ForeINTiFlood: A Novel Framework for Forensic Investigation of Coastal Tidal Floods in The Pekalongan Coastal Area, Central Java, Indonesia.
Saved in:
| Title: | ForeINTiFlood: A Novel Framework for Forensic Investigation of Coastal Tidal Floods in The Pekalongan Coastal Area, Central Java, Indonesia. |
|---|---|
| Authors: | Yulianto, Fajar, Wibowo, Mardi, Yananto, Ardila, Perdana, Dhedy Husada Fadjar, Prabowo, Yudhi, Wiguna, Edwin Adi, Khoirunnisa, Hanah, Aziz, Hilmi, Nurwijayanti, Amalia, Fachrudin, Imam, Kongko, Widjo |
| Source: | Earth Systems & Environment; Jun2025, Vol. 9 Issue 2, p845-880, 36p |
| Abstract: | This study proposes a novel method, the Forensic Investigations Tidal Flood (ForeINTiFlood) framework, to investigate tidal floods in the Pekalongan coast, a region highly susceptible to these events. The key challenge lies in identifying the primary drivers of such floods. ForeINTiFlood addresses this gap by employing a multi-source-temporal geospatial data framework that integrates machine learning and hydrodynamic modeling. This framework offers a robust suite of tools for forensic investigations of coastal tidal flooding. The method leverages a comprehensive set of geo-bio-physical parameters encompassing topographic, bathymetric, shoreline change, land use and land cover, human infrastructure, ecological, and coastal hydrodynamic data. The effectiveness of ForeINTiFlood is assessed through the evaluation of variable importance in machine learning models trained on data spanning 2000 to 2020. The analysis yielded topography, ecology, land use and land cover, and human infrastructure as the most influential variables, with average importance scores of 0.387, 0.245, 0.199, and 0.160, respectively. Accuracy assessments using the k-fold approach demonstrated high performance for Random Forest (RF), Gradient Tree Boosting (GTB), and Classification and Regression Tree (CART) models, achieving average accuracies of 0.927, 0.924, and 0.908, respectively. Furthermore, a positive correlation was observed between the predicted flood probability from the machine learning models and the actual floodwater depth obtained from hydrodynamic modeling for the periods 2000, 2005, 2010, 2015, and 2020, with correlation values ranging from 0.52 to 0.72. These findings suggest a strong link between the model's predictions and real-world flood occurrences. ForeINTiFlood forensic investigation identified three specific areas (central, western, and eastern) as potential entry points for tidal floods. High-resolution satellite imagery further corroborated these findings. These results hold significant value for coastal zone management, flood mitigation strategies, and the development of resilience plans to combat tidal flooding challenges. Additionally, the results of this study can inform decision-makers in formulating strategies to address tidal flooding problems in the study area. [ABSTRACT FROM AUTHOR] |
| Copyright of Earth Systems & Environment is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Complementary Index |
Full Text Finder 
Nájsť tento článok vo Web of Science Be the first to leave a comment!