Text

theses

Water, sediment and fish were sampled at various times and locations between 2008 and 2011 on the Tacony-Frankford Creek, a tributary to the Delaware River in Philadelphia, PA. The study sought to identify spatial and temporal trends, identify the sources and fate of polycyclic aromatic hydrocarbon (PAH) and polychlorinated biphenyl (PCB) compounds, and explore an alternative sampling method for PAH and PCB compounds in the stream. Grab samples of sediments and whole water were collected between November 2008 and April 2010, and fish samples were collected in June 2010 and analyzed for 28 PAH compounds. Total sediment concentrations of ΣPAH ranged from 3,200 ng/g to 72,000 ng/g at various locations in the tidal and non-tidal regions of the tributary and 25,000 ng/L in the Delaware River. The ΣPAH concentration in water ranged between 170 ng/L and 580 ng/L in the tributary and between 130 ng/L and 150 ng/L in the Delaware River. The concentration of ΣPAH varied greatly depending on the type of fish and tissue, and ranged from 43 ng/g in catfish livers to 1,300 ng/g in white perch organs (minus livers). The PAH concentration in water and sediment samples exhibited spatial trends but no identifiable temporal trend. The primary source of PAHs to the stream for the time period was motor vehicle emissions. In June 2011, grab samples of whole water were collected at different times in the tidal cycle at a single location and filtered to separate suspended particles from the aqueous samples. Both fractions were analyzed for PAH and PCB concentrations to determine whether sampling depth and time impact the concentrations in a tidal stream. The sum of 16 PAH compounds ranged from 74 to 160 ng/L and the sum of 41 PCB compounds ranged from 2.7 to 7.9 ng/L in the whole water samples. The results showed that sampling depth impacts both classes of compounds, with concentrations higher near the sediment bed than near the surface and affected mostly by the portion that was associated with particles. Timing of sampling during the tide had a much larger impact on the concentration of PAHs than PCBs. Passive sampling was explored as a technique to mitigate the temporal variability that was evident in the study stream. Polyoxymethylene (POM) was investigated as a material to collect passive samples in small aquatic systems. In the laboratory, 76 μm thick POM sheets exhibited fast sorption, coming to equilibrium with aqueous-phase phenanthrene within 2 weeks at 23o C and 5 weeks at 4°C. Partitioning coefficients log(KPOM) were calculated for 16 PAH and 41 PCB compounds and ranged from 3.1 to 6.4, and 4.1 to 6.1, respectively. The log(KPOM) values were used with collected POM concentrations to predict PAH and PCB concentrations in the aqueous phase and were found to underpredict the aqueous phase for most compounds. Correlations between POM films and sediments or biota were not very strong, except for samples of white sucker fish and catfish livers.

ETD_7707

Ph.D.

Includes bibliographical references

by Kelly L. Francisco

doi:10.7282/T3T43WFN

ETD doctoral

The author owns the copyright to this work.