Targeted Sampling and Bacterial Source Tracking of Potato Creek

1
Targeted Sampling and Bacterial Source Tracking
of Potato Creek

• Peter Hartel, Karen Rodgers, Gwyneth Moody, Sarah
  Hemmings, Jared Fisher, Jenny McDonald, and
  Carolyn Belcher
• University of Georgia

2
Acknowledgements
Karen Rodgers
Jared Fisher
Sarah Hemmings
Gwyneth Moody
Jenny McDonald
Abby Barnes
Carolyn Belcher
Patrick Baber
3
Problem Potato Creek Has Too Many Fecal
Indicator Bacteria

• Whats a fecal indicator bacterium?

• Bacteria from feces of warm-blooded animals

• Present in higher number than pathogens

• Nonpathogenic

• Do not persist in the environment

4
Four Groups of Fecal Indicator Bacteria

• Total coliforms

• Fecal anaerobes

• Fecal enterococci

5
A Problem with Fecal Coliforms

• Some fecal coliforms persist in soil and water

• EPA solution use only Escherichia coli (E.
  coli)

6
A Problem with E. coli

• But how many E. coli per 100 mL (100 mL about a
  cup) before you have a problem?

• EPA recommends 126 E. coli per 100 mL for a
  geometric mean and 235 E. coli per 100 mL for a
  grab sample

• We chose 126 E. coli per 100 mL for a grab
  sample, a very conservative number

7
How Do We Measure E. coli? (Most Probable
Number IDEXX Colilert)
8
How Do We Measure E. coli? (contd) (IDEXX System
24 hours later)

• 26 large wells
• 4 small wells
• 44 E. coli
• per 100 mL

9
All Right, But How Will You Sample Potato Creek
for E. coli?

• Targetedsampling

• Developed by us with GA EPD funds

• Never been done with fresh water

10
Why Separate Baseflow from Stormflow?

• Because typically, counts of E. coli increase 10-
  to 100-fold during stormflow conditions

11
All Right, But How Will You Sample Potato Creek
for E. coli? (contd)

• Targeted sampling

• Playing the childrens game of hot and cold

12
How Do We Identify Sources of E. coli?(
Bacterial Source Tracking or BST)

• Use BST toolbox

• Phenotypic methods based on something expressed
  by the bacterium

• Genotypic methods based on DNA

• Chemical methods based on chemicals (usually
  associated with human sewage)

13
An Example of a Chemical BST Method Fluorometry

• Based on presence of optical brighteners in
  laundry and dishwashing detergents

• Combine targeted sampling with fluorometry

• Combination has never been done in fresh water

14
How Fluorometry Works with E. coli Counts
Optical Brightener
Likely Cause
E. coli Count
High High Leaking sewage
High Low Non-human feces
Low High Gray water
Low Low No contamination
15
Integrated Science Engineering Starts Sampling
Potato Creek

• Three problem reaches
• Griffin, Meansville, and Thomaston

16
Griffin Reach
Base-flow

• Key

17
Griffin Reach
Base-flow
Winter
Summer

• Key

18
Griffin Reach
Base-flow
Storm-flow

• Key

19
Griffin Reach
Base-flow
Storm-flow
Summer 05

• Key

Summer 04
20
Likely Fecal Sources for Griffin Reach
21
Meansville Reach
Sarah Hemmings (top) and Gwyneth Moody (bottom)
22
Meansville Reach
Sarah Hemmings (top) and Gwyneth Moody (bottom)
23
Likely Fecal Sources for Meansville Reach
24
Thomaston Reach
25
Thomaston Reach
26
Thomaston Reach
27
Likely Fecal Source for Thomaston Reach
28
Is Fluorometry a Flop?
http//www.postsecret.blogspot.com
29
How Fluorometry Works
Optical Brighteners
30
Adding a 436-nm Emission Filter to the
Fluorometer
Optical Brightener (mg L-1)
Organic Matter (mg L-1)
No filter
With filter
0 75 150 300
0 75 150 300
--------------------fluorometric units, no
dimension------------------
0 79 165 340
0 0 82 165 339
59
20 126
193 258 374
118 182 297
53
200 164
163 176 191
53 56 59
31
Recommendations for Fluorometry (first time ever
combined with targeted sampling)

• Add 436-nm filter and keep limit at 100 units, or
  

• Increase fluorescence limit to gt250 units

(At gt250 units, only Griffin reach is positive
for human fecal contamination)
32
Three Questions about E. coli Q 1 What about
Tributaries as Sources?

• Cattle likely source to Little Potato Creek

• Tributaries matter as potential sources

33
Three Cautionary Questions about E.
coli Question 2 Is Sediment a Reservoir of E.
coli?
May 17
May 24
June 1
Location
--no. of E. coli per gram of sediment--
Above Griffin WWTP 575 120,226 50,118
Below Griffin WWTP 9,120 9,120 1,175
Thomaston WTP intake 5,011 2,041 3,090

• Average for all Potato Creek sediments 7,762 E.
  coli per gram dry weight of sediment

• Sediment is an ignored reservoir of E. coli

34
Three Cautionary Questions about E.
coli Question 3 Can E. coli Grow and Persist
in Sediment?
Condition
Griffin
Thomaston
Dean Creek
----No. of E. coli per gram of sediment----
Moist sediment 1,698 1,698
12,882
Dried 2 days and rewet 55 71
1,175
24 h after rewet 25 53
8,317
Dried 30 days and rewet lt10 lt10
87
4,786
24 h after rewet lt10 lt10
Dried 60 days and rewet lt10 lt10
lt10
lt10
24 h after rewet lt10 lt10

• Possible regrowth below WWTP, but does not persist

35
Is the Griffin WWTP Responsible for Fecal
Contamination Observed in Thomaston?

• Despite potential regrowth, this result unlikely

• Counts of E. coli the same in Potato Creek
  sediment above and below the plant

• E. coli counts from upstream into Means-ville
  reach are low

• Ignores fecal contamination from cattle near
  Meansville and dog kennel in Thomaston

36
Recommendations for the State

• Switch to E. coli as indicator for fresh water
  

• Establish E. coli limit (e.g., 126 per 100 mL)

• Consider sediment as reservoir of E. coli

• Adopt targeted sampling

• Flow conditions

• Tributaries

• Quick and inexpensive

• Consider BST toolbox (if BST is necessary)

37
Recommendations for McIntosh Trail RDC

• Griffin reach

• Broken sewer pipe fixed, and pet owners
  encouraged to dispose of wastes properly
  

• Meansville reach

• Best management practices needed for agricultural
  animals (i.e., riparian filter strips,
  nonriparian shade, alternative water sources)

• Thomaston reach

• Owners of kennel must clean up dog wastes