ATTACHMENT C - NC Study of E. coli and Fecal Coliform

North Carolina Department of Environmental Quality Division of Water Resources
512 North Salisbury Street 1611 Mail Service Center Raleigh, North Carolina 27699-1611
919.707.9000

January 10, 2024
Memorandum

TO: Environmental Management Commission (EMC)
Water Quality Commitee (WQC)

FROM: North Carolina Department of Environmental Quality (NC DEQ)
Division of Water Resources (DWR)

SUBJECT: North Carolina Study of the E. coli and Fecal Coliform Pathogenic Indicators for
Recrea�onal Waters - Final Report

I. Background

History
The U.S. Environmental Protec�on Agency’s (EPA) 2012 Recrea�onal Water Quality Criteria1 for
Bacterial Indicators recommends states set bacteriological water quality standards for primary
recrea�on2 freshwaters using either E. Coli or Enterococcus pathogenic indicators. North
Carolina’s exis�ng surface water quality standards for freshwaters use the fecal coliform
bacteria group pathogenic indicator. As part of the 2020-2022 Triennial Review, NC DEQ
proposed to adopt the E. coli pathogenic indicator as a site-specific surface water quality
standard for the Class B freshwaters in the 19 coun�es that comprise the NCDEQ Asheville
Regional Office (ARO) opera�onal area.

Over 1,400 comments were received on that proposed ac�on. While the comments supported
moving from a fecal coliform to E. coli pathogenic indicator, there was an overwhelming request
for the E. coli standard to apply statewide. EPA commented that the proposed site-specific
standard did not meet their requirements as it was based on geography and not scien�fic
studies.

In considera�on of the many public comments received, the EMC did not adopt the site-specific
standard as proposed. DEQ commited to proposing a statewide E. coli standard for primary
recrea�on waters and conduc�ng a study that would allow DEQ to evaluate the regulatory and
fiscal impacts of switching to an E. coli standard for all Class B freshwaters, as well as poten�al
implica�ons to exis�ng TMDLs that address current fecal coliform impairments. DWR presented

1 Published at htps://www.epa.gov/sites/default/files/2015-10/documents/rwqc2012.pdf
2 In North Carolina, primary recrea�on freshwaters are classified as Class B. [15A NCAC 02B .0219 (1)]
Page 2 of 6

the workplan to the EMC in May of 2022 and provided updates to the WQC in November 2022,
July 2023, and November 2023.

E. Coli Water Quality Criteria
EPA’s 2012 Recrea�onal Water Quality Criteria recommenda�ons are based on either of two
es�mated illness rates (32 illnesses per 1,000 recreators or 36 illnesses per 1,000 recreators3)
and comprised of two components:
1. the geometric mean of bacteria colony forming units protective of the given illness rate;
and
2. the Statistical Threshold Value (STV) which protects for high level exposure events by
limiting the frequency of exceedances, as a percentage (i.e. 10% or 20%) of samples in a
given period.

All states within EPA’s Region IV and the Eastern Band of Cherokee Indians4 have adopted E. coli
standards for primary recrea�on freshwaters (Table 1).

Table 1 – Summary of E. coli standards for Primary Recreation Freshwaters
State GeoMeanA
(per 100 ml)
10% Exceedance
(per 100 ml)
20% Exceedance
(per 100 ml)
Daily Max
(per 100 ml)
South CarolinaB 126 MPN - - 349 MPN
Tennessee 126 CFU - - 941C cfu
Kentucky 130 CFU - 240 colonies
Florida 126 MPN or MF 410 MPN or MF - -
Georgia 126 CFU 410 CFU - -
Alabama 126 CFU - - 235D colonies
Mississippi 126 CFU 410 CFU - -
Eastern Band of
Cherokee Indians 126 CFU 410 CFU - -
EPA-recommendedE
Water Quality Criteria
126 CFU 410 CFU 274 CFU -
100 CFU 326 CFU 217 CFU -
A Most states have a minimum of five samples within a month or 30 consecu�ve days. Expressed as Most Probable
Number (MPN), Colony Forming Units (CFU), or Membrane Filter (MF).
B Applies to all freshwaters as SC classifies all waters as primary recrea�on.
C Applies to any individual sample from all waterbodies except lakes, reservoirs, state scenic rivers, Excep�onal
Tennessee Water or Outstanding Na�onal Resource waters; for those waters, the standard is 487 cfu per 100 ml.
D Applies to any individual sample.
E EPA recommends either of two values that are associated with different illness rates. 126 CFU is associated with
an es�mated illness rate (NGI) of 36 per 1,000 primary contact recreators while 100 CFU is associated with a NGI
of 32 per 1,000 primary contact recreators.

3 Es�mated illness rates are expressed in terms of the number of highly credible gastrointes�nal illnesses per 1,000
primary contact recreators known as EPA’s Na�onal Epidemiological and Environmental Assessment of
Recrea�onal Water Gastrointes�nal Illnesses (“NEEAR-GI” or “NGI”). A rate of 36 NGI/1,000 recreators
corresponds to a geometric mean of 126/100 ml (MPN or MF count). A rate of 32 NGI/1,000 recreators correspond to a geometric mean of 100/100 ml (MPN or MF count).
4 DWR did not research tribal water quality standards outside of North Carolina.
Page 3 of 6

North Carolina’s current fecal coliform water quality standard for surface waters classified for
primary recrea�on use (Class B) is “Fecal coliforms shall not exceed a geometric mean of
200/100 ml (MF count) based on at least five samples taken over a 30-day period, nor exceed
400/100 ml in more than 20 percent of the samples examined during such period.” [15A NCAC
02B .0219(3)(b)]

DWR evaluated four potential E. coli water quality standards. Options I and II correspond to an
illness rate of 36 per 1,000 recreators; options III and IV correspond to an illness rate of 32 per
1,000 recreators. EPA has stated that all four options are consistent with the latest science
and will meet the National Recommended Water Quality Criteria:
I. E. coli shall not exceed a geometric mean of 126/100 ml (CFU or MPN)5 based on at
least five samples taken over a 30-day period, nor exceed 410/100 ml in more than 10
percent of the samples examined during such period.
II. E. coli shall not exceed a geometric mean of 126/100 ml (CFU or MPN5 based on at least
five samples taken over a 30-day period, nor exceed 274/100 ml in more than 20
percent of the samples examined during such period.
III. E. coli shall not exceed a geometric mean of 100/100 ml (CFU or MPN)5 based on at least
five samples taken over a 30-day period, nor exceed 326/100 ml in more than 10
percent of the samples examined during such period.
IV. E. coli shall not exceed a geometric mean of 100/100 ml (CFU or MPN)5 based on at least
five samples taken over a 30-day period, nor exceed 217/100 ml in more than 20
percent of the samples examined during such period.

II. Methods

Ambient Sampling
24 Class B freshwater ambient monitoring sta�on sites were selected for this study (Figure 1).
Sample holding �me requirements dictated that all sample sites be located within six hours of a
cer�fied lab6. Each of these 24 sites were sampled five �mes during each month of the swim
season7 in 2022 or 2023 for fecal coliform, E. coli, and turbidity. Supplemental measurements of
precipita�on, temperature, DO, standard conduc�vity and pH were also collected with each
sample.

5 The geometric mean is expressed as Colony Forming Units (CFU) or the Most Probable Number (MPN). MPN is a
sta�s�cal method of es�ma�ng the concentra�on of bacteria and allows the use of the popular Colilert test.
6 Sampling sites are not located in the central and northeastern coastal plain due to the distance between Class B
ambient monitoring sta�on sample sites and cer�fied labs and the exclusion of Class SB waters in the study.
7 The swim season runs from April to October, which is when primary recrea�on mainly occurs. Due to hiring
challenges in the Asheville Region, sampling was not completed at the nine western sites during the months of
April and May.
Page 4 of 6

Figure 1 – Ambient Sampling Locations

NPDES Sampling
22 NPDES permited facili�es8 in Class B waters were sent leters reques�ng E. coli samples for
the months of April, May, June, July, and August 2023. 16 NPDES permited facili�es were
requested to sample upstream and downstream of their discharge point in addi�on to their
effluent, while six were requested to sample their effluent only. E. coli data was received from
20 NPDES permited facili�es and compared to their corresponding fecal data.

III. Results and Discussion

Ambient Data
There were 491 individual results for both E. coli and fecal coliform, and 96 “5-in-30” geomeans
were derived from the samples. The exis�ng fecal standard was compared to the four E. coli
op�ons, which showed that none of the E. coli op�ons were significantly different than the
current fecal standard (Table 2).

8 Major and minor NPDES permit holders were included as well as industrial and publicly-owned treatment works.
Page 5 of 6

Table 2– Ambient Data
Existing and Proposed Standards # of Sites
Exceeding
Current fecal standard (>20% of samples >400 or geomean >200) 16
Op�on I: E. coli (>10% of samples >410 or geomean >126) 17
Op�on II: E. coli (>20% of samples >274 or geomean >126) 13
Op�ons III: E. coli (>10% of samples >326 or geomean>100) 18
Op�on IV: E. coli (>20% of samples >217 or geomean >100) 15

NPDES Data
The discharger sampling provided 947 individual results for both E. coli and fecal coliform, and
76 “5-in-30” geomeans were derived from the samples. The effluent data was evaluated and
showed that E. coli Op�ons II and IV were not significantly different than the current fecal
standard, but E. coli Op�ons I and III may result in more exceedances than Op�ons II and IV
(Table 3). The discharger collected instream data was also evaluated and showed that E. coli
Op�ons I and III may result in more exceedances than Op�ons II and IV (Table 4).

Table 3 –-NPDES Effluent Data
Existing and Proposed Standards # of Effluents
Exceeding9
Current fecal standard (>20% of samples >400 or geomean >200) 1
Option I: E. coli (>10% of samples >410 or geomean >126) 7
Option II: E. coli (>20% of samples >274 or geomean >126) 3
Option III: E. coli (>10% of samples >326 or geomean >100) 7
Option IV: E. coli (>20% of samples >217 or geomean >100) 4

9 Note: The water quality standard is established for the waterbody not the effluent. These results do not reflect
non-compliance with the current fecal standard nor any of the proposed E. coli op�ons.
Page 6 of 6

Table 4 –-NPDES Upstream and Downstream Data
Existing and Proposed Standards Upstream Downstream
Exceeds fecal (>20% >400 or geomean >200) 8 3
Exceeds E. coli (>10% >410 or geomean >126) 22 17
Exceeds E. coli (>20% >274 or geomean >126) 12 8
Exceeds E. coli (>10% >326 or geomean >100) 24 18
Exceeds E. coli (>20% >217 or geomean >100) 14 11

IV. Conclusions
The results from DWR’s study confirms that any of the four potential options for E. coli are
consistent with the latest science and will meet the National Recommended Water
Quality Criteria. While there were no significant differences observed between geomeans,
differences were observed between the STVs. Therefore, DWR recommends using a geomean
of a geometric mean of 126/100 ml (CFU or MPN), which is consistent with other Region 4
states. DWR also recommends using a STV exceedance rate of 20%, which is consistent with
the current STV for fecal and allows for statistical anomalies.
V. Proposed E. coli Standard
DWR recommends amending 15A NCAC 02B .0219(3)(b) as follows:

Fecal coliforms Escherichia coli shall not exceed a geometric mean of 200126/100 ml (MF count
colony forming units or most probable number) based on at least five samples taken over a 30-
day period. Escherichia coli also shall not exceed 400274/100 ml (MF count colony forming units
or most probable number) in more than 20 percent of the samples examined during such period.