Preprints
https://doi.org/10.5194/egusphere-2024-551
https://doi.org/10.5194/egusphere-2024-551
15 Mar 2024
 | 15 Mar 2024

Structured exploration of machine learning model complexity for spatio-temporal forecasting of urban flooding

Candace Agonafir and Tian Zheng

Abstract. Urban flooding may lead to significant socio-economic impacts and loss of life. To afford preventative actions, researchers have implemented various modeling techniques to gain insight into urban flood occurrences. Using New York City (NYC) as the study area, data-driven techniques, specifically statistical and neural network models with increasing spatio-temporal complexity, are formulated and tested, assessing the potential relative contribution of different modeling constructs. Zones, based on flood characteristics, are first delineated using the unsupervised machine learning technique of spectral clustering. Then, the models are applied to each cluster, with comprehensive performance evaluation, as to understand which algorithmic, structural aspects contribute to the reduction of prediction errors. A chief discovery of this study is the emergence of the Graph Wavenet (GWN) as the most effective model due to its proficiency in capturing spatio-temporal aspects and implementing dynamic graph creation. Furthermore, it is seen that the enhancement of specific temporal and spatial components within a modeling technique proves beneficial, and a novel adoption of graph-based architectures is additive. Offering a unique exploration of spatio-temporal aspects, emphasizing the benefits of component enhancement and the adoption of graph-based architectures, this paper identifies modification techniques, which would allow for insights to prevail in urban flood modeling despite being confronted with limited data availability.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.
Candace Agonafir and Tian Zheng

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2024-551', Anonymous Referee #1, 14 Apr 2024
    • AC1: 'Reply on RC1', Candace Agonafir, 22 Apr 2024
      • AC2: 'Reply on AC1', Candace Agonafir, 22 Apr 2024
  • RC2: 'Comment on egusphere-2024-551', Anonymous Referee #2, 03 Jun 2024
    • AC3: 'Reply on RC2', Candace Agonafir, 17 Jun 2024

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on egusphere-2024-551', Anonymous Referee #1, 14 Apr 2024
    • AC1: 'Reply on RC1', Candace Agonafir, 22 Apr 2024
      • AC2: 'Reply on AC1', Candace Agonafir, 22 Apr 2024
  • RC2: 'Comment on egusphere-2024-551', Anonymous Referee #2, 03 Jun 2024
    • AC3: 'Reply on RC2', Candace Agonafir, 17 Jun 2024
Candace Agonafir and Tian Zheng
Candace Agonafir and Tian Zheng

Viewed

Total article views: 545 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
398 120 27 545 14 14
  • HTML: 398
  • PDF: 120
  • XML: 27
  • Total: 545
  • BibTeX: 14
  • EndNote: 14
Views and downloads (calculated since 15 Mar 2024)
Cumulative views and downloads (calculated since 15 Mar 2024)

Viewed (geographical distribution)

Total article views: 582 (including HTML, PDF, and XML) Thereof 582 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 09 Dec 2024
Download
Short summary
Our study explores the spatio-temporal capabilities of machine learning models in urban flood prediction. First, zones are delineated via spectral clustering. Then, a detailed exploration of statistical and neural network models, each with varying advances, commences. The Graph Wavenet is demonstrated as having the strongest forecasting ability due to its graph architecture and temporal complexities. The research enhances data-driven approaches in urban flood modeling.