Flood susceptibility mapping in northern regions of Iran using advanced data mining algorithms (Case study: Haraz watershed)

Subject Areas : Regional Planning

Himan Shahabi ¹

1 - . Associate Professor, Department of Geomorphology, Faculty of Natural Resources, University of Kurdistan, Sanandaj, Iran (Department of Zrebar Lake Environmental Research, Kurdistan Studies Institute, University of Kurdistan)

Received: 2020-05-14 Accepted : 2020-07-03 Published : 2021-04-21

Keywords: Bagging, Shannon entropy, Haraz watershed, Flood susceptibility,

Abstract :

Floods are one of the phenomena in nature that human beings have been witnessing for a long time. In Iran, due to the large area, diffident climates, temporal and spatial density of rainfall in most watersheds, we see huge floods every year. Flood susceptibility mapping is one of the basic strategies to reduce the loss of life and property due to floods. In this study, Bagging and Shannon Entropy methods have been used to prepare flood susceptibility maps. In the current study, 201 floodplain locations were prepared. Of the 201 positions, 70% were used for modeling and map preparation. The remaining 30%, which were randomly generated, were used to validate the maps produced. Furthermore, ten effective factors including slope, land curvature, distance to river, elevation, rainfall, stream power index (SPI), topographic wetness index (TWI), lithology, land use and normalized difference vegetation index (NDVI) were used. The mentioned models determined the effect weight of each factor affecting the occurrence of floods. The ROC curve was drawn and the area below the curve (AUC) was calculated to validate the flood susceptibility map. The results showed that Bagging model has a higher accuracy than Shannon Entropy model. Therefore, the high accuracy of this model indicates that it is reliable for preparing a flood susceptibility map in areas without statistics.
The results showed that Bagging model has a higher accuracy than Shannon Entropy model. Therefore, the high accuracy of this model indicates that it is reliable for preparing a flood susceptibility map in areas without statistics.
 
Extended Abstract
Introduction
Due to the importance of flood hazards and Its growing trend in recent years, the preparation of flood maps and flood sensitization zoning has received special attention from researchers and experts. To prepare flood maps There are different hydraulic (by HEC-RAS) and hydrological methods and in recent years Many statistical and probabilistic models have been tested for flood susceptibility maps. also GIS software as a basic analysis tool has been used for spatial management and data manipulation due to its ability to handle large amounts of spatial data and the combination of statistical and probabilistic models with RS and GIS has attracted a lot of attention from researchers.
Northern regions of Iran, due to its special natural and climatic conditions, in terms of population and forest cover, they are one of the most densely populated areas in the country. In addition to high population density and forest cover, in these areas Every year, considerable crops, livestock, orchards, etc. are produced and it shows the importance of this region in various indicators of local and regional development. However, every year there are different natural hazards We are among the floods in the northern regions of the country, including the Haraz watershed in some cases, in addition to extensive financial losses, there are casualties. To prevent these accidents and reduce these damages and casualties, identify places and areas prone to floods and in general, sensitive areas in this area, it seems necessary and logical. In this study, an attempt has been made Based on various factors such as Slope, curvature of the earth, distance from the river, elevation, precipitation, river power index and …, Flood susceptibility zoning map in the watershed of Haraz Using advanced data mining algorithms Be prepared and validated. Therefore, the basic questions that the researcher in this research seeks to investigate as follows: In which part of the Haraz watershed are areas with high flood susceptibility? and to prepare a flood susceptibility map in Haraz watershed, which of entropy Shannon's and Baggind models work best?
 
Methodology
The aim of this applied research, is a preparation flood susceptibility mapping in Haraz watershed using the advanced data mining algorithms that Done with the quantitatively Method, required data According to the objectives of the research Collected from relevant organizations and agencies (Regional Water Company, Natural Resources Department, etc.) and to analyze this data ArcGIS software is used. Overall The research process is as follows First Prepared List of past floods in the study area and so on the effective parameters in the occurrence of flooding have been identified and Using two models, Shannon and Bagging entropy, it is provided Flood susceptibility zoning map in Haraz watershed in northern Iran. Then using the ROC curve, the accuracy and validation of the models have been investigated.
 
Results and discussion
Validation of sensitivity maps prepared in this study Obtained by calculating the relative characteristics index or ROC. This curve Is one of the most efficient methods in providing the ability to determine, identify potential and predict systems that Estimates the accuracy of the model quantitatively. In this way, Area below the curve or AUC It has values ​​between 0.5 and 1, and It is used to evaluate the accuracy of the model. best model Has a level below the curve close to 1, while Values ​​close to 0.5 Indicates the inaccuracy in the model. The results indicate that That Bagging model (0.96) It has a higher accuracy than the Shannon entropy model (0.88).  Although both models are acceptable, But the Bagging Model It has the highest acceptable accuracy in preparing flood susceptibility maps in Haraz watershed.
 
Conclusion
Due to floods in the northern parts of the country and Their growing trend, Preparation of flood susceptibility map, it is a background for cognition Factors affecting flood occurrence, Its occurrence, risk management and risk prevention methods. The purpose of the present study Prioritize the factors that affect the occurrence of floods Using Shanghai bagging and entropy models. after preparing the location map of the floods, 10 factors Included the Slope, curvature of the earth, elevation, Distance from the river, Rainfall, TWI, SPI, lithology, land use and NDVI, they were selected as the factors influencing the flooding of Haraz watershed in Mazandaran province. Prioritize the factors influencing the occurrence of floods Using the Shannon Entropy Index, it showed The NDVI layers weigh (2.03), Distance from the river (1/1), SPI (1.09), Elevation classes (0.995), Slope (0.847), Rainfall (0.54), Lithology (0.421), TWI (0.309), Land use (0.223) and Earth's curvature (0.136) Respectively Have had the most to the least impact on the occurrence of floods. Based on ROC curve results, bagging model It has the highest accuracy in predicting flood susceptibility maps in Haraz watershed and Then there is the Shannon entropy model. According to the final flood susceptibility map, Around the Haraz River Has a high sensitivity to flooding. So it should Refrain from the construction of residential areas or Fruit orchards and even agricultural lands around the river and to Residential areas and existing gardens be made Precautions such as flood walls or graves until the Avoid causing too much damage to these parts. Thus Preparation of sensitivity maps for natural disasters such as floods, landslides, etc. It is necessary for future management and planning until the Be prevented from the loss of life and property to these sectors. Use the results of this research It is necessary to offices and organizations Agricultural Jihad, Natural Resources, Regional Water, Ministry of Energy, Housing and Urban Development, Islamic Revolution Housing Foundation and all researchers and decision makers, and even urban and rural managers, to think about the necessary arrangements to prevent and reduce the destructive effects of floods and their side effects.

References:

1. Bednarik, M., Magulova, B., Matys, M., and Marschalko. M. (2010): Landslide susceptibility assessment of theKralovany–Liptovsky Mikulaš railway case study. Phys. Chem. Earth 35: 162–171.


2. Bronstert, A., (2003): Floods and climate change: interactions and impacts. Risk Anal. 23, 545–557.


3. Bubeck, P., Botzen, W., and Aerts, J., (2012): A review of risk perceptions and other factors that in fl uence fl ood mitigation behavior. Risk Anal. 32, 1481–1495.


4. Fawcett, T., (2006): An introduction to ROC analysis. Pattern Recognition Letters, V. 27, p. 861-874.


5. Feng, CC., and Wang, YC. (2011): GIScience research challenges for emergency   management   in   Southeast   Asia.   Nat   Hazards, 59:597–616.


6. Gokceoglu, C., Sonmez, H., Nefeslioglu, H.A., Duman, T.Y., and Can, T. (2005): The 17 March 2005 Kuzulu landslide (Sivas, Turkey) and landslide-susceptibility map of its near vicinity. Eng. Geol. 81, 65–83.


7. Jamini, Davood., Amini, Abbas., Ghadermarzi, Hamed and Tavakoli, Jafar  (2017): Measurement of Food Security and Investigation of its Challenges in Rural Areas (Case Study: Badr District from Ravansar County), JOURNAL OF REGIONAL PLANNING, 7 (27): pp: 87-102. (in Persian)


8. Kamali, M., Solaimani, K., Shahedi, K., Gord- Noshahri, A., and Gomrokchi, A. (2015): Determining the Flooding Points and Prioritizing Subcatchments of Barajin Catchment of Qazvin Using Hec-HMS and GIS, Iran-Watershed Management Science & Engineering, Vol. 9, No. 29, pp: 27-34. (in Persian)


9. Khosravi, K., Pourghasemi, H.R., Chapi, K., and Bahri, M. (2016): Environ Monit Assess, 188: 656. doi:10.1007/s10661-016-5665-9.


10. Kjeldsen, TR. (2010): Modelling the impact of urbanization on flood frequency relationships in the UK. Hydrol Res 41:391–405.


11. Komac, M.A. (2006): Landslide susceptibility model using the Analytical Hierarchy Process method and multivariate statistics in perialpine Sloveni. Geomorphology 74: 17-28.


12. Kourgialas, N.N., and Karatzas, G.P. (2011): Flood management and a GIS modelling method to assess fl ood-hazard areas—a case study. Hydrol. Sci. J. 56, 212–225.


13. Lee, M.J., Kang, J.E., and Jeon, S. (2012): Application of frequency ratio model and validation for predictive flooded area susceptibility mapping using GIS. In: Geoscience and Remote Sensing Symposium (IGARSS), Munich. 895–898.


14. Manandhar, B. (2010): Flood Plain Analysis and Risk Assessment of Lothar Khola, MSc Thesis, Tribhuvan University, Phokara, Nepal, pp. 64.


15. Merz, B., Thieken, A.H., Gocht, M. (2007): Flood Risk Mapping at the Local Scale: Concepts   and   Challenges, Flood   Risk   Management   in   Europe.   Springer, Netherlands, pp. 231–251.


16. Miller, JR., Ritter, DF., and Kochel, RC. (1990): Morphometric assessment of lithologic controls on drainage basin evolution in the Crawford Upland, south-central Indiana. Am J Sci. 290:569–599.


17. Moore, I.D., Grayson, R.B., and Ladson, A.R. (1991): Digital terrain modelling: a review of hydrological, geomorphological, and biological applications. Hydrol. Proc. 5, 3–30.


18. Naghibi.S.A., Pourghasemi, H.R., Pourtaghi, Z.S., and Rezaei, A. (2014): Groundwater qanat potential mapping using frequency ratio and Shannon’s entropy models in the Moghan watershed, Iran. Earth Sci Inform, DOI 10.1007/s12145-014-0145-7.


19. Nandi, A., and Shakoor, A. (2009): A GIS-based landslide susceptibility evaluation using bivariate and multivariate statistical analyses. Engineering Geology, V. 110, p. 11–20.


20. Nefeslioglu, H.A., Duman, T.Y., and Durmaz, S. (2008): Landslide susceptibility mapping for a part of tectonic Kelkit Valley (Easten Black Sea Region of Turkey), Geomorphology 94: 401-418.


21. Oh, H.  J., and Lee, S. (2010):  Cross-validation of logistic regression model for landslide susceptibility mapping at Geneoung areas, Korea. Disaster Advances, V. 3, p. 44–55.


22. Oh, H.J., and Pradhan, B. (2011): Application of a neuro-fuzzy model to landslide- susceptibility mapping for shallow landslides in a tropical hilly area. Computer and Geoscience, 37, 1264–1276.


23. Pourghasemi, H.R., Moradi, H.R., and Fatemi Aghda, S.M. (2018): Prioritizing Effective Factors in Landslide Occurrence and its Susceptibility Mapping Using Shannon's Entropy Index, Journal of Hydrology and Soil Science, Volume:18 Issue: 4, pp: 181-192. (in Persian)


24. Pourghasemi, H.R., Moradi, H.R., Fatemi Aghda, S.M., Gokceoglu, C., and Pradhan, B. (2012): GIS-based landslide susceptibility mapping with probabilistic likelihood ratio and spatial multi-criteria evaluation models (North of Tehran, Iran). arabian journal of geosciences, 7: 1857-1878.


25. Pradhan, B., Oh, H.J., and Buchroithner, M. (2010): Weights-of-evidence model applied to landslide susceptibility mapping in a tropical hilly area. Geomat. Nat. Haz. Risk, 1, 199-223.


26. Ramakrishna, D., Ghose, M.K., Vinu Chandra, R., and Jeyaram, A. (2005): Probabilistic techniques, GIS and remote sensing in landslide hazard mitigation: a case study from Sikkim Himalayas, India. Geocartography Int. 20 (4): 53–58.


27. Shamsodini, A., Jamini, D., and Jamshidi, A. (2016): Measurement and Analyses of Social Stability in Rural Area (Case Study: Javanrood Township).  Journal of Rural Research. 7(3), 486-503. (in Persian)


28. Sharifi paichoon, M., Omidvar, K., and Motazaker, K. (2019): Assessment of flooding using cluster analysis and multivariable regression methods with emphasis on hydro geomorphological parameters (Case study: Maroon catchment), Journal of Natural Environmental Hazards, Vol 8, Issue 21, pp: 75-92. (in Persian)


29. Sharma, L.P., Patel, N., Ghose, M. K., and Debnath, P. (2010): Influence of Shannon’s entropy on landslide-causing parameters for vulnerability study and zonation-a case study in Sikkim, India. Arab. J. Geosci. 5 (3): 421-431.


30. Swets, J.A. (1988): Measuring the accuracy of diagnostic systems. Sci. 240: 1285-1293.


31. Taylor, J., Davies, M., Clifton, D., Ridley, I., and Biddulph, P. (2011): Flood management:  prediction of microbial contamination in large- scale floods in urban environments. Environ Int 37:1019–1029.


32. Tehrany, M.S., Lee, M.J., Pradhan, B., Jebur, M.N., and Lee, S. (2014b): Flood susceptibility mapping using integrated bivariate and multivariate statistical models. Environ. Earth sci. 72(10): 4001-4015.


33. Tehrany, M.S., Pradhan, B., and Jebur, M.N. (2013): Spatial prediction of flood susceptible areas using rule based decision tree (DT) and a novel ensemble bivariate and multivariate statistical models in GIS. J. Hydrol. 504, 69-79.


34. Tehrany, M.S., Pradhan, B., and Jebur, M.N. (2014a): Flood susceptibility mapping using a novel ensemble weights-of-evidence and support vector machine models in GIS. J. Hydrol. 512: 332-343.


35. Tehrany, M.S., Pradhan, B., and Jebur, M.N. (2015b): Flood susceptibility analysis and its verification using a novel ensemble support vector machine and frequency ratio method. Stoch Environ Res Risk Assess 29:1149–1165.


36. Tehrany, M.S., Pradhan, B., Mansor, Sh., and Ahmad, N. (2015a): Flood susceptibility assessment using GIS-based support vector machine model with different kernel types. Catena 125, 91-101.


37. Theil, H. (1972): Statistical decomposition analysis. North-Holland Publishing Company, Amsterdam.


38. Youssef, A.M., Pradhan, B., and Hassan, A.M. (2011): Flash flood risk estimation along the St. Katherine road, southern Sinai, Egypt using GIS based morphometry and satellite imagery. Environ. Earth Sci. 62, 611–623.


39. Yufeng, S., and Fengxiang, J. (2009): Landslide Stability Analysis Based on Generalized Information Entropy. 2009 International Conference on Environmental Science and Information Application Technology: 83–85.