IHACRES, GR4J and MISD-based multi conceptual-machine learning approach for rainfall-runoff modeling.
Babak Mohammadi1, Mir Jafar Sadegh Safari2, Saeed Vazifehkhah3
1Department of Physical Geography and Ecosystem Science, Lund University, Sölvegatan 12, SE-223 62, Lund, Sweden. babak.mohammadi@nateko.lu.se.
Scientific Reports
|July 15, 2022
Summary
This study combined conceptual and machine learning models for rainfall-runoff (RR) modeling, achieving a 27% improvement in runoff estimation accuracy. The IHACRES-based MLP-WOA model demonstrated superior performance in snow-covered catchments.
Area of Science:
- Hydrology
- Environmental Science
- Data Science
Background:
- Rainfall-runoff (RR) modeling is crucial for water resource management but faces challenges with traditional conceptual and machine learning approaches.
- Conceptual models offer physical insights but may lack predictive power compared to data-driven methods.
- Machine learning models excel in computation but often obscure underlying hydrological processes.
Purpose of the Study:
- To develop a robust and reliable rainfall-runoff (RR) model by integrating conceptual and machine learning techniques.
- To enhance RR modeling accuracy by incorporating hydro-meteorological variables and employing data fusion strategies.
- To evaluate the performance of coupled hydrological models in a snow-covered catchment.
Main Methods:
- Applied three conceptual hydrological models (IHACRES, GR4J, MISD) for runoff simulation in a Swiss snow-covered basin.
- Developed multilayer perceptron (MLP) and support vector machine (SVM) models using conceptual model outputs and hydro-meteorological data.
- Implemented data fusion techniques, including MLP-WOA (whale optimization algorithm), to create advanced evolutionary RR models.
Main Results:
- The IHACRES-based conceptual model outperformed GR4J and MISD in simulating the RR process.
- Incorporating meteorological variables (precipitation, wind speed, humidity, temperature, snow depth) significantly increased model accuracy.
- The coupled IHACRES-based MLP-WOA model achieved an RMSE of 8.49 m³/s, improving performance by approximately 27% over the standard IHACRES model.
- Phase three coupling (hydrological model with machine learning) demonstrated minimal error in runoff estimation compared to other phases.
Conclusions:
- Coupling conceptual and machine learning models offers a superior approach to rainfall-runoff (RR) modeling, enhancing accuracy and reliability.
- The proposed methodology, particularly the IHACRES-based MLP-WOA model, shows significant potential for runoff estimation in complex hydrological environments.
- This integrated approach provides a valuable framework for addressing hydrological challenges and can be adapted for other water-related issues.
Related Concept Videos
01:17Design Example: Analyzing Capacity Contours for Flood Risk Assessment
96
Flood risk assessment involves careful planning and analysis to ensure the safety of communities near water retention structures. Capacity contours are a vital tool in this process, as they illustrate the potential spread of water at specific levels in a given area. In the context of building a bund across a small valley, these contours play a critical role in evaluating the safety of nearby residential areas.In this example, the bund is intended to store stormwater in the valley. The engineers...
96
01:03Precipitation Gravimetry
7.1K
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...
7.1K
01:12Precipitation Processes
576
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...
576
01:17Precipitation and Co-precipitation
2.0K
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...
2.0K
01:27Design Example: Design of an Irrigation Channel
206
Trapezoidal channels are widely used in irrigation systems due to their cost-effectiveness and efficiency in conveying water. Trapezoidal channels feature a flat bottom and sloping sides, making them stable and easier to construct compared to other shapes. The bottom width and side slope ratio are determined based on the required flow capacity and site conditions. The side slope is kept gentle for unlined channels to prevent soil erosion.Hydraulic parameters in channel design include the flow...
206
01:12Modeling and Similitude
325
Scaled modeling is a fundamental technique in engineering, enabling the study of large and complex systems by creating smaller, manageable replicas that recreate critical characteristics of the original. In hydrology and civil infrastructure, for example, scaled models of dams help analyze water flow, turbulence, and pressure. This method allows for accurate predictions of real-world behavior within a controlled environment, significantly reducing the cost and time involved in full-scale...
325