Exploration results April 1988 to April 1989.

Abstract

Initially, 47 samples were collected from the region, 24 of which were collected from significant gold mines, 5 from minor gold occurrences, 9 from a tin mine within the goldfield and the remaining 9 from un-worked quartz veins, presumed to be barren background samples. These samples were crushed, sieved and analysed by decrepitation as described by Burlinson (1988a). To facilitate comparison of the decrepigrams, the total number of counts recorded in the temperature interval 150degrees to 350degreesC was used. This is the temperature region in which CO2 rich inclusions decrepitate and the sum value gives an indication of the abundance of such inclusions in the sample. It is thought that the 002 rich inclusions decrepitate at low temperature because of their high internal pressures and because of the high thermal expansion coefficient of C02. Attempts to normalize this value to allow for variations in the total decrepitation response of each sample were unsuccessful and the logarithm (base 10) of the total number of counts from 150 to 350degreesC was found to be the most useful way of classifying the samples. Although this is an oversimplification of the relation between 002 rich inclusions and low temperature decrepitation, it was found to be a useful and simple interpretative procedure. Nineteen of the samples were from the Enterprise gold mine where 5 stages of quartz veining have been recognized, based on structural considerations (Dann and Delaney, 1984). Of these the two earliest stages are thought to have been formed prior to the gold mineralisation event. Because most of the samples were collected from drillcore it was not possible to allocate the decrepitation samples used in this survey to their particular vein stage and it was not therefore surprising that some samples lacked a low temperature decrepitation peak, these presumably being samples from the pre-mineralization veins. This decrepitation data showed the importance of 002 in the fluid inclusions in this mine (Burlinson, 1984), a fact which had not been recognized in previous microthermometric studies. The presence or absence of CO2 in the inclusions was checked by microscope observations on 9 of the samples, 4 of which had prominent low temperature decrepitation peaks, 3 of which had almost no low temperature decrepitation and the remaining 2 having minor low temperature decrepitation. Standard petrographic sections were prepared and observed both at room temperarure and on a thermoelectrically cooled stage at about 0degreesC. These observations, summarised in table 1, confirm the relationship between the pronounced low temperature decrepitation peaks and the presence of 002 rich fluid inclusions in the samples. This plot shows that 002 rich inclusions are more abundant in areas of known gold mineralisation and less likely to occur in background areas. In addition, other types of mineralisation such as the Jimmy's Knob tin deposit had relatively rare CO rich samples compared with the gold mineralised areas. The background samples plottetl in this diagram were collected from various readily accessible quartz outcrops within the goldfield, but upon which no old gold diggings were known. One of these background samples had a very high CO level, indicating the presence of a C02 rich palaeothermal system in which no gol occurred in the quartz in present surface outcrop. On the theorey that gold may have been deposited in such a system below the level of present day outcrop, it was decided to carry out detailed studies of the area surrounding this sample, as well as continuing to collect regional samples to check the relationship between gold and low temperature decrepitation activity. The area delineated for detailed work was named Cotan and the detailed work done there is described below. The regional work continued by sampling many other mineralised areas throughout the goldfield and a total of 211 samples have now been collected. This data further confirms the association between CO2 rich fluids and gold mineralisation in this region, most of the currently producing mines showing an abundance of samples with total low-temperature decrepitation counts above 1000. Exceptions to this pattern occur at the Goodall and Davies mines, although their lack of CO2 rich samples may be a function of the sampling pattern, both of these mines having been sampled at a time when the ore zones were not exposed in the pits. Although samples remote from mineralisation lack significant levels of CO2, some of the background samples collected near known mineralisation show unexpectedly high CO2 levels. This is so in the area just north of the Enterprise mine, which has been classified as weakly mineralised in this figure, although no workings are known on the quartz veins which were sampled. Other background samples collected near the extensive Union Reefs workings and near the Cotan area also contained high levels of CO2 rich inclusions. These samples were some 2 to 3 km. from the nearest known mineralisations and the presence of 002 in them suggests that there may be very large haloes around the mineralised areas. Such large CO2 haloes have also been described at Timmins (Smith & Kesler, 1985). The more thorough regional study data does confirm the relationship between gold mineralisation and the abundance of CO2 rich fluid inclusions in the Pine Creek goldfield. However because some presumed background samples collected near existing mines also had high CO2 contents it is not possible to reliably discriminate between mineralised and barren veins on the basis of CO2 content alone. The Cotan area was selected on the basis of an unusually high abundance of CO2 rich inclusions in a quartz outcrop within a belt known for its previous gold production. Although there were no old workings on the actual outcrop of interest, several workings occur nearby. The area selected for study was about 3 km. long by 1.5 km. wide, extending approximately north-south and parallel to the direction of the dominant structural features in the area. The area is within the Finniss River group sediments and comprises siltstones and minor greywackes of the Burrell Creek formation. This formation lacks distinctive marker horizons and only one small lithological variation was map able. The rocks strike north westerly at 325 to 335degrees with near vertical dips. There is a strong pervasive axial plane foliation oriented at 325degrees and with near vertical dip, together with evidence of cross shearing in several different directions. The Cotan area is within the Pine Creek shear zone which extends for approximately 50 km. and within which many other gold occurences are located, including the nearby Union Reef and Spring Hill deposits. An anticlinal fold axis runs through the Cotan area but is recognizable only from observation of the bedding-cleavage intersection angles due to the lack of distinctive lithological horizons. A few kilometers to the north, complex folding has been recognized where varying lithologies facilitate the mapping and it is reasonable to expect the Cotan area to be similarly folded. These fold axes plunge southwards at approximately 10 to 15degrees. Quartz veining is common throughout the area, frequently occuring as lenses up to 5 cm. thick and also as massive large quartz veins. Although in situ outcrop of quartz is not always available, quartz rubble (fragments 1 to 3 cm. across) is common. The soils are skeletal and residual and it is not therefore likely that the coarse quartz rubble has been transported. The most likely origin for the quartz is from the discontinuous lensoid quartz veins which are ubiquitous in the bedrock, though they outcrop only poorly. Geophysical and sampling surveys were carried out to explore the area. The geophysical surveys (magnetics and EM methods) were not useful and are not discussed here. The sampling work involved collection of rock samples (mostly of quartz) for routine geochemical analysis and also for decrepitation analysis. The sample sites were located on aerial photographs and were all sampled for decrepitation, with geochemical analyses of only selected sites. The geochemical assay samples were comprised of multiple rock (usually quartz) fragments 2 to 4 cm. across, totalling about 1.5 kg. and collected within a radius of about 1 metre. For the decrepitation analyses, a single piece of quartz about 3 to 4 cm. across was collected at each site. The decrepitation results showed considerable variation between samples collected within centimetres of each other so in subsequent sample collection programmes, 2 or 3 quartz lumps were collected within a metre of each other at each site and analysed seperately in order to observe these variations. Such inhomogeneity in the quartz is common in many areas (Burlinson, 1988b) and is due to growth zoning and multiple depositional episodes which are common features in high level deposits. After collecting and analysing 189 samples, the data were plotted as a cumulative frequency graph. This was rather irregular, but it appeared to be the result of 3 overlapping populations. The characteristics of the 3 constituent populations were derived using the methods described by Sinclair (1976). Using the 2% cumulative frequency points for each population, ranges of decrepitation values were selected which correspond either to individual populations or to overlap of 2 of the populations. The decrepitation values selected are listed in table 2 and these threshold values were used in the subsequent interpretation to define the levels considered to be anomalous. At a later date, after 274 samples had been analysed, the cumulative frequency graph was recalculated on the enlarged data. This differs markedly from the previous graph, despite the fact that 70> of the data used in its derivation is the same. It was not possible to split this graph into constituent populations with any degree of confidence. This indicates that the previous selection of threshold values is not ideal, but as no better selection was possible, those values were used throughout the interpretation of the results. Where multiple samples were collected at a site, there was not usually any difference visible in hand specimen between the quartz fragments collected. However, the decrepitation results were often markedly different, variations from 200 counts to 11000 counts between samples only 1 metre apart being observed. The long term analytical reproducibility of the instrument is better than 5% (Burlinson, 1988a) and the variations are not the result of analytical problems but reflect original differences in the samples. Such differences are due to multiple stages of quartz vein emplacement and growth. The recognition of these depositional conditions is in itself useful as it indicates a high level environment and this can rarely be discerned by mere visual mapping of the quartz. Some 300 samples from the Cotan area have now been analysed by decrepitation. To simplify the diagram, only the seperate sample sites are shown although there are 2 or 3 analyses at most sites. The decrepitation count value plotted is the maximum of the analyses at that sample site. The use of the maximum rather than an average result is in accordance with the results of the regional work discussed above, however, another plot using the average of all results at each site was also prepared. This gave rise to the same pattern of anomalies, but with lower contrast and less preaise definition. The similarity of the 2 plots serves to confirm that the anomalous zone is real and not merely a function of the method of presentation or interpretation of the results. Although the decrepitation results do not define a perfectly continuous zone, when considered together with the geological structure of the area it is reasonable to interpolate an anomalous zone almost the full length of the study area, together with a parallel zone to the west of the river. The rock geochemical samples were analysed either using fire assay or aqua regis digestions, with determination by atomic absorption spectophotometry. The dual assay methods were used as the presence of sulphides in some samples precluded the use of the aqua regia technique. Despite the mixed analytical methods, the results show a low level gold anomaly in quartz which is parallel and close to (but not coincident with) the decrepitation anomaly. Anomalous samples have gold contents of 20 to 500ppb while background levels are less than 20ppb. The anomalous samples often, but not always, contain anomalous levels of arsenic.