Accurate prediction of water levels in dams is very important in planning adequate water supply and power generation from dam to provide sufficient water storage during the critical periods. Effective power planning helps in ensuring steady supply of electric power to consumers to boost industrial activities and water supply to the community. The aim of this study is to develop artificial neural network models for predicting water levels at DadinKowa Dam, which is located 5 kilometers north of the village of DadinKowa on River Gongola in Gombe state. It involves taking of a ten-year record of the daily water levels at the dam from 2007 to 2016. The daily water level data were used to develop five neural network models. The results show that the prediction accuracy of the neural network models increased with increasing input. The Five-input layers neural network model had the lowest relative error (MODEL ERR0R: 0.00165582, RMSE: 0.06671, MARE: 0.000176) while the three-input layers model had the highest relative error (MODEL ERR0R: 2.72107, RMSE: 26087.445, MARE: 3.1272099). The neural network models which involve little mathematics were much simpler to build. The developed models will be very useful in water-use planning for irrigation, municipal uses and predicting power loads and management of power generation. Timely prediction can also help in disaster monitoring, response and control of floods in Nigeria

Keywords: artificial neural network, water level, modeling, feed forward error Back propagation, RMSE, MARE