Ecology Research Project:  

The Effect of Untreated Sewage on

Escherichia coli Population Levels in Lake Michigan

 

Leah Doughty and Brianna Clark

 

BI 341

Fall 2004

 


Abstract

 

In this experiment we tested to see whether or not sewage dumping into Lake Michigan truly has an effect on this aquatic environment’s Escherichia coli (E. coli) levels. Our main focus was to observe if E. coli levels in the lake decreased as the distance away from the Milwaukee Metropolitan Sewerage District’s (MMSD) Jones Island Wastewater Treatment Plant increased. We took a total of ten water samples at designated interval locations south of the water treatment plant and analyzed the water for E. coli as well as other fecal coliforms. Next, a correlation value was determined to be -0.193 between distance from wastewater treatment plant and E. coli population levels. This statistical information allowed us to discover that an increase in distance from MMSD’s Jones Island Wastewater Treatment Plant does lower the E. coli population levels present in Lake Michigan. Although a weak, negative correlation value exists, it still supports our idea that E. coli populations decrease with distance away from the treatment plant.

Keywords: sewage dumping, Lake Michigan, E. coli, fecal coliforms, bacteria, MMSD

 

 

Introduction

 

Contamination of aquatic ecosystems originating from domestic sewage is readily indicated by the presence of elevated Escherichia coli (E. coli) levels (Whitman et al. 2003). Drastically high quantities of this bacterium, which tends to grow in a pH range of 5.5 to 7.5, have been a continual problem to the southern portion of Lake Michigan, especially during the summer months (Kinzelman et al. 2003, Kirakosyan et al. 2004). Numerous beaches along the lake have closed as a result (Kinzelman et al. 2003). Thus, many communities along the coasts of the Great Lakes have adopted beach monitoring programs to protect visitors from potentially harmful microbes in accordance with the requirements of the U.S. Environmental Protection Agency (EPA) (Whitman and Nevers 2003). Increased coastal development and increased recreational use of beaches have resulted in greater threats of water contamination and the associated public health hazard.

In the United States, freshwater beaches are routinely analyzed for E. coli because it provides evidence for the possible presence of human pathogens (bacteria, protozoa, and viruses) (Whitman and Nevers 2003). Potential sources may include: sewage overflows, leaking septic systems, and birds occupying the beach. Rainfall and onshore winds can cause amplification in concentration and can exacerbate the contamination problem.

Overall, many of these beach closures have been speculated by Kinzelman et al. (2003) and Whitman et al. (2003) to be directly associated with local untreated sewage releases by wastewater treatment plants. In southeastern Wisconsin, the main facility is the Milwaukee Metropolitan Sewerage District’s (MMSD) Jones Island Wastewater Treatment Plant. Located along Lake Michigan on a peninsula in Milwaukee's harbor, this plant has been in operation for more than 75 years. Originally designed to treat 85 million gallons of raw sewage per day, recent expansion and renovation as part of the District's $2.3 billion Water Pollution Abatement Program increased the plant's peak capacity to 330 million gallons. Daily average flow to the plant currently is about 112 million gallons (MMSD 2004).

As an environmental agency, the Milwaukee Metropolitan Sewerage District has a strong commitment to pollution prevention and improving the quality of local waters. Thus, MMSD’s mission clearly states that they work to cost-effectively protect public health and the environment, prevent pollution and enhance the quality of area waterways (MMSD 2004). The city's sewer system was so overwhelmed by torrential rains this past spring that MMSD ended up dumping 4.6 billion gallons of untreated, raw sewage into Lake Michigan (Schultze and Rohde 2004).

Overall, recent studies have reported that elevated counts of Escherichia coli indicate the presence of sewage due mainly to local wastewater treatment plant dumping (Whitman et al. 2003). This experiment was designed in order to test whether or not sewage dumping into Lake Michigan truly has an effect on this aquatic environment’s E. coli levels. Our hypothesis was that water samples that were collected closer to the Milwaukee Metropolitan Sewage District’s (MMSD) Jones Island Wastewater Treatment Plant would have higher E. coli population levels than water samples that were taken further south along Lake Michigan’s shoreline and away from this sewage dumping site.

 

Materials

 

  • Petrifilm E. coli / Coliform Count Plate (30)
  • Test tube rack
  • Large test tubes with screw tops (12)
  • Water blanks (distilled) (1)
  • Bacterial colony counter
  • Distilled water
  • Bunsen burner
  • Fume hood
  • Safety gloves
  • Goggles
  • Difco Manual 9th and 10th Edition
  • Wader boots and shoes
  • Sterile pipettes (30)
  • Syringes (2)
  • pHydrion Wide Range pH test paper
  • Incubator (37ºC)
  • Label tape
  • Open type reel tape measure (m)
  • Sharpie markers
  • Cleaning supplies
  • Map of Jones Island Wastewater Treatment Plant

 

 

Methods

 

Our project began by us writing up a research proposal. Shortly thereafter, our idea was approved to be both a practical and ethical experimental study by the Assistant Professor of the Biology Department at Alverno College, Rebecca Burton, Ph.D. Once this was accomplished, we began our project on Wednesday, October 27, 2004. On this day, ten water samples were collected in sterile test tubes from locations of 150-meter intervals downstream (southbound) from the Milwaukee Metropolitan Sewerage District’s (MMSD) Jones Island Wastewater Treatment Plant (700 E. Jones Street, Milwaukee, Wisconsin) (Figure 1). However, there were exceptions to the distance intervals due to private industries, governmental property and a fenced-off Milwaukee County snow-dumping site (Table 1, Figure 2 and Figure 3). Beginning at 805 meters south of Jones Island, water samples were collected by a team, which consisted of our assistant Susan Hagen and us. After all ten lake samples were collected; we returned to school and stored them in the microbiology refrigerator until Friday, October 29, 2004. 

On this second day of experimental work, one mL from each of the ten samples was used to inoculate one 3M Microbiology Products Petrifilm E. coli / Coliform Count Plate #2006-01KC. We then followed the procedures that were outlined in the Carolina 3M Petrifilm E. coli / Coliform Count Plate #2006-01KC instruction manual, exercising sterile aseptic techniques. Two plates were inoculated per sample and labeled according to the test tube sample number (i.e. the two plates created for test tube three were labeled #3a and #3b). Prior to inoculating the plates, all sample tubes were spun in a vortexer.

After inoculation from the water samples, the Petrifilm plates were incubated at 37°C and were read on experimental day three. pH readings of the lake water samples were also taken on this day using the pHydrion Wide Range pH test paper. The results were read within 30 seconds per the instructions on the label.   

On Monday, November 2, 2004 (third day of research), we viewed our Petrifilm plates in order to count E. coli and other fecal coliform colonies. At this time, we evaluated whether or not any Petrifilm plates should be rerun. This decision was mainly based on the colony growth similarity between sample pair plates. If a set had similar numbers of bacterial growth, we did not re-plate. New Petrifilm plates were only created if a pair had very different quantities of bacterial colony growth on them. Thus, we decided that new Petrifilm plates needed to be made from test tubes 2, 5, 6, 8, and 10. Plates that were not being rerun were stored in the refrigerator to minimize further growth. 

In addition to being re-plated, sample number six first was diluted 1:10 in a separate sterile tube, with sterile, distilled water, and then used to inoculate a new pair of plates. This was done because the first pair of Petrifilm plates for sample number six had bacterial growth that was too numerous to count. After inoculation from the five water samples mentioned above, the Petrifilm plates were incubated at 37°C and were read on experimental day four.

On the fourth day of our experiment (Wednesday, November 03, 2004), the final bacteria growth (E. coli and other fecal coliforms) for the five re-plated Petrifilm plates was counted and recorded. Due to the 1:10 dilution of water sample number six, the bacteria colony counts of this sample’s plates were multiplied by ten to get the actual total colony count. 

Overall, this experiment took one week to complete. We made sure to record all of our data and observations in our laboratory notebooks. Lastly, we used Microsoft Excel to make a few tables, graph our results, generate a R2 value by means of a linear trendline equation and determine a correlation value between increased E. coli levels and close proximity to MMSD’s Jones Island Plant. These statistics will additionally demonstrate to us whether or not untreated sewage dumping into Lake Michigan has an effect on this marine ecosystem’s E. coli levels.

 

 

 

    

(a)                                                                                                                                         (b)

 

Figure 1: Jones Island Wastewater Treatment Plant: (a) Entrance; (b) Aerial view (image from MMSD 2004).

 

 

 

 

 

 

Table 1: Specific locations of Lake Michigan water samples.

Sample Location #

Distance South of Jones Island Wastewater Treatment Plant

(Along Lake Michigan)

1

805 meters

2

955 meters

3

1105 meters

4

1255 meters

5

1405 meters

6

1555 meters

7

3165 meters

8

3315 meters

9

3465 meters

10

3615 meters

 

    

Figure 2: Pictures of the unstable U.S. Government peninsula beginning at approximately 1555 meters south of the Jones Island Wastewater Treatment Plant.

    

Figure 3: Pictures of the fenced-off Milwaukee County snow-dumping site beginning at approximately 2700 meters south of the Jones Island Wastewater Treatment Plant.

Results/Data

 

First, we determined that each of the ten water samples collected had a pH of 7.5. Next, we implemented XY scatter plots for both E. coli and other fecal coliform bacteria plate averages (Table 2) and discovered that a correlation was present (Figure 4 and Figure 5). An equation of: y = -0.0003x + 1.0858, and an r-value that was approximately equal to -0.193 existed for E. coli. While we obtained an equation of: y = -0.0029x + 52.504, and an r-value that was approximately equal to -0.070 for other fecal coliform bacteria. These r-values indicate negative correlations between bacterial growth and distance from the water treatment plant (Figure 4 and Figure 5). However, they are not very strong. Instead, they are fairly weak correlations.

Due to the fact that -0.193 and -0.070 are negative values (which indicate negative associations between the variables), E. coli and other fecal coliform levels in the lake did decrease as the distance away from the Milwaukee Metropolitan Sewerage District’s (MMSD) Jones Island Wastewater Treatment Plant increased. Thus, bacterial growth in Lake Michigan is inversely correlated to distance away from the wastewater treatment plant.

 

 

 

 

 

 

 

 

 

Table 2: Average results of bacteria colony counts.

Sample #’s

Distance South of Jones Island Wastewater Treatment Plant

(Along Lake Michigan)

Average of E. coli

Average of other fecal coliforms

1a and 1b

805 meters

0.50

36.50

2a and 2b

955 meters

0.00

45.25

3a and 3b

1105 meters

0.00

1.00

4a and 4b

1255 meters

0.00

5.00

5a and 5b

1405 meters

0.00

70.25

6a and 6b

1555 meters

5.00

163.75

7a and 7b

3165 meters

0.00

14.00

8a and 8b

3315 meters

0.00

52.25

9a and 9b

3465 meters

0.00

28.50

10a and 10b

3615 meters

0.00

47.75

 

 

 

 

 

Figure 4: Correlation between distance from Jones Island Wastewater Treatment Plant and E. coli population levels.

 

 

Figure 5: Correlation between distance from Jones Island Wastewater Treatment Plant and other fecal coliform population levels.

 

 

 

 

Discussion

Although weak, the negative correlation values that we obtained from this experiment do support our hypothesis that water samples which are collected closer to the Milwaukee Metropolitan Sewage District’s (MMSD) Jones Island Wastewater Treatment Plant will have higher E. coli population levels than water samples that are taken further south along Lake Michigan’s shoreline and away from this sewage dumping site.

These results do follow/relate to previously known findings such as, the notion that contamination of aquatic ecosystems originating from domestic sewage is readily indicated by the presence of elevated Escherichia coli (E. coli) levels (Whitman et al. 2003). Recent research has reported that elevated counts of E. coli indicate the presence of sewage due mainly to local wastewater treatment plant dumping (i.e. Jones Island Wastewater Treatment Plant dumping in the Spring of 2004) (Schultze and Rohde 2004, Whitman et al. 2003). Thus, I feel that our evidence and data collected did support our hypothesis. However, our results were of weak support due to the fact that the only significant data points were for samples that were taken within the first 1500 meters south of Jones Island.

Another possible explanation for the weak results gained deals with the time of year that we gathered our water samples. They were collected during the fall season. Although, prior studies indicated that E. coli levels are most prevalent and noticeable in Lake Michigan throughout the summer months of the year (Kinzelman et al. 2003). Consequently, since the optimum temperature for E. coli growth is 37ºC, we believe that the summer heat provides a better, more favorable environment for E. coli to synthesize in than the cool days of fall.

Regarding pH testing in this study, we discovered that the pH of the ten lake samples was consistently 7.5. Since E. coli grows in the pH range of 5.5 to 7.5, it was definitely possible for the bacteria to grow and colonize in Lake Michigan. However, our consistent pH result of 7.5 was at the upper limit of normal E. coli growth range for pH. This may explain why only a trace amount of E. coli was found. Therefore, we have come to conclude that the aquatic pH level could have been a factor in inhibiting the growth of E. coli.

If we were to repeat this experiment in the future, there are a few aspects that we would want to alter before beginning. Sample size would be one of them. Yes, we did collect ten different water samples from a few different places along the shores of Lake Michigan. Yet, we were not able to collect more than this due to time constraints and location accessibility. In our opinion this is a relatively small sample. To gain a better representation, as well as, understanding of bacteria levels in the lake due to wastewater dumping, we truly need more samples out in the field.

Another suggestion for future studies would be to include other directions in which the samples are collected. In addition to going along the shoreline (both north and south), it would be interesting to gather samples in intervals going further east and into Lake Michigan. This would help us to identify which way currents flow away from the wastewater dumping site.

Since E. coli tends to grow best in warm conditions, it would be better to collect the samples during the warm summer months than in the crisp autumn.

The last facet that we would change would be to attempt to decrease human error. Since we were crunched for time, we both helped out where and when we could. Instead we would prefer that in possible future studies, specific tasks would be assigned and carried out at every sample site and in the lab by the same individual they were given to. By task delegation we could possibly lower the human error of us reading more or less measurements than our partner and in turn shifting the statistical results obtained.

 


Literature Cited

 

Kinzelman, J.; Ng, C.; Jackson, E.; S. Gradus, and R. Bagley.  2003.  Enterococci as           indicators of Lake Michigan recreational water quality: comparison of two   methodologies and their impacts on public health regulatory events, Applied and       Environmental Microbiology, 69, 92-96.

 

Kirakosyan, G.; K. Bagramyan, and A. Trchounian.  (2004, December).  Redox sensing       by Escherichia coli: effects of dithiothreitol, a redox reagent reducing     disulphides, on bacterial growth.  Biochemical and biophysical research           communications, 325, 803-806.  Retrieved on November 26, 2004 from                                EBSCOhost database.

 

MMSD: Milwaukee Metropolitan Sewerage District (2004).  Agency and Facility       Information.  Retrieved on October 3, 2004 from the World Wide Web:    http://www.mmsd.com/home/index.cfm.

 

Schultze, S., and M. Rohde.  “Residents talk trash about sewage.”  JS Online                         24 September 2004.  Retrieved on October 3, 2004 from the World Wide Web:

            http://www.jsonline.com/news/metro/sep04/261546.asp.

 

Whitman, R.L., and M.B. Nevers.  2003.  Foreshore sand as a source of Escherichia           coli in nearshore water of Lake Michigan Beach, Applied and Environmental     Microbiology, 69, 5555-5562.

 

Whitman, R.L.; Shively, D.A.; Pawlik, H.; M.B. Nevers, and M.N. Byappanahalli.  (2003,        August).  Occurrence of Escherichia coli and Enterococci in Cladophora           (Chlorophyta) in Nearshore Water and Beach Sand of Lake Michigan, Applied    and Environmental Microbiology, 69, 4714-4719.  Retrieved on October 3,                  2004 from EBSCOhost database.