Artificial Neural Network for the Estimation of Missing Precipitation Data at the EL ENCANTO Station in the Municipality of Tunja, Boyacá
Article Sidebar
Main Article Content
Abstract
Missing and discontinuous precipitation records limit hydrological assessment, water-resource planning, and climaterisk management in mountainous areas such as Tunja, Boyacá. This study developed and evaluated an artificial neural network model to estimate missing monthly precipitation data at the El Encanto station, using neighboring rainfall stations with correlated behavior, geographic variables, and the Oceanic Niño Index (ONI) as predictors. The methodology included DHIME data cleaning, classification of monthly completeness patterns, variable normalization, TensorFlow/Keras training, and validation through MAE, RMSE, R², Nash efficiency, PBIAS, and statistical comparison tests. The results show acceptable performance in reconstructing monthly trend and seasonality (MAE = 19.3 mm; RMSE = 31.6 mm; R² = 0.44; PBIAS = 13.6 %), although the model is limited in reproducing extreme events and high-rainfall variability. The approach is useful as a technical support tool for completing series, provided that its estimates are interpreted with hydrological quality control.
Article Details
Section
DECLARATION OF ORGINIALITY OF SUBMITTED ARTICLE
With this document, I/We certify that the article submitted for possible publication in the institutional journal INGENIO MAGNO of the Research Center Alberto Magno CIIAM of the University Santo Tomás, Tunja campus, is entirely of my(our) own writing, and is a product of my(our) direct intellectual contribution to knowledge.
All data and references to completed publications are duly identified with their respective bibliographical entries and in the citations thus highlighted. If any adjustment or correction is needed, I(we) will contact the journal authorities in advance.
Due to that stated above, I(we) declare that the entirety of the submitted material is in accordance with applicable laws regarding intellectual and industrial property, and therefore, I(we) hold myself(ourselves) responsible for any complaint related to it.
If the submitted article is published, I(we) declare that I(we) fully relinquish publishing rights of the article to the University Santo Tomás, Tunja campus. As remuneration for this relinquishment of rights, I(we) declare my(our) agreement to receive two (2) copies of the edition of the journal in which my(our) article appears.
How to Cite
References
[1] Congreso de la República de Colombia. (2012). Ley 1523 de 2012. Por la cual se adopta la Política Nacional de Gestión del Riesgo de Desastres y se establece el Sistema Nacional de Gestión del Riesgo de Desastres y se dictan otras disposiciones. Diario Oficial No.48.411 del 24 de abril de 2012. https://www.funcionpublica.gov.co/eva/gestornormativo/norma.php?i=47141
[2] Coulibaly, P., Anctil, F., y Bobée, B. (2000). Daily reservoir inflow forecasting using artificial neural networks with stopped training approach. Journal of Hydrology, 301(1-4), 244-257.
[3] Elman, J. L. (1990). Finding structure in time. Cognitive Science, 14(2), 179-211.
[4] Feldkamp, L. A., & Puskorius, G. V. (1998). A signal processing framework based on dynamic neural networks with application to problems in adaptation, filtering, and classification. Proceedings of the IEEE, 86(11), 2259–2277. https://doi.org/10.1109/5.726790
[5] French, M. N., Krajewski, W. F., & Cuykendall, R. R. (1992). Rainfall forecasting in space and time using a neural network. Journal of Hydrology, 137(1–4), 1–31. https://doi.org/10.1016/0022-1694(92)90046-X
[6] Gupta, H. V., Sorooshian, S., y Yapo, P. O. (1999). Status of automatic calibration for hydrologic models: Comparison with multilevel expert calibration. Journal of Hydrologic Engineering, 4(2), 135-143.
[7] Hsu, K.-L., Gupta, H. V., y Sorooshian, S. (1995). Artificial neural network modeling of the rainfall-runoff process. Water Resources Research, 31(10), 2517-2530.
[8] IDEAM (2019). Evaluación de la calidad y continuidad de la red hidrometeorológica nacional. Instituto de Hidrología, Meteorología y Estudios Ambientales, Colombia.
[9] Kisi, O. (2004). Streamflow forecasting using different artificial neural network algorithms. Journal of Hydrologic Engineering, 9(6), 491-496.
[10] Kwon, S. H., & Kim, J. H. (2021). Machine learning and urban drainage systems: State-of-the-art review. Water, 13(24), 3545. https://doi.org/10.3390/w13243545
[11] Nearing, G. S., Kratzert, F., Sampson, A. K., Pelissier, C. S., Klotz, D., Frame, J. M., Prieto, C., & Gupta, H. V. (2021). What role does hydrological science play in the age of machine learning? Water Resources
[12] Nourani, V., y Hosseini B. (2011). Monthly rainfall-runoff modeling using a wavelet-based neural network conjunction model. Journal of Hydrology, 401(3-4), 190-199.
[13] Ordoñez, L., Muñoz, S., Tineo, P., & Mejía, I. (2024). Aplicación de redes neuronales artificiales a la modelación de lluvia-escorrentía en la cuenca del río Chancay Lambayeque. Tecnología Y Ciencias Del Agua , 15 (6), 95-141. https://doi.org/10.24850/j-tyca-2024-06-03
[14] Papailiou, I., Spyropoulos, F., Trichakis, I., & Karatzas, G. P. (2022). Artificial neural networks and multiple linear regression for filling in missing daily rainfall data. Water, 14(18), 2892. https://doi.org/10.3390/w14182892
[15] Pasquier Jaramillo, C. A. (2025) Modelado bayesiano de extremos espacio-temporales de precipitación: aplicaciones en estimación de modelos con covariables.
[16] Pearlmutter, B. (1995). Gradient calculations for dynamic recurrent neural networks: A survey. IEEE Transactions on Neural Networks, 6(5), 1212-1228.
[17] Pinthong, S., Ditthakit, P., Salaeh, N., Hasan, M. A., Son, C. T., Linh, N. T. T., Islam, S., & Yadav, K. K. (2024). Imputation of missing monthly rainfall data using machine learning and spatial interpolation approaches in Thale Sap Songkhla River Basin, Thailand. Environmental Science and Pollution Research, 31, 54044-54060. https://doi.org/10.1007/s11356-022-23022-8
[18] Principe, J. C., Euliano, N. R., & Lefebvre, W. C. (2000). Sistemas neuronales y adaptativos: Fundamentos a través de simulaciones. John Wiley.
[19] Retropropagación: Teoría, arquitecturas y aplicaciones (pp. 433-486). Lawrence Erlbaum Associates.
[20] Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning representations by back-propagating errors. Nature, 323(6088), 533–536. https://doi.org/10.1038/323533a0
[21] Smith, A. B., Jones, C. D., y Johnson, E. F. (2018). Application of artificial neural networks for precipitation estimation in mountainous regions. Journal of Hydrology, 543, 765-776.
[22] Specht, D. F. (1991). A general regression neural network. IEEE Transactions on Neural Networks, 2(6), 568–576. https://doi.org/10.1109/72.97934
[23] Wang, L., y Robertson, A. (2020). Improving precipitation estimation using neural networks in complex terrain. Water Resources Research, 56(2), e2019WR026101.
[24] Wangwongchai, A., Waqas, M., Dechpichai, P., Hlaing, P. T., Ahmad, S., & Humphries, U. W. (2023). Imputation of missing daily rainfall data; A comparison between artificial intelligence and statistical techniques.
[25] Williams, R. J., & Zipser, D. (1995). Algoritmos de aprendizaje basados en gradiente para redes recurrentes y su complejidad computacional. En Y. Chauvin & D. E. Rumelhart (Eds.), Retropropagación: Teoría, arquitecturas y aplicaciones (pp. 433–486). Lawrence Erlbaum Associates, Inc.
[26] Xu, T., & Liang, F. (2021). Machine learning for hydrologic sciences: An introductory overview. Wiley
[27] Yescas, I. A. O. (2016). Elaboración de un protocolo para construir bibliotecas de RNAs pequeños y su implementación para el estudio de la respuesta a sequía crónica en Brachypodium distachyon L (Doctoral dissertation, Universidad del Papaloapan).