Big Dry Creek Annual Water Quality Review for 2011

Pathogens in Urban
Stormwater Systems
Presented by:
Jane Clary and Candice Owens, P.E.
Wright Water Engineers, Inc.
Project Sponsors:
Pathogens in Wet Weather Flows Technical Committee,
ASCE Environmental and Water Resources Institute
Urban Drainage and Flood Control District
Urban Watersheds Research Institute
1. Basic regulatory background
2. Sources of FIB in urban areas
3. Predicting fate and transport
4. Monitoring and source tracking
5. Statistical analysis
6. Source controls
7. Structural controls
8. TMDL case studies
9. Research needs
10. And 40 pages of references
The Problem
Top 10 Causes of Impairment in U.S. by # of 303(d) Listings
• Most strains of E. coli and enterococci do not cause human illness (that
is, they are not human pathogens); rather, they indicate the presence of
fecal contamination.
• Many of these listings are not due to sewage (untreated, CSO, or SSO).
Source: EPA Accessed May 2012,
EPA’s 2012 Recreational Water Quality Criteria
• Frequency of
Exceedances: 0
for geomean &
< 10% for STV
• Duration: 30day assessment
A few highlights:
• Removes use intensity considerations at beaches.
• No national-level exclusion for natural sources.
• Tools for developing alternative RWQC on a site-specific basis (e.g., QMRA).
• Tools for assessing and managing recreational waters, such as predictive
modeling and sanitary surveys.
• New rapid method for enterococci using qPCR method (helpful for beaches).
Health Risk from
Different Fecal Sources (EPA 2012 RWQC)
• “Human pathogens are often present in animal fecal matter, and
thus, there are risks associated with recreating in animal-impacted
• “..quantifying that level of risk associated with animal fecal material
is difficult, and the methods necessary to distinguish between
human and nonhuman fecal sources, with the appropriate level of
confidence, are still under development.”
• “EPA is not developing recommendations that take source of fecal
contamination into account.”
• “…human health risk associated with exposure to waters impacted
by animal sources can vary substantially. In some cases these risks
can be similar to exposure to human fecal contamination, and in
other cases, the risk is substantially lower.”
• “…states interested in adopting different standards to address the
variability in human health risks associated with different sources of
fecal contamination on a site-specific basis should refer to methods
for developing site-specific standards [in the new criteria].
QMRA Basics (Soller et al. 2010)
Implications for Urban Areas
Stormwater Managers/MS4s
• MS4 permit
holders must
address issue due
to TMDLs
• FIB elevated in
urban runoff
• Storm sewer
system can be a
source during dry
weather, too
Bacteria in
Urban Runoff
(Research by Pitt, Maestre and
• Ubiquitously high—
regardless of land
• Absence of firstflush effect
• Look for new
analyses from NSQD
Dry Weather Urban Runoff: Enterococci for
Residential and Commercial Land Uses in San Diego
County (Source: Weston 2009a)
Secondary/Environmental Sources
• Biofilms
– Instream
– In storm sewers
• Sediment
• Decaying
• Soils
• Traditional FIB analysis
• Other source
• Advance molecular
Based on Recommendations in
Griffith et al. 2013
Why is Modeling Bacteria so Difficult?
• Biotic and
abiotic factors
affect survival
• Sediment
Source: Olivieri et al., in WERF 2007
Statistical Analysis:
Games with E. coli Data
Tools to Reduce Bacteria in Runoff & MS4s
Illicit Discharges
Other Source
(pets, urban wildlife,
Dry Weather Screening at MS4 Outfalls
Address Sanitary Sources First
But Instream Issues May Remain
International Stormwater BMP Database
• Statistical analysis updated April
– Includes new statistical
– Hypothesis testing
• FIB data set has grown
– E. coli
– Fecal coliform
– Enterococcus
• Still important to caveat findings
based on # studies & # events
BMP Performance: E. coli
• End of pipe limits are unlikely to be consistently met.
• Volume reduction can help to reduce loads.
General Conclusions Related to
BMP Performance
• Data set remains limited for most BMP
category-FIB combinations.
• Results to date do not support consistent
attainment of numeric effluent limits for FIB
in stormwater at end of pipe.
• Retention (wet) ponds and media filters
appear to provide best performance on a
density/concentration basis.
• Bioretention and other infiltration-oriented
practices can reduce bacteria loads by
reducing frequency and volume of runoff.
• Disinfection works at point of outfall, but not
realistic in many contexts.
• Some BMP types appear to export bacteria.
Nebraska Case Study:
Cost Estimates for E. coli TMDL
• 7.7 sq. mi. Antelope Creek
Watershed, Lincoln
• Source load estimates by land
use & BMP evaluation using
• Curb-cut bioretention retrofits
identified as a key BMP
• Est. Cost: $57 million over 40year plan ($7.4M/sq. mi.)
• City will start w/ source controls
and pilot projects using 5-year
San Diego River, California
Comprehensive Load Reduction Plan
Estimated TMDL Implementation Plan
Costs for the Ballona River TMDL
(Source: City of Beverly Hills et al. 2009)
New and Improved Source
Identification Tools
Candice Owen, P.E.
Source Tracking - Bacteroidales
Using a Bacteria Source Tracking “Toolkit”
• Anaerobic enteric bacteria
– Limited persistence
– Limited re-growth
• Highly abundant in intestines/feces
• Genetic differences specific to host
• Use qPCR to quantify… results in in
gene copies/100 mL
Advanced Methods in Source Tracking
• Microbial Source Tracking
• Quantitative Microbial Risk
Assessment (QMRA)
Courtesy Brandon Steets, Geosyntec Consultants
In Pathogens in Urban Stormwater Systems
What questions can microbial source
tracking answer?
Use a “weight of evidence” approach to answers the following
Which human or non-human
sources are most prevalent
stormwater discharges?
What outfalls display the
highest levels of these
Which areas drain to the
most “problematic”
stormwater outfalls?
How can these conclusions
aid in the implementation of
more effective bacteria
MST Example: Horse Race Track
• Blamed gulls for
bacteria issues
• 3 wet weather
• Used “weight of
• Determined sources
were primarily
QMRA – Quantitative Microbial Risk Assessment
• Multi-step approach that couples risk assessment principles
with statistical computations to estimate consequences of
possible exposure to infectious organisms
• Watershed-specific data is input
• Monte Carlo statistical modeling approach to account for
QMRA Flow Chart
Step 1 Context
Step 2
Step 3 Exposure
Step 5 RISK
Step 4 Health
Step 6 Policy
What questions can QMRA answer?
• What pathogens are useful to
monitor in stormwater and natural
• What pathogens are most prevalent
and likely to cause illness?
• What are the risks to people
recreating in the study waters?
• How can we mitigate risk?
Example: 2011 WERF QMRA Study
• 3 regions of the US
• 7 discharges of concern
• Toolkit of bacteria, viruses
and protozoa
• Risks highest for Norovirus
and rotavirus. Highest in
agricultural and CSO
samples. Risk was highest
for children.
In Closing, We Still Need:
A national policy-level dialogue regarding regulatory
options that are protective of human health, while
recognizing practical economic constraints facing local
• a. What level of control for FIB is practical and
attainable, and reflects “acceptable” levels of public
health protection based on actual pathogenicity and
• b. How can measurable water quality compliance
metrics (e.g., for TMDLs, MS4 permits) be expressed so
that practical constraints are recognized, while still
promoting meaningful water quality improvement?
Download Report Here:
Jane Clary, Wright Water Engineers
[email protected]
Candice Owen, P.E., Wright Water Engineers
[email protected]

similar documents