are bifacially ground and have straight

Archaeology/Human Impacts: Contamination in Intertidal Archaeological Sites

~ 216

ting edges. The collection contains plan-ing adzes of the variety usually inset intoa bone socket and then hafted. Other

diagnostic artifacts in the Lower Compo-nent include a slate fragment with anetched face, a stone saw, and chippedslate preforms for subsequent grindinginto projectile points.

Similar, early Kaniag Phase collec-tions at well dated sites in the regionhave assigned ages of between A.D. 1350and A.D. 1500. Age estimates for theKoniag Phase in the Kodiak Archipelagocite a beginning around A.D. 1100 toA.D. 1200. Age of the incised slate figu-rines or faces is particularly well definedduring the A.D. 1350 to A.D. 1500 period.Triangular ground slate points with aflat ground basa facet typically date fromA.D. 1300 to A.D. 1500, Four radiocar-

bon samples stratigraphically associatedwith the Lower Component range fromA.D. 787+110 to A.D. 1143 &5. The four

dates are acceptably within the earliestestimated range of the Koniag Phase andmay indicate that the developmentalstage of the phase occurred earliest in thenorthern Kodiak area.

The AFG-098 Upper Component col-lection provided an artifact collectionwhich is dearly related to Koniag Phasematerials elsewhere. Ground slate pro-jectile points of several forms were re-covered. Stemmed points, some withbarbs and some with rounded shoul-

ders, were the most common forms re-covered. Medial ridges were present ansome barbed forms but a flattened bicon-

vex cross-section was most common

among the stemmed points. A very dis-tinctive ground slateprojectilepoint witha triangular outline and a sharply de-fined flute or butt facet was found in the

Upper Component. Several ground slateulu forms occur in the component. Both

are bifacially ground and have straightto slightly convex cutting edges. Oneform has a distinctive notch or offset at

the back of the blade which suggestshafting at the back edge near one endwith the other end extending out of thehandle. Other stone artifacts from the

Upper Component include a stone saw,sawn slate fragments, whetstones, plan-ing adzes, a quartz crystal, a hematitenodule for firestarting, and chipped slatepreforms. Fragments of several bane orivory dart heads with unilateral barbs, abarbed hook fragment, and bone awlsare some of the organic artifacts recov-ered. A small jet labret was also found.Fragile artifacts carved from sprucewood, bark and grass matting, and frag-ments of a birch bark container were

recovered from the saturated deposits.More than a dozen species of seeds werefound in floor deposits of the Upper Com-ponent.

Comparison of the Upper Compo-nent collection with Koniag Phase or re-lated site collections on Kodiak Island,both sides of the Alaska Peninsula, andthe outer coast of the Kenai Peninsulademonstrate dose similarities. The tri-

angular point form with a basal fluteconsistently occurs in Koniag Phase orrelated collections dating from betweenA.D. 1500 and Historic times A.D. 1750-

1800!. Other point and ulu forms areconsistent with later Koniag Phase agesas well. Six radiocarbon dates obtained

from the Upper Component levels ofAFG-098 range from A.D. 1343 +60 toA.D. 1490 +125.

The AFG-082 Site is an upland siteeroding slowly into the intertidal zonewhich was tested for contamination fromstorm wave or wind deposited oil. Theground slate ulus, chipped bifaces, plan-ing adze, and small single chamber

217 ~

houses compare reasonably well withmiddle age range Kachemak Traditionremains, approximately 2,000 years old.The ratio of ground slate to chipped stoneremains also supports a Kachemak Tra-dition comparison, Unfortunately, theAFG-082 collectionis toosmall and typo-logically limited to provide a very accu-rate age estimate.

Two radiocarbon samples obtainedfrom the site date to A.D. 203 &5 andA.D. 288 &5. The AFG-082 dates fallwell within the expected age range de-termined from artifact typology. No evi-dence of oil was found in the site and nosediment samples were submitted foranalysis.

The SEL-215 Site on Nuka Island con-tains intact cultural remains within apeat deposit in the intertidal zone. Timespecific traits in the collection are meagerbut grooved splitting adzes suggest anage between A,D. 1000 and A.D, 1500.Indusion of a glass trade bead in thecollection is interpreted as an intrusivehistoric element.

Radiocarbon samples from the siteyielded seven dates ranging from A.D.1142 &0 to A.D. 1442+105. A trace of oilwas detected in one of the two sedimentsamples submitted for HPLV/UV screen-ing. However, the radiocarbon determi-nations compare well with the expectedage of the deposits.

The SEW-068 Site consists of a peatyinter tidal deposit containing cultural re-mains which relate to Kachemak Tradi-tioncollectionselsewhere. Grooved split-ting adzes located nearby may not beassociated however such tools have beendated to that early time from other sitesin Prince William Sound. A general esti-rnate of age for the peaty deposits, basedon artifact typology, is 1000-2000 yearsago. Geological age estimates based on

Exxon Vuldex Oil Spill Symposium Abstracts

rates of isostatic rebound from the 1964Earthquake and from long term regionalsubsidence indicate the deposits shouldbe at least 1500 years old, Two woodsamples from thesaturated deposits pro-vided radiocarbon dates of A.D. '10+65and A.D. 391 +65. The ages obtainedfrom the culturally modified wood frag-ments agree roughly with the expectedage of the cultural deposits.

Conclusions drawn from this studyinclude:

�! Intertidal archaeological deposits atthree of the sites investigated demon-strate that useful and important in-formation is preserved in some inter-tidal sites even though there is some-times no surface evidence of buriedremains.

�! Traces of petroleum hydrocarbons insubsurface remains dooccur althoughthe origin of the contaminants in thesites tested is unknown.

�! Reliable radiocarbon dates can beobtained from oiled deposits. It isuncertain, however, whether that re-sults in the tested sites from cleaningof the samples or simply lack of ac-tual contamination. Examination ofsamples in the radiocarbon labora-tory for oil contaminants should beroutine and cleaningmethods shouldbe modified if necessary to removeidentified contaminants.

�! Dating of intertidal or even exposedupland archaeological remains needsto involve every possible approach todating, not just reliance on a singlemethod. Archaeologists investigat-ing chronological questions in the areaof the Exxon Valdez oil spillneed tobeespecially critical with their conclu-sions. No evidence was found in thesites examined that the Exxon Valdez

Archaeology/Human Impacts: Contamination in Intertidal Archaeological Sites

~ 218

oil spill adversely affected the radio-carbon dating results. Damage tosites appears to be from erosion orvandalism rather than direct oiling,

ReferencesMifflin and Associates. 1991. Exxon Valdez oil

spill damage assessment contamination ofarchaeological materials, Chugach NationalForest: radiocarbon experiments and relatedanalyses, final report U. S. Forest ServiceContract No. 53-0109-1403G5.

Generating Damage Restoration Costs for Archaeological Injuries ofthe ExxonValdez Oil SpillMartin E. McAllisterArchaeological Resource Investigations

3.

4.

5.

Three categories of site specific resto-ration proposals were developed forthe remaining 24 sites: sites recom-mended for monitoring only, sitesrecommended for direct physical res-toration only, and sites with bothtypes of measures recommended.

Standard levels of effort were formu-

lated for direct physical restorationin year one and in years two throughten for oil effect monitoring. Appro-priate site-specific sa1ary estimates

6.

7.

219 ~

This paper summarizes the results ofa monetary assessment of damage forinjuries to archaeological sites docu-mented in the Exxon Valdez oil spiH re-sponse records. Injuries attributable tothe oil spill at 35 archaeological sites inPrince William Sound and the Gulf ofAlaska were analyzed to estimate resto-ration costs for use by the Zxxort ValdezTrustee Council in planning restorationof damages for archaeological resources.The damage assessment was accom-plished in two steps.

First, a damage assessment panel,chaired by the author, met to considerrestoration costs based on the archaeo-

logical injury data. Second, working fromthe findings of the panel, additionalanalyses were conducted by the authorto calculate specific restoration costs forthe archaeological injuries. Two levels ofrestoration costs were produced by thedamage assessment process. First, site-specific restoration costs were developedfor the archaeological sites identified ashaving substantive injuries. Second,gross restoration costs were estimatedfor projected numbers of sites injured bythe oil spill.

The procedures employed by thedamage assessment panel were carriedout in nine steps:1. Thedataonthe35archaeologicalsites

with injuries attributable to the oilspill and other relevant data sourceswere reviewed.

2. A conceptual framework on whichrestoration costs would be based was

Exxon Valdez Oil Spill Symposium Abstracts

developed from the damage assess-ment model contained in the Archaeo-

logical Resources Protection Act of1979, as amended ARPA!.

&m.unented oil spill injuries to ar-chaeological sites were analyzed andgrouped into two major categories,those resulting from oiling and thoseresulting from oil spill response ac-tivities.

Two restoration options were devel-oped, one for ten years of oil effectmonitoring and one for direct physi-cal restoration. These were used to

formulate specific restoration mea-sures appropriate for the two catego-ries of injuries.Ten injured sites were eliminatedfrom further consideration because

appropriate siterestoration work hadalready been accomplished or the sitedamage was not severe enough torequire restoration. Damage at onesite was determined to beunrestorable.

Archaeology/Humean Impacts: Generating Damage Restoration Costs

~ 220

were generated using standard lev-els of effort as a guide.

8. Site-specific support costs necessaryto carry out direct physical restora-tion measures and oil effect monitor-

ing were estimated.9. Restoration costs were estimated for

four different injury scenarios, eachwith a different number of injuredsites, using average per site costs fordirect physical restoration measuresand oil-effect monitoring.

Two principal sets of findings wereproduced. The first consists of the pro-posals for site restoration measures andcosts for the 24 archaeological sites withsubstantive injuries. The second consistsof restoration costs for projected num-bers of sites injured by the oil spill.

Site-specificrestoration proposals arebased on the oil effect monitoring anddirect physical restoration measures.Using the above criteria, there are fivesites in the oil effect monitoring onlycategory, 14 sites in the direct physicalrestoration only category, and five sitesat which both types of measures are rec-ommended.

The first element of the restoration

cost proposals for the 24 sites are person-nel salaries. The salary figures for 17 ofthe 24 sites are based on the three stan-

dard levels of effort defined above. Due

to special circumstances, seven sites havesalary figures based on variations of thestandard computations.

Five sites have cost proposals involv-ing only salaries for ten years of monitor-ing. The standard salary figure for yearone oil effect monitoring is $2,904.51,and $2,202.17 per year for years twothrough ten or $19P19.53 for 9 years.Therefore, the total salary figure for eachof these sites is $22,724.04.

Salary for direct physical restoration

only is proposed for 11 sites. The totalsalary figure for each of these sites is$3,155.79.

Only one site is in the oil effect moni-toring and direct physical restorationsalary category. The total salary figurefor this site consists of the standard sala-ries for ten years oilwffect monitoring,plus costs for directphysicalrestoration-$25P79.83.

Four sites have salary figures for di-rect physical restoration measures abovethe standard level of effort, as well as oileffect monitoring. One has disinterredhuman remains which required the ad-dition of eight days of project supervisortime for consultation with Native Corpo-rations. The result is a salary increase of$1~.24 over the standard salary fordirect physical restoration. The total sal-ary figure for the site consists of theamount for the combined standard sala-

ries shown in the preceding paragraphand the extra cost for consultation,$25,879.83+ $1504.24, or $27,684.07. Atthe other three sites, a test excavation isproposed to fully assess the magnitudeof damage. This required the addition oftwo person days for fieldwork and oneperson day each for analysis and reportpreparation. The result is a salary in-crease of $527.76 over the standard sal-ary for direct physical restoration. Thetotal salary figure for these sites consistsof the amount for the combined standardsalaries and the extra cost for the test

excavation, $25,879.83 + $527.76, or$26,407.59.

One other site is proposed for directphysical restorationsalary abovethestan-dard level, but not for oil effect monitor-ing. Because this site has disinterredhuman remains, eight days of projectsupervisor time were added for consul-tation with Native Corporations at the

221 ~

cost of $1,804.24. Also, the large volumeof disturbance at this site required theaddition of four person days for fiekl-work and two persondays each for analy-sis and report preparation, The result isa salary increase of $1,055.52 over thestandard salary for direct physical resto-ration, The total salary figure for this siteconsists of the standard salary for directphysical restoration and the extra costsfor consultation and restoration mea-

sures, $3,155.79 + $1+04,24 + $1,055,52,or $6,01S.55.

Finally, two sites in the direct physi-cal restoration only category have salaryfigures below the standard level becausetheir injuries require measures differentfrom those proposed for most other sites.For one site, the total salary figure of$2,480.83is for repatriation orreintermentadministration and consultation. The

direct physicalrestoration measures pro-posed for the other site allowed the elimi-nation of two person days for fieldworkand one person day each for analysis andreport preparation. The result is a salarydecrease of $892.08 from the standardsalary for direct physical restoration to atotal salary figure of $2,263.71.

The other elements of the restoration

cost proposals are support costs, Basicsupport costs are for: fieldwork per diem,transportation, supplies and equipment,and processing and duplication. Recov-ery of items requiring expenditures forcuration is anticipated at all but two ofthe sites proposed for direct physicalrestoration measures. The proposals forthree sites involve repatriation orreinterment costs. At one site, the onlysupport costs are for repatriation orreinterment.!

The support cost amounts vary bysite. The average support cost figuresare: for year one oil effect monitoring,


$4,086.83; for years two through ten oileffect monitoring, $33,159.60; and fordirect physical restoration, $10,920.33.

The total figures for the restorationcost proposals for the 24 sites under con-sideration are as follows:

Year one oil

effect monitoring: $69,913,40

Years two through tenoil effect monitoring: 29 791.30

Ten years of oileffect monitoring: $599,704.70

Direct physicalrestoration measures: 272 126.49

Total restoration costs: $871,831.19

Average costs per site for oil effectmonitoring and direct physical restora-tion were calculated from the total cost

figures by dividing them by the numberof sites for which the measures are pro-posed. The average costs are $59,970.47per site for oil effect monitoring and$14,322.45 per site for direct physicalrestoration.

The projections for the numbers ofsites injured on which the estimates ofgross restoration costs are based werederived from four sets of figures:1. Theprojectednumbersof injured sites

in the draft State University of NewYork report entitled Exxon Vatdez OilSpill Archaeological Damage Assess-ment by Albert J. Dekin et al. �992!.

2. The Dekin et al. figures reduced bythe percentages corresponding to thenumber of sites eliminated from con-

sideration by the damage assessmentpanel because the injuries were notsevere enough to require restoration.

3, The number of sites proposed by thepanel for oil effect monitoring and

Archaeology/Human Impacts: Generatirtg Damage Restoration Costs

$4+40,988.10$3,867,061.50

$658+32.70$1,618~.85

~ 222

direct physical restoration as percent-ages of the total number of sites nowincluded in the Alaska Heritage Re-source Survey site inventory recordsfor the oil spill area.

4. The number of sites proposed by thepanel for oil effect monitoring anddirect physical restoration as percent-ages of the number of sites actuallylocated and exMninedby Exxoncrewsduring oil spill response activities.

Each of these four figures was multi-plied by the average per site costs for oileffect monitoring and direct physical res-toration. The resulting gross restorationcosts estimates for oil effect monitoringare:

�! $31+44,319,57�! $28,126,150.43�! $1,439,291.28�! $3,598,228,20

For direct physical restoration measures,they are:

�!�!�!�!

The total gross restoration cost estimatesare:

�! $36~+07.67�! $31,993,211.93�! $2,098,123.98�! $5,216,665.05

The gross cost estimates based on thenumber of sites actually field checked byExxon number four above! are seen asreliable indicators of the overall magni-tude of archaeological restoration needsresulting from the oil spill.

Two important conclusions aredrawnfrom the results of the work summarized

in this paper. First, the ARPA damageassessment model was used successfullyto generate credible restoration cost de-terminations for the documented ar-

chaeological injuries of the Fxxon VatdezOilSpill. Second, the ARPAmodel shouldbe the damage assessment and restora-tion cost determination standard for ar-

chaeological injuries resulting from fu-ture oB spills or other similar situations.ReferencesDekin, A. A. jr., et al. 1992 Exxon Valdex oil spill

archaeological damage assessment, draft re-port. The Research Foundation of the StateUniversity of New York, Binghamton.

Archaeology/Human Impacts: Long-Term Social Psychological Impacts

herring permits and 55% of all PrinceWilliam Sound salmon fishery permits Stratton, 1989!. Subsistence activities i.e., harvesting, giving, receiving fish,moose, deer, berries, etc.! characterize90% of Cordova's households Stratton,1989! and Alaskan natives make up 18percent of the population. These andother data classify Cordova as a naturalresource community Dyer et al., 1992!.The commercial fishing industry andnumerous related businesses link thiscommunity directly to seasonal harvestsof renewable natural resources.

Petersburg, Alaska was selected asthe control communijy for this study.Petersburg is located on an island in thesoutheastern part of the state and is rela-tively isolated with no road connectionsoutside Mitkof Island. Petersburg has apopulation of 3/00 people, with Alas-kan natives comprising approximately20%. Petersburg's economic baseis basedon commercial fishing. Petersburg resi-dents also engage in subsistence activi-ties at a rate similar to Cordova Smythe,1988; Stratton, 1989!,

Data Collection: A stratified random

household sample was selected in theCordova community. In August, 1989,personal interviews with 86 respondentswere conducted, reflecting 70households. Random digit dialingtelephone interviews were conducted inPetersburg and Cordova tocompletedatacollection activities for the first year. In1990, follow-up interviews wereconducted by mail and telephonesurveys. In1991,respondents in Cordovawere reinterviewed by personalinterviews and respondents in Petersburgwere once again contacted by telephone.

Indicators and Measures: Informa-

tion was collected from respondents ondemographics, social attitudes, work,

family and personal disruption and so-cial psychological stress. Social psycho-logical stress was measured by the "im-pact of events scale," which identifiestwo stress components � intrusion andavoidance Horowitz et al., 1979;Horowitz, 1986!.

Statistical Analysis: A series of pathmodels, or structural equation models,were calculated for data available from1989 to 1991 Duncan, 1966; Birnbaum,1981!. In general, these models attemptto identify causal relationships betweensocial structural characteristics, socialdisruption and psychological stress.Results

Higher levels of intrusive stress andavoidance behavior were observed for

the impacted community in 1989, 1990and 1991. These differences were found

to be statistically significant Fr < .05!when t-tests were applied to compareimpact and control community meanscores on the stress indicators. Meanstress scores were found to decline from

1989 to 1991 in the impact communitysuggesting that, over time, a reduction inthe intensity of the social psychologicalimpacts.

The evaluation of the models pro-vided limited support for the hypothesisthat social structural characteristics in-

fluence social and psychological reac-tions of victims of technological disas-ters, The primary predictors of intrusivestress in 1991 included intrusive stress in

'1990 and 1989, work disruption in 1989and continuing social disruption in 1991.Attitudes toward the effectiveness of

cleanup operations were found to pre-dict long-term stress. That is, respon-dents who were most pessimistic aboutcleanup effectiveness in 1990 tended tobe more stressed in 1991. In general,respondents who were male and who

225 ~

experienced both work and family dis-ruption in 1989 held the most pessimisticviews of cleanup effectiveness.

The long-term patterns of social psy-chological stress found in previous stud-ies of a variety of technological disasterswere also observed for residents of the

impact community in this research.Higher levels of intrusive stress andavoidance behavior were found to exist

in 1989, 1990 and 1991 in the impactcommunity. Over time, levels of stresswere found tobe declining. For example,mean intrusive stress scores fell from

24.47 in 1989 to 19.32 in 'l990 and then

further declined to 17.74 in 1991. This

general trend suggests a pattern of re-turn to community equilibrium.

Attempts to develop and evaluatecausal models of long-term stress result-ing from the Exxon Valdez oil spill weremodestly successful. Although someintervening variables were found to pre-dict long-term stress, initial stress levelswere the most important predictors oflater stress levels in the models. Intru-

sive stress existing some 20 months afterthe spill was found to be related to per-ceptions that the cleanup was ineffec-tive. However, the relative effects of

both attitudes toward the cleanup andproblems experienced from litigation onintrusive stress were small when com-

pared to effects of previous stress levelsgenerated from the spill.

In conclusion, alternatives to linear,additive models may be required to fullyunderstand the complexpatternsoflong-term stress created by technological di-sasters in general and the Exxon Valdezoil spill, in particular, Models utilizinginteraction terms may provide more ac-curate explanations. Future analyses ofthese data will evaluate the utility of thisapproach.


ReferencesBaum, A., R. Fleming, and J. E. Singer. 1982.

"Stress at Three Mile Island: Applying psy-chological impact analysis." In: L. Bickman,ed. Applied Social Psychology Annual, 3.Sage: Newbury Park, CA: 217-248,

Baum, A., R. Fleming, and J. E. Singer. 1983."Coping with victimization by technologicaldisaster," Journal of Social Issues,39�!: 117-138.

Birnbaum, I. 1981. An introduction to causalanalysis in sociology. Atlantic Highlands.N.J.: Humanities Press.

Brown, Phil and Edward J. Mikkelsen, 1989. Nosafe place: Toxic waste, leukemia and com-munity action. Berkeley,CA: UniversityofCalifornia Press.

Couch, S. R. and S. Kroll-Smith. 1985. ' Fhechronic technical disaster: Toward a social

scientific perspective." Social Science Quar-terly 66: 564-575.

Duncan, O.D. 1966. "Path analysis: Sociologicalexamples." American Journal of Sociology.72: 1-16.

Dyer, C. L., D. A. Gill, and J. S. Picou. 1992."Social disruption and the Valdez Oil spill:Alaskan natives in a natural resource com-

munity." Sociological Spectrum 12�!:105-126.

Fall, J. A. 1990. "The Division of Subsistence ofthe Alaska Department of Fish and Game:An overview of its research program andfindings: 1980-1990." Arctic Anthropology,27�!: 68-92,

Freudenburg, W. R. and T. R. Jones. 1991. "At-titudes and stress in the presence of techno-logical risk: A test of the supreme courthypothesis." Social Forces 69�!: 1143-'1168.

Gill, D. A. and J. S. Picou. 1989. "Toxic wastedisposal sites as technological disasters." In:Psychosocial Effects of Hazardous ToxicWaste Disposal on Communities. Thomas,Springfield, IL: 81-97.

Gill, D. A. and J. S. Picou. 1991. "The SocialPsychological Impacts of a TechnologicalAccident: Collective Stress and Perceived

Health Risks." Journal of Hazardous Mate-rials 27�!:77-89.

Horowitz, M.J. 1986. Stress response syndromes.Nor thvale, NJ: Jason Aronson, Inc.

~ 226

Axchaeology/Human Impacts:

Horowitz, M.J., N. Milner and W. Alverez. 1979."Impact of event scale: A measure of subjec-tive stress." Psychosomatic Medicine41�!:209-218.

Omohundro, J, D. 1982. 'The impacts of an oilspill." Human Organization, 41I1!: 17-25.

Picou, J.S. and D.D. Rosebrook. 1992. "Techno-logical accident, community class-actionliti-gation, and scienti6c damage assessment: Acase study of court-ordered research." So-ciological Spectrum 13: 117-'i38.

Picou, J, Steven, Duane A. GiH, Christopher L.Dyer and Evans W. Curry. 1992. "Disrup-tion and stress in an Alaskan fishingcommu-nity: Initial and continuing impacts of theExxon Valdezoilspili." IndustNalCrisisQuar-terly 6: 235-257.

Restoration Planning Work Group. 1990. "Res-toration planningfollowing the Exxon VaMezoil spill: August, 1990. Progress Report, U.S,Environmental Protection Agency. Washing-ton, DC 20460.

Smythe, Charles W. 1988. Harvest and use of fishand wildlife resources by residents of Peters-burg, Alaska. Alaska Department of Fishand Game, Division of Subsistence, Techni-cal Paper No. 164, June. Anchorage, AK.

Stratton, Lee. 1989. "Resource usein Cordova: Acoastal community of Southcentral Alaska."Alaska Department of Fish and Game, De-partment of Subsistence Technical Paper No.153, December. Anchorage, AK,

Vyner, Henry M. 1988. Invisible trauma: Thepsychosocial effects of invisible environmen-tal contaminants. Lexington, MA: D. C.Heath and Company.

The Economic Impacts of the Exxon Valdez Oil Spill on SouthcentralAlaska's Commercial Fishing IndustryMaurie J. CohenUniversity of Pennsylvania

The potential of natural disasters togenerate short-term economic benefitsfor impacted individuals and commuru-ties has become an accepted social sci-ence notion Dacy and Kunreuther, 1969;Rogers, 1970; Cochrane, 1975; Wright etal. 1979; Friesema, et al. 1979; Rossi et al.1983!. Unfortunately, the economic di-mensions of human-made disasters havenot received similar treatment despitethe considerable attention that has re-

cently been focused on these events andtheir sequelae e.g., Erickson, 1976, 1991;Levine, 1982; Couch and Kroii-Smith,1985; Shrivastava, 1987!.

In the case of oil spills, economic re-search has been largely confined to theestimation of comprehensive damageassessments Mead and Sorensen, 1970;Burrows et al, 1974a, 1974b; Brown et al.1983; Grigalunas, 1982; Grigalunas et al.1986; Assaf et al. 1986!. With the excep-tion of work completed by Nelson �98l!and Restrepo �982! there has been littleconsideration of the perturbing influ-ences that technological disruptions canhave on local and regional economies.

Within the context of recent techno-

logical accidents e.g., Bhopal, Three MileIsland, Love Canal!, the Exxon Valdez oilspill is distinguished by its widespreadphysical damageof a highly valued natu-ral environment and its extraordinaryeconomic bonanza Cohen, 1993!. In or-der to contain the spilled cargo, collectcontaminated debris, and clean oiledshorelines a massive emergency responseoperation was assembled that eventu-

Exxon Vatdex Oil Spill Symposium Abstracts

ally employed 11,000 local residents andtransient laborers at wages exceeding$16 per hour.

However, while these ephemeralwindfall profit opportunities werebeingexploited, the fundamental componentof the regional economy was experienc-ing a downward realignment. In an ef-fort to measure the magnitude of thisadjustment, this analysis derives ex postestimates of the oil spill's economic im-pacts on southcentral Alaska's commer-cial fishing industry during the years1989 and 1990 in isolation of the consid-

erable financial benefits imparted by theemergency response operation. A three-phase methodology is employed to de-termine the ex-vessel revenue that wouldhave been earned for eachof southcentralAlaska's eleven major commercial fish-ery products chinook, sockeye, coho,pink and chum salmon, king crab, tan-ner crab, Dungeness crab, Pacific herringsac roe, Pacific halibut, and sablefish!during each of these two years had theoilspill not occurred.

First, estimates of the accident'sharvest volume impacts are constructedusing data reported by the AlaskaDepartment of Fish and Game ADF&G!,International PacificHalibutCommission

IPHC!, and the National MarineFisheries Service NMFS!. Thepre-seasonharvest expectations for each commercialfishery in southcentral Alaska PrinceWilliam Sound, Lower Cook Inlet, andKodiak Island! were compared to actualyields. In order to ascertain the extent of

~ 228~

Archaeology/Hujn.m Impacts: Economic Impacts

con',a mutation from the oil spill, physicalmor.i oring and organoleptic test resultswere examined for each regulatoryjuri.;<.iction that evidenced a harvestdefi it in either of the two years undercon i Ceration. Harvest volume impactsfor ~,hach commercial fishery were thenapp 0 ~tioned to the accident according tothe;w data. This method led to theconclusion that the oil spill's harvestvol u me impacts were confinedprii«ipally to Pacific herring and pinkanc hum salmon during the'l989 season.No harvest volume reductionsatt: ibutable to .the accident wereexl: e ~fenced in 1990.

S cond, the Exxon Vatdez oil spiH washylx ithesized to have motivated a fun-dain ntal shift in ex-vessel demand forsoi..t >central Alaska's commercial fish-em ~ roducts as retail consumers becamefearlul of tainted seafood. This effectwas estimated with a price-dependentdemand model that was variouslyad.q >ted to the specific characteristics ofeai.l of the region's major fish and shell-fis,'i species, Independent variables in-cli.c ed in these spe6fications were these i~onal quantity of landed product,national income of major consumingco z >tries, frozen and canned inventoryhoblings, price of substitutes, and ex-ch r. age rates between the United Statesan d major consuming countries.

' Restructuralequations derived fromth.s model were estimated by a biasedle; i~ t squares Btting procedure with datafo- the period 1964-1988. The subse-q..c ntly derived coefficients were thenu:-e C to generate ex post forecasts of thee~- ressel prices that would have pre-v; i ed for southcentral Alaska's commer-ci i fishery products during 1989 and1<~ 0 in the absence of the oil spill. Thep.: a dieted ex-vessel prices were then con-

trasted with their corresponding actualvalues. On the basis of this technique,the largest ex-vessel price impacts weresustained in both years by Pacific her-ring sac roe and coho and chum salmon,Noteworthy is the observation that ac-tual ex-vessel prices for Pacific halibutand sablefish exceeded predicted valuesduring bothseasons, raising the possibil-ity that some demand substitution ofthese products least threatened by oilcontamination may have occurred.

Finally, the full extent of these esti-mated harvestvolume and ex-vessel priceimpacts cannot be attributed to the ExxonValdez oil spill as the commercial fishingindustry was concurrently perturbed byseveral other biological and economicinfluences in addition to the accident,On the biological side, commercial fish-ery yields are generally subject to consid-erable stochastic variability due to nu-merous environmental factors. For in-stance, fluctuations in rates of predation,disease mortality, and water tempera-ture can alter interseasonal commercialfishery harvest volumes. Additionally,various forms of human intervention,including regulatory measures and arti-flcial culbvation, can result in harvestvolume adjustments.

On the economicside,ex-vesselpricesin 1989 and 1990 were influenced byseveral perturbations that occurred si-multaneously to the oil spilL After at-taining unprecedented levels in 1988, ex-vessel prices for most Alaskan commer-cial fishery products began to erode in1989, This trend was motivated by sub-stantial increases in the volume of salmonproduced internationally. Other factorscontributing to this decline included ex-cessive wholesale inventories, reducedconsumer spending, unfavorable ex-change rate adjustments, and suspended

229 ~

speculation among Tokyo fishery prod-uct wholesalers. In order to isolate the

ex-vessel revenue impacts of the ExxonValdez oil spill from these confoundingbiological and economic factors severalalternative scenarios were simulated.

Though each scenario was based ondifferent assumptions regarding themagnitude of the Exxon Valdez oil spillrelative to the array of confounding fac-tors, a consistent conclusion emerges.The economic boom that swept oversouthcentral Alaska as a result of the oil

spill obscured the decline in the profit-abHity of the region's commercial fishingindustry and exasperated the deteriora-tion of international market conditions.

Specifically, the accidentreducedex-ves-sel revenue for southcentral Alaska's

commercial fishers during 1989 by anamount ranging between $6.4 millionand $41.8 million. Ex-vessel revenuereductions were greatest for the region'scommercial harvest of sockeye and pinksalmon while increased ex-vessel values

for Pacific halibut and sablefish margin-ally mitigated these declines. This analy-sis indicates that the oil spill's ex-vesselrevenue impacts in 1990 were between$11.1 miDion and $44.5 million.

Employing 1988 as a baseline, theseamounts represent 3-l9 percent of theex-vessel value of southcentral Alaska's

commercial fishing economy. Given theconsiderable imprecisioninherent in eco-nomic impact analysis of complex per-turbations such as the Exxon Valdez oil

spill, more explicit evaluation is notreadily possible. In spite of this indeter-minacy, this analysis provides aboundedinterval in which one measure of the

accidenYs economic dimensions can be

considered.

The interpretation of these resultsrequires the consideration of at least four


factors. First, the harvest volume impactmethodology lacks scientific rigor and isfurther undermined by the uncertaintythat exists regarding hydrocarbon-basedtoxicity on individual marine species.Even under relatively static circum-stances the determination of fishery har-vest volumes on the basis of historical

trends and biometric forecasting modelsis an imprecise exercise.

Second, theex-vessel price model pro-vides only a partial explanation of de-mand behavior and its empirical spedfi-cation relies on commercial fishery prod-uct data not generally noted for a highdegree of reliability.

Third, changes in the market charac-teristics of many of southcentral Alaska'sfishery products over the course of thespecification period raises concerns as toparameter stability in the ex-vessel pricemodel.

Finally, many of southcentral Alaska'scommercial fishers were employed invarious capacities by the cleanup opera-tion mobilized after the oil spill. Rernu-neration for boat rentals and services

exceeded by several orders of magni-tude the ex-vessel revenue lost due to theinability to conduct scheduled commer-cial fisheries, and overall economic per-formance must be viewed in this light.

References:

Assaf, G, B., B. G. Kroetch, and S. C. Mathur.1986. Non-market valuations of accidental

oil spills: A survey of economic and legalprinciples. Marine Resource Economics 2�!:211-237.

Brown, G, M,, Jr, et al. 1983. Assessing the socialcosts of oil spills: The Amoco Cadiz caststudy. U.S. Department of Commerce, Wash-ington, D. C.

Burrows. P., C. Rowley, D, Owens, 1974a. Theeconomics of accidental oil pollutionby tank-ers in coastal waters. Journal of Public Eco-nomics 3:251-268.

Archaeology/Human Impacts: Economic Impacts

. 1974b. Torry Canyon: A case study inaccidental pollution. Scottish Journal of Po-litical Economy 3:237-258.

Cochrane, H. C. 1975. Natural hazards and theirdistributive effects. Institute of BehavioralScience, Boulder, CO.

Cohen, M. J. 1993. Economic impact of anenvironmental accident: A time-seriesanaly-sis of the Erron Valdez oil spill in SouthcentralAlaska. Sociological Spectrum 1335-63.

Couch, S. R. and J, S. Kroii-Smith. 1985. Thechronic technical disaster. Social ScienceQuarterly 166�! 564-575.

Dacy, D. C. and H. Kunreuther. 1969. Theeconomics of natural disasters: implicationsfor federal pohcy. Free Press, New York.

Erickson, K. T. 1976. Everything in its path:Destruction of community in the BuffaloCreek Flood. Simon and Schuster, New York.

Erickson, K.T. 1991. A new species of trouble. In:S.R. Couch and J, S, Kroll-Smith eds.!. Com-munities at Risk: Collective Responses toTechnological Hazards,pp. 11-29. Peter Lang,New York.

Friesema, H. P., J. Caporaso, G. Goldstein, R.Lineberry, and R. McCleary. 1979. After-math: Communities after natural disasters.Sage Publishers, Beverly Hills, CA.

Grigalunas, T.A. 1982. The geographic distribu-tion of the economic costs of large oil spills:A case study of the Amoco Cadiz. In: Orga-nization of Economic Cooperation and De-velopment ed.!. The Cost of Oil Spills, pp.42-56, OECD, Paris.

Grigalunas, T. A., R. C. Anderson, G. M. Brown,Jr.,R. Congar,N.F. Meade,and P.E. Sorensen.1986. Estimating the cost of oil spills: Les-sons from the Amoco Cadiz incident. Ma-rine Resource Economics 2�!:239-262.

Levine, A.G. 1982. Love Canal: Science, politics,and people. Lexington Books, Lexington,MA.

Meade, W. J. and P. E. Sorensen. 1970. Theeconomic cost of the Santa Barbara oil spill.ln: Santa Barbara Oil Spill Symposium, pp.183-226. University of California, Santa Bar-bara, CA.

Nelson,J, P. 1981. Three Mile Island and residen-tial property values: Empirical analysis andpolicy implications. Land Economics57�!:363-372.

Restrepo and Associates. 1982. IXTOC I oil spilleconomic impact study. National TechnicalInformation Service, Springfield, VA.

Rogers, G, 1970. Economic effects of'the earth-quake. In: National Research Council ed.!,The Great Alaska Earthquake of 1964, pp. 58-76. National Academy of Sciences, Washing-ton, D. C.

Rossi, P.H., J. D, Wright, E. Weber-Burdin, and J.Pereira. 1983. Victims of the environment,Plenum Press, New York.

Shrivastava, P. 1987. Bhopal: Anatomy of acrisis. Ballinger, Cambridge, MA.

Wright, J. D., P. H. Rossi, S. R. Wright, and E.Weber-Burdin. 1979. After the cleanup:Long-range effects of natural disasters. SagePublishers, Beverly Hills, CA.

are bifacially ground and have straight

Documents