May 12, 2015
WEST LAFAYETTE, Ind. –Three new studies suggest
that when communities are hit with disasters that contaminate drinking
water the official decision-making and response often lack scientific
basis.
The result has been an inability to fully
anticipate public health risks and effectively rid plumbing systems of
contaminants, sometimes exposing residents to toxic chemicals, said
Andrew Whelton, an assistant professor in Purdue University's Division of Environmental and Ecological Engineering and Lyles School of Civil Engineering.
Since 2014 more than 1.5 million people across
the nation have received drinking water tainted with crude oil, diesel
fuel, algal toxins and coal-washing chemicals.
"Numerous contamination incidents have been
caused by chemical spills from storage tank ruptures, pipeline breaks,
rail car and truck accidents, as well as algal blooms," Whelton said.
His team has been examining recent disasters in
which tainted drinking water was distributed to homes. The goal is to
develop techniques and tools to help communities respond more
effectively, said Whelton, who will discuss some of the results of three
studies on Wednesday (May 13) during the American Water Works Association Central District spring meeting in Danville, Indiana.
Some of the drinking water catastrophes studied
were a January 2014 chemical spill in West Virginia; an August 2014
toxic algal bloom in Western Lake Erie; a December 2014 accident
involving a petroleum-based solvent in Washington, D.C.; a January 2015
crude oil pipeline accident in Glendive, Montana; and an April 2015
diesel spill in Nibley City, Utah.
In one study, Purdue graduate student Karen
Casteloes discovered that plumbing system flushing procedures did not
account for low-flow faucets and water heaters of all sizes, suggesting
that residents who followed official guidelines may have still had
unsafe water in their homes. Casteloes also worked with Randi Brazeau, a
Metropolitan State University of Denver researcher.
Research results also will be presented at a National Science Foundation sponsored chemical spill workshop, Fostering Advances in Water Resource Protection and Crisis Communication, Lessons Learned from Recent Disasters, on May 27-29 in Shepherdstown, West Virginia.
After reviewing 40 drinking-water emergencies,
Casteloes developed a model that can be used to properly flush plumbing
systems depending on their design. When Casteloes applied the model to
the West Virginia and Montana spills, her results revealed flushing
guidelines issued by the authorities would not have reduced
contamination to safe levels for some homes, specifically manufactured
and single-family homes that contained water-saving fixtures and large
water heaters.
"The larger your water heater and greater number
of low-flow fixtures you have in your home, the longer you need to flush
your plumbing system," Whelton said.
Casteloes said "a significant need" exists for
more analysis following a drinking water chemical contamination incident
because unsafe water can remain in some homes.
In another study of eight oil spills, Purdue
graduate student Xiangning Huang reviewed the response practices of
water utilities, states, and the federal government. She found that
approaches used by authorities varied considerably, and no guidance
exists on which tests should be run in response to certain types of
spills.
Huang said that in the wake of the Nibley City,
Utah, spill and growing use of hydraulic fracturing, there is a need to
address this ambiguity.
The researchers also found that testing is
conducted for chemicals that have established drinking water limits, but
not for unregulated chemicals that also may pose health risks.
"Responders need rapid analytical tools for
characterizing unregulated contaminants in air and water, predicting air
exposure risks and chemical fate in waterways, and in water systems,"
she said, recommending that a full screening of the spilled products be
performed before deciding whether the water is safe and how to
decontaminate water systems.
A third study examined in-home drinking water
testing data from the 2014 West Virginia chemical spill. Whelton and
University of South Alabama graduate student LaKia McMillan collected
and analyzed data from a variety of universities, private companies, and
nonprofit organizations and compared them to data obtained by
responders.
"We found that more than 10 chemicals not listed
on any material safety data sheet were spilled from the Freedom
Industries tank into the Elk River," Whelton said.
The contaminant methyl 4-methylcyclohexanecarboxylate (MMCHC) was present in the drinking water 13 to 30 days after the spill.
"No safe drinking water exposure limit was
established for MMCHC. The long-term health impacts of this chemical
exposure remain unstudied," Whelton said. "Responders only tested
drinking water for three ingredients."
Huang and McMillan's research will be presented at the Association of Environmental Engineering and Science Professors conference on June 13-16 at Yale University.
Whelton said the findings of the three studies
emphasize the need for improvements in chemical spill response, and
in-home water testing is essential. The research was funded by the
National Science Foundation RAPID grant 1424627. Some of the work has
been featured by NSF's Science Nation magazine.
Writer: Emil Venere, 765-494-4709, venere@purdue.edu
Source: Andrew J. Whelton, 540-230-6069, awhelton@purdue.edu
ABSTRACT
Lessons Learned from Water Supply and
Infrastructure Contamination Incidents Caused by Chemical Spills in
North America, 2014-2015
Xiangning Huang†, Karen Casteloes†, Randi Brazeau‡, LaKia McMillanΦ, Devin Kelly Δ,
Andrew J. Whelton† Δ *
† Lyles School of Civil Engineering, Purdue University
‡ Department of Environmental Science, Metropolitan State University of Denver
Δ Division of Environmental and Ecological Engineering, Purdue University
Φ Environmental Toxicology Program, University of South Alabama
*Corresponding author: E-mail awhelton@purdue.edu
Specific objectives were to (1) review chemical spill causes specifically affecting water supplies, (2) identify contaminant types and their fate, (3) examine incident response commonalities and differences, and (4) outline research needs. Results showed that chemical transport accidents (i.e., pipeline, rail, roadway, and shipping) are common causes of drinking water supply and water infrastructure contamination.
Responders have applied a dissimilar approach to investigating past water supply contamination incidents where for the same type of spills (i.e., crude oil) affected waters were monitored for some contaminants but not others. Information is lacking about organic contaminant fate in water supplies and building plumbing systems, and the effectiveness of water treatment processes to remove contaminants common to past chemical spills.
Little information is available about water infrastructure decontamination method effectiveness and determining public health risks posed by contaminated water inside buildings. In some cases plumbing system flushing was recommended without knowledge of contaminants present in the drinking water. Plumbing system decontamination activities have caused clinically diagnosed public health impacts.
Though, for incidents where no illness data were reported, responders commonly did not examine these types of data. Research is needed to develop science-grounded tools that responders can use to rapidly respond to and recover from water supply contamination incidents.