A Thousand Little Endings

Future Tense is The Docket's speculative fiction section. Dar Koehn is a fictional character; the aquifer she drills is not.

Hugoton, Kansas — October 2030

Darlene Koehn's shop sits on a gravel lot south of Hugoton, in Stevens County, Kansas, the southwest corner of the state where the Ogallala Aquifer has lost more water than anywhere else in the High Plains.¹ A mud-rotary rig on a flatbed trailer is parked next to a shipping container she uses for parts storage. Inside, driller's logs fill three filing cabinets and most of a folding table. The oldest is from 1990, her first solo well. The newest is from last week, a domestic well for a family outside Moscow whose old well started pumping sand in August.

Koehn, 66, learned the trade from her father, who started drilling irrigation wells in the 1960s. She has drilled more than 800 wells across five counties. In the early years, most of her work was big irrigation wells: 16-inch casing, 800 gallons a minute, the kind of well that turned arid prairie into corn and cattle country. Now she drills monitoring wells for the Kansas Geological Survey, rehabilitates old wells losing yield, deepens domestic wells for small towns, and plugs and abandons wells that no longer produce.² She estimates that a third of the wells she drilled in the 1990s are now either abandoned or producing at a fraction of their original capacity.

She agreed to talk about what the ground feels like when you've been reading it for forty years and it starts giving you different answers.

You've said you can tell what formation you're in before the cuttings come back up. What does that actually feel like?

Dar: Okay. You're running a mud-rotary rig. Bit's turning, mud's circulating, you've got your hands on the controls. The rate you're penetrating changes. The torque changes. When you go from clay into sand, the bit drops a little. Not falls, just eases. Like the ground lets go. Cemented caliche, you feel resistance, the rig seats differently. Gravel shifts the sound. After enough years you don't think about it. Your hands just know you've changed zones.

The cuttings confirm it when they come up in the mud, but by then I've usually already adjusted.

When did the bit start telling you something different?

Dar: Gradually. Which is the worst way for something to go wrong, because you keep thinking you're just having a bad day.

I'd say mid-2010s. I'd drill a spot I knew. I'd drilled the quarter-section next to it ten years before. The sand would feel right. Same formation, same depth, same texture in the cuttings. But the yield test would come back short. Not dry. Just less.

At first you blame the well. Maybe the screen's not set right. You fiddle with it. But after the third or fourth time, you start thinking maybe the sand isn't lying to you, exactly. It's just not holding what it used to hold.³

What do you mean, the sand isn't lying?

Dar: The formation is still there. The geology hasn't changed. Same sand and gravel the Ogallala's always been. What's changed is how much of it is saturated. You drill into a zone that feels like a productive aquifer through the string, but the water table's dropped below half the interval. You've got twenty feet of beautiful sand sitting above the water and twenty feet below it, and the twenty below is what you're actually pumping from. Used to be forty.⁴

My logs from 1998 and my logs from last month, same spot, same formations, same depths. The only number that's different is where the water starts. And that number's the one that matters.

How much has it moved in your territory?

Dar: In Stevens County? Average is down over a hundred feet since before irrigation started.¹ But averages are useless to me. I've got wells three hundred yards apart where one's still producing and the other pumped sand two years ago. The Ogallala's not a bathtub. It's more like a sponge made by a drunk person. Some parts hold water, some parts don't, and the parts that do aren't where you'd necessarily guess.⁵

You mentioned you now do a lot of well abandonment. What's that like?

Dar: Plugging a well I drilled is a specific feeling I do not recommend.

Can you say more?

Dar: I'd rather not. [pause] No, I'll tell you. I drilled a well for the Grabers in '95. Big irrigation well, beautiful gravel pack, pumped clean from day one. That well ran their whole pivot for twenty years. Last spring I went back out to plug it per code because it went dry in '28 and they'd converted to dryland wheat.⁶

I pulled the pump. Looked down the casing with a light. The screen I set thirty-five years ago was still perfect. The gravel pack was still seated. Everything I built was exactly right. There just wasn't any water behind it anymore.

You also drill monitoring wells for the KGS measurement campaign. What's that like, measuring decline in wells you built?

Dar: Ironic, is what it is. Every January they come out with their steel tape and measure water levels in about fourteen hundred wells across western Kansas.⁷ Some of those are wells I drilled. I built the hole, and now the state sends somebody to stick a tape down it to see how much worse it's getting.

I looked at the card for one of my wells last winter. When I drilled it in 2002, static water level was at 190 feet. January 2030, it was at 242. Fifty-two feet in twenty-eight years. And that well's in a good area.

The KGS database, the GMD monitoring data, satellite tools. Do those help you?

Dar: They tell me the average. The average is useful the way a weather forecast is useful: right about the region, wrong about your backyard.

I check the database before I bid a job now. I didn't used to. I pull up the township trend, then go out and check my logs from nearby wells. I use both. But when they disagree, and they do, I still trust my read first.

The problem is my read is wrong more often than it used to be. Not a lot more. But enough that I notice.

How do you handle being wrong?

Dar: Same way I handle everything. I write it in the log. [laughs]

Honestly, it's humbling. I spent forty years building a map in my head of what's under this ground. Every well I drilled added to it. That map is still mostly right. Mostly. But "mostly" is new. It used to be "almost always." I can't tell you in advance which well is going to be the one where my map is off. That's the part that gets me.

Do you think someone learning the trade now could build the kind of knowledge you have?

Dar: They could build knowledge. I don't know if it would be the same kind.

Mine was built on thirty years of conditions that were stable enough to learn from. The water table moved, but slowly, and in ways consistent enough that your experience accumulated into something reliable. I'm not sure that's true anymore. The ground's still teachable, but the lessons have a shorter shelf life.

My apprentice, he's sharp. Reads the rig well. But I've caught myself telling him things and then adding, "That used to be true." Which is a hell of a thing to teach somebody.

What's the hardest part of your work right now?

Dar: Telling people what I find.

I drill a test hole for a family that just bought forty acres and wants a well. I pull up the cuttings and look at the formation and I know this is marginal. Might produce for five years, might produce for fifteen, might go dry next summer. And they're standing right there, and they've already bought the land.

I'm not a hydrologist. I'm just the person who put a hole in the ground and can tell you what I found. And what I found is that there's less down there than there used to be, and I can't promise it'll stay.

Last question. Is the Ogallala going to run out?

Dar: Parts of it already have. Other parts won't in my lifetime, or my apprentice's. It's a thousand little endings in a thousand different spots, each one somebody's farm or somebody's town.⁸

I'll keep drilling until there's nothing left to drill into. Hasn't happened yet. Getting closer, though. I can feel it.

Dar Koehn is a fictional character created for Future Tense. The Ogallala Aquifer's decline, the Kansas Geological Survey monitoring data, and the groundwater management frameworks described are real.

Footnotes

1. Kansas Geological Survey Technical Series 25, 2023 Status of the High Plains Aquifer in Kansas. GMD 3 (southwest Kansas) has experienced average declines of 101 feet since predevelopment, representing a 45% loss in aquifer thickness. https://kgs.ku.edu/sites/kgs/files/files/TechSeriespdfs/TS25.pdf ↩ ↩²

2. Many western Kansas drillers now offer well rehabilitation, deepening, abandonment per state code, and water testing coordination alongside diminishing new-well construction. Water Well Journal, "Aquifers in the United States — Part 2: The Ogallala Aquifer." https://waterwelljournal.com/aquifers-united-states-part-2/ ↩

3. Kansas Geological Survey Open-File Report 2014-2 found that "two wells in different locations with the same saturated thickness can have very different yields, drawdowns, and estimated useable life left in the aquifer." https://www.kgs.ku.edu/Publications/OFR/2014/OFR14_2/ ↩

4. The Kansas Geological Survey's concept of "practical saturated thickness" distinguishes the net thickness of saturated sediments that significantly contribute to well yield from total saturated thickness, a critical distinction as the productive interval thins. Open-File Report 2005-29. https://www.kgs.ku.edu/Hydro/Publications/2005/OFR05_29/ ↩

5. KGS Bulletin 118 describes the Ogallala as "a heterogeneous complex of elastic deposits whose thickness ranges from more than 300 feet to less than 3 feet, and whose texture ranges from coarse gravel to clay." https://www.kgs.ku.edu/Publications/Bulletins/118/02_intro.html ↩

6. David Condos reported in the Wichita Eagle on western Kansas farmers whose wells began "pumping up a muck of sand instead of clear water" as the aquifer declined, forcing conversion to dryland farming. ↩

7. The annual KGS measurement campaign sends crews across western Kansas each January to measure approximately 1,400 wells using a steel highway tape measure. Allison Kite and Kevin Hardy, Stateline. https://www.fastcompany.com/91018954/were-running-out-of-water-how-kansas-farmers-are-adapting-as-a-crucial-aquifer-goes-dry ↩

8. As of 2025, water levels in parts of GMD 3 were declining by 1.5 to 2+ feet annually. Kansas Governor Laura Kelly stated that "some parts of western Kansas don't have groundwater enough to last another 25 years." The Beacon News, January 2025. https://thebeaconnews.org/stories/2025/01/16/laura-kelly-proposes-millions-in-funding-to-protect-ogallala-aquifer/ ↩