Nancy, Karen and Wes,
I’m passing this along as follow-up to our previous attention to the Flint water changeover situation. The attached was
submitted to me along with a request for de-identified data, which should be no problem.
When you have a few minutes you might want to take a look at it. Sounds like there might be more to this than what we
learned previously. Yikes!
Robert L. Scott
Childhood Lead Poisoning Prevention Program
Michigan Department of Health & Human Services
(517) 335-8178
fax (517) 335-8509
xxxEND_PAGE:dhhs02_b0940_3256_3344_26
003282
A. Project Summary
Overview: Flint, MI, is currently suffering from a “perfect storm” attributable to out-of-control corrosion
of its potable water distribution system. The corrosion is undermining water affordability for residents,
financial viability of city government, water aesthetics, and hygiene/sanitation as revealed in local and
national news reporting. We Aypothesize that these circumstances will also create severe
chemical/biological health risks for Flint residents, including elevated levels of lead and opportunistic
premise plumbing pathogens (OPPPs) in drinking water. Preliminary data collected from a home of a lead
poisoned child in Flint has revealed extraordinarily high levels of lead, with average concentrations over
20 minutes of water use exceeding 2,000 ppb (> 200 times the World Health Organization allowable
levels for lead in potable water).
The main odjectives of this research are to: 1) compare levels of chlorine, iron, fecal indicator bacteria,
OPPPs, and corrosion-inducing bacteria present in water mains of a distribution system with uncontrolled
corrosion (Flint) versus surrounding cities/counties still using non-corrosive water, 2) profile OPPPs
occurrence in hot and cold potable water systems at these same locations, and 3) determine if there is
evidence of elevated lead in Flint homes, and, if so, forensically determine the links to iron corrosion. Our
team is uniquely qualified to do this work given our just published peer reviewed research on this subject
and our extensive collaborations with key stakeholders in Flint.
Intellectual Merit: The four elements of the “perfect storm” currently undermining water quality (and
possibly public health) in Flint mefude: a) chronic underinvestment in water infrastructure, b) under-
appreciation of the role of corrosion control in sustaining urban potable water systems, c) increased
corrosion due to higher chloride in Flint’s new source water, and d) failure to appropriately monitor for
lead and OPPPs. The latter two factors are amongst the most important health problems arising in
modern potable water systems, The high rates of corrosion occurring in Flint are releasing high levels of
iron to water and consuming chlorine disinfectant, which our most recent laboratory testing has indicated
will increase lead release to water and growth of OPPPs in cold and hot water plumbing systems. The
unfortunate but unique opportunity offered by Flint’s current situation, provides an ideal opportunity to
field test our recent discoveries regarding potentially adverse consequences of iron corrosion on
chemical/microbiological water quality at field rather than laboratory scale.
Broader Impacts: This RAPID grant will directly assist residents of Flint in assessing the current safety
of their potable water supply. If the results support recently issued public assurances regarding safety of
water, the current problems in Flint can be considered mainly of aesthetics and perception due to very
distasteful or discolored water. However, if sampling reveals widespread problems, the public will learn
of the potential health threat. Since elements of the “perfect storm” afflicting Flint are occurring at some
level in many other financially stressed U.S. urban centers with decaying drinking water infrastructure,
this Rapid Response Research (RAPID) grant also provides an unprecedented opportunity to advance
fundamental scientific and practical understanding at this emerging nexus of
engineering-public health. The general results and approach used herein can inform residents and
managers of other U.S. cities, who will soon be dealing with similar problems associated with failing
potable water infrastructure exacerbated by increased chloride in water due to excessive use of road salt
and rising sea levels. The research also provides a compelling case study in Citizen Science, since the
experiences of Flint parents in monitoring their children’s health and environmental exposures was a
trigger for our preliminary testing, and Flint consumers will be scientifically empowered by participating
in fundamental research relying on collection of samples from their homes and residences. There is also a
social justice implication of the research, in that these results can help inform the current policy debate
regarding strategies for dealing with cites that have gone bankrupt, as well as the discussion of access to
safe and affordable drinking water as a basic human right.
Synergistic impacts of Corrosive Water and Interrupted Corrosion Control on Chemical/Microbiological §
Water Quality: Flint, MI Pls: Edwards, Pruden, Falkinbam.
xxxEND_PAGE:dhhs02_b0940_3256_3344_27
003283
RAPID: Synergistic Impacts of Corrosive Water and Interrupted Corrosion Control on
Chemical/Microbiological Water Quality: Flint, MI
1. Problem Statement and Objectives
Flint, MI, is currently suffering from a “perfect storm” due to out-of-control corrosion of its potable
water distribution system, undermining the well-being of the community including water affordability for
residents, financial viability of city government, water aesthetics, and hygiene/sanitation.’? Flint’s
problems began in April 2014, when emergency managers hired to deal with the city’s fiscal crisis
determined they could save money by switching to a local river water source as opposed to purchasing
water from Detroit (Table 1). As a result of the change in source water, the Larson Iron Corrosion Index
was raised from 0.54 (low corrosion) to 2.3 (very high corrosion) and the chloride to sulfate mass ratio
(CSMR) index for lead corrosion increased from 0.45 (low corrosion) to 1.6 (very high corrosion).
Concurrently, the managers and state primacy
agency attempted to save even more money by not | Table 1. Water quality parameters for
feeding an orthophosphate corrosion inhibitor to the | drinking water supplied in Flint, MI before
water supply (Table 1). and after the April 2014 switch
Not surprisingly, the combined effect of more Paarnetar Before! After’
corrosive water and removal of the corrosion
inhibitor unleashed unprecedented corrosion in the pH 738 7.61
water main distribution system with cascading | Hardness (mg/L as CaCOs) 101 183
personal, economic, and public health consequences | Alkalinity (mg/L as CaCO;) i: 77
to Flint as tracked by news reports and mandatory Chloride (mg/L) 114 92
chemical/biological monitoring of water in the Sulfate (mg/L) 25.2 41
CSMR? 0.45 L6
distribution system mains.’ Our recent research
also. predicts that these circumstances will
potentially create severe chemical/biological health
risks for residents, due to impacts on water within
building (premise) plumbing systems that include
elevated levels of lead and opportunistic premise
plumbing pathogens (OPPPs). 118 Because the
factors impacting Flint are also occurring at some
level in many other financially-stressed U.S. urban
centers with decaying drinking water infrastructure,
this Rapid Response Research (RAPID) grant provides an unprecedented opportunity to advance
fundamental scientific and practical understanding at this emerging nexus of infrasiructure-
environmental engineering-public health. We view August-September 2015 as the ideal time to first
sample in Flint, as more than 16 months of uncontrolled corrosion have occurred and the water remains
near its seasonal peak temperature, maximizing the likelihood of serious problems with lead and OPPPs if
they exist. ;
Our key irypothesis is that the rapid corrosion of iron water mains will dramatically increase
lead release to water and growth of OPPPs as measured in consumers’ homes. Mechanistically, higher
iron corrosion produces both higher iron in water and lower levels of free chlorine, both of which
dramatically increased lead release and OPPPs regrowth in our just published laboratory research utilizing
simulated distribution systems.'*"? The main objectives of this research ave to: 1) compare levels of
chlorine, iron, fecal indicator bacteria, OPPPs, and corrosion-inducing bacteria present in water mains
of a distribution system with uncontrolled corrosion (Flint) versus controlled corrosion in surrounding
cities/counties still using non-corrosive Detroit water, 2) profile hot and cold potable water systems at the
same sampling locations in #1 for OPPPs, and 3) determine if there is evidence of elevated lead in Flint
homes, and, if so, forensically determine the links to iron corrosion. 5172 The unfortunate but unique
opportunity offered by Flint’s current situation provides an ideal opportunity to field test our recent
discoveries regarding adverse consequences of iron corrosion on OPPPs and lead concentration at the tap.
Inhibitor (mg/L as P) 0.35 NONE
Larson Ratio* 0.5 2.3
TSource: City of Flint Monthly Qperation Report, June
2015. Available from www.cityofilint.com
Source: DWSD 2014 Water Quality Report.
Available from yavw.dwsd.org
34 measure of corrosivity to lead; a value > 0.5 is a
critical trigger [10]
4A measure of corrosivity to mild steel and iron;
corrosion rate increases linearly with Larson Ratio [8]
Synergistic Impacts of Corrosive Water and Interrupted Corrosion Control on Chemicat/
Microbiological Water Quality: Flint, MI Pls: Edwards, Priden, Falkinham
xxxEND_PAGE:dhhs02_b0940_3256_3344_28
003284
D. Project Description
2. Review of Local Events and Intellectual Merit
Local Impacts. After the switch of water sources, residents
of Flint were immediately subject to an outbreak of
corrosion-related drinking water problems including
flooding from large water main breaks and reported health
ailments.’® General Motors, a prime customer of the water
system, reported that the new water was severely
corroding auto parts on its assembly fine and had to begin
importing water, costing the city $400,000 in lost revenue.
? The PI was also alerted by an Environmental Protection
Agency volunteer to a case of childhood lead poisoning in
a Flint home that was certified as “lead free.”* Samples
collected from the home exhibited classic “red water” that
is occurring throughout the city (Figure 1), along with the
highest sustained levels of lead in drinking water that we have encountered in over 25 years of research
on the subject. Specifically, in 30 samples collected over a period of 25 minutes flushing at the kitchen
faucet, lead concentrations averaged over 2,000 ppb and were as high as 13,000 ppb. For perspective,
these levels are more than 200-1,300 times higher than World Health Organization standards (10 ppb) and
several even exceeded the EPA criterion for “hazardous waste” of 5,000 ppb Pb. The city has also
reported unspecified economic losses due to water main breaks and water losses through leaks.°"''"?
The corrosion problems have also had cascading impacts on health parameters monitored under
federal regulations. Because the corrosion is rapidly consuming chlorine disinfectant in the water, the
city violated EPA limits for £. coli.*® The detaching iron rust also has the potential to expose consumers
to other contaminants that pose a serious public health risk, including arsenic and fead that have
accumulated in pipes or sorbed to iron surfaces. 1516 Health effects reported by residents since the switch
include skin rashes, hair loss, vomiting, copper poisoning, and the one confirmed case of lead poisoning.”
47.13 However, the relatively small number of cases reported to date almost certainly underestimates the
full extent of the problem. In response to the sampling showing high lead, the authorities who made the
decision to switch water sources and stop adding corrosion inhibitor publicly stated that “anyone who is
concerned about lead in the drinking water in Flint can relax” and that the water is safe,** but refuse to
sample consumers’ water without pre-flushing the plumbing for at least 5 minutes the night before
sampling. The latter practice is known to miss lead in water problems.
We are also concerned about possible health effects that have not yet been investigated. For
example, in March 2015 Region 5 EPA was provided reports of higher incidence of Legionnaires’ disease
associated with bacteria growth in premise plumbing in the Flint area.”! Legionnaires’ disease has recently
been acknowledged to be the primary source of waterborne disease outbreaks (and associated deaths) in
the U.S.” Despite that acknowledged risk, there is currently no required monitoring for this important
pathogen in consumers’ homes, where it proliferates and can lead to human exposure and infection in
showers.”
Intellectual Merit. The four elements of the “perfect storm” currently undermining water quality
(and possibly public health) in Flint include: a) chronic underinvestment in water infrastructure, b) under-
appreciation of the role of corrosion control in sustaining urban potable water systems, c) increased
corrosivity of water sources nationally due to rising chloride levels from anthropogenic pollution and/or
rising sea levels, and d) failure to appropriately monitor for lead and OPPPs, which are two of the most
important modern-day public health problems arising in building plumbing systems.
Figure 1. Drinking water samples
collected from home of a child who
was lead poisoned by Flint water.
a) Chronic Underinvestment in Water Infrastructure. A large fraction of the nation’s potable
water infrastructure is on the verge of failure, and this problem has been repeatedly voted by
Synergistic Impacts of Corrosive Water and Interrupted Corrosion Control on Chemical/
Microbiological Water Quality: Flint, MIs Pls: Edwards, Pruden, Falkinham
xxxEND_PAGE:dhhs02_b0940_3256_3344_29
003285
b)
qd)
D. Project Description
members of the American Society of Civil Engineers (ASCE) as the most urgent societal
infrastructure challenge with an overall condition grade of “D”.?*"4 Many water main distribution
systems are reaching the end of their design lifetime (60-95 years), with water main breaks
currently at a rate of 240,000 per year nationally and rising.””* Aside from obvious public health
implications associated with compromised delivery of uncontaminated drinking water to the tap,
failure events can cause property damage and water loss through leaks.” Like many post-
industrial manufacturing centers, Flint has a very large potable distribution system constructed to
sustain a large consumer and industrial base that no longer exists. The projected cost to upgrade
the distribution system is $1.5 billion dollars,’ which would transiate to an unbearable cost of
$50,000 per existing customer in Flint.8
Under-appreciation of the role of corrosion control in sustaining potable water systems.
Estimates by ASCE, the American Water Works Association (AWWA), the Environmental
Protection Agency (EPA), the Water Infrastructure Network (WIN), and the National Academy
of Corrosion Engineers (NACE) suggest that direct costs of water pipeline corrosion range
between $8 billion - $36 billion annually and indirect costs are much higher.’’ Leaks result in 7
billion gallons of lost water each day with associated revenue losses of ~ $3 billion per year for
US. utilities.” Problems with leaking potable water plumbing systems in buildings (i.e. premise
plumbing) also cost consumers billions of dollars each year.**"! Water utilities can reduce costs
of potable water system corrosion and extend the lifetime of these invaluable assets by adding
corrosion inhibitors, such as orthophosphate, to the water. Prior research using a relatively low
corrosivity source water determined that each dollar invested in corrosion control produced more
than $5 dollars in financial savings due to reduced corrosion damage and extended lifetime of
pipeline infrastructure.” In Flint, the short-sighted decision to reduce chemical costs by
removing the corrosion inhibitor and introducing corrosive water to the system may have
produced tens if not hundreds of millions of dollars in corrosion damages to its existing potable
water distribution system. We are also aware of many other utilities that are cutting back on their
corrosion inhibitor doses due to cost-cutting pressures.
Increased corrosivity of water sources nationally due to rising chloride levels from
anthropogenic pollution and/or rising sea levels. Chloride. levels in drinking water are rising
nationally in surface water due to use of road salt and seawater intrusion in coastal regions. Road
salt use in winter has risen to 137 [bs per year for every American, with a doubling of salt
application from 1990 to 2014 (10 vs. 22 million tons) associated with a doubling of chioride
levels in northern U.S. waters as monitored by the USGS.3*' There is documented concern about
the damage of salt application to infrastructure such as roads and bridges,” but rising salt levels
in the Potomac (due to road salt} in 2015 also have triggered a spike in consumer complaints of
red or brown water from their main distribution system.**°” and we are currently working with a
utility in Brick, NJ that is reporting high lead in consumers’ water due to higher chloride from
rising sea levels near their intake as weil as road salt use.°*“° The higher corrosivity of water in
Flint due to higher chloride (Table 1), therefore provides an interesting “acute” case study of
higher chloride impacts that can shed light on these important national trends.
Failure of utilities and regulatory agencies to take responsibility for the bve most important
modern day public heaith problems arising in building plumbing systems (i.e¢., lead and
OPPPs). For ten years EPA has acknowledged that utilities are collecting samples in a manner
that “misses” worst case lead in water,’ and to date they have not required utilities to change
monitoring practices to better reveal problems. The EPA LCR sampling protocols have been
under review since 2008 and the EPA is expected to issue new requirements sometime in 2016.
Hence, sampling in Flint without “pre-flushing” to reduce lead, as revealed by the EPA
Synergistic Impacts of Corrosive Water and Interrupted Corrosion Control on Chemical/ Fp ee Wena:
Microbiological Water Quality: Flint, Mis Pls: Edwards, Pruden, Fatkinham
xxxEND_PAGE:dhhs02_b0940_3256_3344_30
003286
D. Project Description
volunteer,’ could inform modifications to the EPA LCR. Likewise, EPA’s current regulations on
Legionella consider only levels that might be present in water leaving the treatment plant, where
it is least likely to be present, and do not yet require monitoring at the point of entry into homes
or within buildings where Legionella is most likely to be present and cause disease.” Our
proposed sampling for Legionella at these locations can therefore inform future regulation and
distribution system management policies for dealing with this emerging public health visk. 19.22
In summary, this proposed RAPID grant characterizing the occurrence of chemical/biological problems in
Flint, MI homes explores a newly emerging nexus between degrading
engineering-public health that can provide insight into problems facing many cities all over the United
States.
3. Approach
The RAPID grant objectives will be achieved through three phases of sampling, using analytical methods
in routine use by the project team, as follows:
Phase 1, Compare levels of chlorine, iron, fecal indicator bacteria, OPPPs, and corrosion-inducing
bacteria present in water mains of a distribution system with uncontrolled corrosion (Flint) versus
controlled corrosion in surrounding cities/counties still using non-corrosive Detroit water. A team
including the PI and at least 3 graduate students will travel to Flint and stay 3-7 days in mid-August 2015,
to collect 8 distribution system samples from surrounding cities still using Detroit water, and to also
sample Flint’s 8 distribution system monitoring locations (Figure 1). We will stay in two hotel rooms, one
located in Flint and one in a surrounding location on Detroit water, to conveniently collect samples for
free chlorine at 2 hour intervals expected to correspond to lowest and highest daily demand. Ali of these
analyses will be conducted using standard methods with the exception of testing for corrosion-inducing
bacteria which will be conducted with Biological Activity Reaction Test (BART) kits. BARTs are
standardized colorimetric culture kits that are semi-quantitative and include testing for Sulfate-Reducing
Bacteria (SRB), Heterotrophic Aerobic Bacteria, Heterotrophic Anaerobic Bacteria, Denitrifiers, Slime
Forming Bacteria, and Acid Producing Bacteria (APB). It is hypothesized that the Flint waters will have
muuch lower levels of free chlorine, higher levels of iron, corrosion-inducing bacteria, and fecal indicator
bacteria than samples collected from locations still on Detroit water.
Phase 2. Profile building hot and cold water plumbing systems for OPPPs at the same sampling
locations used in Phase 1, { oP berma a :
Protocols used previously to sample
for a suite of OPPPs and two host
protozoa’®”* in hot and cold water