作者: Eric N. Aidoo , Ute Mueller , Pierre Goovaerts , Glenn A. Hyndes
DOI: 10.1016/J.FISHRES.2015.03.013
关键词:
摘要: Western Australians are heavily engaged in recreational fishing activities with a participation rate of approximately 30%. An accurate estimation the spatial distribution catch per unit effort (catch rates) is an integral component for monitoring fish population changes and to develop strategies ecosystem-based marine management. Geostatistical techniques such as kriging can provide useful tools characterising distributions rates. However, most fishery data highly skewed, zero-inflated when expressed ratios impacted by small number problem which influence estimates obtained from traditional kriging. The applicability ordinary, indicator Poisson was evaluated three aquatic species different behaviours patterns. prediction performance each estimator assessed based on cross-validation. For all species, accuracy plot (IK) showed better agreement between expected empirical proportions falling within probability intervals increasing size, measured goodness statistic. Also, found be predicting latent compared ordinary maps estimators displayed similar patterns but produced smoother distributions. We show that IK may preferable modelling exhibiting these characteristics, has best regardless life history those species.