Identifying and interpreting extreme rainfall events using image classification
This study presents the first attempt to identify extreme rainfall events based on surrounding sea-level pressure anomalies, using neural network-based classification. Sensitivity analysis was also performed to identify the spatial importance of sea-level pressure anomalies. Three classification mod...
Published in: | Journal of Hydroinformatics |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
IWA Publishing
2021
|
Subjects: | |
Online Access: | https://doi.org/10.2166/hydro.2021.030 https://doaj.org/article/e1b57bc61a6d4a868751b2fcc42d856b |
id |
ftdoajarticles:oai:doaj.org/article:e1b57bc61a6d4a868751b2fcc42d856b |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:e1b57bc61a6d4a868751b2fcc42d856b 2023-05-15T17:29:13+02:00 Identifying and interpreting extreme rainfall events using image classification Andrew Paul Barnes Nick McCullen Thomas Rodding Kjeldsen 2021-11-01T00:00:00Z https://doi.org/10.2166/hydro.2021.030 https://doaj.org/article/e1b57bc61a6d4a868751b2fcc42d856b EN eng IWA Publishing http://jh.iwaponline.com/content/23/6/1214 https://doaj.org/toc/1464-7141 https://doaj.org/toc/1465-1734 1464-7141 1465-1734 doi:10.2166/hydro.2021.030 https://doaj.org/article/e1b57bc61a6d4a868751b2fcc42d856b Journal of Hydroinformatics, Vol 23, Iss 6, Pp 1214-1223 (2021) classification extreme events image classification rainfall extremes sea-level pressure Information technology T58.5-58.64 Environmental technology. Sanitary engineering TD1-1066 article 2021 ftdoajarticles https://doi.org/10.2166/hydro.2021.030 2022-12-31T10:33:58Z This study presents the first attempt to identify extreme rainfall events based on surrounding sea-level pressure anomalies, using neural network-based classification. Sensitivity analysis was also performed to identify the spatial importance of sea-level pressure anomalies. Three classification models were generated: the first classifies the patterns between extreme and regular rainfall events in the North West of England, the second classifies the patterns between extreme and regular rainfall events in the South East of England, and the third classifies between the patterns of extreme events in the North West and South East of England. All classifiers obtain accuracies between 60 and 65%, with precision and recall metrics showing that extreme events are easier to identify than regular events. Finally, a sensitivity analysis is performed to identify the spatial importance of the patterns across the North Atlantic, highlighting that for all three classifiers the local anomaly sea-level pressure patterns around the British Isles are key to determining the difference between extreme and regular rainfall events. In contrast, the pattern across the mid and western North Atlantic shows no contribution to the overall classifications. HIGHLIGHTS Neural networks can distinguish between extreme and regular rainfall events.; The sea-level pressure surrounding the UK is key to distinguishing extreme events.; The western North Atlantic does not contribute to classifying extreme events.; Article in Journal/Newspaper North Atlantic Directory of Open Access Journals: DOAJ Articles Journal of Hydroinformatics 23 6 1214 1223 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
classification extreme events image classification rainfall extremes sea-level pressure Information technology T58.5-58.64 Environmental technology. Sanitary engineering TD1-1066 |
spellingShingle |
classification extreme events image classification rainfall extremes sea-level pressure Information technology T58.5-58.64 Environmental technology. Sanitary engineering TD1-1066 Andrew Paul Barnes Nick McCullen Thomas Rodding Kjeldsen Identifying and interpreting extreme rainfall events using image classification |
topic_facet |
classification extreme events image classification rainfall extremes sea-level pressure Information technology T58.5-58.64 Environmental technology. Sanitary engineering TD1-1066 |
description |
This study presents the first attempt to identify extreme rainfall events based on surrounding sea-level pressure anomalies, using neural network-based classification. Sensitivity analysis was also performed to identify the spatial importance of sea-level pressure anomalies. Three classification models were generated: the first classifies the patterns between extreme and regular rainfall events in the North West of England, the second classifies the patterns between extreme and regular rainfall events in the South East of England, and the third classifies between the patterns of extreme events in the North West and South East of England. All classifiers obtain accuracies between 60 and 65%, with precision and recall metrics showing that extreme events are easier to identify than regular events. Finally, a sensitivity analysis is performed to identify the spatial importance of the patterns across the North Atlantic, highlighting that for all three classifiers the local anomaly sea-level pressure patterns around the British Isles are key to determining the difference between extreme and regular rainfall events. In contrast, the pattern across the mid and western North Atlantic shows no contribution to the overall classifications. HIGHLIGHTS Neural networks can distinguish between extreme and regular rainfall events.; The sea-level pressure surrounding the UK is key to distinguishing extreme events.; The western North Atlantic does not contribute to classifying extreme events.; |
format |
Article in Journal/Newspaper |
author |
Andrew Paul Barnes Nick McCullen Thomas Rodding Kjeldsen |
author_facet |
Andrew Paul Barnes Nick McCullen Thomas Rodding Kjeldsen |
author_sort |
Andrew Paul Barnes |
title |
Identifying and interpreting extreme rainfall events using image classification |
title_short |
Identifying and interpreting extreme rainfall events using image classification |
title_full |
Identifying and interpreting extreme rainfall events using image classification |
title_fullStr |
Identifying and interpreting extreme rainfall events using image classification |
title_full_unstemmed |
Identifying and interpreting extreme rainfall events using image classification |
title_sort |
identifying and interpreting extreme rainfall events using image classification |
publisher |
IWA Publishing |
publishDate |
2021 |
url |
https://doi.org/10.2166/hydro.2021.030 https://doaj.org/article/e1b57bc61a6d4a868751b2fcc42d856b |
genre |
North Atlantic |
genre_facet |
North Atlantic |
op_source |
Journal of Hydroinformatics, Vol 23, Iss 6, Pp 1214-1223 (2021) |
op_relation |
http://jh.iwaponline.com/content/23/6/1214 https://doaj.org/toc/1464-7141 https://doaj.org/toc/1465-1734 1464-7141 1465-1734 doi:10.2166/hydro.2021.030 https://doaj.org/article/e1b57bc61a6d4a868751b2fcc42d856b |
op_doi |
https://doi.org/10.2166/hydro.2021.030 |
container_title |
Journal of Hydroinformatics |
container_volume |
23 |
container_issue |
6 |
container_start_page |
1214 |
op_container_end_page |
1223 |
_version_ |
1766122872973033472 |