Nickel Laterite Mineral Resource Update · Limited_Ni Lat Resource_Report.docx Last Edited: 13/11/2014 10:46:00 AM Number of copies Snowden: 2 Weld Range Metals Limited: 2 . Weld

Weld Range Metals Limited

Range Well Nickel Project

Project No. AU2813

Nickel Laterite Mineral Resource Update

October 2014

Final

Office Locations Perth Level 3, 181 Adelaide Terrace, East Perth WA 6004 AUSTRALIA Tel: +61 8 9213 9213 Fax: +61 8 9322 2576 ABN: 99 085 319 562 [email protected]

Brisbane 2 Burke Street, Woolloongabba QLD 4102 AUSTRALIA PO Box 2207, Brisbane QLD 4001 AUSTRALIA Tel: +61 7 3249 0800 Fax: +61 7 3868 6515 ABN: 99 085 319 562 [email protected]

Johannesburg Technology House ,Greenacres Office Park, Cnr. Victory and Rustenburg Roads, Victory Park JOHANNESBURG 2195 SOUTH AFRICA PO Box 2613, Parklands 2121 SOUTH AFRICA Tel: +27 11 782 2379 Fax: +27 11 782 2396 Reg No. 1998/023556/07 [email protected]

Vancouver Suite 550, 1090 West Pender St, VANCOUVER BC V6E 2N7 CANADA Tel: +1 604 683 7645 Fax: +1 604 683 7929 Reg No. 557150 [email protected]

Calgary Suite 850, 550 11th Avenue SW CALGARY, ALBERTA T2R 1M7 Tel +1 403 452 5559 Fax +1 403 452 5988 [email protected]

Belo Horizonte Afonso Pena 2770, CJ 201 A 205 Funcionários, 30.130-007, BELO HORIZONTE MG BRASIL Tel: +55 (31) 3222-6286 Fax: +55 (31) 3222-6286 [email protected] London 1 Kingdom Street, Paddington Central, LONDON W2 6BD UK Tel: +44 (20) 3402 3022 [email protected]

Website www.snowdengroup.com

This report has been prepared by Snowden Mining Industry Consultants (“Snowden”) on behalf of Weld Range Metals Limited.

2014

All rights are reserved. No part of this document may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of Snowden.

Prepared By John Graindorge BSc(Hons), MAusIMM(CP) Principal Consultant

Reviewed By Richard Sulway

MAppSc, MAusIMM(CP) Divisional Manager – Applied Geosciences

Issued by: Perth Office Doc Ref: 141113_Final_AU2813_Weld Range Metals Limited_Ni Lat Resource_Report.docx

Last Edited: 13/11/2014 10:46:00 AM

Number of copies Snowden: 2 Weld Range Metals Limited: 2

mailto:[email protected]







http://www.snowdengroup.com/

Weld Range Metals Limited: Range Well Nickel Project Nickel Laterite Mineral Resource Update

1 Summary ............................................................................................................................. 6

2 Introduction .......................................................................................................................... 8 2.1 Location ................................................................................................................... 8 2.2 Historical estimates .................................................................................................. 9

3 Geology ............................................................................................................................. 10

4 Source data ....................................................................................................................... 12 4.1 Assays ................................................................................................................... 12

5 QA/QC ............................................................................................................................... 13 5.1 Re-assaying ........................................................................................................... 13 5.2 Re-sampling ........................................................................................................... 13 5.3 Standards ............................................................................................................... 14 5.4 Duplicates .............................................................................................................. 14 5.5 Blanks .................................................................................................................... 15 5.6 Further comments .................................................................................................. 15

6 Interpretation ..................................................................................................................... 16 6.1 Topography ............................................................................................................ 16 6.2 Nickel mineralisation envelope ............................................................................... 16 6.3 Domain coding ....................................................................................................... 18

7 Statistical analysis ............................................................................................................. 20 7.1 Compositing ........................................................................................................... 20 7.2 Summary statistics ................................................................................................. 21 7.3 Declustered and top-cut summary statistics ........................................................... 22

8 Variography ....................................................................................................................... 23

9 Block model ....................................................................................................................... 24 9.1 Model parameters .................................................................................................. 24 9.2 Density ................................................................................................................... 24 9.3 Estimation parameters ........................................................................................... 24

10 Model validation................................................................................................................. 26 10.1 Visual comparison .................................................................................................. 26 10.2 Average input and output grades ........................................................................... 27 10.3 Grade trend plots ................................................................................................... 27

11 Resource classification and reporting ................................................................................ 28 11.1 Resource classification ........................................................................................... 28 11.2 Cut-off grade .......................................................................................................... 28 11.3 Metallurgical test work ............................................................................................ 28 11.4 Resource reporting ................................................................................................. 29

11.4.1 Competent Person’s Statement .............................................................. 29 11.5 Grade tonnage relationship .................................................................................... 29

.Final October 2014 3 of 34


11.6 High grade zones ................................................................................................... 30

12 Comparison with previous resource estimates ................................................................... 32

13 Conclusions and recommendations ................................................................................... 33

14 References ........................................................................................................................ 34

Tables Table 1.1 Weld Range Inferred Mineral Resource at a 0.5% Ni cut-off grade

(JORC, 2012) ........................................................................................... 6 Table 7.1 Summary statistics for composited data from within the

interpreted nickel mineralisation ............................................................. 22 Table 7.2 Naïve and declustered mean grades ...................................................... 22 Table 8.1 Variogram spherical model parameters .................................................. 23 Table 9.1 Block model parameters ......................................................................... 24 Table 9.2 Boundary conditions applied for block grade estimation ......................... 25 Table 9.3 Estimation parameters ........................................................................... 25 Table 10.1 Mean comparison of sample data and block grades estimated in

first pass ................................................................................................. 27 Table 11.1 Inferred Mineral Resource at a 0.5% Ni cut-off grade (JORC,

2012) ...................................................................................................... 29 Table 11.2 Inferred Mineral Resource (JORC, 2012) reported by various Ni

cut-off grades (code 200 blocks only) ..................................................... 30 Table 11.3 Inferred nickel Mineral Resource estimate (JORC, 2012) within

target areas with higher grade nickel mineralisation (code 200 only) ....................................................................................................... 31

Table 12.1 Inferred Mineral Resource reported by Minara 2001 ............................... 32 Table 13.1 Inferred Mineral Resource at 0.5% Ni cut-off grade (JORC, 2012) ......... 33

Figures Figure 2.1 Location of the Weld Range Metals tenements ........................................ 8 Figure 3.1 Geological plan of the Weld Range Complex ......................................... 11 Figure 5.1 Scatter-plot of original and re-assay nickel data ..................................... 13 Figure 5.2 Scatter-plot of original and field duplicate nickel data ............................. 14 Figure 5.3 Scatter-plot of laboratory repeat nickel data ........................................... 15 Figure 6.1 All nickel data – histogram and log probability plot ................................. 16 Figure 6.2 Lateral extents of interpreted nickel mineralisation ................................. 17 Figure 6.3 Cross section illustrating nickel mineralisation (top – blue, base –

cyan) and base of resource model (magenta) – 578140 mE (x2 vertical exaggeration) ............................................................................. 18

Figure 6.4 Interpreted nickel mineralisation horizon and chromium resource .......... 19 Figure 6.5 Cross-section illustrating block coding of the interpreted nickel

mineralisation and chromium mineralisation – 575460 mE (x10 vertical exaggeration) ............................................................................. 19

Figure 7.1 Histogram of sample length for entire dataset ........................................ 20 Figure 7.2 Histogram of sample length for data from within the nickel

mineralisation ......................................................................................... 21



Figure 10.1 Cross-section showing Ni grades – 576940 mE (x10 vertical

exaggeration) ......................................................................................... 26 Figure 10.2 Cross-section showing Ni grades – 578140 mE (x10 vertical

exaggeration) ......................................................................................... 26 Figure 11.1 Global grade tonnage curve (code 200 blocks only) ............................... 29 Figure 11.2 Nickel resource grade contours .............................................................. 30 Figure 11.3 Location of nickel resource and higher grade nickel zones ..................... 31 Figure 12.1 Grade tonnage curves of Snowden 2010 resource (codes 200

and 300) and Minara 2001 polygonal resource ....................................... 32

Appendices Histograms and probability plots Appendix A Variograms for Ni, Cr, Fe and Co Appendix B Grade trend plots for Ni, Cr, Fe and Co Appendix C JORC 2012 Table 1 reporting criteria Appendix D



1 Summary In November 2010, Snowden Mining Industry Consultants (“Snowden”) generated a Mineral Resource estimate at the request of Weld Range Metals Limited (“Weld Range Metals”) for the nickel mineralisation at the Weld Range Project (Snowden, 2010c), located in the Murchison district in the mid-west region of Western Australia. The 2010 nickel laterite Mineral Resource was reported based on the guidelines of the 2004 JORC Code.

In September 2014, Weld Range Metals requested that Snowden update the Mineral Resource reporting of the nickel laterite resource in accordance with the guidelines of the 2012 edition of the JORC Code. No additional drilling or exploration activities have taken place since the 2010 resource estimate and as such the November 2010 resource estimate remains current.

Snowden generated mineralisation interpretations for the nickel enriched horizons within the laterite. Geological logging data is incomplete and the interpreted nickel mineralisation is constrained to within 40 m depth from the surface. Verification of the base of weathering is required to confirm that the nickel resource estimate is restricted to within the laterite horizon.

Snowden completed summary statistics and variography on the composite drillhole data from within the interpreted nickel mineralisation and estimated grades into the block model for Ni, Cr, Fe, and Co using ordinary kriging. Grade estimation was constrained to within the interpreted nickel mineralisation horizons.

Weld Range Metals supplied the density value (1.5 t/m3) to be applied to the model and has advised that an economic cut-off grade of 0.5% Ni should be used as the basis for resource reporting, based on preliminary metallurgical testwork.

In November 2009, Snowden prepared a resource estimate for the chromium mineralisation at Weld Range (Snowden, 2009a). As part of the 2010 estimate the chromium resource was updated for Ni, Co and Fe within the interpreted chromium mineralisation and re-reported in 2010 (Snowden, 2010a). The block model was coded to permit reporting of the resource within the nickel mineralisation horizons and within and outside the chromium mineralisation horizons generated in 2009.

The nickel Mineral Resource has been classified as Inferred in its entirety, in accordance with the 2012 edition of the JORC Code based upon the geology, mineralisation interpretations and drillhole data. Table 1.1 lists the Mineral Resource for the Weld Range nickel laterite deposit above a 0.5% Ni cut-off grade. This nickel Mineral Resource has been reported to exclude the resource contained within the chromium Mineral Resource. The cut-off grade of 0.5% Ni is based on the assumption that the resource would be extracted by conventional open-pit mining methods with the ore beneficiated to upgrade the nickel content during processing.

Table 1.1 Weld Range Inferred Mineral Resource at a 0.5% Ni cut-off grade (JORC, 2012)

Tonnes (Mt) Ni % Cr % Fe % Co %

Total 385.3 0.64 0.78 19.7 0.04 Note: Excludes resource contained within the 2009 chromium Mineral Resource

Snowden has identified two zones that contain higher grade nickel comprising 24.9 million tonnes with an average grade of 0.99% Ni, 0.62% Cr, 19.7% Fe and 0.05% Co above a 0.8% Ni cut-off grade nickel, within the total Inferred Mineral Resource.



Snowden considers that further work is required in terms of definition of the topographic surface, drillhole survey data, quality assurance/quality control (“QA/QC”) data, infill drilling and a program of density determinations to improve the classification of this resource estimate.



2 Introduction 2.1 Location

The Weld Range Project is located approximately 65 km north-northwest of Cue, some 85 km west-southwest of Meekatharra and some 600 km north-northeast of Perth in Western Australia (Figure 2.1).

Figure 2.1 Location of the Weld Range Metals tenements



2.2 Historical estimates Within the project area, there is potential for Ni, Cr and PGE mineralisation in a variety of styles. This 2010 nickel resource estimate is based on the potential for nickel mineralisation within the laterite horizon and is based on drilling undertaken from 1988 to 2000.

Snowden has previously undertaken three resource estimates for the chromium mineralisation (Snowden, 1989; Snowden, 1994; and Snowden 2009a). The 1989 and 1994 resource estimates were restricted to within MLA51/546.

In July 2010 Snowden developed a resource estimate for nickel within the laterite horizon in the vicinity of the high grade chromium mineralisation identified at Chrome Hill (Snowden, 2010a). In September 2010, this model was extended to cover the entire nickel mineralisation within the laterite horizon at the Weld Range Project. All resource reporting for this model excludes the nickel mineralisation identified and reported as part of the chromium Mineral Resource (Snowden, 2009 and 2010b), except where comparisons are made with previous nickel resource estimates in Section 12.

Historical resource estimates of the nickel laterite mineralisation were previously compiled by Sons of Gwalia Ltd in 1997 (Goertz, 1997) and by Minara in 2001. The 1997 estimate reported 94.3 million tonnes with an average grade of 1.03% Ni based on a cut-off grade of 0.8% Ni. This model was based on planar polygons, was constrained to within 40 m of the surface and used a density 1.6 t/m3. Additional drilling was undertaken and in 2001 a polygonal resource estimate prepared by Minara reported 65.7 million tonnes with an average grade of 1.01% Ni based on a cut-off grade of 0.8% Ni and a density of 1.5 t/m3. This estimate is discussed further in Section 12.



3 Geology The Weld Range Complex is a laterally extensive ultramafic/mafic complex hosted within the Gabanintha Formation of the Luke Creek Group and is the lower of two greenstone units contained within the Murchison Supergroup. The Weld Range Complex is bound to the south by concordant dolerites and banded iron formations of the Weld Range, and to the north and west by granite. The complex is generally divisible into an ultramafic (north) and mafic (south) series of layered rock units. The ultramafic units are estimated to be approximately 5 km thick, with the more southern mafic series being approximately 3.5 km thick. The geology of the Weld Range project area is illustrated in Figure 3.1.

The detailed geology and zonation of the Weld Range Complex is not clearly understood due to the deep lateritic weathering and limited outcrop within the area. Overall, the general shape of the complex is that of a recumbent lopolith (reclining triangular funnel shape). The sequence strikes northeast and dips steeply to the southeast.

Primary nickel mineralisation at Weld Range is associated with adcumulate and mesocumulate phases of the ultramafic body. Only rare primary nickel sulphide (millerite) grains have been identified and the only currently known nickel mineralisation is located within the lateritic profile above the main layered ultramafic sequence. Residual laterite outcrops in several areas, most notably toward the southwest of the ultramafic series and at Chrome Hill.

The typical residual laterite profile comprises an upper pisolitic ferricrete horizon up to 15 m thick, that may be locally enriched in chromium, overlying siliceous limonite (>25 m) with elevated nickel and cobalt grades. The profile then passes through a highly siliceous and weakly ferruginous saprolite that is often replaced by massive silicrete, into a serpentinised olivine – chromite adcumulate that may contain stockwork magnesite near the lower saprolite. There is abundant free silica within the laterite profile, especially towards the base of the laterite which is attributed to the presence of olivine adcumulate ultramafic rocks in the substrate.

Data from geological logging of the drillholes is incomplete in the existing databases and only 44% of the intervals that have been assayed for nickel have a weathering code. The current resource model is restricted to within 40 m of the surface (excepted where it is present within the chromium Mineral Resource). Future resource estimation should include modelling of the ferricrete and saprolite horizons.



Figure 3.1 Geological plan of the Weld Range Complex



4 Source data The Weld Range project area has been drilled by a number of separate programs using a combination of reverse circulation (“RC”), aircore (“AC”) and rotary air blast (“RAB”) methods. The majority of drilling occurred in the period from June 1998 to December 2000, but records of historic drilling exist back to around 1971.

Snowden developed a Gemcom database for resource estimation of the chromium mineralisation in 2009 (Snowden, 2009a). This database was used for the 2010 nickel resource estimate and includes data from 773 drillholes. It excludes all RAB drillholes and holes drilled in the 1970s for nickel, as they were designed to target nickel sulphide and are located to the south of the lateritic nickel mineralisation. This database was updated in September 2010 to incorporate the cobalt data.

The spacing between the drill lines varies throughout the project, as shown in Figure 6.2. Holes have been drilled on sections of 200 m to 800 m spacing east-west, and approximately 40 m to 200 m spacing north-south. Within the Chrome Hill area, infill drilling is spaced at 40 m east-west and 20 m north-south, however not all of these drillholes have been assayed for nickel.

A number of adjustments and assumptions were made with regard to collar elevation data and for drillholes with missing survey data. These are discussed in detail in Snowden, 2009a. There are three pairs of coincident drillhole locations (WRA070 and WRA073; WRA026 and WRA027; WRA029 and WPRC001) that are not within the chromium mineralisation interpreted in 2009, but are within the interpreted nickel mineralisation. In each case, data from the first drillhole (i.e. WRA070, WRA026 and WRA029) was excluded from the data coding.

4.1 Assays The 2009 database contained assay data for Cr, Fe and Ni and was updated for Co data. This data was recorded as ppm and Snowden created an additional field to record the Co data expressed in percent. Snowden was also asked to import any available SiO2, Al2O3 and MgO data and determine the coverage of this data within the nickel laterite mineralisation. Snowden reviewed the historical data provided by Weld Range Metals and was unable to locate any assay data for these elements. Snowden located Mg data and determined that 58% of the intervals that have been assayed for nickel have assay results for Mg.

Below detection limit data for cobalt was replaced with values of 0.001% Co. There are 53 Co values of zero which are associated with the RRC series of drillholes. These samples have been assayed for other elements and so do not represent missing samples. It was assumed they represent below detection assay results and were replaced with 0.001% Cr.

As discussed in Snowden 2009a, an unusually high value of 11.6% Ni is present in RRC068. Examination of surrounding data indicates that this is likely to be a date entry error and Snowden replaced this with a Ni value of 1.16%.

Two unusually high values of 7.4% Co and 2.6% Co were noted at the top of RRC057. Adjacent drillholes did not support these high grades and Snowden elected to exclude these from the current database. Assay data from these intervals should be checked.



5 QA/QC Snowden is only aware of limited QA/QC data being collected as part of the various drilling campaigns. As discussed in Snowden 2009a, the QA/QC data is considered to be insufficient to adequately define the quality of the drillhole sampling and assaying.

5.1 Re-assaying As part of Snowden’s review in 2009 (Snowden, 2009a) seven WRRC series drillholes (WRRC0137, 0138, 0143, 0175, 0176, 0200 and 0248) were selected to be re-assayed from the sample pulps available at Weld Range Metals storage facility. The sample pulps were analysed by fusion XRF methods at ALS Chemex’s Brisbane laboratory for Cr, Ni and Fe as well as a suite of additional elements. The results for Cr and Fe are discussed in Snowden, 2009a. The scatter-plot for nickel is included as Figure 5.1 and the results indicate a good correlation between the original nickel assay and the re-assay with a correlation coefficient of 0.995.

Figure 5.1 Scatter-plot of original and re-assay nickel data

5.2 Re-sampling A selection of 31 RWAC series drillholes were re-sampled in the field from the bulk reject by Minara in 1999 to 2000. The re-sampling was primarily for nickel laterite mineralisation. Details of the results could not be located.

Snowden is aware that all 4 m composite samples from drillholes WEC093 to WEC158 with grades greater that 0.4% Ni were re-assayed in 1 m splits. Some 1,312 samples were re-assayed by Ultra Trace laboratories for 11 elements by ICP-OES and merged into the Anaconda (now Minara) database (McConville, 2000). McConville’s analysis of the Ni, Co and Mg data indicates that there is a good correlation between 4 m composites and the 1 m splits.

0

0.5

1

1.5

2

0 0.5 1 1.5 2

Ni r

e-as

say

(%)

Ni vs re-assay Ni (%)

Normal ScatterplotWeld Range

Ni vs re-assay Ni

Correlation Coefficient = 0.995

+/- 10%



5.3 Standards Snowden has only identified ‘certified’ standard material samples being submitted with drillholes WRRC0001 to WRRC0249 where at least one standard was submitted for each batch of 30 samples. Seven different standards of various materials and grades ranging from 120 ppm to 1.68% Ni were used. Snowden is not aware of how or where the standards were manufactured, or the certified standard deviation for the material. Only the nominal standard grade was reported. As such, Snowden has only reviewed the general assay results for the standards rather than a detailed accuracy and precision assessment.

The four higher grade standards [IGS-23 (1.68%), Laterite-83 (0.56%), Laterite-84 (1.61%), and Laterite 88 (1.52%)] show reasonable results with acceptable analytical accuracy (with differences of 1% or less between the mean and the certified value), whereas Laterite-85 (2.15%) and the two lower grade standards [SARM-4 (120 ppm) and SARM-43 (252 ppm)] show relatively poor analytical accuracy.

5.4 Duplicates Snowden has only identified field duplicate samples being collected from drillholes WRRC0001 to WRRC0249, where field duplicates were collected at a ratio of approximately 1:20. The scatter-plot for nickel is included as Figure 5.2 and the results indicate a good correlation between the original nickel assay and the field duplicate assay with a correlation coefficient of 0.986.

Figure 5.2 Scatter-plot of original and field duplicate nickel data

Snowden has only identified laboratory duplicate samples being assayed from drillholes WRRC0001 to WRRC0249. The scatter-plot for nickel is included as Figure 5.3 and the results indicate a good correlation between the original nickel assay and the laboratory duplicate assay with a correlation coefficient of 1.00.



Figure 5.3 Scatter-plot of laboratory repeat nickel data

5.5 Blanks Snowden has only identified blank samples submitted on drillholes WRRC0001 to WRRC0249 where blank samples were submitted at the start of each job and after every 90 samples thereafter. Snowden has not identified nor reviewed any analysis of the blank samples other than a brief statement in the Anaconda, October 2000 Annual Technical Report that states they occur within acceptable tolerance levels.

5.6 Further comments In order to generate a resource model within which a significant proportion of the mineralisation could be reported as Measured and/or Indicated in accordance with the JORC 2012 guidelines requires a comprehensive QA/QC system to be implemented. The QC data needs to be routinely analysed and documented in order to verify the quality of the sample data.

Results from the available data for nickel are good however Snowden considers that appropriate QA/QC procedures may not have been applied during all of the sampling.



6 Interpretation 6.1 Topography

As discussed in Snowden 2009a, 370 of the drillholes appeared to have default or estimated collar elevation values. Snowden used the collar elevation data from the 403 drillholes that were regarded as having measured collar elevations to interpret a topographic surface. This surface was used to estimate collar elevations for the remaining drillholes, code the block model and to constrain the interpreted nickel mineralisation envelope (discussed below) to below the topographic surface.

A more reliable topographic surface will have to be developed for future resource estimation. Snowden recommends that the collar elevation of all the existing drillholes should be measured and incorporated into the drilling database. Additional survey data should be obtained throughout the project area to permit the generation of a reliable topographic surface.

6.2 Nickel mineralisation envelope Snowden examined a histogram and log probability plot of the nickel data composited to 1 m intervals (Figure 6.1) and interpreted a slight inflection in the data at a grade of about 0.4% Ni. Snowden used this mineralisation threshold grade to interpret a nickel mineralisation envelope.

Figure 6.1 All nickel data – histogram and log probability plot



A nominal Ni grade of 0.4% was used to interpret the horizon of nickel mineralisation. Surfaces were developed for the top and base of the nickel mineralisation based on the top of the first and the bottom of the last interval with an assay of 0.4% Ni. These surfaces were amended to maintain a minimum mineralisation thickness of 2 m (to reflect the likely mining selectivity) and to adjust the base of the nickel mineralisation; intersections in the shallow drillholes within the Chrome Hill area were used to determine the top of mineralisation, but where the drillhole had not intersected the full thickness of nickel mineralisation, the base of mineralisation was interpreted from the adjacent deeper drillholes.

The lateral extents of the nickel mineralisation were interpreted to restrict the interpretation within the laterite and to exclude isolated nickel mineralisation intersections. A clipping polygon was used to constrain the interpretation (Figure 6.2). The interpreted mineralisation was extended to about 50 m on section and 100 m along strike beyond the drilling.

Figure 6.2 Lateral extents of interpreted nickel mineralisation

Over 50% of the intervals with nickel assay data do not have lithology or weathering codes in the database. Snowden examined the data that had lithology and weathering data and determined that the deepest logged laterite interval is at 64 m, the average depth to the logged base of laterite is 19.5 m and that the 99th percentile of the base of the logged laterite intersections is at 36 m. Snowden has restricted the interpreted nickel mineralisation and resource estimate to within 40 m of the topographical surface and data was screened to within 45 m of the surface. Weathering and lithological data should be logged for all future drilling and the base of the resource model should be reinterpreted to ensure that it is restricted to within the laterite horizon. Snowden notes there are a number of deeper intersections of nickel mineralisation on Section 578140 mE (Figure 6.3) that lie below 40 m depth and have not been included in the resource model. Additional drilling is recommended to determine the weathering and lithology and define the along strike extent of this mineralisation.



Figure 6.3 Cross section illustrating nickel mineralisation (top – blue, base –

cyan) and base of resource model (magenta) – 578140 mE (x2 vertical exaggeration)

The interpreted mineralised surfaces were used to code and extract the Ni, Cr, Fe and Co data and to code the block model for the resource estimate.

6.3 Domain coding The following codes were applied to the data and block model:

• 100 = within chromium mineralisation envelope and reported for chromium Mineral Resource

• 200 = within nickel mineralisation envelope and reported for nickel Mineral Resource

• 300 = within both chromium and nickel mineralisation envelopes and reported for chromium Mineral Resource.

For the nickel resource estimate, data analysis and resource estimation was restricted to codes 200 and 300. Code 100 is used to identify data and blocks that are outside the interpreted nickel mineralisation and are within the chromium mineralised envelope used for analysis and block grade estimation of the chromium resource as reported in Snowden 2009a and 2010b.

The extents of the interpreted nickel mineralisation surfaces and the extent of the chromium Mineral Resource are illustrated in Figure 6.4 and the domain coding is illustrated in Figure 6.5.



Figure 6.4 Interpreted nickel mineralisation horizon and chromium resource

Figure 6.5 Cross-section illustrating block coding of the interpreted nickel mineralisation and chromium mineralisation – 575460 mE (x10 vertical exaggeration)



7 Statistical analysis 7.1 Compositing

The majority of the nickel assay data has been based on samples taken over intervals of 1 m. For the entire data set (Figure 7.1), the minimum sample interval is 0.09 m and the maximum assayed interval is 8 m. Less than 3% of the data is from sample lengths of less than 1 m, 46% from samples of 1 m, 22% from samples of 2 m and 25% is from samples of 4 m, with the remaining assay data based on sample lengths of 5 m or more.

Within the interpreted nickel mineralisation envelope 52% of the nickel assay data is from 1 m sample lengths and 32% and 16% from sample lengths of 2 m and 4 m respectively (Figure 7.2).

Snowden used a down-hole composite length of 1 m for the resource estimate. The implications of using this composite length for the samples greater than 1 m is that they will be split (i.e. a 2 m sample will be split into two one metre composites), giving greater weight to the sample for statistical analysis. This will affect 48% of the nickel assay data.

Figure 7.1 Histogram of sample length for entire dataset



Figure 7.2 Histogram of sample length for data from within the nickel

mineralisation

7.2 Summary statistics Snowden completed summary statistics on the one metre composites for Ni, Cr, Fe and Co from within the interpreted nickel mineralisation. Summary statistics are included in Table 7.1 and histograms and log probability plots are included in Appendix A.

The Ni and Fe data from within the nickel mineralisation have low (<1) coefficients of variation (CV, ratio of the standard deviation to the mean) of 0.48 and 0.64 respectively. The CV for chromium data from within the nickel mineralisation and outside the chromium mineralisation is higher (1.22). The histogram of the chromium data from within the nickel mineralisation only (200) in Appendix A indicates that the chromium data is bimodal and that a few low and high grade outliers are present. It was decided not to apply a top-cut to the chromium data. Similarly, no top-cuts were applied to the nickel or iron assay data.

The CV for the cobalt data is 1.36 and examination of the histogram and log probability plot indicates that there is a higher grade population above 0.3% Co. Snowden reviewed the disintegration of the histogram tail and elected to top-cut the cobalt data to a grade of 0.3%. This affects less than 1% (54 data points) of the cobalt data.



Table 7.1 Summary statistics for composited data from within the interpreted

nickel mineralisation

Statistics Ni % Cr % Fe % Co %

Samples 7276 6799 6151 6809

Minimum 0.039 0.007 0.56 0.003

Maximum 3.56 11.90 59.20 1.62

Mean 0.59 0.77 20.56 0.04

Standard deviation 0.28 0.95 13.08 0.06

Coefficient of variation 0.48 1.22 0.64 1.36

Variance 0.08 0.90 170.96 0.003

Skewness 2.46 3.76 1.03 10.22

Per

cent

ile a

nd g

rade

at p

erce

ntile

10th 0.35 0.08 7.52 0.01

20th 0.41 0.12 10.20 0.02

30th 0.45 0.17 12.00 0.02

40th 0.48 0.26 14.00 0.03

50th 0.52 0.43 16.40 0.03

60th 0.58 0.68 19.45 0.03

70th 0.64 0.96 23.88 0.04

80th 0.74 1.34 31.00 0.05

90th 0.92 1.90 42.20 0.08

95th 1.12 2.37 49.30 0.12

97.5th 1.36 2.85 52.50 0.18

99th 1.69 4.05 54.90 0.28 Note: Cr is for code 200 data only and excludes data from within interpreted chromium mineralisation (codes 300 and 100).

7.3 Declustered and top-cut summary statistics Snowden examined the declustered means of the composite data to determine an appropriate sample mean against which to validate the model grades. There are only negligible differences between the naive and declustered mean grade within domain codes 200 and 300. The irregular spaced and clustered drilling pattern at Chrome Hill has targeted the chromium mineralisation that is mainly contained within domain code 100.

Table 7.2 summarises declustered statistics for Ni, Cr and Fe and the top-cut and declustered statistics for Co within the interpreted nickel mineralisation.

Table 7.2 Naïve and declustered mean grades

Variable Codes Naïve mean Declustered mean

Ni (%) 200 and 300 0.594 0.589

Cr (%) 200 0.774 0.773

Fe (%) 100, 200 and 300 23.992 24.150

Co (%) 100, 200 and 300 0.041 0.038



8 Variography Snowden investigated variography for the composited nickel data within the interpreted nickel mineralisation. Variography for chromium was based on data from within the chromium mineralised domains and variography for Fe and Co was based on data from within the nickel mineralisation and/or the chromium mineralisation.

The longest range of mineralisation continuity for nickel and chromium was interpreted to be oriented along 060° and is consistent with the orientation of the interpreted mineralisation envelope. The longest range for Fe and Co was interpreted to be oriented along 150°.

Traditional directional variograms were modelled using a spherical model. Three structures were interpreted for Ni, Cr, Fe and Co and the sills were normalised to a value of one. Interpreted variogram parameters are listed in Table 8.1 and plots for the horizontal variogram fan and directional variograms are included in Appendix B for Ni, Cr, Fe and Co.

The along-strike and across-strike variograms for nickel are poorly defined, indicating that additional infill drilling is required for definition of the mineralisation.

Table 8.1 Variogram spherical model parameters

Grade Direction Nugget Sill 1 Range 1 (m) Sill 2 Range 2

(m)

Ni

0°Æ060° 35 230

0°Æ150° 0.10 0.45 100 0.45 160

-90°Æ360° 7 9

Cr

0°Æ060° 120 335

0°Æ150° 0.08 0.36 45 0.56 300

-90°Æ360° 3 10

Fe

0°Æ060° 170 320

0°Æ150° 0.03 0.31 60 0.66 480

-90°Æ360° 25 25

Co

0°Æ060° 50 180

0°Æ150° 0.10 0.37 400 0.53 600

-90°Æ360° 9 12



9 Block model 9.1 Model parameters

Block modelling was completed using Gemcom software, with the surfaces described in Section 6, and according to the parameters listed in Table 9.1.

A parent cell block model (no sub-celling) was developed. Blocks that have a minimum of 50% of the block volume within the nickel mineralisation were coded for grade estimation. A block model code of 200 was used for blocks that are contained within the nickel mineralisation and a block model code of 300 was used for blocks that are within both the nickel and chromium mineralisation.

In addition to this the percentage of each block that lies below the topographical surface was applied for volume and tonnage estimation.

Table 9.1 Block model parameters

Parameter X Y Z

Origin 568400 mE 7027100 mN 570 mRL

Extent 13,500 m 8,625 m 130 m

Block size 100 m 25 m 2 m

Number blocks 135 345 65

9.2 Density Weld Range Metals supplied the average density value of 1.5 t/m3 for conversion of volume to tonnage estimates. This density value was applied by Minara for a polygonal resource estimate of nickel within the Weld Range project area in 2001. Density measurements have not yet been obtained for this area and Weld Range Metals has elected to use the lower density estimate of 1.5 t/m3 for the current resource estimate. The bulk density value is considered by Snowden to be conservative and may be locally higher, however bulk density measurements are required to assess this.

Snowden recommends that a program of density testing should be carried out. This information is required, in conjunction with other detailed data, before any of the resources at Weld Range can achieve a resource classification higher than Inferred.

9.3 Estimation parameters Snowden estimated Ni, Cr, Fe and Co using ordinary kriging with the variogram parameters determined by the variography analysis discussed in Section 8. The kriging estimation parameters are listed in Table 9.3. Ni, Fe and Co grades were estimated into blocks with codes of 100, 200 and 300. For Ni, a hard boundary was applied for blocks with a code of 100 and soft boundary conditions were applied for block with codes of 200 and 300. For Fe and Co soft boundary conditions were applied to blocks with codes of 100, 200 and 300. Blocks with codes of 100 and 300 have not been re-estimated for Cr and these block grades are as estimated by Snowden in 2009 (Snowden, 2009a and 2010b). For Cr grade estimation into blocks with a code of 200 a hard boundary condition was applied data with a code of 200 was used for block grade estimation). The boundary conditions are listed in Table 9.2.



Table 9.2 Boundary conditions applied for block grade estimation

Field Block code 100 Block code 200 Block code 300

Ni Data code 100 Data code 200 and 300 Data code 200 and 300

Cr Data code 100 and 300 (as estimated in Snowden, 2009a) Data code 200 Data code 100 and 300 (as

estimated in Snowden, 2009a)

Fe Data codes 100, 200 and 300 Data codes 100, 200 and 300 Data codes 100, 200 and 300

Co Data codes 100, 200 and 300 Data codes 100, 200 and 300 Data codes 100, 200 and 300

Block grade estimation of Ni, Fe and Co and grade estimation of Cr into blocks with a code of 200, applied search ellipse dimensions based on the Ni variogram parameters. The vertical search was restricted to 4 m to reduce vertical smoothing in the resource model.

For Cr, 61% of the block grades with code 200 were estimated using the first search pass. For Ni and Fe, 62% of the block grades with codes of 100, 200 and 300 were estimated using the first search pass and for Co, 68% of the block grades were estimated using the first search pass. Second and third search passes were applied with dimensions listed in Table 9.3. The third search was applied to estimate grade into the remaining 3% of the blocks and applied soft boundary conditions to all block codes.

Table 9.3 Estimation parameters

Parameter Value

Search 1 160 m x 230 m x 4 m

Gemcom rotation 0, 0, -60 (Z,X,Z rotation)

Minimum samples 4

Maximum samples 20

Discretisation 5 x 5 x 2

Search 2 320 m x 460 m x 8 m

Minimum samples (search 2) 4

Search 3 640 m x 960 m x 16 m

Minimum samples (search 3) 2



10 Model validation Snowden validated the estimate by:

• a visual comparison of drillhole and block grades • comparing the mean input and output grades • compilation of grade trend plots using northing, easting and elevation perspectives.

10.1 Visual comparison Cross-sections of the block grades and the drillhole data were visually examined throughout the resource model. Examples are included in Figure 10.1 and Figure 10.2. This comparison indicated a good correlation between the input drillhole data and the estimated block grades.

Figure 10.1 Cross-section showing Ni grades – 576940 mE (x10 vertical exaggeration)

Figure 10.2 Cross-section showing Ni grades – 578140 mE (x10 vertical exaggeration)



10.2 Average input and output grades For the comparison of the input and output grades, the mean values of the input data and the average grade of the blocks were examined. The results, as presented in Table 10.1, indicate good validation of the block grades with the sample grades with the block grades being within 5% of the sample mean.

Table 10.1 Mean comparison of sample data and block grades estimated in first pass

Grade Codes Sample average

Search pass 1 (only)

Block average Ratio (Block/Sample)

Ni (%) 200 and 300 0.59 0.58 0.98

Ni (%) 100 0.19 0.20 1.05

Cr (%) 200 0.77 0.73 0.95

Fe (%) 100, 200 and 300 23.99 21.67 0.90

Co (%) 100, 200 and 300 0.041 0.039 0.95

10.3 Grade trend plots Snowden generated grade trend plots by easting, northing and elevation slices for Ni, Cr, Fe and Co. The plots for Ni, Cr, Fe and Co are included in Appendix C.

The grade trend plots show a good comparison between the estimated Ni, Cr, Fe and Co block grades and the input sample grades. The validation plots for Ni, Fe and Co are based on all coded data and blocks (codes 100, 200 and 300). The validation for Cr is based on blocks and data from with domain code 200 only. The Cr block grades for domain codes 100 and 300 are as were estimated and validated in 2009 (Snowden 2009a and 2010b).



11 Resource classification and reporting 11.1 Resource classification

The resource has been classified as Inferred in accordance with the guidelines of the 2012 JORC Code based upon the geology, mineralisation interpretations and drillhole data. Snowden’s assessment of the relevant criteria outlined in Table 1 of the 2012 JORC Code is presented in Appendix D.

Snowden considers that further work is required in terms of topographic data, drillhole survey data, QA/QC data, infill drilling and a program of density determinations to improve the classification of this resource estimate.

11.2 Cut-off grade For the purposes of the Inferred Resource, it has been assumed that the Weld Range Project would be mined by conventional open-pit methods and that the ore can be beneficiated during processing. Weld Range Metals advised that metallurgical testwork conducted by Anaconda Nickel in 2000 (Anaconda, 2000), on eight composite RC samples of Weld Range mineralisation, indicates that the mineralisation can be beneficiated to upgrade the nickel content of the ore.

The resource has consequently been reported above a cut-off grade of 0.5% Ni. Snowden cautions that further metallurgical testing is required to confirm these assumptions and that this testing may return a higher economic cut-off grade.

11.3 Metallurgical test work A number of preliminary metallurgical studies have been completed on lateritic material from the Weld Range Project. Anaconda Nickel conducted test work in 2000 (Anaconda, 2000), on eight composite samples, sourced from RC drill chips, to assess beneficiation by dry screening at multiple screen sizes. Results of the dry screening at 0.5 mm, showed an average upgrade for the Ni grade of 44%, recovering 62% of the metal and with a mass recovery of 42% passing -0.5 mm.

Bottle roll beneficiation test work on 439 samples sourced from RC chips, conducted in 2001 by Anaconda (Anaconda, 2001), showed an upgrade in the Ni grade of 82% for the “high upgrade” domain (defined as having a Mg content <6% and Si content >35%) with a mass recovery of 36% passing -0.25 mm. For the “low upgrade” domain (defined as having an Mg content < 6% and a Si content <35%) the Ni content was only upgraded by 16% with a mass recovery of 58% passing -0.25 mm.

Pressure acid leach testing conducted on four composite samples (from RC drill chips) showed average Ni and Co extractions of 97.1% and 97.4% respectively, after 60 minutes of residence time (at 255°C and 4,500 kPa) and with an average acid addition of 273 kg/t (Anaconda, 2001).

Snowden cautions that at this stage the metallurgical testwork is considered preliminary and further testing is required.



11.4 Resource reporting Table 11.1 lists the Inferred Mineral Resource above a 0.5% Ni cut-off grade. This resource estimate excludes mineralisation that is contained within the chromium Mineral Resource model.

Table 11.1 Inferred Mineral Resource at a 0.5% Ni cut-off grade (JORC, 2012)

Tonnes

Mt Ni %

Cr %

Fe %

Co %

Total 385.3 0.64 0.78 19.7 0.04 Note: Excludes resource contained within the 2009 chromium Mineral Resource.

11.4.1 Competent Person’s Statement

John Graindorge is a Chartered Professional and Member of the Australasian Institute of Mining and Metallurgy (“MAusIMM”) and has sufficient experience which is relevant to the style of mineralisation and type of deposit under consideration and to the activity to which he is undertaking to qualify as a Competent Person as defined in the 2012 edition of the “Australasian Code for Reporting of Exploration Results, Mineral Resources and Ore Reserves” (“the JORC Code”). John Graindorge is a full-time employee of Snowden Mining Industry Consultants Pty Ltd. John Graindorge consents to the inclusion in the report of the matters based on his information in the form and context in which it appears.

11.5 Grade tonnage relationship Figure 11.1 and Table 11.2 show the grade tonnage relationship for the global resource for a range of Ni cut-off grades. Ni grade contours based on the extents of blocks with grades of >0.5% Ni and >0.8% Ni are illustrated in Figure 11.2.

Figure 11.1 Global grade tonnage curve (code 200 blocks only)



Table 11.2 Inferred Mineral Resource (JORC, 2012) reported by various Ni cut-

off grades (code 200 blocks only)

Cut-off grade % Ni

Tonnes Mt

Ni %

Cr %

Fe %

Co %

0.5 385.3 0.64 0.78 19.7 0.04

0.8 44.2 0.95 0.75 19.8 0.06

1.0 11.4 1.18 0.77 19.4 0.06

1.3 1.9 1.52 0.79 20.7 0.06

Figure 11.2 Nickel resource grade contours

11.6 High grade zones Weld Range Metals requested Snowden to identify higher grade zones of nickel mineralisation. It is planned to consider these areas for future drilling and resource delineation. Snowden identified two zones of mineralisation that are illustrated in Figure 11.3.

Snowden identified two zones of mineralisation that contain fairly continuous, higher grade nickel, that are located to the north of Chrome Hill. These zones comprise a total of 24.9 million tonnes with an average grade of 0.99% Ni, 0.62% Cr, 19.7% Fe and 0.05% Co above a 0.8% Ni cut-off grade. The higher grade resources are listed in Table 11.3 and the zones are illustrated in Figure 11.3.



Zone 1 and 2 have an average thickness of 13 m and extends to a maximum depth of 40 m (as restricted by the interpreted laterite depth). The overburden thickness at Zone 2 ranges from 2 m to 26 m and has an average thickness of 10 m. Zone 1 intersects the surface and has an average overburden thickness of 10 m and a maximum thickness of 34 m.

Table 11.3 Inferred nickel Mineral Resource estimate (JORC, 2012) within target areas with higher grade nickel mineralisation (code 200 only)

Code Cut-off grade

% Ni Tonnes

Mt Ni %

Cr %

Fe %

Co %

Zone 1 0.0 23.7 0.80 0.71 18.0 0.05

0.8 12.1 0.95 0.72 18.3 0.06

Zone 2 0.0 32.1 0.76 0.40 18.5 0.04

0.8 12.8 1.03 0.52 20.9 0.04

Total 0.0 55.8 0.78 0.53 18.3 0.04

0.8 24.9 0.99 0.62 19.7 0.05

Figure 11.3 Location of nickel resource and higher grade nickel zones



12 Comparison with previous resource estimates In 2001, Minara generated a polygonal resource estimate for the nickel potential at Weld Range. Documentation on this resource estimate is not available in the digital data provided by Minara to Weld Range Metals. Review of the data contained in the file “Inferred Sectional Resource102001.xls” indicates that the Mineral Resource is constrained to within the laterite horizon and extends from 569000 mE to 581400 mE. An average density of 1.5 t/m3 has been applied for tonnage estimation. The Mineral Resource has been classified as Inferred and grades have been estimated for Ni %, Co % and Mg % based on Ni cut-off grades of 0.5%, 0.8%, 1.0% and 1.3%. The resource tabulation prepared by Minara is included in Table 12.1. A comparison of this to the Snowden 2010 resource, for the total resource within the interpreted nickel mineralisation (codes 200 and 300) is illustrated in Table 12.1.

Table 12.1 Inferred Mineral Resource reported by Minara 2001

Cut-off grade % Ni

Tonnes Mt

Ni %

Co %

Mg %

0.5 329 0.75 0.060 2.91

0.8 66 1.01 0.086 2.87

1.0 14 1.39 0.096 3.32

1.3 5 1.64 0.091 3.22

Figure 12.1 Grade tonnage curves of Snowden 2010 resource (codes 200 and 300) and Minara 2001 polygonal resource

The 2010 resource model reports a higher tonnage and lower grade based on a 0.5% Ni cut-off grade. The 2010 resource model extends for an additional 550 m west and an additional 450 m east of the Minara 2001 model which will account for some of the lower grade tonnage.

For higher cut-off grades the 2010 model reports less tonnes at a lower grade. This is expected as polygonal resource estimates are highly selective and do not account for the volume-variance relationship (overestimate the grade and underestimate the tonnes).



13 Conclusions and recommendations Snowden has completed a resource estimate for the mineralised nickel laterite within the Weld Range Project for Weld Range Metals. The resource was classified as Inferred in accordance with the guidelines of the 2012 JORC Code based upon the geology, mineralisation interpretations and drillhole data, and is reported in Table 13.1.

Table 13.1 Inferred Mineral Resource at 0.5% Ni cut-off grade (JORC, 2012)

Tonnes (Mt) Ni % Cr % Fe % Co %

Total 385.3 0.64 0.78 19.7 0.04 Note: Excludes resource contained within the 2009 chromium Mineral Resource.

Snowden considers that further work is required in terms of topographic surveys, drillhole survey data, QA/QC data, infill drilling and a program of density determinations to improve the classification of this resource estimate.

Data from previous drilling programs should only be maintained in the database to be used for resource estimation if the drillholes collars have been accurately surveyed.

Snowden recommends that a systematic infill drilling program is undertaken that implements best practice standards with regards to data collection, QA/QC and data analysis. Based on the nickel variography Snowden recommends that infill drilling is completed based on an 80 m spacing. Initially this should be an off-set grid, with infill to a regular 80 m grid in areas that require greater definition, and be focused on Zones 1 and 2 (higher grade zones). Preliminary metallurgical test work should be designed to test the different geological units and it is suggested that at least six diamond holes are drilled with three holes in each of Zones 1 and 2. Provision should be made for testing of 12 different samples (two geological horizons from each drillhole).

The exploration program should include provision for establishing an industry standard database, along with a survey program to accurately delineate the topography and to resolve drillhole collar uncertainties. Drilling campaigns should include diamond drillholes for density and metallurgical test work and RC drilling for resource delineation and the implementation of industry standard QA/QC practices. The resource definition drilling program should include provision for about 180 RC drillholes for a total of 5,000 m and six diamond drillholes for a total of 200 m to provide samples for density and metallurgical test work.

Future resource estimation needs to incorporate geological definition of the base of laterite and also domain definition within the laterite horizon based on lithological units and geochemical signatures. In addition to grade estimation of Ni, Cr, Fe and Co, estimation of Al2O3, MgO and SiO2 grades should be incorporated into the resource model.



14 References Anaconda, 2000. Weld Range Project Annual Technical Report, Technical Report 921, for the period 16th October 1999 to 15th October 2000. Report prepared by Anaconda Nickel Limited and dated December 2000.

Anaconda, 2001. Weld Range Project Annual Technical Report, Technical Report 1220, for the period 16th October 2000 to 15th October 2001. Report prepared by Anaconda Nickel Limited and dated November 2001.

Goertz, S., 1997. Weld Range JV Project - Supplementary Report on Global Resource Potential for Lateritic Nickel Mineralisation: (Based on the Results of Recent Reverse Circulation Drilling Programmes and Selected Historical Drillhole Data. Prepared by A.G.R.S.S. Pty Ltd for Sons of Gwalia Ltd and Dragon Mining NL, July 1997.

McConville, F., 2000. Comparison of 4 m composites to 1 m splits, Weld Range, WA.

Minara, 2007. Technical Report 1411 - Environmental report, Weld Range project. Field investigation conducted September 2007. Internal report prepared by Minara Resources Limited.

Snowden, 1989. Geostatistical resource estimation of Range Well prospect. Report prepared for Dragon Resources Limited, January 1989.

Snowden, 1994. Resource estimate for the Range Well Cr/Fe project. Report prepared for Dragon Mining NL, May 1994.

Snowden, 2009a. Weld Range Resource Estimate – Chromium, November 2009. Report prepared for Weld Range Metals Project No. 0479.

Snowden, 2009b. Competent persons’ report on Weld Range mineral assets of Dragon Mining Limited. Report prepared for Dragon Mining Limited, July 2009.

Snowden, 2010a. Ni resource update. Memorandum prepared for Weld Range Metals dated 1 July 2010. Project No. 1040.

Snowden, 2010b. Weld Range Project - Chromium resource update. Report prepared for Weld Range Metals Limited, July 2009. Project No. 1040.

Snowden, 2010c. Weld Range – Resource estimate – nickel. Report prepared for Weld Range Metals, November 2010. Project No. 1040.


Histograms and probability plots Appendix A

Variograms for Ni, Cr, Fe and Co Appendix B

Grade trend plots for Ni, Cr, Fe and Co Appendix C

JORC 2012 Table 1 reporting criteria Appendix D

JORC Code (2012) Table 1 – Sampling Techniques and Data

Item Comments

Sampling techniques • The data used for resource estimation is based on the logging and sampling of RC drilling and AC drilling.

• Majority of drilling was completed between June 1998 and December 2000.

• The drilling and sampling processes are poorly documented and Snowden is unable to comment on the quality of the sample collection procedures.

• Sample lengths vary – less than 3% of sample lengths are less than 1 m, 46% of samples are 1 m, 22% of samples are 2 m, 25% of samples are 4 m and 4% of samples >5 m.

Drilling techniques • The majority of the drilling was completed by RC drilling techniques, along with some AC drilling. RAB drilling has also been completed across the project area but these were excluded from the resource estimate.

• Details of the drilling techniques are poorly documented. RC drilling in 2000 utilised a 5¼” face sampling bit. A conventional cross-over bit was used during some of the earlier drilling (e.g. 1991).

Drill sample recovery • Sample recovery data for the RC and AC drilling is not documented. • Documentation refers to difficult drilling conditions in some holes due

to poor ground conditions and free-flowing sands at surface, with hole collars requiring casing.

Logging • Qualitative geological logging of drillholes is incomplete and only 40% of the intervals that have been assayed for nickel have a weathering code logged. Approximately 15% of drillholes within the database have no logging.

• Where geological logging has been completed, the logging was done with sufficient detail.

Sub-sampling techniques and sample

preparation

• The majority of RC and AC samples collected based on a nominal 1 m or 2 m sample interval. Anaconda report that RC drill cuttings in 2000 drilling program were mostly dry.

• RC drilling in 2000 utilised a three-tiered riffle splitter to sub-sample the drill cuttings to produce a nominal 2 kg to 4 kg sub-sample. Where wet samples were returned, a sub-sample was collected by grab sampling.

• Field duplicate samples collected every 20 samples for holes WRRC0001 to WRRC0249, to monitor the precision of the field sampling process. Results show a good comparison to the original samples.

• Sample preparation comprises oven drying and then pulverising using an LM2 or LM5 pulveriser. Chromium-free pulveriser bowls were used for sample preparation in 1999 and 2000, however Snowden is unable to rule out contamination from the pulveriser bowl for earlier drilling.

• Assaying was by mixed acid digest followed by analysis using either ICP-OES or AAS. Samples from drillholes WRRC0093 to WRRC0249 were analysed by XRF (fused bead). Snowden understands that a mixed acid digest may not recover all chromium and may result in a slight underestimation of the chromium content.

• The sample sizes are considered to be appropriate to correctly represent the mineralisation based on the style of mineralisation (nickel/chromium laterite), the thickness and consistency of intersections and the drilling methodology.


Item Comments

Quality of assay data and laboratory tests

• A number of analytical laboratories in Western Australia have been used over the life of the project, including Actlabs (Kalgoorlie), Ultra Trace Laboratories (Perth) and Genalysis Laboratory Services Pty Ltd (Perth).

• Snowden is only aware of limited QC data being collected as part of the various drilling campaigns.

• Field duplicates collected every 20 samples for drillholes WRRC0001 to WRRC0249. Results show reasonable precision has been achieved for these holes.

• Reference materials inserted into the sample batches (nominal rate of 1:30) during 2000 drilling campaign to monitor analytical accuracy. The certified standard deviation of the reference materials is unknown and as such, only general conclusions can be drawn from the results, which suggest that the analytical accuracy was reasonable.

• Results from the available data for nickel are good; however, Snowden considers that appropriate QA/QC procedures may not have been applied during all of the sampling.

Verification of sampling and assaying

• In 2009, as part of the Cr laterite resource estimate, Snowden selected seven WRRC series drillholes (0137, 0138, 0143, 0175, 0176, 0200 and 0248) to be re-assayed for Cr, Ni and Fe from the sample pulps by fused bead XRF at the ALS Chemex laboratory in Brisbane. The re-assaying showed a good comparison between the original assays and the re-assays, with results typically within 10%.

• No twin holes have been completed. • Data entry and database procedures are not documented and

Snowden is unable to comment on their appropriateness. • The following adjustments were made to the assay data:

− Below detection limit values and assay values of “0” were reset to half the detection limit.

− Values in ppm were converted to percentages by dividing through by 10,000.

− An unusually high value of 11.6% Ni in drillhole RRC068 was replaced with a value of 1.16% Ni as examination of surrounding holes does not support this high value and it is assumed to be a data entry error.

− Two unusually high Co values of 7.4% Co and 2.6% Co were noted at the top of RRC057. Adjacent drillholes did not support these high grades and Snowden elected to exclude these from the current database.


Item Comments

Location of data points • The grid is based on the AMG_Z50 grid datum. Collar locations for the 2000 drilling campaign surveyed by a contract surveyor using a DGPS (accuracy reported to be ±10 cm both horizontally and vertically).

• Due to the age of the drillholes and site rehabilitation, drillhole collars are difficult to locate. Field checks by Snowden in 2009, as part of the chromium laterite resource estimate, found the following:

− RWAC0015 (AC hole) was located, providing confidence in the location of RWAC series holes.

− Three drillholes (WEC021, 22 and 23) were found to be out by approximately 138.9 mX and 152.6 mY. The original coordinates were likely located with respect to the MGA grid. Based on this, all WEC series drillholes were adjusted to the AMG_Z50 grid for the resource estimate.

− Three other holes were located which are likely CRC01, 02 and 03, however the coordinates do not conform to the MGA/AMG_Z50 grid transform and confidence in the location of these holes is low. Documents suggest these holes were drilled to the northeast however no down-hole survey data is available. CRC series holes were excluded from the database.

• 370 drillholes in the database have default or estimated collar elevations. A topographic surface was developed by Snowden from the 403 drillholes with measured collar elevations. This surface was used to estimate collar elevations for the remaining 370 drillholes.

• Approximately 50% of the drillholes in the database do not have down-hole survey information. The majority (83%) of the drillholes with down-hole surveys are vertical. It is assumed that the drillholes without down-hole surveys are vertical, which is considered a reasonable assumption as these holes target the laterite mineralisation.

Data spacing and distribution

• The spacing between drill sections varies throughout the project. Holes have been drilled based on sections of 200 m to 800 m spacing east-west and 40 m to 200 m north-south. Within the Chrome Hill area, infill drilling is spaced at 40 mE by 20 mN.

• The section spacing is sufficient to establish the degree of geological and grade continuity necessary to support the resource classifications that were applied.

• The sample data was composited downhole using a 1 m interval.

Orientation of data in relation to geological

structure

• The location and orientation of the Weld Range drilling is appropriate given the strike and morphology of the laterite mineralisation, which is sub-horizontal.

Sample security • Protocols relating to sample security are not documented. Snowden has no reason to believe that sample security poses a material risk to the integrity of the assay data used in the Mineral Resource estimate.

Audits and reviews • Snowden is not aware of any external audits on the sampling techniques and assay data.

JORC Code (2012) Table 1 – Estimation and Reporting of Mineral Resources

Item Comments

Database integrity • No dedicated drillhole database exists and validation procedures are not documented. The database used for the resource estimate was compiled by Snowden based on comma delimited text files supplied by Weld Range Metals.

• Snowden undertook a basic check of the data for potential errors as a preliminary step to compiling the resource estimate. Numerous errors were identified, including:

− Incorrect collar coordinates (corrected or excluded) − Drillholes RHCH09, RWRC01 and RWRC03 have no collar

information (excluded) − Drillholes with no down-hole survey information (assumed to

be vertical) − Drillholes with no geological logging − Overlapping and duplicate samples (removed) − Data entry errors (corrected where possible).

Site visits • Snowden visited the Weld Range project area in June 2009 prior to undertaking a lateritic chromium resource estimate.

Geological interpretation • The geological model for the Weld Range project area is based on surface mapping, geological and geochemical logging of drillholes, and geophysical data.

• A nominal grade of 0.4% Ni, based on assessment of the raw grade distribution, was used to interpret the horizon of nickel mineralisation within the laterite. Surfaces were developed for the top and base of the nickel mineralisation based on the top of the first and the bottom of the last interval with a nominal assay threshold of 0.4% Ni. The surfaces were adjusted where required based on a minimum thickness of 2 m.

• The nickel mineralisation was interpreted as a single mineralised domain. The current data does not allow (due to incomplete chemical results and geological logging) for sub-dividing the mineralisation into different weathering horizons (e.g. ferricrete and saprolite horizons).

• Data from geological logging of the drillholes is incomplete in the existing databases and only 44% of the intervals that have been assayed for nickel have a weathering code. The current resource model is restricted to within 40 m of the surface (excepted where it is present within the chromium Mineral Resource). Verification of the base of weathering is required to confirm that the nickel resource estimate is restricted to within the laterite horizon.

• Over 50% of the intervals with nickel assay data do not have lithology or weathering codes in the database. The average depth to the logged based of laterite is 19.5 m and the 99th percentile of the base of the logged laterite intersections is at 36 m (the deepest logged laterite interval is at 64 m). Snowden has restricted the interpreted nickel mineralisation and resource estimate to within 40 m of the topographical surface.

• Alternative interpretations of the mineralisation are unlikely to significantly change the overall volume of the mineralised envelopes in terms of the reported classified resources.

Dimensions • The mineralisation is sub-parallel to the topography, trends roughly northeast-southwest and has a total strike length of about 14 km. The mineralisation is on average about 15 m to 20 m thick and occurs to a depth of up to 30 m below surface.


Item Comments

Estimation and modelling techniques

• Estimation of Ni, Cr, Fe and Co using ordinary block kriging for all domains with hard domain boundaries. SiO2, Al2O3 and MgO were not estimated due to a lack of assay data.

• No top-cuts were applied to the Ni, Cr and Fe grades. Co grades were top-cut to 0.3% Co, which affects less than 1% of the data.

• Gemcom software was used for the grade estimation. • Block model constructed using a parent cell size of 100 mE by 25 mN

by 2 mRL (no sub-celling). • The search ellipse orientation and radius was based on the results of

the grade continuity analysis, with the same search parameters used for all elements to maintain the metal balance and correlations between elements. An initial search of 230 m along strike by 160 m across-strike by 4 m thick was used, with a minimum of four samples and a maximum of 20 samples.

• Hard and soft boundary conditions were applied depending on the grade and domain being estimated. Chromium grades have not been re-estimated for blocks within the 2009 chromium laterite resource boundaries.

• Block grade estimates were validated against the input composite data both globally and locally by: − Visual comparison of block grade estimates and drillhole

composite data − Generation of section plots of the estimates, declustered and

naïve composite grades, along with the number of composite samples available

− Global comparison of the average composite (naïve) and estimated grades.

• The project is in an exploration phase and no production has taken place.

• No by-products have been considered during estimation. • The deposit was previously estimated by Minara in 2001, who

developed a polygonal estimate of the nickel laterite resources.

Moisture • All tonnages have been estimated as dry tonnages.

Cut-off parameters • The nickel mineralisation was reported above a 0.5% Ni cut-off grade. • The cut-off grade is based on the assumption that the Weld Range

nickel laterite deposit will be mined by conventional open pit mining methods and that the ore can be beneficiated, through screening, during processing. Preliminary metallurgical testwork shows that the mineralisation can be upgraded through screening.

• Snowden believes that the cut-off grade is reasonable for the nickel mineralisation, given the mining assumptions and proposed processing route.

Mining factors and assumptions

• It is assumed the deposit will be mined using traditional open-cut methods.


Item Comments

Metallurgical factors and assumptions

• Metallurgical testwork conducted by Anaconda Nickel Limited in 2000, on composite RC samples of Weld Range mineralisation, indicates that the mineralisation can be beneficiated to upgrade the nickel content of the ore.

− Results of the screening at 0.5 mm on eight composite RC samples, showed an average upgrade for the Ni grade of 44%, recovering 62% of the metal and with a mass recovery of 42% passing -0.5 mm.

− Results of bottle roll beneficiation testwork on 439 samples showed an upgrade in the Ni grade of 82% for the “high upgrade” domain (defined as having <6% Mg and >35% Si) with a mass recovery of 36% passing -0.25 mm. For the “low upgrade” domain (defined as having Mg <6% and Si <35%), the nickel content was only upgraded by 16% with a mass recovery of 58% passing -0.25 mm.

• The metallurgical testwork is considered preliminary and further testing is required.

• No metallurgical modifying factors have been applied to the resource estimate.

Environmental factors and assumptions

• It is assumed that no environmental factors exist that could prohibit any potential mining development at the Weld Range Project.

Density • No bulk density measurements have been collected. • A default bulk density of 1.5 t/m3 was applied to all domains based on

projects with similar styles of mineralisation. The bulk density value is considered by Snowden to be conservative and may be locally higher.

Classification • The nickel laterite resources have been classified as Inferred in their entirety, based on continuity of both the geology and grades, along with the drillhole spacing, data quality and lack of bulk density measurements.

Audits and reviews • Snowden is not aware of any audits or reviews of the Mineral Resource estimate.

Discussion of relative accuracy/confidence

• An assessment of the relative accuracy of the block grade estimates was not deemed appropriate due to the Inferred classification and general data quality issues.