Generation of rainfall data series by using the Markov Chain model in three selected sites in the Kurdistan Region, Iraq

Abstract

Rainfall forecasting can play a significant role in the planning and management of water resource systems. This study employs a Markov chain model to examine the patterns, distributions and forecast of annual maximum rainfall (AMR) data collected at three selected stations in the Kurdistan Region of Iraq using 32 years of 1990 to 2021 rainfall data. A stochastic process is used to formulate three states (i.e., decrease—"d"; stability—"s"; and increase—"i") in a given year for estimating quantitatively the probability of making a transition to any other one of the three states in the following year(s) and in the long run. In addition, the Markov model is also used to forecast the AMR data for the upcoming five years (i.e., 2022–2026). The results indicate that in the upcoming 5 years, the probability of the annual maximum rainfall becoming decreased is 44%, that becoming stable is 16%, and that becoming increased is 40%. Furthermore, it is shown that for the AMR data series, the probabilities will drop slowly from 0.433 to 0.409 in about 11 years, as indicated by the average data of the three stations. This study reveals that the Markov model can be used as an appropriate tool to forecast future rainfalls in such semi-arid areas as the Kurdistan Region of Iraq.

Keywords

Time series / Rainfall / Markov chain / Forecast / Transition Probability

References

[1] Abdi, H., & Molin, P. (2007). Lilliefors/Van Soest’s Test of Normality. Encyclopedia of Measurement and Statistics, pp. 540–544.

[2] AndersonTW, DarlingDA. A test of goodness of fit. Journal of the American Statistical Association, 1954, 49: 765-769

[3] Aziz, F.H., Omar, R, & Ahmed, Q. (2022). Historical Overview of Air Temperature of Kurdistan Region-Iraq from 1973 to 2017. Journal of University of Garmian, the 10th Scientific Conference: Drought and Water Scarcity Management.

[4] Barkotulla M. A. B. (2010). Stochastic Generation of the Occurrence and Amount of Daily Rainfall. Pakistan Journal of Statistics and Operation Research, 61–74.

[5] BoxGE, CoxDR. An analysis of transformations. Journal of the Royal Statistical Society: Series B (methodological), 1964, 26(2):211-243

[6] BrathA, MontanariA, TothE. Neural networks and non-parametric methods for improving real-time flood forecasting through conceptual hydrological models. Hydrology and Earth System Sciences, 2002, 6(4):627-639

[7] CheddadA. On box-cox transformation for image normality and pattern classification. Institute of Electrical and Electronics Engineering (IEEE), 2020, 8: 154975-154983

[8] ChungG, SimKB, JoDJ, KimES. Hourly precipitation simulation characteristic analysis using Markov chain model. Journal of the Korean Society of Hazard Mitigation, 2016, 16(3):351-357

[9] Danilovich A. (2016). New horizons: Iraqi federalism. In: Iraqi Kurdistan in Middle Eastern Politics. Routledge, pp. 49–70.

[10] FadhilRM, RowshonMK, AhmadD, FikriA, AimrunW. A stochastic rainfall generator model for simulation of daily rainfall events in Kurau catchment: model testing. III International Conference on Agricultural and Food Engineering, 2016, 1152: 1-10

[11] GaoS, HuangY, ZhangS, HanJ, WangG, ZhangM, LinQ. Short-term runoff prediction with GRU and LSTM networks without requiring time step optimization during sample generation. Journal of Hydrology, 2020, 589

[12] GargVK, SinghJB. Three-state Markov chain approach on the behaviour of rainfall. New York Science Journal, 2010, 3(12):76-81

[13] Gui, Y., & Shao, J. (2017). Prediction of precipitation based on weighted markov chain in Dangshan. In Proceedings of the International Conference on High Performance Compilation, Computing and Communications, pp. 81–85.

[14] GuptaRS. Hydrology and hydraulic systems, 1989 Prentice Hall 343-350

[15] HajaniE, KlariZ. Trends analysis in rainfall data series in Duhok City, Kurdistan Region, Iraq. Modeling Earth Systems and Environment., 2022, 8: 4177

[16] HajaniE, ShajeeK, KaledF, AbdulhaqH. Characteristics of changes in rainfall data in the Kurdistan Region Iraq. Arabian Journal of Geosciences, 2022, 15(509):1-21

[17] HeR, ZhangL, ChewAW. Modeling and predicting rainfall time series using seasonal-trend decomposition and machine learning. Knowledge-Based Systems, 2022, 251

[18] Holt, C.C. (1957). Forecasting Seasonals and Trends by Exponentially Weighted Averages (O.N.R. Memorandum No. 52). Carnegie Institute of Technology, Pittsburgh USA.

[19] HowardRA. Dynamic probabilistic systems, 1–2, 1971 Wiley

[20] JainS. A Markov chain model and its application. Computers and Biomedical Research, 1986, 19: 374-378

[21] JaleJS, Xavier JrSFA, XavierEFM, StosicT, StosicB, FerreiraTAE. Application of Markov chain on daily rainfall data in Paraíba-Brazil from 1995–2015. Acta Scientiarum Technology, 2019, 41(1):e37186

[22] JimohOD, WebsterP. The optimum order of a Markov chain model for daily rainfall in Nigeria. Journal of Hydrology, 1996, 185(1–4):45-69

[23] Kenton W. (2021). Markov Analysis: What is Markov Analysis. https://www.investopedia.com/terms/m/markov-analysis.asp

[24] KottegodaNT, NataleL, RaiteriE. Some considerations of periodicity and persistence in daily rainfalls. Journal of Hydrology, 2004, 296: 23-37

[25] Kuo M. (2016). Mbd Excel: How to Create a Normally Distributed Set of Random Numbers in Excel. http://www.mbaexcel.com/excel/

[26] Liu, C., Tian, Y. M., & Wang, X. H. (2011). Study of rainfall prediction model based on GM (1, 1)-Markov Chain. In 2011 International Symposium on Water Resource and Environmental Protection (vol. 1, pp. 744–747). Institute of Electrical and Electronics Engineering (IEEE).

[27] MaassA, HufschmidtMM, DorfmanR, ThomasHA, MarglinSA, FairGM. The design of water-resource systems, 1962 Cambridge Harvard University Press 467

[28] MahantaJ, DeyS, KhosroP. Analyzing rainfall condition of Bangladesh: an application of Markov chain. Thailand Statistician, 2018, 16(2):203-212

[29] MakridakisSG, WheelwrightS, HyndmanR. Forecasting: Methods and applications, 1998 3 New York Wiley

[30] Malakoutian, M. M. A., Malakoutian, Y., Mostafapoor, P. & Abed, S. Z. D. (2021). Prediction for monthly rainfall of six meteorological regions and TRNC (Case Study: North Cyprus). Computational Research Progress in Applied Science and Engineering. H&T Publication hal-03228691.

[31] Muhammad M. K. I. B. (2012). Time Series Modelling Using Markov and ARIMA Models. Doctoral dissertation, MSc. Thesis. Faculty of Civil Engineering, University of Technology Malaysia.

[32] Oswal N. (2019). Predicting Rainfall Using Machine Learning Techniques. ARXIV preprint arXiv:1910.13827.

[33] Preacher K. J. (2001). Calculation for the Chi-Square Test. An Interactive Calculation Tool for Chi-Square Tests of Goodness of Fit and Independence. http://quantpsy.org.

[34] RykovVV, BalakrishnanN, NikulinMS. Mathematical and statistical models and methods in reliability: Applications to medicine, finance, and quality control, 2010 Springer Science and Business Media

[35] SakiaRM. The Box-Cox transformation technique: A review. Journal of the Royal Statistical Society: Series D (The Statistician), 1992, 41(2):169

[36] WangDQ, MartizJS. Note on testing a three state Markov chain for independence. Journal Statistics Computation and Simulation, 1990, 37(1–2):61-68

[37] Weather and Climate Kurdistan, WCK (2021). Best Time to Visit Kurdistan, Iraq. https://www.besttimetovisit.com.pk/iraq/kurdistan-2330031/.

[38] World Climate Guide, WCG (2019). Climate-Iraq. https://www.climatestotravel.com/climate.

[39] YusufAU, AdamuL, AbdullahiM. Markov chain model and its application to annual rainfall distribution for crop production. American Journal of Theoretical and Applied Statistics, 2014, 3(2):39-43

[40] ZhangS, WangH, ZhangX. Estimation of channel state transition probabilities based on Markov chains in cognitive radio. Journal of Communications, 2014, 9(6):468-474

[41] Zhou, X., Wang, Y., & Zhou, X. (2017). Precipitation estimation based on weighted Markov chain model. In 2017 Seventh International Conference on Information Science and Technology (ICIST), Institute of Electrical and Electronics Engineering (IEEE), pp. 64–68.