THE BIG SPRING RUN RESTORATION EXPERIMENT: LONG-TERM MONITORING OF SEDIMENT SOURCES AND BUDGETS IN A 1ST ORDER PIEDMONT STREAM

Here we present results of an ongoing multidisciplinary restoration experiment to test hypotheses about the mechanisms controlling landscape processes, channel pattern evolution, and hydrologic connectivity. We recently completed the 8^th yr of pre- & post-restoration monitoring at Big Spring Run (BSR), southeastern PA, a 1^storder Piedmont stream in the Chesapeake Bay watershed. Post-restoration monitoring began in November 2011.

The BSR restoration experiment is designed to understand the mechanisms responsible for development and stability of landscape patterns in shallow, vegetated flow systems, and to test a new paradigm of stream-wetland restoration. This approach is based on the hypothesis that removal of historic sediment, burying a pre-existing Holocene wet meadow anastomosing channel floodplain system, will effectively restore a number of critical zone functions such as nutrient retention. Our research focuses on quantifying pre- and post-restoration sediment transport, sediment and nutrient fluxes, identifying the sources of suspended sediment, surface and ground water interactions, and modeling wetland hydroecological dynamics. These studies benefit from 3 USGS stream gaging stations and 18 USGS piezometers.

We document that the mass of fine sediment (mostly silt and finer) eroded annually (averaged over 5.5 years of data) from stream banks prior to restoration was 164.3 tons, and during the same time interval the average mass of fine sediment deposited (primarily on point bars) prior to restoration was 73.4 tons. The difference between these (i.e., net change from the restoration reach) is 90.9 tons of erosion of fine sediment per year. After restoration, the reduction in suspended sediment load is ~109 tons/yr (based on one yr of data to date); we attribute this reduction to 1) removal of the sediment that formerly was contributed by eroding banks, and 2) deposition and net storage of some of the incoming sediment load within the restoration reach. These results are supported by topographic surveys of 12 monumented cross sections, monitored annually since 2004. Trace element fingerprinting analyses of suspended sediments, stream bank sediments and upland soils within the watershed indicate that >65% of the suspended sediment load prior to restoration was derived from stream bank erosion.