Building use‐inspired species distribution models: using multiple data types to examine and improve model performance
Building use‐inspired species distribution models: using multiple data types to examine and improve model performance
Species distribution models (SDMs) are becoming an important tool for marine conservation and management. Yet while there is an increasing diversity and volume of marine biodiversity data for training SDMs, little practical guidance is available on how to leverage distinct data types to build robust models. We explored the effect of different data types on the fit, performance and predictive ability of SDMs by comparing models trained with four data types for a heavily exploited pelagic fish, the blue shark (Prionace glauca), in the Northwest Atlantic: two fishery dependent (conventional mark-recapture tags, fisheries observer records) and two fishery independent (satellite-linked electronic tags, pop-up archival tags). We found that all four data types can result in robust models, but differences among spatial predictions highlighted the need to consider ecological realism in model selection and interpretation regardless of data type. Differences among models were primarily attributed to biases in how each data type, and the associated representation of absences, sampled the environment and summarized the resulting species distributions. Outputs from model ensembles and a model trained on all pooled data both proved effective for combining inferences across data types and provided more ecologically realistic predictions than individual models. Our results provide valuable guidance for practitioners developing SDMs. With increasing access to diverse data sources, future work should further develop truly integrative modeling approaches that can explicitly leverage the strengths of individual data types while statistically accounting for limitations, such as sampling biases.
climate change, ecological forecasting, highly migratory species, prediction, spatial ecology, species distribution models
Braun, Camrin D.
f53f1c67-16ce-4d16-996e-5b6ed7caae9d
Arostegui, Martin C.
b25b5e24-9e1a-4ff0-ade9-20040a07e91c
Farchadi, Nima
c47af1f0-87f6-4322-b0b4-b8f3039c96ad
Alexander, Michael
0778be4f-a289-45a5-b95c-89de303eba24
Afonso, Pedro
27cf59e8-d472-4533-9bcc-1e746952f34d
Allyn, Andrew
de1e5fbe-e38d-47b2-8f39-0beafb2362f5
Bograd, Steven J.
1f002a82-b8af-4b79-9526-a776daa7d174
Brodie, Stephanie
5753e14e-b941-4e4d-ae45-9246e95abc8f
Crear, Daniel P.
e42b3702-3d8d-49b6-89d0-46db2c0fe6d3
Culhane, Emmett F.
93cef8fc-edac-43bc-914f-f6b54b270009
Curtis, Tobey H.
9c02af9a-0778-4976-8e67-3bf42d0fdf71
Hazen, Elliott L.
5f640051-0c4b-4089-8ea2-a590e59a54ce
Kerney, Alex
500db3b6-b80e-4b55-b651-11106edc3c13
Lezama‐ochoa, Nerea
4d11bea8-ae62-4f92-ab33-431059245665
Mills, Katherine E.
30e07e1e-0d28-43ed-a2cb-47c9419cd9e1
Pugh, Dylan
1107c312-f44b-430c-a09e-d5bf6edfbb34
Queiroz, Nuno
f36209dc-ab54-40b0-8d3c-08a73ba298ef
Scott, James D.
b8564bde-a8ca-4981-971b-0556ff9df03b
Skomal, Gregory B.
c33644a8-c2da-4188-becd-0f368a3a51c9
Sims, David W.
7234b444-25e2-4bd5-8348-a1c142d0cf81
Thorrold, Simon R.
fc28835b-1422-46dd-98e8-af2170f4680c
Welch, Heather
b382cd74-0306-4560-b960-b6b168867374
Young‐morse, Riley
6db9af60-800a-44e8-9efb-accb22c7ea71
Lewison, Rebecca L.
28744946-ab0d-465d-8d2b-984ea7d9d0cc
10 July 2023
Braun, Camrin D.
f53f1c67-16ce-4d16-996e-5b6ed7caae9d
Arostegui, Martin C.
b25b5e24-9e1a-4ff0-ade9-20040a07e91c
Farchadi, Nima
c47af1f0-87f6-4322-b0b4-b8f3039c96ad
Alexander, Michael
0778be4f-a289-45a5-b95c-89de303eba24
Afonso, Pedro
27cf59e8-d472-4533-9bcc-1e746952f34d
Allyn, Andrew
de1e5fbe-e38d-47b2-8f39-0beafb2362f5
Bograd, Steven J.
1f002a82-b8af-4b79-9526-a776daa7d174
Brodie, Stephanie
5753e14e-b941-4e4d-ae45-9246e95abc8f
Crear, Daniel P.
e42b3702-3d8d-49b6-89d0-46db2c0fe6d3
Culhane, Emmett F.
93cef8fc-edac-43bc-914f-f6b54b270009
Curtis, Tobey H.
9c02af9a-0778-4976-8e67-3bf42d0fdf71
Hazen, Elliott L.
5f640051-0c4b-4089-8ea2-a590e59a54ce
Kerney, Alex
500db3b6-b80e-4b55-b651-11106edc3c13
Lezama‐ochoa, Nerea
4d11bea8-ae62-4f92-ab33-431059245665
Mills, Katherine E.
30e07e1e-0d28-43ed-a2cb-47c9419cd9e1
Pugh, Dylan
1107c312-f44b-430c-a09e-d5bf6edfbb34
Queiroz, Nuno
f36209dc-ab54-40b0-8d3c-08a73ba298ef
Scott, James D.
b8564bde-a8ca-4981-971b-0556ff9df03b
Skomal, Gregory B.
c33644a8-c2da-4188-becd-0f368a3a51c9
Sims, David W.
7234b444-25e2-4bd5-8348-a1c142d0cf81
Thorrold, Simon R.
fc28835b-1422-46dd-98e8-af2170f4680c
Welch, Heather
b382cd74-0306-4560-b960-b6b168867374
Young‐morse, Riley
6db9af60-800a-44e8-9efb-accb22c7ea71
Lewison, Rebecca L.
28744946-ab0d-465d-8d2b-984ea7d9d0cc
Braun, Camrin D., Arostegui, Martin C., Farchadi, Nima, Alexander, Michael, Afonso, Pedro, Allyn, Andrew, Bograd, Steven J., Brodie, Stephanie, Crear, Daniel P., Culhane, Emmett F., Curtis, Tobey H., Hazen, Elliott L., Kerney, Alex, Lezama‐ochoa, Nerea, Mills, Katherine E., Pugh, Dylan, Queiroz, Nuno, Scott, James D., Skomal, Gregory B., Sims, David W., Thorrold, Simon R., Welch, Heather, Young‐morse, Riley and Lewison, Rebecca L.
(2023)
Building use‐inspired species distribution models: using multiple data types to examine and improve model performance.
Ecological Applications, 33 (6), [e2893].
(doi:10.1002/eap.2893).
Abstract
Species distribution models (SDMs) are becoming an important tool for marine conservation and management. Yet while there is an increasing diversity and volume of marine biodiversity data for training SDMs, little practical guidance is available on how to leverage distinct data types to build robust models. We explored the effect of different data types on the fit, performance and predictive ability of SDMs by comparing models trained with four data types for a heavily exploited pelagic fish, the blue shark (Prionace glauca), in the Northwest Atlantic: two fishery dependent (conventional mark-recapture tags, fisheries observer records) and two fishery independent (satellite-linked electronic tags, pop-up archival tags). We found that all four data types can result in robust models, but differences among spatial predictions highlighted the need to consider ecological realism in model selection and interpretation regardless of data type. Differences among models were primarily attributed to biases in how each data type, and the associated representation of absences, sampled the environment and summarized the resulting species distributions. Outputs from model ensembles and a model trained on all pooled data both proved effective for combining inferences across data types and provided more ecologically realistic predictions than individual models. Our results provide valuable guidance for practitioners developing SDMs. With increasing access to diverse data sources, future work should further develop truly integrative modeling approaches that can explicitly leverage the strengths of individual data types while statistically accounting for limitations, such as sampling biases.
Text
Ecological Applications - 2023 - Braun - Building use‐inspired species distribution models using multiple data types to
- Accepted Manuscript
Restricted to Repository staff only until 22 May 2024.
Request a copy
More information
Accepted/In Press date: 22 May 2023
e-pub ahead of print date: 7 June 2023
Published date: 10 July 2023
Additional Information:
Funding Information:
We thank all those who supported tagging efforts, the collection of observer program data, and those who contributed to the ICCAT marker tag program, including the NOAA Northeast Fisheries Science Center's Cooperative Shark Tagging Program. We thank the US Atlantic pelagic longline fishery observers and data providers from the NOAA Southeast Fisheries Science Center including L. Beerkircher and S. Cushner. We are grateful to the numerous captains and crews who provided their expertise and ship time and thank J. Suca for helpful comments on an earlier version of this manuscript. This work was supported by a NASA Ecological Forecasting funded project (80NSSC19K0187) and NOAA's Integrated Ecosystem Assessment program. Martin C. Arostegui was supported by the Postdoctoral Scholar Program at Woods Hole Oceanographic Institution with funding provided by the Dr. George D. Grice Postdoctoral Scholarship Fund. Emmett F. Culhane was supported by a Future Investigators in NASA Earth and Space Science and Technology (FINESST) award (80NSSC22K1549).
Publisher Copyright:
© 2023 The Ecological Society of America.
Keywords:
climate change, ecological forecasting, highly migratory species, prediction, spatial ecology, species distribution models
Identifiers
Local EPrints ID: 482046
URI: http://eprints.soton.ac.uk/id/eprint/482046
ISSN: 1051-0761
PURE UUID: e4cd11f1-51ee-4328-a198-3c18773389ff
Catalogue record
Date deposited: 18 Sep 2023 16:39
Last modified: 17 Mar 2024 03:55
Export record
Altmetrics
Contributors
Author:
Camrin D. Braun
Author:
Martin C. Arostegui
Author:
Nima Farchadi
Author:
Michael Alexander
Author:
Pedro Afonso
Author:
Andrew Allyn
Author:
Steven J. Bograd
Author:
Stephanie Brodie
Author:
Daniel P. Crear
Author:
Emmett F. Culhane
Author:
Tobey H. Curtis
Author:
Elliott L. Hazen
Author:
Alex Kerney
Author:
Nerea Lezama‐ochoa
Author:
Katherine E. Mills
Author:
Dylan Pugh
Author:
Nuno Queiroz
Author:
James D. Scott
Author:
Gregory B. Skomal
Author:
Simon R. Thorrold
Author:
Heather Welch
Author:
Riley Young‐morse
Author:
Rebecca L. Lewison
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics