Fall Drought and Mountain Wells: Why Yields Drop in September and What a Low-Yield Pump Setup Costs

September 8, 2026

Every September the phone starts doing the same thing. Somebody in Edneyville runs a shower and a load of laundry back to back and the taps start spitting air. Somebody on a ridge above Zirconia notices the pressure sags by dinner and is back to normal by breakfast. Nothing broke. The well is simply making less water than it did in April, and the pump has found the bottom of it.

This is the most misdiagnosed problem in the trade, and the misdiagnosis is expensive: a homeowner is told the pump is weak, buys a bigger one, and the bigger pump empties the well faster and burns up. Here is what is actually happening under a Henderson County well in the fall, and what the honest fixes cost.

Why there is less water in September

Start with the rock. There is no sand aquifer under the Blue Ridge. The U.S. Geological Survey’s Ground Water Atlas describes the aquifer here as fractured crystalline bedrock, gneiss, schist, and granite, under a blanket of weathered regolith that can run 150 feet thick in places. Water is stored in that regolith and in cracks in the rock below it, and a well produces only what the cracks it happened to intersect will give. The USGS puts average reported yields in these aquifers at roughly 15 to 20 gallons per minute, with a great many wells far below that, and notes that fractures grow scarcer with depth and that ridge tops and steep slopes carry the thinnest weathered zone and the poorest yields, while broad valleys and draws with more than 50 feet of regolith hold the best wells.

Now add the calendar. Recharge is not spread evenly through the year. Rain that falls in the growing season largely goes back to the sky through leaves before it ever reaches rock; rain that falls in the cool months, when the forest is dormant, soaks in. So water levels in mountain wells drift down all summer and reach their annual low in early autumn. Layer a dry spell on top of that and the effect is immediate. The USGS states it plainly: when rainfall is less than normal for several weeks, months, or years, the flow of streams and rivers declines, water levels in lakes and reservoirs fall, and the depth to water in wells increases.

The last ingredient is demand. In this county, September is when second homes fill up, when the orchards are busy, and when everybody who spent August outside is washing something. More draw, less storage in the rock, and the same pump hanging on the same pipe it has hung on for fifteen years.

How to tell a drying well from a broken pump

The pattern gives it away every time.

If your symptoms sit in the left column, the running dry page is the one to read. If they sit in the right, start with the no water page or the low pressure page, because that is a $150 to $450 part and not a water problem at all.

One warning worth repeating: a pump that is intermittently sucking air is running dry in bursts, and dry running damages the mechanical seal and the motor. Grundfos says so about its own submersibles, and it is true of every brand. A well that spits air daily for six weeks in the fall is how a weather problem becomes a $1,200 to $3,500 pump replacement bill in November.

The five fixes, cheapest first

1. Manage the draw. Free.

Spread laundry across days instead of stacking it. Shorten showers. Stop irrigating in a dry September. Fix every running toilet, because a leaking flapper is roughly 200 gallons a day, which on a 3 gallon per minute well is more than an hour of the well’s entire daily production disappearing quietly into the septic field. On a marginal well, demand management alone is often the whole answer, and it costs nothing but attention.

2. Measure before you spend. $150 to $300.

A diagnostic visit on a low-yield complaint is not a parts hunt. It is three measurements: the standing water level with the pump off, the pumping water level after the pump has run, and the recovery rate afterwards. Those numbers tell you how much usable column sits above the pump, how fast the well refills, and whether there is open hole below the pump worth using. Everything after this step is guesswork without them, and the fee is usually credited toward the repair by many crews.

3. Reset the pump lower, with dry-run protection. Priced with the pull.

If the well has hole below the pump, lowering the pump buys usable water column. It is done on the same trip as a pull, so the cost is the labor and pipe rather than a whole new system. Pair it with dry-run protection so the pump shuts itself down instead of grinding when the level does drop. Some pumps have it built in: Grundfos states that the SQ and SQE watch power consumption and switch the pump off for five minutes when it falls below the configured value, then restart automatically. That is exactly the behavior a low-yield well needs in October. More on that on the Grundfos page.

The other half of this step is correct sizing. On a weak well the right pump is usually a smaller one with more stages set deeper, not a bigger one. North Carolina’s construction rules require the pumping capacity of the pump to be consistent with the yield characteristics of the well, so oversizing is not merely unwise, it is out of step with the standard.

4. Add storage. The real fix for a genuinely weak well.

This is the setup that turns a 2 gallon per minute well into a house that behaves normally, and it is standard practice on low-yield mountain wells. A modest pump fills a holding tank slowly all day, and a booster pump serves the house from that tank at full pressure. The well never gets asked for more than it can give, the pump never runs dry, and the household stops living around the well’s schedule. It costs more than a pressure tank at $600 to $1,500 and much less than drilling, and it is sized to your measured yield and your household, which is why step 2 comes first.

For context on why this matters beyond comfort: federal mortgage rules for FHA-insured houses expect an existing well to deliver three to five gallons per minute, with storage explicitly allowed as an alternative on shared wells. If your well is below that and you ever want to sell, storage is not a luxury.

5. Hydrofracture, then drill. $2,000 to $5,000, then $8,000 to $20,000.

Hydrofracturing injects high-pressure water into an existing well to open the fractures it already touches. In fractured mountain rock it works often enough to be the sensible next step before drilling, and at $2,000 to $5,000 it is a fraction of the cost of a second well. It is not guaranteed, and anyone who guarantees it is selling. North Carolina’s certification rules require a Level A or Level B certified well contractor for it.

If that does not deliver, a new well runs $8,000 to $20,000 in mountain rock, at $20 to $45 per foot plus casing plus the pump system, and siting is most of the outcome: the USGS notes that wells drilled along lineaments and fracture traces consistently outperform randomly placed ones. Henderson County Environmental Health charges $350 for the permit plus a $130 new-well water analysis, and the permits and rules page covers the setbacks a new well has to meet. The drilling page covers the rest.

What not to do

Do not buy a bigger pump. It is the single most common mistake and it makes the problem worse, faster.

Do not keep the system running dry while you decide. Shut the well breaker off between uses. That one step has saved more pumps than any replacement part ever has.

Do not assume it is permanent. A well that is marginal in a dry September is often perfectly adequate by December. Measure now, ration now, and make the capital decision when you have watched the well through a wet season as well as a dry one.

If you are buying a house on a well this fall

Ask for the yield off the Well Construction Record before you sign anything. Henderson County Environmental Health keeps searchable septic and well permit records, and the state requires that record to carry the diameter, depth, and yield of the well. A house shown in October with a well that already runs short is telling you something about every October that follows. The pricing page has every number in this trade, and the running dry page has the decision tree in short form.