Question 2 of 5: Layered Vertical-Flow Pressures, Density-Dependent Flow Direction Across an Aquitard, and Confined Storage Volume
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
National Exams — December 2019 — 18-Geol-A2 Hydrogeology. Three-hour, open-book exam; any non-communicating calculator permitted. Five questions constitute a complete paper and all five are of equal value; clarity and organization of the answers, with work shown in detail, are explicitly graded. Unless stated otherwise, water density is taken as 1000 kg/m³, water viscosity as 0.001 kg/m-sec, and g as 9.81 m/s².
Reference texts: Freeze & Cherry, Groundwater (Prentice-Hall, 1979) — Darcy's law and anisotropic conductivity tensors, the three-point head-gradient method, soil phase relations, layered-medium effective conductivity, freshwater-equivalent head across a density interface, elastic storage, the Theis and Hantush-Jacob (leaky) well equations, Cooper-Jacob straight-line analysis, and slug-test analysis (Cooper-Bredehoeft-Papadopulos); Todd & Mays, Groundwater Hydrology — supplementary well-test methods; EGBC Geoscience Professional Practice Guidelines for assumption-disclosure conventions on open-book calculations.
Question 2: Layered Vertical-Flow Pressures, Density-Dependent Flow Direction Across an Aquitard, and Confined Storage Volume (equal value)
Given. (a) Three horizontal formations, each 25 m thick, as below; total head 180 m at the top of the stack, 80 m at the base. (b) 20 m thick aquitard between a freshwater aquifer above ($\rho=1000\ \text{kg/m}^3$, 12 m of water in the top-screened well) and a saline aquifer below ($\rho=1160\ \text{kg/m}^3$, 30 m of water in the bottom-screened well). (c) $S_s=2.5\times10^{-6}\ \text{m}^{-1}$, $b=35$ m, plan area $2200\ \text{m}\times2000\ \text{m}$, target head decline $\Delta H=1.2$ m.
Formation
Thickness
Hydraulic conductivity
Top
25 m
0.001 m/s
Middle
25 m
0.005 m/s
Bottom
25 m
0.001 m/s
Find. (a) water pressure at the two internal boundaries. (b) the direction of flow across the aquitard under each density assumption. (c) the volume of water pumped.
Approach. Part (a) uses continuity of specific discharge through the three series layers to apportion the 100 m total head loss, then converts each boundary's total head to a pressure head using its elevation. Part (b) compares total head at the two screened points; crossing a density interface requires converting the saline-side reading to a freshwater-equivalent head before the comparison is valid. Part (c) combines specific storativity with aquifer thickness to get storativity, then applies $\Delta V=S\,A\,\Delta H$.
Part (a) — specific discharge through the series stack. Steady vertical flow through layers in series carries the same $q$ through each, so the 100 m head loss apportions as $\sum L_i/K_i$:
$$q=\frac{\Delta H}{\sum L_i/K_i}=\frac{180-80}{25/0.001+25/0.005+25/0.001}=\frac{100}{55{,}000}=\boxed{1.818\times10^{-3}\ \text{m/s}}.$$
Head at each internal boundary. Head loss in a layer is $q\,L_i/K_i$: top layer loses $1.818\times10^{-3}\times25{,}000=45.45$ m, giving $h_{1/2}=180-45.45=\boxed{134.5\ \text{m}}$ at the top/middle boundary; the middle layer then loses $1.818\times10^{-3}\times5{,}000=9.09$ m, giving $h_{2/3}=134.5-9.09=\boxed{125.5\ \text{m}}$ at the middle/bottom boundary (check: $125.5-45.45=80.0$ m at the base, matching the given bottom head).
Pressure at the boundaries. With the base of the stack as datum ($z=0$), the top/middle boundary sits at $z=50$ m and the middle/bottom boundary at $z=25$ m, so the pressure head is $\psi=h-z$:
$$p_{1/2}=\rho_wg(h_{1/2}-50)=1000(9.81)(134.5-50)=\boxed{829\ \text{kPa}},$$
$$p_{2/3}=\rho_wg(h_{2/3}-25)=1000(9.81)(125.5-25)=\boxed{985\ \text{kPa}}.$$
Part (b) — total head at the freshwater well. Taking the bottom of the aquitard as datum ($z=0$, top at $z=20$ m), the freshwater well is screened at the top with a 12 m column: $h_1=z_{\text{top}}+\psi_1=20+12=\boxed{32.0\ \text{m}}$.
Case (i) — saline density assumed 1000 kg/m³. Treating the 30 m saline column as if it were fresh water, $h_2=z_{\text{bot}}+\psi_2=0+30=\boxed{30.0\ \text{m}}$. Since $h_1(32.0)\ >\ h_2(30.0)$, water would appear to flow downward, from the fresh aquifer into the saline aquifer.
Case (ii) — true saline density used. Comparing heads across a density interface requires the freshwater-EQUIVALENT head at the saline point, $h_f=z+(\rho_s/\rho_f)\psi$:
$$h_{f,2}=0+\left(\frac{1160}{1000}\right)(30)=\boxed{34.8\ \text{m}}.$$
Now $h_{f,2}(34.8)\ >\ h_1(32.0)$, so the true flow direction is upward — saline water rising into the freshwater aquifer — the OPPOSITE of the naive answer in case (i).
Check: this reversal is real, not a rounding artifact — the denser saline column stores more pressure per metre of water than an equal column of fresh water, so its 30 m reading corresponds to a higher freshwater-equivalent head than its raw water-level elevation suggests. Ignoring the density correction (case i) gets the cross-aquitard flow direction backwards whenever the denser side's actual head is close to or above the lighter side's.
Part (c) — storativity and pumped volume. $S=S_sb=(2.5\times10^{-6})(35)=8.75\times10^{-5}$; plan area $A=2200\times2000=4.40\times10^6\ \text{m}^2$:
$$\Delta V=S\,A\,\Delta H=(8.75\times10^{-5})(4.40\times10^6)(1.2)=\boxed{462\ \text{m}^3}.$$
Quantity
Result
(a) Pressure at top/middle boundary
829 kPa
(a) Pressure at middle/bottom boundary
985 kPa
(b) Flow direction, case (i) – $\rho_{\text{saline}}$ assumed 1000
Downward (fresh → saline)
(b) Flow direction, case (ii) – true $\rho_{\text{saline}}$=1160