Preview	PDF (Front matter) front-burch-the...pdf | Download (164kB) Available under University of Tasmania Standard License.
Preview	PDF (Whole thesis excluding Ch. 7) whole-burch-the...pdf | Download (1MB) Available under University of Tasmania Standard License.
	PDF (Whole thesis contains published material) whole-burch-the...pdf | Document not available for request/download Full text restricted

Abstract

A daily tag return model was developed to estimate shing and natural mortality,
tag reporting rate and catchability with application to lobster sheries. Tag return
data is usually collected with knowledge of the exact date of recapture. By modelling
tags individually, each tag contributes information on shing and natural mortality,
catchability and tag reporting rate to the likelihood. Providing sucient tags are
maintained in the shery to enable recaptures to occur during the shing season,
ner resolution of parameters is possible.
Model performance was tested by simulation of dierent times of release and recapture
as well as a range of dierent seasonal shing patterns typically found in
lobster sheries. Precision and accuracy of estimates were improved when there was
a contrast in shing eort throughout the season or a seasonal closure within the
year. The timing of tag release was not found to aect model performance.
Evaluation of the model was undertaken by comparison of estimates from a previous
study using an identical dataset. Total mortality estimates were equivalent between
models although separation of shing and natural mortality diered between models.
Small improvements in the precision of estimates were obtained for the model that
incorporated exact times of tag release and recapture.
The daily model enabled ner time scale estimates of parameters and this was
explored in the estimate of within season catchability. Penalised spline smoothing
was applied to estimate catchability, resulting in a curve that captured the timing
of biological events such as moulting and mating.
Higher shing mortality estimates were obtained when the shing
eet was separated
into groups based on their reliability at returning tags. This suggests that
shing mortality estimates could be compromised when reporting rate is considered
cosmopolitan across the entire
eet. However, the small number of reliable shers
and their shing patterns, make interpretation of results problematic.
A cost-benet analysis of the use of PIT tags compared to T-bar tags, the standard
tag used in lobster sheries, was undertaken. For a given cost the higher expenses
associated with PIT tags and scanners resulted in fewer lobster being tagged compared
to T-bar tags. The improved tag reporting rate from PIT tags resulted in
improved precision and accuracy of mortality estimates using this technology unless
the tag reporting rate for T-bar tags was substantially increased.

Item Type:	Thesis - PhD
Authors/Creators:	Burch, P
Additional Information:	Copyright the Author
Item Statistics:	View statistics for this item

Actions (login required)

Item Control Page