United States
 Environmental Protection
 Agency
Municipal Environmental Research "^
Laboratory                    "
Cincinnati OH 45268
 Research and Development
EPA-600/S2-82-036  August 1982
 Project Summary
Oil  Spill  Response
Scenarios for Remote
Arctic  Environments
 Robert H. Schulze, William G. Grosskopf, Jack C. Cox, and Lawrence A.
 Schultz
  The  U.S. Environmental Protection
 Agency has recognized that  special
 problems occur during oil spill cleanup
 in remote inland areas in cold climates.
 In Alaska these problems result from
 the harsh arctic and subarctic climate,
 the unusual terrain  features,  and the
 special problems of spills along swift
 rivers. The purpose of this study is to
 describe and analyze these  problems
 using four typical oil spill scenarios.
  The study begins by describing Alaskan
 environmental conditions that affect oil
 spill behavior and oil spill cleanup. This
 description emphasizes  those aspects
 of the environment that have the greatest
 impact on spill behavior and cleanup.
  The study then describes  four spill
 scenarios in remote areas, giving engi-
 neering details of the mechanics of the
 spill movement and the mechanics of
 the cleanup effort. One scenario covers
 a winter blowout of an exploration well
 on the north slope tundra. The next in-
 volves  a Trans-Alaska Pipeline spill in
 the Brooks  Range where the pipeline
 passes under a  sensitive mountain
 stream. The third scenario covers a fuel
 farm spill that contaminates the local
 water  supply of an inland  Alaskan
 village. The fourth scenario involves a
 fuel tank truck spill into a sensitive,
 sport fishing stream.
  The study describes the impact of
 these spills on the environment and the
 circumstances that dictate a particular
 spill response method. Further, it pro-
vides  a  numerical  evaluation of the
effectiveness of the spill response for
the cleanup effort.
  This Project Summary was developed
by EPA's Municipal Environmental Re-
search Laboratory,  Cincinnati, OH, to
announce key findings of the research
project that is fully documented in a
separate report of the same  title (see
Project Report ordering information at
back).

Introduction
  Cold climate oil spill cleanup in remote
inland areas gives rise to a special set of
problems. These problems are particu-
larly significant in the remote areas of
Alaska both  because of the inaccessi-
bility of many areas and because of the
high  level of petroleum development
that is occurring there now. The  prob-
lems of spill response are, in fact, so dif-
ferent from  those that occur in the
lower 48 states that they deserve spe-
cial recognition and analysis. This study
identifies these problems and evaluates
the effectiveness of selected spill re-
sponse methods using four  typical spill
scenarios.
  The problems of  spill response in
Alaska are both complex and diverse.
Because  of this, the analysis of these
problems is divided into seven sections.
The study begins by describing the en-
vironmental  conditions in  Alaska that
affect oil spill behavior and spill cleanup.
To identify these conditions, the natural

-------
environment is described in terms of the
climate, topography, geology, and ecol-
ogy, emphasizing those characteristics
that have the greatest effect on spill im-
pact and on spill behavior and cleanup.
For example, the description of the cli-
mate emphasizes the cold temperatures,
frozen lakes,  rivers, permafrost, and
frozen tundra conditions that affect oil
spill behavior. The description also em-
phasizes the elements  of the northern
environment that are most  sensitive to
the impact of spilled oil such as  the
tundra, bogs, and swift rivers.
  The study of the environment identifies
sets of typical, or  characteristic inland
areas where petroleum development or
handling operations occur  and are,
therefore, most likely to be the sites of
remote spills. Characteristic  environ-
mental  conditions are  then evaluated
along  with  typical spill  situations  to
develop the oil  spill scenarios.  These
scenarios are then  used to develop spill
response methods  showing what could
be  considered  as  the  best available
method to  deal  with each  spill.  These
procedures are analyzed to determine
the  effectiveness  of  the  response
method and the cost effectiveness of
the overall operations.  Finally,  inade-
quacies are identified  in  the spill  re-
sponse procedures, and these are used
to suggest  R&D initiatives  designed to
improve overall performance.
  The paragraphs  that follow  briefly
summarize  the sections  of the study
that develop the background information,
evaluate the spill  response  methods,
and propose R&D  programs to correct
current deficiencies.

Development of the Study
Analysis
  The detailed requirements of the study
were met through a set of systematic
project steps. These steps are described
in the paragraphs that follow.
Alaskan Profiles
  The scenarios for this analysis were
to be selected to illustrate spill response
conditions in four typical remote Alaskan
environments. The Alaskan Profiles were
therefore developed to characterize the
types of environmental and social con-
ditions  that occur in Alaska and the
kinds of problems that would develop
as a result of spills in these areas.
  The information developed in the pro-
files falls into two broad categories, the
natural  environment and the  human
environment. For the purposes of study-
ing the effects of oil spills, the natural
environment includes:
  • Climate
  • Topography
  • Geology
  • Ecology
  Those aspects of the human environ-
ment that are related to oil spill problems
include:
  • Population and Society
  • Economy
  • Transportation Systems
  • Petroleum Development

  The Alaskan Profiles provide a broad
brush picture of the Alaskan environment.
In addition to this broad picture, the pro-
files quickly zero in on those aspects of
the environmental conditions that affect
oil spill  response efforts. For example,
the description of the climate concludes
with an assessment of the impact that
climate  can be expected to have an oil
spill potential, behavior, response, and
site restoration.  Each of the other sec-
tions  covering natural and human envi-
ronmental conditions are also related to
these same aspects of spill impact and
spill response.
  Consider now some examples of the
kinds of information that are developed
in these  environmental  descriptions.
First,  it is shown that oil spill behavior is
profoundly affected by  climatic condi-
tions  such as air temperature, snowfall,
ice, and winds. It is also shown that ex-
treme topographic features present in
Alaska,  such as  steep slopes and swift
rivers, have important effects on oil spill
movement.
  Permafrost is another feature of cold
climates that affects oil spill behavior.
Permafrost affects the permeability of
the soil, the engineering characteristics
of the soil, and the sensitivity of the
terrain to the impact of a spill.
  The physical character of the soil also
determines how the oil is absorbed in
the ground.  A  dry,  porous  soil will
absorb oil quickly; however, a moist or
saturated soil will tend to keep the oil on
the surface. The Profiles show that in
summer Alaska is almost always a wet
area.  In  many areas the soil may be wet
to the point that water is standing or
even  freely flowing  over the surface.
This means that spilled oil will remain
near the surface where it is easier to con-
trol and recover. It also means that the
spill may flow freely over a sloping sur-
face and coat large areas of vegetation.
  The cold climate in Alaska affects the
way in which the spilled oil evaporates,
which  is  one of the most significant
characteristics of the oil spill behavior.
Cold temperatures, ice, and snow all
tend to reduce the rate  of evaporation.
The result is that more of the light ends
remain in spilled crudes,  which makes in
situ burning more successful.
  The Profiles also show how the envi-
ronmental conditions in Alaska affect
spill  response. For example, the study
shows how the climatic conditions of
light, cold,  ice,  snow,  and wind  chill
affect  the methods of spill response
that can  be used and the operation of
the  response equipment.  The study
shows how the topography affects the
spill  response  effort.  Spill  response
operations are also significantly affected
by the accessibility of the areas where
spills may occur and the type of equip-
ment that can  be transported to and
deployed in  these areas.
  Permafrost is a characteristic of arctic/
subarctic  conditions  that  affects the
way in which the spill response effort
can be mounted. Deploying equipment
in remote areas is difficult because off-
road vehicle traffic is not permitted in
many areas during the summer. Arctic
vegetation is fragile and the prints of
vehicles may leave scars that remain for
years.  Because of the permafrost and
arctic vegetation, a massive effort to
recover a spill in an inland area may be
more damaging to the environment than
leaving the spill in place.
  The  preceding  paragraphs provide a
sampling of the kinds of conditions that
occur in Alaska that limit the effective-
ness of inland spill response. The section
covering  the Alaskan Profiles describes
these  and  other conditions in much
greater detail.


Alaskan Inland Oil Spill Behavioi
  To determine the impact of the spillec
oil on the environment  and the way ir
which the spill can  be most effectively
recovered, it is  necessary to describe
the  way  in  which the spilled oil wil
move in the environment.  This  section
therefore,  develops quantitative  rela
tionships  that show how spilled oil be
haves in  remote Alaskan environments
Mathematical relationships are developec
for oil spill behavior for the various char
acteristic Alaskan conditions and com
binations of  these  conditions.  Spil
behavior  is predicted for  oil  on oper
water; oil moving over frozen  ground

-------
snow, and ice; oil in water mixed with
ice; and oil on various soils and ground
water. Evaporation of spilled oil is one
of the most important characteristics in
determining how much oil will be avail-
able for recovery and  what  its condi-
tions  will  be  at the  time of  recovery.
This section also describes spill evapo-
ration characteristics for a wide range
of spill environments. The relationships
developed in describing oil spill behavior
are used in the spill response scenarios
to predict  the movement and  fate of oil
in the environment.

Oil Spill Scenario Selection
  The spill scenarios were selected to
fulfill a number of functions. First, they
illustrate situations that may occur in a
number of typical Alaskan environments.
This was accomplished by  reviewing
sets of typical conditions that may occur
based on environmental conditions de-
scribed in the  Alaskan Profiles. Next,
the scenarios were selected to represent
real spill situations. This is, they must
be typical spill situations that have some
potential to occur in remote Alaskan
areas.  Selecting spill scenarios, then,
involves two  tasks: selecting  typical
environmental situations and selecting
typical spill situations.

Oil Spill Scenario Description
  The study describes four spill scenarios
in remote inland areas of Alaska in engi-
neering terms including both the mechan-
ics of the  spill itself and the mechanics
of the cleanup efforts. The spill scenarios
selected are  of universal  interest be-
cause  of  the  kinds  of spills that are
covered  and  because  of  the  terrain
where they occur.
  The first spill  scenario  describes a
winter blowout of an exploratory well
on the north  slope of Alaska south of
Barrow. The scenario analysis develops
the way in which the  oil could be ex-
pected to  exit the well and the distribu-
tion of the oil on the surrounding terrain.
A complete project plan of the proposed
oil spill cleanup program is presented,
including personnel requirements, logis-
tic requirements, and  a  detailed, line-
item listing of the expected cost of the
response effort.
  The second spill scenario concerns a
Trans-Alaska Pipeline spill on the south
slope of the  Brooks Range where the
pipeline passes under a swift mountain
stream. The spill occurs in late winter
before breakup, and the oil is postulated
to move under the ice of the river along
the riverbanks until breakup  occurs  in
May.  The oil is then released, flowing
with the breakup flood waters until the
waters recede, and a large part of the oil
is  left stranded along the river banks.
This scenario provides the engineering
details of how the oil is expected to
move on the water table along the river-
bank before breakup, and how it moves
downstream and is deposited along the
shoreline. As  in the  first  scenario,  a
complete spill response plan is presented
and costed  as  a  part  of the scenario
description.
  The third spill  scenario traces  the
path of an accumulation of petroleum
products on the water table under a fuel
farm at an inland Alaskan  village. The
spill response procedure shows how the
oil is  recovered from the  water table
where it has been a contaminant to the
village water supply. As before, spill
response methods and costs are provided
in  detail.
  The last spill scenario concerns a fuel
tank truck that overturns on the haul
road north of Fairbanks and spills a re-
fined product into a sensitive sport fish-
ing stream. This scenario is typical of a
great  many spills that occur in Alaska,
and it illustrates the problems involved
in  responding to a small spill in a remote
area.

Spill Response  Techniques and
Devices
  Spill response techniques and devices
were reviewed in terms of their applica-
tion  to  the  selected  spill  scenarios.
Because of the large number of devices
and methods of spill response that are
currently available, the evaluation and
description was limited to the technology
that could be applied to  the scenarios
selected for analysis. The  devices se-
lected are described in detail  giving full
engineering specifications.  In addition,
each device is evaluated in terms of its
expected performance  in  the spill
scenarios.

Spill Response Methods and
Effectiveness Analysis
  This section evaluates  the  effective-
ness  of the spill response methods
described in each of the scenarios. This
analysis  is the most important part of
the study because it provides important
insights into the spill response process
and the  use  of funds to support this
process.
  The effectiveness analysis  begins by
defining and calculating spill impact. For
this study, spill impact is assumed to be
related to the volume of oil spilled, the
area covered by the spill, and the sensi-
tivity of the environment to the spilled
oil. The spill volume and area are mea-
sured using conventional metric units.
The  sensitivity of the  environment to
the spill is represented by a set of weigh-
ing factors that are given values of 0 to
4, with 0 representing the case of no
spill impact and 4 representing the case
of maximum spill impact. Mathematically
this can be shown as:

              Impact  =
 2(time) Volume  x  Area  x Sensitivity

  Next it was necessary to define spill
response effectiveness. Basically, the
concept of  response effectiveness  in-
volves comparing the spill impact with a
response effort to the spill impact if no
response effort  had been  expended.
Taken as a  fraction, the (Impact with
response)/(lmpact with no response)
become smaller as the response effort
increases, which is contrary to what
intuition would expect  of an effective-
ness measurement. To bring logic and
intuition into congruence, the fraction is
subtracted from 1 so that higher values
of response effectiveness  result  from
higher levels of response effort. The ex-
pression  for response  effectiveness,
therefore, becomes:

      Response  effectiveness =
    1 _   Impact with response
         Impact  with no response

  Using this relationship, spill response
effectiveness was computed for three
response  levels:  a minimum level of
response, a  moderate level of response,
and a high level of response, for each of
the  four scenarios developed in the
study. Determining response effective-
ness for each scenario required many
computations  because a separate cal-
culation had to be made for each type of
environment (rivers,  shore, wetland,
and so forth) that occurred in the spill
area. The calculations were also made
for  each of the three spill response
levels. The results of these calculations
were dependent  on the spill response
methods that were selected and the suc-
cess with which  these methods  were
used. The results of each effectiveness
calculation were also determined by the
practical response requirements of the
individual spill situations. For example,

-------
the spill response methods were selected
and the spill response levels were deter-
mined  in consultation with the spill re-
sponse specialists in Alaska. As a result,
the minimum level of response selected
was the minimum practical  level  that
would  be acceptable to local  respon-
sible officials. This level of response is
always likely to be fairly high in terms of
the amount of oil recovered;  therefore,
the minimum level of response in each
case shows  a relatively high level  of
effectiveness. This situation tends  to
show that the minimum level of response
is the preferred course of action in every
case. This result is reinforced in the cost
effectiveness analysis, which is described
next.

Cost Effectiveness of Spill
Response
  In this study, the cost effectiveness
of spill response is determined using the
expression:

      Cost effectiveness =
              Normalized cost
           Response effectiveness

  The normalized cost shown  in  the
formula is the total cost of each spill
response level divided by the total cost
of the entire operation. Since each re-
sponse level describes the actions that
are taken  in addition to the response
level  that preceded it, the costs asso-
ciated with the  response levels  are
cumulative.
  This expression for cost effectiveness
can be visualized conceptually as a rela-
tive cost per unit level of effectiveness.
Ideally, if the cost effectiveness  is plot-
ted against the percent of product re-
covered or levels of spill response, the
cost effective solution would be the min-
imum point on the curve. In practice,
the minimum point on the curve often
corresponds to the lowest acceptable
level  of response, and all other points
are higher. This result can be expected
when one  considers that the  lowest
level  of response generally includes the
heaviest accumulations of oil close to
the spill site. These accumulations can
be cleaned up at the lowest cost per
unit volume recovered,  and therefore
this  recovery is cost effective.  Higher
levels of  response generally   involve
exerting a higher level of effort cleaning
up a smaller amount of oil farther from a
spill site, and therefore these operations
are often less cost effective.
  Consider for a moment the results of
a computation of spill response effec-
tiveness and cost  effectiveness.  The
specific example happens to be from
the pipeline spill scenario, but the details
of the scenario are not important to this
discussion. The  result,  in this case, is
also typical of the analysis that was made
for  each of the  other three scenarios.
Recall that the response effectiveness
and cost effectiveness computations
are made for  three levels of spill re-
sponse, where Response Level I is  the
minimum acceptable level of response.
Response Level II is a moderate level of
response, and Response III is a relatively
high level of response.
  Figure 1  shows  the  results of  the
computation for response effectiveness
for  the  pipeline  scenario. The spill re-
sponse effectiveness shown in Figure 1
is not particularly high, but it can  be
considered to be realistic. Note that the
curve is relatively flat; that is. Response
Level I is high as compared with Levels
II and III. This is a condition that is likely
to be typical of many inland spill situa-
tions. In this scenario, there is a heavy
accumulation of oil at the spill site. A
high percentage  of this  oil must  be
cleaned up to repair the pipeline. Recov-
ering this large volume of oil has a sig-
nificant  effect on the computation  of
spill response effectiveness. In addition,
the next largest volume  of the spill,  the
oil that  coats the shoreline, is also re-
covered in Response  Level I as part of
the minimum acceptable level of response.
This also serves to increase the effec-
tiveness of Response Level I.
  The response  effectiveness for  Re-
sponse  Level II  is only slightly  higher
than that for Response  Level I. This is
because the effort to recover visible oil
downstream only results in a small vol-
ume of  oil being collected. Also, recall
that the response levels  are cumulative;
that is. Response Level II is performed in
addition to Level I  and  so forth. As a
result, the effectiveness  of  Response
Level II  is high, but only slightly higher
than Response Level I.
  Finally,  Figure 1  shows that   the
effectiveness of Response  Level III is
lower than that  for Response Level II.
Although the drop is small, the reasons
for  this  drop are important. Recall that
the spill  impact is the product of the vol-
ume spilled, the area covered, and  the
sensitivity  of the environment to  the
spilled  oil.  Generally, for each  higher
level of  response the volume spilled will
decrease, the area covered will decrease,
and therefore the spill response effec-
tiveness will increase. This is true except
for the factor of sensitivity. The spill sen-
sitivity is expected to decrease with each
response level, but it is not constrained
to do this. In fact, intensive efforts tc
recover the last small amount of oil for an
inland spill are likely to be more destruc-
tive to the environment than leaving the
oil in place. This is what happened in Re-
sponse Level III. The spill sensitivities foi
the various response levels tend to remain
the same,  and even increase in some
cases for the higher  levels of response,
When the sensitivities remain the same,
it may  be that the  oil spill impact is
replaced by an impact caused by the re-
sponse crews and heavy equipment. Ir
cases  in  which  intensive efforts  art
mounted on land to collect small amounts
of oil, the sensitivity  to the response ef
fort may be  considerably greater thar
the sensitivity to the spill. This is whai
happened in Response Level III of  the
pipeline scenario. The  fact that the re
sponse effectiveness for Response Leve
III is only slightly less than Level II can be
accounted for by noting that the sensi
tivities used in computations are an aver
age of a great many  sensitivity factors
and therefore a few  higher sensitivitie;
for Response Level III do not change the
result much. In addition,  the range o
values used for the sensitivities is small
0 to 4, so that the difference that result;
from using a  higher sensitivity is alsc
small. The important  point to note is this
the sensitivity of the environment is
important to planning a spill response ef
fort; and if an intensive effort is launchec
on a highly sensitive inland environment
the impact of the response effort itsel
may be greater than the impact of th<
spilled oil would be if  it were left in place
  Consider now the  cost  effectivenes!
for the sample pipeline  spill scenario
Cost  effectiveness  is defined as  th<
quotient of normalized cost and responst
effectiveness.  The cost  effectivenesi
for the three pipeline spill responst
levels is shown in Figure 2. Since th<
curve has no minimum, Response Leve
I, which gives the lowest value, is thi
most cost-effective  solution. Responst
Level II  is only slightly more costly  ant
also  only slightly more  effective.  Re
sponse Level II represents an essentially
neutral position; it may be worth the ef
fort, but only  marginally. If the effec
tiveness of  Response Level  II  wen
slightly better, it would  be the  mos
desirable solution. Response Level III i
the least desirable solution. The curvi

-------
 shows that it has a  high  incremental
 cost for the additional improvement in
 effectiveness.
   This sample analysis shows that it is
 possible to quantify spill response effec-
 tiveness by determining the  extent of
 the  spill, the sensitivity of the environ-
                                       ment to the spill, and the changes that
                                       various spill response  actions cause in
                                       spill coverage and environmental sensi-
                                       tivity. The advantages of this system
                                       are that both spill volume and area oper-
                                       ate together in determining spill impact.
                                       Also, either of these terms  alone can
    0.5
CO
I
I
w
Cj
I

1
I  0.2
Cb
0.4
0.3
    0.1
                     Response effectiveness = 1 -
                                                Impact with response

                                              Impact with no response
                      I
                                II
                                                  III
                            Spill response level

Figure  1.   Spill response effectiveness vs. spill response level.
    2.5
.1
    2.0
    1.5
                      I              II             III

                           Spill response level

Figure 2.   Cost effectiveness vs. spill response level.
 result in a high level of spill impact. For
 example, a relatively  low volume spill
 that coats a wide area is shown by this
 computation  to  have  a  high level 'of
 impact. Another advantage of the sys-
 tem is that the impact of the spill is a
 function of the spill location; more sen-
 sitive environmental areas are shown to
 suffer a higher level of impact. Also, the
 environmental sensitivity is  not con-
 strained to move with  the spill response
 effort. Thus,  a  massive spill response
 effort on sensitive terrain  can be shown
 to have a greater impact on the environ-
 ment than the spill itself. The volume of
 oil collected is, therefore, not the only
 measure of spill response effectiveness.
  If the cost  of  various levels of spill
 response  can  be determined, then the
 effectiveness  computation can be used
 with the  cost data to show the cost
 effectiveness  of  each  level  of spill
 response.  This  analysis  will  identify
 intensive response efforts that result in
 a very low level of spill  recovery or other
 benefit to the  environment.
  The analysis of spill response  effec-
 tiveness  and  cost  effectiveness has
 many benefits. The system produces a
 numerical result that can provide impor-
 tant  insights  into a  highly complex
 problem. Specifically, this analysis helps
 to identify  response  alternatives that
 are effective as well as those that are
 questionable or even counterproductive.
 Since there are few ways to measure
 the real success or failure of a spill re-
 sponse action, a system that separates
 effective actions from ineffective actions
 is a valuable analytical tool.

Conclusions
  The  major conclusions  that can be
drawn  from this study  are as follows:
  • It is possible to develop a series of
    oil spill scenarios for remote inland
    Alaskan environments that illustrate
    a broad range of physical conditions
    and spill situations.
  • It  is  possible  to characterize spill
    response  effectiveness  quantita-
    tively and  to evaluate the effective-
    ness and the cost  effectiveness of
    specific spill response procedures.
    Specifically, it is possible to evaluate
    spill response efforts in terms of:
    —  Spill impact
    —  Reduction  of spill impact result-
       ing from various response pro-
       cedures
    —  The impact of the spill response
       procedure  itself

-------
  — The cost effectiveness of the re-
    sponse procedure including the
    margin of response effectiveness
    developed  compared with the
    cost of producing that effect

• The analysis of spill response effec-
  tiveness and cost effectiveness has
  many benefits. The system  pro-
  duces a numerical result that can
  provide  important insights  into a
  highly complex problem. For exam-
  ple, the analysis of the spill  impact
  and response effectiveness in the
  four selected scenarios shows that:
  — A relatively high level of spill re-
    sponse effectiveness results from
    the minimum acceptable level of
    response,  and only  minor im-
    provements result from  higher
    levels of response.
  — The minimum level of response
    is the most cost effective, and
    cost increases much more rapidly
    than  effectiveness   as  higher
    levels of response are reached.
  — A final massive spill response ef-
    fort launched to recover a small
    amount of  oil is often  neither
    cost  effective  nor practical.  In
    fact, the final effort may be more
    damaging  to  the environment
    than leaving the oil in place.

• Equipment suitable  for performing
  some routine spill response proce-
  dures in remote inland environments
  may not be available. Development
  of the necessary pieces of response
  equipment has the potential for in-
  creasing the effectiveness and re-
  ducing the cost of spill  cleanup in
  remote areas.

• Wide differences in opinion exist,
  even among qualified  experts  in
  Alaska,  about the kinds of actions
  that  should  be  taken for  site
  restoration.  Important  questions
  that do not appear to be resolved
  include  the treatment  of  heavily
  oiled beaches, the disposal of oiled
  debris and vegetation, and the kinds
  of  policies that should  be adopted
  to minimize the impact on wetlands.
  Decisions concerning the action to
  take for site restoration must  be
  made  on  a  case-by-case basis;
  however,  apparently no generally
  accepted  criteria  exist for  making
  these decisions.  Acceptable meth-
  ods for site restoration need to  be
  identified and described.
General Recommendations
  The  recommendations  that  follow
from this analysis include general rec-
ommendations for the management of
oil spill response programs and specific
recommendations for research and de-
velopment initiatives. The general rec-
ommendations are:

   • This study  has shown that it  is
    possible to evaluate spill response
    effectiveness and  the cost effec-
    tiveness of  specific spill response
    procedures  quantitatively.  It  is,
    therefore, recommended that this
    approach  be  applied to a larger,
    diverse  group of  spill  response
    problems that require analysis and
    quantitative solutions.

   • In each of the four spill scenarios
    evaluated in this analysis, a rela-
    tively high level of  response effec-
    tiveness was  achieved with the
    minimum acceptable level  of  re-
    sponse, and only minor increases in
    effectiveness occurred at  higher
    levels   of response.  Further, the
    minimum level of response was the
    cost effective result in  each case,
    and cost  increased much  more
    rapidly than effectiveness as higher
    levels of response were reached. It
    is recommended that a top level
    policy   analysis  be  conducted to
    establish what could be considered
    a  practical  level of  response  for
    typical  spill situations. This level of
    response  would not  require  an
    unusually costly response  proce-
    dure that may only achieve a rela-
    tively  low  level  of effectiveness
    and would not encourage intensive
    response efforts that could actually
    be damaging to the environment.

   • This study determined that there is
    a  general lack of agreement as to
    the actions that should be taken for
    site restoration after a spill response
    effort  has been completed. It is,
    therefore, recommended that policy
    guidelines  be developed to assist
    field personnel in  determining the
    types and levels of site restoration
    that should be used.


 Research  and Development
 Initiatives
   Specific recommendations for research
 and development initiatives are presented
 here.
Cleaning Oiled Snow
  Cleaning or processing oiled snow is a
routine operation in Alaska, but there is
no  standard,  easy way  of  doing  it.
Sometimes the snow  and oil  are sepa-
rated by melting the snow in an air berm
and skimming the oil off the top of the
water. Although effective, this proce-
dure is relatively slow and labor inten-
sive, and therefore expensive. A single
device is needed to perform all the steps
of the process that are presently per-
formed by the response crew at the air
berm. It is recommended that a project
be established to develop a single, port-
able device that could be used to sepa-
rate snow and oil  with  a  minimum
requirement for outside energy sources
and labor.

Cleaning Sand and Gravel
  The  problem of cleaning sand and
gravel  occurs  repeatedly  in  oil spill
situations.  In this study, oiled gravel in
one  scenario  was cleaned in a snow
melter and returned to its original site.
This procedure is probably effective in
cleaning but not efficient in the use of
men and materials. In  another scenario,
heavily oiled  sand was removed from
the spill site for disposal. This procedure
is also not efficient and is likely to be
damaging  to  the  environment. It  is,
therefore,  recommended that a portable
device be  developed that is capable of
cleaning oiled sand on site so the removal
and disposal is not required.

On-Site  Burning
  Many observers believe that disposal
by burning the spilled product in place is
a good response method in some cases.
The problem is to  identify  the cases in
which this method is best. Burning with
little permanent damage to the environ-
ment is possible in certain seasons of
the year, with certain kinds of vegeta-
tion, terrain  conditions,  and  moisture
conditions.  Additional  information  is
needed to establish the criteria to deter-
mine when burning is the best course of
action. It is recommended that a basic
research program be  initiated to deter-
mine the  impact  of  burning in a wide
variety of conditions. These tests should
be directed toward developing a set of
decision-making criteria for use in  the
field to determine when on-site burning
should be used.

Fast Current Containment Booms
  The swift, debris-laden rivers of Alaska
make deployment of containment booms

-------
a problem. In many cases, the use of
containment booms would be desirable,
but they are not used because existing
booms are not effective in these condi-
tions. The following program is recom-
mended to correct this deficiency:
  • Test existing booms to determine
    which is the best suited for use in
    high current, shallow stream situa-
    tions.
  • Use a hydraulic model to determine
    the best  ways to deploy booms in
    these situations.
  • If none of the booms available are
    suitable for this application, design
    a new boom.

Fast Current Recovery Device
  When  oil  is  successfully contained
behind a boom in a swift river in Alaska,
there remains a problem in  recovering
the slick. The  ideal  device would be
effective in recovering a thin slick in  a
fast  current.  Although many recovery
devices are available, they do not meet
these requirements. It is recommended
that existing devices be tested to deter-
mine whether any are  suitable for this
application. If a suitable device is  not
identified, it  is  recommended that  a
new device be developed.

Portable Flaring Burner
  Disposal of the recovered product is
frequently a problem in Alaska.  In re-
mote inland areas, most of the standard
methods of  disposal are not possible.
Further, transporting the recovered pro-
duct to a site for clean disposal  is also a
problem. A portable device is needed to
provide an environmentally safe method
of disposal in remote inland areas. To
meet this requirement, it is recommend-
ed that a portable, open flame burner be
developed that could be transported to
remote areas and used to provide a safe,
clean burn of the recovered products.

Surface Detection of Product on
the  Water  Table
  The area of the spill on a water table
cannot generally be determined from the
surface. The only way to map the area
of the spill is to drill holes in the ground
until spill boundaries are discovered.
This method is slow,  expensive, and
destructive to the terrain. A device that
could sense  the presence  of under-
ground oil from the surface would be
invaluable. It is recommended, there-
fore, that a device capable of detecting
oil on the water table from the surface
be identified or developed.
Site Restoration
  Wide disagreement exists about the
kinds of actions that should be taken for
Alaskan site restoration. To provide the
basis for making decisions concerning
site restoration, it is recommended that
a research project be initiated to deter-
mine the best ways of responding to
some of the following situations:

  • Oiled beaches and shorelines
  • Oiled debris and vegetation
  • Oiled wetlands
  • Oiled soils


  Decisions concerning the action to
take for site restoration must be  made
on a case-by-case basis; however, no
criteria for  making these decisions ap-
pear to be generally accepted. The pro-
posed  research project should identify
and  describe acceptable methods for
site restoration. These methods should
be accepted and approved by the re-
sponsible state and Federal authorities.
The accepted methods of site restora-
tion could then be used for on-site deci-
sion making by spill  response crews.
Identifying Cost Effective Levels
of Spill Response
  The analysis of scenarios investigated
in this study shows that the final inten-
sive effort to  recover  the last small
amount of  oil  is not  generally cost
effective.  In some cases this effort may
even be damaging to the environment.
It is, therefore,  recommended that an
R&D project be  established to develop
criteria  for  making  the  decision  on
recovering the last traces of a spill. It is
suggested that this project review spill
response records to obtain field data. In
addition, it is recommended that addi-
tional data be developed using a cost
effectiveness simulation of typical spill
problems.  This simulation could be used
to show the expected results of various
response actions and the effectiveness
of these actions. The results of the anal-
ysis could then be  used  to develop
decision-making  criteria  for  use  by
responsible spill control officials.
  The  full report  was submitted  in
fulfillment of Contract No. 68-03-2799
by  ARCTEC,  Incorporated, under  the
sponsorship of the U.S. Environmental
Protection Agency.
  Robert H. Schulze. William G. Grosskopf. Jack C. Cox, and Lawrence A. Schultz
    are with ARCTEC, Incorporated, Columbia, MD 21045.
  Leo T. McCarthy, Jr., is the EPA Project Officer (see below).
  The complete report, entitled "Oil Spill Response Scenarios for Remote Arctic
    Environments," (Order No. PB 82-231  416; Cost: $30.0O, subject to change)
    will be available only from:
          National Technical Information Service
          5285 Port Royal Road
          Springfield, VA 22161
          Telephone: 703-487-4650
  The EPA Project Officer can be contacted at:
          Oil and Hazardous Materials Spills Branch
          Municipal Environmental Research Laboratory-Cincinnati
          U.S.  Environmental Protection Agency
          Edison, NJ 08837
                                                                             •&U. S. GOVERNMENT PRINTING OFFICE: 1982/559-092/0465

-------
United. States                          Center for Environmental Research                                      Fees Paid
Environmental Protection                Information                                                          Environmental
Agency                               Cincinnati OH 45268                                                  Protection
                                                                                                         Agency
                                                                                                         EPA 335
Official Business
Penalty for Private Use $300                                                    PS    0000329

RETURN POSTAGE GUARANTEED

-------