Modeling joint abundance of multiple species using Dirichlet process mixtures
We present a method for modeling the distributions of multiple species simultaneously using Dirichlet process random effects to cluster species into guilds. Guilds are ecological groups of species that behave or react similarly to some environmental conditions. By modeling latent guild structure, we...
Published in: | Environmetrics |
---|---|
Main Authors: | , |
Format: | Article in Journal/Newspaper |
Language: | unknown |
Subjects: | |
Online Access: | https://doi.org/10.1002/env.2440 |
id |
ftrepec:oai:RePEc:wly:envmet:v:28:y:2017:i:3:n:e2440 |
---|---|
record_format |
openpolar |
spelling |
ftrepec:oai:RePEc:wly:envmet:v:28:y:2017:i:3:n:e2440 2023-05-15T15:43:49+02:00 Modeling joint abundance of multiple species using Dirichlet process mixtures Devin S. Johnson Elizabeth H. Sinclair https://doi.org/10.1002/env.2440 unknown https://doi.org/10.1002/env.2440 article ftrepec https://doi.org/10.1002/env.2440 2020-12-04T13:36:37Z We present a method for modeling the distributions of multiple species simultaneously using Dirichlet process random effects to cluster species into guilds. Guilds are ecological groups of species that behave or react similarly to some environmental conditions. By modeling latent guild structure, we capture the cross‐correlations in abundance or occurrence of species over surveys. In addition, ecological information about the community structure is obtained as a by‐product of the model. By clustering species into similar functional groups, prediction uncertainty of community structure at additional sites is reduced over treating each species separately. The proposed model also presents an improvement over previously proposed joint species distribution models by reducing the number of parameters necessary to capture interspecies correlations and eliminating the need to have a priori information on the number of groups or a distance metric over species traits. The method is illustrated with a small simulation demonstration, as well as an analysis of a mesopelagic fish survey from the eastern Bering Sea near Alaska. The simulation data analysis shows that guild membership can be extracted as the differences between groups become larger and if guild differences are small, the model naturally collapses all the species into a small number of guilds, which increases predictive efficiency by reducing the number of parameters to that which is supported by the data. Article in Journal/Newspaper Bering Sea Alaska RePEc (Research Papers in Economics) Bering Sea Environmetrics 28 3 e2440 |
institution |
Open Polar |
collection |
RePEc (Research Papers in Economics) |
op_collection_id |
ftrepec |
language |
unknown |
description |
We present a method for modeling the distributions of multiple species simultaneously using Dirichlet process random effects to cluster species into guilds. Guilds are ecological groups of species that behave or react similarly to some environmental conditions. By modeling latent guild structure, we capture the cross‐correlations in abundance or occurrence of species over surveys. In addition, ecological information about the community structure is obtained as a by‐product of the model. By clustering species into similar functional groups, prediction uncertainty of community structure at additional sites is reduced over treating each species separately. The proposed model also presents an improvement over previously proposed joint species distribution models by reducing the number of parameters necessary to capture interspecies correlations and eliminating the need to have a priori information on the number of groups or a distance metric over species traits. The method is illustrated with a small simulation demonstration, as well as an analysis of a mesopelagic fish survey from the eastern Bering Sea near Alaska. The simulation data analysis shows that guild membership can be extracted as the differences between groups become larger and if guild differences are small, the model naturally collapses all the species into a small number of guilds, which increases predictive efficiency by reducing the number of parameters to that which is supported by the data. |
format |
Article in Journal/Newspaper |
author |
Devin S. Johnson Elizabeth H. Sinclair |
spellingShingle |
Devin S. Johnson Elizabeth H. Sinclair Modeling joint abundance of multiple species using Dirichlet process mixtures |
author_facet |
Devin S. Johnson Elizabeth H. Sinclair |
author_sort |
Devin S. Johnson |
title |
Modeling joint abundance of multiple species using Dirichlet process mixtures |
title_short |
Modeling joint abundance of multiple species using Dirichlet process mixtures |
title_full |
Modeling joint abundance of multiple species using Dirichlet process mixtures |
title_fullStr |
Modeling joint abundance of multiple species using Dirichlet process mixtures |
title_full_unstemmed |
Modeling joint abundance of multiple species using Dirichlet process mixtures |
title_sort |
modeling joint abundance of multiple species using dirichlet process mixtures |
url |
https://doi.org/10.1002/env.2440 |
geographic |
Bering Sea |
geographic_facet |
Bering Sea |
genre |
Bering Sea Alaska |
genre_facet |
Bering Sea Alaska |
op_relation |
https://doi.org/10.1002/env.2440 |
op_doi |
https://doi.org/10.1002/env.2440 |
container_title |
Environmetrics |
container_volume |
28 |
container_issue |
3 |
container_start_page |
e2440 |
_version_ |
1766378023933706240 |