Abstract: Closing Plenary Session  11:00 am (BACK)

Evaluating a New Approach to Aquatic Resource Restoration, Big Spring Run, Pennsylvania

Dorothy Merritts, Ph.D.
Franklin & Marshall College
Lancaster, PA

Robert Walter and Michael Rahnis, Franklin & Marshall College
Daniel Galeone, Allen Gellis, and Michael Langland, US Geological Survey
Jeff Hartranft, PA DEP
William Hilgartner, The Johns Hopkins University
Paul Mayer, Roxanne Adeuya, Bart Falkner and Ken Forshay, US EPA

The original, pre-European settlement wetland soil that was submerged upstream of a low dam and buried rapidly by ~1 m of silt (circa 1710 AD to 1850 AD) provided the physical and ecological targets for restoration at Big Spring Run (BSR) in southeastern Pennsylvania.  BSR is a 2nd-order Piedmont stream (drainage area 15 km2) with low valley slope (~0.004) and relief (~30 m), and is a tributary to the Conestoga River in the Chesapeake Bay (CB) watershed. Because of nearly 8 years of pre-existing scientific research and hydrologic (surface and GW) monitoring data by the USGS, the BSR site was selected by PA Department of Environmental Protection (PADEP) in 2007 to evaluate a new approach to aquatic resources restoration. The Natural Floodplain, Stream, and Riparian Wetland Restoration Best Management Practice (NFSRWR-BMP) proposed by PADEP was included in the recently developed CB Watershed Implementation Plan, part of EPA’s Total Maximum Daily Load (TMDL) assessment for the CB.

The target post-restoration plant community type at BSR is similar to that of the pre-settlement condition, a wet-meadow open canopy palustrine emergent marsh.  The paleoecological record at BSR indicates 3000 years of aquatic ecosystem stability prior to European settlement in 1709. Analysis of fossil seeds and 10 radiocarbon dates acquired from the extensive, organic-rich hydric soil reveals that a sedge meadow wetland dominated by obligate Carex sp persisted from ca. 3000 BP to 300 BP (ca. A.D. 1700). The sedge meadow wetland surface intersected the ground water table at the level of multiple seeps and spring base flow. Notably absent from the paleo-seed record are woody plants typical of palustrine forested and shrub-scrub dominated plant communities that are considered in many riparian plant community restorations.

Geomorphic analysis revealed that small, low-energy channels with minimal bedload sediment transport existed throughout the wet meadow prior to European settlement. Fossil seeds of water plantain, Alisma plantago-aquatica, in silt that buried the wetland indicate that ponding began circa 1730. Burial in wet mud preserved a geologic record of the pre-settlement wetland, but dam breaching sometime in the early 20th c. initiated deep channel incision to the groundwater level, which in turn led to channel bank erosion with lateral channel migration until the time of restoration in 2011.  Channel incision remobilized Pleistocene gravel from beneath the buried wetland soil, and sediment transport of clasts up to ~5-cm in diameter was associated with gravel bar formation within the incised stream corridor.

During restoration in September-October 2011, historic silt (~20,000 tons) was removed by LandStudies, Inc.  The post-restoration stream system was established at the level of the original wetland soil and consists of small channels with low banks that flow overbank frequently (>5 times/year).  Discharge greater than spring base flow is conveyed through the floodplain, with variable channel depth of 0.1-0.2 m and channel boundary shear stresses <1.5 N/m2 to maximize stream stability.  The channel planform increases flow retention and promotes exchange between the stream channel and hyporheic zone across the valley bottom.

Scientific research and monitoring at BSR consist of multiple ecological, hydrological, and geomorphical strategies to (1) identify pre- and post-restoration plant communities and the densities of different plant types, (2) determine sediment sources and quantify pre- and post-restoration sediment transport (suspended load) and fluxes, (3) characterize bedload transport, (4) quantify nutrient fluxes, and (5) model wetland hydroecological dynamics. Three USGS stream gaging stations (with turbidity sensors and suspended sediment sampling) are located at the upstream end of each tributary entering the restoration reach and on the main stem downstream of the restoration reach.  Groundwater is monitored and sampled with18 USGS piezometers. Three years of pre-restoration and 2 years of post-restoration data have been collected to date.