Anthropology Annotated Bibliopgrahy

Coastal Education & Research Foundation, Inc. is collaborating with JSTOR to digitize, preserve and extend access to Journal of Coastal Research.

http://www.jstor.org

Coastal Education & Research Foundation, Inc.

Sea-Level Changes over the past 1,000 Years in the Pacific Author(s): Patrick D. Nunn Source: Journal of Coastal Research, Vol. 14, No. 1 (Winter, 1998), pp. 23-30 Published by: Coastal Education & Research Foundation, Inc. Stable URL: http://www.jstor.org/stable/4298758 Accessed: 14-10-2015 01:55 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

This content downloaded from 198.161.51.4 on Wed, 14 Oct 2015 01:55:54 UTC All use subject to JSTOR Terms and Conditions

Journal of Coastal Research 14 J 1 23-30 Royal Palm Beach, Florida Winter 1998

Sea-Level Changes over the Past 1,000 Years in the Pacific1 Patrick D. Nunn

Department of Geography The University of the South Pacific P.O. Box 1168 Suva, FIJI

$00 ?00O ,S av r???????*

ABSTRACTI

Nunn, P.D., 1998. Sea-level changes over the past 1,000 years in the Pacific. Journal of Coastal Research, 14(1), 23- 30. Royal Palm Beach (Florida), ISSN 0749-0208.

The absence of good information about sea-level change for the past 1,000 years in the Pacific is unfortunate given our much clearer understanding of the earlier Holocene and the past - 100 years. Yet such information is needed if we are to be able to properly understand the causes of late Neogene sea-level changes and to understand the envi- ronmental effects of predicted future changes.

Data are selected from sites in American Samoa, Fiji, the Gambier Islands, Guam, Kosrae, New Zealand, Rota and the Tuamotus which have been tectonically stable for the past thousand years. These data are plotted and a sea-level envelope drawn to characterise the most probable Pacific-wide course of sea-level change for the last millennium.

Sea level was close to its present level - 1000 years BP, then rose to perhaps 0.9 m above around 700 BP. This period of sea-level rise coincided with a period of warming named the Little Climatic Optimum. The transition to the Little Ice Age, when sea level stood lower, was marked by a transition around 690 BP when sea level (and ground temperatures) fell rapidly. A lack of data from the following - 200 years suggest that sea level was lower than the level to which it rose during the early part of the Little Ice Age. A gradual fall occurred during the later part of the Little Ice Age to as much as -0.9 m below present some 200 BP. The last part of the millennium has been charac- terised by a net rise of sea level. Several anomalous dates from the sites examined are most probably explicable by post-emergence contamination of reef limestones.

ADDITIONAL INDEX WORDS: Pacific islands, last millennium, Little Climatic Optimum, Little Ice Age, anomalous dates.

INTRODUCTION

It is somewhat ironic that our knowledge of sea-level vari- ations over the past few thousand years, perhaps even during the latest Neogene (late Quaternary), is better known than those which occurred over the past 1,000 years. Admittedly, the latter changes were of much less magnitude than the for- mer, but it is also an issue of precision. Latest Neogene sea- level changes can be known only comparatively imprecisely and there are often disputes involving several thousand years (or several metres) about particular events. This is obviously not possible with a single 1,000-year period; many of the tech- niques used to establish earlier palaeosea-level changes can not be used with such great effect for the last millennium.

Yet it is doubly ironic to consider that the last 100 years, for which there are comparatively large numbers of contin- ually-monitored series of sea-level data, is also better known than the preceding 900 years. The existence of this "1,000- year hiatus" in our understanding of past sea-level changes is a major hindrance for only by successfully linking the dis- tant to the recent past can we properly hope to understand how sea level might change in the future.

There are other reasons for studying sea-level changes over the past millennium. One is that the study of their environ- mental effects will aid accurate prediction of the conse- quences of future sea-level rise. Another is that an under- standing of last millennium sea levels allows insights into why environments changed. Further, the close relationship between changing sea levels and changing earth surface tem- peratures over many time scales means that the record of changing sea levels over the last millennium can be used as a proxy for contemporary temperature changes which, in turn, permits insights into climate changes.

This paper looks at sea-level changes over the past 1000 years in the Pacific islands, then looks at the environmental consequences of these and associated changes before going on to discuss future changes.

THE HOLOCENE CONTEXT

It is now reasonably well established that all Pacific island coasts experienced inundation associated with a sea level 1- 2 m higher than present around 4000-2000 years BP (NUNN, 1995). It seems increasingly possible, both from interpreta- tions of empirical data and from theoretical considerations, that Holocene sea level has oscillated for the past few thou- sand years (FAIRBRIDGE, 1992; NUNN, 1995).

In some places, such as French Polynesia (PIRAZZOLI and

97060 received and accepted in revision 7 May 1997. 1A contribution to the work of the INQUA Quaternary Shorelines

West Pacific Subcommission

This content downloaded from 198.161.51.4 on Wed, 14 Oct 2015 01:55:54 UTC All use subject to JSTOR Terms and Conditions

24 Nunn

20 N

'o •

HAWAI

? Rota GUAM i NORT I PACIFI- C OCEANAT

SKo•ae 0

PAPUA SNEW

? Wands WSTERN

a SAA ENCH POLYNESIA

I 4 "*ITuhi Ragnroa .

SSOUTHf

PACIFIC OCEAN

wi FM <y

- 40 S

•S

r

wos Boy

SZEAL Bquo"in Bay 140 E 160 E 1800 160W 140 W

I I

Figure 1. The Pacific islands showing the sites from which dates of emerged shorelines have been used to construct the curve of sea-level changes for the past 1,000 years. No error margins are shown.

MONTAGGIONI, 1986), the fall to the present general level within the last 2,000 years occurred comparatively abruptly compared to the fall in other places, such as Fiji and else- where (NUNN, 1990a), which apparently occurred more gent- ly.

In most parts of the Pacific, the sea surface was probably at or close to its present level about 1200-900 years BP. Most records show little variation since that time.

THE LAST MILLENNIUM: FINDING SUITABLE SITES

Many Pacific islands are sites of dramatic, often sporadic, tectonic change which has complicated the derivation of late Neogene sea-level change records (NUNN, 1994a). Yet other islands, commonly in intraplate rather than plate-boundary locations, are believed to have been stable at least for much of the last few thousand years. Their shorelines can then be considered to have registered Holocene eustatic changes rath- er than these confused with tectonic changes; this was one reason why DALY chose to work on Pacific (and other) mid-

ocean islands such as those in Samoa (DALY, 1920, 1924, 1934).

Another major problem is not so much finding suitable sites but finding suitable sites for which some dates of shore- line displacement over the past 1,000 years have been ob- tained. No studies in the Pacific islands to the author's

knowledge have focused primarily on the last 1,000 years. Dates from this time period have been incidental to studies

focusing on the earlier Holocene. The locations of the "stable" sites selected for this study

are shown in Figure 1. Each site is discussed briefly below and the case for its tectonic stability over the last 1000 years presented.

American Samoa

The Samoa islands form a somewhat atypical hotspot is- land chain. Hotspot volcanism is overprinted on a few islands

by anomalous yet superficial volcanism associated with the

tearing of the westward-moving Pacific Plate as it has been

simultaneously pulled down into the trench at the northern

Journal of Coastal Research, Vol. 14, No. 1, 1998

This content downloaded from 198.161.51.4 on Wed, 14 Oct 2015 01:55:54 UTC All use subject to JSTOR Terms and Conditions

Holocene Sea-Level Changes in the Pacific Basin 25

end of the Tonga Trench (NATLAND and TURNER, 1985; KEATING, 1992; NUNN, 1997). Shorelines dating to a Holo- cene sea-level maximum around 1.5 m above present are found around the coasts of the islands in Western Samoa (NUNN, 1991a) and have been observed in parts of American Samoa (DALY, 1924; KIRCH et al., 1990). This is good evidence that these islands have been comparatively stable over the past few thousand years.

The idea that the Samoa islands have been subsiding rap- idly over this time period arose from a misinterpretation of the only Lapita site in the islands at Mulifanua on Upolu (NUNN, 1995). There is no evidence to suggest that signifi- cant subsidence has occurred here although dates from man- grove peats have been used here, as elsewhere, to support such a view (BLOOM, 1980). Mangrove peat dates can only be regarded as indicating minimum sea-level position because of the likelihood that such deposits have become more compact and subsided since they formed.

Recent investigations by the writer on Tutuila found emerged in situ reef at several locations. The most conspic- uous of these is that in Vatia Bay, first reported by DALY (1924), which reaches 3.2 m above modern reef level. Like a similar block at nearby Afono, this reef has yielded only very young dates, the interpretation of which is discussed further below. A lower emerged reef, dated securely to the last mil- lennium, is found near the western extremity of Tutuila about 0.5 km west of Poloa.

Fiji

The tectonic character of Fiji's islands was outlined by NUNN (1990a, 1991b). Many islands are rising or subsiding at rates which probably render suspect any attempt to dis- cern eustatic changes over the past 1,000 years. One of the few places where there appears to be a reasonable chance of stability is the north coast of the largest island, Viti Levu, where large mangrove-filled deltas and an uncommonly com- plex pattern of offshore reefs also suggest late Holocene sta- bility. Emerged notches around headlands and emerged beachrock at Natunuku are at levels consonant with the mid- Holocene sea-level maximum derived for Fiji and found else- where in the region.

Two pieces of emerged reef, representing a maximum 3 m of emergence, are found on the offshore reef named Sucutolu, described by RODDA (1990). Like the dates from Vatia and Afono on Tutuila (see above), those from Sucutolu are sur- prisingly young and are discussed further below.

Gambier Islands (French Polynesia)

The intraplate Gambier Islands form an "almost-atoll" around Mangareva. They formed 7.2-5.2 Ma (BRoussE et al., 1972) and are probably part of a hotspot trace which includes Pitcairn and Moruroa (Mururoa) islands (DUNCAN and CLAGUE, 1985). There is no reason why they should have been affected by significant tectonism during the latest Neo- gene. Evidence for a mid-Holocene sea-level maximum is widespread; the only record of emergence within the last 1,000 years is on the motu (sand island) named Tarauru Roa

on the northeast barrier reef (PIRAZZOLI and MONTAGGIONI, 1987).

Guam

Guam is a large island in the northwest Pacific lying close to the convergent plate boundary in the region. No volcanism has occurred since the early Neogene (middle Miocene) but

uplift associated with forearc deformation occurred within the later Neogene. The record of mid-Holocene sea levels from

Aga Point suggests that late Holocene tectonism has been minimal at this site and that the two dates from the last millennium can be considered reasonable indicators of the

contemporary reef surface (KAYANNE et al., 1988). These dates are considered minima because it cannot be demon- strated that the contemporary reefs had reached their max- imum level (relative to sea level) or were a little below it.

Kosrae, Caroline Islands

The islands of Kosrae in the eastern Caroline islands are the remains of volcanoes which formed some 4 Ma as part of a hotspot island chain. Recent work by KAYANNE et al. (1995) has shown that the maximum Holocene sea level occurred some 3800 BP and reached around 1 m above the present.

The history of later Holocene sea-level change has been determined largely by dates from mangrove peats which, for reasons stated above, yield only minimum levels for contem-

porary sea levels. Yet, assuming that the processes involved in mangrove peat displacement (compaction and subsidence) are continuous, then the younger the date the less the peat would be expected to have been displaced from the level (rel- ative to the present) at which it formed. The peat formed on Kosrae some 720 BP is so close in level to other indicators of

contemporary sea level that it is considered unlikely to have been displaced significantly from the level at which it formed.

New Zealand

Many parts of the islands of New Zealand are tectonically active, but this is well documented (PILLANS, 1986) and there are areas of little or no significant tectonism, particularly along the hinge zones between areas of subsidence and areas of uplift.

GIBB (1986) established the location of such "stable" areas of New Zealand by measuring the relative heights of the Last

Interglacial shoreline. Any place where this was not 3.0 ? 0.3 m above its modern analogue was considered by Gibb to have been tectonically active during the subsequent period. Two places where such stability was found were the Blueskin Bay estuary in South Island, and the Weiti River estuary in North Island. Both these sites proved extremely fruitful in deciphering the course of late Holocene sea-level changes, in- cluding the last 1,000 years.

Rota

Like Guam (see above), Rota is an island along the Mari- anas forearc in the northwest Pacific. Unlike Guam, the high- est Holocene emerged reef reaches almost 4 m above the mod- ern reef, suggesting that the island has experienced late Ho-

Journal of Coastal Research, Vol. 14, No. 1, 1998

This content downloaded from 198.161.51.4 on Wed, 14 Oct 2015 01:55:54 UTC All use subject to JSTOR Terms and Conditions

26 Nunn

3.0_ ? LITTLE CLIMATIC OPTIMUM O LITTLE ICE AGE RECENT ( SEA-LEVEL W 13. RISE . I. z ? ? ? 2.0 O 14. 180 S18 20

1.0

c2

1200 1100 1000 900 800 700 600 500 400 300 200 100 0

Years (BP) Years (BP)

Figure 2. Sea-level changes over the past - 1,000 years. Dates and shoreline displacements are listed in Table 1 and sites shown in Figure 1.

locene uplift (KAYANNE et al., 1988). Yet there is no evidence for uplift within the past few hundred years so the record from just 115 BP is considered likely to be a reliable indicator of contemporary sea level.

Tuamotus

The Tuamotu archipelago in French Polynesia comprises two main chains of islands, mostly atolls, which have been linked to a hotspot to the southeast. Most islands rise from volcanic basements of early Neogene age and have been sub- siding slowly since that time. For short time periods such as the Holocene, these islands can be regarded as effectively sta- ble, which is why there has been so much interest shown in them by researchers concerned with Holocene sea-level change (PIRAZZOLI and MONTAGGIONI, 1986, 1988; PIRAZ- ZOLI et al., 1988). Four dates from the last 1000 years are used in this study.

SEA-LEVEL CHANGES OVER THE PAST 1,000 YEARS

The 24 dates from the "stable" sites described above are plotted (age versus emergence magnitude) in Figure 2, and an envelope drawn around those which appear most repre- sentative. This envelope is intended to enclose the most likely course of sea-level change during this period in the Pacific. Unlike for earlier periods, for which contemporary sea level was perhaps not parallel to the present sea level on account of geoidal change (NUNN, 1986, 1994a), the past 1,000 years has probably not witnessed geoidal changes of magnitude suf- ficient to have caused significant intraregional variation. It would thus be expected that sea level would have changed at approximately the same rates and by the same amounts at

the same times for the past 1,000 years. As a consequence, the envelope is drawn almost as narrowly as it can.

The account of last-millennium sea-level changes is divided for ease of discussion into four: the Little Climatic Optimum marked by sea-level rise (- 1050 BP to 690 BP), the transi- tion period between this and the following period (690 BP to ~ 575 BP), the Little Ice Age marked by generally lower than

present sea levels (- 575 BP to - 150 BP), and the period of recent warming (- 150 BP to present).

The Early Period: Little Climatic Optimum sea-level rise

Around the beginning of this period, sea level was close, perhaps a little above, its present level. The early part is constrained by dates 3 and 5 and by minima from Guam (dates 4 and 6). The later part of this period, around 700 BP, marks the maximum sea level in the Pacific during the last millennium. Available data suggest that sea level may have reached as much as 0.9 m above its present level.

The sea-level rise during this period is coincident (as would be expected) with a period of temperature warming named the Little Climatic Optimum for which evidence has been re- ported parts of the world including the Pacific (NUNN, 1990b, 1992). On many Pacific islands, this time was marked by less rainfall compared to the present, which is believed to have inspired water-conservation strategies, such as terracing, by the contemporary inhabitants of many Pacific islands. It has also been argued that this time was marked by less cyclonic activity than at present and a higher frequency of clear skies which encouraged the successful long-distance colonisation of lands (such as Hawaii, Easter Island [Rapanui], New Zealand and the coast of Panama) remote from the Marquesas is- lands, the easternmost population centre in the Pacific is- lands at the time (BRIDGMAN, 1983; NUNN, 1992).

Journal of Coastal Research, Vol. 14, No. 1, 1998

This content downloaded from 198.161.51.4 on Wed, 14 Oct 2015 01:55:54 UTC All use subject to JSTOR Terms and Conditions

Holocene Sea-Level Changes in the Pacific Basin 27

The Transition Around 690 BP

The available sea-level data for this period points to a com- paratively abrupt fall followed by almost 200 years for which there are no data. Both observations are worthy of comment.

The sea-level fall - 700-650 BP has been found to have coincided with a rapid fall of ground temperature, which pre- sumably drove it (NUNN, 1992). This time is also believed to have coincided with a short-lived rise in precipitation, per- haps associated with increased cyclonic activity arising from reorganisation of wind systems (NuNN, 1994b).

Support for these changes comes from consideration of is- land environments and their human inhabitants at this time. The falling temperature and vastly increased precipitation would have devastated contemporary agricultural systems, leading to catastrophic forest destruction and soil erosion in many places. On many islands, this time marks the approx- imate time at which food resources were perhaps dramati- cally depleted so that people had to compete (as they had never done before) for the few remaining resources. Excellent examples are provided by Easter Island, where the resulting societal strife is well marked in the archaeological record, and island groups like Fiji where a transition from unfortified coastal settlements to fortified hilltop settlements is conspic- uous around this time.

The marked hiatus of around 200 years for which there are no sea-level data suggests that sea level had fallen so low that its subsequent rise (during the Little Ice Age - 430 BP- see below) obscured all records of shoreline change. For ex- ample, the fall of sea level of perhaps as much as 1.4 m dur- ing the transition would have forced most coral reefs to begin growing outwards; the reef-surface corals which grew at this time would have been covered by those which grew up to the subsequent maximum.

The possibility that the transition was coincident with in- creased storminess means that many nearshore sediment transport systems would have been disrupted and may not have recovered until the onset of drier conditions during the Little Ice Age (see below).

The Middle Period: Low Sea Levels of the Little Ice Age Most dates presented in Table 1 suggest that the early part

of the Little Ice Age was marked by a rise to a sea-level max- imum perhaps a little above present sea level - 430 BP (Fig- ure 2). The remainder of the Little Ice Age appears to have been marked by a slowly falling sea level which may have reached almost -0.9 m around 200 BP. This part of the curve is not particularly well constrained. A notable feature of the sea-level data from supposedly stable sites during the Little Ice Age are the dates (13, 14, 18, 20, 21, 23) which are clearly in excess of those contained within the sea-level envelope (Figure 2). A discussion of these is given separately below.

Evidence for the Little Ice Age has been found throughout the world (GROVE, 1988). Temperatures in the Pacific were significantly lower than those both of today and of the Little Climatic Optimum (THOMPSON et al., 1986; QUINN et al., 1993), and it is this which suggests that the sea-level enve- lope should be drawn as it is for the Little Ice Age in Figure 2 rather than at a higher level.

In the Pacific islands, the abrupt cessation of successful

long-distance colonisation and interchange, even within ar-

chipelagoes like the Marquesas, suggests that storminess and cloudiness had both increased during the Little Ice Age com-

pared to the Little Climatic Optimum (BRIDGMAN, 1983; NUNN, 1992). Many Pacific island societies felt the continu-

ing effects of the climate change during the earlier transition (see above).

The Final Period: Recent Sea-Level Rise

There are no publicly available series of continually moni- tored sea-level data for the Pacific for the early part of this

period. Yet SNITNIKOV (1969) reported that data from three

unspecified stations in the Pacific between 143 and 0 BP (AD 1807-1950) indicated a sea-level rise of 11.8 cm.

The most dependable series of continually monitored sea- level data begin around the start of the twentieth century and indicate that Pacific sea level has been rising at net rates of 1.0-1.5 mm/year for the past - 100 years (WYRTKI, 1990; NUNN, 1992). This time has also been marked by rising tem-

peratures and, as for most of the past millennium (NUNN, 1992; SALINGER et al., 1995), it seems probable that temper- ature change has been driving sea-level change in the same direction.

Anomalous Data: Discussion

Dates 1, 13, 14, 18, 20, 21 and 23 (Table 1 and Figure 2) are clearly anomalous in relation to those used to character- ise sea-level changes of the last millennium yet all these dates were assumed to have come from sites which had been

tectonically stable over this period of time. An understanding of how these dates and the sites from which they were ob- tained came to be misinterpreted is instructive for anyone interested in reconstructing the course of past sea-level

changes. Aside from survey error, which is regarded as highly im-

probable, there are five explanations of these anomalous dates.

(a) It is possible that the sites have experienced uplift within the few hundred years since they formed. Given our

present understanding of the longer-term tectonic his-

tory of the areas in question, particularly the levels at which the Holocene sea-level maximum was recorded, such an amount of uplift seems extremely unlikely.

(b) It is possible that sea level in these parts of the Pacific stood significantly higher than that shown within the envelope in Figure 2. For the Little Ice Age, the dates within the envelope are constrained mostly by those from New Zealand. Had the sea surface in the Pacific at the time not been parallel to the present sea level, this would allow such a situation, but it also requires the oceanic geoid in the region (between Samoa and New Zealand) to have deformed in level by - 3 m in the last 520 years BP. Such rates of geoidal deformation are greatly in excess of those which writers such as MORNER (1976) proposed. In addition, it would be remarkable if terrestrial evidence of sea levels - 3 m higher than pres-

Journal of Coastal Research, Vol. 14, No. 1, 1998

This content downloaded from 198.161.51.4 on Wed, 14 Oct 2015 01:55:54 UTC All use subject to JSTOR Terms and Conditions

28 Nunn

Table 1. Data referring to eustatic shoreline displacements at tectonically-stable sites in the Pacific islands over the past -1,000 years. Locations shown in Figure 1, age-displacement data plotted in Figure 2.

Emergence Age Island Place Magnitude (m) (years BP) Source of Information

1 off Viti Levu, Fiji Sucutolu --3.0 1,150 ? 60 this paper 2 Faaite, Tuamotus western coast 0.45 ? 0.2 1,080 ? 55 PIRAZZOLI et al., 1988 3 Tarauru Roa, Gambier - 0.6 950 ? 70 PIRAZZOLI and MONTAGGIONI,

Islands 1987 4 Guam Aga Point -0.66 + 0.031 940 ? 120 MATSUMOTO and KAYANNE,

1988 5 South Island, New Zea- Blueskin Bay estu- 0.02 ? 0.9 907 ? 62 GIBB, 1986

land ary 6 Guam Aga Point -0.72 ? 0.031 900 + 140 MATSUMOTO and KAYANNE,

1988 7 Tutuila, American Samoa Poloa 0.745 740 ? 60 this paper 8 Kosrae, Caroline Islands Utwa 0.61 720 ? 80 KAWANA et al., 1995 9 Rangiroa, Tuamotus Tepaetia 0.6 705 ? 65 PIRAZZOLI and MONTAGGIONI,

1985 10 Hikueru, Tuamotus Tekotaha 0.25 ? 0.1 700 ? 60 PIRAZZOLI et al., 1988 11 Tutuila, American Samoa Poloa 0.89 690 ? 70 this paper 12 North Island, New Zea- Weiti River estu- 0.12 ? 0.23 670 ? 50 GIBB, 1986

land ary 13 off Viti Levu, Fiji Sucutolu -3.0 520 ? 35 this paper 14 Tutuila, American Samoa Vatia Bay 2.11 490 ? 60 this paper 15 North Island, New Zea- Weiti River estu- 0.0 ? 0.24 470 ? 50 GIBB, 1986

land ary 16 South Island, New Zea- Blueskin Bay estu- 0.0 ? 0.5 413 ? 30 GIBB, 1986

land ary 17 North Island, New Zea- Weiti River estu- -0.46 ? 0.24 395 + 34 GIBB, 1986

land ary 18 Tutuila, American Samoa Vatia Bay 2.11 380 ? 60 this paper 19 North Island, New Zea- Weiti River estu- -0.23 ? 0.22 365 ? 30 GIBB, 1986

land ary 20 Tutuila, American Samoa Vatia Bay 2.11 350 ? 50 this paper 21 Tutuila, American Samoa Afono Bay 1.24 270 ? 60 this paper 22 Guam Aga Point -1.06 ? 0.041 260 ? 135 MATSUMOTO and KAYANNE,

1988 23 Marokau, Tuamotus northwest 0.4 ? 0.1 210 + 70 PIRAZZOLI et al., 1988 24 Rota west -0.77 ? 0.02 115 ? 160 MATSUMOTO and KAYANNE,

1988

1 Original data for northwest Pacific islands presented relative to mean sea level were converted relative to the assumed mean level of the modern reef surface by subtracting 0.9 m

ent a few hundred years ago had eluded investigators up until this point.

(c) In some deep embayments along the north coast of Tu- tuila island in American Samoa corals are growing in rock-walled pools regularly sluiced by surf almost 3 m above the main reef level. Were surf unable to reach these pools, their corals would die and could easily be mistaken as emerged reef of much earlier date. The dif- ficulty with this when applied to the sites on Tutuila is that at these there are blocks of massive reef coral, not isolated coral heads attached to volcanic slopes. This ex- planation is inapplicable to the open-ocean sites at Ma- rokau and Sucutolu.

(d) It is also possible that these blocks of emerged reef are parts of the submarine reef which have been broken off during storms and thrown up on the modern reef flat. There are many of these on reefs in the tropical Pacific. Yet, since the corals in these blocks are all in growth position and, in some cases, it is also clear from reef stra- tigraphy that the blocks are the 'right way up', one can only conclude that it is chance which has caused these

storm-tossed blocks to have landed on the reef flat in this way. The chance of this happening is so small that this explanation is likewise dismissed.

(e) The possibility that the dates obtained on these samples are too young because of post-emergence contamination of these reefs cannot be readily dismissed although no single contributory factor can be identified.

In conclusion, there is no satisfactory answer as to the or- igin of anomalous dates from Marokau, Sucutolu and Tutuila. The most likely explanation is that the emerged reefs were contaminated and consequently sample dates were too young. Further work on the sites in question would be useful.

CONCLUSIONS

This study shows that the sea level during the past 1,000 years in the Pacific was not unchanging but oscillated in a manner which may well mimic, albeit at lower amplitudes, the fluctuations of Holocene sea level (FAIRBRIDGE, 1992; NUNN, 1995).

The possibility that these oscillations may be cyclical and

Journal of Coastal Research, Vol. 14, No. 1, 1998

This content downloaded from 198.161.51.4 on Wed, 14 Oct 2015 01:55:54 UTC All use subject to JSTOR Terms and Conditions

Holocene Sea-Level Changes in the Pacific Basin 29

therefore reproducible has led to use of the sea-level rise dur- ing the Little Climatic Optimum as a proxy for the period of recent sea-level rise; such studies suggest that, irrespective of predicted sea-level rise associated with the human-en- hanced greenhouse effect, sea level would probably continue rising until around the end of the twenty-first century (NUNN et al., 1994).

This study has also shown that the record of sea-level change over periods as short as 1,000 years is possible al- though the likelihood of anomalous dates is high. The possi- bility that anomalous dates of the kind described here (dates 13, 14, 18, 20, 21 and 23 in Figure 2) have been inadvertently incorporated into other studies of Holocene and earlier sea- level change is manifest and serves to emphasise the need to ensure that the tectonic history of particular sites is known accurately, especially when dealing with such short time pe- riods.

ACKNOWLEDGEMENTS

Peter Rodda (Fiji Mineral Resources Department) kindly provided samples and field results from Sucutolu. Dr. James Terry (The University of the South Pacific) commented on an early draft of this paper.

The writer's research in American Samoa was funded by National Science Foundation grant 9111566. Radiocarbon dates from Sucutolu were funded by grant 0703-9405 from the University of the South Pacific. I am also grateful to the Pacific Science Association for helping fund my attendance at the Beijing congress where this paper was presented.

LITERATURE CITED

BLOOM, A.L., 1980. Late Quaternary sea level change on South Pa- cific coasts: A study in tectonic diversity. In: MORNER N.A., (ed.), Earth Rheology, Isostasy, and Eustasy. New York: Wiley, pp. 505- 515.

BRIDGMAN, H.A., 1983, Could climatic change have had an influence on the Polynesian migrations? Palaeogeography, Palaeoclimatolo- gy, Palaeoecology, 41, 193-206.

BROUSSE, R.; PHILLIPPET, J.-C.; GUILLE, G., and BELLON, H., 1972. Geochronometrie des iles Gambier (Ocean Pacifique). Comptes Rendus de lAcademie des Sciences, Paris, 274, p. 1995.

DALY, R.A., 1920. A recent worldwide sinking of ocean-level. Geolog- ical Magazine, 57, 246-261.

DALY, R.A., 1924. The geology of American Samoa. Carnegie Institute Publication, 340, 95-145.

DALY, R.A., 1934. The Changing World of the Ice Age. Yale/Oxford University Press.

DUNCAN, R.A. and CLAGUE, D.A., 1985, Pacific plate motions re- corded by linear volcanic chains. In NAIRN, A.E.M.; STEHLI, F.G. and UYEDA, S., (eds.), The Ocean Basins and Margins, Volume 7A, The Pacific, New York: Plenum, pp. 89-121.

FAIRBRIDGE, R.W., 1992. Holocene marine coastal evolution of the United States. In: FLETCHER, C.H. and WEHMILLER, J.F. (eds.) Quaternary Coasts of the United States: marine and lacustrine sys- tems. Tulsa: Society for Sedimentary Geology, pp. 9-20.

GIBB, J.G., 1986. A New Zealand regional holocene eustatic sea-level curve and its application to determination of vertical tectonic movements. Royal Society of New Zealand, Bulletin, 24, 377-395.

GROVE, J.M., 1988. The Little Ice Age. London: Methuen. KAWANA, T.; MIYAGI, T.; FUJIMOTO, K., and KIKUCHI, T., 1995. Late

Holocene sea-level changes and mangrove development in Kosrae island, the Carolines, Micronesia. In: KIKUCHI, T. (ed.) Rapid Sea Level Rise and Mangrove Habitat. Gifu, Japan: Institute for Basin Ecosystem Studies, Gifu University, pp. 1-7.

KAYANNE, H.; YONEKURA, N.; ISHII, T., and MATSUMOTO, E., 1988. Geomorphic and geologic development of Holocene emerged reefs in Rota and Guam, Mariana Islands. In: YONEKURA, N. (ed.) Sea- Level Changes and Tectonics in the Middle Pacific: Report of the HIPAC Project in 1986 and 1987. University of Tokyo, pp. 35-57.

KEATING, B.H., 1992. The geology of the Samoan islands. In: KEAT- ING, B.H. and BOLTON, B.R., (eds.). Geology and Offshore Mineral Resources of the Central Pacific Basin. New York: Springer-Verlag (Circum-Pacific Council for Energy and Mineral Resources, Earth Science Series, Volume 14), pp. 127-178.

KIRCH, P.V.; HUNT, T.L.; NAGAOKA, L., and TYLER, J., 1990. An ancestral Polynesian occupation site at To'aga, Ofu Island, Amer- ican Samoa. Archaeology in Oceania, 25, 1-25.

MATSUMOTO, E. and KAYANNE, H., 1988. Carbon dating results for samples from the HIPAC project in 1988. In: YONEKURA, N. (ed.) Sea-Level Changes and Tectonics in the Middle Pacific: Report of the HIPAC Project in 1986 and 1987. University of Tokyo, pp. 19- 22.

MORNER N.-A., 1976. Eustasy and geoid changes. Journal of Geology, 84, 123-151.

NATLAND, J.H. and TURNER, D.L., 1985. Age progression and pet- rological development of Samoan shield volcanoes: evidence from K-Ar ages, lava compositions and mineral studies. In: BROCHER, T.M., (ed.). Geological Investigations of the Northern Melanesian Borderland. Houston: Circum-Pacific Council for Energy and Min- eral Resources, pp. 139-171.

NUNN, P.D., 1986. Implications of migrating geoid anomalies for the interpretation of high-level fossil coral reefs. Geological Society of America, Bulletin, 97, 946-952.

NUNN, P.D., 1990a. Coastal processes and landforms of Fiji and their bearing on Holocene sea-level changes in the south and west Pa- cific. Journal of Coastal Research, 6, 279-310.

NUNN, P.D., 1990b. Recent environmental changes on Pacific is- lands. The Geographical Journal, 156, 125-140.

NUNN, P.D., 1991a. Sea-level changes during the last 6000 years from Fiji, Tonga and Western Samoa: implications for future coastline development. United Nations ESCAP, CCOP/SOPAC Technical Bulletin 7, 79-90.

NUNN, P.D., 1991b. Tectonic environments of Fiji. United Nations ESCAP, CCOP/SOPAC Technical Bulletin 7, 67-76.

NUNN, P.D., 1995. Holocene sea-level changes in the south and west Pacific. Journal of Coastal Research, Special Issue 17, 311-319.

NUNN, P.D., 1992. Keimami sa vakila na liga ni Kalou (Feeling the hand of God): Human and nonhuman impacts on Pacific island environments. East-West Center, Occasional Paper (2nd revised edition), 13, 69 pp. [3rd edition available from SSED Editorial, The University of the South Pacific, Suva, Fiji]

NUNN, P.D., 1994a. Oceanic Islands. Oxford: Blackwell, 413p. NUNN, P.D., 1994b. Beyond the naive lands: human history and en-

vironmental change in the Pacific Basin. In: WADDELL, E. and NUNN, P.D. (eds.). The Margin Fades: Geographical Itineraries in a World of Islands. Suva, Fiji: Institute of Pacific Studies, The University of the South Pacific, pp. 5-27.

NUNN, P.D., 1997. Pacific Island Landscapes. Suva: Institute of Pa- cific Studies, The University of the South Pacific, in press.

NUNN, P.D.; RAVUVU, A.D.; AALBERSBERG, W.; MIMURA, N., and YAMADA, K., 1994. Assessment of Coastal Vulnerability and Resil- ience to Sea-Level Rise and Climate Change. Case Study: Yasawa Islands, Fiji. Phase 2: Development of Methodology. Apia, Western Samoa: South Pacific Regional Environment Programme, 118 pp.

PILLANS, B., 1986, A late Quaternary uplift map for North Island, New Zealand. Bulletin of the Royal Society of New Zealand, 24, 409-17.

PIRAZZOLI, P.A. and MONTAGGIONI, L.F., 1986. Late Holocene sea- level changes in the northwest Tuamotu islands, French Polyne- sia. Quaternary Research, 25, 350-368.

PIRAZZOLI, P.A. and MONTAGGIONI, L.F., 1987. Les iles Gambier et l'atoll de Temoe (Polynbsie frangaise): anciennes lignes de rivage et comportement gdodynamique. Gdodynamique, 2, 13-25.

PIRAZZOLI, P.A. and MONTAGGIONI, L.F., 1988. Holocene sea-level changes in French Polynesia. Palaeogeography, Palaeoclimatology, Palaeoecology, 68, 153-178.

Journal of Coastal Research, Vol. 14, No. 1, 1998

This content downloaded from 198.161.51.4 on Wed, 14 Oct 2015 01:55:54 UTC All use subject to JSTOR Terms and Conditions

30 Nunn

PIRAZZOLI, P.A.; MONTAGGIONI, L.F.; SALVAT, B., and FAURE, G., 1988. Late Holocene sea level indicators from twelve atolls in the central and eastern Tuamotus (Pacific Ocean). Coral Reefs, 7, 57- 68.

RODDA, P., 1990. Sucutolu (the three sisters). Fiji Mineral Resources Department, Unpublished Note BP1/88.

QUINN, T.M.; TAYLOR, F.W., and CROWLEY, T.J., 1993. A 173 year stable isotope record from a tropical south Pacific coral. Quater- nary Science Reviews, 12, 407-418.

SALINGER, M.J.; BASHER, R.E.; FITZHARRIS, B.B.; HAY, J.E.; JONES, P.D.; MACVEIGH, J.P., and SCHMIDELY-LELEU, I., 1995. Climate

trends in the south-west Pacific. International Journal of Clima-

tology, 15, 285-302. SNITNIKOV, A.V., 1969. Some material on intrasecular fluctuations

of the climate of northwestern Europe and the north Atlantic in the 18th-20th centuries. Geogr. Obs. SSSR, Moscow (Akad.Nauk), 5-27 [in Russian].

THOMPSON, L.G.; MOSLEY-THOMPSON, E.; DANSGAARD, W., and GROOTES, P.M., 1986. The Little Ice Age as recorded in the stra- tigraphy of the tropical Quelccaya ice cap. Science, 234, 361-364.

WYRTKI, K, 1990. Sea level rise: The facts and the future. Pacific Science, 44, 1-16.

Journal of Coastal Research, Vol. 14, No. 1, 1998

This content downloaded from 198.161.51.4 on Wed, 14 Oct 2015 01:55:54 UTC All use subject to JSTOR Terms and Conditions