«By MOLATLHEGI LARTY LOSTMAN MOSEKI STUDENT NO. 208523856 Submitted in fulfillment of the academic requirements For the degree of Master of Science In ...»
4.3 RESULTS 4.3.1 MEGACRYSTIC GRANITE GNEISS (MLM-SRP1, S 21.67581 E 27.36751) This sample yielded 20 zircon grains from which 25 analysis were made. Zircons from this unit have high U concentrations ranging from 85 to 2137ppm and high Th/U ratios ranging from 0.15 to 1.02 which suggest a magmatic origin. Zircons are dark brown with variable opacity and complexity. Some areas of metamictisation (radiation damage due to U and Th decay) and/or alteration obscures detail, but the general characteristic of this population is that they are zoned, magmatic grains with occasional cores - which may or may not be inherited. The last stage of zircon growth is a dark rim obvious on the CL images in Fig. 4.1 (spots 10.1 & 20.1). These rims have a high U content and a low Th/U ratio, suggesting that they are metamorphic in origin. Analyses 10.1 and 20.1 (Fig. 4.2) plot at the lower end of the discordia line (Fig. 4.2) and correspond to ages of 1183.3 ±
9.6 Ma and 1270 ± 12 Ma (Table 4.2) which together with the low Th/U ratio represent a metamorphic event related to Mesoproterozoic orogeny.
Figure 4.1 A-F: CL-images of zircons extracted from the megacrystic granite gneiss unit.
Note some grains with euhedral and oscillatory zoned centres. Numbers refer to the spots at which the analyses were taken
The data are generally discordant and there is additionally slight scatter of the data about a discordia. If one interprets these data in terms of the petrographic details revealed by the CL imaging, then it is possible to differentiate a possible older generation of "cores" [analyses; 8.1, 10.2, 12.1, 15. 2] and a possible younger phase of zoned, magmatic zircon.
This is shown in the Concordia plot (Fig. 4.2) with the light yellow error ellipses representing the older core components. It is difficult and unfortunately somewhat subjective in interpreting some of these structures as older cores, especially for those in grains #2.1, 7.1 and 19.1. These "cores" could also be magmatically zoned, low-U zircons that grew early in the same magma (i.e. coeval) as the more highly zoned overgrowths.
128 Figure 4.2: Wetherill plot of SHRIMP analyses for core and magmatic zircon for a megacrystic granite gneiss sample MLMSRP 1 from Foley East.
In calculating an age from these data all the cores interpreted to be inherited (listed above) are excluded from the regression, which yields an upper intercept age of 2647 ± 24 Ma (Fig. 4.2), interpreted to be the emplacement age of the protolith (porphyritic granite) to the megacrystic granite gneiss. The high MSWD of 9.1 reflects the significant scatter about the discordia. The 4 cores identified as possible inherited grains combine to give an upper intercept date of 2620 ± 39 Ma (MSWD = 0.81; probability = 0.45). This result is within error of that obtained for the zoned magmatic zircons from the general population, suggesting their interpretation as older cores may be incorrect.
129 Table 4.2: U-Th-Pb isotopic compositions and age calculations of zircons from megacrystic granite gneiss, sample MLM-SRP 1.
4.3.2 PINK GNEISSIC GRANITE (MLM-SRP2, S 21.67581 E 27.36751) This sample was taken from a pink gneissic granite dyke that occurs spatially associated with megacrystic granite gneiss (MLM-SRP 1) in the Foley East area. A total of 24 analyses were carried out on the 21 zircon grains obtained from this sample.
Concentrations of U and Th in these grains range from 105 to 916ppm and 57 to 550ppm respectively (Table 4.3). The Th/U ratio varies from 0.14 to 1.43, which is consistent with a magmatic origin for the zircons. The sample produced brown, subhedral to anhedral zircons with some slight rounding of grains (Fig. 4.3). A few grains show significant areas or zones of reworking (e.g. grain #6) with patchy recrystallized (brightCL) zones and some darker embayments (e.g. spot 6.2, Figure 4.3E) that are discordant to the magmatic zoning. There are possible cores to complicate the dating (e.g. spot 10.1)
Fig. 4.3 A-G: CL-images of some selected zircons from pink gneissic granite to illustrate their quality, and some analysed spots shown by white circles. Point labels are keyed to numbers in first column of Table 4.3. Note the euhedral texture and the magmatic zoning.
Most of the analysed spots are on the rim of the grains.
G Figure 4.3 continued.
The results of the SHRIMP analyses show some serious disturbance of the U-Pb systematics, with discordance and scatter making it difficult to define a unique age. There is, however, a group of 10 concordant to subconcordant data (plotted in blue in Figure 4.4) that are used to calculate a weighted mean 207Pb/206Pb age of 2631.5 ± 4.4 Ma (MSWD = 0.87; probability-of-fit = 0.55), which is interpreted as the emplacement age
Sample MLM-SRP 3 collected from a locality SE of Tonota is homogeneous, medium grained grey foliated tonalitic gneiss. The tonalitic gneiss normally occur as bands within the predominant megacrystic granite gneiss unit but locally, xenoliths of the tonalitic gneiss occurs within the megacrystic granite gneiss (Fig. 2.18). Eighteen (18) analyses were obtained from the zircon population of 19 grains. The Th/U ratios vary from 0.24 to 0.5, with many grains having Th/U ratio 0.2 (Table 4.4), compatible with zircon population derived from a magmatic protolith. Uranium concentrations are between 135 to 657 ppm and Th concentrations rage between 32 and 248ppm. The zircons are light brown and generally somewhat elongate with slight rounding of grain tips. Welldeveloped oscillatory zoning indicates that these zircons crystallized from felsic magma (Fig. 4.5). Some dark brown grains are clearly altered/metamict and were avoided for dating purposes.
E Figure 4.5 continued The data as plotted on a conventional Wetherill Concordia plot are highly discordant, but show little scatter about a discordia, which gives upper and lower intercepts of 2698.
9.2 Ma and 633 ± 83 Ma respectively (Fig. 4.6). The MSWD of 2.6 indicates some excess scatter, but compared to other samples from this area the data are relatively simple, with no inheritance observed. The upper intercept age of 2698.9 ± 9.2 Ma is interpreted as the emplacement age of the precursor igneous rock to the tonalitic gneiss.
4.3.4 TONOTA BIOTITE GNEISS (MLM-SRP4, S 21.44802 E 27.44663) Sample MLM-SRP 4 is from the Tonota biotite gneiss which occurs interlayered with amphibolite around Tonota Village. The origin of the protolithic of this unit has been uncertain (Section 1.6). Previous work indicates a sedimentary origin based on their spatial association with metasedimentary rocks of the SFT belt. The contents of U and Th are variable with ranges of 115 to 784 and 11 to 284 ppm (Table 4.5). Th/U ratios of the zircon grains vary from 0.02 to 0.54 with many grains having Th/U ration 0.11, suggesting a magmatic origin. Additional support for a magmatic origin is shown by oscillatory zoning in CL images. This sample produced a very poorly preserved crop of
G Fig. 4.7 continued Finding areas to analyse that are in good enough condition and might give concordant results was difficult and all spots chosen produced highly discordant data. One low-U core (15.1) gives the least discordant (4%) result with a minimum 207Pb/206Pb age of 2978 ± 7.4 Ma. The remaining seventeen spots lie, with unsatisfactory scatter on a Pb-loss trend which an upper intercept age of 2724 ± 48 Ma (Fig. 4.8). The high MSWD = 9.5,
4.3.5 MYLONITIC MEGACRYSTIC GRANITE GNEISS (MLM-SRP 5, S 21.52411 E 27.44227) This was sampled from a megacrystic granite gneiss unit south of Tonota. The rock is petrographically similar to MLM-SRP 1 containing megacrysts of K-feldspar and quartz embedded in a medium grained matrix of biotite and quartz. The megacrysts are deformed, showing evidence for flattening perpendicular to foliation (S1) surface and elongation parallel to the foliation (S1). The rock is flanked to the E by metasedimentary rocks but field/age relationship between the two rock units remains unknown because the contact is not exposed. High U and Th concentrations of the order 72 to 1708 and 38 to 479 ppm and a Th/U ratio range of 0.22 to 1.45 provide evidence for a magmatic origin for the emplacement of the protolith of this rock (Table 4.6). The zircons from this sample have a range of colours from brown to light grey and also show a range of sizes
G Fig. 4.9 continued 144 Most data are highly discordant but follow a discordia trend, albeit with some scatter. A number of analyses of relatively low-U areas are subconcordant. These are from centers of grains (which could be cores) and from embayments such as # 2.1. As with the other samples in this suite, it is difficult to unequivocally conclude that some of the central areas are not inherited cores. But in this case, the "cores" (which would be the oldest component) and the embayments (such as #2.1 which postdate the zoned, outer, magmatic growth) are indistinguishable in age. Regression of the data as shown in the Concordia plot (Fig. 4.10.) produces an upper intercept age of 2625 ± 14 Ma (MSWD = 3.4; probability = 0.000), which is interpreted as the age of the protolith and thus the maximum age of shearing. This age is within error of the age determined for MLM-SRP1 (2647 ± 24 Ma).
Figure 4.10: Wetherill plot of SHRIMP data for core and magmatic zircon for sample megacrystic granite gneiss, MLM-SRP 5 from Tonota south area.
145 The high MSWD and low probability-of-fit reflect the scatter about the single Pb-loss or discordia trend. The regression excluded a number of data points: both analyses from grain #1 and spots 4.1 and 7.1. Grain #1 appears to be a normal strongly zoned magmatic zircon, indistinguishable from the general population. It is, however very different in terms of its U-Pb character to the general population, with the two spots plotting on or near Concordia (yellow ellipses on Fig 4.10). The two analyses are significantly different in terms of the Pb-Pb ages and it is possible that this grain is recording severe Pb-loss at about 2 Ga. The two spots were located in identical, zoned tips at either end of the grain.
The other data for spots 4.1 and 7.1 were excluded from the regression as they clearly fall significantly off the calculated discordia and have a different Pb-loss history and possible origins.
Table 4.6: U-Pb isotopic compositions and age data from zircons from a megacrystic granite gneiss-sample MLM-SRP 5.
Age constraints in NE Botswana are based on crystallization of granitoid rocks and detrital zircons in metasedimentary rocks (McCourt et al., 2004). The zircons from the samples analysed are characterized by severe Pb-loss in some instances and the common resorbed edges and tips would suggest some degree of metamorphic rounding at some stage. The lower intercept ages are imprecise due to the scatter commonly observed in the data, but if the lower intercept dates are reliable indicators of the timing of the Pb-loss, then there appears to be two Pb-loss events at about 1000 Ma and 600-700 Ma (with some considerable uncertainty). The younger of these two Pb-loss events may represent a response to the Neoproterozoic “Pan African” Damara Orogeny while the Mesoproterozoic ages of 1183.3 ± 9.6 Ma and 1270 ± 12 Ma obtained from zircons in sample MLM-SRP 1, could be linked to the “Kibaran event” as documented from the Kgwebe Hills in NW Botswana by Kampunzu et al. (1998), implying co-existence of Archaean, Neoproterozoic and Mesoproterozoic elements in granitoid gneiss in the SFT area. The study area is cut by the Okavango Dyke Swarm and although the great majority of the dykes in this swarm are Phanerozoic (Karoo) in age there are some dykes that are Proterozoic in age (Jourdan et al., 2004). The 600-700Ma lower intercepts on the Concordia plots may be related to this magmatism. Mesoproterozoic ages have not been reported in either the Francistown Complex (Bagai, 2002) or the Mosetse Complex (Majaule and Davis, 1998). The Pb loss event at 2 Ga recorded in sample MLM-SRP 5 (grain # 1 and spots 4.1 and 7) may be related to the Palaeoproterozoic Magondi orogeny or the 2 Ga (reactivation) events reported in the Limpopo belt. The minimum age of the Magondi orogeny is 1997.5 ±2.6 Ma (McCourt et al., 2001).
Deciphering the internal structures of these grains, even with the use of CL and other imaging has been difficult and sometime subjective. Although this is not an uncommon feature of felsic rocks, the small gap (if any) between the magmatic, zoned zircon and the possible cores suggest that if these cores were inherited, then the source of the inheritance was not much different to the age of the host rocks. This suggests that the zircon analysed and the rocks from which they were extracted formed in a short-lived environment, with no significant history before 2700 Ma. Certainly these rocks do not appear to have a
Based on the high Th/U ratios for zircons extracted from the Tonota biotite gneiss together with the limited age range of the grains analysed, the protolith to this gneiss is interpreted to have been an igneous rather than sedimentary as previously proposed (Aldiss, 1991). From the zircon grains analysed, the age of the protolith to the Tonota biotite gneiss is 2724 ± 48 Ma making it the oldest rock in the SFT area. Tonalitic gneiss (sample MLM-SRP 3) gave an upper intercept age 2698.9 ± 9.2 Ma. This is interpreted as the emplacement age of the protolith to the tonalitic gneiss and implies subduction of the Matsitama-Motloutse Complex beneath the SW margin of the Zimbabwe craton (reported by McCourt et al. (2004) was active at 2699 Ma to produce the tonalitic magma.
Megacrystic granite gneiss from Foley East area yielded an upper intercept age of 2647 ± 24 Ma. Pink gneissic granite that crop out as dykes cutting the foliation (S1) in the megacrystic granite gneiss (Figs 2.19 and 2.27) and the banded tonalitic gneiss (Fig.