Page 1 of 28 Notice to ASX Changes to Rio Tinto Aluminium Pacific Operations Ore Reserve and Mineral Resource Estimates 17 February 2021 Rio Tinto today announces changes in estimates of Ore Reserves and Mineral Resources at Rio Tinto Aluminium (RTA) Pacific Operations’ Weipa and Gove assets. These changes will be included in Rio Tinto’s 2020 Annual Report, to be released to the market by 22 February 2021, and are reportable changes compared to those published in the 2019 Annual Report. The Ore Reserve and Mineral Resource estimates are reported in accordance with the Australasian Code for Reporting of Exploration Results, Mineral Resources and Ore Reserves, 2012 (JORC Code) and the ASX Listing Rules. Ore Reserves and Minerals Resources are quoted on a 100 per cent basis (100% Rio Tinto share). The RTA Pacific Operations Ore Reserve estimate has decreased by 305 Mt, from 1,530 Mt to 1,224 Mt. The reduction in Ore Reserve estimate is associated with a routine review of price assumptions over the life of the mine. There has been no material change to the economic cut-off methodology or process. The application of the updated pricing assumptions has removed lower quality material from the Ore Reserve estimate. There has been no material change to the other reserve modifying factors, including: governmental, tenure, environmental, cultural heritage, social, community or operational. RTA Pacific Operations Mineral Resources exclusive of Ore Reserves has increased by 128 Mt, from 1,949 Mt to 2,077 Mt (This includes 1,330 Mt of Inferred Resource at North of Weipa, which remains unchanged from 2019). The increase was mainly due to changes in the forecast pricing assumptions used in the reserving process. Those blocks removed from the 2020 Ore Reserve that are physically located above or below the remaining economic blocks have also been removed from the Mineral Resource. Supporting information surrounding the 2020 update to the Ore Reserve and Mineral Resource estimates has been provided below and in the appendices. EXHIBIT 99.6
Page 2 of 28 RTA Pacific Operations Mineral Resources RTA Pacific Operations Ore Reserves Mineral resources 2020 2019 2020 2019 Tonnage Grade Tonnage Grade Tonnage Grade millions millions millions millions millions Bauxite of tonnes of tonnes of tonnes of tonnes of tonnes Gove (Australia)(b)(c) O/P 23 48.6 9 48.6 2 49.6 34 28 48.7 48.2 100.0 Weipa (Australia)(b) - East Weipa and Andoom(d) O/P 35 51.1 35 11 51.1 52.1 100.0 - Nort of Weipa O/P 1,330 52 1,330 1,330 52.0 52.0 100.0 - Amrun(e) O/P 57 49.0 348 50.3 273 50.5 678 580 50.3 50.1 100.0 TOTAL 115 49.6 357 50.3 1,605 51.7 2,077 1,949 51.4 51.4 100.0 (a) Likely mining method: O/P = open pit; U/G = underground; D/O = dredging operation. Notes % Al2O3 % Al2O3 % Al2O3 % Al2O3 at end 2020 (b) Gove and Weipa Resources are stated as dry tonnes and total alumina grade. (c) Gove Resource tonnes increased following conversion of Reserves to Resources based on updated economic assumptions. (d) East Weipa and Andoom Resource tonnes increased following conversion of Reserves to Resources based on updated economic assumptions. Likely mining method(a) Rio Tinto Interest %Tonnage Grade Measured Indicated resources Inferred resources Total resources 2020 compared with 2019 at end 2020 % Al2O3 (e) Amrun Resource tonnes increased following conversion of Reserves to Resources following updated economic assumptions. at end 2020 Ore reserves 2020 2019 2020 2019 Tonnage Grade Tonnage Grade Tonnage Tonnage Grade Grade Rio Tinto share Recoverable mineral millions millions millions millions millions Bauxite(b) of tonnes of tonnes of tonnes of tonnes of tonnes Gove (Australia)(c) O/P 77 50.4 3.3 49.7 80 131 50.4 49.3 100.0 80 Weipa (Australia)(d) - Amrun(e) O/P 211 54.1 833 53.9 1,044 1,253 54.0 53.1 100.0 1,044 - East Weipa and Andoom(f) O/P 100 51.4 100 146 51.4 50.8 100.0 100 TOTAL 388 52.7 836.3 53.9 1,224 1,530 53.6 52.6 100.0 1,224 Total ore reserves 2020 compared with 2019 Type of mine(a) at end 2020 at end 2020 Interest % (e) Amrun Reserve tonnes decreased following updated economic assumptions and mining depletion. (d) Weipa Reserves are stated as dry tonnes and total alumina grade Notes (a) Type of mine: O/P = open pit, U/G = underground, D/O = dredging operation. (b) Reserves of bauxite, diamonds and iron ore are shown as recoverable Reserves of marketable product after accounting for all mining and processing losses. Mill recoveries are therefore not shown. (c) Gove Reserves are stated as dry tonnes and total alumina grade. Gove Reserve tonnes decreased following updated economic assumptions and mining depletion. (f) East Weipa and Andoom Reserve tonnes decreased following updated economic assumptions and mining depletion. % Al2O3 % Al2O3 % Al2O3% Al2O3 Average mill recovery % Proved ore reserves Probable ore reserves
Page 3 of 28 Summary of information to support the Mineral Resource estimate The RTA Pacific Operations Mineral Resource estimate is supported by a JORC Table 1 (Sections 1, 2 and 3) document provided as Appendices 1 and 2 to this release and also located at www.riotinto.com/investors/reserves- and-resources. An increase in RTA Pacific Operations Mineral Resources coincides with the write back of bauxite ore from Ore Reserves to Mineral Resources due to a routine review of price assumptions over the life of the mine. The methodology of determining Mineral Resources has not changed. The bauxite assets have been in operation for more than fifty years and are well understood. Resource work is currently more focussed on asset evaluation rather than exploration, systematically bringing the bauxite classification to higher levels of confidence. The following summary of information for Mineral Resource estimates is provided in accordance with Chapter 5.8 of ASX Listing Rules. Geology and geological interpretation RTA Pacific Operations have two bauxite deposits, one at Gove (NT) and one at Weipa (QLD). The host rocks of the two mine sites have been converted to bauxite via a continuum of weathering. High annual rainfall and a geologically stable environment has provided the perfect ingredients for these world-class bauxite deposits to form over many millions of years. A deep saprolitic zone overlain by a classic mottled zone below the bauxite ore attests to this. The process of bauxitisation involves the conversion of kaolinite to the bauxite minerals gibbsite and boehmite. The principal influence on the process is the composition, supply and movement of groundwater. The pH of the groundwater is lowered during the process of bauxitisation and we note that the process is still ongoing as we see a low pH regularly throughout the ground water monitoring bores across the RTA mining leases. To a lesser extent there are organic influences such as vegetation, and possibly burrowing organisms and temperature. Pisolitic textures are dominant, with variable cementation. However, variably cemented coarser nodule horizons are also common. Some deeper bauxite, as well as underlying laterite, contain abundant interconnected solution cavities, which may have been created geochemically and/or biologically (associated with root channels). Modern day root channel structures and infill, in the upper part of the bauxite, are common. Gibbsite is the major ore mineral, with boehmite being of lesser significance. Bauxite occurs on laterally extensive plateaus. The plateaus have been variably dissected by erosion into distinguishable main plateau areas, each with their own characteristics, (outlined further in Table 1). The bauxite orebodies are interpreted as flat-lying horizons with topography dictating the geometry. The orebodies are generally overlain by a thin (<1m) overburden cover and occasional red soil. Beneath the bauxite ore is often a transition zone defined by angular and lumpy textures and a geochemical signature of higher silica and lower alumina. The transition zone is often underlain by the ironstone; with a distinct change in physical properties, particularly the colour. Drilling techniques The current drilling method at Weipa utilizes aircore drilling. The typical aircore rig is a Land Cruiser mounted rig with a small enough wheelbase to traverse drill lines cleared with one D-6 dozer blade width. Aircore drilling forces compressed air down a space inside the drill rods to the bit face, where the air is then used to return the sample up the inner tube of the drill rod and out via a cyclone. A three bladed HQ aircore bit is attached to 4-inch rods. The drilling system has been designed to reduce grinding of the sample. Historic drilling utilised auger drilling (until 1991). Drilling at Gove is conducted using a tractor mounted 44 mm diameter vacuum drill rig. Logging, sampling, sub-sampling method and sample analysis method Logging is currently conducted on Panasonic Toughpads and data is captured in an offline acQuire logging package at the drill rig. This system allows for data validation to be applied during logging as well as a streamlined method of exporting the data for importing into the main RTA Geology database. Logging is qualitative in nature,
Page 4 of 28 i.e. based on lithology. Currently there are ~20 lithologies common to the deposits that get modelled into 4 horizons for the estimation of bauxite resources. All sample intervals (0.25 m) are logged. Logged lithologies are vetted against historical drill holes and assay parameters. Samples for geologic logging and analysis are collected on 0.25 m intervals (~2-3 kg) down hole. Whole samples are collected beneath a cyclone return system, i.e. no sample splitting is conducted, or sub samples taken. Multiscreen sampling is undertaken initially to determine optimum screen size for beneficiation at each deposit. Once determined, samples are then beneficiated at the appropriate screen size (1.7 mm for East Weipa, 0.3 mm for Andoom and 0.6 mm for the Amrun deposits). Bauxite at Gove is direct shipping ore (DSO). Samples are processed and XRF analysed for the major oxides: Al2O3, SiO2, Fe2O3, TiO2 and LOI, as well as minor elements and recovery. Estimation methodology Basic geostatistical analysis is used to help with domaining decisions. Most deposits were modelled as a single laterally extensive domain, apart from Moingum (Hey Point), where two lateral extensive domains have been modelled due to difference in source rocks affecting thickness and grade. Four horizon codes, based on the lithology and assays, are assigned for the modelling and estimation of bauxite resources at Weipa and three at Gove vertically. Interpretation and estimation are currently performed using Datamine’s Studio RM software, with variography undertaken in FSS International Consultants (Australia) GS3M software. The bauxite horizon is flattened to the mid-point of the drill hole intersections at East Weipa and Andoom, while drill hole collars are flattened to constant elevation at Amrun, Norman Creek and Moingum (Hey Point). The wireframes are filled with blocks on an in/out basis; there is no sub-blocking or block proportions used. For the bauxite horizon, major oxides; LOI and Recovery are estimated into parent cells using ordinary kriging. Overburden and red soil are assigned 0% recovery for the estimation of resources. Cemented bauxite grade is estimated as part of the bauxite horizon and assigned a 100% recovery; the proportion of cemented bauxite is estimated as an indicator variable. Major oxide chemistry is also estimated for the overburden, red soil and ironstone horizons, where data is available. Ordinary kriging is used for interpolation, using the variogram models for the bauxite. Block sizes are determined by ½ the drill hole spacing for each deposit. A multiple pass search strategy is used to estimate grades utilising different sized search ellipses that include a specified number of samples and drill holes. The lowest category (Pass 4) is not used in the East Weipa and Andoom areas. Maximum extrapolation distance is slightly less than the maximum search radii due to the requirement to use at least two holes to estimate each block. Criteria used for classification Classification within the bauxite horizon is based on the search pass used to estimate grades, using increasing search radii and decreasing numbers of samples for each subsequent pass. Passes 1 and 2 are classified as Measured Mineral Resources (120 to 180 m spacing for Weipa and 50 m to 100 m spacing for Gove), Pass 3 as Indicated (360 m spacing for Weipa and 200 m to 400 m spacing for Gove) and Pass 4 as Inferred. Reasonable prospects for eventual economic extraction RTA Pacific Operations employs a standard approach to identify Mineral Resource volumes with reasonable prospects for eventual economic extraction. Once the Ore Reserves are defined based on applied economic factors in the reserving process, the remaining blocks are evaluated based on grade cut-offs, thickness cut-offs and location (environmental, cultural heritage and infrastructure buffers) for each of the different deposits, and Mineral Resources defined.
Page 5 of 28 Summary of information to support the Ore Reserve estimate The reduction in the RTA Pacific Operations Ore Reserve estimate is associated with a routine review of price assumptions over the life of the mine. There has been no material change to the economic cut-off methodology or process, simply the application of updated pricing assumptions has removed lower quality material from the Ore Reserve estimate. The 2020 Ore Reserve estimate is supported by a JORC Table 1 (Section 4) document provided as Appendices 1 and 2 to this release and located at www.riotinto.com/investors/reserves-and-resources. The following summary of information for the Ore Reserve estimate is provided in accordance with Chapter 5.9 of ASX Listing Rules. Operational / study status Both Weipa and Gove assets have been operating continuously for several decades, and the Ore Reserve estimates and life of mine plans are updated annually. This includes the reconciliation of operating parameters and review of input assumptions into the planning processes. The Ore Reserve estimate for Amrun is based on the completed and approved feasibility study and has now been in continuous operation for two years. The Ore Reserve is mined through shallow, open cut techniques developed over several decades of operations. Once the area is tree cleared and the topsoil / overburden removed, the bauxite is hauled to the processing facility for washing and / or sizing. Product bauxite is stockpiled for shipping to both internal and external customers. Several mining areas are active at any one time to enable blending and to mitigate against operational risk. Economic assumptions Rio Tinto applies a common process to the generation of commodity price assumptions across the group. This involves generation of long-term price forecasts based on current sales contracts, industry capacity analysis, global commodity consumption and economic growth trends (this includes the bonus / penalty adjustments for quality). Exchange rates are also based on internal Rio Tinto modelling of expected future country exchange rates. The reduction in Ore reserve estimate is associated with a routine review of price assumptions over the life of the mine. Capital and operating cost estimates are sourced from internal Rio Tinto financial modelling and / or project capital estimates. Third party payments are reflective of the current agreements in place. Due to the commercial sensitivity of these assumptions, an explanation of the methodology used to determine these assumptions has been provided, rather than the actual figures. Cut-off grades The Ore Reserve cut-off is based on an economic parameter, summarised as the margin realised upon sale of the bauxite. The economic cut-off approach considers revenue (bonus/penalty), fixed / operating / capital costs, royalties and other third-party payments. Bauxite that satisfies this economic cut-off, is considered for inclusion in the Ore Reserve. There has been no material change to the economic cut-off methodology or process. Processing Weipa bauxite is beneficiated through established techniques to improve product quality and handleability. This is achieved through the removal of the finer fraction, leaving the coarser material as product. Expected bauxite recovery and quality from the beneficiation process is assessed through laboratory scale test work of samples generated from the resource drilling process. There is no beneficiation plant at Gove Operations, the bauxite is sized through a primary and secondary crushing system before being stockpiled for shipping. With no crushing loss, 100% of the bauxite mined is planned to be shipped.
Page 6 of 28 Mining and recovery factors Dilution and mining recovery parameters are applied during the Ore Reserve estimation process, based on reconciliation of past performance and are reviewed annually. As the Ore Reserve is shallow, geotechnical risks are low. Stockpile heights and wet road conditions are managed in accordance with standard operating procedures There has been no material change to other Ore Reserve modifying factors, such as: governmental, tenure, environmental, cultural heritage, social or community. Appropriate agreements and approvals remain in place to enable continued operation of these assets. Reserve classification Given the high level of confidence in the reserve modifying factors, all Measured Resources were converted to Proved Ore Reserves and all Indicated Resources were converted to Probable Ore Reserves. Inferred Resources were not considered in the estimation of Ore Reserves. No Probable Ore Reserves have been derived from Measured Resources.
Page 7 of 28 Competent Persons Statement The information in this report that relates to Mineral Resources is based on information compiled under the supervision of Mr Angus C. McIntyre, who is a Member of the Australasian Institute of Mining and Metallurgy (MAusIMM). Mr McIntyre has sufficient experience that 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 JORC Code. Mr McIntyre is a full-time employee of Rio Tinto and consents to the inclusion in this report of RTA Pacific Operations Bauxite Mineral Resources based on the information that he has prepared in the form and context in which it appears. The information in this report that relates to Ore Reserves is based on information compiled under the supervision of Mr William Saba who is a Member of the Australasian Institute of Mining and Metallurgy (MAusIMM). Mr Saba 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 JORC Code. Mr Saba is a full-time employee of Rio Tinto and consents to the inclusion in this report of RTA Pacific Operations Bauxite Ore Reserve based on the information that he has prepared in the form and context in which it appears.
Page 8 of 28 Contacts media.enquiries@riotinto.com riotinto.com Follow @RioTinto on Twitter Media Relations, United Kingdom Illtud Harri M +44 7920 503 600 David Outhwaite T +44 20 7781 1623 M +44 7787 597 493 Media Relations, Americas Matthew Klar T +1 514 608 4429 Media Relations, Asia Grant Donald T +65 6679 9290 M +65 9722 6028 Media Relations, Australia Jonathan Rose T +61 3 9283 3088 M +61 447 028 913 Matt Chambers T +61 3 9283 3087 M +61 433 525 739 Jesse Riseborough T +61 8 6211 6013 M +61 436 653 412 Investor Relations, United Kingdom Menno Sanderse T: +44 20 7781 1517 M: +44 7825 195 178 David Ovington T +44 20 7781 2051 M +44 7920 010 978 Investor Relations, Australia Natalie Worley T +61 3 9283 3063 M +61 409 210 462 Amar Jambaa T +61 3 9283 3627 M +61 472 865 948 Group Company Secretary Steve Allen Rio Tinto plc 6 St James’s Square London SW1Y 4AD United Kingdom T +44 20 7781 2000 Registered in England No. 719885 Joint Company Secretary Tim Paine Rio Tinto Limited Level 7, 360 Collins Street Melbourne 3000 Australia T +61 3 9283 3333 Registered in Australia ABN 96 004 458 404 This announcement is authorised for release to the market by Rio Tinto’s Group Company Secretary.
1 Weipa Table 1 Appendix 1 The following table provides a summary of important assessment and reporting criteria used at Weipa for the reporting of mineral resources and ore reserves in accordance with the Table 1 checklist in The Australasian Code for the Reporting of Exploration Results, Mineral Resources and Ore Reserves (The JORC Code, 2012 Edition). Criteria in each section apply to all preceding and succeeding sections. SECTION 1 SAMPLING TECHNIQUES AND DATA Criteria Commentary Sampling techniques • Samples for geologic logging and analysis are collected on 0.25 m intervals (~2-3 kg) downhole using aircore drilling methods. • Whole samples are collected beneath a cyclone return system (i.e. no sample splitting is conducted). • Multiscreen sampling is undertaken to determine optimum screen size for beneficiation at each deposit. • Once determined, samples are then beneficiated at the appropriate screen size (1.7 mm for East Weipa, 0.3 mm for Andoom and 0.6 mm for the Amrun deposits). Drilling techniques • The current drilling method utilises aircore drilling. • The typical aircore rig used at Weipa is a Land Cruiser mounted rig with a small enough wheelbase to traverse drill lines cleared with one D-6 dozer blade width. Aircore drilling forces compressed air down a space inside the drill rods to the bit face, where the air is then used to return the sample up the inner tube of the drill rod and out via a cyclone. A three bladed HQ aircore bit is attached to 4-inch rods. The drilling system has been designed to reduce grinding of the sample. • Historic drilling utilised auger drilling (until 1991). Drill sample recovery • No direct recovery measurements of aircore drilling samples are performed. • Whole sample is taken. • Holes are re-drilled if there is excessive sample loss (determined visually). • Sample weights are recorded before and after beneficiation in the laboratory. Logging • Standardised Rio Tinto Aluminium Weipa bauxite logging systems are utilized for drilling. • Logging is currently conducted on Panasonic Toughpads and data is captured in an offline acQuire logging package at the drill rig. This system allows for data validation to be applied during logging as well as a streamlined method of exporting the data for importing into the main RTA Geology database. • Logging is qualitative in nature, i.e. based on lithology. Currently there are ~20 lithologies common to the deposits that get modelled into four horizons for the estimation of bauxite resources. • All sample intervals (0.25 m) are logged. • The holes are generally terminated four samples (1 m) into the floor lithologies as observed by the rig geologist. • Logged lithologies are vetted against historical drill holes and assay parameters. Sub-sampling techniques and sample preparation • No sub-sampling undertaken. • Sample preparation at Weipa is carried out at the purpose-built facility. The facility consists of two Kason washing screens, two drying ovens, a multiple screening facility, and grinding units. Beneficiated, un- beneficiated (crude) and multiscreen drill samples pass through this area prior to their being assayed for the major oxides and loss on ignition (LOI). • Sample preparation at ALS, Brisbane was set up with the same specifications of equipment as Weipa, however was scalable to install up to six Kasons, larger drying ovens, more grinding capability and room for multi-screens. • The majority of analyses are undertaken at ALS laboratory in Brisbane since 2015, prior to that the majority of the analyses were done at Weipa. Quality of assay data and laboratory tests • Bauxite industry standard XRF analysis of all major elements and a suite of trace elements are undertaken on all samples. • Matrix matched field standards are systematically used. • Laboratory preparation duplicates and assay standards also form part of the QA/QC procedure. • The Weipa Laboratory participates in a “round robin” process managed through the RTA Process Improvement team. This process includes all the RTA and affiliated laboratories and is reviewed on a quarterly basis to ensure that standards are maintained. The Weipa laboratory analysts also carry out internal checks on the assay data. Results not meeting certain criteria or outside a designated range are re-analysed. Standard and duplicate samples are also used by the Mine Geology Department to monitor the performance of the laboratory via standard QA/QC routines.
2 • The ALS, Brisbane laboratory maintains its NATA accreditation through annual inspections and testing as required. RTA visit and audit both the preparation facility and analytical rooms regularly. • Every assay batch returned from the laboratories is checked through acQuire QA/QC objects before being accepted to the database for use in resource estimation. Major oxides, LOI, and KSiO2 are checked routinely against performance of field standards, lab duplicates, and lab standards. • Any results for the standards that lie outside three (3) standard deviations of the mean are classified as failing QA/QC and assays excluded from modelling. • Analysis of the performance of certified standards, field duplicates, blanks and third-party check assaying has indicated an acceptable level of accuracy and precision with no significant bias or contamination. Verification of sampling and assaying • Twin holes were historically used incorrectly. Samples were combined for the two and tested. Results were not comparable for verification purposes. Twin holes are not in use anymore. • Infill drilling programs for resource definition return results in line with the wider spaced drilling. • Data validation occurs throughout the data collection process: during data capture, during importation into the database, following import into the database and during the modelling process (hole name, locations checks, RL checks, lithology order checks, missing data, incorrect data). Location of data points • Pre-2016 drill hole peg locations were surveyed to Australian Height Datum (AHD) and the Geocentric Datum of Australia 1994 (GDA94) grid (and converted to local mine grids) by contract surveyors using Differential Global Positioning System (DGPS) survey equipment which was accurate to 10 cm in both horizontal and vertical directions. • Post 2016 surveys utilises GNSS GPS systems. Where survey has not been completed, e.g. Amrun 2018- 2019 drilling campaign, LiDAR positioning of drill collar elevations are utilised to assist in the creation of the topographic surface. Data spacing and distribution • All downhole drill sampling is at 0.25 m intervals, and samples are taken of the roof and floor. • No sample compositing is done. Orientation of data in relation to geological structure • Not applicable in lateritic bauxite deposits. All drill holes are vertical, which intersects the horizontal ore body perpendicularly. Sample security • Samples are collected, bagged, ticketed and sealed at the drill sites. Samples are placed in bulk plastic containers, with a capacity of ~ 300 samples, for shipment to the laboratory. All samples are electronically logged into a system for tracking and validation. Samples are placed on a dispatch advise form and verified by the laboratory on arrival. All assay pulps are stored at Weipa or ALS Brisbane in purpose-built sample storage facilities. Audits or reviews • An external Mineral Resource and Ore Reserve audit was completed in 2019 on the Weipa deposit. This audit had an outcome of Satisfactory with one medium and five low rated potential risks to the Mineral Resources and Ore Reserves. Actions were put in place to address all findings. • Numerous internal peer reviews and studies have also been undertaken over the years. These reviews concluded that the fundamental data collection and modelling techniques were appropriate. SECTION 2 REPORTING OF EXPLORATION RESULTS Criteria Commentary Mineral tenement and land tenure status • The Weipa Bauxite deposits are located on the western side of the Cape York Peninsula in far north Queensland, Australia. Mining Lease (ML) 7024 and ML 7031 covers the various deposits. ML 7031 was obtained through the acquisition of Alcan in 2007. ML 6024 is a separate lease that is held in order to provide infrastructure access between the north of Embley and south of Embley operations at Weipa. • ML 7024 was granted by the State Government of Queensland under a separate Act of Parliament, “The Commonwealth Aluminium Corporation Pty. Limited Agreement Act 1957”. The effective date of the lease granted under this act is 1/1/1958 and the expiry date is 31/12/2041 with an option to extend to 31/12/2062. Lease extensions past 2062 can be obtained, beyond the initial renewal period, subject to both parties’ right to terminate on two years notice. • ML 7031 was granted by the State Government of Queensland under a separate Act of Parliament “The Alcan Queensland Pty. Limited Agreement Act of 1965”. The effective date of the lease granted under this act is 1/1/1964 and the expiry date is 31/12/2047 with an option to extend to 31/12/2068. Lease extensions past 2068 can be obtained, beyond the initial renewal period, subject to both parties’ right to
3 terminate on two years notice. Exploration done by other parties • Not applicable. Weipa is a mature mining operation with more than 50 years of operational and orebody knowledge. Geology • The rocks of the Cape York Peninsula are divided into two geological units: the sedimentary rocks on the western side of the peninsula and the igneous and metamorphic rocks exposed in the hills on the eastern
4 side of the peninsula (the Coen Inlier). • The Cape York Peninsula bauxites are confined to a dissected laterite plateau, known officially as the Weipa Plateau on the west coast of Cape York Peninsula. The Weipa Plateau is one of three geomorphologic land units that are of particular interest to the geology of the bauxite and kaolin resources. The other two units are the Merluna Plain and the Mapoon Plain. • The Weipa Plateau is a low plateau, usually no more than a few tens of metres above sea level and has been dissected by various river systems resulting in a series of irregularly shaped islands. It is intensely weathered to a depth of 20 m to 30 m with the upper part of the weathered material reconstituted into various nodules as well as some partially cemented rocks. The flatness of the plateau has meant it has been immune to erosion other than by rivers eating away at the sides. Much of the plateau’s volume was removed in solution in the groundwater, which is also responsible for the formation of the bauxite. The sedimentary rocks of the Weipa Plateau fall into two categories: o The Rolling Downs Group Sediments; and o The Bulimba Formation Sediments (Weipa Beds). • These two groups of sediments are eroded and weathered to form the Weipa bauxites. The different sediments resulted in different types of bauxite formations • The Bulimba Formation sediments lie on top of the Rolling Downs Group and occupy channels that cut down into them. The Rolling Downs Group were uplifted above sea level and weathered before the Bulimba Formation sediments deposited on them. The river sediments are less homogeneous than the marine ones. Deposition occurred as short erratic events rather than a slow continuous one and a changing sea level resulted in a mixture of sands and clays. The greater variability in the sediments is reflected in greater local variability in grade of the Weipa type bauxites. • Andoom type bauxites are derived from shallow marine sediments that are fine grained, with very little quartz, and this material is generally screened at 0.3 mm. The Weipa type bauxites are derived from river deposited sediments that are coarse grained, with abundant quartz, and this material is therefore screened at 1.7 mm. Drilling at Amrun suggests a more intensely braided river system allowing more mixing between the Bulimba and Rolling Downs formations. This fits with the optimum screen size of the area being between the Andoom and Weipa deposits. Amrun is currently screened at 0.6 mm • The Cape York Peninsula bauxites are thin, tabular deposits that vary from zero to 10 m in thickness and are continuous laterally for many kilometres. The unconsolidated pisolites are generally overlain by 0.5 m topsoil and sit on an ironstone base. • The rocks of the Bulimba Formation and Rolling Downs Group have been converted to bauxite via a continuum of weathering. An annual high rainfall and a geologically stable environment has provided the perfect ingredients for a world-class bauxite deposit to form over many millions of years. A deep saprolitic zone overlain by a classic mottled zone below the bauxite mineralisation attests to this. • The process of bauxitisation involves the conversion of kaolinite to the bauxite minerals gibbsite and boehmite. The principal influence on the process is the composition, supply and movement of groundwater. The pH of the groundwater is lowered during the process of bauxitisation and we note that the process is still ongoing as we see a low pH regularly throughout the ground water monitoring bores across the RTA mining leases. To a lesser extent there are organic influences such as vegetation, and possibly burrowing organisms and temperature. • The dissolution of both kaolin and quartz control the distribution of silica grades in the deposits. The combination of kaolin and quartz distributions results in a typical vertical chemical profile that is usually found throughout the deposits and appears to be independent of the bauxite thickness i.e. the same vertical grade trend is found in both thin and thick bauxites. The typical vertical grade profile for silica is high silica at the top of the bauxite, which quickly drops to a much lower silica value that plateaus for the majority of the profile and then rises quickly back to high silica values again right at the base of the bauxite profile. As alumina is left behind by the dissolution of kaolinite, the typical vertical grade profile for alumina is almost the inverse of silica. The relationships between the genetic processes and the resulting grade profiles are displayed in the figure on the next page.
5 Drill hole Information • As this report relates to Mineral Resources and no Exploration Results are being reported, this section is considered not applicable. Resource work is currently more focussed on asset evaluation rather than exploration, systematically bringing the bauxite classification to higher levels of confidence. Data aggregation methods • Not applicable. Weipa is a mature mining operation with more than 50 years of operational and orebody knowledge. Relationship between mineralisation widths and intercept lengths • All drill holes have been drilled perpendicular to the horizontal stratigraphy of the deposit. • All known horizons of the deposits: overburden, red soil, bauxite and ironstone are intersected and sampled during drilling. Drilling continues for 1 m into the ironstone to ensure the transitional boundary between the ore and floor is intersected. Diagrams • Type section for the Weipa deposits. Balanced reporting • Not applicable. Weipa is a mature mining operation with more than 50 years of operational and orebody knowledge. Other substantive exploration data • Not applicable. Weipa is a mature mining operation with more than 50 years of operational and orebody knowledge. Further work • Drilling will continue in the future to further support the five-year and life of mine plans, as well as options for future growth.
6 SECTION 3 ESTIMATION AND REPORTING OF MINERAL RESOURCES Criteria Commentary Database integrity • Data capture is on Panasonic Toughpad digital loggers that have internal validation rules that identify logging errors. • The geological drill hole database (RTA Geology) is managed by the Bauxite Geology Team within Rio Tinto Aluminium. Drilling data is securely stored in a Microsoft SQL Server using an acQuire front end. acQuire is a third-party software product that provides a user-friendly interface to SQL Server and consists of two components: • a Relational Data Model (structured storage tables and links) optimised for the storing of exploration and mining data information; and • a Software System (objects for data collect/importing/exporting, validation, viewing, modification, etc.) to manage the data and provide end user functionality for the optimum use of exploration and mining data. • The database is located on a virtual server hosted in Rio Tinto’s Azure cloud servers in Sydney. They are backed up daily in accordance with Rio Tinto’s standard back up procedure. • The drill hole database used for Mineral Resource estimation has been internally validated. Methods include checking: o acQuire scripts for relational integrity, duplicates, total assay and missing / blank assay values; o Domain names; o Null and negative grade values; o Missing or overlapping intervals; o Duplicate data. • Drill hole data was also validated visually by domain and compared to the geological model. Site visits • The Competent Person for the Weipa Mineral Resource visits the site on a regular basis and is involved in all aspects relating to the ore body knowledge. Geological interpretation • The Weipa bauxite geology is not complex in nature and is well understood as a predominantly pisolitic lateritic weathering profile. • Geological modelling of the bauxite horizon is undertaken using drill hole lithological logging and assay data. Logged lithologies are grouped into four horizons for modelling and estimation purposes, these are: o Overburden (Soil, Overburden, Sand) o Red Soil (Red Soil and Soil Red) o Bauxite (Bauxite, Clay Bauxite, Cemented Bauxite, Transition and Clay Transition) o Ironstone (Ironstone, Clay ironstone, Transition, Clay Transition and Clay) Transition and Clay Transition difficult to log in the field, horizon allocation based on grade • Cross-sectional interpretation of the bauxite stratigraphy is conducted. Three-dimensional wireframes of the sectional interpretations are automatically produced and used to generate and inform the block model. • Geological models using the 1970’s data (4 out of 16 models) are not as reliable as those using post 1995 data due to the difference in the quality of drilling and sampling techniques. These areas currently have a downgraded classification and are systematically being replaced with new holes to increase confidence in the Mineral Resource estimation. Dimensions • The Weipa bauxite deposits are laterally very extensive, covering the majority of ML 7024 and ML 7031 (~380k ha). Deposits vary in average thickness from 1.5 m to around 8 m and vary from 0.3 m to 0.6 m below surface cover. Estimation and modelling techniques • Basic geostatistical analysis is used to help with domaining decisions. Most deposits were modelled as a single domain, apart from Moingum (Hey Point) where two domains have been modelled due to differences in bauxite thickness, grades and source rocks. • Interpretation and estimation are performed using Datamine’s Studio RM software, with variography undertaken in FSS International Consultants (Australia) GS3M software. • Four horizon codes, based on the lithology and assays, are assigned for the modelling and estimation of bauxite resources, see the Geological Interpretation section. Each deposit is a single domain laterally, divided into four horizons vertically. • The bauxite horizon is flattened to the mid-point of the drill hole intersections at East Weipa and Andoom, while drill hole collars are flattened to constant elevation at Amrun, Norman Creek and Moingum (Hey Point). The wireframes are filled with blocks on an in/out basis; there is no sub-blocking or block proportions used. • Major oxides; LOI and Recovery for the bauxite horizon are estimated using ordinary kriging into parent cells. • Overburden and red soil are assigned 0% recovery for the estimation of resources.
7 • Cemented bauxite grade is estimated as part of the bauxite horizon and assigned a 100% recovery; the proportion of cemented bauxite is estimated as an indicator variable. • Major oxide chemistry is also estimated for the overburden, red soil and ironstone horizons, where data is available. Ordinary kriging is used for interpolation, using the variogram models from the bauxite domain. • A multiple pass search strategy is used to estimate grades, as shown in the following table. Pass 4 is not used in the East Weipa and Andoom areas. Maximum extrapolation distance is slightly less than the maximum search radii due to the requirement to use at least two holes to estimate each block. Search Radii Samples Pass X (m) Y (m) Z (m) Min Max Per Hole 1 120 120 1.0 12 24 5 2 180 180 1.5 8 24 5 3 360 360 3.0 8 24 5 4 720 720 6.0 8 24 5 • There are no extreme grade values, so no grade cutting is required. • Estimation parameters and search distances are determined from consideration of the drill hole and sample spacing in each deposit, as well as the anisotropy of the variogram models. • The plan extents of the block models extend at least two blocks past the drilling grid. In the vertical direction, four ‘edge’ blocks are created below the base of drilling. • The block size is set at half the drill hole spacing in the horizontal (40 m x 40 m at Andoom and East Weipa; 50 m x 50 m at Amrun) and at the sample spacing in the vertical (i.e. 0.25 m). • The model block size effectively is the selective mining unit (SMU). • Deleterious element silica is assayed using XRF. Kaolinite (reactive silica) is determined using NIR analysis. Quartz is determined by difference. • No specific assumptions are made regarding the correlation of variables during estimation as each element is estimated independently. Some attributes do show strong positive or negative correlation in the drill hole samples, and the similarity in variogram models for different attributes and identical search parameters effectively guarantee that these correlations are preserved in the estimates. • Routine validation of the block model estimation is completed using global model versus sample statistics, swathe plots, grade tonnage curves, volume checks, and visual cross-section comparisons (block estimates against drill hole samples). Filtering by search volume and number of samples can improve comparisons. • The Mineral Resource estimates take appropriate account of previous estimates and mine production. The new models are broadly comparable with previous estimates despite significant changes in methodology. While detailed reconciliation has not been undertaken, the new models appear to reconcile with mine production within tolerable limits, as previous estimates did. Moisture • All Mineral Resource tonnages are reported on a dry basis. • All Mineral Resources are reported as the beneficiated dry product, except for North of Weipa (reported as DSO product). Cut-off parameters • Grade cut-offs are routinely used to determine the potential Mineral Resources of the modelled horizons. Where lithological contacts are transitional, chemical cut-offs based on alumina and silica are used with <15% SiO2 and > 40% Al2O3. These blocks are then reviewed against location (buffer areas), as well as thickness cut-offs ranging from 0.5 m to 1 m for each deposit. • The estimation of Ore reserves utilises an economic parameter, summarised as the margin realised upon sale of the bauxite. The economic parameter is used as a check to validate technical resource assumptions applied in determining available resources. Mining factors or assumptions • Weipa is mined through shallow open cut techniques developed over several decades of operations. After topsoil is removed, front end loaders excavate the bauxite and belly dump trucks transport the bauxite to the beneficiation plant. • As the Weipa orebodies are shallow, geotechnical risks are extremely low. Pre-production drilling is completed in order to provide better definition of the roof and floor contacts for the five-year mine planning process. • Estimates include internal dilutional but no allowance for external dilution or mining recovery. Dilution and mining recovery are applied during the reserving process, not during estimation. • A minimum mining thickness for the bauxite horizon of 0.5 m is used for the final determination of resource figures. Metallurgical • Weipa bauxite is beneficiated through established techniques to improve product quality and handleability.
8 factors or assumptions This is achieved through the removal of the finer fraction and leaves the coarser material as product. • The beneficiation process typically involves wet screening and may include the use of cyclones and classifiers depending on the part of the deposit being beneficiated. • Expected bauxite recovery and quality from the beneficiation process is assessed through laboratory scale test work of samples generated from the resource drilling process. • Bauxite mineralogy has been investigated through numerous studies, primarily using wet chemical techniques, to understand how it will react in the Bayer Process, which is used to extract the alumina at the refineries. • A proprietary mineralogical calculator “MinCalc” is used to estimate bauxite mineralogy and Bayer processing grade for Cape York ore from the routinely collected elemental chemistry and thermogravimetry, as routine wet chemical techniques are prohibitively expensive. MinCalc calibration is orebody specific and is validated and recalibrated during the life of mine operations. Environmental factors or assumptions • East Weipa, Andoom and Amrun have obtained all relevant environmental approvals required to continue operations. • Operation of tailings dams at Weipa (East Weipa and Andoom) and Amrun are covered by relevant government permits. Bulk density • Bulk density is not measured on drill hole samples and is not modelled. • Bulk density has been determined by using the Sand Replacement Method and Nuclear Density Gauge Testing according to Australian standards AS 1289.5.3.1-1993 and AS 1289.5.8.1-1995. • A number of studies have been conducted over the deposits with the most common test pit spacing being 5,000 m. This produced the bulk density utilised in resource tonne calculations. • Default values are also assigned to cemented bauxite, overburden and floor material for each of the different deposits at Weipa, see table below. Bulk Density Parameters Value (t/m3) Overburden 1.23 Bauxite 1.47 – 1.67* Cemented Bauxite 2.50 Ironstone/Floor 1.42 *Different for each deposit Classification • Drilling is conducted to a 76.2 x 76.2 m (Andoom and East Weipa) or 50 m x 100 m (at Amrun) spacing for grade control purposes. • To be declared a Measured Resource a deposit must be drilled to a 152.4 m x 152.4 m (Andoom and East Weipa) or 100 m x 200 m (at Amrun) spacing. • Indicated Resources are drilled on a 305 m x 305 m (Andoom and Weipa) or 200 m x 400 m (at Amrun) spacing. • Inferred Resources are drilled on a 914.4 m x 914. 4m (Andoom and East Weipa) or 800 m x 1200 m (at Amrun) spacing and utilize multiscreen drilling. • Classification within the bauxite horizon is based on the search pass used to estimate grades, using increasing search radii and decreasing numbers of samples for each subsequent pass. Passes 1 and 2 are classified as Measured Resources, Pass 3 as Indicated and Pass 4 as Inferred. Search Radii Samples Resource Category Pass X (m) Y (m) Z (m) Min Max Per Hole Measured 1 120 120 1.0 12 24 5 2 180 180 1.5 8 24 5 Indicated 3 360 360 3.0 8 24 5 Inferred 4 720 720 6.0 8 24 5 • Appropriate account has been taken of all relevant factors. • The Competent Person is satisfied that the current Mineral Resource classification reflects the relevant factors for the deposit. Audits or reviews • An external Resource and Reserve audit was completed in 2019 on the Weipa deposit. This audit had an outcome of Satisfactory with one medium and five low rated potential risks to the Mineral Resources and
9 Ore Reserves. Actions were put in place to address all findings. • Numerous internal peer reviews and studies have also been undertaken over the years. These reviews concluded that the estimation techniques were appropriate. Discussion of relative accuracy/ confidence • The relative accuracy and confidence level in the Mineral Resource estimates are considered to be in line with the generally accepted accuracy and confidence of the nominated JORC Mineral Resource categories. This has been determined on a qualitative, rather than quantitative, basis, and is based on the estimator’s experience with a number of deposits at Cape York and similar deposits elsewhere. The main factors that affect the relative accuracy and confidence of the estimates are the drill hole spacing and the local definition of the lithological horizons. • The estimates are local, in the sense that they are localised to model blocks of a size considered appropriate for local grade estimation. The tonnages relevant to technical and economic analysis are those classified as Measured and Indicated Mineral Resources. SECTION 4 ESTIMATION AND REPORTING OF ORE RESERVES Criteria Commentary Mineral Resource estimate for conversion to Ore Reserves • The Ore Reserve estimates are developed from the geological models current as at September 2020, and the Mineralogy model updated in 2020. • Mineral Resources are stated exclusive of Ore Reserves. Site visits • The Ore Reserves Competent Person has been employed by Rio Tinto Aluminium for a significant period and has visited Weipa several times in recent years. Study status • Weipa Operations has been operating continuously for several decades, and the Ore Reserve estimate and life of mine plan are updated annually. This includes the reconciliation of operating parameters and review of input assumptions into the planning processes. • The Ore Reserve estimate for Amrun is based on the completed and approved feasibility study and has now been in continuous operation for two years. This includes the reconciliation of operating parameters and review of input assumptions into the planning processes. Cut-off parameters • The Ore Reserve cut-off is based on an economic parameter, summarised as the margin realised upon sale of the bauxite. The economic cut-off approach considers revenue (bonus/penalty), fixed/operating/capital costs, royalties and other third-party payments. Bauxite that satisfies this economic cut-off, is considered for inclusion in the Ore Reserve. Mining factors or assumptions • The Ore Reserve is mined through shallow, open cut techniques developed over several decades of operations. Once the area is tree cleared and the topsoil/overburden removed, the bauxite is hauled to the beneficiation plant for processing. Several mining areas are active at any one time to enable blending and to mitigate against operational risk. • As the Ore Reserve is shallow, geotechnical risks are low. Stockpile heights and wet road conditions are managed in accordance with standard operating procedures • Dilution and mining recovery parameters are applied during the Ore Reserve estimation process, based on reconciliation of past performance and reviewed annually. • Minimum bauxite mining thickness of 0.9 m is used for Amrun Ore Reserve estimation. • Minimum bauxite mining thickness of 0.5m is used for East Weipa & Andoom Ore Reserve estimation, with an upper bench minimum mining thickness of 0.3 m. • Inferred Mineral Resources are not considered in the estimation of Ore Reserves. Metallurgical factors or assumptions • Weipa bauxite is beneficiated through established techniques to improve product quality and handleability. This is achieved through the removal of the finer fraction, leaving the coarser material as product. The beneficiation process typically involves wet screening and may include the use of cyclones and classifiers depending on the part of the deposit being beneficiated. • Expected bauxite recovery and quality from the beneficiation process is assessed through laboratory scale test work of samples generated from the resource drilling process. • Extractable alumina is calculated through application of a Mineralogy model. Environmental • All relevant environmental approvals have been obtained to continue operations. • An EIS has been completed for Amrun with the relevant governmental approvals having been obtained. • Operation of tailings dams at Weipa (East Weipa, Amrun and Andoom) are covered by relevant government permits. Infrastructure • As East Weipa, Andoom and Amrun are all established sites, all appropriate infrastructure for the existing operation is already developed. This includes water, power, sewage, stores, maintenance workshops,
10 administration buildings and the Weipa town-ship. Any infrastructure expansion required in the future is allowed for in the financial modelling that supports the Ore Reserve. Costs • Operating and sustaining capital costs are sourced from the Weipa Operations financial model. • Future capital costs are based on project study estimates or five-year plan sustaining capital amounts. • Traditional owner and carbon tax assumptions are factored into the financial modelling. • Exchange rates are based on internal Rio Tinto modelling of expected future country exchange rates. Revenue factors • Commodity prices are based on internal Rio Tinto modelling of the future supply and demand balance for bauxite, alumina and aluminium. This includes the bonus and penalty adjustments for quality. • Queensland royalties are included in the financial modelling at 10.0% of the bauxite price. • Exchange rates are based on internal Rio Tinto modelling of expected future country exchange rates. Market assessment • Industry analysis is undertaken to assess the existing and future supply and demand balances in bauxite, alumina and aluminium. This includes assessing likely incentive pricing required to bring on new capacity. • Internal Rio Tinto forecasting revises production guidance on an annual basis. Economic • Operating costs are built up from first principles while capital costs are included based on current estimates. Appropriate escalation is built in where capital costs are to be incurred in the future. • The discount rate to be used in the NPV model is supplied from Rio Tinto corporate and is set based on risk adjusted cost of capital. • Sensitivity analysis is carried out to assess key project drivers and the sensitivity of the project economics to movements in these drivers. Social • Weipa has in place the Weipa Community Co-existence Agreement (WCCCA) with local traditional owners. It also has a Community Relations department that seeks to build relationships with the local communities in and around Weipa. Other • Tenure to exploit the Weipa deposit is granted through two state agreements and is held through two mining leases ML 7024 and ML 7031. • The Queensland Government Comalco (ML 7024) lease expires in 2041 with an option of a 21-year extension, then two years’ notice of termination; the Ely Alcan Queensland Pty. Limited Agreement Act 1965 (ML7031) expires in 2048 with a 21-year right of renewal with a two-year notice period. • An EIS process was completed for the Amrun brown field mining expansion. Both the Queensland and Commonwealth governments have approved the EIS subject to several conditions. Classification • Given the high level of confidence in the reserve modifying factors, all Measured Resources were converted to Proved Ore Reserves and all Indicated Resources were converted to Probable Ore Reserves. • Inferred Mineral Resources are not considered in the estimation of Ore Reserves. • The Competent Person is satisfied that the current classification is reasonable for the Weipa Ore Reserves. • No Probable Reserves have been derived from Measured Resources. • The Competent Person is satisfied that the stated Ore Reserve classification reflects the outcome of technical and economic studies. Audits or reviews • A Mineral Resource and Ore Reserve internal audit was completed in 2019 and in 2015 on the Weipa deposit. These audits concluded that there were medium and low rated potential risks to the Mineral Resources and Ore Reserves. All findings from the recent audit have been actioned for completion by end of 2021. Discussion of relative accuracy/ confidence • Ore Reserve estimates are compared with production data on a half yearly and annual basis at Weipa. This reconciliation shows that for all key parameters, production was within ±7% of the estimates for calendar year 2020.
1 Appendix 2 Gove Table 1 The following table provides a summary of important assessment and reporting criteria used at Gove for the reporting of mineral resources and ore reserves in accordance with the Table 1 checklist in The Australasian Code for the Reporting of Exploration Results, Mineral Resources and Ore Reserves (The JORC Code, 2012 Edition). Criteria in each section apply to all preceding and succeeding sections. SECTION 1 SAMPLING TECHNIQUES AND DATA Criteria Commentary Sampling techniques • All sampling has been completed at 0.25 m intervals using a vacuum drill rig. Samples are recovered as drill chips and are typically 0.7kg in weight. • Samples of waste material (overburden, laterite) are not always sent for assay. The decision of which samples to send for assay is made by the logging Geologist. • In every case, the full sample is sent to the laboratory for preparation (including homogenisation) and assay. Drilling techniques • Drilling is completed using a tractor mounted 44 mm diameter vacuum drill rig. Drill sample recovery • No direct recovery measurements of samples are performed on the drill rig. Sample weights are recorded at the laboratory, and the recovery at the rig is visually estimated and recorded as part of the logging as a percentage of what is expected. Logging • Geological logging is performed from top to bottom of every drill hole on 0.25 m intervals. • Geological logging includes specific detail on lithology and estimated recovery as a minimum for each sample. Up to 2 lithologies can be logged for each sample, enabling a more accurate description of the sample than one lithology would. Sub-sampling techniques and sample preparation • No sub-sampling undertaken. • The entire 0.25 m sample interval is submitted for preparation and analysis. Field split duplicates are collected at the rate of 1:100 samples using a manual riffle splitter at the drill site. • The entire sample is oven dried and then pulverised. • XRF determination is used to analyse the major elements Al2O3, SiO2, Fe2O3, TiO2 as well as a suite of minor oxides, and LOI by TGA Leco instrument (1000°C) for each 0.25 m sample. Quality of assay data and laboratory tests • Matrix matched standards and duplicates are submitted at the rate of 1:100. • For each assay batch reported, field standards, field duplicates, lab standards and lab repeats are checked via a database QA/QC object, which displays information on standards and duplicates, contained in the batch before the batch is accepted in the database. • After the assay data is accepted and imported, multiple batches can be viewed using QA/QC objects in the database, to detect overall trends in the data. • If a batch is rejected, a thorough analysis is conducted into the reason, and if no obvious explanation can be found, re-analysis is requested from the laboratory. Verification of sampling and assaying • Twinned holes were historically used to ensure sampling and assay results are consistent over a short distance. They can also be used to double check short-distance variability in the ore. • QA/QC data exists and confirms that the sample assays are likely to be representative and are appropriate for inclusion in the mineral resource estimate. Location of data points • Drill holes are surveyed by mine surveyors. Drilling completed in 2013 was surveyed using real time kinematic GPS (Leica GPS 1200 - GPS & Glonass) +/- 50 mm horizontal, +/- 100 mm vertical precision). Earlier drilling has variable survey methods. A project wide LiDAR DTM survey has been used in conjunction with the collar surveys to provide a complete terrain model. Collar exclusions based on uncertainty of location are managed in the database Collar table with two fields “Collar_Surveyed” and “In_Res_Calc”.
2 Data spacing and distribution • The majority of the drilling for Gove has now been completed on a 50 m x 50 m grid, which is the spacing for a Measured Resource. • Areas not drilled on a 50 m x 50 m grid are either on the edges of the plateaus or have already been mined. Areas that have been drilled to the standard 200 x 200 m grid are scheduled to be drilled on a 50 m x 50 m grid well in advance of mining. Orientation of data in relation to geological structure • All holes are vertical which is perpendicular to the sub-horizontal weathering/laterite domains hosting the bauxite resource. Sample security • Samples are collected and grouped into bulka-bags at the drill site. Bags ready for shipping to the laboratories are stored in company warehouse fenced lots. Transport of samples is via barge to Darwin and road transport to Bureau Veritas, Perth where all bags are matched against dispatch information and stored in the laboratory warehouse prior to preparation. Audits or reviews • Numerous audits and reviews of previous Mineral Resource and Ore Reserve estimates have been undertaken by independent consultants, including Xstract and Snowden. SECTION 2 REPORTING OF EXPLORATION RESULTS Criteria Commentary Mineral tenement and land tenure status • Rio Tinto mines bauxite from its mining lease at Gove, located 650 km east of Darwin in northeast Arnhem Land in the Northern Territory, Australia. The majority of the bauxite product is shipped to external customers, with moderate internal consumption at Rio Tinto alumina refineries in Gladstone. The figure below outlines the location of the operation. 5 km N
3 • The operations at Gove are covered by a number of Special Purposes Leases, a Special Mineral Lease and a Residue Disposal Area Lease. In 1968, an agreement was made between the Commonwealth of Australia and Nabalco Pty Ltd for the granting of a Special Mineral Lease (SML 11) pursuant to the Mining Act (Gove Peninsula Nabalco Agreement). SML 11 has been divided into three parts specifically to reflect the agreement conditions to use the leased land for the purposes granted. The table on the following page summarises the SML 11 use and areas. Lease Use Lease Area (ha) SML 11 Part 1 Mine 20,018 SML 11 Part 2 Conveyor 282 SML 11 Part 3 Refinery 191 • In 2011, Gove Operations renewed all its lease agreements including the Special Mineral Lease covering the mine, conveyor and refinery (SML 11 Parts 1, 2, and 3) and some associated Special Purposes Leases. A new Residue Disposal Area (RDA) lease agreement replaces the previous SPL 403 lease. All leases have been renewed until 2053. The Community & External Relations Department at Gove Operations are responsible for maintaining the lease arrangements. The table and map below summarise the leases that cover the Gove Operations. Lease Use Lease Area (ha) Date of Lease Expiry SML 11 Part 1 Mine 20,018 29 May 2053 SML 11 Part 2 Conveyor 282 29 May 2053 SML 11 Part 3 Refinery 191 29 May 2053 SPL 213 Bulk Cargo Wharf 13 29 May 2053 SPL 217 General Cargo Wharf 14 29 May 2053 SPL 249 Foreshore Protection 8 29 May 2053 SPL 253 Water Discharge 0.1 29 May 2053 SPL 277 Cooling Water Intake 0.7 29 May 2053 Residue Disposal Area Lease Residue Disposal 5,132 29 May 2053 SPL 214 Nhulunbuy Township 660 29 May 2053 SPL 250 Industrial Area 50 29 May 2053 SPL 251 Conveyor Access 23 29 May 2053
4 Exploration done by other parties • Not applicable. Gove is a mature mining operation with more than 40 years of operational and orebody knowledge. Geology • Bauxite deposits at Gove occur at a latitude slightly more than 12 degrees south of the equator. The bauxite appears to be a weathering product of Cretaceous marine sediments, which overlie Proterozoic basement rocks of the Arnhem Inlier. Uplift, weathering, and erosion of the sediments, as well as formation of the bauxite, took place during the Tertiary, but exact timing and detail of events remains uncertain. Marine reworking of weathered products is likely to have taken place, possibly more than once. • Pisolitic textures are dominant, with variable cementation. However, variably cemented coarser nodule horizons are also common. Some deeper bauxite, as well as underlying laterite, contain abundant interconnected solution cavities, which may have been created geochemically and/or biologically (associated with root channels). Modern day root channel structures and infill, in the upper part of the bauxite, are common. Gibbsite is the major ore mineral, with boehmite being of lesser significance. • Bauxite occurs on top of a gently undulating plateau at approximately 30 to 70m above sea level, over an area measuring approximately 20km east-west by 15km north-south as seen in the figure below. The plateau has been variably dissected by erosion into three main plateau areas. Main Plateau is the largest contiguous bauxite area, whilst Rocky Bay and Eldo are two adjacent smaller bauxite areas, figure below. The present day extent of the bauxite likely represents a fraction of what was originally present. SML 11 Part 1 SML 11 Part 3 Residue Disposal Area 5 km N
5 • The bauxite horizon has historically been up to 10m thick but is more typically 2 m to 4 m in thickness. It comprises several distinct layers in either a dominantly cemented (“hard”) profile, or a dominantly poorly cemented to uncemented (loose or soft) profile. The latter tends to be thicker and occupies hollows in the plateau surface and is often considered to be a proximal erosion/deposition product of the “hard” profile bauxite. • Hard profile bauxite generally consists of footwall vuggy laterite (Lat), overlain by nodular bauxite (Lower Nodules – Nod), overlain by extremely vuggy “tubular” bauxite (Tub), overlain by strongly cemented pisolitic bauxite (Cemented Hard - CH), overlain by weakly cemented pisolitic bauxite (Cemented Soft - CS), overlain by a thin loose pisolite layer (LP), overlain by a thin overburden layer (OB) being a mix of loose pisolites and topsoil, figure on next page. • All of the lithologies described above are generally elevated in bauxitic alumina throughout the deposit plateau. The extent of the mineralisation described therefore depends on the chemical grade threshold and thickness limits applied.
6 Drill hole Information • As this report relates to Mineral Resources and no Exploration Results are being reported, this section is considered not applicable. Resource work is currently more focussed on asset evaluation rather than exploration, systematically bringing the bauxite classification to higher levels of confidence. Data aggregation methods • In reporting Exploration Results, weighting averaging techniques, maximum and/or minimum grade truncations (e.g. cutting of high grades) and cut-off grades are usually Material and should be stated. • Where aggregate intercepts incorporate short lengths of high grade results and longer lengths of low grade results, the procedure used for such aggregation should be stated and some typical examples of such aggregations should be shown in detail. • The assumptions used for any reporting of metal equivalent values should be clearly stated. Relationship between mineralisation widths and intercept lengths • All drill holes have been drilled perpendicular to the horizontal stratigraphy of the deposit. • All known horizons of the deposit are intersected and sampled during drilling. Drilling continues for 1 m into the floor horizon to ensure the transitional boundary between the ore and floor is intersected. Diagrams • Refer to previous sections for location and tenements maps, as well as a type section. Balanced reporting • Not applicable. Gove is a mature mining operation with 40 years of operational and orebody knowledge. Other substantive exploration data • Not applicable. Gove is a mature mining operation with 40 years of operational and orebody knowledge. Further work • Drilling will continue in the future to further support the five-year and life of mine plans, as well as options for future growth. SECTION 3 ESTIMATION AND REPORTING OF MINERAL RESOURCES Criteria Commentary Database integrity • Data is stored in an SQL database with the acQuire software as front end. Data management and data access was managed on site up to 2004. From 2004 to 2008 data was managed by Snowden. From 2008 to
7 2012 data was managed by Rio Tinto Alcan. From 2012 to 2013 Snowden was responsible for maintaining the database. From 2014 onwards Rio Tinto Aluminium has again managed the database. • Data is logged directly into an acQuire logging package which retains database rules offline. When the data is uploaded to the database, further checks are run. Spatial validation is completed in both 2D and 3D software packages prior to releasing the data for modelling. Site visits • The Competent Person for the Gove Resources visits the site regularly. Geological interpretation • Interpretation of the bauxite (alumina) mineralisation is based on the physical logging of regolith domains, and on chemical zonation based on results from analysis. Overburden and low alumina lateritic formations bound the alumina enriched bauxite horizon. • Internal dilution was incorporated into each lithological domain to maintain continuity. All bauxite material is oxidised and hydrated aluminium. Dimensions • The Gove deposits extend over an area approximately 20 km east-west by 15 km north-south that includes three separate lateritic bauxite plateaus (Main Plateau, Rocky Bay and Eldo). The bauxite horizon averages 3.5 m in thickness but can be up to 10 m thick and is typically 0.6 m (up to 8.5 m) below surface cover. Estimation and modelling techniques • A three-dimensional (3D) block model was created for all post-2003 drilling. • Basic geostatistical analysis is used to help with domaining decisions. • Transformations are used in the Z (vertical) direction to better model the bauxite horizon chemistry. • Major oxides and LOI for the bauxite horizon are estimated using ordinary kriging. • Major oxide chemistry is also estimated for the overburden and ironstone horizons. Where a suitable variogram model can be established ordinary kriging is used for interpolation, otherwise inverse distance weighting cubed (or squared) is used for estimation. • Estimation parameters and search distances are determined from kriging neighbourhood analysis. • The 3D block model is automatically generated in the RTA macro process (ResMod1999) in Datamine. • The plan extents of the 3D block model extend one block past the drilling grid. In the vertical direction four ‘edge’ blocks are created both at the top and bottom of the bauxite horizon. • The block size is set at half the drill-hole spacing in the horizontal (i.e. 25 m x 25 m) and at the sample spacing in the vertical (i.e. 0.25 m). • Silica is assayed using XRF. Kaolinite (low temperature reactive silica) is determined using NIR analysis (done during 2004 campaign only). • Routine validation of the block model estimation is completed using global model versus sample statistics, swathe plots, grade tonnage curves, volume checks, and visual cross-section comparisons (block estimates against drill hole samples). Filtering by kriging variance, search volume, and number of samples can improve comparison. • A two-dimensional (2D) model was created to include all pre 2004 drilling. • Thickness estimated using Inverse Distance weighted to the power of three (ID3). • Block grades for alumina, silica, iron oxide, titanium oxide and loss on ignition are estimated using ordinary kriging into parent cells of dimensions 25 mE by 25 mN by 0.25 m RL. Density values are assigned based on lithology. • The estimated thickness is applied to the 2D model, creating a 3D model. • Semi-variograms are modelled for each lithology for each area and the best fit variograms selected. They show ranges of continuity between 1,700 m and 7,000 m for alumina and 1,510 m to 4,000 m for silica. Moisture • All Mineral Resource tonnages are reported on a dry basis. • All Mineral Resources are reported as crude (DSO) dry product. Cut-off parameters • No block cut-off is applied to the Mineral Resource estimate reported. • For the 3D model horizon interpretation, a cut-off of < 16% silica and > 35% alumina is specified for a sample to be considered bauxite. This threshold is chosen to better match the lithological logging and to give continuity to the boundaries of the bauxite horizon. • For the 2D model an arbitrary and historical cut-off (threshold) of < 12% silica and > 40% alumina is specified for a sample to be considered bauxite. • No top cuts were applied to the bauxite (alumina and silica) grades. • Top-cut values are applied to the boehmite and gibbsite grades using the 99.5 percentile for each lithology
8 for each area, prior to estimation. Mining factors or assumptions • Model assumes open pit mining of all defined resources and a minimum mining thickness of 1.0 m for the bauxite horizon. • Gove is mined through shallow open cut techniques developed over several decades of operations. After topsoil is removed, dozers rip and push down the ore, followed by front end loaders to excavate the bauxite and rear tipper haul trucks transport the bauxite to the crushing plant. • As the Gove orebody is shallow, geotechnical risks are extremely low. Pre-production drilling is completed in order to provide better definition of the roof and floor contacts for five-year mine planning process. • Dilution and mining recovery parameters are applied in the estimation process based on reconciliation of the reserve model to actual performance. Metallurgical factors or assumptions • Successful and representative beneficiation assessments have been performed throughout the history of the Gove Bauxite deposit, the grade and handleability of the ore allow a crude product to be mined and exported. • Bauxite mineralogy has been investigated through numerous studies, primarily using wet chemical techniques, to understand how it will react in the Bayer Process, which is used to extract the alumina at the refineries. • A proprietary mineralogical calculator “MinCalc” is used to estimate bauxite mineralogy and Bayer processing grade for Gove ore from the routinely collected elemental chemistry and thermogravimetry, as routine wet chemical techniques are prohibitively expensive. MinCalc calibration is orebody specific and is validated and recalibrated during the life of mine operations. Environmental factors or assumptions • There are numerous exclusion zones throughout the lease area. The block model has been coded with these exclusion areas for mine planning. • All overburden removed prior to mining is used to rehabilitate the final mining surface. The overburden is benign and only temporarily stockpiled before the whole amount is used in rehabilitation. • There is no floor or bench waste moved during mining of the Gove ore. Bulk density • In situ bulk density is estimated after individual samples are assigned a density based on lithology, which in turn is based on in situ bulk density test work. Classification • The estimate for the 3D model has been classified as Measured. • The estimate for the 2D model has been classified as a Measured, Indicated and Inferred Mineral Resource based on geological characteristics of the deposit, structural continuity (thickness) and drill hole spacing. • Measured Mineral Resource is nominally informed by 50 mN x 50 mE drill hole spacing with up to 100 m in some cases. • Indicated Mineral Resource is informed by drill hole spacing between 200 mN to 200 mE with up to 400 m in some cases. • Inferred Mineral Resource is the areas surrounding the Measured and Indicated Mineral Resource where continuity is expected and thickness >= 1 m. Audits or reviews • A Resources and Reserves internal audit was completed in 2014 by Rio Tinto Group Audit and Assurance on the Gove deposit. This audit concluded with an overall ranking of Marginal, identifying one high audit finding. A list of actions was developed in response to this, all of which have now been closed. Discussion of relative accuracy/ confidence • Confidence in the Mineral Resource estimate is high due to 40 years of successful mining history and 25 years of reconciliation data as well as appropriately spaced drilling (50 m x 50 m grid). SECTION 4 ESTIMATION AND REPORTING OF ORE RESERVES Criteria Commentary Mineral Resource estimate for conversion to • The Ore Reserve estimate was developed using the geological model finalised in November 2017, and the Mineralogy model finalised in August 2019. • No additional drill / assay information was completed / available for inclusion in the 2020 Reserve & Resource estimation.
9 Ore Reserves • Mineral Resources are reported exclusive of Ore Reserves. Site visits • The Ore Reserves Competent Person has been employed by Rio Tinto Aluminium for a significant period and has visited Gove Operations several times in recent years. Study status • Gove Operations has been operating continuously for several decades, and the Ore Reserve estimate and life of mine plan are updated annually. This includes the reconciliation of operating parameters, and review of input assumptions into the planning processes. Cut-off parameters • The Ore Reserve cut-off is based on an economic parameter, summarised as the margin realised upon sale of the bauxite. The economic cut-off approach considers revenue (bonus/penalty), fixed/operating/capital costs, royalties and other third-party payments. Bauxite that satisfies this economic cut-off, is considered for inclusion in the Ore Reserve. Mining factors or assumptions • The Ore Reserve is mined using a shallow, open cut strip mining technique developed over several decades of operations. Once the area is tree cleared and the topsoil/overburden removed, the bauxite is ripped by dozers before being hauled to the crusher for sizing. Several mining areas are active at any one time to enable blending and to mitigate against operational risk. • As the Ore Reserve is shallow, geotechnical risks are extremely low. Stockpile heights and wet road conditions are managed in accordance with standard operating procedures. • Dilution and mining recovery parameters are applied during the Ore Reserve estimation process, based on reconciliation of past performance and reviewed annually. • Minimum bauxite mining thickness of 0.7 m is used for Ore Reserve estimation. • Inferred Mineral Resources are not considered in the estimation of Ore Reserves. • Apart from maintenance workshops to maintain the mining equipment, stores for consumables and water stands there is no other significant infrastructure required with the mining method. Metallurgical factors or assumptions • There is no beneficiation plant at Gove Operations. The bauxite is sized through a primary and secondary crushing system before being stockpiled for shipping. • With no crushing loss, 100% of the bauxite mined is planned to be shipped. • Extractable alumina is calculated through application of a Mineralogy model. Environmental • Gove Operations has obtained all relevant environmental approvals required to continue operations as outlined in the Gove Mine Management Plan extending until 2030. Infrastructure • As Gove Operations is an established site all appropriate infrastructure for the existing operation is already developed. Any infrastructure expansion required in the future is allowed for in the financial modelling that supports the Ore Reserve. • Following curtailment of the Gove refinery in 2014 there exists an infrastructure surplus, which is being scaled down to meet the new demand of the operation. This includes power, water, labour, accommodation, offices, ancillary equipment, buildings, etc. Costs • Operating and sustaining capital costs are sourced from the Gove Operations financial model. • Future capital costs are based on project study estimates or five-year plan sustaining capital amounts. • Traditional owner and carbon tax assumptions are factored into the financial modelling. • Gove closure costs are based on the Gove Closure Order of Magnitude study. • Exchange rates are based on internal Rio Tinto modelling of expected future country exchange rates. Revenue factors • Commodity prices are based on internal Rio Tinto modelling of the future supply and demand balance for bauxite, alumina and aluminium. This includes the bonus and penalty adjustments for quality. • Northern Territory Government royalties are included in the financial modelling at 10.0% of the bauxite price. • Exchange rates are based on internal Rio Tinto modelling of expected future country exchange rates. Market assessment • Industry analysis is undertaken to assess the existing and future supply and demand balances in bauxite, alumina and aluminium annually. • Internal Rio Tinto forecasting revises production guidance on an annual basis. Economic • Operating costs are built up from first principles while capital costs are included based on current estimates. Appropriate escalation is built in where capital costs are to be incurred in the future. • The discount rate to be used in the NPV model is supplied from Rio Tinto corporate and is set based on risk adjusted cost of capital.
10 • Sensitivity analysis is carried out to assess key project drivers and the sensitivity of the project economics to movements in these drivers. Social • The Gove Mine Management Plan outlines the operating conditions agreed to until 2030. • Mining development must receive approval for archaeological and cultural heritage via the NT Heritage Conservation Act and sacred sites (administered by the Aboriginal Areas Protection Authority) receive approval via the NT Aboriginal Sacred Sites Act. Other • The Gove Operations lease agreements are valid through to 2053, including the Special Mineral Lease covering the mine (SML11 parts 1, 2 & 3) and the associated Special Purpose Leases including the new Residue Disposal Area (RDA) lease agreement. • Following curtailment of the Gove Refinery in 2014, Gove Operations no longer produces alumina or hydrate, only a bauxite product. Marketing arrangements have changed accordingly to enable this transition. • Statutory requirements are managed through the Gove Mine Management Plan. Classification • Given the high level of confidence in the reserve modifying factors, all Measured Resources were converted to Proved Ore Reserves and all Indicated Resources were converted to Probable Ore Reserves. • No Probable Reserves have been derived from Measured Resources. • Inferred Mineral Resources are not considered in the estimation of Ore Reserves. • The Competent Person is satisfied that the current classification is reasonable for the Gove Ore Reserve. Audits or reviews • A Resources and Reserves internal audit was completed in 2014 by Rio Tinto Group Audit and Assurance on the Gove deposit. This audit concluded with an overall ranking of Marginal, identifying one high audit finding. A list of actions was developed in response to this, all of which have now been closed.