Porosity Logs: Sonic, Density, Neutron and Gamma Ray
Sonic (Wyllie, Raymer), density and Pe, neutron hydrogen index, gas crossover, gamma ray, spectral GR and shale volume for CSIR NET Earth Sciences.
- 19 min
On this page · 15 sections
Prerequisites
- Logging Operations: Open-Hole, Cased-Hole, Production Logging and LWD — the borehole environment, invasion and which tools run in which hole.
- Radioactive Decay and Selection Rules — gamma emission, decay series and counting statistics behind the nuclear logs.
- Seismology Basics and Earth’s Internal Structure — P and S waves, velocity and elastic moduli, which the sonic log measures at centimetre scale.
- Facies, Depositional Environments, Basins and Diagenesis — fining-up and coarsening-up successions, read from GR log shapes.
The Hook: Three Tools, Three Porosities
How CSIR NET asks it. Part C Geophysics: porosity from \( \Delta t \) (Wyllie, Raymer–Hunt–Gardner), from bulk density and from neutron logs; matrix values for sandstone, limestone and dolomite; the gas crossover and the shale separation on a neutron–density overlay; lithology from Pe; secondary porosity index; shale volume from GR (linear and Larionov); spectral GR and radioactive sands; GR log shapes.
Porosity: What We Want
Porosity \( \phi \) is the fraction of the bulk volume that is pore space. Total porosity counts all pores including clay-bound water; effective porosity counts the connected pores available to free fluids. Primary (intergranular) porosity forms at deposition; secondary porosity (vugs, moulds, fractures, dissolution) forms later. Porosity logs measure a bulk property and convert it to \( \phi \) through a mixing law: the bulk value is a volume-weighted combination of matrix and pore-fluid values.
| Tool | Measures | Mixing law | Main error sources |
|---|---|---|---|
| Sonic | interval transit time \( \Delta t \) | Wyllie time average (empirical) | compaction, gas, vugs and fractures ignored |
| Density | bulk density \( \rho_b \) | exact linear mixing | matrix density choice, gas, hole rugosity, heavy minerals |
| Neutron | hydrogen index | calibrated, approximately linear | lithology, shale, gas, chlorine |
The Sonic Log
Principle
A transmitter emits a short acoustic pulse; receivers some distance up the tool detect the first compressional (P) arrival refracted along the borehole wall. The interval transit time \( \Delta t \) (slowness) is the time taken to travel a fixed distance in the formation, in µs/ft (or µs/m). It is the reciprocal of velocity:
$$v\ (\text{m/s}) = \frac{304{,}800}{\Delta t\ (\mu\text{s/ft})}.$$The borehole-compensated (BHC) sonic uses two transmitters and two pairs of receivers, firing up and down, and averages the two measurements; this cancels the effects of hole-size changes and tool tilt. Long-spacing sonics read beyond the altered zone near the wall. Dipole sonic tools excite flexural waves to measure shear slowness even in slow formations where the shear velocity is below the mud velocity (where a monopole tool cannot see a refracted shear head wave). Full-waveform tools also record the Stoneley wave, a tube wave along the borehole wall that is attenuated by open fractures and permeability.
Porosity from transit time
Wyllie time-average equation (empirical, for consolidated, clean rocks with intergranular porosity):
$$\boxed{\Delta t = \phi\,\Delta t_f + (1 - \phi)\,\Delta t_{ma}\quad\Longrightarrow\quad \phi_S = \frac{\Delta t - \Delta t_{ma}}{\Delta t_f - \Delta t_{ma}}}$$Raymer–Hunt–Gardner (a field-observation fit that works better over a wider porosity range; commonly used in the simplified form)
$$\phi_S \approx C\,\frac{\Delta t - \Delta t_{ma}}{\Delta t},\qquad C \approx 0.625\text{–}0.70.$$| Material | \( \Delta t \) (µs/ft) | \( v \) (m/s) |
|---|---|---|
| Sandstone matrix | 55.5 (51.3 also used for fast, quartz-rich sands) | about 5,490 |
| Limestone matrix | 47.5 | about 6,420 |
| Dolomite matrix | 43.5 | about 7,010 |
| Anhydrite | 50.0 | about 6,100 |
| Salt (halite) | 67.0 | about 4,550 |
| Water-based mud filtrate | 189 | about 1,610 |
Corrections and limitations:
- Uncompacted sands. Wyllie over-estimates porosity where adjacent shales are slow (high \( \Delta t \)); a compaction correction divides \( \phi_S \) by a factor \( C_p \) estimated from the shale transit time.
- Gas in shallow, uncompacted sands raises \( \Delta t \) and over-estimates porosity.
- Secondary porosity. The first-arrival path goes through the fastest continuous route, the matrix, so the sonic responds mainly to intergranular porosity and largely ignores vugs and fractures. Density and neutron see all pores. The difference gives the secondary porosity index:
- Cycle skipping: in gas zones, fractures or poor-quality holes the far receiver triggers on a later cycle, giving spikes of too-long \( \Delta t \).
Other uses of the sonic
- Seismic tie: combined with density it gives acoustic impedance and reflection coefficients for a synthetic seismogram (the convolutional model of Auto- and Cross-Correlation and LTI Systems); integrated \( \Delta t \) gives time–depth conversion.
- Overpressure detection: in normally compacted shales \( \Delta t \) decreases steadily with depth; shale \( \Delta t \) rising above the trend marks undercompacted, overpressured shale.
- Lithology: the \( V_p/V_s \) ratio separates lithologies (commonly quoted as about 1.6–1.75 for sandstone, about 1.8 for dolomite and about 1.9 for limestone) and drops in gas sands.
- Mechanical properties: \( V_p \), \( V_s \) and density give dynamic Poisson’s ratio and moduli for wellbore stability and fracturing design.
The Density Log
Principle: Compton scattering
A chemical gamma source (caesium-137, 662 keV) on a pad pressed against the wall sends gamma rays into the formation. At these energies they lose energy mainly by Compton scattering off electrons, so the number reaching detectors a short distance away depends on the formation’s electron density \( \rho_e \) (electrons per unit volume). Higher density means more scattering and absorption, and fewer counts at the far detector.
$$\rho_e = \rho_b\,\frac{2\sum Z}{M}$$For most rock-forming minerals \( 2Z/A \) is very close to 1 (quartz 0.9985, calcite 0.9991), so \( \rho_e \approx \rho_b \). Water is the exception (1.1101, because of hydrogen). The tool is calibrated so that its apparent density
$$\rho_a = 1.0704\,\rho_e - 0.1883$$equals the true bulk density for limestone saturated with fresh water (it gives 1.000 for water itself and essentially the true value for calcite and quartz).
Compensation. Two detectors at different spacings respond differently to the mud cake and stand-off between pad and wall. The tool combines them and displays the correction \( \Delta\rho \); a large correction warns that the pad lost contact (rough or washed-out hole) and the reading is unreliable.
Porosity from density
Bulk density is an exact volume average:
$$\rho_b = \phi\,\rho_f + (1 - \phi)\,\rho_{ma}\quad\Longrightarrow\quad\boxed{\phi_D = \frac{\rho_{ma} - \rho_b}{\rho_{ma} - \rho_f}}$$\( \rho_f \) is the density of the fluid in the zone the tool reads (the flushed zone): about 1.0 g/cm³ for fresh filtrate, about 1.1 for salty filtrate.
| Mineral | \( \rho_{ma} \) (g/cm³) | Pe (b/e) |
|---|---|---|
| Quartz (sandstone) | 2.65 | 1.81 |
| Calcite (limestone) | 2.71 | 5.08 |
| Dolomite | 2.87 | 3.14 |
| Anhydrite | 2.98 | 5.05 |
| Halite | 2.04 (log reads about 2.03) | 4.65 |
| Barite (mud additive) | — | about 267 |
| Fresh water | 1.00 | 0.36 |
The photoelectric factor (Pe)
Lower-energy gamma rays are absorbed by the photoelectric effect, whose cross-section rises steeply with atomic number (roughly as \( Z^{3.6} \) per electron). Litho-density tools measure the low-energy part of the spectrum to give Pe in barns per electron. Pe depends mostly on the matrix, little on porosity, so it identifies lithology directly: about 1.8 sandstone, 3.1 dolomite, 5.1 limestone. Barite in the mud swamps the Pe reading. The volumetric cross-section \( U = \text{Pe}\times\rho_e \) mixes linearly by volume and is used in matrix-identification plots.
Errors
- Gas lowers \( \rho_b \): density porosity is too high.
- Wrong matrix: a dolomite logged with a limestone matrix (2.71) gives too low a porosity; a sandstone logged with 2.71 gives too high a value.
- Heavy minerals (pyrite, siderite) raise \( \rho_b \) and lower \( \phi_D \).
- Shale reads as the shale’s density, typically giving moderate apparent porosity.
- Rugose hole: low readings where the pad loses contact.
The Neutron Log
Principle: slowing down by hydrogen
A chemical source (americium–beryllium) emits fast neutrons of a few MeV. They lose energy by elastic collisions with nuclei. The energy lost per collision is greatest when the target has the same mass as the neutron, so hydrogen is by far the most effective moderator: a neutron can lose all its energy in one head-on collision with a proton. Once slowed to thermal energies, neutrons wander until captured (chlorine is a strong absorber).
The count of slowed (epithermal or thermal) neutrons at a detector therefore measures the hydrogen concentration, expressed as the hydrogen index (HI): the hydrogen per unit volume relative to fresh water (HI = 1). In clean rock, hydrogen sits in the pore fluid, so HI ≈ porosity.
Compensated neutron tools use the ratio of counts at near and far detectors, which reduces borehole effects. Logs are usually presented in limestone porosity units: the tool reads true porosity in water-filled limestone.
What the neutron log reads
| Condition | Neutron response | Why |
|---|---|---|
| Water- or oil-filled clean limestone | true porosity | calibration matrix; liquid HI ≈ 1 |
| Clean sandstone on limestone scale | reads somewhat low | matrix effect (and different absorber content) |
| Dolomite on limestone scale | reads high | matrix effect |
| Gas | low | gas has far less hydrogen per volume (low HI), plus an “excavation” effect |
| Shale | high | bound water and hydroxyl in clays |
| Gypsum, coal | very high | water of crystallisation; hydrocarbon-rich organic matter |
| Saline formation and mud | slightly affected (thermal tools) | chlorine captures thermal neutrons |
Neutron–Density Combination
Plotted together on compatible scales (both in limestone porosity units), the two curves separate in characteristic ways.
| Pattern on a limestone-scaled overlay | Interpretation |
|---|---|
| Curves overlay | clean, liquid-filled limestone (true porosity) |
| Density porosity > neutron porosity (crossover) | gas (or, by a few porosity units, clean water-filled sandstone) |
| Neutron porosity ≫ density porosity, with high GR | shale |
| Neutron > density, low GR, Pe about 3 | dolomite |
| Very low density porosity (negative), high Pe | anhydrite |
Porosity from the combination. In liquid-filled zones the average of the two (corrected to the right matrix) is a good estimate. In gas zones, the two errors are partly opposite, and the root-mean-square is used:
$$\boxed{\phi_{ND} \approx \sqrt{\frac{\phi_N^2 + \phi_D^2}{2}}\quad\text{(gas zones)}}$$graph TD
A["Neutron and density
on limestone scale"] --> B{"Which reads higher?"}
B -->|"Overlay"| C["Clean liquid-filled
limestone"]
B -->|"Density higher
(crossover)"| D{"How large?"}
D -->|"Large, low GR"| E["Gas"]
D -->|"A few p.u., low GR, Pe 1.8"| F["Clean sandstone
with liquid"]
B -->|"Neutron higher"| G{"GR and Pe?"}
G -->|"High GR"| H["Shale"]
G -->|"Low GR, Pe about 3"| I["Dolomite"]The Gamma-Ray Log
Principle
The GR tool counts natural gamma rays with a scintillation detector. Almost all natural gamma radiation in rocks comes from three sources:
| Source | Characteristic peak used in spectral tools | Where it is concentrated |
|---|---|---|
| Potassium-40 | 1.46 MeV | illite, micas, K-feldspar, glauconite, potash evaporites |
| Uranium series | 1.76 MeV (bismuth-214) | organic-rich shales, phosphates, some uraniferous sands; uranium moves in solution |
| Thorium series | 2.61 MeV (thallium-208) | clays, heavy minerals (monazite, zircon), volcanic ash |
Total GR is recorded in API units, defined from a calibration facility so that tools from different makers agree. Because radioactive decay is random, the reading has statistical fluctuation; slower logging or longer time constants smooth it at the cost of vertical resolution. Gamma rays pass through steel, so the GR also runs in cased hole and in LWD.
Shale volume
Shales are usually rich in K and Th (clay minerals), and clean sands and carbonates are poor in them. The gamma-ray index is
$$\boxed{I_{GR} = \frac{GR_{\log} - GR_{\min}}{GR_{\max} - GR_{\min}}}$$with \( GR_{\min} \) from a clean bed and \( GR_{\max} \) from a shale. Taking \( V_{sh} = I_{GR} \) (linear) gives the maximum estimate. Non-linear empirical corrections (Larionov) give smaller values:
$$V_{sh} = 0.083\,(2^{3.7 I_{GR}} - 1)\ \ \text{(Tertiary, unconsolidated)};\qquad V_{sh} = 0.33\,(2^{2 I_{GR}} - 1)\ \ \text{(older rocks)}.$$Spectral gamma ray (NGS)
Spectral tools split the count into K (%), U (ppm) and Th (ppm). Uses:
- Radioactive “clean” sands: an arkose (K-feldspar) or a micaceous or glauconitic sand reads high GR though it is not shaly. Spectral data show the K source; the uranium-free GR (computed from K and Th only) gives a better shale estimate.
- Organic-rich source rocks (“hot shales”): high U. Uranium is fixed by organic matter under reducing conditions.
- Clay typing: the Th/K ratio differs between clay minerals (illite K-rich; kaolinite Th-rich and K-poor).
- Fractures and diagenesis: uranium precipitated from moving water along fractures and unconformities.
GR log shapes
In sand–shale successions, GR decreases as grain size coarsens (less clay), so the GR curve tracks grain-size trends:
| Shape | Grain-size trend | Typical setting |
|---|---|---|
| Cylinder (blocky) | uniform, sharp top and base | channel fill, aeolian dune, turbidite sand |
| Bell | fining upward | fluvial point bar, tidal channel, transgressive fill |
| Funnel | coarsening upward | prograding delta front, shoreface, crevasse splay |
| Serrated | interbedded | heterolithic shelf or floodplain |
These shapes are also the basis for correlating parasequences and systems tracts in Stratigraphic Code, Correlation and Sequence Stratigraphy.
Memory Tricks
Don’t Confuse
| Pair | First | Second |
|---|---|---|
| Density vs neutron in gas | porosity reads too high | porosity reads too low |
| Total vs effective porosity | includes clay-bound water | connected pores only |
| Bulk density vs electron density | mass per volume | electrons per volume; equal when \( 2Z/A = 1 \) |
| Pe vs bulk density | lithology (atomic number) | porosity (mass) |
| Sonic porosity vs density porosity | intergranular only (fastest path) | all pore space |
| Wyllie vs Raymer–Hunt–Gardner | linear time average | empirical fit, better at high porosity |
| \( I_{GR} \) vs \( V_{sh} \) | linear index | shale volume after empirical (Larionov) correction; smaller |
| GR vs SP for shale | radioactivity; works in any mud and through casing | electrochemical; needs conductive mud and salinity contrast |
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 radioactive source is used in a density tool, and what interaction does it rely on?
- What is the matrix bulk density of dolomite used in density porosity calculations?
- Which tool reads low porosity in a gas zone, and why?
Answers
- Caesium-137 (662 keV gamma rays); Compton scattering by electrons, so the tool measures electron density.
- 2.87 g/cm³.
- The neutron log: gas has a much lower hydrogen index than water or oil, so fewer neutrons are slowed near the tool.
Level 2: Part B Style
Practice Q1. A clean, water-bearing sandstone has \( \Delta t = 86 \) µs/ft. With \( \Delta t_{ma} = 55.5 \) and \( \Delta t_f = 189 \) µs/ft, the Wyllie sonic porosity is closest to:
(a) 0.16 (b) 0.23 (c) 0.35 (d) 0.45
Answer: (b) 0.23
\( \phi_S = (86 - 55.5)/(189 - 55.5) = 30.5/133.5 = 0.228 \). (The simplified Raymer–Hunt–Gardner form with \( C = 0.625 \) gives \( 0.625 \times 30.5/86 = 0.222 \), close.) (c) divides by \( \Delta t \) without the constant (\( 30.5/86 = 0.355 \)). (d) is \( 86/189 \). (a) divides by \( \Delta t_f \) instead of \( \Delta t_f - \Delta t_{ma} \) (\( 30.5/189 = 0.161 \)).
Practice Q2. A quartz sandstone has bulk density 2.40 g/cm³ and fresh filtrate (1.0 g/cm³) in the flushed zone. Its density porosity is:
(a) 0.09 (b) 0.15 (c) 0.18 (d) 0.25
Answer: (b) 0.15
\( \phi_D = (2.65 - 2.40)/(2.65 - 1.0) = 0.25/1.65 = 0.152 \). (d) forgets the denominator. (c) uses the limestone matrix (\( 0.31/1.71 = 0.181 \)). (a) sets the fluid density to zero (\( 0.25/2.65 = 0.094 \)).
Practice Q3. In a well with a clean sand GR of 25 API and a shale GR of 125 API, a bed reads 75 API. The linear gamma-ray index and the Larionov (Tertiary) shale volume are:
(a) 0.50 and 0.22 (b) 0.50 and 0.33 (c) 0.60 and 0.22 (d) 0.75 and 0.50
Answer: (a) 0.50 and 0.22
\( I_{GR} = (75 - 25)/(125 - 25) = 0.50 \). Larionov Tertiary: \( 0.083(2^{3.7 \times 0.5} - 1) = 0.083(2^{1.85} - 1) = 0.083 \times 2.605 = 0.216 \). (b) uses the older-rock formula \( 0.33(2^{1} - 1) = 0.33 \). (c) computes the index as \( 75/125 \). (d) is \( 75/100 \).
Practice Q4. On a limestone-scaled neutron–density overlay, a low-GR interval shows density porosity well above neutron porosity. The most likely cause is:
(a) shale (b) dolomite (c) gas (d) anhydrite
Answer: (c) gas
Gas lowers bulk density (density porosity up) and lowers hydrogen index (neutron porosity down): crossover. Shale and dolomite both make the neutron read above the density. Anhydrite reads very low (even negative) density porosity and low neutron porosity.
Level 3: Part C Style
Practice Q5. Two low-GR carbonate intervals are logged with a limestone-scaled neutron and a density tool (fluid density 1.0 g/cm³):
| Interval | \( \rho_b \) (g/cm³) | Pe (b/e) | \( \phi_N \) (limestone units) |
|---|---|---|---|
| X | 2.55 | 5.1 | 0.09 |
| Y | 2.70 | 3.1 | 0.14 |
Which statement is correct?
(a) X is limestone with porosity about 9%; Y is dolomite with porosity about 9% (b) X is limestone with porosity about 9%; Y is a tight limestone with porosity under 1% and shale contamination (c) X is dolomite with porosity about 17%; Y is limestone with porosity about 1% (d) Both are limestone; Y has porosity about 14%
Answer: (a)
X: Pe 5.1 is calcite. Limestone density porosity \( = (2.71 - 2.55)/1.71 = 0.094 \), agreeing with the neutron’s 0.09: clean limestone, about 9%.
Y: Pe 3.1 is dolomite. On a limestone scale the density porosity is only \( (2.71 - 2.70)/1.71 = 0.006 \) while the neutron reads 0.14: a large neutron-above-density separation. With low GR that is the dolomite signature, not shale. With the correct matrix, \( \phi_D = (2.87 - 2.70)/1.87 = 0.091 \), about 9% (the limestone-scaled neutron over-reads in dolomite).
(b) reads the separation as shale despite low GR and ignores Pe. (c) swaps the lithologies. (d) ignores Pe and takes the neutron at face value.
Practice Q6. In a carbonate, neutron–density porosity is 0.20 and sonic porosity is 0.14. Consider these statements:
- The secondary porosity index is 0.06.
- The difference suggests vugs or fractures that the sonic first arrival bypasses.
- If the zone contained gas, the neutron–density porosity should be computed as the root-mean-square of the two readings.
- The sonic porosity is the better estimate of total porosity in this rock.
Which are correct?
(a) 1 and 2 only (b) 1, 2 and 3 only (c) 2 and 4 only (d) 1, 3 and 4 only
Answer: (b) 1, 2 and 3 only
- True. SPI \( = 0.20 - 0.14 = 0.06 \).
- True. The sonic responds mainly to intergranular porosity; vugs and fractures are seen by density and neutron.
- True. In gas zones \( \phi \approx \sqrt{(\phi_N^2 + \phi_D^2)/2} \).
- False. Total porosity includes the vugs; the neutron–density value is the better estimate of it.
Practice Q7. A sandstone shows \( \phi_N = 0.10 \) and \( \phi_D = 0.30 \) (both sandstone-scaled), GR 28 API (clean). Which option gives the best porosity estimate and the fluid?
(a) 0.20, oil (b) 0.22, gas (c) 0.30, gas (d) 0.10, water
Answer: (b) 0.22, gas
The crossover (density porosity far above neutron) in a clean sand indicates gas. \( \phi \approx \sqrt{(0.10^2 + 0.30^2)/2} = \sqrt{0.05} = 0.224 \). (a) uses the simple average and misses the gas. (c) and (d) take one tool at face value.
Quick Revision Table
| Idea | Fact or formula |
|---|---|
| Wyllie | \( \phi_S = (\Delta t - \Delta t_{ma})/(\Delta t_f - \Delta t_{ma}) \) |
| Raymer (simplified) | \( \phi_S \approx C(\Delta t - \Delta t_{ma})/\Delta t \), \( C \approx 0.625\text{–}0.70 \) |
| Transit times | sandstone 55.5, limestone 47.5, dolomite 43.5, salt 67, fluid 189 µs/ft |
| Velocity | \( v = 304{,}800/\Delta t \) m/s |
| Density porosity | \( \phi_D = (\rho_{ma} - \rho_b)/(\rho_{ma} - \rho_f) \) |
| Matrix density | 2.65, 2.71, 2.87, anhydrite 2.98 |
| Pe | sandstone 1.8, dolomite 3.1, limestone 5.1; barite mud swamps it |
| Density physics | Cs-137, Compton scattering, electron density, \( \rho_a = 1.0704\rho_e - 0.1883 \) |
| Neutron physics | Am-Be; hydrogen slows neutrons; hydrogen index; limestone units |
| Gas | density porosity high, neutron low: crossover; \( \phi \approx \sqrt{(\phi_N^2 + \phi_D^2)/2} \) |
| Shale | neutron high, high GR |
| SPI | \( \phi_{ND} - \phi_S \) |
| GR index | \( I_{GR} = (GR - GR_{\min})/(GR_{\max} - GR_{\min}) \) |
| Larionov | Tertiary \( 0.083(2^{3.7I} - 1) \); older \( 0.33(2^{2I} - 1) \) |
| Spectral GR | K-40 1.46, U (Bi-214) 1.76, Th (Tl-208) 2.61 MeV |
| GR shapes | cylinder, bell (fining up), funnel (coarsening up), serrated |
Key Takeaways
What’s Next
- Formation Evaluation: Formation Factor, Porosity, Permeability, Saturation and Lithology: porosity from this page and \( R_t \), \( R_w \) from the electrical logs combine into saturation, permeability and lithology.
- Electrical, SP and Temperature Logs: the resistivity half of the open-hole suite.
- Petroleum Geology and Indian Basins: reservoirs and source rocks that these logs evaluate.
- Well Logging: back to the chapter overview.