A Spatially Explicit Comparison of Quantitative and Categorical Modelling Approaches for Mapping Seabed Sediments Using Random Forest
Seabed sediment composition is an important component of benthic habitat and there are many approaches for producing maps that convey sediment information to marine managers. Random Forest is a popular statistical method for thematic seabed sediment mapping using both categorical and quantitative su...
Published in: | Geosciences |
---|---|
Main Authors: | , , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
MDPI AG
2019
|
Subjects: | |
Online Access: | https://doi.org/10.3390/geosciences9060254 https://doaj.org/article/5a78589d878143f0a8deec40ec24451f |
id |
ftdoajarticles:oai:doaj.org/article:5a78589d878143f0a8deec40ec24451f |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:5a78589d878143f0a8deec40ec24451f 2023-05-15T16:18:30+02:00 A Spatially Explicit Comparison of Quantitative and Categorical Modelling Approaches for Mapping Seabed Sediments Using Random Forest Benjamin Misiuk Markus Diesing Alec Aitken Craig J. Brown Evan N. Edinger Trevor Bell 2019-06-01T00:00:00Z https://doi.org/10.3390/geosciences9060254 https://doaj.org/article/5a78589d878143f0a8deec40ec24451f EN eng MDPI AG https://www.mdpi.com/2076-3263/9/6/254 https://doaj.org/toc/2076-3263 2076-3263 doi:10.3390/geosciences9060254 https://doaj.org/article/5a78589d878143f0a8deec40ec24451f Geosciences, Vol 9, Iss 6, p 254 (2019) marine habitat mapping benthic habitat mapping grain size modelling spatial autocorrelation multiscale marine geology Geology QE1-996.5 article 2019 ftdoajarticles https://doi.org/10.3390/geosciences9060254 2022-12-31T11:15:19Z Seabed sediment composition is an important component of benthic habitat and there are many approaches for producing maps that convey sediment information to marine managers. Random Forest is a popular statistical method for thematic seabed sediment mapping using both categorical and quantitative supervised modelling approaches. This study compares the performance and qualities of these Random Forest approaches to predict the distribution of fine-grained sediments from grab samples as one component of a multi-model map of sediment classes in Frobisher Bay, Nunavut, Canada. The second component predicts the presence of coarse substrates from underwater video. Spatial and non-spatial cross-validations were conducted to evaluate the performance of categorical and quantitative Random Forest models and maps were compared to determine differences in predictions. While both approaches seemed highly accurate, the non-spatial cross-validation suggested greater accuracy using the categorical approach. Using a spatial cross-validation, there was little difference between approaches—both showed poor extrapolative performance. Spatial cross-validation methods also suggested evidence of overfitting in the coarse sediment model caused by the spatial dependence of transect samples. The quantitative modelling approach was able to predict rare and unsampled sediment classes but the flexibility of probabilistic predictions from the categorical approach allowed for tuning to maximize extrapolative performance. Results demonstrate that the apparent accuracies of these models failed to convey important differences between map predictions and that spatially explicit evaluation strategies may be necessary for evaluating extrapolative performance. Differentiating extrapolative from interpolative prediction can aid in selecting appropriate modelling methods. Article in Journal/Newspaper Frobisher Bay Nunavut Directory of Open Access Journals: DOAJ Articles Canada Frobisher Bay ENVELOPE(-66.581,-66.581,62.834,62.834) Nunavut Geosciences 9 6 254 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
marine habitat mapping benthic habitat mapping grain size modelling spatial autocorrelation multiscale marine geology Geology QE1-996.5 |
spellingShingle |
marine habitat mapping benthic habitat mapping grain size modelling spatial autocorrelation multiscale marine geology Geology QE1-996.5 Benjamin Misiuk Markus Diesing Alec Aitken Craig J. Brown Evan N. Edinger Trevor Bell A Spatially Explicit Comparison of Quantitative and Categorical Modelling Approaches for Mapping Seabed Sediments Using Random Forest |
topic_facet |
marine habitat mapping benthic habitat mapping grain size modelling spatial autocorrelation multiscale marine geology Geology QE1-996.5 |
description |
Seabed sediment composition is an important component of benthic habitat and there are many approaches for producing maps that convey sediment information to marine managers. Random Forest is a popular statistical method for thematic seabed sediment mapping using both categorical and quantitative supervised modelling approaches. This study compares the performance and qualities of these Random Forest approaches to predict the distribution of fine-grained sediments from grab samples as one component of a multi-model map of sediment classes in Frobisher Bay, Nunavut, Canada. The second component predicts the presence of coarse substrates from underwater video. Spatial and non-spatial cross-validations were conducted to evaluate the performance of categorical and quantitative Random Forest models and maps were compared to determine differences in predictions. While both approaches seemed highly accurate, the non-spatial cross-validation suggested greater accuracy using the categorical approach. Using a spatial cross-validation, there was little difference between approaches—both showed poor extrapolative performance. Spatial cross-validation methods also suggested evidence of overfitting in the coarse sediment model caused by the spatial dependence of transect samples. The quantitative modelling approach was able to predict rare and unsampled sediment classes but the flexibility of probabilistic predictions from the categorical approach allowed for tuning to maximize extrapolative performance. Results demonstrate that the apparent accuracies of these models failed to convey important differences between map predictions and that spatially explicit evaluation strategies may be necessary for evaluating extrapolative performance. Differentiating extrapolative from interpolative prediction can aid in selecting appropriate modelling methods. |
format |
Article in Journal/Newspaper |
author |
Benjamin Misiuk Markus Diesing Alec Aitken Craig J. Brown Evan N. Edinger Trevor Bell |
author_facet |
Benjamin Misiuk Markus Diesing Alec Aitken Craig J. Brown Evan N. Edinger Trevor Bell |
author_sort |
Benjamin Misiuk |
title |
A Spatially Explicit Comparison of Quantitative and Categorical Modelling Approaches for Mapping Seabed Sediments Using Random Forest |
title_short |
A Spatially Explicit Comparison of Quantitative and Categorical Modelling Approaches for Mapping Seabed Sediments Using Random Forest |
title_full |
A Spatially Explicit Comparison of Quantitative and Categorical Modelling Approaches for Mapping Seabed Sediments Using Random Forest |
title_fullStr |
A Spatially Explicit Comparison of Quantitative and Categorical Modelling Approaches for Mapping Seabed Sediments Using Random Forest |
title_full_unstemmed |
A Spatially Explicit Comparison of Quantitative and Categorical Modelling Approaches for Mapping Seabed Sediments Using Random Forest |
title_sort |
spatially explicit comparison of quantitative and categorical modelling approaches for mapping seabed sediments using random forest |
publisher |
MDPI AG |
publishDate |
2019 |
url |
https://doi.org/10.3390/geosciences9060254 https://doaj.org/article/5a78589d878143f0a8deec40ec24451f |
long_lat |
ENVELOPE(-66.581,-66.581,62.834,62.834) |
geographic |
Canada Frobisher Bay Nunavut |
geographic_facet |
Canada Frobisher Bay Nunavut |
genre |
Frobisher Bay Nunavut |
genre_facet |
Frobisher Bay Nunavut |
op_source |
Geosciences, Vol 9, Iss 6, p 254 (2019) |
op_relation |
https://www.mdpi.com/2076-3263/9/6/254 https://doaj.org/toc/2076-3263 2076-3263 doi:10.3390/geosciences9060254 https://doaj.org/article/5a78589d878143f0a8deec40ec24451f |
op_doi |
https://doi.org/10.3390/geosciences9060254 |
container_title |
Geosciences |
container_volume |
9 |
container_issue |
6 |
container_start_page |
254 |
_version_ |
1766004687317762048 |