Short-Term High-Frequency Temperature Forecasting Using Hybrid Stacking Ensemble

Kaan Arık; Burak Ağgül

doi:10.33187/jmsm.1837729

Short-Term High-Frequency Temperature Forecasting Using Hybrid Stacking Ensemble

Abstract

Weather forecasting is valuable for agriculture through crop production, renewable resource management such as solar or wind farms, and disaster management for preparation and response. It enables an understanding of future conditions so that decisions can be made at the right time. To make well-founded decisions about the long-term use of limited resources, it is essential to know accurately when these resources will be available. In this research, we used a high-frequency multivariate sensor dataset recorded every 10 minutes from Max Planck Institute weather station. This dataset includes 20 atmospheric parameters, such as temperature, humidity, pressure, and wind direction. We developed an appropriate preprocessing pipeline to prepare raw data for analysis. Our pipeline included converting timestamps into date/time format, engineering circular wind-direction features, interpolating missing values, splitting the training and test sets chronologically, and normalizing the features using Min--Max scaling. Using the processed dataset, we compared time-series models (SARIMA and Prophet) with tree-based machine learning models (Random Forest, XGBoost, and CatBoost), as well as a hybrid stacking ensemble that combined the strengths of these models. Performance was assessed on a 7-day test set using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and \(R^2\). Our results showed that the Hybrid Stacking Ensemble was the most effective overall model, achieving \(\mathrm{MAE}=0.255\,^{\circ}\mathrm{C}\), \(\mathrm{RMSE}=0.329\,^{\circ}\mathrm{C}\), and \(R^2=0.949\). It significantly outperformed both the individual machine learning models and the baseline models, such as SARIMA and Prophet. Hybrid Stacking Ensemble reduced the error of the best-performing individual model by \(26\%\), while maintaining near-perfect linear agreement with the observed temperatures.

Keywords

High-resolution weather sensing, Data-driven weather modeling, Machine learning, Hybrid stacking ensemble learning, Multivariate climate time series

References

M. H. F. da Silva, G. F. de Jesus, C. Nascimento, et al., Exploring quantum machine learning for weather forecasting, Braz. J. Phys., 56(1) (2026), Article ID 22. https://doi.org/10.1007/s13538-025-01941-4
T. P. Agyekum, P. Antwi-Agyei, A. J. Dougill, The contribution of weather forecast information to agriculture, water, and energy sectors in East and West Africa: A systematic review, Front. Environ. Sci., 10 (2022), Article ID 935696. https://doi.org/10.3389/fenvs.2022.935696
W. Atwa, A. A. Almazroi, N. Ayub, Reliable renewable energy forecasting for climate change mitigation, PeerJ Comput. Sci., 10 (2024), Article ID e2067. https://doi.org/10.7717/peerj-cs.2067
A. C. Gonc¸alves, X. Costoya, R. Nieto, et al., Extreme weather events on energy systems: A comprehensive review on impacts, mitigation, and adaptation measures, Sustainable Energy Res., 11(1) (2024), Article ID 4. https://doi.org/10.1186/s40807-023-00097-6
J. A. Segovia, J. F. Toaquiza, J. R. Llanos, et al., Meteorological variables forecasting system using machine learning and open-source software, Electronics, 12(4) (2023), Article ID 1007. https://doi.org/10.3390/electronics12041007
R. Meenal, D. Binu, K. C. Ramya, et al., Weather forecasting for renewable energy system: A review, Arch. Comput. Methods Eng., 29 (2022), 2875–2891. https://doi.org/10.1007/s11831-021-09695-3
K. Lagouvardos, V. Kotroni, A. Bezes, et al., The automatic weather stations NOANN network of the National Observatory of Athens: Operation and database, Geosci. Data J., 4(1) (2017), 4–16. https://doi.org/10.1002/gdj3.44
H. Lira, L. Mart´ı, N. Sanchez-Pi, A graph neural network with spatio-temporal attention for multi-sources time series data: An application to frost forecast, Sensors, 22(4) (2022), Article ID 1486. https://doi.org/10.3390/s22041486
R. Tsela, S. Maladaki, S. Kolios, An integrated, automated and modular approach for real-time weather monitoring of surface meteorological variables and short-range forecasting using machine learning, Environ. Model. Softw., 183 (2025), Article ID 106203. https://doi.org/10.1016/j.envsoft.2024.106203
V. Papastefanopoulos, P. Linardatos, T. Panagiotakopoulos, et al., Multivariate time-series forecasting: A review of deep learning methods in internet of things applications to smart cities, Smart Cities, 6(5) (2023), 2519–2552. https://doi.org/10.3390/smartcities6050114

R. Wu, Y. Liang, L. Lin, et al., Spatiotemporal multivariate weather prediction network based on CNN-Transformer, Sensors, 24(23) (2024), Article ID 7837. https://doi.org/10.3390/s24237837
Z. Pu, E. Kalnay, Numerical weather prediction basics: Models, numerical methods, and data assimilation, in: Handbook of Hydrometeorological Ensemble Forecasting, Springer, 2018, pp. 67–97. https://doi.org/10.1007/978-3-642-39925-1 11
E. Kalnay, Atmospheric Modeling, Data Assimilation and Predictability, Cambridge University Press, 2003. https://doi.org/10.1017/CBO9780511802270
P. Bauer, A. Thorpe, G. Brunet, The quiet revolution of numerical weather prediction, Nature, 525(7567) (2015), 47–55. https://doi.org/10.1038/nature14956
D. S. Wilks, Statistical Methods in the Atmospheric Sciences, 3rd ed., Academic Press, 2011.
E. P. Agbo, The role of statistical methods and tools for weather forecasting and modeling, Weather Forecasting, 3rd ed., IntechOpen, 2021. https://doi.org/10.5772/intechopen.96854
D. Kampolis, Statistical analysis for long-term weather forecast, Environ. Sci. Proc., 26(1) (2023), Article ID 30. https://doi.org/10.3390/environsciproc2023026030
K. Bi, L. Xie, H. Zhang, et al., Accurate medium-range global weather forecasting with 3D neural networks, Nature, 619(7970) (2023), 533–538. https://doi.org/10.1038/s41586-023-06185-3
K. Han, J. Choi, C. Kim, Comparison of statistical post-processing methods for probabilistic wind speed forecasting, Asia-Pac. J. Atmos. Sci., 54(1) (2018), 91–101.https://doi.org/10.1007/s13143-017-0062-z
C. Kummerow, W. Barnes, T. Kozu, et al., The Tropical Rainfall Measuring Mission (TRMM) sensor package, J. Atmos. Ocean. Technol., 15(3) (1998), 809–817.
T. Kozu, T. Kawanishi, H. Kuroiwa, et al., Development of precipitation radar onboard the Tropical Rainfall Measuring Mission (TRMM) satellite, IEEE Trans. Geosci. Remote Sens., 39(1) (2001), 102–116. https://doi.org/10.1109/36.898669
A. Y. Hou, R. K. Kakar, S. Neeck, et al., The Global Precipitation Measurement mission, Bull. Amer. Meteorol. Soc., 95(5) (2014), 701–722. https://doi.org/10.1175/BAMS-D-13-00164.1
T. Iguchi, T. Kozu, R. Meneghini, et al., Rain-profiling algorithm for the TRMM precipitation radar, J. Appl. Meteorol., 39(12) (2000), 2038–2052.
A. Okazaki, T. Honda, S. Kotsuki, et al., Simulating precipitation radar observations from a geostationary satellite, Atmos. Meas. Tech., 12(7) (2019), 3985–3996. https://doi.org/10.5194/amt-12-3985-2019
S. Siami-Namini, N. Tavakoli, A. S. Namin, A comparison of ARIMA and LSTM in forecasting time series, in: Proc. 17th IEEE Int. Conf. Mach. Learn. Appl. (ICMLA), IEEE, 2018, pp. 1394–1401. https://doi.org/10.1109/ICMLA.2018.00227
J. Brownlee, Multivariate time series forecasting with LSTMs in Keras, Machine Learning Mastery, 2017. Available at: https://machinelearningmastery.com/multivariate-time-series-forecasting-lstms-keras/
G. Naresh, B. Indira, Air pollution prediction using multivariate LSTM deep learning model, Int. J. Intell. Syst. Appl. Eng., 12(8s) (2023), 211–220.
T. Saini, P. Chaturvedi, V. Dutt, Modelling particulate matter using multivariate and multistep recurrent neural networks, Front. Environ. Sci., 9 (2021), Article ID 752318. https://doi.org/10.3389/fenvs.2021.752318
S. Arslan, A hybrid forecasting model using LSTM and Prophet for energy consumption with decomposition of time series data, PeerJ Comput. Sci., 8 (2022), Article ID e1001. https://doi.org/10.7717/peerj-cs.1001
S. Y. Shih, F. K. Sun, H. Y. Lee, Temporal pattern attention for multivariate time series forecasting, Mach. Learn., 108(8) (2019), 1421–1441. https://doi.org/10.1007/s10994-019-05815-0
M. K. Saravana, M. S. Roopa, J. S. Arunalatha, et al., Transformers for multivariate time series forecasting: Comprehensive analysis, challenges, research opportunities and future prospects, IEEE Access, 14 (2026), 11424–11457. https://doi.org/10.1109/ACCESS.2026.3654408
S. Mishra, C. Bordin, K. Taharaguchi, et al., Comparison of deep learning models for multivariate prediction of time series wind power generation and temperature, Energy Reports, 6 (2020), 273–286. https://doi.org/10.1016/j.egyr.2019.11.009
J. Zhang, B. Wang, M. Hua, et al., Spatiotemporal meteorological prediction based on fully convolutional neural network, IEEE Trans. Geosci. Remote Sens., 62 (2024), 1–18. http://doi.org/10.1109/TGRS.2024.3439390
P. Hewage, A. Behera, M. Trovati, et al., Temporal Convolutional Neural (TCN) network for an effective weather forecasting using time-series data from the local weather station, Soft Comput., 24 (2020), 16453–16482. https://doi.org/10.1007/s00500-020-04954-0
M. M. Villia, G. Tsagkatakis, M. Moghaddam, et al., Embedded temporal convolutional networks for essential climate variables forecasting, Sensors, 22(5) (2022), Article ID 1851. https://doi.org/10.3390/s22051851
J. Grigsby, Z. Wang, Y. Qi, Long-range transformers for dynamic spatiotemporal forecasting, arXiv preprint arXiv:2109.12218, 2021. https://arxiv.org/abs/2109.12218
L. Zhao, S. Lu, D. Qi, Improvement of maximum air temperature forecasts using a stacking ensemble technique, Atmosphere, 14(3) (2023), Article ID 600. https://doi.org/10.3390/atmos14030600
X. Li, L. Zhang, X. Wang, B. Liang, Forecasting greenhouse air and soil temperatures: A multi-step time series approach employing attention-based LSTM network, Comput. Electron. Agric., 217 (2024), Article ID 108602. https://doi.org/10.1016/j.compag.2023.108602
H. Zhang, Y. Liu, C. Zhang, et al., Machine learning methods for weather forecasting: A survey, Atmosphere, 16(1) (2025), Article ID 82. https://doi.org/10.3390/atmos16010082
A. King, Weather Long-term Time Series Forecasting [Dataset], Kaggle, 2020. Available at: https://www.kaggle.com/datasets/arashnic/max-planck-weather-dataset
S. J. Taylor, B. Letham, Forecasting at scale, Amer. Statist., 72(1) (2018), 37–45. https://doi.org/10.1080/00031305.2017.1380080
G. James, D. Witten, T. Hastie, et al., An Introduction to Statistical Learning with Applications in Python, 2nd ed., Springer, 2021. https://doi.org/10.1007/978-3-031-38747-0

Details

Primary Language

English

Subjects

Artificial Intelligence (Other)

Journal Section

Research Article

Authors

Kaan Arık ^*
0000-0002-0930-8955
Türkiye

Burak Ağgül
0000-0002-9183-1568
Türkiye

Early Pub Date

June 6, 2026

Publication Date

-

Submission Date

December 7, 2025

Acceptance Date

May 18, 2026

Published in Issue

Year 2026 Number: Advanced Online Publication

DOI

https://doi.org/10.33187/jmsm.1837729

IZ

https://izlik.org/JA58PN36RD

APA

Arık, K., & Ağgül, B. (2026). Short-Term High-Frequency Temperature Forecasting Using Hybrid Stacking Ensemble. Journal of Mathematical Sciences and Modelling, Advanced Online Publication, 120-136. https://doi.org/10.33187/jmsm.1837729

AMA

1.Arık K, Ağgül B. Short-Term High-Frequency Temperature Forecasting Using Hybrid Stacking Ensemble. Journal of Mathematical Sciences and Modelling. 2026;(Advanced Online Publication):120-136. doi:10.33187/jmsm.1837729

Chicago

Arık, Kaan, and Burak Ağgül. 2026. “Short-Term High-Frequency Temperature Forecasting Using Hybrid Stacking Ensemble”. Journal of Mathematical Sciences and Modelling, no. Advanced Online Publication: 120-36. https://doi.org/10.33187/jmsm.1837729.

EndNote

Arık K, Ağgül B (June 1, 2026) Short-Term High-Frequency Temperature Forecasting Using Hybrid Stacking Ensemble. Journal of Mathematical Sciences and Modelling Advanced Online Publication 120–136.

IEEE

[1]K. Arık and B. Ağgül, “Short-Term High-Frequency Temperature Forecasting Using Hybrid Stacking Ensemble”, Journal of Mathematical Sciences and Modelling, no. Advanced Online Publication, pp. 120–136, June 2026, doi: 10.33187/jmsm.1837729.

ISNAD

Arık, Kaan - Ağgül, Burak. “Short-Term High-Frequency Temperature Forecasting Using Hybrid Stacking Ensemble”. Journal of Mathematical Sciences and Modelling. Advanced Online Publication (June 1, 2026): 120-136. https://doi.org/10.33187/jmsm.1837729.

JAMA

1.Arık K, Ağgül B. Short-Term High-Frequency Temperature Forecasting Using Hybrid Stacking Ensemble. Journal of Mathematical Sciences and Modelling. 2026;:120–136.

MLA

Arık, Kaan, and Burak Ağgül. “Short-Term High-Frequency Temperature Forecasting Using Hybrid Stacking Ensemble”. Journal of Mathematical Sciences and Modelling, no. Advanced Online Publication, June 2026, pp. 120-36, doi:10.33187/jmsm.1837729.

Vancouver

1.Kaan Arık, Burak Ağgül. Short-Term High-Frequency Temperature Forecasting Using Hybrid Stacking Ensemble. Journal of Mathematical Sciences and Modelling. 2026 Jun. 1;(Advanced Online Publication):120-36. doi:10.33187/jmsm.1837729