There is a monitoring well in Uvalde County, Texas, designated J-27. On June 16, 2026, its ten-day rolling average read 837.1 feet above mean sea level. The number needs to reach 840 before the Edwards Aquifer Authority will lift Stage 5 restrictions. Three feet of water table. It sounds like nothing. For the irrigated farms of the Winter Garden region, it is the difference between operating and not.
Steve Cargil runs Cargil Produce Company out of Uvalde, a vegetable operation his family started in 1953. Bell peppers, cabbage, onions. Multiple farm locations across the region, near Knippa, Bishop, La Pryor. Seventy-three years of putting seed in this ground.
He is also, as Texas Public Radio identified him in April, an Edwards Aquifer Authority board member.
So Cargil votes on the governance framework that administers the pumping restrictions currently cutting his own operation's water by nearly half. He helps set the rules, then drives home and lives under them. He has sat through the hydrological briefings, read the modeling reports, watched the J-27 readings decline month after month. He sits on the board that produces the information, and then has to decide what to do with it on his own land.
The TPR broadcast put the regional picture simply: "Some farmers are selling water rights or scaling back operations; others are leaving the business altogether." Cargil, at least through the spring of 2026, was doing neither.
Fifty-Six Percent of What You Need
Stage 5 is the most severe tier of the Authority's Critical Period Management plan. It mandates a 44% reduction in permitted pumping for every user in Uvalde County authorized to draw more than three acre-feet per year. Municipal, industrial, agricultural. The restriction does not distinguish between the city watering its parks and a farmer trying to get a pepper crop through July.
The Uvalde Pool entered Stage 5 on April 8, 2024, when J-27 dipped to 839.8 feet. It has not left Stage 5 since. Twenty-six months and counting.
The only other time the Uvalde Pool hit Stage 5 was March 2013 through June 2015. Twenty-seven months. That was supposed to be the anomaly. Now the anomaly has a twin, and the gap between them was nine years. If the next gap is five, or three, the question of whether irrigated agriculture has a future in this region begins to answer itself.
What a 44% pumping cut means in practice depends on what you're growing and when. Texas A&M AgriLife Extension estimates cotton needs 21 to 25 inches of water for maximum yields, with daily demand of a quarter to four-tenths of an inch during blooming and boll development. Vegetables have different critical windows but no less absolute ones. Miss the water window during fruit set on a pepper plant and you get zero. The crop doesn't scale down gracefully.
The specific shape of Cargil's 2026 season is not publicly documented. Whether he shifted acreage, fallowed fields, changed his crop mix, or ran deficit irrigation this spring are decisions the public record doesn't capture. What is documented is the research being done practically next door: the Texas A&M AgriLife Research Center at Uvalde has been studying deficit irrigation strategies and found that a 35% reduction in irrigation water did not affect seed cotton yields, though it reduced fiber quality. A separate finding showed deficit irrigation in early growing stages can save 16% of water over five years without reducing lint yield. These are the margins available. Whether they translate to peppers and cabbage under a 44% cut, and whether Cargil is using them, would require his own account.
The CPM plan was designed, in the Authority's own language, to "help slow the rate of decline" in aquifer levels during dry periods. Slow the rate of decline. The words are chosen carefully. They promise neither stabilization nor recovery. A man who sits on the board that approved those words knows exactly what they offer.
The Arithmetic
The Authority's own data contains a fact that doesn't appear in its public drought communications: the Edwards Aquifer's authorized annual pumping cap is 572,000 acre-feet. The ten-year average annual recharge, 2013 through 2022, is 553,230 acre-feet. In an average year, permitted withdrawals exceed natural replenishment by roughly 19,000 acre-feet. And that's before you account for the water flowing out through Comal and San Marcos Springs whether anyone pumps or not.
Authorized annual pumping: 572,000 acre-feet. Ten-year average annual recharge: 553,230 acre-feet. The deficit exists before anyone turns on a pump.
The gauge and the distance chart don't agree. The gauge is the one telling the truth.
Paul Bertetti, the Authority's senior director of aquifer science, told the San Antonio Report in late 2024 that current conditions show:
"Significantly lower water levels than we've experienced for droughts of these magnitudes over the history of the aquifer system."
A conservation district official was more direct: "What we need is long-term, sustained, above-average rainfall, ideally for years."
Years of above-average rainfall in a region where the last four-plus years have been persistently below average. That's the prescription. Cargil has heard this assessment. He may have been in the room when it was delivered. And he went back to his fields.
The April That Belonged to the City
In April 2026, San Antonio recorded its wettest April day in a century. Four and a half inches on April 20. The J-17 index well in Bexar County jumped from 626.7 to 632.8 feet. The San Antonio Pool dropped from Stage 4 to Stage 3, then kept improving. By mid-June it was at Stage 2. Still restricted, but breathing.
The Uvalde Pool didn't move.
Same aquifer system, same April rains, and J-27 sat at 837 like a patient whose fever won't break. The Edwards is a karst aquifer, permeable limestone riddled with conduits and caves. The U.S. Fish and Wildlife Service describes the upside: karst systems can recharge faster than other aquifer types when heavy rain falls. The downside is the mirror image: constant discharge through large springs makes the system "susceptible to drought events." It fills fast and empties fast. But the western portion recharges through different geology, different drainage, and the rain that fell over San Antonio didn't fall over Uvalde in the same volume.
When the San Antonio Express-News or the Austin Chronicle writes about the Edwards Aquifer recovering, they're writing about the San Antonio Pool, where 2.5 million people get their drinking water and where the political attention concentrates. The Uvalde Pool, where the irrigated agriculture is, operates on its own trigger levels, its own index well, its own increasingly separate trajectory. Cargil, sitting on the board, watches both pools. He sees the headlines about recovery. He sees his own well readings.
The World Around the Crop
A vegetable grower in Uvalde County does not operate in isolation. The buyers and packers, the trucking companies, the equipment dealers, the fertilizer suppliers, the seasonal labor networks, the whole web of commerce that makes a farming region function depends on the aggregate. When cotton operations fold and grain fields go fallow, the infrastructure that a pepper and cabbage operation relies on thins around it.
And the cotton operations are folding. The USDA's season-average farm price for upland cotton in 2025-26 was forecast at 60 cents per pound. The University of Georgia's 2026 cotton outlook examined 28 years of data and found that only four showed positive returns. Average annual losses: $113 per production unit across the entire period.
Texas A&M AgriLife made it official in early 2025 when it published budgets showing that sprinkler-irrigated cotton in the relevant Texas district no longer shows a positive return over total costs. Then nitrogen fertilizer prices rose more than 30% between January and April 2026. AgriLife economists stated what everyone in the Winter Garden already knew: "Most producers are not at a break-even point under these costs-versus-crop price conditions. That means more debt or using savings to get through the season."
The Texas Tribune reported on June 9 that Texas farmers from the Panhandle to the Rio Grande Valley are bracing for another dry summer, with nearly half the state in drought. Each operation that quits removes a piece of the regional infrastructure that the remaining operations need. A vegetable grower's last crop might be a good one. He goes under when the world around it can no longer hold together.
Cargil sees this convergence from both sides of the table. Water restrictions, commodity prices, input costs, infrastructure thinning. Each force is documented separately by a different agency, reported by a different beat, tracked by a different set of experts. Nobody is required to add them up. The board member adds them up in the conference room. The farmer adds them up every morning because he has no choice.
What the Models Were Built For
The Authority uses two computer models for planning: one calibrated to 2001-2015 conditions, the other to the 1947-1958 drought of record, the worst in Texas history. Current conditions have approached that historical worst case. The J-17 well in San Antonio came within 12 feet of its all-time low, set in August 1956.
Whether the current drought represents a return to 1950s conditions or something the models weren't built for is the question the Authority and the South Central Climate Adaptation Science Center are trying to answer with downscaled climate projections. That work is ongoing. No final results have been published. The specific question driving it, per the Fish and Wildlife Service, is "how recharge in the semi-arid western portions of the region will be affected by increasing temperatures and variations in precipitation."
The semi-arid western portions. That's Uvalde.
NOAA's Joel Lisonbee, the Southern Plains drought coordinator, told the Texas Water Resources Institute:
"What we need is long-term, sustained, above-average rainfall, ideally for years... [Forecasters] are predicting a very strong El Niño... occurring at a time when the whole Pacific Ocean is unusually warm. This creates some uncertainty around how the weather patterns will respond, and past El Niño events may not be a good indicator of the next one."
The models are being asked to predict a system that may have moved outside the range of conditions they were calibrated against. Which is an honest description of where the science stands. And it means that a board member sitting in the Authority's conference room, looking at the projections, knows that the best available answer to "When does the aquifer recover?" is: we genuinely don't know.
What Cannot Be Disputed
EAA General Manager Roland Ruiz has acknowledged that "as climate change pushes Texas droughts to be longer and more severe," the Authority is looking at supplementing its enforcement approach with incentive programs to encourage conservation during non-drought periods. The word "incentive" carries freight. It means the current framework, the one that triggers restrictions only after the aquifer is already in crisis, may not be adequate for a future where the intervals between crises keep shortening. It also means the institution Cargil serves on is, at least in principle, looking for new tools.
There are counter-possibilities. El Niño could deliver the sustained rainfall the aquifer needs. The Authority could restructure its approach in ways that give irrigated agriculture more room. Cargil himself may be adapting in ways the public record can't see, running deficit irrigation strategies, shifting planting schedules, making the kind of adjustments that a man with 73 years of family knowledge and a board member's data access is uniquely positioned to make. A private climate analytics firm forecast late-season dryness spawning new drought in the Texas region by September, but forecasts are forecasts. The NOAA seasonal outlook offers the possibility of improvement alongside the warning of reintensification.
What cannot be disputed is the position Cargil occupies. Every piece of information was available to him because he helped produce it. He is the institution. He reads the aquifer science, votes on the management plan, watches the index well, and then drives to his fields and decides whether to plant. Every other farmer in the Winter Garden absorbs the restrictions as something imposed from outside. Cargil helped write them. He knows the recharge arithmetic, the modeling uncertainty, the fact sitting in the Authority's own data that permitted pumping exceeds average recharge. He knows the prescription is years of above-average rainfall and that nobody can promise it.
And he farms. Whether that is the deepest kind of practical knowledge, a man who understands something about this land and this aquifer that the models don't capture, or the weight of 73 years making departure unthinkable, or both of those things fused together in a way no outside observer can separate, the public record doesn't say. It is the question the public record leaves standing.
The aquifer doesn't negotiate. It responds to rainfall and geology and physics. J-27 reads 837.1 feet, three feet short of relief, and has read something like it for twenty-six months.
-
Southern Plains drought economics: NOAA's National Integrated Drought Information System estimated that drought cost agriculture across Kansas, Oklahoma, and Texas more than $23 billion from 2020 through 2024, a figure that continues to climb as the sixth year of drought unfolds.
-
Deficit irrigation research at Uvalde: Texas A&M AgriLife's Uvalde research center is actively studying whether early-season water cuts can save 16% of irrigation without reducing yields, findings that take on new urgency under a 44% pumping restriction that exceeds anything the experiments tested.
-
El Niño uncertainty ahead: NOAA drought coordinator Joel Lisonbee warned that forecasters are predicting a very strong El Niño arriving while the entire Pacific Ocean is unusually warm, creating conditions where past El Niño events may not predict what comes next.
-
National irrigated acreage decline: The 2022 USDA Census of Agriculture found that irrigated cropland nationally hit a 30-year low, down roughly 3 million acres from the 2017 record high, a trend the current drought is accelerating in regions like the Winter Garden.