First published: TJ 9(1):71–92 April 1995 Browse this issue Subscribe to TJ

The failure of U-Th-Pb ‘dating’ at Koongarra, Australia

by Andrew Snelling

Abstract

As with other radiometric ‘dating’ methods, the U-Pb and Pb-Pb isochron methods have been questioned in the open literature, because often an excellent line of best fit between ratios obtained from a set of good cogenetic samples gives a resultant ‘isochron’ and yields a derived ‘age’ that has no geological meaning. At the Koongarra uranium deposit, Australia, there is ample evidence of open system behaviour, or repeated migration, of U and Pb — ore textures, mineral chemistry, supergene alteration, uranium/daughter disequilibrium, and groundwater and soil geochemistry. Yet U-Th-Pb isotopic studies of the uranium ore, host rocks and soils have produced an array of false ‘isochrons’ that yield ‘ages’ which are geologically meaningless. Even a claimed near-concordant U- Pb ‘age’ of 862 Ma (million years) on one uraninite grain is identical to a false Pb-Pb isochron ‘age’ but neither can be connected to any geological event. The open system behaviour of the U-Th-Pb system is clearly the norm, as is the resultant mixing of radiogenic Pb with ‘common’ or background Pb, even in soils in the surrounding region, apparently even up to 17 km away! Because no geologically meaningful results can be interpreted from the U-Th-Pb data at Koongarra (three uraninite grains even yield a ²³²Th/²⁰⁸Pb ‘age’ of 0 Ma), serious questions must be asked about the validity of the fundamental/foundational basis of the U-Th-Pb ‘dating’ method. This makes the task of creationists building their model for the geological record much easier, since claims of U-Th-Pb radiometric ‘dating’ having ‘proven’ the claimed great antiquity of the earth, its strata and fossils can be justifiably ignored.

Introduction

Radiometric dating has now been used for almost 50 years to establish ‘beyond doubt’ the multi-billion year age of the earth’s geological column. Although this column and its ‘age’ was firmly settled well before the advent of radiometric dating, the latter has been used to quantify the, ‘ages’ of the strata and the fossils in the column, so that in many people’s minds today radiometric dating has ‘proved’ the presumed antiquity of the earth.

However, it is important to remember that all radiometric dating methods are based on three main assumptions:-

1. The physico-chemical system must have always been closed. Thus no parent, daughter or other decay products within the system can have been removed, and no parent, daughter or other decay products from outside the system can have been added.
2. The system must initially have contained none of its daughter elements or decay products, or at the very least we need to know the starting conditions/state of the decay system.
3. The decay rate, referred to as the half-life of the radioactive parent element, must have always been the same, that is, constant.

The highly speculative nature of all radiometric dating methods becomes apparent when one realizes that none of the above assumptions is either valid or provable. Put simply, none of these assumptions can have been observed to have always been true throughout the supposed millions of years the radioactive elements have presumed to have been decaying.

Of the various radiometric methods, uranium-thorium- lead (U-Th-Pb) was the first used and it is still widely employed today, particularly when zircons are present in the rocks to be dated. But the method does not always give the ‘expected’ results, leading to fundamental questions about its validity. Indeed, the U- Th-Pb system is well known to be prone to open system behaviour, with U being particularly geochemically mobile, meaning that U is readily lost from the crystal lattices of the minerals used for ‘dating’, including zircons. Pb is also prone to diffusion from minerals. Thus it is questionable as to why this radiometric ‘dating’ method is still used. Instead, it is increasingly being applied in more sophisticated ways to geological ‘dating’ problems.

In the conclusion to a recent paper exposing shortcomings and criticising the validity of the popular rubidium-strontium (Rb-Sr) isochron method, Zheng wrote:

‘. . . some of the basic assumptions of the conventional Rb-Sr isochron method have to be modified and an observed isochron does not certainly define a valid age information for a geological system, even if a goodness of fit of the experimental data points is obtained in plotting ⁸⁷Sr/⁸⁶Sr vs. ⁸⁷Rb/⁸⁶Sr. This problem cannot be overlooked, especially in evaluating the numerical time scale. Similar questions can also arise in applying Sm-Nd and U-Pb isochron methods’¹

Amongst the concerns voiced by Zheng were the problems being found with anomalous isochrons, that is, where there is an apparent linear relationship between ⁸⁷Sr/⁸⁶Sr and ⁸⁷Rb/⁸⁶Sr ratios, even an excellent line of best fit between ratios obtained from good cogenetic samples, and yet the resultant isochron and derived ‘age’ have no distinct geological meaning. Zheng documented the copious reporting of this problem in the literature where various names had been given to these anomalous isochrons, such as apparent isochron, mantle isochron and pseudoisochron; secondary isochron, inherited isochron, source isochron, erupted isochron, mixing line, and mixing isochron.

Similar anomalous or false isochrons are commonly obtained from U- Th-Pb data, which is hardly surprising given the common open system behaviour of the U- Th-Pb system. Yet in the literature these problems are commonly glossed over or pushed aside, but their increasing occurrence from a variety of geological settings does seriously raise the question as to whether U-Th-Pb data ever yields any valid ‘age’ information. One such geological setting that yields these false U -Th -Pb ‘ages’ and ‘isochrons’ is the Koongarra uranium deposit and the surrounding area (Northern Territory, Australia).

Figure 1. Regional geology map showing the location of the Koongarra uranium deposit

Figure 2. Local geology map showing the location of the Koongarra No. 1 and No. 2 orebodies. Because of surficial cover the geological units and outline of the mineralisation are projected to the surface from the base of weathering.

The Koongarra Area

The Koongarra area is 250 km east of Darwin (Northern Territory, Australia) at latitude 12¡52’S and longitude 132¡50’E. The regional geology has been described in detail by Needham and Stuart-Smith² and by Needham^3,4 (see Figure 1), while Snelling⁵ describes the Koongarra uranium deposit and the area’s local geology (see Figure 2).

The Koongarra uranium deposit occurs in a metamorphic terrain that has an Archaean basement consisting of domes of granitoids and granitic gneisses (the Nanambu Complex), the nearest outcrop being 5 km to the north (see Figure 1). Some of the lowermost overlying Lower Proterozoic metasediments were accreted to these domes during amphibolite grade regional metamorphism (estimated to represent conditions of 5-8 kb and 550-630¡ C) at 1800- 1870 Ma (million years ago, according to conventional evolutionary dating). Multiple isoclinal recumbent folding accompanied metamorphism. The Lower Proterozoic Cahill Formation flanking the Nanambu Complex has been divided into two members. The lower member is dominated by a thick basal dolomite and passes transitionally upwards into the psammitic upper member, which is largely feldspathic schist and quartzite. The uranium mineralisation at Koongarra is associated with graphitic horizons within chloritised quartz-mica (±feldspar ±garnet) schists overlying the basal dolomite in the lower member (see Figures 2 and 3). A 150 Ma period of weathering and erosion followed metamorphism. A thick sequence of essentially flat-lying sandstones (the Middle Proterozoic Kombolgie Formation) was then deposited unconformably on the Archaean-Lower Proterozoic basement and metasediments. At Koongarra subsequent reverse faulting has juxtaposed the lower Cahill Formation schists and Kombolgie Formation sandstone.

Figure 3. Simplified cross section through the No. 1 orebody, Koongarra, showing geology, distribution of uranium minerals and alteration, and present groundwater flow.

Owing to the isoclinal recumbent folding of metasedimentary units of the Cahill Formation, the typical rock sequence encountered at Koongarra is probably a tectono-stratigraphy (see Figure 3):-

Polyphase deformation accompanied metamorphism of the original sediments, that were probably dolomite, shales and siltstones. Johnston⁶ identified a D₂ event as responsible for the dominant S₂ foliation of the schist sequence, which at Koongarra dips at 55¡ to the south-east The dominant structural feature, however, is the reverse fault system that dips at about 60¡ to the south-east, sub-parallel to the dominant S₂ foliation and lithological boundaries, just below the mineralised zone.

The Uranium Deposit

There are two discrete uranium orebodies at Koongarra, separated by a 100 m wide barren zone (see Figure 2). The main (No.1) orebody has a strike length of 450 m and persists to 100 m depth. Secondary uranium mineralisation is present in the weathered schists, from below the surficial sand cover to the base of weathering at depths varying between 25 and 30 m (see Figure 3). This secondary mineralisation has been derived from decomposition and leaching of the primary mineralised zone, and forms a tongue-like fan of ore-grade material dispersed down-slope for about 80 m to the southeast. The primary uranium mineralised zone in cross-section is a series of partially coalescing lenses, which together form an elongated wedge dipping at 55¡ to the southeast within the host quartz-chlorite schist unit, sub-parallel to the reverse fault. True widths average 30 m at the top of the primary mineralised zone but taper out at about 100 m below the surface and along strike.

Superimposed on the primary prograde metamorphic mineral assemblages of the host schist units is a distinct and extensive primary alteration halo associated, and cogenetic, with the uranium mineralisation (see Figure 3). This alteration extends for up to 1.5 km from the ore in a direction perpendicular to the host quartz-chlorite schist unit, because the mineralisation is essentially stratabound. The outer zone of the alteration halo is most extensively developed in the semi-pelitic schists, and is manifested by the pseudomorphous replacement of biotite by chlorite, rutile and quartz, and feldspar by sericite. Silicification has also occurred in fault planes and within the Kombolgie Formation sandstone beneath the mineralisation, particularly adjacent to the reverse fault. Association of this outer halo alteration with the mineralisation is demonstrated by the apparent symmetrical distribution of this alteration about the orebody. In the inner alteration zone, less than 50 m from ore; the metamorphic rock fabric is disrupted, and quartz is replaced by pervasive chlorite and phengitic mica, and garnet by chlorite. Uranium mineralisation is only present where this alteration has taken place.

The primary ore consists of uraninite veins and veinlets (1-10 mm thick) that cross-cut the S₂ foliation of the brecciated and hydrothermally altered quartz-chlorite schist host. Groups of uraninite veinlets are intimately intergrown with chlorite, which forms the matrix to the host breccias. Small (10-100 mm) euhedral and subhedral uraninite grains are finely disseminated in the chloritic alteration adjacent to veins, but these grains may coalesce to form clusters, strings and massive uraninite. Coarse colloform and botryoidal uraninite masses and uraninite spherules with internal lacework textures have also been noted, but the bulk of the ore appears to be of the disseminated type, with thin (< 0.5 mm) discontinuous wisps and streaks of uraninite, and continuous strings both parallel and discordant to the foliation (S₂), and parallel to phyllosilicate (001) cleavage planes.

Associated with the ore are minor volumes (up to 5%) of sulphides, which include galena and lesser chalcopyrite, bornite and pyrite, with rare grains of native gold, clausthalite (PbSe), gersdorffite-cobaltite (NiAsS-CoAsS) and mackinawite (Fe, Ni)_1.1S. Galena is the most abundant, commonly occurring as cubes (5-10 mm wide) disseminated in uraninite or gangue, and as stringers and veinlets particularly filling thin fractures within uraninite. Galena may also overgrow clausthalite, and replace pyrite and chalcopyrite. Chlorite, predominantly magnesium chlorite, is the principal gangue, and its intimate association with the uraninite indicates that the two minerals formed together.

Oxidation and alteration of uraninite within the primary ore zone has produced a variety of secondary uranium minerals, principally uranyl silicates.⁷ Uraninite veins, even veins over 1 cm wide, have been completely altered in situ. Within the primary ore zone this in situ replacement of uraninite is most pronounced immediately above the reverse fault breccia, and this alteration and oxidation diminish upwards stratigraphically. It is accompanied by hematite staining of the schists, the more intense hematite alteration in and near the reverse fault breccia being due to hematite replacement of chlorite. The secondary mineralisation of the dispersion fan in the weathered schist above the No.1 orebody is characterised by uranyl phosphates found exclusively in the ‘tail’ of the fan. Away from the tail uranium is dispersed in the weathered schists and adsorbed onto clays and iron oxides.

The ‘age’ of the uranium mineralisation is problematical. The mineralisation, however, must post-date both the Kombolgie Formation sandstone and the Koongarra reverse fault, since it occupies the breccia zones generated by the post Kombolgie reverse faulting. The pattern of alteration which is intimately associated with the ore also crosses the reverse fault into the Kombolgie sandstone beneath the ore zone, so this again implies that the ore was formed after the reverse fault and therefore is younger than both the Kombolgie sandstone and the reverse fault. Because of these geological constraints, Page et al.⁸ suggested the mineralisation was younger than 1600-1688 Ma because of their determination of the timing of the Kombolgie Formation deposition to that period. Sm-Nd isotopic data obtained on Koongarra uraninites^9,10 appears to narrow down the timing of mineralisation to 1550-1650 Ma. It is unclear as to when deep groundwater circulation began to cause oxidation and alteration of the primary uraninite ore at depth, but Airey et al.¹¹ suggest that the weathering of the primary ore to produce the secondary dispersion fan in the weathered schists above the No.1 orebody seems to have begun ‘only’ in the last 1- 3Ma.

Evidence Of An Open System

There are five main lines of independent evidence that the mineral-rock systems at Koongarra have been open to diffusion and migration of U, Th and daughter isotopes including Pb. Such behaviour of these isotopes has crucial implications to all attempts to ‘date’ the Koongarra uranium ore using the U- Th-Pb isotopic systems.

(1) Ore Textures

Mineralogical and textural studies of the ore under both optical and scanning electron microscopes^12,13 indicate that there have been as many as three remobilisations of the uranium during the history of the ore. Pb has likewise been mobile. That is, both the primary U and Pb minerals, uraninite and galena respectively, have been dissolved and redeposited/recrystallised, often some distance away from their original locations. This is shown diagrammatically in Figure 4 as several generations of uraninite and galena.

Figure 4. Paragenesis diagram showing the stages of formation and development of the minerals comprising the Koongarra uranium deposit.

Figures 5-10 illustrate examples of the ore textures under the microscopes, the accompanying descriptions indicating how the textures have been interpreted.

Figure 5. Remobilisation and redeposition of uraninite (white mineral). Photomicrograph shows uraninite veins (left and right) partially destroyed by dissolution of uranium which has been redeposited as scattered veinlets and shapeless masses of a new generation of uraninite (middle). (Magnification 10X).

Figure 6. Uraninite (light grey) has been dissolved and redeposited as thin veinlets and shapeless masses within a chlorite (dark grey) matrix which is also replacing the main uraninite grain. (Magnification 120X).

Figure 7. Two generations of uraninite grains (lighter grey), and more oxidised supergene veins and patches (darker grey). The small scattered white grains are galena. (Magnification 200X).

Figure 8. Two generations of uraninite grains (white, left of photomicrograph) and later thin supergene encrustations (mid grey) around quartz grains (dark grey). The very bright mineral (right) is galena which has similarly dissolved and redeposited. (Magnification 200X).

Figure 9. Remobilised uraninite (light grey) deposited as scattered grains with a chlorite (dark grey) matrix. A remobilised galena vein (white-grey) cuts across the uraninite-chlorite association. (Magnification 50X).

Figure 10. An enlarged view of uraninite (dark grey) sub-grains within a larger vein. Galena (light grey) veinlets which both cross-cut and separate the uraninite sub-grains. The Pb in the galena is supposed to have migrated from the uraninite where it was supposedly produced by radioactive decay. (Magnification 50X).

[^_ denotes not measured; l.d. denotes less than detection limits]

Table 1. Analyses of some representative Koongarra uraninites.

(2) Mineral Chemistry

Uraninite compositions in the ore are never uniform. Electron microprobe analyses of uraninite grains and veins,¹³ that is, micro-analyses of volumes of uraninite between 5 and 10 mm in diameter (see Table 1), reveal that uraninite compositions, particularly U, Pb and Ca contents, vary not only from grain to grain within anyone sample regardless of which generation of uraninite it is, but even at the microscopic level within uraninite grains themselves. Figure 11 illustrates how Pb and Ca have both substituted for U in the UO₂ cubic lattice in varying amounts across the uraninite veins and grains.

Figure 11. Compositional traverse across a uraninite grain similar to those in Figure 10.

Table 2. The secondary uranium minerals at Koongarra.

(3) Supergene Alteration

As has already been briefly noted, supergene alteration (principally oxidation) of uraninite has not only occurred where the zone of surficial weathering has intersected the top of the No.1 orebody, but at depth within the primary ore. Uraninite grains and veins have been replaced by colourful secondary uranium minerals (see Table 2), their occurrence and compositions depending on the chemistries of the immediate rock/mineral environments and the circulating ground waters (see Figures 3 and 12). The net result has been the complete destruction of the uraninite in what was the top of the No.1 orebody, with its replacement (sometimes in situ) by uranyl silicate or uranyl phosphate minerals (usually the latter), and the dispersion of the rest of the U over distances of up to 50 m or more down-slope by ground waters in the weathered zone. Additionally, at the same time there has been yet another remobilisation of both U and Pb in the primary ore zones, with in situ replacement of uraninite (see Figures 13-15) and deposition of supergene uraninite (see Figure 16) and the uranyl silicate minerals sklodowskite and uranophane (see Figures 17 and 18) from the U in solution from circulating ground waters (see Figure 3 again).⁷ Electron microprobe analyses (see Table 3) show that the U and Pb contents have decreased as uraninites were altered to uranyl silicates, while the iron and manganese oxides lining fractures in the host rocks have absorbed the U and Pb that had been dissolved during the oxidation of the uraninites and migrated in the circulating ground waters (see Table 4).

Figure 12. Schematic diagram showing the paths of secondary uranium mineral from uraninite in the Koongarra uranium deposit.

Figure 13. Kasolite (white) and uranophane (grey) replacing a former uraninite vein. Note that the former vein shape, even the sub-grains, have essentially been preserved. (SEM magnification 210X; scale bar microns.)

Figure 14. Globular uraninite mass (black shape just to the left of center) being altered marginally to sklowdowskite (grey concentric sheath). (Magnification 2X; scale bar 3 mm.)

Figure 15. Kasolite (light grey) and sklodowskite (dark grey) replacing a former uraninite vein. (SEM magnification 210X.)

Figure 16. Supergene colloform banded uraninite (grey) deposited in what was originally a void. The banding is produced by a time sequence of uraninite deposition. (SEM magnification 840X.)

Figure 17. A sklodowskite (white) vein composed of radiating aggregates of needle-shaped crystals. (SEM magnification 220X; scale bar 50 microns.)

Figure 18. Uranophane (white) veinlets deposited between quartz (grey) grain boundaries. (SEM magnification 220X; scale bar 50 microns.)

[- denotes not measured; l.d. denotes less than detection limits]

Table 3. Analyses of alteration sequences of uraninites to uranyl silicates at Koongarra.

[- denotes not measured; l.d. denotes less than detection limits]

Table 4. Analyses of iron and manganese oxides in fractures in the Koongarra primary ore.

(4) Uranium/Daughter Disequilibrium

There are two methods of measuring the grade of a uranium ore sample:-

1. by assaying for U directly using standard chemical or related techniques, and
2. by measuring the radioactivity given off by the ore sample, the quantity of such radioactivity being directly related, and proportional, to the U content.

However, because the radioactivity measured is actually the gamma radiation given off by the daughter element bismuth-214 (²¹⁴Bi) far down the ²³⁸U decay chain, any addition or removal of daughter elements between ²³⁸U and ²¹⁴Bi will result in a discrepancy between the above two measurements of the U content of the ore sample. To assess this possibility the two measurements are compared:-

Three possibilities arise:-

1. Ratio = 1. The ore sample is said to be in equilibrium since the two measurements agree, implying that the U and its daughter elements are in equilibrium; neither have apparently migrated.
2. Ratio > 1. The ore sample is said to be in disequilibrium, and since the U content is greater than the daughter element content either U has been added to the sample or daughter elements removed.
3. Ratio < 1. Again the ore sample is aid to be in disequilibrium, but now the U content is less than the daughter element content implying either U removal or daughter element addition to the sample.

Table 5. Summary of disequilibrium patterns in the Koongarra orebodies.

Measurements on ore samples from Koongarra indicate that the ore is in overall disequilibrium (Table 5 and Figure 19).¹⁴ High resolution gamma-ray spectroscopy was then used to determine which daughter elements of ²³⁸U have been mobilised.¹⁵ These investigations showed that even though the high grade uraninite (massive) ore is near equilibrium, radium-226 (²²⁶Ra) and radon-222 (²²²Rn), and the immediate host rocks being relatively enriched in U, having been precipitated from the circulating groundwaters that had dissolved it from the orebody. Figure 20 schematically illustrates these movements of isotopes caused by the present day circulation of groundwaters.

Figure 19. Frequency histogram of disequilibrium ratios measured on Koongarra ore and host rock samples.

Figure 20. Uranium (U) and (Ra) migration and precipitation (ppt) caused by present-day groundwater circulation and chemistry.

(5) Groundwater and Soil Geochemistry

Because of the tropical, monsoonal climate, the ground waters in the Koongarra area are fast moving, annually recharged and low in salinity, the water table rising and falling by as much as 10 m between the wet and the dry seasons. However, U is dissolved by the ground waters from the mineralised aquifer rocks, the level of dissolved U depending on the prevailing pH, Eh, salinity and degree of adsorption. A survey of the chemistry of the ground waters in open drill holes in and near the Koongarra orebodies revealed that a hydrogeochemical halo exists in and around the ore zones reflecting the alteration chemistry of the host rocks and ore, with U levels up to 4100 mg/l.¹⁶ Such measurements confirm the other observations already cited that indicate U is being dissolved from the ore minerals by present day circulating ground waters, dispersed and partly redeposited. Furthermore, the ground waters are also dispersing U- Th decay products such as helium (He) from the ore zone, with measured levels up to 14.2 ml/l.¹⁷

It is hardly surprising, therefore, that the soils overlying the ore zones and the immediate areas of host rocks carry anomalous U concentrations compared to background levels.¹⁸ That the ground waters have been responsible for dispersing U ( and Pb) into the surrounding soils is also clearly demonstrated by analyses down through the soil profile. Furthermore, Dickson et al.^19,20 found the Pb isotopic signature of the U ore in the soils above the No.2 orebody, which is concealed by about 40 m of barren overburden, and in the soils to the south of the No.1 orebody within the hydrogeochemical halo.

Concentration (Wt%)				Atomic Ratios			Ages			Lead Isotope Ratios
Sample No.	%U	%Pb	%Th				t206 m.y.	t207 m.y.
J804/1	62.38	8.07	0.30	0.142	1.312	0.0673	861	862	864	21330	1450	7.10
J804/b	38.21	4.45	0.28	0.126	1.264	0.0727	774	841	1025	9875	731.9	34.84
J801	55.07	3.64	0.34	0.071	0.810	0.0826	447	610	1282	16870	1408	54.20
J807	44.08	5.35	0.33	0.130	1.259	0.0703	796	838	954	12920	921.9	35.49
J809	52.61	5.45	0.39	0.114	1.061	0.0679	699	744	882	105800	7200	62.64
Common lead correction
Mt Isa lead										16.11	15.61	36.72

Table 6. U-Th-Pb concentrations and isotopic compositions of Koongarra uraninites.

Sample No.
J801	10290	1016	55.81
J803	41240	3258	143.9
J804	11530	883	8.539
J809	10540	1261	47.41
J820	4824	709.2	35.15
J821	3399	461.0	43.24

Table 7. Isotopic compositions of Koongarra galenas.

‘Dating’ of the Primary Ore

Hills and Richards^21,22 isotipically analysed individual grains of uraninite and galena that had been hand-picked from drill core (see Table 6 and 7). Only one of the five uraninite samples gave a near-concordant ‘age’ of 862 Ma, that is, the sample plotted almost on the standard concordia curve, and Hills and Richards²² interpreted this as recording fresh formation of Pb-free uraninite at 870 Ma (see Figure 21). The other four uraninite samples all lay well below concordia and did not conform to any regular linear array. Hills and Richards were left with two possible interpretations. On the one hand, preferential loss of the intermediate daughter products of ²³⁸U (that is, escape of radon, a gas) would cause vertical displacement of points below an episodic-loss line, but this would only produce a significant Pb isotopic effect if the loss had persisted for a very long proportion of the life of the uraninite (which is incidentally not only feasible but likely). Alternatively, they suggested that contamination by small amounts of an older (pre-900 Ma) Pb could cause such a pattern as on their concordia plot, to which they added mixing lines that they postulated arose from the restoration to each uraninite sample of the galena which separated from it (see Figure 21 again).

Figure 21. Conventional 206Pb/238U concordia diagram of uraninites from Koongarra. The insert shows the hypothetical directional shift in uraninite data points supposedly explained by contamination from associated galena.

This of course assumes that the Pb in the galenas was also derived predominantly from U decay. They plotted their Pb ratios in all their uraninite samples on a standard ²⁰⁷Pb/²⁰⁶Pb diagram, and contended that the pattern of data points did not conform to a simple age interpretation (see Figure 22). Instead, they contended that the scatter of points could be contained between two lines radiating from the diagram’s origin, lines that essentially represented isochrons for uraninites and galenas from the Ranger and Nabarlek uranium deposits, similar orebodies in the same geological region. From the positions of the Koongarra uraninites and galenas on these diagrams they claimed that the galenas contained left-over radiogenic Pb from earlier uraninites as old as 1700-1800 Ma (the ‘age’ of the Ranger uranium mineralisation), these earlier uraninites being obliterated by the U having remobilised at 870 Ma, the ‘age’ of the lone Pb-free uraninite sample.

Figure 22. Conventional 207Pb/204Pb vs. 206Pb/204Pb plots of galenas and uraninites from Koongarra. Limiting fields of anomalous-lead lines corresponding to ‘ages’ of 1800 Ma and 860 Ma.

In a separate study Carr and Dean²³ isotopically analysed unweathered whole- rock samples from the Koongarra primary ore zone (see Table 8). These were samples of drill core that had been crushed. Their isotopic data on four samples were plotted on a U-Pb isochron diagram and indicated a non-systematic relationship between the ²³⁸U parent and the ²⁰⁶Pb daughter. In other words, the quantities of ²⁰⁶Pb could not simply be accounted for by radioactive decay of ²³⁸U, implying open system behaviour. They also plotted their four results on a standard ²⁰⁷Pb/²⁰⁶Pb isochron diagram (see Figure 23) and found that these samples fell on a poorly defined linear array whose apparent age they did not quantify.

Figure 23. Conventional 207Pb/204Pb vs. 206Pb/204Pb plot of the weathered and unweathered whole-rock samples from Koongarra. The weathered and unweathered samples fall on separate ‘isochrons’.

Sample						Pb (ppm)	U (ppm)
Primary Ore
1	0.0233	0.0752	2438.350	183.370	56.708	80	590.0
2	0.0682	0.0908	1162.990	105.594	79.351		168.0
3	0.0110	0.0692	6845.720	473.718	75.415	112	154.0
4	0.0346	0.0649	5719.990	371.474	198.191	19	17.0
Weathered Zone Ore
5	0.1785	0.1192	387.664	46.210	69.205	5	413.0
6	0.3804	0.2028	124.773	25.310	47.465	5	861.0
7	0.5029	0.2790	72.814	20.315	36.616	50
8	0.9277	0.4118	44.155	18.184	40.964	10
9	0.1608	0.1403	248.526	34.859	39.963	30
10	0.1650	0.1420	241.053	34.225	39.772	30
11	1.0477	0.3534	55.190	19.502	57.822	3
12	0.1213	0.1252	363.622	45.537	44.119	58
13	0.1233	0.1250	357.688	44.709	44.106	10

Table 8. Results of Pb isotopic, U concentration and Pb concentration analyses for Koongarra whole-rock samples.

‘Dating’ of Weathered Rocks and Soils

Carr and Dean²³ also isotopically analysed a further nine whole-rock samples from the weathered schist zone at Koongarra (see Table 8). Some of these samples were again crushed drill core, but the majority were crushed percussion drill chips. When their isotopic data were plotted on a U-Pb isochron diagram, six of the nine samples plotted close to the reference 1000 Ma isochron, while the other three were widely scattered (see Figure 24). However, on the ²⁰⁷Pb/²⁰⁶Pb diagram all nine weathered rock samples plotted on a linear array which gave an apparent isochron ‘age’ of 127050Ma (see Figures 23 and 25).

Figure 24. A U-Pb (238U/204Pb vs. 206Pb/204Pb) isochron diagram with the weathered whole-rock samples plotted on it. Most fall on the 1000 Ma reference isochron, while the 10 Ma reference isochron is also drawn in as a guide to the two outliers.

Figure 25. A conventional 207Pb/204Pb vs. 206Pb/204Pb isochron diagram showing all the weathered whole-rock samples plotted as a linear array which gives an apparent isochron ‘age’ of 1270±50Ma. (This diagram is an expansion of the lower left hand corner of Figure 23.)

In unrelated investigations, Dickson et al.^19,20 collected soil samples from above the mineralisation at Koongarra and from surrounding areas, and these were analysed for Pb isotopes to see if there was any Pb isotopic dispersion halo around the mineralisation sufficiently large enough to warrant the use of Pb isotopic analyses of soils as an exploration technique to find new uranium orebodies. The technique did in fact work, Pb isotopic traces of the deeply buried No.2 orebody mineralisation being found in the soils above, as mentioned earlier. This mineralisation, 40 m below the surface, is blind to other detection techniques.

Dickson et al.²⁰ found that all 113 soil samples from their two studies were highly correlated (r = 0.99986) on a standard ²⁰⁷Pb/²⁰⁶Pb diagram, yielding an apparent (false) isochron representing an ‘age’ of 1445±20 Ma for the samples (see Figure 26). However, most of the soil samples consisted of detritus eroded from the Middle Proterozoic Kombolgie sandstone, so because the samples from near the mineralisation gave a radiogenic Pb signature Dickson et al. interpreted the false ‘isochron’ as being due to mixing of radiogenic Pb from the uranium mineralisation with the ‘common’ Pb from the sandstone.

Figure 26. Plot of 207Pb/206Pb vs. 206Pb/204Pb for all 113 soil samples from the Koongarra area analysed by Dickson et al., indicating the high correlation of r = 0.99986 between the two variables with a fitted regression line yielding an apparent isochron ‘age’ of 1445±20 Ma. The insert shows the distribution of samples about a threshold dividing radiogenic Pb and country rock Pb along this proposed mixing line.

Discussion

Primary Ore Samples

Snelling²⁴ has already highlighted a telling omission by Hills and Richards.²² Having included all the Pb isotopic ratios they had obtained on their five uraninite samples, they tabulated also the derived ‘ages’, except for thos