Hourly UV Index Forecasting Across Climatic Zones in Sri Lanka: A Residual-Correction Hybrid Machine Learning and Deep Learning Approach
DOI:
https://doi.org/10.62411/faith.3048-3719-372Keywords:
Climate adaptation, Deep learning, Environmental forecasting, Machine learning, Public health, Residual-correction hybrid model, Ultraviolet index forecasting, UV radiationAbstract
Ultraviolet (UV) radiation reaching the Earth's surface has shown a long-term increasing trend, raising concerns due to its adverse effects on human health, including an elevated risk of skin cancer. The Ultraviolet Index (UVI) is a standardized measure of UV radiation intensity and plays an important role in public health protection and climate adaptation planning. However, UVI forecasting studies in Sri Lanka remain limited, particularly those comparing traditional statistical, machine learning (ML), deep learning (DL), and hybrid forecasting approaches across different climatic zones. This study develops and evaluates forecasting frameworks for hourly UVI prediction in three representative climatic regions of Sri Lanka: Colombo (wet zone), Badulla (intermediate zone), and Jaffna (dry zone). Hourly UVI observations from January 2018 to December 2023 were obtained from the NASA POWER database. Since UVI values between 18:00 and 05:00 were consistently zero, the analysis was restricted to daylight hours (06:00–17:00). Traditional time-series, ML, and DL models were evaluated for one-hour-ahead forecasting using a univariate framework. In addition, a residual-correction hybrid approach was proposed, in which residuals from the best-performing standalone model were modeled using a secondary learning algorithm to improve prediction accuracy. The results show that ML and DL models consistently outperform traditional statistical approaches, while hybrid models achieve the highest predictive performance across all climatic zones. The CNN–RF hybrid achieved the best performance in Colombo (RMSE = 0.3967, R² = 0.9881), whereas the FNN–RF hybrid performed best in Badulla (RMSE = 0.4265, R² = 0.9847) and Jaffna (RMSE = 0.4291, R² = 0.9844). Furthermore, lag analysis revealed that both short-term dependencies and recurring diurnal patterns contribute substantially to forecasting performance. These findings demonstrate the effectiveness of residual-correction hybrid forecasting for regional UVI prediction and its potential application in public health early warning systems.
Downloads
References
X. Tang, T. Yang, D. Yu, H. Xiong, and S. Zhang, “Current insights and future perspectives of ultraviolet radiation (UV) exposure: Friends and foes to the skin and beyond the skin,” Environ. Int., vol. 185, p. 108535, Mar. 2024, doi: 10.1016/j.envint.2024.108535.
O. Langselius et al., “Global burden of cutaneous melanoma incidence attributable to ultraviolet radiation in 2022,” Int. J. Cancer, vol. 157, no. 6, pp. 1110–1119, Sep. 2025, doi: 10.1002/ijc.35463.
A. Brozyna, B. Zbytek, J. Granese, J. A. Carlson, J. Ross, and A. Slominski, “Mechanism of UV-related carcinogenesis and its contribution to nevi/melanoma,” Expert Rev. Dermatol., vol. 2, no. 4, pp. 451–469, Aug. 2007, doi: 10.1586/17469872.2.4.451.
World Health Organization, “Radiation: The Known Health Effects of Ultraviolet Radiation,” World Health Organization, 2024. https://www.who.int/news-room/questions-and-answers/item/radiation-the-known-health-effects-of-ultraviolet-radiation (accessed Apr. 04, 2026).
World Health Organization, “Global Solar UV Index: A Practical Guide,” World Health Organization, 2002. https://www.who.int/publications/i/item/9241590076 (accessed Apr. 04, 2026).
C. J. Heckman, K. Liang, and M. Riley, “Awareness, understanding, use, and impact of the UV index: A systematic review of over two decades of international research,” Prev. Med. (Baltim)., vol. 123, pp. 71–83, Jun. 2019, doi: 10.1016/j.ypmed.2019.03.004.
V. Fioletov, J. B. Kerr, and A. Fergusson, “The UV Index: Definition, Distribution and Factors Affecting It,” Can. J. Public Heal., vol. 101, no. 4, pp. I5–I9, Jul. 2010, doi: 10.1007/BF03405303.
B. L. Diffey, “Solar ultraviolet radiation effects on biological systems,” Phys. Med. Biol., vol. 36, no. 3, pp. 299–328, Mar. 1991, doi: 10.1088/0031-9155/36/3/001.
U.S. Environmental Protection Agency, “UV Index Overview,” U.S. Environmental Protection Agency, 2026. https://www.epa.gov/enviro/uv-index-overview (accessed Apr. 04, 2026).
K. Fragkos et al., “Twenty-Year Climatology of Solar UV and PAR in Cyprus: Integrating Satellite Earth Observations with Radiative Transfer Modeling,” Remote Sens., vol. 16, no. 11, p. 1878, May 2024, doi: 10.3390/rs16111878.
S. Madronich, “The Atmosphere and UV-B Radiation at Ground Level,” in Environmental UV Photobiology, Boston, MA: Springer US, 1993, pp. 1–39. doi: 10.1007/978-1-4899-2406-3_1.
L. González-Rodríguez et al., “Ultraviolet erythemal radiation in Central Chile: direct and indirect implication for public health,” Air Qual. Atmos. Heal., vol. 14, no. 10, pp. 1533–1548, Oct. 2021, doi: 10.1007/s11869-021-01037-3.
C. Voyant et al., “Machine learning methods for solar radiation forecasting: A review,” Renew. Energy, vol. 105, pp. 569–582, May 2017, doi: 10.1016/j.renene.2016.12.095.
K. P. Arandara, R. A. Attalage, and M. T. R. Jayasinghe, “Thermal Comfort with Evaporative Cooling for Tropical Climates,” in Proc. Int. Conf. Sustainable Built Environment (ICSBE-2010), 2010. [Online]. Available: http://www.civil.mrt.ac.lk/web/conference/ICSBE_2010/vol_01/51.pdf
B. Mayer and A. Kylling, “Technical note: The libRadtran software package for radiative transfer calculations - description and examples of use,” Atmos. Chem. Phys., vol. 5, no. 7, pp. 1855–1877, Jul. 2005, doi: 10.5194/acp-5-1855-2005.
A. Chatzopoulou, K. Tourpali, A. F. Bais, and P. Braesicke, “21st century surface UV radiation changes deduced from CMIP6 models: part I—evolution of major influencing factors,” Photochem. Photobiol. Sci., vol. 24, no. 1, pp. 89–109, Jan. 2025, doi: 10.1007/s43630-024-00675-7.
S. V. D, T. S. V S, and M. Meleet, “Ultraviolet index analysis and forecasting using deep learning methodologies for Bengaluru city,” EPRA Int. J. Multidiscip. Res., vol. 8, no. 7, 2022, [Online]. Available: https://eprajournals.com/IJMR/article/7361
S. S. Prasad, R. C. Deo, S. Salcedo-Sanz, N. J. Downs, D. Casillas-Pérez, and A. V. Parisi, “Enhanced joint hybrid deep neural network explainable artificial intelligence model for 1-hr ahead solar ultraviolet index prediction,” Comput. Methods Programs Biomed., vol. 241, p. 107737, Nov. 2023, doi: 10.1016/j.cmpb.2023.107737.
A. A. M. Ahmed, M. H. Ahmed, S. K. Saha, O. Ahmed, and A. Sutradhar, “Optimization algorithms as training approach with hybrid deep learning methods to develop an ultraviolet index forecasting model,” Stoch. Environ. Res. Risk Assess., vol. 36, no. 10, pp. 3011–3039, Oct. 2022, doi: 10.1007/s00477-022-02177-3.
A. S. Syahab, G. L. Pamungkas, and S. Akmal, “The Classification of Ultraviolet Index Using Logistic Regression and Random Forest methods for Predicting Extreme Conditions,” in Advances in Intelligent Systems Research, 2024, pp. 4–14. doi: 10.2991/978-94-6463-366-5_2.
E. Ervianto et al., “Assessing the Efficacy of the UV Index in Predicting Surface UV Radiation: A Comprehensive Analysis Using Statistical and Machine Learning Methods,” Indones. Rev. Phys., vol. 6, no. 2, pp. 99–121, Dec. 2023, doi: 10.12928/irip.v6i2.8216.
H. Zang, L. Liu, L. Sun, L. Cheng, Z. Wei, and G. Sun, “Short-term global horizontal irradiance forecasting based on a hybrid CNN-LSTM model with spatiotemporal correlations,” Renew. Energy, vol. 160, pp. 26–41, Nov. 2020, doi: 10.1016/j.renene.2020.05.150.
R. Zhang et al., “A hybrid deep learning model with error correction for photovoltaic power forecasting,” Front. Energy Res., vol. 10, Aug. 2022, doi: 10.3389/fenrg.2022.948308.
A. Taha, P. Makeen, and N. Nazih, “Short-term and long-term solar irradiance forecasting with advanced machine learning techniques in Zafarana, Egypt,” Sci. Rep., vol. 15, no. 1, p. 39553, Nov. 2025, doi: 10.1038/s41598-025-24853-4.
A. Taha, N. Nazih, and P. Makeen, “Wind speed prediction based on variational mode decomposition and advanced machine learning models in zaafarana, Egypt,” Sci. Rep., vol. 15, no. 1, p. 15599, May 2025, doi: 10.1038/s41598-025-98543-6.
I. G. N. P. Dharmayasa, C. A. Simatupang, and D. M. Sinaga, “NASA Power’s: an alternative rainfall data resources for hydrology research and planning activities in Bali Island, Indonesia,” J. Infrastruct. Plan. Eng., vol. 1, no. 1, pp. 1–7, Apr. 2022, doi: 10.22225/jipe.1.1.2022.1-7.
H. Kadhim Tayyeh and R. Mohammed, “Analysis of NASA POWER reanalysis products to predict temperature and precipitation in Euphrates River basin,” J. Hydrol., vol. 619, p. 129327, Apr. 2023, doi: 10.1016/j.jhydrol.2023.129327.
G. C. Rodrigues and R. P. Braga, “Evaluation of NASA POWER Reanalysis Products to Estimate Daily Weather Variables in a Hot Summer Mediterranean Climate,” Agronomy, vol. 11, no. 6, p. 1207, Jun. 2021, doi: 10.3390/agronomy11061207.
J. A. Ingio, A. S. Nsang, and A. Iorliam, “Optimizing Rice Production Forecasting Through Integrating Multiple Linear Regression with Recursive Feature Elimination,” J. Futur. Artif. Intell. Technol., vol. 1, no. 2, pp. 96–108, Aug. 2024, doi: 10.62411/faith.2024-17.
W. Aprianti and I. Mukhlash, “Handling Missing Value on Meteorological Data Classification with Rough Set Based Algorithm,” Glob. J. Pure Appl. Math., vol. 11, no. 3, pp. 1147–1155, 2015, [Online]. Available: https://www.ripublication.com/gjpam 2015/gjpamv11n3_01.pdf
M. Hasanpour Kashani and Y. Dinpashoh, “Evaluation of efficiency of different estimation methods for missing climatological data,” Stoch. Environ. Res. Risk Assess., vol. 26, no. 1, pp. 59–71, Jan. 2012, doi: 10.1007/s00477-011-0536-y.
International Commision on Non-Ionizing Radiation Protection, “UV Index: Global Solar Ultraviolet Index (UVI),” ICNIRP. https://www.icnirp.org/en/applications/uv-index/index.html
World Health Organization, “Radiation: The Ultraviolet (UV) Index,” World Health Organization, 2022. https://www.who.int/news-room/questions-and-answers/item/radiation-the-ultraviolet-(uv)-index
M. D. Teja and G. M. Rayalu, “Hybrid time series and machine learning models for forecasting cardiovascular mortality in India: an age specific analysis,” BMC Public Health, vol. 25, no. 1, p. 2150, Jun. 2025, doi: 10.1186/s12889-025-23318-7.
S. Bhuvaneshwari and S. N. S. Rajini, “A Novel Hybrid Model for Stock Price Forecasting: Combining ARIMA, Random Forests, and Gradient Boosting Techniques,” Libr. Prog. Int., vol. 44, no. 2s, 2024, doi: 10.48165/bapas.2024.44.2.1.
G. P. Zhang, “Time series forecasting using a hybrid ARIMA and neural network model,” Neurocomputing, vol. 50, pp. 159–175, Jan. 2003, doi: 10.1016/S0925-2312(01)00702-0.
L. B. Sina, C. A. Secco, M. Blazevic, and K. Nazemi, “Hybrid Forecasting Methods—A Systematic Review,” Electronics, vol. 12, no. 9, p. 2019, Apr. 2023, doi: 10.3390/electronics12092019.
G. H. Bernhard, A. F. Bais, P. J. Aucamp, A. R. Klekociuk, J. B. Liley, and R. L. McKenzie, “Stratospheric ozone, UV radiation, and climate interactions,” Photochem. Photobiol. Sci., vol. 22, no. 5, pp. 937–989, Apr. 2023, doi: 10.1007/s43630-023-00371-y.
F. X. Diebold and R. S. Mariano, “Comparing Predictive Accuracy,” J. Bus. Econ. Stat., vol. 13, no. 3, pp. 253–263, Jul. 1995, doi: 10.1080/07350015.1995.10524599.
P. J. Neale et al., “Environmental consequences of interacting effects of changes in stratospheric ozone, ultraviolet radiation, and climate: UNEP Environmental Effects Assessment Panel, Update 2024,” Photochem. Photobiol. Sci., vol. 24, no. 3, pp. 357–392, Mar. 2025, doi: 10.1007/s43630-025-00687-x.
B. Ladjal et al., “Hybrid deep learning CNN-LSTM model for forecasting direct normal irradiance: a study on solar potential in Ghardaia, Algeria,” Sci. Rep., vol. 15, no. 1, p. 15404, May 2025, doi: 10.1038/s41598-025-94239-z.
A. Chatzopoulou, K. Tourpali, A. F. Bais, and P. Braesicke, “Effects of different aerosol types on surface UV radiation in the 21st century,” Atmos. Environ., vol. 362, p. 121595, Dec. 2025, doi: 10.1016/j.atmosenv.2025.121595.
P. Kumari and D. Toshniwal, “Deep learning models for solar irradiance forecasting: A comprehensive review,” J. Clean. Prod., vol. 318, p. 128566, Oct. 2021, doi: 10.1016/j.jclepro.2021.128566.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Hiruni Navodya Kudagama, Samantha Mathara Arachchi, K. G. Amidu Yohan, C. L. I. S. Fonseka
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
