SavedAdd your journal

Journals / International Journal of Engineering and Geosciences / 2026 / Volume: 11 Issue: 3

Spatiotemporal Analysis and Forecasting of PM2.5 Air Pollution in Azerbaijan Using SARIMA and Prophet Models

SARIMA ve Prophet Modelleri Kullanılarak Azerbaycan’da PM2.5 Hava Kirliliğinin Mekânsal–Zamansal Analizi ve Tahmini

Iraj Teymuri(university of Tabriz)0000-0002-3167-5583
Mirnuh Ismayilov(Institute of Geography, Academician H.A.Aliyev)0000-0003-4189-0401
Mohammad Nemati(University of Tabriz)0000-0003-2054-8535
Pages
—
Publication Date
2026-06-28
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

Overall, the proposed framework—combining freely available reanalysis products, simple spatial interpolation, and interpretable forecasting models—provides a practical and low-cost approach for routine air-quality monitoring, hotspot identification, and policy support in data-scarce regions.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Fine particulate matter (PM2.5) remains a major environmental and public health concern worldwide.
📊
Key Findings & Data:This study investigates the temporal and spatial dynamics of PM2.5 concentrations across Azerbaijan between 2019 and 2024 and develops forecasts using two transparent time-series models.
💡
Conclusion & Impact:Overall, the proposed framework—combining freely available reanalysis products, simple spatial interpolation, and interpretable forecasting models—provides a practical and low-cost approach for routine air-quality monitoring, hotspot identification, and policy support in data-scarce regions.
🔬Design:Modeling & Simulation
📍Scope:2019 Verileri
⚙️Variables:Air pollution · PM2.5 · Time-series analysis

Abstract

Fine particulate matter (PM2.5) remains a major environmental and public health concern worldwide. This study investigates the temporal and spatial dynamics of PM2.5 concentrations across Azerbaijan between 2019 and 2024 and develops forecasts using two transparent time-series models. Monthly PM2.5 fields were derived exclusively from ECMWF CAMS reanalysis products, while Sentinel-5P TROPOMI observations were used only for qualitative spatial interpretation and contextual comparison. All datasets were processed in Google Earth Engine and analyzed in R. Temporal dynamics were modelled using SARIMA and Prophet, while spatial patterns were characterized through stratified random-point sampling followed by inverse distance weighting (IDW) interpolation. The results reveal a pronounced seasonal cycle in which PM2.5 concentrations peak during the warm season rather than in winter. The highest concentrations consistently occurred in June and September, with the absolute maximum reaching 28.9 µg/m³ in June 2019, whereas the lowest values were systematically recorded in January and February. At the interannual scale, 2019 emerged as the most polluted year, followed by a substantial decline during 2020–2023, partly associated with COVID-19–related mobility restrictions, before concentrations increased again during the second half of 2024. Spatial analysis identified persistent hotspots in the central and south-central lowland districts, while the eastern coastal zone—particularly the Absheron Peninsula and greater Baku—experienced intense but episodic pollution events. Getis-Ord hotspot analysis and a highly significant positive Moran’s I statistic (0.922, p < 0.001) confirmed strong and non-random spatial clustering of PM2.5 concentrations. Across most locations and months, Prophet produced lower forecasting errors than SARIMA during the 2024 validation period (MAE = 0.180 µg/m³; RMSE = 0.202 µg/m³). Prophet also reproduced the location and intensity of seasonal pollution maxima more successfully, whereas SARIMA showed systematic spatial displacement of hotspots during several critical months. However, the trend shift observed during late 2024 and the absence of ground-based calibration data indicate that forecasting PM2.5 under non-stationary environmental conditions remains challenging. Overall, the proposed framework—combining freely available reanalysis products, simple spatial interpolation, and interpretable forecasting models—provides a practical and low-cost approach for routine air-quality monitoring, hotspot identification, and policy support in data-scarce regions.

Özet

İnce partikül maddeler (PM2.5), dünya genelinde çevresel ve halk sağlığı açısından başlıca sorunlardan biri olmaya devam etmektedir. Bu çalışma, Azerbaycan genelinde PM2.5 konsantrasyonlarının 2019–2024 dönemindeki zamansal ve mekânsal evrimini incelemekte ve iki şeffaf zaman serisi modeli kullanarak tahminler üretmektedir. Aylık PM2.5 alanları, Sentinel-5P TROPOMI ve ECMWF CAMS yeniden analiz ürünlerinden elde edilmiş, Google Earth Engine ortamında işlenmiş ve R yazılımı kullanılarak analiz edilmiştir. Zamansal dinamikler SARIMA ve Prophet modelleriyle, mekânsal örüntüler ise rastgele nokta örneklemesi ve ardından ters mesafe ağırlıklı enterpolasyon (IDW) yoluyla belirlenmiştir. Bulgular, PM2.5 konsantrasyonlarının kış ayları yerine sıcak mevsimde zirve yaptığı, alışılmadık derecede güçlü bir mevsimsel döngüye işaret etmektedir. En yüksek değerler çoğu yıl Haziran ve Eylül aylarında gözlenmiş; mutlak maksimum Haziran 2019’da 28.9 µg/m³ olarak kaydedilmiştir. Buna karşılık, en düşük konsantrasyonlar sistematik olarak Ocak ve Şubat aylarında görülmüştür. Yıllar arası ölçekte, 2019 en kirli yıl olarak öne çıkarken, 2020–2023 döneminde—kısmen COVID-19’a bağlı hareketlilik kısıtlamalarıyla ilişkili olarak—belirgin bir iyileşme gözlenmiş, ancak 2024’ün ikinci yarısında yeniden artış eğilimi ortaya çıkmıştır. Mekânsal olarak, merkez ve güney-merkez ovalık ilçelerde uzun dönemli ve kalıcı yüksek maruziyet dikkat çekerken, doğu kıyı kuşağı—özellikle Abşeron Yarımadası ve Büyük Bakü—şiddetli ancak dönemsel kirlilik olaylarıyla karakterize edilmektedir. Getis-Ord sıcak nokta analizi ve istatistiksel olarak anlamlı pozitif Moran’s I değeri (p < 0.01), PM2.5 dağılımının güçlü ve rastlantısal olmayan bir mekânsal kümelenme sergilediğini doğrulamaktadır. Tahmin aşamasında, Prophet modeli hem mevsimsel dinamikleri hem de coğrafi örüntüleri yeniden üretmede SARIMA’ya kıyasla tutarlı biçimde daha üstün performans göstermiş ve 2024 doğrulamasında MAE, RMSE, sMAPE ve MASE ölçütlerinde daha düşük hata değerleri elde etmiştir. Prophet özellikle, SARIMA’nın sıcak noktaları mekânsal olarak yanlış konumlandırdığı kritik aylarda, aylık maksimumların yerini ve şiddetini daha doğru biçimde yakalamıştır. Uygulamalı açıdan, bulgular hava kalitesi yönetim politikalarının sıcak mevsime ve sağlık risklerinin en yüksek olduğu merkez ovalar ile doğu kıyı koridoruna öncelik vermesi gerektiğini göstermektedir. Daha genel olarak, serbestçe erişilebilen uydu verilerini, basit mekânsal enterpolasyonu ve yorumlanabilir bir tahmin modelini birleştiren bu çerçeve; veri açısından kısıtlı bölgelerde rutin hava kalitesi izleme, erken uyarı ve politika destek süreçleri için düşük maliyetli ve uygulanabilir bir araç sunmaktadır.

🔖

Cite This Article

Teymuri, I., Ismayilov, M., Nemati, M. (2026). Spatiotemporal Analysis and Forecasting of PM2.5 Air Pollution in Azerbaijan Using SARIMA and Prophet Models. International Journal of Engineering and Geosciences, Volume: 11 Issue: 3. https://doi.org/10.26833/ijeg.1841923

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (58)

  1. Ghiaseddin, M. (2006). Air pollution: Sources, effects, and control. Tehran: University of Tehran Press.
  2. Pope, C. A., 3rd, & Dockery, D. W. (2006). Health effects of fine particulate air pollution: Lines that connect. Journal of the Air & Waste Management Association, 56(6), 709–742.
  3. World Health Organization. (2022). Air pollution: Health impacts. WHO Fact Sheet.
  4. World Health Organization. (2025). World health statistics 2025: Monitoring health for the sustainable development goals. Geneva: WHO.
  5. Burnett, R., Chen, H., Szyszkowicz, M., Fann, N., Hubbell, B., Pope, C. A., 3rd,. & Spadaro, J. V. (2018). Global estimates of mortality associated with long-term exposure to outdoor fine particulate matter. Proceedings of the National Academy of Sciences of the United States of America, 115(38), 9592–9597.
  6. Schoch, M., De Lauriere, C. F., & Bernauer, T. (2026). Large gaps in monitoring urban air pollution in low- and middle-income countries associated with economic conditions and political institutions. PLoS ONE, 21(1), e0321358.
  7. Chief Medical Officer of the United Kingdom. (2017). Annual report of the Chief Medical Officer 2017: Health impacts of all pollution – what do we know? Department of Health and Social Care.
  8. Asian Transport Observatory. (2026). Azerbaijan transport air pollution profile 2026. Asian Development Bank, Transport, Energy, and Climate Change Unit.
  9. World Bank. (2011). Azerbaijan: Country environmental analysis (Report No. 60180). Washington, DC: World Bank Group.
  10. IQAir. (2025). Air quality in Azerbaijan.
  11. AQICN. (2025). Air pollution in Azerbaijan.
  12. McDuffie, E. E., Martin, R. V., Spadaro, J. V., Burnett, R., Smith, S. J., O’Rourke, P.,. & Brauer, M. (2021). Source sector and fuel contributions to ambient PM2.5 and attributable mortality across multiple spatial scales. Nature Communications, 12(1), 3594.
  13. Ismayilov, M., Zeynalova, S., & Guliyeva, S. (2023). Transformation and dynamics of structural-genetic features of the landscapes of the Caspian coast plains. Comptes Rendus de l’Académie Bulgare des Sciences, 76(6), 925–935.
  14. Berrocal, V. J., Guan, Y., Muyskens, A., Wang, H., Reich, Mulholland, J. A., & Chang, H. H. (2020). A comparison of statistical and machine learning methods for creating national daily maps of ambient PM2.5 concentration. Atmospheric Environment, 222, 117130.
  15. Bhatti, U. A., Yan, Y., Zhou, M., Ali, S., Hussain, A., Qingsong, H.,. & Yuan, L. (2021). Time series analysis and forecasting of air pollution particulate matter (PM2.5): An SARIMA and factor analysis approach. IEEE Access, 9, 41019–41031.
  16. Koushesh Vatan, M. A., Nemati, M., Sekertekin, A., & Zehtabi Aslkhiabani, M. (2025). Time series trend detection and forecasting of SUHI in Tabriz, Iran. Urban Climate, 59, 102321.
  17. Zhu, X., Li, B., Cao, Y., & Zhang, Q. (2026). Applicability analysis of tree-based ensemble learning for air pollutant prediction models. Scientific Reports, 16, 9602.
  18. Christian, M. M., & Choi, H. (2024). Air quality forecasting using machine learning: A global perspective with relevance to low-resource settings. arXiv preprint arXiv:2401.04369.
  19. Mogaraju, J. K. (2024). Machine learning assisted prediction of land surface temperature based on major air pollutants. International Journal of Engineering and Geosciences, 9(2), 233–246. https://doi.org/10.26833/ijeg.1394111
  20. Karnati, H., Soma, A., Alam, A., & Kalaavathi, B. (2025). Comprehensive analysis of various imputation and forecasting models for predicting PM2.5 pollutant in Delhi. Neural Computing and Applications, 37(17), 11441–11458.
  21. Gocheva-Ilieva, S. G., Ivanov, A. V., Voynikova, D. S., & Boyadzhiev, D. T. (2014). Time series analysis and forecasting for air pollution in small urban area. Stochastic Environmental Research and Risk Assessment, 28(4), 1045–1060.
  22. Taylor, S. J., & Letham, B. (2018). Forecasting at scale. The American Statistician, 72(1), 37–45.
  23. Shen, J., Valagolam, D., & McCalla, S. (2020). Prophet forecasting model: A machine learning approach to predict air pollutants. PeerJ, 8, e9961.
  24. Hasnain, A., Sheng, Y., Hashmi, M. Z., Bhatti, U. A., Hussain, A., Hameed, M.,. & Zha, Y. (2022). Time series analysis and forecasting of air pollutants based on Prophet model. Frontiers in Environmental Science, 10, 945628.
  25. Chen, M., Liu, B., Liang, M., & Yao, N. (2025). Decoding PM2.5 prediction using SARIMA, Prophet, and LightGBM. Algorithms, 18(3), 167.
  26. Singh, A. (2019). Air pollution forecasting and performance evaluation using advanced time series methods (Doctoral dissertation). National College of Ireland.
  27. Bekkar, A., Hssina, B., Douzi, S., & Douzi, K. (2021). Air pollution prediction in smart city: Deep learning approach. Journal of Big Data, 8(1), 161.
  28. Arnaut, F., Cvetkov, V., Đurić, D., & Samardžić-Petrović, M. (2023). Short-term forecasting of PM10 and PM2.5 with Prophet model. Geofizika, 40(2), 162–177.
  29. Hashemzadeh, M., Borhani, F., & Shahverdi, P. (2026). Comparative forecasting of total ozone column over Iran. Iran Journal of Computer Science, 9(1), 6.
  30. Mani, G., Viswanadhapalli, J. K., & Stonier, A. A. (2022). Prediction and forecasting of air quality index using ARIMA models. Journal of Engineering Research, 10(2), 179–194.
  31. Mohamed, H., Hassan, A., & Elhag, A. (2025). Seasonal dynamics of atmospheric pollutants using Sentinel-5P data. International Journal of Engineering and Geosciences, 10(2), 262–271. https://doi.org/10.26833/ijeg.1587122
  32. Li, Z., Bi, J., Liu, Y., & Hu, X. (2025). Forecasting O3 and NO2 concentrations using machine learning. Environment International, 195, 109249.
  33. Mishra, M. K., Mishra, D. G., & Pulhani, V. (2025). Optimization of IDW interpolation in GIS. IARP National Conference on Radiation Protection for Sustainable Nuclear Energy.
  34. Inness, A., Ades, M., Agustí-Panareda, A., Barré, J., Benedictow, A., Blechschmidt, A. M.,. & Suttie, M. (2019). The CAMS reanalysis of atmospheric composition. Atmospheric Chemistry and Physics, 19(6), 3515–3556.
  35. Veefkind, J. P., Aben, I., McMullan, K., Förster, H., De Vries, J., Otter, G.,. & Levelt, P. F. (2012). TROPOMI on Sentinel-5P. Remote Sensing of Environment, 120, 70–83.
  36. Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., & Moore, R. (2017). Google Earth Engine: Planetary-scale geospatial analysis. Remote Sensing of Environment, 202, 18–27.
  37. Kaplan, G., & Yigit Avdan, Z. (2020). Space-borne air pollution observation from Sentinel-5P. International Journal of Engineering and Geosciences, 5(3), 130–137.
  38. Douros, J., Eskes, H., van Geffen, J., Boersma, K. F., Compernolle, S., Pinardi, G.,. & Veefkind, P. (2023). Sentinel-5P NO2 comparison with CAMS ensemble. Geoscientific Model Development, 16(2), 509–534.
  39. Cochran, W. G. (1977). Sampling techniques. New York: John Wiley & Sons.
  40. Li, Z., Liu, Y., Bi, J., & Hu, X. (2025). Near real-time NO2 forecasting using satellite data. Journal of Hazardous Materials, 494, 138447.
  41. Kotan, B., & Erener, A. (2023). Seasonal analysis of PM10 and SO2 during COVID-19 lockdown. International Journal of Engineering and Geosciences, 8(2), 173–187. https://doi.org/10.26833/ijeg.1111699
  42. Cujia, A., Agudelo-Castañeda, D., Pacheco-Bustos, C., & Teixeira, E. C. (2019). Forecast of PM10 time-series data. Atmospheric Pollution Research, 10(6), 2053–2062.
  43. Taşpınar, F. (2015). Time series models for air pollution. CLEAN – Soil, Air, Water, 43(7), 980–988. Waris, U., & Tariq, S. (2024). Aerosol optical depth forecasting. Atmospheric Pollution Research, 15(7), 102146.
  44. Mall, A., & Singh, S. (2025). Seasonal-ARIMA forecasting of UV radiation. Atmospheric Environment, 362, 121570.
  45. Xu, X., Liu, Y., & Zeng, B. (2025). Grey model for air quality prediction. Atmospheric Pollution Research, 16, 102832.
  46. Zeng, B., Zheng, T., Xu, X., & Wang, J. (2024). Grey prediction model for PM2.5. Engineering Applications of Artificial Intelligence, 136, 108974.
  47. Zhu, X., Dang, Y., & Ding, S. (2021). Grey forecasting models for air quality. Grey Systems: Theory and Application, 11(4), 596–618.
  48. Belt, E. A., Yao, L., Li, X., Wang, N., Shekhar, V., Koch, T., & Dugundji, E. R. (2025). Traffic flow forecasting under road maintenance. Transportation Research Procedia, 82, 1335–1352.
  49. Han, S., Kundhikanjana, W., Towashiraporn, P., & Stratoulias, D. (2022). Interpolation-based fusion for PM2.5 mapping. Atmosphere, 13(2), 161.
  50. Gupta, R. (2025). GIS-based analysis of urban heat islands. International Journal of Engineering and Geosciences, 10(3), 440–455. https://doi.org/10.26833/ijeg.1638818
  51. Nemati, M., Teimouri, I., & Roustaie, S. (2025). Electricity consumption forecasting in urban planning: An energy-oriented urban planning approach in Tabriz. Journal of Geography and Planning. Advance online publication. https://geoplanning.tabrizu.ac.ir/article_20502.html?lang=en
  52. Burnett, R., Chen, H., Szyszkowicz, M., Fann, N., Hubbell, B., Pope, C. A., 3rd,. & Spadaro, J. V. (2018). Global estimates of mortality associated with long-term exposure to outdoor fine particulate matter. Proceedings of the National Academy of Sciences of the United States of America, 115(38), 9592–9597.
  53. McDuffie, E. E., Martin, R. V., Spadaro, J. V., Burnett, R., Smith, S. J., O’Rourke, P.,. & Brauer, M. (2021). Source sector and fuel contributions to ambient PM2.5 and attributable mortality across multiple spatial scales. Nature Communications, 12(1), 3594.
  54. Bhatti, U. A., Yan, Y., Zhou, M., Ali, S., Hussain, A., Qingsong, H.,. & Yuan, L. (2021). Time series analysis and forecasting of air pollution particulate matter (PM2.5): An SARIMA and factor analysis approach. IEEE Access, 9, 41019–41031.
  55. Hasnain, A., Sheng, Y., Hashmi, M. Z., Bhatti, U. A., Hussain, A., Hameed, M.,. & Zha, Y. (2022). Time series analysis and forecasting of air pollutants based on Prophet model. Frontiers in Environmental Science, 10, 945628.
  56. Inness, A., Ades, M., Agustí-Panareda, A., Barré, J., Benedictow, A., Blechschmidt, A. M.,. & Suttie, M. (2019). The CAMS reanalysis of atmospheric composition. Atmospheric Chemistry and Physics, 19(6), 3515–3556.
  57. Veefkind, J. P., Aben, I., McMullan, K., Förster, H., De Vries, J., Otter, G.,. & Levelt, P. F. (2012). TROPOMI on Sentinel-5P. Remote Sensing of Environment, 120, 70–83.
  58. Douros, J., Eskes, H., van Geffen, J., Boersma, K. F., Compernolle, S., Pinardi, G.,. & Veefkind, P. (2023). Sentinel-5P NO2 comparison with CAMS ensemble. Geoscientific Model Development, 16(2), 509–534.