Incorporating Survey Data to Improve Space–Time Geostatistical Analysis of King Prawn Catch Rate

Abstract

Commercial fishing logbook data from the Shark Bay managed prawn fishery in Western Australia provide king prawn catch rate data densely informed and irregularly spaced in both the spatial and temporal domains. Space–time geostatistical analysis for the data from the 2001 to 2004 fishing seasons has shown that short term catch rate prediction is possible with the use of the product-sum covariance model and the subsequent kriging estimation process. However the operation of closure lines within the fishery makes it difficult to capture the high catch rate behaviour in areas as they first open to trawling. One of these regions is the Extended Nursery Area which usually opens in the first week of May. Analysis of the survey trawls from seasons 2001 to 2003 in this region in March and April shows there is a moderate positive correlation between the actual catch rate and the survey catch rate. By using the survey catch rate data as additional data in space–time geostatistical estimation of the catch rates for May 2004, the space–time behaviour of the king prawn catch rate data is more successfully captured.