Dynamic Impact-Based Heavy Rainfall Warning with Multi-classification Machine Learning Approaches
The majority of flood assessment and warning systems primarily focus on the occurrence of floods caused by river overflow, taking into account factors such as intense precipitation. Improving flood resilience, on the other hand, requires a deeper understanding of how these factors affect each other...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Technoscience Publications
2024-12-01
|
| Series: | Nature Environment and Pollution Technology |
| Subjects: | |
| Online Access: | https://neptjournal.com/upload-images/(2)B-4154.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832594019752869888 |
|---|---|
| author | Anand Shankar |
| author_facet | Anand Shankar |
| author_sort | Anand Shankar |
| collection | DOAJ |
| description | The majority of flood assessment and warning systems primarily focus on the occurrence of floods caused by river overflow, taking into account factors such as intense precipitation. Improving flood resilience, on the other hand, requires a deeper understanding of how these factors affect each other and how specific local conditions can have an impact. This study offers impartial tools for estimating the severity of the effects brought on by heavy rainfall to facilitate the prompt communication of effective measures, such as the evacuation of livestock and human settlements and the provision of medical assistance. These tools take into account the cascading effects of various causative factors contributing to heavy rainfall. This article aims to assess the various factors that contribute to the impacts of heavy rainfall, including the timestamp (indicating soil saturation and moisture levels), river gauges (determining water congestion in canal systems), average aerial precipitation (indicating runoff), and the rainfall itself, taking into account both in situ and ex-situ impacts. Support Vector Machine (SVM), Decision Tree (DT), K-Nearest Neighbour (KNN), and Naive Bayes are some of the machine learning methods used in the study to find out how dynamically vulnerable affected districts are to flooding in different compound scenarios. This analysis is conducted by leveraging historical observed datasets. The results demonstrate the feasibility of mitigating the issue of excessive and insufficient flood warnings resulting from the cumulative effects of intense precipitation. By implementing a categorization system that divides the affected areas into various portions, or districts, according to the main factors contributing to flooding, namely rainfall, river discharge, and runoff, The suggested model presents novel insights into the sequential consequences of intense precipitation in the regularly inundated regions of North Bihar, India. Innovative tools can serve as valuable resources for flood forecasters and catastrophe managers to comprehend the extent of flooding and the consequential effects of intense precipitation. |
| format | Article |
| id | doaj-art-a50b470768bc4987b56578340691e188 |
| institution | Kabale University |
| issn | 0972-6268 2395-3454 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Technoscience Publications |
| record_format | Article |
| series | Nature Environment and Pollution Technology |
| spelling | doaj-art-a50b470768bc4987b56578340691e1882025-01-20T07:13:35ZengTechnoscience PublicationsNature Environment and Pollution Technology0972-62682395-34542024-12-012341885190010.46488/NEPT.2024.v23i04.002Dynamic Impact-Based Heavy Rainfall Warning with Multi-classification Machine Learning ApproachesAnand ShankarThe majority of flood assessment and warning systems primarily focus on the occurrence of floods caused by river overflow, taking into account factors such as intense precipitation. Improving flood resilience, on the other hand, requires a deeper understanding of how these factors affect each other and how specific local conditions can have an impact. This study offers impartial tools for estimating the severity of the effects brought on by heavy rainfall to facilitate the prompt communication of effective measures, such as the evacuation of livestock and human settlements and the provision of medical assistance. These tools take into account the cascading effects of various causative factors contributing to heavy rainfall. This article aims to assess the various factors that contribute to the impacts of heavy rainfall, including the timestamp (indicating soil saturation and moisture levels), river gauges (determining water congestion in canal systems), average aerial precipitation (indicating runoff), and the rainfall itself, taking into account both in situ and ex-situ impacts. Support Vector Machine (SVM), Decision Tree (DT), K-Nearest Neighbour (KNN), and Naive Bayes are some of the machine learning methods used in the study to find out how dynamically vulnerable affected districts are to flooding in different compound scenarios. This analysis is conducted by leveraging historical observed datasets. The results demonstrate the feasibility of mitigating the issue of excessive and insufficient flood warnings resulting from the cumulative effects of intense precipitation. By implementing a categorization system that divides the affected areas into various portions, or districts, according to the main factors contributing to flooding, namely rainfall, river discharge, and runoff, The suggested model presents novel insights into the sequential consequences of intense precipitation in the regularly inundated regions of North Bihar, India. Innovative tools can serve as valuable resources for flood forecasters and catastrophe managers to comprehend the extent of flooding and the consequential effects of intense precipitation.https://neptjournal.com/upload-images/(2)B-4154.pdfimpact-based heavy rainfall warning, multi-classification machine learning, impacts of floods, flood assessment, cascading impact |
| spellingShingle | Anand Shankar Dynamic Impact-Based Heavy Rainfall Warning with Multi-classification Machine Learning Approaches Nature Environment and Pollution Technology impact-based heavy rainfall warning, multi-classification machine learning, impacts of floods, flood assessment, cascading impact |
| title | Dynamic Impact-Based Heavy Rainfall Warning with Multi-classification Machine Learning Approaches |
| title_full | Dynamic Impact-Based Heavy Rainfall Warning with Multi-classification Machine Learning Approaches |
| title_fullStr | Dynamic Impact-Based Heavy Rainfall Warning with Multi-classification Machine Learning Approaches |
| title_full_unstemmed | Dynamic Impact-Based Heavy Rainfall Warning with Multi-classification Machine Learning Approaches |
| title_short | Dynamic Impact-Based Heavy Rainfall Warning with Multi-classification Machine Learning Approaches |
| title_sort | dynamic impact based heavy rainfall warning with multi classification machine learning approaches |
| topic | impact-based heavy rainfall warning, multi-classification machine learning, impacts of floods, flood assessment, cascading impact |
| url | https://neptjournal.com/upload-images/(2)B-4154.pdf |
| work_keys_str_mv | AT anandshankar dynamicimpactbasedheavyrainfallwarningwithmulticlassificationmachinelearningapproaches |