There but for GRACE ...
Unless you've been catatonic it's been hard to miss the news coverage given to the latest paper from the GRACE groundwater guru, Jay Famiglietti, and his colleagues: Groundwater Depletion During Drought Threatens Future Water Security of the Colorado River Basin, by Stephanie L. Castle, Brian F. Thomas, John T. Reager, Matthew Rodell, Sean C. Swenson, and James 'Jay' S. Famiglietti. The paper will soon be published in Geophysical Research Letters. Here is a pre-publication copy, followed by the abstract:
Streamflow of the Colorado River Basin is the most over-allocated in the world. Recent assessment indicates that demand for this renewable resource will soon outstrip supply, suggesting that limited groundwater reserves will play an increasingly important role in meeting future water needs. Here we analyze nine years (December 2004 to November 2013) of observations from NASA’s GRACE mission and find that during this period of sustained drought, groundwater accounted for 50.1 km3 of the total 64.8 km3 of freshwater loss. The rapid rate of depletion of groundwater storage (-5.6 ± 0.4 km3 yr-1) far exceeded the rate of depletion of Lakes Powell and Mead. Results indicate that groundwater may comprise a far greater fraction of Basin water use than previously recognized, in particular during drought, and that its disappearance may threaten the long-term ability to meet future allocations to the seven Basin states.
Relating the depletion to the volumes of Lake Mead (see the press release) was a stroke of genius by the authors. Just about everyone knows how big that reservoir is. Kudos - it's a great way to communicate results to the public. I even heard a local CBS-TV news reporter express the groundwater loss as 13 trillion gallons - now that's a number people know is HUGE! But the Lake Mead analogy is spot-on, especially in the West.
What surprised the authors was the amount of groundwater depletion, most of which occurred from the lower Colorado River Basin (CRB). In units commonly used in the Western US, the rate of groundwater depletion over nine years is about 4.5 MAF (million acre-feet) per year.
As Stephanie Castle, the lead author, said in a High Country News blog post by Sarah Jane Keller:
“I think the key phrase has been ‘shocking’.”
A Brief Digression
As an aside - Keller also contributed this comment of her own:
"Since groundwater is a local resource, solutions to its runaway depletion will likely be local too."
That statement is not always true; think of regional aquifers (areal extents exceeding thousands of square miles) like the High Plains aquifer, Madison aquifer, Floridan aquifer, Memphis Sand, and others. Even smaller groundwater basins can underlie several political jurisdictions and pumping in one part of the basin in one jurisdiction can influence the basin's aquifer beneath another political jurisdiction. Is that local? I guess it depends upon what you call 'local'.
How Much Groundwater is There?
But I digress. The reason for this post is to expand upon the statement about the magnitude of the groundwater depletion and cause for concern in the Colorado River Basin. Castle actually hit the nail on the head in this news release:
"We don't know exactly how much groundwater we have left, so we don't know when we're going to run out."
This is an important point. Is the amount of groundwater depletion determined by Castle et al. really a huge amount? I don't know, but I have to say that it did not surprise me all that much until I read it's about 22% of the USA's annual groundwater depletion (during 2000 - 2008) reported by Lenny Konikow in a USGS report. Despite that, I want to know how much the depletion rate is as a percentage of the amount in storage.
Some Things to Consider & Questions to Ask
1) The Colorado River Basin is about 250,000 square miles (about 640,000 square kilometers or 160M acres). Is 4.5 MAF per year a large amount over that expanse? Want to minimize all this? Divide the annual depletion by the CRB's area in acres. You get an average annual depletion of 0.028 AF per acre or about 9200 gallons per acre. The total decline over nine years would be about 0.24 feet or about 0.03 feet per year. Doesn't sound so bad, does it? Disclosure notice: not all the land beneath the CRB has large groundwater reservoirs and the groundwater declines are not uniformly distributed.
2) The resolution of the GRACE information is about 200,000 square kilometers (about 77,000 square miles). It cannot discern changes at scales smaller than that.
3) Does the deficit represent increased pumpage, or decreased recharge, or some combination of the two? See this article by Brett Walton.
4) Recharge reductions caused by the current/recent drought probably have not yet shown up in groundwater, except in shallow aquifers. It takes a long time (decades, centuries, millennia) for water to travel from the surface and become groundwater recharge, especially in dry regions like the CRB.
5) We don't know how much groundwater is stored - the stock of groundwater - beneath the surface of the CRB. GRACE cannot tell us that, but other data can.
6) Of that stored groundwater alluded to in (5): we don't know the quality or the amount we can actually extract. When I say 'extract' I mean in the physical sense (recoverable or drainable groundwater), legal sense (water rights), or environmental sense (e.g., streamflow depletion, land subsidence). These items (water rights, environmental, etc.) will likely limit how much groundwater we can withdraw, so we will likely not come close to extracting all the groundwater that is physically available.
Of Groundwater Stocks and Flows
The GRACE information discerns only changes in storage, not total storage. It provides information on the flows, not the stocks. Recall that as opposed to surface water systems that are flow-dominated, groundwater systems have relatively small flows and large stocks. But these flows are still important because they enable groundwater to: 1) be considered a renewable resource in many instances; and 2) participate in the hydrologic cycle (Margat and van der Gun, 2013, p. 9).
Friend and colleague Michael Wallace, a hydrologist studying for a PhD at the University of New Mexico, reminded me of this issue - stocks v. flows - when he expressed incredulity and dismay a few months ago when he could not find a documented figure for the total volume of groundwater in the Albuquerque-Belen Basin (ABB) of New Mexico. He said that in the absence of that number, the basin is 'not quantified'. Later, with the help of John Hawley, we figured that 100 - 125 MAF was reasonable (Mike had calculated something like 250 MAF). Keep in mind that the ABB is about 1% as extensive (surface area) - 3,100 square miles - as the CRB.
Unfortunately, stocks for aquifers are not commonly calculated or reported, although they can be estimated. The U.S. Geological Survey has done some calculations: I suspect some individual states have done so as well. I should note that Lenny Konikow's excellent USGS report, Groundwater Depletion in the United States, 1900-2008, does not contain estimates of aquifer storage volumes.
In 1980 (albeit 35 years ago), the USGS reported that the High Plains (aka Ogallala) aquifer in Texas had about 390 MAF of drainable water in storage and the Oklahoma portion had about 110 MAF for a total of 500 MAF (nice round number...Hmm). What is it now? Using a figure of 15,000 AF/day of pumpage over 35 years, I calculated 180 MAF of depletion so we are down to 320 MAF. That is a lot of depletion - over 20% more per year (c. 5.5 MAF) than the CRB.
Why is such a number - the groundwater stock - important? In my class I likened it to managing a checking account without knowing the balance (S; the stock) but just the deposits (I; inflows) and withdrawals (O; outflows). As long as I = O, you're fine, regardless of the S (we're assuming no fees or problems with crediting deposits and floating checks). You may feel uncomfortable but you'll be okay. However, as soon as O > I in a given time period you had better start worrying simply because you don't know how long you can keep running a deficit before hitting bottom. That's the problem we have with groundwater in many places.
In the CRB, as in many other groundwater basins, we are likely in the O > I phase, but we don't know how much water is in storage. But we need to know the storage to properly manage the groundwater.
My Ten Cents
So what can we glean form the current GRACE study? There is good and bad news. Here are 'My Ten Cents':
1) Groundwater and groundwater management are getting good press. Maybe those organizations who refuse to acknowledge the importance of groundwater will realize that's downright stupid. And maybe those who think they can keep pumping groundwater like there's no tomorrow (listen up, California and Texas!) with no consequences will realize that they are just kidding themselves.
2) GRACE is a very useful tool (see No More Hydro-Secrets Revisited: GRACE, Groundwater, and Geopolitics) but there is a lot of stuff it cannot do at the present time. The cognoscenti realize its limitations, but not the general public. This can lead to 'irrational exuberance' on the part of the public, government officials and others ('Hey, this GRACE stuff can solve our problems! Let's call these guys up!') Consider this from a 30 July 2014 editorial in the Santa Fe New Mexican (emboldening mine):
Most intriguing for New Mexicans is the knowledge that the Rio Grande Basin — our lifeline — is large enough to study in the same manner. The scientists involved in studying the Colorado River Basin say they can measure New Mexico’s water supply underground as well.
Legislators, the State Engineer’s Office and others who manage water resources in the state should commission such a study.
Just as with surface water allocations, if New Mexicans don’t know what water remains underground, we can’t use it wisely.
What the editorialists and others don't know is that the data probably exist (or can be collected) to ascertain their groundwater supplies. GRACE can't do it for them. My feeling is that some states don't want to know how much groundwater they have.
3) When someone tells me that we're losing a lot of groundwater, I want to know how much is left or what % of the total is represented by the changes. GRACE cannot do that.
4) I hope I don't start hearing about managing aquifers based on groundwater budgets.
5) GRACE is a useful tool, but it is does not present the entire picture, escpecially regarding groundwater. There is no substitute for boots-on-the-ground, old-fashioned (using modern tools, of course), hydrogeological work. Use GRACE in conjunction with other data (I'm sure that is what Jay would say). Don't make it just another BSO ('bright shiny object').
I'm done - for now.
"It's a desert, stupid!" - bumper sticker, Albuquerque, NM, early 1990s (thanks to Jean Witherspoon)