Mark Merrifield & Mike Bevis
School of Ocean and Earth Sciences and Technology
University of Hawaii
markm@soest.hawaii.edu
bevis@soest.hawaii.edu
PROJECT OVERVIEW
Continuous GPS (CGPS) positioning of a select set of global tide gauges is being implemented by the University of Hawaii Sea Level Center (UHSLC) and the Pacific GPS Facility (PGF) in support of altimeter drift monitoring activities. These CGPS-tide gauge stations will constitute a significant portion of the altimeter calibration network proposed recently by Mitchum (1998b). The colocation of GPS receivers at the tide gauge stations will provide local estimates of absolute sea level. Emphasis is therefore placed on selecting tide gauges with relatively long records. All University of Hawaii CGPS stations are scheduled to be in place prior to the Jason launch.
THE TIDE GAUGE CALIBRATION NETWORK
The feasibility of using tide gauges to monitor altimeter drift was demonstrated recently by Mitchum (1998a) using a set of over 50 stations from around the world. The time dependent drift estimate from the altimeter-tide gauge differences successfully identified an artificial drift in the TOPEX altimeter caused by an algorithm error. One of the primary uncertainties remaining in this calibration method is the vertical drift of the individual tide gauges.
Tide gauges measure sea level relative to the land or structure on which the gauge rests (Figure 1). At many locations, tectonic movements or the settling of a pier can obscure trends in absolute sea level. In order to monitor low frequency changes in sea level, associated with for example global sea level rise or decadal climate variations, the vertical land movements at tide gauges must be taken into account. A recent workshop sponsored by the International GPS Service (IGS) and the Permanent Service for Mean Sea Level (PSMSL) has provided guidelines for this long-considered step (Bevis).
Mitchum (1998b) has proposed a network of globally distributed tide gauges with GPS correction for land movement that is designed to detect altimeter drift errors of 1 mm/year over a three year period. The 30 stations initially selected for the calibration network are pictured in Figure 2. Of these, over half have CGPS information available in the region. The University of Hawaii has begun a program to equip 7 of the remaining 14 stations with GPS receivers in support of the Jason mission. The best calibration sites are those where the altimeter and tide gauge data agree closely. For the 7 University of Hawaii CGPS-tide sites, the standard deviation of the differences between the tide gauges and TOPEX is less than 1 cm (Figure 3).
IMPLEMENTATION
The colocation of the GPS with the tide gauge is considered to be a desirable design criterion above the specifications proposed by Mitchum (1998b). We seek to place the tide gauge into the ITRF-96 coordinate system and to correct for vertical movements at the gauge itself. Ideally, the GPS and tide gauge should be attached to the same rigid support structure. Precise leveling surveys are still required to tie the GPS antenna to the tide staff and the system of tide gauge benchmarks to which historic sea level data typically are referenced. This is particularly important for stations with long histories.
Our first tide gauge/CGPS station was established at the Honolulu Harbor tide gauge during 1997. The Honolulu tide gauge is one of the longest continuous records in the Pacific Ocean dating back to 1905 (Figure 4). The linear sea level trend estimated from this record (1.8 cm/decade) differs considerably from other tide gauges along the Hawaiian Chain (e.g., Hilo) highlighting the need for colocated gauges in tectonically active regions.
The GPS antenna at the Honolulu Harbor station is positioned on a building made of reinforced concrete that sits on the same contiguous slab on which the acoustic sea level sensor is attached (Figure 5). Here the primary consideration was sky view for the GPS antenna. Although a uniform station configuration is desired, in practice each station will require different engineering solutions depending upon nearby obstructions to the GPS antenna, the structural integrity of the pier or sea wall, station security, etc.
ADDITIONAL BENEFITS
We will acquire temperature and barometric pressure measurements at each of the CGPS-tide gauge stations. Using this data in combination with the GPS, an estimate of the vertically integrated water vapor is obtainable (Bevis et al., 1992). Although we have not yet processed the data at the Honolulu station, Figure 6 demonstrates the type of information that GPS meteorology can provide.
REFERENCES
Bevis, M., S. Businger, T. A. Herring, C. Rocken, R. A. Anthes, and R. H. Ware, 1992: GPS meteorology: Remote sensing of atmospheric water vapor using the global positioning system. J. Geophys. Res, 97(D14), 15,787-15,801.
Mitchum, G. T., 1998b: Monitoring the stability of satellite altimeters with tide gauges, J. Atmos. and Oceanic Tech., 15, 721-730.
Mitchum, G. T., 1998a: A tide gauge network for altimeter calibration, in review.
