Patterns of island change and persistence offer
alternate adaptation pathways for atoll nations
Paul S. Kench 1, Murray R. Ford1& Susan D. Owen1
Sea-level rise and climatic change threaten the existence of atoll nations. Inundation and
erosion are expected to render islands uninhabitable over the next century, forcing human
migration. Here we present analysis of shoreline change in all 101 islands in the Paciﬁc atoll
nation of Tuvalu. Using remotely sensed data, change is analysed over the past four decades,
a period when local sea level has risen at twice the global average (~3.90 ±0.4 mm.yr−1).
Results highlight a net increase in land area in Tuvalu of 73.5 ha (2.9%), despite sea-level
rise, and land area increase in eight of nine atolls. Island change has lacked uniformity with
74% increasing and 27% decreasing in size. Results challenge perceptions of island loss,
showing islands are dynamic features that will persist as sites for habitation over the next
century, presenting alternate opportunities for adaptation that embrace the heterogeneity of
island types and their dynamics.
DOI: 10.1038/s41467-018-02954-1 OPEN
1School of Environment, University of Auckland, Private Bag, 92010 Auckland, New Zealand. Correspondence and requests for materials should be addressed
to P.S.K. (email: firstname.lastname@example.org)
NATURE COMMUNICATIONS | (2018) 9:605 |DOI: 10.1038/s41467-018-02954-1 |www.nature.com/naturecommunications 1
Understanding of human migration patterns and popula-
tion relocation through the Paciﬁc, since earliest settle-
ment, has been informed by insights into the geologic
template of atoll island formation and the inﬂuence of environ-
mental change (including sea level) in modulating the habitability
of islands1,2. Consequently, islands have been conceptualised as
pedestals for human occupation, presenting opportunities for
resource development and settlement, with their formation critical
in the migration of peoples through the Paciﬁc1. Questions of
contemporary, and near future, atoll island habitability and per-
sistence are equally framed against a backdrop of environmental
change, and in particular, climate-driven increases in sea level3,4.
Climate change remains one of the single greatest environ-
mental threats to the livelihood and well-being of the peoples of
the Paciﬁc5. The fate of small island states confronted with the
spectre of sea-level rise has raised global concern, and prompted a
labyrinth of international programmes to consider how Paciﬁc
nations can and should adapt to the threats of climatic change6.
Islands considered most at risk of physical destabilisation are low-
lying atoll nations7,8. Erosion, combined with increased frequency
of overwash ﬂooding of island margins4is expected to render
islands uninhabitable9,10. Incremental and event-driven climatic
changes to ecological systems also present additional future
stresses for island habitability, including the tolerance of agri-
culture crops to increased soil salinity, as well as concerns about
water security, both in the context of drought and salt water
intrusion of groundwater11–13.
Under these environmental scenarios, conjectures of habit-
ability and mobility become entwined and have driven an urgency
in socio-political discourse about atoll nation futures and human
security14,15. Strategies for adaptation to changing biophysical
conditions are coupled with narratives of environmentally
determined exodus16. Such persistent messages have normalised
island loss and undermined robust and sustainable adaptive
planning in small island nations17. In their place are adaptive
responses characterised by in-place solutions, seeking to defend
the line and include solutions such as reclamation and sea-
walls18,19, potentially reinforcing maladaptive practices. Not-
withstanding the maladaptive outcomes of such approaches15,20
such dialogues present a binary of stay and defend the line or
eventual displacement. There is limited space within these con-
structs to reﬂect on possibilities that a heterogeneous archipelago
(size, number and dynamics of islands) may offer in terms of
sustained habitability, beyond the historic imprint of colonial
agendas and entrenched land tenure systems that may constrain
novel adaptation responses at the national scale7,21,22.
Amid this dispiriting and forlorn consensus, recent commen-
tators have queried whether the loss of islands can be avoided and
ask whether a more optimistic prognosis exists for atoll nations17.
We argue that indeed there are a more nuanced set of options to
be explored to support adaptation in atoll states. Existing para-
digms are based on ﬂawed assumptions that islands are static
landforms, which will simply drown as the sea level rises4,23.
There is growing evidence that islands are geologically dynamic
features that will adjust to changing sea level and climatic con-
ditions24–27. However, such studies have typically examined a
limited number of islands within atoll nations, and not provided
forward trajectories of land availability, thereby limiting the
ﬁndings for broader adaptation considerations17. Furthermore,
the existing range of adaptive solutions are narrowly constrained
and do not reﬂect the inherent physical heterogeneity and
dynamics of archipelagic systems.
Here we present the ﬁrst comprehensive national-scale analysis
of the transformation in physical land resources of the Paciﬁc
atoll nation Tuvalu, situated in the central western Paciﬁc (Sup-
plementary Note 1). Comprising 9 atolls and 101 individual reef
islands, the nation is home to 10,600 people, 50% of whom are
located on the urban island of Fogafale, in Funafuti atoll28.We
speciﬁcally examine spatial differences in island behaviour, of all
101 islands in Tuvalu, over the past four decades (1971–2014), a
period in which local sea level has risen at twice the global average
(Supplementary Note 2). Surprisingly, we show that all islands
have changed and that the dominant mode of change has been
island expansion, which has increased the land area of the nation.
Results are used to project future landform availability and con-
sider opportunities for a vastly more nuanced and creative set of
adaptation pathways for atoll nations.
Planform island change. Analysis of atoll island change aggre-
gated across Tuvalu reveals three striking features of island areal
transformation over the past four decades (Table 1, Fig. 1, Sup-
plementary Data 1). First, only one island has been entirely
eroded from the data set of 101 islands. This island had an initial
size of 0.08 ha and was located on the reef rim of Nukufetau atoll.
Second, total land area of the nation has expanded by 73.5 ha
(2.9%) since 1971. Notably, eight of nine atolls experienced an
increase in land area. Nanumea was the only atoll where a loss in
land was detected, although this totalled less than 0.01%. Third,
there are marked differences in the magnitude and direction of
areal change between islands. A total of 73 islands (of 101) had a
Table 1 Summary of atoll island characteristics and changes in islands, Tuvalu
Atoll/Reef platform (RP) No islds. Atoll land area Change in land
(ha) (ha) (%) Accr. Erod. No. Area (km2) Pop. density
Nanumea 6 356.1 −1.32 −0.004 3 3 1 2.18 281
Niutao (RP) 1 235.2 0.34 0.14 1 —1 2.35 295
Nanumaga (RP) 1 301.0 4.71 1.56 1 —1 3.01 183
Nui 13 342.8 7.61 2.22 13 —1 1.34 544
Vaitupu (RP) 8 522.9 12.27 2.35 6 2 2 5.18 297
Nukufetau 26 314.4 19.40 6.17 15 11 1 0.19 3458
Funafuti 29 261.2 10.06 3.85 19 10 1 1.59 3427
Nukulaelae 19 176.4 10.00 5.67 16 3 1 0.22 1626
Niulakita (RP) 1 42.1 0.05 0.12 1 —1 0.42 109
Island population data obtained from the Tuvalu Census of Population and Housing28
Accr. accreted islands, Erod. eroded islands
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02954-1
2NATURE COMMUNICATIONS | (2018) 9:605 |DOI: 10.1038/s41467-018-02954-1 |www.nature.c om/naturecommunications
net increase in area, totalling 80.7 ha, with a range from <1to
113% growth. These expanding islands had an average increase in
area of 2.18 ha. Largest absolute increases in island area occurred
on the reef platform islands of Vaitupu (11.4 ha, 2.2%) and
Nanumaga (4.7 ha, 1.6%), and the atoll rim islands of Nui (10.4
ha, 7.1%), Nukufetau (5.3 ha, 4.2%) and the capital island of
Funafuti atoll, Fogafale (4.6 ha, 3%). The remaining 28 islands
(27.7% of total) decreased in area, totalling −7.24 ha and ranging
from 1 to 100% reduction. On average, eroding islands decreased
in area by −0.5 ha (22.69%). Of note, erosion was most prevalent
on the smallest islands in the archipelago. Four islands decreased
in area by more than 50%, although these were all islands that
had an initial size of less than 0.5 ha. Largest absolute decreases in
island area occurred on reef rim islands in Nanumea (−2.88 ha,
1.32%), and three islands on the western rim of Funafuti atoll,
Tepuka (−0.89 ha, 8.35%), Fuagea (−0.74 ha, 45.5%) and Fualifeke
(−0.51 ha, 6.16%).
Shoreline dynamics. Analysis of shoreline dynamics at the
transect scale highlights substantive site-speciﬁc changes around
island shorelines (Fig. 2). Of the 19,403 shoreline transects ana-
lysed, 44% (8583) exhibited accretion, 33% (6338) remained
stable and 23% (4482) showed evidence of erosion over the
analysis period. Notably, on the vast majority of islands both
erosion and accretion were recorded on different parts of island
shorelines. Average net shoreline movement (NSM) calculated
from the transect analysis ranged from 3.71 m per decade on
Savave island in Nukufetau to −3.33 m per decade on Fuagea in
Funafuti. Collectively, the balance between erosion and accretion
on each island yields net changes in island area (Fig. 2b) and also
provides the mechanism for effective island migration on the reef
platform surfaces as exhibited in planform analysis (Fig. 3).
Notably greatest variability in shoreline behaviour occurred on
islands located on the rim of larger atolls (Fig. 2), although data
conﬁrm that total land area increased in eight of the nine atolls.
Results challenge existing narratives of island loss showing that
island expansion has been the most common physical alteration
throughout Tuvalu over the past four decades. Of signiﬁcance,
documented increases in island area over this period have
occurred as the sea level has been rising. The sea level at the
0 2.0 4.0 6.0–2.0–4.020 40 60 80 1000–100 –80 –60
Transects eroded (%) Transect accreted (%)
Island shoreline change per decade (m)
Fig. 2 Summary changes in shoreline dynamics between atolls based on Digital Shoreline Analysis System analysis of island shoreline transects. a
Percentage of shoreline transects experiencing erosion (blue bars) and accretion (orange bars) aggregated at the atoll scale, error bars represent maximum
transect erosion and accretion in each atoll. bNet rate (orange circles) and gross rate (red squares) of shoreline movement per decade aggregated at the
atoll scale. Error bars represent minimum and maximum rates within each atoll. Source data: Supplementary Data 2
Decadal rate of change (%)
Reef island size (ha)
0.01 0.1 1.0 10.0 100.0 1000.0
Change in island area (ha)
0.01 0.1 1.0 10.0 100.0 1000.0
Fig. 1 Summary data of physical island change of islands in Tuvalu between
1971 and 2014. aAbsolute changes in island area in hectares with respect
to island size. bPercentage change in islands per decade with respect to
island size. Raw data contained in Supplementary Data 1. Note: square
symbols denote reef platform islands; solid circles denote atoll rim islands;
and light blue circles enclosing symbols denote populated islands
NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02954-1 ARTICLE
NATURE COMMUNICATIONS | (2018) 9:605 |DOI: 10.1038/s41467-018-02954-1 |www.nature.com/naturecommunications 3
Funafuti tide gauge has risen at 3.9 ±0.4 mm y−1over the time-
frame of analysis (total rise of ~ + 0.15 m, Supplementary Fig. 3)
and this rate of change has been spatially coherent across the
archipelago29. Results show that there has been no uniform
morphological response to this increase in the sea level. While
there has been erosion of a subset of smaller-sized islands
(~26.5%, Fig. 1), the majority of islands (73.5%) have expanded in
area. The absence of a uniform or widespread erosion response
indicates that sea-level change alone cannot account for the
observed island changes and suggests that there are a set of
higher-frequency processes that imprint on island change that
may mask the possible effects of incremental sea-level change.
Wave processes and shifts in wave regime have previously been
identiﬁed as critical controls on island morphological adjustment,
and their inﬂuence can be expressed in three ways. First, shifts in
the incident wave climate may reconﬁgure depositional nodes on
reef surfaces30. However, analysis of the 30-year wave hindcast
data from the Tuvalu region shows no appreciable change in wave
climate since 197931,32, implying that this mechanism is unlikely
to be responsible for observed island adjustments. Second, rising
sea levels can allow a greater transfer of wave energy across reef
surfaces, thus enhancing remobilisation of island shorelines and
sediment transfer33–35. There is compelling evidence to indicate
that this process has exerted an inﬂuence on atoll rim islands
throughout the archipelago, expressed as ocean shoreline erosion
and lagoon shoreline accretion (Figs. 3b, c, e) resulting in net
lagoonward migration of islands36,37. However, it is important to
highlight that, in many instances, such migration responses have
also been accompanied by island expansion. Third, storm wave
processes can inﬂuence island morphology and size, although
erosion or accretion trajectories vary depending on storm mag-
nitude and the grade of material comprising islands38,39. While
located outside the primary zone of cyclogenesis, the Tuvalu
archipelago is periodically imﬂuenced by cyclone events that
generate wave heights between 3 and 4 + m40,41. In Tuvalu, it is
possible that extreme wave events can partly explain spatial dif-
ferences in observed island change. For example, Cyclone Bebe
(1972) delivered signiﬁcant volumes of coarse sediment to the
Funafuti reef ﬂat, which were subsequently reworked to the island
shorelines expanding the footprint of the islands on the eastern
rim of Funafuti over the four decades37,40,42. Such episodic events
and their subsequent constructional effects could account for the
predominant expansion mode of mixed sand–gravel and gravel
islands elsewhere in the archipelago. In contrast, the same events
may have destabilised sand islands. Our data show that 54% (13
of 24) of the sand islands reduced in size over the timeframe of
analysis. In Funafuti and Nukufetau these islands are located on
the leeward northwest and northern sectors of atoll rims. While
construction of the islands has occurred under lower-energy
regimes periodic storms may have promoted erosion and desta-
bilisation of these islands.
While wave processes can account for locational shifts in
shorelines, they cannot solely account for the expansion of the
majority of islands. Expansion of islands on reef surfaces indicates
a net addition of sediment. Implications of increased sediment
volumes are profound as they suggest positive sediment genera-
tion balances for these islands and maintenance of an active
linkage between the reef sediment production regime and transfer
to islands, which is critical for ongoing physical resilience of
islands43. Such island reef budgets and their connectivity are
likely to be spatially variable as a consequence of the localised
reefal provenance of island sediments and the temporal dynamics
of reef ecology and sediment generation and transfer mechan-
isms37,43,44. On most windward reef sites such linkages are
modulated by storm-driven wave deposition of new materials and
subsequent reef recovery, whereas at leeward locations, where
sand islands may prevail, supply is likely to be characterised by a
more consistent incremental addition of sediments from reef ﬂat
Direct anthropogenic transformation of islands through
reclamation or associated coastal protection works/development
has been shown to be a dominant control on island change in
other atoll nations24,27,45,46. However, in Tuvalu direct physical
interventions that modify coastal processes are small in scale as a
consequence of much lower population densities. Only 11 of the
study islands have permanent habitation and, of these, only two
100 m 100 m
Fig. 3 Examples of island change and dynamics in Tuvalu from 1971 to 2014. aNanumaga reef platform island (301 ha) increased in area 4.7 ha (1.6%) and
remained stable on its reef platform. bFangaia island (22.4 ha), Nukulaelae atoll, increased in area 3.1 ha (13.7%) and remained stable on reef rim. c
Fenualango island (14.1 ha), Nukulaelae atoll rim, increased in area 2.3 ha (16%). Note smaller island on left Teafuafatu (0.29 ha), which reduced in area
0.15 ha (49%) and had signiﬁcant lagoonward movement. dTwo smaller reef islands on Nukulaelae reef rim. Tapuaelani island, (0.19 ha) top left, increased
in area 0.21 ha (113%) and migrated lagoonward. Kalilaia island, (0.52 ha) bottom right, reduced in area 0.45 ha (85%) migrating substantially lagoonward.
eTeafuone island (1.37 ha) Nukufetau atoll, increased in area 0.04 ha (3%). Note lateral migration of island along reef platform. Yellow lines represent the
1971 shoreline, blue lines represent the 1984 shoreline, green lines represent the 2006 shoreline and red lines represent the 2014 shoreline. Images ©2017
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02954-1
4NATURE COMMUNICATIONS | (2018) 9:605 |DOI: 10.1038/s41467-018-02954-1 |www.nature.c om/naturecommunications
islands sustain populations greater than 600. Notably, there have
been no large-scale reclamations on Tuvaluan islands within the
analysis window of this study (the past four decades). On the
most densely urbanised island Fogafale, there has been minimal
direct shoreline modiﬁcation up to 201447 with observed
increases in island area occurring well beyond the main settle-
ment areas. Elsewhere in the archipelago, direct shoreline mod-
iﬁcation is also limited in scale and includes coastal protection
works along a short length of Savave shoreline in Nukufetau,
dredging of boat access channels across reef ﬂats, and construc-
tion of associated boat-landing structures. Data suggest that these
modiﬁcations have had a negligible direct impact on coastal
change at the construction sites or adjacent sites alongshore with
expansion occurring well outside the footprint of human
Consequently, documented changes in islands throughout
Tuvalu are considered to be driven by environmental rather than
anthropogenic processes. In particular, wave and sediment supply
processes provide the most compelling explanation for the phy-
sical changes documented in islands, most notably the expansion
of the majority of islands, and their locational adjustments over
the past four decades. Collectively, these processes can mask any
incremental effects of rising sea level, making attribution of sea-
level effects elusive, as these processes can promote higher fre-
quency and larger magnitude changes in islands that can persist
in the geomorphic record.
On the basis of empirical changes in islands we project a
markedly different trajectory for Tuvalu’s islands over the next
century than is commonly envisaged. Observations over the past
four decades indicate that the future of Tuvalu’s islands will be
marked by a continual changing mosaic of physical land
Changes expected include the ongoing erosion of smaller sand
islands in the archipelago (<1 ha), continued expansion of the
majority of medium (1–10 ha) and larger-sized islands (>10 ha),
stability of reef platform islands and increased mobility of atoll
reef rim islands. Such changes suggest that the existing footprint
of islands on reef surfaces will continue to change, although the
physical foundation of islands will persist as potential pedestals
for habitation over the coming century. Consequently, while we
recognise habitability rests on an additional set of factors4,11–13
loss of land is unlikely to be a factor in forcing depopulation of
islands or the entire nation. However, changes in land resources
may still stress population sustainability in the absence of
appropriate adaptive initiatives.
Signiﬁcantly, our results show that islands can persist on reefs
under rates of sea-level rise on the order of 3.9 ±0.4 mm yr−1
over the past four decades (Supplementary Note 2, Supplemen-
tary Fig. 3) equating to an approximate total rise of ~0.15 m. This
rate is commensurate with projected rates of sea-level rise across
the next century under the RCP2.6 scenario mid-point rate of 4.4
mm yr−1(range 2.8–6.1 mm yr−1)48. However, under the RCP8.5
the projected rate of sea-level rise will double to 7.4 mm yr−1
(range 5.2–9.8 mm yr−1). Under these higher sea-level projections
it is unclear whether islands will continue to maintain their size,
although the dynamic adjustments observed are expected to occur
at faster rates placing a premium on establishing ongoing mon-
itoring of island morphological dynamics.
Recognition that land resources will remain through the next
century also challenges past and current paradigms of adaptation.
It has been argued that the adaptation experience in atoll coun-
tries to date has been poor6,17. Underpinning past approaches to
adaptation have been a set of time to extinction projections,
implying that habitability of islands is likely to be severely com-
promised in the coming decades4,16. In part, this is due to the lack
of relevant information on the type and scale of changes expected
in the future against which to inform adaptation planning17.
Without such knowledge adaptation solutions have been captured
by the rhetoric of loss, which has foreclosed robust consideration
of sustainable adaption options. Our analysis provides an
empirical basis to reconceptualise alternate and more creative
adaptation pathways in atoll nations with continued habitation of
islands underpinning these approaches.
Quantiﬁed patterns of island physical dynamics provide a sound
basis for new approaches to land use planning. The Tuvalu data
indicate that reef platform islands have remained the most stable
islands and in most instances have increased in area. However,
despite their larger size (>10 ha) and stability these islands remain
among the least densely populated. For example, the reef platform
islands of Nanumaga (3.0 km2) and Vaitupu (5.18km2)have
population densities of 183 and 297 km−2, respectively, which are
much lower than the urban island of Fogafale (area of 1.59 km2)
with a population density of 3427 km−2. Notably, medium-sized
islands (1–10 ha) have largely expanded over recent decades and,
despite the fact that these islands are scarcely populated, they
could provide opportunities for future habitation across the
archipelago. Smaller islands appear the most dynamic, in some
cases experiencing marked erosion and, therefore, do not provide
ideal sites for ongoing habitation.
Insights into island change in Tuvalu parallel observations on
biophysical change made elsewhere25,26,46,49,50 and allow us to
reﬂect more widely on patterns of population distribution and
resource pressures in other atoll nations. Current population
distributions in atoll nations are legacies of economic and social
investment rather than reﬂective of the carrying capacity of the
land and may be considered not well aligned to the changing
mosaic of island adjustments observed over the past four decades.
Contemporary histories of population movement and settlement
in the Paciﬁc are shaped by geopolitical inﬂuences on the dis-
tribution of economic, transportation, health, educational and
livelihood opportunities at a national scale51. Commonly, the
densest populations are located in the economic and political
centres, situated on smaller and less stable islands, which repre-
sent less than 1% of the land available in archipelagoes. The
complexity of habitability in these settings is also coupled with
competing discourses of abandonment, displacement and threats
to human security.
Against this backdrop of patterns of human resettlement,
exploring opportunities presented by the dynamic mosaic of land
availability necessitates a reconsideration of how land-use plan-
ning is undertaken that recognises the heterogeneity of island
changes21, existing land tenure systems, patterns of food secur-
ity52 and approaches to support internal migration within atoll
nations. Such suggestions are by no means novel14 but to date
long-term planning has been constrained by concerns about lack
of data about island change to support informed decisions. Here
we have presented more compelling evidence that islands may
persist and encourage a re-engagement with what alternative
adaptation pathways may look like.
If collective narratives are imagining atoll island futures beyond
geo-political boundaries, destabilising cultural identity and
sovereignty, we ask on the compelling evidence of changing
island dynamics and future land availability, is it inappropriate to
also re-imagine intranational migration and to consider the
development, political and cultural implications of such reloca-
tions? To date, such movement in the Paciﬁc has had varied
outcomes and been driven by formalised relocation agendas and
more informal movement between place, highlighting experiences
of cultural and economic disconnection53,54. However, it has also
been argued that internal relocations can work more effectively
and communities experience less trauma where they are familiar
with the places they ultimately move to, have time to plan and are
NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02954-1 ARTICLE
NATURE COMMUNICATIONS | (2018) 9:605 |DOI: 10.1038/s41467-018-02954-1 |www.nature.com/naturecommunications 5
in control of that planning, have time to accommodate the idea of
movement and move at a time of their choosing in an orderly
manner55. Not least at issue here is the requirement for signiﬁcant
and continued economic investment, including the development
of opportunities for appropriate economic growth and sustained
Embracing such new adaptation pathways will present con-
siderable national-scale challenges to planning, development
goals and land tenure systems. However, as the data on island
change show there is time (decades) to confront these challenges,
which could engender more thoughtful support from interna-
tional agencies. The pursuit of this and other alternate adaptation
pathways does not negate the need to still vigorously support
ongoing mitigation action to curtail future sea level impacts and
climatic changes on small island nations or to undertake robust
efforts to better deﬁne the constraints and thresholds of habit-
ability (such as water resources and food supply) on atoll islands.
These collective efforts provide a more optimistic set of approa-
ches to adaptation, which support the rights of atoll people to
digniﬁed lives and autonomy for future generations and main-
taining the sovereignty of atoll nations.
Data sources. Remotely sensed assessments of shoreline change along coasts
within developed nations typically involve the use of temporally rich collections of
aerial photographs spanning several decades56. However, atoll nations in the Paciﬁc
are remote and have limited collections of aerial imagery. Initial imagery from
Tuvalu was ﬂown in World War II associated with military occupation of Funafuti
atoll. National aerial coverage was ﬁrst ﬂown in 1971. To examine shoreline change
on islands throughout the Tuvalu archipelago, we compare shoreline positions
reconstructed from historic aerial photographs captured between 1943 (fragmen-
tary), 1971 and 1984, and high-resolution panchromatic (WorldView-1) and
multispectral (QuickBird-2, WorldView-2 and WorldView-3) satellite imagery
collected between 2004 and 2015. The principle analysis window (1971–2014) is
~43 years in length.
Image processing. Multispectral satellite imagery was pan-sharpened, a process
through which the coarser resolution multispectral imagery is sharpened using
higher-resolution panchromatic imagery captured simultaneously. The oldest
satellite imagery for each atoll provided the source of ground control points for
georeferencing imagery. Given the paucity of stable anthropogenic features on most
islands, a range of natural features such as cemented conglomerate and beachrock
were used as ground control points following similar studies in the Republic of
the Marshall Islands25. Images were georeferenced in ArcMap and transformed
using a second-order polynomial transformation.
Shoreline interpretation and analysis. The edge of vegetation is widely used as a
proxy for the shoreline within island change studies in atoll settings24,25,56. The
edge of vegetation is readily identiﬁable in all imagery, regardless of image colour
and contrast and irrespective of environmental conditions such as glare and waves,
all of which can impede the interpretation of subtidal and intertidal features such as
the toe of beach. The edge of vegetation represents the vegetated core of the island
and ﬁlters short-term noise associated with the interpretation of dynamic beach
shorelines. Where 1971 shorelines are cloud-obscured, preventing the creation of a
closed polygon, we use previously calculated areas for the vegetated edge of
Three sources of uncertainty were considered when calculating the positional
uncertainty in edge of vegetation, being rectiﬁcation, pixel and digitising errors25.
Rectiﬁcation error was derived from the Root Mean Square Error of
georeferencing. The spatial resolution of scanned aerial photographs and satellite
imagery represents the pixel error. The digitising error was calculated as the SD of
shoreline position from repeated digitisation of the same section of coast by a single
operator. Total shoreline error (Te) was calculated as the root sum of all shoreline
positional errors and ranged between 1.31 and 3.46 m.
Shoreline change analysis was undertaken using the Digital Shoreline Analysis
System (DSAS) an extension within the GIS software package ArcMap58. DSAS
analyses change by recording the intersection of transects cast perpendicular to a
user-generated baseline and the shorelines. In this study, transects were cast every
10 m along the baseline with a total of 19,403 transects analysed for the
archipelago. A range of change statistics were then calculated automatically using
the position of the intersection of shorelines and transects. In environments with
high temporal resolution records of shoreline positions regression-derived
measures of shoreline change rates are widely used56,59. However, due to the
limited number of shorelines used in this study regression-derived shoreline
change rates are unreliable. As a result, two measures of island change are utilised
in this study. First, NSM, the distance between two selected shorelines, was
calculated. Second, the annualised rate of change between two shorelines, known as
the end point rate (EPR) was calculated. Given the multidecadal timeframe of the
data set, the EPR is expressed as decadal rate of change (m per decade). A
conﬁdence interval of 2σ(95.5%) was applied when calculating shoreline change
rates. Transects with statistically signiﬁcant rates of change are considered
erosional (−/ve EPR) or accretionary (+/ve EPR), the remaining transects are
classiﬁed as exhibiting no detectable change.
Data availability. All data are contained in Supplementary Information. Source
ArcMap shapeﬁles are available from the authors on request.
Received: 4 July 2017 Accepted: 8 January 2018
1. Weisler, M. I., Yamano, H. & Hua, Q. A multidisciplinary approach for dating
human colonization of Paciﬁc atolls. J. Isld. Coast. Arch. 7, 102–125 (2012).
2. Nunn, P. D. Sea levels, shorelines and settlements on Paciﬁc reef islands.
Archaeol. Ocean. 51,91–98 (2016).
3. Dickinson, W. R. Paciﬁc atoll living: how long already and until when? GSA
Today 19,4–10 (2009).
4. Storlazzi, C. D., Elias, E. P. L. & Berkowitz, P. Many atolls may be
uninhabitable within decades due to climate change. Sci. Rep. 5, 14546 (2015).
5. Paciﬁc Islands Forum Secretariat. Forum Communiqué (Forty Sixth Paciﬁc
Islands Forum, Ports Moresby, Papua New Guinea, 2015).
6. Barnett, J. & Campbell, J. Climate Change and Small Island States: Power,
Knowledge and the South Paciﬁc(Earthscan, London, 2010).
7. Roy, P. & Connell, J. Climatic change and the future of atoll states. J. Coast.
Res. 7, 1057–1075 (1991).
8. Nicholls, R. J. & Cazenave, A. Sea-level rise and its impact on coastal zones.
Science 328, 1517–1520 (2010).
9. Hubbard, D. et al. Island outlook: warm and swampy. Science 345, 1461
10. Weiss, K. R. Before we drown we may die of thirst. Nature 526, 624–627
11. Connell, J. Food security in the island Paciﬁc: Is Micronesia as far away as
ever? Reg. Environ. Change 15, 1299–1311 (2015).
12. McCubbin, S., Smit, B. & Pearce, T. Where does climate ﬁt? Vulnerability to
climate change in the context of multiple stressors in Funafuti, Tuvalu. Glob.
Environ. Change 30,43–55 (2015).
13. Marotzke, J. et al. Climate research must sharpen its view. Nat. Clim. Change
14. Connell, J. Climatic change: a new security challenge for the atoll states of the
South Paciﬁc. J. Commonw. Comp. Polit. 31, 173–192 (1993).
15. Barnett, J. & O’Neil, S. J. Islands, resettlement and adaptation. Nat. Clim.
Change 2,8–10 (2012).
16. Yamamoto, L. & Esteban, M. Migration as an adaptation strategy for atoll
island states. Int. Migr. 55, 144–158 (2017).
17. Barnett, J. The dilemmas of normalising losses from climate change: towards
hope for Paciﬁc atoll countries. Asia. Pac. Viewp. 58,3–13 (2017).
18. Hay, J. & Mimura, N. Supportingclimate change vulnerability and adaptation
assessments in the Asia-Paciﬁc region: an example of sustainability science.
Sus. Sci. 1,23–35 (2006).
19. Duvat, V. Coastal protection structures in Tarawa atoll, Republic of Kiribati.
Sus. Sci. 8, 363–379 (2013).
20. Kench, P.S. in Pitfalls of Shoreline Stabilization (eds. Cooper, J.A. & Pilkey, O.
H.) Ch. 11, 165–186 (Springer, Netherlands, 2012).
21. Owen, S. D., Kench, P. S. & Ford, M. R. Improving understanding of the
spatial dimensions of biophysical change in atoll countries and implications
for island communities: a Marshall Islands’case study. Appl. Geogr. 72,55–64
22. Warrick, O., Aalbersberg, W., Dumaru, P., McNaught, R. & Teperman, K. The
‘Paciﬁc Adaptive Capacity Analysis Framework’: guiding the assessment of
adaptive capacity in Paciﬁc island communities. Reg. Environ. Change 17,
23. Quaterat, E. et al. The inﬂuence of coral reefs and climate change on wave-
driven ﬂooding of tropical coastlines. Geophys. Res. Lett. 42, 6407–6415
24. Webb, A. & Kench, P. S. The dynamic response of reef islands to sea-level rise:
Evidence from multi-decadal analysis of island change in the Central Paciﬁc.
Glob. Planet. Change 72, 234–246 (2010).
25. Ford, M. R. Shoreline changes interpreted from multi-temporal aerial
photographs and high resolution satellite images: Wotje atoll, Marshall
Islands. Remote Sens. Environ. 135, 130–140 (2013).
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02954-1
6NATURE COMMUNICATIONS | (2018) 9:605 |DOI: 10.1038/s41467-018-02954-1 |www.nature.c om/naturecommunications
26. McLean, R. F. & Kench, P. S. Destruction or persistence of coral atoll islands
in the face of 20th and 21st century sea-level rise? WIREs Clim. Change 6,
27. Duvat, V. K. E. & Pillet, V. Shoreline changes in reef islands of the Central
Paciﬁc: Takapoto Atoll, Northern Tuamotu, French Polynesia. Geomorphology
28. Tuvalu Central Statistics Division. TCSD Census of Population and Housing.
29. Hamlington, B. D. et al. Uncovering an anthropogenic sea-level rise signal in
the Paciﬁc Ocean. Nat. Clim. Change 4, 782–785 (2014).
30. Albert, S. et al. Interactions between sea-level rise and wave exposure on reef
island dynamics in the Solomon Islands. Environ. Res. Lett. 11, 054011 (2016).
31. Trenham, C. E., Hemer, M. A., Durrant, T. H. & Greenslade, D. J. M.
PACCSAP Wind-wave Climate: High resolution wind-wave climate and
projections of change in the Paciﬁc region for coastal hazard assessments.
CAWCR Technical Report No. 068 (Centre for Australian Weather and
Climate Research, CSIRO and Australian Bureau of Meteorology, Australia,
32. Bosserelle, C., Reddy, S. & Lal, D. in WACOP Wave Climate Reports (ed.
Bosserelle, C.) (Secretariat of the Paciﬁc Community, Suva, Fiji Islands; 2015).
33. Sheppard, C. et al. Coral mortality increases wave energy reaching shores
protected by reef ﬂats: examples from the Seychelles. Estuar. Coast. Shelf Sci.
64, 223–234 (2005).
34. Storlazzi, C. D., Elias, E., Field, M. E. & Presto, M. K. Numerical modeling of
the impact of SLR on fringing coral reef hydrodynamics and sediment
transport. Coral Reefs 30,83–96 (2011).
35. Beetham, E., Kench, P. S., O’Callaghan, J. & Popinet, S. Wave transformation
and shoreline water level on Funafuti Atoll, Tuvalu. J. Geophys. Res. Oceans
121, 311–326 (2016).
36. Kench, P. S. & Cowell, P. J. Variations in sediment production and
implications for atoll island stability under rising sea level. In Proc. 9th
International Coral Reef Symposium Vol. 2 (Bali, 23-27 October 2000).
37. Kench, P. S., Thompson, D., Ford, M., Ogawa, H. & McLean, R. Coral islands
defy sea-level rise over the past century: Records from a central Paciﬁc atoll.
Geology 43, 515–518 (2015).
38. Bayliss-Smith, T. P. The role of hurricanes in the development of reef islands,
Ontong Java Atoll, Solomon Islands. Geogr. J. 154, 377–391 (1988).
39. Ford, M. R. & Kench, P. S. Multi-decadal shoreline changes in response to sea
level rise in the Marshall Islands. Anthropocene 11,14–24 (2015).
40. Maragos, J. E., Baines, G. B. K. & Beveridge, P. J. Tropical Cyclone Bebe
creates a new land formation on Funafuti Atoll. Science 181, 1161–1164
41. Fitchett, K. Physical effects of Hurricane Bebe upon Funafuti Atoll, Tuvalu.
Aust. Geogr. 18,1–7 (1987).
42. Baines, G. B. K. & McLean, R. F. Sequential studies of hurricane deposit
evolution at Funafuti Atoll. Mar. Geol. 21,M1–M8 (1976).
43. Perry, C. T. et al. Implications of reef ecosystem change for the stability and
maintenance of coral reef islands. Glob. Change Biol. 17, 3679–3696 (2011).
44. Kench, P. S. & Cowell, P. J. The morphological response of atoll islands to sea-
level rise. Part 2: application of the modiﬁed shoreface translation model
(STM). J. Coast. Res. Special Issue 34, 645–656 (2001).
45. Rankey, E. C. Nature and stability of atoll island shorelines: Gilbert Island
chain, Kiribati, equatorial Paciﬁc. Sedimentology 58, 1831–1859 (2011).
46. Aslam, M. & Kench, P. S. 2017 Reef island dynamics and mechanisms of
change in Huvadhoo Atoll, Republic of Maldives, Indian Ocean. Anthropocene
47. Yamano, H. et al. Atoll island vulnerability to ﬂooding and inundation
revealed by historical reconstruction: Fongafale Islet, Funafuti Atoll, Tuvalu.
Glob. Planet. Change 57, 407–416 (2007).
48. Church, J.A. et al. In Climate Change 2013: The Physical Science Basis.
Contribution of Working Group 1 to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change. (ed. Stocker, T.F. et al.) Ch. 13
(Cambridge University Press, UK, 2013).
49. Biribo, N. & Woodroffe, C. D. Historical area and shoreline change of reef
islands around Tarawa Atoll, Kiribati. Sustain. Sci. 8, 345–362 (2013).
50. Cozannet, G. L., Garcin, M., Salaï, E. & Walker, P. Multidecadal atoll shoreline
change on Manihi and Manuae, French Polynesia. J. Coast. Res. 29, 870–882
51. Connell, J. Vulnerable islands: climate change, tectonic change, and changing
livelihoods in the Western Paciﬁc. Contemp. Pac. 27,1–36 (2015).
52. McCubbin, S. G., Pearce, T., Ford, J. D. & Smit, B. Social–ecological change
and implications for food security in Funafuti, Tuvalu. Ecol. Soc. 22, 53 (2017).
53. Connell, J. Losing ground? Tuvalu, the greenhouse effect and the garbage can.
Asia. Pac. Viewp. 44,89–107 (2003).
54. Green, M. Contested territory. Nat. Clim. Change 6, 817–820 (2016).
55. Wilmsen, B. & Webber, M. What can we learn from the practice of
development-forced displacement and resettlement for organised
resettlements in response to climate change? Geoforum 58,76–85 (2015).
56. Romine, B. M. et al. Are beach erosion rates and sea-level rise related in
Hawaii? Glob. Planet. Change 108, 149–157 (2013).
57. McLean, R. F. & Hosking, P. L. Tuvalu Land Resources Survey. Island Reports
1–10 (FAO, UNDP, University of Auckland, New Zealand, 1992).
58. Thieler, E. R. & Danforth, W. W. Historical shoreline mapping (I):
improving techniques and reducing positioning errors. J. Coast. Res. 10,
59. Genz, A. S. et al. The predictive accuracy of shoreline change rate methods
and alongshore beach variation on Maui, Hawaii. J. Coast. Res. 23,87–105
Research was partially supported by a University of Auckland internal research grant
P.S.K. conceived the project and led analysis and writing of the manuscript. M.R.F. led
remote sensing and DSAS analysis. S.D.O. contributed to the adaptation context for the
article and assisted manuscript writing and preparation.
Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467-
Competing interests: The authors declare no competing ﬁnancial interests.
Reprints and permission information is available online at http://npg.nature.com/
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional afﬁliations.
Open Access This article is licensed under a Creative Commons
Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons license, and indicate if changes were made. The images or other third party
material in this article are included in the article’s Creative Commons license, unless
indicated otherwise in a credit line to the material. If material is not included in the
article’s Creative Commons license and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this license, visit http://creativecommons.org/
© The Author(s) 2018
NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-02954-1 ARTICLE
NATURE COMMUNICATIONS | (2018) 9:605 |DOI: 10.1038/s41467-018-02954-1 |www.nature.com/naturecommunications 7
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at