Knappett et al. 2022 — Arsenic and fluoride remediation in overexploited aquifers, Guanajuato, Mexico
This systems dynamics modelling study evaluates how groundwater overexploitation in the Bajio region of Guanajuato, Mexico interacts with arsenic (As) and fluoride (F) concentrations in drinking water wells. Under baseline scenario S1 (business as usual), arsenic concentrations in pumped groundwater are projected to rise from approximately 14 µg/L to 46 µg/L over 100 years as deep, naturally arsenic-enriched aquifer layers are increasingly drawn upon. The paper proposes managed aquifer recharge (MAR) as the primary remediation lever and demonstrates that aggressive aquifer recharge can maintain As and F concentrations closer to the WHO guideline of 10 µg/L As.
Key numbers
- Current mean As in pumped groundwater (baseline): approximately 14 µg/L
- Projected mean As after 100 years under S1 (no intervention): 46 µg/L — more than 4x the WHO drinking water guideline of 10 µg/L
- Fluoride co-occurs with arsenic in this aquifer system; both are geogenic contaminants from volcanic rock weathering
- WHO drinking water guideline for As: 10 µg/L; Mexico’s official limit: 25 µg/L NOM-127
- Study region: Bajio valley, Guanajuato state, Mexico; population reliant on the aquifer: hundreds of thousands
- Modelling horizon: 100 years; system dynamics model validated against observed trends
Methods (brief)
Systems dynamics modelling using Vensim software. Arsenic concentrations derived from synthesis of existing monitoring data across the Guanajuato aquifer system. Model integrates pumping rates, aquifer recharge, aquifer stratification (shallow vs. deep zones with different As/F profiles), and remediation scenarios. No direct food sampling; drinking water matrices only.
Implications
Certification: Relevant as background context for arsenic in drinking water used for food processing or irrigation; not direct product concentration data. Courses: Illustrative case for how overexploited aquifers increase heavy metal exposure to local populations; good systems-level framing for groundwater-to-food-chain pathways. App: Not directly applicable (no food matrix concentrations). Microbiome: Not addressed.