stores for example). The 9000 level is pretty much a catch-all for all remaining work positions related to that product. For the most part, this is where government facilities, agencies, airports, etc. are located in the classification scheme for SIC coding.
SIC can be related to industries at the GIS level. Although there are similarities that can be found from one industry type or class to the next, SIC can still serve as a starting point for engaging in this type of research, and then modified in order to more carefully apply the SIC classification so it can used to predict the types of releases that come from industrial/landuse sites based on their SIC identifier. To produce a modified SIC for application to industry chemical release monitoring, numerous industries with identical SICs and those with closely related SICs have to analyzed and compared as groups to determine the primary chemicals each type of industry is related to. Once a series of sites are reviewed and considered similar in chemical release type, they can be summed and use to produce an average value per site per chemical group being profiled for whatever sites are being studied. These averages can even be used to replace missing data in sites where studies of release have not been performed and documented in the toxic release site database. A significant amount of time was spent developing profiles for each of the sites and their releases, followed by a reclassification of sites when different SICs appeared to have very similar release histories. In turn, the development of these chemical “signatures” of specific types of sites reclassed into the same group can be applied to other sites that have a SIC which falls into the range of possible SICs with a specific type of release history.
A comparison of chemical risks assigned to 43 SIC Reclass groups in relation to their potential for carcinogenicity and tumorigenicity.
For the state of Oregon, the data on chemicals released by all 450 Confirmed Release Sites were downloaded and evaluated to produce these chemical signature profiles, in relation to SICs. When the SICs were reclassified, 43 signature types or “chemical fingerprints” were developed that were in turn applied to other sites lacking in chemical report data. The following bar graphs, maps and project titles are examples of the outcomes for this stage in this Oregon GIS research project.
Chemical signatures for 3 types of Forest Industry
Likewise, chemical industries in the general sense are different from more specific chemical-based manufacturing settings, which in turn differed from metal-based manufacturing facilities or electronic industries, for example:
A comparison of chemical histories of sites with identical industry class histories but uniquely different SIC and detailed industrial histories: A comparison of different chemical exposure types and exposure risk for the various wood and lumber products industries in Oregon.
An evaluation of sites by SIC in the Trimet Region
By evaluating SIC data we can develop maps depicting specific types of chemical release sites related to specific industry types. This provides more the type of information that is valuable to focused studies on particular forms of waste management, chemical release, etc. The following depicts the automotive industry in the Portland area based on SIC values. This method may be more applicable to studies of sites that result in high public concerns due to their local chemical histories, such as paper industries, dry cleaning businesses or petroleum storage sites. In this case, the automotive sites are reviewed due to their history as fairly hazardous disposal sites, often not at all associated with bearing a particular class of known carcinogens but rather simply residual petrochemicals and metal and paint debris.
Automotive Industry chemical release is dependent upon the major activities taking place at these sites.
Brian Altonen, MPH, MS 