Subsidy Tracker new data released

One of the projects that I work on for the nonprofit group Good Jobs First is their Subsidy Tracker site, "the only national [U.S.] search engine capturing company-specific economic development incentive awards from the federal government, all 50 states and many localities." Here's the press release and the site itself.

You can search this site in all sorts of ways -- including finding all of the subsidies given to major parent companies -- but one of the basic kinds of searches is finding the subsidies to businesses in your home town. You can do this by going to the bottom of the search form, filling in a state, and then choosing a city from a list of cities that will be populated. Here's a search for Northampton, MA. You can see a subsidy for the Coca-Cola plant, a major industrial water user, and one to Kollmorgen (now L-3 KEO), a defense contractor, as well as some subsidies to development firms -- probably the three most controversial businesses where I live.

RTK NET is back up

RTK NET, the Right-To-Know Network is back up (sort of) at rtk.net or rtknet.org. By "sort of" I mean that it currently is up in the form of a page of backend links that I wrote in 2 minutes in vi. But those links do work and provide the previous RTK NET search capability, and the Houston Chronicle (which has taken over RTK NET) is working on putting a front end in place.

RTK NET provides public access to U.S. toxic release / chemical accident / hazardous waste data, including TRI (Toxic Release Inventory), NRC (National Response Center), BRS (RCRAINFO Biennial data), the rest of RCRAINFO (i.e. hazardous waste permits), and RMP (Risk Management Plan) data. Other than the RCRAINFO hazardous waste permits data, the databases are fairly up to date with last update dates as follows:

• TRI: 10/08/2015
• BRS: 10/27/2015
• NRC: 01/05/2016
• RMP: 12/30/2015
• RCRAINFO: 05/19/2013

RTK NET is the only online searchable source that I know of for NRC and RMP data (both chemical accident databases) and its interface to TRI and BRS has some advantages that other sites don't have. I've worked on the project since 1991 and I'm glad that it's going back up.

Toxic 100 Air and Water rankings released

The Political Economy Research Institute at Umass (PERI) has released the Toxic 100 Air and Water rankings of top corporate polluters for 2016, based on Toxic Release Inventory data as weighted by EPA's geographic micro data RSEI model and with facilities assigned to a regularized set of parent companies.

One of the things that becomes apparent in analyses of these kind of data -- whether it's a news story like this one on the super polluters that USA Today did with the Center for Public Integrity or an academic paper published by Mary Collins, Joseph JaJa, and Ian Muñoz -- is that pollution from fixed facilities is dominated by a small number of facilities. To reduce pollution overall you don't necessarily have to reduce it from every facility equally.

Here's an example from the Toxic 100 Air. DuPont, the company ranked second on the list, has 97% of its total U.S. score from a single facility, the DuPont Pontchartrain Works in La Place, LA -- emissions that affect a local population that is over 60% racial and ethnic minorities. So what's going on there?

DuPont vs Chemours facilities

On July 1, 2015 DuPont completed the spin-off of The Chemours Company. There have been many reports about how DuPont loaded Chemours up with a large proportion of both its environmental liabilities and its debt, raising the question of what would happen if Chemours went bankrupt. DuPont is also attempting to merge with Dow Chemical, and there are rumors that the merger, if approved, will be followed by a split of the merged companies into three pieces. If that does happen it will be anyone's guess about who if anyone will pay for past pollution, and how much expensive litigation it would take to even begin to answer that question.

The immediate question, and first step, is to track which DuPont components have become Chemours. Let's say that you have a database divided up into reporting by facility, as most environmental databases are. And let's also say that the data that you have show all facilities reporting as DuPont, because you are tracking past data or because the database you're using last reported before the split (as the 2014 Toxic Release Inventory (TRI) did). Which U.S. DuPont facilities are now Chemours facilities?

Here's the best answer I could turn up to this question from public documents. The list of Chemours facilities is from a June 5, 2015 SEC filing. The number of facilities matches the number of production facilities in the U.S. from the Chemours Web site. Note (1) in the City column indicates a lease, (2) indicates a lease from Du Pont (which I've still treated as belonging to Chemours), and (3) indicates a shared facility between two branches of Chemours.

The facilities below are basically identified by city and state in most sources. But some of the facilities are actually in different cities than the city that they are referred to as being in, either in their mailing address or their facility name. The "2014 TRI" column shows which city the facility is actually in according to that database. If the city is the same as the first column (or if I didn't find it at all), I've marked it with an "x". Otherwise, it shows an alternate city that the facility may be listed under.

City	State	2014 TRI
El Dorado (1)	AR	x
Edge Moor	DE	x
Red Lion (1)	DE	Delaware City
Starke (Mine)	FL	x
Louisville	KY	x
Wurtland	KY	x
Burnside	LA	Darrow
Elkton (1)	MD	x
DeLisle	MS	Pass Christian
Pascagoula (3)	MS	x
Fayetteville	NC	x
Deepwater	NJ	x
Morses Mill (1)	NJ	Linden
Niagara	NY	x
Fort Hill	OH	North Bend
N. Kingstown (1)	RI	x
New Johnsonville	TN	x
Memphis	TN	x
Corpus Christi	TX	Gregory
LaPorte (2)	TX	x
Beaumont	TX	Nederland
Borderland (1)	TX	x
James River	VA	Richmond
Washington	WV	x
Belle (3)	WV	x

Environmental data reports

I mostly work on data-backed sites that present free information to the public. I've done it as a staffer for a nonprofit group from 1991-1997 and as a consultant to nonprofit groups since 1997. From 1998-2005 I also helped to analyze data for a number of white-paper reports from environmental groups -- after that, computer and data literacy became widespread enough so that this kind of work was generally done in-house. These reports are mostly of historical interest now, but some of them may still be useful. There's a page that lists them, but it's a bit much to look through so I've listed them by category here:

• PRTRs (pollutant release and transfer registries): 1991 (waste transfers), 1998 (underground injection), 1998 (water), 2000 (water), 2000 (child development), 2005 (regulatory), 2013 (water)
• chemical accidents: 1998, 1999, 2001, 2005, 2006, 2007
• air toxics: 2000, 2002, 2003, 2004

In addition there were two reports that were pretty much sui generis and were in many ways my favorites: Poisoning Our Future (1998) and Cabinet Confidential (2004). The first of these looked at sources of persistent bioaccumulative toxins (PBTs) by integrating many different documents and databases: something like I'd like some day to go back to and see what we were wrong and right about. The second was a rather inspired but rickety attempt to use a Massachusetts toxics use database, a New Jersey toxics use database, and the U.S. national PRTR to estimate quantities of toxic chemicals within products nationwide, something that none of these databases had been designed to do.

With this and the posts below (copied from my personal blog, when I had a brief enthusiasm for writing posts about this kind of thing in late 2008 / early 2009) I've summed up my previous writing about data. I'm working on a wider variety of projects now and hope to write about them here.

Greenhouse 100

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

The latest project that I've worked on, PERI's Greenhouse 100, is here. It's a publicly accessible EPA database of greenhouse gas polluting facilities, summed up by parent company. I think that it's the first list of its kind that includes both power plants and other industries.

2007 TRI released

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

The latest version of the Toxic Release Inventory (TRI) came out on the 19th, with RTK Net's version open a day later. This year, RTK NET's version also supplies RSEI risk screening numbers -- for the first time, an at least partial answer to the question "How important is this particular release of pollution, anyways?"

EPA continued its recent trend of downplaying the data release. I don't think that they announced they'd be releasing it far in advance -- I had to find out about it through the grapevine after it was already up. I didn't see much news about it, and the news there was was unspecific. For instance, the overall release trend was down, but PCB releases jumped 40%. Why? According to this story, for one example, "EPA said that the jump was probably due to disposal of old equipment or clean up at industrial sites." Probably? The vast majority of the increase seems to be due to one site, Chemical Waste Management in Emelle, Alabama. Why not call that facility and get the actual cause for the jump? That's one of the things that would change this from a contextless, uninvestigated number into a story that people could begin to understand.