Document Type : Research Article

Authors

1 Watershed Management Department, Sari University of Agricultural Sciences and Natural Resources, Sari, Iran

2 Department of Watershed Management, Faculty of Natural Resources and Marine Sciences, Tarbiat Modares University, Noor, Iran

3 Research Center for Geosciences and Social Studies, Hakim Sabzevari University, Sabzevar, Iran

Abstract

Introduction
Flood is one of the most destructive natural disasters that has a negative impact on social, economic and environmental dimensions. Floods usually occur following a prolonged period of rain or snowmelt in combination with unfavorable conditions. In this regard, all over the world, the occurrence of floods has intensified by 40% in the last two decades. In Asia, almost 90% of all human casualties caused by natural disasters are due to floods. The increase in flooding is usually due to increased environmental degradation such as urbanization, increased population growth, and deforestation. Periodic and regular occurrences of floods over a certain timeframe significantly amplify the detrimental impacts on living organisms. Urban areas in close proximity to rivers bear the brunt of these damages, owing to high population density, economic infrastructure, and transportation networks. However, these consequences can be alleviated through meticulous vulnerability analysis. One of the primary objectives pursued by researchers and policymakers is the precise modeling and zoning of floods to mitigate associated risks. Consequently, a myriad of methods and approaches have been devised for flood risk modeling and zoning to address this pressing issue. Among them, hydrological methods such as rainfall-runoff modeling and data-based techniques, which are unable to comprehensively analyze rivers and flood zones due to their one-dimensional nature. This is despite the fact that the morphology of the river is not stable and due to its high erosion potential, it also has a dynamic characteristic. In addition, these methods require fieldwork and large budgets for data collection. Hence, comprehensive flood management is necessary to reduce these effects. Therefore, this study was conducted with the aim of identifying areas sensitive to the risk of flooding in Famnat watershed located in Gilan province. Fomanat watershed is located in Gilan province and is considered a part of the first grade watershed of the Central Plateau. This area is located in the range of 36.89 to 37.57 degrees north latitude and 48.77 to 49.69 degrees east longitude. This region has an area of 3595 square kilometers, the highest point of which is 3088 meters and the lowest point is -69 meters.
 
Materials and Methods
 To carry out the current research, firstly, by reviewing the sources and history of the research, as well as knowing the region, a map and layers of information related to the factors affecting flood susceptibility zoning were prepared. These layers include land use map, slope degree, geology, distance from waterway, digital map of height, direction, shape of land curvature, land curvature profile, rainfall and topographic humidity index, which are created using the collected data and also various additions in the environment. Geographic information system (Arcgis 10.4) was prepared. In this regard, machine learning models such as generalized linear model (GLM), multivariate adaptive regression model (MARS) and classification and regression tree model (CART) were used to zone the sensitivity of the watershed to floods. Also, among 100 flood events, 70% (70) were considered for training and 30% (30) for validation. In the following, using field survey and review of previous studies, 10 factors influencing the occurrence of floods in the watershed area were identified and used. Finally, the area under the ROC curve and the TSS index were used to evaluate the models.
Results and Discussion
 The results of the evaluation of the most important factors affecting the sensitivity of the watershed to floods indicated that the distance from the river, the height and the curvature profile had the greatest impact on the sensitivity of the region, and on the other hand, the factors of slope, geology and topographic humidity index had the least impact. Based on the obtained results, the areas covered by very low, low, medium, high and very high classes in the CART model were 26.6, 17.6, 21.2, 0.1 and 34.0%, respectively. These results for the GLM model were 13.6, 12.7, 16.2, 25.1 and 32.4 percent, respectively. Based on the obtained results, the CART model performed better than other models, so that AUC for MARS model was equal to 0.76, CART model was equal to 0.9 and GLM model was equal to 0.84. Also, the better performance of CART model compared to other models was confirmed by other indicators. So, based on TSS, MARS model equal to 0.52, CART model equal to 0.77 and GLM model equal to 0.66 were obtained.
Conclusion
Implementing the findings of this study can facilitate the adoption of effective management strategies to mitigate losses and casualties. Moreover, in developing nations grappling with restricted access to hydrogeological and soil data, the utilization of geographic information systems (GIS) and data mining techniques assumes a pivotal role in conducting comprehensive studies. These technologies offer valuable insights and support decision-making processes, enabling proactive measures to address flood risks and enhance disaster resilience in vulnerable regions.
 

Keywords

Main Subjects

©2023 The author(s). This is an open access article distributed under Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source.

  1. Adnan, R.M., Liang, Z., & Heddam, S. (2019). Least square support vector machine and multivariate adaptive regression splines for streamflow prediction in mountainous basin using hydro-meteorological data as inputs. Journal of Hydrology, (586), 124371. https://doi.org/10.1016/J.JHYDROL.2019.124371
  2. Atkinson, P., Jiskoot, H., Massari, R., & Murray, T. (1998). Generalized linear modelling in geomorphology. Earth Surface Processes and Landforms. The Journal of the British Geomorphological Group23(13), 1185-1195.
  3. Avand, M., & Moradi, H. (2021). Using machine learning models, remote sensing, and GIS to investigate the effects of changing climates and land uses on flood probability. Journal of Hydrology, 595, 125663. https://doi.org/10.1016/j.jhydrol.2020.125663
  4. Avand, M. T., Moradi, M., Ramazanzadeh Lasboyee, Mehdi. (2021). Spatial prediction of future flood risk: an approach to the effects of climate change. Geosciences, 11(1), 1-20. https://doi.org/10.3390/geosciences11010025
  5. Avand, M.T., Moradi, H.R., & Ramzanzadeh Lesboi, M. (2020). Preparation of flood sensitivity map using Bayesian random forest and linear generalized machine learning models. Environment and Water Engineering, 6(1), 83-95.
  6. Avand, M., Kuriqi, A., Khazaei, M., & Ghorbanzadeh, O. (2022). DEM resolution effects on machine learning performance for flood probability mapping. Journal of Hydro-Environment Research, 40, 1-16. https://doi.org/10.1016/j.jher.2021.10.002
  7. Azadi, F., Sadouq, S.H., Ghahrodi, M., & Shahabi, H. (2020). Flood risk sensitivity zoning in Kashkan watershed using WOE and EBF models. Scientific Journal of Geography and Environmental Hazards, 9(1), 45-60.
  8. Chapi, K., Singh, V.P., Shirzadi, A., Shahabi, H., Bui, D.T., Pham, B.T., & Khosravi, K. (2017). A novel hybrid artificial intelligence approach for flood susceptibility assessment. Environmental Modelling & Software, 95, 229-245. https://doi.org/10.1016/j.envsoft.2017.06.012
  9. Hasanuzzaman, M., Islam, A., Bera, B., & Shit, P.K. (2022). A comparison of performance measures of three machine learning algorithms for flood susceptibility mapping of river Silabati (tropical river, India). Physics and Chemistry of the Earth, Parts A/B/C,127, 103198. https://doi.org/10.1016/j.pce.2022.103198
  10. Kalantar, B., Ueda, N., Saeidi, V., Ahmadi, K., Halin, A. A., & Shabani, F. (2020). Landslide susceptibility mapping: Machine and ensemble learning based on remote sensing big data. Remote Sensing12(11), 1737. https://doi.org/10.3390/rs12111737
  11. Khosravi, K., Nohani, E., Maroufinia, E., Pourghasemi, H.R. (2016). A GIS-based flood susceptibility assessment and its mapping in Iran: a comparison between frequency ratio and weights-of-evidence bivariate statistical models with multi-criteria decision-making technique. Natural Hazards, 83, 947–987. https://doi.org/10.1007/s11069-016-2357-2
  12. Kourgialas, N.N., Karatzas, G.P. (2011). Flood management and a GIS modelling method to assess flood-hazard areas—a case study. Hydrological Sciences Journal, 56, 212–225. https://doi.org/10.1080/02626667.2011.555836
  13. Lazarus, E.D., Constantine, J.A. (2013). Generic theory for channel sinuosity. Proceedings of the National Academy of Sciences USA, 110, 8447–8452. https://doi.org/10.1073/pnas.1214074110
  14. Marmion, M., Luoto, M., Heikkinen, R.K., & Thuiller, W. (2009). The performance of state-of-the-art modelling techniques depends on geographical distribution of species. Ecological Modelling220(24), 3512-3520. https://doi.org/10.1016/j.ecolmodel.2008.10.019
  15. Mosavi, A., Golshan, M., Janizadeh, S., Choubin, B., Melesse, A.M., & Dineva, A.A. (2020). Ensemble models of GLM, FDA, MARS, and RF for flood and erosion susceptibility mapping: a priority assessment of sub-basins. Geocarto International, 37(9): 1-20. https://doi.org/10.1080/10106049.2020.1829101
  16. Mirchooli, F., Motevalli, A., Pourghasemi, H.R., Mohammadi, M., Bhattacharya, P., Maghsood, F.F., & Tiefenbacher, J.P. (2019). How do data-mining models consider arsenic contamination in sediments and variables importance? Environmental Monitoring and Assessment, 191. https://doi.org/10.1007/s10661-019-7979-x
  17. Mohammadi, M., Darabi, H., Mirchooli, F., Bakhshaee, A., & Torabi Haghighi, A. (2021). Flood risk mapping and crop-water loss modeling using water footprint analysis in agricultural watershed, northern Iran. Natural Hazards105, 2007-2025. https://doi.org/10.1007/s11069-020-04387-w
  18. Nachappa, T.G., Ghorbanzadeh, O., & Gholamnia, K. (2020). Multi-Hazard Exposure Mapping Using Machine Learning for the State of Salzburg, Austria. Remote Sensing, 12(17), 2757. https://doi.org/10.3390/rs12172757
  19. Mosavi, A., Golshan, M., Janizadeh, S., Choubin, B., Melesse, A.M., & Dineva, A.A. (2022). Ensemble models of GLM, FDA, MARS, and RF for flood and erosion susceptibility mapping: a priority assessment of sub-basins. Geocarto International37(9), 2541-2560. https://doi.org/10.1080/10106049.2020.1829101
  20. Pourghasemi, H.R., & Beheshtirad, M. (2015). Assessment of a data-driven evidential belief function model and GIS for groundwater potential mapping in the Koohrang Watershed, Iran. Geocarto International, 30, 662-685. https://doi.org/10.1080/10106049.2014.966161
  21. Pourghasemi, H.R., Kariminejad, N., Amiri, M., Edalat, M., Zarafshar, M., Blaschke, T., & Cerda, A. (2020). Assessing and mapping multi- hazard risk susceptibility using a machine learning technique. Scientific Reports, 10, 1–11. https://doi.org/10.1038/s41598-020-60191-3
  22. Rahmati, O., Pourghasemi, H.R., & Zeinivand, H. (2016). Flood susceptibility mapping using frequency ratio and weights-of-evidence models in the Golastan Province, Iran. Geocarto International, 31, 42–70. https://doi.org/10.1080/10106049.2015.1041559
  23. Rutkowski, L., Jaworski, M., Pietruczuk, L., & Duda, P., (2014). The CART decision tree for mining data streams. Information Sciences, 266, 1–15. https://doi.org/10.1016/j.ins.2013.12.060
  24. Seydi, S. T., Kanani-Sadat, Y., Hasanlou, M., Sahraei, R., Chanussot, J., & Amani, M. (2022). Comparison of machine learning algorithms for flood susceptibility mapping. Remote Sensing, 15(1), 192
  25. Skilodimou, H.D., Bathrellos, G.D., Chousianitis, K., Youssef, A.M., Pradhan, B. (2019). Multi-hazard assessment modeling via multi-criteria analysis and GIS : a case study. Environmental Earth Sciences, 78, 42. https://doi.org/10.1007/s12665-018-8003-4
  26. Tehrany, M.S., Pradhan, B., & Jebur, M.N. (2015). Flood susceptibility analysis and its verification using a novel ensemble support vector machine and frequency ratio method. Stochastic Environmental Research and Risk Assessment, 29, 1149–1165. https://doi.org/10.1007/s00477-015-1021-9
  27. Tehrany, M.S., Pradhan, B., & Jebur, M.N. (2014). Flood susceptibility mapping using a novel ensemble weights-of-evidence and support vector machine models in GIS. Journal of Hydrology, 512, 332–343. https://doi.org/10.1016/j.jhydrol.2014.03.008
  28. Tehrany Shafapour, M., Pradhan, B., Mansor, S.H., & Noordin, A. (2015). Flood susceptibility assessment using GIS-based support vector machine model with different kernel types. Catena, 125, 91-101.
  29. Yousefi, H., Yonsei, H. A., Davoudi-Moghadam, D., Arshiya, A., & Shamsi, Z. (2022). Determination of flood potential using CART, GLM and GAM machine learning models (case study: Kashkan Basin). Scientific Research Journal of Irrigation and Water Engineering of Iran, 12(48), 84-105.
  30. Yousefi, H., Yonsei, H.A., Arshiya, A., Yarahamdi, Y., & Guderzi, A. (2021). Determining flood-prone areas with models to reduce flood risks (case study: Kashkan watershed). Ecohydrology, 8(1), 307-319.

 

CAPTCHA Image