Journal of the Royal Society of Western Australia, 87(1), March 2004 
extended recurrence periods for very large earthquakes 
in these zones (e.g. Weber et al. 1998), the inevitable 
extrapolation of data collected over the last few decades 
introduces a great deal of uncertainty in their analysis. 
Therefore, there is the need for independent, yet 
complementary, estimates of the rates of deformation in 
addition to those inferred from seismology. 
The most effective means of quantifying contemporary 
surface deformation at discrete points over large areas is 
through high-precision geodetic measurements. In a 
deforming region, the amount of deformation can be 
quantified using repeated measurements of position, 
angles, distances, or a combination of these among a 
network of stable ground points. Historically, the 
measurements were made using conventional terrestrial 
surveying techniques (i.e. triangulation by measurement 
of angles, trilateration by measurement of distances, or a 
combination of both). Now, measurements from the 
Global Positioning System (GPS) have taken over as the 
primary means with which to quantify regional 
deformation (e.g. Bevis et al. 1997; Bock et al. 1997; Clarke 
et al. 1998; Weber et al. 1998; Pan et al. 2001). 
Due to the small amount of expected deformation in 
the SWSZ (if indeed there is any surface expression of it) 
and the low precision of the terrestrial and ad-hoc GPS 
measurement methods then used, previous studies have 
proved inconclusive. Therefore, a more rigorous 
approach is required. Accordingly, a 48-ground-point 
GPS-geodetic deformation monitoring network has been 
established across the SWSZ (Fig 1), for which epoch-one 
GPS-geodetic measurements were made during May 
2002. This collaborative venture involves funding and 
scientists from GA (Minerals and Geohazards, and 
National Mapping Divisions), the Western Australian 
Department of Land Information (DLI), the New Zealand 
Institute of Geological and Nuclear Sciences (GNS), 
Curtin University of Technology, and the University of 
Western Australia. This paper summarises the scientific 
rationale for this joint venture, describes the permanent 
network of ground monuments, gives results of the May 
2002 GPS campaign and discusses the future work, 
including issues pertaining to the likely amount of 
surface deformation that could be detected using the GPS 
techniques described. 
Previous geodetic deformation estimation in the SWSZ 
There have been several attempts to geodetically 
measure surface deformation in various parts of 
Australia. Wellman (1981) presented an analysis and 
interpretation of repeat geodetic survey data for 
monitoring horizontal surface deformation throughout 
south-east and south-west Australia. In part of the SWSZ, 
Wellman (1981) utilised results from a combination of 
first-order (Anon 2002) triangulation observations and 
resurvey trilateration observations. Note that the repeat 
survey used a different geodetic surveying technique 
than the initial survey, as well as different 
instrumentation. 
Coleman & Lambeck (1983) seriously questioned the 
validity of Wellman's (1981) conclusions, arguing that the 
interpreted deformation is not significant because several 
critical factors were neglected. Clearly, this raises doubt 
as to the significance of any claimed crustal movement in 
the SWSZ, though admittedly Wellman (1981) states that 
'it is irregular in magnitude and direction' in the SWSZ, 
which concurs with the later analysis performed by 
Featherstone (1998). Clearly, it is uncertain whether 
Wellman's (1981) estimates of horizontal surface 
deformation in the SWSZ are real or are simply an 
artefact of measurement, reduction and adjustment 
errors. This uncertainty is compounded by the inclusion 
of the same data in the least squares adjusted coordinates 
between measurement epochs (Coleman & Lambeck 
1983). 
Soon after the 1968 Meckering earthquake, DLI 
initiated a programme of episodic repeat-geodetic 
measurements to monitor surface deformation around 
the Meckering region, until 1995. This involved nine 
horizontal geodetic monitoring cells over parts of the 
SWSZ, and conducting second-order geodetic levelling 
over a large proportion of the SWSZ, which was used by 
Wellman & Tracey (1987). Changes in funding and 
advances in geodetic measurement technology dictated 
that the horizontal monitoring used both terrestrial- 
geodetic and GPS-geodetic techniques, which can exhibit 
scale differences (cf Savage et al. 1996). Featherstone 
(1998) analysed the horizontal DLI data and argued that 
these investigations of the SWSZ were also inconclusive 
(cf Coleman & Lambeck, 1983). Based on comparisons of 
the measured distances (i.e. the primary observations), 
no statistically significant changes were detected in 
relation to the expected precision of the measurement 
techniques used. Moreover, there were contradictory 
estimates of extension and compression for the same 
baselines. Although one of DLLs monitoring cells (to the 
west of Meckering) did show some significant 
differences, these could be simply attributed to 
instrumentation differences (Featherstone 1998). 
In 2000, the first-named author reoccupied parts of this 
monitoring cell with Leica CRS1000 GPS receivers. 
Unfortunately, the GPS data collected were not of 
sufficient quality to resolve any motion. Nevertheless, the 
field survey was useful because it was discovered that 
several of the ground monuments were difficult to 
accurately centre over (e.g. 10 mm spikes set in concrete 
with no drill-hole; cf the monument to the right in Fig 2), 
and many of the monuments were not set on bedrock. At 
least one was demonstrably unstable, moving when 
kicked very gently. Therefore, the apparent statistically 
significant deformation observed in this area could 
simply be due to one or all of different GPS instruments, 
GPS-antenna centring errors over the ground marks, and 
disturbance of the marks between observation epochs. 
From all these previous studies, there is clearly no 
consensus as to whether any surface deformation has 
actually been detected in the SWSZ, or whether 
observation and data reduction errors have been 
misinterpreted as deformation (cf Coleman & Lambeck 
1983; Featherstone 1998). Accordingly, the consortium 
has taken a fresh approach to geodetic deformation 
monitoring in the SWSZ, as follows. 
The new 48-point SWSZ network 
The most effective means of quantifying contemporary 
surface deformation at discrete points over large areas by 
GPS is through a network of continuously operating 
geodetic GPS receivers. Such an approach has been 
adopted in tectonically active regions such as Southern 
2 
