Modelling of Flood in Lagos using Artificial Neural Network

Abstract

Flooding events have caused havoc to live and properties in recent times. Flooding could be caused by several factors some of which are rainfall, temperature, and relative humidity, which lead to rise in water level. The method of artificial neural networks (ANN) was used in this work to model flood occurrence in Lagos, Nigeria. Meteorological data were collected from the Nigeria Meteorological Agency (NIMET) and Nigeria Hydrological survey Agency (NHSA) for 54 years. Artificial neural networks are known to have capacity for pattern recognition and have been proven to be reliable predictive tools for modelling of flood. The Levenberg-Marquardt backpropagation algorithm was used with NARX (non-linear autoregressive network with exogenous inputs). Predictions from the neural network model were checked and validated using tests of correlation coefficients (R), coefficients of determination (R²), and mean square errors (MSE) between the observations and predictions. Forecast predictions from the neural network model reveal on the average that there will be increase in rainfall (and corresponding increase in the rise in water level) for the period from 2017 to 2020. 

Keywords: ANN-artificial neural network, flood, rainfall, temperature, relative humidity, Lagos

Cite this Article

I.D. Ekpa, S.O. Udo, Obu J.A. et al. Modelling of Flood in Lagos using Artificial Neural Network. Research & Reviews: Journal of Space Science & Technology. 2018; 7(2): 50–57p.