[01] David Lomeling, Department of Agricultural Sciences, College of Natural Sciences and Environmental Studies (CNRES), University of Juba, Juba, South Sudan. [02] Salah Joseph Huria, Department of Agricultural Sciences, College of Natural Sciences and Environmental Studies (CNRES), University of Juba, Juba, South Sudan.

A simple Feed-Forward Neural Network (FFNN) model with a learning back-propagation algorithm was applied to forecast drought patterns derived from rainfall data of Juba County, South Sudan from 1997-2016. The annual rainfall data were aggregated into three seasons MAMJ, JAS and OND and later trained for best predictions for the period 2017-2034 using the Alyuda Forecaster XL software. Best training was attained once the minimum error of the weight ∆W and expressed as Mean Square Error between the measured and estimated values. Drought expressed as SPI was derived by fitting the respective CDFs to the rainfall amounts of each season. The results showed that for MAMJ and JAS months, the number forecasts were over 85% whereas this was between 60-80% for OND months. Rainfall forecast showed that the MAMJ months for the years 2019 to 2027 will be moderately wet with near to normal drought except in April 2021 which will experience some severe wetness. Interdecadal severe drought is expected between 2028 to 2033 after almost two decades. Declining trend per decade of SPI for all seasons was significant at p<0.01 whereas JAS and OND seasonal decrease in the next 100 years is forecasted to remain within the near to normal range while MAMJ is forecasted to have moderate drought.

Feed Forward Neural Network, Drought Forecasting, Cumulative Distribution Function, Training Set, Standard Precipitation Index

[01] Fer, I., Tietjen, B., Jeltsch, F. and Wolff, C. (2017). The influence of El Niño–Southern Oscillation regimes on eastern African vegetation and its future implications under the RCP8.5 warming scenario. Biogeosciences, 14, 4355–4374. https://doi.org/10.5194/bg-14-4355-2017. [02] Li, B., Wang, L., Kaseke, K. F., Li, L., Seely, M. K. (2016). The Impact of Rainfall on Soil Moisture Dynamics in a Foggy Desert. PLoS ONE 11 (10) e0164982. doi: 10.1371/journal.pone.0164982. [03] Narasimhan, B., Srinivasan, R. (2005). Development and evaluation of Soil Moisture Deficit Index (SMDI) and Evapotranspiration Deficit Index (ETDI) for agricultural drought monitoring. Agric. For. Meteorol. 133, 69–88. [04] Dutra, E., Viterbo, P. and Miranda P. M. A. (2008). Era-40 reanalysis hydrological applications in the characterization of regional drought Geophys. Res. Lett. 35, L19402. [05] Njoku, E., Jackson, T., Lakshmi, V., Chan, T., Nghiem, S. (2003). Soil moisture retrieval from AMSR-E, IEEE Geosc. Remote Sens. Let. 41 (2), 215–229. doi: 10.1109/TGRS.2002.808243. [06] Owe, M., de Jeu, R., Holmes, T. (2008). Multi-sensor historical climatology of satellite-derived global land surface moisture, J. Geophys. Res., 113 F01002, doi: 10.1029/2007JF000769. [07] Abbot, J. and Marohasy, J. (2017a). Forecasting extreme monthly rainfall events in regions of Queensland, Australia using artificial neural networks. Int. J. Sustain. Dev. Plan. 12 (7), 1117–1131. [08] Hartmann H, Snow JA, Su B (2016) Seasonal predictions of precipitation in the Aksu-Tarim River basin for improved water resources management. Glob. Planet. Chang. 147: 86–96. [09] Devi, S. R. P., Arulmozhiyarman, C., Venktash, C. and Pramay, A. (2016). Performance comparison of artificial neural network models for daily rainfall prediction. International Journal of Automation and Computing, 13 (5), 417-427. [10] Hung, N. Q, Babel, M. S, Weesakul, S. and Tripathi, N. K. (2009). An artificial neural network model for rainfall forecasting in Bangkok Thailand Hydrol. Earth Syst. Sci. 13, 1413–1425. [11] Warsito, B., Gernowo, R., Sugiharto, A. 2016. Rainfall prediction by using wavelet general regression neural network. Int. J. Appl. Math. Stat. 54 (3), 32–41. [12] Haviluddin, M., Hardwinarto, S., Sumaryono, M. A. (2015). Rainfall Monthly Prediction Based on Artificial Neural Network: A Case Study in Tenggarong Station, East Kalimantan -Indonesia Procedia Computer Science 59, 142 – 151. [13] Abbot, J. and Marohasy, J. (2017b). Skillful rainfall forecasts from artificial neural networks with long duration series and single month optimization. Atmospheric Research 197, 289-299. [14] Elsanabary, M. H. and Gan, T. Y. (2014). Wavelet analysis of seasonal rainfall variability of the Upper Blue Nile Basin, its teleconnection to global sea surface temperature, and its forecasting by an artificial neural network. Mon. Weather Rev., 142, 1771–1791. [15] Benmahdjoub, K., Amour, Z. and Boulifa, M.- (2013). Forecasting of rainfall using time delay neural network in Tizi-Ouzou (Algeria). Terra-green 13 international conference 2013- advancements in renewable energy and clean environment Energy Procedia 36, 1138–1146. [16] Badr, H. S., Zaitchik, B. F. and Guikema, S. D. (2014). Application of statistical models to the prediction of seasonal rainfall anomalies over the Sahel. J. Appl. Meteorol. Climatol. 53 (3), 614–636. [17] Hayes, M., Svoboda, M., Wall, N. and Widhalm, M. (2011). The Lincoln Declaration on Drought Indices: Universal Meteorological Drought Index Recommended. Bull. Amer. Meteor. Soc. 92, 485–488. doi: http://dx.doi.org/10.1175/2010BAMS3103.1. [18] McKee, T. B., Doesken, N. J. and Kleist, J. (1993). The Relationship of Drought Frequency and Duration to Time Scales. Eighth Conference on Applied Climatology, 17-22 January 1993, Anaheim, California. [19] Christiansen, N. H, Voie, N. H., Winther, P. E. T., Høgsberg, O. J. (2014). Comparison of Neural Network Error Measures for Simulation of Slender Marine Structures, Journal of Applied Mathematics, Article ID 759834, 11 pp. http://dx.doi.org/10.1155/2014/759834. [20] Nakama T. Comparisons of Single- and Multiple Hidden-Layer Neural Networks. In: Liu D, Zhang H, Polycarpou M, Alippi C, He H. (eds) Advances in Neural Networks – ISNN 2011. ISNN 2011. Lecture Notes in Computer Science, vol 6675. Springer, Berlin, Heidelberg. [21] Sonntag, E. D. (1992). Feedback stabilization using two-hidden-layer nets. IEEE Transactions on Neural Networks 3, 981–990. [22] Lolli, F., Gamberini, R., Regattieri, A., Balugani, E., Gatos, T., Gucc, S. 2016. Single-hidden layer neural networksfor forecasting intermittent demands. International Journal of Production Economics 183 (A), (116-128). https://doi.org/10.1016/j.ijpe.2016.10.021. [23] Mahmoud, O., Anwar, O., Jimoh F. E. S. M. 2007. Learning Algorithm Effect on Multilayer Feed Forward Artificial Neural Network Performance in Image Coding. Journal of Engineering Science and Technology, 2 (2), 188–199. [24] Shaikhina, T., Khovanova, N. A. 2017. Handling limited datasets with neural networks in medical applications: A small-data approach. Artificial Intelligence in Medicine, 75, 51-63. http://dx.doi.org/10.1016/j.artmed.2016.12.003. [25] Forman, G., Cohen, I. 2004. Learning from little: comparison of classifiers given little training. Proc. PKDD 19, 161–72. [26] Rowell, D, P., Booth, B. B. B., Nicholson, S. E., Good, P. 2015. Reconciling past and future rainfall trends over East Africa. J. Climate, 28, 9768–9788. https://doi.org/10.1175/JCLI-D-15-0140.1. [27] Lomeling, D., Modi, A. L., Moti, S. K., Kenyi, M. C., Silvestro, G. M., Yieb, J. L. L. (2016). Comparing the Macro-aggregate Stability of Two Tropical Soils: Clay Soil (Eutric vertisol) and Sandy Loam Soil (Eutric leptosol). International Journal of Agriculture and Forestry, 6 (4): 142-151.