Monitoring sea-level change in Oceania
The issue of sea-level change in Oceania has received considerable political and media attention. Most of the debate has been focused on the scientific predictions of sea-level rise produced by the Intergovernmental Panel on Climate Change (IPCC). The most recent estimates come from the IPCC's 2001 Assessment Report which gives a range for its predicted global sea-level rise to the year 2100, with a central value of 48 cm.
A key concern for the Oceania region is the significant number of low-lying atolls, which are already susceptible to regional sea-level fluctuations and storm events. A recent article in Tiempo by Paul Kench and Peter Cowell (Issue 46, December 2002) provided some very useful information about how low-lying atolls respond to changes in sea level. In particular, they used sediment transport modelling to demonstrate how these coasts are unique in their response to sea-level rise. The authors showed how sediments on low-lying atoll islands can be rolled over toward the lagoon rather than the textbook examples of sandy beaches that show a simple readjustment of the beach profile. Sediment transport models were used to provide estimates of the magnitude of change which would occur under different sea-level rise scenarios for specific sites such as South Tarawa in Kiribati. This type of information is important for management purposes when examining atoll island vulnerability.
In addition to these computer-based simulation models, it is also essential to collect real information on what is actually happening to the coasts of these atoll islands. Thus, it is necessary to have very accurate monitoring of sea-level change and coastal change. This article discusses both of these needs.
Monitoring sea-level change
At a recent Pacific Islands conference in Rarotonga in the Cook Islands in 2000, Bill Mitchell and others discussed the issue of sea-level rise in the Pacific region. They first examined long-term tide-gauge records for the region (covering more than 50 years) collected by the University of Hawaii. This gave a positive relative sea-level trend of 1.07 mm/yr. When they used stations with more than 25 years of record it was possible to analyse 27 sites in total giving a positive relative sea-level trend of 0.8 mm/yr.
However, Mitchell and others commented on the poor quality of some of the tide-gauge records. They suspected datum shifts, gaps in the records and a lack of gauge maintenance, although they did recognize the fact that the gauges were set up to record sea-level fluctuations rather than sea-level trends. It should be noted that Mitchell and others did not adjust their sea-level trends for land movement.
A more sophisticated SEAFRAME (Sea level Fine Resolution Acoustic Measuring Equipment) tide-gauge array has now been deployed at 12 sites to monitor sea level in the region (Figure 1). Feasibility studies have been conducted for an additional two sites. At each site, sea level is measured with an accuracy of up to 1 mm by an acoustic sensor together with a back-up measurement using a pressure sensor.
The tide-gauge array is part of a South Pacific monitoring project funded by the Australian Agency for International Development. In addition to the precision of the recording instrument, each site is accurately surveyed to monitor for any datum changes and continuous global positioning system (CGPS) equipment provides accurate locational data.
In order to produce reliable sea-level trends, it is necessary to collect continuous data from the SEAFRAME gauges over a period capturing several El Niño events. Global analyses of long-term tidal data have used sites that are not affected by local land movements and have tide gauge records of 70 years or greater. The resultant mean sea-level trends are then corrected using a global isostatic adjustment to allow for the variable coastal response to the global redistribution of water (and ice in the Northern Hemisphere) following the last glacial cycle. This effect has been modelled geophysically and verified at a number of sites, including some preliminary analyses in the Oceania region.
Unfortunately, the SEAFRAME tide-gauge array has only been operating for around ten years and it is still too early to obtain reliable long-term estimates of trends, although it is clear that some local influences are affecting the record such as in Tonga, where an abnormally high sea-level change is being recorded. See the Australian National Tidal Facility website for trend updates.
The SEAFRAME gauges can already provide useful information on regional changes. The ten-year record of sea-level anomalies (Figure 2) shows a dramatic drop in sea level associated with the 1997/98 El Niño. Some countries such as Tuvalu experienced a drop in sea level of almost 40 cm during that event. Other countries such as Nauru experienced a sea-level rise in the year preceding this event, resulting in a total sea-level variation of over 50 cm from one year to the next. Such a sea-level change over a short period is likely to impact coastal processes, but there are no accurate measurements of these effects.
Monitoring the effects of sea-level change on the coast
While there has been considerable monitoring of coastal change in the Oceania region, the timing, purpose and results of these studies varies considerably. Consequently, there has been no systematic monitoring programme. There has been some excellent long-term monitoring completed for the islands of Tuvalu where a detailed re-survey was conducted in 1991 so that it could be compared with the early expedition survey conducted in 1881. However, only small coastal changes were seen to have occurred during the 100-year period between the two surveys.
There have also been a number of short- to medium-term monitoring studies conducted elsewhere in the region but these lack a systematic approach and are often ad hoc or have irregular times between surveys. Moreover, most have been conducted in areas affected by human impact such as causeway construction, dredging or reclamation. In Kiribati, for example, there has been extensive monitoring conducted for South Tarawa with a particular focus on the effects that the Nippon causeway has had on nearby coastal erosion.
Regional organizations such as SOPAC (South Pacific Applied Geo-Science Commission) have provided important assistance to countries requesting help for specific coastal monitoring programmes to identify coastal changes resulting from areas where human impact has altered coastal processes. The problem with this is that much of this coastal change caused by human impact is likely to outweigh any impacts which could be attributable to sea-level change by itself. For this reason, it is often very difficult to separate the human impact from any sea-level change impacts on the coast.
In order to address this problem, the Asia-Pacific Network for Global Change Research funded a seeding project in 2002 to establish a systematic coastal monitoring network in Oceania.
The significance of this particular project is that the monitoring network is focused on the three low-lying atoll countries of Kiribati, the Marshall Islands and Tuvalu and, once the network has been set up, it will be run by the countries themselves. Eventually it will provide some real evidence about what is happening to the coast as a result of sea-level change. Preliminary results from this project were recently presented by Paul Kench at the Australasian Ports and Harbours conference in New Zealand in 2003.
The Asia-Pacific Network project set up the Pacific Atoll Coastal Monitoring and Analysis Network (PACMAN). The aim of the project is to conduct systematic coastal monitoring on atoll islands and to test if any observed coastal changes are related to corresponding sea-level changes.
In order to do this, PACMAN needed to ensure that there was sufficient institutional and human capacity in each of the three participating countries and that there was an appropriate coordination mechanism located within the region. It was decided that START-Oceania, based in Suva, Fiji, was best placed to do the coordination since it was the regional focus for a broader international network with a mandate for capacity building and training in global change issues.
The first PACMAN meeting took place in Fiji in December 2002 with representatives from Fiji, Tuvalu and Kiribati (flight cancellations prevented the early participation from the Marshall Islands). A training workshop provided participants with a background to coastal monitoring and surveying techniques relevant for atoll coasts.
The meeting also established linkages and initiated cooperation with key government agencies in the region. Subsequently, PACMAN has been supported by the Lands and Survey Department of the Tuvalu government and the Ministry of Natural Resources Development of the Kiribati government. Without such cooperation, this monitoring network would not be possible.
The next stage was to conduct a monitoring survey in Tuvalu with the cooperation of government surveyors. Although previous two-dimensional survey data existed for Funafuti Atoll, most of this was focused on the main island of Fongafale where there had been considerable human impact. In order to avoid these artificial influences on coastal change it was decided to use the uninhabited island of Fatato for the inaugural PACMAN survey (Figure 3).
In January 2003, a PACMAN team from Fiji, Kiribati and Tuvalu, assisted by coastal research experts from Australia and New Zealand, carried out a detailed survey of Fatato Island. Monitoring was conducted using both cross-island profiling using survey equipment and mapping of the island shoreline using global positioning system (GPS) equipment. This GPS survey on Tuvalu was then linked to data from the CGPS equipment described above as part of the SEAFRAME sea-level monitoring programme. This meant that the GPS survey of Fatato could be corrected to sub-centimetre accuracy. In addition, fixed benchmarks were located on the island so that all the cross-island and island coast mapping surveys can be accurately located and re-surveyed at a later date. The survey shows both the topography of the island and its position on the reef surface so that changes in either of these can be determined by subsequent surveys. With sufficient data it will be possible to link these to any sea-level changes monitored from the SEAFRAME records.
Although there has been a lot of discussion and debate linking the vulnerability of atoll islands to the impacts of global change, this has suffered for several reasons.
There is scientific uncertainty about sea-level rise predictions. Problems exist with interpretations of sea-level trends from tidal data for the last century. There is a poor understanding of change and dynamics in atoll environments. Erosion is often quoted as an impact of sea-level rise but there are very few studies giving accurate assessments of how much erosion will occur and how the low-lying reef island will respond. Not enough is known about the natural variability of atoll islands in order to identify global change-induced effects.
The result has been that many assessments are, to a large extent, based on speculation. This has not helped the small-island states of the atoll countries, particularly when discussing the impacts of global change in international fora.
While there is always likely to be debate over the scientific accuracy of sea-level rise predictions, it is possible to address some of the uncertainties related to monitoring and a better understanding of island change.
As discussed above, the SEAFRAME project is already providing accurate sea-level monitoring data indicating significant regional variations in the sea-level record for the past ten years in the Oceania region. The sea-level anomalies from these data indicate that sea level can vary by as much as 50 cm from one year to the next in association with El Niño events. Eventually, the SEAFRAME programme will be able to provide accurate estimates of mean sea-level trends for the region.
The PACMAN project has only just begun but promises to provide accurate data on atoll- island change, which is not affected by localized human impact. Once sufficient data are collected, it will be possible to assess whether or not the coastal change is related to sea-level change.
An important aspect of this project is that the PACMAN network is intended to be run by the atoll-based countries themselves and will not need to rely on foreign aid.
In conclusion, the lack of hard data means that it is difficult for low-lying atoll countries, which already have less scientific resources and funding compared to the developed nations of Oceania, to present their case for global change impacts such as the scenarios of total island disappearance. In order to present accurate statements of island vulnerability, it is important that assistance is given to these countries to obtain accurate monitoring data and documented evidence of island shoreline change linked to sea-level change.