Predicting Seasonal Rainfall Patterns and Trends in Juba County, South Sudan Using Artificial Neural Networks



Abstract Views
309

Downloads
388

Citations

Share

##plugins.themes.bootstrap3.article.sidebar##

Submitted Mar 30, 2020
Published Apr 16, 2020
10.24018/ejfood.2020.2.2.32

Abstract viewed = 309 times

Table of Contents

##plugins.themes.bootstrap3.article.main##

A simple Feed Forward Neural Network (FFNN) model with a learning back-propagation algorithm was applied to forecast rainfall data from 1997-2016 of Juba County, South Sudan. 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 or cost function of the weight  was attained during gradient descent and expressed as Mean Square Error (MSE) and AE of the input variable.  The results showed that for MAMJ and JAS months, the number good forecasts were over 97% whereas this was between 60-80% for OND months. The Seasonal Kendal (SK) test on future rainfall forecasts as well as the Theil-Sen slope showed a declining monotonic trend in the mean amounts in all three seasons with MAMJ, JAS at OND at 100, 150 and 80 mm respectively towards the end of 2034.  Declining onset of MAMJ rains is expected to significantly affect the timing for land preparation and crop planting. The forecast accuracy of the FFNN can be used as a vital tool for decision makers in projecting future rainfall events.

Keywords: Feed Forward Neural Network, Land Preparation, Monotonic Trend, Rainfall Prediction.

References

  1. D. Lomeling, A. L. Modi, S. K. Moti, M. C. Kenyi, G. M. Silvestro, J. L. L. Yieb. 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 vol. 6 no. 4:142-151. 2016.
     Google Scholar
  2. B. Li, L. Wang, K. F. Kaseke, L. Li, M. K. Seely. The Impact of Rainfall on Soil Moisture Dynamics in a Foggy Desert.
     Google Scholar
  3. PLOS ONE, vol. 11 no. 10 e0164982.
     Google Scholar
  4. doi: 10.1371/journal.pone.0164982. 2016.
     Google Scholar
  5. M. Burke, A. de Janvry, J. Quintero-. Providing index-based
     Google Scholar
  6. Agricultural insurance to smallholders: Recent progress and
     Google Scholar
  7. future promise. 2015. Available at:
     Google Scholar
  8. http://siteresources.worldbank.org/EXTABCDE/Resources/7455676- 1292528456380/7626791-1303141641402/7878676-1306270833789/Parallel-Session-5-Alain_de_Janvry.pdf. Accessed 18 Nov 2015.
     Google Scholar
  9. B. Narasimhan and R. Srinivasan. Development and evaluation of soil moisture deficit index (SMDI) and evapotranspiration deficit index (ETDI) for agricultural drought monitoring. Agric. For. Meteorol., vol. 133: 69–88. 2005.
     Google Scholar
  10. E. Dutra, P. Viterbo, P. M. A. Miranda. Era-40 reanalysis hydrological applications in the characterization of regional drought. Geophys. Res. Lett. 35, L19402. 2008.
     Google Scholar
  11. E. Njoku, T. Jackson, V. Lakshmi, T. Chan, S. Nghiem. Soil
     Google Scholar
  12. moisture retrieval from AMSR-E, IEEE Geosc. Remote Sens. Let., vol. 41 no. 2: 215–229. doi:10.1109/TGRS.2002.808243. 2003.
     Google Scholar
  13. M. Owe, R. de Jeu, T. Holmes. Multi-sensor historical climatology of satellite-derived global land surface moisture, J. Geophys. Res., vol. 113, F01002. doi:10.1029/2007JF000769.2008.
     Google Scholar
  14. P. Singh. Indian summer monsoon rainfall (ISMR) forecasting
     Google Scholar
  15. using time series data: A fuzzy-entropy-neuro based expert system. Geoscience Frontiers, vol. 9: 1243-1257. 2018.
     Google Scholar
  16. H. Hartmann, J. A. Snow, B. Su. Seasonal predictions of precipitation in the Aksu-Tarim River basin for improved water
     Google Scholar
  17. resources management. Glob. Planet. Chang., vol. 147: 86–96. 2016.
     Google Scholar
  18. S. R. P. Devi, C. Arulmozhiyarman, C. Venktash, A. Pramay.A.
     Google Scholar
  19. Performance comparison of artificial neural network models for daily
     Google Scholar
  20. rainfall prediction. International Journal of Automation and
     Google Scholar
  21. Computing, vol. 13 no. 5: 417-427. 2016.
     Google Scholar
  22. N. Q. Hung, M. S. Babel, S. Weesakul, N. K. Tripathi. An artificial
     Google Scholar
  23. neural network model for rainfall forecasting in Bangkok, Thailand,
     Google Scholar
  24. Hydrol. Earth Syst. Sci., vol. 13: 1413– 1425. 2009.
     Google Scholar
  25. B. Warsito, R. Gernowo, A. Sugiharto. Rainfall prediction by using
     Google Scholar
  26. wavelet general regression neural network. Int. J. Appl. Math. Stat.,
     Google Scholar
  27. vol. 54 no. 3: 32–41. 2016.
     Google Scholar
  28. M. Haviluddin, S. Hardwinarto, M. A. Sumaryono. 2015. Rainfall
     Google Scholar
  29. Monthly Prediction Based on Artificial Neural Network: A Case
     Google Scholar
  30. Study in Tenggarong Station, East Kalimantan –Indonesia. Procedia
     Google Scholar
  31. Computer Science, vol. 59: 142 – 151. 2015.
     Google Scholar
  32. J. Abbot, J and J. Marohasy. Skilful rainfall forecasts from artificial
     Google Scholar
  33. neural networks with long duration series and single month
     Google Scholar
  34. optimization. Atmospheric Research, vol. 197: 289-299. 2017b.
     Google Scholar
  35. M. H. Elsanabary and T. Y. Gan. Wavelet analysis of seasonal
     Google Scholar
  36. rainfall variability of the Upper Blue Nile Basin, its teleconnection to
     Google Scholar
  37. global sea surface temperature, and its forecasting by an artificial
     Google Scholar
  38. neural network. Mon. Weather Rev., vol.142 no. 5: 1771–1791.
     Google Scholar

  39.  Google Scholar
  40. K. Benmahdjoub, Z. Amour, M. Boulifa. Forecasting of rainfall using
     Google Scholar
  41. time delay neural network in Tizi-Ouzou (Algeria). Terragreen 13
     Google Scholar
  42. International Conference 2013- Advancements in Renewable Energy
     Google Scholar
  43. and Clean Environment. Energy Procedia, vol. 36: 1138–1146.
     Google Scholar

  44.  Google Scholar
  45. H. S. Badr, B. F. Zaitchik, S. D. Guikema. Application of
     Google Scholar
  46. statistical models to the prediction of seasonal rainfall anomalies over the Sahel. J. Appl. Meteorol. Climatol., vol. 3: 614–636. 2014.
     Google Scholar
  47. M. Zeynoddin, H. Bonakdari, A. Azari, I. Ebtehaj, B. Gharabaghi,
     Google Scholar
  48. H. R. Madavar. Novel hybrid linear stochastic with non-Linear extreme machine learning method. J Environ. Manage. vol. 15 no.
     Google Scholar
  49. : 190-206. doi: 10.1016/j.jenvman.2018.05.072. 2018.
     Google Scholar
  50. J. L. Ng, S. A. Aziz, H. F. Huang, Y. Wayayok, M. K. Rowshon.
     Google Scholar
  51. Stochastic modelling of seasonal and yearly rainfalls with low-
     Google Scholar
  52. frequency variability. Stoch. Environ. Res. Risk A. vol, 31: 2215–
     Google Scholar
  53. https://doi.org/10.1007/s00477-016- 1373-9. 2017.
     Google Scholar
  54. Z. M. Yaseen, M I. Ghareb, I. Ebtehaj, J. Bonakdari, R. Siddique,
     Google Scholar
  55. S. Heddam, A. A. Yusif, R. Deo. Rainfall pattern forecasting using
     Google Scholar
  56. novel hybrid intelligent model based ANFIS-FFA. Water Resour.
     Google Scholar
  57. Manag., vol. 32 no. 1: 105-122.
     Google Scholar
  58. https://doi.org/10.1007/s11269-017-1797-0. 2018.
     Google Scholar
  59. O. I. Osarumwense. Applicability of box Jenkins SARIMA model in
     Google Scholar
  60. rainfall forecasting: a case study of Port-Harcourt South Nigeria.
     Google Scholar
  61. Can. J. Comput. Math. Nat. Sci. Eng. Med., vol. 4: 1–4. 2013.
     Google Scholar
  62. D. Lomeling, S. W. Kenyi. Forecasting solid waste generation in Juba
     Google Scholar
  63. Town, South Sudan using Artificial Neural Networks (ANNs) and
     Google Scholar
  64. Autoregressive Moving Averages (ARMA). Journal of Environment
     Google Scholar
  65. and Waste Management, vol. 4 no. 2: 211-223. 2017.
     Google Scholar
  66. H, Moeeni, B. Hossein, I. Ebtehaj. Integrated SARIMA with
     Google Scholar
  67. Neuro-Fuzzy Systems and Neural Networks for Monthly
     Google Scholar
  68. Inflow Prediction. Water Resources Manage., vol. 37 no. 1: 1- 17.
     Google Scholar
  69. doi 10.1007/s11269-017-1632-7. 2017.
     Google Scholar
  70. M. A. Khamis and K. A. Fattah. Estimating oil–gas ratio
     Google Scholar
  71. for volatile oil and gas condensate reservoirs: artificial neural
     Google Scholar
  72. network, support vector machines and functional network approach.
     Google Scholar
  73. Journal of Petroleum Exploration and Production Technology, vol. 9:
     Google Scholar
  74. -582. doi.org/10.1007/s13202-018-0501-0. 2018.
     Google Scholar
  75. P. Mirzaie and S. Salavati. The Application of Artificial Neural
     Google Scholar
  76. Networks for the Prediction of Oil Production Flow Rate, Energy
     Google Scholar
  77. Sources, Part A: Recovery, Utilization, and Environmental Effects,
     Google Scholar
  78. vol. 34 no. 19: 1834-1843. doi:10.1080/15567036.2010.492386.2012.
     Google Scholar
  79. R. H. Abiyev and M. K. S. Ma’aitah MKS. Deep Convolutional
     Google Scholar
  80. Neural Networks for Chest Diseases Detection. Journal of
     Google Scholar
  81. Healthcare Engineering. Article ID 4168538, 11 pages.
     Google Scholar
  82. https://doi.org/10.1155/2018/4168538. 2018.
     Google Scholar
  83. S. P. Mohanty, D. Hughes, M. Salathé. Using Deep Learning
     Google Scholar
  84. for Image-Based Plant Disease Detection. Front. Plant Sci.,
     Google Scholar
  85. vol. 7 Article ID 1419, 10 pages. doi: 10.3389/fpls.2016.01419. 2016.
     Google Scholar
  86. L. Liu, M. Ji, M. Buchroithner. Transfer Learning for Soil
     Google Scholar
  87. Spectroscopy Based on Convolutional Neural Networks and Its
     Google Scholar
  88. Application in Soil Clay Content Mapping Using Hyperspectral
     Google Scholar
  89. Imagery. Sensors 18: 3169, 18 pages. doi:10.3390/s18093169. 2018.
     Google Scholar
  90. T. Shaikhina and N. A. Khovanova. 2017. Handling limited datasets
     Google Scholar
  91. with neural networks in medical applications: A small data approach.
     Google Scholar
  92. Artificial Intelligence in Medicine, vol. 75: 51-63.
     Google Scholar
  93. http://dx.doi.org/10.1016/j.artmed.2016.12.003. 2016.
     Google Scholar
  94. M. Hayes, M. Svoboda, N. Wall, M. Widhalm. The Lincoln
     Google Scholar
  95. Declaration on Drought Indices: Universal Meteorological
     Google Scholar
  96. Drought Index Recommended. Bull. Amer. Meteor. Soc. vol. 92:
     Google Scholar
  97. –488. doi: http://dx.doi.org/10.1175/2010BAMS3103.1. 2011.
     Google Scholar
  98. T. B. McKee, N. J. Doesken, J. Kleist J. The Relationship of
     Google Scholar
  99. Drought Frequency and Duration to Time Scales. Eighth
     Google Scholar
  100. Conference on Applied Climatology, 17-22 January 1993,
     Google Scholar
  101. Anaheim, California.
     Google Scholar
  102. R. M. Hirsch, J. R. Slack, R. A. Smith. Techniques of Trend
     Google Scholar
  103. Analysis for Monthly Water Quality Data. Water Resources
     Google Scholar
  104. Research, vol. 18 no. 1: 107-121. 1982.
     Google Scholar
  105. D. R. Helsel and R. M. Hirsch RM. Statistical Methods in Water
     Google Scholar
  106. Resources Elsevier, NY. 1995.
     Google Scholar
  107. T. Salmi, A. Maatta, P. Anttila, T. Ruoho-Airola, T. Amnell.
     Google Scholar
  108. Detecting Trends of Annual Values of Atmospheric
     Google Scholar
  109. Pollutants by the Mann-Kendall Test and Sen’s Slope
     Google Scholar
  110. Estimates—The Excel Template Application MAKESENS;
     Google Scholar
  111. Finnish Meteorological Institute: Helsinki, Finland. 2002.
     Google Scholar
  112. N. H. Christiansen, N. H. Voie, P. E. T. Winther, O. J. Høgsberg.
     Google Scholar
  113. Comparison of Neural Network Error Measures for
     Google Scholar
  114. Simulation of Slender Marine Structures. Journal of Applied
     Google Scholar
  115. Mathematics vol. 2014, Article ID 759834, 11 pages.
     Google Scholar
  116. http://dx.doi.org/10.1155/2014/759834. 2014.
     Google Scholar
  117. T. Nakama. Comparisons of Single- and Multiple-Hidden-
     Google Scholar
  118. Layer Neural Networks. In: Liu D, Zhang H, Polycarpou
     Google Scholar
  119. M, Alippi C, He H. (eds) Advances in Neural Networks –
     Google Scholar
  120. ISNN 2011. ISNN 2011. Lecture Notes in Computer
     Google Scholar
  121. Science, vol 6675. Springer, Berlin, Heidelberg.
     Google Scholar
  122. E. D. Sonntag. Feedback stabilization using two-hidden-
     Google Scholar
  123. layer nets. IEEE Transactions on Neural Networks, vol.3:
     Google Scholar
  124. –990. 1992.
     Google Scholar
  125. F. Lolli, R. Gamberini, A. Regattieri, E. Balugani, T. Gatos,
     Google Scholar
  126. S. Gucc. Single-hidden layer neural networksfor forecasting
     Google Scholar
  127. intermittent demands. International Journal of Production
     Google Scholar
  128. Economics, vol. 183 no. A: 116-128. 2016.
     Google Scholar
  129. https://doi.org/10.1016/j.ijpe.2016.10.021. 2016.
     Google Scholar
  130. O. Mahmoud, O. Anwar, F. E. S. M. Jimoh, Learning Algorithm
     Google Scholar
  131. Effect on Multilayer Feed Forward Artificial Neural Network
     Google Scholar
  132. Performance in Image Coding. Journal of Engineering Science and
     Google Scholar
  133. Technology, vol. 2 no. 2: 188–199. 2007.
     Google Scholar