Machine Learning Methods for Flood Hazard Mapping

Authors: S. Zappacosta, C. Bove, M. Carmela Marinelli, P. di Lauro, K. Spasenovic, L. Ostano, G. Aiello, M. Pietrosanto

Abstract:

This paper proposes a neural network approach for assessing flood hazard mapping. The core of the model is a machine learning component fed by frequency ratios, namely statistical correlations between flood event occurrences and a selected number of topographic properties. The classification capability was compared with the flood hazard mapping River Basin Plans (Piani Assetto Idrogeologico, acronimed as PAI) designed by the Italian Institute for Environmental Research and Defence, ISPRA (Istituto Superiore per la Protezione e la Ricerca Ambientale), encoding four different increasing flood hazard levels. The study area of Piemonte, an Italian region, has been considered without loss of generality. The frequency ratios may be used as a standalone block to model the flood hazard mapping. Nevertheless, the mixture with a neural network improves the classification power of several percentage points, and may be proposed as a basic tool to model the flood hazard map in a wider scope.

Keywords: flood modeling, hazard map, neural networks, hydrogeological risk, flood risk assessment

References:

[1] D. Judi, C. Rakowski, S. Waichler, Y. Feng and M. Wigmosta, «Integrated Modeling Approach for the Development of Climate-Informed, Actionable Information,» Water, vol. 10, n. 6, p. 775, 13 June 2018.
[2] S. A. Schumm and R. W. Lichty, «Time, space, and causality in geomorphology,» American Journal of Science, vol. 263, n. 2, pp. 110-119, February 1965.
[3] K. Li, S. Wu, E. Dai and Z. Xu, «Flood loss analysis and quantitative risk assessment in China,» Natural Hazards, vol. 63, n. 2, pp. 737-760, September 2012.
[4] G. Testa, D. Zuccalà, F. Alcrudo, J. Mulet and S. Soares-Frazão, «Flash flood flow experiment in a simplified urban district,» Journal of Hydraulic Research, vol. 45, pp. 37-44, 2007.
[5] E. van Beek and J. van Alphen, «From flood defence to flood management - prerequisites for sustainable flood management,» in Floods, from defence to management: Proceedings of the 3rd International Symposium on Flood Defence, Leiden, 2006.
[6] W.-H. Teng, M.-H. Hsu, C.-H. Wu and A. S. Chen, «Impact of Flood Disasters on Taiwan in the Last Quarter Century,» Natural Hazards, vol. 37, n. 1-2, pp. 191-207, 2006.
[7] C. Kousky, «Financing Flood Losses: A Discussion of the National Flood Insurance Program,» vol. 21, n. 1, pp. 11-32, 2018.
[8] D. R. Dassanayake, A. Burzel and H. Oumeraci, «Methods for the Evaluation of Intangible Flood Losses and Their Integration in Flood Risk Analysis,» Coastal Engineering Journal, vol. 57, n. 1, pp. 1-35, 2015.
[9] V. Meyer, S. Scheuer e and D. Haase, «A multicriteria approach for flood risk mapping exemplified at the Mulde river, Germany,» Natural Hazards, vol. 48, n. 1, pp. 17-39, 2009.
[10] M.-J. Lee, J.-e. Kang and S. Jeon, «Application of frequency ratio model and validation for predictive flooded area susceptibility mapping using GIS,» in 2012 IEEE International Geoscience and Remote Sensing Symposium, July, 2012.
[11] A. K. Kar, A. Lohani, N. Goel and G. Roy, «Rain gauge network design for flood forecasting using multi-criteria decision analysis and clustering techniques in lower Mahanadi river basin, India,» Journal of Hydrology: Regional Studies, vol. 4, pp. 313-332, September 2015.
[12] M. Ahmadlou, A. Al‐Fugara, A. R. Al‐Shabeeb, A. Arora, R. Al‐Adamat, Q. B. Pham, N. Al‐Ansari, N. T. T. Linh and H. Sajedi, «Flood susceptibility mapping and assessment using a novel deep learning model combining multilayer perceptron and autoencoder neural networks,» Journal of Flood Risk Management, vol. 14, n. 1, March 2021.
[13] H. Mojaddadi Rizeei, B. Pradhan, H. Nampak, N. Ahmad and A. Halim bin Ghazali, «Ensemble machine-learning-based geospatial approach for flood risk assessment using multi-sensor remote-sensing data and GIS,» Geomatics, Natural Hazards and Risk, 1 3 2017.
[14] M. E. Ritter, The physical environment: An introduction to physical geography, 2003.
[15] A. Piloyan and M. Konečný, «Semi-Automated Classification of Landform Elements in Armenia Based on SRTM DEM using K-Means Unsupervised Classification,» Quaestiones Geographicae, vol. 36, n. 1, pp. 93-103, March 2017.
[16] I. D. Moore and G. J. Burch, «Physical Basis of the Length-slope Factor in the Universal Soil Loss Equation,» vol. 50, n. 5, pp. 1294-1298, 1986.
[17] S. J. Riley, S. D. Degloria e R. Elliot, «A Terrain Ruggedness Index that Quantifies Topographic Heterogeneity,» Intermountain Journal of Science, vol. 5, pp. 23-27, 1999.
[18] A. R. Rasyid, N. P. Bhandary and R. Yatabe, «Performance of frequency ratio and logistic regression model in creating GIS based landslides susceptibility map at Lompobattang Mountain, Indonesia,» Geoenvironmental Disasters, vol. 3, n. 1, p. 19, December 2016.
[19] Soil Science Division Staff, «Soil survey manual,» United States Department of Agriculture, Washington, D.C., 1993.
[20] «Soil Grids,» (Online). Available: https://soilgrids.org/