The Mystery of a Lifeless Creek: Investigating Dissolved Oxygen and Fecal-Coliform Bacteria

Part of the Young Naturalist Awards Curriculum Collection.

by Anna, Grade 9, Pennsylvania - 2004 YNA Winner

Standing at our deck rail, I look down at the creek below. It is picturesque at any season. In the spring and summer, flowers grow on the bank, trees of several different species stand straight and proud. In the fall when the leaves turn crimson, orange, and gold, my creek looks royal. Even in the winter when the leaves are gone and the trees are bare, the water still looks peaceful and tranquil. After living here half my life, I do not always notice the Dunkin Doughnut cups, soda cans, Styrofoam containers, and other litter strewn on the banks. But it is there, a sad reminder of human neglect of the environment. Standing on the bank looking into the water, I can see the glint of a metal soda can squashed and lying on the creek's bottom. 

I have many memories of the creek. When I was younger, my friend and I would eagerly pull on rubber boots and splash into the creek and follow it downstream. Sometimes we would sit on the big rocks in the middle. On Earth Day the neighborhood would gather to clean the banks and stream bottom, clear out brush, and plant flowers. I remember the time a gasoline truck spilled its contents into the water, causing the firemen to evacuate the neighborhood. They were afraid the gasoline would ignite and explode. However, for the years I have lived by it, I cannot remember a time when it contained aquatic life. Although further downstream there were some fish, there were never any fish swimming or frogs sunning on the banks near my house. All signs of a healthy aquatic environment were and are nonexistent. 

At the beginning of my block, Mill Creek emerges from underneath the city of Philadelphia. When it rains, torrents of water and trash from the storm sewers dump into the creek. Some of the house sewage pipes drain into the creek. The Philadelphia Water Department has been working on correcting this but has not yet succeeded. Sometimes the creek rises over its banks and floods the yard. Afterward we can tell how far it came by the line of litter in the grass. During all types of weather, however, human feces and sewage are also dumped into the creek. Though not always apparent, they sometimes turn the creek from blue to brown, and produce an unpleasant odor. These unwanted additions to the creek repulsed me, yet I desired to find out how polluted the creek really was. In the Clean Water Act of 1972, standards were put in place for treating municipal waste or sewage (Smith, p. 8). Even though the Act doubled the number of bodies of water that were safe for fishing and swimming, today 40 percent of the nation's waterways remain unclean for these uses. In Pennsylvania, roughly 25 percent of the water remains polluted (Cooper, p. 955-956). My creek is part of that 25 percent. I wanted to find out how polluted it really was.

At the start of my project, I borrowed a water-testing kit from my high school. I also obtained access to a probe that would precisely measure the amounts of dissolved oxygen in the creek. Examining the equipment, I noticed that most of the tests would at best give me an estimate of the amount of that substance present in the stream. I was greatly disappointed that the test for fecal-coliform bacteria, which shows the presence of fecal content in the creek, would only give me a positive or negative reading. Researching online at a community Web site about our creek, I found a table showing the fecal-coliform bacteria content at different locations between the years of 1998 and 2001. I was amazed to find that in the short distance the creek traveled from its emergence from underneath the city, very close to my backyard, to about a mile downstream, the fecal-coliform bacteria count ranged from more than 200,000 colony counts per 100 milliliters (cfu) to 260 (Rudin chart). I began to wonder if other tests would also vary if tested at different locations as the creek traveled further away from the city. Mainly, I wondered if there was a connection between fecal-coliform bacteria levels and dissolved oxygen levels in the water. I didn't have the instruments to test bacteria levels, but I could use the oxygen probe and the data from the prior tests. I decided to go ahead and do all the tests but to focus on the connection, if any, between fecal-coliform bacteria and dissolved oxygen content. I also wanted to find out what factors were preventing the creek from supporting aquatic life. 

The creek behind the home of Anna, 14, a 2004 Young Naturalist Awards winner.
Site 1: the creek behind Anna's house.

Before I began testing, I researched the importance of dissolved oxygen in water. It is important in water because it affects the ability of aquatic life-forms to live in water. The oxygen gets into the water by several methods: diffusion from the surrounding air, rapid movement (which is also known as aeration), or as a waste product of photosynthesis. If the dissolved oxygen level in the water is too high, fish and other aquatic invertebrates can suffer from a disease in which burgeoning gas bubbles in their blood vessels block their blood flow. However, this condition is rare, and the oxygen level would have to be extremely high. On the other hand, if the level of oxygen drops too low, below 5.0 ppm (parts per million), aquatic life is put under stress—stress that increases as the oxygen levels drop. If the oxygen level is lower than 1-2 ppm for a long period of time, large numbers of aquatic organisms may die (Dissolved Oxygen, par. 1-4).

I also researched fecal-coliform bacteria. I found that it is a type of coliform bacteria found inside the intestinal tract of warm-blooded animals. If this type of bacteria is found in a body of water, it means that the water has been fecally contaminated by a warm-blooded animal. If this bacteria is present, other harmful bacteria associated with it may also be in the water. Humans can be a source of fecal-coliform bacteria in water through leaking sewer lines, sewage spills, and malfunctioning septic systems. Fecal contamination can cause various diseases in humans who come into contact with polluted water (Chattahoochee Riverway Project, par. 1-3). One of my neighbors got a life-threatening infection through creek water contact with cuts in his feet. Levels of contamination in water are usually higher after rainstorms, in the summer, and in streams with higher turbidity, or water cloudiness (Chattahoochee Riverway Project, par. 7).When I began to test the water, I chose three sites to get water samples. These sites had been tested before for fecal-coliform bacteria content. Site number one, my backyard, is between 50 and 75 feet from where the water emerges from the storm sewer. The water is fairly deep; if it were clean, it would be deep enough to swim in. The depth abruptly changes to about six inches, with solitary and grouped rocks breaking the surface. The bank across from our yard is a somewhat steep rocky embankment. On our side there is grass all year and flowers in the spring and summer. There is also bamboo growing near the water, and a one-and-a-half-foot wall built to prevent erosion from the yard during storms. There is no animal or plant life in the water. The water moves slowly (but is not stagnant) except during storms, when it surges and moves so fast that it creates crashing white waves as it hits the rock embankment. On most days the water's turbidity is low, making it easy to see the trash lying at the bottom of the deep section.

Site 2

Site two is approximately six-tenths of a mile downstream from the first site. It is about six inches from the surface to the water's rocky dirt bottom. The water is clear and moves at a moderate pace. One bank is a gravel and stone beach with lots of litter. The other is an almost vertical embankment of brush and trees, with big rocks sitting right along the water. There is no plant or animal life in the water except for some moss. About 25 feet upstream is a 15-inch-high waterfall.


Site 3

Site three is about another three-tenths of a mile downstream. The banks have turned into rock and concrete walls, and the bottom of the creek is also concrete. Along the edges of the creek are rust-colored deposits. Here, the water is about two inches deep and moves swiftly. The only way to gather the water samples is by bucket. Pipes are placed in the wall to channel runoff rainwater from the nearby lawns and golf course. There are no rocks, plant life, or animal life in the creek. 

As well as testing dissolved oxygen, I also tested the following: pH to find out if the water was acidic or basic (A Citizen's Guide to Understanding and Monitoring Lakes and Streams, par. 1); nitrate because it enters water through human and animal waste (Nitrate—A Drinking Water Concern, par. 4); hardness to see if there was a high concentration of harmful substances dissolved in the water (the harder the water, the more difficult it is for metals to dissolve); and chlorine. Chlorine is used by water-treatment plants to treat bacteria in water (Pringle, p. 35). Its presence in the creek water would indicate whether the fecal-coliform bacteria were being treated with chlorine. To confirm that fecal-coliform bacteria were still present in the stream, I decided to test the sites for its presence. I learned from my research that for a water source to be considered safe for human consumption, the pH level should be between 6-9, the dissolved oxygen level should be higher than 5 ppm (parts per million and the fecal-coliform bacteria level should be less than 200 cfu(On-Line Volunteer Water Quality Monitoring Guidebook, par. 3). Also, nitrate levels above 45 ppm are not safe for human consumption (Nitrate—A Drinking Water Concern, par. 3). 

I decided to test the water at different times of day to see if the measurements I took would change as the surrounding temperatures changed. I had found in my research that tests measuring pH and dissolved oxygen can have different readings depending on the time of day. I also wanted to see if pollution levels would change when people were at home instead of at work. 

When I started to do the testing, it became clear that I could not carry all the equipment to each site because of the difficulty in accessing the water at the last source, the bulk of the equipment, the cold temperatures in mid-December, and the need for a computer connection for the dissolved oxygen probe. To solve this problem, I collected water samples in glass jars and brought them back to the house. This method was especially helpful when I tested the creek water at night, because many of the tests required good lighting to see the results. Because the water contained raw sewage, I wore hospital gloves while doing the testing.

The different tests I did all required different amounts of time. The pH was the easiest test because the paper immediately changed color. I would then compare it to a color chart to find the water's pH. The test for fecal-coliform bacteria, however, required 30 hours before it showed results. If the test was positive, as it always was, the water would turn yellow, and bubbles and a thick brownish gel would rise to the top. Many of the other tests required dissolving a powder or a pill in a specified amount of water and then noting the intensity of the color change. 

I had to take great care and precaution when doing the tests to make sure I did them all the same way. When using the probe I had to be careful not to stir the water too fast while measuring, or aeration would cause the dissolved oxygen reading to increase drastically. Other samples required specific amounts of water that had to be right, or else the results might vary for reasons other than differences in the water's origin. One test I had to be extremely careful with was the water hardness test. It involved adding liquid with a dropper. I had to be careful to add only one drop at a time, or the test would not be precise. Several of the tests, including the chlorine and nitrate tests, only contained one test tube that had to be cleaned between each use. 

After completing the testing, I was ready to analyze the data. One of the first things I noticed was that there was a pH of 7 for every test done. Seven is the neutral number for pH and means that the substance is neither acidic nor basic. Another result that was the same for each source was the chlorine content. Every time it was 0 ppm. This meant no chlorine was being used to treat the bacteria in the water. The fecal-coliform bacteria tests all tested positive, which means the levels at all three sites were probably above 200 cfu. The nitrate levels were always between 3 and 5 ppm. I noted, however, that the nitrate level was always lowest at the second site and highest at the third test site. None of the sites consistently showed higher or lower readings. The tests for water hardness came to approximately 9.3 gpg (grains per gallon) for each site when I averaged each site's results. The water was hard making it difficult for metals to dissolve in it. Only the tests for fecal-coliform bacteria showed levels of contamination high enough for the creek water to be considered polluted.

Because I especially wanted to see if there was a connection between dissolved oxygen and fecal-coliform bacteria content, I was particularly interested in that data. I noted that as the creek moved farther away from the city, the dissolved oxygen in milligrams per liter increased. I also noted that the readings for dissolved oxygen levels were the lowest at night. When my data for dissolved oxygen levels is correlated with the fecal-coliform bacteria data from 1998-2001, it can be inferred that the dissolved oxygen and fecal-coliform bacteria measurements are inversely related. The level of oxygen increases as the fecal-coliform bacteria content drops. When thought about, however, this makes perfect sense. When the dissolved oxygen levels are too low, the source cannot support aquatic life (Dissolved Oxygen, par. 4). The same is true when the level of fecal-coliform bacteria is too high. As the creek moves farther away from the city, it becomes cleaner. This is either because as the water is exposed longer to the air it gains more dissolved oxygen, or because at each successive site the pace of the water quickened. If I were to redo this experiment, I would test at sites with the same water speed to see if there was a change in the levels of dissolved oxygen.

This graph shows the inverse relationship between dissolved oxygen and fecal-coliform bacteria as the stream progresses from the city. The values of fecal-coliform bacteria are plotted in units of 100 colony counts per 100 milliliters (data collected 10/11/00. Rudin, chart). The dissolved oxygen is in units of 1 milligram per liter (my data, collected 12/23/03).

Although the level of dissolved oxygen in the creek does increase as the water moves farther away from the city, the oxygen levels I found at each test site were high enough to sustain aquatic life. I have come to the conclusion that the reason my creek does not support aquatic life is because of its fecal-coliform bacteria content. There may, of course, be other things polluting the creek that I did not have the means to test. I am glad that the city is taking steps to correct the sewage problem. Someday I will look into my creek and see the fish swimming, scales flashing in the sun. Until that day I will continue to be interested in the health of my creek.




Smith, Zachary A. and Grenetta Thomassey. Freshwater Issues: A Reference Handbook. Denver: ABC-CLIO, 2002.

Pringle, Laurence. Water: The Next Great Resource Battle. New York: Macmillan, 1982.


Journal Articles

Cooper, Mary H. "The Issues." CQ Researcher, 24 November 2000: 955-964.


Web Sites

A Citizen's Guide to Understanding and Monitoring Lakes and Streams. 1 December 2003. Washington State Department of Ecology. Retrieved from the World Wide Web on 3 December 2003.

Chattahoochee Riverway Project. 27 January 2003. Retrieved from the World Wide Web on 3 December 2003.

Dissolved Oxygen. Retrieved from the World Wide Web on 3 December 2003.

Nitrate—A Drinking Water Concern. 2 July 2003. Michigan State University Extension. Retrieved from the World Wide Web on 6 January 2004.

On-Line Volunteer Water Quality Monitoring Guidebook. North Carolina Cooperative Extension. Retrieved from the World Wide Web on 4 December 2003.

Rudin, Andy. Melrose Park Neighbors Association: Cleaning up Mill Run... an Ongoing Challenge. 26 March 2002. Retrieved from the World Wide Web on 14 December 2003.

Water Hardness. Retrieved from the World Wide Web on 3 January 2004.