Go Back Email this Link to a friend


HARMONY GOLD MINING COMPANY LIMITED - Appendix to the Updated Wafi-Golpu Feasibility Study

Release Date: 12/04/2018 16:52:00      Code(s): HAR       PDF(s):  
Appendix to the Updated Wafi-Golpu Feasibility Study

Harmony Gold Mining Company Limited
Registration number 1950/038232/06
Incorporated in the Republic of South Africa
ISIN: ZAE000015228
JSE share code: HAR
(“Harmony” or “the company”)


Appendix to the Updated Wafi-Golpu Feasibility Study

Johannesburg, Thursday, 12 April 2018. Harmony Gold Mining Company Limited
(Harmony) hereby releases an appendix to the announcement “Updated Wafi-Golpu
Feasibility Study” released on 19 March 2018. This release includes the JORC
Code 2012 Edition – Table 1 which underpins the Golpu Mineral Resource and
Ore Reserve. The updated production target which is planned to be mined is
based solely on the Probable Ore Reserves.

Golpu Mineral Resource

The Mineral Resources for Golpu Project remain unchanged from the release
titled “Golpu feasibility study confirms robust investment case” dated 15
February 2016 (the original release). Harmony confirms that it is not aware
of any new information or data that materially affects the information
included in the original release in regard to Mineral Resources, and that all
material assumptions and technical parameters underpinning the Mineral
Resource estimate in the original release continue to apply and have not
materially changed. Harmony confirms that the form and context in which the
Competent Person’s findings are presented have not been materially modified
from the original release.

Mineral Resources are reported inclusive of Ore Reserves.

Golpu Ore Reserve

As stated in the market release “Updated Wafi-Golpu Feasibility Study” dated
19 March 2018, the Feasibility Study Update Ore Reserve is estimated to
contain 5.5 million ounces of gold and 2.5 million tonnes of copper
(Harmony’s 50% interest). This estimate is materially in line with previous
estimates and reflects updated long term cost and metal price assumptions and
optimised designs in the Golpu Feasibility Study Update (refer Golpu Ore
Reserve Table below).

Golpu Ore Reserve1

                     Tonnes Gold Grade Copper Grade Insitu Gold Insitu Copper

                     (Mt)   (g/t Au)   (% Cu)      (Moz)       (Mt)

Probable Ore         200    0.86       1.2         5.5         2.5

Reserve


1   The Ore Reserve shown represents Harmony’s 50% interest.
JORC Code 2012 Edition – Table 1

Section 1 - Sampling Techniques and Data

-   Sampling techniques:

Diamond drill holes are the principal source of geological and grade
information for the Golpu deposit. Diamond core drilling was used to obtain
continuous samples ranging in size from PQ3 to NQ3 with rare intervals of BQ
which were cut into half (in the case of HQ, NQ and BQ) and into half or
quarter (in the case of PQ) using a diamond core saw, from which half or
quarter is prepared for assay and the remaining core retained in the core
farm as reference.

The half or quarter core sent for assay was bagged in labelled calico sample
bags with the sample number scribed on an aluminium strip included in the
bag. The calico bags were placed in larger polyweave bags and transported by
road or helicopter to Lae by company employees. Sampling intervals are
typically 1m or 2m fixed intervals. The entire half or quarter core is
dispatched for sample preparation. Core recovery is recorded to ensure a
representative sample is obtained.

All core was logged and photographed prior to cutting. Some core was wrapped
in tape during sampling to maintain core quality. Oriented core is cut along
the orientation line at the bottom of hole to reduce the possibility of
sample bias. Sample numbers and drill hole intervals were recorded by the
responsible geologist and used by technicians for cutting and sampling. A
sample despatch sheet documenting the sample numbers and required assay work
was sent with each sample batch to the laboratory.

All drill core is sampled and assayed over the entire hole length. However
empirical rock strength data is required for geotechnical input to mine
designs - since 2011 approximately 20cm of whole core was taken at 50m
intervals from all holes for Unconfined Compressive Strength testing which
were not assayed. This is not considered to present a material impact on
sample quality due to the disseminated, stockwork and micro-fracture infill
nature of the mineralisation.

-   Drilling techniques:

Diamond core drilling, PQ, HQ, NQ and BQ in diameter, triple tube core
barrels and oriented typically using the ACE core orientation system.

-   Drill sample recovery:

Core recovery is recorded for all diamond drilling on a metre by metre basis
as a percentage. Sample recovery was 96.4% over the entire drilling dataset
including oxide material and the adjacent Wafi epithermal mineralisation but
increases to 98.4% within the Golpu Mineral Resource volume. All drilling is
conducted using triple tube core barrels and appropriate core handling
protocols. No material relationship has been identified between core recovery
and grade due to the diffuse nature of the mineralisation in the Golpu
porphyry-style deposit.
-   Logging:

All diamond drill core has been geologically and geotechnically logged to
support appropriate Mineral Resource estimation, mining studies and
metallurgical studies. Geological logging is both qualitative and
quantitative and records lithology, mineralisation, alteration mineralogy,
weathering, structures and other physical characteristics of the core.

-   Sub-sampling techniques and sample preparation:

Sample preparation protocols for drilling in the Golpu area has varied over
time. However, all core is sawn in half or quarter typically cut beside the
orientation line. Only minor intervals of second half core submission has
been conducted. The entire sawn half or quarter core is submitted for the
first stage of sample preparation. All subsequent sampling is by riffle or
rotary splitters to ensure sub-sample representivity until homogenised in the
pulverisers. Grind size screening was typically applied also to maximise sub-
sampling representivity and to ensure compliance to sub-sampling sample mass
requirements.

Historic sampling from CRAE/Elders drilling 1990-1996 was prepared at Pilbara
(Analabs) Laboratories in Lae. All samples were dried and jaw crushed to a
nominal 5mm, then entirely pulverised to 180 microns. A sub sample of 500g
was obtained with a riffle splitting device which was pulverised in a LM2 to
nominal 75 micron. A 100g sub sample (pulp) was obtained and despatched for
analysis.

Sample preparation for Harmony and WGJV drill holes 2005-2011 was carried out
at Intertek Lae sample preparation facility with pulps sent to Intertek
Jakarta for assay. All samples were dried at 60°C and then jaw crushed to
nominal 2mm. A sub sample of 1.5kg was obtained with a riffle splitting
device which was pulverised to 75 micron using LM2 mill. An approximate 250g
sub sample (pulp) was obtained and despatched for analysis.

Sample preparation for WGJV drill holes 2012-2014 was carried out at the
Intertek sample preparation facility in Lae. All samples were dried at 60°C,
then crushed in a Boyd Crusher to a minimum 95% passing 2.8mm. A sub sample
of 3.5kg (±0.5kg) is obtained using a Rotary Splitting Device (RSD) and
pulverised in a LM5 mill with a minimum 95% passing 106 microns. An
approximate 250g sub sample (pulp) was obtained and despatched for analysis.
Representative pulverised material and crushate reject is retained for all
samples. Repeat samples are obtained from pulverised material at the rate of
1 in 20 samples and check crusher duplicates have also been analysed.

The sampling techniques used over the history of the project are considered
appropriate for assessment of porphyry mineralised systems.

-   Quality of assay data and laboratory tests:

QAQC protocols for Golpu drilling have varied over the project’s 24 year
history.

CRAE and Elders (1990-1996) sampled at Pilbara Laboratories Lae did not have
the support of regular submission standards or duplicates and but were
supported by regular submission of pulp splits to a second laboratory. Gold
was determined by 50g Fire Assay with AAS finish and multi-element analyses
including copper, silver, molybdenum, arsenic and iron were determined using
AAS. Approximately 20% of composites used in the Golpu Mineral Resource model
are derived from CRAE / Elders drilling – this are located in the upper Golpu
Porphyry where there is also significant drill data acquired by Harmony.

Drilling by Harmony and WGJV 2005-2014 was analysed at Intertek Laboratories
Jakarta and included submission of certified standards, blanks, quarter core
duplicates and re-assay of selected pulp splits at a second laboratory. Gold
was determined by 50 or 30g Fire Assay with AAS finish, multi-element
analyses including copper, silver, molybdenum, arsenic and iron was
determined by 2 acid ICPMS\OES finish analyses. From October 2013 multi-
element analyses have been determined by 4-acid (full) digest with ICPMS/OES
finish. From 2013 gold has been analysed at the Intertek Lae Laboratory.
Total sulphur was determined by Leco.

Pulp samples shipped to Jakarta are re-dried in their original pulp packets
at <60°C for a minimum of 4 hours or until dry before analysis. Certified
reference materials were inserted at the rate of 1 in 20 samples. Matrix-
matched samples from coarse reject Wafi-Golpu sample material were
homogenised, independently certified and implemented into the QA sample
stream from April 2013. Pulp samples (second sample from LM5 bowl) within
each sample batch are submitted at the rate of 1 in 20 samples. Coarse
duplicates have also been analysed and additionally 5% of all pulps with
accompanying new standards are checked at an independent laboratory.

Assay results are assessed on a per-batch basis on receipt of assays to
determine appropriate levels of accuracy and bias in gold and copper
analyses. The acceptance tolerance must be within defined site QAQC
protocols. Routine check assay programmes are conducted on a periodic basis.
All preparation and analytical laboratories including check laboratories have
been reviewed for operational procedures and QAQC compliance by project
geologist and QAQC chemists.

The analysis methods employed are considered appropriate for the material and
mineralisation. No material issues have been identified that invalidate the
use of primary assays now held in the Wafi-Golpu Datashed database for Golpu
deposit grade estimations.

-   Verification of sampling and assaying:

All field data is captured digitally into a Logchief logging system, stored
electronically in a Datashed database, and exported to a Lae based Datashed
database, which is maintained by the Database Manager. Digital assay files
are received directly from the laboratory and input directly to Datashed.
Significant intersections are reported by the geology team, and verified by
the Geology Manager.

No specific holes have been twinned at Golpu. However due to the drilling
configuration (typically towards grid west or to grid west on the common
sections and multiple holes from a single drill pad with small variation in
dip), multiple holes cross in close proximity. No inconsistency in sampling
and assaying have been identified.

No adjustment has been made to reported assays for use in the estimation of
the Mineral Resource.
-   Location of data points:

The local grid called Wafi Local Grid (WLG) is a planar grid oriented
approximately 45 degrees from north which is used up for block modelling and
geology databases. The height datum is Mean Sea Level but 5,000m is added for
WLG. WLG is datum point referenced to PNG Geodetic Datum 1994. Topographic
control is by digital terrain models derived from a high resolution Lidar
survey of the Golpu area conducted in 2007 with a reported spatial accuracy
of 0.2m.

Drill hole collar locations are located using hand held global positioning
system (GPS) and completed drillhole collars surveyed in the Wafi Local Grid
by a qualified and competent surveyor using a theodolite or differential GPS.

Downhole surveys were completed on all holes typically at 18m and then every
30m down the hole. Elders and CRAE drillholes were surveyed using an Eastman
single shot camera and Harmony / WGJV drillholes were surveyed using a Reflex
downhole survey tool. From 2011 surveys have been conducted by a fully
competent and licensed contractor using a north-seeking gyroscope instrument.

For all periods of the drilling programme, downhole surveying was determined
using the latest available methodology. These are considered sufficiently
accurate to locate all assays to the level of precision required for
classification as an Indicated or Inferred Mineral Resource.

-   Data spacing and distribution:

Drillhole spacing within the Golpu deposit ranges from less than 100m x 100m
in the upper portion of the deposit and up to 200m x 200m in the lower
portions of the deposit. The drill spacing is considered sufficient to
establish the degree of geological and grade continuity appropriate for
Mineral Resource classification of a large porphyry gold/copper system.
Drillholes are entirely sampled at regular 1m or 2m intervals regardless of
lithological or mineralogical boundaries. Assays are composited to 10m
downhole intervals for use in grade estimation.

-   Orientation of data in relation to geological structure:

The Golpu mineralised system is approximately elliptical in plan elongated
towards 345 degrees WLG with a steep west to sub-vertical dip. The majority
of drilling is oriented across this orientation, but the dataset does include
holes drilled parallel to the long axis. Most holes are complete transects
through the porphyry and enclosing mineralised host sediments. The
orientation of sampling is considered unbiased toward known structures and
adequate for the diffuse nature of the mineralisation style i.e. porphyry
gold copper mineralisation.

?   Sample security

Diamond drill core is delivered directly from the drill rig at the end of
each shift by the drill crew to the logging shed within the fenced and
patrolled Wafi Camp security compound. Core is marked up and photographed as
soon as possible to identify any core loss and ensure size and consistency of
the samples. Historically all core was sawn in half at the Wafi site and half
core for assay bagged into calico bags and in turn secured in plastic bags.
Samples are identified by both internal aluminium tags and external
labelling. Some whole core was directly shipped as plastic-wrapped and
secured trays to the dedicated core farm within the security patrolled
compound at Nine Mile, Lae. Core is sawn, bagged and identified as for the
Wafi site procedures.

Whether transported as whole core in trays or bagged sawn core samples, all
transport is always under the direct supervision of WGJV employees within
tamper evident packaging from site until delivery to the Intertek Laboratory
in Lae. Pulps and crusher residues are returned from the Lae laboratory to
the Nine Mile core farm for long term storage again under direct supervision
of WGJV staff.

Since 2005, core samples were prepared in Intertek, Lae within their secured
premises and pulps are air-freighted by international couriers to Intertek
Laboratory in Jakarta, Indonesia for assaying. A detailed labelling,
documentation and tamper evident packing protocol is in place for this
transfer. Pulps are stored on a long term basis in Jakarta. Assay results
from Intertek Jakarta are returned to WGJV network and loaded to the Wafi
database by dedicated administrators after correlation against despatch
records and after passing QAQC protocols.

-   Audits or reviews:

Internal reviews of core handling, sample preparation and assays laboratories
were conducted on a regular basis by both project personnel and owner
representatives. External consultants also reviewed sampling protocols and
based on heterogeneity studies for sampling mass and sampling precision
provided recommendations to improve QAQC early in the drilling program.

In the Competent Persons opinion, the sample preparation, security and
analytical procedures are consistent with current industry standards and are
entirely appropriate and acceptable for the styles of mineralisation
identified and will be appropriate for use in Mineral Resource estimates.
There are no identified drilling, sampling or recovery factors that
materially impact the adequacy and reliability of the results of the drilling
programme in place on the Wafi-Golpu Property.



Section 2 - Reporting of Exploration Results

-   Mineral tenement and land tenure status

The Wafi-Golpu project is located in Exploration Licence (EL440) within the
Morobe Province of Papua New Guinea. The property is located at approximately
6º52'S latitude, 146º27'E longitude approximately 60 km southwest of Lae, the
nearest commercial centre within the region with a population of about
90,000.

The owner of the project is the Wafi-Golpu unincorporated joint venture
(WGJV), one of three unincorporated joint ventures in the Morobe Province of
Papua New Guinea between subsidiaries of Newcrest Mining Limited (Newcrest)
(50%) and Harmony (50%) referred to collectively as the Morobe Mining Joint
Ventures (WGJV). The WGJV holds two exploration licences covering a total
area of approximately 129 km2, registered in the name of the WGJV
participants Newcrest PNG2 Ltd (50%) (a wholly owned Newcrest subsidiary) and
Wafi Mining Limited (50%) (a wholly owned Harmony subsidiary). Key proposed
infrastructure areas are located on adjoining EL1105.

EL440 tenement licence expires in March 2018 and a renewal was lodged in
December 2017 which is currently pending. The EL1105 tenement licence expires
in January 2019. Both tenements remain in good standing.

Subject to the project being developed, a royalty of 2% of net smelter
revenue and a Mining Levy of 0.25% is payable to the Government of Papua New
Guinea.

A compensation agreement with local landowners is in place whereby specified
payments are made due to impacts of exploration activities including loss of
trees, impact on water resources, access restrictions, and disturbance to
sacred sites and burial sites.

Consistent with the current administrative practice of the Government of
Papua New Guinea and under the terms of the Wafi-Golpu exploration licences,
the Government of Papua New Guinea has reserved the right to acquire up to a
30% equitable interest in the project. In January 2011, the PNG Government
indicated an intention to exercise the option, nominating the State-owned
Petromin PNG Holdings Limited to take up the interest. The option is
exercisable at any time prior to commencement of mining. Under the terms of
the State option set out in the Wafi-Golpu exploration licences, the price
payable for the interest is the proportionate share of the accumulated
exploration expenditure at the point of exercise. Post-exercise, the State
holding entity will be responsible for their proportionate share of
continuing exploration, development and project costs.

-   Exploration done by other parties:

Exploration has been conducted by the WGJV since 2008. Previous exploration
activity has been documented by many workers, and notably includes Harmony,
Abelle, Elders and CRA during their tenure since the 1970’s. The Golpu
Porphyry was discovered by Elders in 1991 and the high grade Hornblende
(Livana) Porphyry by WGJV in 2010. Data transferred from previous exploration
programmes has been assessed for quality and risk associated with inclusion
of this data evaluated in the Mineral Resource estimation.

-   Geology

The Golpu deposit lies in a block of deformed Upper Mesozoic to Middle
Miocene metasedimentary to sedimentary rocks cut by Miocene-Pliocene calc-
alkaline dioritic intrusives. Copper and gold mineralisation results from a
multiple intrusive porphyry system with the upper portion overprinted by high
sulphidation epithermal alteration. Post mineral faulting has displaced and
rotated the original intrusive configuration.

The deformational history of PNG is characterised by accretion during oblique
collision of the Australian and Pacific plates. A series of arc-normal
transfer structures formed across PNG which taped mantle derived melts to
high crustal levels. One of these structures termed the Wafi Transfer
Structure is interpreted to have facilitated the emplacement of the Golpu
intrusives.
The Golpu Porphyry system consists of multiple, hornblende-bearing diorite
porphyries intruded into host sediments. The porphyries are separated based
on their spatial position, and where not texturally destroyed, into coarse
hornblende-rich, feldspathic-rich or quartz ‘eye’ inclusions variants.
Intrusives range from small dykes to small stocks and apopheses. Individual
intrusions pinch and swell vertically over tens of metres and form stocks,
pipes and dykes.

The Golpu deposit is approximately 800m by 400m elliptical in plan and
extends from 200m below surface to approximately 2,000m. Hydrothermal
alteration related to the porphyry gol-copper mineralisation forms a
predictable zonal arrangement grading from potassic core to propylitic
margins. A high-sulphidation epithermal system is ‘telescoped’ over the upper
portion of the porphyry system forming a central alunite–quartz (advanced
argillic) core grading out to dickite–kaolinite (argillic) with an outer
margin of sericite alteration. This results in either epithermal-dominant,
interaction (mixed) or porphyry-only zones within the Golpu deposit.

Mineralisation is derived from either the porphyry or epithermal systems.
Within the porphyry environment, mineralisation is disseminated,
microfracture and stockwork vein controlled. The dominant copper-gold
sulphide species varies laterally and vertically within the deposit from an
inner bornite (plus chalcopyrite) core, chalcopyrite as the dominant copper
sulphide grading to a pyrite only shell. The porphyry system is mineralised
with gold, copper, silver and molybdenum. The Livana Porphyry is the main
mineralised porphyry. Other porphyries act either as weak mineralisers or as
benign hosts.

In the high sulphidation epithermal system which is ‘telescoped’ over the
upper portion of the Golpu Porphyry, gold occurs within pyrite or as electrum
associated with pyrite-enargite-tetrahedrite. Abundant arsenian pyrite
results in high sulphur and elevated arsenic levels in the epithermal altered
volume. Mineralisation broadly follows the metasedimentary and volcanic host
rocks stratigraphy (40° dip to east).

Post-mineral thrust (reverse) faulting has dismembered the original porphyry
and epithermal systems with offsets of up to 200m within the mineralised
column and rotated the high grade porphyry core between faults to dip 70
degrees to grid west.

-   Drill hole Information:

No exploration has been reported in this statement, therefore there is no
drill hole information to report. This section is not relevant to this report
on Mineral Resources.

Comments relating to drill hole information relevant to the Mineral Resource
estimate can be found in Section 1 – “Sampling techniques”, “Drilling
techniques” and “Drill sample recovery”.

-   Data aggregation methods:

No exploration has been reported in this release, therefore there are no
drill hole intercepts to report. This section is not relevant to this report
on Mineral Resources.
Comments relating to data aggregation methods relevant to the Mineral
Resource estimate can be found in Section 1 – “Sampling techniques”,
“Drilling techniques” and “Drill sample recovery”.

-   Relationship between mineralisation widths and intercept lengths:

No exploration has been reported in this release, therefore there are no
relationships between mineralisation widths and intercept lengths to report.
This section is not relevant to this report on Mineral Resources.

-   Diagrams:

No exploration has been reported in this release; therefore, no exploration
diagrams have been produced. This section is not relevant to this report on
Mineral Resources.

-   Balanced reporting:

No exploration has been reported in this release, therefore there are no
results to report. This section is not relevant to this report on Mineral
Resources.

?   Other substantive exploration data:

No exploration has been reported in this release, therefore there are no
results to report. This section is not relevant to this report on Mineral
Resources.

-   Further work:

No further exploration is planned for the Golpu Mineral Resource volume.
Specific underground drill programs have been designed within the 2015 Golpu
FS BC1 and BC2 volumes to increase confidence in local grade precision and
refine geotechnical conditions at critical mine and supporting infrastructure
locations. Such programs would be implemented after establishment of access
declines. Proposed additional surface drilling is confined to infill
geotechnical investigations of access paths. These proposed drill programmes
are not exploration related.



Section 3 - Estimation and Reporting of Mineral Resources

-   Database integrity:

Data from the Golpu Project is stored within the WGJV ‘Datashed’ software
database located at the Lae office, PNG. Drill core is logged directly into
laptops in the core shed with periodic integration to the WGJV database.
Assay data is received from the laboratory in digital format which is
subsequently uploaded to the WGJV database using import templates. All data
uploaded to the database must pass a data integrity checks and reviews. User
access to the database is controlled by a hierarchy of permissions and are
controlled by WGJV database administrators with oversight of data integrity
by an external Datashed software specialist.
Historical assay data collated by CRAE was imported into the WGJV database
from an existing MS Access database. The process used by CRAE to transfer
assay data into their database is not recorded, however checks of the assay
data in the database with the original hardcopy results indicate they are
satisfactory for use in a Mineral Resource estimate.

Detailed data review was completed before the estimation of the Golpu
December 2015 Mineral Resource estimate. Checks included validation of collar
surveys against planned locations and downhole surveys consistency of hole
path. Assays were reviewed and compared against observed mineralisation.
Logging records were reviewed against core photographs as part of the
interpretative geology compilation. All corrections were completed before
final data extraction for input to the Mineral Resource estimation.

-   Site visits:

The Competent Person is an employee of Newcrest Mining and travelled to site
on a regular basis as a former member of the WGJV team during the last
resource drilling campaign. Site visits validated the documented mapping,
drilling, logging and sampling processes and on-site data management.
Laboratory visits to the Lae preparation laboratory, Lae fire assay
laboratory and Jakarta assaying facilities were conducted to verify assaying
and QAQC procedures.

-   Geological interpretation:

The December 2015 geology model for the Golpu deposit includes lithology,
alteration, oxidation, sulphide distribution and structures wireframes. Fault
wireframes include major thrust faults which displace mineralisation. The
most significant thrust is the Reid Fault which displaces the upper Golpu
mineralisation approximately 200m up-dip with a small displacement to the
north.

All lithological, porphyry-related alteration and fault models were
constructed in Leapfrog software using implicit modelling interpolations from
primary logging codes and modified for interpretative correlations of logged
intervals. The implicit modelling methodology is considered less subjective
than traditional sectional interpretations.

ASD and ‘Corescan’ spectral mineralogical data was used in conjunction with
the current logged alteration dataset. This enables a higher level of
resolution of the layered epithermal system especially for subdivision of
clays and other difficult to distinguish indicator minerals of alteration
type and intensity.

Estimation domains are a combination of mineralised porphyry or host units,
alteration type and fault partition.

All geological contacts are honoured in the geological interpretations used
for grade estimation domains. The confidence in the geological volumes and
lithological and faulted contact correlations that were used in the
estimation domains is reflected in the resource classification. The
geological and structural framework used in the Mineral Resource has also
been externally reviewed. It concluded that the current model is supported by
contacts seen in core and makes kinematic and geometric sense with no obvious
flaws. There is sufficient drill data to constrain the geological model that
alternative interpretations will not be materially different from the
framework used in the 2015 Mineral Resource. The geological and structural
model has defined the major structural and lithological contacts that impact
grade continuity.

-   Dimensions:

The Golpu deposit is approximately 800m by 400m elliptical in plan and
extends from 200m below surface to approximately 2,000m depth.

-   Estimation and modelling techniques:

The Golpu Mineral Resource grades were estimated with Ordinary Kriging using
pairwise variograms of 10m composites for seven elements: gold, copper,
silver, molybdenum, sulphur, arsenic, and iron with Vulcan software using
domain specific variograms and search for informing 10m composites using the
variogram anisotropy. The grades were estimated into a block model with 40m x
40m x 40m parent cells with 10m x 10m x 10m resolution on domain margins.
This reflects the estimation precision available from the drillhole spacing
of less than 100m x 100m in the upper portion of the deposit and up to 200m x
200m in the lower portions of the deposit and the planned mining method
(block caving). Variograms are typically low nugget (7-17% for gold and 5-30%
for copper) with long ranges. Search parameters vary by element and
estimation domain but reflect the orientation and ranges of the variograms.
The maximum number of samples per block typically restricts the actual
distance of informing samples to substantially less than the search limits.
While there are spatial associations between elements, all are estimated
independently.

The grade estimation is based on an underlying ‘diffusion’ model where grade
trends from lower to higher values from the mineralisation margin to the
porphyry core in a relatively continuous relationship. Domain drift is
apparent for the porphyry system and pairwise variograms were used for
modelling grade continuity. Contact analyses indicated the Hornblende
(Livana) Porphyry has abrupt grade contacts and is modelled independently.
Estimation domains are also bounded at all major thrust faults where drilling
has demonstrated clear grade truncations. Most other estimation domains are
continuous ‘diffusive’ transitions from mineralised porphyry margins to the
mineralisation limit regardless of host lithology. All porphyry-related
domains are modelled with an orientation defined by the elongation of the
porphyry system. All epithermal, oxidation and cover sequence domains have
shallow dips to grid east again reflecting their overall orientation.

Top-cuts were applied to gold and copper composite grades but have no impact
on global estimated Mineral Resources. No top cuts were applied to arsenic
composites. This is a potential contaminant in copper concentrate and
sensitivity to high grade arsenic composites is required to evaluate the
final As content potentially delivered to concentrate.

Silver and molybdenum are modelled as they may reach potentially extractable
by-products however silver and molybdenum are not included in the revenue
estimation. Sulphur and iron are estimated as they inform sulphide speciation
and gold:sulphur and copper:sulphur ratios are included in metallurgical
recovery models.
The model has been validated by comparison with informing composite
declustered statistics and alternative modelling methods including
conditional simulations. Alternative models constructed included nearest
neighbour, inverse distance, raw variogram Ordinary Kriging, Discrete
Gaussian Model, and Conditional Simulation models with Sequential Gaussian
into nodes and Direct Block Simulation using Turning Bands into 10m blocks.
The impact of independently domaining the Livana Porphyry as a ‘hard’
boundary compared to incorporation into a continuous grade trend was also
evaluated.

The risk associated with the inclusion of historical data has been evaluated
by re-modelling without the non-QAQC validated data - there is no material
change between models. Historical assays have been included in the Mineral
Resource estimate to improve local estimation precision only.

There are no selective mining units applied to the Mineral Resource
reflecting the planned mining method.

The grade, recovery and value models used to quantify the Golpu Mineral
Resource are considered appropriate for the style of mineralisation and are
suitable for the required estimation precision for the planned mining method
– block caving.

-   Moisture:

All tonnages are calculated and reported on a dry tonnes basis.

-   Cut-off parameters:

The Mineral Resource estimate is reported within a break-even value shell
using the 2017 (unchanged from 2015) Mineral Resource revenues from gold and
copper only and the cost structure from the 2015 Stage 2 PFS (Life-of-Mine-
Plan based on 14Mt/year from block cave mining with processing by sulphide
flotation producing a copper concentrate for pumping to Lae port and shipment
to overseas smelters). Costs include block cave mining, treatment /
processing and General and Administration (G&A). Net Smelter Return (NSR)
includes metallurgical recoveries and off-site realisation (TCRC) including
royalties. Gold revenues assumptions are US$1,300/oz and copper US$3.40/lb.

The value of each in-situ block is estimated and a smoothed shell generated
at the break-even margin. The shell includes internal below value cut-off
blocks and excludes isolated above cut-off blocks. While not a block-cave
design, the shell is representative the bulk mining method planned – block
caving. All Mineral Resources are constrained within the margin breakeven
‘value’ shell representing the limit to eventual economic extraction.

-   Mining factors or assumptions:

The Mineral Resource estimate is reported within a notional constraining
shell at the marginal break-even cut-off, based on mass mining by block
caving with no internal selectivity. The 40m x 40m x 40m block-model size and
the application of a constraining spatial shell that includes all internal
materials and excludes above margin break-even blocks outside the notional
shell reflects the non-selective planned mining method.
-   Metallurgical factors or assumptions:

The metallurgical recovery included in the margin estimation is based on ore
processing by copper flotation with copper and gold recovery to copper
sulphide concentrate.

Significant test-work has been completed to establish recovery algorithms for
copper and gold. Metallurgical domains are based on the host lithology and
alteration type. Each metallurgical domain is assigned a recovery algorithm
further subdivided on copper:sulphur and gold:sulphur ratios. Estimated
metallurgical recovery is included in the quantification of the Mineral
Resource reporting margin value cut-off. For the Mineral Resource cut-off,
recovery models are applied for porphyry, high chalcopyrite porphyry,
sediments and epithermal alteration domains.

Silver and molybdenum are included in the Mineral Resource reporting volume
but revenues are not included in the margin value estimation in line with the
2015 Stage 2 PFS. There is no dedicated recovery and revenue path in the 2015
Life-of-Mine PFS for these elements but both have reasonable prospects of
eventual economic extraction with only minor changes to the metallurgical
flow-sheet. Current modelling indicates silver in copper concentrate will not
consistently be above payable grades but this can be potentially achieved
during concentrate marketing negotiations. Molybdenum will similarly not
always be above cut-off grades however potentially economic grades are
present within the block-cave volume.

-   Environmental factors or assumptions:

Based on environmental characterisation completed to date, there are no
recognised physico-chemical or biological environmental factors that will
limit potential mining or milling operations. Geochemical assessment of rock
and tailings has been completed to quantify acid forming characteristics and
composition of the material. Waste rock locations, construction and dump
design alternatives have been evaluated and designed given this information,
with adequate controls allowed for acid rock drainage management.
Hydrological models have been undertaken and test water bores have been
constructed to evaluate mine vicinity water flows. Mine water will require
treatment for both entrained silt contents and acid rock drainage and pH
management before eventual discharge to the receiving environment. Treatment
of water will ensure a quality that meets PNG Receiving Water Criteria to
mitigate potential impacts to downstream communities and the environment.
Options for terrestrial tailings dams have been evaluated and viable options
designed. All development and production activities will be permitted by the
PNG Department of Environment and Conservation under the Environment Act
(2000).

-   Bulk density:

Bulk density has been determined on 10cm core samples typically at 10m
intervals down all holes. Methods used to derive bulk density values include
air/water (approximately 95%) and wax/water (approximately 5%) where samples
are friable. The average bulk density, after statistical review and removal
of outliers, is assigned to domains derived from a combination of oxidation,
alteration and lithology. The assignment of a constant bulk density per
domain assumes limited internal variation within the domain.
No elements reflecting sulphide mineralogy are considered significantly
abundant to correlate bulk density and grade within the reported Mineral
Resource volume.

-   Classification:

The Mineral Resource is classified based on: geological confidence as a
function of continuity and complexity of geological features; data spacing
and distribution; and estimation quality parameters including distance to
informing samples for block grade estimation.

Indicated Mineral Resource, where the geological framework can be modelled
with confidence and mineralisation continuity can be assumed, is classified
from below the intense epithermal alteration zone to the 4,100m Wafi Grid
Level (WGL) - approx. 1,400m below surface or to a major interpreted fault at
similar depth. Below this fault and above 3780m WGL, drillhole spacing is
increased and geological and grade continuity is less reliable – this volume
is classified as Inferred Mineral Resource. All Mineral Resources are
constrained within the margin breakeven ‘value’ shell representing the limit
to eventual economic extraction. It is the Competent Person’s view that the
classifications used for the Mineral Resources are appropriate for the
deposit.

-   Audits or reviews

The geological and structural framework used in the Mineral Resource has also
been externally reviewed. It concluded that the current model is supported by
contacts seen in core and makes kinematic and geometric sense with no obvious
flaws. The Mineral Resource estimate was the subject of independent external
review by AMC. No material issues were identified in these reviews and AMC
concluded that the estimates had been prepared using accepted industry
practice and classified and reported in accordance with the JORC 2012 Code.

-   Discussion of relative accuracy/confidence:

For an Indicated Resource estimate it is considered reasonable for the local
relative uncertainty to be +/- 15% in tonnage, grade and metal (exclusive of
each other, i.e., each variable has to satisfy the criteria) for an annual
production volume at a 90% confidence level. Direct block co-simulations
(gold and copper) of the annual production volumes represented by the average
height of draw for BC1 and BC2 in the 2015 Golpu FS were evaluated to
demonstrate confidence intervals. This evaluation indicates this criterion
can be satisfied. Relative uncertainties and confidence level estimates are
considered for both copper and gold as they are both significant economic
contributors. There is no production from the Golpu deposit to compare
relative accuracy and confidence.
Section 4 - Estimation and Reporting of Ore Reserves

-   Mineral Resource Estimate for conversion to Ore Reserves

The Golpu deposit lies in a block of deformed Upper Mesozoic to Middle
Miocene metasedimentary to sedimentary rocks cut by Miocene-Pliocene calc-
alkaline dioritic intrusives. Copper and gold mineralisation results from a
multiple intrusive porphyry system with the upper portion overprinted by high
sulphidation epithermal alteration. Post mineral faulting has displaced and
rotated the original intrusive configuration.

The Golpu Mineral Resource grades were estimated with Ordinary Kriging using
pairwise variograms of 10m composites for seven elements: gold, copper,
silver, molybdenum, sulphur, arsenic, and iron. The grades were estimated
into a block model with 40m x 40m x 40m parent cells with 10m x 10m x 10m
resolution on domain margins. This reflects the estimation precision
available from the drillhole spacing and the planned mining method (block
caving).

The Mineral Resource is classified based on: geological confidence as a
function of continuity and complexity of geological features; data spacing
and distribution; and estimation quality parameters including distance to
informing samples for block grade estimation. Indicated and Inferred Mineral
Resources were constrained within a margin breakeven ‘value’ shell
representing the limit to eventual economic extraction.

The reported Golpu Mineral Resources are inclusive of Ore Reserves.

-   Site Visits:

The Competent Person for the Ore Reserve estimate travelled to site on the
following occasions:

•     April 2015 – Site familiarisation to confirm suitability for
infrastructure and inspect core.

•      June 2016 – Inspect progress on geotechnical drilling programme.

•      June 2017 – Select a site for the proposed Nambonga decline portal.

-   Study Status:

A Feasibility Study Update was completed in March 2018 for the development of
an underground mine comprising of three block caves (BC44, BC42 & BC40). The
Feasibility Study Update provides supporting basis for this Ore Reserve
estimate. BC44 and BC42 are at a Feasibility confidence level, while BC40 is
at a Pre-Feasibility confidence level.

These studies show that the mine plan is technically achievable and
economically viable taking into consideration all material Modifying Factors.
-   Cut-off Parameters:

The Golpu Ore Reserve employs a value based cut-off determined from the Net
Smelter Return (NSR) and site operating costs based on the outcomes of the
Feasibility Study Update. The cut-off values applied for the estimation of
Ore Reserves are:

Activity                                           Units              USD(real)

Development prior to first BC44

crusher commissioning                              USD/t ore milled   10

BC44                                               USD/t ore milled   60

BC42                                               USD/t ore milled   40

BC40                                               USD/t ore milled   19.15



The NSR calculation takes into account reserve revenue factors, metallurgical
recovery assumptions, transport costs, refining charges and royalty charges.

The site operating costs include mining cost, processing cost, relevant site
general and administration costs and relevant sustaining capital costs.

-   Mining factors or assumptions:

Estimation of the Golpu Ore Reserve involved standard steps of mine
optimisation, mine design, production scheduling and financial modelling.
Factors and assumptions have been based on numerical modelling as well as
experience and performance in similar caving operations. The basis of the
analysis is considered at Feasibility (BC44 and BC42) and Pre-Feasibility
(BC40) study levels.

Preceding Pre-Feasibility and Feasibility studies completed in 2012, 2014 and
2015 deemed block caving to be the appropriate underground mining method to
maximise the economic output of the Mineral Resource. The Feasibility Study
Update (on which this Ore Reserve statement is based) defined a three lift
block cave mine plan. Extraction levels for the three block caves are;
4400mRL (BC44), 4200mRL (BC42), and the 4000mRL (BC40). Geotechnical
assessment during the studies has resulted in the following key block cave
mine design parameters in the Feasibility Study Update:

Mine Design Parameter                Value

Undercutting Strategy                Advanced Undercut

Undercut Design                      W Cut with Apex level

Extraction Level Layout              El Teniente

Extraction Spacing                   30m x 18m
Draw Column Height                  Average

                                    BC44      320m

                                    BC42      490m

                                    BC40      590m

                                    Maximum

                                    BC44      530m

                                    BC42      805m

                                    BC40      1,120m



Additional drilling will be required to collect further data for further
geological, geotechnical and metallurgical studies to inform final design.

Grade control during the production phase will be in the form of block cave
drawpoint sampling.

The in situ grade model estimated in July 2014 was the basis for the Ore
Reserve estimate.

The Feasibility Study Update proposes the following mining approach:

•     Secondary/Initial underground access via the Nambonga Decline to
provide earlier and quicker access to underground drill platforms, second
means of egress and ventilation.

•     Primary underground access via the Watut Portal and the twin Watut
Declines to the underground block cave mine. The Watut Declines also form
part of the primary ventilation circuit and materials handling system
conveying ore to the Watut Process Plant.

•     A ‘Cave Engineering Level’ established above the Reid Fault at 4870mRL
for data gathering, further refinement of the rock mass, monitoring of the
cave and potentially dewatering.

•     Ore extracted via three block caves producing at 17Mtpa (design
capacity) using an inclined conveying system to discharge on a stockpile on
the surface.

The following Modifying Factors have been applied:

•     All development has mining factors for dilution and recovery applied to
accurately represent the expected mined tonnes.

•     Decline, access and infrastructure shapes for BC42 and BC40 outside of
the Mineral Resource have tonnes contributing but not metal; these tonnes are
allocated to unclassified material.

•     PCBC™ software is used for cave production scheduling and estimation of
grade for material drawn from the block caves.

The total Life of Mine dilution is approximately 17%.
The geological model is classified as Indicated and Inferred Mineral
Resources. There is no Measured Mineral Resource. Mine plans are based on the
extraction of caving blocks solely delineated on the basis of Indicated
Mineral Resources.

Ore Reserves estimates and statements are required to include estimates of
dilution.   The dilution included in the total Ore Reserve (400Mt on 100%
basis) is approximately 79Mt due to the block cave mining method. The
dilution included in the Ore Reserve contains 7% of the gold metal and 5% of
the copper metal of the Ore Reserve and does not have a material impact upon
the estimate. Even without consideration of the metal contained in the
dilution incorporated in the Ore Reserve, the economic analysis indicates an
economic Probable Ore Reserve. The Wafi-Golpu Project is a greenfield block
caving project and will require the following mining infrastructure to
support the block caves:

•     ventilation fans and refrigeration equipment;

•     dewatering equipment;

•     crushing and conveying equipment; and

•     underground workshop, service and personnel facilities.

-   Metallurgical factors or assumptions:

The ore will be processed on site at the proposed treatment plant with a
design capacity of 17 Mtpa using conventional single stage SAG and ball mill
grinding, recycle crushing and flotation methods that are incrementally sized
to match the mining rate to produce a copper and gold concentrate. The
technology associated with the ore processing is industry standard for this
style of deposit.

The key metallurgical testwork for the Golpu deposit can be grouped into five
main programmes as follows:

•     Testwork completed prior to 2011 on samples from above 5120mRL.

•     2012 PFS Variability testwork and Metallurgical Domain Model completed
on samples over the vertical extent of the known Golpu deposit from 5120mRL
to 3850mRL across 14 exploration drill holes.

•     2013/14 Variability and flowsheet development testwork from 102
composites in the 2012 PFS programme. Variability samples were prepared from
material selected from exploration drill holes to provide spatial and grade
variability within the respective domains. The testwork samples were obtained
from 14 exploration drill holes across seven metallurgical domains.

•      2015 Feasibility Study testwork programme executed testwork through the
chosen process flowsheet using bulk samples from a mine plan targeting the
development of two block caves. Based on the mine development, the ore types
identified in the early years of production included domains 29 (Sericite
metasediment), domain 30 (Sericite porphyry) and domain 33 (Actinolite
porphyry) and account for 92% of material mined within the planned block
caves.
•     2018 Feasibility Study Update testwork programme including comminution
testwork to determine milling characteristics for ore from 4000mRL. This ore
is characterised by higher work indices than the ore higher in the orebody,
thus additional work will be executed in the near future to confirm
characteristics and if necessary alter the mill specifications during
detailed design.

A total of 13 geometallurgical domains were assigned to represent an improved
geological interpretation of the Golpu deposit and increase the understanding
of the copper and gold recoveries in the deposit. Gold and copper recoveries
are calculated for each domain. The geometallurgical domains are based upon
103 composite samples assembled from 17 exploration drillholes through the
entire deposit. Life of Mine metallurgical recoveries are:

•      Gold 68%

•      Copper 95%

Final concentrate derived from the testwork was utilised to conduct a product
quality assessment, which incorporated chemical analysis for major elements
and potential deleterious elements. The analysis indicated that the levels of
deleterious elements in concentrate did not exceed any of the typical
concentration restrictions for sale.

-   Environmental:

Feasibility study level analysis is in progress assessing the potential
environmental impacts of the mining and processing operations required for
the mining of the Golpu deposit and an Environmental Impact Statement is
proposed to be submitted by the WGJV to the PNG government by the end of June
2018.

NAF (Non-Acid Forming) waste rock would be produced from the first 300m of
the Nambonga Decline and the first 2,000m of the twin Watut Declines. This
material would be used to construct the retaining wall, base and access road
for the PAF (Potentially Acid Forming) cells. PAF would be expected to be
encountered from below these points for the remaining scope of the mine. This
material will either be stored in cells encapsulated in impervious material
or treated via the processing plant.

Deep Sea Tailings Placement (DSTP) has been identified in the Feasibility
Study Update as the preferred method for tailings management.

-   Infrastructure:

The Wafi-Golpu Project is a greenfield project and currently does not have
infrastructure to support mining operations. Major Infrastructure is required
and included in the Feasibility Study Update, including:

•      access road;

•      ventilation and refrigeration plant;

•      processing plant (copper concentrator);

•     Deep Sea Tailings Placement system including tailings pipeline from
site to the discharge point near Lae;
•     concentrate export pipeline plus associated dewatering and loading
facilities at the existing Lae Port;

•      accommodation camp; and

•      on site power station.

The land in which the Project is located is mostly under customary land
title, some of which has been in dispute between customary land title holders
since mineral exploration began in the early 1980s. The compensation of
landholders is a requirement to the start of work however, Section 160 of the
Mining Act 1992 means that a dispute between customary land title holders of
this nature will not impede Project execution.

-   Costs:

Capital and Operating costs have been determined as part of the Feasibility
Study Update.

Capital cost estimates are based on multiple market prices across all
technical disciplines. Provision has been made for capital expenditure
requirements for new equipment, infrastructure and replacement of
infrastructure and equipment during the life of the mine. Contingency has
also been factored into the capital cost estimate consistent with the level
of accuracy of the study.

Operating cost estimate first principles cost modelling expenses have been
quantified as far as possible and where practicable supported by quotations.

Long term metal prices and exchange rate assumptions adopted in the Ore
Reserve estimate are the WGJV approved long term assumptions for the Project.

No cost impact is expected from deleterious elements. It has therefore not
been necessary to include realisation penalties (additional costs) relating
to minor elements when preparing the Ore Reserve estimate.

Transport and refining charges have been based on forecast supply and demand
assumptions.

The following allowances have been made for royalties payable in the
preparation of the Ore Reserve estimate:

•     Royalty of 2.00% of net smelter revenue (i.e. gross revenue from all
mining sales adjusted for realisation and freight charges).

•      Mining Levy of 0.25% of gross revenue from all mining sales.

-   Revenue factors:

Long term metal prices and exchange rate assumptions adopted in the Ore
Reserve estimate are the WGJV approved long term assumptions for the Project
and are:

•      USD1,200/oz for gold

•      USD3.00/lb for copper

•      USD/AUD 0.75
•      PGK/USD 3.10

The NSR calculation takes into account reserve revenue factors, metallurgical
recovery assumptions, transport costs and refining charges and royalty
charges.

Metallurgical test work analysis has indicated that the levels of deleterious
elements in concentrate did not exceed any of the typical concentration
restrictions for sale.

-   Market assessment:

Third party forecasts were used in the Feasibility Study Update (noting that
this information is commercial in confidence).

The Wafi-Golpu Project’s natural market for concentrate is Asia due to the
proximity of the mine to Asian region smelters.

The Wafi-Golpu Project is expected to achieve first ore milled approximately
4.75 years post SML grant. At such time, the Wafi-Golpu Project may face
competition from both new and established mines.

Concentrate volume forecasts were derived from the Feasibility Study Update
production schedule.

-   Economic:

The Ore Reserve has been evaluated through a financial model. All operating
and capital costs as well as revenue factors stated in this document were
included in the financial model. A discount factor of 8.5% real was applied.
This process demonstrated the Golpu Ore Reserve to have a positive NPV.

Sensitivities were conducted on the key input parameters including commodity
prices, capital and operating costs, ore grade, discount rate, exchange rate
and recovery which confirmed the estimate to be robust. The NPV range has not
been provided as it is commercially sensitive.

-   Social:

To assess the social and economic impacts of the Project upon communities,
the Feasibility Study Update included an in depth Social and Cultural
assessment, including leveraging off historical assessment work completed. In
addition an assessment of the potential economic impacts of the Project (if
developed) was undertaken by WGJV.

The land in which the Project is located is mostly under customary land
title, some of which has been in dispute since mineral exploration began in
the early 1980s. The compensation of landholders is a requirement to the
start of work however, Section 160 of the Mining Act 1992 means that a
dispute of this nature will not impede Project execution.

Other key agreements that will be required for project development include:
the Memorandum of Agreement (a Development Forum process) in relation to
benefits that might be accrued to effective landholders and a Mining
Development Contract with the Independent State of PNG.
The respect for all landowners, and regular engagement with them, will be
vital to the maintenance of a social licence to operate.

-   Other:

A Level 2B Environment Permit (EP) has been granted for exploration
activities.

Applications for a Special Mining Lease (SML) and related ancillary tenements
have been submitted by WGJV to the Mineral Resources Authority.

Approval of the Project by the PNG Government will be founded on the
assessment of the Environmental Impact Statement (EIS) due to be submitted by
the WGJV to the PNG government by the end of June 2018. This EIS will inform
government’s decision to grant an SML, related ancillary tenements and a
Level 3 EP. The grant of these key instruments is a prerequisite for
execution of the Project.

The Golpu deposit is located in a seismically active area in a region close
to a source of earthquakes that can produce seismic accelerations at the
site. This risk has been taken into account in infrastructure and mine
design.

-   Classification:

The Ore Reserve classification is based on Indicated Mineral Resources only.
No Measured Mineral Resources are stated for this deposit. This
classification is based on geological confidence as a function of continuity
and complexity of geological features; data spacing and distribution and
estimation quality parameters including distance to informing samples for
block grade estimation.

Diluting material has been included within the Probable Ore Reserve as mined
dilution due to the non-selective nature of block cave mining. This
represents 7% of the gold metal and 5% of the copper metal in the Ore
Reserve. Even without consideration of the metal contained in the dilution
incorporated in the Ore Reserve, the economic analysis indicates an economic
Probable Ore Reserve.

It is the Competent Person’s view that the classifications used for the Ore
Reserves are appropriate.

-   Audits or reviews:

SRK Consulting (Australasia) Pty Ltd (SRK) was commissioned to conduct an
independent review of the mining section of the Feasibility Study Update,
which included the Ore Reserve estimation processes and results.

SRK concluded that the Ore Reserve estimates had been prepared using normal
industry practice and has been appropriately classified as Probable Ore
Reserve. SRK did not identify any material issues with the estimate.

-   Discussion of relative accuracy/ confidence:

The accuracy of the estimates within this Ore Reserve is mostly determined by
the order of accuracy associated with the Mineral Resource model.
BC44 and BC42 are at a Feasibility confidence level (+/-15% accuracy), while
BC40 is at a Pre-Feasibility confidence level (+/- 25% accuracy).

The Competent Person views the Golpu Ore Reserve a reasonable assessment of
the global estimate.

The remaining areas of uncertainty at the current study stage are with the
geotechnical parameters for the mining area below 4200mRL (i.e. BC40) that
has been investigated to a Pre-Feasibility Study (PFS) level of confidence
and is constrained by an incomplete set of orebody data with rock strength
only being confirmed in the north east quadrant of the planned BC40
footprint. An analysis has indicated that based on the known rock strength
being extrapolated across the remaining areas of lower orebody knowledge,
mining is feasible. The Modifying Factors (key inputs) applied within PCBC™
cave scheduling software relies upon geology and geotechnical data such as
structural geology and rock mass strength.

Further orebody data is required to confirm the geological and geotechnical
information and is planned as part of the Forward Works Programme.

Golpu is a greenfields site and there is no previous production from the
Golpu deposit to compare relative accuracy and confidence.



About the Wafi-Golpu Project

Harmony and Newcrest each currently own 50% of Wafi-Golpu through the WGJV.

The State of PNG retains the right to purchase, at a pro rata share of
accumulated exploration expenditure, up to 30% equity interest in any mineral
discovery at Wafi-Golpu, at any time before the commencement of mining. If
the State of PNG chooses to take-up its full 30% interest, the interest of
each of Newcrest and Harmony will become 35%.

The Golpu deposit is located approximately 65km south-west of Lae in the
Morobe Province of PNG which is the second largest city in PNG and will host
the Wafi-Golpu export facilities. The proposed mine site sits at an elevation
of approximately 200 metres above sea level in moderately hilly terrain and
is located near the Watut River approximately 30km upstream from the
confluence of the Watut and Markham rivers.



For further information please contact

Lauren Fourie

+27 (0)71 607 1498

lauren.fourie@harmony.co.za

Marian van der Walt

+27 (0)82 888 1242

Marian@harmony.co.za
This information is available on our website at www.harmony.co.za

Competent Person’s Statement

The information in this report that relates to Golpu Mineral Resources is
based on information compiled by the Competent Person, Mr David Finn, who is
a member of The Australasian Institute of Mining and Metallurgy. Mr David
Finn, is a full-time employee of Newcrest Mining Limited or its relevant
subsidiaries. Mr David Finn has sufficient experience which is relevant to
the styles of mineralisation and type of deposit under consideration and to
the activity which he is undertaking to qualify as a Competent Person as
defined in the JORC Code 2012 and SAMREC 2016 (materially the same as the
JORC code).. Mr David Finn consents to the inclusion of material of the
matters based on his information in the form and context in which it appears.

The information in this report that relates to Golpu Ore Reserves is based on
information compiled by the Competent Person, Mr Pasqualino Manca, who is a
member of The Australasian Institute of Mining and Metallurgy. Mr Pasqualino
Manca, is a full-time employee of Newcrest Mining Limited or its relevant
subsidiaries, Mr Pasqualino Manca has sufficient experience which is relevant
to the styles of mineralisation and type of deposit under consideration and
to the activity which he is undertaking to qualify as a Competent Person as
defined in the JORC Code 2012 and SAMREC 2016 (materially the same as the
JORC code). Mr Pasqualino Manca consents to the inclusion of material of the
matters based on his information in the form and context in which it appears.

Mr Gregory Job, BSc, MSc, who has 29 years’ relevant experience and a member
of the Australian Institute of Mining and Metallurgy (AusIMM), is Harmony’s
competent person for Papua New Guinea.

Mr Jaco Boshoff, BSc (Hons), MSc, MBA, Pr. Sci. Nat, MSAIMM, MGSSA is
Harmony’s lead competent person. Mr Boshoff who has 22 years’ relevant
experience, is registered with the South African Council for Natural
Scientific Professions (SACNASP) and is a member of the South African
Institute of Mining and Metallurgy (SAIMM) and a member of the Geological
Society of South Africa (GSSA).
Forward Looking Statements

This report contains forward-looking statements within the meaning of the
safe harbor provided by Section 21E of the Securities Exchange Act of 1934,
as amended, and Section 27A of the Securities Act of 1933, as amended, with
respect to our financial condition, results of operations, business
strategies, operating efficiencies, competitive positions, growth
opportunities for existing services, plans and objectives of management,
markets for stock and other matters. These include all statements other than
statements of historical fact, including, without limitation, any statements
preceded by, followed by, or that include the words “targets”, “believes”,
“expects”, “aims”, “intends”, “will”, “may”, “anticipates”, “would”,
“should”, “could”, “estimates”, “forecast”, “predict”, “continue” or similar
expressions or the negative thereof.

These forward-looking statements, including, among others, those relating to
our future business prospects, revenues and income, wherever they may occur
in this report and the exhibits to this report, are essentially estimates
reflecting the best judgment of our senior management and involve a number of
risks and uncertainties that could cause actual results to differ materially
from those suggested by the forward-looking statements. As a consequence,
these forward-looking statements should be considered in light of various
important factors, including those set forth in this presentation. Important
factors that could cause actual results to differ materially from estimates
or projections contained in the forward-looking statements include, without
limitation: overall economic and business conditions in South Africa, Papua
New Guinea, Australia and elsewhere, estimates of future earnings, and the
sensitivity of earnings to the gold and other metals prices, estimates of
future gold and other metals production and sales, estimates of future cash
costs, estimates of future cash flows, and the sensitivity of cash flows to
the gold and other metals prices, statements regarding future debt
repayments, estimates of future capital expenditures, the success of our
business strategy, development activities and other initiatives, estimates of
reserves statements regarding future exploration results and the replacement
of reserves, the ability to achieve anticipated efficiencies and other cost
savings in connection with past and future acquisitions, fluctuations in the
market price of gold, the occurrence of hazards associated with underground
and surface gold mining, the occurrence of labor disruptions, power cost
increases as well as power stoppages, fluctuations and usage constraints,
supply chain shortages and increases in the prices of production imports,
availability, terms and deployment of capital, changes in government
regulation, particularly mining rights and environmental regulation,
fluctuations in exchange rates, the adequacy of the Group’s insurance
coverage and socio-economic or political instability in South Africa and
Papua New Guinea and other countries in which we operate.

For a more detailed discussion of such risks and other factors (such as
availability of credit or other sources of financing), see the Company’s
latest Integrated Annual Report and Form 20-F which is on file with the
Securities and Exchange Commission, as well as the Company’s other Securities
and Exchange Commission filings. The Company undertakes no obligation to
update publicly or release any revisions to these forward-looking statements
to reflect events or circumstances after the date of this presentation or to
reflect the occurrence of unanticipated events, except as required by law.
Ends.



12 April 2018



Corporate office:

Randfontein Office Park

P O Box 2

Randfontein

South Africa 1760

T +27 (11) 411 2000



Listing codes:

JSE:HAR

NYSE:HMY



ISIN no:

ZAE000015228

Registration no: 1950/038232/06

JSE Sponsor: J.P. Morgan Equities South Africa Propriety Limited

Date: 12/04/2018 04:52:00 Produced by the JSE SENS Department. The SENS service is an information dissemination service administered by the JSE Limited ('JSE'). 
The JSE does not, whether expressly, tacitly or implicitly, represent, warrant or in any way guarantee the truth, accuracy or completeness of
 the information published on SENS. The JSE, their officers, employees and agents accept no liability for (or in respect of) any direct, 
indirect, incidental or consequential loss or damage of any kind or nature, howsoever arising, from the use of SENS or the use of, or reliance on,
 information disseminated through SENS.

                                        
Email this JSE Sens Item to a Friend.

Send e-mail to
© 2019 SHARENET (PTY) Ltd, Cape Town, South Africa
Home     Terms & conditions    Privacy Policy
    Security Notice    Contact Details
Market Statistics are calculated by Sharenet and are therefore not the official JSE Market Statistics. The calculation/derivation may include underlying JSE data.