Implementation of a Machine Learning-Based Predictive Model for Assessing pH Variability in Coastal Marine Waters

Abstract

The increase in temperature, along with the rise in carbon dioxide (CO₂) concentration in the marine environment, poses a challenge for the marine environment. The decrease in pH (acidification) in oceans and seas has direct effects such as species loss and shifts in dominant species, as well as indirect effects such as the reorganisation of ecological functions and changes in community organisation patterns. This study aims to evaluate the predictive ability of the Gaussian Process Regression (GPR) model, a machine learning method. In this study, sea water parameters (01.06.2022-28.03.2025) from the Black Sea Region (Giresun coast) obtained from the Copernicus Marine Environment Monitoring Service (CMEMS) were used as the data set. Sea water pH was predicted using important parameters such as temperature and partial carbon dioxide surface pressure (spCO₂) in sea water. The findings of this study demonstrate that the Rational Quadratic Kernel GPR, trained with high spatial resolution SST and spCO₂ data provided by Copernicus, offers high accuracy and strong generalisation in pH estimation for the Giresun coast, demonstrating superior performance compared to other kernels and thus establishing a robust scientific foundation for the development of climate-sensitive decision support and early warning systems.

Keywords

• Machine learning algorithms
• Artificial intelligence-based modeling
• Black Sea
• Climate change impacts
• Acidification

Kıyı Deniz Sularında pH Değişkenliğinin Değerlendirilmesi için Makine Öğrenimi Tabanlı Tahmin Modelinin Uygulanması

Abstract

Deniz ortamındaki karbondioksit (CO2) konsantrasyonundaki artışla birlikte sıcaklıktaki artış, deniz ortamı için bir zorluk teşkil etmektedir. Okyanus ve denizlerde pH'ın düşmesinin (asitleşme) tür kaybı gibi doğrudan ve baskın türlerde kaymalar, ekolojik işlevlerin yeniden düzenlenmesi ve topluluk organizasyonu modellerinde değişiklikler gibi dolaylı etkileri vardır. Bu çalışma, bir makine öğrenmesi yöntemi olan Gauss Süreci Regresyonu (GPR) modelinin tahmin yeteneğini değerlendirmeyi amaçlamaktadır. Bu çalışmada, Copernicus Deniz Çevresi İzleme Servisi'nden (CMEMS) elde edilen Karadeniz Bölgesi (Giresun kıyıları) deniz suyu parametreleri (01.06.2022-28.03.2025) veri seti olarak kullanılmıştır. Sıcaklık ve deniz suyundaki kısmi karbondioksit yüzey basıncı (spCO2) gibi önemli parametreler kullanılarak, deniz suyu pH’sı tahmin edilmiştir. Bu çalışmanın bulguları, Copernicus’tan sağlanan yüksek mekânsal çözünürlüklü SST ve spCO₂ verileriyle eğitilen Rasyonel Kuadratik çekirdekli GPR’nin Giresun kıyıları için pH tahmininde yüksek doğruluk ve güçlü genelleme sunduğunu, diğer çekirdeklere kıyasla üstün performans gösterdiğini ve böylece iklim duyarlı karar destek ile erken uyarı sistemlerinin geliştirilmesine sağlam bir bilimsel temel oluşturduğunu göstermektedir.

Keywords

• Makine öğrenmesi algoritmaları
• Yapay zeka tabanlı modelleme
• Karadeniz
• İklim değişikliği etkileri
• Asidifikasyon

References

1. Akkan, T., Mutlu, T., & Baş, E. (2022). Forecasting sea surface temperature with feed-forward artificial networks in combating the global climate change: The sample of Rize, Türkiye. Ege Journal of Fisheries & Aquatic Sciences (EgeJFAS)/Su Ürünleri Dergisi, 39(4). http://doi.org/10.12714/egejfas.39.4.06
2. Allison, N., Cole, C., Hintz, C., Hintz, K., Rae, J., & Finch, A. (2021). Resolving the interactions of ocean acidification and temperature on coral calcification media pH. Coral Reefs, 40(6), 1807-1818. https://doi.org/10.1007/s00338-021-02170-2
3. Alver, D. O., Isik, H., Palabiyik, S., Akkan, B. E., & Akkan, T. (2025). pH acidification in the Red Sea: A machine learning-based validation study. Journal of Sea Research, 102613. https://doi.org/10.1016/j.seares.2025.102613
4. Asante, F., Bento, M., Broszeit, S., Bandeira, S., Chitará-Nhandimo, S., Amoné-Mabuto, M., & Correia, A. M. (2023). Marine macroinvertebrate ecosystem services under changing conditions of seagrasses and mangroves. Marine Environmental Research, 189, 106026. https://doi.org/10.1016/j.marenvres.2023.106026
5. Banza, M., & Rutto, H. (2023). Modelling of adsorption of nickel (II) by blend hydrogels (cellulose nanocrystals and corn starch) from aqueous solution using adaptive neuro‐fuzzy inference systems (ANFIS) and artificial neural networks (ANN). The Canadian Journal of Chemical Engineering, 101(4), 1906-1918. https://doi.org/10.1002/cjce.24603
6. Baş, E., & Eğrioğlu, E. (2025). A New Automatic Forecasting Method Based on Explainable Deep Dendritic Artificial Neural Network. https://doi.org/10.21203/rs.3.rs-7628747/v1
7. Bengil, F., Mavruk, S., Polat, S., & Akbulut, G. (2024). İskenderun Körfezi kıyı alanlarında sıcaklık ve klorofil-a için uydu ve model temelli veri setlerinin temsil yeteneği üzerine bir değerlendirme. Ege Journal of Fisheries & Aquatic Sciences (EgeJFAS)/Su Ürünleri Dergisi, 41(3). https://doi.org/10.12714/egejfas.41.3.07
8. CMEMS, 2025 E.U. Copernicus Marine Service Information (CMEMS). Marine Data Store (MDS), Retrieved on 25 March, 2025 from https://data.marine.copernicus.eu/products

9. Dong, C., Xu, G., Han, G., Bethel, B. J., Xie, W., & Zhou, S. (2022). Recent developments in artificial intelligence in oceanography. Ocean-Land-Atmosphere Research. https://doi.org/10.34133/2022/9870950
10. Egrioglu, E., & Bas, E. (2024). A new deep neural network for forecasting: Deep dendritic artificial neural network. Artificial Intelligence Review, 57(7), 171. https://doi.org/10.1007/s10462-024-10790-7
11. Friedlingstein, P., O'sullivan, M., Jones, M. W., Andrew, R. M., Hauck, J., Olsen, A., ... & Zaehle, S. (2020). Global carbon budget 2020. Earth System Science Data Discussions, 2020, 1-3. https://doi.org/10.5194/essd-12-3269-2020
12. Gruber, N. (2011). Warming up, turning sour, losing breath: ocean biogeochemistry under global change. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 369(1943), 1980-1996. https://doi.org/10.1098/rsta.2011.0003
13. Gruber, N., Clement, D., Carter, B. R., Feely, R. A., Van Heuven, S., Hoppema, M., et al. (2019). The oceanic sink for anthropogenic CO2 from 1994 to 2007. Science 363, 1193–1199. doi: 10.1126/science.aau5153
14. Huang, Y., Chen, S., Tang, X., Sun, C., Zhang, Z., & Huang, J. (2024). Dynamic patterns and potential drivers of river water quality in a coastal city: Insights from a machine-learning-based framework and water management. Journal of Environmental Management, 370, 122911. https://doi.org/10.1016/j.jenvman.2024.122911
15. IPCC (2014) Climate Change 2013: The Physical Science Basis
16. Irwan, D., Ali, M., Ahmed, A. N., Jacky, G., Nurhakim, A., Ping Han, M. C., ... & El-Shafie, A. (2023). Predicting water quality with artificial intelligence: a review of methods and applications. Archives of Computational Methods in Engineering, 30(8), 4633-4652. https://doi.org/10.1007/s11831-023-09947-4
17. Ishizu, M., Miyazawa, Y., Tsunoda, T., & Ono, T. (2019). Long-term trends in pH in Japanese coastal seawater. Biogeosciences, 16(24), 4747-4763. https://doi.org/10.5194/bg-16-4747-2019
18. Isık, H., & Akkan, T. (2025). Water quality assessment with artificial neural network models: Performance comparison between SMN, MLP and PS-ANN methodologies. Arabian Journal for Science and Engineering, 50(1), 369-387. https://doi.org/10.1007/s13369-024-09238-5
19. Işık, H., Bas, E., Egrioglu, E., & Akkan, T. (2024). A new single multiplicative neuron model artificial neural network based on black hole optimization algorithm: forecasting the amounts of clean water given to metropolis. Stochastic Environmental Research and Risk Assessment, 38(11), 4259-4274. https://doi.org/10.1007/s00477-024-02802-3
20. Jia, X., Ji, Q., Han, L., Liu, Y., Han, G., & Lin, X. (2022). Prediction of sea surface temperature in the East China Sea based on LSTM neural network. Remote Sensing, 14(14), 3300. https://doi.org/10.3390/rs14143300
21. Kisi, O., Karimi, S., Shiri, J., Makarynskyy, O., & Yoon, H. (2014). Forecasting sea water levels at Mukho Station, South Korea using soft computing techniques. The International Journal of Ocean and Climate Systems, 5(4), 175-188. https://doi.org/10.1260/1759-3131.5.4.175
22. Lancheros, E., Camps, A., Park, H., Rodriguez, P., Tonetti, S., Cote, J., & Pierotti, S. (2019). Selection of the key earth observation sensors and platforms focusing on applications for polar regions in the scope of copernicus system 2020–2030. Remote Sensing, 11(2), 175. https://doi.org/10.3390/rs11020175
23. Le Quéré, C., Andrew, R. M., Friedlingstein, P., Sitch, S., Julia Pongratz, J., Manning, A. C., et al. (2018). Global Carbon Budget 2017. Earth Syst. Sci. Data 10, 405–448. https://doi.org/10.5194/essd-10-405-2018
24. Lee, Y. W., Park, M. O., Kim, S. G., Kim, T. H., Oh, Y. H., Lee, S. H., & Joung, D. (2025). Long-term variations in pH in coastal waters along the Korean Peninsula. Biogeosciences, 22(3), 675-690. https://doi.org/10.5194/egusphere-2024-1836
25. Li, X., Zhou, S., Wang, F., & Fu, L. (2024). An improved sparrow search algorithm and CNN-BiLSTM neural network for predicting sea level height. Scientific reports, 14(1), 4560. https://doi.org/10.1038/s41598-024-55266-4
26. Macagga, R. A. T., & Hsu, P. C. (2025). Spatiotemporal Dynamics of Marine Heatwaves and Ocean Acidification Affecting Coral Environments in the Philippines. Remote Sensing, 17(6), 1048. https://doi.org/10.3390/rs17061048
27. Mahato, K. D., & Kumar, U. (2024). Optimized Machine learning techniques Enable prediction of organic dyes photophysical Properties: Absorption Wavelengths, emission Wavelengths, and quantum yields. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 308, 123768. https://doi.org/10.1016/j.saa.2023.123768
28. Maier, H. R., Jain, A., Dandy, G. C., & Sudheer, K. P. (2010). Methods used for the development of neural networks for the prediction of water resource variables in river systems: Current status and future directions. Environmental modelling & software, 25(8), 891-909. https://doi.org/10.1016/j.envsoft.2010.02.003
29. Matias, Í. O., Genovez, P. C., Torres, S. B., de Araújo Ponte, F. F., de Oliveira, A. J. S., de Miranda, F. P., & Avellino, G. M. (2021). Improved classification models to distinguish natural from anthropic oil slicks in the Gulf of Mexico: seasonality and Radarsat-2 beam mode effects under a machine learning approach. Remote Sensing, 13(22), 4568. https://doi.org/10.3390/rs13224568
30. Mihailov, M. E., Chirosca, A. V., & Chirosca, G. (2025). Fusion of In-Situ and Modelled Marine Data for Enhanced Coastal Dynamics Prediction Along the Western Black Sea Coast. Journal of Marine Science and Engineering, 13(2), 199. https://doi.org/10.3390/jmse13020199
31. Minnett, P. J., Alvera-Azcárate, A., Chin, T. M., Corlett, G. K., Gentemann, C. L., Karagali, I., ... & Vazquez-Cuervo, J. (2019). Half a century of satellite remote sensing of sea-surface temperature. Remote Sensing of Environment, 233, 111366. https://doi.org/10.1016/j.rse.2019.111366
32. Monaco, C. L., Metzl, N., Fin, J., Mignon, C., Cuet, P., Douville, E., ... & Tribollet, A. (2021). Distribution and long-term change of the sea surface carbonate system in the Mozambique Channel (1963–2019). Deep Sea Research Part II: Topical Studies in Oceanography, 186, 104936. https://doi.org/10.1016/j.dsr2.2021.104936
33. Palabıyık, S., & Akkan, T. (2024). Evaluation of water quality based on artificial intelligence: performance of multilayer perceptron neural networks and multiple linear regression versus water quality indexes. Environ Dev Sustain, https://doi.org/10.1007/s10668-024-05075-6
34. Pan, Y., Zeng, X., Xu, H., Sun, Y., Wang, D., & Wu, J. (2021). Evaluation of Gaussian process regression kernel functions for improving groundwater prediction. Journal of Hydrology, 603, 126960. https://doi.org/10.1016/j.jhydrol.2021.126960
35. Pourzangbar, A., Jalali, M., & Brocchini, M. (2023). Machine learning application in modelling marine and coastal phenomena: a critical review. Frontiers in Environmental Engineering, 2, 1235557. https://doi.org/10.3389/fenve.2023.1235557
36. Rasmussen, C. E. (2003). Gaussian processes in machine learning. In Summer school on machine learning (pp. 63-71). Berlin, Heidelberg: Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-540-28650-9_4
37. Rérolle, V. M., Floquet, C. F., Mowlem, M. C., Connelly, D. P., Achterberg, E. P., & Bellerby, R. R. (2012). Seawater-pH measurements for ocean-acidification observations. TrAC Trends in Analytical Chemistry, 40, 146-157. https://doi.org/10.1016/j.trac.2012.07.016
38. Santana-Casiano, J. M., & González-Dávila, M. (2010). pH decrease and effects on the chemistry of seawater. In Oceans and the atmospheric carbon content (pp. 95-114). Dordrecht: Springer Netherlands. https://doi.org/10.1007/978-90-481-9821-4_5
39. Uluocak, I. (2025). Comparative study of multivariate hybrid neural networks for global sea level prediction through 2050. Environmental Earth Sciences, 84(3), 79. https://doi.org/10.1007/s12665-025-12090-x
40. Vytla, V., Baduru, B., Kolukula, S. S., Ragav, N. N., & Kumar, J. P. (2025). Forecasting of sea surface temperature using machine learning and its applications. Journal of Earth System Science, 134(1), 25. https://doi.org/10.1007/s12040-024-02483-0
41. Wang, S., Gong, J., Gao, H., Liu, W., & Feng, Z. (2023). Gaussian process regression and cooperation search algorithm for forecasting nonstationary runoff time series. Water, 15(11), 2111. https://doi.org/10.3390/w15112111
42. Widdicombe, S., Dashfield, S. L., McNeill, C. L., Needham, H. R., Beesley, A., McEvoy, A., ... & Berge, J. A. (2009). Effects of CO2 induced seawater acidification on infaunal diversity and sediment nutrient fluxes. Marine ecology progress series, 379, 59-75. https://doi.org/10.3354/meps
43. Zhang, G. P., & Qi, M. (2005). Neural network forecasting for seasonal and trend time series. European journal of operational research, 160(2), 501-514. https://doi.org/10.1016/j.ejor.2003.08.037
44. Zhang, N., Xiong, J., Zhong, J., & Leatham, K. (2018). Gaussian process regression method for classification for high-dimensional data with limited samples. In 2018 Eighth International Conference on Information science and technology (ICIST) (pp. 358-363). IEEE. https://doi.org/10.1109/ICIST.2018.8426077