Application of machine learning algorithms for flood susceptibility assessment and risk management

Assessing floods and their likely impact in climate change scenarios will enable the facilitation of sustainable management strategies. In this study, five machine learning (ML) algorithms, namely (i) Logistic Regression, (ii) Support Vector Machine, (iii) K-nearest neighbor, (iv) Adaptive Boosting (AdaBoost) and (v) Extreme Gradient Boosting (XGBoost), were tested for Greater Hyderabad Municipal Corporation (GHMC), India, to evaluate their clustering abilities to classify locations (flooded or non-flooded) for climate change scenarios. A geo-spatial database, with eight flood influencing factors, namely, rainfall, elevation, slope, distance from nearest stream, evapotranspiration, land surface temperature, normalised difference vegetation index and curve number, was developed for 2000, 2006 and 2016. XGBoost performed the best, with the highest mean area under curve score of 0.83. Hence, XGBoost was adopted to simulate the future flood locations corresponding to probable highest rainfall events under four Representative Concentration Pathways (RCPs), namely, 2.6, 4.5, 6.0 and 8.5 along with other flood influencing factors for 2040, 2056, 2050 and 2064, respectively. The resulting ranges of flood risk probabilities are predicted as 39–77%, 16–39%, 42–63% and 39–77% for the respective years.

[1]  G. Jenks.  The Data Model Concept in Statistical Mapping , 1967 .

[2]  Bahram Choubin,et al.  Ensemble Boosting and Bagging Based Machine Learning Models for Groundwater Potential Prediction , 2020, Water Resources Management.

[3]  Suresh Sankaranarayanan,et al.  Flood prediction based on weather parameters using deep learning , 2019, Journal of Water and Climate Change.

[4]  C. Onyutha,et al.  Contribution of climatic variability and human activities to stream flow changes in the Haraz River basin, northern Iran , 2019, Journal of Hydro-environment Research.

[5]  Zongxue Xu,et al.  Downscaling of daily extreme temperatures in the Yarlung Zangbo River Basin using machine learning techniques , 2018, Theoretical and Applied Climatology.

[6]  A. Das,et al.  Spatio-temporal dynamics of water resources of Hyderabad Metropolitan Area and its relationship with urbanization , 2020 .

[7]  Shiyuan Xu,et al.  A review of advances in urban flood risk analysis over China , 2015, Stochastic Environmental Research and Risk Assessment.

[8]  B. Pradhan.  Flood susceptible mapping and risk area delineation using logistic regression, GIS and remote sensing , 2010 .

[9]  Ramavarapu S. Sreenivas,et al.  Machine learning and price-based load scheduling for an optimal IoT control in the smart and frugal home , 2021 .

[10]  Nguyen Thi Thuy Linh,et al.  Flood susceptibility modelling using advanced ensemble machine learning models , 2020 .

[11]  J. Yazdi,et al.  Evaluation of data-driven models to downscale rainfall parameters from global climate models outputs: the case study of Latyan watershed , 2018, Journal of Water and Climate Change.

[12]  T. Andualem,et al.  Groundwater potential assessment using GIS and remote sensing: A case study of Guna tana landscape, upper blue Nile Basin, Ethiopia , 2019, Journal of Hydrology: Regional Studies.

[13]  Omid Bozorg-Haddad,et al.  Runoff Projection under Climate Change Conditions with Data-Mining Methods , 2017 .

[14]  C. C. Carneiro,et al.  Synthetic geochemical well logs generation using ensemble machine learning techniques for the Brazilian pre-salt reservoirs , 2021 .

[15]  Ibrahim Gad,et al.  A comparative study of prediction and classification models on NCDC weather data , 2020, International Journal of Computers and Applications.

[16]  Dong Wang,et al.  Streamflow forecasting using extreme gradient boosting model coupled with Gaussian mixture model , 2020 .

[17]  Mahyat Shafapour Tehrany,et al.  Flood susceptibility assessment using GIS-based support vector machine model with different kernel types , 2015 .

[18]  Xiaohong Chen,et al.  Flood hazard risk assessment model based on random forest , 2015 .

[19]  Mohammad Ali Ghorbani,et al.  Evaluation of daily solar radiation flux using soft computing approaches based on different meteorological information: peninsula vs continent , 2018, Theoretical and Applied Climatology.

[20]  Shahab Araghinejad,et al.  A Comparative Assessment of Artificial Neural Network, Generalized Regression Neural Network, Least-Square Support Vector Regression, and K-Nearest Neighbor Regression for Monthly Streamflow Forecasting in Linear and Nonlinear Conditions , 2018, Water Resources Management.

[21]  Romulus Costache,et al.  Flash-flood potential assessment and mapping by integrating the weights-of-evidence and frequency ratio statistical methods in GIS environment – case study: Bâsca Chiojdului River catchment (Romania) , 2017, Journal of Earth System Science.

[22]  José A. Sobrino,et al.  Changes in land surface temperatures and NDVI values over Europe between 1982 and 1999 , 2006 .

[23]  Chengquan Huang,et al.  Quality assessment of Landsat surface reflectance products using MODIS data , 2012, Comput. Geosci..

[24]  B. Pradhan,et al.  Machine learning algorithm for flash flood prediction mapping in Wadi El-Laqeita and surroundings, Central Eastern Desert, Egypt , 2021, Arabian Journal of Geosciences.

[25]  S. Sannigrahi,et al.  Analyzing the role of biophysical compositions in minimizing urban land surface temperature and urban heating , 2017, Urban Climate.

[26]  Massimo Menenti,et al.  Reconstruction of global MODIS NDVI time series: performance of harmonic analysis of time series (HANTS). , 2015 .

[27]  Nadhir Al-Ansari,et al.  Flood Detection and Susceptibility Mapping Using Sentinel-1 Remote Sensing Data and a Machine Learning Approach: Hybrid Intelligence of Bagging Ensemble Based on K-Nearest Neighbor Classifier , 2020, Remote. Sens..

[28]  Swathi Vemula,et al.  Urban floods in Hyderabad, India, under present and future rainfall scenarios: a case study , 2018, Natural Hazards.

[29]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1997, EuroCOLT.

[30]  Paresh Chandra Deka,et al.  Support vector machine applications in the field of hydrology: A review , 2014, Appl. Soft Comput..

[31]  B. Pradhan,et al.  Flood susceptibility prediction using four machine learning techniques and comparison of their performance at Wadi Qena Basin, Egypt , 2020, Natural Hazards.

[32]  C. A. Morales Rodriguez,et al.  Flash Flood Forecasting in São Paulo Using a Binary Logistic Regression Model , 2020 .

[33]  Junliang Fan,et al.  Machine learning models for the estimation of monthly mean daily reference evapotranspiration based on cross-station and synthetic data , 2019, Hydrology Research.

[34]  Alaa M. Al-Abadi,et al.  Mapping flood susceptibility in an arid region of southern Iraq using ensemble machine learning classifiers: a comparative study , 2018, Arabian Journal of Geosciences.

[35]  H. Pourghasemi,et al.  Application of GIS-based data driven random forest and maximum entropy models for groundwater potential mapping: A case study at Mehran Region, Iran , 2016 .

[36]  D. A. Sachindra,et al.  Multi-model ensemble predictions of precipitation and temperature using machine learning algorithms , 2020 .