Projected changes in habitat suitability for 33 marine species on the Northeast US shelf based on species distribution models fit to bottom trawl survey data from the NOAA Northeast Fisheries Science Center

Website: https://www.bco-dmo.org/dataset/765386
Data Type: model results
Version: 1
Version Date: 2019-04-22

Project
» Adaptations of fish and fishing communities to rapid climate change (CC Fishery Adaptations)
ContributorsAffiliationRole
Pinsky, MalinRutgers UniversityPrincipal Investigator
Rogers, LaurenStanford UniversityPrincipal Investigator, Contact
Griffin, RobertStanford UniversityCo-Principal Investigator
St. Martin, KevinRutgers UniversityCo-Principal Investigator
Fuller, EmmaPrinceton UniversityScientist
Young, TaliaRutgers UniversityScientist
Rauch, ShannonWoods Hole Oceanographic Institution (WHOI BCO-DMO)BCO-DMO Data Manager

Abstract
Projected changes in habitat suitability for 33 marine species on the Northeast US shelf. Changes in habitat suitability are calculated based on species distribution models fit to bottom trawl survey data from the NOAA Northeast Fisheries Science Center. Positive values indicate an increase in habitat suitability by 2040-2050 relative to historical (1963-2005). The spatial resolution of projections is 0.25 x 0.25 degrees.


Coverage

Spatial Extent: N:44.875 E:-64.875 S:33.625 W:-76.875
Temporal Extent: 1963 - 2050

Dataset Description

Projected changes in habitat suitability for 33 marine species on the Northeast US shelf. Changes in habitat suitability are calculated based on species distribution models fit to bottom trawl survey data from the NOAA Northeast Fisheries Science Center. Positive values indicate an increase in habitat suitability by 2040-2050 relative to historical (1963-2005). The spatial resolution of projections is 0.25 x 0.25 degrees.


Methods & Sampling

The following methods are excerpted from Rogers et al. (in press):
Bottom trawl data from the NOAA Northeast Fisheries Science Center (NEFSC) fall (1963-2014) surveys were used to characterize the realized thermal niches of species. At each survey station, fish of each species were counted and weighed, and surface and bottom temperature measurements were taken. Correction factors were applied to standardize catch rates for changes in vessel and gear type. A total of 33 species were selected based on their near continuous presence in the survey as well as relative importance to commercial fisheries. For 4 species, data from 1972 onwards were used because observations were irregular prior to that year.

Generalized Additive Models were used to estimate the realized thermal niches of species. We restricted k (number of knots) to 4 or 6 for each of our covariates to ensure biologically meaningful responses. Our response variable was probability of occurrence in a trawl haul, and we used a binomial response with logit transform:

p(occurᵧ,ⱼ) ~ logit-1 (s(STᵧ,ⱼ)+s(BTᵧ,ⱼ)+s(meanbiomassᵧ)+s(rugosityⱼ))

where STᵧ,ⱼ and BTᵧ,ⱼ are sea surface temperature and bottom temperature measured at each haul location j in year y, and meanbiomassᵧ is the average annual catch across all hauls to account for interannual changes in abundance due to, e.g., fishing. Rugosityᵧ is a measure of benthic habitat roughness, measured as the Terrain Ruggedness Index, using the GEBCO 2014 30-arcsecond bathymetry data (downloaded 4 Feb 2015 from http://www.gebco.net/). The resulting estimated smooth functions describing the relationship between probability of occurrence and temperature can be interpreted as realized thermal niches.

For each species, the change in predicted probability of occurrence under future (2040-2050) projected climate conditions was compared to historical (1963-2005) conditions for each cell within a 0.25°x0.25° spatial grid. Because the modeled probability of occurrence included a component of catchability, values for each species were scaled by dividing by the maximum observed or predicted probability of occurrence across the study area. Positive values for a grid square indicated a projected increase in probability of occurrence, whereas negative values indicated a projected decrease in probability of occurrence.

See related dataset for NEFSC bottom trawl data: https://www.bco-dmo.org/dataset/753142 (doi: 10.1575/1912/bco-dmo.753142.1)


Data Processing Description

Data were processed using R version 3.4.4.

BCO-DMO Processing:
- replaced spaces with underscores in the column headers.


[ table of contents | back to top ]

Data Files

File
projected_changes_habitat.csv
(Comma Separated Values (.csv), 137.82 KB)
MD5:9e19330380f8ef8717c570ace36ef7c1
Primary data file for dataset ID 765386

[ table of contents | back to top ]

Related Publications

Pinsky, M., & Selden, R. (2019). Trawl survey data and species distribution model predictions for presence, absence and abundance. Biological and Chemical Oceanography Data Management Office. doi:10.1575/1912/bco-dmo.753142.1
General
Rogers, L. A., Griffin, R., Young, T., Fuller, E., St. Martin, K., & Pinsky, M. L. (2019). Shifting habitats expose fishing communities to risk under climate change. Nature Climate Change, 9(7), 512–516. https://doi.org/10.1038/s41558-019-0503-z
Results

[ table of contents | back to top ]

Parameters

ParameterDescriptionUnits
brosme_brosmeChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
centropristis_striataChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
gadus_morhuaChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
glyptocephalus_cynoglossusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
hippoglossoides_platessoidesChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
hippoglossus_hippoglossusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
leucoraja_ocellataChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
limanda_ferrugineaChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
lophius_americanusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
zoarces_americanusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
melanogrammus_aeglefinusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
merluccius_albidusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
merluccius_bilinearisChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
paralichthys_dentatusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
peprilus_triacanthusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
pollachius_virensChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
pseudopleuronectes_americanusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
scomber_scombrusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
scophthalmus_aquosusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
sebastes_fasciatusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
squalus_acanthiasChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
stenotomus_chrysopsChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
urophycis_chussChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
urophycis_tenuisChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
homarus_americanusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
placopecten_magellanicusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
loligo_pealeiiChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
clupea_harengusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
illex_illecebrosusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
menticirrhus_americanusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
pomatomus_saltatrixChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
cynoscion_regalisChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
micropogonias_undulatusChange in habitat suitability for species. Positive values for a grid square indicate a projected increase in probability of occurrence by 2040-2050 relative to historical (1963-2005), whereas negative values indicate a projected decrease in probability of occurrence. unitless
lonLongitude decimal degrees
latLatitude decimal degrees

[ table of contents | back to top ]

Project Information

Adaptations of fish and fishing communities to rapid climate change (CC Fishery Adaptations)

Coverage: Northeast US Continental Shelf Large Marine Ecosystem


Description from NSF award abstract:
Climate change presents a profound challenge to the sustainability of coastal systems. Most research has overlooked the important coupling between human responses to climate effects and the cumulative impacts of these responses on ecosystems. Fisheries are a prime example of this feedback: climate changes cause shifts in species distributions and abundances, and fisheries adapt to these shifts. However, changes in the location and intensity of fishing also have major ecosystem impacts. This project's goal is to understand how climate and fishing interact to affect the long-term sustainability of marine populations and the ecosystem services they support. In addition, the project will explore how to design fisheries management and other institutions that are robust to climate-driven shifts in species distributions. The project focuses on fisheries for summer flounder and hake on the northeast U.S. continental shelf, which target some of the most rapidly shifting species in North America. By focusing on factors affecting the adaptation of fish, fisheries, fishing communities, and management institutions to the impacts of climate change, this project will have direct application to coastal sustainability. The project involves close collaboration with the National Oceanic and Atmospheric Administration, and researchers will conduct regular presentations for and maintain frequent dialogue with the Mid-Atlantic and New England Fisheries Management Councils in charge of the summer flounder and hake fisheries. To enhance undergraduate education, project participants will design a new online laboratory investigation to explore the impacts of climate change on fisheries, complete with visualization tools that allow students to explore inquiry-driven problems and that highlight the benefits of teaching with authentic data. This project is supported as part of the National Science Foundation's Coastal Science, Engineering, and Education for Sustainability program - Coastal SEES.

The project will address three questions:
1) How do the interacting impacts of fishing and climate change affect the persistence, abundance, and distribution of marine fishes?
2) How do fishers and fishing communities adapt to species range shifts and related changes in abundance? and
3) Which institutions create incentives that sustain or maximize the value of natural capital and comprehensive social wealth in the face of rapid climate change?

An interdisciplinary team of scientists will use dynamic range and statistical models with four decades of geo-referenced data on fisheries catch and fish biogeography to determine how fish populations are affected by the cumulative impacts of fishing, climate, and changing species interactions. The group will then use comprehensive information on changes in fisher behavior to understand how fishers respond to changes in species distribution and abundance. Interviews will explore the social, regulatory, and economic factors that shape these strategies. Finally, a bioeconomic model for summer flounder and hake fisheries will examine how spatial distribution of regulatory authority, social feedbacks within human communities, and uncertainty affect society's ability to maintain natural and social capital.



[ table of contents | back to top ]

Funding

Funding SourceAward
NSF Division of Ocean Sciences (NSF OCE)

[ table of contents | back to top ]