Mesocosms - Cutting Edge in Research?

How do we know whether a particular concentration of a pesticide in an estuary may pose risks to that ecosystem? Or to the various species that might be exposed?

How do we know that a concentration over a period of, let’s say, six days may pose little or no risk? But that the same concentration – or even a lesser concentration – may pose significant risk over a longer period of time? And, if so, how long?

Can we assume that a concentration, or a duration, with no negative impact on one species is benign also to others? Might juvenile fish or clams differ, for instance, from oysters or grass shrimp in their responses to concentrations, or to duration of the exposure?

Far being rhetorical or academic questions, those issues and others can be central to the well-being and even long-term survival of some estuarine resources, including commercially valuable species.

Marine scientists at NOAA’s Center for Coastal Environmental Health and Biomolecular Research, CCEHBR– one of the five centers in the agency’s National Centers for Coastal Ocean Science, NCCOS – are tackling those and similar issues head-on in a series of ongoing experiments.

They’re using a unique tool called “mesocosms” to closely simulate real-world natural ecosystems.

What ARE ‘Mesocosms’? Just WHAT Do They Do?

The layperson might think of a mesocosm – pronounced as in “mē-zo-cozm” – as a scientifically sound aquarium. That analogy may be helpful, but in this case the 12 mesocosms at CCEHBR’s Charleston, S.C., laboratory are finely tuned and meticulously maintained to simulate the specific ecosystem being studied.

The word “mesocosm” comes from Latin and literally means “middle world,” just as the more familiar term “microcosm” means “little world.”

The term dates back to one of the great American ecologists in recent history – the late Dr. Eugene Odom, of the University of Georgia, often described as the father of modern ecology.

Think of mesocosm experiments from one perspective as simplified field experiments containing some of the same structure and function of natural ecosystems. From another perspective, think of them as a laboratory experiment expanded to incorporate some environmental structure and function. That image gives real meaning to the “middle” – or “meso” – part of the term – about halfway between lab and field.

The 12 mesocosms at the CCEHBR Charleston, S.C., facility give the NCCOS marine scientists a self-contained and self-supporting system for simulating an estuary ecosystem much as it exists in nature. They use that system to evaluate short- and long-term effects of varying doses of a pollutant, for instance a pesticide, in a contained setting.

Marine biologist Paul L. Pennington, Ph.D., of the NCCOS/CCEHBR laboratory generally oversees the 12 – soon to become 36 – mesocosms. They’ve been an integral part of his research since 1998. The mesocosms had been the subject of a two-year pilot or demonstration project done in conjunction with scientists at Virginia Tech.

Taken collectively, the 12 replicate mesocosms at CCEHBR let researchers concurrently perform three experiments, each with, for instance, a control group and three separate treatment groups, with varying controlled doses of an insecticide, heavy metal, or even pharmaceutical or other material being tested. That approach lets researchers simulate a specific estuarine system, determine the stability of their simulation, and evaluate effects of various chemical contaminants, for instance a particular agricultural pesticide, on the ecology of each system.

Located in a greenhouse on-site, the mesocosms – each measuring four feet long by two feet wide and two feet deep – are used in particular for doing eco-toxicological work on small tidal creeks. The tidally controlled systems feature virtually identical flow rates to avoid even subtle discrepancies among the mesocosms that could influence test results. The mesocosms generally are used in specific experiments about nine months out of a year, with the remaining three months dedicated to system cleansing and recovery as they are readied for the next series of experiments.

In setting up and operating the 12 existing mesocosms – and another 24 expected to go online over the next year or so –the Charleston scientists had in mind a way of providing reliable and cost-effective simulations of coastal near-shore ecosystems, allowing results to be both reliable and consistent, and allowing for replication.

They say they are confident the existing system meets those criteria, sufficiently confident, in fact, that they are moving ahead with plans for another two-dozen microcosms as part of a greenhouse expansion just about to get under way.

Over the past six years, the scientists have used the mesocosms primarily for simulating releases and runoff events of numerous pesticides, as often occur during when it rains. Such releases have led to high mortality rates to a range of commercially important invertebrate and fish populations and also to other estuarine organisms. The researchers have found that the mesocosm results closely track actual in-the-field responses, providing results closely comparable to what happens with acute field runoffs.

Fipronil Test Illustrates Care Taken in Experiments

It’s helpful to look at a specific example, and Pennington points to the work with the pesticide fipronil as being illustrative.

Registered for use in the U.S. in 1996, fipronil over time degrades into several products, two of which are toxic to invertebrates and are relatively persistent in the environment, being absorbed onto soils and particulate matter. Fipronil is used to control various pests, including termite control, control of water rice weevil on rice, control of fire ants and mole crickets in turf grass management, and use on dogs and cats to control fleas. Effective against arthropods, fipronil poses a high risk to non-target crustacean species

In a peer-reviewed 2004 article in Environmental Pollution, E.F. Wirth, Pennington, and their co-authors reported on their testing of fipronil in the mesocosms. They dosed replicated mesocosms with intact marsh plots and seawater with “environmentally realistic” fipronil concentrations of 150, 355, and 5,000 nanograms (one-billionth of a gram) per liter.

The fipronil experiment was planned to run for 672 hours – 28 consecutive days – replicating a southeastern U.S. salt marsh dominated by the common Spartina alterniflora. In each of the 12 replicate mesocosm tanks, total volume, including the sump, was maintained at 300 liters, with a high-tide volume of approximately 250 liters and a residual low tide volume of 55 liters. Scientists added sediment and marsh grasses so that each tank consisted of two plots of low marsh and two plots of mid-marsh, along with a simulated stream bottom.

The Charleston scientists in March 2001 had taken plots of inter-tidal salt marsh from an undeveloped reference site along Leadenwah Creek, Wadmalaw Island, South Carolina. They collected low- and middle-tidal range marsh samples, removing mussels and crabs, and placed the marsh corps in sediment trays for transport back to the laboratory. They programmed the mesocosms’ tidal cycles for 10 a.m. and 10 p.m. high tides and 4 a.m. and 4 p.m. low tides.

Once the mesocosms were set up, the scientists let them acclimate to the new conditions for 90 days, all the time conducting daily monitoring of temperature, relative humidity, photoactive radiation, and water quality criteria such as salinity, pH, temperature, and dissolved oxygen. Each day, they added deionized water to each tank to maintain salinity and adjust for losses caused by evaporation.

The Testing Protocol Gets Under Way…

Three days before dosing the mesocosms with fipronil treatments determined after careful literature reviews, the scientists added 25 fish and 24 grass shrimp to each tank. The fish, C. variegates and the shrimp, both randomly selected, had been laboratory-reared from embryos – 16 of them males and eight females. Thirty Leadenwah Creek oysters, scrubbed clean of any sediment and algae, were added to each mesocosm, and also juvenile clams bought from a local commercial firm. The scientists placed the clams in jars with a consistent depth of sieved sediment, and they then placed the jars in a sediment tray to represent the creek bed.

Every seven days, the scientists assessed the survival of 10 field-collected shrimp, which they had caged in each mesocosm water sump. They removed the caged shrimp and added another 10 to each test – days 7, 14, 21 and all the way through day 49. At that point, they determined, the shrimp mortality no longer differed from the control group.

At the end of the planned 28-day run, they assessed organism survival for each species, having assured themselves that there were no statistically significant differences in water quality among the individual mesocosms.

…and Results Begin to Become Available

 For the clams, the scientists found no statistical differences among the control and three different dose treatments. Oyster survival also showed no statistical differences, although oysters exposed at the highest concentration, 5,000 ng/L, were substantially longer than those in the control group. Weights showed no statistical differences for the oysters.

The sheepshead minnows used in the tests and exposed for 28 days “appeared to tolerate fipronil exposure at the concentrations used,” the scientists say, and they found no significant differences of weight or length in the various tests.

The most pronounced response to the fipronil exposures came in terms of grass shrimp survival. Seventy- nine percent of grass shrimp in the control group survived after 28 days, but none in the 5,000 ng/L test.

In addition to evaluating grass shrimp mortality, the scientists examined surviving adult shrimp after 28 days of exposure and sorted them by sex and by reproductive status (male, ovigerous female, non-ovigerous female). They found that the average number of ovigerous females decreased as fipronil doses increased. Furthermore, the percentage of surviving male grass shrimp was also influenced by dose: as doses increased, males became a larger part of the total surviving shrimp population.

One Key Finding: Longer than Expected Persistent Toxicity

Among the important insights gleaned from this particular study, Pennington points to the finding that a single fipronil exposure was still toxic to grass shrimp between two and six weeks after does of 355 and 5,000 ng/L. The researchers’ finding pointed to “significant toxicity in the highest treatment” for six to seven weeks after the dosing.

Pointing to maximum concentrations of 8,000 ng/L reported by the Louisiana Department of Agriculture and Forestry in reviewing the onset of crayfish mortalities in Louisiana bayous, the Charleston scientists say their maximum concentration level of 5,000 ng/L is “well within the range of environmental concentrations” found in some Louisiana waters. They say surface water concentrations in those areas raise concerns of “significantly impacting grass shrimp populations” there. And they recommend further analysis of long-term fipronil exposures “to better understand the possible effects on chronic and sublethal endpoints” such as fish reproduction.

A Bright Future for Mesocosm Tests?

With recent fipronil registrations issued for use in residential fire ant control, increased applications seem likely in the future, and so too increased exposures of non-target species seem likely, the scientists say.

Over the next several months, Pennington expects to see the results of the first mesocosm research testing sediment quality guidelines for DDT, the pesticide long banned in the U.S. but still found in tissue sediments and in the environment.

By the summer of 2005, he is optimistic that the 24 new mesocosm systems will be up and running, and he envisions future testing may turn from the general focus on agricultural pesticides to look also at increasingly definitive studies of sediment quality guidelines… “to the point that we have some reasonable confidence” in the findings, as he says. He foresees a day too, perhaps three to five years down the road, when the increased mesocosm capacity will be useful also in evaluating pharmaceuticals, lipid-reducing compounds, antibiotics, and drugs controlling things such as heart rate.

“Those are in laboratory research now,” he says, “but eventually they may also be evaluated in mesocosms.”

With the southeastern U.S. increasingly trending from an agrarian economy to one facing the pressures associated with a growing coastal population, coastal sprawl, and increased applications of treated sewage on a growing number of golf courses (and with it the increased nutrient concentrations resulting from fertilizer supplements), Pennington, his colleagues, and their growing number of in-house mesocosms should have plenty to keep them busy in coming years.