Determination of flood probability and prioritization of sub-watersheds: A comparison of game theory to machine learning

Avand, Mohammadtaghi; Khiavi, Ali Nasiri; Khazaei, Majid; Tiefenbacher, John P.

doi:10.1016/j.jenvman.2021.113040

Cited by 45 publications

(15 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…6. Considering the complex nature of ood and its spatial changes, mapping its intensity and frequency at the watershed scale can be vital for making correct decisions in ood management, as well as understanding changes and predicting its occurrence (Avand et al 2021). The spatial mapping of ood events and its seasonal analysis of ood occurrence is useful and optimal tool for investigating the hydrological aspects of ooding (Ruiz et al 2014;Fischer et al 2016).…”

Section: Resultsmentioning

confidence: 99%

Iso-severity flood hazard patterns in diverse relief and climatic regions to support flood management

Mostafazadeh

Khiavi

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Floods are one of the most devastating natural events that threaten social and economic conditions more than any other natural disaster. Therefore, the aim of this study was to map the Iso-severity flood patterns and determining the seasonality of flood events over different return periods in in Ardabil province, Iran with diverse relief and climatic regions. Thus, the data of 32 River Gauge Stations (RGS) were used in the spatiotemporal analysis in a 40-year recorded period. EasyFit software was used to determine the most appropriate statistical distribution for the discharge data at 2-, 5-, 10-, 25-, 50- and 100-year return periods. Then, the frequency of monthly and seasonal flood events was quantified in each RGS. Then, the significant trend of flood frequency in each RGS was tested, and the seasonality of flood events was explored. According to the results, it can be said that in the studied RGS, about 63.63% of the floods are occurred in April. On the other hand, the seasonality of flood events showed that the maximum number and percentage of flood events are assigned to spring season over different return periods. As in some stations such as Yamchi, Nouran, Naneh-Karan, Mashiran, Amouqin, Doroud and Pole- Almas, almost 100% of the floods are related to spring and this shows that the floods in Ardabil province are spring floods. Iso-severity flood mapping in different return periods showed that the frequency of floods was higher than the central and northern regions in the southern and highlands of the study area. It is necessary for the southern and Iso-severity flood areas of the province, while conducting the necessary studies, proper planning should be considered with the focus on Integrated Watershed Management (IWM) in order to reduce the flood occurrence in the future and minimize its human and economic losses.

show abstract

Section: Resultsmentioning

confidence: 99%

Iso-severity flood hazard patterns in diverse relief and climatic regions to support flood management

Mostafazadeh

Khiavi

2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Following the literature review, numerous ML models were formulated and developed to map flood susceptibility [ 27 , 57 , 60 , 98 , 99 , 100 , 101 , 102 , 103 ]. In this study, four potential ML models of MLP-NN, NB, RF, and GBM, were used to assess the flood susceptibility in two different watersheds in Canada.…”

Section: Discussionmentioning

confidence: 99%

Spatial Prediction of Current and Future Flood Susceptibility: Examining the Implications of Changing Climates on Flood Susceptibility Using Machine Learning Models

Gharakhanlou

Pérez

2022

Entropy

View full text Add to dashboard Cite

The main aim of this study was to predict current and future flood susceptibility under three climate change scenarios of RCP2.6 (i.e., optimistic), RCP4.5 (i.e., business as usual), and RCP8.5 (i.e., pessimistic) employing four machine learning models, including Gradient Boosting Machine (GBM), Random Forest (RF), Multilayer Perceptron Neural Network (MLP-NN), and Naïve Bayes (NB). The study was conducted for two watersheds in Canada, namely Lower Nicola River, BC and Loup, QC. Three statistical metrics were used to validate the models: Receiver Operating Characteristic Curve, Figure of Merit, and F1-score. Findings indicated that the RF model had the highest accuracy in providing the flood susceptibility maps (FSMs). Moreover, the provided FSMs indicated that flooding is more likely to occur in the Lower Nicola River watershed than the Loup watershed. Following the RCP4.5 scenario, the area percentages of the flood susceptibility classes in the Loup watershed in 2050 and 2080 have changed by the following percentages from the year 2020 and 2050, respectively: Very Low = −1.68%, Low = −5.82%, Moderate = +6.19%, High = +0.71%, and Very High = +0.6% and Very Low = −1.61%, Low = +2.98%, Moderate = −3.49%, High = +1.29%, and Very High = +0.83%. Likewise, in the Lower Nicola River watershed, the changes between the years 2020 and 2050 and between the years 2050 and 2080 were: Very Low = −0.38%, Low = −0.81%, Moderate = −0.95%, High = +1.72%, and Very High = +0.42% and Very Low = −1.31%, Low = −1.35%, Moderate = −1.81%, High = +2.37%, and Very High = +2.1%, respectively. The impact of climate changes on future flood-prone places revealed that the regions designated as highly and very highly susceptible to flooding, grow in the forecasts for both watersheds. The main contribution of this study lies in the novel insights it provides concerning the flood susceptibility of watersheds in British Columbia and Quebec over time and under various climate change scenarios.

show abstract

“…In the flood probability estimation based on geo-environment indicators [40], the weight of indicator order is demonstrated as "slope", "distance to rivers", "altitude" (referring to the "elevation"), and then followed by the "terrain ruggedness index" (TRI) and "drainage density". Moreover, in the study of prioritization of sub-watershed flood probability based on physical, hydrological, and climatological parameters [41], the high-risk sub-watersheds have higher "permeability" and "rainfall" and greater "drainage network density" at a shorter "distance from rivers". In [42], the flood hazard estimation in the vicinity of the main channels of the Kifisos and Ilisos Rivers indicated that the highest flood hazard areas were total covered "streams", expansion of "impermeable formations" and "intense urbanization".…”

Section: The Choice Of Converging Indicators and The Related Influenc...mentioning

confidence: 99%

An Extended Watershed-Based AHP Model for Flood Hazard Estimation: Constraining Runoff Converging Indicators via MFD-Derived Sub-Watershed by Maximum Zonal Statistical Method

et al. 2022

View full text Add to dashboard Cite

Floods threaten the sustainable development of areas with a high probability of hazard. A typical analytic hierarchy process (pixel-based AHP) based flood hazard estimation method may ignore the similar threat caused by neighborhood cells at the sub-watershed scale. This study proposed an extended Watershed-based Zonal Statistical AHP for flood hazard estimation: Constraining converging related indicators by the sub-watersheds (WZSAHP-RC) model to improve this gap. Before calculating the flood hazard index, the proposed model uses the sub-watershed derived by the multiple flow direction method as a based unit to calculate the maximum zonal statistical value of runoff converging indicators. Moreover, taking the Chaohu basin of Anhui in China as the case study, the validation flooding ground-truthing was constructed from GF-3, and Landsat OLI images of the flood event from 20 July to 24 July 2020, which is the biggest flood recorded by the Zhongmiao station, which recorded a new water level, 0.82 m higher than the historical record. Compared with the validation, the results indicated the proposed method could improve the correct ratio by 38% (from 22% to 60%) and the fit ratio by 17% (from 18% to 35%) when considering the predicted flood hazard levels of “High” and “Very High” as flooded areas. Moreover, the flood hazard map derived by WZSAHP-RC demonstrated greater consistency in the flooded districts filtered by Baidu News than the pixel-based AHP. It revealed that considering two- or even multi-dimensional homogeneity may help to improve the accuracy of flood hazard maps on a catchment scale.

show abstract

Determination of flood probability and prioritization of sub-watersheds: A comparison of game theory to machine learning

Cited by 45 publications

References 53 publications

Iso-severity flood hazard patterns in diverse relief and climatic regions to support flood management

Iso-severity flood hazard patterns in diverse relief and climatic regions to support flood management

Spatial Prediction of Current and Future Flood Susceptibility: Examining the Implications of Changing Climates on Flood Susceptibility Using Machine Learning Models

An Extended Watershed-Based AHP Model for Flood Hazard Estimation: Constraining Runoff Converging Indicators via MFD-Derived Sub-Watershed by Maximum Zonal Statistical Method

Contact Info

Product

Resources

About