A preliminary investigation into the bacteriological water quality problems of Stonelick Lake State Park, Ohio.

Table 1
Stonelick Lake bathing area fecal coliform results
          Fecal coliform         Log          Geometric
            Densities      Transformation       Means
Date      (per 100 ml)                       (per 100 ml)
06/05/90         340               2.53
06/12/90         100               2.00
06/19/90          42               1.62
06/26/90         250               2.39
06/28/90          90               1.95
07/02/90          47               1.647          102
07/10/90       1,400               3.14
07/17/90       1,100               3.04
07/23/90         340               2.53
07/25/90         370               2.56
07/30/90       1,600               3.20           494
08/07/90         230               2.36
08/14/90         130               2.11
08/22/90       3,700               3.56
08/28/90         310               2.49           430
09/04/90          17               1.23
09/11/90          85               1.92
09/18/90          54               1.73
09/26/90         630               2.79            84
Descriptive Statistics:
n = 19   Mean = 570  Std. Dev. = 890
Mean = 2.35   Std. Dev. = 0.61
Overall Geometric Mean = 231
The bathing area fecal coliform results for Harsha Lake are
only displayed to provide the reader general information. No
comparisons should be made between the bathing area fecal
coliform results of Stonelick Lake and Harsha Lake since the
watershed characteristics of each lake and the respective ages
of each lake differ greatly. The primary focus of this article
is on the Stonelick Lake bacterial water quality problems.
Watershed description
Stonelick Lake is located in the northeastern corner of
Clermont County, Ohio. Clermont County, located in southwestern
Ohio, lies just east of Hamilton County and the city of
Cincinnati. The lake was constructed by the Ohio Department of
Natural Resources in the early 1950s and it is now more than 35
years old. Water surface area at normal pool is 181 acres. The
original depth of the lake was 25 feet, but soil erosion
throughout the watershed over the years has caused the lake to
lose more than one-half of this original depth (2). Large
amounts of sediment have deposited in the upper one-third of
the lake as evident by the many emergent plants growing in and
around these submerged islands of sediment.
The 14,242 acre watershed area of Stonelick Lake encompasses
four Ohio counties: Warren, Brown, Clinton and Clermont.
Watershed land use consists of 10,541 acres of crop land (74%);
1,314 acres of woodland (9.2%); 1,259 acres of other land
(8.8%) with uses such as Stonelick State Park area, motorcross
area, farm out-buildings, paddock areas, etc.; 763 acres of
grassland (5.4%); and 375 acres of residential and developed
property (2.6%) (2).
Sixty-four percent of the soil types occurring in the watershed
are composed of the Avonburg-Clermont soil association. These
nearly level and poorly draining soils are derived from
Illinoisan glacial till and windblown loess. These soils are
further characterized by seasonal water tables and occasional
fragipans. Thus, these soils have severe limitations for
on-site sewage disposal (3).
Study methods
The staff of the CCGHD began sampling the bathing areas of
Stonelick and Harsha Lakes on June 5, 1990. Samples were
collected weekly using a standard protocol. The samples were
taken in 3 feet of water, upstream from all bathers and the
sampler. The samples were collected at a depth of 1 foot below
the water surface. Sterile water sampling bags were used to
collect the 100 ml samples. During sampling, physical water
conditions were noted. The conditions included turbidity, lake
level, number of waterfowl and general weather conditions. The
time of sampling, sample identification number, and the
sampler's name were recorded. The samples were immediately
placed in an ice-packed cooler and taken directly to the
laboratory by 2:00 p.m. each day, thereby maintaining
chain-of-custody as well as meeting the six hour holding time.
All sample information was reviewed in the laboratory and then
the lake samples were placed in a refrigerator. The membrane
filter technique as outlined in Standard Methods for the
Examination of Water and Wastewater, number 9222D, was followed
to determine fecal coliform densities (4). Fecal coliform
colonies were counted after the required incubation period and
reported in the number of fecal coliform colonies per 100 ml.
All fecal coliform data collected was entered into a Lotus
1-2-3 computer program. Regression and correlation analysis was
conducted on the CCGHD data. This was done to determine the
strength of relationship between variables.
Results and discussion
The fecal coliform densities for the Stonelick Lake Beach
bathing area collected by the CCGHD are shown in Table 1. The
fecal coliform results for the bathing areas of Harsha Lake are
shown in Table 2. Table 3 displays the Stonelick watershed
fecal coliform results which includes the bathing area data.
Unfortunately, over the eight week period, samples could not be
taken at all locations due to no stream flow or
inaccessibility. Graph 1 depicts the fluctuations in fecal
coliform densities during the entire sampling period for the
bathing area.
The bathing area fecal coliform results in Table 1 are highly
variable, which is not unexpected. The arithmetic mean is 570
per 100 ml and the standard deviation is 890 per 100 ml. All
results were transformed into logarithmic values before
conducting regression and correlation analysis in order to
provide for a more symmetrical distribution and to eliminate
extreme values which affect the statistical analysis of the
data.
The high fecal coliform value of 3,700 colonies per 100 ml,
sampled on August 22, was preceded by a large rainfall. A 3.1
inch rain had fallen the day before. The monthly geometric
means for the months of June, July, August and September were
102, 949, 430 and 84 colonies per 100 ml, respectively. The
overall geometric mean is 231 per 100 ml. The safe bathing
water standard of 200 fecal coliform colonies per 100 ml was
exceeded for the months of July and August. In September, the
bacterial levels had fallen to within safe limits. The high
fecal count found on September 26 is somewhat puzzling but may
be associated with the lake mixing effects that occur with fall
turnover. The lake water temperature was noticeably cooler than
when sampling began.
Graph 2 compares the density of fecal coliform levels between
bathing area waters (STLK 5) and the waters of the main
tributaries (Stonelick and Locust Creeks) at the lake inlet
(STLK 7). It is readily apparent that as the bacterial levels
of the lake inlet change, the bathing area bacterial levels
change accordingly. Differences in bacterial counts can be
explained by the dilutive and mixing effects of the lake.
Further explanations include the lag time for water movement
between the two sample locations, allowing for bacterial
die-off.
In calculating regression and correlation analysis on sample
locations STLK 7 vs. STLK 5, a strong relationship is evident.
A positive correlation coefficient of r = .72 at a level of
significance of alpha = .05 and n = 8 gave a p |is less than~
.05, indicating the observed fluctuations are statistically
significant.
There are several sources for the fecal coliform bacteria
levels in Stonelick and Locust Creeks. In considering the
watershed above sample location STLK 7, Figure 3 reveals four
on-site sewage system concentrations (25 plus homes) and three
livestock operations (less than 1,000 animal units) operating
in the watershed. Several hundred homes scattered throughout
the watershed have on-site sewage disposal systems 20 years old
and older. The Village of Blanchester started construction in
1990 on a public wastewater system. This project should
eventually eliminate all of the malfunctioning septic systems
in the Fairground Acres subdivision area of Blanchester.
Table 2
Harsha Lake bathing area fecal coliform results
             Fecal coliform        Log         Geometric
               Densities      Transformation       Means
Date         (per 100 ml)                      (per 100 ml)
06/12/90        7.7              0.88
06/19/90       23.0              1.36
06/26/90       16.0              1.20
07/02/90        4.3              0.63
07/10/90       30.0              1.47
07/17/90       14.0              1.14
07/25/90        2.8              0.44
07/30/90        4.3              0.63
08/09/90        1.0              0.00               8
08/15/90        2.1              0.32
Descriptive Statistics:
n = 10  Arithmetic Mean = 10.5   Std. Dev. = 9.9
Sum = 8.07  Mean = 0.81  Std. Dev. = 0.48
Overall Geometric Mean = 6

Sample location STLK 3 is located in the lake, near the outlet of Stonelick's lake wastewater lagoon system. This two-stage lagoon system treats the sewage from the state park campgrounds. Table 3 shows that the majority of the fecal coliform results from STLK 3 are higher than the bathing area STLK 5. The regression and correlation analysis on sample locations STLK 3 vs. STLK 5 gave one of the strongest correlation coefficients of samples analyzed. The correlation coefficient r = .80 at a level of significance of alpha = .05 and a sample size of n = 7 gave probability value utilizing the student's t test of p |is less than~ .05. The bacterial levels between STLK 3 and STLK 5 are statistically significant.

Stonelick State Park's wastewater system is in need of repair. The sludge has accumulated in the lagoons without maintenance since the park's inception in 1953. Inspection of the wastewater system revealed the system was discharging directly into the lake. The lagoon outlet is submerged under the lake and there is no disinfection prior to discharge. The wastewater treatment system is not being operated properly and is discharging to the waters of the state without a National Pollutant Discharge Elimination System (NPDES) permit.

Sample location STLK 4 is located in an unnamed tributary adjacent to the park's wastewater treatment system. Samples were collected from this tributary approximately 100 yards from the lake. The park's wastewater collection system crosses the tributary just above the sampling point. This tributary outlets into Stonelick Lake about 50 feet downstream from sample location STLK 3. Graph 3 displays the fecal coliform results from both STLK 3 and STLK 4 in comparison with the fecal coliform results from the bathing area (STLK 5). The close association of the fecal coliform levels between sample locations STLK 3 and STLK 4 is readily apparent in Graph 3. In conducting regression and correlation analysis on STLK 4 vs. STLK 3, an r value equal to .95 is calculated. In conducting regression and correlation analysis on STLK 4 vs. STLK 5, r = .75, at a level of significance of alpha = .05, n = 7, p = .05. Both STLK 3 and STLK 4 fecal coliform fluctuations are statistically significant, relative to those levels found at STLK 5.

The strong relationship between the fecal levels found at STLK 3 vs. STLK 4 could be due to leakage from the park's wastewater collection system into the tributary near location STLK 4. There are no known on-site sewage disposal systems or livestock operations within the watershed of the STLK 4 sampling site. Practically all of the watershed of this unnamed tributary lies within the state park property.

In conducting regression and correlation analysis (alpha = .05) on all other sample locations--STLK 1, STLK 2, STLK 6, STLK 8 and STLK 9--to the bathing area (STLK 5), no statistical significance was found.

The r values and the resulting p values for these locations, in comparison to STLK 5, are as follows:

* STLK 1 vs. STLK 5 -- could not analyze due to few samples

* STLK 2 vs. STLK 5 -- r = .049

* STLK 6 vs. STLK 5 -- r = .38; p = .32 not significant

* STLK 8 vs. STLK 5 -- r = .37; p |is greater than~ .40 not significant

* STLK 9 vs. STLK 5 -- r = .64; p |is greater than~ .2 not significant

It is interesting to note that the largest source of fecal coliform bacteria entering the lake was from location STLK 6 but the levels did not strongly correlate with those bacterial levels found at STLK 5. This may be due to the location of STLK 6 relative to its outlet at the lake. Referring to Figure 1, there is a sizeable distance between STLK 6 and the lake.

Bacterial levels most likely are dropping off during low flow conditions by the time the water reaches the lake. Furthermore, there is small cove where STLK 6 enters the lake which could allow for quiescent settling, since it also seems plausible the main lake current is to the opposite side of the lake at this point. The probable source of the fecal coliform bacteria are from a swine livestock operation located about 100 yards upstream from the sample point.

Conclusions

There is no single or primary source of the fecal coliform pollution of Stonelick Lake. Throughout the watershed, there are many sources which are loading fecal coliform bacteria to the bathing area.

Figure 3 shows a few of the larger sources of fecal contamination. There are five on-site septic system clusters, five livestock operations, and one public wastewater system (ODNR) draining wastewaters into Stonelick Lake.

There is no question that the evidence of fecal coliform bacteria in bathing waters represents a health threat to the exposed individuals. Pathogenic bacteria are commonly found when a species enumeration or analytical profile index is conducted on the fecal coliform filter sample (5). Even though the fecal coliform test is not the perfect indicator for fecal pollution, it is still widely used and accepted. The use of the fecal coliform/fecal streptococcus ratio as an investigatory tool for determination of human versus animal fecal contamination of water is no longer recommended by the Ohio Department of Health (4). Testing for Escherichia coli would be a more specific test for warm-blooded fecal pollution.

To abate the many sources of fecal pollution in the Stonelick Lake Watershed will require inter-agency governmental co-operation along with cooperation from the private landowners. The respective health districts in each of the four counties will need to inspect and issue repair orders on the malfunctioning on-site sewage disposal systems. The Division of Soil and Water Conservation along with the respective TABULAR DATA OMITTED soil and water conservation districts in each of the counties will need to work with the livestock farmers in abating and managing the resulting animal waste. Also, the soil and water conservation districts need to assist farmers in controlling excessive soil erosion.

The Division of Parks and Recreation will need to repair their wastewater system so it no longer discharges directly into the lake. The Division of Parks and Recreation should also proceed to dredge out the sediment that has accumulated in Stonelick Lake over the years. Research studies have argued that fecal coliform bacteria can survive for several days in the sediment bottoms of lakes rich in organic matter, especially when a continuing recharge of fecal coliform bacteria is occurring (6). During sampling, a common observation noted was that when the lake was turbid due to rainfall, the fecal coliform counts were higher than when the lake was less turbid.

In order to systematically address and abate the fecal bacteria problems of Stonelick Lake, a watershed protection action plan needs to be written by the agencies mentioned above with implementation dates. Failure to take action will result in future water quality advisories and beach closures at Stonelick Lake.

Timothy I. Ingram, Clermont County General Health District, 2400 Clermont Center Dr., Suite 203, Batavia, OH 45103.

References

1. Health Education Service (1990), Recommended Standards for Bathing Beaches, A Report of the Committee on the Great Lakes--Upper Mississippi River Board of State Public Health and Environmental Managers, Albany, NY.

2. Ohio-Kentucky-Indiana Regional Council of Governments (1986), Stonelick Lake: An Analysis of Pollution Sources, Cincinnati, OH.

3. United States Department of Agriculture--Soil Conservation Service (1975), Soil Survey of Clermont County, Ohio.

4. American Public Health Association (1989), Standard Methods for the Examination of Water and Wastewater -- 17th Edition, Washington, D.C.

5. National Academy of Sciences (1977), Drinking Water and Health, Washington, D.C.

6. Struck, Philip H. (1988), The Relationship Between Sediment and Fecal Coliform Levels in a Puget Sound Estuary, J. Env. Health 50(7):403-407.