Journal
WATER RESOURCES MANAGEMENT
Volume 33, Issue 9, Pages 3239-3256Publisher
SPRINGER
DOI: 10.1007/s11269-019-02301-z
Keywords
Flood potential index; Adaptive neuro-fuzzy inference system; Fuzzy support vector machine; Bivariate statistics
Categories
Funding
- Research Institute of the University of Bucharest
Ask authors/readers for more resources
In Romania, as in the rest of the world, the flood frequency has increased considerably. Prahova river basin is among the most exposed catchments of the country to flood risk. It also represents the area of the present study for which the identification of surfaces with high susceptibility to flood phenomena was attempted by applying 2 hybrid models (adaptive neuro-fuzzy inference system and fuzzy support vector machine hybrid) and 2 bivariate statistical models (certainty factor and statistical index). The computation of Flood Potential Index (FPI) was possible by considering a number of 10 flood conditioning factors together with a number of 158 flood pixels and 158 non-flood pixels. Generally, the high and very high flood potential appears on around 25% of the upper and middle basin of Prahova river. The validation of the results was made through the ROC Curve model. One of the novelties of this research is related to the application of Fuzzy Support Vector Machine ensemble for the first time in a study concerning the evaluation of the susceptibility to a certain natural hazard.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available