KaydedilenlerDerginizi ekleyin

Dergiler / Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi / 2026 / Cilt: 15

Makine öğrenmesi ile ileriye yönelik kuraklıkların tahmin edilmesi

Forecasting future droughts with machine learning

Türker Tuğrul(GAZİ ÜNİVERSİTESİ, TEKNOLOJİ FAKÜLTESİ, İNŞAAT MÜHENDİSLİĞİ BÖLÜMÜ)0000-0001-7547-847X
Sayfa
1–11
Yayın Tarihi
2026-01-15
🏛️Bu makalenin bibliyografik verileri ve tam metin bağlantısı DergiPark üzerinden sağlanmıştır.
DergiPark Kaydı ↗
✨

Yapay Zeka Hızlı Bakış & Çıkarımlar

AI Insight
⚡ TL;DR(1 Cümlede Hızlı Çıkarım)

Elde edilen sonuçlarda, WT’den SVM’nin M02 (SVMWM02) modelinin en iyi sonucu verdiği tespit edilmiştir.

📌 3 Maddede Bu Makale

🎯
Amaç & Hipotez:İnsanlar ve yaşam alanları için tehdit oluşturan ciddi bir doğal afet olan kuraklık, ilk başta fark edilmesi zor olup, asıl etkileri gerçekleştiğinde fark edilebilmektedir.
📊
Temel Bulgular & Veri:Kuraklığı tespit etmek, önlemek, izlemek ve etkilerini azaltmak için çok sayıda yöntem önerilmiştir.
💡
Sonuç & Katkı:Elde edilen sonuçlarda, WT’den SVM’nin M02 (SVMWM02) modelinin en iyi sonucu verdiği tespit edilmiştir.
🔬Tür:Modelleme & Sayısal Simülasyon
📍Kapsam:2023 Verileri
⚙️Değişkenler:Kuraklık · Makine öğrenmesi · Destek vektör makineleri

Özet

İnsanlar ve yaşam alanları için tehdit oluşturan ciddi bir doğal afet olan kuraklık, ilk başta fark edilmesi zor olup, asıl etkileri gerçekleştiğinde fark edilebilmektedir. Kuraklığı tespit etmek, önlemek, izlemek ve etkilerini azaltmak için çok sayıda yöntem önerilmiştir. Bunların arasından araştırmacıların sıklıkla tercih ettiği yöntemlerden birisi Standartlaştırılmış Yağış Evapotranspirasyon İndeksi (SPEI)’dir. Bu yöntem kuraklıkların tespiti ve izlenmesi amacıyla kullanıldığı için, ileriye yönelik herhangi bir bilgi sunmamaktadır. Dolayısıyla, ileriye yönelik tahminlerin yapılabilmesi amacıyla makine öğrenmesi gibi tekniklerden yararlanılmaktadır. Bu çalışmada da Türkiye’nin batı bölgesinde yer alan Edirne için ileriye yönelik kuraklık tahmin modelleri oluşturulmuş ve bu modeller makine öğrenme yöntemleri ile analiz edilmiştir. Uzun-Kısa Süreli Bellek (LSTM) ve Destek Vektör Makinesi (SVM) tercih edilen yöntemler olmuştur. 4 farklı kuraklık modeli oluşturmak için cross-correlation yönteminden yararlanılmıştır. Hidrolik kuraklıkları en iyi temsil ettiği varsayılan 1901-2023 yılları arasındaki SPEI12 verisinden yararlanılmıştır. Ayrıca, model performanslarının artırılması amacıyla Dalgacık Dönüşümü (WT) kullanılmıştır. Model sonuçlarının irdelenebilmesi için Korelasyon Katsayısı (r), Kök Ortalama Kare Hatası (RMSE), Nash–Sutcliffe Verimliliği (NSE), Kling–Gupta Verimliliği (KGE) ve Performans İndeks (PI), performans metriklerinden yararlanılmıştır. Elde edilen sonuçlarda, WT’den SVM’nin M02 (SVMWM02) modelinin en iyi sonucu verdiği tespit edilmiştir. Bu model, Edirne için yapılacak olan ileriye yönelik kuraklık modellinde kullanılmalıdır.

Abstract

Drought, a severe natural disaster that poses a threat to humans and their living environments, is difficult to detect at first and its actual impacts often become noticeable only after they occur. Numerous methods have been proposed to detect, prevent, monitor, and mitigate the effects of drought. Among these, one of the most commonly preferred by researchers is the Standardized Precipitation Evapotranspiration Index (SPEI). Since this method is used for detecting and monitoring droughts, it does not provide any future information. Therefore, techniques such as machine learning are utilized to make future predictions. In this study, future drought prediction models were developed for Edirne, which is located in the western region of Türkiye, and these models were analyzed using machine learning methods. The Long Short-Term Memory (LSTM) and Support Vector Machine (SVM) were selected as the modeling approaches. Additionally, the cross-correlation method was used to construct four different drought models. SPEI12 data for the period 1901–2023, which are assumed to best represent hydrological droughts, were employed in the analysis. The Wavelet Transform (WT) was also used to improve model performance. To evaluate the model results, several performance metrics were utilized, including the Correlation Coefficient (r), the Root Mean Square Error (RMSE), the Nash–Sutcliffe Efficiency (NSE), the Kling–Gupta efficiency (KGE), and the Performance Index (PI). The findings revealed that the WT-based SVMM02 (SVMWM02) model produced the best results. Therefore, this model should be used in future drought modeling studies for Edirne.

🔖

Bu Makaleye Atıf Yap

Tuğrul, T. (2026). Makine öğrenmesi ile ileriye yönelik kuraklıkların tahmin edilmesi. Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi, Cilt: 15, 1–11. https://doi.org/10.28948/ngumuh.1819926

📄 Tam Metin PDF Okuyucu

📄

PDF görüntüleyici hazırlanıyor...

Kaynakça (50)

  1. B. Ghazi, S. Dutt, ve A. Torabi Haghighi, Projection of future meteorological droughts in lake Urmia Basin Iran. Water, 15(8),1558, 2023. https://doi.org/ 10.3390/w15081558.
  2. Y. Gao, X. Li, L. R. Leung, D. Chen, ve J. Xu, Aridity changes in the Tibetan Plateau in a warming climate. Environmental Research Letters, 10(3), 034013, 2015. https://doi.org/10.1088/1748-9326/10/3/034013
  3. J. Spinoni et al., Future Global Meteorological Drought Hot Spots: A Study Based on CORDEX Data, (in English). Journal of Climate, 33(9),3635-3661, 01 May. 2020 2020. https://doi.org/10.1175/JCLI-D-19-0084.1
  4. S. M. Vicente-Serrano, S. M. Quiring, M. Peña-Gallardo, S. Yuan, ve F. Domínguez-Castro, A review of environmental droughts: Increased risk under global warming?. Earth-Science Reviews, 201, 102953, 2020. https://doi.org/10.1016/j.earscirev.2019.102953
  5. T. Wang, X. Tu, V. P. Singh, X. Chen, K. Lin, ve Z. Zhou, Drought prediction: Insights from the fusion of LSTM and multi-source factors. Science of the Total Environment, 902, 166361, 2023. https://doi.org/10.1016/j.scitotenv.2023.166361
  6. CRED, Disasters in Numbers 2021. Brussels, Belgium: Centre for Research on the Epidemiology of Disasters (CRED), 2021.
  7. D. F. Balting, A. AghaKouchak, G. Lohmann, ve M. Ionita, Northern Hemisphere drought risk in a warming climate. NPJ Climate and Atmospheric Science, 4(1), 61, 2021. https://doi.org/10.1038/s41612-021-00218-2
  8. J. Yin et al., Future socio-ecosystem productivity threatened by compound drought–heatwave events. Nature Sustainability, 6(3),.259-272, 2023.
  9. K. Forrest, B. Tarroja, F. Chiang, A. AghaKouchak, ve S. Samuelsen, Assessing climate change impacts on California hydropower generation and ancillary services provision. Climatic Change, 151(3), 395-412, 2018/12/01 2018. https://doi.org/10.1007/s10584-018-2329-5
  10. J. Laimighofer ve G. Laaha, How standard are standardized drought indices? Uncertainty components for the SPI & SPEI case. Journal of Hydrology, 613,. 128385,2022.https://doi.org/10.1016/j.jhydrol.2022.128385
  11. F. Chiang, O. Mazdiyasni, ve A. AghaKouchak, Evidence of anthropogenic impacts on global drought frequency, duration, and intensity. Nature communications, 12(1), 2754, 2021. https://doi.org/10.1038/s41467-021-22314-w
  12. Y. H. Essa, M. Hirschi, W. Thiery, A. M. El-Kenawy, ve C. Yang, Drought characteristics in Mediterranean under future climate change. npj Climate and Atmospheric Science, 6(1), 133, 2023/09/07 2023. https://doi.org/10.1038/s41612-023-00458-4
  13. T. B. McKee, N. J. Doesken, ve J. Kleist, The relationship of drought frequency and duration to time scales. in Proceedings of the 8th Conference on Applied Climatology,, 17(22), Boston, 179-183, 1993
  14. M. Naresh Kumar, C. Murthy, M. Sesha Sai, ve P. Roy, On the use of Standardized Precipitation Index (SPI) for drought intensity assessment, Meteorological Applications: A journal of forecasting, practical applications, training techniques and modelling, 16(3), 381-389, 2009. https://doi.org/10.48550/ arXiv.1509.04808
  15. T. Caloiero, S. Veltri, P. Caloiero, ve F. Frustaci, Drought analysis in Europe and in the Mediterranean basin using the standardized precipitation index. Water, 10(8), 1043, 2018. https://doi.org/10.3390/w10081043
  16. W. M. Kim ve C. C. Raible, Dynamics of the Mediterranean droughts from 850 to 2099 CE in the Community Earth System Model. Climate of the Past, 17(2), 887-911, 2021. https://doi.org/10.5194/cp-17-887-2021
  17. T. Tuğrul, S. Oruc, ve M. A. Hınıs, Transforming wind data into insights: A comparative study of stochastic and machine learning models in wind speed forecasting. Applied Sciences, 15(7), 3543, 2025. https://doi.org/10.3390/app15073543
  18. S. Oruc, T. Tugrul, ve M. A. Hinis, Beyond traditional metrics: exploring the potential of hybrid algorithms for drought characterization and prediction in the tromso region, Norway, Applied Sciences, 14(17),. 7813, 2024. https://doi.org/10.3390/app14177813
  19. W. Almikaeel, L. Čubanová, ve A. Šoltész, Hydrological drought forecasting using machine learning—Gidra river case study, Water, 14(3), 387, 2022. https://doi.org/10.3390/w14030387
  20. K. En-Nagre et al., Assessment and prediction of meteorological drought using machine learning algorithms and climate data. Climate Risk Management, 45, 100630, 2024. https://doi.org/10.1016/j.crm.2024.100630
  21. C. Lalika, A. U. H. Mujahid, M. James, ve M. C. Lalika, Machine learning algorithms for the prediction of drought conditions in the Wami River sub-catchment, Tanzania. Journal of Hydrology: Regional Studies, 53, 101794, 2024.
  22. T. Tuğrul ve M. A. Hinis, Improvement of drought forecasting by means of various machine learning algorithms and wavelet transformation. Acta Geophysica, 2024/07/01 2024. https://doi.org/10.1007/s11600-024-01399-z
  23. A. Devanand et al., Australia’s Tinderbox Drought: An extreme natural event likely worsened by human-caused climate change. Science Advances, 10(10),. eadj3460, 2024. https://doi.org/10.1126/ sciadv.adj3460
  24. C. B. Pande et al., Forecasting of SPI and meteorological drought based on the artificial neural network and M5P model tree. Land, 11(11), 2040, 2022. https://doi.org/10.3390/land11112040
  25. Y. Zhao et al., Drought monitoring and performance evaluation based on machine learning fusion of multi-source remote sensing drought factors. Remote Sensing, 14(24), 6398, 2022. https://doi.org/10.3390/rs14246398
  26. M. Achite, O. M. Katipoğlu, M. Jehanzaib, N. Elshaboury, V. Kartal, ve S. Ali, Hydrological drought prediction based on hybrid extreme learning machine: Wadi Mina Basin Case Study, Algeria. Atmosphere, 14(9), 1447, 2023. https://doi.org/10.3390/ atmos14091447
  27. S. M. Vicente-Serrano, S. Beguería, J. I. López-Moreno, M. Angulo, ve A. El Kenawy, A new global 0.5 gridded dataset (1901–2006) of a multiscalar drought index: comparison with current drought index datasets based on the Palmer Drought Severity Index, Journal of Hydrometeorology, 11(4), 1033-1043, 2010.
  28. A. Ahmadalipour, H. Moradkhani, ve M. C. Demirel, "A comparative assessment of projected meteorological and hydrological droughts: elucidating the role of temperature. Journal of Hydrology, 553, 785-797, 2017. https://doi.org/10.1016/ j.jhydrol.2017.08.047
  29. Z. Hao ve V. P. Singh, Drought characterization from a multivariate perspective: A review. Journal of Hydrology, 527, 668-678, 2015. https://doi.org/10.1016/j.jhydrol.2015.05.031
  30. D. Haro-Monteagudo, A. Daccache, ve J. Knox, Exploring the utility of drought indicators to assess climate risks to agricultural productivity in a humid climate. Hydrology Research, 49(2), 539-551, 2018. https://doi.org/10.2166/nh.2017.010
  31. S. S. Wahla, J. H. Kazmi, A. Sharifi, S. A. Shirazi, A. Tariq, ve H. Joyell Smith, Assessing spatio-temporal mapping and monitoring of climatic variability using SPEI and RF machine learning models. Geocarto International, 37(27), 14963-14982, 2022. https://doi.org/10.1080/10106049.2022.2093411
  32. R. G. Allen, L. S. Pereira, D. Raes, ve M. Smith, Crop evapotranspiration-Guidelines for computing crop water requirements-FAO Irrigation and drainage paper 56. Fao, Rome, 300(9), D05109, 1998.
  33. B. E. Boser, I. M. Guyon, ve V. N. Vapnik, A training algorithm for optimal margin classifiers. in Proceedings of the fifth annual workshop on Computational learning theory, 144-152, 1992.https://doi.org/10.1145/130385.13040.
  34. E. Shabani, M. A. Ghorbani, ve S. Inyurt, The power of the GP-ARX model in CO2 emission forecasting. in Risk, Reliability and Sustainable Remediation in the Field of Civil and Environmental Engineering: Elsevier, 79-91, 2022
  35. M. Achite, O. M. Katipoglu, S. Şenocak, N. Elshaboury, O. Bazrafshan, ve H. Y. Dalkılıç, Modeling of meteorological, agricultural, and hydrological droughts in semi-arid environments with various machine learning and discrete wavelet transform. Theoretical and Applied Climatology, 154(1-2), 413-451, 2023. https://doi.org/10.1007/s00704-023-04564-4
  36. S. Saha, A. Saha, T. K. Hembram, B. Kundu, ve R. Sarkar, Novel ensemble of deep learning neural network and support vector machine for landslide susceptibility mapping in Tehri region, Garhwal Himalaya. Geocarto International, 37(27), 17018-17043, 2022. https://doi.org/10.1080/ 10106049.2022.2120638
  37. M. Klaus-Robert, M. Sebastian, R. Gunnar, T. Koji, ve S. Bernhard, An introduction to kernel-based learning algorithms. IEEE TNN, 2001.
  38. A. Belayneh, J. Adamowski, B. Khalil, ve B. Ozga-Zielinski, Long-term SPI drought forecasting in the Awash River Basin in Ethiopia using wavelet neural network and wavelet support vector regression models. Journal of Hydrology, 508, 418-429, 2014. https://doi.org/10.1016/j.jhydrol.2013.10.052
  39. S. R. Gunn, Support vector machines for classification and regression. Technical Report, Image Speech and Intelligent Systems Research Group, 1997, 1(1), 1-52, 1997.
  40. V. N. Vapnik, An overview of statistical learning theory. IEEE transactions on neural networks, 10(5), 988-999, 1999.
  41. M. Panahi, N. Sadhasivam, H. R. Pourghasemi, F. Rezaie, ve S. Lee, Spatial prediction of groundwater potential mapping based on convolutional neural network (CNN) and support vector regression (SVR). Journal of Hydrology, 588, 125033, 2020. https://doi.org/10.1016/j.jhydrol.2020.125033
  42. A. Belayneh, J. Adamowski, ve B. Khalil, Short-term SPI drought forecasting in the Awash River Basin in Ethiopia using wavelet transforms and machine learning methods. Sustainable Water Resources Management, 2(1), 87-101, 2016. https://doi.org/10.1007/s40899-015-0040-5
  43. S. Hochreiter ve J. Schmidhuber, Long short-term memory. Neural computation, 9(8), 1735-1780, 1997.
  44. T. Mikolov, A. Joulin, S. Chopra, M. Mathieu, ve M. A. Ranzato, Learning longer memory in recurrent neural networks. arXiv preprint arXiv:1412.7753, 2014. https://doi.org/10.48550/arXiv.1412.7753
  45. O. Kisi, Wavelet regression model as an alternative to neural networks for river stage forecasting. Water resources management, 25(2), 579-600, 2011. https://doi.org/10.1007/s11269-010-9715-8
  46. B. Yusuf, A. Shabri, ve R. Rezaiy, An innovative hybrid drought modelling approach using wavelet-long-short term memory (W-LSTM) with standardised precipitation index (SPI) data for Badeggi district, Niger State-Nigeria. Discover Applied Sciences, 7(10), 1129, 2025/10/03 2025. https://doi.org/10.1007/s42452-025-07649-z
  47. H. Citakoglu ve Ö. Coşkun, Comparison of hybrid machine learning methods for the prediction of short-term meteorological droughts of Sakarya Meteorological Station in Turkey. Environmental Science and Pollution Research, 29(50), 75487-75511, 2022. https://doi.org/10.1007/s11356-022-21083-3
  48. M. Erişmiş, Standartlaştırılmış yağış buharlaşma indeksine göre Meriç Nehri Havzasında uzun dönem kuraklık analizi. International Journal of Geography and Geography Education, 50, 313-328, 2023. https://doi.org/10.32003/igge.1297107.
  49. T. B. McKee, N. J. Doesken, ve J. Kleist, The relationship of drought frequency and duration to time scales. in Proceedings of the 8th Conference on Applied Climatology,, 17(22), Boston, 179-183, 1993
  50. A. Devanand et al., Australia’s Tinderbox Drought: An extreme natural event likely worsened by human-caused climate change. Science Advances, 10(10),. eadj3460, 2024. https://doi.org/10.1126/ sciadv.adj3460