ARTIFICIAL NEURAL NETWORK TECHNIQUE FOR RAINGAUGE BASED RAINFALL NOWCASING

Rainfall forecasting and nowcasting are of great importance, for instance, in real-time flood early warning systems. Long term rainfall forecasting demands global climate, land, and sea data, thus, large computing power and storage capacity are required. Rainfall nowcasting's computing requirement, on the other hand, is much less demanding. Rainfall nowcasting may use data captured by radar and/or weather stations. This paper presents the application of Artificial Neural Network (ANN) on rainfall nowcasting using data purely observed at weather and/or rainfall stations. The study focuses on the North-East monsoon period (December, January and February) for certain locations in Singapore. Rainfall and weather data of ten stations, between 2000 and 2006, were selected and divided into three groups, training, over-fitting test, and validation data sets. Several neural network architectures were tried. Two architectures, Backpropagation ANN and Group Method of Data Handling ANN, yielded better rainfall nowcasting, up to two hours, than the other architectures. The obtained rainfall forecasts were then used by a catchment model to forecast catchment runoff. Results are encouraging and promising; together with ANN's high computational speed, the proposed approach may be considered for the real-time flood early warning system.