Metaheuristic-enhanced deep learning for monthly pan evaporation prediction under limited climatic data
Ozgur Kisi1,2,3, Rana Muhammad Adnan4,5, Mohammad Zounemat-Kermani6
1Department of Civil Engineering, Lübeck University of Applied Sciences, 23562, Lübeck, Germany. ozgur.kisi@th-luebeck.de.
Scientific Reports
|May 2, 2026
Summary
This study introduces novel Artificial Protozoa Optimizer (APO) and Dung Beetle Optimizer (DBO) algorithms to improve long short-term memory (LSTM) networks for monthly pan evaporation prediction, showing significant accuracy gains.
Area of Science:
- Hydrology
- Artificial Intelligence
- Metaheuristic Optimization
Background:
- Accurate monthly pan evaporation prediction is crucial for water resource management.
- Limited climatic data poses challenges for traditional hydrological models.
- Bio-inspired metaheuristic algorithms offer potential for optimizing complex models.
Purpose of the Study:
- To integrate Artificial Protozoa Optimizer (APO) and Dung Beetle Optimizer (DBO) into Long Short-Term Memory (LSTM) networks for monthly pan evaporation prediction.
- To evaluate the performance of LSTM-APO and LSTM-DBO against standard LSTM and other hybrid models using limited climatic data.
- To assess the effectiveness of these novel hybrid models in enhancing prediction accuracy and generalization.
Main Methods:
- Developed hybrid models: LSTM integrated with APO (LSTM-APO) and DBO (LSTM-DBO).
- Benchmarked performance against standard LSTM, LSTM-GWO, and LSTM-HHO.
- Utilized a 40-year dataset from two stations in southeast China across three data-splitting scenarios (M1, M2, M3).
- Evaluated models using metrics such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), coefficient of determination (R²), and Nash-Sutcliffe Efficiency (NSE).
Main Results:
- Both LSTM-APO and LSTM-DBO consistently outperformed benchmark models across all scenarios.
- LSTM-APO demonstrated substantial error reductions (up to 47.2% in RMSE/MAE) and improved R²/NSE (up to 9%) compared to the best LSTM.
- LSTM-DBO also yielded significant error reductions (20-30%) and showed robust predictive stability.
- Visual analyses confirmed LSTM-APO's predictions closely matched observed data, with LSTM-DBO performing comparably.
Conclusions:
- APO and DBO effectively optimize LSTM hyperparameters for improved monthly pan evaporation prediction, especially with limited data.
- These novel hybrid models show strong promise for reliable hydrological forecasting, outperforming established methods.
- Further research is recommended to explore the real-time applicability and cross-climate transferability of LSTM-APO and LSTM-DBO.
Related Concept Videos
01:17Precipitation and Co-precipitation
4.8K
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...
4.8K
01:12Precipitation Processes
5.0K
The experimental conditions in a gravimetric analysis should be optimized to maximize the particle size and purity of the obtained precipitate. Ideally, the concentration of the precipitating reagent should be low with effective stirring to maintain low relative supersaturation for the growth of large crystals. In homogeneous precipitation, the precipitant is slowly generated by a chemical reaction in the solution to avoid local reagent excesses. For example, urea decomposes gradually to...
5.0K
01:03Precipitation Gravimetry
12.8K
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...
12.8K
01:57Adaptations that Reduce Water Loss
24.2K
Though evaporation from plant leaves drives transpiration, it also results in loss of water. Because water is critical for photosynthetic reactions and other cellular processes, evolutionary pressures on plants in different environments have driven the acquisition of adaptations that reduce water loss.
24.2K