Open research questions in Hydrological Forecasting Using AI

What the literature leaves open

• The lack of a comparative investigation of machine learning architectures for emulating reference evapotranspiration. The need for a study that evaluates the performance of machine learning models using chronological data partitioning and time-series-aware cross-validation.

  Physics-Based Target Emulation of Reference Evapotranspiration Using Machine Learning Surrogates: A Conceptual Digital Twin Perspective · 2026 · DOI

• The paper identifies the need for improved irrigation management and drought-resilient cropping systems in central Arizona. It highlights the urgency for accurate prediction of daily actual crop evapotranspiration for effective irrigation scheduling and sustainable water resources management.

  Machine Learning-Based Prediction of Guayule (Parthenium argentatum) Evapotranspiration Under Flood Irrigation and Arid Conditions · 2026 · DOI

• The gap is the difficulty of studying wind effects on water distribution uniformity in field conditions. Another gap is the need for efficient irrigation systems to ensure food security and sustainable water use. The paper also identifies the gap in the development of computational models that can accurately simulate water distribution by sprinkler irrigation systems working under windy conditions.

  Modeling Water Distribution Uniformity of Medium-Sized Sprinklers Using Artificial Neural Networks · 2025 · DOI

• CNNs are also limited temporal models and lack the inherent ability to capture temporal relationships. LSTMs and GRUs require high-quality datasets for effective performance and careful and time-consuming tuning for optimal performance. They still cannot perform well if there are missing data and noisy datasets.

  Recent Advances in Remote Sensing and Artificial Intelligence for River Water Quality Forecasting: A Review · 2025 · DOI

• Further effort is needed to explore the spatiotemporal evolution of nitrate concentrations and identify the controlling variables in real-world aquatic systems. How to improve the LSTM model to capture low-frequency, high-impact water quality extremes may remain an open research question. Adding properties like basin density might enhance the LSTM's ability to capture extreme flow events.

  Analyzing Multi-Year Nitrate Concentration Evolution in Alabama Aquatic Systems Using a Machine Learning Model · 2025 · DOI

• In situ nitrate observations typically lack the spatiotemporal resolution needed for a detailed understanding of biogeochemical processes. Empirical estimates based on temperature, salinity, chlorophyll, and other properties often perform poorly due to complex relationships between nitrate and these variables.

  Analyzing Multi-Year Nitrate Concentration Evolution in Alabama Aquatic Systems Using a Machine Learning Model · 2025 · DOI

• Forecasting wet extremes is a challenge. The ability to forecast precipitation in arid regions is limited. The method struggles with predicting precipitation in regions with isolated thunderstorm activity and infrequent convective systems.

  A Novel k-Nearest Neighbors Approach for Forecasting Sub-Seasonal Precipitation at Weather Observing Stations · 2025 · DOI

• Existing forecasting methods have limited predictability and inadequate spatial scale. Sub-seasonal to seasonal forecasting remains a challenge. There is a need for more accurate and reliable forecasting methods.

  A Novel k-Nearest Neighbors Approach for Forecasting Sub-Seasonal Precipitation at Weather Observing Stations · 2025 · DOI

• The need for efficient machine learning approaches for forecasting IWQI. The lack of comprehensive assessments of groundwater quality in Naama, Algeria.

  Groundwater Quality Assessment and Irrigation Water Quality Index Prediction Using Machine Learning Algorithms · 2024 · DOI

• The need to avoid overfitting or underfitting the data, producing inaccurate inferences.

  Groundwater Quality Prediction and Analysis Using Machine Learning Models and Geospatial Technology · 2024 · DOI

• In the meantime, most future flood estimations are conducted only at specific gauge stations due to limited data availability and are unable to support basin‐wide water resources planning and management.

  Improving Basin‐Wide Flood Estimation From a Global Hydrological Model Through Spatiotemporal‐Pattern‐Based Machine Learning · 2026 · DOI

• Future research should explore the integration of hybrid models that combine machine learning approaches with physically based hydrological models. Future research should focus on developing hybrid models that combine the strengths of various machine learning techniques with physical process-based approaches.

  A comparative analysis of machine learning models for predicting groundwater and surface water in a stressed semi-arid watershed: The Khanmirza case study · 2026 · DOI

• The lack of effective methods for detecting anomalies in water quality data. The need for a deep learning-based approach for unsupervised anomaly detection.

  Seasonal Anomaly Detection in the Halda River Using a Multivariate Deep Learning Framework · 2026 · DOI

• Integration with real-time sensor networks: Future implementations could integrate real-time IoT sensor data streams to enable continuous monitoring and live forecasting of water quality indices, enhancing responsiveness to pollution events.

  A new machine learning technique for predicting river water quality using AVOA-RNN · 2026 · DOI

• In summary, the combination of ML and DL models improved streamflow prediction in basins with limited data and successfully restored missing hydrometeorological data.

  Stream flow prediction utilizing deep learning models in the Lake Abaya-Chamo sub-basin, South Ethiopia · 2026 · DOI

• There has been no research done on the soil infiltration process employing machine learning methodologies. The study aims to fill this gap by demonstrating the potential of machine learning models for precise infiltration modeling.

  Approximation of cumulative infiltration of soils in arid and semi-arid regions · 2026 · DOI

• The limited availability of data in higher turbidity ranges. The sparsity and increased complexity of data beyond the 150 NTU threshold. Mitigating bias and minimizing the impact of variables on predictions.

  Random Forest Model for Optimizing Coagulant Doses in Drinking Water Treatment: Application at the Miguel de la Cuba Ibarra Plant · 2025 · DOI

• Investigating the application of machine learning models for ET0 estimation in other climates and regions, - Developing operational Digital Twins or integrated hydrological systems, - Exploring the use of other machine learning architectures for ET0…

  Physics-Based Target Emulation of Reference Evapotranspiration Using Machine Learning Surrogates: A Conceptual Digital Twin Perspective · 2026 · DOI

• Further analysis of the feature importance scores for input variables, - Investigation of other ML algorithms for predicting guayule ETa, - Application of the ML-based ETa models to other crops…

  Machine Learning-Based Prediction of Guayule (Parthenium argentatum) Evapotranspiration Under Flood Irrigation and Arid Conditions · 2026 · DOI

• The study area is representative of data-scarce regions, where monitoring is sparse and data quality presents ongoing difficulties, - Pumping volumes were estimated from agricultural electricity consumption records, which may incorporate systematic biases, - The influence of uncertainty on forecasting outcomes will be explicitly evaluated in the sensitivity analysis, - The model's performance may be affected…

  Application of a Hybrid CNN-LSTM Model for Groundwater Level Forecasting in Arid Regions: A Case Study from the Tailan River Basin · 2025 · DOI

• The model is limited to raw water with turbidity levels below 150 NTU, - The sparsity and increased complexity of data beyond 150 NTU introduce significant bias, - The study used a historical dataset of 2556 jar test…

  Random Forest Model for Optimizing Coagulant Doses in Drinking Water Treatment: Application at the Miguel de la Cuba Ibarra Plant · 2025 · DOI

• Using GIS and soft computing tools to monitor the GQI - Developing more accurate machine learning models for water quality…

  Groundwater Quality Prediction and Analysis Using Machine Learning Models and Geospatial Technology · 2024 · DOI

• The approach simultaneously optimizes input lag times and model hyperparameters, addressing a key limitation in previous studies.

  Hybrid machine learning for predicting groundwater level: A comparison of boosting algorithms with neural networks · 2025 · DOI

• This study proposed a hybrid AVOA-RNN framework for predicting river water quality using the Cauvery River dataset. By integrating advanced imbalance-handling strategies (SMOTE, SMOGN) with African Vulture Optimization for hyper-parameter tuning, the model demonstrated superior predictive capability in both regression and classification tasks.

  A new machine learning technique for predicting river water quality using AVOA-RNN · 2026 · DOI

• Developing optimized hybrid models for other environmental forecasting applications. Evaluating the performance of the XGBoost-FDA hybrid model using different datasets and case studies.

  Enhancing suspended sediment load forecasting using a physics-based optimized hybrid model: a regional study in the Mazandaran basins · 2026 · DOI

Most-cited papers in Hydrological Forecasting Using AI

• What Role Does Hydrological Science Play in the Age of Machine Learning? · Water Resources Research · 2020 · 728 citations
• Temporal convolutional neural (TCN) network for an effective weather forecasting using time-series data from the local weather station · Soft Computing · 2020 · 720 citations
• A Rainfall‐Runoff Model With LSTM‐Based Sequence‐to‐Sequence Learning · Water Resources Research · 2020 · 708 citations
• Root mean square error or mean absolute error? Use their ratio as well · Information Sciences · 2021 · 694 citations
• Short-term runoff prediction with GRU and LSTM networks without requiring time step optimization during sample generation · Journal of Hydrology · 2020 · 647 citations
• Ensemble machine learning paradigms in hydrology: A review · Journal of Hydrology · 2021 · 608 citations
• Hybrid decision tree-based machine learning models for short-term water quality prediction · Chemosphere · 2020 · 599 citations
• A survey on river water quality modelling using artificial intelligence models: 2000–2020 · Journal of Hydrology · 2020 · 549 citations
• Short-term water quality variable prediction using a hybrid CNN–LSTM deep learning model · Stochastic Environmental Research and Risk Assessment · 2020 · 545 citations
• Extreme gradient boosting (Xgboost) model to predict the groundwater levels in Selangor Malaysia · Ain Shams Engineering Journal · 2021 · 521 citations

Most recent work

• A hybrid WPD–SSA machine learning transformer for modeling nonlinear dynamics of flow systems · Journal of Water Process Engineering · 2026
• Autonomous inverse modeling of complex groundwater systems via a physics-integrated large language model multi-agent framework · Water Research · 2026
• A physics informed deep learning framework for rainfall forecasting in diverse climatic regions · Discover Artificial Intelligence · 2026
• Explainable Artificial Intelligence in Hydrology: A Review · Water Resources Management · 2026
• Operational agricultural water management in river-based aquaculture: A machine-learning approach to predict dissolved oxygen in the Halda River, Bangladesh · Aquacultural Engineering · 2026
• Multi-scale feature fusion and uncertainty quantification in streamflow prediction: A temporal convolutional network approach with hybrid denoising · Environmental Modelling & Software · 2026
• Application of LSTM coupled models in runoff simulation and prediction: a review · Applied Water Science · 2026
• Application of Hybrid Machine Learning for Groundwater Level Prediction: A Comprehensive Review · Archives of Computational Methods in Engineering · 2026
• Machine learning-driven rainfall forecasting model for sustainable and adaptive infrastructure planning · Discover Sustainability · 2026
• Machine learning-driven water quality index prediction in the Dau Tieng reservoir, southern Vietnam, with interpretability via SHapley additive exPlanations: A vision for water quality management strategies · Environmental Research · 2026

Related topics in Environmental Science