Ecosystem Services ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Ecosystem Services
journal homepage: www.elsevier.com/locate/ecoser

Global values of coastal ecosystem services: A spatial economic

analysis of shoreline protection values
Nalini S. Rao a,n, Andrea Ghermandi b, Rosimeiry Portela a, Xuanwen Wang c
a
Betty and Gordon Moore Center for Science and Oceans, Conservation International, 2011 Crystal Drive Suite 500, Arlington, VA 22202, USA
b
Department of Natural Resources and Environmental Management, University of Haifa, 199 Abba Hushi Blvd., Mount Carmel, Haifa 3498838, Israel
c
Center for Construction Research and Training, 8484 Georgia Avenue, Suite 1000, Silver Spring, MD 20910, USA

art ic l e i nf o a b s t r a c t

Article history: A global study to estimate the ecosystem service value of specific coastal ecosystems is developed.
Received 31 March 2014 Specific variables are identified and used to develop a global multivariate regression function that
Received in revised form supports the identification of important drivers of the value of ecosystem service of coastal protection
30 October 2014
around the world, and the Caribbean is examined in detail. Variables hypothesized to affect the
Accepted 22 November 2014
ecosystem service value fall into three categories, and were informed by a meta-analysis of existing
economic literature. Site characteristics include ecosystem type and size. Study characteristics include
Keywords: valuation method. Context variables include measures of development, anthropogenic pressures,
Benefit transfer biodiversity, and population density. Results of the meta-analytic regression show that variables
Coast
significantly affecting the ecosystem service value included size, level of development, storm frequency,
Ecosystem service
valuation method and gross domestic product per capita. A benefit transfer function is then generated to
Meta-analysis
Shoreline protection extrapolate values to other sites around the world where coastal wetlands, mangrove and coral reefs
Spatial analysis exist. This function is used to derive a global map of the value of a set of coastal ecosystem services
worldwide. The Caribbean region is discussed as a case study in this global analysis.
& 2015 Published by Elsevier B.V.

1. Introduction (Brander et al., 2012). This is particularly important in the case

One third of the world's population lives in coastal commu-
nities and coastal zones are twice as densely populated as inland
areas (MA, 2005; Barbier et al., 2008). Although coastal commu-
nities may interact in very different ways with the natural
ecosystems they are located in and surrounded by, it is increas-
ingly well understood that natural ecosystems play a crucial role in
determining the well-being of human populations (TEEB, 2010).
Thus, preserving the continuous flows of benefits is increasingly
recognized as one of the most important catalysts for the con-
servation and sustainable management of natural ecosystems
(Chan et al., 2006; Turner et al., 1998; Fisher et al., 2009).
Coastal wetlands, mangroves and near-shore coral reefs pro-
vide crucial benefits to many coastal communities by protecting
them from flooding and storm surges, both seasonal and idiosyn-
cratic storm events. The benefits from this ecosystem service may
include prevention of loss of life, damages to housing, infrastruc-
ture and food sources, as prevention of saltwater intrusion
n Capturing the economic value of ecosystem services in a
Corresponding author at: Electric Power Research Institute (EPRI), 3420 Hill-
view Avenue, Palo Alto, CA 94304, USA. Tel.: þ 1 650 855 2044.
E-mail address: nrao@epri.com (N.S. Rao).
of poor, vulnerable communities, which recent research shows to
be often the most critically dependent on the provision of
ecosystem services (Ghermandi et al., 2013; McGranahan et al.,
2007). A widely accepted common ground in most definitions of
vulnerability is the identification of the inability to cope with
adverse effects, whether these are natural disasters, war, food
shortages, or others (McCarthy et al., 2001; Adger, 2009). Vulner-
ability may be linked to geographical limitations, which constrain
market access, for example. Communities which show in general a
more strong direct reliance on ecosystem services and the absence
of substitutes are more sensitive to impairment in their provision
(WRI, 2005; Ghermandi et al., 2013). One important way to
investigate vulnerability and dependence on coastal ecosystem
services is to examine their estimated values. While coastal
ecosystems offer a wide variety of benefits which are difficult to
quantify, making estimations of the value can lead to under-
standing the drivers of the high and low values, and in turn
inform policy. While this is a global analysis, it can be used for
informing regional policy as well, especially in areas which are
data poor.
specific monetary unit or welfare measure is a challenging ende-
avor. Although valuation research is an active area, the lack of a

http://dx.doi.org/10.1016/j.ecoser.2014.11.011
2212-0416/& 2015 Published by Elsevier B.V.

Please cite this article as: Rao, N.S., et al., Global values of coastal ecosystem services: A spatial economic analysis of shoreline
protection values. Ecosystem Services (2015), http://dx.doi.org/10.1016/j.ecoser.2014.11.011i
2 N.S. Rao et al. / Ecosystem Services ∎ (∎∎∎∎) ∎∎∎–∎∎∎
standardized methodology and common type of unit makes
comparisons difficult. Ideally, the determination of the economic
values of services should be preceded by the biophysical assess-
ment of their availability, which in turn should be distinguished
from their overall provision or availability, to their actual use, in
the form of benefits that humans enjoy from the services. This is
however rarely the case, and for a number of reasons. First, some
ecosystem services are more easily and directly quantified than
others. Second, ecosystem services are inherently spatial and
assessments need to rely on modeling of their flow in space and
time (TEEB, 2010). As a general rule, biophysical assessments are
highly dependent on the status of scientific knowledge and data
availability, often relying on proxies to identify service provision,
as opposed to benefits. This is particularly true in the case where
there is lack of consensus on the best/ideal measurement units for
these services. Finding a common metric with which to compare a
wide variety of studies and management outcomes is however
crucial to inform policy decisions, and hence, monetary valuation,
even in the absence of biophysical assessment, can be useful as a
common language and framework in which the available informa-
tion can be analyzed and tradeoffs can be evaluated.
Further compounding the difficulty in quantifying and valuing
ecosystem services is the fact that many ecosystem services are
public goods and as a result, subject to a failure of the market to
reflect their economic value.1 This is the case, for instance, of the
service of shoreline protection, which, in spite of its significance to
human lives, is rarely quantified, valued or taken into account in
management decisions (de Groot, 1994). Careful quantification and
valuation of the services can lead to better informed policy
decisions. For that reason, and in an attempt to address different
policy questions, researchers have attempted the valuation of
services at different spatial scales of assessment, from local
(Badola and Hussain, 2005) to global scale (Costanza et al., 1997;
TEEB, 2010; de Groot et al., 2012; Ghermandi and Nunes, 2013).
These types of studies look at the value of a wide range of
ecosystem services, measuring them with a variety of different
methods, with the ultimate goal of alleviating the shortcoming
associated with the common exclusion of nature's values into
policy and decision-making.
This study is the first systematic attempt to examine and map
the economic value of shoreline protection as provided by three
major coastal ecosystem types – coastal wetlands, mangroves and
coral reefs—worldwide. We examine the available information on
their economic value and investigate their dependence from a
series of study-, site- and context-specific driving factors in a
geographically explicit manner by means of meta-analysis. Identi-
fying and understanding the drivers of the ecosystem services
values for specific services is a first step towards their eventual
integration into governmental policies and accounting. Next, the
results of the meta-analytic regression are used to infer an
estimate of the value of ecosystem services in other areas for
which direct valuation has not been performed and produce a map
of the worldwide distribution of shoreline protection values from
the considered ecosystems. Finally, we discuss how the results of
this study can be interpreted for the Caribbean region, with
specific respect to coastal conservation and development policy.
1.1. The services of mangroves, coral reefs and wetlands
One ecosystem service provided by mangroves, coral reefs and
coastal marshes in wide coastal areas worldwide devoid of other,
1
The public good nature of certain ecosystem services implies the property of
non-exclusivity, in which there is no possibility for one user to preclude another's
usage of the service, and the property of non-rivalry, in which use by one does not
result in insufficient goods or services for another.
Please cite this article as: Rao, N.S., et al., Global values of coastal ecosystem services: A spatial economic analysis of shoreline
protection values. Ecosystem Services (2015), http://dx.doi.org/10.1016/j.ecoser.2014.11.011i
artificial defenses, is storm regulation or coastal protection. All
three ecosystems mitigate the full effect of the storm surges,
slowing or preventing them from reaching coastal human
populations.
In general, when reefs and mangroves are damaged or
destroyed, the absence of this natural barrier has been shown to
increase the damage to coastal communities from normal wave
action and violent storms. This storm protection that coastal
ecosystems provide prevents both the loss of life and property
for communities living in near-shore areas. The roots of mangrove
plants help to hold the sediment in place (Orth et al., 2006).
Mangrove forests protect inland communities and freshwater
resources from saltwater intrusion during storms, and they protect
near shore settlements from erosion (Semesi, 1998; Badola and
Hussain, 2005). The root systems of mangroves prevent the
resuspension of sediment and slow water flow in areas where
the protection of shoreline-based activities are important
(Spaninks and van Beukering, 1997; Gilbert and Janssen, 1998).
Mangroves protect areas from storms, have some recreational and
fishing service value, and protect water quality (Aburto-Oropeza
et al., 2008). In general, mangroves serve as “natural barriers” to
protect life, infrastructure and property of coastal communities
(Badola and Hussain, 2005). In addition the protection of property
and infrastructure will indirectly benefit the tourism and recrea-
tion industries, but this indirect effect is not measured in this
study. Valuation of mangroves has primarily focused on their
storm protection services, though there is a small growing
literature on the direct uses of mangroves. In major storms,
research has shown that coastal communities experienced greater
damage and higher mortality rates from many types of natural
disasters when mangroves had been removed, and the value of
these damages ranges widely (Danielsen et al., 2005; Barbier,
2007a, 2007b; Das and Vincent, 2009).
Coral reefs (and mangroves) also minimize the impact of
storms by reducing wind action, wave action, and currents and
coral reef structures buffer shorelines against waves, storms and
floods (Done et al., 1996; Moberg and Folke, 1999; Adger et al.,
2005). In general, the structure of coral reefs provides a significant
barrier to storm surges (UNEP-WCMC, 2006). They are increasingly
under human and climatic threat to due to water pollution, sea
temperature rise and ocean acidification (Bruno and Bertness,
2001), and regional studies have shown that the threats that coral
reefs are facing affect their ability to provide ecosystem services
(Bruno and Selig, 2007). Coral reefs are generally undervalued due
to their open-access nature and to the fact that the ecosystem
service of storm protection they provide is a (quasi) public good,
and as a result, often disregarded in policy and decision-making
contexts (Brander et al., 2007). Economic studies on coral reefs
have included their diverse uses, which include direct uses such as
fishing and diving, as well as indirect uses such as storm
protection.
Wetlands found in coastal areas also function as storm buffers
distinctly from how open water or land dissipates storm energy
(Simpson and Riehl, 1981). This mechanism reduces the area of
open water over which wind can form waves and simultaneously
decreasing the storm surge, and absorbing the energy of waves
(Costanza et al., 2006; Costanza et al., 2008). Moreover, coastal
wetlands provide a physical barrier for storm protection. They also
serve functions such as water purification, habitat for birds and
fish, as well as the prevention of saltwater intrusion from sea
water. Wetland ecosystem service studies have been performed
more extensively than those from the previous two ecosystems,
perhaps due to the wide range of services they provide, as well as
possibly because they serve as a link of freshwater—marine
systems. Research has been performed on a site scale, and several
meta-analyses have been recently published (Brouwer et al., 1999;
 N.S. Rao et al. / Ecosystem Services ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Woodward and Wui, 2001; Brander et al., 2006; Ghermandi et al.,
2010). This study adds value to the current meta-analysis literature
in two ways. First, it provides a first global and spatially explicit
analysis of the economic value of coastal ecosystems for shoreline
protection, which may constitute an important tool in the hands of
coastal and marine planners and managers. Second, the integra-
tion with Geographic Information Systems helps to put the values
elicited in primary valuation studies into their context, a dimen-
sion which is often lacking in primary valuation study reporting.
1.2. Valuation and meta-analysis
Common methods for assessment of shoreline protection
values are based on market prices, revealed preference analysis,
stated preference surveys, avoided damage, and replacement cost.
Different methods are applicable to different ecosystem services,
and the results of different valuation techniques will differ
depending on the welfare measure used (Freeman, 2003), and
some methods do not have a sound foundation in welfare theory,
which may lead to either over- or underestimation (Brander et al.,
2006). In the shoreline protection literature, avoided damages and
replacement cost are very common methods used for shoreline
protection studies (Vo et al., 2012).
Many site level studies have examined the value of shoreline
protection provided by coral reefs, mangroves, and coastal wet-
lands around the world, often coming up with very wide ranges of
values.2 The statistical analysis of this wide range of studies and
research outcomes through meta-regression analysis can be a
useful tool to help pointing out the principal drivers of value for
this ecosystem service. Although such exercise cannot in general
replace the more precise and site-specific information that can be
obtained from a primary valuation study, it can provide the policy-
and decision-makers with a suitable framework within which to
evaluate the complex multi-dimensional aspects involved in the
management of these coastal ecosystems. Moreover, the meta-
analytical framework can be used as a cost-effective tool to draw
generalized conclusions and transfer ecosystem service value
estimates to policy sites for which primary valuation studies are
not available. This is often the case in less-developed regions,
including those where ecosystems such as mangroves and coastal
wetlands are more abundant. There is by now a large number of
studies that have used meta-regression analysis of ecosystem
values for such purposes (for a comprehensive overview see
Nelson and Kennedy, 2009).
2. Materials and methods
2.1. Collection of primary valuations of shoreline protection
A wide variety of studies were examined during the phase of
literature review for this study. The studies were focused on
shoreline protection by coral reefs, mangroves and wetlands.3
The studies used were all economic valuation studies which were
associated with a specific site. The studies were from a variety of
coastal sites, including studies on marine protected areas (MPAs).
From the 100 studies collected using Web of Science and EconLit,
54 studies were selected for the meta-analysis based on having
sufficiently detailed, site-specific primary ecosystem service valua-
tion methodologies. The remaining studies were discarded,
because they lacked site-specific economic values, which were
2
See for instance the case of coral reefs values in the TEEB Ecosystem Services
Valuation Database (de Groot et al., 2012).
3 5
In this study, the definition of wetland is adopted from Ramsar Convention,
and specifically those in marine areas (Ramsar Convention, Articles 1.1 and 2.1).
Please cite this article as: Rao, N.S., et al., Global values of coastal ecosystem services: A spatial economic analysis of shoreline
protection values. Ecosystem Services (2015), http://dx.doi.org/10.1016/j.ecoser.2014.11.011i
3
necessary for the benefit transfer. From the 54 studies, 92
observations were generated. The 92 observations were distrib-
uted rather homogeneously among the three ecosystem types (30
observations for mangroves, 36 for coral reefs, and 23 for coastal
wetlands). In total, studies from 27 countries are included. The
studies are listed in Appendix A.
2.2. Explanatory Variables
The dependent variable is the economic value of coastal
ecosystem measured by the natural log of 2003 US dollars per
hectare per year, and we corrected for inflation using the Purchas-
ing Power Parity (PPP) index (see Woodward and Wui, 2001;
Brander et al., 2006; TEEB, 2010; Ghermandi et al., 2010). When
the original study did not present the data in such terms, then it
was standardized based on information given in the paper itself.
The value was estimated in per area units to account for the fact
that some of the studies used presented total values and others
presented marginal values. Performing the analysis using wetland
values in per area terms assumes the marginal wetland value and
the average wetland value are equal (Brander et al., 2006).
The independent variables are classified within three cate-
gories: study variables, site variables, and contextual variables. A
variety of different variables were chosen and tested for each
category, though the final regression expression contains a subset
of the list of explanatory variables. Variables were chosen because
of their perceived relevance to the value of the ecosystem service,
their importance in other similar ecosystem service value meta-
analyses, or their importance in terms of biophysical or socio-
economic effects on natural systems.
Study-specific variables include the valuation method used in
the primary study (i.e., replacement cost, damage cost avoided,
contingent valuation and market price). Binary variables were
included for each of the methods in order to control for metho-
dological heterogeneity among the observed values (Carson et al.,
1996; Bateman et al., 2004). The year of the study's publication
and the year the study was conducted were also included in the
development of the regression, in order to capture any temporal
issues that affected valuation studies, such as weather or disaster
related events, or any large scale demographic and developmental
changes.
Site-specific variables used in the regression analysis include
the size of coastal ecosystem, the ecosystem type and the latitude
and longitude of the center point of the study site. The information
on ecosystem size was taken directly from what was reported in
the study. Latitude and longitude of the study site were collected
from the reported coordinates of the study, and from the spatial
data source GeoHack4 if the original study provided the name of
the location but did not provide the coordinates.
Context-specific variables include Gross Domestic Product
(GDP) per capita, corrected for purchasing power parity (PPP) for
each study country, population of the study area as defined in the
study, a development index and the average monthly temperature
in the cell of the study site.5 The GDP per capita in PPP is collected
from CIA World Fact Book and adjusted to 2003 US$ using the
Penn World Table (Heston et al., 2006). Income measures have
been found to be important in terms of estimating the value of
natural resources and ecosystem services, and have been included
and found significant in many meta-analyses of ecosystem service
values. The human development intensity measures were
4
GeoHack is a web-based toolserver which links to maps on the internet. It is a
geographic information system extension. http://toolserver.org/  geohack/geo
hack.php?params=0_0_0_N_0_0_0_W
The cell size was 0.51  0.51 uniformly for each of the explanatory
variables used.
4 N.S. Rao et al. / Ecosystem Services ∎ (∎∎∎∎) ∎∎∎–∎∎∎
estimated using the results of the GLOBIO3 model which average
Mean Species Abundance—the relationship between development
pressures and species abundance—to categorize the impacts of
infrastructure on biodiversity, and summarize them at a 0.51
resolution (Alkemade et al., 2009). In this index, biodiversity was
defined as “the remaining mean species abundance, MSA, of
original species, relative to their abundance in pristine or primary
vegetation which are assumed to be not disturbed by human
activities for a prolonged period” (Alkemade et al., 2009). We used
this explanatory variable to examine whether infrastructure
development along coastlines would have an impact on the value
of the ecosystem service. Specifically, we were exploring whether
or not the presence of high, medium and low impact development
intensity6 would affect the value of the nearby ecosystem service.
The GLOBIO3 model data is spatially explicit and we used the
values for the specific coastal zones. The temperature was esti-
mated from the MOM4-TOPAZ Earth System Model, version 2.1
(IPCC SRES A2 scenario) from the Geophysical Fluid Dynamics
Laboratory.7 Storm frequency and wind speed were determined
for each point using meteorology data from the International Best
Track Archive for Climate Stewardship (IBTrACS) dataset (Knapp
et al., 2010). Population density of the cells containing the study
sites were obtained from Center for International Earth Science
Information Network (CIESIN; CIESIN, 2005). Regional variables
are introduced in the model to capture fixed regional effect, such
as historical, cultural, demographic, geographic characteristics. The
27 countries in the database are divided into five regions, Africa,
Asia, North America, Caribbean, and Oceania.
2.3. Meta-analytic regression: Single and multivariable analysis
Single variable regressions are conducted to test whether the
economic value of the coastal ecosystem is correlated with the size
of ecosystem, population and GDP per capita of the study area. A
multivariable analysis was also conducted using an Ordinary Least
Square (OLS) model. Explanatory variables were selected using
stepwise removal starting from the whole set of variables
described in the previous section. Various model specifications
were tested, including testing for interactions between the eco-
system type and the ecosystem service. According to the BOX–COX
test, the LOG–LOG functional form is the best fit for this study. A
regression with robust standard errors was performed to account
for heteroskedasticity (p o0.05) and deviation of the distribution
of residuals from normality. Multicollinearity was not found to
affect the model (max VIF o 10).
The OLS model for meta-analytic regression is defined in Eq. (1):
ln Y i ¼ α þ β1 ln X 1i þ β2 ln X 2i þ β3 ln X 3i þ β4 X 4i þ εi
where: Yi represents the economic value of coastal ecosystem
measured by dollar per hectare per year (2003 US$); X1i, X2i, X3i
and X4i are, respectively, the vectors for study-, site-, context-
specific characteristics, and regional binary variables.8 The coeffi-
cients to be estimated are β1, β2, β3 and β4 and the error term is
denoted by εi. The explanatory variables described above were all
included in the regression and then the model was improved by
removing the insignificant variables one-by-one to determine an
6 analysis, observations from Indonesia, the United States, and
The low, medium, and high impact development indices refer to the intensity
of infrastructure and general development pressure on the area. An area of low
impact development has a low level of infrastructure and development intensity.
7
Temperature was initially included because we hypothesized that the
temperature of the sea surface close to the ecosystem might affect the ecosystem
function, and thus the ecosystem service valued—especially with respect to coral
health, which has been found to suffer due to thermal stress (Selig et al., 2010).
8
These regional binary variables included Asia, Africa, North America, Car-
ibbean, Europe, and Oceania, and the Caribbean was used as the base.
Please cite this article as: Rao, N.S., et al., Global values of coastal ecosystem services: A spatial economic analysis of shoreline
protection values. Ecosystem Services (2015), http://dx.doi.org/10.1016/j.ecoser.2014.11.011i
expression of best fit. To evaluate the out of sample forecast
accuracy of the model we first used the n-1 data splitting technique.
2.4. Benefit transfer
After data-processing, the meta-regression model described
above was applied as a benefit transfer value function to provide
an estimate of the values of the investigated ecosystem services in
areas where primary valuation results are not available. In the
context of the valuation of ecosystem goods and services, value
transfer refers to the procedure of drawing inferences on the
unobserved monetary value of an ecosystem good or service in a
policy site by borrowing existing valuation estimates from com-
parable study sites. The reasons for performing benefit transfers
are varied, as are the potential pitfalls, which are described in the
following sections. However, for areas which are data poor, which
describe the majority of coastlines, particularly in lower income
countries, benefit transfer methods are one option to estimate the
value of ecosystem services in different areas. The advantages
provided by benefit transfer include that it is relatively rapid to
assess, compared with collecting information at the sites and
hence it is also less expensive (Hanley et al., 2006).
The actual benefit transfer procedure was performed according
to the following steps. The spatially explicit, continuous, explana-
tory values used in this analysis were then calculated for all coastal
areas, using the global data sets available. To create a spatial
dataset to which we would transfer our regression function, we
used the following existing datasets for each ecosystem and
processed them to form a spatial layer where at least one of the
ecosystem types was present. The layer covers the land area
adjacent to the ecosystem, cell area was at 0.51, in order to include
areas where human communities live and access the resources,
and the current available resolution of the spatial information at
the global scale prevented working with a more detailed spatial
scale. For the mangrove ecosystem type, we used the global
dataset showing the distribution of mangroves compiled by
UNEP-WCMC in collaboration with the International Society for
Mangrove Ecosystems (ISME—Spalding et al., 1997). For the coral
reef data we used spatial data from 1997 from United Nations
Environment Program—World Conservation Monitoring Center
(Spalding et al., 2001; United Nations Environment Program—
World Conservation Monitoring Center (2003)). For the wetlands
data, we used the data described in the category of “coastal
wetlands” in the Global Lakes and Wetlands Database (GLWD;
Lehner and Döll, 2004). Then the remaining data sets listed above
were clipped to the cover the same areas as the ecosystem layer
created. For the non-spatial regional information, a proportion
ð1Þ based on the studies used in the meta-analytic regression was
calculated and included. The size of the ecosystem was calculated
using the average size of ecosystems in the dataset. The benefit
transfer function was run using a raster calculator function in
ArcGIS 9.0, and the values were transferred to areas delineated by
the ecosystem type layer created above.
3. Results: general results
In terms of characteristics of the studies used in the meta-
Thailand occur the most frequently, at 17, 18, and 7 times,
respectively. Fifty observations come from Asia, eighteen observa-
tions from the Americas, twelve observations from Caribbean,
seven observations from Oceania, three from Africa and two from
Europe. Thirty-four of the observations come from peer reviewed
journals, and 58 are from “gray literature,” working papers or
technical reports. Based on single correlations, and noting that the
 N.S. Rao et al. / Ecosystem Services ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5

Ecosystem Service Value (ln $/ha/year 2003)

values are on a per-hectare basis, in the larger ecosystems the 15
average per-hectare value is lower than in small ecosystems
(Fig. 1). The GDP per capita has a positive and significant correla-
tion with the economic value of coastal ecosystem (Fig. 2). The
10
population density of the study area does not significantly corre-
late with the economic value of coastal ecosystem, with the slope
and R2 close to zero (Fig. 3).
5
3.1. Econometric results

Econometric results of the OLS regressions are reported in

0
Table 1, with the model chosen for a benefit transfer denoted as
Model 1. Model 1 is the preferred model, with an R2 of 0.45 7 8 9 10 11

(F(12,77)¼7.26; Prob4 F¼0.00). The statistical result of the F-test GDP (ln)

rejects the null hypothesis and suggests that all the coefficients are lny Fitted values
R-squared=0.12
jointly significantly different from zero. Given the LOG–LOG
functional form, the coefficients of each continuous independent Fig. 3. Single correlation between the economic value and the GDP per capita
in GDP.
variable refer to the elasticity, the percentage by which the
dependent variable changes with respect to a one percent change
in the independent variable.
The coefficient for the valuation method variable was insignif- service values. The size of ecosystem has a negative and significant
icant for replacement cost and damage cost avoided, but signifi- impact on the economic value of coastal ecosystem. The economic
cant for contingent valuation (at the 1% level) when compared to value of one hectare of coastal ecosystem is expected to decrease
the category which included market price valuations. When by 0.32% with respect to a one percent increase in the size of
compared to using market prices, using replacement cost or ecosystem. Mangroves have a negative and significant impact on
damage cost avoided methods does not bring a different impact ecosystem service value, while coral reefs have a negative but not
on the economic value of coastal ecosystem, however, using significant impact. Real GDP per capita affects the economic value
contingent valuation will yield significantly higher ecosystem of coastal ecosystem positively, and significantly (p ¼0.007), which
confirms the results of the single variable correlations. The
regression results show that the study sites which have a low
Ecosystem Service Value (ln $/ha/year 2003)

15
impact development index from the GLOBIO3 model have lower
economic value than the study sites having medium or high levels
of development. Several of the datasets which had been processed
10 and used for the analysis were not included in the final model
specification, including the regional indicator variables, the biodi-
versity indices, latitude/longitude, and the publication status
because they were insignificant. A second model was estimated,
5
and the results presented in Table 1. The objective was firstly to
test the robustness of the final model specification to another level
of development from the GLOBIO3 model, as well as to examine
0 whether the temperature variable was masking the significance of
0 5 10 15 the existing variables. The R2 values of models 1 and 2 were
Size (ln) similar, at 0.45 and 0.46, respectively, though specific coefficients
varied in significance between Model 1 and Model 2.
lny Fitted values
R-squared=0.08 The results of the benefit transfer performed using the coeffi-
Fig. 1. Single correlation between the economic value and the size of the cient estimated for Model 1 and are presented spatially in Fig. 4.
ecosystem. The results for the Caribbean region are highlighted in Fig. 5. The
final explanatory variables included in the meta-analytic regres-
sion of the study site were: valuation method, ecosystem type,
Ecosystem Service Value (ln $/ha/year 2003)

15 ecosystem size, GDP per capita, number of storms, wind speed,

average temperature, development impact, and wind speed and
ecosystem interaction terms. The values range from  1.0 to 7.6
((ln)2003 US$/ha/yr), which correspond to 0.4-1998.2 US$/ha/year
10
in 2003 and 0.51-2529.9 US$/ha/year in 2013.

5 4. Discussion

The negative and significant correlation between the economic

value of coastal ecosystem and the size of ecosystem, suggests the
0 presence of diminishing marginal returns to ecosystem size. The
5 10 15 20 reason for this finding could illustrate an issue of scarcity of the
Population Density (ln) ecosystem resource; the smaller the resource, the more it is valued in
terms of providing critical ecosystem benefits. Other studies, primarily
R-squared=0.00 lny Fitted values
of wetlands, have shown that the size of the ecosystem has a negative
Fig. 2. Single correlation between the economic value and the population density. correlation with the value (see Brander et al., 2006; Ghermandi et al.,

Please cite this article as: Rao, N.S., et al., Global values of coastal ecosystem services: A spatial economic analysis of shoreline
protection values. Ecosystem Services (2015), http://dx.doi.org/10.1016/j.ecoser.2014.11.011i
6 N.S. Rao et al. / Ecosystem Services ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Table 1
Meta-analysis regression results.

Model 1 Model 2

Category Variable Coef. P4t Coef. P4 t

Constant 2.9972 0.28 1.373022 0.569
Valuation method Replacement cost 0.56304 0.38 0.521758 0.402
Damage cost avoided approach 0.6435 0.237 0.744342 0.181
Contingent valuation 3.06231 0.002nnn 2.982701 0.001nnn
Site characteristics Size (log)  0.3156 0.001nnn  0.34026 0.001nnn
Mangroves  1.4726 0.067n  1.56847 0.058n
Coral reefs  0.6517 0.376  0.70467 0.362
Context variables GDP/capita (log) 0.77233 0.007nnn 0.679175 0.024nn
Low impact development  1.5153 0.011nn  1.00793 0.224
High impact development 0.598313 0.431
Average temperature (ln) 0.669321 0.064n
Storms (number) 0.05591 0.049nn 0.05378 0.024
Wind speed  0.0564 0.04nn  0.04062 0.233
Mangrove and wind speed 0.0256 0.431 0.004806 0.908
Coral reef and wind speed 0.07643 0.016nn 0.059393 0.109
Observations 90 90
R2 0.4454 0.4563
F (12,77) 7.26 8.40

n
Represents significant at 10%, respectively.
nn
Represents significant at 5%, respectively.
nnn
Represents significant at 1%, respectively.

Fig. 4. Spatial results of benefit transfer developed from meta-analytic regression function showing the global, coastal distribution of ecosystem service value in regions with
coral reefs, mangroves and coastal wetlands.

2010), but some studies have shown the opposite or constant result
(Woodward and Wui, 2001). We assume that the values of the
ecosystem service are spread constantly across the entire ecosystem
for this analysis. However, the hectares which are immediately near
the coastline may protect coastal communities more than the portions
of the ecosystems furthest from the communities. This distance issue
could affect the ecosystem service value of very large ecosystems.
The single variable correlation, suggests the higher the GDP per
capita, the higher the per-hectare per-year dollar value of the
coastal ecosystem. Countries with a higher per capita GDP might
value the resources at a higher level over and above the PPP
adjusted levels for a variety of reasons, including the quality of
infrastructure, development and livelihoods which are protected
from storms in these areas. Higher property values in some of
these areas could affect this result. Higher adjusted per capita GDP
contributing to higher ecosystem service value was mirrored in
the meta-analytic regression results, at a high level of significance.
The coefficient on the method of contingent valuation method
is large, positive, and significant, consistent with studies that show
that stated preference methods produce higher values than
replacement cost and avoided damages (see Jerath et al., 2012).
Storm frequency and wind speed were significant in the
analysis, positively increasing the ecosystem service value. This
result might indicate that the ecosystem service benefit of shore-
line protection is higher in areas where there is more risk of storm
damage from these physical drivers of storm damage. The inter-
action of wind speed and coral reefs was positive, which could
indicate that wind-driven erosion and strong waves lead to higher
ecosystem service values. In addition, it could be that coral reefs
protect higher value recreational beaches while mangroves are not
usually surrounding beach recreation areas. Mangroves them-
selves have a negative and significant effect on ecosystem service
value. Coral reefs and coastal wetlands may be surrounding areas
of higher value, more infrastructure, and higher GDP than
mangroves.
Globally, it has been estimated that windstorms have caused
considerable property damage, and these include hurricanes, cyclones
and tropical storms (Costanza et al., 2008; Arkema et al., 2013).

Please cite this article as: Rao, N.S., et al., Global values of coastal ecosystem services: A spatial economic analysis of shoreline
protection values. Ecosystem Services (2015), http://dx.doi.org/10.1016/j.ecoser.2014.11.011i
 N.S. Rao et al. / Ecosystem Services ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 7

Fig. 5. Spatial results showing the coastal distribution of ecosystem service value in regions with coral reefs, mangroves and coastal wetlands in the Caribbean.

The inclusion of the physical parameters of storm frequency and wind
speed is a unique contribution to this study, as these physical
parameters are directly mitigated by each of the coastal ecosystems
examined. Indeed the inclusion of storm frequency and wind speed in
a meta-analytic regression shows that the physical parameters have an
effect on the value of the ecosystem service, and this effect would not
be easily captured through primary valuation research. As tropical
storms and cyclones are a major factor for coastal communities,
especially countries in hurricane and storm zones, physical storm
information may become more important to integrate into valuation
research. As discussed above, the size of the ecosystem and wave
attenuation may be linked. Koch et al. (2009) note that the edges of
the biotic structures receive the majority of the wave energy and thus
provide the majority of wave attenuation. For this study, we assumed
linearity of the ecosystem service provision, but this assumption can
be explored further with a decay function, in order to determine how
the value may change with distance to the ecosystem edge. In addition
to size, the ecosystem service value may vary by season as well, and
some of this variation would be captured by the storm frequency
variable. Storm and hurricane frequency and impact is a major issue in
the Caribbean region, and will be discussed in Section 4.1.
Geographical indicator variables of region were not significant—
and in fact were correlated with GDP per capita—and hence were
excluded from the final analysis. As with similar meta-analyses (see
Brander et al., 2006), the distribution of sites is not a reflection of the
distribution of these ecosystems on the globe. Further, more than one
observation can be taken from the same study site, which further
compounds the effects of geography. Nevertheless, geographical
trends can be seen in the map representing the results of the benefit
transfer. Poorer areas of the world, as defined by GDP and low
development status, are associated with areas of lower ecosystem
service value in our model. Compared to high and medium develop-
ment sites, low impact development sites have lower values. High
development sites are well developed with hotels, houses and facil-
ities. Therefore, the dollar value of storm protection of ecosystem is
higher because the value at risk is higher and in addition replacement
cost or damage avoided approach have been used. While coastal
populations continue to grow, which fuels development on coasts, the
cost of this development varies greatly between different locations. In
general some types of coastal infrastructure increase the risks of
flooding. In studies of coastal zones, there are some population effects
which might be affecting the geographic results. For example the
Small Island States have a similar level of urban percentage in the
coastal zone as Africa (13%), but have only a slightly higher total
population in the coastal zone (16%) as world averages (13%;
McGranahan et al., 2007). Of course these States have a higher
proportion of land itself in the coastal zone, and are at generally low
development status and a lack of resources inland, so will have a
difficult time building resilience to storm events. Conversely, Asia has
very large populations in delta regions despite having a relatively
small percentage of land in the coastal zone (McGranahan et al.,
2007); the large inland areas and stronger economies can help Asian
populations withstand storm effects, given similar reliance on ecosys-
tems as populations in Small Island States.

Please cite this article as: Rao, N.S., et al., Global values of coastal ecosystem services: A spatial economic analysis of shoreline
protection values. Ecosystem Services (2015), http://dx.doi.org/10.1016/j.ecoser.2014.11.011i
8 N.S. Rao et al. / Ecosystem Services ∎ (∎∎∎∎) ∎∎∎–∎∎∎
In terms of the benefit transfer, from Fig. 4 we can see some
geographic variation. Low values for ecosystem services can be seen
in most of the coastlines of Africa which are included. This may be
due to the low impact development variable from the GLOBIO3
model, which will drive the ecosystem service value lower. Indeed in
Africa and parts of Asia, the lower development might lead to low
values, especially if damage cost avoided is a main valuation
method, because the coastal infrastructure is less developed. Higher
values can be seen in the parts of North America, Europe the Middle
East and Australia. These values are likely driven by GDP and
development, which affects the ecosystem service value positively.
Intermediate values are observed in the Caribbean, South and
Southeast Asia, driven by a combination of factors, such as varied
GDP per capita values, and development levels, and presence of the
different ecosystem types, which interact in opposite directions.
Additionally, perhaps whether or not the site is located on an island
might affect the ecosystem service value, because the island nations
seem to present intermediate values compared to the values of
mainland continents. As noted above, the interaction between
proportion of land in coastal zone, population in coastal zone and
urban population in coastal zone might affect the results of the
regression and benefit transfer.9 Finally, given the results of the
regression it was expected that the highest values are found in
developed countries with high GDP per capita and high level of
coastal development. Contrary to expectations we did not find a
statistically significant effect of population density on the values;
this may be due to the fact that coastal protection values will
primarily be felt by the population living within a few hundred
meters from the coast, but given the current resolution of the
underlying GIS layers, we are unable to capture the population
density at such level of detail. Finally, the notable contribution of this
benefit transfer is to estimate ecosystem service values in policy
sites where shoreline protection studies have not been performed.
However, we do note that the map estimates must be applied with
care in regions where there are fewer studies, such as Africa or
Europe in our study.
4.1. The Caribbean region
The result of the meta-analytic regression and benefit transfer
for the Caribbean region is shown in Fig. 5. The variation and
trends across the Caribbean are similar to the variation around the
world, and broadly due to reasons outlined above in Section 3.1.
This region is a natural example of a subset of the global analysis,
as it contains a large proportion of the studied ecosystems,
seasonal hurricanes, and a high proportion of coastal zones per
country. Additionally, the Caribbean region has been the subject of
many ecosystem service valuations to date (Schuhmann, 2012;
MESP, 2013), some of which were used in this study. Hence, the
Caribbean region offers a unique location to examine how the
results of this research can be integrated into policy. Indeed, in
most cases, ecosystem service valuations are performed for a
specific policy reason, such as to estimate compensation or the
effect of a regulation. The purposes in the Caribbean are similar,
where ecosystem service valuation has recently been examined in
the context of decision-making in the Caribbean (Waite et al.,
2014). With respect to shoreline protection policy, one example of
9
The analysis would be improved by using a decay function for the estimated
ecosystem service capacity based on the distance of the ecosystem from the
community site. In the future, the relationship could be examined using finer scale
income measures such as the GCP data (Nordhaus, 2006) within the regression
analysis, or local, municipal-level coastal income indicators could be used. It is
possible also that the GDP of the entire country per capita might not reflect
accurately those areas which depend on ecosystem services, and not engineering
solutions, for storm protection.
Please cite this article as: Rao, N.S., et al., Global values of coastal ecosystem services: A spatial economic analysis of shoreline
protection values. Ecosystem Services (2015), http://dx.doi.org/10.1016/j.ecoser.2014.11.011i
the use of valuation is to support potential national level marine
protection legislation in Belize (Clarke et al., 2013).
Shoreline ecosystem integrity, and hence the value of shoreline
protection, is highly affected by human pressure in the Caribbean. For
example, pressure is exerted through overfishing, development, pollu-
tion, and climate change on over 75% of Caribbean coral reefs (Waite
et al., 2014; Burke et al., 2011). Development along the coastal zone in
the Caribbean will affect the ecosystem service values in different
ways from our model results. Increasing development on the coastline
may decrease the extent of the ecosystem, and while we found
diminishing marginal returns to size of the ecosystem, the increase
in total size will affect ecosystem service value positively. Also, when
the coastal ecosystem size is diminished, then vulnerable communities
who depend on the coastal resource will be adversely affected. While
regional policy is more difficult to implement than country-level
policy, the importance of the coastal ecosystem service to protect
property from storm surge through wave attenuation, which has been
valued in several studies used in this meta-analysis, gives a good
incentive for environmental planners to preserve the size of the
existing habitat, which is a policy implication of our study. This is
especially true for countries in the eastern part of Central America, as
well as island countries, such as Haiti and the Dominican Republic,
where very low values of ecosystem service value were found.10
Policies which protect ecosystems from destruction or degra-
dation, especially in areas with lower ecosystem service values,
would allow those coastal residents who depend heavily on the
ecosystem protection. Habitat integrity may affect Caribbean
shoreline protection values, for example, pollution has increased
coral reef die-off in the region, and Barbados is one country which
instituted management to address this (DeGeorges et al. 2010).
Increased development might increase the estimated ecosystem
service value of the area, based on the regression results (though
not significantly). Policy-makers could examine development,
conservation, and size of the ecosystem when designing policies
with respect to shoreline protection ecosystem service.
In the context of decision-making and policy in the coming
years, climate change and climate change adaptation options will be
very important. Jones et al. (2012) found that ecosystem-based
adaptation in coastal zones is a critical component to protect coastal
communities from climate change impacts. While the results of this
study used the methods described in Section 2, there are comple-
mentary modeling efforts using either process-based, engineering,
or geospatial information that can be integrated with a meta-
analytic regression analysis approach to inform policy at a lower
spatial resolution. For example, there have been modeling efforts to
estimate damages from sea level rise using geospatial and economic
information (Simpson et al., 2010), general climate change impacts
on Caribbean countries (Haites et al., 2002).
4.2. General policy implications
The study has implications for policy in terms of coastal
populations. In Southeast Asia, coral reefs and mangroves are
extensive, yet the model predicted low to intermediate ecosystem
service values, adjusted for 2003 levels. In future studies, it might
be important to explore the implications of the level of the impact
10
Policies which protect ecosystems from destruction or degradation, espe-
cially in areas with lower ecosystem service values, would allow those coastal
residents who depend heavily on the ecosystem protection. Habitat integrity may
affect Caribbean shoreline protection values, for example, pollution has increased
coral reef die-off in the region, and Barbados is one country which instituted
management to address this (DeGeorges et al. 2010). Increased development might
increase the estimated ecosystem service value of the area, based on the regression
results (though not significantly). Policy-makers could examine development,
conservation, and size of the ecosystem when designing policies with respect to
shoreline protection ecosystem service.
 N.S. Rao et al. / Ecosystem Services ∎ (∎∎∎∎) ∎∎∎–∎∎∎
development index in at these study sites; populations in some
areas might not construct high value homes in developing coun-
tries, and this will affect the damage costs, for example, and have
implications for resilience to natural events. The combination of
meta-analytic regression and spatial benefit transfer is a method
which is less expensive than conducting a series of primary
studies, and hence used for environmental decision-making.
Further research to inform applications to environmental decision-
making must account for the combination of ecosystem services
and the ultimate effects on ecosystem service value; the impor-
tance of including finer scale income measures to understand
issues of vulnerability and resilience to ecosystem function; the
interaction between income, coastal settlements and development
levels; and how different information on storm frequency and
intensity effects ecosystem service value—and on what time scales.
5. Conclusions
This study presented the novel use of meta-analytic regression
and benefit transfer to examine issues of ecosystem service values
associated with coastal ecosystems, and the benefits that accrue to
coastal populations. After performing a meta-analytic regression
of the economic value of coastal and shoreline protection, the
results show that size, level of development, GDP, type of ecosys-
tem, wind speed, storm frequency, and method all affect the
economic value of the ecosystem service. Inclusion of different
types of such data may lead to better model prediction. Benefit
transfer was used to estimate values for ecosystem services in
coastal areas where mangroves, coral reefs and coastal wetlands
exist. The results were mapped spatially, and the different impor-
tant driving factors of different values were discussed as they
occurred spatially.
We expect that this type of study can be useful for both
furthering the research on ecosystem service provision, as well
as inform regional development and conservation policies. The
results could be ranked in terms of which coastal ecosystems
would be most important to conserve due to their importance for
direct and indirect use and the associated value estimates. The
baseline values could feed into coupled ecological and economic
modeling and/or inform policy makers for decision-making. The
pervasive losses of coastal ecosystems and their important con-
tribution to coastal protection, mostly associated with increasing
coastal populations, habitat loss, nutrient pollution and invasive
species, are likely to be compounded by climate related changes.
These climatic changes include increasing sea temperature and
ocean acidification, and the increase in storm frequency and
severity. With this baseline, we will be able to adjust the drivers
and test for the sensitivity of the values to climate variations or
shocks. Indeed the impact of climate change on coastal areas, and
their capacity to buffer coastal community from the effects of
storms, is an important aspect that policymakers can anticipate in
advance, and take measures to protect such ecosystems. The
general framework used in this global study could be used to
identify tradeoffs in terms of development and conservation, areas
for further regional analyses using finer-scale data, and examine
the impacts of climate change on the ecosystem service values.
Acknowledgments
The authors would like to thank Conservation International for
support of this work, as well as Leah Karrer and Emily Pidgeon.
The authors also thank Keith Alger, Paulo Nunes, Laura Onofri,
Elizabeth Selig, Fabiano Godoy, Andres Cano, Erica Ashkenazi,
Kellee Koenig, Karyn Tabor, Andre Boustany, John Fay, and Venetia
Please cite this article as: Rao, N.S., et al., Global values of coastal ecosystem services: A spatial economic analysis of shoreline
protection values. Ecosystem Services (2015), http://dx.doi.org/10.1016/j.ecoser.2014.11.011i
9
Hargreaves-Allen for methodological, technical and editorial
support.
Appendix A. Bibliography of studies used in the meta-analysis
Badola and Hussain (2005), Bann (1997, 1999), Barbier (2007b),
Berg et al. (1998), Burke et al. (2002, 2011), Carleton and Lawrence
(2005), Cesar (2002, 1996), Cesar et al. (2003, 2000), Christensen
(1982), Cooper et al. (2009), Costanza et al. (1989), de Lopez (2003),
Emerton and Kekulandala (2003), Farber (1987, 1996), Farber and
Costanza (1987), Gunawardena and Rowan (2005), Gustavson
(1998), Hargreaves-Allen (2004), Howorth (1982), IUCN (2006),
JacobsGIBB Ltd (2004), King and Lester (1995), Ledoux (2003),
Naylor and Drew (1998), Öhman and Cesar (2000), Pet-Soede et al.
(1999), Ranasinghe and Kallesoe (2006), Ruitenbeek and Cartier
(1999), Samonte-Tan et al. (2007), Sathirathai (1998), Sathirathai and
Barbier (2001), Seenprachawong (2002), Shabman and Batie (1987),
Spurgeon, (2002), Tri et al. (1996), van Beukering (2007, 2006), van
Beukering et al. (2003), and White et al. (2000a, 2000b)
References
Aburto-Oropeza, O., Ezcurra, E., Danemann, G., Valdez, V., Murray, J., Sala, E., et al.,
2008. Mangroves in the Gulf of California increase fishery yields. Proc Natl Acad
Sci USA 105 (30), 10456–10459.
Adger, W.N., 2009. Vulnerability. Glob. Environ. Change 16, 268–281.
Adger, W.N., Hughes, T.P., Folke, S.R., Carpenter, S.R., Rockstrom, J., et al., 2005.
Socialecological resilience to coastal disasters. Science 309, 1036–1039.
Alkemade, R., van Oorschot, M, Miles, L., Nellemann, C., Bakkenes, M., ten Brink, B.,
2009. GLOBIO3: A Framework to Investigate Options for Reducing Global
Terrestrial Biodiversity Loss. Ecosystems 12, 374–390.
Arkema, K.K., et al., 2013. Coastal habitats shield people and property from sea-
level rise and storms. Nat. Clim. Change 3 (10), 913–918.
Badola, R., Hussain, S.A., 2005. Valuing ecosystem functions: an empirical study on
the storm protection function of Bhitarkanika mangrove ecosystem. India
Environm. Conserv. 32 (1), 85–92.
Bann, C., 1997b. An Economic Analysis of Tropical Forest Land Use Options,
Ratanakiri Province. International Development Research Center, Cambodia.
Bann, C., 1999. A contingent valuation of the mangroves of Benut, Report for Johor
State Forestry Department/DANCED/Darudec. Johor State, Malaysia.
Barbier, E.B., Koch, E.W., Silliman, B.R., Hacker, S.D., Wolanski, E., Primavera, J.,
Granek, E., Polasky, S., Aswani, S., Cramer, L.A., Stoms, D.M., Kennedy, C.J., Bael, D.,
Kappel, C.V., Perillo, G.M., Reed, D.J., et al., 2008. Coastal ecosystem–based
management with nonlinear ecological functions and values. Science 319,
321–323.
Barbier, E.B., 2007a. Valuing ecosystem services as productive inputs. Econ. Policy
49, 177–229.
Barbier, E.B., 2007b. Do mangroves reduce expected storm damages? Valuing the
storm protection service of coastal wetlands. In: Proceedings of the 15th
Annual Conference of the EAERE Thessaloniki, Greece.
Bateman, I.J., Lovett, A.A., Brainard, J.S., 2004. Applied environmental economics: A
GIS approach to cost-benefit analysis. Cambridge University Press, Cambridge.
Berg, H., Öhman, M.C., Troeng, S., Lindén, O., 1998. Environmental economics of
coral reef destruction. in Sri Lanka. AMBIO 7 (9), 627–634.
Brander, L.M., RJGM, Florax, Vermaat, J.E., 2006. The empirics of wetland valuation:
A comprehensive summary and a meta-analysis of the literature (doi: 101007/
s10640-005-3104-4). Environ Resour. Econ.
Brander, L.M., van Beukering, P., Cesar, H.S.J., 2007. The recreational value of coral
reefs: a meta-analysis. Ecol. Econ. 63 (1), 209–218.
Brander, L.M., Wagtendonk, A.J., Hussain, S.S., McVittie, A., Verburg, P.H., de Groot, R.S.,
van der Ploeg, S., 2012. Ecosystem service values for mangroves in Southeast Asia: a
meta-analysis and value transfer application. Ecosyst. Services 1 (1), 62–69.
Brouwer, R., Langford, I.H., Bateman, I.J., Turner, R.K., 1999. A meta-analysis of
wetland contingent valuation studies. Reg. Environ. Change 1 (1), 47–57.
Bruno, J.F., Bertness, M.D., 2001. Habitat modification and facilitation in benthic
marine communities. In: Bertness, M.D., Gaines, S.D., Hay, M.E. (Eds.), Marine
Community Ecology. Sinauer, Sunderland, MA.
Bruno, J.F., Selig, E.R., 2007. Regional decline of coral cover in the Indo-Pacific:
Timing, extent, and subregional comparisons. PLoS One 2 (8), e711.
Burke, L., Selig, E., Spalding, M., 2002. Reefs at risk in Southeast Asia. World
Resource Institute, Washington DC.
Burke, L., Reytar, K., Spalding, M., Perry, A., 2011. Reefs at Risk Revisited. World
Resources Institute, Washington, DC.
Carleton, C., Lawrence, K.S., 2005. Economic Valuation of Environmental Resource
Services in the Turks and Caicos Islands. Nautilus Consultants Ltd, Peebles, UK.
Carson, R.T., Flores, N.E., Martin, K.M., Wright, J.L., 1996. Contingent valuation and
revealed preference methodologies: Comparing the estimates for quasi-public
goods. Land Econ. 72, 80–99.
10 N.S. Rao et al. / Ecosystem Services ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Cesar, H.J.S., 2002. The Biodiversity Benefits of Coral Reef Ecosystems: Values and
markets. OECD, Paris.
Cesar, H.J.S., 1996. Economic Analysis of Indonesian coral reefs. Working
Paper Series Work in Progress World Bank, Washington DC.
Cesar, H.J.S., Burke, L., Pet-Soede, L., 2003. The Economics of Worldwide Coral Reef
Degradation. Cesar Environmental Economics Consulting.
Cesar, H.J.S., Öhman, M.C., Espeut, P., Honkanen, M., 2000. An Economic Valuation
of Portland Bight, Jamaica: An Integrated Terrestrial and Marine Protected Area.
Working paper 00/03. Institute for Environmental Studies, Free University,
Amsterdam.
Center for International Earth Science Information Network (CIESIN), Centro
Internacional de Agricultura Tropical (CIAT), 2005. Gridded population of the
world version 3 (GPWv3): Population density grids. Socioeconomic Data and
Applications Center (SEDAC), Columbia University Palisades, NY. Available at
〈http://sedac.ciesin.columbia.edu/gpw〉. (10.01.2011).
Chan, K.M.A., Shaw, M.R., Cameron, D.R., Underwood, E.C., Daily, G.C., 2006.
Conservation planning for ecosystem services. PLoS Biol. 4 (11), e379.
Christensen, B., 1982. Management and utilization of mangroves in Asia and the
Pacific, FAO Environment Paper No.3, Rome. 160 pp.
Clarke, C., Canto, M., Rosado, S., 2013. Belize integrated Coastal Zone management
plan. Coastal Zone Management Authority and Institute (CZMAI), Belize City.
Cooper, E., Burke, L., Bood, N., 2009. Coastal capital: Belize, The economic
Contribution of Belize's Coral Reefs and Mangroves. World Resource Institute,
Washington, DC.
Costanza, R., Mitsch, W.J., Day Jr, J.W., 2006. A new vision for New Orleans and the
Mississippi Delta: applying ecological economics and ecological engineering.
Front. Ecol Environ. 4, 465–472.
Costanza, R., Perez-Maqueo, O., Martinez, M.L., Sutton, P., Anderson, S.J., Mulder, K.,
2008. The value of coastal wetlands for hurricane protection. Ambio 37 (4),
241–248.
Costanza, R., d'Arge, R., deGroot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K.,
Naeem, S., Paruelo, J., Raskin, R.G., Sutton, P., vandenBelt, M., 1997. The value of
the world's ecosystem services and natural capital. Nature 387, 253–260.
Costanza, R., Farber, S.C., Maxwell, J., 1989. Valuation and management of wetland
ecosystems. Ecol. Econ.1, 335–361.
de Groot, R.S., 1994. Environmental functions and the economic value of natural
ecosystems. In: Jansson, A., Hammer, M., Folke, C., Costanza, R. (Eds.), Investing
in Natural Capital: the Ecological Economics Approach to Sustainability. Island
press, Washington DC (504 pp.).
de Groot, R., Brander, L., van der Ploeg, S., Costanza, R., Bernard, F., Braat, L., Christie, M.,
Crossman, N., Ghermandi, A., Hein, L., Hussain, S., Kumar, P., McVittie, A., Portela, R.,
Rodriguez, L.C., ten Brink, P., van Beukering, P., 2012. Global estimates of the value
of ecosystems and their services in monetary units. Ecosyst. Serv. 1, 50–61.
Das, S., Vincent, J.R., 2009. Mangroves protected villages and reduced death toll
during Indian super cyclone. Proc. Natl. Acad. Sci. USA 106 (18), 7357.
Danielsen, F., Sorensen, M.K., Olwig, M.F., Selvam, V., Parish, F., Burgess, N.D.,
Hiraishi, T., Karunagaran, V.M., et al., 2005. The Asian tsunami: a protective role
for coastal vegetation. Science 310, 643.
de Lopez, T.T., 2003. Economics and stakeholders of Ream National Park, Cambodia.
Ecol. Econ. 46, 269–282.
DeGeorges, A., Goreau, T.J., Reilly, B., 2010. Land-Sourced pollution with an
emphasis on domestic sewage: lessons from the caribbean and implications
for coastal development on Indian Ocean and Pacific coral reefs. Sustainability
2, 2919–2949.
Done, T.J., Ogden, J.C., Wiebe, W.J., Rosen, B.R., 1996. Diversity and ecosystem
function of coral reefs. In: Mooney, J.H. (Ed.), Functional Roles of Biodiversity: A
Global Perspective. John Wiley, New York.
Emerton, L. and Kekulandala, L.D.C.B., 2003. Assessment of the economic value of
Muthurajawela Wetland. Occasional Papers of IUCN. Colombo, Sri Lanka.
Farber, S., 1987. The value of coastal wetlands for protection of property against
hurricane wind damage. J. Environ. Econ. Manag. 14 (2), 143–151.
Farber, S., 1996. Welfare loss of wetlands disintegration: A Louisiana study.
Contemp. Econ. Policy 14, 92–106.
Farber, S.C., Costanza, R., 1987. The economic value of wetlands systems. J. Environ.
Manag. 24, 41–51.
Fisher, B., Turner, R.K., Morling, P., 2009. Defining and classifying ecosystem
services for decision making. Ecol. Econ. 68, 643–653.
Freeman, A.M., 2003. Economic Valuation: What and Why. The Economics of Non-
Market Goods and Resources 3, 1–25.
Ghermandi, A., Ding, H., Nunes, P., 2013. The social dimension of biodiversity policy
in the European Union: valuing the benefits to vulnerable communities.
Environ. Sci. Policy 33, 196–208.
Ghermandi, A., Nunes, P.A.L.D., 2013. A global map of coastal recreation values:
Results from a spatially explicit meta-analysis. Ecol. Econ. 86, 1–15.
Ghermandi, A., van den Bergh, J., Brander, L., de Groot, H., Nunes, P., 2010. The
values of natural and human-made wetlands: a meta-analysis. Water Resour.
Res. 46, W12516.
Gilbert, A.J., Janssen, R., 1998. Use of environmental functions to communicate the
values of a mangrove ecosystem under different management regimes. Ecol.
Econ. 25 (3), 323–346.
Gunawardena, M., Rowan, J.S., 2005. Economic valuation of a mangrove ecosystem
threatened by shrimp aquaculture in Sri Lanka. Environ. Manag. 36 (4),
535–550.
Gustavson, K.R., 1998. Values associated with the yse of the Montego Bay Marine
Park. World Bank Research Committee. Project #RPO 681-05.
Please cite this article as: Rao, N.S., et al., Global values of coastal ecosystem services: A spatial economic analysis of shoreline
protection values. Ecosystem Services (2015), http://dx.doi.org/10.1016/j.ecoser.2014.11.011i
Haites, E., Pantin, D., Attzs, M., Bruce, J., McKinnon, J., 2002. Assessment of the
Economic Impact of Climate Change on CARICOM Countries. Margaree Con-
sultants, Toronto, Canada.
Hanley, N., Wright, R.E., Alvarez-Farizo, B., 2006. Estimating the economic value of
improvements in river ecology usingchoice experiments. An application to the
water framework directive. J. Environ. Manag. 78, 183–193.
Hargreaves-Allen, V., 2004. Estimating the total economic value of coral reefs for
residents of Sampela, a Bajau Community in Wakatobi Marine National,
Sulawesi: a case study (Master thesis). Imperial College of Science, Technology
and Medicine, London.
Heston, A., Summers, R., Atons, B., 2006. Penn world table version 62 Center of
International Comparisons of Production. Income and Prices at the University of
Pennsylvania.
Howorth, R., 1982. Technical report on coastal erosion in Kiribati; visit to South
Tarawa January 22–February 10, 1982. SOPAC Technical Report 22 (61 pp.).
IUCN, 2006. Ecological and Socio-Economic Values of Mangrove Ecosystems in
Tsunami affEcted Areas: Rapid Ecological-Economic-Livelihood Assessment of
Ban Naca and Ban Bangman in Ranong Province. Thailand. Available at 〈http://
cmsdata.iucn.org/downloads/thailand_socioeconomic_value_report.pdf〉
(10.06.2011).
Knapp, K.R., Kruk, M.C., Levinson, D.H., Diamond, H.J., Neumann, C.J., 2010. The
international best track archive for climate stewardship (IBTrACS): Unifying
tropical cyclone best track data. Bull. Am. Meteor. Soc. 91, 363–376.
Jerath, M., Bhat, M.G., and Rivera-Monroy, V.H., June 16–19, 2012. Alternative
approaches to valuing carbon sequestration in mangroves. In: Proceedings of
the of the Conference on Ecological Economics (ISEE) and Rio þ 20. Rio de
Janeiro.
Jones, H.P., Hole, D.G., Zavaleta, E., 2012. Harnessing nature to help people adapt to
climate change. Nat. Clim. Change , http://dx.doi.org/10.1038/NCLIMATE1463.
King, S.E., Lester, J., 1995. The value of salt marsh as a sea defense. Mar. Pollut. Bull.
30 (3), 180–189.
Koch, E.W., et al., 2009. Non-linearity in ecosystem services: temporal and spatial
variability in coastal protection. Front. Ecol. Environ. 7 (1), 29–37.
Ledoux, L., 2003. Wetland values in the context of flood and coastal defense
appraisal. CSERGE School of Environmental Sciences University of East Anglia
Norwich NR4 7TJ.
Lehner, B., Döll, P., 2004. Development and validation of a global database of lakes
reservoirs and wetlands. J. Hydrol. 296, 1–22.
McCarthy, J.J., Canziani, O.F., Leary, N.A., Dokken, D.J., White, K.S. (Eds.), 2001.
Climate Change: Impacts Adaptation andVulnerability. Cambridge University
Press, Cambridge.
McGranahan, G., Balk, D., Anderson, B., 2007. The rising tide: assessing therisks of
climate change and human settlements in low elevation coastal zones. Environ.
Urban. 19 (1), 17–37.
Marine Ecosystem Services Partnership (MESP), 2013. Marine Ecosystem Services
Partnership. Online at: 〈http://www.marineecosystemservices.org/〉. (12.03.
2014).
Millennium Assessment (MA), 2005. Ecosystems and Human Well-Being: Biodi-
versity Synthesis. World Resources Institute, Washington DC.
Moberg, F., Folke, C., 1999. Ecological goods and services of coral reef ecosystems.
Ecol. Econ. 29, 215–233.
Naylor, R., Drew, M., 1998. Valuing mangrove resources in Kosrae Micronesia.
Environ. Dev. Econ. 3, 471–490.
Nelson, J., Kennedy, P., 2009. The use (and abuse) of meta-analysis in environ-
mental and natural resource economics: an assessment. Environ. Resour. Econ.
42 (3), 345–377.
Nordhaus, W.D., 2006. Geography and macroeconomics: New data and new
findings. Proc Natl. Acad. Sci. USA 103 (10), 3510–3517.
Öhman, M., Cesar, H., 2000. Costs and Benefits of Coral Mining. In: Cesar, H. (Ed.),
Collected Essays on the Economics of Coral Reefs. CORDIO. Kalmar University,
Kalmar, Sweden.
Orth, R.J., Carruthers, T.J.B., Dennison, W.C., Duarte, C.M., 2006. A global crisis for
seagrass ecosystems. Bioscience 56 (12), 987–997.
Pet-Soede, C., Cesar, H.J.S., Pet, J.S., 1999. An economic analysis of blast fishing on
Indonesian coral reefs. Environ. Conserv. 26 (2), 83–93.
Ranasinghe, T., Kallesoe, M., 2006. Valuation rehabilitation and conservation of
mangroves in tsunami affected areas of Hambantota. IUCN, Sri Lanka.
Ruitenbeek, H.J. Cartier, C., 1999. Issues in applied coral reef biodiversity valuation:
results for Montego Bay Jamaica. World Bank Research Committee Project
RPO# 682-22.
Samonte-Tan, G., White, A.T., Tercero, M.A., Diviva, J., Tabara, E., Caballes, C., 2007.
Economic valuation of coastal and marine resources: Bohol Marine Triangle
Philippines. Coast. Manag. 35, 319–338.
Sathirathai, S., 1998. Economic Valuation of Mangroves and the Roles of Local
Communities in the Conservation of Natural Resources: Case study of Surat
Thani, South of Thailand, Research Report, Environment and Economics
Program for South East Asia. International Development Research Center.
Ottawa, Canada.
Sathirathai, S., Barbier, E.B., 2001. Valuing mangrove conservation in southern
Thailand. Contemp. Econ. Policy 19 (2), 109–122.
Schuhmann, P., 2012. The valuation of marine ecosystem goods and services in the
Caribbean: a review and framework for future work. Regional Governance
Framework for the CLME Project. Center for Resource Management and
Environment Studies, University of the West Indies. Cave Hill, Barbados.
 N.S. Rao et al. / Ecosystem Services ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Seenprachawong, U., 2002. An economic valuation of coastal ecosystems in Phang
Nga Bay, Thailand, Research Report, Environment and Economics Program for
South East Asia. International Development Research Center Ottawa, Canada.
Selig, E.R., Casey, K.S., Bruno, J.F., 2010. New insights into global patterns of ocean
temperature anomalies: implications for coral reef health and management.
Glob. Ecol. Biogeogr. 19 (3), 39–411.
Semesi, A.K., 1998. Mangrove management and utilization in Eastern Africa. Ambio
27 (8), 620–626.
Shabman, L.A., Batie, S.S., 1987. Mitigating damages from coastal wetland devel-
opment: policy, economics and financing. Mar. Resour. Econ. 4, 227–248.
Simpson, R.H., Riehl, H., 1981. The Hurricane and its Impact. Louisiana State
University Press, Baton Rouge, LA.
Simpson, M.C., D., Scott, M., Harrison, N., Silver, E., O'Keeffe, R., Sim, S., Harrison, M.,
Taylor, G., Lizcano, M., Rutty, H., Stager, J.Oldham, M., Wilson, M., New, J., Clarke, O.J.,
Day, N., Fields, J.Georges, R., Waithe, and P., McSharry. 2010. Quantification and
magnitude of losses and damages resulting from the impacts of climate change:
modelling the transformational impacts and costs of sea level rise in the Caribbean.
Barbados: United Nations Development Program(UNDP).
Spalding, M.D., Blasco, F., Field, C.D. (Eds.), 1997. World Mangrove Atlas. The
International Society for Mangrove Ecosystems, Okinawa, Japan.
Spalding, M.D., Ravilious, C., Green, E.P., 2001. World Atlas of Coral Reefs. University
of California Press, Berkeley.
Spaninks, F., van Beukering, P., 1997. Economic valuation of mangrove ecosystems:
potential and limitations. CREED Working Paper Series 14 IIED. London, UK.
Spurgeon J., 2002. Rehabilitation, conservation and sustainable utilization of
mangroves in Egypt. FAO Environment Paper.
TEEB, 2010. The economics of ecosystems and biodiversity. Mainstreaming the
economics of nature: a synthesis of the approach, conclusions and recommen-
dations of TEEB. United Nation Environment Program.
Tri N.H., Adger N., Kelly M., Granich S., Ninh N.H., 1996. The role of natural resource
management in mitiagting climate impacts: mangroves restoration in Vietnam.
CSERGE Working Paper GEC 96-06.
Please cite this article as: Rao, N.S., et al., Global values of coastal ecosystem services: A spatial economic analysis of shoreline
protection values. Ecosystem Services (2015), http://dx.doi.org/10.1016/j.ecoser.2014.11.011i
11
Turner, R.K., Adger, W.N., et al., 1998. Ecosystem services value, research needs, and
policy relevance: a commentary. Ecol. Econ. 25 (1), 61–65.
United Nations Environment Program—World Conservation Monitoring Center,
2003. Global Distribution of Coral Reefs 1 km data. Cambridge, UK.
UNEP-WCMC, 2006. In the front line: Shoreline protection and other ecosystem
services from mangroves and coral reefs. United Nations Environmental
Program–World Conservation Monitoring Center, Cambridge UK.
van Beukering P.J.H. (Ed.), 2007. The economic value of Guam's Coral Reefs.
University of Guam Marine Laboratory Technical Report no. 116. Available at
〈http://www.guammarinelab.com/publications/uogmltechrep116.pdf〉
(Accessed 24.06.2011).
Economic value of the coral reefs of Saipan Commonwealth of the Northern
Mariana Islands (CNMI). In: van Beukering, P.J.H. (Ed.), 2006. Report, Cesar
Environmental Economics Consulting, Washington, DC.
van Beukering, P.J.H., Cesar, H.J.S., Janssen, M.A., 2003. Economic valuation of the
Leuser National Park on Sumatra, Indonesia. Ecol. Econ. 44, 43–62.
Vo, Q.T., Kuenzer, C., Vo, Q.M., Moder, F., Oppelt, N., 2012. Review of valuation
methods for mangrove ecosystem services. Ecol. Indic. 23, 431–446.
Waite, R., et al., 2014. Coastal Capital: Ecosystem Valuation for Decision Making in
the Caribbean (Accessible at). World Resources Institute, Washington, DC.
White, A.T., Vogt, H.P., Arin, T., 2000a. Philippine coral reefs under threat: the
economic losses caused by reef destruction. Mar. Pollut. Bull. 40 (7), 598–605.
White, A.T., Ross, M., Flores, M., 2000b. Benefits and costs of coral reef and wetland
management, Olango Island, Philippines. In: Cesar, H.S.J. (Ed.), Collected Essays
on the Economics of Coral Reefs. CORDIO, Kalmar University, Sweden.
Woodward, R.T., Wui, Y., 2001. The economic value of wetland services: a meta-
analysis. Ecol. Econ. 37, 257–270.
WRI, 2005. The Wealth of the Poor: Managing Ecosystem to Fight Poverty. World
Resources Institute Press, Washington DC.