Skip common site navigation and headers
United States Environmental Protection Agency
Great Lakes Monitoring
Begin Hierarchical Links EPA Home > Great Lakes > Monitoring > Indicators > Water Indicators > Oxygen Depletion
R/V Lake Guardian
Great minds,
   Great Lakes

Beach closings
Benthic invertebrates
Data Projects


Great Lakes National Program Office Environmental Indicators

Dissolved Oxygen Depletion in Lake Erie


Oxygen is essential for all plants and animals to survive, whether they live on the land or in the water. Aquatic organisms rely on oxygen that is dissolved in the water. In most lakes and streams, the amount of oxygen in the water is continually being replenished by oxygen from the air. Sometimes, however, conditions exist in which the dissolved oxygen in the water is used up by organisms faster than it can be replaced from the air. If all the oxygen used up, the organisms will suffocate. Such conditions exist in Lake Erie in the summer. This report explains why the Central Basin of Lake Erie is susceptible to low dissolved oxygen concentrations, and it presents current data and historical trends.

The overall goal for the Great Lakes Water Quality Agreement (GLWQA) “ . . . is to restore and maintain the chemical, physical and biological integrity of the waters of the Great Lakes Basin Ecosystem.” More specific goals for the control of phosphorus include, “Restoration of year-round aerobic conditions in the bottom waters of the Central Basin of Lake Erie,” and “Substantial reduction in the present [1978] level of algal biomass to a level below that of nuisance condition in Lake Erie.”  The United States and Canada have spent billions of dollars toward the reduction of phosphorus loadings into Lake Erie in order to reduce the amount of algal growth, and subsequently, to restore aerobic conditions to the Central Basin during the summer season. 

The EPA Great Lakes National Program Office (GLNPO) monitors the status of dissolved oxygen in the water column at a fixed network of stations several times each summer. These data are then used to assess the timing, extent and severity of reduced oxygen conditions. Over several years, trends in the rate of oxygen depletion over the summer should reflect the impact of management programs to limit both point sources and non-point sources of phosphorus loadings. A description of the program and sampling design are available from GLNPO.

Why does Lake Erie have a problem with low levels of oxygen?

In recent decades, the bottom waters in the Central Basin of Lake Erie become anoxic (without oxygen) in the late summer. Aquatic creatures, including fish and bottom-dwelling animals, need oxygen in the water to live or they suffocate. The fish may be able to swim to better waters, but most of the other animals cannot.

The configuration of this part of Lake Erie is largely responsible for the problem. However, too many nutrients, especially phosphorus, make the problem much worse. Most of the excess phosphorus comes from human activities, including sewage treatment plants and agriculture.

Monitoring Program for Dissolved Oxygen in Lake Erie

Ten sampling stations for the dissolved oxygen monitoring program were selected to cover most of the affected area. In most years, each station is visited in early June, late June, mid-July, early August, late August and mid-September. The most severe oxygen depletion is observed in late August and/or mid-September.

How Does the Dissolved Oxygen Problem Develop?

In the summer, the water in the Great Lakes separates into two layers. The top layer is warmer than the bottom one, it receives the sunlight, and it mixes with oxygen from the air. The bottom layer is cooler than the top layer, it is usually dark, and it is cut off from the air so it cannot re-supply its oxygen. 

The Western Basin of Lake Erie is shallower than the typical thickness of the upper layer of water. Winds are able to keep the whole water column stirring, so there is plenty of opportunity for oxygen to dissolve into the water from the air. The bottom of the Central Basin is a broad, shallow plain that is slightly deeper than the typical thickness of the upper layer. Therefore, the bottom layer is relatively thin and contains a relatively small volume of water. In the deeper Eastern Basin, the bottom layer is much thicker and contains more water.

If there is much phosphorus in the water, it acts like fertilizer, and more algae will grow in the warm, sunlit top layer. Eventually these algae will sink into the dark bottom layer, where they stop growing and begin dying. Bacteria and fungi then decompose the plant organic matter. Bacteria and fungi also need oxygen to live, and they use what is available in the water. Because the bottom layer is cut off from the air, over the summer, less and less oxygen remains in the water. If there is a small volume of water and a lot of decomposition going on, as exists in the Central Basin, the oxygen will be used up faster than if there is much water and/or little decomposition. This is not a problem in the Eastern Basin largely because of the much greater volume of water in the bottom layer.

Each autumn, the top layer cools, and the wind mixes it deeper and deeper into the bottom layer. Eventually the whole water column is the same temperature, and the wind can again mix it from top to bottom and the oxygen can be restored from the air. In the Central Basin, this phenomenon typically occurs in early to mid-September, when oxygen is restored from top to bottom.

How Much Dissolved Oxygen is Enough?

There is no absolute value for how much oxygen in the water is enough. Depending on the water temperature, oxygen concentrations in equilibrium with the air can range from about 8 mg/liter (summer warm water) to over 14 mg/l (winter cold water). Some organisms are much more sensitive to low oxygen levels than others. Generally, the less oxygen available, the more stressed the organisms will be. Few species tolerate oxygen concentrations less than 1 or 2 mg/liter for very long. 

Seasonal Dissolved Oxygen Levels

Dissolved Oxygen Concentration
Lake Erie Central Basin Hypolimnion

Over the course of the summer, dissolved oxygen levels in the bottom waters steadily decline. In some years, e.g., 2001, 2002, and 2003, the decline was quite rapid and the minimum average concentration in the Central Basin was less than 1 mg/liter. In other years, e.g., 1993, 1997 and 2004, the decline was much slower, and in late August there remained an average concentration of 2.5 mg/L or greater. The apparent return of the rapid rate of decline of dissolved oxygen concentrations in 2001, 2002, and 2003 is of concern to many scientists and will continue to be watched closely. In most years much of the Central Basin had “turned over” between late August and mid-September, and the oxygen was restored. In 1997, however, the two water layers remained intact through mid-September, and the dissolved oxygen concentrations continued to decline to less than 1 mg/L. In 2004 in mid-September, there was no longer any true lower layer, and the thermocline (boundary layer between the upper and lower layers) was intersecting the bottom.

In general, less phosphorus in the water in the spring will result in fewer algae growing, which in turn means less organic matter to decompose. Less decomposition activity would take less dissolved oxygen from the water. As a result of phosphorus control programs, we could expect that the severity of oxygen depletion, the duration of the minimum oxygen levels, and the amount of the Central Basin area affected would all be reduced.

Other factors contribute to the variability from year to year, however, especially the thickness and the temperature of the bottom layer. A thicker bottom layer holds a larger volume of water and the rate of oxygen depletion will be less. Likewise, a cooler bottom layer will slow down the rate of decomposition and oxygen loss.

Lake Erie Central Basin Dissolved Oxygen Concentrations

The reduction in average dissolved oxygen concentrations in the Lake Erie Central Basin progresses from early June through at least late August. In 1985, average concentrations were less than 2 mg/l (red dot) by early August, and they were below 1 mg/l (black dot) from late August through mid-September. By 1993 conditions had improved so that average concentration remained above 6 mg/l (green dots) through early August. Only in mid-September did the oxygen concentration average less than 2 mg/l (red dot).

From 1997 to 2004, considerable variability was observed in the year-to-year dissolved oxygen concentrations in late summer. The rate and severity of oxygen depletion in 1997 was similar to that for 1993, although the average concentration was worse in mid-September. In 1998, 2001, and 2002, the oxygen was depleted more rapidly, approaching the rate observed in 1985. The average concentration was less than 1 mg/l by late August, but some recovery had begun by mid-September. The few data that are available for 2000 indicate that conditions were among the poorest in recent years. An additional survey was conducted in early September 2003, at which time little or no oxygen was present in the bottom waters. In 2004, there was suffiicient mixing of the upper water layer into the lower one that average oxygen concentrations remained above 2 mg/l from mid-August through mid-September.

Lake Erie Central Basin Dissolved Oxygen Concentrations

The reduction in dissolved oxygen concentrations from early June through mid-September is not uniform across all sampling stations in the Central Basin, and averaging the data together from all 10 stations can be deceiving. The similarity in the rate of oxygen depletion between 1993 and 1997 is reflected in the pattern of oxygen concentrations among the stations. In mid-September, the upper and lower water layers remained separated at most locations in 1997, while in 1993 the bottom water had mixed with the upper layer at most locations.

In 1998, 2001, 2003 and perhaps in 2000, oxygen depletion was well underway by mid-July, and most locations were nearly without oxygen by late August. In contrast, in 2004, oxygen concentrations in the bottom waters remained above 1 mg/l at all stations throughout the entire season. The bottom layer was not unusually thick, nor unusually cool in 2004. Therefore, some other factors may have been involved, e.g., reduced algal growth in the upper water layer or perhaps reduced influence of zebra mussels on Lake Erie nutrient recycling.

The three most westerly stations (left side of the dot clusters) exhibited a pattern of more rapid oxygen depletion than the other stations, and at these stations the two water layers were less likely to have mixed together by mid-September. For all years presented here, low oxygen conditions (less than 1 mg/l, black dots) were observed at least part of the season at all three stations. Also, for a least one of these stations each year except 2002, the low oxygen condition persisted from late August through mid-September.

The three most easterly stations (right side of the dot clusters) exhibited a pattern of less rapid oxygen depletion than at the other stations, and the two water layers were more likely to have mixed together by mid-September. Although low oxygen conditions were observed at one or two of these stations in all years except 1993, the condition did not persist over two sampling periods.

The individual station data are not available for the 1985 samples. In 1999, no data were available for early June, and in late June the data were collected by Environment Canada at slightly different locations than used by U.S. EPA. The differences in sampling operations do not affect the interpretation of the data for later in the season. For 2000, data are available only for August, but they indicate severe oxygen deficiency for that time. Environment Canada conducted 5 of the 6 surveys in 2004, but the same station locations were visited as during previous years.

In general, there appeared to be worsening conditions from 1993 to 2003. Oxygen concentrations were declining more rapidly earlier in the season in recent years, and the area of oxygen depletion (< 1 mg/l) seemed to be holding steady or increasing. The situation will continue to be monitored closely.

Thickness and Temperature of Bottom Water Layer

Variability in the rate of dissolved oxygen depletion, its severity, and its duration are related to year-to-year differences in the thickness and temperature of the bottom water layer in the Central Basin of Lake Erie. These differences are determined by the climate over Lake Erie in the spring, i.e., average air temperature and wind velocity. Rapidly climbing air temperature with calm winds will result in a thinner, warmer top water layer and a thicker, cooler bottom layer. A cooler, windy spring will permit the entire water column to warm to some extent before the top layer separates at a deeper depth, resulting in a warmer, thin bottom layer.

More oxygen will be retained longer into the season if the bottom layer is thicker and/or cooler than average, as in 1989 and 1997. In years that the bottom water layer is thinner and/or warmer than usual, as in 1990 and 1991, oxygen will be lost more quickly and the reduction may be more severe. For the 2003 season, the bottom layer was generally close to average thickness but warmer than average. These conditions would be expected to contribute to the relatively rapid rate of oxygen depletion over the season, and to its severity during August. In 2004, the bottom waters continually warmed and became thinner as the season progressed, probably from the upper water layer mixing into the lower one.

The influence of nutrients and algal growth on the rate and severity of oxygen depletion are superimposed on the natural variability due to springtime weather.

Oxygen Depletion Rate in Lake Erie Central Basin

To reduce the amount of variability in the year-to-year data about dissolved oxygen in Lake Erie, the data can be statistically adjusted to approximate the oxygen concentrations and the rate of depletion as if the bottom layer were always one thickness (4.6 meters) and one temperature (10 degrees Celsius). The resultant rate of dissolved oxygen depletion (mg/liter/month) is artificial for any given year, but the rates can be fairly compared between years.

From 1970 through 1989, the average oxygen depletion rate declined. The lowest rates were observed for the 1988 and 1989 data. From 1990 through 2004, however, the average adjusted oxygen depletion rate has remained nearly steady or perhaps has increased slightly. The calculated depletion rate for 2001 was encouraging, being the 3rd lowest in this time series and consistent with the long term trend. The depletion rates for 2003 and 2004, however, were again near the average from the years 1990 – 2004, and they were was not consistent with the long term trend observed from 1970 – 1989.

At least two factors may be influencing the rate of oxygen depletion in recent years. The concentration of total phosphorus in the springtime waters of the Central Basin of Lake Erie have been increasing slightly since 1990. Steady or increasing phosphorus levels could support continued algal growth which would generate enough organic matter to lead to oxygen depletion in the bottom waters.

The influence of the zebra mussels and quagga mussels (a close relative) on nutrient cycling and subsequently dissolved oxygen depletion in the Central Basin of Lake Erie is not well understood. They could have an ecological effect that obscures the rate of oxygen depletion from decomposition of algal organic matter.


The EPA Great Lakes National Program Office (GLNPO) monitors the status of Great Lakes waters each year in cooperation with other federal agencies. A description of the program and sampling design are available from GLNPO.


Begin Site Footer

EPA Home | Privacy and Security Notice | Contact Us