Saint-Gobain, Northampton, MA

This is a copy of an old post from my personal blog, placed here so that all data posts will be in one place.

The fallout from the USA Today report mentioned in the last post is still settling, with politicians and Federal and state regulators promising various responses. I've heard that there's the usual rush of facilities, after a TRI report like this is released, to correct their reporting errors and/or explain their reports. But I'm getting involved with this locally too; a school that I've often biked past, and that some of my friends' kids go to, is listed as the third most potentially polluted from these industrial sources in Massachusetts.

Update: after new information was supplied by the facility, this is probably nothing to worry about: see below.

Looking at the site, 99% of the estimated risk is from a single polluter here, a Saint-Gobain facility. Looking up their TRI data here, as well as their RSEI data, it's clear that the estimated risk is due to their report of an air release of chromium compounds.

I've seen many, many facilities get themselves bad publicity by reporting in the way that this facility did, when they report something like chromium compounds or diisocyanates, something with a high toxicity or carcinogenicity. They reported an air release using a release range: 11-499 pounds. EPA routinely transforms this into the midpoint of the range: 250 pounds. That's a lot of chromium. In addition, EPA doesn't have respondents distinguish between hexavalent chromium, a known human carcinogen (and the chemical that Erin Brockovich campaigned against), and trivalent chromium, which isn't anywhere near as bad.

The facility should clarify this so that everyone can get a better idea of how much concern there should be. I called their TRI public contact on the phone, left a message, and got no response -- that's no surprise; I've been working with TRI data since 1991 and have never, ever gotten a response from a public contact. So I sent the following Email to their technical contact:

Dear Mr. (redacted): I am Emailing you because you are listed as the TRI (Toxic Release Inventory) technical contact for the Saint-Gobain Ceramic Materials facility in Northampton. I am an independent researcher living in Northampton who has worked with TRI data for some time, as well as the RSEI data used to make the USA Today report that references your facility at: http://content.usatoday.com/news/nation/environment/smokestack/polluter/465 As you can see from this USA Today report, the Saint-Gobain facility causes the Montessori School of Northampton to rank in the top two percentile of schools nationwide in terms of schools whose air is potentially polluted by industrial facilities, as well as potentially affecting other Northampton schools. Looking at your TRI report at: http://data.rtknet.org/tri/tri.php?facility_id=01060SNTGB175IN&reporting_year=2005&datype=T&reptype=f&detail=4 and at the RSEI data, it is clear that the potential pollution in question is listed in your report as a release range of 11-499 pounds of chromium and chromium compounds to the air in 2005. EPA routinely treats release ranges of this sort as being equivalent to the midpoint of the range: 250 pounds. I have some questions about your TRI report and the facility's operations: 1. Do you have a better idea of how much chromium and / or chromium in compounds is actually released than 11-499 pounds? Getting a more accurate number could affect the RSEI risk screening calculation quite dramatically. 2. Do you have an idea, through testing or other means, of how much of the chromium is hexavalent chromium -- a known human carcinogen -- vs how much is trivalent [note: I originally wrote this as "chromium trioxide", oops.]? That would also lead to a better understanding of the potential risk involved. 3. Are there any plans possible or underway to reduce use of, or emissions of, chromium? Your 2006 TRI report, the latest publicly available, shows the same chromium release as in 2005. Thank you for your attention to this request for information. I plan on sharing your reply with other people who may be concerned about the chromium emissions from the Saint-Gobain Ceramic Materials facility. If you wish to contact me, I can be reached through Email, or by phone at xxx-xxx-xxxx. Best regards, Rich Puchalsky cc: (redacted), Montessori School of Northampton

I'll see what I response I get. Given that Saint-Gobain is a multinational, I'd guess that my missive may well get passed up quite a chain. But I plan on continuing to pursue this.

My guess is that the range reporting may well have inflated their reported number to be greater than what they actually released. But there's no way to know without asking them.

Update (12/12/2008): The facility says that the level of hexavalent chromium in the chromium they use is very low, so I'd think that this means there's no reason for undue concern. Their reply was:

We produce various ceramic powders used by our customers in coating applications. One of our products is trivalent chromium oxide used in wear resistance applications. This chromium product contains well over 99% trivalent chromium oxide. We have tested for hexavalent chromium oxide and the concentration tends to be around 50ppm in our product. As stated in the USA Today website that you reference “Chromium 3 (trivalent) is an essential nutrient and helps the body process proteins, sugars, and fats.” We believe that the emission values that you reference are a conservative estimate and the actual values are likely much less. The trivalenet chromium is present as a small particle like dust. We process our internal process air through “dust collectors” that are designed to remove 99.99% of the dust in the air. This treated air is then vented back into the building. We also utilize a plant wide central vacuum system for cleaning floors and equipment to minimize dust generation during cleaning. Therefore any trivalent chromium emissions are simply any minimal dust that may escape through open doors. As you can imagine the actual number would be difficult to measure, but we feel that the actual number is closer to 11 lbs than to 499 lbs per year.

My reply was:

Thank you for your reply. The RSEI model used to estimate the risk from your facility's chromium emissions assumes that air emissions of chromium are particulates, as is the case for your facility, but it also assumes, based on industry averages, that the percentage of hexavalent chromium in the chromium released would be 34%. An actual hexavalent chromium percentage of 1% (to round up) would then reduce the estimated risk to 1/34 of the original calculation. That would lower the concern about emissions from your facility to the point where the Montessori school, and other schools in Northampton, would no longer score high on a nationwide or statewide comparison of the kind used in the USA Today report. I should mention, however, that as far as I know, TRI estimates of releases are supposed to be made for the amount leaving the facility, not the amount "released" internally and then recollected before it reaches the outside. It's good to know that you have processes in place to reduce emissions, but the 11-499 pound estimate is presumably your estimate for the amount released to the air after your emission-control equipment has worked, not before. I encourage you to develop a better estimate of how much chromium is actually released, sufficient for you to report to TRI with your best estimate in pounds rather than as a range. Until that is done, the public can only assume that 250 pounds is the best guess, which leads to more concern about your facility than is warranted if the true number is actually much lower. Hexavalent chromium is by far the primary concern for air emissions, but there are some environmental processes that can change trivalent to hexavalent chromium under some conditions, so it's still of interest to people to know how much is being released.

And that is that. Really what is needed is some way to change TRI reporting to discourage this kind of report from happening -- reporting chromium speciation would be good (people already report dioxin speciation to TRI, so it can be done), and there might be a category of highly toxic chemicals that range reporting is not used for, or at least warned about. I'm going to take this up with EPA once the new administration is in place.

Toxic air and America's schools

This is a copy of an old post from my personal blog, placed here so that data posts will be all in one place.

There's an excellent report by USA Today that uses EPA reported pollution data and air modeling to estimate air quality at schools nationwide. It has a good database-backed Web site, too, that lets you look up your school, the schools with the worst air quality from those sources in your state, and so on.

I was involved very tangentially in this project -- I helped to work on data for PERI that USA Today used. I'm impressed by what USA Today did with it; it's better than the usual environmental toxics story. What PERI did was a bit complicated to explain, but I'll make the attempt -- there's a database, TRI, in which large industrial sources report their toxic pollution. A project within EPA, RSEI, takes the reported pounds of air pollution from each facility and runs an air model to see where the chemicals are going, geographically. They can then use Census data to see how many people live in each area affected by the pollution. They add all of this up into an overall risk screening score for each facility. PERI realized that these data could be obtained for each location on the ground, instead of being all added up to a single score for each facility. That lets you find the contribution of each polluter to a particular place where people live. (Or, in this case, where a school is.)

I only wish that I had the resources that USA Today does; their database-backed Web site is significantly more polished and user-friendly than I can generally make mine in the time that I have available to work on them. Learning how to embed databases into Google Maps is something I'm going to have to pick up. But there is one map graphic they created that I particularly wanted to call attention to: this -- or as a screenshot rather than an interactive map, so that I can show it here, the map below.

Compare this with the map of electric power generation by fuel source that I copied from eGRID for this earlier post. Or, for that matter, compare it with Joel Garreau's division of North America into "nine nations":

Industrial pollution near schools seems to be heavily concentrated in the areas of the country that Garreau refers to as Dixie and the Foundry. That's surprising, in some ways. Other places such as California certainly have industrial pollution too, but perhaps it's more spread out, or further away from city centers. Of course, this particular pollution map doesn't include non-point sources such as cars -- Los Angeles' smog problem gives it much worse overall air quality throughout its region than the vast majority of point-source-polluted sites. And this map doesn't include all sorts of other sources. But it's an example of how pollution is largely affected by regional patterns of development. If you were mapping environmental damage from mining waste, you'd get a different region of the U.S. highlighted -- national solutions to these problems often come down to regional politics.