Bay Journal

Legacy sediments may pose threat to Bay cleanup

Researchers find that myriad old mills piled up problems for streams

  • By Karl Blankenship on March 01, 2007
  • Comments are closed for this article.
Flood plain reconnection projects, like this one on Lititz Run in Lancaster County excavate accumulated sediment to allow streams to easily overflow their banks. (LandStudies) In this flood plain reconnection project, Much of the nutrients and sediment can be stored, protecting downstream areas. (LandStudies) At the flood plain reconnection projects on Lititz Run in Lancaster County, native vegetation grew back on its own when the original flood plain was exposed. (LandStudies) The Denlinger’s Mill Dam in Lancaster County, which was built in the early 1700s, was 20 feet high and backed up water for 2.3 miles. Since breaching in 1901, it has lost 46 percent of its stored sediment, but its steep banks still hold about 171,659 tons.  (Dorothy Merritts and Bob Walter)

When early settlers needed power, they often looked no further than the nearest stream. By erecting a small dam, they harnessed power to turn giant waterwheels that would grind grain into flour, saw logs into boards, produce gunpowder, and perform a host of other tasks essential for commerce.

The demand for water power was nearly insatiable. U.S. Census figures show that more than 65,000 water-powered mills were operating in the eastern United States by 1840. Streams in Pennsylvania’s Lancaster County averaged a dam every two miles.

“Within about 50 years of settlement, the streams were literally converted into a series of linked slackwater ponds,” said Bob Walter, an associate professor in the Department of Earth and Environment at Franklin and Marshall College. “And now, our generation has forgotten there were dams and millponds there.”

Some mills remain—a few have become historic attractions—but most slowly disappeared, along with many of their dams. While the structures are gone, Walter, and Dorothy Merritts, a professor in the Department of Earth and Environment at Franklin and Marshall, say the legacy of those mill dams continue to haunt the region’s waterways—and may pose a threat to the Chesapeake.

The dams trapped sediment washing off farms and clear-cut forest lands. When breached—often because of lack of maintenance, severe storms or both—they left behind piles of sediment that were 20 feet high in places. The stream, no longer restrained by a dam, would cut through the accumulated sediment like a knife through butter, creating deeply incised banks. “I grew up thinking this was the way streams are supposed to look,” Walter said.

Instead, the two researchers say, streams look that way because the region had so many sediment-filled dams. In Lancaster County alone, historical records indicate nearly 400 mill dams were constructed.

In a recent report for the Pennsylvania Department of Environmental Protection, Merritts and Walter estimate that the 644 miles of streams in the Conestoga River watershed—the county’s main drainage—continue to store about 84 million cubic yards of “legacy sediment.” That’s roughly the equivalent of stacking sediment 9 miles high on a football field. Some is still behind existing dams, but where dams have breeched, much of the sediment is stored in high stream banks, which may be eroding into the river.

Merritts and Walter estimate that more than half of the 180,000 tons of sediment flushed out of the Conestoga and into the Susquehanna River each year originates from legacy sediment—not “new” sources being washed off farms or development sites. They further estimate that legacy sediment each year carries with it about 135,000 pounds of phosphorus. That’s a bit more than 2 percent of the phosphorus reduction goal for the entire 64,000-square-mile Chesapeake Bay watershed.

Merritts and Walter concluded in their report that “stream bank erosion is an important source of sediment and nutrients to tributaries of the Chesapeake Bay, and is at least as significant as runoff from upland sources in some watersheds.”

That could have a major implication for the Bay cleanup effort.

Past estimates of the amount of nutrients reaching streams—and ultimately the Chesapeake—have assumed that sediment and nutrients were coming from sources on the land, such as farms and development sites. The nutrient reduction “tributary strategies” prepared by states to meet Bay Program nutrient and sediment reduction goals called for a variety of actions to curb runoff from the land, such as reducing fertilizer use, changing farm tilling practices, planting streamside buffers and other actions.

But those actions alone may not meet the goals if a large amount of the phosphorus and sediment reaching the Chesapeake is already in stream channels.

The role that “legacy sediments” play in affecting Bay water quality will be getting a close look in the coming months as state and federal officials consider ways to better parse the relative amount of pollution coming from land uses versus what may be coming from material already in stream channels and flood plains.

In addition, they will be examining the effectiveness of new cleanup strategies aimed at dealing with sediment and phosphorus already in the stream. Such practices could include identifying erosion “hot spots,” then removing built-up sediments to restore the original wide flood plains.

Though their most extensive work was conducted in the Conestoga watershed, Merritts and Walter also examined streams at other sites in Pennsylvania and Maryland, and concluded that the potential impact from old dams may be widespread in the region. “The Chesapeake Bay watershed is like the bull’s-eye of milling activity,” Walter said. And, beyond dams built for mills, scores more were built to supply water for canals, drinking water, farming and other purposes, he noted.

In a nutshell, this is the picture that Merritts and Walker have painted: Prior to European settlement, valley floors were mostly forested wetlands, covered with a rich organic layer built up over thousands of years from falling leaves, decaying trees and other material. That layer, roughly a foot and a half thick and usually grayish in color, rested upon a layer of gravels left from earlier ice ages. Small streams wound through the forested organic layer, frequently overflowing onto the wide flood plain of the valley floor.

When Europeans arrived, they cleared the land, sending large amounts of sediment down hilly slopes, smothering the flood plain. The settlers also began building dams, trapping much of the sediment as a darker layer above the gray organic material. Over time, those dams breached. Stream water that once flowed over the top began cutting narrow channels down through the accumulated sediment. Initially, the amount of sediment eroded was relatively small—the fine, rich soils washing off fields were very cohesive, so banks remained intact as the stream dug deeper.

But when the stream cuts through the old organic wetland layer, it hits the ancient layer of coarse gravel. Unlike the sediment above, the gravel quickly erodes, undercutting the steep stream slopes above and dramatically increasing the amount of sediment going into the stream.

The idea that historic sediments cover the landscape is not new. Geologists have recognized for decades that immense amounts of sediment smothered the original valley bottoms and streams in the decades immediately after settlement.

Studies show that old flood plains throughout the Piedmont are typically covered with anywhere from 1.5 to 3 feet of sediment, whether dams were present or not. The initial slug of sediment was so immense that some colonial era ports along the Bay, such as Joppatown and Port Tobacco, were closed as shipping channels were filled with dirt from farms.

But the work by Merritts and Walter suggests that, because of the abundance of small dams, typically 8 to 10 feet high, more of the sediment dislodged by early settlers may still be stored in stream channels than previously realized.

“Dorothy and Bob have added to the story in terms of documenting that small dams were perhaps more prevalent than previously thought,” said Peter Wilcock, a professor of geography and environmental engineering at Johns Hopkins University who specializes in river sedimentation processes. “They have changed the flavor of it. But it hasn’t been a sea change.”

Wilcock cautioned against making direct links between erosion in streams and the Chesapeake Bay’s sediment and phosphorus pollution. Just because a stream bank has high banks, he said, doesn’t mean it has a high erosion rate. Factors other than bank height affect erosion rates, so the water may simply be passing through.

Even if erosion is taking place, the sediment is not necessarily making it to the Chesapeake. Many rivers and streams have a huge capacity to store sediment, especially in wide flood plains of downstream areas.

Studies have shown that often, the erosion rates of upstream areas are not matched by sediment discharge at the mouth. “All that sediment is in the river valley, but it can stay there for a million years,” Wilcock said. “It can turn to rock. If you have a sediment source in one place, you still don’t know what proportion is going to get to the Bay.”

To better understand that, Allen Gellis, a scientist with the U.S. Geological Survey, has used sediment “fingerprinting” to determine the origin of sediments in streams. By looking at the presence of certain types of radioactive particles attached to the soil—byproducts of Cold War era nuclear tests —he can determine when sediments were on fields receiving trace amounts of nuclear fallout.

Old sediments, stored in the stream banks centuries ago, have none at all, while newer sediment can be “aged” based on the types of radioactive particles released during tests at certain times.

His results show that in the Conestoga watershed, most of the sediment in the streams is old—predating the nuclear weapon tests. Sediment from more recent agricultural activities is the second largest source.

But his studies, limited to a handful of small watersheds, also indicate that the amount of sediment actually transported out of a watershed can vary dramatically. In slowly moving tidal coastal plain rivers, for instance, large amount of sediment are eroded from stream banks, but are quickly redeposited on flood plains—little seems to make it to the Bay.

“Legacy sediment may be important, but there is only one way to know for sure, and that is to do this fingerprinting to look at all the sediment sources,” Gellis said. “If it turns out the banks are important, then we can determine where in the watershed the banks are eroding.”

Pinpointing where sediment and phosphorus that reaches the Bay originates is critical because almost all of the more than 100,000 miles of stream in the watershed have been impacted by human activities over the last 400 years. Stream restoration is one of the most expensive ways to control nutrients and sediment—often costing hundreds of thousands of dollars per stream mile, or more.

“The places where we would like to put our restoration money would be those places that have tall banks that are actively eroding—and are very closely plugged into the Bay,” Wilcock said.

Pennsylvania started that process by using aerial photography to identify areas with high banks where the erosion potential may be greatest and targeting them for management actions.

But some caution that concern about legacy sediments should not overshadow traditional nutrient and sediment control actions aimed at keeping sediment and nutrients on the land and out of streams.

“We can’t keep creating more legacy sediments, that would be dumb,” said Tom Simpson, a soil scientist with the University of Maryland and chair of the Bay Program’s Nutrient Subcommittee. “It’s never going to get better unless you stop the source.”

Harry Campbell, a scientist with the Chesapeake Bay Foundation and a member of a Legacy Sediment Workgroup created by the DEP, said that most public funding should promote lower cost programs such as those that reduce nutrients from farms or control stormwater runoff, as opposed to costly stream restoration projects.

“It is a best management practice that can be advantageous under certain conditions,” he said. “But there are still questions technically as to whether or not we are receiving the best bang for our buck in terms of meeting the Bay goal. I think we need an integrated approach that includes legacy sediments but is not exclusive to legacy sediments.”

Also, controlling legacy sediment probably has a proportionately smaller impact on nitrogen—the more problematic nutrient for the Chesapeake—than phosphorus. Unlike phosphorus, much of which binds to particles, many forms of nitrogen are water soluble and readily move through sediment: Much of the nitrogen in the legacy sediment is likely of more recent origin and just passing through rather than nitrogen buried centuries ago.

Legacy sediments do contain some forms of organic nitrogen from old material such as leaves, algae and other living material that was buried with the sediment, but many scientists say those forms may not be as biologically available to algae and other organisms as the inorganic nitrogen used in fertilizers, and therefore are not as significant a contributor to the Bay’s water quality problems.

Nonetheless, recognizing the impact of the deep piles of sediment left by dams could have other far-reaching implications.

For instance, areas with deep layers of legacy sediment may be inappropriate places for streamside forest plantings because the roots of the trees may never reach shallow groundwater where they are supposed to absorb nitrogen headed toward streams. Merritts said trees planted in deep legacy sediment by her watershed group died because they couldn’t get water.

Understanding where deep legacy sediments are located is also important because there may be locations unsuitable for infrastructure such as roads, bridges and sewer lines, which may be susceptible to future erosion.

“They were great places to dig and to run sewer lines, because it was easy to excavate, and it followed the natural gradient of the stream,” Walter said. “The problem is, now they are blowing out. We see many, many examples of sewer lines being exposed and hung out to dry.”

It may also have ramifications for dams removed to improve fish migration—a major Bay region objective—especially in Pennsylvania, which leads the nation in dam removals. Past studies have shown that initially, dam removals have little impact on sediment movement. The work by Merritts and Walter suggests that could change over time, if the stream cuts through accumulated sediment and reaches long-buried layers of gravel. That may mean future dam removals could require increased efforts to either stabilize banks or remove stored sediment.

An even more sobering revelation, though, may that early settlers had a far greater impact on the landscape than generally realized. Huge areas of riparian wetland flood plains were rapidly buried—what people often assume are “natural” flood plains today are actually terraces created by sediments that have accumulated behind thousands of dams.

“There are few streams that have not been impacted in some way,” said Jeff Hartranft, of DEP’s Bureau of Waterways Engineering and chair of the Legacy Sediment Work Group. “But understanding the legacy sediment problem is the first step in proposing a fix.”

  • Category:
  • 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