School of Mathematics and Statistics

Postgraduate research

Research being undertaken by our current postgraduates:

Contact

Ross Marriott

Phone: (+61 4) 2877 1407


Start date

Jan 2013

Submission date

Jan 2017

Ross Marriott

Ross Marriott profile photo

Thesis

Alternative Spatiotemporal Imputation Methods for Catch Rate Standardisation.

Summary

Analyses of historical trends in data reported by commercial fishers (e.g., catch landings and time spent fishing or “effort”) is a key feature of assessments of exploited fish populations. In these assessments, commercial catch-per-unit-of-effort (cpue) data are often used to develop an index of historical fish abundance. Commercial cpue data are typically affected by “fishery-dependent” factors such as skipper skill, targeted fishing behaviour and weather conditions. Accordingly, it is good practice to correct for the influence of fishery-dependent factors on cpue data through statistical standardisation methods involving application of generalized linear models.

However it is not clear what should be done when data are missing from a time series of cpue data. For instance, commercial fleets are highly mobile, and may discontinue fishing in an area when the fish abundance has been depleted to such low levels that fishing is no longer economically viable there. This can result in a biased index of fish abundance.

This study seeks to explore and evaluate the suitability of applying alternative spatiotemporal imputation methods to cpue standardisations in order to account for missing data. This will be done through simulation and cross-validation analyses of real datasets for two key finfish fisheries in Western Australia. The method producing the least biased and most precise index of fish abundance can then be put forward as the preferred method for such assessments.

Why my research is important

This research addresses an important aspect of catch rate standardisation analyses required for fisheries stock assessments. In the absence of other available information on the trend of fish abundance over time, these analyses make use of available logbook data for providing valuable indices reflecting historical changes in fish abundance. The resulting information is valuable for estimating the current fished status of a fish population and determining the sustainability of current fishing practice.

Funding

  • In-kind funding support from, and data provision by the Department of Fisheries, Western Australia.


Statistics clinic

Assistance in statistics is available for Postgraduates students by research at the UWA Centre for Applied Statistics.

 

School of Mathematics and Statistics

This Page

Last updated:
Monday, 16 June, 2014 2:41 PM

http://www.maths.uwa.edu.au/275877