Journal of the Royal Society of Western Australia, 90(3), September 2007 
geodetic surveyors in Western Australia are well-served 
by AUSGeoid98. 
The outliers, acknowledging the simplicity of the 
three-sigma test, should not be treated blindly. It is 
conceivable that the curvature and torsion of the 
plumbline is larger than anticipated. For instance, in 
areas of complex geology or high elevation, it is 
conceivable that mass-density contrasts will cause large 
curvature and torsion in the plumbline or steep gravity 
field gradients that are not be modelled by AUSGeoid98. 
A key example is in the proximity of the Darling Fault, 
where low-density sediments juxtapose the high-density 
Yilgarn Craton. 
Kearsley (1976) shows that gravimetric vertical 
deflections, computed using Vening-Meinesz's integral, 
are highly sensitive to nearby gravity anomalies, which 
themselves are greatly affected by the topography. 
4 1 
In order to test this and to ascertain any effect of using 
quasigcoid, rather than geoid, gradients to approximate 
Pizetti vertical deviations, the north-south and east-west 
deviation differences (astronomic minus gravimetric) are 
plotted as a function of Australian Height Datum (AHD) 
height of the astrogeodetic stations (Figures 6 and 7). 
Unweighted linear regression, coupled with the 
correlation coefficient (R-squared statistic), show that the 
differences are uncorrelated with AHD height. This 
observation justifies the earlier assumptions that the 
curvature and torsion of the plumbline are negligible in 
Western Australia and the use of the quasigeoid as 
opposed to geoid to compute Pizetti vertical deflections 
is acceptable, certainly in relation to the expected one 
arc-second precision of the astrogeodetic vertical 
deflections. It also shows that using quasigeoid gradients, 
as opposed to Vening-Meinesz's integral, is less sensitive 
to the topography (c/. Kearsley 1976). 
o 
0 ) 
03 
I 
(D 
o 
C 
<D 
0 ) 
C£ 2- 
0 
-1 
c 
o 
*4= .o 
> 
a> 
"O 
-3 - 
LJJ -4 - 
y = 0.0002X - 0.3008 
r2 = 0.0026 
0 100 200 300 400 500 600 700 800 900 1000 1100 
AHD height (metres) 
Figure 7. Differences between astronomic-Helmert deviations and AUSGeoid98-derived deviations in the east-west component as a 
function of AHD height (420 points). 
AHD height (metres) 
Figure 8. Differences between astronomic-Helmert deviations and AUSGeoid98-derived deviations in the north-south component as a 
function of AHD height (420 points). 
148 
