PROGRAM commissioned report are not

Transcription

1 LOG TERM EVIROMETL MOITORIG PROGRM RESULTS D ITERPRETTIOS ROM SMPLIG, 8 PWSR ontract o. 9.. The opinions expressed in this commissioned report are not necessarily those of PWSR

2 LTEMP Report 8 TBLE O OTETS bstract Introduction Methods Results and Discussion... Port Valdez Sediments... 9 Sediment Biomarkers... Sediment Grain Size... 9 Tissues in Port Valdez and Knowles Head... Prince William Sound and GO Mussel Tissues... 7 onclusions cknowledgements References ppendices 9 ppendix PH and SH Plots... Sediment... Tissues... 8 ppendix Biomarker Plots... 7 MT July GO July MT July... 8 GO July... 8 ppendix LTEMP PH interpretation using multivariate scoring methods... 8 ii

3 LTEMP Report 8 List of igures igure Map of the LTEMP sites. ircled regions represent sites with similar hydrocarbon signatures and events (discussed later).... igure Time series of TPH in sediments at lyeska Terminal and Gold reek.... igure Example PH and SH signatures of sediments at lyeska Terminal between July 8 and July showing the progression from a primarily petrogenic to pyrogenic PH signature with increasing terrestrial biogenic SH and decreasing higher-molecular-weight n-alkane residuals. Red dashed line is samplespecific MDL. Sample ID code comprises station-matrix-year-season-rep.... igure Example PH and SH signatures of sediments at Gold reek between July 8 and July showing the essentially invariant background naphthalene components and pyrogenic, parent-dominated, PH. SH patterns reflect constant terrestrial and marine biogenic input. Red dashed line is sample-specific MDL.... igure G/MS m/z 9 ion traces for S reference oil from uke Bay Laboratory (upper plot) and a similar orth Slope oil (lower plot) (Lilis et al., 998) demonstrating comparability of detection.... igure Representative BL-quantified biomarker plots of EVOS (S) rude Oil standard (top), MT sediments from July (center), and GO sediments from July (bottom). Red line denotes S reference normalized to the sample s hopane (highlighted in gold).... igure 7 omparison of average GO and MT diagnostic ratios with S reference. Only the three best ratios were used for final evaluations.... igure 8 Diagnostic biomarker parameters showing discrete D clustering of MT (spheres) and GO (cubes) sediment samples; oblique, front, top and right perspectives. Reference oils as pink cylinders; other colors indicate sampling years (989, & ) igure 9 Ion traces of hydrocarbon sources found in PWS. dapted from Wang and Stout, igure umulative grain size curves (%) for GO and MT, igure D plots of grain size components from GO and MT -. Sampling years color coded (blue, red ). ote shifts to higher clay content at MT plus higher sand content at both and then retro-shift in GO samples back to original cluster.... igure Time series of mussel tissues TPH from lyeska Terminal, Gold reek and Knowles Head sites.... igure Representative PH and SH profiles from MT mussel tissues between July 8 and July showing primarily combustion products, perylene, and planktonic biogenic SH and a trace of crude oil (in July ). Red dashed line is sample-specific MDL.... igure Representative PH and SH profiles from GO mussel tissues between July 8 and July showing primarily combustion products, perylene, and planktonic/terrestrial biogenic SH. ote the occurrence of perylene in and was also observed in the MT mussels. Red dashed line is sample-specific MDL.... igure Representative PH and SH profiles from KH mussel tissues between July 8 and July showing primarily trace-level combustion products, perylene (in ), and planktonic/terrestrial biogenic SH.... igure Mussel-tissue TPH trends for all stations (99-) igure 7 Mussel-tissue PH histogram plots showing within-site fidelity among replicates and regional similarities between eastern PWS sites igure 8 Mussel-tissue PH histogram plots showing within-site fidelity among replicates and regional similarities between central and previously EVOS-oiled PWS sites igure 9 Mussel-tissue PH histogram plots showing within-site fidelity among replicates and regional similarities between outer coast sites.... igure Mussel-tissue PH histogram plots showing within-site fidelity among replicates.... iii

4 LTEMP Report 8 igure Status and trends result from national Mussel Watch data (Kimbrough et al., 8). ll laskan sites characterized as low concentrations.... igure Summary page of laska regional Mussel Watch results and trends based on - data from Kimbrough et al., List of Tables Table List of polycyclic aromatic hydrocarbon (PH), saturated hydrocarbon (SH), and biomarkers analytes in this study, along with analyte abbreviations used in figures throughout this report, and internal and surrogate standards.... Table Historic and current method detection limits (MDL) for quantifying hydrocarbon analytes at uke Bay Laboratory. ll values in dry weight (based on average 8- dry wt/wet wt measurements).... Table verage grain size components for GO and MT, -... Table urrent TPH concentrations in regional mussel tissues (ppb, DW) relative to - O Mussel Watch monitoring data and a recovered laskan oil-spill event.... over image view from the mussel community at Disk Island, PWS, July ; logistics provided by orthwind viation, Homer. oto by Bill Driskell. iv

5 LTEMP Report 8 Stations: MT IB OH DII GO KH SHB SHH SLB WIB ZB BL H S BWT DSI DW EVOS EVTHD EMP G/ID G/MS GERG KLI MDL IST MS O PH PEI PGS PSI PWS R S ST SH SIM SRM TH TI TO TPH TSH UM BBREVITIOS lyeska Marine Terminal, Port Valdez ialik Bay, west of Seward onstantine Harbor, Hinchinbrook Entrance, PWS Disk Island, Knight Island Group, western PWS Gold reek, Port Valdez Knowles Head, eastern PWS Sheep Bay, eastern PWS Shuyak Harbor, Kodiak Sleepy Bay, Latouche Island, western PWS Windy Bay, Outer Kenai Peninsula Zaikof Bay, Montague Island, central PWS O/MS uke Bay Laboratory, Juneau K aliphatic hydrocarbons (same as saturated hydrocarbons SH) laskan orth Slope lyeska Terminal s Ballast Water Treatment acility Dissolved Signal Index Dry Weight Exxon Valdez oil spill Exxon Valdez Trustees Hydrocarbon Database US EP Environmental Monitoring and ssessment Program gas chromatography/flame ionization detector gas chromatography/mass spectrometry Geochemical and Environmental Research Group, Texas &M University Kinnetic Laboratories, Inc., nchorage K analytic method detection limit ational Institute of Standards and Technology ational Marine isheries Service ational Oceanographic and tmospheric dministration polycyclic (or polynuclear) aromatic hydrocarbons Payne Environmental onsultants, Inc., Encinitas, particle grain size Particulate Signal Index Prince William Sound Regional itizens dvisory ouncil Scientific dvisory ommittee for PWSR Shoreline leanup ssessment Team saturated hydrocarbons (same as H: n-alkanes + pristane and phytane) selected ion monitoring IST standard reference material total H total inorganic carbon total organic carbon total PH total saturated hydrocarbons (same as total alkanes) unresolved complex mixture v

6 LTEMP Report BSTRT REPORT OR LTEMP 8 Petrogenic hydrocarbon inputs from the lyeska Marine Terminal (MT) and tanker operations are declining in mussels and sediments. The decrease likely reflects a combination of reduced ballast water treatment facility (BTW) discharge volumes from decreased orth Slope oil production, the transition to double hulled tankers with segregated ballast tanks, and improved BWT efficiency at removing particulate/oil phase polycyclic aromatic hydrocarbons (PH) but also reflects an apparent regional decline evident in Knowles Head mussels. omposition is shifting away from petrogenic to pyrogenic in mussels and sediments; even the sediments at the terminal are beginning to look more like the background reference, pyrogenic profiles at Gold reek (GO). Low levels of S oil are evident in Port sediments based on biomarkers; concentrations of these recalcitrant, source specific indicators are seven times higher at MT than GO. Beyond Port Valdez, replicates at each of the seven remote sites show remarkable site fidelity in both PH and saturated hydrocarbon (SH) patterns, even at trace levels. ompared to West oast Mussel Watch data and the more recent laskan Mussel Watch sites, LTEMP results for the Prince William Sound (PWS) and Gulf of laska (GO) sites demonstrate that the region is exceptionally clean. ITRODUTIO In July 8 and pril 9, all ten LTEMP stations (igure ) were visited in a sampling scheme designed, in part, to monitor for lingering impacts from the Exxon Valdez oil spill (EVOS). In subsequent years, the grand round sampling was reduced to only once every five years, and in Port Valdez and Knowles Head, the spring and fall collections were discontinued. nnual summer collections continued at the Port and nearby Knowles Head sites specifically monitoring terminal and tanker operations. This report summarizes findings for July 8 through July, a period comprising the previous contract s final year (8 9) for visiting all sites plus years of the current year monitoring cycle. s appropriate, the results are presented with the overall perspective/trend analysis from the inception of the program. ext year s report will cover the recently completed July visit to all sites and summarize the program s results and findings since 99. Some of these data have been previously presented in a interim data report (Driskell and Payne ). METHODS ollection and analytical methods have been described in previous LTEMP reports (Payne et al., 8a, a). Briefly, three replicates of mussels are collected by hand at each site while triplicate sediment samples are collected from two locations within the Port using a modified Van Veen grab. Sampling protocols have remained constant, and the MS uke Bay Laboratory has continued with the analytical chemistry measurements using their Standard Operating Procedures (SOPs) as detailed in our previous reports.

7 LTEMP Draft Report igure Map of the LTEMP sites. ircled regions represent sites with similar hydrocarbon signatures and events (discussed later). ew to this contract period is the inclusion of biomarkers for sediment samples along with the usual polycyclic aromatic hydrocarbons (PH) and saturated hydrocarbon (SH) data (Table ). Petroleum biomarkers are unique for a given oil source and thus facilitate detection of MT derived, laska orth Slope (S) crude oil constituents in Port sediments. Since the inceptionn of the program, low level, combustion derived PH have been observed in the GO sediments; oil specific biomarkers would be much more useful at teasing out an S signal against the Port s glacial and pyrogenic inputs than the ubiquitous PH. Thee biomarkers would not be as effective for the mussel tissues since unlike the accumulating sediments, mussels regularly purge and at most stations are currently only carrying trace level, dissolved components. Table List of polycyclic aromatic hydrocarbon (PH), saturated hydrocarbon (SH), and biomarkers analytes in this study, along with analyte abbreviations used in figures throughout this report, and internal and surrogate standards. LYTES PH aphthalene aphthalene aphthalene aphthalene aphthalene bbreviation Internal Standard Surrogate Standard

8 LTEMP Report 8 LYTES bbreviation Internal Standard Surrogate Standard Biphenyl cenaphthylene cenaphthene E uorene uorenes uorenes uorenes Dibenzothiophene D Dibenzothiophene D Dibenzothiophene D Dibenzothiophene D Dibenzothiophene D nthracene enanthrene P enanthrene/nthracene P enanthrene/nthracene P enanthrene/nthracene P enanthrene/nthracene P uoranthene L rene PYR uoranthene/rene uoranthene/rene P uoranthene/rene uoranthene/rene Benzo(a)nthracene B hrysene hrysenes hrysenes hrysenes hrysenes Benzo(b)fluoranthene BB Benzo(k)fluoranthene BK Benzo(e)pyrene Benzo(a)pyrene BP Perylene Indeno(,, cd)pyrene IP Dibenzo(a,h)anthracene D Benzo(g,h,i)perylene BP Total PH TPH Saturated hydrocarbons (SH or n alkanes) n Decane B 7 n Undecane B 7 n Dodecane B 7 n Tridecane B 7

9 LTEMP Report 8 LYTES bbreviation Internal Standard Surrogate Standard n Tetradecane B 8 n Pentadecane B 8 n Hexadecane B 8 n Heptadecane B 8 tane tane B 8 n Octadecane B 9 ane ane B 9 n onadecane B 9 n Eicosane B 9 n Heneicosane B 9 n Docosane B n Tricosane B n Tetracosane B n Pentacosane B n Hexacosane B n Heptacosane 7 B n Octacosane 8 B n onacosane B n Triacontane B n Hentriacontane B n Dotriacontane B n Tritriacontane B n Tetratriacontane B Total n lkanes TLK 7 alibrated analytes are identified by boldface. Internal standards: = hexamethylbenzene; B = dodecylcyclohexane. Surrogate standards: = naphthalene-d8, = acenaphthene-d, = phenanthrene-d, = chrysene-d, = benzo[a]pyrene-d, = perylene-d, 7 = dodecane-d, 8 = hexadecane-d, 9 = eicosane-d, = tetracosane-d, and = triacontane-d. Petroleum Biomarkers lass Biomarker bbrev Terpanes tricyclic terpane T tricycilic terpane T tricyclic terpane (a) T tricyclic terpane (b) T tetracyclic terpane Ta tricyclic terpane (a) Tc tricyclic terpane (b) Tb 8 tricyclic terpane (a) T8 8 tricyclic terpane (b) T7 tricyclic terpane (a) T tricyclic terpane (b) T9

10 LTEMP Report 8 8 Hopanes 8α(H),β(H),9, trisnorhopane Ts 7α(H),β(H),9, trisnorhopane Tm 7α(H),8α(H),β(H) 8, bisnorhopane a 7α(H),β(H) norhopane b 7α(H),β(H) norhopane T 8α(H),β(H) norneohopane T 7β(H),α(H) norhopane (normoretane) T7 8α(H) and 8β(H) oleanane T8 7α(H),β(H) hopane T9 7α(H) nor 9 homohopane 7β(H),α(H) hopane (moretane) T S 7α(H),β(H) homohopane T R 7α(H),β(H) homohopane T Gammacerane Ta S 7α(H),β(H), bishomohopane T R 7α(H),β(H), bishomohopane T7 S 7α(H),β(H),, trishomohopane T R 7α(H),β(H),, trishomohopane T S 7α(H),β(H),,, tetrakishomohopane T R 7α(H),β(H),,, tetrakishomohopane T S 7α(H),β(H),,,, pentakishomohopane T R 7α(H),β(H),,,, pentakishomohopane T Steranes α(h),β(h),7â(h) sterane 7 S β(h),7α(h) diasterane (diacholestane) S 7 R β(h),7α(h) diasterane (diacholestane) S 7 S α(h),α(h),7α(h) cholestane S 7 R α(h),β(h),7β(h) cholestane S 7 S α(h),β(h),7β(h) cholestane S 7 R α(h),α(h),7α(h) cholestane S7 8 S α(h),α(h),7α(h) ergostane (methylcholestane) S 8 R α(h),β(h),7β(h) ergostane (methylcholestane) S 8 S α(h),β(h),7β(h) ergostane (methylcholestane) S 8 R α(h),α(h),7α(h) ergostane (methylcholestane) S S α(h),α(h),7α(h) stigmastane (ethylcholestane) S R α(h),β(h),7β(h) stigmastane (ethylcholestane) S S α(h),β(h),7β(h) stigmastane (ethylcholestane) S7 R α(h),α(h),7α(h) stigmastane (ethylcholestane) S8

11 LTEMP Report RESULTS D DISUSSIO s in previous LTEMP reports, all analytes are reported on a dry weight (DW) basis, and corrected for surrogate recoveries. Surrogates are novel or deuterated compounds added in known amounts to each sample in order to assess the efficiency of extraction and analysis. In the uke Bay Laboratory SOPs, surrogate recoveries are considered acceptable if they are between % and %. If more than recoveries from an entire string (analytic batch) fall outside the recovery targets, the string fails Q criteria and is reprocessed. If the majority of the failed recoveries occur in an individual sample, only that sample is reprocessed. Surrogate recovery standards were met for 9.% of all surrogate hydrocarbons analyzed during this reporting period. or the new biomarker analytes, median surrogate recovery was.% for analyses. or different Q reasons, the 9 sediment batch was re extracted and reanalyzed. Laboratory method blanks for each analytic sample batch demonstrated no significant background interference from analytical procedures, thus assuring that the analytes in the field samples represented environmental constituents and not analytical artifacts. In addition, new method detection limits (MDLs), established in, dramatically improved the confidence in accurately quantifying low level PH. Dropping an order of magnitude below previous MDLs, PH limits are now down to a miniscule median. dry weight in sediments and.8 dry weight in tissues (Table ). Likewise, MDLs for SH dropped to a median. dry weight in sediments and dry weight in tissues. Table Historic and current method detection limits (MDL) for quantifying hydrocarbon analytes at uke Bay Laboratory. ll values in dry weight (based on average 8- dry wt/wet wt measurements). Tissue Sediment nalyte bbrev naphthalene methylnaphthalene methylnaphthalene , dimethylnaphthalene ,, trimethylnaphthalene biphenyl* acenaphthylene acenaphthene E fluorene dibenzothiophene phenanthrene P methylphenanthrene anthracene fluoranthene L pyrene PYR benzo(a)anthracene B chrysene benzo(b)fluoranthene

12 LTEMP Report 8 benzo(k)fluoranthene Benzo(e)pyrene Benzo(a)pyrene BP Perylene Indeno(,, cd)pyrene ID dibenzo(a,h)anthracene DB benzo(ghi)perylene min max median extracted sample mass (g wet weight) 8 n nonane.97 n decane n undecane n dodecane n tridecane n tetradecane n pentadecane 9... n hexadecane n heptadecane pristane t n octadecane phytane n nonadecane n eicosane n heneicosane n docosasne n tricosane n tetracosine n pentacosane n hexacosane n heptacosane n octacosane n nonacosane n triacontane n hentriacontane... 7

13 LTEMP Report 8 n dotriacontane 9... n tritriacontane n tetratriacontane n pentatriacontane 7... n hexatriacontane min 7... max median extracted sample mass (g wet weight) 8 8 * Biphenyl is sometimes encountered as a laboratory artifact (as it was during the MDL study) thereby elevating the MDL for this constituent compared to the other analytes. It is not utilized to a great extent in the LTEMP data analysis. 8

14 LTEMP Report PORT VLDEZ SEDIMETS In sediments from the Berth site at the lyeska Marine Terminal (MT), average TPH levels are very low, plateaued around DW since March (igure ), but there has been a shift in the patterns. PH patterns (examples in igure ; all samples in ppendix ) suggest that the higher molecular weight PH became more pyrogenic (i.e. from combustion sources, less petrogenic) starting in July 9. pyrogenic pattern is recognized by the dominant parent PH relative to the alkylated homologues (e.g., in the bottom of igure, the, and are the non alkylated parents); in petrogenic signatures, the analyte groups form a hump pattern in which the parent PH no longer dominates. In the SH plots, by the next year, July, petrogenic alkanes begin to diminish relative to odd carbon numbered biogenic n alkanes (, 7, 9 ) derived from terrestrial plant waxes. By, the intermediate molecular weight PH (fluorenes and phenanthrenes/anthracenes) also became more pyrogenic in character with the SH still dominated by biogenic constituents. The Gold reek reference site (GO) sediments had lower TPH concentrations than MT throughout the duration of the program (Payne et al., 8b) excepting two spill events (igure ), most recently with extremely low concentrations, DW. GO PH profiles were dominated by pyrogenics (Payne et al., 8a,b; a) from pril through this 8 reporting period (see igure and ppendix ). In addition to the highermolecular weight pyrogenic PH, Gold reek sediments also contain a dominant and invariant suite of naphthalenes that are believed to derive from glacial and riverine sediment input to the Port (Payne et al. a,b). SH in the sediments are almost exclusively dominated by biogenic sources from marine phytoplankton and terrestrial plant waxes (i.e., n, n 7, pristane, and an odd carbon numbered suite of n alkanes between n and n, respectively). In comparing the MT and GO sediment profiles, it is interesting that as the petrogenic profiles at MT are disappearing, by July its signatures (igure bottom) begin to resemble the Gold reek reference sediments (igure ). Only traces of the higher molecular weight petroleum waxes (> n ) remain whereas they used to dominate the SH profiles (Payne et al., 8a,b). In our LTEMP report (Payne et al., a), we cautioned that at extremely low concentrations using our algorithmic PH source phase assignment model is problematic whereby a very minute change in a trace pattern could flip a phase assignment. or the 8 data where nearly all patterns at trace levels and near MDLs, we ve abandoned that approach in favor of pattern descriptions and graphics. We also include the interpretations of Mark arls (ppendix ) in which he s applied a multivariate scoring procedure (arls, ) to assign type and sources. His results generally corroborate what we re seeing in the histogram patterns. We also noted that without biomarker data, attributing the very minor petrogenic signal observed in Gold reek sediments to lyeska Marine Terminal discharges would be confounded. rom their EMP biomarker data, Shaw et al. () concluded that EMP Stations and (near Gold reek but at greater depths in the Port) contained petrogenic components derived from laska orth Slope crude oil. To confirm this finding that our data only hinted at, a suite of biomarkers was added to the sediment analyte list with the 9 LTEMP contract. 9

15 LTEMP Report 8 MT S GO S TPH dry wt Mar 9 Mar 9 Mar 97 Mar 99 Mar Mar Mar Mar 7 Mar 9 Mar Mar Mar igure Time series of TPH in sediments at lyeska Terminal and Gold reek.

16 LTEMP Report 8 7 7//8 MT S 8 SEDIMET 97 7 TPH MT S TSH Y P P P P P B BP IDP DH //9 MT S 9 SEDIMET MT S 9. TSH. 7R TPH... Y P P P P P B BP IDP DH // MT S SEDIMET 7 TPH 7 MT S. TSH.. Y P P P P P B BP IDP DH // MT S Sediment 8 MT S 8. TSH 7 TPH. 8. Y P P P P P B BP IDP DH // MT S SEDIMET 8 TPH MT S 8.7 TSH 7 Y P P P P P B BP IDP DH igure Example PH and SH signatures of sediments at lyeska Terminal between July 8 and July showing the progression from a primarily petrogenic to pyrogenic PH signature with increasing terrestrial biogenic SH and decreasing higher-molecular-weight n-alkane residuals. Red dashed line is sample-specific MDL. Sample ID code comprises station-matrixyear-season-rep.

17 LTEMP Report 8. 7//8 GO S 8 SEDIMET 9 TPH GO S TSH.. Y P P P P P B BP IDP DH //9 GO S 9 SEDIMET 7R TPH GO S 9.9 TSH Y P P P P P B BP IDP DH // GO S SEDIMET 7 TPH GO S 7.8 TSH.. Y P P P P P B BP IDP DH 7 8. // GO S SEDIMET TPH GO S 7.8 TSH.. Y P P P P P B BP IDP DH // GO S SEDIMET GO S. TSH. 8 TPH... Y P P P P P B BP IDP DH 7 8 igure Example PH and SH signatures of sediments at Gold reek between July 8 and July showing the essentially invariant background naphthalene components and pyrogenic, parent-dominated, PH. SH patterns reflect constant terrestrial and marine biogenic input. Red dashed line is sample-specific MDL.

18 LTEMP Report SEDIMET OMRKERS t our request, in, the uke Bay Laboratory (BL) modified their analytical procedures to report petroleum biomarkers. The method is an extension to the normal G/MS 87 method used to acquire PH data; conceptually, three new ions are added to the mass spectrometer s list for selective ion monitoring (SIM). s a lab calibration standard, BL runs the original Exxon Valdez oil (from the ship s hold, an S crude oil), in addition to a IST standard reference material (SRM). But there are no certified standard values for biomarkers in these oils; we can only compare BL results to other lab results for validation. We ve done this for both S and Deepwater Horizon oils (also analyzed at BL and now a IST standard). The results are favorable; BL s S reference oil results compare quite well with published S results (igure ); however, the lab has issues separating the triplet T terpane compounds, Ta, Tb and Tc (identified as peaks 8, 9 & (igure center). This is unfortunate since the triplets are often used as diagnostic markers. In BL quantified plots of the field samples (igure ), there also seem to be deficits in the mid terpane analytes (left third of the plot) but with this small number of samples from only two locations (and only two analytic strings), it is difficult to know, at this stage, if these results reflect actual environmental conditions or analytic artifacts. It is also possible that the biomarker patterns for S crude oil have changed slightly since 989 as oil from several new orth Slope fields are blended into the pipeline mixture. More samples of S oil are being requested from lyeska to assess the PH and biomarker composition of the current blend. Presumably, this will also reflect the biomarker composition in oil residues in the treated ballast water currently being discharged into the Port. Multiple approaches have been suggested for interpreting biomarker data but some degree of expert guided pattern matching must be employed. Most approaches involve various diagnostic ratios (Wang and Stout, 7) with several ratios normalized on the highly conservative 7α(H),β(H) hopane (also labeled T9 or hopane). But depending on the local environs and despite the purported persistence of biomarkers, all ratios are not equally effective and must be individually evaluated for a given spill/habitat. With LTEMP data, we initially screened results graphically with an overlaid S reference (igure ). We then used the frequently reliable, Ts/Tm and norhopane (T)/hopane plus a suggested R homohopane (T)/hopane ratios to confirm the visual similarities. ormally, the T triplets ratio also would be added to this list but not if the lab is having analytic issues. The final suite of diagnostic ratios can then be the basis for a rigorous methodology advocated by European agencies (ordtest plots) or adapted for other multivariate approaches (various authors in Wang & Stout, 7). Mark arls is currently exploring a novel scoring method.

19 LTEMP Draft Report 8 hopane norhopane (T) homohopane (T) T triplet Tm Ts hopane norhopane (T) homohopane (T) Ts Tm T triplet Tricyclic terpane [R] Tricyclic terpane 7 7baa Moretane [] Tricyclic terpane [S] Tricyclic terpane 8 S Homohopane [] Tricyclic terpane 8a Trisnorneohopane [7 Ts] 9 R Homohopane [] Tricyclic terpane 7 [R] Tricyclic terpane Gammacerane [] Tricyclic terpane 8 7a Trisnorneohopane [7 Tm] S Bishomohopane [] Tricyclic terpane 9 [S] Tricyclic terpane R Bishomohopane [] 7 Tricyclic terpane [R] Tricyclic terpane S Trishomohopane [] 8 [ S] Tricyclic terpane Bisnorhopane [8] R Trishomohopane [] 9 [ R] Tricyclic terpane orhopane [] S Tetrakishomohopane [] Tetracyclic terpane 8a eonorhopane [] R Tetrakishomohopane [] 8 [ S] Tricyclic terpane 7ba ormoretane [] 7 S Pentakishomohopane [] 8 [ R] Tricyclic terpane Oleananee [] 8 R Pentakishomohopane [] [ S] Tricyclic terpane Hopane [] igure G/MS m/z 9 ion traces for S reference oil from uke Bayy Laboratory (upper plot) and a similar orth Slope oil (lower plot) (Lilis et al., 998) demonstrating comparability of detection.

20 LTEMP Report S Source T T T T Ta Tc Tb T8 T7 T T9 Ts Tm a b T T T7 T8 T9 T9a T T T Ta T T7 T T T T T T S S S S S S S7 S S S S S S S7 S8... MT S T T T T Ta Tc Tb T8 T7 T T9 Ts Tm a b T T T7 T8 T9 T9a T T T Ta T T7 T T T T T T S S S S S S S7 S S S S S S S7 S8... GO S T T T T Ta Tc Tb T8 T7 T T9 Ts Tm a b T T T7 T8 T9 T9a T T T Ta T T7 T T T T T T S S S S S S S7 S S S S S S S7 S igure Representative BL-quantified biomarker plots of EVOS (S) rude Oil standard (top), MT sediments from July (center), and GO sediments from July (bottom). Red line denotes S reference normalized to the sample s hopane (highlighted in gold). rom the and LTEMP collections, eighteen samples were analyzed. Visual inspection of individual plots for sediment biomarkers and their associated diagnostic ratios (ppendix ) show general agreement by stations and across sampling years. The MT sites are consistently similar to the S reference profile in the mid hopane suite of analytes (Ts through T) (igure ). t GO, the fit is similar but looser, and there are more nondetected analytes because the concentrations of biomarkers are lower (total biomarkers avg:. DW at MT vs.. at GO). Thus, there are some concerns with detection limits. However, even with a looser fit, the

21 LTEMP Report 8 selected GO ratios are in agreement with the reference (igure 7) for the three selected diagnostic ratios (T, T and Ts/Tm), and when GO s total biomarker concentrations peak (GO S D, 7. DW and GO S, 9. DW), the fit is excellent (ppendix ) S ref MT avg GO avg.... T/T9 T/T9 D/P D/P Ts/Tm S/s7 with std err of means igure 7 omparison of average GO and MT diagnostic ratios with S reference. Only the three best ratios were used for final evaluations. Plotting the diagnostic ratios in D space (igure 8) shows samples clustering by station. Both stations differ from the EVOS era S reference (in pink) with MT (spheres) being more similar (closer) to the reference. lso note that GO tends to remain tightly clustered across years (in these three parameters), more so than at MT, which surprisingly suggests more a variable background influence in the MT biomarker signature. dditional reference samples (described below) are also shown. In these plots, the primary distinction between stations is driven by norhopane (T) levels. But these are just the fine scale differences; overall, these profiles are significantly different from two other petrogenic sources found in the Sound. s depicted by Bence et al. (99) (igure 9), the S origin EVOS oil differs from Katalla and Yakataga regional sources east of PWS, primarily by the absence of the oleanane biomarker (seen just before the hopane) plus various subtler differences (e.g., note Ts and Tm proportions in all non EVOS plots). In the Gulf of laska, the Katalla/Yakataga materials become finely fragmented and get transported westward by Gulf currents eventually reaching all the way to Shelikof Straits and down the laska Peninsula but with some deposited in the PWS basin to become the much disputed (during EVOS era) PWS background. lso found in intertidal regions of the Sound are tarballs from Monterey formation oil (Kvenvolden et al., 99), remnants from oil spilled during the 9 earthquake. Imported in the pre pipeline era from alifornia, Monterey oil is also distinct from S oil by its inclusion of both oleanane and 8, bisnorhopane (just before norhopane igure 9). Oleanane comes from the flowering angiosperms which only appeared in the later Triassic formations; apparently the orth Slope deposits never saw the bloom. In one of the eighteen LTEMP sediment samples, a trace level of oleanane was detected, again confirming that like the dispersion seen in the D plots, the biomarker patterns represent mixtures with other background biomarker sources.

22 LTEMP Draft Report 8 GO BL S MT 8 BL SS 87 7

23 LTEMP Draft Report 8 BL SS 88 BL S igure 8 Diagnostic biomarker parameters showing discrete D clustering of MT (spheres) and GO (cubes) sediment samples; oblique, front, top and right perspectives. Reference oils as pink cylinders; otherr colors indicate sampling years (989, & ). 8

24 LTEMP Draft Report 8 9 Exxon Valdez Oil Monterey Oil Katalla Seep Oil = Hopane 9 = orhopane (T) O = Oleanane (T8) IS = labb internal standard igure 9 Ion traces of hydrocarbon sourcess found in PWS. dapted from Wang and Stout, 7. Our preliminary assessments, pending accumulating a larger time series of samples, additional lab experience, and a more current S source sample, lead us to conclude that the mid hopane centric diagnostic ratios confirm the presence of S derivedd biomarkers in both MT and GO sediments. urther, the form of variation in diagnostic biomarkers is site specific with both locations showing variants of the S source (i.e., weathering and/or mixtures with background biomarkers). These findings are corroborated byy Shaw et al. () in lyeska s EMP, who, using hopane and norhopane, reported finding S biomarkers in deep Valdezz basin sediments. SEDIMET GRI SIZE Sediment grain size samples were recently analyzed for all collections, comprising both historic and future samplings. These data are presented in two formats: the standard cumulative (%) grain size curves and in D plots (igure & igure ). or this project, the grain size data only serve to demonstrate the constancy and comparability of the sampling site environs. Both sites are dominated by glacial flour inputs, showing approximately equal portions of clay and silt with minor sand components (Table ) and both sites show minor trends and outliers. In the D plots, note there are annual shifts to higher sand content at both locations albeit still a minor component (<%) and with a return to original conditions at GO in. More dramatically, at MT, there has been a steady increase in clay content. The locations occupy a heterogeneous fjord floor dynamically swept by tidal currents (and prop wash at MT) plus with sampling guided by GPS, the sites have been accruing grab sampler pock marks and drag scars at the same locations for years. Reassuringly, when we get off sitconsidering the non at GO, we begin to see gravel in the grab. These shiftss are noted with only modest confidence rigorous collection methods, i.e., spooning up ml of sample remnants after collecting the less consolidated surface floc for hydrocarbons. 9

25 LTEMP Report 8 Table verage grain size components for GO and MT, - % lay % Silt % Sand MT GO GO grain size 8 7 lay Silt Sand MT grain size 8 rep 8 lay Silt Sand 9 igure umulative grain size curves (%) for GO and MT, -.

26 LTEMP Draft Report 8 GO Grain Size Replicates MT Grain Size Replicates igure D plots of grain size components from GO and MT -. Sampling years color coded (blue, red ). ote shifts to higher clay content at MT plus higher sand content at both and then retro-shift in GO samples back to original cluster.

27 LTEMP Report S I PORT VLDEZ D KOWLES HED In July, average TPH levels in MT tissue samples approached all time lows, ~7 DW (igure ), and although they appear to have increased slightly (~ DW) in July, the majority of the PH are pyrogenic not petrogenic in nature (see igure and other profiles in ppendix ). lso note the appearance of perylene in July and. This is a naturally occurring ring unsubstituted PH generated by biologic processes or in the early stages of diagenesis in marine sediments (Bence et al., 7) and potentially being neither a nonpetrogenic or pyrogenic PH, is not included in TPH calculations. Most of the saturated hydrocarbons (SH) in MT mussel tissues throughout this period are dominated by below MDL biogenic constituents (e.g., n, n 7, pristane, and n 7 ). In July, two of the three replicate tissue samples also showed traces of below MDL, higher molecular weight SH in the range suggesting a possible petrogenic input, but this was not reflected in the PH patterns. verage TPH levels in GO mussel tissues have consistently been very low except for the all diesel spill (cleared by the following July igure ). Beginning in the 99 s, mixed dissolved phase, petrogenic, and pyrogenic signals were common at this site, roughly trending with similar patterns or discharge events at the lyeska Marine Terminal (Payne et al., 8a,b; a). Between March and July 8, with the exception of the aforementioned diesel spill, the signals became largely pyrogenic with occasional petrogenic components (s) (igure ). rom July to present, the TPH levels have consistently been in the 8 7 DW range. SH patterns at this site over the 8 period have been almost exclusively trace level biogenic (e.g., n, n 7, pristane, and n 7 ). Exceptions included two of three replicates in pril 9, and one of three replicates in June/July when traces of petrogenic SH appeared (see ppendix ). In none of these cases, however, were petrogenic sources suggested by the PH profiles. Interestingly, traces of perylene noted in the MT mussels in July and were also observed at GO during the same collections. verage TPH levels in mussel samples from the tanker anchorage at KH have been consistently low, ranging from DW since March (igure ). With the exception of one sample in July that showed a petrogenic PH profile (Payne et al., 8a,b; ), the samples from this site have been dominated by trace level (at or just above detection limit) naphthalenes and combustion derived anthracenes/phenanthrenes since July 999; this pattern has continued throughout the July 8 July period (see igure and additional profiles in the ppendix ). lso note the appearance of perylene that was also observed in the mussels at MT and GO in and. Presumably this reflects some region wide marine event. The near or just below MDL SH patterns at KH between July 8 and July are almost exclusively biogenic (e.g., n, n 7, and pristane) except for one of three replicates in July when a slightly fuller, but below MDL pattern suggested the possibility of a petrogenic source. The PH profile, however, did not support this observation (igure ).

28 LTEMP Report 8 tanker spill TPH dry wt diesel spill all stations MT B GO B KH B Mar 9 Mar 9 Mar 97 Mar 99 Mar Mar Mar Mar 7 Mar 9 Mar Mar 7 8 igure Time series of mussel tissues TPH from lyeska Terminal, Gold reek and Knowles Head sites.

29 LTEMP Report //8 MT B 8 9 TPH MT B 9. TSH Y P P P P P B BP IDP DH 7 8 Y P P P P P B BP IDP DH /7/9 MT B 9 D 8 TPH MT B 9. TSH // MT B 8 MT B.8 TSH TPH... Y P P P P P B BP IDP DH Y P P P P P B BP IDP DH 7// MT B TPH MT B. TSH // MT B Tissue 87 TPH MT B 7.9 TSH Y P P P P P B BP IDP DH igure Representative PH and SH profiles from MT mussel tissues between July 8 and July showing primarily combustion products, perylene, and planktonic biogenic SH and a trace of crude oil (in July ). Red dashed line is samplespecific MDL.

30 LTEMP Report 8 7 7//8 GO B 8 9 TPH GO B 9.9 TSH Y P P P P P B BP IDP DH 7 8 Y P P P P P B BP IDP DH 7//9 GO B 9 7 TPH GO B 9.8 TSH // GO B TPH 8 GO B 9. TSH Y P P P P P B BP IDP DH // GO B TPH 8 GO B. TSH 8 8 Y P P P P P B BP IDP DH // GO B Tissue GO B 9. TSH TPH 7 Y P P P P P B BP IDP DH igure Representative PH and SH profiles from GO mussel tissues between July 8 and July showing primarily combustion products, perylene, and planktonic/terrestrial biogenic SH. ote the occurrence of perylene in and was also observed in the MT mussels. Red dashed line is sample-specific MDL.

31 LTEMP Report 8 8 igure Representative PH and SH profiles from KH mussel tissues between July 8 and July showing primarily 9 trace-level combustion products, perylene (in ), and planktonic/terrestrial biogenic SH Y P P P P P B BP IDP DH KH B 8 D 9 TPH 7// KH B 9.8 TSH.... Y P P P P P B BP IDP DH KH B 9 77 TPH 7// KH B TSH... Y P P P P P B BP IDP DH KH B TPH 7// KH B 8. TSH.... Y P P P P P B BP IDP DH KH B TPH 7// KH B.7 TSH 8 Y P P P P P B BP IDP DH KH B 8 Tissue TPH 7/9/ KH B. TSH

32 LTEMP Report PRIE WILLIM SOUD D GO MUSSEL S Only one additional sampling (pril 9) has been added to the time series since the publication of our data report, and as of that sampling, overall TPH levels at these central PWS and GO sites continued to decline (igure ) although more recently, the annually sampled Valdez sites have marginally rebounded. TPH dry wt tanker spill Mar 9 Mar 9 Mar 97 Mar 99 Mar Mar Mar Mar 7 Mar 9 Mar Mar Mar diesel spill IB B OH B GO B SHB B SLB B ZB B MT B DII B KH B SHH B WIB B igure Mussel-tissue TPH trends for all stations (99-). ll of the PH levels measured in the pril 9 samples are extremely low (mostly near or just below even the new MDLs), yet remarkable site fidelity exists within the PH and SH patterns for each stations replicates (igure 7 through igure ). nd to a degree, regional patterns are still visible although weaker than in the previously reported 8 data set (Payne et al., a). We believe that despite the at or below MDL character of the data, the within site and within region consistency would not be achievable if data quality were uncontrolled at these trace levels. The laboratory method blanks were also very clean (primarily only showing below MDL traces of the problematic biphenyl), which again supports the notion that rather than laboratory noise, the observed patterns were truly due to low level constituents in the mussel tissues at each site. 8 7

33 LTEMP Report igure 7 Mussel-tissue PH histogram plots showing within-site fidelity among replicates and regional similarities between eastern PWS sites Y P P P P P B BP IDP DH KH B TPH /7/ KH B 9. TSH Y P P P P P B BP IDP DH KH B 9 8 TPH /7/ KH B 9. TSH..... Y P P P P P B BP IDP DH KH B 9 9 TPH /7/ KH B 9.88 TSH Y P P P P P B BP IDP DH SHB B 9 TPH /8/ SHB B 9.98 TSH.... Y P P P P P B BP IDP DH SHB B 9 7 TPH /8/ SHB B TSH... Y P P P P P B BP IDP DH SHB B 9 8 TPH /8/ SHB B 9 9. TSH Knowles Head Sheep Bay

34 LTEMP Report igure 8 Mussel-tissue PH histogram plots showing within-site fidelity among replicates and regional similarities between central and previously EVOS-oiled PWS sites Y P P P P P B BP IDP DH DII B 9 TPH /8/ DII B 9.7 TSH... Y P P P P P B BP IDP DH DII B 9 TPH /8/ DII B TSH... Y P P P P P B BP IDP DH DII B 9 D TPH /8/ DII B 9.8 TSH... Y P P P P P B BP IDP DH SLB B 9 TPH /8/ SLB B 9.7 TSH... Y P P P P P B BP IDP DH SLB B 9 TPH /8/ SLB B 9.9 TSH... Y P P P P P B BP IDP DH SLB B 9 TPH /8/ SLB B TSH Disk Island Sleepy Bay

35 LTEMP Report 8 9 igure 9 Mussel-tissue PH histogram plots showing within-site fidelity among replicates and regional similarities between outer coast sites Y P P P P P B BP IDP DH IB B 9 TPH /9/ IB B 9 7. TSH 7 8 Y P P P P P B BP IDP DH IB B 9 8 TPH /9/ IB B TSH... Y P P P P P B BP IDP DH IB B 9 TPH /9/ IB B 9. TSH... Y P P P P P B BP IDP DH WIB B 9 9 TPH /9/ WIB B TSH... Y P P P P P B BP IDP DH WIB B 9 TPH /9/ WIB B TSH... Y P P P P P B BP IDP DH WIB B 9 TPH /9/ WIB B TSH ialik Bay Windy Bay

36 LTEMP Report 8 9 igure Mussel-tissue PH histogram plots showing within-site fidelity among replicates Y P P P P P B BP IDP DH SHH B 9 TPH // SHH B 9. TSH.... Y P P P P P B BP IDP DH SHH B TPH // SHH B 9. TSH.... Y P P P P P B BP IDP DH SHH B 9 8 TPH // SHH B 9 9. TSH... Y P P P P P B BP IDP DH ZB B 9 7 TPH /8/ ZB B 9. TSH... Y P P P P P B BP IDP DH ZB B 9 8 TPH /8/ ZB B TSH... Y P P P P P B BP IDP DH ZB B 9 9 TPH /8/ ZB B 9.9 TSH Zaikof Bay Shuyak Harbor

37 LTEMP Report ll of the PWS and GO stations (igure 7 through igure ) show ubiquitous low level naphthalene signals that we think are part of the overall background pattern (Payne et al., a,b). Likewise, Knowles Head (KH), Sheep Bay (SHB), and Windy Bay (WIB) show pyrogenic phenanthrenes/anthracenes along with lower concentrations of fluorenes and isolated peaks for parent fluoranthene, pyrene, and chrysene (all combustion products). The relative phenanthrene/anthracene homologue distribution patterns at Disk Island (DII) and Sleepy Bay (SLB) (both formerly EVOS oiled sites) show a little more complexity and within site variability suggesting the possibility of a more petrogenic influence, e.g., oil was still observed in disturbed intertidal substrates at DII in pril 9 (Payne et al., a). Zaikof Bay (ZB) also shows evidence of a minor petrogenic phenanthrene/anthracene pattern, and its fluoranthene/chrysene/perylene profile is unique among all the stations. inally, Shuyak Harbor (SHH) and ialik Bay (IB) (with the exception of one replicate) show the least complex PH patterns of all the stations examined. The SH patterns at all of these sites are mostly below the sample specific MDLs, but there appears to be sporadic indications of petrogenic SH (n n ), again with remarkable site fidelity at several of the stations. In the Sound, with the bulk of EVOS oil either dissipated or the remnants buried and sequestered, the regional sites that are physically remote from the chronic low input of BWT discharge are currently looking very clean. s stated in our last report, TPH levels in both the PWS and GO regions are dropping at nearly identical rates. This parallelism suggests the decreases are influenced by similar (oceanographic scale?) processes, but the occasional asynchrony of peak events also suggests regional variation in the dynamics. So what broad scale sources might appear as an ambient background signal? Possibilities include atmospheric deposits from forest fires and volcanic activity, leaching of water soluble constituents from the pervasive source rock (oil shale or coal that constitutes much of the PH bearing material being transported through the region), some upwelling/climate driven events, a combination of these or some novel mechanism relating to decadal oscillations and/or global warming. Based on a limited scale partitioning experiment with clean seawater and intertidal sediment from onstantine Harbor (Payne et al. b), we now believe that sedimentary material previously believed to only contain sequestered and nonbioavailable PH, is a component of the ubiquitous dissolved phase (naphthalene) signals observed throughout Prince William Sound. nother question is how low will this generally declining trend in TPH values go? Obviously, at some point, the trend must level out. ollow up sampling in for oil residues from the 997 M/V Kuroshima grounding in Summer Bay, Unalaska, found TPH levels between and 8 DW, with an average of 7 DW (Table, Helton et al., ). This compares favorably and is actually higher than LTEMP s pril 9 range of DW inside the Sound and 8 DW at Gulf of laska sites (overall average DW). These data also suggest a natural dissolved phase background TPH somewhere below DW a range in which analytical sensitivity can be highly susceptible to procedural artifacts. It could easily be the case that the LTEMP data are currently tracking subtle variations in the background PH, and that we are near or at the minimum. or relative comparisons, data from the ational Status and Trends, Mussel Watch Program (Table, igure ) and 8 laskan sites (igure ) (now summing 8 parent and alkylated PH homologues versus LTEMP PH analytes) show that PH concentration in mussels for other West oast sites is nearly times higher at 8 DW. The highest level reported on the West oast was,9 DW in Seattle, Washington. The lowest, DW, was from mussels collected on Santa atalina Island, miles offshore of Orange ounty in Southern alifornia. In, the average TPH concentration in mussels from the five laskan Mussel Watch sites (Ketchikan, ahku Bay, Port Valdez, Unakwik Inlet, and ook Inlet) was 7 DW with levels ranging from DW (Kimbrough et al., 8). onsidering these and even more recent 8 values from the MW data portal (Table ), the LTEMP results for the PWS and GO sites demonstrate that these remote locations are exceptionally clean.

38 LTEMP Report Table urrent TPH concentrations in regional mussel tissues (ppb, DW) relative to - O Mussel Watch monitoring data and a recovered laskan oil-spill event. LTEMP Port Valdez 7 9 PWS 9 GO 8 West oast Mussel Watch average (Kimbrough et al., 8) 8 So. alif. Santa atalina Island Seattle Elliot Bay, W,9 laska Mussel Watch 8 ook Inlet Bear ove 9 ook Inlet Homer Spit 8 Port Valdez lyeska Marine Terminal Port Valdez Gold reek 9 Ketchikan Mountain Point ahku Bay East Side 9 Port Valdez Mineral reek ats Resurrection Bay ash Road Resurrection Bay ash Road 7 Resurrection Bay ash Road 99 Resurrection Bay ash Road 7 Unakwit Inlet Siwash Bay 7 Resurrection Bay ash Road,7 anwalek anwalek 9 Port Graham Port Graham 7 Port Graham Murphy's Slough 8 Seldovia Bay Powder Island M/V Kuroshima (997) Unalaska followup 8

39 LTEMP Draft Report 8 igure Status and trends result from national Mussel Watch data (Kimbrough et al., 8). ll laskan sites characterized as low concentrations.

40 LTEMP Draft Report igure Summary page of laska regional Mussel Watch results and trends based onn - data from Kimbrough et al., 8. inally, a EVOS Trustees Program, Long term Monitoring of nthropogenic Hydrocarbons in the Exxon Valdez Oil Spill Region, examined ten intertidal sites within the aked Knight Southwest Island complex to measure the extent of buried oil still present years after the spill. t previously heavily oiled EVOS sites, to random pits (depending on the beach width) were excavated to a depthh of ~. m to look for residual oil. If oil was detected, available nearby mussels were also collected. The results have been published elsewhere (Short et al., 7) but, as co authors, PH and SH sample profiles were included in ppendix E of our / Report (Payne et al., 8a). Briefly, TPH levels in the oiled pits ranged from a low of (on Knight Island) to a high of 7, (on Latouche Island) with the oil showing states of weathering varying from very fresh to

41 LTEMP Report extensively degraded. On the other hand, nearby mussel samples only showed low dissolved phase TPH ( DW, derived primarily from naphthalenes and phenanthrenes/anthracenes) that were in the same concentration range but compositionally different from the signals observed at LTEMP PWS and GO sites sampled in pril 9. rom these studies, it was concluded that although in, there were still persistent buried EVOS residues at a number of beaches, they were highly sequestered and did not appear to be bioavailable unless disturbed. Rates of disappearance had diminished to an estimated % yr. If left undisturbed, Short et al. (7) predicted they would be there for decades. To address these residual deposits, for the last three years EVOS Trustees have sponsored beach remediation projects OLUSIOS Based on the sediment and mussel data, petrogenic hydrocarbon inputs from the lyeska Marine Terminal and tanker operations continue to decline. Within Port Valdez, and as concluded in our last LTEMP report (Payne et al., a), the decrease likely reflects a combination of reduced BTW discharge volumes from decreased orth Slope oil production, the transition to double hulled tankers with segregated ballast tanks, and improved BWT efficiency at removing particulate/oil phase PH. In the tissue collections at the Terminal, the mussels showed a predominant pyrogenic, not petrogenic, profile, and even the sediments at this site are beginning to look more like Gold reek s background reference, pyrogenic profiles. Biomarker data show evidence of low level S oil in the Port sediments. t GO, concentrations are seven times lower than those at MT and the profiles indicate mixing with background sources. With the pril 9 mussel collections at the central PWS and GO sites, the overall TPH levels at these stations continued to decline. ll of the PH levels measured in the pril 9 samples are extremely low (mostly near or just below even the new MDLs), yet remarkable site fidelity exists within the PH and SH patterns for each stations replicates. The SH patterns at all of these sites are mostly below the sample specific MDLs, but there appears to be sporadic indications of petrogenic SH (n n ), again with remarkable site fidelity at several of the stations. However, in corresponding PH fractions when trace level petrogenic SH signals were observed, there was little or no evidence of oil contamination. ompared to West oast Mussel Watch data and the more recent laskan Mussel Watch sites, the LTEMP results for the PWS and GO sites demonstrate that the region is exceptionally clean KOWLEDGEMETS This success of this project reflects the efforts of many more individuals than currently appear in the author s list. rom our stalwart vessel operator, tireless supporter and colleague, Dave Janka (and nnette Janka), our primary pilots, Terry Kennedy and Jose Dereft, our former and current colleagues at uke Bay Lab (including the project newbie, Dr. L. Schaufler), project managers, Lisa Ka aihue and Joe Banta, our many prior and current S advisors, and the every ready folks on R staff and our families for the years of laskan adventures and productive science G THKS! REEREES Bence,.E., K.. Kvenvolden, M.. Kennicutt II 99. Organic geochemistry applied to environmental assessments of Prince William Sound, laska, after the Exxon Valdez oil spill a review. Organic Geochemistry, Volume, Issue, January 99, Pages 7