A sequential multi-level framework to improve habitat suitability modelling
A sequential multi-level framework to improve habitat suitability modelling
Context: habitat suitability models (HSM) can improve our understanding of a species’ ecology and are valuable tools for informing landscape-scale decisions. We can increase HSM predictive accuracy and derive more realistic conclusions by taking a multi-scale approach. However, this process is often statistically complex and computationally intensive.
Objectives: we provide an easily implemented, flexible framework for sequential multi-level, multi-scale HSM and compare it to two other commonly-applied approaches: single-level, multi-scale HSM and their post-hoc combinations.
Methods: our framework implements scale optimisation and model tuning at each level in turn, from the highest (population range) to the lowest (e.g. foraging habitat) level, whilst incorporating output habitat suitability indices from a higher level as a predictor. We used MaxEnt and a species of conservation concern in Britain, the lesser horseshoe bat (Rhinolophus hipposideros), to demonstrate and compare multi-scale approaches.
Results: integrating models across levels, either by applying our framework, or by multiplying single-level model predictions, improved predictive performance over single-level models. Moreover, differences in the importance and direction of the species-environment associations highlight the potential for false inferences from single-level models or their post-hoc combinations. The single-level summer range model incorrectly identified a positive influence of heathland cover, whereas sequential multi-level models made biological sense and underlined this species’ requirement for extensive broadleaf woodland cover, hedgerows and access to buildings for roosting in rural areas.
Conclusions: we conclude that multi-level HSM appear superior to single-level, multi-scale approaches; models should be sequentially integrated across levels if information on species-environment relationships is of importance.
Lesser horseshoe bat, MaxEnt, Multi-scale, Nested model, Rhinolophus hipposideros, Species distribution model
10.1007%2Fs10980-020-00987-w
1001-1020
Bellamy, Chloe
02a9be14-27ea-49fa-93ab-094fa9a17341
Boughey, K.L.
ad77e8d6-4731-4eba-b51f-228662196203
Hawkins, Charlotte
ead165d7-7144-477f-9095-be0f8aef3850
Reveley, Sonia
5618035c-1fb3-49bf-b793-98c219c350ec
Spake, Rebecca
1cda8ad0-2ab2-45d9-a844-ec3d8be2786a
Williams, Carol
8b326737-4a5d-4949-a68d-4fd4fe75bb8c
Altringham, John
ab2b584c-45ca-411b-9ee9-58b85c24c1b7
1 April 2020
Bellamy, Chloe
02a9be14-27ea-49fa-93ab-094fa9a17341
Boughey, K.L.
ad77e8d6-4731-4eba-b51f-228662196203
Hawkins, Charlotte
ead165d7-7144-477f-9095-be0f8aef3850
Reveley, Sonia
5618035c-1fb3-49bf-b793-98c219c350ec
Spake, Rebecca
1cda8ad0-2ab2-45d9-a844-ec3d8be2786a
Williams, Carol
8b326737-4a5d-4949-a68d-4fd4fe75bb8c
Altringham, John
ab2b584c-45ca-411b-9ee9-58b85c24c1b7
Bellamy, Chloe, Boughey, K.L., Hawkins, Charlotte, Reveley, Sonia, Spake, Rebecca, Williams, Carol and Altringham, John
(2020)
A sequential multi-level framework to improve habitat suitability modelling.
Landscape Ecology, 35 (4), .
(doi:10.1007%2Fs10980-020-00987-w).
Abstract
Context: habitat suitability models (HSM) can improve our understanding of a species’ ecology and are valuable tools for informing landscape-scale decisions. We can increase HSM predictive accuracy and derive more realistic conclusions by taking a multi-scale approach. However, this process is often statistically complex and computationally intensive.
Objectives: we provide an easily implemented, flexible framework for sequential multi-level, multi-scale HSM and compare it to two other commonly-applied approaches: single-level, multi-scale HSM and their post-hoc combinations.
Methods: our framework implements scale optimisation and model tuning at each level in turn, from the highest (population range) to the lowest (e.g. foraging habitat) level, whilst incorporating output habitat suitability indices from a higher level as a predictor. We used MaxEnt and a species of conservation concern in Britain, the lesser horseshoe bat (Rhinolophus hipposideros), to demonstrate and compare multi-scale approaches.
Results: integrating models across levels, either by applying our framework, or by multiplying single-level model predictions, improved predictive performance over single-level models. Moreover, differences in the importance and direction of the species-environment associations highlight the potential for false inferences from single-level models or their post-hoc combinations. The single-level summer range model incorrectly identified a positive influence of heathland cover, whereas sequential multi-level models made biological sense and underlined this species’ requirement for extensive broadleaf woodland cover, hedgerows and access to buildings for roosting in rural areas.
Conclusions: we conclude that multi-level HSM appear superior to single-level, multi-scale approaches; models should be sequentially integrated across levels if information on species-environment relationships is of importance.
Text
Bellamy2020_Article_ASequentialMulti-levelFramewor
- Version of Record
More information
Accepted/In Press date: 26 February 2020
e-pub ahead of print date: 16 March 2020
Published date: 1 April 2020
Additional Information:
Funding Information:
The ‘Putting Woodland Bats on the Map’ project has been funded and supported by various organisations including: Bat Conservation Trust (BCT), Ernest Kleinwort, Forestry England (South England Forest District), Forest Research, J & JR Wilson, the Margaret Joan Tottle Deceased Will Trust, the Martin Wills Wildlife Maintenance Trust, Scottish Forestry Trust, Scottish Natural Heritage, the Edith Murphy Foundation, the Late Miss Eileen Margaret Tyler's Charitable Trust and the Woodland Trust. We are grateful for the many individuals and organisations involved in the project and to those submitting species records directly and via the NBN Gateway and BCT (listed in the Supplementary Information (SI4)). These data came, in part, from the National Bat Monitoring Programme (NBMP), which is run by Bat Conservation Trust in partnership with the Joint Nature Conservation Committee, and supported and steered by Natural England, Natural Resources Wales, Northern Ireland Environment Agency, and Scottish Natural Heritage. The NBMP is indebted to all volunteers who contribute data to the programme. Thanks also for permission to use the woody linear features dataset (Paul Scholefield (Centre for Ecology and Hydrology)). Thanks to Laura Graham (University of Southampton) and Kevin Watts (Forest Research) for reviewing earlier drafts of the manuscript and providing useful feedback, from expert advice on lesser horseshoe ecology from Henry Schofield (Vincent Wildlife Trust), and for the advice from three anonymous reviewers.
Publisher Copyright:
© 2020, The Author(s).
Keywords:
Lesser horseshoe bat, MaxEnt, Multi-scale, Nested model, Rhinolophus hipposideros, Species distribution model
Identifiers
Local EPrints ID: 439151
URI: http://eprints.soton.ac.uk/id/eprint/439151
ISSN: 0921-2973
PURE UUID: f9f23115-3211-4929-8e9e-53e8b770a4f7
Catalogue record
Date deposited: 06 Apr 2020 16:30
Last modified: 16 Mar 2024 07:23
Export record
Altmetrics
Contributors
Author:
Chloe Bellamy
Author:
K.L. Boughey
Author:
Charlotte Hawkins
Author:
Sonia Reveley
Author:
Carol Williams
Author:
John Altringham
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics