2000•Fisheries OceanographyRequires access

Patterns in the spawning of cod (Gadus morhuaL.), sole (Solea soleaL.) and plaice (Pleuronectes platessaL.) in the Irish Sea as determined by generalized additive modelling

Fox, O'Brien O'Brien, Dickey‐Collas, Nash

Open publisher page 68 citations

Abstract

Eleven ichthyoplankton cruises were undertaken covering most of the Irish Sea during the period February to June, 1995. To identify spawning localities and investigate temporal trends in egg production, the data on stage 1 A egg distributions of cod (Gadus morhua), plaice (Pleuronectes platessa) and sole (Solea solea) have been modelled using generalized additive models (GAMs). A two‐stage approach was adopted where presence/absence was firstly modelled as a binary process and a GAM surface subsequently fitted to egg production (conditional on presence). We demonstrate that this approach can be used to model egg production both in space and in time. The spawning sites for cod, plaice and sole in the Irish Sea were defined in terms of the probability of egg occurrence. For cod, we demonstrate that by integrating under predicted egg production surfaces, a cumulative production curve can be generated and used to define percentiles of production and thus delimit the extent of the spawning season. However, for plaice and sole, the surveys did not fully cover the spawning season and the limitations that this imposes on GAM modelling of these data are discussed. Comparison of the spawning sites in 1995 with historical data suggests that the locations of cod, plaice and sole egg production in the Irish Sea have probably remained relatively constant over the last 30 years.

About this research paper

What this paper is about

Eleven ichthyoplankton cruises were undertaken covering most of the Irish Sea during the period February to June, 1995. To identify spawning localities and investigate temporal trends in egg production, the data on stage 1 A egg distributions of cod (Gadus morhua), plaice (Pleuronectes platessa) and sole (Solea solea) have been modelled using generalized additive models (GAMs). A two‐stage approach was adopted where presence/absence was firstly modelled as a binary process and a GAM surface subsequently fitted to egg production (conditional on presence). We demonstrate that this approach can be used to model egg production both in space and in time. The spawning sites for cod, plaice and sole in the Irish Sea were defined in terms of the probability of egg occurrence. For cod, we demonstrate that by integrating under predicted egg production surfaces, a cumulative production curve can be generated and used to define percentiles of production and thus delimit the extent of the spawning season. However, for plaice and sole, the surveys did not fully cover the spawning season and the limitations that this imposes on GAM modelling of these data are discussed. Comparison of the spawning sites in 1995 with historical data suggests that the locations of cod, plaice and sole egg production in the Irish Sea have probably remained relatively constant over the last 30 years.

Why it matters

OpenAlex reports 68 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Eleven ichthyoplankton cruises were undertaken covering most of the Irish Sea during the period February to June, 1995. To identify spawning localities and investigate temporal trends in egg production, the data on stage 1 A egg distributions of cod (Gadus morhua), plaice (Pleuronectes platessa) and sole (Solea solea) have been modelled using generalized additive models (GAMs). A two‐stage approach was adopted where presence/absence was firstly modelled as a binary process and a GAM surface subsequently fitted to egg production (conditional on presence). We demonstrate that this approach can be used to model egg production both in space and in time. The spawning sites for cod, plaice and sole in the Irish Sea were defined in terms of the probability of egg occurrence. For cod, we demonstrate that by integrating under predicted egg production surfaces, a cumulative production curve can be generated and used to define percentiles of production and thus delimit the extent of the spawning season. However, for plaice and sole, the surveys did not fully cover the spawning season and the limitations that this imposes on GAM modelling of these data are discussed. Comparison of the spawning sites in 1995 with historical data suggests that the locations of cod, plaice and sole egg production in the Irish Sea have probably remained relatively constant over the last 30 years.

Key concepts: Pleuronectes, Gadus, Fishery, Irish sea, Oceanography, Atlantic cod, Biology, Generalized additive model

Related papers

Back to paper searchBrowse research topicsOriginal source
Patterns in the spawning of cod (Gadus morhuaL.), sole (Solea soleaL.) and plaice (Pleuronectes platessaL.) in the Irish Sea as determined by generalized additive modelling — Research Paper | ScholarLens