Hydrogeology: Darcy's Law, Aquifers, Groundwater Exploration and Pollution
CSIR NET Earth Sciences Part C geology: Darcy's law, aquifer properties, pumping tests, resistivity exploration, contaminant transport, India's aquifers.
- 32 min
On this page · 15 sections
- Prerequisites
- The Hook: Safe from Germs, Not from Arsenic
- Hydraulic Head and Darcy’s Law, Properly
- Aquifers in Different Rocks
- Well Hydraulics and Pumping Tests
- Groundwater Exploration
- Groundwater Pollution
- Indian Groundwater Regimes
- Memory Tricks
- Don’t Confuse
- Common Mistakes
- Practice
- Quick Revision Table
- Key Takeaways
- What’s Next
Prerequisites
- The Part B foundation: hydrological cycle, porosity and specific yield, aquifer vocabulary, Darcy’s law, the Thiem equation and the Ghyben–Herzberg relation are all in Hydrological Cycle, Water Resources and Management. This page assumes them and goes to Part C depth.
- Electrical resistivity arrays from Mineral Exploration: Methods, Sampling and Evaluation: the same equipment is the hydrogeologist’s main tool.
- Lineaments, drainage density and GIS overlay from Remote Sensing, GIS, Photogrammetry and GPS.
- Eh–pH chemistry of iron, arsenic and other redox-sensitive species: Redox, Eh–pH Diagrams and the Geochemistry of Waters and Aerosols.
The Hook: Safe from Germs, Not from Arsenic
Hydraulic Head and Darcy’s Law, Properly
Head
Groundwater flows from higher to lower hydraulic head, the mechanical energy per unit weight of water:
$$\boxed{h = z + \frac{p}{\rho g}\ \left(+\frac{v^2}{2g}\right)}$$The velocity term is negligible in groundwater (velocities are metres per day, not per second), so \( h = z + \psi \), elevation head plus pressure head. A piezometer (a pipe open only at its base) measures \( h \) at one point.
Two piezometers at one site: the shallow one has its screen at 100 m elevation with a pressure head of 15 m (\( h = 115 \) m); the deep one at 80 m with 30 m (\( h = 110 \) m). Head falls with depth, so flow is downward: this is a recharge area, with vertical gradient \( (115 - 110)/20 = 0.25 \). In a discharge area (valley bottoms, springs, wetlands) head increases with depth and flow is upward; a deep well there may flow at the surface.
Darcy’s law and its limits
$$\boxed{q = -K\,\frac{dh}{dl}} \qquad \boxed{K = \frac{k\,\rho g}{\mu}} \qquad v = \frac{q}{n_e}$$- \( K \) (hydraulic conductivity, m s\(^{-1}\)) depends on fluid and medium; the intrinsic permeability \( k \) (m\(^2\)) on the medium alone. One darcy is about \( 9.87\times 10^{-13} \) m\(^2\). For water at about 20 °C (\( \mu \approx 10^{-3} \) Pa s), \( K = 10^{-4} \) m s\(^{-1}\) corresponds to \( k \approx 1.0\times 10^{-11} \) m\(^2\), about 10 darcy.
- Warm water has lower viscosity and therefore higher \( K \) in the same rock.
- Darcy’s law is linear and needs laminar flow. The Reynolds number for porous media is \( Re = \rho q d/\mu \) with \( d \) a representative grain diameter (often \( d_{10} \)); Darcy’s law holds up to \( Re \) of about 1–10. For \( q = 10^{-4} \) m s\(^{-1}\) and \( d = 1 \) mm, \( Re = 0.1 \): Darcian. In karst conduits, fractured basalt near wells and coarse gravel at high gradients flow becomes non-linear.
- An empirical estimate for clean, uniform sand is Hazen’s formula \( K\ (\text{cm s}^{-1}) \approx C\,d_{10}^2 \) with \( d_{10} \) in cm and \( C \approx 100 \): \( d_{10} = 0.2 \) mm gives \( K \approx 0.04 \) cm s\(^{-1}\) \( = 4\times 10^{-4} \) m s\(^{-1}\).
Heterogeneity and anisotropy: layered aquifers
For horizontal layers of thickness \( b_i \) and conductivity \( K_i \):
$$\boxed{K_h = \frac{\sum K_i b_i}{\sum b_i}} \qquad \boxed{K_v = \frac{\sum b_i}{\sum b_i/K_i}}$$Flow parallel to the layers takes the arithmetic mean (dominated by the most permeable layer); flow across them takes the harmonic mean (dominated by the least permeable layer). Hence \( K_h \ge K_v \) always for layered sequences. Example: 5 m of sand (10 m d\(^{-1}\)), 2 m of clay (0.01 m d\(^{-1}\)) and 3 m of sand (5 m d\(^{-1}\)) give \( K_h = 6.50 \) m d\(^{-1}\) and \( K_v = 10/201.1 = 0.050 \) m d\(^{-1}\): an anisotropy of about 130 from one thin clay bed.
Refraction of flow lines at a boundary between media obeys \( \tan\theta_1/\tan\theta_2 = K_1/K_2 \) (angles from the normal): flow lines bend towards the normal on entering the less permeable medium, so flow crosses clay beds nearly vertically and runs along sand beds nearly horizontally.
Storage and the groundwater flow equation
| Term | Definition | Typical value |
|---|---|---|
| Specific storage \( S_s \) | water released per unit volume of aquifer per unit head decline, by compression of the skeleton and expansion of water: \( S_s = \rho g(\alpha + n\beta) \) | \( 10^{-6} \) to \( 10^{-4} \) m\(^{-1}\) |
| Storativity \( S \) (confined) | \( S = S_s b \), dimensionless | \( 10^{-5} \) to \( 10^{-3} \) |
| Specific yield \( S_y \) (unconfined) | gravity drainage of pores | 0.01 to 0.3 |
| Transmissivity \( T \) | \( T = Kb \) | m\(^2\) d\(^{-1}\) |
Mass conservation with Darcy’s law gives the groundwater flow equation. For a homogeneous, isotropic confined aquifer in two dimensions:
$$\boxed{\frac{\partial^2 h}{\partial x^2} + \frac{\partial^2 h}{\partial y^2} = \frac{S}{T}\,\frac{\partial h}{\partial t}}$$At steady state the right side is zero and the head obeys Laplace’s equation. \( T/S \) is the hydraulic diffusivity, which controls how fast a pressure change spreads: a confined aquifer, with small \( S \), feels a pumping well kilometres away within hours.
Flow nets
Solutions of Laplace’s equation are drawn as flow nets: equipotential lines (constant head) and flow lines crossing at right angles to form curvilinear squares in an isotropic medium. For a total head loss \( H \) across the net with \( N_f \) flow channels and \( N_d \) head drops, the discharge per unit width is
$$\boxed{q' = K H\,\frac{N_f}{N_d}}$$With \( K = 10^{-5} \) m s\(^{-1}\), \( H = 12 \) m, \( N_f = 4 \) and \( N_d = 12 \): \( q' = 4\times 10^{-5} \) m\(^2\) s\(^{-1}\), about 3.5 m\(^3\) per day per metre width. Flow nets are used for seepage under dams and through embankments and for well capture zones.
Aquifers in Different Rocks
| Rock | Porosity type | Behaviour |
|---|---|---|
| Unconsolidated sand and gravel (alluvium, coastal, glacial) | primary, intergranular | the most productive aquifers; layered, often multi-aquifer |
| Sandstone | primary, reduced by cement; secondary in fractures | moderate to good (for example Gondwana and Tertiary sandstones) |
| Limestone and dolomite | secondary, solution enlarged (karst) | extremely heterogeneous; conduit flow, rapid transport, high vulnerability |
| Basalt | vesicles, flow-top breccia, cooling joints, fractures; weathered zones | layered: productive vesicular and fractured units separated by massive units; red boles may confine |
| Granite, gneiss, schist (hard rock) | secondary only: weathered mantle (regolith) and fractures | two-tier system: the weathered zone stores water and the fractured zone below transmits it to wells; yields fall with depth as fractures close |
| Shale, clay | high total porosity, very low \( K \) | aquitards and aquicludes |
Leaky aquifers are confined aquifers that receive water through an overlying aquitard. The leakage factor \( B = \sqrt{T b'/K'} \) (\( b' \), \( K' \) of the aquitard) measures how far from a well leakage becomes important; pumping-test curves in leaky aquifers flatten to a steady state (Hantush–Jacob solution) instead of declining indefinitely as Theis predicts.
Springs occur where the water table meets the surface: depression springs (in a valley cutting the water table), contact springs (above an impermeable bed), fault and fracture springs, karst springs (often large, from conduits), artesian springs from confined aquifers, and thermal springs where deep circulation brings heated water up along faults.
Well Hydraulics and Pumping Tests
Steady state
| Aquifer | Equation |
|---|---|
| Confined (Thiem) | \( Q = \dfrac{2\pi T(s_1 - s_2)}{\ln(r_2/r_1)} \) |
| Unconfined (Dupuit–Thiem) | \( Q = \dfrac{\pi K(h_2^2 - h_1^2)}{\ln(r_2/r_1)} \) |
In the Dupuit form \( h \) is the saturated thickness above the aquifer base; the Dupuit assumptions (horizontal flow, gradient equal to the slope of the water table) break down very close to the well, where a seepage face forms. Example: \( K = 20 \) m d\(^{-1}\), \( h_1 = 18 \) m at 10 m and \( h_2 = 20 \) m at 100 m: \( Q = \pi(20)(400 - 324)/\ln 10 \approx 2{,}070 \) m\(^3\) d\(^{-1}\).
Transient: Theis
For a fully penetrating well pumping at constant \( Q \) from an infinite, homogeneous, confined aquifer:
$$\boxed{s = \frac{Q}{4\pi T}\,W(u)} \qquad \boxed{u = \frac{r^2 S}{4Tt}}$$\( W(u) \) is the well function (the exponential integral \( E_1(u) \)): \( W(u) = -0.5772 - \ln u + u - \frac{u^2}{2\cdot 2!} + \dots \). Values: \( W(1) = 0.219 \), \( W(0.1) = 1.82 \), \( W(0.01) = 4.04 \), \( W(0.001) = 6.33 \). Field drawdown versus time is matched against the type curve \( W(u) \) versus \( 1/u \) on log–log paper to obtain \( T \) and \( S \).
Example: \( Q = 2{,}000 \) m\(^3\) d\(^{-1}\), \( T = 500 \) m\(^2\) d\(^{-1}\), \( S = 2\times 10^{-4} \), \( r = 100 \) m, \( t = 1 \) d: \( u = 10^4\times 2\times 10^{-4}/2000 = 10^{-3} \), \( W = 6.33 \), \( s = (2000/6283)\times 6.33 \approx 2.0 \) m.
Cooper–Jacob straight line
For small \( u \) (commonly \( u < 0.01 \); some texts allow 0.05) the series reduces to its first two terms:
$$\boxed{s = \frac{2.303\,Q}{4\pi T}\log_{10}\frac{2.25\,T t}{r^2 S}}$$Drawdown plotted against \( \log t \) becomes a straight line. If \( \Delta s \) is the drawdown per log cycle of time and \( t_0 \) the intercept at zero drawdown:
$$\boxed{T = \frac{2.303\,Q}{4\pi\,\Delta s}} \qquad \boxed{S = \frac{2.25\,T t_0}{r^2}}$$Example: \( Q = 2{,}500 \) m\(^3\) d\(^{-1}\), \( \Delta s = 0.8 \) m, observation well at 50 m, \( t_0 = 0.5 \) min: \( T = 2.303\times 2500/(4\pi\times 0.8) = 573 \) m\(^2\) d\(^{-1}\) and \( S = 2.25\times 573\times(0.5/1440)/2500 = 1.8\times 10^{-4} \) (a confined aquifer). The same analysis on drawdown versus \( \log r \) at one time (distance–drawdown) uses \( \Delta s \) per log cycle of distance with \( T = 2.303Q/(2\pi\Delta s) \).
Reading departures from Theis
| Late-time behaviour on the log plot | Interpretation |
|---|---|
| Drawdown flattens to a steady value | recharge boundary (river, lake) or leakage through an aquitard |
| Slope doubles | impermeable (barrier) boundary, for example a fault or the valley wall; image-well theory |
| S-shaped curve in an unconfined aquifer | delayed yield (gravity drainage lags): early elastic response, then a flat segment, then a later Theis-like curve with \( S_y \) |
Other well terms: specific capacity \( Q/s \) (a quick index of \( T \)); recovery test (residual drawdown after pumping stops, analysed by the Theis recovery method); step-drawdown test separating aquifer loss (\( \propto Q \)) from turbulent well loss (\( \propto Q^2 \)); radius of influence; well interference when cones overlap.
Groundwater Exploration
graph TD
A[Desk study: geology, hydrogeological maps, well inventory] --> B[Remote sensing and GIS: lineaments, landforms, drainage density, palaeochannels]
B --> C[Groundwater potential zones]
C --> D[Surface geophysics: VES, resistivity imaging, magnetics, seismic refraction]
D --> E[Test drilling and borehole logging]
E --> F[Pumping tests and water quality]
F --> G[Production well design]Surface and remote sensing methods
- Hydrogeomorphological mapping from satellite images classifies landforms by groundwater prospect: valley fills, buried pediments, alluvial plains, palaeochannels and flood plains are good; structural hills and pediments with thin cover are poor.
- Lineaments (fractures, dykes, faults) and their intersections are the main targets in hard rock. Dykes can act as barriers (holding water on their upstream side) or, when fractured, as conduits.
- Vegetation anomalies, springs, seepage lines and soil moisture (thermal imagery) mark shallow water.
- GIS overlay of geology, geomorphology, lineament density, slope, drainage density and soils gives groundwater potential zone maps.
Electrical resistivity: the main geophysical tool
Resistivity of a water-bearing rock depends on the resistivity of the pore water and on porosity. For clay-free rock, Archie’s law:
$$\boxed{\rho_0 = a\,\rho_w\,\phi^{-m}} \qquad F = \frac{\rho_0}{\rho_w} = a\,\phi^{-m}, \qquad \rho_t = \rho_0\,S_w^{-n}\ \text{(partly saturated)}$$with \( a \approx 1 \), cementation exponent \( m \approx 1.3 \) (loose sand) to about 2 (consolidated rock) and saturation exponent \( n \approx 2 \). A sand with \( \phi = 0.25 \), \( m = 2 \) and fresh water of 10 Ω m has \( F = 16 \) and \( \rho_0 = 160 \) Ω m; the same sand with saline water of 1 Ω m drops to 16 Ω m, which is in the range of clay. Low resistivity can mean clay or saline water: the commonest ambiguity, resolved by drilling, IP (clay is polarisable) or water sampling.
Vertical electrical sounding (VES), usually with the Schlumberger array, expands the current electrodes about a fixed centre and plots \( \rho_a \) against AB/2 on log–log paper. Three-layer curves have four shapes:
| Type | Layer resistivities | Hydrogeological example |
|---|---|---|
| H | \( \rho_1 > \rho_2 < \rho_3 \) (minimum) | dry topsoil, saturated weathered zone, fresh hard rock: the classic hard-rock target |
| K | \( \rho_1 < \rho_2 > \rho_3 \) (maximum) | resistive dry sand or gravel over a conductive saturated or clayey layer |
| A | \( \rho_1 < \rho_2 < \rho_3 \) (ascending) | soil, weathered rock, fresh rock with little water |
| Q | \( \rho_1 > \rho_2 > \rho_3 \) (descending) | dry layer, fresh aquifer, saline aquifer or clay below |
The Dar Zarrouk parameters of a layer are its longitudinal conductance \( S = h/\rho \) (siemens) and transverse resistance \( T_r = h\rho \) (Ω m\(^2\)). By the principle of equivalence, a thin conductive layer between resistive ones (H type) is resolved only through \( S \): 10 m at 5 Ω m and 20 m at 10 Ω m (both \( S = 2 \) S) give nearly the same curve. A thin resistive layer between conductors (K type) is resolved only through \( T_r \). The principle of suppression: a thin layer whose resistivity lies between those of its neighbours (A and Q types) may not show on the curve at all. Interpretations are therefore non-unique and must be tied to boreholes.
Electrical resistivity tomography (ERT) with multi-electrode cables gives 2D or 3D images of fractures, weathered-zone thickness and the freshwater–saltwater interface.
Other geophysical methods
| Method | Hydrogeological use |
|---|---|
| Seismic refraction | depth to bedrock, thickness of weathered zone and alluvium; water table in unconsolidated sediments (P velocity jumps on saturation) |
| Magnetics | mapping dykes and faults that compartmentalise hard-rock and basalt aquifers |
| Gravity | buried valleys and thick alluvial basins |
| Electromagnetic (FEM, TEM, airborne EM) | rapid conductivity mapping: saline intrusion, clay layers, aquifer geometry over large areas |
| Ground-penetrating radar | shallow water table and sediment structure in sand |
| Magnetic resonance sounding | responds directly to hydrogen in free water: water content and a proxy for permeability |
Borehole logging and other tools
Logs in uncased holes identify aquifer zones: resistivity and SP logs separate clay from sand and fresh from saline water; natural gamma picks out clays (K-bearing); caliper, temperature and flowmeter logs locate productive fractures. Hydrochemistry and isotopes (\( \delta^{18}\)O, \( \delta \)D, tritium, \( ^{14}\)C) show where water was recharged and how old it is. Tracer tests measure velocity and dispersivity directly.
Groundwater Pollution
Sources
| Point sources | Non-point (diffuse) sources |
|---|---|
| septic tanks and leaking sewers, landfills and dumps, industrial effluent pits, leaking underground fuel tanks, mine tailings, injection wells | fertilisers (nitrate), pesticides, irrigation return flow (salinity), urban runoff, atmospheric deposition |
Geogenic contaminants come from the aquifer itself: arsenic, fluoride, uranium, iron and manganese, salinity of connate or evaporated water.
Transport: advection, dispersion, retardation
The one-dimensional advection–dispersion equation for a solute of concentration \( C \):
$$\boxed{R\,\frac{\partial C}{\partial t} = D_L\frac{\partial^2 C}{\partial x^2} - v\,\frac{\partial C}{\partial x}}$$with \( R = 1 \) for a non-sorbing solute; a first-order decay term is added when the solute degrades or decays.
| Process | Effect | Parameter |
|---|---|---|
| Advection | plume moves with the average linear velocity \( v = Ki/n_e \) | \( v \) |
| Mechanical dispersion | spreading from velocity variations at pore and aquifer scale; larger along flow than across it | dispersivity \( \alpha_L \): \( D = \alpha_L v + D^* \) |
| Molecular diffusion | spreading along concentration gradients; dominates only at very low velocity (clays) | \( D^* \) |
| Sorption | solute held on mineral and organic surfaces moves slower than water | retardation factor \( R \) |
| Decay or degradation | radioactive decay, biodegradation | rate constant \( \lambda \) |
The retardation factor for linear sorption with distribution coefficient \( K_d \):
$$\boxed{R = 1 + \frac{\rho_b}{n}K_d} \qquad v_c = \frac{v}{R}$$With \( \rho_b = 1.6 \) g cm\(^{-3}\), \( n = 0.3 \) and \( K_d = 0.5 \) mL g\(^{-1}\): \( R = 3.67 \). If water moves at 0.3 m d\(^{-1}\), the solute moves at 0.082 m d\(^{-1}\); water crosses 100 m in 333 days, the solute in about 1,220 days. Chloride and nitrate (\( K_d \approx 0 \)) travel with the water and are used as conservative tracers.
A pulse spreads with a standard deviation \( \sigma = \sqrt{2Dt} \): with \( \alpha_L = 5 \) m and \( v = 0.3 \) m d\(^{-1}\), \( D = 1.5 \) m\(^2\) d\(^{-1}\), and after 333 days \( \sigma \approx 32 \) m. Field dispersivities grow with the scale of observation (the scale effect), because larger distances sample more heterogeneity.
Non-aqueous phase liquids
LNAPLs (petrol, diesel; lighter than water) float on the water table and spread as a pancake, dissolving slowly into the water below. DNAPLs (chlorinated solvents such as trichloroethene and tetrachloroethene, creosote, coal tar; denser than water) sink through the aquifer, pool on clay layers and follow the slope of the clay rather than the hydraulic gradient. DNAPL sources are the hardest to find and to remove.
Geogenic contaminants in India
| Contaminant | Mechanism | Setting |
|---|---|---|
| Arsenic | reductive dissolution of Fe(III) oxyhydroxides that adsorb As, driven by microbial oxidation of buried organic matter; arsenite As(III) dominates under reducing conditions and is more mobile and more toxic than arsenate As(V). Oxidation of arsenic-bearing pyrite is a secondary mechanism, important where water tables are lowered | young (Holocene) alluvial and deltaic aquifers of the Ganga–Brahmaputra plains and delta; older (Pleistocene) oxidised sediments are generally low in arsenic |
| Fluoride | dissolution of fluorite, fluorapatite, micas and amphiboles in granitic and gneissic rocks; high in alkaline, sodium-bicarbonate water with low calcium, because calcite precipitation removes Ca\(^{2+}\) and lets more fluorite dissolve; enhanced by evaporation and long residence | hard-rock and arid terrains of peninsular and north-western India |
| Salinity | seawater intrusion, evaporative concentration, connate water, irrigation return flow | coastal aquifers (Saurashtra, Kachchh and other coasts), arid Rajasthan and south-west Punjab–Haryana |
| Uranium | oxidising, bicarbonate-rich water dissolves U(VI) from granitic rocks and sediments; mobilised further where intensive pumping and fertiliser use change water chemistry | reported in parts of north-western India |
| Iron and manganese | reductive dissolution of oxides | many alluvial and hard-rock aquifers |
| Nitrate (anthropogenic) | fertiliser and sewage; persists under oxidising conditions, removed by denitrification under reducing ones | intensively farmed plains and towns |
Indian drinking water is regulated by IS 10500 (BIS). Values commonly applied: arsenic 0.01 mg L\(^{-1}\); fluoride 1.0 mg L\(^{-1}\) acceptable and 1.5 mg L\(^{-1}\) permissible in the absence of an alternative source; nitrate 45 mg L\(^{-1}\) (as NO\(_3^-\)); total dissolved solids 500 mg L\(^{-1}\) acceptable and 2,000 mg L\(^{-1}\) permissible.
Vulnerability: the DRASTIC index
DRASTIC rates seven hydrogeological factors (each 1–10) and multiplies by fixed weights:
| Factor | Weight |
|---|---|
| Depth to water | 5 |
| Net Recharge | 4 |
| Aquifer media | 3 |
| Soil media | 2 |
| Topography (slope) | 1 |
| Impact of the vadose zone | 5 |
| Hydraulic Conductivity | 3 |
The index ranges from 23 to 230. Higher means more vulnerable to contamination from the surface. A shallow water table, high recharge and permeable vadose zone dominate because their weights are largest.
Remediation
| Approach | How it works |
|---|---|
| Source control | remove leaking tanks, excavate contaminated soil, cap landfills |
| Pump-and-treat | extract contaminated water and treat it at the surface; contains plumes but is slow to clean aquifers (tailing, rebound) because of sorption and diffusion into low-permeability zones |
| Permeable reactive barrier | a trench of reactive material across the plume; zero-valent iron reduces chlorinated solvents and Cr(VI) |
| Bioremediation | stimulate microbes with oxygen, nutrients or electron donors |
| Air sparging, soil vapour extraction | strip volatile compounds |
| Monitored natural attenuation | rely on dispersion, sorption and degradation, with monitoring |
| For geogenic arsenic and fluoride | switch to safe aquifers (often deeper or older), treatment (adsorption on activated alumina or iron oxides; Nalgonda technique for fluoride), rainwater harvesting and surface supply |
Indian Groundwater Regimes
Hydrogeological provinces
| Province | Aquifers | Character |
|---|---|---|
| Indo-Gangetic–Brahmaputra alluvial plains | thick Quaternary sands and gravels with clay lenses, several hundred metres or more in places | high yields, multi-tier (unconfined above, semi-confined to confined below); over-exploited in the north-west; arsenic in parts of the middle and lower Ganga plain and the delta |
| Bhabar and Terai belts (Himalayan foothills) | Bhabar: coarse boulder–gravel piedmont fans; Terai: finer sediments downslope | the Bhabar is a recharge zone with a deep water table where streams sink; groundwater re-emerges as springs and seepage in the Terai, where shallow and flowing artesian conditions are common |
| Peninsular hard rocks (Archaean and Proterozoic granites, gneisses, schists, charnockites) | weathered zone over fractured zone | low to moderate yields, highly variable over short distances; heavily used by dug wells and borewells, so many blocks are stressed |
| Deccan Traps basalt | vesicular and fractured flow units, weathered mantle | moderate yields; compartmentalised by massive flow interiors and dykes; groundwater in the upper flows and weathered zone |
| Consolidated and semi-consolidated sedimentary rocks (Gondwana, Vindhyan, Cuddapah, Tertiary) | sandstones, limestones | moderate to good; Gondwana and Tertiary sandstones can form confined aquifers |
| Coastal and deltaic aquifers | alluvium and Tertiary sediments of the east and west coasts | productive but vulnerable to seawater intrusion |
| Arid western Rajasthan | alluvium, Tertiary sediments, palaeochannels | deep water tables, much saline groundwater; fresh water in some buried channels |
| Himalayan and other hilly terrain | fractured rocks, colluvium, valley fill | springs are the main source; many are reported to be declining |
The Central Ground Water Board (CGWB) maps aquifers nationally under its aquifer-mapping programme (NAQUIM) and groups them into principal aquifer systems by rock type (alluvium, laterite, basalt, sandstone, shale, limestone, granite, schist, quartzite, charnockite, khondalite, banded gneissic complex, gneiss and intrusives).
Monitoring and assessment
- CGWB measures water levels in a national network of observation wells several times a year, including pre-monsoon and post-monsoon rounds; the rise between them reflects monsoon recharge.
- Water-table fluctuation method: recharge \( = S_y\,\Delta h\,A \) (plus net draft during the period). An area of 50 km\(^2\) with \( S_y = 0.12 \) and a monsoon rise of 3 m stores \( 50\times 10^6\times 0.12\times 3 = 1.8\times 10^7 \) m\(^3\) (18 million cubic metres).
- Dynamic groundwater resources are assessed jointly by CGWB and the states, assessment unit by unit (blocks, mandals, talukas), using the Groundwater Estimation Committee methodology of 2015 (GEC 2015).
| Category | Stage of extraction |
|---|---|
| Safe | ≤ 70% |
| Semi-critical | > 70% and ≤ 90% |
| Critical | > 90% and ≤ 100% |
| Over-exploited | > 100% |
(Long-term water-level decline is also checked, and units with saline groundwater are reported separately.) A unit extracting 102 million m\(^3\) against an extractable 120 million m\(^3\) is at 85%: semi-critical. Over-exploited units are concentrated in Punjab, Haryana, Delhi, Rajasthan and western Uttar Pradesh, and in hard-rock parts of the south.
Management
Demand-side measures (efficient irrigation such as drip and sprinkler, crop choice, metering, regulating new wells) and supply-side measures (managed aquifer recharge: check dams, percolation tanks, recharge shafts and wells, rooftop rainwater harvesting, spreading basins) together with participatory, aquifer-level planning. In the alluvial north-west, conjunctive use of canal water and groundwater addresses both falling water tables (where groundwater is over-used) and waterlogging and salinity (where canal water is over-used).
Memory Tricks
Don’t Confuse
| This | vs | That |
|---|---|---|
| Hydraulic conductivity \( K \): fluid + medium (m s\(^{-1}\)) | Intrinsic permeability \( k \): medium only (m\(^2\), darcy) | |
| Specific discharge \( q \) | Average linear (seepage) velocity \( v = q/n_e \) | |
| Storativity (confined, \( 10^{-5} \)–\( 10^{-3} \)) | Specific yield (unconfined, 0.01–0.3) | |
| \( K_h \): arithmetic mean | \( K_v \): harmonic mean | |
| Recharge area: head falls with depth | Discharge area: head rises with depth | |
| Thiem: steady, confined, uses \( s \) | Dupuit–Thiem: steady, unconfined, uses \( h^2 \) | |
| Aquifer loss \( \propto Q \) | Well loss \( \propto Q^2 \) | |
| H-type VES (minimum) | K-type VES (maximum) | |
| Equivalence: different layers, same curve | Suppression: a layer that does not appear | |
| Mechanical dispersion: velocity variations | Molecular diffusion: concentration gradients | |
| LNAPL floats | DNAPL sinks and pools on clay | |
| As(III), arsenite: reducing, more mobile and toxic | As(V), arsenate: oxidising, adsorbs strongly on Fe oxides | |
| Bhabar: recharge, deep water table | Terai: discharge, springs and shallow water |
Common Mistakes
Practice
All questions below are original practice questions written for this page. They are not from any CSIR NET paper.
Level 1: Foundation
- Which mean (arithmetic or harmonic) gives the effective vertical hydraulic conductivity of a layered sequence?
- Which three-layer VES curve type has \( \rho_1 > \rho_2 < \rho_3 \)?
- What are the DRASTIC weights of depth to water and of topography?
Answers
- Harmonic: \( K_v = \sum b_i/\sum(b_i/K_i) \).
- H type (a minimum).
- 5 for depth to water and 1 for topography.
Level 2: Part B Style
Practice Q1. An aquifer consists of 5 m of sand (\( K = 10 \) m d\(^{-1}\)), 2 m of clay (\( K = 0.01 \) m d\(^{-1}\)) and 3 m of sand (\( K = 5 \) m d\(^{-1}\)). The effective horizontal and vertical conductivities are about:
(a) 6.5 and 0.05 m d\(^{-1}\) (b) 5.0 and 5.0 m d\(^{-1}\) (c) 6.5 and 6.5 m d\(^{-1}\) (d) 0.05 and 6.5 m d\(^{-1}\)
Answer: (a)
\( K_h = (50 + 0.02 + 15)/10 = 6.50 \) m d\(^{-1}\). \( K_v = 10/(0.5 + 200 + 0.6) = 10/201.1 = 0.0497 \) m d\(^{-1}\). (d) swaps the two means.
Practice Q2. In a pumping test (\( Q = 2{,}500 \) m\(^3\) d\(^{-1}\)) the semi-log plot of drawdown against time at an observation well becomes a straight line with 0.8 m of drawdown per log cycle. The transmissivity is about:
(a) 249 m\(^2\) d\(^{-1}\) (b) 573 m\(^2\) d\(^{-1}\) (c) 1,320 m\(^2\) d\(^{-1}\) (d) 3,125 m\(^2\) d\(^{-1}\)
Answer: (b) 573 m² d⁻¹
\( T = 2.303Q/(4\pi\Delta s) = 2.303\times 2500/(4\pi\times 0.8) = 5757.5/10.05 = 573 \) m\(^2\) d\(^{-1}\). (a) omits the 2.303 (\( 2500/(4\pi\times 0.8) \approx 249 \)); (d) is \( Q/\Delta s \) without \( 4\pi \) or 2.303.
Practice Q3. Sorption of a pesticide in a sandy aquifer is described by \( K_d = 0.5 \) mL g\(^{-1}\); bulk density is 1.6 g cm\(^{-3}\) and porosity 0.30. Compared with groundwater, the pesticide front moves:
(a) at the same speed (b) about 1.8 times slower (c) about 3.7 times slower (d) about 5.3 times slower
Answer: (c)
\( R = 1 + (\rho_b/n)K_d = 1 + (1.6/0.3)(0.5) = 1 + 2.67 = 3.67 \). (d) is \( \rho_b/n \) alone; (b) leaves out the porosity (\( 1 + 1.6\times 0.5 = 1.8 \)).
Practice Q4. Which statement about arsenic in the alluvial aquifers of the Bengal delta is correct?
(a) It is released mainly by industrial effluents (b) It is released mainly by reductive dissolution of iron oxyhydroxides, and dominantly occurs as As(III) (c) It is highest in oxidised, brown Pleistocene sands (d) It is removed from water by reducing conditions
Answer: (b)
Microbial oxidation of organic matter in young (Holocene) grey sediments reduces Fe(III) oxyhydroxides, releasing adsorbed arsenic; under these reducing conditions arsenite dominates. Oxidised Pleistocene sediments generally retain arsenic on their iron oxides, so (c) is the reverse of the usual pattern.
Level 3: Part C Style
Practice Q5. A well pumps 2,500 m\(^3\) d\(^{-1}\) from a confined aquifer. Drawdown at an observation well 50 m away plots on a semi-log graph as a straight line with 0.8 m per log cycle, extrapolating to zero drawdown at 0.5 min. After 600 min the drawdown curve flattens to a constant value. Which statements are correct?
- \( T \approx 573 \) m\(^2\) d\(^{-1}\) and \( S \approx 1.8\times 10^{-4} \).
- At 100 min, \( u < 0.01 \), so the Cooper–Jacob method is valid there.
- The flattening indicates an impermeable boundary such as a fault.
- The value of \( S \) is consistent with confined, not unconfined, conditions.
(a) 1, 2 and 4 only (b) 1 and 3 only (c) 2, 3 and 4 only (d) 1, 2, 3 and 4
Answer: (a) 1, 2 and 4 only
- 1 true: \( T = 573 \) m\(^2\) d\(^{-1}\); \( t_0 = 0.5/1440 = 3.47\times 10^{-4} \) d; \( S = 2.25\times 573\times 3.47\times 10^{-4}/50^2 = 1.79\times 10^{-4} \).
- 2 true: at \( t = 100 \) min \( = 0.0694 \) d, \( u = 2500\times 1.79\times 10^{-4}/(4\times 573\times 0.0694) = 0.0028 \).
- 3 false: flattening (steady drawdown) indicates a recharge boundary (a river or lake) or leakage; a barrier boundary would steepen the slope (double it).
- 4 true: \( 10^{-4} \) is in the confined range; unconfined aquifers have \( S_y \) of about 0.01–0.3.
Practice Q6. A Schlumberger VES in granite terrain gives a curve that falls from 300 Ω m at small spacings to a minimum near 40 Ω m and then rises steeply above 1,000 Ω m. A laboratory test shows a clean sample of the saturated weathered granite has porosity 0.25 and Archie constants \( a = 1 \), \( m = 2 \). Which statements are correct?
- The curve is H type.
- If the minimum represents clean saturated material, the pore water has a resistivity of about 2.5 Ω m.
- Another model with the middle layer twice as thick and twice as resistive would fit the curve almost equally well.
- The deepest layer is probably fresh, unfractured granite, a poor aquifer.
(a) 1, 3 and 4 only (b) 1, 2, 3 and 4 (c) 1 and 2 only (d) 2 and 4 only
Answer: (b) 1, 2, 3 and 4
- 1 true: high, low, high is a minimum: H type.
- 2 true: \( F = a\phi^{-m} = 1/0.25^2 = 16 \); \( \rho_w = \rho_0/F = 40/16 = 2.5 \) Ω m. (In practice clay in the weathered zone also lowers resistivity, so this is an upper estimate of salinity; field water should be tested.)
- 3 true: equivalence: a conductive layer between resistive ones is resolved only through its longitudinal conductance \( S = h/\rho \), which is unchanged when both \( h \) and \( \rho \) double.
- 4 true: very high resistivity at depth means unweathered, unfractured granite with little water.
Practice Q7. A watershed of 50 km\(^2\) in hard rock (\( S_y = 0.12 \)) shows an average water-level rise of 3 m between pre- and post-monsoon measurements, with negligible pumping during the monsoon. The assessment unit’s annual extractable resource is 120 million m\(^3\) and annual extraction is 102 million m\(^3\). A DRASTIC assessment of one village gives ratings D 9, R 6, A 8, S 6, T 10, I 8, C 4. Which statements are correct?
- Monsoon recharge in the watershed is about 18 million m\(^3\).
- The assessment unit is in the “critical” category.
- The village’s DRASTIC index is 167.
- Lowering the depth-to-water rating from 9 to 3 would reduce the index by 30.
(a) 1, 3 and 4 only (b) 1 and 2 only (c) 2, 3 and 4 only (d) 1, 2, 3 and 4
Answer: (a) 1, 3 and 4 only
- 1 true: \( 50\times 10^6\times 0.12\times 3 = 1.8\times 10^7 \) m\(^3\).
- 2 false: \( 102/120 = 85\% \): semi-critical (> 70% and ≤ 90%).
- 3 true: \( 5(9) + 4(6) + 3(8) + 2(6) + 1(10) + 5(8) + 3(4) = 45 + 24 + 24 + 12 + 10 + 40 + 12 = 167 \).
- 4 true: the change is \( 5\times(9 - 3) = 30 \).
Practice Q8. A conservative tracer and a sorbing contaminant (\( R = 3.67 \)) enter an aquifer together. The average linear velocity is 0.3 m d\(^{-1}\) and the longitudinal dispersivity is 5 m (diffusion negligible). Which statement is correct about a well 100 m downgradient?
(a) Both arrive in about 333 days, because both are carried by the same water (b) The tracer peak arrives in about 333 days with a longitudinal spread (one standard deviation) of about 32 m; the contaminant peak arrives in about 1,220 days (c) The contaminant arrives first because sorption speeds it up (d) The tracer arrives in about 1,220 days and the contaminant in about 333 days
Answer: (b)
Tracer: \( t = 100/0.3 = 333 \) d. \( D = \alpha_L v = 1.5 \) m\(^2\) d\(^{-1}\); \( \sigma = \sqrt{2Dt} = \sqrt{2\times 1.5\times 333} = 31.6 \) m. Contaminant: \( v_c = 0.3/3.67 = 0.082 \) m d\(^{-1}\); \( t = 100/0.082 \approx 1{,}222 \) d. The front of the tracer arrives well before 333 days because of dispersion, which is why early detection at monitoring wells is possible.
Quick Revision Table
| Idea | Fact or formula |
|---|---|
| Head | \( h = z + p/\rho g \); flow from high to low \( h \) |
| Recharge vs discharge area | head falls vs rises with depth |
| Darcy | \( q = -K\,dh/dl \); \( v = q/n_e \); \( K = k\rho g/\mu \); 1 darcy ≈ \( 9.87\times 10^{-13} \) m\(^2\) |
| Validity | laminar, \( Re = \rho qd/\mu \) below about 1–10 |
| Hazen | \( K \approx 100\,d_{10}^2 \) (cm s\(^{-1}\), cm) |
| Layers | \( K_h \) arithmetic; \( K_v \) harmonic |
| Refraction | \( \tan\theta_1/\tan\theta_2 = K_1/K_2 \) |
| Storage | \( S = S_sb \) (\( 10^{-5} \)–\( 10^{-3} \)); \( S_y \) 0.01–0.3; \( T = Kb \) |
| Flow equation | \( \nabla^2h = (S/T)\partial h/\partial t \); Laplace at steady state |
| Flow net | \( q' = KHN_f/N_d \) |
| Leaky | \( B = \sqrt{Tb'/K'} \); Hantush–Jacob |
| Thiem | \( Q = 2\pi T(s_1 - s_2)/\ln(r_2/r_1) \) |
| Dupuit | \( Q = \pi K(h_2^2 - h_1^2)/\ln(r_2/r_1) \) |
| Theis | \( s = QW(u)/4\pi T \), \( u = r^2S/4Tt \) |
| Cooper–Jacob | \( T = 2.303Q/4\pi\Delta s \); \( S = 2.25Tt_0/r^2 \); \( u < 0.01 \) |
| Boundaries | flatten: recharge or leakage; slope doubles: barrier |
| Well loss | \( \propto Q^2 \); aquifer loss \( \propto Q \) |
| Archie | \( \rho_0 = a\rho_w\phi^{-m} \); \( F = \rho_0/\rho_w \) |
| VES | H min, K max, A ascending, Q descending |
| Dar Zarrouk | \( S = h/\rho \), \( T_r = h\rho \); equivalence, suppression |
| ADE | advection, dispersion \( D = \alpha_Lv + D^* \), retardation, decay |
| Retardation | \( R = 1 + (\rho_b/n)K_d \); \( v_c = v/R \) |
| Spread | \( \sigma = \sqrt{2Dt} \) |
| NAPL | LNAPL floats; DNAPL sinks |
| Arsenic | reductive dissolution of Fe oxyhydroxides; Holocene alluvium; As(III) |
| Fluoride | Na–HCO\(_3\), low Ca, granitic hard rock, arid |
| IS 10500 | As 0.01; F 1.0/1.5; NO\(_3^-\) 45; TDS 500/2,000 mg L\(^{-1}\) |
| DRASTIC | weights 5, 4, 3, 2, 1, 5, 3; range 23–230 |
| India | alluvial plains, Bhabar–Terai, hard rock, Deccan basalt, sedimentary, coastal, arid, Himalayan springs |
| Recharge | \( S_y\Delta hA \) |
| Stage of extraction | safe ≤ 70; semi-critical ≤ 90; critical ≤ 100; over-exploited > 100% |
Key Takeaways
What’s Next
- Applied Geology: Quick Revision: every formula and trap from this unit on one page.
- Engineering Geology: Dams, Tunnels, Landslides and Seismic Design: seepage, uplift and pore pressure under dams and in slopes.
- Physiography and River Basins of India: the landscapes that host these aquifers.
- Natural Hazards: land subsidence and other hazards linked to groundwater.
- Back to the Applied Geology overview.