How Do Continuous High-Resolution Models of Patchy Seabed Habitats Enhance Classification Schemes?
Predefined classification schemes and fixed geographic scales are often used to simplify and cost-effectively map the spatial complexity of nature. These simplifications can however limit the usefulness of the mapping effort for users who need information across a different range of thematic and spa...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2019-05-01
|
Series: | Geosciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3263/9/5/237 |
id |
doaj-f64af74322d54d69a365374a1225eafc |
---|---|
record_format |
Article |
spelling |
doaj-f64af74322d54d69a365374a1225eafc2020-11-25T01:30:20ZengMDPI AGGeosciences2076-32632019-05-019523710.3390/geosciences9050237geosciences9050237How Do Continuous High-Resolution Models of Patchy Seabed Habitats Enhance Classification Schemes?Gustav Kågesten0Dario Fiorentino1Finn Baumgartner2Lovisa Zillén3Geological Survey of Sweden, 752 36 Uppsala, SwedenAlfred Wegener Institute Helmholtz Centre for Polar and Marine Research, 27570 Bremerhaven, GermanyGeological Survey of Sweden, 752 36 Uppsala, SwedenGeological Survey of Sweden, 752 36 Uppsala, SwedenPredefined classification schemes and fixed geographic scales are often used to simplify and cost-effectively map the spatial complexity of nature. These simplifications can however limit the usefulness of the mapping effort for users who need information across a different range of thematic and spatial resolutions. We demonstrate how substrate and biological information from point samples and photos, combined with continuous multibeam data, can be modeled to predictively map percentage cover conforming with multiple existing classification schemes (i.e., HELCOM HUB; Natura 2000), while also providing high-resolution (5 m) maps of individual substrate and biological components across a 1344 km<sup>2</sup> offshore bank in the Baltic Sea. Data for substrate and epibenthic organisms were obtained from high-resolution photo mosaics, sediment grab samples, legacy data and expert annotations. Environmental variables included pixel and object based metrics at multiple scales (0.5 m−2 km), which improved the accuracy of models. We found that using Boosted Regression Trees (BRTs) to predict continuous models of substrate and biological components provided additional detail for each component without losing accuracy in the classified maps, compared with a thematic model. Results demonstrate the sensitivity of habitat maps to the effects of spatial and thematic resolution and the importance of high-resolution maps to management applications.https://www.mdpi.com/2076-3263/9/5/237habitat mappingHELCOM HUBNatura 2000Baltic Seaspatial scaleseascapemachine learningboosted regression treespercent coversubstratebiotamultibeambackscatter |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Gustav Kågesten Dario Fiorentino Finn Baumgartner Lovisa Zillén |
spellingShingle |
Gustav Kågesten Dario Fiorentino Finn Baumgartner Lovisa Zillén How Do Continuous High-Resolution Models of Patchy Seabed Habitats Enhance Classification Schemes? Geosciences habitat mapping HELCOM HUB Natura 2000 Baltic Sea spatial scale seascape machine learning boosted regression trees percent cover substrate biota multibeam backscatter |
author_facet |
Gustav Kågesten Dario Fiorentino Finn Baumgartner Lovisa Zillén |
author_sort |
Gustav Kågesten |
title |
How Do Continuous High-Resolution Models of Patchy Seabed Habitats Enhance Classification Schemes? |
title_short |
How Do Continuous High-Resolution Models of Patchy Seabed Habitats Enhance Classification Schemes? |
title_full |
How Do Continuous High-Resolution Models of Patchy Seabed Habitats Enhance Classification Schemes? |
title_fullStr |
How Do Continuous High-Resolution Models of Patchy Seabed Habitats Enhance Classification Schemes? |
title_full_unstemmed |
How Do Continuous High-Resolution Models of Patchy Seabed Habitats Enhance Classification Schemes? |
title_sort |
how do continuous high-resolution models of patchy seabed habitats enhance classification schemes? |
publisher |
MDPI AG |
series |
Geosciences |
issn |
2076-3263 |
publishDate |
2019-05-01 |
description |
Predefined classification schemes and fixed geographic scales are often used to simplify and cost-effectively map the spatial complexity of nature. These simplifications can however limit the usefulness of the mapping effort for users who need information across a different range of thematic and spatial resolutions. We demonstrate how substrate and biological information from point samples and photos, combined with continuous multibeam data, can be modeled to predictively map percentage cover conforming with multiple existing classification schemes (i.e., HELCOM HUB; Natura 2000), while also providing high-resolution (5 m) maps of individual substrate and biological components across a 1344 km<sup>2</sup> offshore bank in the Baltic Sea. Data for substrate and epibenthic organisms were obtained from high-resolution photo mosaics, sediment grab samples, legacy data and expert annotations. Environmental variables included pixel and object based metrics at multiple scales (0.5 m−2 km), which improved the accuracy of models. We found that using Boosted Regression Trees (BRTs) to predict continuous models of substrate and biological components provided additional detail for each component without losing accuracy in the classified maps, compared with a thematic model. Results demonstrate the sensitivity of habitat maps to the effects of spatial and thematic resolution and the importance of high-resolution maps to management applications. |
topic |
habitat mapping HELCOM HUB Natura 2000 Baltic Sea spatial scale seascape machine learning boosted regression trees percent cover substrate biota multibeam backscatter |
url |
https://www.mdpi.com/2076-3263/9/5/237 |
work_keys_str_mv |
AT gustavkagesten howdocontinuoushighresolutionmodelsofpatchyseabedhabitatsenhanceclassificationschemes AT dariofiorentino howdocontinuoushighresolutionmodelsofpatchyseabedhabitatsenhanceclassificationschemes AT finnbaumgartner howdocontinuoushighresolutionmodelsofpatchyseabedhabitatsenhanceclassificationschemes AT lovisazillen howdocontinuoushighresolutionmodelsofpatchyseabedhabitatsenhanceclassificationschemes |
_version_ |
1725092049416355840 |