PREDICTION OF TÜRKİYE'S BURNED FOREST AREAS USING ARIMA MODEL

Year 2023, Volume: 33 Issue: 1, 347 - 355, 19.01.2023

Kübra Bağcı

https://doi.org/10.18069/firatsbed.1176961

Abstract

Abstract: Large-scale forest fires can cause significant ecological losses. Additionally, preserving forest areas may help to slow down climate change. Statistical models are one of the tools used in planning fire management strategies. In this study, the burned forest area of Türkiye is modeled using the Autoregressive Integrated Moving Average (ARIMA) method following the identification, estimation, validation, and forecasting steps. As is known the ARIMA analysis is one of the popular techniques used in time series analysis. Annual total burned forest areas in Türkiye over the period 1940-2021 are considered in the analysis. Three preliminary models are considered for evaluation of their modeling and prediction performances. The models' validities are investigated with Ljung–Box statistics, residual analysis, and cross-validation. According to the results, the ARIMA (3,1,0) model is found to be the most suitable model for predicting the future values of the burned forest area time series in Türkiye. Forecasts for Türkiye’s burned forest areas series are obtained for the next 3 years accordingly.

Keywords

ARIMA model, burned forest area, forest fires, time series, Türkiye

Türkiye'nin Yanan Orman Alanının ARIMA Modeli ile Tahmini

Abstract

Büyük ölçekli orman yangınları önemli ekolojik kayıplara neden olmaktadır. Ayrıca ormanlık alanların korunması iklim değişikliğini yavaşlatmaya yardımcı olabilmektedir. İstatistiksel modeller, yangın yönetimi stratejilerinin planlanmasında kullanılan araçlardan biridir. Bu çalışmada, Türkiye'nin yanan orman alanı, tanımlama, tahmin, doğrulama ve öngörü adımları izlenerek Otoregresif Bütünleşik Hareketli Ortalama (ARIMA) yöntemi kullanılarak analiz edilmiştir. ARIMA yöntemi zaman serileri analizinde kullanılan popüler tekniklerden biridir. Analizde 1940-2021 yılları arasında Türkiye'deki yıllık toplam yanan orman alanı ölçümleri kullanılmıştır. Modelleme ve tahmin performanslarının değerlendirilmesi için üç model ele alınmıştır. Modellerin geçerliliği, Ljung–Box istatistikleri ve çapraz doğrulama ile araştırılmıştır. Sonuçlara göre, ARIMA (3,1,0) modeli Türkiye'nin yanmış orman alanı zaman serilerinin gelecek değerlerinin tahmin edilmesi için en uygun model olarak bulunmuş ve öngörüler önümüzdeki 3 yıl için elde edilmiştir.

Keywords

ARIMA, orman yangını, Türkiye, yanan orman alanı, zaman serisi

References

• Amatulli, G., Camia, A., & San-Miguel-Ayanz, J. (2013). Estimating future burned areas under changing climate in the EU-Mediterranean countries. Science of The Total Environment, 450–451, 209–222. Elsevier.
• Baş, R. (2014). Türkiye’de orman yangınları nedenleri, zararları ve yangınlara karşı alınacak önlemler. Journal of the Faculty of Forestry Istanbul University, 27(2), 52–73.
• Boubeta, M., Lombardía, M. J., González-Manteiga, W., & Marey-Pérez, M. F. (2016). Burned area prediction with semiparametric models. International Journal of Wildland Fire, 25(6), 669–678. CSIRO Publishing.
• Boubeta, M., Lombardía, M. J., Marey-Pérez, M. F., & Morales, D. (2015). Prediction of forest fires occurrences with area-level Poisson mixed models. Journal of Environmental Management, 154, 151–158. Academic Press.
• Çekim, H. Ö., Kadilar, C., & Özel, G. (2013). Characterizing forest fire activity in Turkey by compound Poisson and time series models. AIP Conference Proceedings, 1558(1), 1442. American Institute of PhysicsAIP. Retrieved January 30, 2022, from https://aip.scitation.org/doi/abs/10.1063/1.4825789
• Chen, W., Moriya, K., Sakai, T., Koyama, L., & Cao, C. X. (2016). Mapping a burned forest area from Landsat TM data by multiple methods. Geomatics, Natural Hazards and Risk, 7(1), 384–402. Taylor & Francis. Retrieved from https://doi.org/10.1080/19475705.2014.925982
• Çolak, E., & Sunar, F. (2020). Evaluation of forest fire risk in the Mediterranean Turkish forests: A case study of Menderes region, Izmir. International Journal of Disaster Risk Reduction, 45, 101479. Elsevier.
• EAA. (n.d.). EAA. 2021. Date of access: 03.05.2022. https://www.eea.europa.eu/ims/forest-fires-in-europe
• Fernández-Manso, A., Quintano, C., & Fernández-Manso, O. (2011). Forecast of NDVI in coniferous areas using temporal ARIMA analysis and climatic data at a regional scale. International Journal of Remote Sensing, 32(6), 1595–1617. Taylor & Francis. Retrieved from https://doi.org/10.1080/01431160903586765
• Giannakopoulos, C., le Sager, P., Bindi, M., Moriondo, M., Kostopoulou, E., & Goodess, C. M. (2009). Climatic changes and associated impacts in the Mediterranean resulting from a 2 °C global warming. Global and Planetary Change, 68(3), 209–224. Retrieved from https://www.sciencedirect.com/science/article/pii/S0921818109001131
• Hyndman, R. J. (2014). Measuring forecast accuracy. Business forecasting: Practical problems and solutions, 177–183. Wiley.
• Kouassi, J.-L., Wandan, N., & Mbow, C. (2020). Predictive Modeling of Wildfire Occurrence and Damage in a Tropical Savanna Ecosystem of West Africa. Fire, 3(3). Retrieved from https://www.mdpi.com/2571-6255/3/3/42
• Küçük Matcı, D., & Avdan, U. (2020). Comparative analysis of unsupervised classification methods for mapping burned forest areas. Arabian Journal of Geosciences, 13(15), 711. Retrieved from https://doi.org/10.1007/s12517-020-05670-7
• Küçükosmanoğlu, A. (1987). İstatistiklerle Türkiye’de orman yangınları. Journal of the Faculty of Forestry Istanbul University, 37(3), 103–106.
• Kwiatkowski, D., Phillips, P. C. B., Schmidt, P., & Shin, Y. (1992). Testing the null hypothesis of stationarity against the alternative of a unit root: How sure are we that economic time series have a unit root? Journal of Econometrics, 54(1–3), 159–178. North-Holland.
• Makridakis, S., Spiliotis, E., & Assimakopoulos, V. (2018). Statistical and Machine Learning forecasting methods: Concerns and ways forward. (A. R. Hernandez Montoya, Ed.)PLOS ONE, 13(3), e0194889. Public Library of Science. Retrieved November 5, 2022, from https://dx.plos.org/10.1371/journal.pone.0194889
• Maleki, A., Nasseri, S., Aminabad, M. S., & Hadi, M. (2018). Comparison of ARIMA and NNAR Models for Forecasting Water Treatment Plant’s Influent Characteristics. KSCE Journal of Civil Engineering, 22(9), 3233–3245. Retrieved from https://doi.org/10.1007/s12205-018-1195-z
• Mohammadi, F., Bavaghar, M. P., & Shabanian, N. (2014). Forest Fire Risk Zone Modeling Using Logistic Regression and GIS: An Iranian Case Study. Small-scale Forestry, 13(1), 117–125. Retrieved from https://doi.org/10.1007/s11842-013-9244-4
• Mueller, S. E., Thode, A. E., Margolis, E. Q., Yocom, L. L., Young, J. D., & Iniguez, J. M. (2020). Climate relationships with increasing wildfire in the southwestern US from 1984 to 2015. Forest Ecology and Management, 460, 117861. Elsevier.
• OGM. (2018). General Directorate of Forestry (OGM). Date of access: 03.05.2022. https://www.ogm.gov.tr/tr/e-kutuphane/resmi-istatistikler
• Oncel Cekim, H., Güney, C. O., Şentürk, Ö., Özel, G., & Özkan, K. (2021). A novel approach for predicting burned forest area. Natural Hazards, 105(2), 2187–2201. Springer Science and Business Media B.V.
• Özbayoǧlu, A. M., & Bozer, R. (2012). Estimation of the Burned Area in Forest Fires Using Computational Intelligence Techniques. Procedia Computer Science, 12, 282–287. Elsevier.
• Papakosta, P., & Straub, D. (2017). Probabilistic prediction of daily fire occurrence in the Mediterranean with readily available spatio-temporal data. iForest - Biogeosciences and Forestry, 10(1), 32–40. SISEF - Italian Society of Silviculture and Forest Ecology. Retrieved from https://iforest.sisef.org/contents/?id=ifor1686-009
• Pena, D., Tiao, G. C., & Tsay, R. S. (2011). A course in time series analysis (Vol. 322). John Wiley & Sons.
• Podur, J. J., Martell, D. L., & Stanford, D. (2010). A compound Poisson model for the annual area burned by forest fires in the province of Ontario. Environmetrics, 21(5), 457–469. John Wiley & Sons, Ltd. Retrieved from https://doi.org/10.1002/env.996
• Sari, F. (2021). Forest fire susceptibility mapping via multi-criteria decision analysis techniques for Mugla, Turkey: A comparative analysis of VIKOR and TOPSIS. Forest Ecology and Management, 480, 118644. Elsevier.
• Satir, O., Berberoglu, S., & Donmez, C. (2016). Mapping regional forest fire probability using artificial neural network model in a Mediterranean forest ecosystem. Geomatics, Natural Hazards and Risk, 7(5), 1645–1658. Taylor & Francis. Retrieved from https://doi.org/10.1080/19475705.2015.1084541
• Shumway, R. H., Stoffer, D. S., & Stoffer, D. S. (2000). Time series analysis and its applications (Vol. 3). Springer.
• Tasker, K. A., & Arima, E. Y. (2016). Fire regimes in Amazonia: The relative roles of policy and precipitation. Anthropocene, 14, 46–57. Elsevier.
• Tedim, F., Leone, V., Amraoui, M., Bouillon, C., Coughlan, M. R., Delogu, G. M., Fernandes, P. M., et al. (2018). Defining extreme wildfire events: difficulties, challenges, and impacts. Fire, 1(1), 9. Multidisciplinary Digital Publishing Institute.
• Xie, Y., & Peng, M. (2019). Forest fire forecasting using ensemble learning approaches. Neural Computing and Applications, 31(9), 4541–4550. Retrieved from https://doi.org/10.1007/s00521-018-3515-0