Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
April 10, 2026HydrologyOpen Access

Interpretable Deep Learning for Characterizing Sinkhole to Supply Well Transfer Dynamics in Karst Aquifers

View Full Paper
Ask AI
Bookmark
Share

Authors

BNBenoit NigonMGMathieu GodardAJAbderrahim Jardani

Discussion

Loading...

Member takes

Overview

Quantifies sinkhole impacts on turbidity peaks at a water supply well, indicating effective mitigation strategies in karst environments.

Key Points

  • The aim is to quantify the impact of sinkholes on turbidity peaks at supply wells in karst aquifers.
  • Simulated surface erosion and runoff using WaterSed.
  • Developed deep learning models based on hydroclimatic time series and erosion outputs.
  • Optimized various deep learning architectures to identify the best-performing model.
  • Conducted interpretability analyses to understand model predictions.
  • Turbidity is mostly influenced by seasonal conditions and rainfall accumulation.
  • Multiple sinkholes have a joint effect on turbidity with short time lags.
  • Temporal analyses indicate a rapid response from the karst, followed by attenuation.

Cite This Study

Nigon et al. (2026) studied this question.

synapsesocial.com/papers/69d8948f6c1944d70ce057e2https://doi.org/10.3390/hydrology13040102
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Sinkhole risk forecasting in the Lithuania–Latvia Karst region using artificial intelligence2026
  2. 2Deep-Learning-Based Automatic Sinkhole Recognition: Application to the Eastern Dead Sea2024 · 21 citations
  3. 3Comparative Analysis of Sinkhole Susceptibility Models Using Ensemble Machine Learning and Local Interpretable Model‐Agnostic Explanations ( <scp>LIME</scp> )2026 · 3 citations
  4. 4Identification of thresholds to conduct efficient soils and water conservation strategies against erosion impacts: new insights from a modelling prospective in Normandy (France)2024
  5. 5Explainable Spatially Distributed Hydrologic Modeling of a Snow Dominated Mountainous Karst Watershed Using Attention2024