Salt-Loving Algae Wipe Out Fish in Appalachian Stream
Alga’s sudden move from Texas prompts fears for other vulnerable
streams; many ask whether the invader hitched a ride on natural gas
drilling equipment.
Environmental Science & Technology
11 November 2009
By Rebecca Renner
A salt-loving alga that killed tens of millions of fish in Texas has
struck for the first time in an Appalachian stream that flows along the
border of Pennsylvania and West Virginia. Prymnesium parvum or “golden
algae” caused the sudden death of thousands of fish, mussels, and
salamanders in early September along some 30 miles of Dunkard Creek.
University and government scientists fear the disaster could presage
further kills in the region. Streams at risk due to high concentrations
of total dissolved solids (TDS) include portions of the northern branch
of the Potomac River and 20 other streams in West Virginia, according
to state scientists. Pennsylvania, Maryland, Virginia, and Kentucky
also have many vulnerable rivers and streams, according to U.S. EPA
scientists.
Dunkard Creek is a tributary to the Monongahela River, where last year
high TDS levels fouled industrial equipment and ruined the taste of
drinking water. Faced with projected increases in TDS as a result of
the burgeoning and water-intensive natural gas hydraulic fracturing
activity at the Marcellus Formation, Pennsylvania Department of
Environmental Protection (PA DEP) recently proposed TDS standards for
end-of-pipe discharges of 500 parts per million (ppm) TDS and 250 ppm
each for sulfate and chloride.
Despite historically high TDS levels, the creek was a good fishing
stream with small mouth bass, muskie, mussels, and salamanders,
according to biologist Frank Jernejcic with the West Virginia
Department of Natural Resources. In just a few days the algal bloom
wiped out the creek’s 18 species of fish and 14 species of freshwater
mussels—the most diverse population of mussels in the Monongahela
basin. “This is the worst fish kill I’ve experienced in 21 years in
West Virginia,” says Paul Ziemkiewicz, director of the Water Research
Institute at West Virginia University.
Relatively high levels of sulfate and other dissolved salts have been
common in Dunkard Creek over the past 10 years as a result of active
and abandoned coal mine discharges, according to West Virginia
monitoring data. But immediately before the bloom, chloride (300 ppm),
sodium (>3000 ppm), TDS (9500 ppm), and electrical conductivity
(>50,000 microsiemens per centimeter) all skyrocketed to
unprecedented levels, prompting biologists to initially blame the
chemical contamination for the aquatic devastation.
Now that the algae have been identified as the immediate cause of the
fish kills, biologists wonder if the soaring salt levels somehow
initiated the bloom. If so, there is evidence that Dunkard Creek is not
alone in recently receiving record high chloride, associated with
hydraulic fracturing or coal-bed methane wastewaters, not coal mine
water, according to Ziemkiewicz. Using water monitoring and stream flow
data, he calculated chemical loadings to the Monongahela and its
tributaries. “Our mass balances can account for most of the sulfate but
not the sodium and chloride,” he says. “Concentrations of sodium and
chloride were much higher than usual in Dunkard Creek during the fish
kill—but even these high levels do not account for the loads in the
main stem of the Monongahela River. We are missing major sources of
those ions” he says. It is not currently possible to track the fate of
hydraulic fracturing wastewater because, unlike coal mine discharges,
it is not subject to permitted discharge controls under the Clean Water
Act.
Midas touch
First identified in the 1930s, P. parvum is a microscopic
flagellated organism that caused massive fish kills in the Sea of
Galilee and in Israeli fish farms in the 1950s. Toxic blooms have also
occurred in brackish waters in Europe as far north as Scandinavia and
in China. The algae thrive in naturally brackish water typical of
rivers and reservoirs in East Texas, Oklahoma, and Wyoming. Since the
first documented fish kill in Texas in 1985, when more than 100,000
fish died in the Pecos River, the organism has killed more than 18
million fish valued at more than $7 million. In 2001, P. parvum killed
the entire year’s production of striped bass in Texas’s Dundee State
Fish Hatchery.
P. parvum’s numbers usually remain low. But sometimes it rapidly
reproduces with blooms that give the affected water a golden color. In
Texas, blooms usually occur during the cooler months when the alga
seems to have an advantage over other algae that grow best in warm
waters.
The algae’s toxins do not threaten humans, livestock, or wild animals.
But the algal toxins break gill membranes so that unwanted chemicals,
in particular calcium, pour in, says James Grover, a biologist at the
University of Texas Arlington. “The cells fill up with calcium and
explode,” he says.
The exact conditions that bring on an algal bloom are unknown, “That’s
the $64,000 question,” says biologist Carmelo Tomas at the University
of North Carolina Wilmington. “Studies have looked at nitrogen,
vitamins, and trace metals, but these present a confusing story,” he
adds.
Texas’s experience suggests that the algae need an ecological
opportunity, according to Luci Cook-Hildreth, Golden Alga Coordinator
with the Texas Parks and Wildlife Department. In Texas, blooms occur
mainly in the winter months. “When temperatures cool down, that knocks
back other algae. Then if we have a couple of clear sunny days, that’s
when we tend to get a bloom,” she says. Ziemkiewicz says conditions
were similar on Dunkard Creek just before the bloom.
Many streams vulnerable
West Virginia estimates that 21 streams could be at risk, based on
having electrical conductivities greater than 1500 microsiemens per
centimeter, which converts to about 750 to 1000 ppm TDS. This is a good
preliminary assessment, says Grover. In Texas, blooms appear to require
water with at least 500 to 1000 ppm TDS. “Specific ions are not
crucial,” says Grover, “but I get the impression that the more calcium
in the mix, the better it grows,” he says. Calcium also enhances the
toxicity.
Blooms are not as toxic in slightly acidic water. “We think this is
because there are ionizable groups on the toxin, so that the toxin
becomes more potent as pH rises,” Grover says. “There’s a lot we don’t
know, but I believe that we know enough about the overall salinity and
the role of calcium and pH to offer guidance, and West Virginia’s on
the right track.”
Blooms can occur when surface water temperatures are as low as 12
°C.
They are unpredictable, but Grover, who is trying to develop a
predictive model for P. parvum, believes this may be due to variable
rainfall. “If we have a few wet months, the higher flow flushes out the
algae. I think that’s why it comes and goes,” he says.
Fresh water
The algae’s need for high TDS suggests a control measure in the
Appalachians because, unlike East Texas, the high TDS is a man-made
problem, according to Tomas. He notes that in 2001, a North Carolina
striped bass business excavated and filled acres of ponds with
artisanal well water. The bass did not grow well because the water was
too soft. So the fish farmer changed the ponds to brackish water with
TDS of about 4 parts per thousand. The fish thrived at first, but soon
they died as a result of a P. parvum bloom. It wasn’t until the fish
farm went back to very low salinity water that the cycle of algal
blooms ended.
The case study makes it clear that a switch from water high in TDS to
low TDS water can prevent the algae from blooming, but the lower bound
for the TDS is unknown. Pennsylvania’s proposed discharge standards of
500 ppm TDS and 250 ppm each for sulfate and chloride would appear to
keep the salinity below levels likely to cause an outbreak.
Hitching a ride
In early November, scientists found P. parvum in another stream
close to Dunkard Creek. While some workers are extending this quest,
others are trying to explain how a salt-loving organism common in East
Texas and coastal environments found its way north of the Mason−Dixon
line. Sea birds such as seagulls could have transported the algae, but
many wonder whether P. parvum could have hitched a ride in a water
tanker truck or other drilling equipment that has moved from the fields
of the Barnett Shale in East Texas to the Marcellus Shale in
Appalachia. “The movement of water in tankers for coastal species has
been shown to be a powerful way for alien species to invade,” says
Grover. “Cells or spores can survive in ships’ ballast water. I don’t
know anybody who has looked at smaller tanker trucks or other such
equipment, but it must be a consideration,” he says. A recent New York
state environmental impact study likewise sounds a now somewhat
prescient warning: “potential mechanisms for the possible transfer of
invasive aquatic species may include trucks, hoses, pipelines, and
other equipment used for water withdrawal and transport,” the report
states. Residents who live along Dunkard Creek have seen drilling
tankers remove water from the creek; this suggests the physical
possibility of such a transfer. In Pennsylvania, such withdrawals do
not require a permit as long as they are small in comparison to the
volume of the stream, according to PA DEP spokesperson Helen Humphreys.
Genetic studies are currently under way to see if P. parvum from
Dunkard Creek is genetically similar to strains from Texas. “The
genetic studies may show that there is a lot of variability. This might
suggest multiple introductions from multiple sources,” says Grover.
“But if the Dunkard Creek strain matches those from Texas, well maybe
that looks a little fishy,” he says sadly.
Copyright © 2009 American Chemical Society