Browse by author
Lookup NU author(s): Dr Fabrice StephensonORCiD
This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
© 2025 The Author(s). Diversity and Distributions published by John Wiley & Sons Ltd. Aim: Understanding the distribution of marine biodiversity is critical for evidence-based identification of areas for protection and restoration. Taxonomic richness is a key, intuitive component of biodiversity and is often used to guide marine spatial planning and protection. In this study, we explore the relative merits of two spatial modelling approaches, stacked species distribution models (S-SDMs) and macro-ecological models (MEMs), for mapping the richness of benthic invertebrate taxa. Location: New Zealand Exclusive Economic Zone. Methods: Two hundred and seven individual layers from SDMs of benthic invertebrate genera were pooled from an existing database and stacked to create a single genera richness layer. The same occurrence data used to develop the SDMs, comprising over 120k occurrences, was used to fit MEMs using an ensemble modelling approach. Results: The S-SDM layer performed poorly when validated against a database of observed genera richness, while the MEM approach performed well. While there were some consistencies in the areas predicted as high richness, substantial differences between the methods were also apparent, with the MEM seemingly better able to discern nuanced, fine-scale patterns in richness. Areas of high richness predicted by the MEM include parts of the Chatham Rise, a large component of the sub-Antarctic region, continental-shelf and coastal habitat in the south of the South Island, the north-east coast of the North Island, around East Cape and the Kermadec, Lau-Colville and Macquarie Ridges. Main Conclusions: Spatial and catchability biases in the underlying occurrence data may contribute to the poor performance of the S-SDM and suggest the approach may not be appropriate when using occurrence datasets with limited systematic sampling. The predictions from the MEM provide the best available information for the distribution of benthic invertebrate richness for New Zealand waters and thus offer important information for current and future marine spatial planning processes.
Author(s): Brough T, Stephenson F, Leunissen E, Lundquist C
Publication type: Article
Publication status: Published
Journal: Diversity and Distributions
Year: 2025
Volume: 31
Issue: 2
Online publication date: 18/02/2025
Acceptance date: 17/01/2025
Date deposited: 10/03/2025
ISSN (print): 1366-9516
ISSN (electronic): 1472-4642
Publisher: John Wiley and Sons Inc.
URL: https://doi.org/10.1111/ddi.70006
DOI: 10.1111/ddi.70006
Data Access Statement: All datasets underpinning the analysis undertaken in this study are freely available. The individual species distribution models used to generate the S-SDM and the raw data underpinning the MEM – counts of number of genera per 1km, matched with environmental variables – are available via the Dryad data repository at doi: https://doi.org/10.5061/dryad.fttdz091c (https://datadryad.org/stash/share/BppENOck1iQT0b_ikpoY0YbobMdEt1mh63IyAtDDu2U). In the same repository are R scripts used to generate and stack the individual SDMs that underpin the S-SDM richness layer, and a script used to generate the gear-class MEMs. Geotiff files for the final S-SDM, MEM and their associated uncertainty are also provided.
Altmetrics provided by Altmetric
