If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.



Fluoride hot spots identified in groundwater aquifers

Researchers use machine learning to predict fluoride exposure in areas with little to no monitoring

by Krystal Vasquez
August 12, 2022

Researchers from the Swiss Federal Institute of Aquatic Science and Technology (Eawag) have created a machine learning algorithm that can predict where groundwater might contain unsafe levels of naturally occurring fluoride. Using the model, the scientists calculated that approximately 180 million people worldwide could potentially be exposed to fluoride concentrations above the World Health Organization’s recommended guideline of 1.5 mg/L (Nat. Commun. 2022, DOI: 10.1038/s41467-022-31940-x).

Fluoride hot spots
Global map showing where fluoride concentrations above 1.5 mg/L are most likely located. Hot spots can be seen in the western US, central Australia, eastern Brazil, and many areas of Africa and Asia.
Credit: Nat. Commun.

This map shows where the probability of fluoride concentrations in groundwater exceeds the World Health Organization's recommended limit of 1.5 mg/L.

Fluoride is abundant in the Earth’s crust and enters groundwater through the weathering of minerals. Although scientists have shown that fluoride concentrations within the range of 0.5 to 1 mg/L in drinking water are safe and can help prevent tooth decay, consuming too much can negatively impact human health. People chronically exposed to fluoride above the WHO’s recommended limit can develop dental fluorosis, which causes brown staining of the teeth. Fluoride consumption above 6 mg/day can also cause skeletal fluorosis, a disabling disease characterized by bone deformities and joint pain.

Joel Podgorski, a senior scientist at Eawag who led the work, trained the algorithm by using 400,000 publicly available fluoride measurements collected across 77 countries. He and his colleague Michael Berg, head of Eawag’s Department of Water Resources and Drinking Water, combined these data with spatially continuous maps of environmental parameters that are known to influence fluoride accumulation in groundwater, such as aridity, groundwater pH, and topography.

The resulting output was a global-scale map that can predict fluoride concentrations even in areas where no groundwater data had been collected previously. “No one else has really done this, recently, on a global scale,” Podgorski says. Although Eawag researchers had created another global fluoride map in 2008, Podgorski explains that that model’s resolution was much coarser because less fluoride data was available at the time (Environ. Sci. Technol. 2008, DOI: 10.1021/es071958y).

For this study, Podgorski also identified which areas were at highest risk of exposure by combining the fluoride map with global population data, as well as estimates of each country’s groundwater usage. Although fluoride hot spots were seen around the world, the top 20 at-risk countries were all located in Asia and Africa, likely due to these areas’ geology, large population densities, and high reliance on groundwater, Podgorski explains.

Many of the identified hot spots also coincide with places where little to no monitoring takes place, says Katherine Ransom, a hydrologist at the US Geological Survey, California Water Science Center. “This information can benefit the health of people living in areas with high concentrations of fluoride in their groundwater,” she says. “The maps and population estimates may be used to define or refine where future groundwater monitoring” or groundwater treatment, is necessary.


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.