Festering Questions from the West Virginia Chemical Spill

A post by Anne Jefferson
The massive impact of last week’s chemical spill into the Elk River in West Virginia continues to cause hardship for the up to 300,000 people affected by the water ban and to pose tough questions for scientists and authorities involved with assessing and mitigating the spill’s effects on central West Virginia’s water supply.

Before Thursday morning, West Virginia’s Elk River was a well-known river for fishing and the sole water supply for the city of Charleston and populations in parts of nine counties. On Thursday morning, West Virginia Department of Environmental Quality Division of Air Quality was alerted to licorice or anise odor and traced the source to a leaking tank at a Freedom Industries facility located on the bank of the Elk River. The leaking tank contained 4-methylcyclohexane methanol, a chemical used in frothing during the coal washing process. The tank has been described as “antique” and was made of riveted stainless steel, which is an old construction technique. Unfortunately, the leaking tank was located near the river bank, and up to 7500 gallons of the chemical made it into the river. (Here’s what the site looks like from the river) Even more unfortunately, the water intake for West Virginia American Water was located only one mile downstream. West Virginia American Water is the major water supplier for 9 counties in central and southwestern Virginia, including the state’s capital, Charleston. Water treatment plants typically aren’t designed to remove industrial chemicals, and West Virginia America Water wouldn’t even have known to test for this one. At some point Thursday afternoon, the dots were connected and a water ban was put in place for anyone served by the water supply. Residents, businesses, and institutional facilities were told not to use the water for drinking, cooking, bathing, or washing clothes or dishes. The only allowed uses are toilet flushing and fire fighting. As of Sunday afternoon, three days after the spill occurred, the water use ban is still in place, state and federal emergencies have been declared, and FEMA and the National Guard are distributing bottled water and water from tanker trucks.

Google Maps view of white industrial tanks adjacent to a river.

Freedom Industries site adjacent to the Elk River, upstream of Charleston, West Virginia. Image from Google Maps.

Google Maps of Elk River and surroundings.

Point A indicates the Freedom Industries site and Point B indicates the water treatment plan. Image from Google Maps.

There aren’t a lot of data on the toxicology or safe levels of 4-methylcyclohexane methanol (MCMH) in water. We know that it is an organic solvent and a light non-aqueous phase liquid, which might be why there have been no reports of fish kills, since fish are hanging out in warmer, deeper pools at this time of year. Deborah Blum did some sleuthing and surmising from related chemicals, and so far the health effects appear to consist of a lot of people worried about symptoms but few people having problems clearly related to the spill. But the axiom of not ingesting industrial chemicals still stands. Officials are scrambling to find out what they can from the chemical’s manufacturer and to test samples to determine what level of contaminant is actually in the water supply. At some point, the state decided that a level of less than 1 part per million (1 ppm) of the contaminant was safe, and now the goal seems to be to have the water treatment plant consistently producing water at with contaminant levels lower than 1 ppm. There will also be some delay between samples being collected and the test results coming in, though it sounds like they may be making rapid on-the-ground strides in the rate at which they can test samples in four labs that have been set up for this task. Some reports say the state reported river water concentrations of MCMH had declined to 1.7 ppm on Friday, from 3 ppm. No numbers seem to have been released Saturday or Sunday, though NBC News is reporting on a news conference indicating 0 ppm going into and out the water treatment plant on Sunday morning. This article from the Bluefield Gazette gives some idea of the plan to flush parts of the massive water distribution system, which goes over mountains. All indications are that it will be days more before the water ban is lifted for everyone.

As an outside observer to this disaster, gleaning what I can from the media reports, here are my questions about the past, present, and future of the central West Virginia water supply.

  • Why was a tank farm allowed to locate on the river bank only a mile upstream of a major municipal water supply intake? And if the tank farm was upriver before the water treatment plant was built, what safety concerns were raised in selecting that site?
  • Were any special emergency procedures in place at either the industrial facility or the water treatment plant given this proximity? (I can find a risk management plan for the water treatment plant to deal with its risk of spilling chloride, but I can’t find anything similar for Freedom Industries or how the water treatment plant planned for risks of spills upstream from its facility. These plans may exist off-line.)
  • How was an acceptable maximum contaminant level of 1 part per million decided upon?
  • When the water treatment plant is producing less than 1 part per million of MCMH in the water supply, how will they test and ensure that all parts of the water distribution network are at that level?
  • What will the flushing recommendations be for water users? How will the recommended flushing times be determined?
  • Will there be longer term monitoring of MCMH levels at various points in the distribution network to ensure that levels remain below the acceptable maximum contaminant level?
  • Will there be long term studies of the environmental and human health effects of the exposure, particularly on vulnerable populations like children?
  • Most importantly, how will this incident change policy and procedures in this region, in West Virginia, in the coal processing industry, and in the nation?

Based on what I’ve learned, the risks of this human-caused disaster could potentially have been substantially mitigated by more stringent watershed planning that prevented the tank farm from operating so close to the river bank so close upstream of the water treatment plant and tighter regulation and routine inspection of the tank farm and its safety procedures. If the spill had still occurred even with those risk reductions, it would have been useful to have more information on the toxicology of the chemical that spilled (and all industrial chemicals) to provide better guidance on what maximum contaminant level is acceptable. As Deborah Blum points out, “Our Toxic Substances Control Act is more than 35 years old and we (by which I mean Congress) haven’t conjured up the backbone to update and strengthen it as of this date.”

This spill and its disruption of West Virginians lives shows what can happen when environmental regulations are not stringent, rigorous, and routinely enforced.  The spectre of 300,000 US residents unable to drink or bathe in their tap water for days on end should stick in our collective consciousness for a long time. We don’t need to invoke the threat of terrorists or climate extremes to disrupt our water lifelines; West Virginia reminds us that business as usual can do that too.

Categories: by Anne, hydrology, public science, society

Comments (12)

  1. Readers: Please note that this is still a rapidly developing story, with news articles being published hourly. Everything in here was accurate when I wrote, but may be subject to later revision as more becomes known about the spill and its aftermath. At a certain point, I gave up trying to link every source I’d read in researching this story, because so many articles have been published and I wanted the focus of this one to be on the big picture questions.

  2. Cujo359 says:

    the risks of this human-caused disaster could potentially have been substantially mitigated by more stringent watershed planning that prevented the tank farm from operating so close to the river bank so close upstream of the water treatment plant and tighter regulation and routine inspection of the tank farm and its safety procedures.

    Unfortunately, as we learned in the West, Texas fertilizer plant explosion last year, safety inspections are nowhere near “routine”, as in “happening within a reasonable time”. To me, part of the challenge is that there is a mix of state, local, and federal authority for these things, and all too often no level of government puts the time and money necessary into inspections and enforcement.

    I don’t know what the answer is, but I think it’s going to have to involve a substantial change of priorities or thinking among the people who run things.

  3. agesilaus says:

    OK from the viewpoint of someone who spent 31 years in a powerplant where we had to stock tanks full of various industrial chemicals, my questions are these:

    1)There are Federal laws governing the storage of chemicals, these require that tanks over a certain volume have a containment area around the storage tanks that will hold 100+% of the tank max volume.

    2) See 40 CFR part 264.193(b)

    Secondary storage requirements:

    ” Designed or operated to contain 100% of the capacity of the larges tank within its boundary.
    Designed or operated to prevent run-on or infiltration of precipitation into the secondary containment system unless the collection system has sufficient excess capacity to contain run-on or infiltration. Such additional capacity must be sufficient to contain precipitation from a 25-year, 24-hour rainfall event.
    Free of cracks or gaps.
    Designed and installed to surround the tank completely and to cover all surroundings likely to come into contact with the waste if the waste is released from the tank(s) (i.e., capable of preventing lateral as well as vertical migration of the waste).”

    3) In Florida the state enforces these laws and in our area the county also had it’s own inspectors. We would get inspections from all three levels of government at least once a year. These inspections were on site and the inspectors would inspect the actual tanks and containment areas.

    4) There are strict record keeping rules where the tank owners must record their periodic inspections of the tank system and keep those forms for the inspectors. At our site all tanks got a walk by every hour and a detailed inspection once a month.

    5) Leaks must be reported within a few hours of discovery to the entire chain of government agencies. A remediation plan has to be in place and put into action. We had a chemical cleanup company on call 24 hours a day.

    That aerial shot shows not only are there no containments around those tanks they are actually uphill from a waterway.

    The finger pointing should at a minimum include the 1) the company 2) The state of West Virginia and 3) the regional EPA office.

    As a chemist in the environmental field for all those years I would say that 1 ppm standard came frome what we called the anal extraction method, pardon the crudity. Not based on anything but someone’s wild guess.

  4. Thomas wells says:

    This whole mess is a clusterflush.

  5. carol Jefferson says:

    Keep in mind that West Virginia is a state that condoned mountain top removal coal mining and filling nearby valleys with overburden.

  6. Jill Marshall says:

    ditto to what agesilaus says. I worked for a water quality agency in CA for close to 20 years and industrial and now even-non industrial sites (think refinery vs. wine production) are required to have secondary containment. In our region the city, county and water quality agency all had inspectors that would monitor these sites. Even if one agency could only visit once every few years, the overlap ensured that regular site inspections happened. Even if the tank was old and ready to go, and the site location is a legacy issue from when public policy wanted industrial site near flushing rivers, secondary containment should have prevented or at least mitigated this issue.

  7. Lab Lemming says:

    Is that river iced over this time of year?

  8. I appreciate the insights of those of you working in the field. My perspective is from by my experience as an academic and on the water planning side, so I’m glad for the extra information about the regulations and how inspections work in other states.

  9. Don Atkinson says:

    It looks like there is a containment wall around the tanks. It obviously failed or the leak was in piping outside the wall as the stuff hit the river but when the place was built it looks like the installed the wall.

    http://images.politico.com/global/2014/01/10/10914-west-virginia-chemical-spill-ap-328.jpg

  10. Nancy Ball says:

    The text above has a typo: “(I can find a risk management plan for the water treatment plant to deal with its risk of spilling chloride,…” should of course be “chlorine” – the toxic gas, not the simple anion.

    Links (2)
  1. Pingback: RiskScience Tracking: MCHM Spill - Charleston,West Virginia - RiskScience

  2. Pingback: Festering Questions from the West Virginia chemical spill