Well Yield Test: How to Find What a Borehole Can Really Give
Author
Hisham Abdalla
Date Published

Disclaimer: Research and analysis by the engineering team. Sources referenced below.
A well yield test pumps a borehole at known rates while you record the water level, and the record tells you how much the well can give without the level falling to the pump. Three tests do the work. A step test shows how drawdown grows as the pumping rate rises, a constant rate test shows whether the level settles, and a recovery test shows how the level comes back once pumping stops. For a solar pump the result sets the highest flow the pump may be allowed to reach.
Our guide to solar borehole pumping explains why the borehole is the constraint. This article covers how each test is run, how to read the numbers, and how they become a pump specification.
The Words You Need First
Five terms recur in every test report.
Static water level. The depth to water, measured from a fixed point such as the top of the casing, with the pump off and the level at rest. It moves with the seasons, so measure it again at the end of the dry season.
Pumping water level. The depth to water once pumping has run long enough to settle at a given rate.
Drawdown. The fall in water level at a stated pumping rate, in metres: the pumping level minus the static level.
Specific capacity. Yield divided by drawdown, in m3 per hour per metre. It falls as the rate rises, so compare two wells at the same rate.
Safe yield. The rate you are prepared to pump every day for years. Nobody measures it directly: you set it below what the test achieved.
The international standard for these tests is ISO 14686:2003, on pumping tests for water wells. ISO lists it as Edition 1, published in July 2003 and confirmed in 2022, with a second edition now at draft stage. The text is paywalled, so this article follows open field guides.
What You Need on Site
Five things have to be in place before the pump starts.
A test pump bigger than the well. The test pump must be able to pump at least the yield you expect, according to the solar guide from Water Mission. Set it deep, with the intake a few metres below the lowest level you expect during the test.
A way to set and measure the rate. A gate valve on the discharge sets the flow, and a flow meter, weir tank, or bucket and stopwatch measures it. The ICRC's field guide advises trying the valve settings the day before and noting the turns needed for each rate.
A dipper. An electric probe on a graduated tape that beeps when it touches water. Tape stretch causes errors, so use the same dipper and the same datum for the whole test.
A discharge pipe. The water must be carried far enough from the borehole that it cannot soak back in and change the levels you are measuring, as both the ICRC and Water Mission stress.
A foot valve. For the recovery test, the rising main needs a non-return valve at its base. Without one, the water in the pipe falls back into the borehole when the pump stops and spoils the readings.
The ICRC also says the level must be at rest before a test starts, so never test on a day the borehole is being drilled or developed.

The three pumping tests, how each is run and what it shows. Source: ICRC, Practical Guidelines for Test Pumping in Water Wells, 2011; Water Mission, Solar Powered Water Systems, 2021.
The Step Drawdown Test
A step test starts at a low rate. Hold it for a fixed time, then raise the rate without switching off, and repeat. The ICRC describes four or five steps of equal length, commonly 60 to 120 minutes each. Water Mission divides the range of expected flows into four to six equal increments and holds each for at least 60 minutes. The last step should reach or pass the rate you intend to use in service.
Record the level every minute for the first 15 minutes of each step and every 5 minutes after that, as Water Mission specifies, and measure the flow at each reading. If the flow drifts by more than 10 percent, adjust the valve gently, because over-correcting makes it worse.
The ICRC analyses the result with Jacob's equation for drawdown: s = BQ + CQ2, where s is drawdown, Q is the pumping rate, and B and C are constants for that well. Dividing by Q gives s/Q = B + CQ, a straight line. Take the drawdown at the very end of each step, plot s/Q against Q, fit a line, and read B from the intercept and C from the slope.
With Q in m3/h and s in metres, B is in metres per m3/h and C in metres per (m3/h) squared. Work in litres a second and both numbers change, so write the units beside them.
B is the part of drawdown that grows in proportion to the rate. C is the part that grows with its square, where turbulent flow near the screen and the pump intake shows up. The ICRC warns that splitting the two into aquifer loss and well loss is less clean than it looks, so use B and C to compare tests.
Take a hypothetical well with a static level of 28.0 m. Four steps of 60 minutes give these drawdowns at the end of each step.
At 3 m3/h the drawdown is 2.2 m. At 6 m3/h it is 5.0 m. At 9 m3/h it is 8.6 m, and at 12 m3/h it is 13.0 m. Specific capacity, the rate divided by drawdown, falls from 3 / 2.2 = 1.36 m3/h per metre at the first step through 1.20 and 1.05 to 12 / 13.0 = 0.92 at the last, because the well gets less productive as you push it. A best-fit line through s/Q gives B = 0.61 and C = 0.039, and those two constants return all four drawdowns to within 0.1 m.
If you repeat the test later and B has not moved but C has risen, the well itself has deteriorated, probably because the screen is clogging. If B has shifted and C has not, the aquifer has changed. A step test is a poor predictor of long-term behaviour, though, and the ICRC says plainly to use a constant rate test for that.

A hypothetical step test, four steps of 60 minutes, showing specific capacity falling as the rate rises. Q is in m3/h and s in metres. Source: worked example built on Jacob's equation as described in the ICRC test pumping guide, 2011.
The Constant Rate Test
A constant rate test holds one rate for long enough to see whether the level settles. It is the test that decides safe yield.
Choose the rate from the step test. The ICRC says the usual choice is the intended operating rate, though some hydrogeologists add 25 to 50 percent. In our hypothetical the owner hopes to run at up to 8 m3/h, so a test at 10 m3/h is 25 percent above that.
Duration comes next. Water Mission treats 24 hours as the minimum for a full yield test. The ICRC says one or two days suit a small or medium borehole, and one or two weeks are common for a large one serving a big population. Where 24 hours is impossible, Water Mission accepts a test as long as the daily pumping hours, 7 to 8 hours for solar, while warning that an hour or two of steady flow does not show a well can sustain it. We would shorten it only if forced, and then design to the lower end of the safe yield range below.
Season matters too. The ICRC and Water Mission both favour testing when yield and levels are at their lowest, which usually means the end of the dry season. Keep this test on a separate day from the step test, with the level back at rest.
The ICRC's reading schedule is every 30 seconds for the first 10 minutes, every minute until 30 minutes, and every 5 minutes until 2 hours. After that, read it every 30 minutes or every hour, depending on how fast it is still falling.
Plot drawdown against the logarithm of time as you go. A straight line is the normal shape. The ICRC reads a line that steepens, with drawdown growing faster, as a sign the aquifer is limited, and a sudden jump as a possible dewatered fracture, which is a serious warning. A line that eases off suggests the aquifer is gaining water from somewhere.
At 10 m3/h the step curve predicts 0.61 x 10 + 0.039 x 100 = 10.0 m of drawdown after the first hour. Suppose drawdown reaches 10.9 m by hour 24 and eases off over the last six hours. That well is coping. A level still falling steeply at hour 24 means the rate is too high, however much water came out.
The Recovery Test
Switch off, and keep measuring. Start the stopwatch the moment the pump stops and use the same reading schedule as at the start of pumping. Do not pull the pump while this test is running, because removing it changes the water level.
The ICRC points out that switching a pump off is a cleaner step than starting it, and that recovery data smooth out small changes in pumping rate. It says the test should ideally last as long as the pumping did, though cost often cuts it short.
Water Mission stops when the level is back at the static level or within 10 percent of it, and treats a well that misses that mark as inconclusive. On a 28.0 m static level, 10 percent is 2.8 m, which is generous against 10.9 m of drawdown, so we would keep reading past that point if the crew can stay. A level that stops well short is a warning in itself: one explanation the ICRC gives is an aquifer of limited extent with no recharge.
Reading the Results: Safe Yield and Pumping Level
Three numbers come out of the tests: a safe yield, a daily volume and a pumping level.
Safe yield. No formula gives this. Water Mission notes that governing bodies commonly hold a new solar system to 80 to 90 percent of the yield test result, and some to as little as 60 percent. That is a permit convention, not a property of the aquifer. Use your regulator's figure where there is one, and go lower if the level did not quite settle or the test was short.
In the hypothetical, 10 m3/h held for 24 hours, and 80 percent of that is 8 m3/h, which confirms the rate the owner hoped for. Testing 25 percent high and taking 80 percent are the same margin seen from two ends.
Daily volume. A pump limited to 8 m3/h at full sun delivers about 8 x 5.5 = 44 m3 in a month with 5.5 peak sun hours. If the farm or village needs 60 m3 a day, a bigger pump will not help, because its extra flow would pass the safe yield. The answer is lower demand, a second source, or more hours at 8 m3/h. A larger array buys some of those hours by reaching the cap earlier and holding it later: at the 7 to 8 hours of daily running Water Mission assumes for solar, 8 m3/h gives 56 to 64 m3. More than that needs a generator or grid supply.
Pumping level. Take the end-of-dry-season static level, add the drawdown at the maximum rate, and add the creep seen in the long test. Drawdown at 8 m3/h is 0.61 x 8 + 0.039 x 64 = 7.4 m. The design pumping level is 28.0 + 7.4 + 0.9 = 36.3 m. Set the pump intake below that with a margin of several metres.
That level feeds straight into the head. Suppose the delivery point is 10 m above ground, with 3 m of friction. A designer who uses the static level works with 28 + 10 + 3 = 41 m of head. The real head is 36.3 + 10 + 3 = 49.3 m. For the same energy, volume falls in proportion to head, so the array designed for 41 m delivers 41 / 49.3 = 83 percent of its design volume, a shortfall of 17 percent. A centrifugal pump can lose more than that, because the extra head moves it back along its curve. Our pump sizing guide shows what head does to array size, and the design method puts it in sequence.

How hypothetical test results become pump limits and design head. Safe yield uses 80 percent of the tested rate, a common permit limit and not a law of the aquifer. Source: Water Mission, Solar Powered Water Systems, 2021 (safe yield range); MIMAH engineering practice.
What Is Different About a Solar Pump
A solar pump does not run at one rate. It starts slowly at dawn, reaches its highest flow near midday and tails off, and then the well rests for 16 hours or more.
The step test's drawdown curve gives the pumping level at each flow through the day. The rest period may help the well, but do not count on it without evidence. A 24-hour test at 10 m3/h takes out 240 m3, more than five times the 44 m3 of a sunny day, which gives margin in the short term and proves nothing about years of daily draw. The ICRC allows a test either at the rate the pump delivers while switched on or at the long-term average rate, and says the average is better when sustainability is the question.
Fix the maximum with the equipment. A pump that cannot exceed the safe yield cannot pull the level below the design pumping level, whatever the weather does. Capping the controller's top frequency does the same job on an existing pump. The motor's cooling flow at low speed and the dry-run protection still need checking, as set out in our borehole guide.
What Yields Look Like in Practice
The British Geological Survey's Africa Groundwater Atlas gives typical yields in the Ilo/Gundumi Formation of Nigeria's Sokoto Basin of 2 to 8 litres a second, which is 7 to 29 m3/h, and says basement aquifers tend not to be high yielding. Water Mission notes that two boreholes of the same depth and diameter can behave differently, so a test is the only way to know which one you have.
A test shows what the well delivers now. It cannot show whether recharge will sustain years of pumping, and the GIZ and FAO Toolbox warns of depletion where abstraction keeps exceeding natural recharge. That is why you keep measuring the level after commissioning.
Test First, Then Order the Pump
Before you buy anything, commission a step test, a constant rate test and a recovery test in the dry season, and keep the record sheets. Then repeat the step test each year and compare B and C. A rising C is an early sign of a clogging screen, and sand is what wears out pump ends, as our repair guide explains. The yield also shapes the price, as our article on solar pump cost shows.
If you have a borehole to equip or a test report to interpret, our renewable energy team can read your data and tell you what the pump may safely deliver. Contact us with your static level, step results and the daily volume you need.
