SavedAdd your journal

Journals / Journal of International Environmental Application and Science / 2026 / Volume: 21 Issue: 3

Global Forecasting of Multiple Short Water-Quality Time Series Across River, Estuary, and Lagoon Monitoring Stations

Chioma Howard(University of Africa, Toru-Orua, Bayelsa State, Nigeria.)0009-0005-4129-1291
Love Nnoka(Captain Elechi Amadi Polytechnic, Rumuola, Port Harcourt, Rivers State, Nigeria)
Pages
—
Publication Date
2026-10-02
DOI
—
🏛️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)

Monitoring agencies should retain persistence as a safeguard, deploy pooled forecasts only after subgroup-specific validation, and prioritize hydrological, tidal, meteorological, and pollution-source covariates for lagoon and extreme-event warning systems.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Abstract Environmental early-warning systems are most difficult to develop where contamination risks are substantial but monitoring histories are short.
📊
Key Findings & Data:The open dataset contains 26 variables measured monthly during 2021 at 10 Niger Delta River, estuary, and lagoon stations, yielding 260 twelve-observation series and 3,120 measurements.
💡
Conclusion & Impact:Monitoring agencies should retain persistence as a safeguard, deploy pooled forecasts only after subgroup-specific validation, and prioritize hydrological, tidal, meteorological, and pollution-source covariates for lagoon and extreme-event warning systems.
🔬Design:Lab & Field Experiment
📍Scope:2021 Verileri
⚙️Variables:Niger Delta · partial pooling · ridge regression

Abstract

Abstract Environmental early-warning systems are most difficult to develop where contamination risks are substantial but monitoring histories are short. This study evaluates whether global forecasting can overcome that constraint by pooling temporal information across water-quality variables, stations, and ecosystems. The open dataset contains 26 variables measured monthly during 2021 at 10 Niger Delta River, estuary, and lagoon stations, yielding 260 twelve-observation series and 3,120 measurements. A ridge-regularized global autoregression incorporating two lags, recent change, annual harmonics, and variable and ecosystem indicators was compared with persistence and separately estimated local AR(1) models. Four expanding forecast origins produced 1,040 leakage-resistant one-month-ahead predictions. Performance was evaluated using mean absolute scaled error (MASE), relative MAE, log-scale diagnostics, calibration, clustered bootstrap intervals, trimming, subgroup analysis, and leave-one-station-out sensitivity. Local AR(1) achieved the lowest aggregate MASE (1.7546), but its improvement over persistence (1.7720) was small and uncertain; global ridge was virtually identical to persistence (1.7721). The 95% bootstrap intervals included zero for both fitted-model differences. Nevertheless, global ridge improved 14 of 26 variables, reduced Pb, Hg, Zn, and Cu errors substantially, performed best in rivers, improved estuary forecasts, and lowered root mean squared log error by approximately 6%. These gains were offset by poorer lagoon forecasts and weak performance for several irregular variables, while calibration revealed compression of extreme observations. The study contributes a reproducible demonstration that partial pooling is an ecosystem- and variable-dependent resource rather than a universal remedy for short records. Monitoring agencies should retain persistence as a safeguard, deploy pooled forecasts only after subgroup-specific validation, and prioritize hydrological, tidal, meteorological, and pollution-source covariates for lagoon and extreme-event warning systems.

🔖

Cite This Article

Howard, C., Nnoka, L. (2026). Global Forecasting of Multiple Short Water-Quality Time Series Across River, Estuary, and Lagoon Monitoring Stations. Journal of International Environmental Application and Science, Volume: 21 Issue: 3. https://ajindex.com/articles/1112201

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (21)

  1. Ahmed AN, Othman FB, Afan HA, Ibrahim RK, Fai CM, Hossain MS, Ehteram M, Elshafie A. (2019). Machine learning methods for better water quality prediction. J. Hydrology, 578, 124084. https://doi.org/10.1016/j.jhydrol.2019.124084
  2. Anyanwu I, Beggel S, Sikoki F, Okuku E, Unyimadu J-P, Geist J, (2024) Total petroleum hydrocarbons, heavy metals and nutrients in a monthly time series collected at different ecosystems of the Niger Delta, Africa. PANGAEA. https://doi.org/10.1594/PANGAEA.964588
  3. Anyanwu, I. N, Beggel, S, Sikoki, FD, Okuku EO, Unyimadu, J-P, Geist J, (2023) Pollution of the Niger Delta with total petroleum hydrocarbons, heavy metals and nutrients in relation to seasonal dynamics. Sci. Reports, 13, 14079. https://doi.org/10.1038/s41598-023-40995-9
  4. Bergmeir C, Hyndman RJ, Koo B, (2018). A note on the validity of cross-validation for evaluating autoregressive time series prediction. Computational Statistics & Data Analysis, 120, 70–83. https://doi.org/10.1016/j.csda.2017.11.003
  5. Bi J, Lin Y, Dong Q, Yuan H, Zhou M. (2021). Large-scale water quality prediction with integrated deep neural network. Information Sciences, 571, 191–205. https://doi.org/10.1016/j.ins.2021.04.057
  6. Hewamalage H, Bergmeir C, Bandara K, (2022) Global models for time series forecasting: A simulation study. Pattern Recognition, 124, 108441. https://doi.org/10.1016/j.patcog.2021.108441
  7. Howard CC, Howard IC. (2026a) Hybrid and physics-based time series models for forecasting produced water quality: A comparative study in the Niger Delta. IJS J. Phys. Sci., 3(1), 122–133. https://doi.org/10.54117/ijps.v3i1.16
  8. Howard CC, Howard IC. (2026b) Hybrid neural network approaches to produced water quality forecasting and management in Nigerian oil fields. Int. J. Math., Statis., & Comp. Sci. 4, 527–543. https://doi.org/10.59543/7fjxt923
  9. Hyndman RJ, Koehler AB. (2006) Another look at measures of forecast accuracy. Int. J. Forecast., 22(4), 679–688. https://doi.org/10.1016/j.ijforecast.2006.03.001
  10. Januschowski T, Gasthaus J, Wang Y, Salinas D, Flunkert V, Bohlke-Schneider M, Callot L, (2020) Criteria for classifying forecasting methods. Int. J. Forecast., 36(1), 167–177. https://doi.org/10.1016/j.ijforecast.2019.05.008
  11. Li W, Zhao Y, Zhu Y, Dong Z, Wang F, Huang F, (2024). Research progress in water quality prediction based on deep learning technology: A review. Environ. Sci. & Pollut. Res., 31(18), 26415–26431. https://doi.org/10.1007/s11356-024-33058-7
  12. Lim B, Arik, SO, Loeff N, Pfister T, (2021) Temporal fusion transformers for interpretable multi-horizon time series forecasting. Int. J. Forecast., 37(4), 1748–1764. https://doi.org/10.1016/j.ijforecast.2021.03.012
  13. Lokman A, Ismail WZW, Aziz NAA. (2025). A review of water quality forecasting and classification using machine learning models and statistical analysis. Water, 17(15), 2243. https://doi.org/10.3390/w17152243
  14. Makridakis S, Spiliotis E, Assimakopoulos V, (2022) M5 accuracy competition: Results, findings, and conclusions. Int. J. Forecast., 38(4), 1346–1364. https://doi.org/10.1016/j.ijforecast.2021.11.013
  15. Montero-Manso P, Hyndman RJ, (2021) Principles and algorithms for forecasting groups of time series: Locality and globality. Int. J. Forecast., 37(4), 1632–1653. https://doi.org/10.1016/j.ijforecast.2021.03.004
  16. Nie Y, Nguyen NH, Sinthong P, Kalagnanam J. (2023) A time series is worth 64 words: Long-term forecasting with transformers. The Eleventh International Conference on Learning Representations. https://openreview.net/forum?id=Jbdc0vTOcol
  17. Salinas D, Flunkert V, Gasthaus J, Januschowski T, (2020) DeepAR: Probabilistic forecasting with autoregressive recurrent networks. International Journal of Forecasting, 36(3), 1181–1191. https://doi.org/10.1016/j.ijforecast.2019.07.001
  18. Tashman LJ, (2000) Out-of-sample tests of forecasting accuracy: An analysis and review. Int. J. Forecast., 16(4), 437–450. https://doi.org/10.1016/S0169-2070(00)00065-0
  19. Wang Z, Wang Q, Wu T. (2023). A novel hybrid model for water quality prediction based on VMD and IGOA optimized for LSTM. Front. of Environ. Sci. & Engin., 17, 88. https://doi.org/10.1007/s11783-023-1688-y
  20. Wu J, Wang Z. (2022). A hybrid model for water quality prediction based on an artificial neural network, wavelet transform, and long short-term memory. Water, 14(4), 610. https://doi.org/10.3390/w14040610
  21. Zeng A, Chen M, Zhang L, Xu Q, (2023) Are transformers effective for time series forecasting? Proceedings of the AAAI Conference on Artificial Intelligence, 37(9), 11121–11128. https://doi.org/10.1609/aaai.v37i9.26317