Daily rainfall prediction for Bihar using artificial neural networks
Prediction of rainfall using ANN
DOI:
https://doi.org/10.21921/jas.v9i04.%2011596Keywords:
Rainfall, Streamflow, ANN model, Flood controlAbstract
Accurate daily rainfall prediction is required for accurate stream flow prediction, flooding risk analysis and construction of reliable flood control and early warning system. However, because of its nonlinearity, the prediction of daily rainfall with high accuracy and long prediction lead time is difficult. In this study, the artificial neural network (ANN) model was applied to predict the daily rainfall using six meteorological parameters (minimum and maximum temperature, morning and evening relative humidity, pan evaporation and rainfall) for a period of 2015-19 were used. The three ANN models were trained and tested by using 1 (ANN-1), 3 (ANN-3) and 5 (ANN-5) days preceding meteorological parameters. For the development of the ANN models, 70% of data is chosen for the training process, and the remaining 30% of data is chosen for the testing phase. It was found that the ANN-3 and ANN-5were a promising algorithm to predict daily rainfall. The model trained with one-day preceding information was a very poor performance to predict daily rainfall. The ANN-5 model had RMSE, MARE and MAE as 0.42, 0.04, 0.10 for training and 1.60, 0.07, and 0.24 for testing, respectively, which was the lowest as compared to ANN-1 and ANN-3 models. Nonetheless, the NSE and KGE were greater than 0.92 for both training and testing. The PBias were only positive for the ANN-5model which was 1.35 and 3.90 for training and testing, respectively. The ANN model would be helpful in the quick and accurate prediction of daily rainfall.
References
Abadi M, Agarwal A, Barham P, Brevdo E, Chen Z, Citro C, and Zheng X. 2016. Tensorflow: Large-scale machine learning on heterogeneous distributed systems. arXiv preprint arXiv,1603.04467.
Agarwal A and Singh R D. 2004. Runoff modelling through back propagation artificial neural network with variable rainfall-runoff data. Water resources management 18(3): 285-300.
Altunkaynak A and Nigussie T A. 2015. Prediction of daily rainfall by a hybrid wavelet-season-neuro technique. Journal of Hydrology 529:287-301.
Castellano-Méndez M, González-Manteiga W, Febrero-Bande M, Prada-Sánchez J M and Lozano-Calderón R. 2004. Modelling of the monthly and daily behaviour of the runoff of the Xallas river using Box–Jenkins and neural networks methods. Journal of hydrology 296(1-4): 38-58.
Chollet F and et al. 2015. Keras. GitHub. Retrieved from https://github.com/fchollet/keras.
Commeh M K, Agyei-Agyemang A, Tawiah P O and Asaaga, B A. 2022. CFD analysis of a flat bottom institutional cookstove. Scientific African 16: e01117.
Ejigu Y B and Nigatu H M. 2021. Rainfall Prediction and Cropping Pattern Recommendation Using Artificial Neural Network. In International Conference on Advances of Science and Technology: Springer, p:500-516.
Gaume E and Gosset R. 2003. Over-parameterisation, a major obstacle to the use of artificial neural networks in hydrology? Hydrology and Earth System Sciences 7(5): 693-706.
Ghumman A R, Ghazaw Y M, Sohail A R and Watanabe K. 2011. Runoff forecasting by artificial neural network and conventional model. Alexandria Engineering Journal 50(4): 345-350.
Hall T, Brooks H E and Doswell C A. 1999. Precipitation forecasting using a neural network. Weather and Forecasting 14(3):338-345.
Harris C R, Millman K J, vanderWalt S J et al. 2020. Array programming with NumPy. Nature585:357–362.
Hunter J D. 2007. Matplotlib: A 2D Graphics Environment, Computing in Science & Engineering9(3):90-95.
Lee J, Kim C G, Lee J E, Kim N W and Kim H. 2018. Application of artificial neural networks to rainfall forecasting in the Geum River basin, Korea. Water 10(10):1448.
McKinney W and et al. 2010. Data structures for statistical computing in python. In Proceedings of the 9th Python in Science Conference, 445:51-56.
Ojo O S and Ogunjo S T. 2022. Machine learning models for prediction of rainfall over Nigeria. Scientific African 16: e01246.
Parthsarthi P and Singh A K. 2013. A simple approach about the characteristics of available surface water in Bihar state of India. Geosciences 3(2):68-76.
Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M and Duchesnay E. 2011. Scikit-learn: Machine Learning in Python, J. Mach. Learn. Res.,12:2825–2830.
Radda I A, Kumar B M and Pathak P. 2021. Land degradation in Bihar, India: an assessment using rain-use efficiency and residual trend analysis. Agricultural Research 10(3):434-447.
Roushangar K, Alizadeh F and Adamowski J. 2018. Exploring the effects of climatic variables on monthly precipitation variation using a continuous wavelet-based multiscale entropy approach. Environmental Research 165:176-192.
Sarangi A and Bhattacharya A K. 2005. Comparison of artificial neural network and regression models for sediment loss prediction from Banha watershed in India. Agricultural Water Management 78(3):195-208.
Sivakumar B, Jayawardena A W and Fernando T M K G. 2002. River flow forecasting: use of phase-space reconstruction and artificial neural networks approaches. Journal of Hydrology 265: 225-245.
Sohail A, Watanabe K and Takeuchi S. 2008. Runoff analysis for a small watershed of Tono area Japan by back propagation artificial neural network with seasonal data. Water Resources Management 22(1): 1-22.
Suryaningtyas L S, Ery S and Rispiningtati R. 2020. Hydrological analysis of TRMM (Tropical rainfall measuring mission) data in Lesti sub watershed. Civil and Env. Science Journal3:18-30.
Tripathy S S, Karmakar S and Ghosh S. 2021. Hazard at weather scale for extreme rainfall forecast reduces uncertainty. Water Security 14:100106.
VanRossum G and DrakeJ F L. 1995. Python tutorial. Centrum voorWiskundeen Informatica Amsterdam, The Netherlands.
Yen M H, Liu D W, Hsin Y C, Lin C E and Chen C C. 2019. Application of the deep learning for the prediction of rainfall in Southern Taiwan. Scientific Reports 9(1): 1-9.
