Fast high-fidelity flood inundation map generation by super-resolution techniques
Flooding is one of the most frequent natural hazards and causes more economic loss than all the other natural hazards. Fast and accurate flood prediction has significance in preserving lives, minimizing economic damage, and reducing public health risks. However, current methods cannot achieve speed...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IWA Publishing
2024-01-01
|
| Series: | Journal of Hydroinformatics |
| Subjects: | |
| Online Access: | http://jhydro.iwaponline.com/content/26/1/319 |
| _version_ | 1797205100793430016 |
|---|---|
| author | Zeda Yin Yasaman Saadati Beichao Hu Arturo S. Leon M. Hadi Amini Dwayne McDaniel |
| author_facet | Zeda Yin Yasaman Saadati Beichao Hu Arturo S. Leon M. Hadi Amini Dwayne McDaniel |
| author_sort | Zeda Yin |
| collection | DOAJ |
| description | Flooding is one of the most frequent natural hazards and causes more economic loss than all the other natural hazards. Fast and accurate flood prediction has significance in preserving lives, minimizing economic damage, and reducing public health risks. However, current methods cannot achieve speed and accuracy simultaneously. Numerical methods can provide high-fidelity results, but they are time-consuming, particularly when pursuing high accuracy. Conversely, neural networks can provide results in a matter of seconds, but they have shown low accuracy in flood map generation by all existing methods. This work combines the strengths of numerical methods and neural networks and builds a framework that can quickly and accurately model the high-fidelity flood inundation map with detailed water depth information. In this paper, we employ the U-Net and generative adversarial network (GAN) models to recover the lost physics and information from ultra-fast, low-resolution numerical simulations, ultimately presenting high-resolution, high-fidelity flood maps as the end results. In this study, both the U-Net and GAN models have proven their ability to reduce the computation time for generating high-fidelity results, reducing it from 7–8 h down to 1 min. Furthermore, the accuracy of both models is notably high.
HIGHLIGHTS
In our study area, our models have demonstrated the capability to dramatically decrease the computation time required to generate high-fidelity results, reducing it from 7–8 h to 1 min.;
The GAN model displays a lower sensitivity to changes in input resolution compared with the U-Net model.;
The proposed method effectively recovers lost information because of the large grid size in the low-resolution geometry.; |
| first_indexed | 2024-03-08T03:18:00Z |
| format | Article |
| id | doaj.art-c79c5ee877b44818b01eebadadd1cfb5 |
| institution | Directory Open Access Journal |
| issn | 1464-7141 1465-1734 |
| language | English |
| last_indexed | 2024-04-24T08:45:45Z |
| publishDate | 2024-01-01 |
| publisher | IWA Publishing |
| record_format | Article |
| series | Journal of Hydroinformatics |
| spelling | doaj.art-c79c5ee877b44818b01eebadadd1cfb52024-04-16T13:35:26ZengIWA PublishingJournal of Hydroinformatics1464-71411465-17342024-01-0126131933610.2166/hydro.2024.228228Fast high-fidelity flood inundation map generation by super-resolution techniquesZeda Yin0Yasaman Saadati1Beichao Hu2Arturo S. Leon3M. Hadi Amini4Dwayne McDaniel5 Department of Civil and Environmental Engineering, College of Engineering and Computing, Florida International University, Miami, Florida, USA Knight Foundation School of Computing and Information Sciences, College of Engineering and Computing, Florida International University, Miami, Florida, USA Department of Mechanical and Materials Engineering, College of Engineering and Computing, Florida International University, Miami, Florida, USA Department of Civil and Environmental Engineering, College of Engineering and Computing, Florida International University, Miami, Florida, USA Knight Foundation School of Computing and Information Sciences, College of Engineering and Computing, Florida International University, Miami, Florida, USA Department of Mechanical and Materials Engineering, College of Engineering and Computing, Florida International University, Miami, Florida, USA Flooding is one of the most frequent natural hazards and causes more economic loss than all the other natural hazards. Fast and accurate flood prediction has significance in preserving lives, minimizing economic damage, and reducing public health risks. However, current methods cannot achieve speed and accuracy simultaneously. Numerical methods can provide high-fidelity results, but they are time-consuming, particularly when pursuing high accuracy. Conversely, neural networks can provide results in a matter of seconds, but they have shown low accuracy in flood map generation by all existing methods. This work combines the strengths of numerical methods and neural networks and builds a framework that can quickly and accurately model the high-fidelity flood inundation map with detailed water depth information. In this paper, we employ the U-Net and generative adversarial network (GAN) models to recover the lost physics and information from ultra-fast, low-resolution numerical simulations, ultimately presenting high-resolution, high-fidelity flood maps as the end results. In this study, both the U-Net and GAN models have proven their ability to reduce the computation time for generating high-fidelity results, reducing it from 7–8 h down to 1 min. Furthermore, the accuracy of both models is notably high. HIGHLIGHTS In our study area, our models have demonstrated the capability to dramatically decrease the computation time required to generate high-fidelity results, reducing it from 7–8 h to 1 min.; The GAN model displays a lower sensitivity to changes in input resolution compared with the U-Net model.; The proposed method effectively recovers lost information because of the large grid size in the low-resolution geometry.;http://jhydro.iwaponline.com/content/26/1/319flood inundation mapflood predictionneural networks |
| spellingShingle | Zeda Yin Yasaman Saadati Beichao Hu Arturo S. Leon M. Hadi Amini Dwayne McDaniel Fast high-fidelity flood inundation map generation by super-resolution techniques Journal of Hydroinformatics flood inundation map flood prediction neural networks |
| title | Fast high-fidelity flood inundation map generation by super-resolution techniques |
| title_full | Fast high-fidelity flood inundation map generation by super-resolution techniques |
| title_fullStr | Fast high-fidelity flood inundation map generation by super-resolution techniques |
| title_full_unstemmed | Fast high-fidelity flood inundation map generation by super-resolution techniques |
| title_short | Fast high-fidelity flood inundation map generation by super-resolution techniques |
| title_sort | fast high fidelity flood inundation map generation by super resolution techniques |
| topic | flood inundation map flood prediction neural networks |
| url | http://jhydro.iwaponline.com/content/26/1/319 |
| work_keys_str_mv | AT zedayin fasthighfidelityfloodinundationmapgenerationbysuperresolutiontechniques AT yasamansaadati fasthighfidelityfloodinundationmapgenerationbysuperresolutiontechniques AT beichaohu fasthighfidelityfloodinundationmapgenerationbysuperresolutiontechniques AT arturosleon fasthighfidelityfloodinundationmapgenerationbysuperresolutiontechniques AT mhadiamini fasthighfidelityfloodinundationmapgenerationbysuperresolutiontechniques AT dwaynemcdaniel fasthighfidelityfloodinundationmapgenerationbysuperresolutiontechniques |