Steep headwater streams may account for much of a watershed's total stream length in hilly country. Placid most of the time, these streams can generate surprisingly powerful flows during heavy rain or rapid snowmelt, even in hydrologically healthy areas. When hilly regions urbanize, these streams can become raging torrents, capable of wreaking havoc through local flooding and erosion. This problem is growing in many areas as development moves into hillier terrain.
For steep creeks, “step-pool” architecture creates a tumbling flow that reinforces channel stability and reduces the downstream flood peak by dissipating flow energy and reducing stream velocities. When the extra runoff from impervious urban surfaces is introduced into these channels they adjust through erosion and can lose their stable channel structure. What were once bucolic little forest streams can soon become gully-like channels that speed floodwaters to downstream areas.
Direct intervention is required to stabilize these streams when they threaten property or infrastructure. Although stormwater management practices that emphasize onsite infiltration are a worthwhile approach to reducing storm flows, this cannot stabilize already unraveled channels, and these techniques are rarely useful on thinner hillside soils anyway.
A conventional approach to stabilizing steep streams is to line them uniformly with very large rocks. This can be effective at halting erosion, but it requires a lot of rock. Since all but the highest flows are below the surface when a small stream is paved, this is hardly conducive to a natural appearance and practically eliminates the stream's habitat value.
Such conventional engineering approaches to channel stabilization are fast becoming unacceptable where public works agencies are increasingly trying to cope with stormwater problems and improve the habitat and amenity value of their stream channels. State highway departments are also increasingly required to enhance the streams they intersect. Finally, state and federal regulatory agencies are now requiring a natural stream channel design (NSCD) approach, based on fluvial geomorphological principles, during any modifications to natural watercourses. (Fluvial geomorphology is the science of river land-forms and processes.)
For these reasons, attempts to restore a more natural step-pool morphology to degraded steep creeks are increasingly common. Even man-made drainage channels can be designed in this way. Compared to pipes, concrete chutes, or ordinary rock-lined channels, a step-pool stormwater channel can attenuate storm runoff while simultaneously providing an attractive, naturalistic landscape feature.
Research in fluvial geomorphology has shown that, unlike lower gradient streams, the stability of step-pool streams is more dependent on the arrangement of the biggest particles in the streambed than on the precise geometry of the channel. Following the NSCD principle that stable natural analogs should serve as templates for restoration, such observations can be applied to steep channel stabilization in any land-use context.
One commonly observed problem in many attempts to construct natural-appearing step-pool watercourses is that the reconstructed channel often ends up looking more engineered than natural. While this is partly because successful environmental construction always requires a strong dose of “art,” it is also a function of design. Many projects specify a relatively narrow range of rock sizes for channel-spanning structures, even though natural boulder steps generally include a wide range of rock sizes.
It turns out that this diversity of rock sizes is central to the stability of these structures. Empirical studies in mountainous terrain indicate that the stable steps in steep creeks do not so much conform to some regular spacing, as some investigators have contended, than to a more random pattern, with steps tending to form where large boulders have become lodged along the watercourse. These largest boulders are called “keystones” because smaller boulders and cobbles tend to naturally accumulate and pack tightly together in the stable channel “pinchpoints” that they create.
Tried-and-true channel stabilizing structures such as boulder weirs or cross vanes (commonly constructed using uniformly sized large rock) can be modified as a keystone-based step structure. Depending on the application, such channel-spanning structures also can be configured as even more irregular rock structures with labels like “M” weirs or “O” weirs.
Natural form is reinforced still further by mixing these step types along the channel profile, just as nature would do. As part of this natural stream channel design approach, a spreadsheet routine has been developed that provides a basis for the spacing and sizing of pools and steps, depending mainly on stream gradient. This again stems from the geometric relationships between step height and step length observed in natural step-pool streams. Combining both appropriate spacing with the use of keystones in the individual structures at least partly ensures that the reconstructed channel will be both natural appearing and stable. We have found, however, that since humans cannot pack sediments together as well as flowing water can, careful construction will always be equally vital to the long-term stability of these high-energy stream channels.
— Todd Moses, senior restoration specialist, and Mike Lower, E.I.T., environmental engineer, are with Skelly and Loy Inc., Harrisburg, Pa.
Stream reconstruction in the Pocono Mountains
Skelly and Loy recently completed a step-pool stream reconstruction project in what had been a severely eroding ravine slicing through a newly urbanized area in the Pocono Mountains of eastern Pennsylvania. The channel had been rerouted years ago as part of a coal-mining operation, long since abandoned. Rapid erosion was sending massive amounts of sediment downstream and threatening a municipal water supply line.
The reconstructed reach of this stream is an extremely steep, confined channel with a gradient of about 11%. Classified as an A2a+ type channel according to the Rosgen classification system, this stream is almost steep enough to be termed a “cascade.” Because of this extremely steep gradient, the boulder steps in the reconstructed channel were generally spaced about one to two channel widths apart. (The step, or pool, spacing in most step-pool streams usually varies from between about one to four channel widths.) Although disguised by the surrounding large boulders within the reconstructed streambed, keystone boulders as large as 8 feet in diameter were used to construct the rock steps in this stream reach.
The first phase was completed late in 2004, and the final phase will be completed later in 2005. This work was expensive, in part because of the amount of earth that had to be removed to stabilize the side slopes of the 30-foot-deep ravine and because of the very large size of the rock (and large number of steps) required to ensure stability in such a very steep creek. The cost for the design, permitting, and construction of this first phase of work was on the order of $400 per foot. More typical step-pool channel reconstructions would likely cost only about half this amount. Installation costs can also be expected to decline as contractors gain experience with this type of construction.