Hybrid modeling of karstic springs: Error correction of conceptual reservoir models with machine learning
Bibliographic record
Abstract
ABSTRACT Accurate spring discharge modeling and prediction is crucial for water management, helping authorities optimize use, manage variability, and prepare for droughts. Developing reliable simulation and forecasting tools is essential for effective management of groundwater resources from karstic springs. Although hybrid modeling approaches have been explored in hydrology, their application to spring discharge modeling is underexplored. Previous studies have focused on conceptual/distributed or data-driven models separately, missing the potential advantages of combining them. This creates a research gap in exploring the benefits of hybrid models for spring discharge. This study developed a hybrid model combining a conceptual GR5J model with Random Forests to simulate spring discharge from Bordeaux's largest karst aquifer. Model performance was assessed through comparison with the individual GR5J, RF, and benchmark models (weekly average of observed values). The hybrid model outperformed all models. Evaluation using actual meteorological data found the hybrid model achieved the highest accuracy by reducing GR5J simulation errors by 22%. When considering meteorological uncertainty, the hybrid model outperformed the individual GR5J, RF and benchmark models by 11, 30 and 47% respectively. The study findings suggest combining conceptual and machine learning approaches can improve spring discharge simulations, opening promising opportunities for enhanced forecasting in karst aquifers.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".