
In the traditional picture of plant growth, roots absorb water from soil. But new research revealed that woody plants across the United States in fact regularly rely on a deeper source of water—that found in bedrock. These results highlight a relatively unexplored component of the terrestrial water cycle, the researchers suggested.
Digging Deep
The root systems of some shrubs and trees can run deep—up to tens of meters into the ground. Such downward diving roots pass right through any soil that’s present and head straight into the weathered bedrock below. And there’s potentially a very important resource to be found there, in the heart of the near-surface, life-sustaining region of the planet known as the critical zone.
The fractured structure of weathered bedrock traps moisture, said Daniella Rempe, a hydrologist at the Jackson School of Geosciences at the University of Texas at Austin. “It has the capacity to store water.”
At a field site in California, Rempe and her colleagues recently showed that the roots of woody plants make use of water in bedrock. But it’s important to get a handle on whether such a phenomenon is widespread, said Rempe. “We’ve done a lot of on-the-ground, site-specific work,” she said. “We have a limit on what we can say…until we scale up.”
Data Wrangling
“This would have taken [as much time as] a whole Ph.D.”
Rempe and her collaborators have now done just that. The team started by mining existing literature related to plants’ use of rock moisture and assembling data sets related to precipitation, evapotranspiration, soil depth, and more. Team member Erica L. McCormick, an ecohydrologist also at the Jackson School of Geosciences, then analyzed much of those data using Google Earth Engine, an online platform for processing geospatial information. That technology was indispensable for this project, said McCormick. Without it, the data-intensive analysis could have stretched out over years, she said. “This would have taken [as much time as] a whole Ph.D.”
The researchers adopted a so-called deficit-based method to estimate, over the continental United States, how much water plant roots pull from bedrock. They tracked, on a daily basis for pixels measuring 500 meters on a side, the amount of water going into the system (i.e., precipitation) and the amount of water going out (i.e., evapotranspiration). When there was more evapotranspiration than precipitation occurring in a particular pixel, that represented a deficit that must have originated from subsurface storage.

The true water storage capacity of the subsurface is unknown, the authors acknowledged, but their calculations yielded a lower limit. “You know that that amount of water was pulled from storage; therefore your storage capacity must be at least that large,” said McCormick.
To home in on just the bedrock component of that storage capacity, the team used a U.S. Department of Agriculture data set to subtract out the soil water storage capacity in the top 1.5 meters of the subsurface. At the same time, the researchers restricted their analysis to pixels with shallow soil depths. “We wanted to make sure that 100% of the soil was occurring in the upper 1.5 meters,” said McCormick.
All of California’s Reservoirs
“It’s a huge volume of water.”
The researchers found that woody plants over wide swaths of the continental United States pulled water from bedrock. Every year from 2003 to 2017—the interval spanned by the team’s data—vegetation in California alone pulled roughly the same amount of water from bedrock as is stored in all of the state’s reservoirs combined. “It’s a huge volume of water,” said McCormick.
And woody plants elsewhere in the western United States and the Appalachian region relied on bedrock water some years during the 2003–2017 interval, the team found. Usage of bedrock-derived water is therefore “widespread and routine,” the researchers concluded. These results were published last month in Nature.
The existence of significant stores of water in bedrock has the potential to rewrite our understanding of the terrestrial water cycle, the team suggested. That’s because bedrock is the conduit through which precipitation must first flow before ending up in aquifers. “Plants are taking substantial amounts of this water straight from bedrock, which has big implications for our predictions of how much water will be left in storage,” McCormick told Eos.
“We’re going to have a really hard time if we don’t get this right,” said Rempe.
In the future, it’ll be critical to extend this analysis beyond the United States. “It would be interesting to see a larger, global-scale study,” said Lan Wang-Erlandsson, a hydrologist at the Stockholm Resilience Centre at Stockholm University who was not involved in the research. “My wish list is more data.”
—Katherine Kornei (@KatherineKornei), Science Writer
This news article is included in our ENGAGE resource for educators seeking science news for their classroom lessons. Browse all ENGAGE articles, and share with your fellow educators how you integrated the article into an activity in the comments section below.