How Do You Measure an Aquifer You Can't See? The Pumping Test

The third and last of the laboratory sequence. The permeameter measured a handful of repacked material; this measures the real aquifer, by disturbing it and watching it respond. It is the only method here that returns two properties instead of one: transmissivity T, and storativity S, which cannot be obtained from a core at all. Built to stand in for a test you will almost certainly never get to run yourself. The kit is laid out and named — submersible pump, gate valve, flow meter, discharge line, water-level dipper, stopwatch, steel tape, printed data sheet — and then the two things that have to be right before the pump starts: a measured datum and a distance, and somewhere for fifteen hundred cubic meters a day of water to go where it will not come back. The whole twenty-four hours then runs as an animation, with the stopwatch, the flow meter being held steady, and the log-spaced readings filling in. Then both standard analyzes. Theis type-curve matching is shown as an animation — the field sheet slides over the fixed type curve with its axes held parallel until the points land, which is the part that is impossible to picture from a book — and then Cooper–Jacob's straight line. Both are worked through to numbers. Then the failures, deliberately demonstrated rather than described. Fitting the straight line through all twenty-two points looks perfectly respectable and gives T only 7 % high, while S comes out 25 % low — an asymmetry the video explains rather than asks you to memorise. Plus a non-constant discharge, boundaries, the u ‹ 0.05 condition and how to check it when T and S are the unknowns, and the recovery test you should run before packing the van. CHAPTERS 0:00 Why fitting all the data is wrong 0:54 The set-up: two wells, and why the second one matters 2:34 The equipment, and the logistics of a 24-hour test 3:42 Before the pump: the datum and the discharge 4:53 Running it: 22 readings across a day 6:57 Looking at it: log-log and semi-log 7:28 The type curve: Theis and the match point 10:48 The straight line: Cooper–Jacob 11:57 Fitting all 22 points, a drifting rate, boundaries 15:28 The recovery: a second test, free 16:02 Comparison, and the link back to K 17:10 Ask the aquifer, then say what you used SUBTITLES English, Español, हिन्दी, 日本語, 한국어 and 中文（简体） — choose one from the CC button. THE NUMBERS ON SCREEN *The test.* Q = 1500 m³/day, r = 60 m, b = 20 m, 22 time–drawdown pairs over 24 h. The dataset was generated from the Theis solution with T = 250 m²/day and S = 2.0 × 10⁻⁴, so the true answer is known and the methods can be scored against it. *Type curve.* Match point W(u) = 1, 1/u = 1, s = 0.477 m, t = 1.04 min → T = 250 m²/day, S = 2.0 × 10⁻⁴. *Cooper–Jacob, late points only* (t ≥ 30 min): Δs = 1.09 m per log cycle → T = 252 m²/day (+0.7 %); t₀ = 1.74 min → S = 1.91 × 10⁻⁴ (−4.7 %). *Cooper–Jacob, all 22 points* (the demonstrated mistake): T = 269 m²/day (+7.4 %), S = 1.50 × 10⁻⁴ (−25.2 %). **K from T**: 252 / 20 = 12.6 m/day, compared on screen with the 14.4 m/day measured on the bench in video 06. The well function W(u) is computed from its series expansion and agrees with published tables (W(1) = 0.21938, W(10⁻⁴) = 8.63322). WHERE IT SITS IN THE SERIES Video **07**, the last of the laboratory sequence and the end of the scale progression: a handful of grains (05), a repacked column (06), a real aquifer (07). Section 20 explicitly compares its field K with the bench K from video 06. #hydrogeology #groundwater #pumpingtest