What a microresistivity image actually shows
A borehole formation microresistivity image, often described using the term FMI, is a high-resolution map of electrical variation along the borehole wall. It is not an ordinary photograph. It does not directly record fracture aperture, porosity, salinity, permeability or the chemical identity of a mineral. To read it well, start with the button-electrode measurement, the scale of the colour display, the orientation and gaps between pads, and the physical conditions needed for electrical contact. Then ask what interpretation of each pattern can be independently tested. This guide turns bright and dark image textures into a disciplined sequence of observations, candidate explanations and quantitative checks.
The immediate practical questions are often whether a dark sinusoid represents a conductive fracture and whether repeated bright and dark bands record depositional layers. A useful answer requires more than memorising that dark equals conductive. Display palettes can be inverted, dynamic normalisation changes the meaning of brightness between depth windows, and clay-rich rock can look conductive without being an open fluid pathway. A repeatable structural geometry may be established even when the hydraulic meaning remains unknown. Distinguish those two kinds of result before the data are used to make completion, geological or environmental decisions.
Think of the word translate as a constraint. An instrument returns currents and inferred local electrical contrasts; processing turns them into image pixels; interpretation turns pixels into language. Every step needs an explicit correspondence. The guide to meaning and context in translation offers a parallel: an attractive word substitution is not necessarily a faithful sentence. Likewise, a dark image feature is not automatically an open fracture. The surrounding rock type, tool response and independent evidence determine how much the word fracture can mean.
The physical measurement: button electrodes and current
Formation microresistivity tools press arrays of small electrodes, commonly called buttons, against the borehole wall using expandable pads. Applied voltage drives current through the adjacent conductive path toward a return electrode. The response depends on electrical conductivity along the current path, pad contact, mud properties and rock near the wall. A button that draws more current in a comparable measurement geometry generally encounters a more conductive path. Multiple buttons arranged around the instrument collect narrowly spaced samples as the tool moves along the borehole, creating a depth-by-azimuth map.
It is useful to separate resistance from resistivity before looking at the image. Electrical resistance in ohms describes a complete path and includes geometry. Material resistivity in ohm-metres is a geometric property under the relevant idealisations. A microresistivity image is often a processed representation of relative conductance or resistivity variation around the wall, not a direct series of laboratory resistivity specimens. Tool manufacturers apply calibrations and focusing methods, so generic current-to-colour arithmetic must not replace the instrument’s own response model. The first thing to read in a real report is the measurement scale, not the visual texture.
Because the buttons investigate a relatively shallow region, features that are small and close to the wall can appear clearly. That advantage comes with a limitation: a shallow, high-resolution electrical picture cannot establish the behaviour of a structure metres away. Saline mud invasion, alteration near the wall and disturbed surfaces may influence the image. A beautiful microresistivity pattern could therefore be an excellent record of a disturbed borehole annulus but a poor stand-in for undisturbed formation properties. Add an invasion and borehole-condition check whenever a claim extends beyond the sampled region.
Unroll the cylinder before interpreting the pattern
Each electrode samples the wall at a known relative position. The combined image is displayed as if the cylindrical borehole surface had been cut lengthwise and flattened. The vertical axis represents measured depth, while the horizontal axis represents azimuth when orientation information is available. Many display systems use north at the left edge, then progress through east, south and west before returning to north at the other edge, but this is a convention to verify, not assume. Misreading the azimuth axis can rotate a reported fracture direction by tens or hundreds of degrees.
Unlike a camera with continuous visual coverage, a pad-based device can leave unsampled strips between pads. Those strips may be displayed as pale blanks, interpolation bands, stretched pixels or another colour assigned by software. Do not count an unsampled region as uniform rock, and do not claim continuity of a narrow feature through a gap without stating the interpolation. Coverage can vary with borehole diameter, pad deployment and local wall roughness. A feature that appears discontinuous only at a pad boundary deserves a processing check before a geological explanation is preferred.
The image should be reviewed with depth, logging speed, tool inclination, azimuth reference and borehole-diameter information. If the sonde is in a deviated well, the displayed sinusoid geometry must be corrected from borehole-relative coordinates to earth coordinates. If a conductive band wraps through the image seam, the left and right edge portions belong to the same physical neighbourhood. Before measuring dip or counting fractures, imagine the flat display wrapped back into a cylinder. That simple reversal catches many duplicated or truncated feature interpretations.
How a sinusoid appears on a resistivity image
A planar surface intersecting a cylindrical borehole traces a curve that becomes sinusoidal on an unrolled azimuth-by-depth image. In an ideal vertical circular hole, the peak-to-trough depth excursion H across the full circumference and borehole diameter D give the geometric relation tan(dip) approximately H/D. The dip is measured from horizontal. This relation is a useful conceptual check, not an adequate substitute for full orientation correction in deviated wells. The tool may acquire dip from fitted curves using trajectory data, selected pixels and statistical processing rather than a hand-drawn measurement.
Worked example: in a 0.120 m diameter vertical borehole, a continuous conductive sinusoid varies through 0.090 m in measured depth from its shallowest to deepest point. The dimensionless ratio is 0.090/0.120 = 0.75. The idealised dip is arctan(0.75), approximately 36.9 degrees. Reverse calculation gives 0.120 m times tan(36.9 degrees), about 0.090 m. Both lengths refer to the same physical units and the full sinusoid excursion. The reverse check verifies arithmetic; it does not prove that the pixels identify one geological plane.
A reader who uses the radius instead of the diameter obtains arctan(1.50), about 56.3 degrees, which is a large error made from apparently simple numbers. Another common mistake is deriving dip from the angle of a line on a stretched screenshot rather than the labelled depth excursion. A third is treating image trough direction as geographic dip direction when the panel is not correctly oriented. Report the geometric inputs and correction convention beside the result so that another interpreter can reproduce or challenge it.
Colour palettes, static scales and dynamic normalisation
On common displays, conductive intervals are shown darker and resistive intervals brighter, but some palettes invert that mapping. A resistivity curve and a colour-bar legend must accompany interpretation. More subtly, a dynamically normalised image may remap local highs and lows within a moving depth window. A dark pixel in one window need not represent the same absolute resistivity as a dark pixel several metres away. This processing is useful for seeing fine structure across variable background lithology, yet it is dangerous for quantitative comparisons.
Suppose one window represents a relative scale of 2 to 20 arbitrary units and another represents 20 to 200. A feature at 15 may appear bright in the first but dark in the second if each is stretched locally. The plotted colours would differ even if the underlying values were compared on a consistent physical scale. The numbers in this illustration are hypothetical display units, not calibrated rock resistivities. The correct response is to inspect the underlying channel, colour mapping and normalisation settings before interpreting relative brightness as a change in material composition or pore fluid.
A static normalisation allows comparison across depth where the response and calibration support it, but may obscure subtle fine structure in intervals dominated by very conductive or very resistive material. Dynamic normalisation highlights bedding and fractures but can manufacture visual equivalence between intervals with very different baseline values. Use both representations where possible and explain which conclusions depend on image texture versus a calibrated bulk log. If a report states ‘more conductive at 80 m than 120 m’, that claim needs a common electrical reference, not two separately stretched colour panels.
Recognise conductive fractures without mistaking every dark line for flow
A conductive feature can result from an open, fluid-filled discontinuity, a clay-lined fracture, an altered seam or a thin mineral band. A dark line on an FMI panel is evidence of a relative electrical response at the wall, assuming the palette is understood. It does not uniquely identify water movement. Saline fluids often increase electrical conductivity, but conductive clay surfaces can produce a similar appearance without a hydraulically open crack. Some natural fractures may be resistive when filled or cemented. Qualify the observation before moving from appearance to geology.
Trace continuity, orientation and context are useful. Does the dark line produce a coherent sinusoid around enough of the borehole circumference? Does it cut across bedding rather than follow a depositional contact? Is there a corresponding discontinuity in core, optical or acoustic imagery? Does the feature occur within a zone of borehole breakouts where pads may have poor contact? A convincing interpretation often combines several modest observations rather than one dramatic dark stripe. Where the feature is only visible on a single pad, the proper label may be candidate fracture trace with incomplete circumferential coverage.
Even a well-supported geological fracture does not have a measured transmissivity until hydraulic data support it. The heat-pulse flowmeter article explains how vertical flow evidence differs from electrical imaging, and the fluid temperature and conductivity guide explains complementary inflow clues. If a fracture image and a flow anomaly appear near one depth, first check datum alignment and interval resolution. Proximity strengthens a hypothesis only after it is established that both observations refer to the same physical zone.
Read bedding contacts and sedimentary texture
Not every sinusoid is a fracture. Thin beds, cross-bedding, erosional contacts and lithological boundaries may form continuous geometric traces on the borehole surface. A succession of similarly dipping sinusoids may record stratification. Cross-cutting dark lines can indicate fractures or later structures, but their geological identity still needs context. The most useful visual questions are how the traces relate to one another, whether they are repeated and whether their internal patterns change systematically. A reader who labels every curved interface as a fracture loses the depositional history the image may contain.
Microresistivity textures can reflect changes in grain size, cementation, clay content, fluid composition and pore distribution. Thin laminated sediment may produce fine parallel stripes, while heterogeneous carbonate or volcanic textures may create irregular patches. These are interpretations of electrical contrast, not direct microscopic observations of grains. An analyst might use calibrated core and thin sections to associate an image texture with a lithofacies, then apply that association cautiously where core is missing. The calibration should state how many reference intervals support the relationship and whether they represent similar conditions.
Suppose a logged interval has twenty thin dark laminae over 0.50 m of measured depth. An average visible spacing of 0.50/20 = 0.025 m, or 25 mm, is only an approximate depth-axis spacing. If the beds dip, their true normal spacing differs from vertical spacing and may require a cosine correction under a suitable model. If layers are irregular or some are obscured, counting and division yield an observational descriptor rather than a guaranteed geological bed thickness. The difference matters when a depositional model is built on millimetre-scale interpretations.
Pad coverage and apparent gaps
A microresistivity pad images only the wall that its electrodes sample. Four or more pads may provide substantial circumferential coverage, but a full uninterrupted 360-degree image should not be assumed. The blank sectors between pads can hide small fractures, lithological contacts or other fine structures. Interpretation software may interpolate across them to make a smoother display; that mathematical gap fill is not a measurement. Users should inspect a coverage plot or raw pad channels, especially where a conclusion depends on a line apparently crossing an unsampled region.
Imagine a theoretical instrument whose sampled wall sectors total 280 degrees of 360 degrees. Its angular coverage fraction is 280/360 = 0.778, or about 77.8 percent. That simple number does not mean 77.8 percent of geological fractures are detected, because their orientations, widths and contrasts affect visibility. It describes instrument coverage, not detection probability. A 10-mm feature could fall entirely between pads at one azimuth even when most of the circumference is measured. Treat coverage and feature detection as different quantities.
Coverage may decline in enlarged holes when pads do not contact the wall well. A caliper measurement can reveal washouts and ovality that help explain poor image quality. The borehole caliper guide provides the supporting geometry. If a report records few fractures in a large washout interval, ask whether the count reflects intact rock or inadequate measurement. Missing electrical contact should be flagged in the same feature table as image confidence; otherwise downstream readers may mistake blanks for negative findings.
Borehole conditions can dominate the response
Conductive mud is generally important for galvanic microresistivity measurements because a usable current path is required. Excessively resistive fluid, poor pad contact or a rough wall may degrade the signal. Drilling-fluid invasion alters the near-borehole saturation and chemistry, while mud cakes and filter cakes modify contact. A feature may appear electrically dark because the near-wall material has different fluid content from the deeper rock. These possibilities must be considered before image brightness is used to infer undisturbed formation properties.
Pad pressure and borehole geometry also matter. A tool in an oval or inclined hole may contact different walls unequally. An irregular section can concentrate contact in a few points and leave poor measurements elsewhere. If repeated button channels show systematic streaks at pad boundaries, suspect acquisition or processing artefacts. Check the caliper, acquisition diagnostics and repeat logging when uncertainty matters. A good geological model cannot repair an unacknowledged instrument-contact failure merely by fitting a plausible sinusoid to noisy pixels.
Depth registration can compound the difficulty. A microresistivity image may be recorded on a continuous logging pass while core is measured in physical pieces and another geophysical sonde reports depth from a different reference. Before declaring correspondence, verify common datums and apply documented corrections. A 0.20-m mismatch could place a thin conductive zone beside the wrong core fracture. Do not quietly drag curves until one interpretation looks satisfying. Preserve raw depth markers and write down the logic for every shift used in integration.
Worked interpretation: dark trace, bright rock, uncertain permeability
Consider a hypothetical 110-mm-diameter hole with a continuous dark sinusoid from 47.20 to 47.26 m. The 60-mm peak-to-trough excursion implies H/D = 60/110 = 0.545. The simplified vertical-hole dip is arctan(0.545), about 28.6 degrees. If corrected azimuth marks place the low point near 210 degrees, one may report a candidate planar discontinuity dipping approximately 29 degrees toward 210 degrees, subject to navigation quality. The geometry comes from the trace shape and measured diameter, not from its dark colour.
Now suppose optical imaging shows a thin pale vein at nearly the same depth and no evident opening. That evidence raises the possibility of a mineralised or altered plane whose microresistivity response is nevertheless conductive. If a hydraulic flowmeter has no detectable change at this depth under ambient conditions, the most defensible statement is that a conductive planar feature is present but its present-day hydraulic activity is undetermined. No flow response may reflect the absence of a sufficient gradient, limited detection sensitivity or isolation from the main conducting network, not necessarily a sealed plane.
Reverse-check the geometry: tan(28.6 degrees) is about 0.545; 110 mm times 0.545 equals 59.95 mm, in agreement with the observed 60 mm. Next check whether the trace can be followed across enough pads to support a unique sinusoid. Finally inspect the mud properties and dynamic scaling to make sure the darkness is physically comparable across the interval. A numerical result is valuable only after image coverage and electrical meaning pass these separate checks. This three-layer verification prevents a neat dip calculation from masquerading as proof of a flowing fracture.
Worked interpretation: a normalisation trap
Suppose a borehole microresistivity panel has two depth windows, one with a background raw value near 10 arbitrary response units and another near 100. Each window is dynamically stretched so that its local lower decile is dark and upper decile is pale. A local line reading 12 in the first window may appear visually similar to a line reading 120 in the second even though the raw magnitudes differ tenfold. The colours are comparable in rank within each window, not in absolute electrical response across windows.
If a report calls the two lines ‘equally conductive’, the equivalence is unjustified. To compare electrical behaviour one must examine calibrated measurements or a common fixed scale, taking tool-specific corrections into account. The arithmetic 120/12 = 10 reveals the raw-value ratio in this hypothetical instrument channel but is not automatically a tenfold formation-conductivity ratio. A manufacturer’s processing definition determines whether and how current readings map into calibrated electrical properties. The correct written claim is that both lines contrast with their local backgrounds, and the absolute comparability remains unresolved.
This error is common in many data visualisations. A colour can encode a local rank, a transformed number, or a physical quantity with units. The legend and algorithm decide. Before transferring a visual conclusion to another discipline, ask what the colour actually represents and whether the scale is constant. Good technical vocabulary prevents phrases such as similar appearance, similar measured response and similar material property from being used interchangeably. Each phrase lives at a different level of evidence.
A different worked case: laminations rather than fractures
Imagine a 3-m interval with thirty nearly parallel conductive and resistive stripes that form repeated gentle sinusoids, while recovered core shows interbedded fine sandstone and shale. The image supports a layered depositional structure; it does not justify counting thirty hydraulically open fractures. If eight of the stripes are judged to represent distinctive geological bed contacts, their feature frequency is eight divided by 3 m, approximately 2.67 contacts per metre. That statistic describes a selected class of interpreted boundaries and should not be silently compared with fracture frequencies from another well.
The same interval might contain two steep sinusoids that cross the layered pattern. Those two may be candidate fractures, giving a separate fracture-trace frequency of 2/3 = 0.67 per metre, provided the image coverage and selection rules support both. Reporting both sets separately preserves different geological questions. Combining ten total lines into one count of 3.33 per metre produces a higher number but destroys the interpretation. The units match, yet the nouns do not. A meaningful denominator therefore needs a meaningful numerator.
Suppose one crossing sinusoid is missing for 90 degrees of azimuth because of an unsampled pad gap. That loss of coverage reduces the confidence of its fitted orientation; it does not automatically make the feature invalid. The analyst can fit the visible portions, quantify uncertainty, seek an acoustic or optical counterpart and flag the result as provisional. Declaring a perfect angle from a partly unseen trace exaggerates certainty; deleting every incomplete feature may introduce bias. The appropriate middle course is to retain tentative evidence with a visible confidence label.
Fracture aperture, electrical width and hydraulic aperture
An image may allow a dark feature to be assigned a pixel width. That apparent electrical width is not automatically the physical opening of a fracture. Button-electrode size, sampling, pad contact, the fluid conductivity contrast and image processing all influence how broad the response appears. A sharp, very thin geological feature may be widened by the instrument’s point-spread response; a wide irregular opening may be represented by several different electrical textures. Quantitative aperture claims need appropriate calibration or independent geometric observations.
For illustration, a dark stripe spans six pixels on a display with nominal vertical sampling of 2 mm per pixel, suggesting 12 mm of displayed depth-axis width. That product, six times 2 mm, is mathematically correct, but it describes apparent signal extent along the depth axis. It is not a 12-mm fracture aperture normal to the plane. The plane’s dip and orientation, the tool’s spatial response and the geometry of the feature would all matter in a physical-width calculation. Never use a convenient pixel conversion to label an unmeasured mechanical quantity.
Hydraulic aperture is different again. It is a conceptual opening inferred from flow behaviour under a chosen model, and can differ greatly from physical openings in rough, partially sealed fracture networks. Transmissivity integrates how effectively a zone transmits water; electrical conductance measures charge transport. The words contain a similar root but their physical mechanisms differ. The vocabulary distinction is central to interpreting one observation without claiming a second that was never measured. The semantic feature article explains how slight differences between terms change the meaning of a statement.
Orientation correction and uncertainty
Dip direction from a microresistivity image requires reliable orientation data, commonly tool heading and well deviation. If the well is vertical and north registration is secure, the low point of a planar sinusoid supplies an interpretable dip direction under the stated display convention. If the hole deviates or the magnetic reference is disturbed, correction is required. In an inclined borehole the displayed curve expresses the plane’s relationship to the borehole axis, not automatically to the horizontal geographic reference system.
Suppose a feature is measured at an apparent 40-degree dip in borehole coordinates but the hole itself is inclined 25 degrees toward a different azimuth. It would be wrong to add or subtract 25 degrees without considering the three-dimensional directions. Planar orientations require vector geometry or appropriate structural software and the actual borehole trajectory. The error may be moderate or severe depending on relative azimuth. This is a conceptual rather than a numerical shortcoming: two angles do not uniquely define a correction when the orientation of their planes is ignored.
For consequential reports, include the reference north definition, inclination data source, correction method and orientation uncertainty. If the tool crosses magnetically disturbed rock, a magnetic heading may be unreliable even when image quality is excellent. Comparisons with core orientation, gyroscopic surveys or independent structural observations may be needed. The sound conclusion might be ‘trace confidently detected; geographic strike and dip uncertain’ rather than a detailed but unsupported orientation rose diagram.
Integrate with gamma, sonic, acoustic and optical records
The strongest interpretation often places microresistivity alongside other measurements that respond to different physical properties. A natural gamma log may help distinguish clay-rich from cleaner lithological intervals; an acoustic televiewer may resolve wall roughness and reflectivity; an optical televiewer can supply colour and texture in clear conditions; a sonic log evaluates seismic slowness and related wave behaviour. Agreement can narrow alternative explanations, while disagreement can reveal a processing problem or a physically meaningful contrast. The goal is convergence on a well-defined question, not unanimous-looking plots.
For example, a thin dark microresistivity line within a high gamma interval may be related to clay-rich fabric rather than a large open flow conduit. A co-located acoustic-wall irregularity could support a discontinuity, yet the hydraulic role remains a separate test. The natural and spectral gamma article develops the mineralogical caveat, and the acoustic televiewer guide explains a complementary image type. Do not merge different instrument signals into a single unsupported image story.
Depth matching is essential. A core fracture at 75.4 m and an electrical trace at 75.6 m might correspond after documented depth correction, but one cannot simply claim that they do. Establish top-of-casing reference, logging direction, tool spacing and survey registration. Review more than one nearby landmark and preserve original depths. Where separate probes have different vertical responses, a broad anomaly should not be forced to match a sharp thin line exactly. Describe the measurement footprints and the level of positional confidence.
Build a transparent feature inventory
An auditable feature inventory is more useful than an unlabeled montage of colour panels. For each accepted trace, record its identifier, depth, available azimuth coverage, image processing, contrast polarity, geological classification, confidence, estimated orientation and independent observations. Include a reason for exclusion when a tempting line is rejected as pad artefact, drilling mark or unreliable coverage. This structure allows another analyst to reproduce the logic and to revise one interpretation without destroying the whole record.
A fracture-density calculation must include its denominator and completeness rule. Suppose eight accepted candidate natural fractures occur in 12 m of good-quality image. The apparent image-trace frequency is 8/12 = 0.67 per readable metre. If the total borehole interval is 15 m but the remaining 3 m is unusable, reporting 8/15 = 0.53 per total interval metre answers a different question. Neither number describes the complete three-dimensional fracture network. Reporting both alongside coverage and classification is more honest than selecting whichever gives the preferred narrative.
When images are assessed by multiple interpreters, compare matches at specific depths rather than only totals. Two reviewers could each identify eight traces but disagree about half of them. A consensus should record which markings are observable, which labels remain provisional and what evidence would resolve disagreement. Fitting curves to the same pixels can produce apparently precise angles while retaining a shared mistaken assumption that the marks are natural fractures. Independent review is valuable only when competing explanations can actually be considered.
Common mistakes and repairs
Mistake one is equating dark with open, water-bearing fracture. Repair it by identifying the response scale and checking with optical, acoustic and hydraulic evidence. Mistake two is comparing dynamic-image brightness across separate depth windows. Repair it with a common response scale or calibrated measurement. Mistake three is interpreting pad gaps as solid rock. Repair it by checking azimuth coverage and acquisition diagnostics. These three errors have different causes, yet all arise when a visual output is mistaken for a direct observation of a richer physical property.
Mistake four is calculating dip from a stretched screen image. Use measured excursion and diameter in the same units, or trajectory-corrected software outputs. Mistake five is treating apparent electrical width as physical fracture aperture. Write ‘response width’ until a suitable independent geometric or hydraulic measurement is available. Mistake six is mixing feature categories in a frequency statistic. Label numerator, denominator, quality rule and geological class. An apparently small wording correction often prevents a serious downstream engineering misunderstanding.
Mistake seven is quietly correcting depth offsets until logs appear to agree. Document raw depth, reference datums and the evidence for a registration shift. Mistake eight is trusting a single processed image without checking original channel behaviour or pad contact. Preserve raw data and acquisition conditions. Mistake nine is assuming higher visual resolution means greater depth of investigation. Electrode buttons may see fine wall textures while offering little evidence about connections metres away. A source of apparent clarity can be a source of overconfidence unless its measurement footprint is understood.
Independent transfer problems
Problem one: an ideal vertical hole is 100 mm wide and a candidate planar trace spans 80 mm vertically. Find the simplified dip and reverse-check. Problem two: a tool records 300 degrees of usable circumferential coverage. Find its coverage fraction and explain why it cannot automatically be called the percentage of fractures detected. Problem three: an interval contains fifteen candidate fractures over 25 m of good image. Calculate the apparent frequency and state what this statistic excludes. Problem four: two dynamically scaled panels contain equally dark lines, but different raw backgrounds. State what can and cannot be concluded.
Answers: arctan(80/100) = arctan(0.80), approximately 38.7 degrees. Reversing, 100 mm times tan(38.7 degrees) is close to 80 mm. Coverage is 300/360 = 0.833, or 83.3 percent of circumference under the simplified angular calculation; detection efficiency depends on feature width, contrast and geometry. The frequency is 15/25 = 0.60 interpreted traces per metre of readable borehole. That is not an unbiased measure of fractures outside the hole or below instrument resolution. For problem four, equal display darkness is evidence only of a selected colour mapping; calibrated electrical comparability requires an appropriate common scale.
Problem five: core shows a healed calcite-filled seam at a depth where the FMI image records a dark sinusoid, while ambient flow logs show no change. Write one defensible sentence. A suitable answer is: ‘A conductive planar image feature coincides approximately with a calcite-filled discontinuity; an open flowing fracture is not established by the available core and ambient-flow observations.’ The explanation names the measurement, the correspondence, the competing evidence and the remaining uncertainty. It is stronger than the confident but unsupported sentence ‘A highly permeable fracture has been found.’
Frequently asked questions
Is FMI a photograph? No. It is an electrical image generated from many small electrode measurements and processed according to the tool response and display settings. Can the image reveal bedding orientation? Often yes, where planar traces are sufficiently visible and reliable borehole orientation and geometry are available. Can it find a fracture that does not intersect the hole? It is primarily a near-wall imaging technique; it cannot reliably map remote fractures merely because its local wall image is detailed. Are dark areas always saline water? No. Clay, alteration, mud contact and other conductive pathways can also affect the measurement.
Can microresistivity imaging work without conductive contact? The galvanic measurement depends on a suitable electrical path, so fluid, wall condition and pad contact can limit applicability. Can a conductive image reveal permeability directly? No. Permeability and transmissivity need hydraulic evidence or carefully qualified models. Why do some parts of an image look blank? Pad spacing, poor contact, borehole enlargement or processing may create unsampled or unreliable sectors. Why are some traces broken at image edges? The image is an unrolled cylinder and the left-right edges may meet at a seam.
Can two logs have different colours and still agree? Yes, if their palettes or normalisation differ. Conversely two visually similar panels can encode very different raw responses. Why include an image-quality grade with each geological feature? Because a provisional interpretation should remain provisional when it enters a database or later engineering report. The practical reading habit is to record measurement, geometry, interpretation and test separately. That separation lets future evidence refine a conclusion without disguising which parts were observed in the first place.
A compact procedure for real field decisions
Begin with the actual decision: identify thin bedding, classify candidate fractures, evaluate borehole-wall integrity or support hydraulic-zone selection. Verify the tool model, conductive fluid, pad deployment, coverage, depth datum and orientation correction. Review static and dynamic displays and raw channel behaviour. Mark observable traces without first assigning fluid or geological labels. For each planar candidate, fit the visible sinusoid, check the seam, calculate orientation under the correct borehole trajectory and state the angular confidence. Distinguish dark relative conductance from physical aperture and from hydraulic transmissivity.
Then connect the observations to core, gamma, caliper, optical, acoustic and fluid-flow evidence. Keep each source’s depth registration, vertical resolution and uncertainty visible. Report an alternative explanation when a strong conclusion depends on a single image feature. Summarise the results in an inventory that another analyst can audit. For actions such as setting screen intervals, choosing perforation targets or designing further tests, state the independent evidence required before moving from candidate feature to final selection. A traceable interpretation is much more valuable than a long list of visually compelling labels.
Related reading includes normal resistivity and invasion, which concerns a different electrical measurement geometry, and borehole induction logging, which operates by electromagnetic induction rather than pad-contact galvanic imaging. For translating evidence into clear technical language, consult the context stack and the lexical-gap article. They reinforce the same habit: name only the distinction the evidence actually supports.
Primary methodological references
US EPA, Borehole Formation Microresistivity Imaging (FMI).