Assessing the Potential to Operationalize Shoreline Sensitivity Mapping: Classifying Multiple Wide Fine Quadrature Polarized RADARSAT-2 and Landsat 5 Scenes with a Single Random Forest Model
The Random Forest algorithm was used to classify 86 Wide Fine Quadrature Polarized RADARSAT-2 scenes, five Landsat 5 scenes, and a Digital Elevation Model covering an area approximately 81,000 km2 in size, and representing the entirety of Dease Strait, Coronation Gulf and Bathurst Inlet, Nunavut. Th...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , , , |
Format: | Text |
Language: | English |
Published: |
Multidisciplinary Digital Publishing Institute
2015
|
Subjects: | |
Online Access: | https://doi.org/10.3390/rs71013528 |
id |
ftmdpi:oai:mdpi.com:/2072-4292/7/10/13528/ |
---|---|
record_format |
openpolar |
spelling |
ftmdpi:oai:mdpi.com:/2072-4292/7/10/13528/ 2023-08-20T04:04:39+02:00 Assessing the Potential to Operationalize Shoreline Sensitivity Mapping: Classifying Multiple Wide Fine Quadrature Polarized RADARSAT-2 and Landsat 5 Scenes with a Single Random Forest Model Sarah Banks Koreen Millard Jon Pasher Murray Richardson Huili Wang Jason Duffe agris 2015-10-19 application/pdf https://doi.org/10.3390/rs71013528 EN eng Multidisciplinary Digital Publishing Institute https://dx.doi.org/10.3390/rs71013528 https://creativecommons.org/licenses/by/4.0/ Remote Sensing; Volume 7; Issue 10; Pages: 13528-13563 RADARSAT-2 Landsat 5 classification Random Forest Arctic shorelines Text 2015 ftmdpi https://doi.org/10.3390/rs71013528 2023-07-31T20:47:16Z The Random Forest algorithm was used to classify 86 Wide Fine Quadrature Polarized RADARSAT-2 scenes, five Landsat 5 scenes, and a Digital Elevation Model covering an area approximately 81,000 km2 in size, and representing the entirety of Dease Strait, Coronation Gulf and Bathurst Inlet, Nunavut. The focus of this research was to assess the potential to operationalize shoreline sensitivity mapping to inform oil spill response and contingency planning. The impact of varying the training sample size and reducing model data load were evaluated. Results showed that acceptable accuracies could be achieved with relatively few training samples, but that higher accuracies and greater probabilities of correct class assignment were observed with larger sample sizes. Additionally, the number of inputs to the model could be greatly reduced without impacting overall performance. Optimized models reached independent accuracies of 91% for seven land cover types, and classification probabilities between 0.77 and 0.98 (values for latter represent per-class averages generated from independent validation sites). Mixed results were observed when assessing the potential for remote predictive mapping by simulating transferability of the model to scenes without training data. Text Arctic Bathurst Inlet Coronation Gulf Nunavut MDPI Open Access Publishing Arctic Bathurst Inlet ENVELOPE(-108.051,-108.051,66.840,66.840) Coronation Gulf ENVELOPE(-112.003,-112.003,68.134,68.134) Dease Strait ENVELOPE(-107.502,-107.502,68.834,68.834) Nunavut Remote Sensing 7 10 13528 13563 |
institution |
Open Polar |
collection |
MDPI Open Access Publishing |
op_collection_id |
ftmdpi |
language |
English |
topic |
RADARSAT-2 Landsat 5 classification Random Forest Arctic shorelines |
spellingShingle |
RADARSAT-2 Landsat 5 classification Random Forest Arctic shorelines Sarah Banks Koreen Millard Jon Pasher Murray Richardson Huili Wang Jason Duffe Assessing the Potential to Operationalize Shoreline Sensitivity Mapping: Classifying Multiple Wide Fine Quadrature Polarized RADARSAT-2 and Landsat 5 Scenes with a Single Random Forest Model |
topic_facet |
RADARSAT-2 Landsat 5 classification Random Forest Arctic shorelines |
description |
The Random Forest algorithm was used to classify 86 Wide Fine Quadrature Polarized RADARSAT-2 scenes, five Landsat 5 scenes, and a Digital Elevation Model covering an area approximately 81,000 km2 in size, and representing the entirety of Dease Strait, Coronation Gulf and Bathurst Inlet, Nunavut. The focus of this research was to assess the potential to operationalize shoreline sensitivity mapping to inform oil spill response and contingency planning. The impact of varying the training sample size and reducing model data load were evaluated. Results showed that acceptable accuracies could be achieved with relatively few training samples, but that higher accuracies and greater probabilities of correct class assignment were observed with larger sample sizes. Additionally, the number of inputs to the model could be greatly reduced without impacting overall performance. Optimized models reached independent accuracies of 91% for seven land cover types, and classification probabilities between 0.77 and 0.98 (values for latter represent per-class averages generated from independent validation sites). Mixed results were observed when assessing the potential for remote predictive mapping by simulating transferability of the model to scenes without training data. |
format |
Text |
author |
Sarah Banks Koreen Millard Jon Pasher Murray Richardson Huili Wang Jason Duffe |
author_facet |
Sarah Banks Koreen Millard Jon Pasher Murray Richardson Huili Wang Jason Duffe |
author_sort |
Sarah Banks |
title |
Assessing the Potential to Operationalize Shoreline Sensitivity Mapping: Classifying Multiple Wide Fine Quadrature Polarized RADARSAT-2 and Landsat 5 Scenes with a Single Random Forest Model |
title_short |
Assessing the Potential to Operationalize Shoreline Sensitivity Mapping: Classifying Multiple Wide Fine Quadrature Polarized RADARSAT-2 and Landsat 5 Scenes with a Single Random Forest Model |
title_full |
Assessing the Potential to Operationalize Shoreline Sensitivity Mapping: Classifying Multiple Wide Fine Quadrature Polarized RADARSAT-2 and Landsat 5 Scenes with a Single Random Forest Model |
title_fullStr |
Assessing the Potential to Operationalize Shoreline Sensitivity Mapping: Classifying Multiple Wide Fine Quadrature Polarized RADARSAT-2 and Landsat 5 Scenes with a Single Random Forest Model |
title_full_unstemmed |
Assessing the Potential to Operationalize Shoreline Sensitivity Mapping: Classifying Multiple Wide Fine Quadrature Polarized RADARSAT-2 and Landsat 5 Scenes with a Single Random Forest Model |
title_sort |
assessing the potential to operationalize shoreline sensitivity mapping: classifying multiple wide fine quadrature polarized radarsat-2 and landsat 5 scenes with a single random forest model |
publisher |
Multidisciplinary Digital Publishing Institute |
publishDate |
2015 |
url |
https://doi.org/10.3390/rs71013528 |
op_coverage |
agris |
long_lat |
ENVELOPE(-108.051,-108.051,66.840,66.840) ENVELOPE(-112.003,-112.003,68.134,68.134) ENVELOPE(-107.502,-107.502,68.834,68.834) |
geographic |
Arctic Bathurst Inlet Coronation Gulf Dease Strait Nunavut |
geographic_facet |
Arctic Bathurst Inlet Coronation Gulf Dease Strait Nunavut |
genre |
Arctic Bathurst Inlet Coronation Gulf Nunavut |
genre_facet |
Arctic Bathurst Inlet Coronation Gulf Nunavut |
op_source |
Remote Sensing; Volume 7; Issue 10; Pages: 13528-13563 |
op_relation |
https://dx.doi.org/10.3390/rs71013528 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3390/rs71013528 |
container_title |
Remote Sensing |
container_volume |
7 |
container_issue |
10 |
container_start_page |
13528 |
op_container_end_page |
13563 |
_version_ |
1774715025910398976 |