Testing the Waters: Ohio

Ohio monitors 63 public and semipublic beaches along 7.3 miles of Lake Erie shoreline. The state's beachwater quality monitoring program is administered by the Ohio Department of Health (ODH).

Tracking the 2011 Harmful Algal Bloom in Lake Erie

The western part of Lake Erie sometimes experiences blooms of Microcystis, a type of cyanobacteria. When close to shore, these blooms foul the beaches, producing a rotten smell. More important, some forms of Microcystis produce liver toxins; although these usually cause nothing more than skin and intestinal problems in humans, they have in some cases caused death in pets, wildlife, and livestock. These blooms can also contribute to the Lake Erie dead zone, an area of depleted oxygen that threatens the lake's billion-dollar fishery.

The Microcystis bloom that developed in Lake Erie in 2011 was by far the largest one in recent years.1 Serendipitously, 2011 was the year that NOAA began testing a bulletin that provided a weekly forecast for Microcystis blooms in western Lake Erie based on satellite imagery. The bulletin depicted the current location of the bloom, predicted its future movement, and categorized its intensity.2 The NOAA bulletin archives3 show that a suspected bloom was beginning to develop on July 22, 2011. A high concentration of algal toxins in the vicinity of Toledo Light #2 confirmed this bloom by July 28. By late summer, the western portion of the lake and part of the central portion were covered by this bloom, which at one point extended more than 12 miles from shore and was more than 60 feet deep.4 The bloom continued to spread until September 8, when it stretched along the Ohio coast of Lake Erie from Maumee Bay to Catawba Island and along the Michigan coast from Northern Maumee Bay to the mouth of the Detroit River. The easternmost portion of the bloom was observed past Point Pelee and to the northeast in Rondeau Provincial Park in Ontario. Although it persisted around the Bass Islands, Pelee Island, and Kelleys Island to Cleveland, the bloom began dying because of falling water temperatures by October 20. By the following week only the coast off Lorain, Ohio, still showed signs of a bloom.3 In Ohio, the bloom resulted in 203 days of beach advisories at Kelleys Island State Park in Ottawa County, Maumee Bay State Park (Erie) in Lucas County, and Lion's Park and Battery Park in Erie County.

What Does Beachwater Monitoring Show?

In 2011, Ohio reported 63 coastal beaches. Of these, 9 (14%) were assigned a monitoring frequency of daily, 34 (54%) a frequency of more than once a week, and 18 (29%) once a week. There was no information for 2 beaches (3%). In 2011, 22%5 of all reported beach monitoring samples exceeded the state's daily maximum bacterial standard of 235 colonies/100 ml. Samples taken at 29 of Ohio's beaches exceeded the standard at least 20% of the time. The beaches with the highest percent exceedance rates of the state standard in 2011 were Edson Creek in Erie County (52%); Lakeview Beach in Lorain County (51%); and Villa Angela State Park (50%), Edgecliff Beach (50%), Moss Point Beach (47%), and Sims Beach (47%) in Cuyahoga County. Beaches in Cuyahoga County had the highest exceedance rate of the state standard in 2011 (33%), followed by Lorain (33%), Erie (18%), Ashtabula (16%), Lucas (16%), Ottawa (16%), and Lake (15%) counties. NRDC considers all reported samples individually (without averaging) when calculating the percent exceedance rates in this analysis. This includes duplicate samples and samples taken outside the official beach season, if any.

* Please note that only samples from a common set of beaches monitored each year from 2007-2011 are included in the bar chart.

What Are Ohio's Sampling Practices?

In 2011, for the first time, all sampling activities in Ohio were conducted by local entities and ODH did not itself directly monitor any beaches. The monitoring season varies from location to location, depending on which entity is conducting the monitoring, but generally runs from Memorial Day through Labor Day.

Ohio is a "home rule" state, and the state can only recommend sampling practices, standards, and notification protocols and procedures to local entities that participate in the beachwater quality monitoring program. Guidance recommends that samples be taken in water that is 3 feet deep, 1 foot below the surface. For the most part, monitoring is conducted at the area of the beach used most by the public. Beaches are prioritized for monitoring based on visitor use and water quality history, so beaches visited the most frequently and/or having a potential for contamination (Tier 1) are sampled the most frequently. ODH states that all of the Lake Erie beaches identified by the Ohio Department of Natural Resources are monitored at least weekly, with the majority sampled at least four times a week.4

Many of the beaches in the program are already sampled daily or as frequently as laboratory availability will allow. The monitoring frequency of these beaches does not increase after a bacterial exceedance has been found, but if an exceedance is discovered at a beach that is monitored only once a week, resampling may be conducted on the next business day.

How Many Beach Advisories Were Issued in 2011?

Total advisory days for approximately 579 events lasting six consecutive weeks or less increased 3% to 1,301 days in 2011 from 1,259 days in 2010. For prior years, there were 1,012 days in 2009, 783 days in 2008, 657 days in 2007, 629 days in 2006, and 182 days in 2005. In addition, there were 3 extended events (165 days total), all due to the algal bloom, and no permanent events in 2011. Extended events are those in effect more than six weeks but not more than 13 consecutive weeks; permanent events are in effect for more than 13 consecutive weeks.

For the 579 events lasting six consecutive weeks or less, 93% (1,208) of advisory days were due to monitoring that revealed elevated bacteria levels, 4% (51) were preemptive based on the results of computer modeling, 3% (38) were preemptive for other reasons (algal bloom), and <1% (4) were preemptive due to known sewage spills or leaks.

How Does Ohio Determine When to Warn Visitors About Swimming?

Local jurisdictions have the authority to close beaches and to issue advisories. Beaches are rarely closed because of elevated bacterial counts alone.4 Ohio uses an E. coli single-sample maximum standard of 235 cfu/100 ml for beach advisory decisions. No geometric mean standard is applied when making advisory decisions.4

The state recommends that local authorities issue advisories when the bacterial standard is exceeded. In Cuyahoga County, resamples are taken to confirm an exceedance before an advisory is issued.4

Ohio uses a predictive model called Nowcast at three of its Lake Erie beaches: Huntington Beach and Edgewater State Park in Cuyahoga County and Maumee Bay State Park in Lucas County. The model relies on environmental factors including rainfall, turbidity, and/or wave height to predict E. coli levels. It is continuously under improvement and can be tailored to provide predictions that are appropriate for a particular beach and even a particular portion of the swimming season for a particular beach. Predictive models are useful because they allow advisories to be issued the day that bacteria levels are suspected to be high. In contrast, when advisories are issued on the basis of E. coli counts determined by culture methods, they are issued the day after standards are exceeded because it generally takes 24 hours for culture results to be available. Many times, the culture results of samples taken on the day a beach is placed under advisory reveal that the water quality was actually acceptable on the day of the advisory.

Bacterial monitoring at Huntington and Edgewater State Park beaches has shown that Nowcast-based decisions about notifications are more protective of public health than decisions based on bacterial monitoring.6 The beach at Maumee Bay State Park began using Nowcast to issue advisories for the first time during the 2011 swim season. The model did not perform as well as the models at Huntington and Edgewater Beach and will probably be adjusted for the 2012 season in order to improve its accuracy.4

Work is under way to expand the number of beaches where predictive modeling is used to issue advisories. During 2011, Century Beach and Lakeview Beach, both in Lorain, were monitored seven days a week in order to develop a Nowcast model that will predict water quality at these beaches.4

There are no preemptive rainfall standards at beaches in Ohio, but beach managers may issue preemptive rainfall advisories if they feel that rain has compromised water quality.7 Beach managers may also restrict beach access because of sewage or other pollution spills, or because of any other threat to public health.

Notes

1. Thomas Bridgeman, "Harmful Algal Blooms in Western Lake Erie," presented October 12, 2011, accessed at www.glc.org/announce/11/pdf/Bridgeman-UToledo.pdf, March 2012.
2. National Oceanic and Atmospheric Administration (NOAA), "Harmful Algal Blooms in Lake Erie-Experimental HAB Bulletin," accessed at www.glerl.noaa.gov/res/Centers/HABS/lake_erie_hab/lake_erie_hab.html, March 2012.
3. NOAA, "Harmful Algal Blooms in Lake Erie—Experimental HAB Bulletin Archive," accessed at www.glerl.noaa.gov/res/Centers/HABS/lake_erie_hab/lake_erie_hab_archive.html, March 2012.
4. Ohio Department of Health, "2011 Bathing Beach Monitoring and Notification Program Results," not dated.
5. Why don't the 2011 percent exceedance values in this summary match? Only samples from a common set of beaches monitored each year from 2007–2011 are included in the bar chart. Because some beaches were not monitored in each of those years, the percent exceedance for this subset of beaches (25%) did not have the same value as the percent exceedance for all of the beaches monitored in 2011 (22%).
6. Ohio Nowcast, "How Did the Nowcast System Perform in Past Years?" accessed at www.ohionowcast.info/nowcast_perform.asp, March 2012.
7. Mary Clifton, Ohio Department of Health, personal communication, May 2012.
8. Reported advisory days are for events lasting six consecutive weeks or less. Days in parentheses are for events lasting more than six consecutive weeks.