Finding the right fit: Comparative cetacean distribution models using multiple data sources and statistical approaches
International audience Aim: Accurate predictions of cetacean distributions are essential to their conservation but are limited by statistical challenges and a paucity of data. This study aimed at comparing the capacity of various statistical algorithms to deal with biases commonly found in nonsystem...
Published in: | Diversity and Distributions |
---|---|
Main Authors: | , , , |
Other Authors: | , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
HAL CCSD
2018
|
Subjects: | |
Online Access: | https://doi.org/10.1111/ddi.12782 https://hal.sorbonne-universite.fr/hal-01829624/file/Derville%20et%20al.%202018_Diversity_and_Distributions_sans%20marque.pdf https://hal.sorbonne-universite.fr/hal-01829624 |
id |
fttriple:oai:gotriple.eu:10670/1.jsa0q9 |
---|---|
record_format |
openpolar |
spelling |
fttriple:oai:gotriple.eu:10670/1.jsa0q9 2023-05-15T17:10:52+02:00 Finding the right fit: Comparative cetacean distribution models using multiple data sources and statistical approaches Derville, Solene Torres, Leigh G. Iovan, Corina Garrigue, Claire Université Pierre et Marie Curie - Paris 6 (UPMC) Ecologie marine tropicale dans les Océans Pacifique et Indien (ENTROPIE Réunion ) Centre National de la Recherche Scientifique (CNRS)-Université de La Réunion (UR)-Institut de Recherche pour le Développement (IRD) Oregon State University (OSU) 2018-01-01 https://doi.org/10.1111/ddi.12782 https://hal.sorbonne-universite.fr/hal-01829624/file/Derville%20et%20al.%202018_Diversity_and_Distributions_sans%20marque.pdf https://hal.sorbonne-universite.fr/hal-01829624 en eng HAL CCSD Wiley hal-01829624 doi:10.1111/ddi.12782 10670/1.jsa0q9 https://hal.sorbonne-universite.fr/hal-01829624/file/Derville%20et%20al.%202018_Diversity_and_Distributions_sans%20marque.pdf https://hal.sorbonne-universite.fr/hal-01829624 Hyper Article en Ligne - Sciences de l'Homme et de la Société ISSN: 1366-9516 EISSN: 1472-4642 Diversity and Distributions Diversity and Distributions, Wiley, 2018, ⟨10.1111/ddi.12782⟩ envir geo Journal Article https://vocabularies.coar-repositories.org/resource_types/c_6501/ 2018 fttriple https://doi.org/10.1111/ddi.12782 2023-01-22T17:00:14Z International audience Aim: Accurate predictions of cetacean distributions are essential to their conservation but are limited by statistical challenges and a paucity of data. This study aimed at comparing the capacity of various statistical algorithms to deal with biases commonly found in nonsystematic cetacean surveys and to evaluate the potential for citizen science data to improve habitat modelling and predictions. An endangered population of humpback whales (Megaptera novaeangliae) in their breeding ground was used as a case study. Location: New Caledonia, Oceania. Methods: Five statistical algorithms were used to model the habitat preferences of humpback whales from 1,360 sightings collected over 14 years of nonsystematic research surveys. Three different background sampling approaches were tested when developing models from 625 crowdsourced sightings to assess methods accounting for citizen science spatial sampling bias. Model evaluation was conducted through cross-validation and prediction to an independent satellite tracking dataset. Results: Algorithms differed in complexity of the environmental relationships modelled , ecological interpretability and transferability. While parameter tuning had a great effect on model performances, GLMs generally had low predictive performance , SVMs were particularly hard to interpret, and BRTs had high descriptive power but showed signs of overfitting. MAXENT and especially GAMs provided a valuable complexity trade-off, accurate predictions and were ecologically intelligible. Models showed that humpback whales favoured cool (22–23°C) and shallow waters (0–100 m deep) in coastal as well as offshore areas. Citizen science models converged with research survey models, specifically when accounting for spatial sampling bias. Main conclusions: Marine megafauna distribution models present specific challenges that may be addressed through integrative evaluation, independent testing and appropriately tuned statistical algorithms. Specifically, controlling overfitting is a ... Article in Journal/Newspaper Megaptera novaeangliae Unknown Diversity and Distributions 24 11 1657 1673 |
institution |
Open Polar |
collection |
Unknown |
op_collection_id |
fttriple |
language |
English |
topic |
envir geo |
spellingShingle |
envir geo Derville, Solene Torres, Leigh G. Iovan, Corina Garrigue, Claire Finding the right fit: Comparative cetacean distribution models using multiple data sources and statistical approaches |
topic_facet |
envir geo |
description |
International audience Aim: Accurate predictions of cetacean distributions are essential to their conservation but are limited by statistical challenges and a paucity of data. This study aimed at comparing the capacity of various statistical algorithms to deal with biases commonly found in nonsystematic cetacean surveys and to evaluate the potential for citizen science data to improve habitat modelling and predictions. An endangered population of humpback whales (Megaptera novaeangliae) in their breeding ground was used as a case study. Location: New Caledonia, Oceania. Methods: Five statistical algorithms were used to model the habitat preferences of humpback whales from 1,360 sightings collected over 14 years of nonsystematic research surveys. Three different background sampling approaches were tested when developing models from 625 crowdsourced sightings to assess methods accounting for citizen science spatial sampling bias. Model evaluation was conducted through cross-validation and prediction to an independent satellite tracking dataset. Results: Algorithms differed in complexity of the environmental relationships modelled , ecological interpretability and transferability. While parameter tuning had a great effect on model performances, GLMs generally had low predictive performance , SVMs were particularly hard to interpret, and BRTs had high descriptive power but showed signs of overfitting. MAXENT and especially GAMs provided a valuable complexity trade-off, accurate predictions and were ecologically intelligible. Models showed that humpback whales favoured cool (22–23°C) and shallow waters (0–100 m deep) in coastal as well as offshore areas. Citizen science models converged with research survey models, specifically when accounting for spatial sampling bias. Main conclusions: Marine megafauna distribution models present specific challenges that may be addressed through integrative evaluation, independent testing and appropriately tuned statistical algorithms. Specifically, controlling overfitting is a ... |
author2 |
Université Pierre et Marie Curie - Paris 6 (UPMC) Ecologie marine tropicale dans les Océans Pacifique et Indien (ENTROPIE Réunion ) Centre National de la Recherche Scientifique (CNRS)-Université de La Réunion (UR)-Institut de Recherche pour le Développement (IRD) Oregon State University (OSU) |
format |
Article in Journal/Newspaper |
author |
Derville, Solene Torres, Leigh G. Iovan, Corina Garrigue, Claire |
author_facet |
Derville, Solene Torres, Leigh G. Iovan, Corina Garrigue, Claire |
author_sort |
Derville, Solene |
title |
Finding the right fit: Comparative cetacean distribution models using multiple data sources and statistical approaches |
title_short |
Finding the right fit: Comparative cetacean distribution models using multiple data sources and statistical approaches |
title_full |
Finding the right fit: Comparative cetacean distribution models using multiple data sources and statistical approaches |
title_fullStr |
Finding the right fit: Comparative cetacean distribution models using multiple data sources and statistical approaches |
title_full_unstemmed |
Finding the right fit: Comparative cetacean distribution models using multiple data sources and statistical approaches |
title_sort |
finding the right fit: comparative cetacean distribution models using multiple data sources and statistical approaches |
publisher |
HAL CCSD |
publishDate |
2018 |
url |
https://doi.org/10.1111/ddi.12782 https://hal.sorbonne-universite.fr/hal-01829624/file/Derville%20et%20al.%202018_Diversity_and_Distributions_sans%20marque.pdf https://hal.sorbonne-universite.fr/hal-01829624 |
genre |
Megaptera novaeangliae |
genre_facet |
Megaptera novaeangliae |
op_source |
Hyper Article en Ligne - Sciences de l'Homme et de la Société ISSN: 1366-9516 EISSN: 1472-4642 Diversity and Distributions Diversity and Distributions, Wiley, 2018, ⟨10.1111/ddi.12782⟩ |
op_relation |
hal-01829624 doi:10.1111/ddi.12782 10670/1.jsa0q9 https://hal.sorbonne-universite.fr/hal-01829624/file/Derville%20et%20al.%202018_Diversity_and_Distributions_sans%20marque.pdf https://hal.sorbonne-universite.fr/hal-01829624 |
op_doi |
https://doi.org/10.1111/ddi.12782 |
container_title |
Diversity and Distributions |
container_volume |
24 |
container_issue |
11 |
container_start_page |
1657 |
op_container_end_page |
1673 |
_version_ |
1766067530563059712 |