Combining fisheries surveys to inform marine species distribution modelling

Meadhbh Moriarty, Debbi Pedreschi, Suresh Sethi, Bradley Harris, Simon Greenstreet, Nathan Wolf, Scott Smeltz, C McGonigle

Research output: Contribution to journalArticle

Abstract

Ecosystem-scale examination of fish communities typically involves creating spatio-temporally explicit relative abundance distribution maps using data from multiple fishery-independent surveys. However, sampling performance varies by vessel and sampling gear, which may influence estimated species distribution patterns. Using GAMMs, the effect of different gear–vessel combinations on relative abundance estimates at length was investigated using European fisheries-independent groundfish survey data. We constructed a modelling framework for evaluating relative efficiency of multiple gear–vessel combinations. 19 northeast Atlantic surveys for 254 species-length combinations were examined. Space-time variables explained most of the variation in catches for 181/254 species-length cases, indicating that for many species, models successfully characterized distribution patterns when combining data from disparate surveys. Variables controlling for gear efficiency explained substantial variation in catches for 127/254 species-length data sets. Models that fail to control for gear efficiencies across surveys can mask changes in the spatial distribution of species. Estimated relative differences in catch efficiencies grouped strongly by gear type, but did not exhibit a clear pattern across species’ functional forms, suggesting difficulty in predicting the potential impact of gear efficiency differences when combining survey data to assess species’ distributions and highlighting the importance of modelling approaches that can control for gear differences.
LanguageEnglish
Article numberfsz254
Number of pages14
JournalICES Journal of Marine Science
DOIs
Publication statusPublished - 20 Jan 2020

Fingerprint

fishery survey
biogeography
fisheries
modeling
vessel
relative abundance
distribution
sampling
fishery
spatial distribution
ecosystems
ecosystem

Keywords

  • Catchability
  • gear efficiency
  • fisheries independent assessment
  • Generalised Additive Mixed 49 Model (GAMM)
  • survey standardisation
  • species distribution modelling

Cite this

Moriarty, Meadhbh ; Pedreschi, Debbi ; Sethi, Suresh ; Harris, Bradley ; Greenstreet, Simon ; Wolf, Nathan ; Smeltz, Scott ; McGonigle, C. / Combining fisheries surveys to inform marine species distribution modelling. In: ICES Journal of Marine Science. 2020.
@article{2d60e170a9194c81a3f1d46f3199dd2b,
title = "Combining fisheries surveys to inform marine species distribution modelling",
abstract = "Ecosystem-scale examination of fish communities typically involves creating spatio-temporally explicit relative abundance distribution maps using data from multiple fishery-independent surveys. However, sampling performance varies by vessel and sampling gear, which may influence estimated species distribution patterns. Using GAMMs, the effect of different gear–vessel combinations on relative abundance estimates at length was investigated using European fisheries-independent groundfish survey data. We constructed a modelling framework for evaluating relative efficiency of multiple gear–vessel combinations. 19 northeast Atlantic surveys for 254 species-length combinations were examined. Space-time variables explained most of the variation in catches for 181/254 species-length cases, indicating that for many species, models successfully characterized distribution patterns when combining data from disparate surveys. Variables controlling for gear efficiency explained substantial variation in catches for 127/254 species-length data sets. Models that fail to control for gear efficiencies across surveys can mask changes in the spatial distribution of species. Estimated relative differences in catch efficiencies grouped strongly by gear type, but did not exhibit a clear pattern across species’ functional forms, suggesting difficulty in predicting the potential impact of gear efficiency differences when combining survey data to assess species’ distributions and highlighting the importance of modelling approaches that can control for gear differences.",
keywords = "Catchability, gear efficiency, fisheries independent assessment, Generalised Additive Mixed 49 Model (GAMM), survey standardisation, species distribution modelling",
author = "Meadhbh Moriarty and Debbi Pedreschi and Suresh Sethi and Bradley Harris and Simon Greenstreet and Nathan Wolf and Scott Smeltz and C McGonigle",
year = "2020",
month = "1",
day = "20",
doi = "10.1093/icesjms/fsz254",
language = "English",
journal = "ICES Journal of Marine Science",
issn = "1054-3139",

}

Combining fisheries surveys to inform marine species distribution modelling. / Moriarty, Meadhbh; Pedreschi, Debbi; Sethi, Suresh; Harris, Bradley; Greenstreet, Simon ; Wolf, Nathan; Smeltz, Scott; McGonigle, C.

In: ICES Journal of Marine Science, 20.01.2020.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Combining fisheries surveys to inform marine species distribution modelling

AU - Moriarty, Meadhbh

AU - Pedreschi, Debbi

AU - Sethi, Suresh

AU - Harris, Bradley

AU - Greenstreet, Simon

AU - Wolf, Nathan

AU - Smeltz, Scott

AU - McGonigle, C

PY - 2020/1/20

Y1 - 2020/1/20

N2 - Ecosystem-scale examination of fish communities typically involves creating spatio-temporally explicit relative abundance distribution maps using data from multiple fishery-independent surveys. However, sampling performance varies by vessel and sampling gear, which may influence estimated species distribution patterns. Using GAMMs, the effect of different gear–vessel combinations on relative abundance estimates at length was investigated using European fisheries-independent groundfish survey data. We constructed a modelling framework for evaluating relative efficiency of multiple gear–vessel combinations. 19 northeast Atlantic surveys for 254 species-length combinations were examined. Space-time variables explained most of the variation in catches for 181/254 species-length cases, indicating that for many species, models successfully characterized distribution patterns when combining data from disparate surveys. Variables controlling for gear efficiency explained substantial variation in catches for 127/254 species-length data sets. Models that fail to control for gear efficiencies across surveys can mask changes in the spatial distribution of species. Estimated relative differences in catch efficiencies grouped strongly by gear type, but did not exhibit a clear pattern across species’ functional forms, suggesting difficulty in predicting the potential impact of gear efficiency differences when combining survey data to assess species’ distributions and highlighting the importance of modelling approaches that can control for gear differences.

AB - Ecosystem-scale examination of fish communities typically involves creating spatio-temporally explicit relative abundance distribution maps using data from multiple fishery-independent surveys. However, sampling performance varies by vessel and sampling gear, which may influence estimated species distribution patterns. Using GAMMs, the effect of different gear–vessel combinations on relative abundance estimates at length was investigated using European fisheries-independent groundfish survey data. We constructed a modelling framework for evaluating relative efficiency of multiple gear–vessel combinations. 19 northeast Atlantic surveys for 254 species-length combinations were examined. Space-time variables explained most of the variation in catches for 181/254 species-length cases, indicating that for many species, models successfully characterized distribution patterns when combining data from disparate surveys. Variables controlling for gear efficiency explained substantial variation in catches for 127/254 species-length data sets. Models that fail to control for gear efficiencies across surveys can mask changes in the spatial distribution of species. Estimated relative differences in catch efficiencies grouped strongly by gear type, but did not exhibit a clear pattern across species’ functional forms, suggesting difficulty in predicting the potential impact of gear efficiency differences when combining survey data to assess species’ distributions and highlighting the importance of modelling approaches that can control for gear differences.

KW - Catchability

KW - gear efficiency

KW - fisheries independent assessment

KW - Generalised Additive Mixed 49 Model (GAMM)

KW - survey standardisation

KW - species distribution modelling

U2 - 10.1093/icesjms/fsz254

DO - 10.1093/icesjms/fsz254

M3 - Article

JO - ICES Journal of Marine Science

T2 - ICES Journal of Marine Science

JF - ICES Journal of Marine Science

SN - 1054-3139

M1 - fsz254

ER -