·      Drill hole data is stored in a Plexer database. Collar, survey, sample dispatch data and analytical results are uploaded from .csv files as they become available. The upload process includes validation checks for consistency and anomalous values.

·      Drill logs have been entered directly into Fusion from paper-based records. This process was improved by the introduction of digital logging in October 2018 whereby this data is generated as .csv files for upload and validation.

·      The Competent Person has reviewed the Taylor deposit Mineral Resource Estimate, visits the site regularly and is a full-time employee of South32.

·      The objectives of the site visit are to understand all inputs and processes contributing to the FY23 Mineral Resource estimate, including core drilling, changes in core logging procedures, digital core logging, database audits and resampling programs to improve confidence in geological interpretation, density estimation and geometallurgical inputs.

·      The Competent Person discussed sample preparation and laboratory procedures with ALS representatives to ensure that these procedures are applied.

·      The findings of site visits indicate the data and procedures are of sufficient quality for Mineral Resource estimation and reporting. Review and required improvement are continuously discussed and any required changes are implemented.

·      'Mineralisation domains' are created within bounding lithologies using indicator modelling methods of the cumulative in-situ value of metal content. The metal content descriptors, termed 'Metval' and 'Oxval' are calculated by summing the multiplication of economic analyte grades for Mn, Zn, Pb, Cu and Ag, price and recovery. Metval and Oxval cut-off ranges for mineralisation domains ranged from US$6 to US$17 for the different litho-structural domains. Material above the Metval and Oxval cut-off is modelled utilising the indicator numerical model function in Leapfrog Geo™ to create volumes.

·      Indicator models are guided using geologic trends based on modelled lithologic contacts and structures within a post mineralisation fault block model. Constraints on these domains include known bounding structures, stratigraphy and manually digitised limits on the extents of mineralisation. In addition to drill hole data, historic underground mine plans and mapping and surface geologic mapping is used to help extend geologic features to the topography. The purpose of these domains is to provide mineralised volumes within the larger lithologic boundaries and to ensure relevant geological controls and constraints are considered. Indicator cut-offs are selected to create continuous volumes consistent with the overall modelling approach for CRD-style mineralisation.

·      Mineralised domains are evaluated against multiple indicator scenarios for parameters such as inherent dilution, exclusion and volumetric changes. These evaluations aim to balance the parameters with the understood continuity of mineralisation from site geological staff interpretation.

·      Alternate geological interpretations have not been used; however, the model is continually evolving as new data is collected.

·      The mineralising system is yet to be fully drill tested in multiple directions. The Taylor sulphide mineralisation is constrained up-dip where it transitions to oxide mineralisation, representing a single contiguous mineralised system. Taylor is open in multiple directions.

·      The north-bounding edge of the thrusted carbonate rock is marked by a thrust fault where it ramps up over the Jurassic/Triassic 'Older Volcanics' and 'Hardshell Volcanics'. This interpreted pre-mineral structure that created the sequence of carbonates also appears to be a key conduit for mineralisation.

·      The Taylor deposit has an approximate strike length of 2,500m and width of 1,900m. The stacked profile of the thrusted host stratigraphy extends 1,200m from near-surface and is open in several directions. (Figure 5 and Figure 6).

·      Geologic modelling was performed using Leapfrog Geo™ 2022.1.0 and grade estimations using Maptek Vulcan.

·      Elemental estimation includes Zn, Pb, Ag and Cu. As and Mg are estimated as potential deleterious analytes and Fe, Ca, S, and Mg are estimated as tonnage inputs.

·      The specific gravity is also estimated using a restricted search guided by geologic trends.

·      Estimation and modelling techniques reflect the interpreted structural and lithological controls on mineralisation apparent in the core and in data. These align with the current understanding of the formation of CRD style mineralisation. Key assumptions include:

o  The relative importance of structure and lithology in either facilitating or constraining the deposition of mineralisation.

o  Geological domaining according to these controls; and

o  All boundaries are considered "hard."

·      Search orientations are aligned with mineralised structures and lithological contacts using locally varying anisotropy to assign directions on a block-by-block basis. Search distances and variography parameters are interpolated into 'parent' blocks of 9m by 9m by 4.5m from 3D geological wireframes taken from the geological model.

·      Assay data is composited to a nominal interval of 1.5m within mineralisation domains for the purpose of exploratory data analysis to derive estimation parameters for ordinary kriging.

·      To manage the risk of local grade overestimation, high-grade outliers in the drill holes are capped prior to compositing. Cap values are determined using log probability plots for each domain. Selected thresholds are typically above the 99.5 percentile where the distribution or sample support deteriorates and to reduce the coefficient of variation. No bottom caps are applied.

·      The outputs of geostatistical analysis, including variography and Quantitative Kriging Neighbourhood Analysis (QKNA), are used to optimise grade estimation parameters such as search distances, sample selection criteria, and block dimension. A parent block size of 9m by 9m by 4.5m is selected relative to a data spacing of between 25m and 150m. However typically a data spacing of approximately 50m within the core of mineralisation is used to support mining study selectivity within the minimum Selective Mining Unit (SMU) dimension.

·      Sub-cells to a minimum of 1.5m are built along the contacts of the estimation domains to reduce the volume variance between wireframe models and the orthogonal block model.

·      The dimensions of the anisotropic search ellipses for each estimation pass are generally matched to the ranges of the first and second structures of the variograms per domain using ranges of the overall structure of grade continuity for the zinc variogram models. The search ellipse ranges vary between estimation domains but remain the same for all elements within individual domains. While related elements (e.g. Pb-Ag, Pb-Zn, Ag-Zn) are not co-kriged, their correlated nature is validated and confirmed that the relationship is preserved in block estimates.

·      Minimum and maximum sample criteria, an octant search strategy and a restriction of the number of samples used from each drill hole are applied to assist with reduction of local grade bias. A second search pass, set at the entire range of the zinc variogram, is used to estimate lower confidence areas of the model.

·      Kriging tests with visual and statistical validation of results indicate whether it is appropriate to apply an initial top cap, which is then adjusted up or down to counter any global bias introduced. The degree of grade smoothing between data and block values is analysed through a comparison of mean differences, histograms, q-q plots and swath plots.

·      Classification criteria constrain the reporting of estimates to within demonstrated grade and geological continuity ranges. As all estimation passes rely on at least two holes to inform the estimate, there is no extrapolation from single holes in any classified material.

·      The appropriateness of estimation techniques contributes to the high confidence estimation outcome achieved in areas of data spacing within the full ranges of grade continuity.

·      The grade estimations are compared against previous estimates and reviewed locally for differences in data/interpretation and globally using graded tonnage plots and waterfall analysis.

·      The Mineral Resource is reported for Zn, Pb and Ag without any assumptions relating to recovery of by-products.

·      Resource estimation is well established and reasonable for the deposit.

·      Moisture content of the core appears to be minimal, based on logging observations and pre-and post-dried sample weights tested by ALS on assay samples from July 2019 to February 2022 on over 50,000m. A dry bulk density is assumed for estimation purposes.

·      NSR reporting cut-off values are based on relevant project study operational costs and pricing scenarios. Application of a nominal lower limit of breakeven economics from these costs is considered as the reasonable prospects for eventual economic extraction under current economic modelling.

·      The calculations for each block are used to determine resource block cut-off according to variability of physical costs such as logistics, treatment costs, refining costs and economic factors such as metal pricing.

·      The NSR cut-off values for reporting the FY23 Taylor deposit Mineral Resource is US$80/dmt for material considered extractable by underground open-stope methods.

·      The input parameters for the NSR calculation include South32 long-term forecasts for Zn, Pb and Ag pricing, haulage, treatment, shipping, handling and refining charges.

·      Underground mining factors and assumptions are based on pre-feasibility level project studies and are calibrated against South32's Cannington zinc, lead and silver mine production. Longhole stopes on a sub- or full-level basis with subsequent paste backfill is the assumed mining method.

·      Reasonable prospects for eventual economic extraction are determined through assessment of the model at Prefeasibility Study (PFS) levels using processes ranging from stope optimisation and mine scheduling through to detailed financial modelling.

·      The NSR block value incorporates metallurgical recovery based on test work for composite and individual mineralisation domains.

·      Total metallurgical recovery assumptions vary for sulphide geological domains from 87% to 94% for zinc; 94% to 95% for lead; and 87% to 92% for silver. These assumptions have been verified through extensive metallurgical test work.

·      Pre-Feasibility level environmental assumptions, including possible waste and process residue disposal options, are factored into physical and financial models that are used to evaluate reasonable prospects for eventual economic extraction.

·      Dry bulk density is estimated for mineralisation domains where data density is sufficient to estimate Zn on the first pass. Zn variograms and first pass search criteria are applied to density measurements. The current database records 25,272 Specific Gravity (SG) measurements.

·      SG was originally calculated beyond the range of the first pass using Zn, Pb, Ag, Fe, Ca and Mg using a regression formula. Measurements from previous campaigns, low numbers of which were taken from sulphide and oxide mineralisation in carbonates, are excluded from the analysis because assaying did not include the full complement of elements used for the regression formulae.

·      A final pass of assigned average density values is applied to fill blocks on the outskirts without grade.

·      Historically SG measurements were taken from an approximate 20cm representative section of competent core within a 1.5m sample interval. Since May 2021, to improve the SG regression analysis, SG measurements are broken out with an associated assay interval of approximately >60cm. The measurement technique determines a specific gravity using the core weight in air and weight immersed in water. Routine calibration of scales and duplicate measurements are undertaken for quality control.

·      The core is not oven dried or coated to prevent water ingress prior to immersion unless porosity is noted in the sample. If porosity is noted, the core was coated in plastic film.

·      Lithology outside of mineralisation domains have an average bulk density assigned by rock type.

·      The FY23 Mineral Resource has been independently audited by Golder Associates Pty Ltd. The audit concluded, in general, that modelling has been conducted in a manner consistent with industry standards and supporting documentation has been adequate.

·      Geological modelling is such that there is a moderate-to-high degree of predictability in the position and quality of mineralisation where infill drilling is being conducted. Geostatistical analysis indicates a low nugget effect and ranges of grade continuity are beyond drill spacing in Measured and Indicated areas of the deposit.

·      Measured Resources of the FY23 Taylor deposit Mineral Resource global estimate is expected to be within 15% accuracy for tonnes and grade when reconciled over any quarterly production volume using mining assumptions matched to the determination of reasonable prospects for eventual economic extraction. Indicated Mineral Resource uncertainty should be limited to ±30% quarterly and ±15% annually. It is expected that Inferred Mineral Resources will be converted to higher confidence classifications prior to extraction.

·      The Competent Person is satisfied that the accuracy and confidence of Mineral Resource estimation is well established and reasonable for the deposit.