Decline in Mercury Pollution from Human Activities Found

Exciting environmental news emerges from MIT: Mercury emissions resulting from human activities have seen a notable decline over the last 20 years, contrary to what global emissions inventories suggest.

In a groundbreaking study, researchers assessed data from all available monitoring stations throughout the Northern Hemisphere, revealing that atmospheric mercury concentrations dropped by approximately 10 percent from 2005 to 2020.

Using two distinct modeling techniques, the researchers pinpointed a reduction in mercury emissions due to human activities as the primary factor behind this positive trend. This stands in stark contrast to global inventories, which often indicate rising emissions by basing estimates on average emission rates of pollutants and the scope of these activities worldwide.

“Our findings underscore the importance of utilizing actual, ground-level data to refine our models and emissions estimates. This is crucial for effective policy-making since incorrect estimations of past emissions hinder our ability to predict future mercury pollution developments,” explains Ari Feinberg, the study’s lead author and former postdoc at the Institute for Data, Systems, and Society (IDSS).

The implications of these results are significant for scientists engaged in a collaborative global initiative aimed at refining pollution models and gaining a deeper insight into what influences global atmospheric mercury concentrations.

However, the research highlighted challenges stemming from limited data from global monitoring stations and a general lack of understanding regarding mercury pollution, preventing the researchers from identifying a precise cause for the discrepancies between emissions inventories and observed measurements.

“It appears that mercury emissions are trending positively, which is encouraging. However, this is only the beginning. Continued measurement and scientific advancement are essential,” adds Noelle Selin, co-author of the study and an MIT professor in IDSS and the Department of Earth, Atmospheric, and Planetary Sciences (EAPS).

Feinberg and Selin, along with an international team of researchers, contributed to this work, which was published in the Proceedings of the National Academy of Sciences this week.

Understanding the Mercury Mismatch

The Minamata Convention is a vital global treaty aimed at reducing anthropogenic mercury emissions, a harmful neurotoxin emitted by sources such as coal-fired power plants and small-scale gold mining operations. Signed in 2013 and enacted in 2017, this treaty undergoes evaluation every five years. The first review coincided with troubling reports indicating that global mercury emissions had increased, contrary to international reduction efforts.

This confusing situation sparked concern among environmental scientists like Selin, particularly given that monitoring stations indicated a decrease in atmospheric mercury levels during the same timeframe.

Bottom-up inventories derive emission estimates by combining emission factors, such as the amount of mercury released when coal is burned, with data on related activities like coal combustion in power plants.

“The primary question we aimed to address was: What is occurring with mercury in the atmosphere, and what does it reveal about anthropogenic emissions?” Selin states.

Modeling mercury emissions poses unique challenges. Notably, mercury uniquely exists in liquid form at room temperature, lending it special characteristics. Additionally, mercury removed from the atmosphere may be re-emitted later by sinks such as oceans and land, complicating the identification of primary sources.

Moreover, mercury’s toxicity makes it more difficult to study in controlled environments compared to many other air pollutants, resulting in limited understanding of its chemical reactions. Furthermore, the network of monitoring stations for mercury is significantly smaller than for pollutants like methane and nitrous oxide.

“One of the main hurdles we faced was devising statistical methods to reconcile the data gaps, given that measurements came from various periods and networks,” Feinberg explains.

Innovative Modeling Techniques

The research team gathered data from 51 Northern Hemisphere monitoring stations and utilized statistical methods to merge data from nearby locations, allowing them to address data gaps and analyze regional trends.

Their findings indicated that atmospheric mercury concentrations in the Northern Hemisphere decreased by about 10 percent between 2005 and 2020. Researchers then employed two modeling approaches: biogeochemical box modeling and chemical transport modeling, to explore possible origins of this decline. The box modeling technique conducted extensive simulations, while chemical transport modeling enabled an analysis of how weather patterns and spatial variations affected trends.

For example, the researchers evaluated a hypothesis suggesting that an additional “sink” may be capturing more mercury from the atmosphere than previously recognized. Their models assessed the feasibility of such an undiscovered sink.

“As we systematically explored each hypothesis, we were quite surprised to identify declines in anthropogenic emissions as the likely cause,” notes Selin.

The importance of long-term mercury monitoring stations is emphasized by their findings, as many of the stations analyzed are no longer operational due to funding shortages.

While the team could not determine the exact reasons for the inconsistencies between inventory data and actual measurements, they proposed several hypotheses. One possibility is that global inventories might be lacking vital data from certain countries. They found that using a more thorough regional inventory from China clarified some discrepancies, yet a gap remained.

Additionally, the researchers speculate that changes involving two major and uncertain mercury sources—small-scale gold mining and mercury-laden products—might contribute to the observed discrepancies. Small-scale gold mining, often conducted in remote regions of developing countries, accounts for approximately 40 percent of anthropogenic emissions, making it difficult to estimate accurately. Furthermore, the timeframe for how long it takes for pollutants to be released from discarded products, such as thermometers or scientific instruments, is hard to determine.

“We have a long way to go before we can accurately identify the source of this discrepancy,” Feinberg remarks.

Moving forward, researchers from various countries, including MIT, will collaborate on studies aimed at improving emissions estimation models, which will be instrumental in enhancing mercury monitoring efforts, according to Feinberg.

This research was funded by the Swiss National Science Foundation, the U.S. National Science Foundation, and the U.S. Environmental Protection Agency.