Lipid-free polyethylene membrane tubing (LFT) has been further developed in response to a growing need for an inexpensive and simple time-integrative sampling device for dissolved hydrophobic contaminants in water. The LFT sampler is based on the diffusion of dissolved hydrophobic target compounds through the aqueous boundary layer and into the polyethylene membrane, mimicking uptake by organisms. We demonstrate through laboratory and field validation studies that LFT provided the same benefits as many other passive sampling devices, withoutthe potential of analytical interference from lipid impurities. A total of 370 LFTs and semipermeable membrane devices were deployed for 21 days in paired studies at highly urbanized, undeveloped, and two Superfund sites, representing several river conditions. A simple internal surrogate spiking method served as an in situ calibration indicator of the effects of environmental conditions on the uptake rates. A modified extraction method for the LFT increased recoveries while decreasing solvent use and labor compared to other organic extraction procedures. LFT sampling rates were estimated using ratios, in situ calibration and modeling for over 45 target analytes, including PAHs, PCBs, and pesticides.

Twenty-five PCBs (polychlorinated biphenyls), 15 organochlorine (OC) pesticides, and mercury were determined in fish from the Willamette River in Oregon, including a Portland Harbor superfund site. Fish were collected during the summer of 2000 along a 20-mile stretch of the lower Willamette River. Concentrations of sumPCBs (sum of 25 individually determined PCB congeners) and sumDDTs (sum of p,p'-DDT, p,p'-DDE, and p,p'-DDD) in fish ranged from 14 to 530 and from 18 to 510 ng/g-wet weight, respectively. SumPCBs concentrations at all sites exceeded US EPA fish advisory's screening values. Hexachlorobiphenyl congener 153 was the most abundant of the PCBs detected and p,p'-DDE was the most abundant OC pesticide detected. Low levels of dieldrin were detected in fish at all sites with the highest concentration at the superfund site (4.6 ng/g-wet weight), while other OC pesticides tested were near or below detection limits (approximately 2 ng/g). In general, organic chemical contaminant concentrations were highest in fish from the superfund site and were lower further upriver. Smallmouth bass had the highest levels of OC compounds of three fish species examined. They also had the largest site-to-site variations whereas black crappie had little variation throughout the study area. Mercury levels in fish ranged from 13 to 520 ng/g. Historical fish residue data are limited from the Portland Harbor superfund site, what data is available is over a decade old, generally consisted of only a few fish (< or = 3) and analyses quantified only a few PCB congeners (< 3).

The Willamette River, one of 14 American Heritage Rivers, flows through the most densely populated and agriculturally productive region of Oregon. Previous biological monitoring of the Willamette River detected elevated frequencies of skeletal deformities in fish from certain areas of the lower (Newberg pool [NP], rivermile [RM] 26 - 55) and middle (Wheatland Ferry [WF], RM 72 - 74) river, relative to those in the upper river (Corvallis [CV], RM 125-138). The objective of this study was to determine the likely cause of these skeletal deformities. In 2002 and 2003, deformity loads in Willamette River fishes were 2-3 times greater at the NP and WF locations than at the CV location. There were some differences in water quality parameters between the NP and CV sites, but they did not readily explain the difference in deformity loads. Concentrations of bioavailable metals were below detection limits (0.6 - 1 microg/ L). Concentrations of bioavailable polychlorinated biphenyls (PCBs) and chlorinated pesticides were generally below 0.25 ng/L. Concentrations of bioavailable polycyclic aromatic hydrocarbons were generally less than 5 ng/L. Concentrations of most persistent organic pollutants were below detection limits in ovary/oocyte tissue samples and sediments, and those that were detected were not significantly different among sites. Bioassay of Willamette River water extracts provided no evidence that unidentified compounds or the complex mixture of compounds present in the extracts could induce skeletal deformities in cyprinid fish. However, metacercariae of a digenean trematode were directly associated with a large percentage of deformities detected in two Willamette River fishes, and similar deformities were reproduced in laboratoryfathead minnows exposed to cercariae extracted from Willamette River snails. Thus, the weight of evidence suggests that parasitic infection, not chemical contaminants, was the primary cause of skeletal deformities observed in Willamette River fish.

Because PCBs and organochlorine pesticides continue to be of global concern, studies that address information gaps, such as factors and influences of spatial and temporal effects on contaminant bioavailability, are valuable. The present study focused on the spatial and temporal distribution of bioavailable organochlorine pesticides and PCBs in surface waters of a contaminated harbor. Passive sampling devices were intensively deployed adjacent to various land uses on the Willamette River, OR, including Portland Harbor and McCormick and Baxter Superfund sites, during summer and fall, extreme conditions, 2001-2004. An increase of bioavailable sigmaDDTs (sum of p,p'-DDT, p,p'-DDD, and p,p'-DDE) concentrations was strongly affected bythe local historic production of DDTs and temporal changes in river conditions. The increase of bioavailable p,p'-DDD and high DDD/DDE ratios observed during summer indicates conditions favoring anaerobic reductive processes. In contrast to sigmaDDTs, the bioavailable concentrations and daily loads of dieldrin and PCBs increased during fall, especially during episodic rainstorms. On the basis of the PCB congener profiles, PCB inputs from urban runoff /sewer overflows were considered likely current sources of bioavailable PCB into the Harbor. The exceedence of the U.S. national and Oregon water quality criteria was a function of the temporal variability of each bioavailable contaminant. This illustrates the impacts associated with temporal changes of bioavailable organochlorine distributions in surface waters and the significance of considering realistic temporal, bioavailability, and site-specific conditions in risk assessment and water quality management.