CBS Tabanlı Melez Makine Öğrenmesi Uygulamalarının Ani Sel Duyarlılık Haritalamasında Kullanımı

Öz

Bu çalışmada Kentucky Nehri havzasında son yirmi yılda meydana gelen ani sel baskınları kayıtlarına dayanarak makine öğrenmesi yöntemleri kullanılarak taşkın tehlike haritalamasının yapılması amaçlanmıştır. Tahminlerin gerçekleştirilebilmesi için yaygın olarak kullanılan ve pratik bir algoritma olan rastgele orman (RF) yöntemi kullanılmıştır. Ayrıca, bu yöntemin içsel parametreleri (ağaç sayısı ve maksimum ağaç derinliği) ise parçacık sürü optimizasyonu (PSO) algoritması ile optimize edilmiştir. Bu bağlamda 343 adet geçmiş ani sel kayıtlarına ilaveten havza sınırları içerisinde yer alacak şekilde aynı sayıda rastgele nokta atanmıştır. Tüm bu noktalara 12 adet ani sel tehlikesini tetikleyecek faktörler tanıtılmış olup, tahminler bu doğrultuda gerçekleştirilmiştir. Tahmin sonuçları birçok performans değerlendirme indikatörü göz önüne alınarak analiz edildiğinde melez PSO-RF modelinin test veri setinde oldukça başarılı sonuçlar gösterdiği görülmüştür. Öyle ki hem ani sel olan noktalar hem de ani sel gerçekleşmeyen noktalar %70 oranında doğruluk ile tahmin edilmiştir. Yapılan detaylı değerlendirmeler sonucu ise ikili sınıflandırma problemlerinde önemli bir gösterge olan AUROC değeri ise 0.79 olarak hesaplanmıştır. Ayrıca, ani selleri tetikleyen faktörlerin sonuçlar üzerindeki tekil etkileri incelendiğinde şiddetli yağış faktörü en etkili değişken olarak bulunmuş olup, onu sırasıyla topoğrafya, NDVI ve eğri numarası faktörleri izlemiştir. Öte yandan, litoloji faktörünün ani sellerin modellenmesi üzerindeki etkisi ise diğer faktörlere göre oldukça az olduğu sonucuna varılmıştır. Tüm bu bulgular ışığında elde edilen sonuçlar hem taşkın tehlike haritalaması literatürüne katkı yapacak, hem de ilgili bölgede yaşanacak gelecek ani sel olayları meydana gelmeden alınması gereken tedbirler ile ilgili yol gösterici nitelikte olacaktır.

Anahtar Kelimeler

• taşkın haritalama
• rastgele orman
• parçacık sürü optimizasyonu
• makine öğrenmesi
• coğrafi bilgi sistemleri

Incorporating the GIS-Based Hybrid Machine Learning Applications into the Flash Flood Susceptibility Mapping

Öz

This study chiefly aimed to perform flash flood susceptibility mapping by means of machine learning methods based on the records attained in the Kentucky River basin over the last two decades. To carry out analysis, one of the widely adopted practical tree-based machine learning tools, i.e., the random forest (RF) method, was utilized, while the hyperparameters (i.e., number of trees and maximum tree depth) of the RF algorithm were tuned via the particle swarm optimization (PSO) strategy. In this vein, a total of 343 flash-flooded and the same number of random (non-flash flooded) points were assigned within the Kentucky River basin boundaries. In addition, a total of 12 factors triggering flash floods have been introduced to the corresponding points and the predictions were conducted in this regard. Many performance evaluation indicators considered within the scope of this study illustrated that the hybrid PSO-RF model revealed quite accurate predictive results based on the blinded testing set; such that both flash-flooded and non-flash flooded points exist in the test set were estimated with an accuracy of 70%. In addition, one of the promising performance indicators in assessing binary classification implementations, called AUROC, was calculated as 0.79. Further analysis regarding the individual impacts of the triggering factors also highlighted that the heavy rainfall probability factor was found to be the most effective variable, followed by topography, NDVI, and curve number, respectively. On the other hand, it was concluded that the effect of the lithology on the flash flood modeling is considerably lower compared to its counterparts. Overall, the results acquired in the light of all these findings have important potential in terms of both contributing to the flood susceptibility mapping literature and guiding with respect to the measures that should be taken prior to the flash flood incidents in the corresponding region.

Anahtar Kelimeler

• flood mapping
• random forest
• particle swarm optimization
• machine learning
• geographical information system

Kaynakça

1. Abedi, R., Costache, R., Shafizadeh-Moghadam, H., & Pham, Q. B. (2021). Flash-flood susceptibility mapping based on XGBoost, random forest and boosted regression trees. Geocarto International, 0(0), 1–18. https://doi.org/10.1080/10106049.2021.1920636
2. Ali, S. A., Parvin, F., Pham, Q. B., Vojtek, M., Vojteková, J., Costache, R., … Ghorbani, M. A. (2020). GIS-based comparative assessment of flood susceptibility mapping using hybrid multi-criteria decision-making approach, naïve Bayes tree, bivariate statistics and logistic regression: A case of Topľa basin, Slovakia. Ecological Indicators, 117(June), 106620. https://doi.org/10.1016/j.ecolind.2020.106620
3. Andaryani, S., Nourani, V., Haghighi, A. T., & Keesstra, S. (2021). Integration of hard and soft supervised machine learning for flood susceptibility mapping. Journal of Environmental Management, 291(April), 112731. https://doi.org/10.1016/j.jenvman.2021.112731
4. Arabameri, A., Saha, S., Mukherjee, K., Blaschke, T., Chen, W., Ngo, P. T. T., & Band, S. S. (2020). Modeling spatial flood using novel ensemble artificial intelligence approaches in northern Iran. Remote Sensing, 12(20), 1–30. https://doi.org/10.3390/rs12203423
5. Arora, A., Arabameri, A., Pandey, M., Siddiqui, M. A., Shukla, U. K., Bui, D. T., … Bhardwaj, A. (2021). Optimization of state-of-the-art fuzzy-metaheuristic ANFIS-based machine learning models for flood susceptibility prediction mapping in the Middle Ganga Plain, India. Science of the Total Environment, 750, 141565. https://doi.org/10.1016/j.scitotenv.2020.141565
6. Breiman, L. (2001). Random forests. Machine Learning, 45, 5–32. https://doi.org/10.1023/A:1010933404324
7. Catani, F., Lagomarsino, D., Segoni, S., & Tofani, V. (2013). Landslide susceptibility estimation by random forests technique: Sensitivity and scaling issues. Natural Hazards and Earth System Sciences, 13(11), 2815–2831. https://doi.org/10.5194/nhess-13-2815-2013
8. Chen, W., & Li, Y. (2020). GIS-based evaluation of landslide susceptibility using hybrid computational intelligence models. Catena, 195(December 2019), 104777. https://doi.org/10.1016/j.catena.2020.104777

9. Costache, R., Hong, H., & Pham, Q. B. (2020). Comparative assessment of the flash-flood potential within small mountain catchments using bivariate statistics and their novel hybrid integration with machine learning models. Science of the Total Environment, 711, 134514. https://doi.org/10.1016/j.scitotenv.2019.134514
10. Costache, R., Pham, Q. B., Sharifi, E., Linh, N. T. T., Abba, S. I., Vojtek, M., … Khoi, D. N. (2020). Flash-flood susceptibility assessment using multi-criteria decision making and machine learning supported by remote sensing and GIS techniques. Remote Sensing, 12(1). https://doi.org/10.3390/RS12010106
11. Costache, R., & Tien Bui, D. (2019). Spatial prediction of flood potential using new ensembles of bivariate statistics and artificial intelligence: A case study at the Putna river catchment of Romania. Science of the Total Environment, 691, 1098–1118. https://doi.org/10.1016/j.scitotenv.2019.07.197
12. Darabi, H., Torabi Haghighi, A., Rahmati, O., Jalali Shahrood, A., Rouzbeh, S., Pradhan, B., & Tien Bui, D. (2021). A hybridized model based on neural network and swarm intelligence-grey wolf algorithm for spatial prediction of urban flood-inundation. Journal of Hydrology, 603(PA), 126854. https://doi.org/10.1016/j.jhydrol.2021.126854
13. Ekmekcioğlu, Ö., Başakın, E. E., & Özger, M. (2020). Tree-based nonlinear ensemble technique to predict energy dissipation in stepped spillways. European Journal of Environmental and Civil Engineering, 0(0), 1–19. https://doi.org/10.1080/19648189.2020.1805024
14. Ekmekcioğlu, Ö., Koc, K., & Özger, M. (2021). Stakeholder perceptions in flood risk assessment: A hybrid fuzzy AHP-TOPSIS approach for Istanbul, Turkey. International Journal of Disaster Risk Reduction, 60(May). https://doi.org/10.1016/j.ijdrr.2021.102327
15. Ekmekcioğlu, Ö., & Koc, K. (2022). Explainable step-wise binary classification for the susceptibility assessment of geo-hydrological hazards. CATENA, 216, 106379. https://doi.org/10.1016/j.catena.2022.106379
16. Ekmekcioğlu, Ö., Koc, K., Özger, M., & Işık, Z. (2022). Exploring the additional value of class imbalance distributions on interpretable flash flood susceptibility prediction in the Black Warrior River basin, Alabama, United States. Journal of Hydrology, 610, 127877. https://doi.org/10.1016/j.jhydrol.2022.127877
17. Fang, Z., Wang, Y., Peng, L., & Hong, H. (2020). Integration of convolutional neural network and conventional machine learning classifiers for landslide susceptibility mapping. Computers and Geosciences, 139(February), 104470. https://doi.org/10.1016/j.cageo.2020.104470
18. Gigović, L., Pamučar, D., Bajić, Z., & Drobnjak, S. (2017). Application of GIS-interval rough AHP methodology for flood hazard mapping in Urban areas. Water (Switzerland), 9(6), 1–26. https://doi.org/10.3390/w9060360
19. Goswami, S., Murthy, C. A., & Das, A. K. (2018). Sparsity measure of a network graph: Gini index. Information Sciences, 462, 16–39. https://doi.org/10.1016/j.ins.2018.05.044
20. Habba, M., Ameur, M., & Jabrane, Y. (2018). A novel Gini index based evaluation criterion for image segmentation. Optik, 168, 446–457. https://doi.org/10.1016/j.ijleo.2018.04.045
21. Hou, C., Xie, Y., & Zhang, Z. (2022). An improved convolutional neural network based indoor localization by using Jenks natural breaks algorithm. China Communications, 19(4), 291–301. https://doi.org/10.23919/JCC.2022.04.021
22. Ikeuchi, H., Hirabayashi, Y., Yamazaki, D., Muis, S., Ward, P. J., Winsemius, H. C., … Kanae, S. (2017). Compound simulation of fluvial floods and storm surges in a global coupled river-coast flood model: Model development and its application to 2007 Cyclone Sidr in Bangladesh. Journal of Advances in Modeling Earth Systems, 9(4), 1847–1862. https://doi.org/10.1002/2017MS000943
23. Jaafar, H. H., Ahmad, F. A., & El Beyrouthy, N. (2019). GCN250, new global gridded curve numbers for hydrologic modeling and design. Scientific Data, 6(1), 1–9. https://doi.org/10.1038/s41597-019-0155-x
24. Liu, X., Zhang, Z., Jiang, T., Li, X., & Li, Y. (2021). Evaluation of the Effectiveness of Multiple Machine Learning Methods in Remote Sensing Quantitative Retrieval of Suspended Matter Concentrations: A Case Study of Nansi Lake in North China. Journal of Spectroscopy, 2021, 1–17. https://doi.org/10.1155/2021/5957376
25. Long, Y., Song, Y., & Chen, L. (2022). Identifying subcenters with a nonparametric method and ubiquitous point-of-interest data: A case study of 284 Chinese cities. Environment and Planning B: Urban Analytics and City Science, 49(1), 58–75. https://doi.org/10.1177/2399808321996705
26. Lu, Y., He, T., Xu, X., & Qiao, Z. (2021). Investigation the Robustness of Standard Classification Methods for Defining Urban Heat Islands. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 14, 11386–11394. https://doi.org/10.1109/JSTARS.2021.3124558
27. Kennedy, J., & Eberhart, R. (1995). Particle swarm optimization. Proceedings of ICNN’95 - International Conference on Neural Networks, 4, 1942–1948. IEEE. https://doi.org/10.1109/ICNN.1995.488968
28. Khosravi, K., Pham, B. T., Chapi, K., Shirzadi, A., Shahabi, H., Revhaug, I., … Tien Bui, D. (2018). A comparative assessment of decision trees algorithms for flash flood susceptibility modeling at Haraz watershed, northern Iran. Science of the Total Environment, 627, 744–755. https://doi.org/10.1016/j.scitotenv.2018.01.266
29. Kim, J., Jeong, S., & Regueiro, R. A. (2012). Instability of partially saturated soil slopes due to alteration of rainfall pattern. Engineering Geology, 147–148, 28–36. https://doi.org/10.1016/j.enggeo.2012.07.005
30. Koc, K., Ekmekcioğlu, Ö., & Özger, M. (2021). An integrated framework for the comprehensive evaluation of low impact development strategies. Journal of Environmental Management, 294, 113023. https://doi.org/10.1016/j.jenvman.2021.113023
31. Marchi, L., Borga, M., Preciso, E., & Gaume, E. (2010). Characterisation of selected extreme flash floods in Europe and implications for flood risk management. Journal of Hydrology, 394(1–2), 118–133. https://doi.org/10.1016/j.jhydrol.2010.07.017
32. NCEI. (2021). NCEI. Erişim adresi: https://www.ncei.noaa.gov/ (Erişim tarihi: 10 Kasım, 2022)
33. Ngo, P. T. T., Hoang, N. D., Pradhan, B., Nguyen, Q. K., Tran, X. T., Nguyen, Q. M., … Bui, D. T. (2018). A novel hybrid swarm optimized multilayer neural network for spatial prediction of flash floods in tropical areas using sentinel-1 SAR imagery and geospatial data. Sensors (Switzerland), 18(11). https://doi.org/10.3390/s18113704
34. Nhu, V. H., Ngo, P. T. T., Pham, T. D., Dou, J., Song, X., Hoang, N. D., … Bui, D. T. (2020). A new hybrid firefly-pso optimized random subspace tree intelligence for torrential rainfall-induced flash flood susceptible mapping. Remote Sensing, 12(17), 1–19. https://doi.org/10.3390/RS12172688
35. Norallahi, M., & Seyed Kaboli, H. (2021). Urban flood hazard mapping using machine learning models: GARP, RF, MaxEnt and NB. Natural Hazards, 106(1), 119–137. https://doi.org/10.1007/s11069-020-04453-3
36. Pham, B. T., Luu, C., Phong, T. Van, Trinh, P. T., Shirzadi, A., Renoud, S., … Clague, J. J. (2021). Can deep learning algorithms outperform benchmark machine learning algorithms in flood susceptibility modeling? Journal of Hydrology, 592(July 2020), 125615. https://doi.org/10.1016/j.jhydrol.2020.125615
37. Panahi, M., Jaafari, A., Shirzadi, A., Shahabi, H., Rahmati, O., Omidvar, E., … Bui, D. T. (2021). Deep learning neural networks for spatially explicit prediction of flash flood probability. Geoscience Frontiers, 12(3), 101076. https://doi.org/10.1016/j.gsf.2020.09.007
38. Pourghasemi, H. R., Gayen, A., Edalat, M., Zarafshar, M., & Tiefenbacher, J. P. (2020). Is multi-hazard mapping effective in assessing natural hazards and integrated watershed management? Geoscience Frontiers, 11(4), 1203–1217. https://doi.org/10.1016/j.gsf.2019.10.008
39. Rahmati, O., Falah, F., Naghibi, S. A., Biggs, T., Soltani, M., Deo, R. C., … Tien Bui, D. (2019). Land subsidence modelling using tree-based machine learning algorithms. Science of the Total Environment, 672, 239–252. https://doi.org/10.1016/j.scitotenv.2019.03.496
40. Riley, S. J., De Gloria, S. D., & Elliot, R. (1999). A Terrain Ruggedness that Quantifies Topographic Heterogeneity. Intermountain Journal of Sciences, 5(1–4), 23–27.
41. Saha, S., Sarkar, R., Thapa, G., & Roy, J. (2021). Modeling gully erosion susceptibility in Phuentsholing, Bhutan using deep learning and basic machine learning algorithms. Environmental Earth Sciences, 80(8), 1–21. https://doi.org/10.1007/s12665-021-09599-2
42. Shahabi, H., Shirzadi, A., Ronoud, S., Asadi, S., Pham, B. T., Mansouripour, F., … Bui, D. T. (2021). Flash flood susceptibility mapping using a novel deep learning model based on deep belief network, back propagation and genetic algorithm. Geoscience Frontiers, 12(3), 101100. https://doi.org/10.1016/j.gsf.2020.10.007
43. Talukdar, S., Ghose, B., Shahfahad, Salam, R., Mahato, S., Pham, Q. B., … Avand, M. (2020). Flood susceptibility modeling in Teesta River basin, Bangladesh using novel ensembles of bagging algorithms. Stochastic Environmental Research and Risk Assessment, 34(12), 2277–2300. https://doi.org/10.1007/s00477-020-01862-5
44. Tang, X., Li, J., Liu, M., Liu, W., & Hong, H. (2020). Flood susceptibility assessment based on a novel random Naïve Bayes method: A comparison between different factor discretization methods. Catena, 190(March), 104536. https://doi.org/10.1016/j.catena.2020.104536
45. Thieken, A. H., Petrow, T., Kreibich, H., & Merz, B. (2006). Insurability and Mitigation of Flood Losses in Private Households in Germany. Risk Analysis, 26(2), 383–395. https://doi.org/10.1111/j.1539-6924.2006.00741.x
46. Tien Bui, D., Hoang, N. D., Pham, T. D., Ngo, P. T. T., Hoa, P. V., Minh, N. Q., … Samui, P. (2019). A new intelligence approach based on GIS-based Multivariate Adaptive Regression Splines and metaheuristic optimization for predicting flash flood susceptible areas at high-frequency tropical typhoon area. Journal of Hydrology, 575(April), 314–326. https://doi.org/10.1016/j.jhydrol.2019.05.046
47. Tien Bui, D., Hoang, N. D., Martínez-Álvarez, F., Ngo, P. T. T., Hoa, P. V., Pham, T. D., … Costache, R. (2020). A novel deep learning neural network approach for predicting flash flood susceptibility: A case study at a high frequency tropical storm area. Science of the Total Environment, 701, 134413. https://doi.org/10.1016/j.scitotenv.2019.134413
48. Xu, H., Fan, G., & Song, Y. (2022). Novel Key Indicators Selection Method of Financial Fraud Prediction Model Based on Machine Learning Hybrid Mode. Mobile Information Systems, 2022, 1–12. https://doi.org/10.1155/2022/6542652