Gw. Smith et al., A COMPARISON OF THEORETICAL ESTIMATES OF THE ERRORS ASSOCIATED WITH ULTRASONIC TRACKING USING A FIXED HYDROPHONE ARRAY AND FIELD-MEASUREMENTS, Hydrobiologia, 372, 1998, pp. 9-17
Consideration of position fixing errors associated with fixed array ac
oustic tracking techniques has generally concentrated upon the consequ
ences of inaccuracies in the timing of the arrival of the ultrasonic p
ulse at each hydrophone, the overall accuracy of the system being dete
rmined by combining timing errors with information on the position of
the transmitter in the array. This paper presents a mathematical treat
ment which develops previous studies, allowing the distribution of pos
ition errors to be predicted from this same information and also makin
g it possible to quantify multi-path effects and the resulting errors
in position fixing. These theoretical considerations are compared with
the distributions of 1000 position fixes generated from each of five
stationary transmitters using a fixed hydrophone array. In each case t
he transmitters were on the sea-bed within the boundary of the hydroph
one array. Four of them were monitored in what would generally be cons
idered unfavourable tracking conditions on an underwater reef characte
rised by deep crevices and a dense covering of Laminaria sp. Although
the total area encompassing all estimated positions obtained during th
e field measurements varied considerably between tags, all exceeded 15
0 m(2). The position errors of the four tags associated with the reef
were large compared to theoretical estimates based upon expected error
s in time measurement. It is concluded that multi-path effects and ran
dom noise can make a substantial contribution to the errors associated
with fixing the positions of tagged individuals, particularly when tr
acking conditions are difficult and the signal level reaching any give
n hydrophone is low relative to the background noise. Thus timing erro
rs and the general position within the hydrophone array are only contr
ibutory factors in the determination of the total error, which may be
more strongly influenced by the micro topography in the vicinity of th
e tag.