Combining Regional Habitat Selection Models for Large-Scale Prediction: Circumpolar Habitat Selection of Southern Ocean Humpback Whales
Machine learning algorithms are often used to model and predict animal habitat selection—the relationships between animal occurrences and habitat characteristics. For broadly distributed species, habitat selection often varies among populations and regions; thus, it would seem preferable to fit regi...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , , , , , , , , , , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
MDPI AG
2021
|
Subjects: | |
Online Access: | https://archimer.ifremer.fr/doc/00696/80845/84431.pdf https://archimer.ifremer.fr/doc/00696/80845/84432.zip https://doi.org/10.3390/rs13112074 https://archimer.ifremer.fr/doc/00696/80845/ |
id |
ftarchimer:oai:archimer.ifremer.fr:80845 |
---|---|
record_format |
openpolar |
spelling |
ftarchimer:oai:archimer.ifremer.fr:80845 2023-05-15T16:35:53+02:00 Combining Regional Habitat Selection Models for Large-Scale Prediction: Circumpolar Habitat Selection of Southern Ocean Humpback Whales Reisinger, Ryan R. Friedlaender, Ari S. Zerbini, Alexandre N. Palacios, Daniel M. Andrews-goff, Virginia Dalla Rosa, Luciano Double, Mike Findlay, Ken Garrigue, Claire How, Jason Jenner, Curt Jenner, Micheline-nicole Mate, Bruce Rosenbaum, Howard C. Seakamela, S. Mduduzi Constantine, Rochelle 2021-06 application/pdf https://archimer.ifremer.fr/doc/00696/80845/84431.pdf https://archimer.ifremer.fr/doc/00696/80845/84432.zip https://doi.org/10.3390/rs13112074 https://archimer.ifremer.fr/doc/00696/80845/ eng eng MDPI AG https://archimer.ifremer.fr/doc/00696/80845/84431.pdf https://archimer.ifremer.fr/doc/00696/80845/84432.zip doi:10.3390/rs13112074 https://archimer.ifremer.fr/doc/00696/80845/ info:eu-repo/semantics/openAccess restricted use Remote Sensing (2072-4292) (MDPI AG), 2021-06 , Vol. 13 , N. 11 , P. 2074 (23p.) ensembles habitat selection machine learning prediction resource selection functions telemetry humpback whale Megaptera novaeangliae text Publication info:eu-repo/semantics/article 2021 ftarchimer https://doi.org/10.3390/rs13112074 2021-09-23T20:37:42Z Machine learning algorithms are often used to model and predict animal habitat selection—the relationships between animal occurrences and habitat characteristics. For broadly distributed species, habitat selection often varies among populations and regions; thus, it would seem preferable to fit region- or population-specific models of habitat selection for more accurate inference and prediction, rather than fitting large-scale models using pooled data. However, where the aim is to make range-wide predictions, including areas for which there are no existing data or models of habitat selection, how can regional models best be combined? We propose that ensemble approaches commonly used to combine different algorithms for a single region can be reframed, treating regional habitat selection models as the candidate models. By doing so, we can incorporate regional variation when fitting predictive models of animal habitat selection across large ranges. We test this approach using satellite telemetry data from 168 humpback whales across five geographic regions in the Southern Ocean. Using random forests, we fitted a large-scale model relating humpback whale locations, versus background locations, to 10 environmental covariates, and made a circumpolar prediction of humpback whale habitat selection. We also fitted five regional models, the predictions of which we used as input features for four ensemble approaches: an unweighted ensemble, an ensemble weighted by environmental similarity in each cell, stacked generalization, and a hybrid approach wherein the environmental covariates and regional predictions were used as input features in a new model. We tested the predictive performance of these approaches on an independent validation dataset of humpback whale sightings and whaling catches. These multiregional ensemble approaches resulted in models with higher predictive performance than the circumpolar naive model. These approaches can be used to incorporate regional variation in animal habitat selection when fitting range-wide predictive models using machine learning algorithms. This can yield more accurate predictions across regions or populations of animals that may show variation in habitat selection Article in Journal/Newspaper Humpback Whale Megaptera novaeangliae Southern Ocean Archimer (Archive Institutionnelle de l'Ifremer - Institut français de recherche pour l'exploitation de la mer) Southern Ocean Remote Sensing 13 11 2074 |
institution |
Open Polar |
collection |
Archimer (Archive Institutionnelle de l'Ifremer - Institut français de recherche pour l'exploitation de la mer) |
op_collection_id |
ftarchimer |
language |
English |
topic |
ensembles habitat selection machine learning prediction resource selection functions telemetry humpback whale Megaptera novaeangliae |
spellingShingle |
ensembles habitat selection machine learning prediction resource selection functions telemetry humpback whale Megaptera novaeangliae Reisinger, Ryan R. Friedlaender, Ari S. Zerbini, Alexandre N. Palacios, Daniel M. Andrews-goff, Virginia Dalla Rosa, Luciano Double, Mike Findlay, Ken Garrigue, Claire How, Jason Jenner, Curt Jenner, Micheline-nicole Mate, Bruce Rosenbaum, Howard C. Seakamela, S. Mduduzi Constantine, Rochelle Combining Regional Habitat Selection Models for Large-Scale Prediction: Circumpolar Habitat Selection of Southern Ocean Humpback Whales |
topic_facet |
ensembles habitat selection machine learning prediction resource selection functions telemetry humpback whale Megaptera novaeangliae |
description |
Machine learning algorithms are often used to model and predict animal habitat selection—the relationships between animal occurrences and habitat characteristics. For broadly distributed species, habitat selection often varies among populations and regions; thus, it would seem preferable to fit region- or population-specific models of habitat selection for more accurate inference and prediction, rather than fitting large-scale models using pooled data. However, where the aim is to make range-wide predictions, including areas for which there are no existing data or models of habitat selection, how can regional models best be combined? We propose that ensemble approaches commonly used to combine different algorithms for a single region can be reframed, treating regional habitat selection models as the candidate models. By doing so, we can incorporate regional variation when fitting predictive models of animal habitat selection across large ranges. We test this approach using satellite telemetry data from 168 humpback whales across five geographic regions in the Southern Ocean. Using random forests, we fitted a large-scale model relating humpback whale locations, versus background locations, to 10 environmental covariates, and made a circumpolar prediction of humpback whale habitat selection. We also fitted five regional models, the predictions of which we used as input features for four ensemble approaches: an unweighted ensemble, an ensemble weighted by environmental similarity in each cell, stacked generalization, and a hybrid approach wherein the environmental covariates and regional predictions were used as input features in a new model. We tested the predictive performance of these approaches on an independent validation dataset of humpback whale sightings and whaling catches. These multiregional ensemble approaches resulted in models with higher predictive performance than the circumpolar naive model. These approaches can be used to incorporate regional variation in animal habitat selection when fitting range-wide predictive models using machine learning algorithms. This can yield more accurate predictions across regions or populations of animals that may show variation in habitat selection |
format |
Article in Journal/Newspaper |
author |
Reisinger, Ryan R. Friedlaender, Ari S. Zerbini, Alexandre N. Palacios, Daniel M. Andrews-goff, Virginia Dalla Rosa, Luciano Double, Mike Findlay, Ken Garrigue, Claire How, Jason Jenner, Curt Jenner, Micheline-nicole Mate, Bruce Rosenbaum, Howard C. Seakamela, S. Mduduzi Constantine, Rochelle |
author_facet |
Reisinger, Ryan R. Friedlaender, Ari S. Zerbini, Alexandre N. Palacios, Daniel M. Andrews-goff, Virginia Dalla Rosa, Luciano Double, Mike Findlay, Ken Garrigue, Claire How, Jason Jenner, Curt Jenner, Micheline-nicole Mate, Bruce Rosenbaum, Howard C. Seakamela, S. Mduduzi Constantine, Rochelle |
author_sort |
Reisinger, Ryan R. |
title |
Combining Regional Habitat Selection Models for Large-Scale Prediction: Circumpolar Habitat Selection of Southern Ocean Humpback Whales |
title_short |
Combining Regional Habitat Selection Models for Large-Scale Prediction: Circumpolar Habitat Selection of Southern Ocean Humpback Whales |
title_full |
Combining Regional Habitat Selection Models for Large-Scale Prediction: Circumpolar Habitat Selection of Southern Ocean Humpback Whales |
title_fullStr |
Combining Regional Habitat Selection Models for Large-Scale Prediction: Circumpolar Habitat Selection of Southern Ocean Humpback Whales |
title_full_unstemmed |
Combining Regional Habitat Selection Models for Large-Scale Prediction: Circumpolar Habitat Selection of Southern Ocean Humpback Whales |
title_sort |
combining regional habitat selection models for large-scale prediction: circumpolar habitat selection of southern ocean humpback whales |
publisher |
MDPI AG |
publishDate |
2021 |
url |
https://archimer.ifremer.fr/doc/00696/80845/84431.pdf https://archimer.ifremer.fr/doc/00696/80845/84432.zip https://doi.org/10.3390/rs13112074 https://archimer.ifremer.fr/doc/00696/80845/ |
geographic |
Southern Ocean |
geographic_facet |
Southern Ocean |
genre |
Humpback Whale Megaptera novaeangliae Southern Ocean |
genre_facet |
Humpback Whale Megaptera novaeangliae Southern Ocean |
op_source |
Remote Sensing (2072-4292) (MDPI AG), 2021-06 , Vol. 13 , N. 11 , P. 2074 (23p.) |
op_relation |
https://archimer.ifremer.fr/doc/00696/80845/84431.pdf https://archimer.ifremer.fr/doc/00696/80845/84432.zip doi:10.3390/rs13112074 https://archimer.ifremer.fr/doc/00696/80845/ |
op_rights |
info:eu-repo/semantics/openAccess restricted use |
op_doi |
https://doi.org/10.3390/rs13112074 |
container_title |
Remote Sensing |
container_volume |
13 |
container_issue |
11 |
container_start_page |
2074 |
_version_ |
1766026189019807744 |