Bay Journal

Air regs have helped water quality more than previously thought

Nitrate concentrations in 9 forested watersheds decreased between 31.9 percent and 72.6 percent.

  • By Karl Blankenship on January 01, 2014
  • Comments are closed for this article.
Nitrate runoff from forested watersheds has declined sharply since late 1990s, when stricter air regulations were implemented.  (Dave Harp)

The Chesapeake Bay may have reaped greater benefits from air pollution reductions over the last two decades than previously estimated, new research suggests.

Emissions of nitrogen oxides, or NOX, have declined sharply since the late 1990s as a result of increasingly stringent regulations aimed primarily at reducing acid rain and ground-level ozone, or smog. When those pollutants fall to the ground, they also contribute nitrogen pollution to the Bay.

But until now, it’s been thought that the water quality benefits to the Chesapeake from air pollution controls were relatively modest, largely because the greatest reductions in nitrogen deposition took place in the heavily forested western edge of the watershed, which is downwind of large Midwestern power plants.

But a new study suggests otherwise.

Researchers with the University of Maryland Center for Environmental Science reported in a recent paper that nitrate (the most common form of nitrogen) concentrations in streams and rivers draining nine forested watersheds along the Appalachian spine have declined sharply since the late 1990s — a decline that corresponded with implementation of tougher air pollution control regulations.

“It really kind of blew us away,” said Keith Eshleman, the paper’s lead author. “I don’t think this is really what people predicted.”

It was unexpected because scientists typically consider forests to be effective nitrogen sponges that readily absorb almost all of the nitrogen that lands on them.

After all, a forest already contains, and naturally recycles, a huge amount of nitrogen, leaking on average only a few pounds of nitrogen per acre into streams. If a forest already contains, say, 2,000 pounds of nitrogen per acre in its trees, soils and other plants, it seems that another 10 pounds from atmospheric deposition wouldn’t make much of a difference.

“So it makes perfect sense to think that the system would be largely unresponsive to a change in nitrogen deposition based on that argument,” Eshleman said.

But data analyzed by he and his colleagues Robert Sabo and Kathleen Kline showed that nitrate concentrations in the nine watersheds decreased between 31.9 percent and 72.6 percent, with an average of 46.4 percent. Generally, the greatest nitrogen reductions were in areas that had the greatest reductions in atmospheric deposition — especially in the headwaters of the Potomac River in Western Maryland.

Eshleman said the reason appears to be that much of the nitrogen deposition arrives during times when forests have little opportunity to absorb it. That includes heavy rains, when more nitrogen has the potential to be flushed into streams before it can be absorbed, and especially during the winter when there is little biological activity to use any nitrogen falling to the ground.

“We know a lot of this nitrogen comes in as pulses. It is coming in during extreme events,” Eshleman said. “It may be coming in during periods where the forest has very little capability to take it up, either because the soils are frozen or the infiltration capacity of the soil is such that it is exceeded by the rate of rainfall.”

Some of the watersheds monitored by the team were small, only about a square mile, but some were large, including all of the Potomac River above Hancock, MD. All were more than three-quarters forested, so there were limited nitrogen inputs from other sources.

With one exception, the watersheds had at least 20 years of monitoring data available. And in all of them, the scientists observed declining nitrate levels in streams beginning in the late 1990s and accelerating in the last decade. That coincides with monitored reductions in nitrogen deposition resulting from the implementation of a series of Clean Air Act regulations.

“This is really a freebie,” Eshleman said “The lion’s share of it was done for tropospheric ozone — human health — as much as anything. The fact that you are getting any bang for the buck with regard to water quality is just an extra.”

Those findings also suggest that the state-federal Bay Program partnership may have underestimated the amount of nitrogen reductions that stemmed from air pollution controls. Recent computer model-based estimates from the Bay Program show that air pollution controls have reduced the amount of nitrogen entering the Chesapeake by 10 percent since 1985. That 35 million-pound reduction accounts for about a third of the estimated nitrogen reductions achieved to date.

But the Bay Program models generally assume that only about 10 percent of the airborne nitrogen landing on forests actually makes it to streams. That’s in line with published literature, but significantly less than what Eshleman observed in all of the nine watersheds. “You are getting way more bang here than the Bay Program is proposing,” he said.

Rich Batiuk, associate director for science with the EPA’s Bay Program Office, said the Bay Program is already examining Eshleman’s data, and his work may spur changes as its models and other decision-support tools are updated in the next several years. That is part of a planned 2017 review of progress toward meeting nutrient reduction goals set in the Chesapeake Bay Total Maximum Daily Load.

“Keith’s work could have a direct influence at how we look at forests and the amount of nitrogen that comes from them,” Batiuk said.

But he cautioned that what’s good news for air pollution could be problematic for some other sources of nutrients. The Bay models are calibrated to reflect actual nitrogen concentrations observed in rivers throughout the watershed. If air pollution reductions account for a bigger portion of the downward nitrogen trends observed in stream monitoring data than is currently thought, it means other sources may be sending more of the nutrient into waterways than is being estimated.

“That gives us some feedback that may lead us to rethink these other sources as well,” Batiuk said. “Are these efforts to reduce other sources of nitrogen doing what we think they should be doing?”

The extent to which further air pollution reductions will result in additional water quality benefits is unclear. EPA officials have said that while some additional nitrogen reductions are anticipated, the majority of the benefits have already been realized through actions taken by states to meet federal Clean Air Act requirements.

Eshleman, though, thinks further significant reductions are possible, particularly if efforts to reduce greenhouse gas emissions reduce consumption of fossil fuels, which are the major source of NOx emissions.

If that happens, Eshleman said, “the implications for NOx are huge. We could get still some fairly big reductions in nitrate losses in these catchments, I think, in the next 10 or 15 years.”

The study, “Surface Water Quality is Improving Due to Declining Atmospheric N Depositions,” appeared in the Nov. 5, 2013, issue of Environmental Science & Technology.

  • Google+
  • LinkedIn
  • Reddit
  • StumbleUpon
About Karl Blankenship

Karl Blankenship is editor of the Bay Journal and Executive Director of Chesapeake Media Service. He has served as editor of the Bay Journal since its inception in 1991. .(JavaScript must be enabled to view this email address).

Read more articles by Karl Blankenship

Comments

Comments are now closed for this article. Comments are accepted for 60 days after publication.

Ad for rainbarrel depot

Copyright ©2014 Bay Journal / Chesapeake Media Service / Advertise with Us

Terms of use | Privacy Policy