During droughts, groundwater pumping is increased to make up for losses from surface water. This is especially true in California’s Central Valley, which stretches roughly 400 miles from Redding to just south of Bakersfield, and is the heart of the state’s $47 billion-a-year agricultural industry.
For decades, many parts of the Central Valley aquifer have been overdrafted, but recent work by scientists from the University of California, Los Angeles, and the University of Houston have attempted to put a more precise number on how much water is being pumped.
They found that over the past five years of drought, California withdrew 9.5 cubic miles of groundwater – around seven times the amount of water in Lake Shasta, the state’s largest reservoir.
The study, published in the journal Geophysical Research Letters, also estimated that during another drought, from 2007 to 2009, the valley lost 4 cubic miles of groundwater. These numbers aren’t particularly surprising, said Dennis Lettenmaier, a professor of geography at UCLA and an author of the study. What is news, he said, is that the study puts numbers to something that is difficult, but important, to calculate. He points out that pumping up groundwater during a drought can make sense, but there needs to be a strategy for replenishing that resource.
Community water wells can run dry, Lettenmaier noted. “And there’s a practical issue with subsidence, the fact that the land drops” when water is extracted, he added. “It has been a problem in some parts of the Central Valley for a long time, but it certainly accelerated by the groundwater depletion.”
The Central Valley groundwater basin is the region’s largest single source of water. It may hold more than 1 billion acre-feet of water, said Thomas Harter, a professor of hydrology at the University of California, Davis, who was not involved in the study. Water from rain, irrigation and surface water can soak into the ground and replenish this underground aquifer, but, in dry years, the state might pump 5 to 7 million acre-feet more groundwater than is being replenished by the natural system. In wet years, though, California probably only refills aquifers by 1 to 5 million acre-feet.
This Dec. 22, 2015, photograph shows a buckle in the lining of the Delta Mendota Canal near Dos Palos, California, caused by sinking land. Years of drought and heavy reliance on pumping of groundwater have made the land sink faster than ever up and down California’s Central Valley, requiring repairs to infrastructure that experts say are costing billions of dollars. (AP/Scott Smith)
“So, you need a number of these above-average wet years to make up for a long drought like what we had in the last five years,” said Harter. “Until we’ve had those, the drought, from a groundwater perspective, isn’t really over.”
One way to measure this loss in underground water is by looking at changes in gravity. NASA’s Gravity Recovery and Climate Experiment (GRACE) uses satellites to detect gravitational changes in an area of land. As water is pumped out of the ground, the loss of mass decreases the area’s gravitational pull, and GRACE can estimate how much was lost.
But the system’s resolution is large, and the readings can often include more than the valley.
So Lettenmaier and the other researchers looked at both GRACE data and calculated groundwater levels over time using the “water balance method,” which estimates groundwater based on how much water flows into an area, and how much leaves it. It’s analogous to estimating the amount of water in a bathtub by comparing the water pouring in from the tap to what’s flowing down the drain, Lettenmaier said. But it’s a little more complicated than that, and the researchers had to make some assumptions, like how much water was lost through evapotranspiration. So although it has a better resolution than GRACE, it has “a host of different uncertainties,” said Lettenmaier.
Still, despite such limitations, GRACE and water balance method results matched surprisingly well – both showed about the same amount of groundwater lost during the droughts. However, between the droughts there was some discrepancy. The researchers estimated that the Central Valley recovered around 5 cubic miles of groundwater from 2009 to 2012 – substantial, though not enough to make up for the overdraft created during the 2007 to 2009 drought. GRACE, meanwhile, showed no recovery at all.
Lettenmaier said that he doesn’t know exactly why those numbers don’t reconcile.
Still, the new study was an important step towards understanding Central Valley groundwater, said Peter Gleick, president emeritus and chief scientist of the Pacific Institute, and a reviewer on the paper. Now, he said, these numbers need to be refined. For one, he’d like to see an increase in resolution from GRACE.
“It would be nice to have more detail about the southern San Joaquin, versus the central part of the San Joaquin, versus the northern part of the San Joaquin,” said Gleick. “The Central Valley is not a single aquifer; it is many different groundwater basins, and ultimately every one of them has to be evaluated separately.”
Increased resolution isn’t currently possible, said Carmen Boening, a project scientist on GRACE at NASA’s Jet Propulsion Laboratory, but it is something that NASA is working on for the next version, called GRACE Follow On, to launch in late 2017 or early 2018.
As for water balancing, Gleick said that what researchers like Lettenmaier really need is more data.
“Ideally, we need to measure, monitor and report every use of groundwater,” said Gleick. “That’s the first step towards sustainable management, and we’re moving forward too slowly in that direction.”