Forecasting Drought Indices using Artificial Neural Network and M5 Model Tree Techniques in Middle Gujarat Region of India
DOI:
https://doi.org/10.52151/jae2024613.1856Abstract
Developing accurate drought prediction models for drought risk assessment and management, and comprehending their effectiveness is a challenging task. This study employed artificial neural network (ANN) and M5 model tree models to forecast two drought indices, i.e., one month timescale standardized precipitation index (SPI-1) and one month timescale standardized precipitation evapotranspiration index (SPEI-1) with one-month lead time for the middle Gujarat, India using 30-year (1986-2015) gridded dataset of rainfall and temperature. The models were developed considering one to twenty hidden neurons, and trained using Levenberg-Marquardt algorithm to minimize the prediction error. Log-sigmoidal transfer functions were applied in both the hidden and output layers. Of the total data (Jan 1986-Dec 2015), 70% data (Jan 1986-Dec 2006) were used for model training, 15% data (Jan 2007-Jun 2011) for cross-validation and remaining 15% data (Jul 2011 - Dec 2015) for model testing. Results indicated that prediction accuracy of SPI-1 was better than that of SPEI-1 at one-month lead time as revealed in terms of five performance indicators, namely, coefficient of determination (R2), Nash-Sutcliffe efficiency (NSE), root mean square error (RMSE), mean absolute error (MAE), and percentage peak deviation (Pdv). It was found that ANN model for each grid was having considerably different performance in forecasting SPI-1 and SPEI-1 with R2, NSE (%), RMSE, Pdv (%), and MAE having highest values as 0.693, 46.93, 0.844, 62.65, and 0.65, respectively, for SPI-1, whereas these were 0.469, 20.06, 1.07, 90.25, 0.87, respectively, for SPEI-1. It was revealed that performance of ANN models in forecasting SPI-1 was better in comparison to SPEI-1. Amongst ANN and M5 models, ANN models performed better than the M5 model tree for most of the grids selected in this study. Different ANN models with log sigmoidal activation function in the hidden as well as in the output layer for SPI-1 drought index forecast with one-month lead time were suggested for use by scientists, irrigation planners, and policy makers.
Downloads
References
Bacanli, U. G., Firat, M., & Dikbas, F. (2009). Adaptive Neuro Fuzzy inference system for drought forecasting. Stochastic Environmental Research and Risk Assessment, 23(8), 1143-1154. https://doi.org/10.1007/s00477-008-0288-5
Bandyopadhyay, N., Bhuiyan, C., & Saha, A. K. (2015). Temperature extremes, moisture deficiency and their impacts on dryland agriculture in Gujarat, India. In: Andreu J, Solera A, Paredes-Arquiola-J, Haro-Monteagudo D, van Lanen H (eds) Drought: research and science-policy interfacing. CRC Press, Balkema, pp 119–124. https://doi.org/10.1201/b18077-20
Berhail, S., & Katipoğlu, O. M. (2023). Comparison of the SPI and SPEI as drought assessment tools in a semi-arid region: case of the Wadi Mekerra basin (northwest of Algeria). Theoretical and Applied Climatology, 154, 1373–1393. https://doi.org/10.1007/s00704-023-04601-2
Bhakar, S. R., Bansal, A. K., Chhajed, N., & Singh, R. V. (2006). Rainfall analysis for drought estimation of Udaipur region. Journal of Agricultural Engineering (India), 43(3), 40-43. https://doi.org/10.52151/jae2006433.1189
Bhalme, H. N., & Mooley, D. A. (1980). Large-scale droughts/floods and monsoon circulation. Monthly Weather Review, 108(8), 1197-1211. https://doi.org/10.1175/1520-0493(1980)108<1197:LSDAMC>2.0.CO;2
Bhattacharya, B., & Solomatine, D.P. (2005). Neural networks and M5 model trees in modelling water level-discharge relationship. Neurocomputing, 63(10), 381–396. https://doi.org/10.1016/j.neucom.2004.04.016
Danandeh Mehr, A., Torabi Haghighi, A., Jabarnejad, M., Safari, M. J. S., & Nourani, V. (2022). A new evolutionary hybrid random forest model for SPEI forecasting. Water, 14, 755. https://doi.org/10.3390/w14050755
Dawson, C .W., Abrahart R. J, & See, L. M. (2007). HydroTest: A web-based toolbox of evaluation metrics for the standardised assessment of hydrological forecasts. Environmental Modelling & Software, 22(7), 1034-1052. https://doi.org/10.1016/j.envsoft.2006.06.008
Deo, R. C., & Şahin, M. (2015). Application of the artificial neural network model for prediction of monthly standardized precipitation and evapotranspiration index using hydro-meteorological parameters and climate indices in eastern Australia. Atmospheric Research, 161, 65-81. https://doi.org/10.1016/j.atmosres.2015.03.018
Deo, R. C., Tiwari, M. K., Adamowski, J. F., & Quilty, J. M. (2017). Forecasting effective drought index using a wavelet extreme learning machine (W-ELM) model. Stochastic Environmental Research and Risk Assessment, 31(5), 1211-1240. https://doi.org/10.1007/s00477-016-1265-z
Gommes, R., & Petrassi, F. (1994). Rainfall variability and drought in Sub-Saharan Africa since 1960. Agro-meteorology series working paper 9, Food and Agriculture Organization, Rome, Italy.
Govindaraju, R. S., & Rao, A. R. (2010). Artificial Neural Networks in Hydrology. Springer Dordrecht. https://doi.org/10.1007/978-94-015-9341-0
Hagan, M. T., & Menhaj, M. B. (1994). Training feedforward networks with the Marquardt algorithm. IEEE Transactions on Neural Networks, 5(6), 989-993. https://doi.org/10.1109/72.329697
Hassoun, M. H. (1995). Fundamentals of Artificial Neural Networks, MIT press, London.
Khanna, M., Kumar, M., Rani, R., & Pathak, H. (2018). Spatio-temporal variation of hydrological drought in the Indo-Gangetic plains. Journal of Agricultural Engineering (India), 55(1), 64-71.https://doi.org/10.52151/jae2018551.1647
Kim, T. W., & Valdes, J. B. (2003). Nonlinear model for drought forecasting based on a conjunction of wavelet transforms andneural networks. Journal of Hydrologic Engineering, 8(6), 319-328. https://doi.org/10.1061/(ASCE)1084-0699(2003)8:6(319)
McKee, T. B., Doesken, N. J., & Kleist, J. (1993). The relationship of drought frequency and duration to time scales. Proceedings of the 8th Conference on Applied Climatology. Anaheim, California, 17-22 January 1993, 179-184.
Mishra, A. K., Desai, V. R., & Singh, V. P. (2007). Drought forecasting using a hybrid stochastic and neural network model. Journal of Hydrologic Engineering, 12(6), 626-638. https://doi.org/10.1061/(ASCE)1084-0699(2007)12:6(626)
Mitchell, T. M. (1997). Machine Learning. Computer Science Series. McGraw-Hill, Burr Ridg.
Mundetia, N., & Sharma, D. (2015). Analysis of rainfall and drought in Rajasthan State, India. Global NEST Journal, 17(1), 12-21. https://doi.org/10.30955/gnj.001379
Palmer, W. C. (1965). Meteorological Drought. Research Paper 45, US Weather Bureau, Office of Climatology, Washington, DC.
Pandya, P., & Gontia, N.K . (2023). Development of drought severity–duration–frequencycurves for identifying drought proneness in semi-arid regions. Journal of Water and Climate Change, 14(3), 824-842. https://doi.org/10.2166/wcc.2023.438
Paulo, A., & Pereira, L. S. (2007). Prediction of SPI drought class transitions using Markov chains. Water Resources Management , 21(10), 1813-1827. https://doi.org/10.1007/s11269-006-9129-9
Pei, Z., Fang, S., Wang, L., & Yang, W. (2020). Comparative analysis of drought indicated by the SPI and SPEI at various timescales in Inner Mongolia, China. Water, 12, 1925. https://doi.org/10.3390/w12071925
Planning Commission Report (2007) Groundwater management and ownership: Report of the expert group. Government of India, New Delhi.
Quinlan, J. R. (1992). Learning with continuous classes. In Proceedings of AI’92 (Adams & Sterling, Eds), 343-348, Singapore: World Scientific, 1992. https://sci2s.ugr.es/keel/pdf/algorithm/congreso/1992-Quinlan-AI.pdf
Shang, J., Zhao, B., Hua, H., Wei, J. Qin, G., & Chen, G. (2023). Application of informer model based on SPEI for drought forecasting. Atmosphere, 14, 951. https://doi.org/10.3390/atmos14060951
Shortridge, J. E., Guikema, S. D., & Zaitchik, B.F. (2016). Machine learning methods for empirical streamflow simulation: A comparison of model accuracy, interpretability, and uncertainty in seasonal watersheds. Hydrology and Earth System Sciences, 20(7), 2611-2628. https://doi.org/10.5194/hess-20-2611-2016
Sigdel, M., & Ikeda, M. (2010). Spatial and temporal analysis of drought in Nepal using standardized precipitation index and its relationship with climate indices. Journal of Hydrology and Meteorology, 7(1), 59-74. https://doi.org/10.3126/jhm.v7i1.5617
Tefera, A. S., Ayoade, J. O., & Bello, N. J. (2019). Comparative analyses of SPI and SPEI as drought assessment tools in Tigray Region, Northern Ethiopia. SN Applied Sciences, 1, 1265. https://doi.org/10.1007/s42452-019-1326-2
Thornthwaite, C. W. (1948). An approach toward a rational classification of climate. Geographical Review, 38(1), 55-94. http://dx.doi.org/10.2307/210739
Torabi Haghighi, A., Abou Zaki, N., Rossi, P. M., Noori, R., Hekmatzadeh, A. A., Saremi, H., & Kløve, B. (2020). Unsustainability syndrome - from meteorological to agricultural drought in arid and semi-arid regions. Water, 12, 838. https://doi.org/10.3390/w12030838
Van Loon, A. F., Gleeson, T., Clark, J., Van Dijk, A. I. J. M., Stahl, K., Hannaford, J., … & Van Lanen, H.A.J. (2016). Drought in the anthropocene. Nature Geoscience, 9, 89. https://doi.org/10.1038/ngeo2646
Vicente-Serrano, S. M., Begueria, S., & Lopez-Moreno, J. I. (2010). A multiscalar drought index sensitive to global warming: the standardized precipitation evapotranspiration index. Journal of Climate, 23(7), 1696-1718. https://doi.org/10.1175/2009JCLI2909.1
Wable, P. S., Jha, M. K., Adamala, S., Tiwari, M. K., & Biswal, S. (2023). Application of hybrid ANN techniques for drought forecasting in the semi-arid region of India. Environmental Monitoring and Assessment, 195, 1090. https://doi.org/10.1007/s10661-023-11631-w
Wagener, T., Wheater, H. S., & Gupta, H. V. (2004). Rainfall-Runoff Modelling in Gauged and Ungauged Catchments. Imperial College Press., London, UK.
Wilhite, D. A., & Glantz, M. H. (1985). Understanding: the drought phenomenon: the role of definitions. Water International, 10(3), 111-120.
Xu, D., Zhang, Q., Ding, Y., & Zhang, D. (2022). Application of a hybrid ARIMA-LSTM model based on the SPEI for drought forecasting. Environmental Science and Pollution Research, 29, 4128–4144. https://doi.org/10.1007/s11356-021-15325-z.
