User menu
IOTC CPUE workshop (2013) recommended that when environmental covariates is incorporated in CPUE standardization, it should be conducted in sub-area where ecological process producing good habitat areas and the variability pattern of the environmental signature is well identified. We attempted to implement this recommendation using one case study with Indian Ocean yellowfin tuna incorporating oceanographic variables into CPUE standardization using HSI (Habitat Suitability Index). We used four oceanographic variables affecting YFT habitat, i.e., thermocline depth and depth-specific sea temperature, salinity and vertical shear currents.
Then SI (Suitable Index) was estimated for each oceanographic variable through spatial correlations with YFT CPUE. SI is % frequency distribution representing the most suitable sea temperature range for YFT (for example) as 1.0 then for other ranges, proportional scales (0 to less than 1) are assigned. Then, HSI integrated four SI using geometric means and was represented as one scale from 0 (worst habitat)-1 (best habitat). As SI is based on CPUE, we cannot use it in GLM due to violation of assumption of GLM (CPUE will be in both sides in GLM). Thus, we changed to use operation-based SI as the proxy CPUE-based SI.
Then we attempted CPUE standardization in sub-area (higher HSI score areas instead of the whole area) as recommended by the CPUE workshop. Then effectiveness of HSI was tested by GLM with and without HSI in that sub-area. It was resulted that HSI effect was the highest significant term when it was incorporated. As this is very preliminary study with only one case study, we cannot make any general conclusion. We need to explore more case studies using different areas, species and fleet to provide reliable conclusions in the future. In addition, it is the critical point that we need to verify if operation-based SI is the proxy of CPUE-based SI. Otherwise, we cannot use this approach.
