Skill target: Recognising evaporites using conventional well logs
Summary: Many evaporite properties can be ascertained by examining a suite of conventional well logs. Many evaporite beds contain only one or two dominant saline minerals; they lack free pore fluids and have negligible total porosity. This dramatically simplifies log interpretation and enhances the reliability of inferences with respect to mineralogy. Thick clean evaporite intervals show the same characteristic set of log responses, not only locally on a basin but worldwide (e.g., Serra, 1984; Warren, 2016). The most useful of the conventional well log suites for the study of evaporites are the tools measuring hole diameter, electrical properties, bulk density, neutron porosity, sonic, and if significant levels of potash salts are present, both gamma and multispectral gamma logs.
1. Basic set-up for wireline logging
2. Measurement while drilling
What is well logging? Cable tool vs MWD
Well-logs are a continuous recording of a geophysical parameter along a borehole, where the value of the measurement is continuously plotted against depth in the borehole. The well-logging industry is currently transitioning from wireline or cable-based well logging tools to the increasing use of well-log (MWD) tools designed for use in directional drilling. Wireline or cable tools are lowered on the end of a steel cable or wireline into the borehole and can only be utilised in vertical to steeply inclined wells (Figure 1). The same set of conventional well-log measurements are now increasingly collected using MWD (measurement while drilling) and LWD (logging while drilling (Figure 2).)
With MWD/LWD, geophysical measurements are made by a suite of well-logging tools that reside immediately behind the advancing drill-bit. Part of the data collected by these tools is sent to the surface in real-time (MWD) by mud pulsing or some other telemetry method. The remaining portion of the collected data (LWD) is stored on a hard disk and recovered typically when a worn drill-bit is pulled to the surface to be replaced.
Although there are numerous well-logging tools and measurements that can be used in the study of evaporites, our introductory discussion deals with only a few of the more conventional logging methods. For comprehensive analyses of the geological applications of well-logs both cable and MWD to the interpretation of evaporites and other lithologies, there are many logging-company manuals, as well as excellent books and articles such as Kruger (2014), Crain (2010, 2014), Ellis and Singer (2007), Nelson (2007), Rider (1996), Nurmi (1978), and Alger and Crain (1966), Crain and Anderson (1966).
Measurement parameters
The geophysical parameters measured by a set of well-logging tools include:
1) natural or spontaneous radioactivity using a gamma tool (GR); this can be the total gamma output from a formation, or the measurement of a rocks natural radioactivity. Outputs can be broken down into the proportions from the three main natural radioactive elements present in all rocks (Uranium, Potassium and Thorium) as measured by a spectral gamma tool.
2) The rocks response to induced radiation as in the Compton scattering range measured in a density tool (RHOB) or to neutron bombardment in the neutron tool (NPHI). 3) The time taken for the passage of sound waves through rock, as measured by a sonic tool (DT) or the rock's response to an induced passage of current, as in the resistivity and conductivity tools.
Table 1 summarises the pertinent ranges of these properties in evaporite salts, as well as other lithologies and brines.
Table 1. Typical well-log and physical properties of evaporite salts and associated sediments and brines. Compiled by the author from a variety of sources.
3. Schematic illustrating the conceptual basis for a Vclay or a Vshale calculation, using the Total Gamma log
4. Apparent K2O to Gamma response (API units) is linear (see also Table 1)
Total and spectral gamma log
The gamma-ray or gamma log is a record of the formation's radioactivity. The radiation emanates from uranium, thorium and potassium which occur naturally in the formation. A simple gamma-ray log measures the radioactivity of the three radiogenic elements (U, K, Th) combined. In contrast, the spectral gamma log shows the amount of each of the three radiogenic elements contributing to a formation's radioactivity.
As the first indicator of lithology in non-evaporitic siliciclastic intervals, the gamma log is extremely useful in suggesting where shale may be expected in a formation; worldwide, elevated gamma readings opposite a sandstone-mudstone succession are typically used to indicate shaliness of the formation (Vclay - Figure 3) and is a first-order indicator of reservoir cut-offs. Clays can contain high levels of potassium-containing minerals, while thorium tends to be fixed in shales compared with most quartzose sands).
Uranium in seawater is typically "fixed" into the clayey sediment in three main ways:
1. Chemical precipitation at acid (pH 2.5 - 4.0) to reducing interfaces just below the sediment-water contact.
2. Adsorption by organic matter in the clays.
3 Adsorption by phosphates in the muds.
In the subsurface, uranium can also be mobilised and "re-fixed in the subsurface at redox interfaces. High gamma readings can also be due to the high potassium content in glauconite-rich sands and arkoses or indicate uranium's secondary movement to form "hot" cement and fissure fills as in several Middle East carbonate reservoirs.
More importantly for our purposes, high gamma signatures are not only associated with shales, especially organic-rich shales, but elevated values can also be associated with those evaporite salts that contain high proportions of potassium such as sylvite, carnallite, and polyhalite and grouped as potash salts (Table 1; Figure 4). In the potash-entraining salts, depending on mineralogy, there is between 10% and 50% potassium by weight. When it is considered that the average shale contains only 2.7% potassium, the very strong radioactivity indicative of the potassium salts in an evaporite suite is understandable. It means potash beds can be distinguished from the somewhat-elevate uranium-derived kicks of marine shale and the low radiogenic content of adjacent halite, anhydrite or carbonate beds.
As a general rule of thumb, potash salts have a much higher potassium content than the clay minerals, and no thorium content, since thorium is insoluble and can be considered as an indicator of detrital origin. So in front of intervals entraining potassium evaporites, the Th curve will be flat and near zero, while the K curve will show a high percentage of potassium and a shape generally very similar to that of the total gamma-ray, at least if at the same time the uranium curve is flat and near-zero (likely indicating little or no organic material in the rock). In contrast, opposite intervals composed of the dominant evaporite salts halite or anhydrite, the gamma output shows very low readings (Figure 4).
Once an initial tie-back to core-determined assay values is done in a potash interval, it is sometimes possible to estimate the percentage of K2O from the gamma response (Figure 5). As a general "rule of thumb," Edwards et al. (1967) showed that for a 6.25-inch, liquid-filled hole, there was a correlation of 12.6 API units per 1% K2O. However, a strong tie from gamma reading to ore grade depends on ore bed thickness and mineralogical consistency in the potash unit. In many cases where mineralogy varies, the tie-back is not as reliable (Nelson, 2007; Hill 2015).
Elevated gamma values at the exploration stage are a useful screening tool or a pointer to the occurrence and distribution of potential potash zones in a basin. But the use of gamma logs as reliable ore quality indicator requires ongoing quantification and ongoing comparison to rock/ore sampling of mineralogical variation, as well as quantification of thin-bed effects in the potential ore zone.
Figure 6 lists K2O and K percentages for various pure thick intervals of a variety of potash and non-potash salts plotted against density, neutron and sonic log values.
5. Conversion chart for approximate K2O calculations using a total GR log run through an interval of potash salts (Schlumberger). The chart shows a worked example: 150 API in a 7-inch borehole with 10 lb/gallon diesel fuel-based drilling mud is the equivalent of ≈ 12% K2O. This chart follows the approximating rule of thumb mentioned in the text, so that each increase of 15 API units is equivalent to an increase in ore grade of around 1% K2O.
6. Crossplots of apparent K2O and mineral specific readouts of the bulk density, neutron and sonic logs. If a particular mineral dominates in the interval of interest, across a sufficiently thick layer containing the mineral, then the log values are considered diagnostic.
Bulk density or density log (RHOB)
The bulk density log is a measure of the degree of scattering or attenuation of gamma rays by electrons in the formation (Compton scattering). The bulk density or density log is related to the electron density of a formation and is the near-numerical equivalent of the formation's specific gravity (gm/cc); that is it is considered to measure variations in the average total density of the formation. A tool-measured value includes the solid rock matrix density and the density of fluids enclosed in the pores.
The electron density of a formation (electrons/cc) is closely related to the expected density (gm/cc) and is typically used as a direct indicator of standard density (Table 1). Some minerals, including halite and sylvite, have electron densities that are not directly proportional to their specific gravities. Such minerals require apparent bulk density for interpretation, and these values differ slightly from the actual density. More importantly, many evaporite minerals have sufficient differences in bulk density to be recognised, especially when cross plotted against gamma (GR), Pef and neutron (NPHI) values (Figures 7, 8 and 9).
Most evaporite units in the subsurface are relatively pure and often mono- or bi-mineralic. Hence, their lithological composition can be suspected, if not positively identified from the density log. However, when units are impure, the densities will fluctuate according to levels of other minerals. Fortunately, most common evaporites (halite and anhydrite) are thicker than a metre and more than 90% pure and plot as intervals of constant density with only minor variation. When this occurs, densities near the expected value in a clean evaporite unit can be easily identified and correlated to mineralogy using the bulk density log values as listed in Figures 7 and 8.
7. Gamma GR versus RHOB (density). The gamma values clearly differentiate non-potash (non-radiogenic) from potash salts (radiogenic), while the characteristic bulk densities can define and separate most near mono-mineralic evaporite salt beds.
8. Density (RHOB) versus Pef crossplot
Neutron log
The neutron porosity index or neutron log provides a continuous record of a formation’s reaction to fast neutron bombardment. It is primarily a measurement of the hydrogen concentration in the formation, whether from the water of hydration, as in the case of hydrated salts such as gypsum and carnallite, or from water or oil in the more commonly understood non-evaporite situation. Qualitatively in clean limestones, dolomites and sandstones, the neutron log is used to measure porosity (in limestone-equivalent porosity units). It can also be a good discriminator between oil and gas in reservoir intervals without hydrated salts or other minerals. Geologically, it can be used to identify gross lithology, and so define evaporites (negative porosity values), hydrated minerals, and volcanic rocks and zeolites.
A crossplot of formation bulk density versus neutron-log measurement is an extremely valuable tool for identifying various subsurface evaporite lithologies (Figure 9). For example, in thick evaporite successions, a neutron log can distinguish between various evaporite salts on the basis of water of crystallisation. Gypsum is the most common of the evaporite salts containing water of crystallisation. However, carnallite, polyhalite, and kainite also hold a water radical (Table 1). In a neutron-density (NPHI-RHOB) crossplot, all these hydrated salts have high neutron-log values and characteristic tightly-clustered apparent bulk densities, which separate them from other anhydrous evaporites such as salt or anhydrite, which contain no water and hence have NPHI values near zero (Figure 9).
9. Crossplot of neutron (NPHI Ø) versus density (RHOB gm/cc) for a variety of evaporite salts. Also shown are the typical field lines for most sandstones, limestones and dolomites with variable porosities.
Sonic log
The sonic or acoustic log measures a formation’s interval transit time, designated ∆t, measured in microseconds/ft or microseconds/m (∆t is the reciprocal of sonic velocity) qnd so is a measure of a formation’s capacity to transmit sound waves. Geologically this capacity typically varies with lithology and rock texture, notably porosity. Once again, because most subsurface evaporites have extremely low porosities and are often relatively pure, the sonic log value can be used to reliably identify evaporites, once an initial identification has been inferred by some other means. In this case, the narrow acoustic ranges of the evaporite salts can be diagnostic, especially if crossplotted against the neutron or density values (Figure 10, Table 1).
The seeming precision of the values given in Table 1 is illusory as the actual transit times in thick evaporites are influenced by compositional variation, temperature and confining pressure. Rock salt is mainly composed of the mineral halite and if it is more than 90% pure is a lithology with a density that is near constant with depth under hydrostatic pressure (Warren, 2016; Chapters 1, 10). Since density is probably the most critical factor in determining acoustic velocity, the ∆Tma of a thick halite unit tends to be relatively constant over a wide depth range. For pure halite, the interval transit time is 68 µsec/ft (14,625 ft/sec). However, many halite units contain varying levels of impurities, usually anhydrite, either as nodules or layers.
10. Sonic versus neutron log crossplot illustrating the narrow range of plot points for commonplace evaporites and caprock sulphur (all lack porosity) versus the broad range of possibilities for sandstones limestones and dolomites due to variable levels of porosity.
Electrical Properties (Resistivity and conductivity logs)
Electrical resistivity, the reciprocal of electrical conductivity, is the degree with which a formation opposes the flow of electrical current. Onshore, a log of the spontaneous potential of a formation is run at the same time as a resistivity log. In reality, the measured resistivity is dependent on the combined resistivity of both the rock matrix and any contained fluids. Most solid rock materials are insulators, while their enclosed fluids are conductors. Hydrocarbons are the exception to subsurface fluid conductivity; they are infinitely resistive, and this is the basis for the quick look-identification of hydrocarbons and the use of Archies Law to determine water saturation levels in potential hydrocarbon reservoirs. In terms of evaporite identification, most evaporite units contain little if any pores or free water and so possess very high resistivities compared to other more porous units often over 20,000 ohm.m (Figure 11; Table 1).
When the evaporite unit is relatively pure and monomineralogic, it creates a distinctive blocky log shape, whereas when it entrains beds of thin more porous lithologies (mudstones, shale, sands, limestone, dolomite) or perhaps contains brine-filled cavities and vugs, a much spikier log is seen across a saline interval. The actual wireline log signature depends on the content of brine, sand, clay, bitumen and other variables. Within a local area in a basin, an elevated resistivity signature, although it does not allow a first indication of the presence of evaporite salts, can subsequently confirm it. When bitumens and salts are present in the same interval (as in salt-encased Eocambrian carbonate-slither reservoirs in the South Oman Salt Basin), the co-occurrence of halite cement, anhydrite cement and bitumen complicate a reliable interpretation of movable hydrocarbons.
Figure 11. Typical conventional log signatures showing elevated shallow and deep reading resistivities values in both anhydritic and halite units due to a lack of interconnected pores filled with conductive fluids. Note that the deep reading signature is well over 20,000 ohm.m.
Basic identification approach
The gamma log (aka as the lithology log) measures the natural or spontaneous radioactivity of a formation. In a sand-shale basin, the measured gamma values are used to infer clay content. Opposite the contacts between common evaporites with low radiogenic content (e.g. anhydrite and halite @ x605m in Figure 11), it can be flat and so not of much use in defining changes in dominant salt mineralogies. In contrast, the gamma log (especially the spectral gamma log) in a potash-entraining basin is a reliable indication of the presence or absence of potash salts withing the otherwise low gamma signal opposite halite and anhydrite.
12. Typical density and gamma ray log signatures that can be related to spatial changes in polyhalite/anhydrite mineralogy within the Z2sl2 unit in the Zechstein of NW Europe (After Biehl et al. 2014). . Well D shows a signature that is typical for the Groningen High, Germany. Typical log values for the Z2 halite are density around 2.1 g/cm3 and a natural radioactivity <20 API units.
In a classic quick-look analysis of any potential hydrocarbon reservoir, the sonic, density and neutron logs are used both individually and in combination to estimate the porosity of likely reservoir strata (as illustrated by the clean lithology porosity lines in Figures 9, 10). These three logs are referred to as the “porosity logs”. Although they are typically used to infer porosity indirectly, they actually reflect variations in rock properties related to the passage of sound, induced gamma radiation and high energy neutron bombardment. The fact there is negligible porosity in most subsurface evaporites means the “porosity logs” in combination with each other, or with a spectral gamma log, can be used to identify evaporite mineralogies.
In the potash ore zones of the Prairie Evaporite in the Devonian of Canada, the magnitude of gamma log values follows the assayed K2O content (Figure 13; Fuzesy, 1982). In the under-development Zechstein polyhalite horizons in the Zechstein of NW Europe, an overlaid combination of a density and a gamma log in the same track can also be useful in separating what are often two co-associated high density subsurface sulphate salts, namely anhydrite and polyhalite (Figure 12). Polyhalite is a potash salt with elevated density (2.8 gm/cc) and elevated gamma values. Anhydrite has an even higher density but lacks potash and hence exhibits relatively low GR values. This difference in potash response in what are both characteristically high-density minerals allows for their differentiation.
13. Typical wireline characteristics, thicknesses and inferred ore grades of the potash members in the Upper Prairie Evaporite (after Fuzesy, 1982).
With conventional wireline log suites in bedded and halokinetic sequences worldwide I use following quick-look procedure to identify various major anhydrite, halite and potash units (refer to Table 1 and figures 11, 13, 14):
• Tentatively define evaporite intervals as zones dominated by lowest gamma-ray values (some carbonates (non organic-rich) also show very low, but typically slightly higher gamma values). Remember potash beds encased in halite, or less often anhydrite, will have high gamma values.
• Confirm pure anhydrite intervals (thicker than a metre -tool resolution dependent), a) Sonic - ∆t ≈ 50 microseconds), b) Bulk Density - Log value of >2.95 gm/cc. Anhydrite densities in log curve greater than 2.95 typically indicate anhydrite (but be aware of possible metal sulphides (pyrite, galena) and barite cement in some evaporite masses, especially in the caprock to halokinetic structures. NPHI porosities of anhydrite tend to hover at zero or slightly on the negative side (in standard limestone porosity units).
• Confirm pure halite intervals (thicker than a metre) using a combination of density and NPHI (neutron) logs. Halite-dominated zones show a consistent blend of bulk densities around 2.1 gm/cc, negative NPHI porosities and sonic (delta T) values around 64 - 70 µs/ft.
• Use the caliper - a curve tracking the nominal bit size to confirm anhydrite versus halite (±potash salts). A caliper value much larger than bit-size indicates borehole wash-out, it is probably due to the intersection of salt using an undersaturated mud, not the less-soluble anhydrite. Anhydrite beds tend to show an "in-gauge" caliper profiles and also tend to be slower-drilled units compared to halite (penetration data can be seen in a mudlog or well completion report). However, carbonate intrasalt beds can also show slow drill penetration rates and an "in-gauge" profile.
• Use the resistivity log - Salt like anhydrite has a high resistance to current flow (Table 1), but, due to wash-out, salt units often shows lower apparent resistivity values, especially in the microresistivity and shallow reading curve outputs.
• If there are high gamma (K-rich) intervals within thick halite beds or high-density values (>2.8gm/cc) adjacent to anhydrite, consider these intervals to be possible zones with elevated levels of salts such as sylvite, carnallite or polyhalite.
• Zones of very-low apparent bulk densities (<1.4 gm/cc) and low gamma values within a thick halite unit may indicate beds dominated by non-potash evaporite minerals, such as bischofite.
• A lack of porosity and the near-linear response of the spectral gamma log to mineral proportions means GR outputs when tied to assay values can be reliably used to infer K2O ore grades (Figure 7). In summary, most nonporous, thick relatively monomineralic evaporite units are readily identified using wireline logs, and often the proportions of minerals can be reliably determined using relevant crossplots.
Remember, any mineralogical interpretation based on well log outputs is just that, an interpretation, and whenever possible, should be checked against rock evidence such as chips or core. When confirming a log suite interpretation of an evaporite interval, you should keep in mind that chips composed of the more soluble evaporite minerals are often completely dissolved in the drilling mud before they reach the shale shaker. In this case, the wireline logs can give a better indication of actual mineralogy than the mud chips.
14. Conventional log suite opposite the upper contact of a thick clean halite interval in the Middle East.
Relevant readings
Alger, R.P. and Crain, E.R., 1966. Defining evaporite deposits with electrical well logs. In: L.L. Raymer, W.R. Hoyle and M.P. Tixier (Editors), Second Syposium on Salt. North Ohio Geol. Soc. , pp. 116-130.
Biehl, B.C., Reuning, L., Strozyk, F. and Kukla, P.A., 2014. Origin and deformation of intra-salt sulphate layers: an example from the Dutch Zechstein (Late Permian). International Journal of Earth Sciences, 103(3): 697-712.
Crain, E.R., 2010. Potash redux. InSite CWLS Magazine, 29(2): 17-26.
Crain, E.R. and Anderson, W.B., 1966. Quantitative log evaluation of the Prairie Evaporite Formation in Saskatchewan. J. Can. Pet. Technol., 5(3): 145-152.
Ellis, D.V. and Singer, J.M., 2007. Well logging for Earth Scientists (Second Edition). Elsevier.
Fuzesy, L.M., 1982. Petrology of potash ore in the Esterhazy Member of the Middle Devonian Prairie Evaporite in southeastern Saskatchewan. NDGS/SKGS-AAPG; Fourth International Williston Basin Symposium, October 5-7, 1982: 67-73.
Hill, D.G., 1993. Multiple log potash assay. Journal of Applied Geophysics, 30(4): 281-295.
Kruger, N., 2014. The Potash Members of the Prairie Formation in North Dakota. Report of Investigation - North Dakopta Geologica; Survey No. 113.
Nelson, P.H., 2007. Evaluation of potash grade with gamma-ray logs. U.S. Geological Survey Open-File Report 2007-1292, 14 p.
Nurmi, R.D., 1978. Use of well logs in evaporite sequences. In: W.E. Dean and B.C. Schreiber (Editors), Marine evaporites. . Soc. Econ. Paleontol. Mineral., Short Course notes, Tulsa, OK, pp. 144-176.
Rider, M., 1996. The geological interpretation of well-logs, second edition. Whittles Publishing, Caithness, 268 pp.
Warren, J.K., 2016. Evaporites: A compendium (ISBN 978-3-319-13511-3). Springer, Berlin, 1854 pp.
This is a sampling of some of the content in reservoir skill sets offered across various modules in our in-house and workshop training sessions
At Saltworks, the aim of all our training modules and workshops is two-fold. 1) give an understanding of the relevant process, 2) train participants in the application of the skill sets tied to the concept and prioritise the skill sets needed to apply this understanding. Below we illustrate the skills and knowledge necessary to recognise subsurface evaporite salts using a conventional suite of well logs.
If you want to know more, please download the relevant saline geosystems or carbonate geosystems catalogue and choose a combination of the various training modules that best suite your company needs.