No Cover Image

Journal article 277 views 156 downloads

Predicting shoreline changes using deep learning techniques with Bayesian optimisation

Tharindu Manamperi, Alma Rahat, Doug Pender, Demetra Cristaudo, Rob Lamb Orcid Logo, Harshinie Karunarathna Orcid Logo

Coastal Engineering, Volume: 203, Start page: 104856

Swansea University Authors: Tharindu Manamperi, Alma Rahat, Harshinie Karunarathna Orcid Logo

  • 70210.VoR.pdf

    PDF | Version of Record

    © 2025 The Authors. This is an open access article under the CC BY license.

    Download (18.36MB)

Check full text

DOI (Published version): 10.1016/j.coastaleng.2025.104856

Abstract

Accurate prediction of shoreline change is vital for effective coastal planning and management, especially under increasing climate variabilities. This study explores the applicability of deep learning (DL) techniques, particularly Long Short-Term Memory (LSTM) and Convolutional Neural Network-LSTM...

Full description

Published in: Coastal Engineering
ISSN: 0378-3839
Published: Elsevier BV 2026
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa70210
first_indexed 2025-08-21T11:41:42Z
last_indexed 2025-09-30T08:55:59Z
id cronfa70210
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2025-09-29T15:23:36.3311592</datestamp><bib-version>v2</bib-version><id>70210</id><entry>2025-08-21</entry><title>Predicting shoreline changes using deep learning techniques with Bayesian optimisation</title><swanseaauthors><author><sid>fc003f5b52ece11a6eb261f0fd12b112</sid><firstname>Tharindu</firstname><surname>Manamperi</surname><name>Tharindu Manamperi</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>6206f027aca1e3a5ff6b8cd224248bc2</sid><firstname>Alma</firstname><surname>Rahat</surname><name>Alma Rahat</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>0d3d327a240d49b53c78e02b7c00e625</sid><ORCID>0000-0002-9087-3811</ORCID><firstname>Harshinie</firstname><surname>Karunarathna</surname><name>Harshinie Karunarathna</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2025-08-21</date><abstract>Accurate prediction of shoreline change is vital for effective coastal planning and management, especially under increasing climate variabilities. This study explores the applicability of deep learning (DL) techniques, particularly Long Short-Term Memory (LSTM) and Convolutional Neural Network-LSTM (CNN-LSTM) models, for shoreline forecasting at monthly to inter-annual timescales, under two modelling approaches&#x2014;direct input (DI) and autoregressive (AR). All models demonstrated the ability to reproduce temporal shoreline variability, while the autoregressive DL models were performing better.Further, a noise impact assessment revealed that seasonal decomposition and noise filtering significantly enhanced the model performance. In particular, the models using 52-week data decomposition and residual noise reduction improved the model performance. The reduction of data noises also resulted in narrower ensemble prediction envelopes, indicating that ensemble candidate models behave with low diversity. The temporal data resolution analysis showed that lower data resolutions reduce the predictive performance of the model and at least fortnightly data are required to satisfactorily capture the trend of variability of the shoreline position at this beach.The use of ensemble predictions, derived from a selected subset of model trials based on their collective performance, proved beneficial by capturing diverse temporal behaviours, thereby offering a quasi-probabilistic forecast with minimal computational cost. Overall, the study underscores the potential of DL models, particularly with autoregressive architectures, for reliable and transferable shoreline change prediction. It also emphasizes the importance of data quality, resolution, and preprocessing in improving model robustness, laying the groundwork for future research into use of DL in multi-scale shoreline predictions.</abstract><type>Journal Article</type><journal>Coastal Engineering</journal><volume>203</volume><journalNumber/><paginationStart>104856</paginationStart><paginationEnd/><publisher>Elsevier BV</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>0378-3839</issnPrint><issnElectronic/><keywords>Shoreline prediction; Deep Learning; LSTM; Bayesian Optimisation</keywords><publishedDay>15</publishedDay><publishedMonth>1</publishedMonth><publishedYear>2026</publishedYear><publishedDate>2026-01-15</publishedDate><doi>10.1016/j.coastaleng.2025.104856</doi><url/><notes/><college>COLLEGE NANME</college><CollegeCode>COLLEGE CODE</CollegeCode><institution>Swansea University</institution><apcterm>SU Library paid the OA fee (TA Institutional Deal)</apcterm><funders>The Authors would like to acknowledge the JBA Trust (project No. W22-1128), UK &amp; Engineering and Physical Sciences Research Council (EPSRC) - Doctoral Training Partnerships (DTP) (EP/W524694/1) for funding the Doctoral Research study of the first author.</funders><projectreference/><lastEdited>2025-09-29T15:23:36.3311592</lastEdited><Created>2025-08-21T12:35:32.9951372</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Civil Engineering</level></path><authors><author><firstname>Tharindu</firstname><surname>Manamperi</surname><order>1</order></author><author><firstname>Alma</firstname><surname>Rahat</surname><order>2</order></author><author><firstname>Doug</firstname><surname>Pender</surname><order>3</order></author><author><firstname>Demetra</firstname><surname>Cristaudo</surname><order>4</order></author><author><firstname>Rob</firstname><surname>Lamb</surname><orcid>0000-0002-9593-621x</orcid><order>5</order></author><author><firstname>Harshinie</firstname><surname>Karunarathna</surname><orcid>0000-0002-9087-3811</orcid><order>6</order></author></authors><documents><document><filename>70210__35198__8ed141ed6a8e49f7861db72b15930a4a.pdf</filename><originalFilename>70210.VoR.pdf</originalFilename><uploaded>2025-09-29T15:21:00.2984071</uploaded><type>Output</type><contentLength>19247964</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>&#xA9; 2025 The Authors. This is an open access article under the CC BY license.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>http://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2025-09-29T15:23:36.3311592 v2 70210 2025-08-21 Predicting shoreline changes using deep learning techniques with Bayesian optimisation fc003f5b52ece11a6eb261f0fd12b112 Tharindu Manamperi Tharindu Manamperi true false 6206f027aca1e3a5ff6b8cd224248bc2 Alma Rahat Alma Rahat true false 0d3d327a240d49b53c78e02b7c00e625 0000-0002-9087-3811 Harshinie Karunarathna Harshinie Karunarathna true false 2025-08-21 Accurate prediction of shoreline change is vital for effective coastal planning and management, especially under increasing climate variabilities. This study explores the applicability of deep learning (DL) techniques, particularly Long Short-Term Memory (LSTM) and Convolutional Neural Network-LSTM (CNN-LSTM) models, for shoreline forecasting at monthly to inter-annual timescales, under two modelling approaches—direct input (DI) and autoregressive (AR). All models demonstrated the ability to reproduce temporal shoreline variability, while the autoregressive DL models were performing better.Further, a noise impact assessment revealed that seasonal decomposition and noise filtering significantly enhanced the model performance. In particular, the models using 52-week data decomposition and residual noise reduction improved the model performance. The reduction of data noises also resulted in narrower ensemble prediction envelopes, indicating that ensemble candidate models behave with low diversity. The temporal data resolution analysis showed that lower data resolutions reduce the predictive performance of the model and at least fortnightly data are required to satisfactorily capture the trend of variability of the shoreline position at this beach.The use of ensemble predictions, derived from a selected subset of model trials based on their collective performance, proved beneficial by capturing diverse temporal behaviours, thereby offering a quasi-probabilistic forecast with minimal computational cost. Overall, the study underscores the potential of DL models, particularly with autoregressive architectures, for reliable and transferable shoreline change prediction. It also emphasizes the importance of data quality, resolution, and preprocessing in improving model robustness, laying the groundwork for future research into use of DL in multi-scale shoreline predictions. Journal Article Coastal Engineering 203 104856 Elsevier BV 0378-3839 Shoreline prediction; Deep Learning; LSTM; Bayesian Optimisation 15 1 2026 2026-01-15 10.1016/j.coastaleng.2025.104856 COLLEGE NANME COLLEGE CODE Swansea University SU Library paid the OA fee (TA Institutional Deal) The Authors would like to acknowledge the JBA Trust (project No. W22-1128), UK & Engineering and Physical Sciences Research Council (EPSRC) - Doctoral Training Partnerships (DTP) (EP/W524694/1) for funding the Doctoral Research study of the first author. 2025-09-29T15:23:36.3311592 2025-08-21T12:35:32.9951372 Faculty of Science and Engineering School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Civil Engineering Tharindu Manamperi 1 Alma Rahat 2 Doug Pender 3 Demetra Cristaudo 4 Rob Lamb 0000-0002-9593-621x 5 Harshinie Karunarathna 0000-0002-9087-3811 6 70210__35198__8ed141ed6a8e49f7861db72b15930a4a.pdf 70210.VoR.pdf 2025-09-29T15:21:00.2984071 Output 19247964 application/pdf Version of Record true © 2025 The Authors. This is an open access article under the CC BY license. true eng http://creativecommons.org/licenses/by/4.0/
title Predicting shoreline changes using deep learning techniques with Bayesian optimisation
spellingShingle Predicting shoreline changes using deep learning techniques with Bayesian optimisation
Tharindu Manamperi
Alma Rahat
Harshinie Karunarathna
title_short Predicting shoreline changes using deep learning techniques with Bayesian optimisation
title_full Predicting shoreline changes using deep learning techniques with Bayesian optimisation
title_fullStr Predicting shoreline changes using deep learning techniques with Bayesian optimisation
title_full_unstemmed Predicting shoreline changes using deep learning techniques with Bayesian optimisation
title_sort Predicting shoreline changes using deep learning techniques with Bayesian optimisation
author_id_str_mv fc003f5b52ece11a6eb261f0fd12b112
6206f027aca1e3a5ff6b8cd224248bc2
0d3d327a240d49b53c78e02b7c00e625
author_id_fullname_str_mv fc003f5b52ece11a6eb261f0fd12b112_***_Tharindu Manamperi
6206f027aca1e3a5ff6b8cd224248bc2_***_Alma Rahat
0d3d327a240d49b53c78e02b7c00e625_***_Harshinie Karunarathna
author Tharindu Manamperi
Alma Rahat
Harshinie Karunarathna
author2 Tharindu Manamperi
Alma Rahat
Doug Pender
Demetra Cristaudo
Rob Lamb
Harshinie Karunarathna
format Journal article
container_title Coastal Engineering
container_volume 203
container_start_page 104856
publishDate 2026
institution Swansea University
issn 0378-3839
doi_str_mv 10.1016/j.coastaleng.2025.104856
publisher Elsevier BV
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Civil Engineering{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Civil Engineering
document_store_str 1
active_str 0
description Accurate prediction of shoreline change is vital for effective coastal planning and management, especially under increasing climate variabilities. This study explores the applicability of deep learning (DL) techniques, particularly Long Short-Term Memory (LSTM) and Convolutional Neural Network-LSTM (CNN-LSTM) models, for shoreline forecasting at monthly to inter-annual timescales, under two modelling approaches—direct input (DI) and autoregressive (AR). All models demonstrated the ability to reproduce temporal shoreline variability, while the autoregressive DL models were performing better.Further, a noise impact assessment revealed that seasonal decomposition and noise filtering significantly enhanced the model performance. In particular, the models using 52-week data decomposition and residual noise reduction improved the model performance. The reduction of data noises also resulted in narrower ensemble prediction envelopes, indicating that ensemble candidate models behave with low diversity. The temporal data resolution analysis showed that lower data resolutions reduce the predictive performance of the model and at least fortnightly data are required to satisfactorily capture the trend of variability of the shoreline position at this beach.The use of ensemble predictions, derived from a selected subset of model trials based on their collective performance, proved beneficial by capturing diverse temporal behaviours, thereby offering a quasi-probabilistic forecast with minimal computational cost. Overall, the study underscores the potential of DL models, particularly with autoregressive architectures, for reliable and transferable shoreline change prediction. It also emphasizes the importance of data quality, resolution, and preprocessing in improving model robustness, laying the groundwork for future research into use of DL in multi-scale shoreline predictions.
published_date 2026-01-15T05:32:01Z
_version_ 1856986909608247296
score 11.096295

Similar Items