NivaarExam PrepOfficial exam papers ↗

24-Pet-B1 Natural Gas Engineering · December 2015

Question 7 of 10: Neutron-Density Porosity and Sxo in a Gas Sand

Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)

Notes on this paper

National Exams, 98-Pet-B1, Well Logging and Formation Evaluation — December 2015, 3 hours, closed book (approved calculators permitted), 10 questions, all marked.

Reference texts: Bassiouni, Theory, Measurement, and Interpretation of Well Logs (SPE Textbook Series Vol. 4); Asquith & Krygowski, Basic Well Log Analysis, 2nd ed. (AAPG); Ellis & Singer, Well Logging for Earth Scientists, 2nd ed.; Schlumberger, Log Interpretation Charts / Log Interpretation Principles and Applications.

Question 7: Neutron-Density Porosity and Sxo in a Gas Sand (15 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Given. The paper supplies the gas-sand crossplot on the Attachment page (density porosity φD against neutron porosity φN, contoured for porosity and for Sxo), and that chart prints its own fluid and matrix constants, so none of them has to be assumed:

QuantityValueWhere it comes from
Apparent neutron porosity, φNa (sandstone matrix)5%question
Bulk density, ρb2.0 g/cm³question
Matrix density, ρma2.65 g/cm³printed on the attachment chart
Mud-filtrate density, ρmf1.00 g/cm³printed on the attachment chart (fresh mud)
Gas density, ρg0 g/cm³printed on the attachment chart ("low density" gas)

Find. True porosity φ and flushed-zone filtrate saturation Sxo, both without and with the neutron excavation-effect correction.

Approach. In a gas-bearing flushed zone the two porosity logs see two different mixtures of the same three components (matrix, filtrate, residual gas), so they give two independent equations in the two unknowns φ and Sxo — no resistivity log is needed. Without the excavation effect the pair solves in closed form. With it, the neutron equation picks up the Segesman–Liu excavation term and no longer inverts by hand, which is exactly why the paper supplies the chart: it is that same pair of equations drawn out, so the answer is read off it.

  1. Apparent density porosity from ρb. Scaling the density log on the chart's own fluid density (ρf = ρmf = 1.00): $$\phi_{Da}=\frac{\rho_{ma}-\rho_b}{\rho_{ma}-\rho_{mf}}=\frac{2.65-2.0}{2.65-1.00}=\frac{0.65}{1.65}=\boxed{39.4\%}$$ The pair (φNa, φDa) = (5.0, 39.4) p.u. is the entry point on the chart, and the 34-p.u. separation is the classic gas signature: gas's near-zero hydrogen index makes the neutron tool under-read, while its near-zero density makes the density tool over-read.
  2. The two tool responses, written out. With the flushed zone holding filtrate (Sxo) plus residual gas (1−Sxo), and ρg = 0 so the gas carries neither mass nor hydrogen: $$\rho_b=(1-\phi)\rho_{ma}+\phi S_{xo}\rho_{mf}+\phi(1-S_{xo})\rho_g \;\Longrightarrow\; \rho_b=\rho_{ma}-\phi\,\rho_{ma}+\phi S_{xo}$$ $$\phi_{Na}=\phi S_{xo}-\Delta\phi_{ex},\qquad \Delta\phi_{ex}=\left(2\phi^2S_{xo}+0.04\,\phi\right)\!\left(1-S_{xo}\right)$$ Δφex is the excavation effect: the gas-filled part of the pore space not only contains no hydrogen, it also removes matrix atoms that would themselves have slowed neutrons, so the neutron tool under-reads by MORE than the hydrogen-index deficit alone. Setting Δφex = 0 gives the "without excavation effect" case.
  3. φ and Sxo WITHOUT the excavation effect. With Δφex = 0 the neutron equation is simply φSxo = φNa = 0.05, and substituting that product straight into the density equation eliminates Sxo: $$\rho_b=\rho_{ma}-\phi\rho_{ma}+\phi_{Na}\quad\Longrightarrow\quad \phi=\frac{\rho_{ma}-\rho_b+\phi_{Na}}{\rho_{ma}}=\frac{2.65-2.00+0.05}{2.65}=\frac{0.70}{2.65}=\boxed{26.4\%}$$ $$S_{xo}=\frac{\phi_{Na}}{\phi}=\frac{0.05}{0.2642}=\boxed{18.9\%}$$
  4. φ and Sxo WITH the excavation effect — read from the attachment chart. Entering the chart at φN = 5.0 p.u. on the horizontal axis and φD = 39.4 p.u. on the vertical axis, the point falls between the Sxo = 30% and Sxo = 40% curves (about one-third of the way across) and between the 25% and 30% porosity contours (about one-third of the way up): $$\phi\approx\boxed{26.6\%},\qquad S_{xo}\approx\boxed{33\%}$$
  5. Cross-check on the chart reading. Substituting φ = 0.266, Sxo = 0.333 back into the excavation term reproduces the logged neutron reading: $$\Delta\phi_{ex}=\left(2(0.266)^2(0.333)+0.04(0.266)\right)(1-0.333)=0.0385$$ $$\phi_{Na}=\phi S_{xo}-\Delta\phi_{ex}=0.0885-0.0385=0.0500\;\checkmark$$ Solving the same two equations numerically with the printed ρg = 0 gives φ = 28.0%, Sxo = 32.9% — the same saturation to within a point, and a porosity 1.4 p.u. above the chart, which is the chart's own construction tolerance.
QuantityWithout excavation effectWith excavation effect
Apparent density porosity, φDa39.4%
True porosity, φ26.4%≈ 26.6%
Flushed-zone filtrate saturation, Sxo18.9%≈ 33%
Check: the "with excavation effect" pair is a chart reading, good to roughly ±1 p.u. in φ and ±3 saturation points; the "without" pair is exact arithmetic on the chart's own printed constants. The lesson the question is built around is the contrast between the two columns: ignoring the excavation effect barely changes the POROSITY (26.4% vs. 26.6%) but nearly halves the SATURATION (18.9% vs. 33%). Because the excavation term makes the neutron tool under-read still further, crediting all of that deficit to "less filtrate" instead of "excavation" leaves a residual-gas saturation that is far too high — and residual gas is what an Sxo/Sw movable-hydrocarbon analysis turns on.