Forecasting and Spatially Interpolating Rainfall Data- A case study on Thiruvananthapuram District

IJEP 44(5): 428-436 : Vol. 44 Issue. 5 (May 2024)

Surendar Natarajan1*, T. Sree Sharmila2 and A. Jegan Bharath Kumar3

1. Sri Sivasubramaniya Nadar (SSN) College of Engineering, Department of Civil Engineering, Kalavakkam – 603 110, Tamil Nadu, India
2. Anna University, Department of Computer Science and Engineering, College of Engineering Guindy, Chennai – 600 025 , Tamil Nadu, India
3. KSCSTE- National Transportation Planning and Research Centre (NATPAC), Thiruvananthapuram – 695 011, Kerala, India


Hydrologists have a formidable challenge when trying to forecast rainfall, the major source of water supplies. The ability to foresee it is crucial for drought and flood management. Agriculture, hydroelectric power generation, fisheries and the dairy industry all rely on reliable precipitation forecasts. Rainfall prediction also involves recording various parameters, like wind speed, direction and temperature. Conventional statistical and advanced machine learning (ML) techniques are used for rainfall predictions. In last few years, it has been seen that ML techniques have achieved better performance in weather predictions than the traditional statistical methods. In this work, advanced ML techniques, like decision tree (DT), linear regression (LR) and XG-boost were used to predict rainfall in Thiruvananthapuram district, Kerala, India. The rainfall data was collected for the period of 50 years and it was predicted till the year 2035. From the adopted ML techniques, linear regression algorithm gave closer results with observed rainfall data and in same way, the predicted results also followed the same trend. The predicted rainfall results were plotted spatially through Arc-GIS process for spatially accurate information.


Rainfall prediction, Machine learning techniques, Arc-GIS spatial maps, XG-boost



  1. Mosavi, A., T. Rabezuk and A.R. Varkonyi-Koczy. 2012. Reviewing the novel machine learning tools for missive data by novel artificial neural networks. Expert Syst. Appl., 39(10): 456-464.
  2. Ren, X., et al. 2020. Deep learning based weather predictions: A survey. Big Data Res., 23:100178.
  3. Yu, P.S., et al. 2017. Comparison of random forests and support vector machine for real-time radar-derived rainfall forecasting. J. Hydrol., 532: 92-104.
  4. Demeke, E., H. Getamesay and W. Taye. 2022. Deep learning model for daily rainfall prediction: Case study of Jimma, Ethiopia. Water Supply. 22(3): 3448. DOI: 10.2166/ws. 2021.391.
  5. Karmakar, S., M.K. Kowar and P. Guhathakurta. 2009. Long range monsoon rainfall pattern recognition and prediction for the sub-division ‘EPMB’ Chhattisgarh using deterministic and probabilistic neural network. 7th International Conference on Advanced pattern recognition. Proceedings, pp 367-370.
  6. Balamunegan, M.S. and R. Manojkumar. 2021. Study of short term rain forecasting using machine learning based approach. Wireless Netw., 27: 5429- 5434. DOI: 10.1007/s11276-019-02168-3.
  7. Rahman, A.U., et al. 2022. Rainfall prediction system using machine learning fusion for smart cities. Sensors. 22:3504. DOI: 10.3390/sz2093504.
  8. Choubin, B., et al. 2018. Precipitation forecasting using classification and regression trees (CART) model: A comparative study of different approa-ches. Env. Earth Sci., 77: 314.
  9. Kim, H.U. and T.S. Bae. 2017. Preliminery study of deep learining-based precipitation. J. Korean Soc. Surv. Geodesy Photogram. Cartography. 35:423-430. DOI: 10.7848/ksgpc.2017.35.5.423.
  10. Chao, Z., et al. 2018. Research on real-time local rainfall prediction based on MEMS sensors. J. Sensors. 6184713. DOI: 10.1155/2018/6184713.
  11. Krajewski, W.F. and A.S. James. 2002. Radar hydrology: Rainfall estimation. Adv. Water Resour., 25:1387-1394. DOI: 10.106/S0309-1708(02)000 62-3.
  12. Sattari, M.T., et al. 2018. Prediction of groundwater level in Ardebil plain using support vector regression and M5 tree model. Ground Water. 56(4): 636-646. DOI: 10.1111/gwat.
  13. Tharun, V.P., R. Prakash and S.R. Devi. 2018. Prediction of rainfall using data mining techniques. Second International Conference on Inventive communication and computational technologies (ICICCT). Proceedings, pp 1507-1512.
  14. Michaelides, S.C., F.S. Tymvios and T. Michaelidou. 2009. Spatial and temporal characteristics of the annual rainfall frequency distribution in Cyprus. Atmos. Res., 94(4): 606-615. DOI:10.106/j.atm osres.
  15. Vega-Garcia, C., M. Decuyper and J. Aleazar. 2019. Applying cascade-correlation natural networks to in-fill gaps in mediterranean daily flow data series. Water. 11:1691. DOI: 10.3390/w1108 1691.
  16. Abhishek, K., et al. 2012. A rainfall prediction model using artificial neural network. IEEE control and system graduate research colloquium (ICSGRC)-Selangor, Malaysia. Proceedings, pp 82-87.
  17. Nikam, V.B. and B.B. Meshram. 2013. Modelling rainfall predication using data mining method: A Bayesian approach. International Conference on Computational intelligence, modelling and simulation. Bangkok, Thailand. Proceedings, pp 132-
  18. Zeelan, B., et al. 2020. Rainfall prediction using machine learning and deep learning techniques. International Conference on Electronics and sustainable communication systems (ICESC 2020). Midlesex University. Proceedings, pp 92-97.