Gorfine et al.: Two methods for estimating abundance of Haliotis rubra 
439 
divers, each represented in the data for relatively 
few years of a long time series, is problematic. Strip 
transects can be time consuming and may traverse 
nonabalone habitat, such as sand; however they pro- 
vide a consistent base sampling unit among research 
divers. Patch-size estimates have the advantage of 
providing a measure of the number and size of ag- 
gregations of abalone but are disadvantaged by dif- 
ficulties in selecting appropriate patch-size catego- 
ries and maintaining consistency in the area searched 
during surveys. Change-in-ratio (CIR) has the ad- 
vantage of not being dependent upon a particular 
base sampling unit, such as time or area. However, 
CIR requires that there is no change in one of the 
two animal categories in the ratio, in this instance 
prerecruits. This means that the time interval be- 
tween surveys must be sufficiently short that the 
effects of growth and natural mortality on the rela- 
tive abundance of prerecruit abalone can be consid- 
ered negligible. Mark-recapture (M-R) surveys are 
time consuming, require that many abalone be 
tagged, require an estimate of tag loss, may incur 
incidental mortality due to tagging, assume that the 
tagged abalone will disperse randomly within the 
untagged population, and can be affected by migra- 
tion to and from the study site and into and out of 
cryptic habitat. If these limitations can be overcome, 
then M-R can be an effective abundance estimation 
technique that is free from the research diver effects 
that characterize other methods. 
We introduce a new radial transect survey method 
for estimating the abundance of abalone and furnish 
the results of a preliminary comparison with timed 
searches previously used in Victoria. This new 
method was developed as an alternative to timed 
searches because 1) standardization of the effects of 
research divers who conducted surveys prior to 1992 
with those involved in surveys from 1992 onwards 
was not possible and 2) the depths at many of the 
reefs surveyed were sufficient to warrant limiting 
the number of ascents made by research divers each 
day in order to reduce the divers’ risk of decompres- 
sion illness. We present an analysis of investigations 
into the effect of research divers on abundance esti- 
mates and the probability of detecting interannual 
changes in abundance with both survey methods. In 
Victoria, trends in interannual abundance for each 
management zone have been used as part of an adap- 
tive management strategy. Several successive years 
of decrease in abundance will trigger consideration 
for decreasing annual quotas and similarly consid- 
eration will be give to increasing quotas where trends 
in abundance appear to be increasing. Whether quo- 
tas are changed and by how much will largely de- 
pend on the size of the change in abundance. It is 
Figure 1 
Location of study sites and management zones along the Victo- 
rian coast. CS = Cape Schanck; GI = Gabo Island; PC = Point 
Cook; WI = Williamstown (includes Sticks Reef); SP = Sandpatch 
Point; CZ = central zone; EZ = eastern zone; WZ = western zone. 
anticipated that this approach will lead to the 
achievement of quotas that will minimize inter- 
annual changes in abalone abundance and thereby 
optimize sustainable yields from the fishery. 
Materials and methods 
Survey methods 
Radial transects Estimates of abalone abundance 
with radial transects were performed by research 
divers. At a particular site, a buoyed shot-line (a 20- 
mm-diameter line anchored by a 20-kg lead weight) 
was deployed. A stainless steel ring and swivel were 
used to connect a 35-m long transect-line of 20-mm- 
diameter line marked at 5-, 15-, and 20-m distances 
to the lead weight. The transect lines were made from 
a form of nylon rope that tends to float to the sur- 
face, thus decreasing the likelihood of the lines be- 
coming tangled. Dives were made with a surface-sup- 
