Related Experiment Videos
Daily water level forecasting with limited data using cluster- and season-based transfer learning
Zexi Huang1, Wei Sun1, Yutong Xie1
1School of Geography and Planning, Sun Yat-Sen University, Guangzhou, Guangdong, 510275, China; Southern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Zhuhai, 519082, China.
The Science of the Total Environment
|June 9, 2026
Summary
Transfer learning combined with clustering or seasonal classification improves hydrological prediction accuracy, especially for short-term water level forecasting with limited data. This approach enhances model performance and interpretability for river forecasting.
Area of Science:
- Hydrology
- Water Resource Management
- Machine Learning in Environmental Science
Background:
- Deep learning combined with clustering or classification offers a knowledge-guided approach for hydrological prediction.
- Limited data in short-term water level forecasting poses challenges for developing robust deep learning models within clusters.
Purpose of the Study:
- To develop and evaluate a framework combining transfer learning with seasonal classification or Fuzzy C-means (FCM) clustering for short-term water level forecasting.
- To address data scarcity issues in hydrological prediction models.
- To improve the accuracy and interpretability of river forecasting.
Main Methods:
- Developed a framework integrating transfer learning with seasonal classification or FCM clustering.
- Pre-trained a Long Short-Term Memory (LSTM) model using various data treatments and hyperparameters.
- Fine-tuned the pre-trained LSTM model using seasonal data (four and two seasons) and FCM clustering with precipitation or water level data.
- Applied the framework to daily water level prediction at Lechang Gorge, China.
Main Results:
- The optimally pre-trained LSTM model showed minimal bias (-0.003 m) and strong validation performance (R=0.943, NSE=0.457, RMSE=0.889 m).
- Seasonal classification fine-tuning showed varied performance, outperforming in dry periods but underperforming in flood seasons.
- FCM clustering with precipitation data yielded the best results, outperforming the baseline with a 6.2% improvement (RMSE=0.412 m) in validation.
Conclusions:
- Transfer learning effectively overcomes data limitations in hydrological forecasting by leveraging broader datasets.
- Combining transfer learning with FCM clustering, particularly using precipitation data, significantly enhances short-term water level prediction accuracy.
- The proposed framework improves model interpretability and identifies pathways for future advancements in river forecasting systems.
Related Concept Videos
01:20Cluster Sampling Method
Appropriate sampling methods ensure that samples are drawn without bias and accurately represent the population. Because measuring the entire population in a study is not practical, researchers use samples to represent the population of interest.
To choose a cluster sample, divide the population into clusters (groups) and then randomly select some of the clusters. All the members from these clusters are in the cluster sample. For example, if you randomly sample four departments from your...
To choose a cluster sample, divide the population into clusters (groups) and then randomly select some of the clusters. All the members from these clusters are in the cluster sample. For example, if you randomly sample four departments from your...
01:19Survival Tree
Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
Building a Survival Tree
Constructing a survival tree begins...
Building a Survival Tree
Constructing a survival tree begins...
01:03Precipitation Gravimetry
Precipitation gravimetry is based on converting an analyte into a sparingly soluble precipitate, which is separated by filtration and weighed. An ideal precipitate should be pure, insoluble, of known composition, and easily filtered from the reaction mixture.
In determining nickel by gravimetric analysis, a precipitant of ethanolic dimethylglyoxime is added to a hot nickel salt solution. This is quickly followed by the dropwise addition of dilute ammonia solution until precipitation occurs. A...
In determining nickel by gravimetric analysis, a precipitant of ethanolic dimethylglyoxime is added to a hot nickel salt solution. This is quickly followed by the dropwise addition of dilute ammonia solution until precipitation occurs. A...
01:17Precipitation and Co-precipitation
Precipitation and coprecipitation methods can be used to separate a mixture of ions in a solution. In qualitative inorganic analysis, ions that form sparingly soluble precipitates with the same reagent are separated based on the differences in solubility products. For example, consider the separation of Cu(II) and Fe(II) ions by precipitation as insoluble sulfides. First, copper(II) sulfide is precipitated by the addition of acidic H2S, where the dissociation of H2S is suppressed. Adding H2S...
01:18Uniform Depth Channel Flow: Problem Solving
To calculate the flow rate for a trapezoidal channel, first, identify the bottom width, side slope, and flow depth of the channel. The cross-sectional area (A) corresponding to the depth of flow (y), channel bottom width (B), and side slope (θ) is determined by:Next, calculate the wetted perimeter, which includes the bottom width and the sloped side lengths in contact with the water. Using the values of the cross-sectional area and the wetted perimeter, determine the hydraulic radius by...
02:41Responses to Drought and Flooding
Water plays a significant role in the life cycle of plants. However, insufficient or excess of water can be detrimental and pose a serious threat to plants.