The longline CPUE series for the three outer islands are GLMM standardised. Year, month, area, trap-type, soak time, depth and year-area interactions are treated as fixed effects, and year-month interactions treated as a random effect. For Tristan, for which the available powerboat data are more limited, a GLM with year and month as fixed effects is applied. After initial increase, the standardised CPUE indices show drops over the most recent years for all islands except Gough, for which there is a steep recent increase.
Reference:
Johnston, S. J., Brandao, A., & Butterworth, D. S. (2010). GLMM-and GLM-standardised lobster CPUE from the Tristan da Cunha group of islands for the 1997-2008 period. MARAM/Tristan/2010/May/04. CPUE series for Nightingale Island. Both series have been renormalised to a mean of, 1.
Johnston, S. J., Brandão, A., & Butterworth, D. S. (2009). GLMM- and GLM-standardised lobster CPUE from the Tristan da Cunha group of islands for the 1997-2007 period University of Cape Town ,Faculty of Science ,Marine Resource Assessment and Management Group. Retrieved from http://hdl.handle.net/11427/18533
Johnston, Susan J, Anabela Brandão, and Doug S Butterworth GLMM- and GLM-standardised lobster CPUE from the Tristan da Cunha group of islands for the 1997-2007 period. University of Cape Town ,Faculty of Science ,Marine Resource Assessment and Management Group, 2009. http://hdl.handle.net/11427/18533
Johnston SJ, Brandão A, Butterworth DS. GLMM- and GLM-standardised lobster CPUE from the Tristan da Cunha group of islands for the 1997-2007 period. 2009 [cited yyyy month dd]. Available from: http://hdl.handle.net/11427/18533