CICEET Progress Report for the period 9/01/08 Through 2/15/09

Project Title: Field Validation of General Methodology for Evaluating Narcosis Toxicity using the Sediment Profile Imaging and Micro-sampling System (SPIMS)
Principal Investigator(s): Marion Nipper
Additional Investigator(s): R. Scott Carr, Philip M. Gschwend
Project Start Date: 9/01/07

Figures

Figure 1

Tasks to meet objectives
To perform chemical analyses of collected sediments and conduct a thorough analysis of the data gathered so far.

Progress on Tasks
Chemical analyses were performed with (a) the pore waters collected from subsamples of the sediments to which passive samplers were exposed, and (b) polyethylene strips (PE) strips that were equilibrated with the same sediments.

Have the results/data gathered during this reporting period changed the project objectives when compared to your original proposal? Please explain.
No.

Dissemination activities during this reporting period (please include the number of participants where applicable).
A poster entitled “Passive samplers as sources of hydrophobic organic chemicals in sediment toxicity tests” was presented at the 29th Annual Meeting of the Society of Environmental Toxicology and Chemistry, on 16-20 November, 2008, in Tampa, Florida, with an attendance of approximately 3,000 people.

Difficulties
None so far. However, due to the delay in the completion of the sediment profile imaging and micro-sampling system (SPIMS), which was supposed to be used for sediment profile imaging and site delineation in the current project, these activities are delayed. We will now move forward and use the REMOTS® camera for this purpose, with assistance of Germano and Associates, Inc., but a 1-year unfunded extension of this project may be necessary and will be requested.

Data Generated to date
As part of our efforts to interpret the integrated narcosis doses from mixtures of hydrophobic organic compounds in sediments, we have been working to assess the mixtures accumulated in polyethylene strips, used as passive samplers. The sediments are mixed with the passive sampler for 12 days, then the passive sampler is solvent extracted, and finally, the extracts are analyzed by GC/MS and/or GCxGC. Comparisons of the results for sediment pore waters (PW) and corresponding passive samplers exposed to the same sediments were also performed (See Nipper_Figure1_March2009.doc). In the previous report it had been established that sediments collected from our initial set of sampling sites yielded extracts from the passive samplers which exhibited chromatographic traces that were substantially different from one another. Interestingly, the PW extracts from Tabbs Bay, Formosa and Witco Cut, gave chromatographic traces which appear to be similar to one another in their major components. The Tabbs Bay trace shows signs of an underlying unresolved complex mixture (UCM) which matches the findings with the passive sampler. The Corpus Christi Bay porewater extract does not seem to match the pattern from the above samples so clearly. Some of the components match the spectra and retention times of the other traces, but it does not seem that the overall signal fingerprint matches up particularly well. The similarity among porewater extracts when the passive sampler extracts were so different is surprising. We theorize that the water soluble components at these sites are similar in composition, and these compounds are not concentrated by the passive sampler since this chiefly absorbs the more hydrophobic compounds available in the sediment. Not surprisingly, the passive samplers of contaminated sites (e.g., Tabbs Bay) show very different chromatographic outputs than reference sites (e.g., TX Ref Site) (See Nipper_Figure2_March2009.pdf). To interpret these complex mixtures, we need to weight the analytes for their differential tendencies to bioaccumulate. As a first approach, we have worked to calibrate the two-dimensional GC outputs in terms of the octanol-water partitioning properties associated with each retention window (i.e., at each (x,y) elution position on the 2-D chromatograms where x = the retention time on our apolar column, t₁, and y = the retention time on our polar column, t₂). To date, we have analyzed 29 diverse alkanes, PCBs, PAHs, phthalates, chlorinated pesticides, and nitrogen compounds (aniline and nitrobenzene), all with known log K_ow values, to seek a correlation between retention behavior and octanol-water partitioning. For this training set, we have found:

log K_ow = 0.140 + 0.014 * t₁ - 1.449 + 0.220 * t₂ + 3.942 N=29, R² = 0.80

where t₁ is the retention time of the analyte on our apolar column, and
t₂ is the retention time of the analyte on our polar column.

These results suggest we can look at analytes with log K_ow values ranging from about -2 to 12. This expression can now be used to estimate the octanol-water partition coefficients of even the unidentified substances we detect by GCxGC. In the future, we will use this ability to combine quantitative analyses of the mixtures in each retention window with what this correlation suggests the lipid-uptake would be based on previous relations between lipid-water partitioning and octanol-water partitioning.

Based on data generated to date we plan on returning to Tabbs Bat and to the Corpus Christi Bay sampling sites for further studies. However, hurricane Ike is likely to have strongly revolved the sediments in Galveston Bay and, consequently, Tabbs Bay. Therefore, if the sediment profile imaging (SPI) survey reveals a strongly modified environment which lacks benthic community gradients from the contamination source, we may return to Lavaca Bay to conduct a SPI survey at the Formosa site. The Texas NERR will be the reference site.

Project Objectives for Next Reporting Period

Objectives
1) To define the stations at which passive samplers will be deployed in situ; 2) To acquire better understanding of the loading and unloading dynamics of passive sampler in the presence of chemicals with a range of log K_ow.

Work Plan to Meet Objectives
1) Sediment profile imaging studies will be performed at a gradient from sources of HOCs using the REMOTS® camera; 2) Laboratory analyses will be performed with passive samplers using chemicals with a range of log K_ow. Chemical analyses will be performed temporally using passive samplers exposed to the different chemicals, and water exposed to the loaded passive samplers. Ultimately, toxicity tests will be performed using the time frame of unloading established with the chemical experiments.

Dissemination Objectives for next reporting period
A webpage for this project and linked to the pages of the Texas PIs will be created in GulfBase.org. This is a website dedicated to the dissemination of information on Gulf of Mexico research.

Expenditures
Expenditures are within the anticipated range for the activities performed to date.

End User Advisor Feedback
End User Advisor: Robert M Burgess
Organization: U.S. EPA Office of Research and Development/NHEERL
Location: Atlantic Ecology Division - Narragansett
Phone number: 401-782-3106
E-mail: burgess.robert@epa.gov

At this stage, what are the potential applications for this research? Please discuss how you and others could potentially use the technology.
The project appears to be on course and the researchers are communicating their interim findings to the appropriate audiences. Right now, and at this stage of the project, it is too early to address potential applications of the research. When completed and field validated, the results of this work will be very applicable in a wide range of environmental management uses including the relatively inexpensive mapping of the distribution and bioavailability of organic contaminants in sediments. This application could be very interesting to U.S. EPA programs like Superfund.

What are the key challenges to application of this technology? Please consider the technology itself as well as issues related to regulation, politics, socio-economic pressures, trends in the field etc.
For the technology being developed and field validated in this project, the key challenges to application are likely to be generating a tool that is scientifically robust with regard to the data it produces but user friendly (and inexpensive) such that it is readily adopted by environmental managers. Successful field validation of the methodology discussed in the proposal will make this adoption more likely.

Has anything changed about this project's potential applicability since the last reporting period (not applicable to the first Progress Report)?
No.

Questions/comments/ suggestions for the researchers?
(1) General comment: This progress report presents some very interesting results. I look forward to the next update.

(2) Section 1.e: Authors may want to indicate the SETAC meeting was held in Tampa, Florida.

(3) Section 1.g: First paragraph: Are there any concerns about the 12 days being insufficient to achieve equilibrium in the sediment-polyethylene system?

(4) Section 1.g: Second paragraph: For the contaminated sediments, have you considered that, in some cases (e.g., Tabbs Bay), the method for extraction of the interstitial water may result in the loss of some compounds in the final extract While in contrast, these compounds do appear in the passive sampler extract ­ especially the higher molecular weight compounds? Also, the peak at ~10.5 minutes in the interstitial water scans appears in every sample, is it possible this is a contaminant introduced during the extraction?

(5) In the next report, it would be useful to have a table showing the compounds, log K_ows, and retention time windows for the chemicals making-up the 29 used to develop the regression equation presented on page 2.

(6) Figure 2: This figure needs more text describing what is being presented. It is difficult to interpret.

PI Response to End User Advisor Feedback
2. The location of the SETAC Meeting was added to Section 1.e of the report.

3. Yes, we are concerned that 12 days is not sufficient. Empirically, PE strips have appeared to equilibrate with excess sediment in a tumbling slurry (~12 days) for many PAHs, PCBs and dioxins in our previous work (Lohmann et al., ES&T, 39, 141-148, 2005). However, more recent observations on highly chlorinated PCBs exchanging into PE from sediment suggest this time may prove insufficient for compounds that exhibit very low diffusivities within the polyethylene (e.g., octa-, nona-, and deca- chlorobiphenyls). Since the PE/sediment ratio is such that we insignificantly deplete the contaminants from the sediment, the slow step in the exchange involves diffusion into the polymer, and so compounds with D_PE > (12.5x10^-3 cm)² /(12 days x 86400 sec/day) = 1.5x10^-12 cm²/sec should equilibrate. While most PAHs of interest have diffusivities in PE that are greater than this (e.g., Rusina et al., Chemosphere, 2007), we are finding that highly chlorinated PCBs have lower diffusivities in the PE, so these large compounds may be under-represented.

4. Yes, we do believe that the porewater extraction method may promote loss of some compounds, particularly the more hydrophobic ones, which should be present in the passive samplers. We expect this to be the source of some of the differences in chemistry and toxicity that we have observed. However, the low concentrations of HOCs in the pore water are not expected to play a big role in the aggregate narcosis dose and, therefore, not contribute as much to porewater toxicity.

GCMS analysis indicates the peak we see in our porewater extracts at ~10.5 minutes is di-n-butyl phthtalate, a common contaminant.

5. The compounds used to train the retention time-log Kow relations are shown in the Figure 1.