Memphis, Tennessee, the northern terminus and urban hub of the Mississippi Delta, was a foul place in the mid-19th century. Yellow fever swept the city three times in the 1870s, killing 8,000 residents and scaring away thousands more. The city’s water system, patchwork and decrepit, was a key culprit, offering perfect breeding grounds for mosquitoes. By the 1880s, the city was drawing water from the Wolf River, muddy and polluted. According to old state health bulletins, complaints about the lack of filtration went unanswered.

Then, in 1887, an ice company drilled a deep well downtown. Out rushed clear water. The digger had lucked upon a deep aquifer whose water is still touted in local press releases as some of the sweetest in the world.

The city was ecstatic. “Crowds soon collected about the flowing fountain,” one local official recalled a few years later. “Policemen were in requisition. The news spread like wildfire. The elixir of life had been found.”

That’s a highfalutin’ phrase, perhaps, but when it comes to groundwater—the water that fills the pores in underground layers of sand, gravel, and rock—grandiosity is common. In 1861 the Ohio Supreme Court declared such water “so secret, occult, and concealed” that any attempt to regulate its use would be senseless.

Hydrological knowledge has advanced plenty in the subsequent 150 years, but groundwater retains an air of mystery. We know that there is a lot of it—20 to 100 times more than fresh surface water—and it is generally cheap, clean, and easy to find: Just drill down. And so, as the world goes dry, more and more people are turning to groundwater. Already, more than a third of the world’s farmland is equipped for groundwater irrigation, and between 25 and 40 percent of the world’s population relies on groundwater as their primary drinking source. Since 1950, the rate of pumping in the U.S. has more than doubled, reaching 80 billion pounds a day in 2005.

But since our understanding of the thickness and storage of some aquifers is imprecise, estimates of worldwide groundwater volume can differ by several orders of magnitude. And even if we had a precise figure, we couldn’t be sure when groundwater pumping would become too expensive—or too destructive—to be practical.

“Unfortunately, when it comes to mustering political and financial support for even the most basic data collection, groundwater is usually a low priority,” says Bill Alley, the former chief of the U.S. Geological Survey’s Office of Groundwater. Alley serves now as the director of science and technology with the National Ground Water Association and is currently working on a book on the issue.

“Groundwater is very important and is likely to become more important,” he says. Yet groundwater policy varies wildly from state to state, and where aquifers cross state borders, policy is even less clear. No national legislation or Supreme Court decision has established how groundwater should be shared across states.

Memphis still pumps from that same aquifer, and there is no sign that it will run out anytime soon. But in Mississippi, farmers are worried about their irrigation wells, and the state is suing Memphis for drawing up drinking water from across the border. The Mississippi River Alluvial Aquifer—the aquifer used by farmers, closer to the surface than the source of drinking water—has according to some estimates been declining by 1 to 1.5 feet a year over the past four decades. The aquifer is measured twice each year by the local water-management district, which turns the data over to the Mississippi Department of Environmental Quality (MDEQ). “You can see recovery in the aquifer between the fall, after irrigation is over, and when you measure again in the spring,” Kay Whittington, the director of the MDEQ’s Office of Land and Water Resources, told me. “It’s just a matter that it’s not necessarily recovering enough.

This story could take place nearly anywhere: Indeed, recent groundwater depletion in California’s Central Valley is nearly twice as intense as in the Mississippi Embayment (the geological system that contains the Delta’s aquifers). This summer, when NASA examined data about shifts in the planet’s gravitational field, they found alarming news: 21 of the world’s 37 largest aquifers are now running a deficit, with 13 categorized as particularly concerning. In dry, developing regions of the world like North Africa and Pakistan, groundwater crises are coming fast.

But it’s in the water-rich Mississippi Delta that the extent of this problem is fully revealed. Parts of the Delta average nearly 55 inches of rain each year; the Mississippi River, the largest river on the continent and one of the largest in the world, slips along the region’s flank. When planters first arrived in the Mississippi Delta soon after the Civil War, the big concern was how to get the water out of the area: Before the river levees were completed, much of the land would be underwater for half the year.

In recent years the Mississippi Embayment system has been depleted at a faster rate than any aquifer system in the country save one. When you factor in the size of the system, the region is still better off than arid parts of the west, but the depletion is alarming enough that in 2014 Mississippi launched the Delta Sustainable Water Resources Task Force by state executive order. The task force is now gathering data through a metering program and encouraging farmers to apply basic conservation techniques. “We recognize there is a problem,” Whittington said. “It’s just a matter of what’s best to do about it.”

* * *

Sheahan Pumping Station is tucked into a quiet Memphis neighborhood of sprawling, manicured mansions. The building’s architecture suits the city’s history: With clean stone siding and a massive, arched entryway, it looks like a kind of temple in honor of groundwater.

Here, just north of the Delta, water remains a selling point. Over the past few decades, the city has attracted beverage-makers and biotech firms by touting the price, quality, and seemingly endless supply of its pumped groundwater. A water-protection task force in Memphis estimates that 58 trillion gallons sit under Shelby County, stored deep beneath a kind of geological layer cake. In this metaphor, the layers are made up of water-storing sand and gravel separated by layers of less permeable clay. The alluvial aquifer used by farmers is just below the icing. But while it’s easy to access, it’s too high in iron for humans too drink.

The good drinking water is found far deeper, in a layer known to geologists as the Middle Claiborne Aquifer. As of 2005, the city-run pumping stations in Memphis churned out the aquifer’s water at a daily rate of more than187 million gallons.

But whose water is it? Over the objections of the U.S. solicitor general, the Supreme Court is now considering whether to rule on that question. If it does, it will be the first time the high court has made a decision solely about transboundary groundwater—a fact that has water wonks across the world suddenly paying attention to the Mississippi Delta.

To understand the conflict, one must know that a well displaces water not just vertically, but horizontally, too. Picture the shape of bathwater heading down a drain: In a similar way, the decrease in groundwater pressure forms a “cone of depression” over time, deep and narrow at the bottom and wider toward the surface, like a funnel. The cones of depression that originate in Memphis are now large enough to extend across the state line—which means some of the water being pumped through stations like Sheahan once sat beneath Mississippi.

According to the state’s lawyers, between 1965 and 2006 Memphis removed more than 363 billion gallons of water that belonged to Mississippi, up to 20 percent of the city’s water supply. As of 1995, in fact, the city was the largest user of Mississippi’s groundwater, according to the Mississippi Department of Environmental Quality. In 2005 Mississippi sued the city and Memphis Gas, Light, and Water, seeking up to $1.2 billion in damages. According to a report compiled by Mississippi’s legal counsel, the utility company knew of the problem from several papers prepared by its own scientists, but it refused to participate in a cooperative study. A 2000 report sent to the Tennessee state congress by two faculty members at the University of Tennessee supported the claim that the utility company was liable for the water removed from Mississippi.

For a decade now, the case has bounced through the courts, subject to the vagaries and inconsistencies of U.S. water law. Memphis claimed from the start that the suit should be dismissed. The case was appealed to the Fifth District Court, which in 2009 found that the suit must include Tennessee, not just Memphis, because only states have the sovereign power to negotiate over water. And cases between U.S. states can only be tried by the Supreme Court.

* * *

In Mississippi, the aquifers used to be a farmer’s dream: Just tap through a thin layer of silt, sand, and clay, and there it was, water that went 150 feet deep. And surface aquifers recharged quickly, which in theory made them near-limitless resources. In 1929, when wells in Arkansas were showing concerning declines, the Mississippi farmlands were still doing fine. It wasn’t until the 1970s that large-scale irrigation begin, and even then, plenty of surface water was still available.

Nevertheless, by the mid-1980s water levels in the aquifer had dropped below the streambeds of some local rivers—a condition that can drain water out of the rivers, which in turn can damage wildlife habitat, make it harder for communities to manage wastewater, and complicate disputes over water rights. Since then, the number of irrigation permits in the Delta has shot up, from less than 3,000 in the late 1980s to nearly 18,000 today. Over the past two decades, the portion of the Delta experiencing the effects of dewatering rose from 26 percent to 40 percent.

The Delta farmer David Arant Sr. says that when he took over his family’s farm in the 1970s, most of his neighbors would run their pumps almost nonstop.

“In the Delta, with our heat, we go through a drought every summer,” he told me. Though there is plenty of precipitation, most falls in the winter, when the fields are fallow. “If you don’t irrigate, you’re not going to be able to stay in business.”

Rory Doyle

It was a hot morning when we spoke, and Arant and his son, also named David Arant, were taking a break from harvesting soybeans. With only five combined inches of rain in June, July, and August,  this year had been particularly bad. Now it was September, the heat remained, and the there had only been one additional inch of rain that month. “We haven’t had a real rain here since July 4,” David Jr. said. In late October, the U.S. Department of Agriculture named many counties in the Delta drought disaster areas.

The Arant family came to the region 90 years ago, migrating from hill country east of the Delta onto a tract of land that straddles the border of Leflore and Sunflower counties. Like most farms here, theirs is a large-scale operation, focused on the bottom line; most of their soybeans, corn, and rice are sold as commodity crops. But, unusual for this region, they also mill some rice for small-scale local distribution and are seeking organic certification for some crops.

And they are thinking carefully about water. In the past, farmers relied on visual signs and the gut sense to determine their water-management strategies, but the Arants have installed monitors on their pumps, among other technological solutions. They’ve also seen particular benefits from soil-moisture indictors, electronic probes that quantify the water in their soil. With these techniques, “you can stretch it out,” David Jr. says—even when the soybeans sit dry for a week, which might set other farmers in a panic, he can verify that his plants still have enough water.

Technology is not always cheap; while a simple hand-read moisture sensor costs only $150, a more robust system—with data sent to a farmer’s phone, technical support, and so on—goes up to $3,000 per sensor. But Jason Krutz, an agricultural researcher with Mississippi State University, says that he can use one sensor to make effective decisions for 80 to 160 acres of land. And because more water efficiency means less money spent on fuel to pump water, he says, such tools pay for themselves in a few years.

Krutz works in Stoneville, 45 minutes south of the Arant farm, at the state’s largest agricultural-research center. The center houses the work of three dozen Mississippi State Ph.D.s, and the U.S. Department of Agriculture has another 60 researchers and support staff on campus.

“In the middle of a cotton field in the middle of nowhere, it’s like the Silicon Valley of agricultural research,” Krutz said. He was driving past experimental rice fields, where the flood water—typically kept four inches deep atop a rice field—had been left to recede until it was up to a foot underground. This would make a conventional farmer squirm, Krutz told me. But his preliminary research shows that even if the water is four inches underground, rice can still thrive.

Eighty percent of the Delta’s farmland is irrigated by furrows, or ditches that run alongside the mounded rows of dirt that hold the plants. It’s notoriously inefficient; much of the water runs off the edge of the field or soaks too deep into the ground to be captured by a plant’s roots. There are now free computer programs that let farmers calculate exactly how large a hole they should cut in each portion of irrigation piping in order to ensure each section of a field is watered at a consistent pace. For another $3,000, a farmer can buy a “surge valve,” which irrigates in quick bursts that cover the field more efficiently than one steady stream.

With such technologies, Krutz has been able to produce a typical soybean yield with 30 percent less water. For corn, he’s discovered, water use can be cut by nearly 50 percent. “We’re talking about potentially doubling the span of the aquifer,” Krutz says. But that requires farmers to take the time and expense of learning new techniques—including mastering new technology, a challenge for many in an aging industry. “Basically none of our producers are using these three tools in concert,” Krutz says, “in Mississippi, or anywhere in the mid-South.”

The state’s water task force has surveyed farmers twice, in 2013 and 2015, and while Whittington couldn’t provide precise numbers, she said she’s seen increased awareness of the problem. To add some motivation, the task force is working with the National Resources Conservation Service, a part of the U.S. Department of Agriculture, to distribute $5 million in incentives to farmers who  implement water-saving techniques.

But even if farmers change their practices, Krutz worries that the cost-savings offered by conservation will lead to the cultivation of more land, negating any reduction in water use. More than a million acres in the region sit fallow but could be farmed if the price is right. “I don’t want to call it a Band-Aid, but conservation alone is not going to take care of this aquifer problem,” Krutz said.

The task force is already discussing other solutions—the U.S. Army Corps of Engineers, for example, is working on a $3-million study of the possibility of diverting the Yazoo River, which runs along the region’s edge, into the central Delta. It is even possible, Whittington told me, that the state will pump water out of the Mississippi River and back into the aquifer, reversing the longstanding direction in which locals have moved the water.

This practice, called artificial recharge, has become a popular water-storage method across the country. The alluvial aquifer might seem like an unlikely candidate for such a technique, since it naturally recharges at a fast rate. But local use is so high that what happens naturally matters little now.

The models of the aquifer can’t indicate how much longer the Delta has. “We’re still collecting data,” Whittington told me. “We’re gathering information to really better understand what the status quo is before we make any major decisions about what needs to be done …. It’s pressing, but we have an opportunity.”

* * *

Imagine drilling a hole 200 years ago, and, like magic, finding water bubbling up. Is it any surprise that groundwater was once considered mystical? In arid regions of the country, some landowners today still hire water dowsers, who use sticks and other charms to find an underground water source, rather than having geologists pinpoint where to drill.

Early law was tainted by this awe. Since the extent of underground water seemed hard to ascertain, the simplest policy was just to grant ownership of “percolating waters” to the owner of the land on which they are found. This was the “rule of capture:” In essence, if you pump it, it’s yours.

Some courts cited these early precedents as late as 1997, said Gabriel Eckstein, a professor at the Texas A&M University School of Law who specializes in water issues—despite the fact that the later part of the 19th century had seen great leaps in hydrological knowledge. “The courts said, ‘We realize we know a lot more about groundwater today, and that this is not valid, but we’re going to stick with it because we’ve done it this way for 100 years.’ Their argument was that it’s created property rights—even though it was based on bad science—and they’re ingrained.”

Legislatures did realize that limits had to be set; in some states, the rule of capture was eventually supplemented by the idea of “reasonable use.” A landowner could not, for example, pump water and sell it off to someone else; reasonable use requires that groundwater must be used on the land right above it. Some laws went further still, requiring that the water be used in ways that balanced society’s competing needs; given the difficulties of observing and measuring groundwater, proponents of these laws argued, it was never easy to declare that someone’s use was unreasonable. Still, by 1934, as drilling was hitting its prime, a legal encyclopedia declared reasonable use the “American rule.”

When it comes to water policy, though, there is never one simple, consistent rule. Legal scholars divide current state policies into five distinct doctrines. “It’s a hodgepodge,” Eckstein told me. “There are a number of folks who have tried to do nationwide surveys and analyses across regions. I tried to 10 years ago, and I gave up in frustration.” As groundwater becomes an increasingly important issue, he’s picking up that research again now.

* * *

In 2010, the Supreme Court denied Mississippi’s request for appeal. All of the case law cited in previous decisions had been about surface water, or aquifers directly connected to surface rivers and lakes. So when the Supreme Court declined to hear Mississippi’s case, some legal scholars interpreted it as a tacit suggestion that the same doctrines and precedents would apply to groundwater.

Surface water that crosses state boundaries can be divvied up by act of Congress, or, more commonly, by a compact between states (which also must be approved by Congress). When neither exists, it is up to the Supreme Court to decide what constitutes an “equitable apportionment” of the transboundary water.

After the court’s dismissal, Mississippi and Tennessee—along with Arkansas, which also lies atop portions of the aquifer—began to study the water and discuss its shared use. But Mississippi was unhappy with the pace of discussions. The state filed a new motion with the Supreme Court in June 2014, claiming that Tennessee declined to negotiate a settlement.

And now Mississippi is suing again, seeking $615 million. The 300-page motion never asks for equitable apportionment by the Court. Instead, it argues that the aquifer shouldn’t be treated like surface water at all. The water isn’t just a lake that sit underground, in other words; the water is trapped in rock, and moves very slowly. Turn on a tap in Memphis, and the water in your glass has likely not seen the earth’s surface for two or three millennia. If left undisturbed by our meddling, the water in the Middle Claiborne would move as little as an inch per day, or 30 feet each year. Even a glacier moves faster.

Because the water below is static, Mississippi argues, what’s below Mississippi’s surface is Mississippi’s, and what’s below Tennessee is Tennessee’s. The groundwater, the state says, has “resided” in Mississippi for “thousands of years.”

But the models of this aquifer remain imprecise. A few years ago, after looking at historical documentation of local wells, researchers at the University of Memphis concluded that water actually naturally moves from Mississippi into Tennessee. Brian Waldron, one of the researchers, told the Memphis Commercial Appeal that he believed the previous science had been incorrect, and that drilling in Mississippi was actually capturing water that “would be moving into Tennessee.” (Waldron is the head of the university’s Groundwater Institute, which receives funding from Memphis Gas, Light, and Water, but he told the paper that this research was completed independently of that institute. Waldron did not respond to emailed requests for comment.)

More troubling, perhaps, is that Mississippi’s claims run counter to the recommendations of most scientists and water-policy experts. The dream for years has been a “conjunctive” system of water management—one that treats all of water as “one water.” In such a system, surface water and groundwater are both viewed as parts of a single, continuous system.

“In many states the two quantities are kept separate and never the twain shall meet, which is ridiculous,” Michael Campana, a professor of hydrogeology at Oregon State University and the technical director of the American Water Resources Association, told me. Like Eckstein, he said that most people involved in the debate over groundwater law understand how ridiculous the current policy is—but that the path to new legislation is difficult. Water rights are considered property rights, which are protected by both the U.S. and state constitutions. That leaves courts wary of entering decisions and legislatures wary of enacting new laws.

This past June, after considering various motions and responses from both Mississippi and the defendants—Memphis Gas, Light, and Power; the city of Memphis; and now the state of Tennessee—the Supreme Court granted leave to Mississippi to file a bill of complaint. Responding briefs have been filed through the summer and fall. The Court, it seems, remains open to the possibility that a state can claim absolute ownership of groundwater.

“From the standpoint of a water wonk like myself, this is absolutely fascinating because of the implications it has,” Campana said. “I know a number of my colleagues feel the same.” When one of his friends found out that the Supreme Court might hear the case, Campana said, he “went absolutely nuts.”

A ruling that affirms Mississippi’s claims—that a state can own water—would be a sharp reversal of modern jurisprudence and policy. (Even Mississippi itself, when regulating groundwater use within its boundaries, does not consider overlying property as a signal of ownership). If, alternatively, the case prompts the states to enter into a deal that lays out how the water will be shared, it will be the first such agreement to deal solely with groundwater.

That could have a ripple effect on state groundwater policy across the nation. “The apportionment rule is based on equity, and that means fairness,” Eckstein said. “To judge that, they put a lot of emphasis on efficiency and economics: ‘Are you efficient with that water use? Are you putting it to good economic use?’ You need a whole bunch of information to figure that out.” The Court, if it takes the case, will need to appoint a “special master” to sift through the science and the facts of the case; that appointee’s answers to certain questions will shape how future courts analyze the use of groundwater that crosses state lines.

Rory Doyle

The Supreme Court’s decisions on water policy have international implications, too. “Who owns the Rio Grande? Who owns the Nile River? That is in part based on U.S. Supreme Court doctrine,” Eckstein said. “International law has borrowed from equitable apportionment.” The same question that lies at the core of the Mississippi case—whether equitable apportionment applies to transboundary aquifers—is a “big debate” in the international community, he added.

Eckstein helped the United Nations draft articles about international transboundary aquifers, which were published in 2008. These remain only a proposal, not international law. A 2011 agreement about the management of the Guarani Aquifer in South America was based on these principles, but the agreement has not yet been ratified by all of the countries involved. That kind of hesitation is typical of international groundwater management. “There’s a no-pumping rule on one aquifer between Sonora and Arizona,” Eckstein said. “Then the next paragraph says it’s pending the development of a border-wide agreement, and this was in 1973. There’s been nothing else.”

There’s currently  only one true international treaty about transboundary groundwater management: an agreement between France and Switzerland that sets pumping formulas and explicitly addresses artificial recharge for a shared aquifer. By contrast, Eckstein notes, there have been more than 400 international agreements over rivers and lakes over the past 200 years.

As a result, groundwater pumping is a free-for-all in many places—meaning  those with the money to do so are free to just pump away, taking all the water they can. Before too long, aquifers across the world, like the Mississippi Embayment’s surface aquifer, will begin to show their limits. And if that happens without clear international policies in place, the potential for conflict is high. Since international lawyers often take their cues from U.S. law, a Supreme Court decision could set important precedent, creating what a NASA scientist, writing in Nature Climate Change, called “the global civil and policy infrastructure required to peaceably share groundwater across political boundaries.”

* * *

Memphis, meanwhile, is in an enviable position. While the alluvial aquifer is draining away, water in the deep aquifer remains plentiful—so much so that a report complied by Mississippi’s lawyers references internal Memphis documents predicting“the possibility of barges and tank trucks of pure Memphis groundwater leaving town headed for distant cities to be sold for profit.” The same report says that the MDEQ has also been approached by a “private concern” about the possibility of shipping groundwater stores to arid countries.

If there is a lesson from the surface aquifers in Mississippi, though, it’s that no resource is as infinite as it seems. Elsewhere in the mid-South, where the sand composition of the aquifer differs just slightly, the Middle Claiborne has already shown alarming declines.

“It’s a big deal,” Eckstein says of groundwater overuse. “You can tie this to climate change; you can tie this to so many things. It’s really frustrating the degree to which people aren’t paying attention.”

Whatever the Supreme Court’s decision, this case could be a chance to wipe away the old mysticism shrouding contemporary ideas of groundwater. “[I hope] the states will have the wisdom to go back and say, ‘We don’t want to do this again,’” Campana said. “‘This is costly and doesn’t get us anywhere. Let’s sit down and work this out, let’s come to an agreement about how we should best manage the system here.’” But given the big questions involved, he added, it won’t be easy. And if Mississippi’s claims are dismissed again, Memphis will have little motivation to negotiate.

The city’s residents live in a world far removed from yellow fever and muddy water. In their lifetime, a turn of the tap has always yielded water, clean and sweet. It would be easy to assume that Memphis was simply water-blessed. And like a symbol of that blessing, Sheahan Station, that stone temple to water, rises calmly over its green lawn. The municipal wells, more than 175 spread across 10 well fields, send water to a city where a million people drink and wash and bathe—never thinking about how as the water flows, the ground is shifting beneath their feet.