Subscriber access provided by University of Sussex Library
Ecotoxicology and Human Environmental Health
Lead levels at the tap and consumer exposure from legacy and recent lead service line replacements in 6 utilities Elise Deshommes, Benjamin Trueman, Ian Douglas, Daniel Huggins, Laurent Laroche, Jeff Swertfeger, Abby Spielmacher, Graham A. Gagnon, and Michèle Prévost Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b02388 • Publication Date (Web): 20 Jul 2018 Downloaded from http://pubs.acs.org on July 24, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 33
Environmental Science & Technology
1
Lead levels at the tap and consumer exposure
2
from legacy and recent lead service line
3
replacements in 6 utilities
4
Elise Deshommes*, a, Benjamin Truemanb, Ian Douglasc, Dan Hugginsd, Laurent
5
Larochee, Jeff Swertfegerf, Abby Spielmacherg, Graham A. Gagnonb and Michèle
6
Prévosta
7 8
*
9
4711 (2236), fax number: (+1) 514-340-5918
Corresponding Author, e-mail:
[email protected], phone number: (+1) 514-340-
10
a
11
Montreal, Quebec, H3T 1J4, Canada
12
b
13
Scotia, B3H4R2, Canada
14
c
15
d
16
e
17
f
18
g
19
Canada
Department of Civil, Geological and Mining Engineering, Polytechnique Montreal,
Department of Civil and Resource Engineering, Dalhousie University, Halifax, Nova
City of Ottawa, Drinking Water Services, Ottawa, Ontario, K1P 1J1, Canada City of London, Water Operations, London, Ontario, N6A 4L9, Canada
City of Montreal, Technical Expertise Division, Montreal, Quebec, H8N 2K2, Canada
Greater Cincinnati Water Works, Cincinnati, Ohio, 45232, USA City of Guelph, Water Services, Environmental Services Guelph, Ontario, N1E 6P7
1 ACS Paragon Plus Environment
Environmental Science & Technology
Page 2 of 33
20
KEYWORDS: lead service line replacement, flushing, profile sampling, corrosion
21
control, exposure
22
TOC/GRAPHICAL ABSTRACT
23 24
ABSTRACT
25
Profile, regulatory, and investigative sampling were completed in six utilities to study the
26
impact of partial and full lead service line replacements (LSLRs) on water lead levels
27
(WLLs) and consumer’s exposure. As compared to households with no replacement, lead
28
release after partial LSLR (PLSLR) was generally greater in the short-term (3-50 days),
29
comparable or lower in the medium (2 years). This was
30
mainly explained by insufficient time elapsed to stabilize scales after disturbances to the
31
service line. One utility showed sustained lead release over 18 months after PLSLR.
32
Moreover, the reduction in WLLs was small when analyzing results for the same
33
households. As a comparison, full LSLR decreased WLLs drastically and immediately.
34
The occurrence of low (0-5 µg/L) to high (≥50 µg/L) WLLs in the profiles varied between
35
households and reflected the variability of exposure amongst households in the same
36
system. Using this probability of occurrence, the distribution of WLLs of exposure was
37
estimated for households with or without a PLSLR, and used to model young children
38
blood lead levels (BLLs) for both groups of households. The range of modeled BLLs 2 ACS Paragon Plus Environment
Page 3 of 33
Environmental Science & Technology
39
decreased slightly for households with PLSLR, but still overlapped the range estimated
40
for households with no replacement. This analysis suggests that, in a system, PLSLRs do
41
not reduce young children blood lead levels except in a fraction of households.
42
INTRODUCTION
43
Lead service lines (LSLs) contribute to increased water lead levels (WLLs) and
44
ultimately to increased blood lead levels (BLLs) of young children1-3. While corrosion
45
control is applied in the United States and in Ontario (Canada) for systems exceeding the
46
Action Levels of 15 µg/L and 10 µg/L respectively4,5, the lead crises in Flint, MI, and
47
Washington, DC, are stark reminders that the legacy of LSLs combined with human error
48
represents a serious risk for lead exposure through water1,2. Corrosion control may also
49
be insufficient to achieve the recently proposed Canadian health-based value of 5 µg/L6.
50
LSL replacements (LSLRs) are frequent in distribution systems upgrading their water
51
mains, they are also conducted proactively. Considering the legal constraints related to
52
dual ownership, and the lack of funding, utilities are however usually unable to complete
53
full LSLRs (FLSLRs)7. Partial LSLRs (PLSLRs) can increase WLLs due to the
54
disturbances caused to the LSL scales during replacement work. Moreover, galvanic
55
corrosion deposits have been observed at the copper-lead junction of field-collected
56
LSLs, and related to increased WLLs over the long-term in pilot studies7-12. Full-scale
57
monitoring in Halifax and Montreal measured acute particulate lead release immediately
58
after PLSLR; however, after two years the WLLs were generally lower than before
59
replacement13,14. Nonetheless, 61% of the samples exceeded 10 µg/L short- and long-
60
term (>2 years) after PLSLR in Montreal, while about 25% of the samples collected six
61
months post-PLSLR were >15 µg/L in Halifax13. In these studies, the post-LSLR flushing
3 ACS Paragon Plus Environment
Environmental Science & Technology
Page 4 of 33
62
procedure, the type of LSLR, and the water main material and corrosion state were
63
evidenced as key parameters driving lead release following LSLR13-16.
64
Regulated lead sampling includes the collection of the first draw after overnight
65
stagnation without any pre-flush in the US (Lead and Copper Rule or LCR)4; two liters
66
after 30 minutes of stagnation preceded by a pre-flush in Ontario (regulation 170/03)5;
67
one liter after 5 minutes of flushing in Quebec17; and a random daytime sample in
68
Europe18. The US and European protocols aim to evaluate system-wide corrosion control
69
efficiency, although the latter can estimate consumer exposure at the system level. The
70
Ontario protocol is a compromise because it includes a pre-flush that limits the
71
probability of capturing water originating from the LSL and particulate lead. Its shorter
72
stagnation time is however more typical of water use intervals in households18. Finally,
73
the Quebec protocol aims to identify worst-case households >10 µg/L after flushing, but
74
also underestimates the consumer’s exposure18,19. Profile (sequential) sampling consists
75
of collecting consecutive samples (typically four or more samples) at the tap after
76
stagnation. The volume collected covers the piping volume from the tap to the main19,20.
77
Profile sampling has been used to detect lead sources in the piping, to quantify WLL
78
variability, to measure corrosion control efficiency, and to evaluate galvanic corrosion in
79
PLSLRs13,19-21. Although informative, profile sampling is not realistically applicable at
80
large scale. Nonetheless, WLLs measured from profile sampling have been successfully
81
linked to those measured after flushing, indicating that a few profiles may be sufficient to
82
evaluate consumers’ potential risk of exposure in a full-scale system14,19.
83
The main objective of this study was to investigate and document the impacts of PLSLR
84
and FLSLR over short- and long-term for many water qualities, using full-scale results
4 ACS Paragon Plus Environment
Page 5 of 33
Environmental Science & Technology
85
from profile and regulatory sampling. Based on the probability of occurrence of WLLs in
86
the profiles, plausible exposure scenarios are drawn to better understand the impact of
87
LSLR on exposure assessment.
88
MATERIAL AND METHODS
89
Utilities and monitoring design. Sampling for WLLs was completed at the kitchen tap
90
of 3460 households before and/or after LSLR in five Canadian utilities (three provinces)
91
and one US utility. Among these utilities, utilities A, B, and F are regulated using a 90th
92
percentile action level of 10 µg/L after flushing and 30 minutes of stagnation (1st and 2nd
93
liter); utilities C and E are regulated using a maximum acceptable concentration of 10
94
µg/L after 5 minutes of flushing; utility D is regulated using a 90th percentile action level
95
of 15 µg/L and a 1st draw sample collected after at least 6 hours of stagnation. Moreover,
96
as the impact of PLSLR has been extensively studied for utilities C and E13,14,19, it will
97
not be detailed in this paper. Corrosion control consisted in pH adjustment for utilities A
98
and D and resulted in WLLs below the action level; the other utilities did not have
99
corrosion control or not an optimized one as WLLs exceeded the local regulation. Table 1
100
presents the main water quality parameters for the utilities studied and their monitoring
101
design.
102
The configurations of the households monitored included full LSLs (100% Pb), PLSLRs
103
and FLSLRs (100% Cu). The effect of LSLR was evaluated over time after recent
104
PLSLRs and FLSLRs. Moreover, old PLSLRs present in the distribution system were
105
sampled. For purposes of this study, old PLSLRs are defined as legacy PLSLRs of
106
greater than two years, while recent PLSLRs are PLSLRs of less than two years (0-50
107
days in utility A, 3 days to 6 months in utility E, 0-1 year in utility D; 3 days to 2 years in
5 ACS Paragon Plus Environment
Environmental Science & Technology
Page 6 of 33
108
other utilities; Table 1). Households monitored were all single-family homes except for a
109
subsample of nineteen multi-unit homes. Finally, PLSLRs consisted of a copper pipe on
110
the public side of the service line and a lead pipe on the private side, except a fraction of
111
old PLSLRs in utilities C and F (lead on the public side, copper on the private side).
112
Profile sampling was accomplished in five of the six utilities, and consisted in the
113
collection of four to sixteen consecutive liters following a stagnation of either 30 minutes
114
(30MS) or greater than 6 hours (6HS). Other protocols included the collection of a
115
sample after 5 minutes of flushing (5MF, 5 utilities); a first-draw after 6HS according to
116
the LCR (utility D), and two consecutive liters after 30MS according to Ontario 170/03
117
regulation (utility F). Finally, samples were collected after 3 or 10 minutes of flushing
118
(3MF, 10MF) in utility D following LSLR, while samples were collected in utility A by
119
opening and closing the tap at maximum flow rate according to the particulate
120
stimulation sampling (PSS) detailed in Deshommes et al.22. Samples were analyzed for
121
total lead in water with ICP/MS by licensed laboratories (3450 households) according to
122
the USEPA 200.8 method and Standard Method 3125 B. Finally, a subset of samples was
123
analyzed with ASV analyzer23 in utility C (40 households). Collection periods were
124
equally distributed to cover cold and warm seasons except for a subset of 30MS profiles
125
collected in utility C (summer only), and for 6HS profiles collected in utility E
126
(monitoring in summer for 100% Pb households, monitoring over time after LSLR from
127
summer to winter).
128 129
6 ACS Paragon Plus Environment
Page 7 of 33
130
Environmental Science & Technology
Table 1. Summary of the utilities and monitoring design. Utility and water quality parameters
A (pH 9.2; Alk. 33 mg/L as CaCO3, CSMR 0.2) B (pH 8.1; Alk. 90 mg/L as CaCO3, CSMR 0.44) C (pH 8.0; Alk. 95 mg/L as CaCO3, CSMR 1.1) D (pH 8.8; Alk. 74 mg/L as CaCO3, CSMR 0.45)
Number of households (number of samples in italics) Recent * PLSLR
Old * PLSLR
n=207 (n=1034) n=111 (n=111) n=11 (n=1195) n=1245 (n=1245) n=17 (n=170)
n=15 ǂ (n=160) n=15 (n=64) n=1 ǂ (n=63)
n=187 (n=935) n=81 (n=81) n=3 (n=405) n=506 (n=506) n=20 (n=211)
n=5 (n=132)
n=13 ǂ (n=417)
100% Pb
n=12 (n=303) n=16 (n=50)
Seasons
Sampling protocol
Summer and winter Summer and winter
5MF (1L), 30MS profile (4L)
*
Full LSLR -
PSS (1 or 2L)
-
All seasons
5MF (1L), 30MS profile (8L)
-
All seasons
5MF (1L)
n=3 (n=19)
Summer
5MF (1L), 30MS profile † (≥8L)
n=7 (n=331)
n=3 ǂ (n=113)
All seasons
5MF (2L), 6HS profile ** (≥6L)
-
-
-
All seasons
Profile after 6HS (4-10L)
n=11 ǂ (n=245)
-
n=5 ǂ (n=183)
All seasons
1 draw 6HS, ǂǂ 3MF&10MF (0.75L)
-
st
E (pH 7.3; Alk. 20 mg/L as n=45 n=74 Summer to 5MF (1L), 6HS profile n=57 ǂ ǂ †† CaCO3; CSMR 1.1; 0.5 (n=225) (n=755) (n=975) Winter (4L) mg/L as OPO4) F st nd n=108 n=99 n=185 n=439 5MF (0.125L), 1 and 2 (pH 7.9; Alk. 277 mg/L All seasons L after 30MS (n=245) (n=288) (n=276) (n=1096) as CaCO3, CSMR 1.1) Notes: 5MF—5 minutes of flushing; 30MS—30 minutes of stagnation; 6HS—at least 6 hours of stagnation; PSS— Particulate Stimulation Sampling22; *PLSLR stands for partial LSLR; Recent PLSLR defined as