Current Projects at Roaring Creek
Physical and chemical conditions in streams are influenced by a variety of processes operating in the terrestrial environment and along hydrologic flow paths in groundwater and surface water systems. Bucknell University Environmental Center established the Roaring Creek Field Station for Watershed Studies in the Roaring Creek Tract of Weiser State Forest in January of 2010. South Branch Roaring Creek flows through beautiful mature mixed deciduous forest in the Roaring Creek Tract, but the stream in this section has been dammed in 3 locations to provide water for citizens in Mount Carmel and Shamokin. Since the establishment of this research and teaching facility, several students from my lab have conducted research projects at Roaring Creek to understand the changes that occur in water quality as Roaring Creek flows from high elevation beaver ponds and meadows through forested reaches and several water-supply reservoirs. We have also used this site for a teaching lab in "Watershed Systems Science" and "Bucknell on the Susquehanna" courses.

Patterns of dissolved organic carbon in Roaring Creek (with Joanna Freeman, BS Environmental Studies '12)
Joanna has been sampling several sites along Roaring Creek to determine how DOC quantity and quality vary along the stream due to interactions with groundwater and flow through reservoirs. Joanna's research demonstrated that both DOC quantity and quality decline from upstream to downstream but that reservoirs serve as sources for high quality DOC to the stream along the way. Seasonal patterns in DOC quantity and quality were related to low summer stream flow, rain in late summer, and leaching from fresh leaves in fall.

Seasonal changes in physical and chemical conditions and plankton populations in McWilliams Reservoir (Molly Clark, BS Environmental Studies '12)
Molly studied the largest reservoir on Roaring Creek from October 2010 to October 2011 to see how the lake changes seasonally. Molly found that the reservoir showed typical seasonal patterns of a dimictic mesotrophic lake. The lake was thermally stratified from May until October in 2011 and was covered with ice from December 2010 until March 2011. Spring turnover occurred from March to May, and fall turnover occurred from October to December. Light penetrated into the hypolimnion of the lake until late summer, when the epilimnion deepened and algae became more abundant. As a result, oxygen concentrations remained fairly high to the bottom of the lake until August and September 2011 when hypolimnetic oxygen depletion was obvious. Coupled with this anoxic hypolimnion was an increase in ammonium and phosphorus in the deep water of McWilliams Reservoir, which affected biogeochemistry of Roaring Creek (as revealed in Joanna's study). Algal biomass and zooplankton densities were fairly low throughout the year in the reservoir. Zooplankton densities tracked algal biomass in the water column, and both appeared to cluster near the thermocline in the lake during warm periods of summer.

Assessment of artificial substrate samplers (Haley Coffin, BS Biology '12)
Haley collected benthic macroinvertebrates from several sites along a short reach of Roaring Creek to see how variable community composition is among microhabitat types. She also compared benthic macroinvertebrate samples from rock baskets with kick nets to determine if microhabitat type affected the ability of rock baskets to provide representative benthic macroinvertebrate samples. Haley found that benthic communities in rock baskets were similar to kick net samples from deep sites with slower water but that rock basket samples from shallow riffle sites were dramatically different than kick net samples from the same sites. Estimated densities of benthic macroinvertebrates were higher in rock baskets than in kick net samples. Rock baskets in riffles were absolutely full of black fly larvae (97.5% of organisms), which made up a significant but not overwhelming proportion (41.5%) of benthic macroinvertebrates in kick net samples. Mayflies, stoneflies, and caddisflies comprised over 28% of kick net samples but just over 3% of rock basket samples. Rock baskets therefore seemed to collect a biased sample of certain benthic macroinvertebrates, as they were strongly preferred by black flies and other filter feeders and avoided by larger benthic organisms (e.g., hellgrammites).