Sea levels are an integral part of a complex system representing the integration of a series of coupled ocean-atmosphere and ocean-land processes. Not only does the sea surface respond to short term diurnal, seasonal and inter-annual changes, but it also responds to changes in freshwater inflow, heat flux and other factors that are linked to climate change processes. However, the critical signals resulting from these climate change processes are often obscured by the natural variability that occurs over shorter time scales thus giving particular importance to reliable data collected over long periods of time.

New Zealand has been monitoring sea levels since the last decade of the 19th Century and has some of the longest and most reliable tide gauge records in the southern hemisphere. In addition, it has traditionally had good administrative systems at all levels which, until recent years, have enabled these records to be retained. Because of these systems, the tide gauges from which these records have been collected have been able to be referenced to stable benchmarks in their vicinity and thence linked into a national coordinate system. This data, although collected by a variety of authorities, have traditionally been maintained and made available as an integral part of their contribution to a greater national good.

Economic and administrative changes in the last 12 years have altered this picture dramatically. The introduction of commercial entities in place of Harbour Boards (the former owners and administrators of most tide gauges), together with the ongoing restructuring of government agencies, has led to a period of time during which the importance of tide gauges and their associated records have been largely overlooked. In this process, the "national good" ethic has been largely lost, lines of communication ruptured, and data of greatly diminished quality collected.

It has only been in the last two years that new initiatives have resulted in the possibility of an integrated sea level monitoring programme being developed in New Zealand. This paper describes one of these initiatives.

A Global Perspective

The influence on the world's climate of increasing amounts of CO2, aerosols and various other greenhouse gases being released into the atmosphere, has been of concern to the scientific and civilian communities for the last two decades. A number of scientific studies covering a broad spectrum of environmental conditions have been made in order to assess possible anthropogenic changes induced on the global climate. These studies have predominantly been made with a view towards determining trends and patterns in the world's climate and sea level, including the rate at which temperatures have increased, changes in precipitation levels, the degree of sea level rise and any increase in the variability of climate. See for example Houghton et al. (1990) and Houghton et al. (1995).

Rising global sea levels are considered to be one indicator of global warming. Historically, tide gauge data indicates that sea levels have risen between 10 and 25 cm over the last century, with a significant component of this rise being attributed to increasing mean global temperatures. Most studies, conducted using global tide gauge data archived by the Permanent Service for Mean Sea Level (PSMSL), (Spencer and Woodworth (1993)), indicate that the current range of sea level rise estimates is between 1 and 3 mm per year with the mean trend approaching 2 mm per year. Some notable estimates are given by Peltier and Tushingham (1989), Hannah (1990), Douglas (1991) and Douglas (1997). Both Gorntiz (1995) and Douglas (1995) provide review papers which summarise global sea level rise estimates.

However, not all investigators consider that global sea level rise can be determined accurately using tide gauge records. The two most significant limitations, as pointed out by Gröger and Plag (1993), for example, are the predominantly northern hemisphere spatial distribution of tide gauges and the extremely small number of long term records (with particularly few in the southern hemisphere). Additionally, most global analysis have only used sites which have records that are of high data integrity, are not in close proximity to tectonically active plate boundaries and cover more than a 50 years period. For example, Douglas (1991) used 21 records (minimum 60 years, average 76) while Douglas (1997) was able to increase the number sites to 24 (minimum 60 year, average 83). At the very best, and even if the spatial distribution covers most oceanographic groups, this is a statistically small number of sites on which to base global estimates.

The biggest unknown component, apart from the record length and data integrity for which little can be done, is the vertical crustal movement at each site. The problem with any tide gauge record is that it only gives a relative change in sea level. That is, an apparent rise in sea level over time can be explained either by an actual increase in sea level or equally, a subsidence in the land on which the tide gauge is attached. In reality, the observed sea level record is most likely to be a combination of both effects and in practice it is difficult to decouple these two movements. The most probable causes of vertical land movement, as suggested by Warrick and Ahmad (1996), are isostatic adjustment caused by Post Glacial Rebound (PGR), tectonic effects at plate boundaries, sedimentation and man induced factors such as ground water extraction and oil or gas extraction. PGR is the most common cause of regional subsidence or uplift with notable examples in Scandinavia and North America, particularly the East Coast of America. Thus, an attempt to determine global sea level rise must take PGR into account, the most used model for doing so being the ICE-3G of Tushingham and Peltier (1991).

However, PGR models cannot take into account any other vertical crustal movements. Thus it has been apparent for some time that to realise the full potential of the PSMSL sea level data, more attention should be made to accurately measuring the vertical component of any regional or local movement, (Carter (1993). Although the space based techniques to measure vertical movements such as VLBI or SLR have been operational for 10-20 years, the systems are large, not easily transportable and are expensive to operate. This has restricted such applications to a few geodetic observatories. However, in the last five years, the emergence of GPS as a ubiquitous and relatively cheap positioning tool, has provided the potential to make space-based measurements at individual tide gauge sites. As recommended by the combined IGS and PSMSL meeting on monitoring sea level change, (Neilan and Woodworth (1998)), this short coming can be over come by providing a height time series and vertical velocity estimates in terms of the global International Terrestrial Reference System (ITRS).

Recent Experiments to Determine Vertical Crustal Movement

There are currently only a few sites where vertical land movement is being measured for the purpose of decoupling the crustal and sea level rise signals. Most analyses rely on sites being located at tectonically stable sites where the local subsidence is to be assumed zero and any PGR is computed using a model.

One of the first examples of this type of measurement was using the Swedish permanent GPS network (SWEPOS) which has been operational since 1993, (Johansson et al., 1994). The 16 station network with inter-station distances of 200-2000 km is used to contribute towards studies of the Fennoscandian post glacial uplift where rates as high as 12 mm per year have been observed.

The European Commission has funded a long term project to measure mean sea level variations at sixteen tide gauge sites along the Atlantic Coast of Europe, (Ashkenazi et al., 1994). Initially two measurement campaigns, (November 1993 and March 1994), were been observed and processed. This provides first epoch measurements to demonstrate zero vertical land movement and further observation campaigns will be carried out in the future.

A large collaborative project between European Countries is being carried out in the Mediterranean and Black Sea region to determine variations in sea level between well established tide gauges, (Zerbini et al., 1996). In addition to GPS measurements, many of the sites have included, or are close to, SLR and VLBI sites as well as Water Vapour Radiometer (WVR) and absolute gravity measurements. The relative sea level trends determined from records longer than 30 years are less than 1.5 mm per year. However, the crustal movements so far determined at the tide gauge sites, are small compared to the decadal sea level variability, although geological evidence suggests that there are significant temporal and spatial variations.

Recent work by Nerem et al. (1998) has been conducted on the East Coast of the Untied States, specifically around Chesapeake Bay. This area includes a particularly fragile wetlands ecosystem being lost to the rising oceans. The sea level trend in this area is approximately 3.5 mm per year, about twice the global average, for which PGR accounts for 1 mm of this effect. The small network has been established over the last four years and is only now just beginning to produce results.

The New Zealand Situation

Sea level data are used in New Zealand for a wide variety of purposes, some of which require only a short time series of data, and others of which require long term reliable historical records. The key uses for long term, reliable, sea level data may be summarised as follows.

1. To confirm the impacts of global climate change on New Zealand.

2. As a basis for the definition of New Zealand's national heighting systems.

3. As a basis upon which to design a host of land development and engineering projects, storm water reticulation systems being an obvious case in point.

4. As a basis upon which to anticipate changes in coastal cadastral and/or jurisdictional boundaries.

5. To support a range of scientific studies such as crustal deformation, changes in ocean circulation patterns and understanding the El Nino/Southern Oscillation (ENSO) phenomena.

For many years there has been no one agency responsible for overall co-ordination of the collection, verification and storage of sea level data for New Zealand. While, in the first half of this century, the various port authorities traditionally collected the data and, with some prompting, passed it on to the Department of Lands and Survey for analysis, interest quickly waned once the initial analyses had been performed. At that time the data was used to help define vertical datums and to assist in the development of models for tidal prediction.

In the 1970's, and with the various port authorities still collecting the tidal data, the RNZN Hydrographic Office began to undertake a role in its microfilming and storage. They subsequently extended this role to include some of the tidal predictions used in the Nautical Almanac. The Department of Lands and Survey continued to maintain an involvement in these processes, albeit at a rather nominal level. This situation continued until the late 1980's at which time the RNZN assumed this role in full.

In 1987, the Departments of Survey and Land Information and Conservation jointly funded an attempt to collect, verify and digitise the hard copy sea level records from the Ports of Auckland, Wellington, Lyttleton and Dunedin with the aim of ascertaining the extent to which sea levels around the coasts of New Zealand had changed over a period of 90 years. On average it was found that since the start of the 20th Century, sea level had risen by 1.7 mm per year, (Hannah, 1990) - a figure which has since been shown to be almost identical to the best global estimates for sea level rise. It was on the basis of these data that the long term sea level change predictions were prepared for the New Zealand Climate Change Committee.

An assessment of the present unsatisfactory data collection and administration systems, including a historical view, present and potential future developments, has been made by Hannah (1998) in a report commissioned by the National Science Strategy (NSS) Committee for Climate Change. The key recommendations include the adoption of appropriate government policy and the provision for administrative and financial commitments from New Zealand. This is required for the ongoing operation of specific tide gauges in order for New Zealand to meets its international obligations towards climate change programmes.

Tide gauges in New Zealand

The New Zealand tide gauges are operated by a number of different agencies, each of which have different operational requirements and interests in the collected data. Currently the Ports Authorities operate gauges at a total of sixteen Standard Ports and send the data to the Hydrographic Office for collation and archiving in the tidal database. These ports, shown in Figure 1, are Auckland, Bluff, Dunedin, Gisborne, Lyttleton, Marsden Point, Napier, Nelson, Onehunga, Picton, Port Taranaki, Timaru, Wellington, Westport, and Whangarei, (Blick et al., 1997).

Most of the gauge data from the Standard Ports is of limited use for scientific purposes such ocean circulation studies, calibration of satellite altimeter missions, tsunami warnings and storm surges. These limitations are caused by a combination of the gauges being poor quality or poorly maintained, and not being sited in open sea locations and therefore being subject to the effects caused by the local port facilities and adjacent river flows. The longest New Zealand records, which date back to the turn of the century, have been recorded at the ports of Auckland, Wellington, Lyttelton and Dunedin.

Over the last four years the National Institute for Water and Atmosphere (NIWA) has funded a programme to install open-coast sea level recorders at sites to give coverage around New Zealand's coastline. At present, these have been established at Sumner Head, Kaikoura , Jackson Bay, Dog Island, Charleston, Kapiti Island and Riversdale. Additional sites are planned for Mokohinau Island, and Anawhata.

Offshore from mainland New Zealand, a gauge is operated by the Civil Defence on the Chatham Island, and at Cape Roberts (Ross Dependency) by Victoria University.

Future Scientific Plans in New Zealand

In order to make use of sea level data, and especially for regions with a history of seismic and tectonic instability, it will be necessary to determine the rate of land uplift or land subsidence. Such is the case in New Zealand. A collaborative project, with the Institute of Geological and Nuclear Sciences, has recently been funded by the Foundation for Research, Science and Technology (FoRST) to establish four continuously operated GPS receivers at the four tide gauge sites with the longest records, namely Auckland, Wellington, Lyttleton and Dunedin. The project has been (initially) funded for a total of five years beginning in July 1999. These receivers will be run in conjunction with the current permanent IGS stations located at Auckland and the Chatham Island as well as two other permanent receivers located at Wellington and Dunedin. Other permanent stations are planned and should come online over the next few years.

Each receiver will be located as close as possible to the structure on which each tide gauge is mounted. However, since gauges are historically located on wharf frontages, due consideration must be taken of the environmental surrounds. One gauge, (Auckland), is located on a busy wharf with a considerable amount of shipping and vehicular traffic close by, while the Wellington gauge is inside a large corrugated iron building.

The New Zealand network will be integrated into the International Terrestrial Reference Frame (ITRF) based on the International Terrestrial Reference System (ITRS). This will be achieved by using a selection of close IGS stations such as Dededo (GUAM), Kwajalein Atol (KWJ1), Tidbinbilla (TIDB), Hobart (HOB2), MacQuarie Island (MAC1), Casey (CAS1), Ross Island (MCM4), Papeete (TAHI), Easter Island (EISL). Other permanent stations are included in the network such as OUSD, (Dunedin, New Zealand), and any new stations that are established in the future.

Current investigations are being carried out to establish processing procedures to improve the reliability of the vertical component of each station, particularly for the receivers at the tide gauge site. Data between October 1997 and February 1998 has been processed as part of the above regional network using the Bernese Post-processing software (Version 4.0). An example of the repeatability in the vertical component which is being achieved in given in Figure 2. The range in vertical component from this short period of data, is 66 mm with an rms of ñ12 mm. Further processing will include incorporating ground based meteorological observations at some of the sites and using IGS troposphere and ionosphere products.

Summary

It is now apparent that good, long term sea level data is an essential component in the development of models which both predict long term climate change and its impacts. It is also clear that it is now regarded as a key factor in the development of long term coastal planning policy as well as the definition of cadastral and international boundaries and the more traditional use of tide gauge data in the definition of vertical reference systems. However, in many places around the globe, including New Zealand, sea level records are contaminated by localised vertical movements at individual tide gauges, and it has only been in recent years, with the advent of GPS, that it has become both possible and realistic to accurately measure this movement.

New Zealand started the 20th Century with reliable and well maintained tide gauges. In more recent years the long term operation and maintenance of many of these tide gauges has been compromised due to factors such as privatisation of port authorities, government re-organisation and reduced funding. Only now are these deficiencies showing some signs of being rectified. Government departments are now recognising the importance of maintaining these long term records both for national planning decisions, but also the importance of contributing to international scientific programmes.

A medium term project to measure the vertical land movement at four key tide gauges has been funded by New Zealand's Foundation for Research, Science and Technology. It will be some time before results of this project will become available, but it is hoped that a reliable height time series will be produced.

References

Ashkenazi, V., R. M. Bingley, C. C. Chang, A. H. Dodson, J. A. Torres, C. Boucher, H. Fagard, J. L. Caturla, R. Quiros, J. Capdevila, C. Calvert, T. F. Baker, A. Rius and P. A. Cross (1994). EUROGAUGE: The West European Tide Gauge Monitoring Project. International Symposium on Marine Positioning (INSMAP94), 224-234.

Blick, G., D. Mole, M. Pearse and B. Wallen (1997). Land Information New Zealand's Role in and Needs for Sea Level Data. Wellington, New Zealand, Land Information New Zealand: 23.

Carter, W. E. (1993). Surrey Workshop of the IAPSO Tide Gauge Bench Mark Fixing Committee, Institute of Oceanographic Sciences, Deacon Laboratory, Surrey, UK: 75.

Douglas, B. C. (1991). "Global Sea Level Rise." Journal of Geophysical Research 96(C4): 6981-6992.

Douglas, B. C. (1995). "Global Sea Level Change: Determination and Interpretation." Reviews of Geophysics (Supplement): 1425-1432.

Douglas, B. C. (1997). "Global Sea Rise: A Redetermination." Surveys in Geophysics 18: 279-292.

Gorntiz, V. (1995). "Sea Level Rise: A Review of Recent, Past and Near Future Trends." Earth Surface Processes and Landforms 20: 7-20.

Gröger, M. and H.-P. Plag (1993). "Estimations of a Global Sea Level Trend: Limitations from the Structure of the PSMSL Global Sea Level Data Set." Global and Planetary Change 8: 161-171.

Hannah, J. (1990). "Analysis of mean sea level data from New Zealand for the period 1899-1988." Journal of Geophysical Research 95(B8): 12399-12405.

Hannah, J. (1998). New Zealand Sea Level Monitoring. Dunedin, New Zealand, Department of Surveying, Otago University: 20.

Houghton, J. T., L. G. M. Filho, B. A. Callander, N. Harris, A. Kattenberg and K. Maskell (1995). Climate Change 1995: Contribution of Working Group I to the Second Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, UK, Cambridge University Press.

Houghton, J. T., G. J. Jenkins and J. J. Ephraums (1990). Climate Change: The IPCC (Intergovernmental Panel on Climate Change) Scientific Assessment. Cambridge, England, Cambridge University Press.

Johansson, J. M., R. T. K. Jaldehag, T. R. Carlsson, T. M. Carlsson, G. Elgered, P. O. J. Jarlemark, B. I. Nilsson, B. O. Rönnäng, H.-G. Scherneck, J. L. Davis, P. Elosegui and J. X. Mitrovica (1994). Fennoscandian Postglacial Land Uplift: Results from Seven Months of Continuous GPS Measurements. The Sixth General Assembly of WEGENER, St. Petersburg, Russia, Smithsonian Astrophysical Observatory, 153-166.

Neilan, R. and P. L. Woodworth (1998). Workshop on Methods for Monitoring Sea Level: GPS and Tide Gauge Benchmark Monitoring, GPS Altimeter Calibration., Workshop organised by the IGS and PSMSL, Jet Propulsion Laboratory.

Nerem, R. S., T. M. van Dam and M. S. Schenewerk (1998). Chesapeake Bay Subsidence Monitored as Wetlands Loss Continues. EOS Transactions. 79: 149-157.

Peltier, W. R. and A. M. Tushingham (1989). "Global Sea Level Rise and the Greenhouse Effect: Might they be Connected?" Science 244: 806-810.

Spencer, N. E. and P. L. Woodworth (1993). Data Holdings of the Permanent Service for Mean Sea Level. Bidston, Birkenhead, England, Permanent Service for Mean Sea Level.

Tushingham, A. M. and W. R. Peltier (1991). "ICE-3G: A New Global Model of late Pleistocene Deglaciation based upon Geophysical Predictions of Post Glacial Relative Sea Level Change." Journal of Geophysical Research 96: 4497-4523.

Warrick, R. A. and Q. K. Ahmad (1996). The Implications of Climate and Sea Level Change for Bangladesh. Dordrecht, Kluwer Academic Publishers.

Zerbini, S., H.-P. Plag, T. Baker, M. Becker, H. Billiris, B. Bürki, H.-G. Kahle, I. Marson, L. Pezzoli, B. Richter, C. Romagnoli, M. Sztobryn, P. Tomasi, M. Tsimplis, G. Veis and G. Verrone (1996). "Sea Level in the Mediterranean: A First Step Toward Separating Crustal Movements and Absolute Sea-level Variations." Global and Planetary Change 14: 1-48.