The application of a generalized linear mixed model to the Area 8+ trapboat data

Working Paper


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

Two papers by Glazer and Butterworth (2011a, 2011b) were recently submitted to the West Coast Rock Lobster Scientific Working Group meeting for consideration. The first of those included sub-area as an explanatory variable in a Generalized Linear Mixed Model (GLMM) applied to standardize the Area 8 trapboat CPUE data. The second paper incorporated data from Area 8+ (rather than data from Area 8 only as used in the past) in the standard Generalized Linear Model (GLM) applied to those data. It was agreed at the meeting that the analyses presented in the aforementioned papers be amalgamated by conducting a GLMM on the Area 8+ data, which incorporates sub-area as an explanatory variable in the model. The results of such an analysis are presented in this paper. It should be noted that Area 8+ for trapboat fishing comprises data from Areas 8, 10 and 11. No trapboat fishing takes place in the area East of Hangklip (Areas 12-14) and a procedure has therefore been adopted to adjust the standardized index to allow for the movement of lobster into the East of Hangklip area over a period of time (1987-1995). The index obtained above is then extended back to 1985 by scaling the pre- 1992 indices from the model applied in the past to standardize the Area 8 CPUE data, so that they can be incorporated with the GLMM-based index.