Socio-Economic Planning Sciences 51 (2015) 34e45

Contents lists available at ScienceDirect

Socio-Economic Planning Sciences

journal homepage: www.elsevier.com/locate/seps

Adaptation analysis for environmental change in coastal communities

Hooman Mosto Camare a, Daniel E. Lane b, *
a
b

Morneau Sheppel Toronto, ON, Canada

Telfer School of Management, 55 Laurier Ave E, University of Ottawa, Ottawa, ON K1N 6N5, Canada

a r t i c l e i n f o

a b s t r a c t

Article history:
Available online 24 June 2015

Small islands and coastal communities around the world are particularly vulnerable to climate change
impacts, mainly from storm surge attributed to more frequent and severe coastal storms, and mounting
sea-level rise. Coastal hazards including inundation, salinisation of the water supply, and land erosion all
threaten vital infrastructure that support coastal communities. This research, part of the International
Community-University Research Alliance (ICURA) C-Change project Managing Adaptation to Environmental Change in Coastal Communities: Canada and the Caribbean, develops and applies a multicriteria
decision evaluation and support system for evaluating adaptation options for coastal communities. The
paper estimates vulnerability, resilience, and adaptive capacity measures associated with adaptation
strategies in coastal communities with respect to their environmental, economic, social, and cultural
dimensions. Results are determined using a multi-participant formulation of the Analytical Hierarchy
Process (AHP) for identifying multicriteria decisions as adaptation strategies in a specic coastal context.
The application of the framework is conducted for the coastal community of Little Anse on Isle Madame,
Nova Scotia. Specically, the state of the Little Anse breakwater is analysed and adaptation options for
protecting, accommodating, and retreating are presented and evaluated in the face of predicted storm
scenarios. The results indicate that, in the case of Little Anse, the strategic decision to protect the
community by a new breakwater arm provides preferred measures for resilience and adaptive capacity.
2015 Elsevier Ltd. All rights reserved.

Keywords:
Coastal community vulnerability
Climate change adaptation
Resilience
Adaptive capacity
Analytical hierarchy process
Group decision making

1. Introduction
Coastal communities around the globe are experiencing
changing climate impacts from increased frequency and severity of
coastal storms, e.g., Typhoon Haiyan, Hurricane Sandy, Hurricane
Katrina, rising sea levels threatening island states, e.g., the
Maldives, permafrost thawing in polar regions, and widespread
coastal erosion [7,14,15]. Canada has the longest coastline in the
world and the Atlantic region of Canada is subject to impacts from a
wide range of interannual hurricanes, winter cyclonic storms, and
ooding [33]. The hurricane season of 2010 was especially harsh
with frequent storm incidents in the region [8,9]. The predictable
and mounting impacts of changing coastal environments require
further investigation of coastal vulnerability and adaptive capacity
and the preparation of communities in the Canadian coastal zone to
face of impacts of climate change [1,4,5].

* Corresponding author. Tel.: 1 613 562 5800x4795.

E-mail addresses: Hooman.Mosto@gmail.com (H. Mosto Camare), dlane@
uOttawa.ca (D.E. Lane).
http://dx.doi.org/10.1016/j.seps.2015.06.003
0038-0121/ 2015 Elsevier Ltd. All rights reserved.

The purpose of this paper is to present an evaluation framework

for coastal adaptation decisions applied in a dened coastal community context. The evaluative process proceeds from the development of a prole of the coastal community that includes its
vulnerability to environmental change, as well as the community's
identied priorities. A multicriteria decision framework is formulated that accounts for the expected impacts of severe coastal
storms to rank alternative adaptation strategies. The results of the
strategy rankings provide support for coastal communities' efforts
to adapt strategically to the pending environmental changes. This
research is part of the International Community-University
Research Alliance (ICURA) C-Change project Managing Adaptation to Environmental Change in Coastal Communities: Canada and
the Caribbean [3,19]. C-Change seeks to raise awareness of coastal
climate vulnerability and promote improved adaptive capacity and
coastal community preparedness in selected communities in Canada and the Caribbean region. Evaluation of adaptation decisions
options and the development of measures for vulnerability and
adaptive capacity are applied in this paper to the C-Change coastal
community of Little Anse, Richmond County, Cape Breton, Nova
Scotia, Canada.

H. Mosto Camare, D.E. Lane / Socio-Economic Planning Sciences 51 (2015) 34e45

2. Methods
The evaluation of coastal communities' adaptation strategies to
environmental change is a problem characterized by multiple
community criteria and the participation of multiple communitybased agents. This problem is modelled for decision support using a multicriteria, multi-participant framework that captures
community vulnerabilities (measured by storm impacts), and the
suite of community priorities embraced by contributing community leaders and decision makers.
The research process comprises the following four steps:
I) Community Prole and Problem Formulation. Prole the resources of the coastal community with respect to the community's pillars of sustainability criteria identied as: (1)
Economic; (2) Environmental; (3) Cultural; and (4) Social
dimensions. The pillars of sustainability were formally
adopted by municipalities throughout Canada in the development of the 2009e2010 Integrated Community Sustainability Plans (ICSPs) [21].
II) Community Vulnerability and Storm Impacts. Collect historical
data on severe storms impacts and damages to determine
coastal community vulnerability on related indicators of the
multiple criteria of the community prole.
III) Community Adaptation Data and Problem Analysis. Prepare
the multicriteria problem formulation for analysis; obtain
feedback from the representative community group participants on criteria priorities; design the severe storm scenarios
and associated data analyses; and prepare the adaptation
strategy alternatives for evaluation.
IV) Community Evaluation of Adaptation Strategies. For each of
the storm scenarios, compare the contribution of the
alternative adaptation strategies as a function of the expected storm impacts and the community priorities; rank
the overall effectiveness of the alternatives; compute
community vulnerability, resilience, and adaptive capacity
measures.
I. Community Prole. Four dimensions, dened as pillars of
community sustainability are dened to categorize the status of
coastal communities' assets position through the associated values
of related indicators. The indicators, i are resource asset measures
expressed in monetary units (with the exception of numbers of
demographic groups, i.e., seniors and children, at risk) and representing community characteristics of interest. Selected indicators
by pillar are denoted below:
1) Economic pillar e EC(i), i 1, nec; e.g., value of built environment (residential, commercial/industrial), public works;
2) Environmental pillar e EN(i), i 1, nen; e.g., value of land use,
residential, commercial/industrial, green space;
3) Cultural pillar e C(i), i 1, nc; e.g., value of community resources, historical sites, faith-based infrastructure (church
grounds), schools; and
4) Social pillar e S(i), i 1, ns; e.g., value of labour income, identication of vulnerable demographic groups (seniors, children).
These four community pillars and their associated indicators are
the key community dimensions by which the community denes
itself [27,34].
II. Community Vulnerability and Storm Impacts. Vulnerability is
dened by local community threats from rising seas and more
frequent and severe storms. Historical storms are described to
provide the local coastal community context. Two primary sources

35

are used for extracting historical storms incidences along with their
impacts:
(i) National Oceanic and Atmospheric Administration (NOAA)'s
HURDAT2, or best track database [17,18]; and
(ii) Environment Canada's Climatology of Hurricanes for Eastern
Canada [9].
The HURDAT2 database reports on historical storms relevant to
selected coastal communities, and provides information via an
online tool to query the HURDAT2 database. Storm and wind speeds
are measured in kilometres per hour and the pressure is measured
in millibars. The ranges of these indicators are the basis for categorizing storm scenarios. Storm Scenarios are sorted based on the
intensity of observed storms.
Storm surge is an unusual rise of water generated by a storm,
over and above the predicted astronomical tides. Approximately
1500 people lost their lives as the result of Hurricane Katrina and
majority of those casualties, directly or indirectly, were the result of
storm surge [24]. The combination of surge with normal tides
creates hurricane storm tide, which has the potential to increase
the mean water level to critical heights causing signicant damage
to critical infrastructure and people [24]. The Canadian Tides and
Water Levels Data Archive [11] provides maximum water levels at
the time of storms from marine observation stations aligned with
Canadian coastal communities. Fig. 1 illustrates storm surge and
storm tide.
Data on historical coastal storms' tracks, severity, and water
level are the drivers for coastal community impacts. GIS mapping
is used in C-Change coastal community storm impact analysis [19]
to analyse the community assets impact by storms of different
severity. Limited data exist on the impacts of storms to coastal
community assets, homes, infrastructure and businesses. Much of
these data are reported in the local media [32]. However, given
that there are no formal community-by-community reporting
mechanisms for natural storm damage, then the impacts of storms
are estimated based on aggregate values dependent on storm
severity.
III. Community Adaptation Data and Problem Analysis. The Analytic Hierarchy Process (AHP) is adopted to formulate and analyse
the multicriteria problem [31]. The AHP problem hierarchy for
Adaptation Decision Support (top level of hierarchy) is constructed from the Community Prole and the 4 pillars of sustainability (second level of criteria in the hierarchy). The third level of
the hierarchy presents the indicators associated with each pillar.
Fig. 2 presents an illustration of the coastal community adaptation
problem hierarchy.
Coastal communities are represented in the decision making
process by stakeholders with different perspectives on the importance of the dimensions of the community prole. These
stakeholder-participants are represented as follows:
1) Community group members: these are the representatives of the
community, including community social clubs such as Community Seniors, Youth members, the Knights of Columbus, the
Ladies Auxiliary, and faith-based church groups.
2) Local Government: representatives of local government, namely,
the Regional Municipality, responsible for community services
and governance, and governance liaison (municipal, provincial,
and federal levels of government).
3) Business/Industry: delegates of the local coastal industries
including sheries industry managers, local contractors,
restaurant and hotel owners, hardware and grocery store
owners and operators.

36

H. Mosto Camare, D.E. Lane / Socio-Economic Planning Sciences 51 (2015) 34e45

Fig. 1. Storm surge and storm tide.

Source: [24]

Fig. 2. Illustration of the coastal community adaptation problem hierarchy.

4) Community Professionals: Community professionals including

but not limited to lawyers, nurses, doctors and hospice workers,
engineers and academics.
Each community group holds a unique perspective regarding
the relative importance of the criteria of the coastal community
prole as described in the problem hierarchy (Fig. 2 above). For the
application presented below, community representatives of the 4
participant groups completed questionnaires in workshops that
provided pairwise comparisons of Fig. 2 hierarchy elements [19].
Pairwise comparisons are assigned using the nine-point scale
illustrated in the AHP software [10]. The pairwise comparisons data
are used to calculate groupwise tradeoffs between the criteria of
the hierarchy resulting in participants' priority weights illustrated
in Table 1 for the 4 pillars of the community prole.
As illustrated in Table 1, for the case study, Local Government
representatives' pairwise comparisons results in the maximum
weight to the Community Social pillar (0.39) while the Business
sector gives maximum weight to the Economic pillar of the

Table 1
Participant priorities of the community prole pillars.
Participant group

Community prole pillars

Economic

Environmental

Cultural

Social

Community groups
Local government
Business/Industry
Professionals
Combined (Fig. 7)

0.25
0.22
0.45
0.25
0.28

0.24
0.21
0.16
0.30
0.22

0.25
0.19
0.16
0.18
0.20

0.27
0.39
0.24
0.28
0.29

The bold text denotes the maximum for each column (dimension). These uniquely
correspond to the preferences of the four participant groups.

community (0.45). Community groups have a more balanced

perspective toward the Community Prole pillars with each pillar
receiving an approximately equal weighting. The Professionals
group provides maximum weight to the Environment pillar (0.30).
Group AHP modelling combines the weights of each stakeholder
into a single weighted hierarchy set by pre-weighting the individual group inputs through the calculation of geometric means of the
pairwise comparison in the hierarchy [31]. (The combined participant groups weights of Table 1 are the geometric means also presented in the AHP output of Fig. 7.)
In a similar manner, for each pillar, pairwise comparisons of the
suite of indicators are collected to determine priority weights for
the indicators. In AHP, these pairwise comparisons determine the
relative importance of the indicators in the form of priority weights
(see also Fig. 7 for the case study). To complete the AHP problem
data requirements, values for the suite of indicators for each
community prole pillar are required. These data values are measures of the indicator status under the different storm scenarios
and for each adaptation alternative. For a given storm scenario and
adaptation strategy, indicator values, as illustrated in Table 2 below
for the case study, are transformed to utility values (in the AHP Data
Grid of the Expert Choice software to reect the preferences of decision makers), and then weighted with the indicator weights and
the pillar weights (Fig. 7) to result in an overall weight for
comparative ranking of strategies.
The following variables, sets, and functions are dened that
summarize the general AHP model formulation and results for this
problem:
(i) Community Status Quo e No Storm Scenario, SQ0. The concept
of Community Status Quo is quantied by the multidimensional
valuation of the community of interest in terms of each of the indicators corresponding to the four pillars of the Community Prole,

H. Mosto Camare, D.E. Lane / Socio-Economic Planning Sciences 51 (2015) 34e45

37

Table 2
Estimated assets value (TA) for Little Anse.a
Community prole pillar

Indicator,SQ0(i)

Number/Length/Area

Estimated (CDN2010) asset value ($)

Economic

Built Envt: houses

Built Envt: out-buildings
Public works: roads
Public works: wharf
Public works: wells
Breakwater (adaptation element)
Residential land
Water bodies/lake
Greenspace/trees
Community centre
Church grounds
Income
People > 60 years
Children < 14 years
Total Assets, TA

100
8
2093 m
30 m
45
1
155,411 sq m
19,221 sq m
200,000 sq m
1
1
100 households
67
33

$3,175,000
$64,000
$2,272,998
$25,000
$427,500
$1,350,000
$1,864,932
$19,221
$54, 800
$242,700
$800,000
$3,497,000
e
e
$13,738,351

Environmental land use

Cultural
Social

a
Asset valuation measures summarized here are detailed for each item of the Community Prole in Ref. [23]. For further information on these valuations contact the
authors.

namely, Economic (EC), Environmental (EN), Cultural (C), and Social

(S). The Community Status Quo is a value vector assigned from the
community assets for each pillar indicator, e.g., residential housing
structures are an economic asset indicator as part of the EC pillar.
The total value of all residential housing in the community, independent of storms, is included as an element of the total asset value
vector for the Community Status Quo position, e.g., in the community of Little Anse, the 100 residential buildings are valued at
$3.175CDN (2010) based on local real estate valuations (see also
Table 2 below). Let SQ0 be the vector constructed from the four
vectors (pillars), i.e., EC0 [EC(1), , EC(nec)], EN0, C0, S0 that denote
the initial status quo values of each pillar's designated nec, nen, nc,
and ns indicators. The vector of Community Status Quo asset
values for all indicators, at initial time, t 0, and prior to any
severe storm is denoted by the N-dimensional vector, where
N nec nen nc ns:

i
h
SQ 0 EC 0 ; EN 0 ; C 0 ; S 0

(1)

For example, estimated status quo monetary asset values (demographic groups excepted) for the application are provided in
Table 2 for each of the pillar indicators.
(ii) Severe Storm Damage Impacts, Dj. Severe coastal storms
inict damage from high winds and coastal ooding. Estimated
damage from severe storms is a function of the severity of the
storm. Storm severity is indicated by index j, j 0,I,II, NS and
j 0 denotes the no storm or Status Quo scenario as in Equation
(1) above. Severe storm damage from Storm Scenario j, Dj is a

vector representing the costs and losses to the indicator values

in the Community Prole vector. The vectors for the Status
Quo or the pre-storm position of the community, SQ0, and
the value losses resulting from the storm damage estimates,
Dj, are directly comparable. Thus, over all pillars' indicators,
0  Dj  SQ0. Table 4 presents damage estimates by storm
scenario for houses, an Economic pillar indicator, for the Little
Anse application.
(iii) Post-Storm Status, SQj. Severe storm damage is represented
by a loss position in the Status Quo Community Prole indicators.
Damages from storms result in a new and reduced state of Community Prole valued assets, denoted vectorially by SQj. Storm
damages and the move to a new Community Prole status position
assumes no previous application of any adaptation strategies that
may be considered or expressly put in place to alleviate potential
severe storm damages. Thus, SQj SQ0Dj.
(iv) Total Asset and Damage Valuations, TA, TDj. The asset position
of the Community Prole is determined by the value-based measures (in monetary units) over all pillars' indicators (demographic
groups excepted). The indicators are summed over all pillars to
obtain the total asset value and the total damage estimate as scalars. Thus, the total initial status quo asset position indicator, TA and
the total damage valuation indicator, TDj following storm j are: TA
Pnec 0
Pnen 0
Pnc 0
Pns 0
P j
EC i
EN i
C i
S i and TDj
D i
i
i
i
i
over all nec nen nc ns pillar indicators, i. The estimatedi total
asset position for the Little Anse application is presented in Table 2
below. Total damage estimates by storm scenario and community
pillar are given in Table 5.

Table 3
Little Anse assets-at-risk from Storm Scenarios I, III, and VI.
Community prole pillar

Assets-at-risk

Storm Scenario I

Storm Scenario III

Storm Scenario VI

Economic

Built Envt: houses

Built Envt: out-buildings
Public works: roads
Public works: wharf
Public works: wells
Residential land
Water bodies/lake
Greenspace/trees
Community centre
Church grounds
Income at risk
Seniors at risk of ooding (over 60)
Children at risk of ooding (under 14)

21
$1000/bldg
558 m
20%
10
22,243 sq m
5%
2000 sq m
Not impacted
10%
$13,660
14
7

29
$2000/bldg
740 m
25%
14
33,035 sq m
15%
2850 sq m
Not impacted
20%
$27,320
18
9

40
$4000/bldg
1014 m
32.5%
20
49,222 sq m
30%
4125 sq m
Not impacted
40%
$47,810
27
14

Environment land use

Cultural
Social

Source: [23]

38

H. Mosto Camare, D.E. Lane / Socio-Economic Planning Sciences 51 (2015) 34e45

Table 4
Assets-at-risk and damage estimations for Little Anse: Houses.
Storm
Number of
Scenario houses
ooded

Total asset value at Damage/House Total estimated

risk, ARj(House)
damage, Dj(House)

I
III
VI

$666,750
920,750
1,270,000

21
29
40

$4000
$6000
$9000

$84,000
$174,000
$360,000

reect the general risk attitude of the community with respect to a

particular indicator. Information is gathered from C-Change community partners over several workshop settings and sample storm
scenario cases. This information is used to develop representative
composite utility functions for the indicators of the problem
formulation.
Let the upper bound of the utility function for a single indicator
be represented by u{SQ*(i)}, the pre-storm community status

Source: [23]

Table 5
Little Anse community prole pillar assets-at-risk and damage estimates.
Community prole pillar

Economic
Environment
Cultural
Social
Total, TDj

Storm Scenario I

Storm Scenario III

Storm Scenario VI

At risk ARI

Damages DI

At risk ARIII

Damages DIII

At risk ARVI

Damages DVI

$1,456,738
$286,685
$800,000
$13,660
$2,557,083

$249,412
$17,031
$80,000
$13,660
$360,103

$1,946,390
$416,422
$800,000
$27,320
$3,190,132

$472,750
$26,902
$160,000
$27,320
$686,972

$2,650,204
$611,015
$800,000
$47,810
$4,109,029

$931,521
$41,884
$280,000
$47,810
$1,301,215

Source: [23]

(v) Community Asset-at-risk valuation, ARj. The valuation of

community assets affected by severe storm j is denoted by ARj. For
example, a house, situated along the shore is an economic asset
whose value can be considered as an at risk value. While the
entire house is considered at risk, depending on storm severity j,
actual storm damage may be considerably less, and reaches a
maximum only if the house is completely destroyed. In most cases,
the actual damage to the property is a portion of at risk value, so
that, overall TDj  ARj  TA. For the Little Anse application, asset-atrisk values are presented in Table 3 by indicator and storm scenario.
Table 4 estimates asset-at-risk in monetary units for houses by
storm scenario. Finally, Table 5 presents asset-at-risk totals by
community pillar.
(vi) Utility Valuation of Indicator Shifts. Utility is an abstract
measure of the relative strength of preference/desirability for a
particular outcome [20]. Utility curve valuations standardize indicators at different scales of measurement, and permit their direct
comparison. Alternative risk prole functional forms are used to

Fig. 3. Risk averse utility function for a single indicator.

Source: [23].

where all assets are assumed to be intact and at their highest

realizable target level denoted as the ideal community position
over all indicators, SQ*. This indicator value, considered as the
ideal state, represents the maximum utility anchor point for the
construction of the utility curve, i.e., u{SQ*(i)} 1. Let the lower
bound of this indicator's utility function be represented by u{SQj(i)}
which is the status of the indicator value i following the severe
storm, j, j 0,I,II, NS. The worst case indicator value for the most
severe storm scenario j Ns represents the minimum utility anchor
point, i.e., u{SQNs(i)} 0. Over all community indicators, adaptation
strategies are designed to move the community away from SQj, the
status of the community following severe storm j, to approach SQ0,
the pre-storm community status benchmark.
Fig. 3 depicts the general marginal decreasing risk averse
(concave) utility function adopted for determining the weighted
valuation of selected Community Prole indicators in the case
study based on preference information collected from C-Change
community partners. Given that the assumption of the relevant risk

H. Mosto Camare, D.E. Lane / Socio-Economic Planning Sciences 51 (2015) 34e45

prole is not strictly based on empirical information among the

decision makers for all indicators, it is understood that this
assumption about risk functional form is a strong assumption that
may need to be explored in more detail [16].
Fig. 3 also illustrates example utility indicator values for the
alternative adaptation strategies labelled Strategy 1 and denoted
A1, and Strategy 2 and denoted A2. Strategy 2 has an improved
utility indicator value (u{SQj(A2)}  u{SQj(A1)}). Indicator utility
values transform the indicator scale to a unique and comparable
dimensionless scale. Utility values over all indicators, together with
the AHP Community Prole pillar priority weights (Table 1),
determine the overall utility value over all indicators for each
strategy. These values enable the comparison and ranking of the
adaptation strategies discussed below.
IV. Community Evaluation of Adaptation Strategies. Adaptation
strategies for coastal communities under threat from coastal storms
and sea-level rise are divided into four broad categories [22,29].
These strategies are articulated as follows:
A1) Protect e Adaptation strategies that employ man-made
engineered structures such as seawalls, breakwaters, revetments and groins are included as a means of overcoming
negative storm damage impacts designed to protect human
infrastructure currently in place and arguably, in the way of
the storms. Protection measures can range from small scale
efforts by individual property owners, to large scale public
projects [22,25].
A2) Accommodate e Accommodation strategies seek continued
occupancy of coastal areas through adjustments made to
infrastructure and human activities designed to accept sealevel rise and storm surge to moderate impact severity
[26]. Accommodation strategies include redesigning and
reinforcing existing structures, encouraging appropriate land
use, and development by implementing legislation and soft
measures for increasing natural resilience by rehabilitating
coastal dunes and wetlands. Examples are elevation of
buildings, prohibiting or controlling the removal of beach
sediment, promoting mangrove growth and natural barriers,
shifting agriculture production to salt-tolerant crops, and
development of warning systems for extreme events [22,25].
A3) Retreat e Retreat means avoiding direct storm impacts by
moving away from the threat [22,25]. Retreat can mean
migration, abandoning the land when impacts are not
tolerable, or restricting construction and development in
ood and erosion hazard zones so that impacts are reduced.
Retreat can be promoted or enforced by public education,
land swapping or management strategies such as zoning,
insurance or tax policies.
A4) Status Quo e This option, most often applied in practice,
recognizes the strategy of doing nothing to prepare for
coastal storms impacts.
Let Ak, k 1,2,3,4 denote the four adaptation strategies as
described above. Data are assigned to the indicators of the community prole to estimate the storm scenario j impacts on the
community prole indicators denoted, SQj(Ak) for each adaptation
strategy, Ak. These data estimates provide the basis for ranking of
the alternative strategies using AHP through the utility values, u
{SQj(Ak)}. These utility values associated with the adaptation
strategy rankings are used in turn to determine utility-related
measures of vulnerability, u{Vj}, resilience, u{Rj(Ak)}, and adaptive
capacity, u{ACj(Ak)}, as described below. Specic adaptation strategies are dened for the Little Anse application presented below.
Community Vulnerability, Resilience and Adaptive Capacity. The
concept of vulnerability is presented as the difference between pre-

39

storm and post-storm Community Prole states in terms of the

weighted utility values of the indicators. Let u{Vj} denote a singlevalued measure of the vulnerability of the community to severe
Storm Scenario j in terms of their respective utility values, u, and
the AHP hierarchy weights, then

o
n
o
n
ufVjg u SQ 0  u SQ j  0

(2)

When adaptation strategies are introduced to moderate the

impacts of storm damages, then a reduction is expected in the
coastal community's level of vulnerability. These strategies are
designed to protect community assets from storm impacts, and
therefore result in less loss on initial asset values. Thus, applying
adaptation strategy Ak means that the new status of the community
in the face of severe storm j will be given by u{SQj(Ak)}, the modied
community status in utility terms, such that: u{SQj}  u
{SQj(Ak)}  u{SQ0}.
Let u{Rj(Ak)} denote the resilience of the community under the
adaptation strategy, Ak in terms of the utility measures. Resilience is
a value-added measure arising from the adoption of the particular
adaptation strategy, and denoting the reduction in vulnerability
loss compared to the do nothing strategy under storm scenario j.
Thus, u{Rj(Ak)} is given by:

o
n
o
n
o
n
u Rj Ak u SQ j Ak  u SQ j

(3)

where 0 < u{Rj(Ak)} < u{Vj}. Fig. 4 illustrates the effects on community status and the concept of vulnerability to storm scenario j, u
{Vj} as the difference between the current Best CaseeNo Storm
status of the Community Prole indicators, u{SQ0} (or the Ideal
state, SQ*) versus the impact of Storm Scenario j, (and no adaptation
strategy) status of the community indicators, u{SQj}.
Finally, community adaptive capacity, u{ACj(Ak)} is determined
for each adaptation strategy, Ak, and for all adaptation options of
the community, as the ratio of resilience to vulnerability for the
specied storm scenario, j, j I,II, VI, and the alternative adaptation strategy, Ak, i.e.,

n
o
n
o. n o
u AC j Ak u Rj Ak
u Vj

(4)

where 0 < u{ACj(Ak)} < 1 or 100%. Average u{ACj(Ak)} or AC is

determined by averaging the adaptive capacity results of all available strategies in a given context for Storm Scenario j I,II, NS.
The framework presented here is an inclusive approach with
respect to all pillars of the community prole, namely, Economic,
Environmental, Cultural and Social. The measures of vulnerability,
resilience, and adaptive capacity dened above can be applied to
each of the Community Prole pillars. Moreover, being framed as a
multicriteria and multiple participant formulation, this decision
analysis also permits various stakeholders' to provide their inputs
on key values toward developing a quantitative comparative analysis of the adaptation strategies. Community Vulnerability
(Table 6), and Resilience and Adaptive Capacity measures (Table 8)
are provided for the Little Anse application presented below.
The following section applies the model process and illustrates
model analyses and results for the case of the failed breakwater in
the community of Little Anse on Isle Madame, Richmond County,
Nova Scotia.
3. Results e application
The community of Little Anse is located at the southeastern end
of Petit de Grat Island on Isle Madame, Richmond County, Cape
Breton. The map of Fig. 5 locates the coastal community of Little

40

H. Mosto Camare, D.E. Lane / Socio-Economic Planning Sciences 51 (2015) 34e45

u{V*}
Vulnerability at ideal to SSj
u{Vj}

Vulnerability at Current Status to SSj

Resilience of Ak

Vulnerability Loss of Ak

u{R (Ak)}

u{V (Ak)}

u{SQ }

u{SQ (Ak)}

Worst Case-SSj/No Strategy

u{SQ }

SSj/Apply Strategy Ak

Best Case-No Storm

u{SQ*}=1
Ideal State

Fig. 4. Community status axis for adaptation strategy, Ak, k 1,2,3,4 and storm scenario j effects, SSj, j I,II, Ns.

Table 6
Relative utility vulnerability measures for Little Anse current and ideal community
prole status.
Scenario/Vulnerability

u{Vj} (vs current)

u{Vj} (vs ideal)

Ideal status
Current (No Storm)
I
II
III
IV
V
VI

0.116
e
0.072
0.102
0.136
0.176
0.232
0.316

e
0.116
0.186
0.216
0.248
0.286
0.340
0.422

storm waters that over-top the road from the sea and spread into
the brackish pond across the road. Little Anse is an aging community with more senior residents than juveniles. There are no schools
or shopping in the community. The asset base, estimated at $14
million (2010) includes residential buildings and land, and associated roads, wells, and septic systems, as well as greenspace, as are
presented in Table 2. Table 2 values are the Community Prole indicator values, SQ0(i) of Equation (1) as inputs to the AHP model for
Little Anse [30].
II. Community Vulnerability and Storm Impacts. The Little Anse
harbour and the community itself are regularly ooded as the result

Table 7
Comparison of adaptation strategy rankings for combined inputs and average of rankings by participant group.

Storm Scenario I
Status Quo
New BW Arm
Retreat
Road build up
Storm Scenario III
Status Quo
New BW Arm
Retreat
Road build up
Storm Scenario VI
Status Quo
New BW Arm
Retreat
Road build up

Local government

Business

Community

Professional

Average (output)

Combined (input)

0.247
0.251
0.249
0.253

0.267
0.236
0.247
0.25

0.252
0.248
0.249
0.251

0.254
0.245
0.252
0.249

0.255
0.245
0.249
0.251

0.246
0.253
0.251
0.251

0.237
0.263
0.245
0.255

0.26
0.244
0.245
0.251

0.244
0.258
0.246
0.251

0.247
0.256
0.249
0.248

0.247
0.255
0.246
0.251

0.237
0.263
0.248
0.252

0.208
0.307
0.24
0.245

0.242
0.273
0.236
0.249

0.222
0.299
0.238
0.242

0.229
0.293
0.239
0.238

0.225
0.293
0.238
0.243

0.211
0.304
0.241
0.244

The bold text corresponds to the preferred decision row indicated by the maximum Average and Combined output columns for each Storm Scenario I, III, and VI.
Source: [23]
Table 8
Relative utility values for resilience and adaptive capacity measures for New
Breakwater Arm adaptation strategy for Little Anse.
Scenario

Resilience of
NBA, u{Rj(A1)}

Vulnerability, u{Vj}

Adaptive capacity of
NBA, u{ACj(A1)}

I
II
III
IV
V
VI

0.014
0.032
0.052
0.078
0.118
0.192

0.072
0.102
0.136
0.176
0.232
0.316

0.194
0.314
0.382
0.443
0.508
0.607

Anse. The application of Little Anse to the evaluation strategy follows the research process and is presented below.
I. Community Prole. Little Anse is a small community comprised
of approximately 100 residents. Most of these buildings are in the
area beyond the community wharf and are in danger of being
isolated when the single road through the community is cut off by

of storms from the seaward side and the failure of the breakwater to
protect the shoreline. Floods result in closure of the seaside road
cutting off the inhabitants to the south, inundating properties,
basements, and local water sources (wells), and damaging houses
and out-buildings. The deciencies of the failed breakwater are the
results of displaced armour stone, deteriorated concrete and stone
quality, and lack of maintenance.
ArcGIS was used to examine Maximum Observed Water Level
(MOWL) associated with storms. MOWLs were estimated to range
between 1 m and 2 m for the community, based on historical storm
data. The 1 m water levels are illustrated by the buffer line in
Fig. 6 [28].
The buffer line situates the limit of the ooding scenario based
on the community elevation data provided. The 1 m ood line
implies that properties, roads, and the wharf, on the seaward side
are ooded. Assets are therefore at risk from the storm surge and
likely are damaged as a consequence of being on the seaward side
of the buffer line.

H. Mosto Camare, D.E. Lane / Socio-Economic Planning Sciences 51 (2015) 34e45

Fig. 5. Little Anse, Richmond County, Cape Breton, Nova Scotia, Canada.
Source: [13]

Assets-at-risk from storms are estimated for Storm Scenario

j I,II, VI, with increasing maximum water level. Table 3 summarizes these impacts for Storm Scenarios I, II and VI for selected
assets.
Asset damages for Little Anse House assets are presented in
Table 4 as an illustration. With respect to Little Anse residential
houses, the average value of a house in Little Anse is determined
from 2010 Multiple Listing Service (MLS) real estate values. The
number of houses ooded by the Storm Scenario is estimated from
ArcGIS ood lines (Fig. 6). Residential home valuations are taken
from the Property Valuation Service Corporation (PVSC) database
with the average 2010 value of $31,750 (CDN) for Little Anse
properties. Table 4 records the number of houses and buildings
impacted as the result of Storm Scenarios I, III, and VI. According to
Canada Mortgage and Housing Corporation [6], one of the most
frequent damages to any building as the result of ooding, is
damage to standing water in basements. According to the same
source [6], the average cost per ooding incident to a basement is

Fig. 6. Little Anse Breakwater (black arm) with 1 m ood buffer line.
Source: [12,23,28]

41

between $3000 and $5000 CDN (average of $4000 CDN).

Increasing estimates by row in Table 4 reect the increased
severity and ood water levels of the storm scenarios. Asset
damages for Little Anse House assets are presented in Table 4 as an
illustration. With respect to Little Anse residential houses, the
average value of a house in Little Anse is determined from 2010
Multiple Listing Service (MLS) real estate values. The number of
houses ooded by the Storm Scenario is estimated from ArcGIS
ood lines (Fig. 6). Residential home valuations are taken from the
Property Valuation Service Corporation (PVSC) database with the
average 2010 value of $31,750 (CDN) for Little Anse properties.
Table 4 records the number of houses and buildings impacted as
the result of Storm Scenarios I, III, and VI. According to Canada
Mortgage and Housing Corporation [6], one of the most frequent
damages to any building as the result of ooding, is damage to
standing water in basements. According to the same source [6],
the average cost per ooding incident to a basement is between
$3000 and $5000 CDN (average of $4000 CDN). Increasing estimates by row in Table 4 reect the increased severity and ood
water levels of the storm scenarios.
Asset damages for Little Anse House assets are presented in
Table 4 as an illustration. With respect to Little Anse residential
houses, the average value of a house in Little Anse is determined
from 2010 Multiple Listing Service (MLS) real estate values. The
number of houses ooded by the Storm Scenario is estimated from
ArcGIS ood lines (Fig. 6). Residential home valuations are taken
from the Property Valuation Service Corporation (PVSC) database
with the average 2010 value of $31,750 (CDN) for Little Anse
properties. Table 4 records the number of houses and buildings
impacted as the result of Storm Scenarios I, III, and VI. According to
Canada Mortgage and Housing Corporation [6], one of the most
frequent damages to any building as the result of ooding, is
damage to standing water in basements. According to the same
source [6], the average cost per ooding incident to a basement is
between $3000 and $5000 CDN (average of $4000 CDN). Increasing
estimates by row in Table 4 reect the increased severity and ood
water levels of the storm scenarios.
Similar valuations of Little Anse assets, assets-at-risk, and
damage estimates are provided for all indicators of Table 2.

42

H. Mosto Camare, D.E. Lane / Socio-Economic Planning Sciences 51 (2015) 34e45

Table 5 summaries the estimates of assets-at-risk, and damage

by Storm Scenario for all Little Anse indicators of Table 2 by Community Prole pillar and for Storm Scenarios I, III, and IV.
It is noted from Table 5 that the Economic pillar has the highest
Assets-At-Risk, ARj and Damage, Dj estimates values in comparison
to the other Community Prole pillar values.
III. Community Adaptation Problem Formulation. The AHP problem hierarchy for Little Anse is presented in Fig. 7 along with the
pairwise comparison weights for the combined participants
(Table 1). Fig. 7 presents the Little Anse AHP hierarchy from the [10]
software download used for the analysis of the AHP model results.
The hierarchy corresponds to the assets of the coastal community
and in so doing, provides the local context for this analysis.
Little Anse is represented by multiple stakeholders in the
community participatory decision process. Little Anse stakeholders
are represented by members in the Seniors Community, faith-based
(Catholic) church, and local media (Telile); in local government by
the Administration of the Municipality of the County of Richmond;
the local sheries harvesting and processing industry; and
Community professionals in public health, and hospice. Table 1
provided the Little Anse Community Prole priority weights
applied to this problem.
Community prole indicator assets, in monetary values (Table 2)
(demographic groups excepted), were transformed using the AHP
Data Grid of the Expert Choice software to utility values (Fig. 3), and
then weighted with the AHP indicator and corresponding pillar
weights to obtain overall measures for comparative ranking. Utility
proles for specic indicators were derived from feedback of CChange community partners who were polled on their preferences
(utilities) for alternative levels of damages to community assets,
e.g., the length of road closures, ooding of properties, etc.
Measures of the vulnerability, u{Vj} of the Little Anse coastal
community are evaluated using the difference Equation (1) for the
comparison of: (i) the current (no storm) status quo, and (ii) the
ideal status quo state of the Community Prole indicators versus

Treeview

Goal:Adaptation of Little Anse community to storm surge events

Economic (L: .285)
Built Environment (L: .278)
Houses (L: .668)
Buildings (L: .332)
Public Works (L: .484)
Roads (L: .451)
Wharfs (L: .245)

the impacts of Storm Scenarios I through VI. Fig. 8 compares the

AHP model results for the current No Storm Status Quo scenario,
u{SQ0}, with each of the Ideal scenario, u{SQ*} 1, and the estimated post-storm Status Quo positions, u{SQj} for the six Storm
Scenarios j I,II, VI, for NS VI, the most severe storm.
Fig. 8 determines coastal community vulnerability with respect
to each Storm Scenario and relative to the current (no storm)
overall community relative utility status. The lower bars denote the
ranked value of the current status quo position. The comparable
upper bars denote the ideal (rst bar) and the decreasing Storm
Scenario weights. Increasing storm severity implies greater storm
damage, and overall lower community asset (and utility) values in
comparison with the status quo. Accordingly, the relative value of
the no storm status quo position increases as the severity of the
storm increases. The simple differences measure community
vulnerability in terms of the weighted utility rankings of the paired
options over all indicators and community pillars. For example, in
Fig. 8, the difference between the ranking of the current No Storm
status and Storm Scenario I yields vulnerability measure u
{VI} 0.5360.464 0.072. For Storm Scenario VI, u{VVI} is
increased to 0.316 (0.6580.342).
Table 6 below summarizes the vulnerability measures, u{Vj} as
the differences in each of the cases of Fig. 8 (i.e., AHP model differences between (in the rst column) the Ideal scenario utility, u
{SQ*} 0.558, versus the current No Storm Status Quo scenario, u
{SQ0} 0.442, and, in subsequent columns, the estimated poststorm Status Quo positions, u{SQj} for the six Storm Scenarios
j I,II, VI compared with the current No Storm Status Quo
scenario). As noted in Table 6, vulnerability increases with the
severity of the storm scenario. As expected, vulnerability is higher
with respect to the higher targeted position of the Ideal status of
the community compared with its current status.
IV. Community Evaluation of Adaptation Strategies. The existing
breakwater structure was constructed between 1946 and 1948 with
an initial 10 tonnes of concrete blocks. At a later unknown date,
concrete blocks were reinforced with a poured-in-place slab and
covered entirely with armour stones [2]. Despite these reinforcements, in 2011 the existing structure was in a state of
disrepair and is no longer providing the desired protection to the
community of Little Anse that now regularly experiences ooding
from storms (The Reporter 2011).
The current state of Little Anse breakwater, new breakwater
strategies, and their cost estimations were studied by coastal engineers [2] who developed options for this site. A set of specic
engineering options designed to protect the community from
storm impacts were considered for Little Anse breakwater based on
the analysis of the coastal engineers. These options included: (1)

Wells (L: .303)

Adaptation cost (L: .238)
Environment (L: .224)
Land Use (L: 1.000)

1
0.9
0.8

Residential Land (L: .511)

0.7

Lake (L: .260)

0.6

Trees (L: .229)

0.5

Cultural (L: .197)

Community Center (L: .507)
Church grounds (L: .493)
Social (L: .294)
Income loss (L: .272)
Safety at risk (people over 60) (L: .412)
Safety at risk (people under 14) (L: .316)
Fig. 7. Little Anse AHP Hierarchy with combined weights for all participants.
Source: [10]

0.558

0.464

0.449

0.432

0.412

0.384

0.342

0.536

0.551

0.568

0.588

0.616

0.658

0.4
0.3
0.2

0.442

0.1
0
No Storm Vs No Storm Vs No Storm Vs No Storm Vs No Storm Vs No Storm Vs No Storm Vs
Ideal
SSI
SSII
SSIII
SSIV
SSV
SSVI
Fig. 8. Relative utility values comparison of current community status quo (No
Storm) scenario, u{SQ0} versus Ideal, u{SQ*} 1, and Storm Scenarios I through VI
impacts, u{SQj}.
Source: [23]

H. Mosto Camare, D.E. Lane / Socio-Economic Planning Sciences 51 (2015) 34e45

rehabilitation of the existing breakwater; (2) closing of the existing

opening and creation of another; and (3) construction a new
breakwater arm. Prior analysis suggested that the third option
provided a preferable protection strategy.
Specic adaptation strategies for Little Anse are presented
below:
A1) Protect e New breakwater arm: Build a new arm opposite to
the existing breakwater and rehabilitation of the existing
breakwater. This option narrows the current opening in a
way that waves would go through a signicant energy
reduction before they reach the inner harbour area. This
option is expected to mitigate signicantly the surge events
originating from the south and east. With the staggered
structure in place, waves will be further contained before
reaching the shoreline. Therefore, this strategy can provide
the best protection to the community from ooding (fewer
houses ooded and reduced ooding costs, road more protected from over-topping). However, this option is also the
most expensive capital cost strategy among all breakwater
engineering alternatives. Estimated costs are over $5 million.
A2) Accommodate e Road Build Up: Build a new road further from
the shoreline. The new road is estimated to be 1400 m in
length. Road building includes: capital cost (material and
labour), mobilization cost (machinery), and engineering.
Estimated total costs are approximately $2 million. Considering that the current road is the only access route to Little
Anse with the outside, building a new road which is more
immune to storm surge impacts seems to be a long term and
potentially justiable option.
A3) Retreat: This strategy, moves a selection of houses away from
the coastline to somewhere safer from ooding. For Storm
Scenario III, 29 Little Anse houses are in the ood zones and
are therefore candidates for removal (Fig. 6). The cost of the
retreat strategy includes: compensation for the value of
property, plus mobilization cost (uncompensated cost of
relocating a household to a new location). The fully costed
estimates for this strategy approach $2.5 million.
A4) Do Nothing Status Quo: The suite of adaptation strategies
includes the Do Nothing (Status Quo) option.
The AHP ranking of adaptation strategies are presented for three
modelled Storm Scenarios I, III, and VI. The summary of the results
by participant group, and the combined input are presented in
Table 7 and discussed briey below.
Storm Scenario I. For this least severe storm scenario, all four
adaptation strategies are closely ranked (stakeholder rankings
range from 0.245 to 0.254) with ranking range of only 0.009
(Table 7). For this least severe storm scenario all options appear only
marginally better than the Status Quo (Do Nothing) option. These
results would appear to vindicate the often adopted decision of
not-doing-anything-at-all, i.e., in the event that decision makers
appear condent that elevated severe storms will not occur on
their watch, then it is rationalized that the least severe outcomes
will not require augment preparation and strategy development.
Storm Scenario III. In this more severe storm scenario, the New
breakwater arm strategy becomes slightly more preferred among
most participants, with the exception of the Business stakeholders.
It is noted that the Road build up option is generally preferred to
the Retreat option for stakeholders, and the Status Quo Do
Nothing option is least preferred for all participants with the
exception of the Business stakeholders (Table 7).
Storm Scenario VI. From Table 7 it is evident that New breakwater arm strategy becomes dominant among strategies for all
participants. The Road build up strategy is generally more

43

preferred to the Retreat option in all Storm Scenarios due principally to its low relative cost. Although compared to Road build
up, the Retreat strategy makes more contribution on indicators
overall (houses, buildings, income, well and safety versus only road
and safety in the case of road build up strategy) but the higher
cost of retreat and the signicance of the road (in terms of pairwise comparison weights in the hierarchy) makes the Road build
up alternative preferable to the Retreat strategy. By taking into
account the intangible costs a moving out, e.g., reluctance to
abandon your house, the Retreat strategy would be expected to
become even more inferior to other adaptation strategies. At this
most severe storm scenario, the Status Quo Do Nothing option
has a signicantly lower rank than all other adaptation strategies
(Table 7). This indicates that the elevated storm severity encourages
a trend toward adaptation action in the event that coastal storms
are becoming more frequent and severe.
Finally, the concept of resilience, u{Rj(Ak)} (Equation (3)) and
adaptive capacity, u{ACj(Ak)} (Equation (4)) by Storm Scenario j and
adaptation strategy Ak are analysed with respect to the contribution
of adaptation strategies to reduced storm impacts. The New
Breakwater Arm adaptation strategy A1 is used to illustrate the
resilience and adaptive capacity. In Fig. 9 below, NBA denotes the
New Breakwater Arm adaptation strategy, and the impact of this
strategy is compared against the Do Nothing strategy for each of
the six Storm Scenarios, SSI-SSVI.
In Fig. 9, the lower bars denote the weighted utility of the New
Breakwater Arm adaptation strategy. The upper bars denote the
relative utility value of the Do Nothing strategy for each of the
increasing storm scenarios. For the lowest severe storm, SSI, the
New Breakwater Arm provides little additional utility compared
to the no adaptation strategy. However, the differences between
the weighted utility values of the New Breakwater Arm strategy,
and the Do Nothing strategy increase as the Storm Scenarios
become more severe. The increasing differential weighted utility
values, or resilience (Equation (3) and Fig. 4) indicate increasing
contribution of the adaptation strategy to the resilience of the
community with increasing storm severity. As the Storm Scenarios
become more severe, the contribution of this differential becomes
more signicant, therefore indicating more community resilience.
Adaptive capacity, captured as the ratio of resilience to vulnerability
(Equation (4)), indicates the fraction of the community vulnerability is covered by the reduction in storm impacts through the
adoption of the adaptation strategy in question. The corresponding
resilience and adaptive capacity measures for the New Breakwater
Arm adaptation strategy under the six Storm Scenarios are presented in Table 8. For the New Breakwater Arm strategy, resilience and adaptive capacity measures increase as the severity of the
storm increases.

1
0.9
0.8
0.7

0.493

0.484

0.474

0.461

0.441

0.404

0.507

0.516

0.526

0.539

0.559

0.596

NBA Vs SSI

NBA Vs SSII

NBA Vs SSIII

NBA Vs SSIV

NBA Vs SSV

NBA Vs SSVI

0.6
0.5
0.4
0.3
0.2
0.1
0

Fig. 9. Relative Utility Value Comparison of New Breakwater Arm (NBA) adaptation
strategy results, u{SQj(Ak)}with Storm Scenarios I through VI impacts, u{SQj} for SSI
through SSVI.
Source: [23]

44

H. Mosto Camare, D.E. Lane / Socio-Economic Planning Sciences 51 (2015) 34e45

4. Conclusion
The paper develops and applies a multicriteria, group decision
making model in support of the evaluation of alternative adaptation strategies for coastal communities facing changing environmental conditions from more frequent severe storms and sea-level
rise. The procedure determines context-based measures for
coastal community vulnerability, resilience, and adaptive capacity
based on a multidimensional prole of the community. These
logical, local context measures are presented as alternatives to
aggregate measures of vulnerability that do provide decision
support for adaptation strategies. The formulation of the multicriteria and group participant AHP model framework is readily
applicable to the coastal adaptation decision problem, and
dependent on subjective trade-off information from participants,
as well as data estimates on local storm impacts. In the context of
the application presented here, it is encouraging to note that
coastal community participants are eager to engage in the discussion of adaptation strategies, and to provide feedback and information to assist in decision making. The challenge is to
convince decision makers that decision evaluation is a feasible and
necessary task in support of adaptation action for these complex
problems.
Analytical Hierarchy Process (AHP) was utilized as the decision
support tool for this study based on its ease of application and its
treatment of qualitative and quantitative data sources. The model
can be used to evaluate a multitude of adaptation strategies that
could overlap several strategy categories presented here according
to the heterogeneity to the coastal decision problem. AHP sensitivity analysis can be conducted to examine the impacts of different
indicator value estimates, altered priority weights, the changes in
decision makers' utility and risk proles, and alternative approaches to aggregating multiple participant information toward
improved strategy evaluation.
The research has exposed the need for improving the amount
and availability of local storm data, and it promotes the analysis of
these data while structuring the evaluation of adaptation strategies at the local community level. The imminent creep of coastal
climate change and the urgency of coastal communities to be
better prepared to adapt to the changing coastal environment
requires further effort in the strategic analysis of adaptation
strategies.

[2]
[3]

[4]

[5]

[6]
[7]

[8]

[9]

[10]
[11]

[12]

[13]
[14]

[15]

[16]
[17]

[18]

Acknowledgements
The authors gratefully acknowledge the support of the Canadian
Funding Agencies, the Social Sciences and Humanities Research
Council (SSHRC) and the International Development Research
Centre (IDRC), and the Telfer School of Management of the University of Ottawa in Ottawa, Canada, for their support. The helpful
and extensive comments from journal reviewers are acknowledged
with thanks, as are the contributions of our C-Change colleagues,
community partners, collaborators, and students in developing the
work described here. C-Change is deeply appreciative of the time
and effort spent of municipal staff and community members of the
C-Change partner community of Isle Madame, Richmond County,
Nova Scotia. It is thanks to their valued participation, commitment,
and contribution to managing adaptation that this research has
relevance.

References
[1] Adger WN, Brooks N, Bentham G, Agnew M, Eriksen S. New indicators
of vulnerability and adaptive capacity. Norwich: Tyndall Centre for

[19]

[20]
[21]

[22]

[23]

[24]

[25]

[26]

[27]

Climate Change Research, University of East Anglia; 2004. Technical

Report 7.
Baird & Associates. Coastal engineers, Ottawa. 2010 [accessed 12.10.13],
http://www.baird.com/.
C-Change. Managing adaptation to environmental change in coastal communities: Canada and the Caribbean. 2013. Project website. [accessed
22.11.13], www.coastalchange.ca.
Clark G, Moser S, Ratick S, Dow K, Meyer W, Emani S, et al. Assessing the
vulnerability of coastal communities to extreme storms: the case of Revere,
MA, USA. Mitig Adapt Strateg Glob Change 1998;31:59e82.
Cutter SL, Emrich CT, Webb JJ, Morath D. Social vulnerability to climate variability hazards: a review of the literature. 2009. Final report to Oxfam
America. Available at: http://adapt.oxfamamerica.org/resources/Literature_
Review.pdf [accessed 05.02.12].
CMHC. Canadian mortgage and housing corporation (CMHC). 2010 [accessed
20.11.12], http://www.cmhc-schl.gc.ca/en/co/maho/gemare/gemare_002.cfm.
Dolan AH, Walker IJ. Understanding vulnerability of coastal communities to
climate change related risks. J Coast Res 2004. Special Issue 39. Proceedings of
the 8th International Coastal Symposium, Itajai, SC e Brazil, ISSN 0749e0208.
Available
at:
http://www.geog.uvic.ca/blast/Publication%20PDFs/Dolan%
26Walker-ICS042.pdf [accessed 23.11.13].
Drye W. 2010 Hurricane season Strange. National geographic, daily news.
2010. Available at: http://news.nationalgeographic.com/news/2010/11/
101130-2010-atlantic-hurricane-season-science-environment/
[accessed
17.04.12].
Environment Canada. Canadian hurricane centre ready. For the 2010 hurricane season. 2010. Available at: http://www.ec.gc.ca/default.asp?
langEn&n714D9AAE-1&news9CE46DEE-DF7F-479B-8A105670DBC7CDC9 [accessed 27.05.12].
Expert Choice. Analytic hierarchy process software. 2010 [accessed 10.10.13],
http://www.expertchoice.com/.
Fisheries and Oceans Canada. Canadian tides and water levels data archive.
2013. Available at: http://www.meds-sdmm.dfo-mpo.gc.ca/isdm-gdsi/twlmne/index-eng.htm [accessed 07.09.13].
GeoNova. Geographic gateway to Nova Scotia. 2010 [accessed 20.06.10],
http://www.gov.ns.ca/geonova/home/products/softpage/data_locator.
asp.
Google Earth, 2011. http://www.google.com/earth/ [accessed 10.09.11].
IPCC. Summary for policymakers. In: Stocker TF, Qin D, Plattner G-K,
Tignor M, Allen SK, Boschung J, et al., editors. Climate change 2013: the
physical science basis. Contribution of working group I to the fth
assessment report of the intergovernmental panel on climate change.
Cambridge, United Kingdom and New York, NY, USA: Cambridge University
Press; 2013. http://www.climatechange2013.org/images/report/WG1AR5_

SPM_FINAL.pdf.
IPCC. Summary for policymakers. In: Parry ML, Canziani OF, Palutikof JP, van
der Linden PJ, Hanson CE, editors. Climate change 2007: impacts, adaptation
and vulnerability. Contribution of working group ii to the fourth assessment
report of the intergovernmental panel on climate change. Cambridge, UK:
Cambridge University Press; 2007. p. 7e22.
Kahneman D, Tversky A. Prospect theory: an analysis of decision under risk.
Econometrica 1979;47(2):263e92.
Landsea C, Franklin J, Beven J. The revised Atlantic hurricane database
(HURDAT2). 2013. February. [accessed 22.11.13], http://www.aoml.noaa.gov/
hrd/hurdat/newhurdat-format.pdf.
Landsea C, Franklin J. Atlantic Hurricane database uncertainty and presentation of a new database format. Mon Weather Rev October 2013;141:3576e92.
http://dx.doi.org/10.1175/MWR-D-12-00254.1.
Lane DE, Mercer Clarke C, Forbes D, Watson P. The gathering Storm: managing
adaptation to environmental change in coastal communities and small
islands. Sustain Sci July 2013;8(3):469e89.
Lane DE, Stephenson RL. A framework for risk analysis in sheries decision
making. ICES J Mar Sci 1998;55:1e13.
Ling C, Dale A, Hanna K. Integrated community sustainability planning tool.
2007 [accessed 23.11.13], http://crcresearch.org/sites/default/les/icspplanning-tool.pdf.
McCulloch M, Forbes DL, Shaw WR, CCAF scientic team. Coastal impacts of
climate change and sea level rise on Prince Edward Island. Geological Survey
of Canada, open le 4261, synresearch report. 2002.
Mosto Camare H. Multicriteria decision evaluation of adaptation strategies
for vulnerable coastal communities. Ottawa, Canada: University of Ottawa;
2011. MSc thesis, Systems Science [accessed 30.09.12], http://www.
coastalchange.ca/images/stories/Documents_Tab/Mosto_FinalThesis_2011.
pdf.
NHC. Storm surge overview. National Hurricane Center; 2010. Available
at:
http://www.nhc.noaa.gov/ssurge/index.shtml#INTRO
[accessed
09.09.12].
NRC. Climate change impacts and adaptation. Natural Resources Canada;
2010a [accessed 20.11.12], http://adaptation.nrcan.gc.ca/perspective/coastal_
4_e.php.
NRC. Storm surges. Natural Resources Canada; 2010b [accessed 20.11.12],
http://atlas.nrcan.gc.ca/site/english/maps/environment/naturalhazards/
storm_surge/1.
O'Brien K, Eriksen S, Schjolden A, Nygaard L. What's in a word?. 2004. Conicting interpretations of vulnerability in climate change research. Working

H. Mosto Camare, D.E. Lane / Socio-Economic Planning Sciences 51 (2015) 34e45

[28]

[29]
[30]

[31]
[32]

[33]

[34]

paper. Centre for International Climate and Environmental Research Oslo,

University of Oslo, Oslo.
Pakdel S. Spatialetemporal modelling for estimating impacts of storm surge
and sea level rise on coastal communities: the Case of Isle Madame in Cape
Breton, Nova Scotia, Canada. Ottawa, Canada: University of Ottawa; 2011.
M.Sc Thesis in Systems Science.
Pilkey HO, Young R. The rising sea. Island Press, A Shearwater Book; 2009.
County Richmond. Richmond integrated community sustainability plan. Cape
Breton, Nova Scotia. 2010 [accessed 10.05.11], http://www.richmondcounty.
ca/default.asp?mn1.48.144.
Saaty TL. Analytical hierarchy process. New York: McGraw Hill; 1980.
The Reporter. No help for wharf and breakwater. 2010 [accessed 12.11.10],
http://www.porthawkesburyreporter.com/porthawkesburyreporter.com/
stories.asp?id2979.
Vasseur L, Catto N. Atlantic Canada. In: Lemmen DS, Warren FJ, Lacroix J,
Bush E, editors. From impacts to adaptation: Canada in a changing climate.
Ottawa, ON: Government of Canada; 2008.
Yamin F, Rahman A, Huq S. Vulnerability, adaptation and climate disasters:
a conceptual overview. Inst Dev Stud 2005;36(4):1e14.

45

Hooman Mosto Camare, B.Sc., MBA, M.Sc. (Systems Science) Hooman received an
B.Sc in industrial engineering in 2002 from Azad University of Tehran. He worked two
years in the IT industry as a programmer and analyst. He also worked as a contractor
for Shell Development Iran and as part of environmental impact assessment study for a
gas pipeline project. In 2006, he moved to Malaysia to pursue his MBA certicate
graduating in 2008 from MMU University. He moved to Canada in 2009 and completed
his M.Sc. in Systems Science at the University of Ottawa in 2011. He is living and
working in Toronto.

Daniel E. Lane, Full Professor, Telfer School of Management, University of Ottawa Full
Professor since 1996, Lane is currently Coordinator of the Operations Management and
Information Systems (OMIS) section of Telfer. He is Co-Director (Canada) of the CChange ICURA project, Chair of the Ocean Management Research Network (OMRN),
and Director of C-FOAM, the Canadian Fisheries, Oceans, and Aquaculture Management
research group at the Telfer School of Management of the University of Ottawa. Professor Lane is a member of the Canadian Operational Research Society (CORS), the Institute for Operations Research and Management Science (INFORMS), and the Natural
Resource Modelling Association (NRMA).