A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Quantifying the Impact of Future Climate Change on Flood Susceptibility: An Integration of CMIP6 Models, Machine Learning, and Remote Sensing
https://orcid.org/0000-0002-3416-4551
In recent years, the frequency of floods has escalated due to global warming and human activities. Addressing this challenge, our study investigates how future climate change scenarios will affect flood susceptibility in the Tajan watershed, northern Iran. The primary objective is to quantify and map the evolving risk of flooding in this region under different future climate scenarios. We applied machine learning techniques, coupled model intercomparison project phase 6 (CMIP6) climatic models, and remote sensing to achieve this goal. The CanESM5 climate model was chosen for its accuracy among four global climate models in CMIP6 to estimate future precipitation trends under shared socioeconomic pathways (SSP 2.6, 4.5, and 8.5) over two-time horizons: future (2020–2060) and far future (2061–2100). These scenarios encompass various influential factors, such as greenhouse gas emissions, urbanization, deforestation, and socioeconomic development, which played crucial roles in modulating flood susceptibility. Flood susceptibility maps were generated considering future precipitation patterns and scenarios using random forest (RF) and support vector machine (SVM) algorithms, 432 flood locations, and 15 flood influencing factors. The accuracy of our prediction was validated through multiple statistical measures, including the area under the receiver operating characteristic (AUC-ROC) curve. The results indicated that the proposed models performed well, with the RF model () demonstrating higher accuracy compared to the SVM model (). From a spatial perspective, increased future precipitation under all SSP scenarios enhances the likelihood of flood occurrences in the central and downstream regions. In the far future, intensified precipitation due to changes in regional topography and climate, coupled with higher greenhouse gas concentrations, is expected to heighten flood risks, especially at higher altitudes. We hope that our study findings will inform effective flood risk management strategies and adaptation plans in response to climate-induced flood risks, both in our study area and in similar regions globally.
Quantifying the Impact of Future Climate Change on Flood Susceptibility: An Integration of CMIP6 Models, Machine Learning, and Remote Sensing
https://orcid.org/0000-0002-3416-4551
In recent years, the frequency of floods has escalated due to global warming and human activities. Addressing this challenge, our study investigates how future climate change scenarios will affect flood susceptibility in the Tajan watershed, northern Iran. The primary objective is to quantify and map the evolving risk of flooding in this region under different future climate scenarios. We applied machine learning techniques, coupled model intercomparison project phase 6 (CMIP6) climatic models, and remote sensing to achieve this goal. The CanESM5 climate model was chosen for its accuracy among four global climate models in CMIP6 to estimate future precipitation trends under shared socioeconomic pathways (SSP 2.6, 4.5, and 8.5) over two-time horizons: future (2020–2060) and far future (2061–2100). These scenarios encompass various influential factors, such as greenhouse gas emissions, urbanization, deforestation, and socioeconomic development, which played crucial roles in modulating flood susceptibility. Flood susceptibility maps were generated considering future precipitation patterns and scenarios using random forest (RF) and support vector machine (SVM) algorithms, 432 flood locations, and 15 flood influencing factors. The accuracy of our prediction was validated through multiple statistical measures, including the area under the receiver operating characteristic (AUC-ROC) curve. The results indicated that the proposed models performed well, with the RF model () demonstrating higher accuracy compared to the SVM model (). From a spatial perspective, increased future precipitation under all SSP scenarios enhances the likelihood of flood occurrences in the central and downstream regions. In the far future, intensified precipitation due to changes in regional topography and climate, coupled with higher greenhouse gas concentrations, is expected to heighten flood risks, especially at higher altitudes. We hope that our study findings will inform effective flood risk management strategies and adaptation plans in response to climate-induced flood risks, both in our study area and in similar regions globally.
Quantifying the Impact of Future Climate Change on Flood Susceptibility: An Integration of CMIP6 Models, Machine Learning, and Remote Sensing
https://orcid.org/0000-0002-3416-4551
In recent years, the frequency of floods has escalated due to global warming and human activities. Addressing this challenge, our study investigates how future climate change scenarios will affect flood susceptibility in the Tajan watershed, northern Iran. The primary objective is to quantify and map the evolving risk of flooding in this region under different future climate scenarios. We applied machine learning techniques, coupled model intercomparison project phase 6 (CMIP6) climatic models, and remote sensing to achieve this goal. The CanESM5 climate model was chosen for its accuracy among four global climate models in CMIP6 to estimate future precipitation trends under shared socioeconomic pathways (SSP 2.6, 4.5, and 8.5) over two-time horizons: future (2020–2060) and far future (2061–2100). These scenarios encompass various influential factors, such as greenhouse gas emissions, urbanization, deforestation, and socioeconomic development, which played crucial roles in modulating flood susceptibility. Flood susceptibility maps were generated considering future precipitation patterns and scenarios using random forest (RF) and support vector machine (SVM) algorithms, 432 flood locations, and 15 flood influencing factors. The accuracy of our prediction was validated through multiple statistical measures, including the area under the receiver operating characteristic (AUC-ROC) curve. The results indicated that the proposed models performed well, with the RF model () demonstrating higher accuracy compared to the SVM model (). From a spatial perspective, increased future precipitation under all SSP scenarios enhances the likelihood of flood occurrences in the central and downstream regions. In the far future, intensified precipitation due to changes in regional topography and climate, coupled with higher greenhouse gas concentrations, is expected to heighten flood risks, especially at higher altitudes. We hope that our study findings will inform effective flood risk management strategies and adaptation plans in response to climate-induced flood risks, both in our study area and in similar regions globally.
Quantifying the Impact of Future Climate Change on Flood Susceptibility: An Integration of CMIP6 Models, Machine Learning, and Remote Sensing
J. Water Resour. Plann. Manage.
Gholami, Farinaz (author) / Li, Yue (author) / Zhang, Junlong (author) / Nemati, Alireza (author)
2024-09-01
Article (Journal)
Electronic Resource
English
Artificial Neural Networks for Flood Prediction in Current and CMIP6 Climate Change Scenarios
| Wiley | 2025
Artificial Neural Networks for Flood Prediction in Current and CMIP6 Climate Change Scenarios
| DOAJ | 2025
Artificial Neural Networks for Flood Prediction in Current and CMIP6 Climate Change Scenarios
| Wiley | 2025