Latent Gaussian models to decide on spatial closures for bycatch management in the Barents Sea shrimp fishery

Abstract

14In the Barents Sea and adjacent water, fishing grounds are closed for shrimp

Quantifying multiple pressure interactions affecting populations of a recreationally and commercially important freshwater fish

“…In addition, R‐INLA contains methods to model Gaussian random fields (GRFs) and construct surfaces that are able to estimate parameters for datasets with complex spatial structures (Beguin, Martino, Rue, & Cumming, ; Bivand, Gomes‐Rubio, & Rue, ; Lindgren, Rue, & Lindström, ), that is, potential spatial dependence among adjacent lakes in our study. The flexibility and application of R‐INLA have recently been demonstrated in several nontechnical ecology papers that explore spatial and temporal characteristics of fisheries bycatch (Breivik, Storvik, & Nedreaas, ; Cosandey‐Godin, Krainski, Worm, & Flemming, ) and potential for conservation (Pennino et al, ).…”

Latent Gaussian models to predict historical bycatch in commercial fishery

“…The 61 survey data used in this research have previously been used by MSS to regulate the shrimp 62 shery (Breivik et al, 2016). See Little et al (2015) for a summary of management methods 63 with respect to bycatch in several other large sheries.…”

“…Bycatch was also predicted in Breivik et al (2016) for regulation purposes. Our research diers 65 mainly because we utilize huge amounts of shery data, resulting in new computational di-66 culties, and that the data distribution is changed from log-Gaussian to zero-inated negative 67 binomial.…”

“…Our research diers 65 mainly because we utilize huge amounts of shery data, resulting in new computational di-66 culties, and that the data distribution is changed from log-Gaussian to zero-inated negative 67 binomial. Furthermore, the target catch is in this research a given covariate since it is included 68 in both the shery data and the survey data, while in Breivik et al (2016) where future predic-69 tions was the focus, the shrimp catch was stochastic. To adapt to the information given in the 70 shery data, the response variable for bycatch in Breivik et al (2016) is changed from bycatch 71 per nautical mile to total bycatch, and with duration trawled included as an oset.…”

“…This selection of random structure is the same as 275 in Breivik et al (2016). See Table 4 a) The spatial eect.…”