https://doi.org/10.65770/ZIWD5587
ABSTRACT
Produced water from oil and gas operations may contain heavy metals, total petroleum hydrocarbons (TPH), and production chemicals that create substantial environmental-management challenges in the Niger Delta. This study develops a hybrid convolutional neural network–long short-term memory–attention (CNN–LSTM–Attention) model for multi-horizon prediction of lead (Pb), cadmium (Cd), chromium (Cr), nickel (Ni), copper (Cu), mercury (Hg), and TPH. The analysis uses reported Nigerian Upstream Petroleum Regulatory Commission (NUPRC) monitoring records from three flow stations for 2018–2024, together with operational and environmental covariates. Models were evaluated with chronological train–validation–test partitions, walk-forward validation, rolling-origin cross-validation, and comparisons with standalone LSTM, gated recurrent unit (GRU), and autoregressive integrated moving average (ARIMA) models. At the seven-day horizon, the hybrid model achieved RMSE = 0.0847, MAE = 0.0623, and R2 = 0.9756, outperforming the baselines. Antecedent TPH, production flow rate, water cut, rainfall, and seasonality were the most influential predictors. Paired error tests and the Diebold Mariano test indicated that the reported forecast improvements were statistically significant. The results support the model’s use as a decision-support component for risk-based sampling and early warning; they do not, by themselves, establish regulatory compliance or causal effects. External validation at additional facilities and uncertainty-calibrated forecasts are required before operational deployment.
References
- [1] Abdelhamid, C., Latrach, A., Rabiei, M., & Venugopal, K. (2025). Produced water treatment technologies: A review. Energies, 18(1), 63. https://doi.org/10.3390/en18010063
- [2] Agbozu, I. E., Ekweozor, I. K. E., & Opuene, K. (2007). Survey of heavy metals in the catfish Synodontis clarias. International Journal of Environmental Science and Technology, 4, 93–97. https://doi.org/10.1007/BF03325966
- [3] Al-Ghouti, M. A., Al-Kaabi, M. A., Ashfaq, M. Y., & Da’na, D. A. (2019). Produced water characteristics, treatment and reuse: A review. Journal of Water Process Engineering, 28, 222–239. https://doi.org/10.1016/j.jwpe.2019.02.001
- [4] Amakiri, K. T., Canon, A. R., Molinari, M., & Angelis-Dimakis, A. (2022). Review of oilfield produced water treatment technologies. Chemosphere, 298, 134064. https://doi.org/10.1016/j.chemosphere.2022.134064
- [5] Amakiri, K. T., Ogolo, N. A., Angelis-Dimakis, A., & Albert, O. (2023). Physicochemical assessment and treatment of produced water: A case study in Niger Delta Nigeria. Petroleum Research, 8(1), 87–95. https://doi.org/10.1016/j.ptlrs.2022.05.003
- [6] Bai, S., Kolter, J. Z., & Koltun, V. (2018). An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv:1803.01271. https://doi.org/10.48550/arXiv.1803.01271
- [7] Barzegar, R., Aalami, M. T., & Adamowski, J. (2020). Short-term water quality variable prediction using a hybrid CNN–LSTM deep learning model. Stochastic Environmental Research and Risk Assessment, 34, 415–433. https://doi.org/10.1007/s00477-020-01776-2
- [8] Box, G. E. P., Jenkins, G. M., Reinsel, G. C., & Ljung, G. M. (2015). Time Series Analysis: Forecasting and Control (5th ed.). Wiley.
- [9] César, S. D., De Jager, D., & Njoya, M. (2024). Environmental trade-offs in energy production:A review of the produced water life cycle and environmental footprint. Marine Pollution Bulletin, 203, 116480. https://doi.org/10.1016/j.marpolbul.2024.116480
- Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9, 1735– https://doi.org/10.1162/neco.1997.9.8.1735
- Howard, C. C., Etuk, E. H., & Howard, I. C. (2021). Application of a hybrid multiple linear regression (MLR) and artificial neural network (ANN) models to produced water quality prediction. African Scholar Journal of African Sustainable Development, 23(2), 1–15.
- Howard, C. C., & Howard, I. C. (2026a). Hybrid neural network approaches to produced water quality forecasting and management in Nigerian oil fields. International Journal of Mathematics, Statistics, and Computer Science, 4, 527–543. https://doi.org/10.59543/7fjxt923
- Howard, C. C., & Howard, I. C. (2026b). Temporal dynamics and predictive modeling of produced water quality in Niger Delta oil fields. Journal of Statistical Sciences and Computational Intelligence, 2(2), 601–619. https://doi.org/10.64497/jssci.140
- Karahan, S. M., Vandenbruwaene, W., Castano, J. A., & Verwaeren, J. (2025). Explainable AI for aquatic environmental intelligence: A SHAP-enhanced LSTM approach using high-frequency water quality data in a river system. Journal of Hydroinformatics, 27(12), 1918–1928. https://doi.org/10.2166/hydro.2025.118
- Kundu, S., Datta, P., Pal, P., Ghosh, K., Das, A., & Das, B. K. (2025). Unveiling the hiddenconnections: Using explainable artificial intelligence to assess water quality criteria in nine giant rivers. Journal of Cleaner Production, 492, 144861. https://doi.org/10.1016/j. 2025.144861
- Neff, J., Lee, K., & DeBlois, E. M. (2011). Produced water: Overview of composition, fates, and effects. In K. Lee and J. Neff (Eds.), Produced Water, pp. 3–54. Springer. https://doi.org/10.1007/978-1-4614-0046-2_1
- Ordinioha, B., & Brisibe, S. (2013). The human health implications of crude oil spills in the Niger Delta, Nigeria: An interpretation of published studies. Nigerian Medical Journal, 54, 10–16. https://doi.org/10.4103/0300-1652.108887
Download all article in PDF
