Van İli Sıcaklık Tahmininde SARIMA Modeli: Geleceğe Dönük Bir Öngörü

Year 2025, Issue: Van Gölü Havzası Özel Sayısı, 398 - 418, 30.10.2025

Burak Uyar , Ferayi Güzel Urcan

https://doi.org/10.53568/yyusbed.1678607

Abstract

İklim, dünyadaki yaşamın sürdürülebilirliği açısından kritik bir rol oynamaktadır ve sürekli değişen iklim koşulları, gezegen için büyük bir tehdit oluşturmaktadır. Bu durum, gelecekteki iklim verilerinin doğru bir şekilde tahmin edilmesini zorlaştırmaktadır. Birçok algoritma geliştirilmiş olsa da, henüz tam anlamıyla iklim verileriyle ilgili doğru tahminler yapılabilmiş değildir. Zaman serisi analizi, özellikle ARIMA algoritması, gelecekteki sıcaklık verilerini tahmin etmek için kullanılmaktadır. Potansiyel modeller, otokorelasyon fonksiyonları ve kısmi otokorelasyon fonksiyonları ile belirlenirken, en uygun model AIC gibi performans ölçütlerine dayalı olarak seçilmektedir. Seçilen modelin doğruluğu, Dickey Fuller testi ile doğrulanmakta ve SARIMA (p, d, q) × (P, D, Q) m modeli, en düşük AIC değerine sahip olan ile belirlenmektedir. Son olarak, SARIMA modeli kullanılarak gelecekteki sıcaklık tahminleri yapılmakta ve tahmin doğruluğu, kök ortalama kare hatası (RMSE) ve ortalama kare hatası (MSE) gibi ölçütlerle değerlendirilmektedir. Bu çalışmada, Van iline ait sıcaklık verilerinin tahmini için mevsimsel ARIMA modeli kullanılmıştır. Veri analizi sürecinde, ADF ve KPSS testleriyle serinin durağan olup olmadığı kontrol edilmiş ve gerekli dönüşümler yapılmıştır. Elde edilen en uygun model SARIMA(1, 1, 1) × (1, 0, 1, 12) olarak belirlenmiş ve modelin doğruluğu test edilmiştir. Modelin tahmin doğruluğu RMSE ve MSE değerleri ile doğrulanmış ve gelecekteki sıcaklık tahminlerinin doğruluğu oldukça yüksek bulunmuştur. SARIMA modeli, Van ilindeki sıcaklık değişimlerinin doğru tahmin edilmesinde etkili bir araçtır ve gelecekteki iklim tahminleri için güvenilir bir model olarak kullanılabilir.

Keywords

ARIMA modeli , SARIMA modeli , sıcaklık tahmini , sınıflandırma , zaman serileri.

SARIMA Model in Temperature Forecasting for Van Province: A Forward-Looking Prediction

Abstract

Climate plays a critical role in the sustainability of life on Earth, and the constantly changing climate conditions pose a significant threat to the planet. This makes it difficult to accurately predict future climate data. Although many algorithms have been developed, accurate predictions regarding climate data have not yet been fully achieved. Time series analysis, particularly the ARIMA (Autoregressive Integrated Moving Average) algorithm, is used to predict future temperature data. Potential models are determined using autocorrelation functions (ACF) and partial autocorrelation functions (PACF), while the best model is selected based on performance metrics such as the Akaike Information Criterion (AIC). The accuracy of the selected model is validated through the Dickey-Fuller (ADF) test, and the Seasonal Auto-Regressive Integrated Moving Average (SARIMA) (p, d, q) × (P, D, Q) m model is determined based on the lowest AIC value. Finally, future temperature predictions are made using the SARIMA model, and prediction accuracy is evaluated using metrics such as Root Mean Squared Error (RMSE) and Mean Squared Error (MSE). In this study, the seasonal ARIMA model was used to predict temperature data for Van Province. During the data analysis process, stationarity of the series was tested using the ADF and KPSS tests, and necessary transformations were applied. The best model obtained was identified as SARIMA(1, 1, 1) × (1, 0, 1, 12), and the model's accuracy was tested with high precision. The prediction accuracy of the model was validated using RMSE and MSE values, and the accuracy of the future temperature predictions was found to be very high. The SARIMA model is an effective tool for accurately predicting temperature changes in Van Province and can be used as a reliable model for future climate predictions.

Keywords

ARIMA model , classification , SARIMA model , temperature prediction , time series

References

• Adhikari, R., Agrawal, R. K., (2013). An introductory study on time series modeling and forecasting. International Congress on Engineering-Engineering for Evolution. 27–29 November 2019, Covilhã, Portugal. 857-871.
• Bilgili, M., Pinar, E. & Durhasan, T. Global monthly sea surface temperature forecasting using the SARIMA, LSTM, and GRU models. Earth Sci Inform 18, 10 (2025). https://doi.org/10.1007/s12145-024-01585-z
• Box, G. E., Jenkins, G. M., Reinsel, G. C., Ljung, G. M., (2015). Time Series Analysis: Forecasting and Control. John Wiley & Sons, Inc. No: 5, Hoboken, New Jersey. 712.
• Brath, A., Montanari, A., Toth, E., (1999). Short-term rainfall prediction with time series analysis techniques for real-time flash flood forecasting. Mediterranean Storms, Proceedings EGS Plinius Conference, Maratea. 14-16 October 1999, Bologna, Italy. 176-189.
• Charles Z., 2022. Seasonal ARIMA (SARIMA). https://www.real-statistics.com/time-series-analysis/seasonal-arima-sarima/ Real Statistics Using Excel, Italian. Erişim tarihi: 08.03.2022.
• Dimri, T., Ahmad, S., Sharif, M., (2020). Time series analysis of climate variables using seasonal ARIMA approach. Journal of Earth System Science, 129 (1): 1-16.
• Hardwinarto, S., Haviluddin, M., Aipassa, M., (2015). Rainfall monthly prediction based on artificial neural network: a case study in Tenggarong Station. Procedia Computer Science, 59 (1): 142-151.
• Hossain, M. L., Shams, S. M. N., & Ullah, S. M. M. (2025). Time series and deep learning approaches for renewable energy forecasting in Dhaka: a comparative study of ARIMA, SARIMA, and LSTM models. Discover Sustainability, 6, Article 775. https://doi.org/10.1007/s43621 025 01733 5
• Hyndman, R. J., Athanasopoulos, G., (2018). Forecasting: principles and practice. https://otexts.com/fpp2/seasonal-arima.html. OTexts, Australia. Erişim tarihi: 12.03.2022.
• Wuzhe Huang, Fa Si, Feifei Han, Jiahao Liu, Jingshi Zheng, and Yuwen Wei "Global temperature prediction based on SARIMA+LSTM model", Proc. SPIE 12756, 3rd International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2023), 127563R (28 July 2023); https://doi.org/10.1117/12.2686399
• John, A., Marohasy, J., (2017). Application of artificial neural networks to forecasting monthly rainfall one year in advance for locations within the Murray Darling Basin, Australia. International Journal of Sustainable Development and Planning, 12 (8): 1282-1298.
• Luk, K. C., Ball, J. E., Sharma, A., (2001). An application of artificial neural networks for rainfall forecasting. Mathematical and Computer odelling, 33 (6-7): 683-693.
• Nhita, F., Saepudin, D., Wisesty, U. N., (2015). Comparative Study of Moving Average on Rainfall Time Series Data for Rainfall Forecasting Based on Evolving Neural Network Classifier. Computational and Business Intelligence (ISCBI). 7-9 December 2015, Bali, Indonesia. 112-116.
• Nikam, V. B., Meshram, B. B., (2013). Modeling rainfall prediction using data mining method: A Bayesian approach. Fifth International Conference on Computational Intelligence, Modelling and Simulation. 24-25 September 2013, Seoul, Korea. 132-136.
• Raicharoen, T., Lursinsap, C., Sanguanbhokai, P., (2003). Application of critical support vector machine to time series prediction. Proceedings of the 2003 International Symposium on Circuits and Systems, 2003. ISCAS '03. 25-28 May 2003, Bangkok, Thailand. 5-5.
• Salman, A. G., Kanigoro, B., (2021). Visibility forecasting using autoregressive integrated moving average (ARIMA) models. Procedia Computer Science, 179 : 252-259.
• Sawale, G. J., Gupta, S. R., (2013). Use of artificial neural network in data mining for weather forecasting. International Journal Of Computer Science And Applications, 6 (2): 383-387.
• Shoba, G., Shobha, G., (2014). Rainfall prediction using data mining techniques: A survey. International Journal of Engineering and Computer Science, 3 (5): 6206-6211.
• Tektaş, M., (2010). Weather forecasting using ANFIS and ARIMA models. Environmental Research, Engineering and Management, 51 (1): 5-10.
• Wang, L., Wu, J., (2012). Application of hybrid RBF neural network ensemble model based on wavelet support vector machine regression in rainfall time series forecasting. 2012 Fifth International Joint Conference on Computational Sciences and Optimization. 23-26 June 2012, Harbin. 867-871.
• Xu, D., Min, J., Shen, F., Ban, J., Chen, P., (2016). Assimilation of MWHS radiance data from the FY‐3B satellite with the WRF Hybrid‐3DVAR system for the forecasting of binary typhoons. Journal of Advances in Modeling Earth Systems, 8 (2): 1014-1028.