Vol 19 No 4 (2025): BAREKENG: Journal of Mathematics and Its Application
Articles

IMPUTATION OF MISSING DAILY RAINFALL DATA USING CONVOLUTIONAL NEURAL NETWORKS (CNN) WITH SPATIAL INTERPOLATION

pdf
  • Lilis Sriwahyuni,
  • Sri Nurdiati,
  • Endar Hasafah Nugrahani,
  • Ihwan Sukmana,
  • Mohamad Khoirun Najib
Lilis Sriwahyuni
Master’s Student of Applied Mathematics, Cluster of Mathematics, IPB University, Indonesia
Sri Nurdiati
Cluster of Mathematics, IPB University, Indonesia
Endar Hasafah Nugrahani
Cluster of Mathematics, IPB University, Indonesia
Ihwan Sukmana
Master’s Student of Applied Mathematics, Cluster of Mathematics, IPB University, Indonesia
Mohamad Khoirun Najib
Cluster of Mathematics, IPB University, Indonesia
DOI: https://doi.org/10.30598/barekengvol19iss4pp2921-2936
Published September 1, 2025
Keywords
  • Convolutional Neural Network,
  • Imputation Missing Data,
  • Interpolation Spline,
  • Machine Learning,
  • Mean Absolute Error
How to Cite
[1]
L. Sriwahyuni, S. Nurdiati, E. H. Nugrahani, I. Sukmana, and M. K. Najib, “IMPUTATION OF MISSING DAILY RAINFALL DATA USING CONVOLUTIONAL NEURAL NETWORKS (CNN) WITH SPATIAL INTERPOLATION”, BAREKENG: J. Math. & App., vol. 19, no. 4, pp. 2921-2936, Sep. 2025.

Copyright (c) 2025 Lilis Sriwahyuni, Sri Nurdiati, Endar Hasafah Nugrahani, Ihwan Sukmana, Mohamad Khoirun Najib

Creative Commons License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Abstract

Accurate rainfall estimation is crucial in climate analysis and water resource planning. Observational data from weather stations play a vital role in climatological analysis as they represent actual conditions at specific locations. However, many observation stations in Indonesia need more complete data, hindering analysis and data-driven decision-making. To address this issue, this study aims to impute missing rainfall data for BMKG stations in East Java using the Convolutional Neural Network (CNN) method. Satellite data used in this study include ERA5 without interpolation and ERA5 with interpolation. The study employs a spatial interpolation approach. Data were split into training and testing datasets with various ratios: 95:5%, 90:10%, 80:20%, 70:30%, and 50:50%. The results show that the CNN method with spatially interpolated satellite data yields better results, with a Mean Absolute Error (MAE) of 7.50 on the training data and 7.05 on the testing data, indicating better generalization capability than the method without interpolation. The combination of CNN and ERA5 with interpolation was chosen for imputing missing rainfall data at BMKG stations in East Java due to its lower MAE.

pdf

Downloads

Download data is not yet available.

References

  1. H. R. Bedane, K. T. Beketie, E. E. Fantahun, G. L. Feyisa, and F. A. Anose, “THE IMPACT OF RAINFALL VARIABILITY AND CROP PRODUCTION ON VERTISOLS IN THE CENTRAL HIGHLANDS OF ETHIOPIA,” Environmental Systems Research, vol. 11, no. 1, p. 26, Dec. 2022, doi: https://doi.org/10.1186/s40068-022-00275-3 .
  2. B. Hamududu and A. Killingtveit, “ASSESSING CLIMATE CHANGE IMPACTS ON GLOBAL HYDROPOWER,” Energies (Basel), vol. 5, no. 2, pp. 305–322, Feb. 2012, doi: https://doi.org/10.3390/en5020305
  3. T. D. Fletcher, H. Andrieu, and P. Hamel, “UNDERSTANDING, MANAGEMENT AND MODELLING OF URBAN HYDROLOGY AND ITS CONSEQUENCES FOR RECEIVING WATERS: A STATE OF THE ART,” Adv Water Resour, vol. 51, pp. 261–279, Jan. 2013, doi: https://doi.org/10.1016/j.advwatres.2012.09.001
  4. C. J. Walsh, A. H. Roy, J. W. Feminella, P. D. Cottingham, P. M. Groffman, and R. P. Morgan, “THE URBAN STREAM SYNDROME: CURRENT KNOWLEDGE AND THE SEARCH FOR A CURE,” J North Am Benthol Soc, vol. 24, no. 3, pp. 706–723, Sep. 2005, doi: https://doi.org/10.1899/04-028.1
  5. T. Schnepper, J. Groh, H. H. Gerke, B. Reichert, and T. Pütz, “EVALUATION OF PRECIPITATION MEASUREMENT METHODS USING DATA FROM A PRECISION LYSIMETER NETWORK,” Hydrol Earth Syst Sci, vol. 27, no. 17, pp. 3265–3292, Sep. 2023, doi: https://doi.org/10.5194/hess-27-3265-2023
  6. F. J. Tapiador et al., “GLOBAL PRECIPITATION MEASUREMENTS FOR VALIDATING CLIMATE MODELS,” Atmos Res, vol. 197, pp. 1–20, Nov. 2017, doi: https://doi.org/10.1016/j.atmosres.2017.06.021
  7. M. Bernard and C. Gregoretti, “THE USE OF RAIN GAUGE MEASUREMENTS AND RADAR DATA FOR THE MODEL‐BASED PREDICTION OF RUNOFF‐GENERATED DEBRIS‐FLOW OCCURRENCE IN EARLY WARNING SYSTEMS,” Water Resour Res, vol. 57, no. 3, Mar. 2021, doi: https://doi.org/10.1029/2020WR027893
  8. L. C. Sieck, S. J. Burges, and M. Steiner, “CHALLENGES IN OBTAINING RELIABLE MEASUREMENTS OF POINT RAINFALL,” Water Resour Res, vol. 43, no. 1, Jan. 2007, doi: https://doi.org/10.1029/2005WR004519
  9. S. A. Rafhida, S. Nurdiati, R. Budiarti, and M. K. Najib, “BIAS CORRECTION OF LAKE TOBA RAINFALL DATA USING QUANTILE DELTA MAPPING,” CAUCHY: Jurnal Matematika Murni dan Aplikasi, vol. 9, no. 2, pp. 297–309, Nov. 2024, doi: https://doi.org/10.18860/ca.v9i2.29124
  10. D. Desmonda, T. Tursina, and M. A. Irwansyah, “PREDIKSI BESARAN CURAH HUJAN MENGGUNAKAN METODE FUZZY TIME SERIES,” Jurnal Sistem dan Teknologi Informasi (JUSTIN), vol. 6, no. 4, p. 141, Oct. 2018, doi: https://doi.org/10.26418/justin.v6i4.27036
  11. E. Ardiyani, S. Nurdiati, A. Sopaheluwakan, P. Septiawan, and M. K. Najib, “PROBABILISTIC HOTSPOT PREDICTION MODEL BASED ON BAYESIAN INFERENCE USING PRECIPITATION, RELATIVE DRY SPELLS, ENSO AND IOD,” Atmosphere (Basel), vol. 14, no. 2, p. 286, Jan. 2023, doi: https://doi.org/10.3390/atmos14020286
  12. L. Li, “A ROBUST DEEP LEARNING APPROACH FOR SPATIOTEMPORAL ESTIMATION OF SATELLITE AOD AND PM2.5,” Remote Sens (Basel), vol. 12, no. 2, p. 264, Jan. 2020, doi: https://doi.org/10.3390/rs12020264
  13. X. Yang, R. Cui, C. Tian, S. Hu, J. Jiang, and P. Xu, “LINEAR SPLINE AND CNN-LSTM FOR MISSING VALUES IMPUTATION OF BEIDOU SATELLITE RADIATION DOSE DATA,” Chinese Journal of Space Science, vol. 42, no. 1, p. 163, 2022, doi: https://doi.org/10.11728/cjss2022.01.201116100
  14. M. K. Najib and S. Nurdiati, “KOREKSI BIAS STATISTIK PADA DATA PREDIKSI SUHU PERMUKAAN AIR LAUT DI WILAYAH INDIAN OCEAN DIPOLE BARAT DAN TIMUR,” Jambura Geoscience Review, vol. 3, no. 1, pp. 9–17, Jan. 2021, doi: https://doi.org/10.34312/jgeosrev.v3i1.8259
  15. A. Gelman and J. Hill, DATA ANALYSIS USING REGRESSION AND MULTILEVEL/HIERARCHICAL MODELS. Cambridge University Press, 2006. doi: https://doi.org/10.1017/CBO9780511790942
  16. A. Wangwongchai, M. Waqas, P. Dechpichai, P. T. Hlaing, S. Ahmad, and U. W. Humphries, “IMPUTATION OF MISSING DAILY RAINFALL DATA; A COMPARISON BETWEEN ARTIFICIAL INTELLIGENCE AND STATISTICAL TECHNIQUES,” MethodsX, vol. 11, p. 102459, Dec. 2023, doi: https://doi.org/10.1016/j.mex.2023.102459
  17. J. M. Jerez et al., “MISSING DATA IMPUTATION USING STATISTICAL AND MACHINE LEARNING METHODS IN A REAL BREAST CANCER PROBLEM,” Artif Intell Med, vol. 50, no. 2, pp. 105–115, Oct. 2010, doi: https://doi.org/10.1016/j.artmed.2010.05.002
  18. Y. Lops, A. Pouyaei, Y. Choi, J. Jung, A. K. Salman, and A. Sayeed, “APPLICATION OF A PARTIAL CONVOLUTIONAL NEURAL NETWORK FOR ESTIMATING GEOSTATIONARY AEROSOL OPTICAL DEPTH DATA,” Geophys Res Lett, vol. 48, no. 15, Aug. 2021, doi: https://doi.org/10.1029/2021GL093096
  19. Q. Zhang, Q. Yuan, C. Zeng, X. Li, and Y. Wei, “MISSING DATA RECONSTRUCTION IN REMOTE SENSING IMAGE WITH A UNIFIED SPATIAL–TEMPORAL–SPECTRAL DEEP CONVOLUTIONAL NEURAL NETWORK,” IEEE Transactions on Geoscience and Remote Sensing, vol. 56, no. 8, pp. 4274–4288, Aug. 2018, doi: https://doi.org/10.1109/TGRS.2018.2810208
  20. L. Sriwahyuni, S. Nurdiati, E. H. Nugrahani, and M. K. Najib, “PERFORMANCE OF MACHINE LEARNING FOR IMPUTING MISSING DAILY RAINFALL DATA IN EAST JAVA UNDER MULTIPLE SATELLITE DATA MODELS,” Geographia Technica, vol. 20, no. 1/2025, pp. 346–368, Mar. 2025, doi: https://doi.org/10.21163/GT_2025.201.23
  21. T. C. H. Lux, L. T. Watson, T. H. Chang, Y. Hong, and K. Cameron, “INTERPOLATION OF SPARSE HIGH-DIMENSIONAL DATA,” Numer Algorithms, vol. 88, no. 1, pp. 281–313, Sep. 2021, doi: https://doi.org/10.1007/s11075-020-01040-2
  22. T. Wang, D. J. Wu, A. Coates, and A. Y. Ng, “END-TO-END TEXT RECOGNITION WITH CONVOLUTIONAL NEURAL NETWORKS.”
  23. Z. Chen and Y. Li, “FDSPC: FAST AND DIRECT SMOOTH PATH PLANNING VIA CONTINUOUS CURVATURE INTEGRATION,” May 2024.
  24. S. Moghtadernejad, Y. Jin, and B. T. Adey, “ESTIMATING THE VALUES OF MISSING DATA RELATED TO INFRASTRUCTURE CONDITION STATES USING THEIR SPATIAL CORRELATION,” Journal of Infrastructure Systems, vol. 29, no. 1, Mar. 2023, doi: https://doi.org/10.1061/(ASCE)IS.1943-555X.0000726
  25. M. N. Arefin and A. K. M. Masum, “A PROBABILISTIC APPROACH FOR MISSING DATA IMPUTATION,” Complexity, vol. 2024, pp. 1–15, Jan. 2024, doi: https://doi.org/10.1155/2024/4737963
  26. S. Nurdiati, M. K. Najib, F. Bukhari, R. Revina, and F. N. Salsabila, “PERFORMANCE COMPARISON OF GRADIENT-BASED CONVOLUTIONAL NEURAL NETWORK OPTIMIZERS FOR FACIAL EXPRESSION RECOGNITION,” BAREKENG: Jurnal Ilmu Matematika dan Terapan, vol. 16, no. 3, pp. 927–938, Sep. 2022, doi: https://doi.org/10.30598/barekengvol16iss3pp927-938
  27. D. M. Hawkins, “THE PROBLEM OF OVERFITTING,” J Chem Inf Comput Sci, vol. 44, no. 1, pp. 1–12, Jan. 2004, doi: https://doi.org/10.1021/ci0342472
  28. K. Fukushima, “Neocognitron: A SELF-ORGANIZING NEURAL NETWORK MODEL FOR A MECHANISM OF PATTERN RECOGNITION UNAFFECTED BY SHIFT IN POSITION,” Biol Cybern, vol. 36, no. 4, pp. 193–202, Apr. 1980, doi: https://doi.org/10.1007/BF0034425
  29. Z. Li, F. Liu, W. Yang, S. Peng, and J. Zhou, “A SURVEY OF CONVOLUTIONAL NEURAL NETWORKS: ANALYSIS, APPLICATIONS, AND PROSPECTS,” IEEE Trans Neural Netw Learn Syst, vol. 33, no. 12, pp. 6999–7019, Dec. 2022, doi: https://doi.org/10.1109/TNNLS.2021.3084827
  30. V. Nair and G. E. Hinton, “RECTIFIED LINEAR UNITS IMPROVE RESTRICTED BOLTZMANN MACHINES.”
  31. H. Junninen, H. Niska, K. Tuppurainen, J. Ruuskanen, and M. Kolehmainen, “METHODS FOR IMPUTATION OF MISSING VALUES IN AIR QUALITY DATA SETS,” Atmos Environ, vol. 38, no. 18, pp. 2895–2907, Jun. 2004, doi: https://doi.org/10.1016/j.atmosenv.2004.02.026
  32. D. P. Kingma and J. Ba, “ADAM: A METHOD FOR STOCHASTIC OPTIMIZATION,” Dec. 2014.
  33. T. Dozat, “WORKSHOP TRACK-ICLR 2016 INCORPORATING NESTEROV MOMENTUM INTO ADAM.”
  34. I. Loshchilov and F. Hutter, “SGDR: STOCHASTIC GRADIENT DESCENT WITH WARM RESTARTS,” Aug. 2016.
  35. Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner, “GRADIENT-BASED LEARNING APPLIED TO DOCUMENT RECOGNITION,” Proceedings of the IEEE, vol. 86, no. 11, pp. 2278–2324, 1998, doi: https://doi.org/10.1109/5.726791
  36. A. Krizhevsky, I. Sutskever, and G. E. Hinton, “IMAGENET CLASSIFICATION WITH DEEP CONVOLUTIONAL NEURAL NETWORKS,” Commun ACM, vol. 60, no. 6, pp. 84–90, May 2017, doi: https://doi.org/10.1145/3065386
  37. S. Park and N. Kwak, “ANALYSIS ON THE DROPOUT EFFECT IN CONVOLUTIONAL NEURAL NETWORKS,” 2017, pp. 189–204. doi: https://doi.org/10.1007/978-3-319-54184-6_12