Broad-scale species distribution models applied to data-poor areas.
10 pages International audience Species distribution models (SDMs) have been increasingly used over the past decades to characterise the spatial distribution and the ecological niche of various taxa. Validating predicted species distribution is important, especially when producing broad-scale models...
Published in: | Progress in Oceanography |
---|---|
Main Authors: | , , , , , , , |
Other Authors: | , , , , , , , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
HAL CCSD
2019
|
Subjects: | |
Online Access: | https://hal.science/hal-02270076 https://hal.science/hal-02270076/document https://hal.science/hal-02270076/file/S0079661118301939.pdf https://doi.org/10.1016/j.pocean.2019.04.007 |
id |
ftmuseumnhn:oai:HAL:hal-02270076v1 |
---|---|
record_format |
openpolar |
institution |
Open Polar |
collection |
Muséum National d'Histoire Naturelle (MNHM): HAL |
op_collection_id |
ftmuseumnhn |
language |
English |
topic |
Boosted Regression Trees (BRTs) Presence-only Cross-validation Extrapolation Modelling evaluation [SDE.BE]Environmental Sciences/Biodiversity and Ecology |
spellingShingle |
Boosted Regression Trees (BRTs) Presence-only Cross-validation Extrapolation Modelling evaluation [SDE.BE]Environmental Sciences/Biodiversity and Ecology Guillaumot, Charlène Artois, Jean Saucède, Thomas Demoustier, Laura Moreau, Camille Eléaume, Marc Agüera, Antonio Danis, Bruno Broad-scale species distribution models applied to data-poor areas. |
topic_facet |
Boosted Regression Trees (BRTs) Presence-only Cross-validation Extrapolation Modelling evaluation [SDE.BE]Environmental Sciences/Biodiversity and Ecology |
description |
10 pages International audience Species distribution models (SDMs) have been increasingly used over the past decades to characterise the spatial distribution and the ecological niche of various taxa. Validating predicted species distribution is important, especially when producing broad-scale models (i.e. at continental or oceanic scale) based on limited and spatially aggregated presence-only records. In the present study, several model calibration methods are compared and guidelines are provided to perform relevant SDMs using a Southern Ocean marine species, the starfish Odontaster validus Koehler, 1906, as a case study. The effect of the spatial aggregation of presence-only records on modelling performance is evaluated and the relevance of a target-background sampling procedure to correct for this effect is assessed. The accuracy of model validation is estimated using k-fold random and spatial cross-validation procedures. Finally, we evaluate the relevance of the Multivariate Environmental Similarity Surface (MESS) index to identify areas in which SDMs accurately interpolate and conversely, areas in which models extrapolate outside the environmental range of occurrence records.Results show that the random cross-validation procedure (i.e. a widely applied method, for which training and test records are randomly selected in space) tends to over-estimate model performance when applied to spatially aggregated datasets. Spatial cross-validation procedures can compensate for this over-estimation effect but different spatial cross-validation procedures must be tested for their ability to reduce over-fitting while providing relevant validation scores. Model predictions show that SDM generalisation is limited when working with aggregated datasets at broad spatial scale. The MESS index calculated in our case study show that over half of the predicted area is highly uncertain due to extrapolation. Our work provides methodological guidelines to generate accurate model assessments at broad spatial scale when using ... |
author2 |
Laboratoire de Biologie Marine Université libre de Bruxelles (ULB) Spatial Epidemiology Lab (SpELL) Biogéosciences UMR 6282 (BGS) Université de Bourgogne (UB)-Centre National de la Recherche Scientifique (CNRS) Institut de Systématique, Evolution, Biodiversité (ISYEB ) Muséum national d'Histoire naturelle (MNHN)-École Pratique des Hautes Études (EPHE) Université Paris Sciences et Lettres (PSL)-Université Paris Sciences et Lettres (PSL)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Université des Antilles (UA) Danish Shellfish Center DTU-aqua Work supported by a “Fonds pour la formation à la Recherche dans l’Industrie et l’Agriculture” (FRIA) grant and by the Belgian Science Policy Office(BELSPO, contract n°BR/132/A1/vERSO). |
format |
Article in Journal/Newspaper |
author |
Guillaumot, Charlène Artois, Jean Saucède, Thomas Demoustier, Laura Moreau, Camille Eléaume, Marc Agüera, Antonio Danis, Bruno |
author_facet |
Guillaumot, Charlène Artois, Jean Saucède, Thomas Demoustier, Laura Moreau, Camille Eléaume, Marc Agüera, Antonio Danis, Bruno |
author_sort |
Guillaumot, Charlène |
title |
Broad-scale species distribution models applied to data-poor areas. |
title_short |
Broad-scale species distribution models applied to data-poor areas. |
title_full |
Broad-scale species distribution models applied to data-poor areas. |
title_fullStr |
Broad-scale species distribution models applied to data-poor areas. |
title_full_unstemmed |
Broad-scale species distribution models applied to data-poor areas. |
title_sort |
broad-scale species distribution models applied to data-poor areas. |
publisher |
HAL CCSD |
publishDate |
2019 |
url |
https://hal.science/hal-02270076 https://hal.science/hal-02270076/document https://hal.science/hal-02270076/file/S0079661118301939.pdf https://doi.org/10.1016/j.pocean.2019.04.007 |
genre |
Southern Ocean |
genre_facet |
Southern Ocean |
op_source |
ISSN: 0079-6611 Progress in Oceanography https://hal.science/hal-02270076 Progress in Oceanography, 2019, 175, pp.198-207. ⟨10.1016/j.pocean.2019.04.007⟩ https://www.sciencedirect.com/science/article/pii/S0079661118301939 |
op_relation |
info:eu-repo/semantics/altIdentifier/doi/10.1016/j.pocean.2019.04.007 hal-02270076 https://hal.science/hal-02270076 https://hal.science/hal-02270076/document https://hal.science/hal-02270076/file/S0079661118301939.pdf doi:10.1016/j.pocean.2019.04.007 PII: S0079-6611(18)30193-9 |
op_rights |
http://creativecommons.org/licenses/by-nc/ info:eu-repo/semantics/OpenAccess |
op_doi |
https://doi.org/10.1016/j.pocean.2019.04.007 |
container_title |
Progress in Oceanography |
container_volume |
175 |
container_start_page |
198 |
op_container_end_page |
207 |
_version_ |
1799467405013942272 |
spelling |
ftmuseumnhn:oai:HAL:hal-02270076v1 2024-05-19T07:49:00+00:00 Broad-scale species distribution models applied to data-poor areas. Guillaumot, Charlène Artois, Jean Saucède, Thomas Demoustier, Laura Moreau, Camille Eléaume, Marc Agüera, Antonio Danis, Bruno Laboratoire de Biologie Marine Université libre de Bruxelles (ULB) Spatial Epidemiology Lab (SpELL) Biogéosciences UMR 6282 (BGS) Université de Bourgogne (UB)-Centre National de la Recherche Scientifique (CNRS) Institut de Systématique, Evolution, Biodiversité (ISYEB ) Muséum national d'Histoire naturelle (MNHN)-École Pratique des Hautes Études (EPHE) Université Paris Sciences et Lettres (PSL)-Université Paris Sciences et Lettres (PSL)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Université des Antilles (UA) Danish Shellfish Center DTU-aqua Work supported by a “Fonds pour la formation à la Recherche dans l’Industrie et l’Agriculture” (FRIA) grant and by the Belgian Science Policy Office(BELSPO, contract n°BR/132/A1/vERSO). 2019-07 https://hal.science/hal-02270076 https://hal.science/hal-02270076/document https://hal.science/hal-02270076/file/S0079661118301939.pdf https://doi.org/10.1016/j.pocean.2019.04.007 en eng HAL CCSD Elsevier info:eu-repo/semantics/altIdentifier/doi/10.1016/j.pocean.2019.04.007 hal-02270076 https://hal.science/hal-02270076 https://hal.science/hal-02270076/document https://hal.science/hal-02270076/file/S0079661118301939.pdf doi:10.1016/j.pocean.2019.04.007 PII: S0079-6611(18)30193-9 http://creativecommons.org/licenses/by-nc/ info:eu-repo/semantics/OpenAccess ISSN: 0079-6611 Progress in Oceanography https://hal.science/hal-02270076 Progress in Oceanography, 2019, 175, pp.198-207. ⟨10.1016/j.pocean.2019.04.007⟩ https://www.sciencedirect.com/science/article/pii/S0079661118301939 Boosted Regression Trees (BRTs) Presence-only Cross-validation Extrapolation Modelling evaluation [SDE.BE]Environmental Sciences/Biodiversity and Ecology info:eu-repo/semantics/article Journal articles 2019 ftmuseumnhn https://doi.org/10.1016/j.pocean.2019.04.007 2024-04-25T00:45:05Z 10 pages International audience Species distribution models (SDMs) have been increasingly used over the past decades to characterise the spatial distribution and the ecological niche of various taxa. Validating predicted species distribution is important, especially when producing broad-scale models (i.e. at continental or oceanic scale) based on limited and spatially aggregated presence-only records. In the present study, several model calibration methods are compared and guidelines are provided to perform relevant SDMs using a Southern Ocean marine species, the starfish Odontaster validus Koehler, 1906, as a case study. The effect of the spatial aggregation of presence-only records on modelling performance is evaluated and the relevance of a target-background sampling procedure to correct for this effect is assessed. The accuracy of model validation is estimated using k-fold random and spatial cross-validation procedures. Finally, we evaluate the relevance of the Multivariate Environmental Similarity Surface (MESS) index to identify areas in which SDMs accurately interpolate and conversely, areas in which models extrapolate outside the environmental range of occurrence records.Results show that the random cross-validation procedure (i.e. a widely applied method, for which training and test records are randomly selected in space) tends to over-estimate model performance when applied to spatially aggregated datasets. Spatial cross-validation procedures can compensate for this over-estimation effect but different spatial cross-validation procedures must be tested for their ability to reduce over-fitting while providing relevant validation scores. Model predictions show that SDM generalisation is limited when working with aggregated datasets at broad spatial scale. The MESS index calculated in our case study show that over half of the predicted area is highly uncertain due to extrapolation. Our work provides methodological guidelines to generate accurate model assessments at broad spatial scale when using ... Article in Journal/Newspaper Southern Ocean Muséum National d'Histoire Naturelle (MNHM): HAL Progress in Oceanography 175 198 207 |