Keepataw Preserve
Will County, Illinois
The Keepataw restoration and a series of related projects grew out of a biological assessment that I wrote for the Illinois State Toll Highway Authority as part of the Interstate 355 South Extension project. Initially there were a few federally endangered species under consideration: The Hine's emerald dragonfly, the Indiana bat, and the leafy prairie clover. We were able to establish that there would be no project-related impacts to two of these. However, there was potential dragonfly breeding habitat within the footprint of a proposed bridge over the Des Plaines River floodplain. The client chose to mitigate the potential impact rather than lose the time required to do thorough species surveys.
The biological opinion issued by Fish and Wildlife Service included a commitment by ISTHA to create or restore 200 linear meters of dragonfly breeding streamlet, as well as a stated acreage of upland habitat management. What was perhaps unusual was that FWS encouraged an experimental approach to restoration, with a provision to monitor results and use the findings to improve future restoration efforts.
The first step was formation of a Hine's emerald dragonfly working group (HEDWG). I was asked to chair the group, based in part on my long experience (since 1994) of annual monitoring of the species. Numerous stakeholders would be part of the group: regulatory agencies, three county Forest Preserve Districts (which owned most of the potential restoration sites), and representatives from academia and the private sector. The group met monthly, in conjunction with a progress report by construction managers for the highway and bridge projects.
Initially the group brainstormed potential restoration sites. Then pros and cons were identified for each site; several proved to be impractical for various reasons and were dropped from further consideration.
A few of the potential restoration sites were within the Forest Prerve District of Will County's Keepataw Preserve, just west of the bridge alignment and on the northwest bank of the Des Plaines River. Here a series of seeps and springs issued from 75-foot high bluffs, feeding extensive marshes on the floodplain below. Much of the site had been mined for flagstone for several decades beginning in the 1880s, and the remains of two lime furnaces still stood within the preserve. The shallow quarry pits formed ponds on the floodplain, and discarded dolomite rubble extended like a man-made moraine parallel to the bluff.
The western part of the bluff and the narrow ribbon of sedge meadow and wetland along it's base had already been restored by clearing of invasive shrubs. But this represented lass than half of the available bluff face, so one day I attempted to walk the other half of the bluff face... something I hadn't done since 1989, when I'd conducted an amphibian and reptile inventory of the site for the Forest Preserve District. Walking that half-mile or so took me several hours, most of it spent fighting through tangled thickets of bush honeysuckle, buckthorn, and cottonwood.
Along much of the way, restoration opportunities appeared to be limited. If there were seeps present, they were either small or submerged beneath standing water. Those stagnant pools largely precluded Hine's emerald dragonfly breeding, which occurs in small, slowly flowing streamlets. The best opportunities seemed to be at the extreme ends of the overgrown segment of bluff base, where there were clearly defined seeps and a few rivulets.
I returned with a few other members of HEDWG, and we evaluated each area. The western end was relatively simple, there was only one potential streamlet and all that was required was a simple excavation of the edge of a spoil pile to double its length. The eastern end was more troublesome. Here we found a series of seeps and springs (nine were ultimately identified within a small area), and a few very small sedge meadow/fen openings persisted close to the bluff. None of these measured more than about 15 by 30 feet. The springs and seeps fed a well incised streamlet, a little larger and deeper and colder than the ones that typically supported dragonfly larvae. The rest of the surrounding area was a honeysuckle thicket with scattered large cottonwoods projecting above the shrub canopy. We noticed clear evidence of hydrology among the shrub cover, driftlines of debris and eroded channels in the nearly bare soil. But there seemed no clear pattern. The driftlines seemed to point in multiple directions. This place didn't play by the rules, and it was difficult to understand what was happening because the dense vegetation limited sightlines.
Several of us were intrigued by the site. So a few months later, as winter set in and leaf cover had fallen from the trees and shrubs, I returned with a pair of loppers and spent the better part of a day on that roughly two acre parcel. The site was defined on the east by a stream coming under Bluff Road in a large concrete culvert, on the south by the densely overgrown dolomite spoil pile, on the west by the bluff (which turned sharply to the west at the qouthwest corner of the site), and on the north by Bluff Road and a 50-foot high rip-rap embankment below the road.
I spent the first few hours patiently cutting narrow transects across the site. Gradually, with the ability to see came understanding.
What was happening was really fairly simple. During more severe storm events, runoff from the upper watershed of the small stream poured through the culvert under Bluff Road. This scoured the stream down to dolomite bedrock, just a few feet down. A bit below the culvert, beginning perhaps after a few hundred feet, the stream diverged into several braided channels defined by deposited gravel bars. Eventually the streams fed into a reed canary grass marsh, and then one of the quarry ponds, both of these well south of the area that I was interested in.
During the highest flow events, the stream was unable to hold all of it's water, which flowed overbank. Here the suspended fine silts which other wise were carried well downstream were able to drop out. Over time, one to two feet of fine sediment had been deposited over what presumably had once been a more extensive sedge meadow/fen complex. Only those few small remnants close to the bluff had been left intact and unburied.
Equally damaging, as the floodwaters found their way out the low southwest corner of the site, where they were able to skirt around the turn in the bluff in the narrow gap between it and the spoil pile, a channel had been eroded. This channel had then captured all of the flow from the nine seeps and springs, and it had headcut all the way up to the uppermost spring, close to the road.
Initially this location was only one of several with promise. But most of the others dropped out one by one. Some were technically infeasible, some were too expensive, some were too small to offer much benefit, some depended on off-site water sources. Eventually the group reached consensus on pursuing three locations. Two of them were at Keepataw, with the third at nearby Waterfall Glen in DuPage County. Dan Soluk of the University of South Dakota took the lead on the Waterfall Glen site, which would include an innovative experimental design allowing manipulation of flow rate and temperature to test larval growth rate and survival. The Keepataw sites were mine to design.
That same winter we had the eastern Keepataw site cleared of woody vegetation. A crew consisting of one guy with a chainsaw and three laborers removed all shrubs and about half of the trees (the remaining cottonwoods were removed later). The small remnant high quality areas near the bluff were delineated with flagging and boundary tape to prevent foot traffic into the soft wet soils near the springs and seeps. Woody material was cut and burned on site; at first this was done by hand, and once openings were made a bobcat was brought in to speed the process of moving the branches to central burn pits.
The clearing effort took about a week. Toward the end of that time a major storm moved in, with sustained heavy rainfall. It turned out to be a 100-year event. I spent most of the day on site, taking full advantage of the opportunity to observe hydrology in action. By afternoon much of the site was underwater, and our theories proven to be correct. The next day, a fresh layer of silt covered much of the site, with a few localized areas of scour.
With the site cleared and detailed topography available, we moved forward with the design. The concept would be to first, separate the stormwater from the restoration site to prevent future silt (and weed seed) deposition, by building a low berm along the stream. Then, the deposited silt would be excavated and removed and the area replanted with native vegetation. As part of this process the streamlet would be moved away from the bluff and into the interior of the restoration site, and the headcut channel filled in. This more circuitous course would allow time for the spring water to warm a little, into ranges suitable for dragonfly larvae. Streamlet design and bordering vegetation were based on a reference site less than a mile away which was known to support successful Hine's emerald dragonfly breeding.
The restoration site was graded and planted in the late winter and early spring of 2008. As of the summer of 2009 it was performing as expected. The streamlet meandered through the now lower central portion of the site, and in lower reaches it diffused into multiple shallow branches much as at the reference site. Native vegetation was already dense and well established. Crayfish burrows (necessary for overwintering of dragonfly larva) had been present only in the extreme southwestern corner of the site pre-project, but now the crayfish had moved on their own nearly a third of the way across the site. Probably the presence of slightly warmer water and greater hydrologic stability encouraged this.
After completion of the wetland component of the restoration extensive work was done to restore the oak woodlands atop the bluff. Buckthorn and other shrubs were removed from the understory of an extensive area resulting in a much more open woodland. As the herbaceous layer recovers, precipitation runoff will be slowed and more water will infiltrate to feed the seeps and springs below, resulting in a more reliable water source.
The restoration site will require management. Similar sites in this region, in the absence of fire or other management, eventually become nearly monotypic cattail marshes. The responsibility for this falls on the Forest Preserve District once ISTHA has met it's five year mitigation monitoring requirements.
The Keepataw restoration project straddled my time with two different firms. The conceptual work took place during my time with Earth Tech, and after I moved to Winzler and Kelly in March 2007 I was able to continue through the detailed final design. The project was completed efficiently in part because the HEDWG meetings kept all of the players involved in the process and built consensus, and also because of the frequent and close interaction between environmental staff and construction managers. ISTHA's commitment to success was also essential, as was the Forest Preserve District of Will County's willingness to assume management responsibility post-project.
XXXX
XXXX
Copyright and all other rights reserved © 2010
Kenneth S. Mierzwa ken.mierzwa@gmail.com
April 10, 2010
April 10, 2010